From fddfdeafa8396f85c666bfc5e1e920eb535514cf Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 31 Jan 2006 15:24:34 +0100 Subject: [BLOCK] A few kerneldoc fixups Signed-off-by: Jens Axboe --- fs/bio.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/bio.c b/fs/bio.c index bbc442b8c867..1f3bb501c262 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -411,6 +411,7 @@ static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page /** * bio_add_pc_page - attempt to add page to bio + * @q: the target queue * @bio: destination bio * @page: page to add * @len: vec entry length -- cgit v1.2.3-59-g8ed1b From 3a69c7dc6f3d58aeb9ce5051fc7060d55e05c003 Mon Sep 17 00:00:00 2001 From: Yingping Lu Date: Wed, 1 Feb 2006 12:14:34 +1100 Subject: [XFS] Interim solution for attribute insertion failure during file creation due to ENOSPC. The current solution removes the inode when the attribute insertion fails. Long term solution would be to make the inode creation and attribute insertion atomic. SGI-PV: 947610 SGI-Modid: xfs-linux-melb:xfs-kern:205193a Signed-off-by: Yingping Lu Signed-off-by: Nathan Scott --- fs/xfs/linux-2.6/xfs_iops.c | 50 +++++++++++++++++++++++++++++---------------- 1 file changed, 32 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 76c6df34d0db..eda7919b70a1 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -262,6 +262,31 @@ has_fs_struct(struct task_struct *task) return (task->fs != init_task.fs); } +STATIC inline void +cleanup_inode( + vnode_t *dvp, + vnode_t *vp, + struct dentry *dentry, + int mode) +{ + struct dentry teardown = {}; + int err2; + + /* Oh, the horror. + * If we can't add the ACL or we fail in + * linvfs_init_security we must back out. + * ENOSPC can hit here, among other things. + */ + teardown.d_inode = LINVFS_GET_IP(vp); + teardown.d_name = dentry->d_name; + + if (S_ISDIR(mode)) + VOP_RMDIR(dvp, &teardown, NULL, err2); + else + VOP_REMOVE(dvp, &teardown, NULL, err2); + VN_RELE(vp); +} + STATIC int linvfs_mknod( struct inode *dir, @@ -316,30 +341,19 @@ linvfs_mknod( } if (!error) + { error = linvfs_init_security(vp, dir); + if (error) + cleanup_inode(dvp, vp, dentry, mode); + } if (default_acl) { if (!error) { error = _ACL_INHERIT(vp, &va, default_acl); - if (!error) { + if (!error) VMODIFY(vp); - } else { - struct dentry teardown = {}; - int err2; - - /* Oh, the horror. - * If we can't add the ACL we must back out. - * ENOSPC can hit here, among other things. - */ - teardown.d_inode = ip = LINVFS_GET_IP(vp); - teardown.d_name = dentry->d_name; - - if (S_ISDIR(mode)) - VOP_RMDIR(dvp, &teardown, NULL, err2); - else - VOP_REMOVE(dvp, &teardown, NULL, err2); - VN_RELE(vp); - } + else + cleanup_inode(dvp, vp, dentry, mode); } _ACL_FREE(default_acl); } -- cgit v1.2.3-59-g8ed1b From fad3aa1e8e2e4123a19b926fefd91ec63dd56497 Mon Sep 17 00:00:00 2001 From: Nathan Scott Date: Wed, 1 Feb 2006 12:14:52 +1100 Subject: [XFS] Fix regression in xfs_buf_rele dealing with non-hashed buffers, as occur during log replay. Novell bug 145204, Fedora bug 177848. SGI-PV: 948860 SGI-Modid: xfs-linux-melb:xfs-kern:25064a Signed-off-by: Nathan Scott --- fs/xfs/linux-2.6/xfs_buf.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'fs') diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index e44b7c1a3a36..a36a8e3b703f 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -822,6 +822,13 @@ xfs_buf_rele( XB_TRACE(bp, "rele", bp->b_relse); + if (unlikely(!hash)) { + ASSERT(!bp->b_relse); + if (atomic_dec_and_test(&bp->b_hold)) + xfs_buf_free(bp); + return; + } + if (atomic_dec_and_lock(&bp->b_hold, &hash->bh_lock)) { if (bp->b_relse) { atomic_inc(&bp->b_hold); -- cgit v1.2.3-59-g8ed1b From 3fb803a990cd17546bd89c38e0e29a891f71ce7d Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 1 Feb 2006 03:04:34 -0800 Subject: [PATCH] knfsd: Restore recently broken ACL functionality to NFS server A recent patch to Allow run-time selection of NFS versions to export meant that NO nfsacl service versions were exported. This patch restored that functionality. Signed-off-by: Andreas Gruenbacher Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfssvc.c | 76 ++++++++++++++++++++++++++++++++++++-------------------- 1 file changed, 49 insertions(+), 27 deletions(-) (limited to 'fs') diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 89ed04696865..1d163b616915 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -64,6 +64,32 @@ struct nfsd_list { }; static struct list_head nfsd_list = LIST_HEAD_INIT(nfsd_list); +#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) +static struct svc_stat nfsd_acl_svcstats; +static struct svc_version * nfsd_acl_version[] = { + [2] = &nfsd_acl_version2, + [3] = &nfsd_acl_version3, +}; + +#define NFSD_ACL_MINVERS 2 +#define NFSD_ACL_NRVERS (sizeof(nfsd_acl_version)/sizeof(nfsd_acl_version[0])) +static struct svc_version *nfsd_acl_versions[NFSD_ACL_NRVERS]; + +static struct svc_program nfsd_acl_program = { + .pg_prog = NFS_ACL_PROGRAM, + .pg_nvers = NFSD_ACL_NRVERS, + .pg_vers = nfsd_acl_versions, + .pg_name = "nfsd", + .pg_class = "nfsd", + .pg_stats = &nfsd_acl_svcstats, + .pg_authenticate = &svc_set_client, +}; + +static struct svc_stat nfsd_acl_svcstats = { + .program = &nfsd_acl_program, +}; +#endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */ + static struct svc_version * nfsd_version[] = { [2] = &nfsd_version2, #if defined(CONFIG_NFSD_V3) @@ -79,6 +105,9 @@ static struct svc_version * nfsd_version[] = { static struct svc_version *nfsd_versions[NFSD_NRVERS]; struct svc_program nfsd_program = { +#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) + .pg_next = &nfsd_acl_program, +#endif .pg_prog = NFS_PROGRAM, /* program number */ .pg_nvers = NFSD_NRVERS, /* nr of entries in nfsd_version */ .pg_vers = nfsd_versions, /* version table */ @@ -147,6 +176,26 @@ nfsd_svc(unsigned short port, int nrservs) nfsd_program.pg_vers[i] = nfsd_version[i]; } + +#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) + found_one = 0; + + for (i = NFSD_ACL_MINVERS; i < NFSD_ACL_NRVERS; i++) { + if (NFSCTL_VERISSET(nfsd_versbits, i)) { + nfsd_acl_program.pg_vers[i] = + nfsd_acl_version[i]; + found_one = 1; + } else + nfsd_acl_program.pg_vers[i] = NULL; + } + + if (!found_one) { + for (i = NFSD_ACL_MINVERS; i < NFSD_ACL_NRVERS; i++) + nfsd_acl_program.pg_vers[i] = + nfsd_acl_version[i]; + } +#endif + atomic_set(&nfsd_busy, 0); error = -ENOMEM; nfsd_serv = svc_create(&nfsd_program, NFSD_BUFSIZE); @@ -411,30 +460,3 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) nfsd_cache_update(rqstp, proc->pc_cachetype, statp + 1); return 1; } - -#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) -static struct svc_stat nfsd_acl_svcstats; -static struct svc_version * nfsd_acl_version[] = { - [2] = &nfsd_acl_version2, - [3] = &nfsd_acl_version3, -}; - -#define NFSD_ACL_NRVERS (sizeof(nfsd_acl_version)/sizeof(nfsd_acl_version[0])) -static struct svc_program nfsd_acl_program = { - .pg_prog = NFS_ACL_PROGRAM, - .pg_nvers = NFSD_ACL_NRVERS, - .pg_vers = nfsd_acl_version, - .pg_name = "nfsd", - .pg_class = "nfsd", - .pg_stats = &nfsd_acl_svcstats, - .pg_authenticate = &svc_set_client, -}; - -static struct svc_stat nfsd_acl_svcstats = { - .program = &nfsd_acl_program, -}; - -#define nfsd_acl_program_p &nfsd_acl_program -#else -#define nfsd_acl_program_p NULL -#endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */ -- cgit v1.2.3-59-g8ed1b From caf736085f2f0d22a992a855d9caae14973f7ea4 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 1 Feb 2006 03:04:39 -0800 Subject: [PATCH] smbfs readdir vs signal fix An old patch designed to fix http://bugme.osdl.org/show_bug.cgi?id=4497, "getdents gives empty/random result upon signal". If smbfs's readdir() is interupted by a signal, smb_readdir() failed to noticed that and proceeded to treat the unread-into page as valid directory contents. Fix that up by handling the -ERESTARTSYS. Thanks to Stian Skjelstad for reporting and testing. Cc: Stian Skjelstad Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/smbfs/dir.