diff options
Diffstat (limited to 'fs')
568 files changed, 21096 insertions, 13524 deletions
diff --git a/fs/9p/cache.c b/fs/9p/cache.c index 64c58eb26159..9eb34701a566 100644 --- a/fs/9p/cache.c +++ b/fs/9p/cache.c @@ -55,42 +55,27 @@ int v9fs_random_cachetag(struct v9fs_session_info *v9ses) return scnprintf(v9ses->cachetag, CACHETAG_LEN, "%lu", jiffies); } -static uint16_t v9fs_cache_session_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - struct v9fs_session_info *v9ses; - uint16_t klen = 0; - - v9ses = (struct v9fs_session_info *)cookie_netfs_data; - p9_debug(P9_DEBUG_FSC, "session %p buf %p size %u\n", - v9ses, buffer, bufmax); - - if (v9ses->cachetag) - klen = strlen(v9ses->cachetag); - - if (klen > bufmax) - return 0; - - memcpy(buffer, v9ses->cachetag, klen); - p9_debug(P9_DEBUG_FSC, "cache session tag %s\n", v9ses->cachetag); - return klen; -} - const struct fscache_cookie_def v9fs_cache_session_index_def = { .name = "9P.session", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = v9fs_cache_session_get_key, }; void v9fs_cache_session_get_cookie(struct v9fs_session_info *v9ses) { /* If no cache session tag was specified, we generate a random one. */ - if (!v9ses->cachetag) - v9fs_random_cachetag(v9ses); + if (!v9ses->cachetag) { + if (v9fs_random_cachetag(v9ses) < 0) { + v9ses->fscache = NULL; + return; + } + } v9ses->fscache = fscache_acquire_cookie(v9fs_cache_netfs.primary_index, &v9fs_cache_session_index_def, - v9ses, true); + v9ses->cachetag, + strlen(v9ses->cachetag), + NULL, 0, + v9ses, 0, true); p9_debug(P9_DEBUG_FSC, "session %p get cookie %p\n", v9ses, v9ses->fscache); } @@ -99,45 +84,15 @@ void v9fs_cache_session_put_cookie(struct v9fs_session_info *v9ses) { p9_debug(P9_DEBUG_FSC, "session %p put cookie %p\n", v9ses, v9ses->fscache); - fscache_relinquish_cookie(v9ses->fscache, 0); + fscache_relinquish_cookie(v9ses->fscache, NULL, false); v9ses->fscache = NULL; } - -static uint16_t v9fs_cache_inode_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct v9fs_inode *v9inode = cookie_netfs_data; - memcpy(buffer, &v9inode->qid.path, sizeof(v9inode->qid.path)); - p9_debug(P9_DEBUG_FSC, "inode %p get key %llu\n", - &v9inode->vfs_inode, v9inode->qid.path); - return sizeof(v9inode->qid.path); -} - -static void v9fs_cache_inode_get_attr(const void *cookie_netfs_data, - uint64_t *size) -{ - const struct v9fs_inode *v9inode = cookie_netfs_data; - *size = i_size_read(&v9inode->vfs_inode); - - p9_debug(P9_DEBUG_FSC, "inode %p get attr %llu\n", - &v9inode->vfs_inode, *size); -} - -static uint16_t v9fs_cache_inode_get_aux(const void *cookie_netfs_data, - void *buffer, uint16_t buflen) -{ - const struct v9fs_inode *v9inode = cookie_netfs_data; - memcpy(buffer, &v9inode->qid.version, sizeof(v9inode->qid.version)); - p9_debug(P9_DEBUG_FSC, "inode %p get aux %u\n", - &v9inode->vfs_inode, v9inode->qid.version); - return sizeof(v9inode->qid.version); -} - static enum fscache_checkaux v9fs_cache_inode_check_aux(void *cookie_netfs_data, const void *buffer, - uint16_t buflen) + uint16_t buflen, + loff_t object_size) { const struct v9fs_inode *v9inode = cookie_netfs_data; @@ -154,9 +109,6 @@ fscache_checkaux v9fs_cache_inode_check_aux(void *cookie_netfs_data, const struct fscache_cookie_def v9fs_cache_inode_index_def = { .name = "9p.inode", .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .get_key = v9fs_cache_inode_get_key, - .get_attr = v9fs_cache_inode_get_attr, - .get_aux = v9fs_cache_inode_get_aux, .check_aux = v9fs_cache_inode_check_aux, }; @@ -175,7 +127,13 @@ void v9fs_cache_inode_get_cookie(struct inode *inode) v9ses = v9fs_inode2v9ses(inode); v9inode->fscache = fscache_acquire_cookie(v9ses->fscache, &v9fs_cache_inode_index_def, - v9inode, true); + &v9inode->qid.path, + sizeof(v9inode->qid.path), + &v9inode->qid.version, + sizeof(v9inode->qid.version), + v9inode, + i_size_read(&v9inode->vfs_inode), + true); p9_debug(P9_DEBUG_FSC, "inode %p get cookie %p\n", inode, v9inode->fscache); @@ -190,7 +148,8 @@ void v9fs_cache_inode_put_cookie(struct inode *inode) p9_debug(P9_DEBUG_FSC, "inode %p put cookie %p\n", inode, v9inode->fscache); - fscache_relinquish_cookie(v9inode->fscache, 0); + fscache_relinquish_cookie(v9inode->fscache, &v9inode->qid.version, + false); v9inode->fscache = NULL; } @@ -203,7 +162,7 @@ void v9fs_cache_inode_flush_cookie(struct inode *inode) p9_debug(P9_DEBUG_FSC, "inode %p flush cookie %p\n", inode, v9inode->fscache); - fscache_relinquish_cookie(v9inode->fscache, 1); + fscache_relinquish_cookie(v9inode->fscache, NULL, true); v9inode->fscache = NULL; } @@ -236,12 +195,18 @@ void v9fs_cache_inode_reset_cookie(struct inode *inode) old = v9inode->fscache; mutex_lock(&v9inode->fscache_lock); - fscache_relinquish_cookie(v9inode->fscache, 1); + fscache_relinquish_cookie(v9inode->fscache, NULL, true); v9ses = v9fs_inode2v9ses(inode); v9inode->fscache = fscache_acquire_cookie(v9ses->fscache, &v9fs_cache_inode_index_def, - v9inode, true); + &v9inode->qid.path, + sizeof(v9inode->qid.path), + &v9inode->qid.version, + sizeof(v9inode->qid.version), + v9inode, + i_size_read(&v9inode->vfs_inode), + true); p9_debug(P9_DEBUG_FSC, "inode %p revalidating cookie old %p new %p\n", inode, old, v9inode->fscache); @@ -367,7 +332,8 @@ void __v9fs_readpage_to_fscache(struct inode *inode, struct page *page) const struct v9fs_inode *v9inode = V9FS_I(inode); p9_debug(P9_DEBUG_FSC, "inode %p page %p\n", inode, page); - ret = fscache_write_page(v9inode->fscache, page, GFP_KERNEL); + ret = fscache_write_page(v9inode->fscache, page, + i_size_read(&v9inode->vfs_inode), GFP_KERNEL); p9_debug(P9_DEBUG_FSC, "ret = %d\n", ret); if (ret != 0) v9fs_uncache_page(inode, page); diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 8fb89ddc6cc7..e622f0f10502 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -292,6 +292,10 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts) #ifdef CONFIG_9P_FSCACHE kfree(v9ses->cachetag); v9ses->cachetag = match_strdup(&args[0]); + if (!v9ses->cachetag) { + ret = -ENOMEM; + goto free_and_return; + } #endif break; case Opt_cache: @@ -471,6 +475,9 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses, return fid; err_clnt: +#ifdef CONFIG_9P_FSCACHE + kfree(v9ses->cachetag); +#endif p9_client_destroy(v9ses->clnt); err_names: kfree(v9ses->uname); diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index bdabb2765d1b..42e102e2e74a 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -579,6 +579,24 @@ static int v9fs_at_to_dotl_flags(int flags) } /** + * v9fs_dec_count - helper functon to drop i_nlink. + * + * If a directory had nlink <= 2 (including . and ..), then we should not drop + * the link count, which indicates the underlying exported fs doesn't maintain + * nlink accurately. e.g. + * - overlayfs sets nlink to 1 for merged dir + * - ext4 (with dir_nlink feature enabled) sets nlink to 1 if a dir has more + * than EXT4_LINK_MAX (65000) links. + * + * @inode: inode whose nlink is being dropped + */ +static void v9fs_dec_count(struct inode *inode) +{ + if (!S_ISDIR(inode->i_mode) || inode->i_nlink > 2) + drop_nlink(inode); +} + +/** * v9fs_remove - helper function to remove files and directories * @dir: directory inode that is being deleted * @dentry: dentry that is being deleted @@ -621,9 +639,9 @@ static int v9fs_remove(struct inode *dir, struct dentry *dentry, int flags) */ if (flags & AT_REMOVEDIR) { clear_nlink(inode); - drop_nlink(dir); + v9fs_dec_count(dir); } else - drop_nlink(inode); + v9fs_dec_count(inode); v9fs_invalidate_inode_attr(inode); v9fs_invalidate_inode_attr(dir); @@ -805,28 +823,21 @@ struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, if (IS_ERR(dfid)) return ERR_CAST(dfid); - name = dentry->d_name.name; - fid = p9_client_walk(dfid, 1, &name, 1); - if (IS_ERR(fid)) { - if (fid == ERR_PTR(-ENOENT)) { - d_add(dentry, NULL); - return NULL; - } - return ERR_CAST(fid); - } /* * Make sure we don't use a wrong inode due to parallel * unlink. For cached mode create calls request for new * inode. But with cache disabled, lookup should do this. */ - if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) + name = dentry->d_name.name; + fid = p9_client_walk(dfid, 1, &name, 1); + if (fid == ERR_PTR(-ENOENT)) + inode = NULL; + else if (IS_ERR(fid)) + inode = ERR_CAST(fid); + else if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) inode = v9fs_get_inode_from_fid(v9ses, fid, dir->i_sb); else inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb); - if (IS_ERR(inode)) { - p9_client_clunk(fid); - return ERR_CAST(inode); - } /* * If we had a rename on the server and a parallel lookup * for the new name, then make sure we instantiate with @@ -835,12 +846,14 @@ struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, * k/b. */ res = d_splice_alias(inode, dentry); - if (!res) - v9fs_fid_add(dentry, fid); - else if (!IS_ERR(res)) - v9fs_fid_add(res, fid); - else - p9_client_clunk(fid); + if (!IS_ERR(fid)) { + if (!res) + v9fs_fid_add(dentry, fid); + else if (!IS_ERR(res)) + v9fs_fid_add(res, fid); + else + p9_client_clunk(fid); + } return res; } @@ -1024,12 +1037,12 @@ clunk_newdir: if (S_ISDIR(new_inode->i_mode)) clear_nlink(new_inode); else - drop_nlink(new_inode); + v9fs_dec_count(new_inode); } if (S_ISDIR(old_inode->i_mode)) { if (!new_inode) inc_nlink(new_dir); - drop_nlink(old_dir); + v9fs_dec_count(old_dir); } v9fs_invalidate_inode_attr(old_inode); v9fs_invalidate_inode_attr(old_dir); diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index af03c2a901eb..48ce50484e80 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -94,7 +94,7 @@ v9fs_fill_super(struct super_block *sb, struct v9fs_session_info *v9ses, if (v9ses->cache) sb->s_bdi->ra_pages = (VM_MAX_READAHEAD * 1024)/PAGE_SIZE; - sb->s_flags |= SB_ACTIVE | SB_DIRSYNC | SB_NOATIME; + sb->s_flags |= SB_ACTIVE | SB_DIRSYNC; if (!v9ses->cache) sb->s_flags |= SB_SYNCHRONOUS; diff --git a/fs/Kconfig b/fs/Kconfig index bc821a86d965..ac4ac908f001 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -196,7 +196,7 @@ config HUGETLBFS help hugetlbfs is a filesystem backing for HugeTLB pages, based on ramfs. For architectures that support it, say Y here and read - <file:Documentation/vm/hugetlbpage.txt> for details. + <file:Documentation/admin-guide/mm/hugetlbpage.rst> for details. If unsure, say N. diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c index 29444c83da48..e18eff854e1a 100644 --- a/fs/adfs/dir.c +++ b/fs/adfs/dir.c @@ -146,20 +146,6 @@ adfs_dir_lookup_byname(struct inode *inode, const struct qstr *name, struct obje obj->parent_id = inode->i_ino; - /* - * '.' is handled by reserved_lookup() in fs/namei.c - */ - if (name->len == 2 && name->name[0] == '.' && name->name[1] == '.') { - /* - * Currently unable to fill in the rest of 'obj', - * but this is better than nothing. We need to - * ascend one level to find it's parent. - */ - obj->name_len = 0; - obj->file_id = obj->parent_id; - goto free_out; - } - read_lock(&adfs_dir_lock); ret = ops->setpos(&dir, 0); @@ -266,17 +252,17 @@ adfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) error = adfs_dir_lookup_byname(dir, &dentry->d_name, &obj); if (error == 0) { - error = -EACCES; /* * This only returns NULL if get_empty_inode * fails. */ inode = adfs_iget(dir->i_sb, &obj); - if (inode) - error = 0; + if (!inode) + inode = ERR_PTR(-EACCES); + } else if (error != -ENOENT) { + inode = ERR_PTR(error); } - d_add(dentry, inode); - return ERR_PTR(error); + return d_splice_alias(inode, dentry); } /* diff --git a/fs/affs/namei.c b/fs/affs/namei.c index d8aa0ae3d037..41c5749f4db7 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -201,14 +201,16 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) struct super_block *sb = dir->i_sb; struct buffer_head *bh; struct inode *inode = NULL; + struct dentry *res; pr_debug("%s(\"%pd\")\n", __func__, dentry); affs_lock_dir(dir); bh = affs_find_entry(dir, dentry); - affs_unlock_dir(dir); - if (IS_ERR(bh)) + if (IS_ERR(bh)) { + affs_unlock_dir(dir); return ERR_CAST(bh); + } if (bh) { u32 ino = bh->b_blocknr; @@ -222,11 +224,12 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) } affs_brelse(bh); inode = affs_iget(sb, ino); - if (IS_ERR(inode)) - return ERR_CAST(inode); } - d_add(dentry, inode); - return NULL; + res = d_splice_alias(inode, dentry); + if (!IS_ERR_OR_NULL(res)) + res->d_fsdata = dentry->d_fsdata; + affs_unlock_dir(dir); + return res; } int diff --git a/fs/affs/super.c b/fs/affs/super.c index e602619aed9d..d1ad11a8a4a5 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -241,6 +241,7 @@ parse_options(char *options, kuid_t *uid, kgid_t *gid, int *mode, int *reserved, affs_set_opt(*mount_opts, SF_NO_TRUNCATE); break; case Opt_prefix: + kfree(*prefix); *prefix = match_strdup(&args[0]); if (!