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/smbfs/dir.c b/fs/smbfs/dir.c index c6c33e15143a..0424d06b147e 100644 --- a/fs/smbfs/dir.c +++ b/fs/smbfs/dir.c @@ -209,6 +209,8 @@ init_cache: ctl.valid = 1; read_really: result = server->ops->readdir(filp, dirent, filldir, &ctl); + if (result == -ERESTARTSYS && page) + ClearPageUptodate(page); if (ctl.idx == -1) goto invalid_cache; /* retry */ ctl.head.end = ctl.fpos - 1; @@ -217,7 +219,8 @@ finished: if (page) { cache->head = ctl.head; kunmap(page); - SetPageUptodate(page); + if (result != -ERESTARTSYS) + SetPageUptodate(page); unlock_page(page); page_cache_release(page); } -- cgit v1.2.3-59-g8ed1b From 9cd684551124e71630ab96d238747051463f5b56 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 1 Feb 2006 03:04:40 -0800 Subject: [PATCH] fuse: fix async read for legacy filesystems While asynchronous reads mean a performance improvement in most cases, if the filesystem assumed that reads are synchronous, then async reads may degrade performance (filesystem may receive reads out of order, which can confuse it's own readahead logic). With sshfs a 1.5 to 4 times slowdown can be measured. There's also a need for userspace filesystems to know whether asynchronous reads are supported by the kernel or not. To achive these, negotiate in the INIT request whether async reads will be used and the maximum readahead value. Update interface version to 7.6 If userspace uses a version earlier than 7.6, then disable async reads, and set maximum readahead value to the maximum read size, as done in previous versions. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/file.c | 9 +++++++-- fs/fuse/fuse_i.h | 3 +++ fs/fuse/inode.c | 14 ++++++++++++-- include/linux/fuse.h | 16 ++++++++++++++-- 4 files changed, 36 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index a7ef5e716f3c..296351615b00 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -335,9 +335,14 @@ static void fuse_send_readpages(struct fuse_req *req, struct file *file, loff_t pos = page_offset(req->pages[0]); size_t count = req->num_pages << PAGE_CACHE_SHIFT; req->out.page_zeroing = 1; - req->end = fuse_readpages_end; fuse_read_fill(req, file, inode, pos, count, FUSE_READ); - request_send_background(fc, req); + if (fc->async_read) { + req->end = fuse_readpages_end; + request_send_background(fc, req); + } else { + request_send(fc, req); + fuse_readpages_end(fc, req); + } } struct fuse_readpages_data { diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 46cf933aa3bf..4a83adfec968 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -272,6 +272,9 @@ struct fuse_conn { reply, before any other request, and never cleared */ unsigned conn_error : 1; + /** Do readpages asynchronously? Only set in INIT */ + unsigned async_read : 1; + /* * The following bitfields are only for optimization purposes * and hence races in setting them will not cause malfunction diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index c755a0440a66..879e6fba9480 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -473,6 +473,16 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) if (req->out.h.error || arg->major != FUSE_KERNEL_VERSION) fc->conn_error = 1; else { + unsigned long ra_pages; + + if (arg->minor >= 6) { + ra_pages = arg->max_readahead / PAGE_CACHE_SIZE; + if (arg->flags & FUSE_ASYNC_READ) + fc->async_read = 1; + } else + ra_pages = fc->max_read / PAGE_CACHE_SIZE; + + fc->bdi.ra_pages = min(fc->bdi.ra_pages, ra_pages); fc->minor = arg->minor; fc->max_write = arg->minor < 5 ? 4096 : arg->max_write; } @@ -496,6 +506,8 @@ static void fuse_send_init(struct fuse_conn *fc) arg->major = FUSE_KERNEL_VERSION; arg->minor = FUSE_KERNEL_MINOR_VERSION; + arg->max_readahead = fc->bdi.ra_pages * PAGE_CACHE_SIZE; + arg->flags |= FUSE_ASYNC_READ; req->in.h.opcode = FUSE_INIT; req->in.numargs = 1; req->in.args[0].size = sizeof(*arg); @@ -552,8 +564,6 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) fc->user_id = d.user_id; fc->group_id = d.group_id; fc->max_read = d.max_read; - if (fc->max_read / PAGE_CACHE_SIZE < fc->bdi.ra_pages) - fc->bdi.ra_pages = fc->max_read / PAGE_CACHE_SIZE; /* Used by get_root_inode() */ sb->s_fs_info = fc; diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 528959c52f1b..5425b60021e3 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -14,7 +14,7 @@ #define FUSE_KERNEL_VERSION 7 /** Minor version number of this interface */ -#define FUSE_KERNEL_MINOR_VERSION 5 +#define FUSE_KERNEL_MINOR_VERSION 6 /** The node ID of the root inode */ #define FUSE_ROOT_ID 1 @@ -58,6 +58,9 @@ struct fuse_kstatfs { __u32 spare[6]; }; +/** + * Bitmasks for fuse_setattr_in.valid + */ #define FATTR_MODE (1 << 0) #define FATTR_UID (1 << 1) #define FATTR_GID (1 << 2) @@ -75,6 +78,11 @@ struct fuse_kstatfs { #define FOPEN_DIRECT_IO (1 << 0) #define FOPEN_KEEP_CACHE (1 << 1) +/** + * INIT request/reply flags + */ +#define FUSE_ASYNC_READ (1 << 0) + enum fuse_opcode { FUSE_LOOKUP = 1, FUSE_FORGET = 2, /* no reply */ @@ -247,12 +255,16 @@ struct fuse_access_in { struct fuse_init_in { __u32 major; __u32 minor; + __u32 max_readahead; + __u32 flags; }; struct fuse_init_out { __u32 major; __u32 minor; - __u32 unused[3]; + __u32 max_readahead; + __u32 flags; + __u32 unused; __u32 max_write; }; -- cgit v1.2.3-59-g8ed1b From cb82a6cdf994d6656ad0a25ed28395af3416a27c Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 1 Feb 2006 03:04:49 -0800 Subject: [PATCH] compat_sys_pselect7() fix fs/compat.c: In function `compat_sys_pselect7': fs/compat.c:1820: warning: passing arg 5 of `compat_core_sys_select' from incompatible pointer type Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/compat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/compat.c b/fs/compat.c index ff0bafcff720..cc58a20df57a 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1781,7 +1781,7 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, { compat_sigset_t ss32; sigset_t ksigmask, sigsaved; - long timeout = MAX_SCHEDULE_TIMEOUT; + s64 timeout = MAX_SCHEDULE_TIMEOUT; struct compat_timespec ts; int ret; -- cgit v1.2.3-59-g8ed1b From 537421be79b94bcf620467f50dd9e38b739c2a00 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 1 Feb 2006 03:05:24 -0800 Subject: [PATCH] Mark CONFIG_UFS_FS_WRITE as BROKEN OpenBSD doesn't see "." correctly in directories created by Linux. Copying files over several KB will buy you infinite loop in __getblk_slow(). Copying files smaller than 1 KB seems to be OK. Sometimes files will be filled with zeros. Sometimes incorrectly copied file will reappear after next file with truncated size. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/Kconfig b/fs/Kconfig index ef78e3a42d32..93b5dc4082ff 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1327,7 +1327,7 @@ config UFS_FS config UFS_FS_WRITE bool "UFS file system write support (DANGEROUS)" - depends on UFS_FS && EXPERIMENTAL + depends on UFS_FS && EXPERIMENTAL && BROKEN help Say Y here if you want to try writing to UFS partitions. This is experimental, so you should back up your UFS partitions beforehand. -- cgit v1.2.3-59-g8ed1b From 4e6a510a74145585f4111d60d1b5fd450d795dd8 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 1 Feb 2006 03:05:31 -0800 Subject: [PATCH] mm: hugepage accounting fix 2.6.15's hugepage faulting introduced huge_pages_needed accounting into hugetlbfs: to count how many pages are already in cache, for spot check on how far a new mapping may be allowed to extend the file. But it's muddled: each hugepage found covers HPAGE_SIZE, not PAGE_SIZE. Once pages were already in cache, it would overshoot, wrap its hugepages count backwards, and so fail a harmless repeat mapping with -ENOMEM. Fixes the problem found by Don Dupuis. Signed-off-by: Hugh Dickins Acked-By: Adam Litke Acked-by: William Irwin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index f568102da1e8..b35195289945 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -72,8 +72,8 @@ huge_pages_needed(struct address_space *mapping, struct vm_area_struct *vma) unsigned long start = vma->vm_start; unsigned long end = vma->vm_end; unsigned long hugepages = (end - start) >> HPAGE_SHIFT; - pgoff_t next = vma->vm_pgoff; - pgoff_t endpg = next + ((end - start) >> PAGE_SHIFT); + pgoff_t next = vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT); + pgoff_t endpg = next + hugepages; pagevec_init(&pvec, 0); while (next < endpg) { -- cgit v1.2.3-59-g8ed1b From e965f9630c651fa4249039fd4b80c9392d07a856 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 1 Feb 2006 03:05:41 -0800 Subject: [PATCH] Direct Migration V9: Avoid writeback / page_migrate() method Migrate a page with buffers without requiring writeback This introduces a new address space operation migratepage() that may be used by a filesystem to implement its own version of page migration. A version is provided that migrates buffers attached to pages. Some filesystems (ext2, ext3, xfs) are modified to utilize this feature. The swapper address space operation are modified so that a regular migrate_page() will occur for anonymous pages without writeback (migrate_pages forces every anonymous page to have a swap entry). Signed-off-by: Mike Kravetz Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 60 +++++++++++++++++++++++++++++++++++++++++++++ fs/ext2/inode.c | 2 ++ fs/ext3/inode.c | 2 ++ fs/xfs/linux-2.6/xfs_aops.c | 1 + fs/xfs/linux-2.6/xfs_buf.c | 1 + include/linux/fs.h | 8 ++++++ include/linux/swap.h | 5 ++++ mm/rmap.c | 1 + mm/swap_state.c | 1 + mm/vmscan.c | 20 ++++++++++++++- 10 files changed, 100 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 3dc712f29d2d..8bcbac87a28c 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -3049,6 +3049,66 @@ asmlinkage long sys_bdflush(int func, long data) return 0; } +/* + * Migration function for pages with buffers. This function can only be used + * if the underlying filesystem guarantees that no other references to "page" + * exist. + */ +#ifdef CONFIG_MIGRATION +int buffer_migrate_page(struct page *newpage, struct page *page) +{ + struct address_space *mapping = page->mapping; + struct buffer_head *bh, *head; + + if (!mapping) + return -EAGAIN; + + if (!page_has_buffers(page)) + return migrate_page(newpage, page); + + head = page_buffers(page); + + if (migrate_page_remove_references(newpage, page, 3)) + return -EAGAIN; + + bh = head; + do { + get_bh(bh); + lock_buffer(bh); + bh = bh->b_this_page; + + } while (bh != head); + + ClearPagePrivate(page); + set_page_private(newpage, page_private(page)); + set_page_private(page, 0); + put_page(page); + get_page(newpage); + + bh = head; + do { + set_bh_page(bh, newpage, bh_offset(bh)); + bh = bh->b_this_page; + + } while (bh != head); + + SetPagePrivate(newpage); + + migrate_page_copy(newpage, page); + + bh = head; + do { + unlock_buffer(bh); + put_bh(bh); + bh = bh->b_this_page; + + } while (bh != head); + + return 0; +} +EXPORT_SYMBOL(buffer_migrate_page); +#endif + /* * Buffer-head allocation */ diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index e7d3f0522d01..a717837f272e 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -706,6 +706,7 @@ struct address_space_operations ext2_aops = { .bmap = ext2_bmap, .direct_IO = ext2_direct_IO, .writepages = ext2_writepages, + .migratepage = buffer_migrate_page, }; struct address_space_operations ext2_aops_xip = { @@ -723,6 +724,7 @@ struct address_space_operations ext2_nobh_aops = { .bmap = ext2_bmap, .direct_IO = ext2_direct_IO, .writepages = ext2_writepages, + .migratepage = buffer_migrate_page, }; /* diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 8824e84f8a56..3fc4238e9703 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1559,6 +1559,7 @@ static struct address_space_operations ext3_ordered_aops = { .invalidatepage = ext3_invalidatepage, .releasepage = ext3_releasepage, .direct_IO = ext3_direct_IO, + .migratepage = buffer_migrate_page, }; static struct address_space_operations ext3_writeback_aops = { @@ -1572,6 +1573,7 @@ static struct address_space_operations ext3_writeback_aops = { .invalidatepage = ext3_invalidatepage, .releasepage = ext3_releasepage, .direct_IO = ext3_direct_IO, + .migratepage = buffer_migrate_page, }; static struct address_space_operations ext3_journalled_aops = { diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 120626789406..9892268e3005 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1462,4 +1462,5 @@ struct address_space_operations linvfs_aops = { .commit_write = generic_commit_write, .bmap = linvfs_bmap, .direct_IO = linvfs_direct_IO, + .migratepage = buffer_migrate_page, }; diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index a36a8e3b703f..bfb4f2917bb6 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -1521,6 +1521,7 @@ xfs_mapping_buftarg( struct address_space *mapping; static struct address_space_operations mapping_aops = { .sync_page = block_sync_page, + .migratepage = fail_migrate_page, }; inode = new_inode(bdev->bd_inode->i_sb); diff --git a/include/linux/fs.h b/include/linux/fs.h index 84bb449b9b01..e059da947007 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -363,6 +363,8 @@ struct address_space_operations { loff_t offset, unsigned long nr_segs); struct page* (*get_xip_page)(struct address_space *, sector_t, int); + /* migrate the contents of a page to the specified target */ + int (*migratepage) (struct page *, struct page *); }; struct backing_dev_info; @@ -1719,6 +1721,12 @@ extern void simple_release_fs(struct vfsmount **mount, int *count); extern ssize_t simple_read_from_buffer(void __user *, size_t, loff_t *, const void *, size_t); +#ifdef CONFIG_MIGRATION +extern int buffer_migrate_page(struct page *, struct page *); +#else +#define buffer_migrate_page NULL +#endif + extern int inode_change_ok(struct inode *, struct iattr *); extern int __must_check inode_setattr(struct inode *, struct iattr *); diff --git a/include/linux/swap.h b/include/linux/swap.h index 229b6d04b4b6..f3e17d5963c3 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -193,13 +193,18 @@ extern int isolate_lru_page(struct page *p); extern int putback_lru_pages(struct list_head *l); extern int migrate_page(struct page *, struct page *); extern void migrate_page_copy(struct page *, struct page *); +extern int migrate_page_remove_references(struct page *, struct page *, int); extern int migrate_pages(struct list_head *l, struct list_head *t, struct list_head *moved, struct list_head *failed); +extern int fail_migrate_page(struct page *, struct page *); #else static inline int isolate_lru_page(struct page *p) { return -ENOSYS; } static inline int putback_lru_pages(struct list_head *l) { return 0; } static inline int migrate_pages(struct list_head *l, struct list_head *t, struct list_head *moved, struct list_head *failed) { return -ENOSYS; } +/* Possible settings for the migrate_page() method in address_operations */ +#define migrate_page NULL +#define fail_migrate_page NULL #endif #ifdef CONFIG_MMU diff --git a/mm/rmap.c b/mm/rmap.c index f4b91d7aa5cf..df2c41c2a9a2 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -233,6 +233,7 @@ void remove_from_swap(struct page *page) delete_from_swap_cache(page); } +EXPORT_SYMBOL(remove_from_swap); #endif /* diff --git a/mm/swap_state.c b/mm/swap_state.c index 7b09ac503fec..db8a3d3e1636 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -27,6 +27,7 @@ static struct address_space_operations swap_aops = { .writepage = swap_writepage, .sync_page = block_sync_page, .set_page_dirty = __set_page_dirty_nobuffers, + .migratepage = migrate_page, }; static struct backing_dev_info swap_backing_dev_info = { diff --git a/mm/vmscan.c b/mm/vmscan.c index 5e98b86feb74..5a610804cd06 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -614,6 +614,15 @@ int putback_lru_pages(struct list_head *l) return count; } +/* + * Non migratable page + */ +int fail_migrate_page(struct page *newpage, struct page *page) +{ + return -EIO; +} +EXPORT_SYMBOL(fail_migrate_page); + /* * swapout a single page * page is locked upon entry, unlocked on exit @@ -659,6 +668,7 @@ unlock_retry: retry: return -EAGAIN; } +EXPORT_SYMBOL(swap_page); /* * Page migration was first developed in the context of the memory hotplug @@ -674,7 +684,7 @@ retry: * Remove references for a page and establish the new page with the correct * basic settings to be able to stop accesses to the page. */ -static int migrate_page_remove_references(struct page *newpage, +int migrate_page_remove_references(struct page *newpage, struct page *page, int nr_refs) { struct address_space *mapping = page_mapping(page); @@ -749,6 +759,7 @@ static int migrate_page_remove_references(struct page *newpage, return 0; } +EXPORT_SYMBOL(migrate_page_remove_references); /* * Copy the page to its new location @@ -788,6 +799,7 @@ void migrate_page_copy(struct page *newpage, struct page *page) if (PageWriteback(newpage)) end_page_writeback(newpage); } +EXPORT_SYMBOL(migrate_page_copy); /* * Common logic to directly migrate a single page suitable for @@ -815,6 +827,7 @@ int migrate_page(struct page *newpage, struct page *page) remove_from_swap(newpage); return 0; } +EXPORT_SYMBOL(migrate_page); /* * migrate_pages @@ -914,6 +927,11 @@ redo: if (!mapping) goto unlock_both; + if (mapping->a_ops->migratepage) { + rc = mapping->a_ops->migratepage(newpage, page); + goto unlock_both; + } + /* * Trigger writeout if page is dirty */ -- cgit v1.2.3-59-g8ed1b From d739b42b82773206297db1fc0d96ef895a5d9688 Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Wed, 1 Feb 2006 03:06:43 -0800 Subject: [PATCH] reiserfs: remove kmalloc wrapper Remove kmalloc() wrapper from fs/reiserfs/. Please note that a reiserfs /proc entry format is changed because kmalloc statistics is removed. Signed-off-by: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/dir.c | 16 +++----- fs/reiserfs/fix_node.c | 50 ++----------------------- fs/reiserfs/journal.c | 84 ++++++++++++++---------------------------- fs/reiserfs/namei.c | 16 ++++---- fs/reiserfs/procfs.c | 3 +- fs/reiserfs/super.c | 6 --- fs/reiserfs/xattr.c | 11 ++---- include/linux/reiserfs_fs.h | 16 -------- include/linux/reiserfs_fs_sb.h | 1 - 9 files changed, 49 insertions(+), 154 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c index 9dd71e807034..d71ac6579289 100644 --- a/fs/reiserfs/dir.c +++ b/fs/reiserfs/dir.c @@ -150,18 +150,15 @@ static int reiserfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (d_reclen <= 32) { local_buf = small_buf; } else { - local_buf = - reiserfs_kmalloc(d_reclen, GFP_NOFS, - inode->i_sb); + local_buf = kmalloc(d_reclen, + GFP_NOFS); if (!local_buf) { pathrelse(&path_to_entry); ret = -ENOMEM; goto out; } if (item_moved(&tmp_ih, &path_to_entry)) { - reiserfs_kfree(local_buf, - d_reclen, - inode->i_sb); + kfree(local_buf); goto research; } } @@ -174,15 +171,12 @@ static int reiserfs_readdir(struct file *filp, void *dirent, filldir_t filldir) (dirent, local_buf, d_reclen, d_off, d_ino, DT_UNKNOWN) < 0) { if (local_buf != small_buf) { - reiserfs_kfree(local_buf, - d_reclen, - inode->i_sb); + kfree(local_buf); } goto end; } if (local_buf != small_buf) { - reiserfs_kfree(local_buf, d_reclen, - inode->i_sb); + kfree(local_buf); } // next entry should be looked for with such offset next_pos = deh_offset(deh) + 1; diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c index 45829889dcdc..aa22588019ec 100644 --- a/fs/reiserfs/fix_node.c +++ b/fs/reiserfs/fix_node.c @@ -2021,38 +2021,6 @@ static int get_neighbors(struct tree_balance *p_s_tb, int n_h) return CARRY_ON; } -#ifdef CONFIG_REISERFS_CHECK -void *reiserfs_kmalloc(size_t size, gfp_t flags, struct super_block *s) -{ - void *vp; - static size_t malloced; - - vp = kmalloc(size, flags); - if (vp) { - REISERFS_SB(s)->s_kmallocs += size; - if (REISERFS_SB(s)->s_kmallocs > malloced + 200000) { - reiserfs_warning(s, - "vs-8301: reiserfs_kmalloc: allocated memory %d", - REISERFS_SB(s)->s_kmallocs); - malloced = REISERFS_SB(s)->s_kmallocs; - } - } - return vp; -} - -void reiserfs_kfree(const void *vp, size_t size, struct super_block *s) -{ - kfree(vp); - - REISERFS_SB(s)->s_kmallocs -= size; - if (REISERFS_SB(s)->s_kmallocs < 0) - reiserfs_warning(s, - "vs-8302: reiserfs_kfree: allocated memory %d", - REISERFS_SB(s)->s_kmallocs); - -} -#endif - static int get_virtual_node_size(struct super_block *sb, struct buffer_head *bh) { int max_num_of_items; @@ -2086,7 +2054,7 @@ static int get_mem_for_virtual_node(struct tree_balance *tb) /* we have to allocate more memory for virtual node */ if (tb->vn_buf) { /* free memory allocated before */ - reiserfs_kfree(tb->vn_buf, tb->vn_buf_size, tb->tb_sb); + kfree(tb->vn_buf); /* this is not needed if kfree is atomic */ check_fs = 1; } @@ -2095,24 +2063,15 @@ static int get_mem_for_virtual_node(struct tree_balance *tb) tb->vn_buf_size = size; /* get memory for virtual item */ - buf = - reiserfs_kmalloc(size, GFP_ATOMIC | __GFP_NOWARN, - tb->tb_sb); + buf = kmalloc(size, GFP_ATOMIC | __GFP_NOWARN); if (!buf) { /* getting memory with GFP_KERNEL priority may involve balancing now (due to indirect_to_direct conversion on dcache shrinking). So, release path and collected resources here */ free_buffers_in_tb(tb); - buf = reiserfs_kmalloc(size, GFP_NOFS, tb->tb_sb); + buf = kmalloc(size, GFP_NOFS); if (!buf) { -#ifdef CONFIG_REISERFS_CHECK - reiserfs_warning(tb->tb_sb, - "vs-8345: get_mem_for_virtual_node: " - "kmalloc failed. reiserfs kmalloced %d bytes", - REISERFS_SB(tb->tb_sb)-> - s_kmallocs); -#endif tb->vn_buf_size = 0; } tb->vn_buf = buf; @@ -2619,7 +2578,6 @@ void unfix_nodes(struct tree_balance *tb) } } - if (tb->vn_buf) - reiserfs_kfree(tb->vn_buf, tb->vn_buf_size, tb->tb_sb); + kfree(tb->vn_buf); } diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 4491fcf2a0e6..16b526fd20b9 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -152,18 +152,16 @@ static struct reiserfs_bitmap_node *allocate_bitmap_node(struct super_block struct reiserfs_bitmap_node *bn; static int id; - bn = reiserfs_kmalloc(sizeof(struct reiserfs_bitmap_node), GFP_NOFS, - p_s_sb); + bn = kmalloc(sizeof(struct reiserfs_bitmap_node), GFP_NOFS); if (!bn) { return NULL; } - bn->data = reiserfs_kmalloc(p_s_sb->s_blocksize, GFP_NOFS, p_s_sb); + bn->data = kzalloc(p_s_sb->s_blocksize, GFP_NOFS); if (!bn->data) { - reiserfs_kfree(bn, sizeof(struct reiserfs_bitmap_node), p_s_sb); + kfree(bn); return NULL; } bn->id = id++; - memset(bn->data, 0, p_s_sb->s_blocksize); INIT_LIST_HEAD(&bn->list); return bn; } @@ -197,8 +195,8 @@ static inline void free_bitmap_node(struct super_block *p_s_sb, struct reiserfs_journal *journal = SB_JOURNAL(p_s_sb); journal->j_used_bitmap_nodes--; if (journal->j_free_bitmap_nodes > REISERFS_MAX_BITMAP_NODES) { - reiserfs_kfree(bn->data, p_s_sb->s_blocksize, p_s_sb); - reiserfs_kfree(bn, sizeof(struct reiserfs_bitmap_node), p_s_sb); + kfree(bn->data); + kfree(bn); } else { list_add(&bn->list, &journal->j_bitmap_nodes); journal->j_free_bitmap_nodes++; @@ -276,8 +274,8 @@ static int free_bitmap_nodes(struct super_block *p_s_sb) while (next != &journal->j_bitmap_nodes) { bn = list_entry(next, struct