*prefix) return 0; diff --git a/fs/afs/Makefile b/fs/afs/Makefile index 45b7fc405fa6..532acae25453 100644 --- a/fs/afs/Makefile +++ b/fs/afs/Makefile @@ -12,6 +12,8 @@ kafs-objs := \ cell.o \ cmservice.o \ dir.o \ + dir_edit.o \ + dynroot.o \ file.o \ flock.o \ fsclient.o \ diff --git a/fs/afs/addr_list.c b/fs/afs/addr_list.c index fd9f28b8a933..7587fb665ff1 100644 --- a/fs/afs/addr_list.c +++ b/fs/afs/addr_list.c @@ -121,7 +121,7 @@ struct afs_addr_list *afs_parse_text_addrs(const char *text, size_t len, p = text; do { struct sockaddr_rxrpc *srx = &alist->addrs[alist->nr_addrs]; - char tdelim = delim; + const char *q, *stop; if (*p == delim) { p++; @@ -130,28 +130,33 @@ struct afs_addr_list *afs_parse_text_addrs(const char *text, size_t len, if (*p == '[') { p++; - tdelim = ']'; + q = memchr(p, ']', end - p); + } else { + for (q = p; q < end; q++) + if (*q == '+' || *q == delim) + break; } - if (in4_pton(p, end - p, + if (in4_pton(p, q - p, (u8 *)&srx->transport.sin6.sin6_addr.s6_addr32[3], - tdelim, &p)) { + -1, &stop)) { srx->transport.sin6.sin6_addr.s6_addr32[0] = 0; srx->transport.sin6.sin6_addr.s6_addr32[1] = 0; srx->transport.sin6.sin6_addr.s6_addr32[2] = htonl(0xffff); - } else if (in6_pton(p, end - p, + } else if (in6_pton(p, q - p, srx->transport.sin6.sin6_addr.s6_addr, - tdelim, &p)) { + -1, &stop)) { /* Nothing to do */ } else { goto bad_address; } - if (tdelim == ']') { - if (p == end || *p != ']') - goto bad_address; + if (stop != q) + goto bad_address; + + p = q; + if (q < end && *q == ']') p++; - } if (p < end) { if (*p == '+') { @@ -243,9 +248,9 @@ void afs_merge_fs_addr4(struct afs_addr_list *alist, __be32 xdr, u16 port) xport == a->sin6_port) return; if (xdr == a->sin6_addr.s6_addr32[3] && - xport < a->sin6_port) + (u16 __force)xport < (u16 __force)a->sin6_port) break; - if (xdr < a->sin6_addr.s6_addr32[3]) + if ((u32 __force)xdr < (u32 __force)a->sin6_addr.s6_addr32[3]) break; } @@ -280,7 +285,7 @@ void afs_merge_fs_addr6(struct afs_addr_list *alist, __be32 *xdr, u16 port) xport == a->sin6_port) return; if (diff == 0 && - xport < a->sin6_port) + (u16 __force)xport < (u16 __force)a->sin6_port) break; if (diff < 0) break; diff --git a/fs/afs/afs.h b/fs/afs/afs.h index b94d0edc2b78..b4ff1f7ae4ab 100644 --- a/fs/afs/afs.h +++ b/fs/afs/afs.h @@ -67,10 +67,14 @@ typedef enum { } afs_callback_type_t; struct afs_callback { - struct afs_fid fid; /* file identifier */ - unsigned version; /* callback version */ - unsigned expiry; /* time at which expires */ - afs_callback_type_t type; /* type of callback */ + unsigned version; /* Callback version */ + unsigned expiry; /* Time at which expires */ + afs_callback_type_t type; /* Type of callback */ +}; + +struct afs_callback_break { + struct afs_fid fid; /* File identifier */ + struct afs_callback cb; /* Callback details */ }; #define AFSCBMAX 50 /* maximum callbacks transferred per bulk op */ @@ -123,21 +127,20 @@ typedef u32 afs_access_t; * AFS file status information */ struct afs_file_status { - unsigned if_version; /* interface version */ -#define AFS_FSTATUS_VERSION 1 + u64 size; /* file size */ + afs_dataversion_t data_version; /* current data version */ + time_t mtime_client; /* last time client changed data */ + time_t mtime_server; /* last time server changed data */ + unsigned abort_code; /* Abort if bulk-fetching this failed */ afs_file_type_t type; /* file type */ unsigned nlink; /* link count */ - u64 size; /* file size */ - afs_dataversion_t data_version; /* current data version */ u32 author; /* author ID */ - kuid_t owner; /* owner ID */ - kgid_t group; /* group ID */ + u32 owner; /* owner ID */ + u32 group; /* group ID */ afs_access_t caller_access; /* access rights for authenticated caller */ afs_access_t anon_access; /* access rights for unauthenticated caller */ umode_t mode; /* UNIX mode */ - time_t mtime_client; /* last time client changed data */ - time_t mtime_server; /* last time server changed data */ s32 lock_count; /* file lock count (0=UNLK -1=WRLCK +ve=#RDLCK */ }; diff --git a/fs/afs/afs_fs.h b/fs/afs/afs_fs.h index d47b6d01e4c0..ddfa88a7a9c0 100644 --- a/fs/afs/afs_fs.h +++ b/fs/afs/afs_fs.h @@ -31,10 +31,12 @@ enum AFS_FS_Operations { FSGETVOLUMEINFO = 148, /* AFS Get information about a volume */ FSGETVOLUMESTATUS = 149, /* AFS Get volume status information */ FSGETROOTVOLUME = 151, /* AFS Get root volume name */ + FSBULKSTATUS = 155, /* AFS Fetch multiple file statuses */ FSSETLOCK = 156, /* AFS Request a file lock */ FSEXTENDLOCK = 157, /* AFS Extend a file lock */ FSRELEASELOCK = 158, /* AFS Release a file lock */ FSLOOKUP = 161, /* AFS lookup file in directory */ + FSINLINEBULKSTATUS = 65536, /* AFS Fetch multiple file statuses with inline errors */ FSFETCHDATA64 = 65537, /* AFS Fetch file data */ FSSTOREDATA64 = 65538, /* AFS Store file data */ FSGIVEUPALLCALLBACKS = 65539, /* AFS Give up all outstanding callbacks on a server */ diff --git a/fs/afs/cache.c b/fs/afs/cache.c index f62ff71d28c9..b1c31ec4523a 100644 --- a/fs/afs/cache.c +++ b/fs/afs/cache.c @@ -12,167 +12,39 @@ #include <linux/sched.h> #include "internal.h" -static uint16_t afs_cell_cache_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t buflen); -static uint16_t afs_volume_cache_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t buflen); - -static uint16_t afs_vnode_cache_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t buflen); -static void afs_vnode_cache_get_attr(const void *cookie_netfs_data, - uint64_t *size); -static uint16_t afs_vnode_cache_get_aux(const void *cookie_netfs_data, - void *buffer, uint16_t buflen); static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, const void *buffer, - uint16_t buflen); + uint16_t buflen, + loff_t object_size); struct fscache_netfs afs_cache_netfs = { .name = "afs", - .version = 1, + .version = 2, }; struct fscache_cookie_def afs_cell_cache_index_def = { .name = "AFS.cell", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = afs_cell_cache_get_key, }; struct fscache_cookie_def afs_volume_cache_index_def = { .name = "AFS.volume", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = afs_volume_cache_get_key, }; struct fscache_cookie_def afs_vnode_cache_index_def = { - .name = "AFS.vnode", - .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .get_key = afs_vnode_cache_get_key, - .get_attr = afs_vnode_cache_get_attr, - .get_aux = afs_vnode_cache_get_aux, - .check_aux = afs_vnode_cache_check_aux, + .name = "AFS.vnode", + .type = FSCACHE_COOKIE_TYPE_DATAFILE, + .check_aux = afs_vnode_cache_check_aux, }; /* - * set the key for the index entry - */ -static uint16_t afs_cell_cache_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct afs_cell *cell = cookie_netfs_data; - uint16_t klen; - - _enter("%p,%p,%u", cell, buffer, bufmax); - - klen = strlen(cell->name); - if (klen > bufmax) - return 0; - - memcpy(buffer, cell->name, klen); - return klen; -} - -/*****************************************************************************/ -/* - * set the key for the volume index entry - */ -static uint16_t afs_volume_cache_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct afs_volume *volume = cookie_netfs_data; - struct { - u64 volid; - } __packed key; - - _enter("{%u},%p,%u", volume->type, buffer, bufmax); - - if (bufmax < sizeof(key)) - return 0; - - key.volid = volume->vid; - memcpy(buffer, &key, sizeof(key)); - return sizeof(key); -} - -/*****************************************************************************/ -/* - * set the key for the index entry - */ -static uint16_t afs_vnode_cache_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct afs_vnode *vnode = cookie_netfs_data; - struct { - u32 vnode_id[3]; - } __packed key; - - _enter("{%x,%x,%llx},%p,%u", - vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version, - buffer, bufmax); - - /* Allow for a 96-bit key */ - memset(&key, 0, sizeof(key)); - key.vnode_id[0] = vnode->fid.vnode; - key.vnode_id[1] = 0; - key.vnode_id[2] = 0; - - if (sizeof(key) > bufmax) - return 0; - - memcpy(buffer, &key, sizeof(key)); - return sizeof(key); -} - -/* - * provide updated file attributes - */ -static void afs_vnode_cache_get_attr(const void *cookie_netfs_data, - uint64_t *size) -{ - const struct afs_vnode *vnode = cookie_netfs_data; - - _enter("{%x,%x,%llx},", - vnode->fid.vnode, vnode->fid.unique, - vnode->status.data_version); - - *size = vnode->status.size; -} - -struct afs_vnode_cache_aux { - u64 data_version; - u32 fid_unique; -} __packed; - -/* - * provide new auxiliary cache data - */ -static uint16_t afs_vnode_cache_get_aux(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct afs_vnode *vnode = cookie_netfs_data; - struct afs_vnode_cache_aux aux; - - _enter("{%x,%x,%Lx},%p,%u", - vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version, - buffer, bufmax); - - memset(&aux, 0, sizeof(aux)); - aux.data_version = vnode->status.data_version; - aux.fid_unique = vnode->fid.unique; - - if (bufmax < sizeof(aux)) - return 0; - - memcpy(buffer, &aux, sizeof(aux)); - return sizeof(aux); -} - -/* * check that the auxiliary data indicates that the entry is still valid */ static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, const void *buffer, - uint16_t buflen) + uint16_t buflen, + loff_t object_size) { struct afs_vnode *vnode = cookie_netfs_data; struct afs_vnode_cache_aux aux; @@ -189,12 +61,6 @@ static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, return FSCACHE_CHECKAUX_OBSOLETE; } - if (vnode->fid.unique != aux.fid_unique) { - _leave(" = OBSOLETE [uniq %x != %x]", - aux.fid_unique, vnode->fid.unique); - return FSCACHE_CHECKAUX_OBSOLETE; - } - if (vnode->status.data_version != aux.data_version) { _leave(" = OBSOLETE [vers %llx != %llx]", aux.data_version, vnode->status.data_version); diff --git a/fs/afs/callback.c b/fs/afs/callback.c index f4291b576054..571437dcb252 100644 --- a/fs/afs/callback.c +++ b/fs/afs/callback.c @@ -23,36 +23,55 @@ /* * Set up an interest-in-callbacks record for a volume on a server and * register it with the server. - * - Called with volume->server_sem held. + * - Called with vnode->io_lock held. */ int afs_register_server_cb_interest(struct afs_vnode *vnode, - struct afs_server_entry *entry) + struct afs_server_list *slist, + unsigned int index) { - struct afs_cb_interest *cbi = entry->cb_interest, *vcbi, *new, *x; + struct afs_server_entry *entry = &slist->servers[index]; + struct afs_cb_interest *cbi, *vcbi, *new, *old; struct afs_server *server = entry->server; again: + if (vnode->cb_interest && + likely(vnode->cb_interest == entry->cb_interest)) + return 0; + + read_lock(&slist->lock); + cbi = afs_get_cb_interest(entry->cb_interest); + read_unlock(&slist->lock); + vcbi = vnode->cb_interest; if (vcbi) { - if (vcbi == cbi) + if (vcbi == cbi) { + afs_put_cb_interest(afs_v2net(vnode), cbi); return 0; + } + /* Use a new interest in the server list for the same server + * rather than an old one that's still attached to a vnode. + */ if (cbi && vcbi->server == cbi->server) { write_seqlock(&vnode->cb_lock); - vnode->cb_interest = afs_get_cb_interest(cbi); + old = vnode->cb_interest; + vnode->cb_interest = cbi; write_sequnlock(&vnode->cb_lock); - afs_put_cb_interest(afs_v2net(vnode), cbi); + afs_put_cb_interest(afs_v2net(vnode), old); return 0; } + /* Re-use the one attached to the vnode. */ if (!cbi && vcbi->server == server) { - afs_get_cb_interest(vcbi); - x = cmpxchg(&entry->cb_interest, cbi, vcbi); - if (x != cbi) { - cbi = x; - afs_put_cb_interest(afs_v2net(vnode), vcbi); + write_lock(&slist->lock); + if (entry->cb_interest) { + write_unlock(&slist->lock); + afs_put_cb_interest(afs_v2net(vnode), cbi); goto again; } + + entry->cb_interest = cbi; + write_unlock(&slist->lock); return 0; } } @@ -72,13 +91,16 @@ again: list_add_tail(&new->cb_link, &server->cb_interests); write_unlock(&server->cb_break_lock); - x = cmpxchg(&entry->cb_interest, cbi, new); - if (x == cbi) { + write_lock(&slist->lock); + if (!entry->cb_interest) { + entry->cb_interest = afs_get_cb_interest(new); cbi = new; + new = NULL; } else { - cbi = x; - afs_put_cb_interest(afs_v2net(vnode), new); + cbi = afs_get_cb_interest(entry->cb_interest); } + write_unlock(&slist->lock); + afs_put_cb_interest(afs_v2net(vnode), new); } ASSERT(cbi); @@ -88,35 +110,18 @@ again: */ write_seqlock(&vnode->cb_lock); - vnode->cb_interest = afs_get_cb_interest(cbi); + old = vnode->cb_interest; + vnode->cb_interest = cbi; vnode->cb_s_break = cbi->server->cb_s_break; + vnode->cb_v_break = vnode->volume->cb_v_break; clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); write_sequnlock(&vnode->cb_lock); + afs_put_cb_interest(afs_v2net(vnode), old); return 0; } /* - * Set a vnode's interest on a server. - */ -void afs_set_cb_interest(struct afs_vnode *vnode, struct afs_cb_interest *cbi) -{ - struct afs_cb_interest *old_cbi = NULL; - - if (vnode->cb_interest == cbi) - return; - - write_seqlock(&vnode->cb_lock); - if (vnode->cb_interest != cbi) { - afs_get_cb_interest(cbi); - old_cbi = vnode->cb_interest; - vnode->cb_interest = cbi; - } - write_sequnlock(&vnode->cb_lock); - afs_put_cb_interest(afs_v2net(vnode), cbi); -} - -/* * Remove an interest on a server. */ void afs_put_cb_interest(struct afs_net *net, struct afs_cb_interest *cbi) @@ -150,6 +155,7 @@ void afs_break_callback(struct afs_vnode *vnode) write_seqlock(&vnode->cb_lock); + clear_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); if (test_and_clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) { vnode->cb_break++; afs_clear_permits(vnode); @@ -190,13 +196,24 @@ static void afs_break_one_callback(struct afs_server *server, if (cbi->vid != fid->vid) continue; - data.volume = NULL; - data.fid = *fid; - inode = ilookup5_nowait(cbi->sb, fid->vnode, afs_iget5_test, &data); - if (inode) { - vnode = AFS_FS_I(inode); - afs_break_callback(vnode); - iput(inode); + if (fid->vnode == 0 && fid->unique == 0) { + /* The callback break applies to an entire volume. */ + struct afs_super_info *as = AFS_FS_S(cbi->sb); + struct afs_volume *volume = as->volume; + + write_lock(&volume->cb_break_lock); + volume->cb_v_break++; + write_unlock(&volume->cb_break_lock); + } else { + data.volume = NULL; + data.fid = *fid; + inode = ilookup5_nowait(cbi->sb, fid->vnode, + afs_iget5_test, &data); + if (inode) { + vnode = AFS_FS_I(inode); + afs_break_callback(vnode); + iput(inode); + } } } @@ -207,21 +224,23 @@ static void afs_break_one_callback(struct afs_server *server, * allow the fileserver to break callback promises */ void afs_break_callbacks(struct afs_server *server, size_t count, - struct afs_callback callbacks[]) + struct afs_callback_break *callbacks) { _enter("%p,%zu,", server, count); ASSERT(server != NULL); ASSERTCMP(count, <=, AFSCBMAX); + /* TODO: Sort the callback break list by volume ID */ + for (; count > 0; callbacks++, count--) { _debug("- Fid { vl=%08x n=%u u=%u } CB { v=%u x=%u t=%u }", callbacks->fid.vid, callbacks->fid.vnode, callbacks->fid.unique, - callbacks->version, - callbacks->expiry, - callbacks->type + callbacks->cb.version, + callbacks->cb.expiry, + callbacks->cb.type ); afs_break_one_callback(server, &callbacks->fid); } diff --git a/fs/afs/cell.c b/fs/afs/cell.c index 3d2c5e0e854e..fdf4c36cff79 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -18,7 +18,7 @@ #include <keys/rxrpc-type.h> #include "internal.h" -unsigned __read_mostly afs_cell_gc_delay = 10; +static unsigned __read_mostly afs_cell_gc_delay = 10; static void afs_manage_cell(struct work_struct *); @@ -75,7 +75,7 @@ struct afs_cell *afs_lookup_cell_rcu(struct afs_net *net, cell = rcu_dereference_raw(net->ws_cell); if (cell) { afs_get_cell(cell); - continue; + break; } ret = -EDESTADDRREQ; continue; @@ -130,6 +130,8 @@ static struct afs_cell *afs_alloc_cell(struct afs_net *net, _leave(" = -ENAMETOOLONG"); return ERR_PTR(-ENAMETOOLONG); } + if (namelen == 5 && memcmp(name, "@cell", 5) == 0) + return ERR_PTR(-EINVAL); _enter("%*.*s,%s", namelen, namelen, name, vllist); @@ -334,8 +336,8 @@ int afs_cell_init(struct afs_net *net, const char *rootcell) return PTR_ERR(new_root); } - set_bit(AFS_CELL_FL_NO_GC, &new_root->flags); - afs_get_cell(new_root); + if (!test_and_set_bit(AFS_CELL_FL_NO_GC, &new_root->flags)) + afs_get_cell(new_root); /* install the new cell */ write_seqlock(&net->cells_lock); @@ -411,7 +413,7 @@ static void afs_cell_destroy(struct rcu_head *rcu) ASSERTCMP(atomic_read(&cell->usage), ==, 0); - afs_put_addrlist(cell->vl_addrs); + afs_put_addrlist(rcu_access_pointer(cell->vl_addrs)); key_put(cell->anonymous_key); kfree(cell); @@ -522,7 +524,9 @@ static int afs_activate_cell(struct afs_net *net, struct afs_cell *cell) #ifdef CONFIG_AFS_FSCACHE cell->cache = fscache_acquire_cookie(afs_cache_netfs.primary_index, &afs_cell_cache_index_def, - cell, true); + cell->name, strlen(cell->name), + NULL, 0, + cell, 0, true); #endif ret = afs_proc_cell_setup(net, cell); if (ret < 0) @@ -547,7 +551,7 @@ static void afs_deactivate_cell(struct afs_net *net, struct afs_cell *cell) spin_unlock(&net->proc_cells_lock); #ifdef CONFIG_AFS_FSCACHE - fscache_relinquish_cookie(cell->cache, 0); + fscache_relinquish_cookie(cell->cache, NULL, false); cell->cache = NULL; #endif diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c index 41e277f57b20..c332c95a6940 100644 --- a/fs/afs/cmservice.c +++ b/fs/afs/cmservice.c @@ -133,21 +133,10 @@ bool afs_cm_incoming_call(struct afs_call *call) } /* - * clean up a cache manager call + * Clean up a cache manager call. */ static void afs_cm_destructor(struct afs_call *call) { - _enter(""); - - /* Break the callbacks here so that we do it after the final ACK is - * received. The step number here must match the final number in - * afs_deliver_cb_callback(). - */ - if (call->unmarshall == 5) { - ASSERT(call->cm_server && call->count && call->request); - afs_break_callbacks(call->cm_server, call->count, call->request); - } - kfree(call->buffer); call->buffer = NULL; } @@ -161,14 +150,14 @@ static void SRXAFSCB_CallBack(struct work_struct *work) _enter(""); - /* be sure to send the reply *before* attempting to spam the AFS server - * with FSFetchStatus requests on the vnodes with broken callbacks lest - * the AFS server get into a vicious cycle of trying to break further - * callbacks because it hadn't received completion of the CBCallBack op - * yet */ - afs_send_empty_reply(call); + /* We need to break the callbacks before sending the reply as the + * server holds up change visibility till it receives our reply so as + * to maintain cache coherency. + */ + if (call->cm_server) + afs_break_callbacks(call->cm_server, call->count, call->request); - afs_break_callbacks(call->cm_server, call->count, call->request); + afs_send_empty_reply(call); afs_put_call(call); _leave(""); } @@ -178,9 +167,8 @@ static void SRXAFSCB_CallBack(struct work_struct *work) */ static int afs_deliver_cb_callback(struct afs_call *call) { + struct afs_callback_break *cb; struct sockaddr_rxrpc srx; - struct afs_callback *cb; - struct afs_server *server; __be32 *bp; int ret, loop; @@ -201,7 +189,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) call->count = ntohl(call->tmp); _debug("FID count: %u", call->count); if (call->count > AFSCBMAX) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); call->buffer = kmalloc(call->count * 3 * 4, GFP_KERNEL); if (!call->buffer) @@ -218,7 +206,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) _debug("unmarshall FID array"); call->request = kcalloc(call->count, - sizeof(struct afs_callback), + sizeof(struct afs_callback_break), GFP_KERNEL); if (!call->request) return -ENOMEM; @@ -229,7 +217,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) cb->fid.vid = ntohl(*bp++); cb->fid.vnode = ntohl(*bp++); cb->fid.unique = ntohl(*bp++); - cb->type = AFSCM_CB_UNTYPED; + cb->cb.type = AFSCM_CB_UNTYPED; } call->offset = 0; @@ -245,7 +233,7 @@ static int afs_deliver_cb_callback(struct afs_call *call) call->count2 = ntohl(call->tmp); _debug("CB count: %u", call->count2); if (call->count2 != call->count && call->count2 != 0) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); call->offset = 0; call->unmarshall++; @@ -260,22 +248,13 @@ static int afs_deliver_cb_callback(struct afs_call *call) cb = call->request; bp = call->buffer; for (loop = call->count2; loop > 0; loop--, cb++) { - cb->version = ntohl(*bp++); - cb->expiry = ntohl(*bp++); - cb->type = ntohl(*bp++); + cb->cb.version = ntohl(*bp++); + cb->cb.expiry = ntohl(*bp++); + cb->cb.type = ntohl(*bp++); } call->offset = 0; call->unmarshall++; - - /* Record that the message was unmarshalled successfully so - * that the call destructor can know do the callback breaking - * work, even if the final ACK isn't received. - * - * If the step number changes, then afs_cm_destructor() must be - * updated also. - */ - call->unmarshall++; case 5: break; } @@ -286,10 +265,9 @@ static int afs_deliver_cb_callback(struct afs_call *call) /* we'll need the file server record as that tells us which set of * vnodes to operate upon */ rxrpc_kernel_get_peer(call->net->socket, call->rxcall, &srx); - server = afs_find_server(call->net, &srx); - if (!server) - return -ENOTCONN; - call->cm_server = server; + call->cm_server = afs_find_server(call->net, &srx); + if (!call->cm_server) + trace_afs_cm_no_server(call, &srx); return