reiserfs_bitmap_node, list); list_del(next); - reiserfs_kfree(bn->data, p_s_sb->s_blocksize, p_s_sb); - reiserfs_kfree(bn, sizeof(struct reiserfs_bitmap_node), p_s_sb); + kfree(bn->data); + kfree(bn); next = journal->j_bitmap_nodes.next; journal->j_free_bitmap_nodes--; } @@ -581,7 +579,7 @@ static inline void put_journal_list(struct super_block *s, jl->j_trans_id, jl->j_refcount); } if (--jl->j_refcount == 0) - reiserfs_kfree(jl, sizeof(struct reiserfs_journal_list), s); + kfree(jl); } /* @@ -1818,8 +1816,7 @@ void remove_journal_hash(struct super_block *sb, static void free_journal_ram(struct super_block *p_s_sb) { struct reiserfs_journal *journal = SB_JOURNAL(p_s_sb); - reiserfs_kfree(journal->j_current_jl, - sizeof(struct reiserfs_journal_list), p_s_sb); + kfree(journal->j_current_jl); journal->j_num_lists--; vfree(journal->j_cnode_free_orig); @@ -2093,21 +2090,15 @@ static int journal_read_transaction(struct super_block *p_s_sb, } trans_id = get_desc_trans_id(desc); /* now we know we've got a good transaction, and it was inside the valid time ranges */ - log_blocks = - reiserfs_kmalloc(get_desc_trans_len(desc) * - sizeof(struct buffer_head *), GFP_NOFS, p_s_sb); - real_blocks = - reiserfs_kmalloc(get_desc_trans_len(desc) * - sizeof(struct buffer_head *), GFP_NOFS, p_s_sb); + log_blocks = kmalloc(get_desc_trans_len(desc) * + sizeof(struct buffer_head *), GFP_NOFS); + real_blocks = kmalloc(get_desc_trans_len(desc) * + sizeof(struct buffer_head *), GFP_NOFS); if (!log_blocks || !real_blocks) { brelse(c_bh); brelse(d_bh); - reiserfs_kfree(log_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); - reiserfs_kfree(real_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); + kfree(log_blocks); + kfree(real_blocks); reiserfs_warning(p_s_sb, "journal-1169: kmalloc failed, unable to mount FS"); return -1; @@ -2145,12 +2136,8 @@ static int journal_read_transaction(struct super_block *p_s_sb, brelse_array(real_blocks, i); brelse(c_bh); brelse(d_bh); - reiserfs_kfree(log_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); - reiserfs_kfree(real_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); + kfree(log_blocks); + kfree(real_blocks); return -1; } } @@ -2166,12 +2153,8 @@ static int journal_read_transaction(struct super_block *p_s_sb, brelse_array(real_blocks, get_desc_trans_len(desc)); brelse(c_bh); brelse(d_bh); - reiserfs_kfree(log_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); - reiserfs_kfree(real_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); + kfree(log_blocks); + kfree(real_blocks); return -1; } memcpy(real_blocks[i]->b_data, log_blocks[i]->b_data, @@ -2193,12 +2176,8 @@ static int journal_read_transaction(struct super_block *p_s_sb, get_desc_trans_len(desc) - i); brelse(c_bh); brelse(d_bh); - reiserfs_kfree(log_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); - reiserfs_kfree(real_blocks, - get_desc_trans_len(desc) * - sizeof(struct buffer_head *), p_s_sb); + kfree(log_blocks); + kfree(real_blocks); return -1; } brelse(real_blocks[i]); @@ -2217,12 +2196,8 @@ static int journal_read_transaction(struct super_block *p_s_sb, journal->j_trans_id = trans_id + 1; brelse(c_bh); brelse(d_bh); - reiserfs_kfree(log_blocks, - le32_to_cpu(desc->j_len) * sizeof(struct buffer_head *), - p_s_sb); - reiserfs_kfree(real_blocks, - le32_to_cpu(desc->j_len) * sizeof(struct buffer_head *), - p_s_sb); + kfree(log_blocks); + kfree(real_blocks); return 0; } @@ -2472,13 +2447,11 @@ static struct reiserfs_journal_list *alloc_journal_list(struct super_block *s) { struct reiserfs_journal_list *jl; retry: - jl = reiserfs_kmalloc(sizeof(struct reiserfs_journal_list), GFP_NOFS, - s); + jl = kzalloc(sizeof(struct reiserfs_journal_list), GFP_NOFS); if (!jl) { yield(); goto retry; } - memset(jl, 0, sizeof(*jl)); INIT_LIST_HEAD(&jl->j_list); INIT_LIST_HEAD(&jl->j_working_list); INIT_LIST_HEAD(&jl->j_tail_bh_list); @@ -3042,14 +3015,12 @@ struct reiserfs_transaction_handle *reiserfs_persistent_transaction(struct } return th; } - th = reiserfs_kmalloc(sizeof(struct reiserfs_transaction_handle), - GFP_NOFS, s); + th = kmalloc(sizeof(struct reiserfs_transaction_handle), GFP_NOFS); if (!th) return NULL; ret = journal_begin(th, s, nblocks); if (ret) { - reiserfs_kfree(th, sizeof(struct reiserfs_transaction_handle), - s); + kfree(th); return NULL; } @@ -3067,8 +3038,7 @@ int reiserfs_end_persistent_transaction(struct reiserfs_transaction_handle *th) ret = -EIO; if (th->t_refcount == 0) { SB_JOURNAL(s)->j_persistent_trans--; - reiserfs_kfree(th, sizeof(struct reiserfs_transaction_handle), - s); + kfree(th); } return ret; } diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index 8f8d8d01107c..c8123308e060 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -456,7 +456,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, /* get memory for composing the entry */ buflen = DEH_SIZE + ROUND_UP(namelen); if (buflen > sizeof(small_buf)) { - buffer = reiserfs_kmalloc(buflen, GFP_NOFS, dir->i_sb); + buffer = kmalloc(buflen, GFP_NOFS); if (buffer == 0) return -ENOMEM; } else @@ -490,7 +490,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, retval = reiserfs_find_entry(dir, name, namelen, &path, &de); if (retval != NAME_NOT_FOUND) { if (buffer != small_buf) - reiserfs_kfree(buffer, buflen, dir->i_sb); + kfree(buffer); pathrelse(&path); if (retval == IO_ERROR) { @@ -515,7 +515,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, reiserfs_warning(dir->i_sb, "reiserfs_add_entry: Congratulations! we have got hash function screwed up"); if (buffer != small_buf) - reiserfs_kfree(buffer, buflen, dir->i_sb); + kfree(buffer); pathrelse(&path); return -EBUSY; } @@ -535,7 +535,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, &entry_key); if (buffer != small_buf) - reiserfs_kfree(buffer, buflen, dir->i_sb); + kfree(buffer); pathrelse(&path); return -EBUSY; } @@ -546,7 +546,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, reiserfs_paste_into_item(th, &path, &entry_key, dir, buffer, paste_size); if (buffer != small_buf) - reiserfs_kfree(buffer, buflen, dir->i_sb); + kfree(buffer); if (retval) { reiserfs_check_path(&path); return retval; @@ -1065,7 +1065,7 @@ static int reiserfs_symlink(struct inode *parent_dir, goto out_failed; } - name = reiserfs_kmalloc(item_len, GFP_NOFS, parent_dir->i_sb); + name = kmalloc(item_len, GFP_NOFS); if (!name) { drop_new_inode(inode); retval = -ENOMEM; @@ -1079,14 +1079,14 @@ static int reiserfs_symlink(struct inode *parent_dir, retval = journal_begin(&th, parent_dir->i_sb, jbegin_count); if (retval) { drop_new_inode(inode); - reiserfs_kfree(name, item_len, parent_dir->i_sb); + kfree(name); goto out_failed; } retval = reiserfs_new_inode(&th, parent_dir, mode, name, strlen(symname), dentry, inode); - reiserfs_kfree(name, item_len, parent_dir->i_sb); + kfree(name); if (retval) { /* reiserfs_new_inode iputs for us */ goto out_failed; } diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index fc2f43c75df4..ef6caed9336b 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -88,7 +88,6 @@ static int show_super(struct seq_file *m, struct super_block *sb) seq_printf(m, "state: \t%s\n" "mount options: \t%s%s%s%s%s%s%s%s%s%s%s\n" "gen. counter: \t%i\n" - "s_kmallocs: \t%i\n" "s_disk_reads: \t%i\n" "s_disk_writes: \t%i\n" "s_fix_nodes: \t%i\n" @@ -128,7 +127,7 @@ static int show_super(struct seq_file *m, struct super_block *sb) "SMALL_TAILS " : "NO_TAILS ", replay_only(sb) ? "REPLAY_ONLY " : "", convert_reiserfs(sb) ? "CONV " : "", - atomic_read(&r->s_generation_counter), SF(s_kmallocs), + atomic_read(&r->s_generation_counter), SF(s_disk_reads), SF(s_disk_writes), SF(s_fix_nodes), SF(s_do_balance), SF(s_unneeded_left_neighbor), SF(s_good_search_by_key_reada), SF(s_bmaps), diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 