afs_queue_call_work(call); } @@ -303,7 +281,8 @@ static void SRXAFSCB_InitCallBackState(struct work_struct *work) _enter("{%p}", call->cm_server); - afs_init_callback_state(call->cm_server); + if (call->cm_server) + afs_init_callback_state(call->cm_server); afs_send_empty_reply(call); afs_put_call(call); _leave(""); @@ -315,7 +294,6 @@ static void SRXAFSCB_InitCallBackState(struct work_struct *work) static int afs_deliver_cb_init_call_back_state(struct afs_call *call) { struct sockaddr_rxrpc srx; - struct afs_server *server; int ret; _enter(""); @@ -328,10 +306,9 @@ static int afs_deliver_cb_init_call_back_state(struct afs_call *call) /* we'll need the file server record as that tells us which set of * vnodes to operate upon */ - server = afs_find_server(call->net, &srx); - if (!server) - return -ENOTCONN; - call->cm_server = server; + call->cm_server = afs_find_server(call->net, &srx); + if (!call->cm_server) + trace_afs_cm_no_server(call, &srx); return afs_queue_call_work(call); } @@ -341,8 +318,6 @@ static int afs_deliver_cb_init_call_back_state(struct afs_call *call) */ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call) { - struct sockaddr_rxrpc srx; - struct afs_server *server; struct afs_uuid *r; unsigned loop; __be32 *b; @@ -398,11 +373,11 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call) /* we'll need the file server record as that tells us which set of * vnodes to operate upon */ - rxrpc_kernel_get_peer(call->net->socket, call->rxcall, &srx); - server = afs_find_server(call->net, &srx); - if (!server) - return -ENOTCONN; - call->cm_server = server; + rcu_read_lock(); + call->cm_server = afs_find_server_by_uuid(call->net, call->request); + rcu_read_unlock(); + if (!call->cm_server) + trace_afs_cm_no_server_u(call, call->request); return afs_queue_call_work(call); } @@ -500,9 +475,9 @@ static int afs_deliver_cb_probe_uuid(struct afs_call *call) b = call->buffer; r = call->request; - r->time_low = ntohl(b[0]); - r->time_mid = ntohl(b[1]); - r->time_hi_and_version = ntohl(b[2]); + r->time_low = b[0]; + r->time_mid = htons(ntohl(b[1])); + r->time_hi_and_version = htons(ntohl(b[2])); r->clock_seq_hi_and_reserved = ntohl(b[3]); r->clock_seq_low = ntohl(b[4]); diff --git a/fs/afs/dir.c b/fs/afs/dir.c index ba2b458b36d1..7d623008157f 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -1,6 +1,6 @@ /* dir.c: AFS filesystem directory handling * - * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2002, 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -10,27 +10,26 @@ */ #include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> #include <linux/fs.h> #include <linux/namei.h> #include <linux/pagemap.h> +#include <linux/swap.h> #include <linux/ctype.h> #include <linux/sched.h> -#include <linux/dns_resolver.h> +#include <linux/task_io_accounting_ops.h> #include "internal.h" +#include "xdr_fs.h" static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags); -static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentry, - unsigned int flags); static int afs_dir_open(struct inode *inode, struct file *file); static int afs_readdir(struct file *file, struct dir_context *ctx); static int afs_d_revalidate(struct dentry *dentry, unsigned int flags); static int afs_d_delete(const struct dentry *dentry); -static void afs_d_release(struct dentry *dentry); -static int afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, +static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); +static int afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, + loff_t fpos, u64 ino, unsigned dtype); static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl); static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); @@ -43,6 +42,14 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags); +static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags); +static void afs_dir_invalidatepage(struct page *page, unsigned int offset, + unsigned int length); + +static int afs_dir_set_page_dirty(struct page *page) +{ + BUG(); /* This should never happen. */ +} const struct file_operations afs_dir_file_operations = { .open = afs_dir_open, @@ -67,15 +74,10 @@ const struct inode_operations afs_dir_inode_operations = { .listxattr = afs_listxattr, }; -const struct file_operations afs_dynroot_file_operations = { - .open = dcache_dir_open, - .release = dcache_dir_close, - .iterate_shared = dcache_readdir, - .llseek = dcache_dir_lseek, -}; - -const struct inode_operations afs_dynroot_inode_operations = { - .lookup = afs_dynroot_lookup, +const struct address_space_operations afs_dir_aops = { + .set_page_dirty = afs_dir_set_page_dirty, + .releasepage = afs_dir_releasepage, + .invalidatepage = afs_dir_invalidatepage, }; const struct dentry_operations afs_fs_dentry_operations = { @@ -85,91 +87,38 @@ const struct dentry_operations afs_fs_dentry_operations = { .d_automount = afs_d_automount, }; -#define AFS_DIR_HASHTBL_SIZE 128 -#define AFS_DIR_DIRENT_SIZE 32 -#define AFS_DIRENT_PER_BLOCK 64 - -union afs_dirent { - struct { - uint8_t valid; - uint8_t unused[1]; - __be16 hash_next; - __be32 vnode; - __be32 unique; - uint8_t name[16]; - uint8_t overflow[4]; /* if any char of the name (inc - * NUL) reaches here, consume - * the next dirent too */ - } u; - uint8_t extended_name[32]; -}; - -/* AFS directory page header (one at the beginning of every 2048-byte chunk) */ -struct afs_dir_pagehdr { - __be16 npages; - __be16 magic; -#define AFS_DIR_MAGIC htons(1234) - uint8_t nentries; - uint8_t bitmap[8]; - uint8_t pad[19]; -}; - -/* directory block layout */ -union afs_dir_block { - - struct afs_dir_pagehdr pagehdr; - - struct { - struct afs_dir_pagehdr pagehdr; - uint8_t alloc_ctrs[128]; - /* dir hash table */ - uint16_t hashtable[AFS_DIR_HASHTBL_SIZE]; - } hdr; - - union afs_dirent dirents[AFS_DIRENT_PER_BLOCK]; -}; - -/* layout on a linux VM page */ -struct afs_dir_page { - union afs_dir_block blocks[PAGE_SIZE / sizeof(union afs_dir_block)]; +struct afs_lookup_one_cookie { + struct dir_context ctx; + struct qstr name; + bool found; + struct afs_fid fid; }; struct afs_lookup_cookie { - struct dir_context ctx; - struct afs_fid fid; - struct qstr name; - int found; + struct dir_context ctx; + struct qstr name; + bool found; + bool one_only; + unsigned short nr_fids; + struct afs_file_status *statuses; + struct afs_callback *callbacks; + struct afs_fid fids[50]; }; /* * check that a directory page is valid */ -bool afs_dir_check_page(struct inode *dir, struct page *page) +static bool afs_dir_check_page(struct afs_vnode *dvnode, struct page *page, + loff_t i_size) { - struct afs_dir_page *dbuf; - struct afs_vnode *vnode = AFS_FS_I(dir); - loff_t latter, i_size, off; + struct afs_xdr_dir_page *dbuf; + loff_t latter, off; int tmp, qty; -#if 0 - /* check the page count */ - qty = desc.size / sizeof(dbuf->blocks[0]); - if (qty == 0) - goto error; - - if (page->index == 0 && qty != ntohs(dbuf->blocks[0].pagehdr.npages)) { - printk("kAFS: %s(%lu): wrong number of dir blocks %d!=%hu\n", - __func__, dir->i_ino, qty, - ntohs(dbuf->blocks[0].pagehdr.npages)); - goto error; - } -#endif - /* Determine how many magic numbers there should be in this page, but * we must take care because the directory may change size under us. */ off = page_offset(page); - i_size = i_size_read(dir); if (i_size <= off) goto checked; @@ -178,112 +127,222 @@ bool afs_dir_check_page(struct inode *dir, struct page *page) qty = PAGE_SIZE; else qty = latter; - qty /= sizeof(union afs_dir_block); + qty /= sizeof(union afs_xdr_dir_block); /* check them */ - dbuf = page_address(page); + dbuf = kmap(page); for (tmp = 0; tmp < qty; tmp++) { - if (dbuf->blocks[tmp].pagehdr.magic != AFS_DIR_MAGIC) { + if (dbuf->blocks[tmp].hdr.magic != AFS_DIR_MAGIC) { printk("kAFS: %s(%lx): bad magic %d/%d is %04hx\n", - __func__, dir->i_ino, tmp, qty, - ntohs(dbuf->blocks[tmp].pagehdr.magic)); - trace_afs_dir_check_failed(vnode, off, i_size); + __func__, dvnode->vfs_inode.i_ino, tmp, qty, + ntohs(dbuf->blocks[tmp].hdr.magic)); + trace_afs_dir_check_failed(dvnode, off, i_size); + kunmap(page); goto error; } + + /* Make sure each block is NUL terminated so we can reasonably + * use string functions on it. The filenames in the page + * *should* be NUL-terminated anyway. + */ + ((u8 *)&dbuf->blocks[tmp])[AFS_DIR_BLOCK_SIZE - 1] = 0; } + kunmap(page); + checked: - SetPageChecked(page); + afs_stat_v(dvnode, n_read_dir); return true; error: - SetPageError(page); return false; } /* - * discard a page cached in the pagecache + * open an AFS directory file */ -static inline void afs_dir_put_page(struct page *page) +static int afs_dir_open(struct inode *inode, struct file *file) { - kunmap(page); - unlock_page(page); - put_page(page); + _enter("{%lu}", inode->i_ino); + + BUILD_BUG_ON(sizeof(union afs_xdr_dir_block) != 2048); + BUILD_BUG_ON(sizeof(union afs_xdr_dirent) != 32); + + if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(inode)->flags)) + return -ENOENT; + + return afs_open(inode, file); } /* - * get a page into the pagecache + * Read the directory into the pagecache in one go, scrubbing the previous + * contents. The list of pages is returned, pinning them so that they don't + * get reclaimed during the iteration. */ -static struct page *afs_dir_get_page(struct inode *dir, unsigned long index, - struct key *key) +static struct afs_read *afs_read_dir(struct afs_vnode *dvnode, struct key *key) + __acquires(&dvnode->validate_lock) { - struct page *page; - _enter("{%lu},%lu", dir->i_ino, index); - - page = read_cache_page(dir->i_mapping, index, afs_page_filler, key); - if (!IS_ERR(page)) { - lock_page(page); - kmap(page); - if (unlikely(!PageChecked(page))) { - if (PageError(page)) - goto fail; - } + struct afs_read *req; + loff_t i_size; + int nr_pages, nr_inline, i, n; + int ret = -ENOMEM; + +retry: + i_size = i_size_read(&dvnode->vfs_inode); + if (i_size < 2048) + return ERR_PTR(-EIO); + if (i_size > 2048 * 1024) + return ERR_PTR(-EFBIG); + + _enter("%llu", i_size); + + /* Get a request record to hold the page list. We want to hold it + * inline if we can, but we don't want to make an order 1 allocation. + */ + nr_pages = (i_size + PAGE_SIZE - 1) / PAGE_SIZE; + nr_inline = nr_pages; + if (nr_inline > (PAGE_SIZE - sizeof(*req)) / sizeof(struct page *)) + nr_inline = 0; + + req = kzalloc(sizeof(*req) + sizeof(struct page *) * nr_inline, + GFP_KERNEL); + if (!req) + return ERR_PTR(-ENOMEM); + + refcount_set(&req->usage, 1); + req->nr_pages = nr_pages; + req->actual_len = i_size; /* May change */ + req->len = nr_pages * PAGE_SIZE; /* We can ask for more than there is */ + req->data_version = dvnode->status.data_version; /* May change */ + if (nr_inline > 0) { + req->pages = req->array; + } else { + req->pages = kcalloc(nr_pages, sizeof(struct page *), + GFP_KERNEL); + if (!req->pages) + goto error; } - return page; -fail: - afs_dir_put_page(page); - _leave(" = -EIO"); - return ERR_PTR(-EIO); -} + /* Get a list of all the pages that hold or will hold the directory + * content. We need to fill in any gaps that we might find where the + * memory reclaimer has been at work. If there are any gaps, we will + * need to reread the entire directory contents. + */ + i = 0; + do { + n = find_get_pages_contig(dvnode->vfs_inode.i_mapping, i, + req->nr_pages - i, + req->pages + i); + _debug("find %u at %u/%u", n, i, req->nr_pages); + if (n == 0) { + gfp_t gfp = dvnode->vfs_inode.i_mapping->gfp_mask; + + if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_stat_v(dvnode, n_inval); + + ret = -ENOMEM; + req->pages[i] = __page_cache_alloc(gfp); + if (!req->pages[i]) + goto error; + ret = add_to_page_cache_lru(req->pages[i], + dvnode->vfs_inode.i_mapping, + i, gfp); + if (ret < 0) + goto error; + + set_page_private(req->pages[i], 1); + SetPagePrivate(req->pages[i]); + unlock_page(req->pages[i]); + i++; + } else { + i += n; + } + } while (i < req->nr_pages); -/* - * open an AFS directory file - */ -static int afs_dir_open(struct inode *inode, struct file *file) -{ - _enter("{%lu}", inode->i_ino); + /* If we're going to reload, we need to lock all the pages to prevent + * races. + */ + ret = -ERESTARTSYS; + if (down_read_killable(&dvnode->validate_lock) < 0) + goto error; - BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); - BUILD_BUG_ON(sizeof(union afs_dirent) != 32); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + goto success; - if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(inode)->flags)) - return -ENOENT; + up_read(&dvnode->validate_lock); + if (down_write_killable(&dvnode->validate_lock) < 0) + goto error; - return afs_open(inode, file); + if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { + ret = afs_fetch_data(dvnode, key, req); + if (ret < 0) + goto error_unlock; + + task_io_account_read(PAGE_SIZE * req->nr_pages); + + if (req->len < req->file_size) + goto content_has_grown; + + /* Validate the data we just read. */ + ret = -EIO; + for (i = 0; i < req->nr_pages; i++) + if (!afs_dir_check_page(dvnode, req->pages[i], + req->actual_len)) + goto error_unlock; + + // TODO: Trim excess pages + + set_bit(AFS_VNODE_DIR_VALID, &dvnode->flags); + } + + downgrade_write(&dvnode->validate_lock); +success: + return req; + +error_unlock: + up_write(&dvnode->validate_lock); +error: + afs_put_read(req); + _leave(" = %d", ret); + return ERR_PTR(ret); + +content_has_grown: + up_write(&dvnode->validate_lock); + afs_put_read(req); + goto retry; } /* * deal with one block in an AFS directory */ static int afs_dir_iterate_block(struct dir_context *ctx, - union afs_dir_block *block, + union afs_xdr_dir_block *block, unsigned blkoff) { - union afs_dirent *dire; + union afs_xdr_dirent *dire; unsigned offset, next, curr; size_t nlen; int tmp; _enter("%u,%x,%p,,",(unsigned)ctx->pos,blkoff,block); - curr = (ctx->pos - blkoff) / sizeof(union afs_dirent); + curr = (ctx->pos - blkoff) / sizeof(union afs_xdr_dirent); /* walk through the block, an entry at a time */ - for (offset = AFS_DIRENT_PER_BLOCK - block->pagehdr.nentries; - offset < AFS_DIRENT_PER_BLOCK; + for (offset = (blkoff == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); + offset < AFS_DIR_SLOTS_PER_BLOCK; offset = next ) { next = offset + 1; /* skip entries marked unused in the bitmap */ - if (!(block->pagehdr.bitmap[offset / 8] & + if (!(block->hdr.bitmap[offset / 8] & (1 << (offset % 8)))) { _debug("ENT[%zu.%u]: unused", - blkoff / sizeof(union afs_dir_block), offset); + blkoff / sizeof(union afs_xdr_dir_block), offset); if (offset >= curr) ctx->pos = blkoff + - next * sizeof(union afs_dirent); + next * sizeof(union afs_xdr_dirent); continue; } @@ -291,34 +350,34 @@ static int afs_dir_iterate_block(struct dir_context *ctx, dire = &block->dirents[offset]; nlen = strnlen(dire->u.name, sizeof(*block) - - offset * sizeof(union afs_dirent)); + offset * sizeof(union afs_xdr_dirent)); _debug("ENT[%zu.%u]: %s %zu \"%s\"", - blkoff / sizeof(union afs_dir_block), offset, + blkoff / sizeof(union afs_xdr_dir_block), offset, (offset < curr ? "skip" : "fill"), nlen, dire->u.name); /* work out where the next possible entry is */ - for (tmp = nlen; tmp > 15; tmp -= sizeof(union afs_dirent)) { - if (next >= AFS_DIRENT_PER_BLOCK) { + for (tmp = nlen; tmp > 15; tmp -= sizeof(union afs_xdr_dirent)) { + if (next >= AFS_DIR_SLOTS_PER_BLOCK) { _debug("ENT[%zu.%u]:" " %u travelled beyond end dir block" " (len %u/%zu)", - blkoff / sizeof(union afs_dir_block), + blkoff / sizeof(union afs_xdr_dir_block), offset, next, tmp, nlen); return -EIO; } - if (!(block->pagehdr.bitmap[next / 8] & + if (!(block->hdr.bitmap[next / 8] & (1 << (next % 8)))) { _debug("ENT[%zu.%u]:" " %u unmarked extension (len %u/%zu)", - blkoff / sizeof(union afs_dir_block), + blkoff / sizeof(union afs_xdr_dir_block), offset, next, tmp, nlen); return -EIO; } _debug("ENT[%zu.%u]: ext %u/%zu", - blkoff / sizeof(union afs_dir_block), + blkoff / sizeof(union afs_xdr_dir_block), next, tmp, nlen); next++; } @@ -330,13 +389,14 @@ static int afs_dir_iterate_block(struct dir_context *ctx, /* found the next entry */ if (!dir_emit(ctx, dire->u.name, nlen, ntohl(dire->u.vnode), - ctx->actor == afs_lookup_filldir ? + (ctx->actor == afs_lookup_filldir || + ctx->actor == afs_lookup_one_filldir)? ntohl(dire->u.unique) : DT_UNKNOWN)) { _leave(" = 0 [full]"); return 0; } - ctx->pos = blkoff + next * sizeof(union afs_dirent); + ctx->pos = blkoff + next * sizeof(union afs_xdr_dirent); } _leave(" = 1 [more]"); @@ -349,8 +409,10 @@ static int afs_dir_iterate_block(struct dir_context *ctx, static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, struct key *key) { - union afs_dir_block *dblock; - struct afs_dir_page *dbuf; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct afs_xdr_dir_page *dbuf; + union afs_xdr_dir_block *dblock; + struct afs_read *req; struct page *page; unsigned blkoff, limit; int ret; @@ -362,45 +424,54 @@ static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, return -ESTALE; } + req = afs_read_dir(dvnode, key); + if (IS_ERR(req)) + return PTR_ERR(req); + /* round the file position up to the next entry boundary */ - ctx->pos += sizeof(union afs_dirent) - 1; - ctx->pos &= ~(sizeof(union afs_dirent) - 1); + ctx->pos += sizeof(union afs_xdr_dirent) - 1; + ctx->pos &= ~(sizeof(union afs_xdr_dirent) - 1); /* walk through the blocks in sequence */ ret = 0; - while (ctx->pos < dir->i_size) { - blkoff = ctx->pos & ~(sizeof(union afs_dir_block) - 1); + while (ctx->pos < req->actual_len) { + blkoff = ctx->pos & ~(sizeof(union afs_xdr_dir_block) - 1); - /* fetch the appropriate page from the directory */ - page = afs_dir_get_page(dir, blkoff / PAGE_SIZE, key); - if (IS_ERR(page)) { - ret = PTR_ERR(page); + /* Fetch the appropriate page from the directory and re-add it + * to the LRU. + */ + page = req->pages[blkoff / PAGE_SIZE]; + if (!page) { + ret = -EIO; break; } + mark_page_accessed(page); limit = blkoff & ~(PAGE_SIZE - 1); - dbuf = page_address(page); + dbuf = kmap(page); /* deal with the individual blocks stashed on this page */ do { dblock = &dbuf->blocks[(blkoff % PAGE_SIZE) / - sizeof(union afs_dir_block)]; + sizeof(union afs_xdr_dir_block)]; ret = afs_dir_iterate_block(ctx, dblock, blkoff); if (ret != 1) { - afs_dir_put_page(page); + kunmap(page); goto out; } - blkoff += sizeof(union afs_dir_block); + blkoff += sizeof(union afs_xdr_dir_block); } while (ctx->pos < dir->i_size && blkoff < limit); - afs_dir_put_page(page); + kunmap(page); ret = 0; } out: + up_read(&dvnode->validate_lock); + afs_put_read(req); _leave(" = %d", ret); return ret; } @@ -414,23 +485,23 @@ static int afs_readdir(struct file *file, struct dir_context *ctx) } /* - * search the directory for a name + * Search the directory for a single name * - if afs_dir_iterate_block() spots this function, it'll pass the FID * uniquifier through dtype */ -static int afs_lookup_filldir(struct dir_context *ctx, const char *name, - int nlen, loff_t fpos, u64 ino, unsigned dtype) +static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, + int nlen, loff_t fpos, u64 ino, unsigned dtype) { - struct afs_lookup_cookie *cookie = - container_of(ctx, struct afs_lookup_cookie, ctx); + struct afs_lookup_one_cookie *cookie = + container_of(ctx, struct afs_lookup_one_cookie, ctx); _enter("{%s,%u},%s,%u,,%llu,%u", cookie->name.name, cookie->name.len, name, nlen, (unsigned long long) ino, dtype); /* insanity checks first */ - BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); - BUILD_BUG_ON(sizeof(union afs_dirent) != 32); + BUILD_BUG_ON(sizeof(union afs_xdr_dir_block) != 2048); + BUILD_BUG_ON(sizeof(union afs_xdr_dirent) != 32); if (cookie->name.len != nlen || memcmp(cookie->name.name, name, nlen) != 0) { @@ -447,15 +518,15 @@ static int afs_lookup_filldir(struct dir_context *ctx, const char *name, } /* - * do a lookup in a directory + * Do a lookup of a single name in a directory * - just returns the FID the dentry name maps to if found */ -static int afs_do_lookup(struct inode *dir, struct dentry *dentry, - struct afs_fid *fid, struct key *key) +static int afs_do_lookup_one(struct inode *dir, struct dentry *dentry, + struct afs_fid *fid, struct key *key) { struct afs_super_info *as = dir->i_sb->s_fs_info; - struct afs_lookup_cookie cookie = { - .ctx.actor = afs_lookup_filldir, + struct afs_lookup_one_cookie cookie = { + .ctx.actor = afs_lookup_one_filldir, .name = dentry->d_name, .fid.vid = as->volume->vid }; @@ -482,70 +553,265 @@ static int afs_do_lookup(struct inode *dir, struct dentry *dentry, } /* - * Probe to see if a cell may exist. This prevents positive dentries from - * being created unnecessarily. + * search the directory for a name + * - if afs_dir_iterate_block() spots this function, it'll pass the FID + * uniquifier through dtype */ -static int afs_probe_cell_name(struct dentry *dentry) +static int afs_lookup_filldir(struct dir_context *ctx, const char *name, + int nlen, loff_t fpos, u64 ino, unsigned dtype) { - struct afs_cell *cell; - const char *name = dentry->d_name.name; - size_t len = dentry->d_name.len; + struct afs_lookup_cookie *cookie = + container_of(ctx, struct afs_lookup_cookie, ctx); int ret; - /* Names prefixed with a dot are R/W mounts. */ - if (name[0] == '.') { - if (len == 1) - return -EINVAL; - name++; - len--; - } + _enter("{%s,%u},%s,%u,,%llu,%u", + cookie->name.name, cookie->name.len, name, nlen, + (unsigned long long) ino, dtype); - cell = afs_lookup_cell_rcu(afs_d2net(dentry), name, len); - if (!IS_ERR(cell)) { - afs_put_cell(afs_d2net(dentry), cell); - return 0; + /* insanity checks first */ + BUILD_BUG_ON(sizeof(union afs_xdr_dir_block) != 2048); + BUILD_BUG_ON(sizeof(union afs_xdr_dirent) != 32); + + if (cookie->found) { + if (cookie->nr_fids < 50) { + cookie->fids[cookie->nr_fids].vnode = ino; + cookie->fids[cookie->nr_fids].unique = dtype; + cookie->nr_fids++; + } + } else if (cookie->name.len == nlen && + memcmp(cookie->name.name, name, nlen) == 0) { + cookie->fids[0].vnode = ino; + cookie->fids[0].unique = dtype; + cookie->found = 1; + if (cookie->one_only) + return -1; } - ret = dns_query("afsdb", name, len, "ipv4", NULL, NULL); - if (ret == -ENODATA) - ret = -EDESTADDRREQ; + ret = cookie->nr_fids >= 50 ? -1 : 0; + _leave(" = %d", ret); return ret; } /* - * Try to auto mount the mountpoint with pseudo directory, if the autocell - * operation is setted. + * Do a lookup in a directory. We make use of bulk lookup to query a slew of + * files in one go and create inodes for them. The inode of the file we were + * asked for is returned. */ -static struct inode *afs_try_auto_mntpt(struct dentry *dentry, - struct inode *dir, struct afs_fid *fid) +static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, + struct key *key) { - struct afs_vnode *vnode = AFS_FS_I(dir); - struct inode *inode; - int ret = -ENOENT; + struct afs_lookup_cookie *cookie; + struct afs_cb_interest *cbi = NULL; + struct afs_super_info *as = dir->i_sb->s_fs_info; + struct afs_iget_data data; + struct afs_fs_cursor fc; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct inode *inode = NULL; + int ret, i; + + _enter("{%lu},%p{%pd},", dir->i_ino, dentry, dentry); + + cookie = kzalloc(sizeof(struct afs_lookup_cookie), GFP_KERNEL); + if (!cookie) + return ERR_PTR(-ENOMEM); + + cookie->ctx.actor = afs_lookup_filldir; + cookie->name = dentry->d_name; + cookie->nr_fids = 1; /* slot 0 is saved for the fid we actually want */ - _enter("%p{%pd}, {%x:%u}", - dentry, dentry, vnode->fid.vid, vnode->fid.vnode); + read_seqlock_excl(&dvnode->cb_lock); + if (dvnode->cb_interest && + dvnode->cb_interest->server && + test_bit(AFS_SERVER_FL_NO_IBULK, &dvnode->cb_interest->server->flags)) + cookie->one_only = true; + read_sequnlock_excl(&dvnode->cb_lock); - if (!test_bit(AFS_VNODE_AUTOCELL, &vnode->flags)) + for (i = 0; i < 50; i++) + cookie->fids[i].vid = as->volume->vid; + + /* search the directory */ + ret = afs_dir_iterate(dir, &cookie->ctx, key); + if (ret < 0) { + inode = ERR_PTR(ret); goto out; + } - ret = afs_probe_cell_name(dentry); - if (ret < 0) + inode = ERR_PTR(-ENOENT); + if (!cookie->found) goto out; - inode = afs_iget_pseudo_dir(dir->i_sb, false); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); + /* Check to see if we already have an inode for the primary fid. */ + data.volume = dvnode->volume; + data.fid = cookie->fids[0]; + inode = ilookup5(dir->i_sb, cookie->fids[0].vnode, afs_iget5_test, &data); + if (inode) + goto out; + + /* Need space for examining all the selected files */ + inode = ERR_PTR(-ENOMEM); + cookie->statuses = kcalloc(cookie->nr_fids, sizeof(struct afs_file_status), + GFP_KERNEL); + if (!cookie->statuses) goto out; + + cookie->callbacks = kcalloc(cookie->nr_fids, sizeof(struct afs_callback), + GFP_KERNEL); + if (!cookie->callbacks) + goto out_s; + + /* Try FS.InlineBulkStatus first. Abort codes for the individual + * lookups contained therein are stored in the reply without aborting + * the whole operation. + */ + if (cookie->one_only) + goto no_inline_bulk_status; + + inode = ERR_PTR(-ERESTARTSYS); + if (afs_begin_vnode_operation(&fc, dvnode, key)) { + while (afs_select_fileserver(&fc)) { + if (test_bit(AFS_SERVER_FL_NO_IBULK, + &fc.cbi->server->flags)) { + fc.ac.abort_code = RX_INVALID_OPERATION; + fc.ac.error = -ECONNABORTED; + break; + } + afs_fs_inline_bulk_status(&fc, + afs_v2net(dvnode), + cookie->fids, + cookie->statuses, + cookie->callbacks, + cookie->nr_fids, NULL); + } + + if (fc.ac.error == 0) + cbi = afs_get_cb_interest(fc.cbi); + if (fc.ac.abort_code == RX_INVALID_OPERATION) + set_bit(AFS_SERVER_FL_NO_IBULK, &fc.cbi->server->flags); + inode = ERR_PTR(afs_end_vnode_operation(&fc)); } - *fid = AFS_FS_I(inode)->fid; - _leave("= %p", inode); - return inode; + if (!IS_ERR(inode)) + goto success; + if (fc.ac.abort_code != RX_INVALID_OPERATION) + goto out_c; +no_inline_bulk_status: + /* We could try FS.BulkStatus next, but this aborts the entire op if + * any of the lookups fails - so, for the moment, revert to + * FS.FetchStatus for just the primary fid. + */ + cookie->nr_fids = 1; + inode = ERR_PTR(-ERESTARTSYS); + if (afs_begin_vnode_operation(&fc, dvnode, key)) { + while (afs_select_fileserver(&fc)) { + afs_fs_fetch_status(&fc, + afs_v2net(dvnode), + cookie->fids, + cookie->statuses, + cookie->callbacks, + NULL); + } + + if (fc.ac.error == 0) + cbi = afs_get_cb_interest(fc.cbi); + inode = ERR_PTR(afs_end_vnode_operation(&fc)); + } + + if (IS_ERR(inode)) + goto out_c; + + for (i = 0; i < cookie->nr_fids; i++) + cookie->statuses[i].abort_code = 0; + +success: + /* Turn all the files into inodes and save the first one - which is the + * one we actually want. + */ + if (cookie->statuses[0].abort_code != 0) + inode = ERR_PTR(afs_abort_to_error(cookie->statuses[0].abort_code)); + + for (i = 0; i < cookie->nr_fids; i++) { + struct inode *ti; + + if (cookie->statuses[i].abort_code != 0) + continue; + + ti = afs_iget(dir->i_sb, key, &cookie->fids[i], + &cookie->statuses[i], + &cookie->callbacks[i], + cbi); + if (i == 0) { + inode = ti; + } else { + if (!IS_ERR(ti)) + iput(ti); + } + } + +out_c: + afs_put_cb_interest(afs_v2net(dvnode), cbi); + kfree(cookie->callbacks); +out_s: + kfree(cookie->statuses); out: - _leave("= %d", ret); - return ERR_PTR(ret); + kfree(cookie); + return inode; +} + +/* + * Look up an entry in a directory with @sys substitution. + */ +static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry, + struct key *key) +{ + struct afs_sysnames *subs; + struct afs_net *net = afs_i2net(dir); + struct dentry *ret; + char *buf, *p, *name; + int len, i; + + _enter(""); + + ret = ERR_PTR(-ENOMEM); + p = buf = kmalloc(AFSNAMEMAX, GFP_KERNEL); + if (!buf) + goto out_p; + if (dentry->d_name.len > 4) { + memcpy(p, dentry->d_name.name, dentry->d_name.len - 4); + p += dentry->d_name.len - 4; + } + + /* There is an ordered list of substitutes that we have to try. */ + read_lock(&net->sysnames_lock); + subs = net->sysnames; + refcount_inc(&subs->usage); + read_unlock(&net->sysnames_lock); + + for (i = 0; i < subs->nr; i++) { + name = subs->subs[i]; + len = dentry->d_name.len - 4 + strlen(name); + if (len >= AFSNAMEMAX) { + ret = ERR_PTR(-ENAMETOOLONG); + goto out_s; + } + + strcpy(p, name); + ret = lookup_one_len(buf, dentry->d_parent, len); + if (IS_ERR(ret) || d_is_positive(ret)) + goto out_s; + dput(ret); + } + + /* We don't want to d_add() the @sys dentry here as we don't want to + * the cached dentry to hide changes to the sysnames list. + */ + ret = NULL; +out_s: + afs_put_sysnames(subs); + kfree(buf); +out_p: + key_put(key); + return ret; } /* @@ -554,16 +820,13 @@ out: static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { - struct afs_vnode *vnode; - struct afs_fid fid; + struct afs_vnode *dvnode = AFS_FS_I(dir); struct inode *inode; struct key *key; int ret; - vnode = AFS_FS_I(dir); - _enter("{%x:%u},%p{%pd},", - vnode->fid.vid, vnode->fid.vnode, dentry, dentry); + dvnode->fid.vid, dvnode->fid.vnode, dentry, dentry); ASSERTCMP(d_inode(dentry), ==, NULL); @@ -572,28 +835,37 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(-ENAMETOOLONG); } - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { + if (test_bit(AFS_VNODE_DELETED, &dvnode->flags)) { _leave(" = -ESTALE"); return ERR_PTR(-ESTALE); } - key = afs_request_key(vnode->volume->cell); + key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { _leave(" = %ld [key]", PTR_ERR(key)); return ERR_CAST(key); } - ret = afs_validate(vnode, key); + ret = afs_validate(dvnode, key); if (ret < 0) { key_put(key); _leave(" = %d [val]", ret); return ERR_PTR(ret); } - ret = afs_do_lookup(dir, dentry, &fid, key); - if (ret < 0) { + if (dentry->d_name.len >= 4 && + dentry->d_name.name[dentry->d_name.len - 4] == '@' && + dentry->d_name.name[dentry->d_name.len - 3] == 's' && + dentry->d_name.name[dentry->d_name.len - 2] == 'y' && + dentry->d_name.name[dentry->d_name.len - 1] == 's') + return afs_lookup_atsys(dir, dentry, key); + + afs_stat_v(dvnode, n_lookup); + inode = afs_do_lookup(dir, dentry, key); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); if (ret == -ENOENT) { - inode = afs_try_auto_mntpt(dentry, dir, &fid); + inode = afs_try_auto_mntpt(dentry, dir); if (!IS_ERR(inode)) { key_put(key); goto success; @@ -611,10 +883,9 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, _leave(" = %d [do]", ret); return ERR_PTR(ret); } - dentry->d_fsdata = (void *)(unsigned long) vnode->status.data_version; + dentry->d_fsdata = (void *)(unsigned long)dvnode->status.data_version; /* instantiate the dentry */ - inode = afs_iget(dir->i_sb, key, &fid, NULL, NULL, NULL); key_put(key); if (IS_ERR(inode)) { _leave(" = %ld", PTR_ERR(inode)); @@ -623,9 +894,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, success: d_add(dentry, inode); - _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%u }", - fid.vnode, - fid.unique, + _leave(" = 0 { ino=%lu v=%u }", d_inode(dentry)->i_ino, d_inode(dentry)->i_generation); @@ -633,67 +902,23 @@ success: } /* - * Look up an entry in a dynroot directory. - */ -static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentry, - unsigned int flags) -{ - struct afs_vnode *vnode; - struct afs_fid fid; - struct inode *inode; - int ret; - - vnode = AFS_FS_I(dir); - - _enter("%pd", dentry); - - ASSERTCMP(d_inode(dentry), ==, NULL); - - if (dentry->d_name.len >= AFSNAMEMAX) { - _leave(" = -ENAMETOOLONG"); - return ERR_PTR(-ENAMETOOLONG); - } - - inode = afs_try_auto_mntpt(dentry, dir, &fid); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); - if (ret == -ENOENT) { - d_add(dentry, NULL); - _leave(" = NULL [negative]"); - return NULL; - } - _leave(" = %d [do]", ret); - return ERR_PTR(ret); - } - - d_add(dentry, inode); - _leave(" = 0 { ino=%lu v=%u }", - d_inode(dentry)->i_ino, d_inode(dentry)->i_generation); - return NULL; -} - -/* * check that a dentry lookup hit has found a valid entry * - NOTE! the hit can be a negative hit too, so we can't assume we have an * inode */ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) { - struct afs_super_info *as = dentry->d_sb->s_fs_info; struct afs_vnode *vnode, *dir; struct afs_fid uninitialized_var(fid); struct dentry *parent; struct inode *inode; struct key *key; - void *dir_version; + long dir_version, de_version; int ret; if (flags & LOOKUP_RCU) return -ECHILD; - if (as->dyn_root) - return 1; - if (d_really_is_positive(dentry)) { vnode = AFS_FS_I(d_inode(dentry)); _enter("{v={%x:%u} n=%pd fl=%lx},", @@ -729,14 +954,25 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) goto out_bad_parent; } - dir_version = (void *) (unsigned long) dir->status.data_version; - if (dentry->d_fsdata == dir_version) - goto out_valid; /* the dir contents are unchanged */ + /* We only need to invalidate a dentry if the server's copy changed + * behind our back. If we made the change, it's no problem. Note that + * on a 32-bit system, we only have 32 bits in the dentry to store the + * version. + */ + dir_version = (long)dir->status.data_version; + de_version = (long)dentry->d_fsdata; + if (de_version == dir_version) + goto out_valid; + + dir_version = (long)dir->invalid_before; + if (de_version - dir_version >= 0) + goto out_valid; _debug("dir modified"); + afs_stat_v(dir, n_reval); /* search the directory for this vnode */ - ret = afs_do_lookup(&dir->vfs_inode, dentry, &fid, key); + ret = afs_do_lookup_one(&dir->vfs_inode, dentry, &fid, key); switch (ret) { case 0: /* the filename maps to something */ @@ -789,7 +1025,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) } out_valid: - dentry->d_fsdata = dir_version; + dentry->d_fsdata = (void *)dir_version; dput(parent); key_put(key); _leave(" = 1 [valid]"); @@ -840,7 +1076,7 @@ zap: /* * handle dentry release */ -static void afs_d_release(struct dentry *dentry) +void afs_d_release(struct dentry *dentry) { _enter("%pd", dentry); } @@ -854,6 +1090,7 @@ static void afs_vnode_new_inode(struct afs_fs_cursor *fc, struct afs_file_status *newstatus, struct afs_callback *newcb) { + struct afs_vnode *vnode; struct inode *inode; if (fc->ac.error < 0) @@ -871,6 +1108,8 @@ static void afs_vnode_new_inode(struct afs_fs_cursor *fc, return; } + vnode = AFS_FS_I(inode); + set_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); d_add(new_dentry, inode); } @@ -885,6 +1124,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) struct afs_vnode *dvnode = AFS_FS_I(dir); struct afs_fid newfid; struct key *key; + u64 data_version = dvnode->status.data_version; int ret; mode |= S_IFDIR; @@ -901,8 +1141,8 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, dvnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = dvnode->cb_break + dvnode->cb_s_break; - afs_fs_create(&fc, dentry->d_name.name, mode, + fc.cb_break = afs_calc_vnode_cb_break(dvnode); + afs_fs_create(&fc, dentry->d_name.name, mode, data_version, &newfid, &newstatus, &newcb); } @@ -916,6 +1156,11 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) goto error_key; } + if (ret == 0 && + test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, + afs_edit_dir_for_create); + key_put(key); _leave(" = 0"); return 0; @@ -939,6 +1184,7 @@ static void afs_dir_remove_subdir(struct dentry *dentry) clear_nlink(&vnode->vfs_inode); set_bit(AFS_VNODE_DELETED, &vnode->flags); clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); } } @@ -950,6 +1196,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry) struct afs_fs_cursor fc; struct afs_vnode *dvnode = AFS_FS_I(dir); struct key *key; + u64 data_version = dvnode->status.data_version; int ret; _enter("{%x:%u},{%pd}", @@ -964,14 +1211,19 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry) ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, dvnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = dvnode->cb_break + dvnode->cb_s_break; - afs_fs_remove(&fc, dentry->d_name.name, true); + fc.cb_break = afs_calc_vnode_cb_break(dvnode); + afs_fs_remove(&fc, dentry->d_name.name, true, + data_version); } afs_vnode_commit_status(&fc, dvnode, fc.cb_break); ret = afs_end_vnode_operation(&fc); - if (ret == 0) + if (ret == 0) { afs_dir_remove_subdir(dentry); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_edit_dir_remove(dvnode, &dentry->d_name, + afs_edit_dir_for_rmdir); + } } key_put(key); @@ -1036,6 +1288,7 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode; struct key *key; unsigned long d_version = (unsigned long)dentry->d_fsdata; + u64 data_version = dvnode->status.data_version; int ret; _enter("{%x:%u},{%pd}", @@ -1061,8 +1314,9 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, dvnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = dvnode->cb_break + dvnode->cb_s_break; - afs_fs_remove(&fc, dentry->d_name.name, false); + fc.cb_break = afs_calc_vnode_cb_break(dvnode); + afs_fs_remove(&fc, dentry->d_name.name, false, + data_version); } afs_vnode_commit_status(&fc, dvnode, fc.cb_break); @@ -1071,6 +1325,10 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) ret = afs_dir_remove_link( dentry, key, d_version, (unsigned long)dvnode->status.data_version); + if (ret == 0 && + test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_edit_dir_remove(dvnode, &dentry->d_name, + afs_edit_dir_for_unlink); } error_key: @@ -1092,6 +1350,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct afs_vnode *dvnode = AFS_FS_I(dir); struct afs_fid newfid; struct key *key; + u64 data_version = dvnode->status.data_version; int ret; mode |= S_IFREG; @@ -1112,8 +1371,8 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, dvnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = dvnode->cb_break + dvnode->cb_s_break; - afs_fs_create(&fc, dentry->d_name.name, mode, + fc.cb_break = afs_calc_vnode_cb_break(dvnode); + afs_fs_create(&fc, dentry->d_name.name, mode, data_version, &newfid, &newstatus, &newcb); } @@ -1127,6 +1386,10 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, goto error_key; } + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, + afs_edit_dir_for_create); + key_put(key); _leave(" = 0"); return 0; @@ -1148,10 +1411,12 @@ static int afs_link(struct dentry *from, struct inode *dir, struct afs_fs_cursor fc; struct afs_vnode *dvnode, *vnode; struct key *key; + u64 data_version; int ret; vnode = AFS_FS_I(d_inode(from)); dvnode = AFS_FS_I(dir); + data_version = dvnode->status.data_version; _enter("{%x:%u},{%x:%u},{%pd}", vnode->fid.vid, vnode->fid.vnode, @@ -1176,9 +1441,9 @@ static int afs_link(struct dentry *from, struct inode *dir, } while (afs_select_fileserver(&fc)) { - fc.cb_break = dvnode->cb_break + dvnode->cb_s_break; - fc.cb_break_2 = vnode->cb_break + vnode->cb_s_break; - afs_fs_link(&fc, vnode, dentry->d_name.name); + fc.cb_break = afs_calc_vnode_cb_break(dvnode); + fc.cb_break_2 = afs_calc_vnode_cb_break(vnode); + afs_fs_link(&fc, vnode, dentry->d_name.name, data_version); } afs_vnode_commit_status(&fc, dvnode, fc.cb_break); @@ -1194,6 +1459,10 @@ static int afs_link(struct dentry *from, struct inode *dir, goto error_key; } + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_edit_dir_add(dvnode, &dentry->d_name, &vnode->fid, + afs_edit_dir_for_link); + key_put(key); _leave(" = 0"); return 0; @@ -1217,6 +1486,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, struct afs_vnode *dvnode = AFS_FS_I(dir); struct afs_fid newfid; struct key *key; + u64 data_version = dvnode->status.data_version; int ret; _enter("{%x:%u},{%pd},%s", @@ -1240,8 +1510,9 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, dvnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = dvnode->cb_break + dvnode->cb_s_break; - afs_fs_symlink(&fc, dentry->d_name.name, content, + fc.cb_break = afs_calc_vnode_cb_break(dvnode); + afs_fs_symlink(&fc, dentry->d_name.name, + content, data_version, &newfid, &newstatus); } @@ -1255,6 +1526,10 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, goto error_key; } + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, + afs_edit_dir_for_symlink); + key_put(key); _leave(" = 0"); return 0; @@ -1277,6 +1552,8 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, struct afs_fs_cursor fc; struct afs_vnode *orig_dvnode, *new_dvnode, *vnode; struct key *key; + u64 orig_data_version, new_data_version; + bool new_negative = d_is_negative(new_dentry); int ret; if (flags) @@ -1285,6 +1562,8 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, vnode = AFS_FS_I(d_inode(old_dentry)); orig_dvnode = AFS_FS_I(old_dir); new_dvnode = AFS_FS_I(new_dir); + orig_data_version = orig_dvnode->status.data_version; + new_data_version = new_dvnode->status.data_version; _enter("{%x:%u},{%x:%u},{%x:%u},{%pd}", orig_dvnode->fid.vid, orig_dvnode->fid.vnode, @@ -1307,10 +1586,11 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, } } while (afs_select_fileserver(&fc)) { - fc.cb_break = orig_dvnode->cb_break + orig_dvnode->cb_s_break; - fc.cb_break_2 = new_dvnode->cb_break + new_dvnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(orig_dvnode); + fc.cb_break_2 = afs_calc_vnode_cb_break(new_dvnode); afs_fs_rename(&fc, old_dentry->d_name.name, - new_dvnode, new_dentry->d_name.name); + new_dvnode, new_dentry->d_name.name, + orig_data_version, new_data_version); } afs_vnode_commit_status(&fc, orig_dvnode, fc.cb_break); @@ -1322,9 +1602,68 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, goto error_key; } + if (ret == 0) { + if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags)) + afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name, + afs_edit_dir_for_rename); + + if (!new_negative && + test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags)) + afs_edit_dir_remove(new_dvnode, &new_dentry->d_name, + afs_edit_dir_for_rename); + + if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags)) + afs_edit_dir_add(new_dvnode, &new_dentry->d_name, + &vnode->fid, afs_edit_dir_for_rename); + } + error_key: key_put(key); error: _leave(" = %d", ret); return ret; } + +/* + * Release a directory page and clean up its private state if it's not busy + * - return true if the page can now be released, false if not + */ +static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags) +{ + struct afs_vnode *dvnode = AFS_FS_I(page->mapping->host); + + _enter("{{%x:%u}[%lu]}", dvnode->fid.vid, dvnode->fid.vnode, page->index); + + set_page_private(page, 0); + ClearPagePrivate(page); + + /* The directory will need reloading. */ + if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_stat_v(dvnode, n_relpg); + return 1; +} + +/* + * invalidate part or all of a page + * - release a page and clean up its private data if offset is 0 (indicating + * the entire page) + */ +static void afs_dir_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) +{ + struct afs_vnode *dvnode = AFS_FS_I(page->mapping->host); + + _enter("{%lu},%u,%u", page->index, offset, length); + + BUG_ON(!PageLocked(page)); + + /* The directory will need reloading. */ + if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_stat_v(dvnode, n_inval); + + /* we clean up only if the entire page is being invalidated */ + if (offset == 0 && length == PAGE_SIZE) { + set_page_private(page, 0); + ClearPagePrivate(page); + } +} diff --git a/fs/afs/dir_edit.c b/fs/afs/dir_edit.c new file mode 100644 index 000000000000..8b400f5aead5 --- /dev/null +++ b/fs/afs/dir_edit.c @@ -0,0 +1,505 @@ +/* AFS filesystem directory editing + * + * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/pagemap.h> +#include <linux/iversion.h> +#include "internal.h" +#include "xdr_fs.h" + +/* + * Find a number of contiguous clear bits in a directory block bitmask. + * + * There are 64 slots, which means we can load the entire bitmap into a + * variable. The first bit doesn't count as it corresponds to the block header + * slot. nr_slots is between 1 and 9. + */ +static int afs_find_contig_bits(union afs_xdr_dir_block *block, unsigned int nr_slots) +{ + u64 bitmap; + u32 mask; + int bit, n; + + bitmap = (u64)block->hdr.bitmap[0] << 0 * 8; + bitmap |= (u64)block->hdr.bitmap[1] << 1 * 8; + bitmap |= (u64)block->hdr.bitmap[2] << 2 * 8; + bitmap |= (u64)block->hdr.bitmap[3] << 3 * 8; + bitmap |= (u64)block->hdr.bitmap[4] << 4 * 8; + bitmap |= (u64)block->hdr.bitmap[5] << 5 * 8; + bitmap |= (u64)block->hdr.bitmap[6] << 6 * 8; + bitmap |= (u64)block->hdr.bitmap[7] << 7 * 8; + bitmap >>= 1; /* The first entry is metadata */ + bit = 1; + mask = (1 << nr_slots) - 1; + + do { + if (sizeof(unsigned long) == 8) + n = ffz(bitmap); + else + n = ((u32)bitmap) != 0 ? + ffz((u32)bitmap) : + ffz((u32)(bitmap >> 32)) + 32; + bitmap >>= n; + bit += n; + + if ((bitmap & mask) == 0) { + if (bit > 64 - nr_slots) + return -1; + return bit; + } + + n = __ffs(bitmap); + bitmap >>= n; + bit += n; + } while (bitmap); + + return -1; +} + +/* + * Set a number of contiguous bits in the directory block bitmap. + */ +static void afs_set_contig_bits(union afs_xdr_dir_block *block, + int bit, unsigned int nr_slots) +{ + u64 mask, before, after; + + mask = (1 << nr_slots) - 1; + mask <<= bit; + + before = *(u64 *)block->hdr.bitmap; + + block->hdr.bitmap[0] |= (u8)(mask >> 0 * 8); + block->hdr.bitmap[1] |= (u8)(mask >> 1 * 8); + block->hdr.bitmap[2] |= (u8)(mask >> 2 * 8); + block->hdr.bitmap[3] |= (u8)(mask >> 3 * 8); + block->hdr.bitmap[4] |= (u8)(mask >> 4 * 8); + block->hdr.bitmap[5] |= (u8)(mask >> 5 * 8); + block->hdr.bitmap[6] |= (u8)(mask >> 6 * 8); + block->hdr.bitmap[7] |= (u8)(mask >> 7 * 8); + + after = *(u64 *)block->hdr.bitmap; +} + +/* + * Clear a number of contiguous bits in the directory block bitmap. + */ +static void afs_clear_contig_bits(union afs_xdr_dir_block *block, + int bit, unsigned int nr_slots) +{ + u64 mask, before, after; + + mask = (1 << nr_slots) - 1; + mask <<= bit; + + before = *(u64 *)block->hdr.bitmap; + + block->hdr.bitmap[0] &= ~(u8)(mask >> 0 * 8); + block->hdr.bitmap[1] &= ~(u8)(mask >> 1 * 8); + block->hdr.bitmap[2] &= ~(u8)(mask >> 2 * 8); + block->hdr.bitmap[3] &= ~(u8)(mask >> 3 * 8); + block->hdr.bitmap[4] &= ~(u8)(mask >> 4 * 8); + block->hdr.bitmap[5] &= ~(u8)(mask >> 5 * 8); + block->hdr.bitmap[6] &= ~(u8)(mask >> 6 * 8); + block->hdr.bitmap[7] &= ~(u8)(mask >> 7 * 8); + + after = *(u64 *)block->hdr.bitmap; +} + +/* + * Scan a directory block looking for a dirent of the right name. + */ +static int afs_dir_scan_block(union afs_xdr_dir_block *block, struct qstr *name, + unsigned int blocknum) +{ + union afs_xdr_dirent *de; + u64 bitmap; + int d, len, n; + + _enter(""); + + bitmap = (u64)block->hdr.bitmap[0] << 0 * 8; + bitmap |= (u64)block->hdr.bitmap[1] << 1 * 8; + bitmap |= (u64)block->hdr.bitmap[2] << 2 * 8; + bitmap |= (u64)block->hdr.bitmap[3] << 3 * 8; + bitmap |= (u64)block->hdr.bitmap[4] << 4 * 8; + bitmap |= (u64)block->hdr.bitmap[5] << 5 * 8; + bitmap |= (u64)block->hdr.bitmap[6] << 6 * 8; + bitmap |= (u64)block->hdr.bitmap[7] << 7 * 8; + + for (d = (blocknum == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); + d < AFS_DIR_SLOTS_PER_BLOCK; + d++) { + if (!