397d9590c8f2..77891de0e02e 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -472,12 +472,6 @@ static void reiserfs_put_super(struct super_block *s) print_statistics(s); - if (REISERFS_SB(s)->s_kmallocs != 0) { - reiserfs_warning(s, - "vs-2004: reiserfs_put_super: allocated memory left %d", - REISERFS_SB(s)->s_kmallocs); - } - if (REISERFS_SB(s)->reserved_blocks != 0) { reiserfs_warning(s, "green-2005: reiserfs_put_super: reserved blocks left %d", diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index cc061bfd437b..4f0db4e54517 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -368,15 +368,13 @@ static int __xattr_readdir(struct file *filp, void *dirent, filldir_t filldir) if (d_reclen <= 32) { local_buf = small_buf; } else { - local_buf = - reiserfs_kmalloc(d_reclen, GFP_NOFS, inode->i_sb); + local_buf = kmalloc(d_reclen, GFP_NOFS); if (!local_buf) { pathrelse(&path_to_entry); return -ENOMEM; } if (item_moved(&tmp_ih, &path_to_entry)) { - reiserfs_kfree(local_buf, d_reclen, - inode->i_sb); + kfree(local_buf); /* sigh, must retry. Do this same offset again */ next_pos = d_off; @@ -399,13 +397,12 @@ static int __xattr_readdir(struct file *filp, void *dirent, filldir_t filldir) if (filldir(dirent, local_buf, d_reclen, d_off, d_ino, DT_UNKNOWN) < 0) { if (local_buf != small_buf) { - reiserfs_kfree(local_buf, d_reclen, - inode->i_sb); + kfree(local_buf); } goto end; } if (local_buf != small_buf) { - reiserfs_kfree(local_buf, d_reclen, inode->i_sb); + kfree(local_buf); } } /* while */ diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index e276c5ba2bb7..7d51149bd793 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -1971,22 +1971,6 @@ extern struct file_operations reiserfs_file_operations; extern struct address_space_operations reiserfs_address_space_operations; /* fix_nodes.c */ -#ifdef CONFIG_REISERFS_CHECK -void *reiserfs_kmalloc(size_t size, gfp_t flags, struct super_block *s); -void reiserfs_kfree(const void *vp, size_t size, struct super_block *s); -#else -static inline void *reiserfs_kmalloc(size_t size, int flags, - struct super_block *s) -{ - return kmalloc(size, flags); -} - -static inline void reiserfs_kfree(const void *vp, size_t size, - struct super_block *s) -{ - kfree(vp); -} -#endif int fix_nodes(int n_op_mode, struct tree_balance *p_s_tb, struct item_head *p_s_ins_ih, const void *); diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index 3e68592e52e9..31b4c0bd4fa0 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -382,7 +382,6 @@ struct reiserfs_sb_info { on-disk FS format */ /* session statistics */ - int s_kmallocs; int s_disk_reads; int s_disk_writes; int s_fix_nodes; -- cgit v1.2.3-59-g8ed1b From 8c777cc4be1390862d053cbc002246e87572147b Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Wed, 1 Feb 2006 03:06:43 -0800 Subject: [PATCH] reiserfs: use __GFP_NOFAIL instead of yield and retry loop for allocation This patch replaces yield and retry loop with __GFP_NOFAIL in alloc_journal_list(). Signed-off-by: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/journal.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 16b526fd20b9..2d04efb22eea 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2446,12 +2446,8 @@ static int journal_read(struct super_block *p_s_sb) static struct reiserfs_journal_list *alloc_journal_list(struct super_block *s) { struct reiserfs_journal_list *jl; - retry: - jl = kzalloc(sizeof(struct reiserfs_journal_list), GFP_NOFS); - if (!jl) { - yield(); - goto retry; - } + jl = kzalloc(sizeof(struct reiserfs_journal_list), + GFP_NOFS | __GFP_NOFAIL); INIT_LIST_HEAD(&jl->j_list); INIT_LIST_HEAD(&jl->j_working_list); INIT_LIST_HEAD(&jl->j_tail_bh_list); -- cgit v1.2.3-59-g8ed1b From e5dd259f78ba0fd0c7bfc5c52179dbbff3eb48aa Mon Sep 17 00:00:00 2001 From: Diego Calleja Date: Wed, 1 Feb 2006 03:06:44 -0800 Subject: [PATCH] reiserfs: missing kmalloc failure check According to http://bugzilla.kernel.org/show_bug.cgi?id=5778 fs/reiserfs/file.c is missing this check. Signed-off-by: Diego Calleja Cc: Jeff Mahoney Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/file.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index ad6fa964b0e7..1cad5d066a5c 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -192,6 +192,8 @@ static int reiserfs_allocate_blocks_for_region(struct reiserfs_transaction_handl allocated_blocks = kmalloc((blocks_to_allocate + will_prealloc) * sizeof(b_blocknr_t), GFP_NOFS); + if (!allocated_blocks) + return -ENOMEM; /* First we compose a key to point at the writing position, we want to do that outside of any locking region. */ -- cgit v1.2.3-59-g8ed1b From c87d0c07ea198db1ce451421904edd60b7d385ee Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 1 Feb 2006 03:06:45 -0800 Subject: [PATCH] reiserfs: remove reiserfs_permission_locked This function is completely unused since the xattr permission checking changes. Remove it and fold __reiserfs_permission into reiserfs_permission. Signed-off-by: Christoph Hellwig Cc: Jeff Mahoney Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/xattr.c | 30 ++++++++---------------------- include/linux/reiserfs_xattr.h | 2 -- 2 files changed, 8 insertions(+), 24 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 4f0db4e54517..2f085845f670 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -1319,9 +1319,7 @@ int reiserfs_xattr_init(struct super_block *s, int mount_flags) return err; } -static int -__reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd, - int need_lock) +int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd) { umode_t mode = inode->i_mode; @@ -1357,15 +1355,14 @@ __reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd, if (!(mode & S_IRWXG)) goto check_groups; - if (need_lock) { - reiserfs_read_lock_xattr_i(inode); - reiserfs_read_lock_xattrs(inode->i_sb); - } + reiserfs_read_lock_xattr_i(inode); + reiserfs_read_lock_xattrs(inode->i_sb); + acl = reiserfs_get_acl(inode, ACL_TYPE_ACCESS); - if (need_lock) { - reiserfs_read_unlock_xattrs(inode->i_sb); - reiserfs_read_unlock_xattr_i(inode); - } + + reiserfs_read_unlock_xattrs(inode->i_sb); + reiserfs_read_unlock_xattr_i(inode); + if (IS_ERR(acl)) { if (PTR_ERR(acl) == -ENODATA) goto check_groups; @@ -1414,14 +1411,3 @@ __reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd, return -EACCES; } - -int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd) -{ - return __reiserfs_permission(inode, mask, nd, 1); -} - -int -reiserfs_permission_locked(struct inode *inode, int mask, struct nameidata *nd) -{ - return __reiserfs_permission(inode, mask, nd, 0); -} diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h index c84354e8374c..87280eb6083d 100644 --- a/include/linux/reiserfs_xattr.h +++ b/include/linux/reiserfs_xattr.h @@ -43,8 +43,6 @@ int reiserfs_delete_xattrs(struct inode *inode); int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs); int reiserfs_xattr_init(struct super_block *sb, int mount_flags); int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd); -int reiserfs_permission_locked(struct inode *inode, int mask, - struct nameidata *nd); int reiserfs_xattr_del(struct inode *, const char *); int reiserfs_xattr_get(const struct inode *, const char *, void *, size_t); -- cgit v1.2.3-59-g8ed1b From ec191574b9c3cb7bfb95e4f803b63f7c8dc52690 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 1 Feb 2006 03:06:46 -0800 Subject: [PATCH] reiserfs: use generic_permission Use the generic_permission code with a proper wrapper and callback instead of having a local copy. Signed-off-by: Christoph Hellwig Cc: Jeff Mahoney Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/xattr.c | 103 +++++++++++++--------------------------------------- 1 file changed, 26 insertions(+), 77 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 2f085845f670..ffb79c48c5bf 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -1319,95 +1319,44 @@ int reiserfs_xattr_init(struct super_block *s, int mount_flags) return err; } -int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd) +static int reiserfs_check_acl(struct inode *inode, int mask) { - umode_t mode = inode->i_mode; - - if (mask & MAY_WRITE) { - /* - * Nobody gets write access to a read-only fs. - */ - if (IS_RDONLY(inode) && - (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode))) - return -EROFS; - - /* - * Nobody gets write access to an immutable file. - */ - if (IS_IMMUTABLE(inode)) - return -EACCES; - } - - /* We don't do permission checks on the internal objects. - * Permissions are determined by the "owning" object. */ - if (is_reiserfs_priv_object(inode)) - return 0; - - if (current->fsuid == inode->i_uid) { - mode >>= 6; -#ifdef CONFIG_REISERFS_FS_POSIX_ACL - } else if (reiserfs_posixacl(inode->i_sb) && - get_inode_sd_version(inode) != STAT_DATA_V1) { - struct posix_acl *acl; - - /* ACL can't contain additional permissions if - the ACL_MASK entry is 0 */ - if (!(mode & S_IRWXG)) - goto check_groups; + struct posix_acl *acl; + int error = -EAGAIN; /* do regular unix permission checks by default */ - reiserfs_read_lock_xattr_i(inode); - reiserfs_read_lock_xattrs(inode->i_sb); - - acl = reiserfs_get_acl(inode, ACL_TYPE_ACCESS); + reiserfs_read_lock_xattr_i(inode); + reiserfs_read_lock_xattrs(inode->i_sb); - reiserfs_read_unlock_xattrs(inode->i_sb); - reiserfs_read_unlock_xattr_i(inode); + acl = reiserfs_get_acl(inode, ACL_TYPE_ACCESS); - if (IS_ERR(acl)) { - if (PTR_ERR(acl) == -ENODATA) - goto check_groups; - return PTR_ERR(acl); - } + reiserfs_read_unlock_xattrs(inode->i_sb); + reiserfs_read_unlock_xattr_i(inode); - if (acl) { - int err = posix_acl_permission(inode, acl, mask); + if (acl) { + if (!IS_ERR(acl)) { + error = posix_acl_permission(inode, acl, mask); posix_acl_release(acl); - if (err == -EACCES) { - goto check_capabilities; - } - return err; - } else { - goto check_groups; - } -#endif - } else { - check_groups: - if (in_group_p(inode->i_gid)) - mode >>= 3; + } else if (PTR_ERR(acl) != -ENODATA) + error = PTR_ERR(acl); } - /* - * If the DACs are ok we don't need any capability check. - */ - if (((mode & mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == mask)) - return 0; + return error; +} - check_capabilities: +int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd) +{ /* - * Read/write DACs are always overridable. - * Executable DACs are overridable if at least one exec bit is set. + * We don't do permission checks on the internal objects. + * Permissions are determined by the "owning" object. */ - if (!(mask & MAY_EXEC) || - (inode->i_mode & S_IXUGO) || S_ISDIR(inode->i_mode)) - if (capable(CAP_DAC_OVERRIDE)) - return 0; + if (is_reiserfs_priv_object(inode)) + return 0; /* - * Searching includes executable on directories, else just read. + * Stat data v1 doesn't support ACLs. */ - if (mask == MAY_READ || (S_ISDIR(inode->i_mode) && !(mask & MAY_WRITE))) - if (capable(CAP_DAC_READ_SEARCH)) - return 0; - - return -EACCES; + if (get_inode_sd_version(inode) == STAT_DATA_V1) + return generic_permission(inode, mask, NULL); + else + return generic_permission(inode, mask, reiserfs_check_acl); } -- cgit v1.2.3-59-g8ed1b From d62b1b87a7d1c3a21dddabed4251763090be3182 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Wed, 1 Feb 2006 03:06:47 -0800 Subject: [PATCH] resierfs: fix reiserfs_invalidatepage race against data=ordered After a transaction has closed but before it has finished commit, there is a window where data=ordered mode requires invalidatepage to pin pages instead of freeing them. This patch fixes a race between the invalidatepage checks and data=ordered writeback, and it also adds a check to the reiserfs write_ordered_buffers routines to write any anonymous buffers that were dirtied after its first writeback loop. That bug works like this: proc1: transaction closes and a new one starts proc1: write_ordered_buffers starts processing data=ordered list proc1: buffer A is cleaned and written proc2: buffer A is dirtied by another process proc2: File is truncated to zero, page A goes through invalidatepage proc2: reiserfs_invalidatepage sees dirty buffer A with reiserfs journal head, pins it proc1: write_ordered_buffers frees the journal head on buffer A At this point, buffer A stays dirty forever Signed-off-by: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/inode.c | 4 +++- fs/reiserfs/journal.c | 13 +++++++++++++ 2 files changed, 16 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index ffa34b861bdb..60e2f2344703 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2743,6 +2743,7 @@ static int invalidatepage_can_drop(struct inode *inode, struct buffer_head *bh) int ret = 1; struct reiserfs_journal *j = SB_JOURNAL(inode->i_sb); + lock_buffer(bh); spin_lock(&j->j_dirty_buffers_lock); if (!buffer_mapped(bh)) { goto free_jh; @@ -2758,7 +2759,7 @@ static int invalidatepage_can_drop(struct inode *inode, struct buffer_head *bh) if (buffer_journaled(bh) || buffer_journal_dirty(bh)) { ret = 0; } - } else if (buffer_dirty(bh) || buffer_locked(bh)) { + } else if (buffer_dirty(bh)) { struct reiserfs_journal_list *jl; struct reiserfs_jh *jh = bh->b_private; @@ -2784,6 +2785,7 @@ static int invalidatepage_can_drop(struct inode *inode, struct buffer_head *bh) reiserfs_free_jh(bh); } spin_unlock(&j->j_dirty_buffers_lock); + unlock_buffer(bh); return ret; } diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 2d04efb22eea..bc8fe963b3cc 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -877,6 +877,19 @@ static int write_ordered_buffers(spinlock_t * lock, if (!buffer_uptodate(bh)) { ret = -EIO; } + /* ugly interaction with invalidatepage here. + * reiserfs_invalidate_page will pin any buffer that has a valid + * journal head from an older transaction. If someone else sets + * our buffer dirty after we write it in the first loop, and + * then someone truncates the page away, nobody will ever write + * the buffer. We're safe if we write the page one last time + * after freeing the journal header. + */ + if (buffer_dirty(bh) && unlikely(bh->b_page->mapping == NULL)) { + spin_unlock(lock); + ll_rw_block(WRITE, 1, &bh); + spin_lock(lock); + } put_bh(bh); cond_resched_lock(lock); } -- cgit v1.2.3-59-g8ed1b From fc5cd582e9c934ddaf6f310179488932cd154794 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Wed, 1 Feb 2006 03:06:48 -0800 Subject: [PATCH] reiserfs: zero b_private when allocating buffer heads The b_private field in buffer heads needs to be zero filled when the buffers are allocated. Thanks to Nathan Scott for finding this. It was causing problems on systems with both XFS and reiserfs. Signed-off-by: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 8bcbac87a28c..5e4a90ee103f 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1022,6 +1022,7 @@ try_again: bh->b_state = 0; atomic_set(&bh->b_count, 0); + bh->b_private = NULL; bh->b_size = size; /* Link the buffer to its page */ -- cgit v1.2.3-59-g8ed1b From e0e851cf30f1a9bd2e2a7624e9810378d6a2b072 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Wed, 1 Feb 2006 03:06:49 -0800 Subject: [PATCH] reiserfs: reiserfs hang and performance fix for data=journal mode In data=journal mode, reiserfs writepage needs to make sure not to trigger transactions while being run under PF_MEMALLOC. This patch makes sure to redirty the page instead of forcing a transaction start in this case. Also, calling filemap_fdata* in order to trigger io on the block device can cause lock inversions on the page lock. Instead, do simple batching from flush_commit_list. Signed-off-by: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/inode.c | 7 +++++++ fs/reiserfs/journal.c | 19 ++++++++++++++----- 2 files changed, 21 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 60e2f2344703..b33d67bba2fd 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2363,6 +2363,13 @@ static int reiserfs_write_full_page(struct page *page, int bh_per_page = PAGE_CACHE_SIZE / s->s_blocksize; th.t_trans_id = 0; + /* no logging allowed when nonblocking or from PF_MEMALLOC */ + if (checked && (current->flags & PF_MEMALLOC)) { + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + /* The page dirty bit is cleared before writepage is called, which * means we have to tell create_empty_buffers to make dirty buffers * The page really should be up to date at this point, so tossing diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index bc8fe963b3cc..1b2402a9a8e1 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -988,6 +988,7 @@ static int flush_commit_list(struct super_block *s, struct reiserfs_journal *journal = SB_JOURNAL(s); int barrier = 0; int retval = 0; + int write_len; reiserfs_check_lock_depth(s, "flush_commit_list"); @@ -1037,16 +1038,24 @@ static int flush_commit_list(struct super_block *s, BUG_ON(!list_empty(&jl->j_bh_list)); /* * for the description block and all the log blocks, submit any buffers - * that haven't already reached the disk + * that haven't already reached the disk. Try to write at least 256 + * log blocks. later on, we will only wait on blocks that correspond + * to this transaction, but while we're unplugging we might as well + * get a chunk of data on there. */ atomic_inc(&journal->j_async_throttle); - for (i = 0; i < (jl->j_len + 1); i++) { + write_len = jl->j_len + 1; + if (write_len < 256) + write_len = 256; + for (i = 0 ; i < write_len ; i++) { bn = SB_ONDISK_JOURNAL_1st_BLOCK(s) + (jl->j_start + i) % SB_ONDISK_JOURNAL_SIZE(s); tbh = journal_find_get_block(s, bn); - if (buffer_dirty(tbh)) /* redundant, ll_rw_block() checks */ - ll_rw_block(SWRITE, 1, &tbh); - put_bh(tbh); + if (tbh) { + if (buffer_dirty(tbh)) + ll_rw_block(WRITE, 1, &tbh) ; + put_bh(tbh) ; + } } atomic_dec(&journal->j_async_throttle); -- cgit v1.2.3-59-g8ed1b From 3d4492f81dd7b486f1be0616a1ce7f73760f406e Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Wed, 1 Feb 2006 03:06:49 -0800 Subject: [PATCH] reiserfs: reiserfs write_ordered_buffers should not oops on dirty non-uptodate bh write_ordered_buffers should handle dirty non-uptodate buffers without a BUG() Signed-off-by: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/journal.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 1b2402a9a8e1..47c9f432dea5 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -846,6 +846,14 @@ static int write_ordered_buffers(spinlock_t * lock, spin_lock(lock); goto loop_next; } + /* in theory, dirty non-uptodate buffers should never get here, + * but the upper layer io error paths still have a few quirks. + * Handle them here as gracefully as we can + */ + if (!buffer_uptodate(bh) && buffer_dirty(bh)) { + clear_buffer_dirty(bh); + ret = -EIO; + } if (buffer_dirty(bh)) { list_del_init(&jh->list); list_add(&jh->list, &tmp); @@ -1030,9 +1038,12 @@ static int flush_commit_list(struct super_block *s, } if (!list_empty(&jl->j_bh_list)) { + int ret; unlock_kernel(); - write_ordered_buffers(&journal->j_dirty_buffers_lock, - journal, jl, &jl->j_bh_list); + ret = write_ordered_buffers(&journal->j_dirty_buffers_lock, + journal, jl, &jl->j_bh_list); + if (ret < 0 && retval == 0) + retval = ret; lock_kernel(); } BUG_ON(!list_empty(&jl->j_bh_list)); -- cgit v1.2.3-59-g8ed1b From 6ae1ea447d21c4fecf5df8d0e1022461274fb4e8 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Wed, 1 Feb 2006 03:06:50 -0800 Subject: [PATCH] reiserfs: reiserfs fix journal accounting in journal_transaction_should_end reiserfs: journal_transaction_should_end should increase the count of blocks allocated so the transaction subsystem can keep new writers from creating a transaction that is too large. Signed-off-by: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/journal.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 47c9f432dea5..b7a179560ab4 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2823,6 +2823,9 @@ int journal_transaction_should_end(struct reiserfs_transaction_handle *th, journal->j_cnode_free < (journal->j_trans_max * 3)) { return 1; } + /* protected by the BKL here */ + journal->j_len_alloc += new_alloc; + th->t_blocks_allocated += new_alloc ; return 0; } -- cgit v1.2.3-59-g8ed1b From fa385bef256077f3b820b241e8f3755ef3905b74 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Wed, 1 Feb 2006 03:06:51 -0800 Subject: [PATCH] reiserfs: reiserfs: check for files > 2GB on 3.5.x disks When a filesystem has been converted from 3.5.x to 3.6.x, we need an extra check during file write to make sure we are not trying to make a 3.5.x file > 2GB. Signed-off-by: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/file.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'fs') diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 1cad5d066a5c..f3473176c83a 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -1287,6 +1287,23 @@ static ssize_t reiserfs_file_write(struct file *file, /* the file we are going t struct reiserfs_transaction_handle th; th.t_trans_id = 0; + /* If a filesystem is converted from 3.5 to 3.6, we'll have v3.5 items + * lying around (most of the disk, in fact). Despite the filesystem + * now being a v3.6 format, the old items still can't support large + * file sizes. Catch this case here, as the rest of the VFS layer is + * oblivious to the different limitations between old and new items. + * reiserfs_setattr catches this for truncates. This chunk is lifted + * from generic_write_checks. */ + if (get_inode_item_key_version (inode) == KEY_FORMAT_3_5 && + *ppos + count > MAX_NON_LFS) { + if (*ppos >= MAX_NON_LFS) { + send_sig(SIGXFSZ, current, 0); + return -EFBIG; + } + if (count > MAX_NON_LFS - (unsigned long)*ppos) + count = MAX_NON_LFS - (unsigned long)*ppos; + } + if (file->f_flags & O_DIRECT) { // Direct IO needs treatment ssize_t result, after_file_end = 0; if ((*ppos + count >= inode->i_size) -- cgit v1.2.3-59-g8ed1b From 7045f37b17ffa6e85435ca980122b46a74caa7e4 Mon Sep 17 00:00:00 2001 From: Martin Waitz Date: Wed, 1 Feb 2006 03:06:57 -0800 Subject: [PATCH] DocBook: fix some kernel-doc comments in fs and block Update some parameter descriptions to actually match the code. Signed-off-by: Martin Waitz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inode.c | 2 +- fs/namei.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/inode.c b/fs/inode.c index 108138d4e909..d0be6159eb7f 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1179,7 +1179,7 @@ EXPORT_SYMBOL(bmap); /** * touch_atime - update the access time * @mnt: mount the inode is accessed on - * @inode: inode accessed + * @dentry: dentry accessed * * Update the accessed time on an inode and mark it for writeback. * This function automatically handles read only file systems and media, diff --git a/fs/namei.c b/fs/namei.c index 4acdac043b6b..7ac9fb4acb2c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1161,6 +1161,7 @@ static int __path_lookup_intent_open(int dfd, const char *name, /** * path_lookup_open - lookup a file path with open intent + * @dfd: the directory to use as base, or AT_FDCWD * @name: pointer to file name * @lookup_flags: lookup intent flags * @nd: pointer to nameidata @@ -1175,6 +1176,7 @@ int path_lookup_open(int dfd, const char *name, unsigned int lookup_flags, /** * path_lookup_create - lookup a file path with open + create intent + * @dfd: the directory to use as base, or AT_FDCWD * @name: pointer to file name * @lookup_flags: lookup intent flags * @nd: pointer to nameidata -- cgit v1.2.3-59-g8ed1b