((bitmap >> d) & 1)) + continue; + de = &block->dirents[d]; + if (de->u.valid != 1) + continue; + + /* The block was NUL-terminated by afs_dir_check_page(). */ + len = strlen(de->u.name); + if (len == name->len && + memcmp(de->u.name, name->name, name->len) == 0) + return d; + + n = round_up(12 + len + 1 + 4, AFS_DIR_DIRENT_SIZE); + n /= AFS_DIR_DIRENT_SIZE; + d += n - 1; + } + + return -1; +} + +/* + * Initialise a new directory block. Note that block 0 is special and contains + * some extra metadata. + */ +static void afs_edit_init_block(union afs_xdr_dir_block *meta, + union afs_xdr_dir_block *block, int block_num) +{ + memset(block, 0, sizeof(*block)); + block->hdr.npages = htons(1); + block->hdr.magic = AFS_DIR_MAGIC; + block->hdr.bitmap[0] = 1; + + if (block_num == 0) { + block->hdr.bitmap[0] = 0xff; + block->hdr.bitmap[1] = 0x1f; + memset(block->meta.alloc_ctrs, + AFS_DIR_SLOTS_PER_BLOCK, + sizeof(block->meta.alloc_ctrs)); + meta->meta.alloc_ctrs[0] = + AFS_DIR_SLOTS_PER_BLOCK - AFS_DIR_RESV_BLOCKS0; + } + + if (block_num < AFS_DIR_BLOCKS_WITH_CTR) + meta->meta.alloc_ctrs[block_num] = + AFS_DIR_SLOTS_PER_BLOCK - AFS_DIR_RESV_BLOCKS; +} + +/* + * Edit a directory's file data to add a new directory entry. Doing this after + * create, mkdir, symlink, link or rename if the data version number is + * incremented by exactly one avoids the need to re-download the entire + * directory contents. + * + * The caller must hold the inode locked. + */ +void afs_edit_dir_add(struct afs_vnode *vnode, + struct qstr *name, struct afs_fid *new_fid, + enum afs_edit_dir_reason why) +{ + union afs_xdr_dir_block *meta, *block; + struct afs_xdr_dir_page *meta_page, *dir_page; + union afs_xdr_dirent *de; + struct page *page0, *page; + unsigned int need_slots, nr_blocks, b; + pgoff_t index; + loff_t i_size; + gfp_t gfp; + int slot; + + _enter(",,{%d,%s},", name->len, name->name); + + i_size = i_size_read(&vnode->vfs_inode); + if (i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || + (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + return; + } + + gfp = vnode->vfs_inode.i_mapping->gfp_mask; + page0 = find_or_create_page(vnode->vfs_inode.i_mapping, 0, gfp); + if (!page0) { + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + _leave(" [fgp]"); + return; + } + + /* Work out how many slots we're going to need. */ + need_slots = round_up(12 + name->len + 1 + 4, AFS_DIR_DIRENT_SIZE); + need_slots /= AFS_DIR_DIRENT_SIZE; + + meta_page = kmap(page0); + meta = &meta_page->blocks[0]; + if (i_size == 0) + goto new_directory; + nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; + + /* Find a block that has sufficient slots available. Each VM page + * contains two or more directory blocks. + */ + for (b = 0; b < nr_blocks + 1; b++) { + /* If the directory extended into a new page, then we need to + * tack a new page on the end. + */ + index = b / AFS_DIR_BLOCKS_PER_PAGE; + if (index == 0) { + page = page0; + dir_page = meta_page; + } else { + if (nr_blocks >= AFS_DIR_MAX_BLOCKS) + goto error; + gfp = vnode->vfs_inode.i_mapping->gfp_mask; + page = find_or_create_page(vnode->vfs_inode.i_mapping, + index, gfp); + if (!page) + goto error; + if (!PagePrivate(page)) { + set_page_private(page, 1); + SetPagePrivate(page); + } + dir_page = kmap(page); + } + + /* Abandon the edit if we got a callback break. */ + if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + goto invalidated; + + block = &dir_page->blocks[b % AFS_DIR_BLOCKS_PER_PAGE]; + + _debug("block %u: %2u %3u %u", + b, + (b < AFS_DIR_BLOCKS_WITH_CTR) ? meta->meta.alloc_ctrs[b] : 99, + ntohs(block->hdr.npages), + ntohs(block->hdr.magic)); + + /* Initialise the block if necessary. */ + if (b == nr_blocks) { + _debug("init %u", b); + afs_edit_init_block(meta, block, b); + i_size_write(&vnode->vfs_inode, (b + 1) * AFS_DIR_BLOCK_SIZE); + } + + /* Only lower dir pages have a counter in the header. */ + if (b >= AFS_DIR_BLOCKS_WITH_CTR || + meta->meta.alloc_ctrs[b] >= need_slots) { + /* We need to try and find one or more consecutive + * slots to hold the entry. + */ + slot = afs_find_contig_bits(block, need_slots); + if (slot >= 0) { + _debug("slot %u", slot); + goto found_space; + } + } + + if (page != page0) { + unlock_page(page); + kunmap(page); + put_page(page); + } + } + + /* There are no spare slots of sufficient size, yet the operation + * succeeded. Download the directory again. + */ + trace_afs_edit_dir(vnode, why, afs_edit_dir_create_nospc, 0, 0, 0, 0, name->name); + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + goto out_unmap; + +new_directory: + afs_edit_init_block(meta, meta, 0); + i_size = AFS_DIR_BLOCK_SIZE; + i_size_write(&vnode->vfs_inode, i_size); + slot = AFS_DIR_RESV_BLOCKS0; + page = page0; + block = meta; + nr_blocks = 1; + b = 0; + +found_space: + /* Set the dirent slot. */ + trace_afs_edit_dir(vnode, why, afs_edit_dir_create, b, slot, + new_fid->vnode, new_fid->unique, name->name); + de = &block->dirents[slot]; + de->u.valid = 1; + de->u.unused[0] = 0; + de->u.hash_next = 0; // TODO: Really need to maintain this + de->u.vnode = htonl(new_fid->vnode); + de->u.unique = htonl(new_fid->unique); + memcpy(de->u.name, name->name, name->len + 1); + de->u.name[name->len] = 0; + + /* Adjust the bitmap. */ + afs_set_contig_bits(block, slot, need_slots); + if (page != page0) { + unlock_page(page); + kunmap(page); + put_page(page); + } + + /* Adjust the allocation counter. */ + if (b < AFS_DIR_BLOCKS_WITH_CTR) + meta->meta.alloc_ctrs[b] -= need_slots; + + inode_inc_iversion_raw(&vnode->vfs_inode); + afs_stat_v(vnode, n_dir_cr); + _debug("Insert %s in %u[%u]", name->name, b, slot); + +out_unmap: + unlock_page(page0); + kunmap(page0); + put_page(page0); + _leave(""); + return; + +invalidated: + trace_afs_edit_dir(vnode, why, afs_edit_dir_create_inval, 0, 0, 0, 0, name->name); + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + if (page != page0) { + kunmap(page); + put_page(page); + } + goto out_unmap; + +error: + trace_afs_edit_dir(vnode, why, afs_edit_dir_create_error, 0, 0, 0, 0, name->name); + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + goto out_unmap; +} + +/* + * Edit a directory's file data to remove a new directory entry. Doing this + * after unlink, rmdir or rename if the data version number is incremented by + * exactly one avoids the need to re-download the entire directory contents. + * + * The caller must hold the inode locked. + */ +void afs_edit_dir_remove(struct afs_vnode *vnode, + struct qstr *name, enum afs_edit_dir_reason why) +{ + struct afs_xdr_dir_page *meta_page, *dir_page; + union afs_xdr_dir_block *meta, *block; + union afs_xdr_dirent *de; + struct page *page0, *page; + unsigned int need_slots, nr_blocks, b; + pgoff_t index; + loff_t i_size; + int slot; + + _enter(",,{%d,%s},", name->len, name->name); + + i_size = i_size_read(&vnode->vfs_inode); + if (i_size < AFS_DIR_BLOCK_SIZE || + i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || + (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + return; + } + nr_blocks = i_size / AFS_DIR_BLOCK_SIZE; + + page0 = find_lock_page(vnode->vfs_inode.i_mapping, 0); + if (!page0) { + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + _leave(" [fgp]"); + return; + } + + /* Work out how many slots we're going to discard. */ + need_slots = round_up(12 + name->len + 1 + 4, AFS_DIR_DIRENT_SIZE); + need_slots /= AFS_DIR_DIRENT_SIZE; + + meta_page = kmap(page0); + meta = &meta_page->blocks[0]; + + /* Find a page that has sufficient slots available. Each VM page + * contains two or more directory blocks. + */ + for (b = 0; b < nr_blocks; b++) { + index = b / AFS_DIR_BLOCKS_PER_PAGE; + if (index != 0) { + page = find_lock_page(vnode->vfs_inode.i_mapping, index); + if (!page) + goto error; + dir_page = kmap(page); + } else { + page = page0; + dir_page = meta_page; + } + + /* Abandon the edit if we got a callback break. */ + if (!test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + goto invalidated; + + block = &dir_page->blocks[b % AFS_DIR_BLOCKS_PER_PAGE]; + + if (b > AFS_DIR_BLOCKS_WITH_CTR || + meta->meta.alloc_ctrs[b] <= AFS_DIR_SLOTS_PER_BLOCK - 1 - need_slots) { + slot = afs_dir_scan_block(block, name, b); + if (slot >= 0) + goto found_dirent; + } + + if (page != page0) { + unlock_page(page); + kunmap(page); + put_page(page); + } + } + + /* Didn't find the dirent to clobber. Download the directory again. */ + trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_noent, + 0, 0, 0, 0, name->name); + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + goto out_unmap; + +found_dirent: + de = &block->dirents[slot]; + + trace_afs_edit_dir(vnode, why, afs_edit_dir_delete, b, slot, + ntohl(de->u.vnode), ntohl(de->u.unique), + name->name); + + memset(de, 0, sizeof(*de) * need_slots); + + /* Adjust the bitmap. */ + afs_clear_contig_bits(block, slot, need_slots); + if (page != page0) { + unlock_page(page); + kunmap(page); + put_page(page); + } + + /* Adjust the allocation counter. */ + if (b < AFS_DIR_BLOCKS_WITH_CTR) + meta->meta.alloc_ctrs[b] += need_slots; + + inode_set_iversion_raw(&vnode->vfs_inode, vnode->status.data_version); + afs_stat_v(vnode, n_dir_rm); + _debug("Remove %s from %u[%u]", name->name, b, slot); + +out_unmap: + unlock_page(page0); + kunmap(page0); + put_page(page0); + _leave(""); + return; + +invalidated: + trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_inval, + 0, 0, 0, 0, name->name); + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + if (page != page0) { + unlock_page(page); + kunmap(page); + put_page(page); + } + goto out_unmap; + +error: + trace_afs_edit_dir(vnode, why, afs_edit_dir_delete_error, + 0, 0, 0, 0, name->name); + clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + goto out_unmap; +} diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c new file mode 100644 index 000000000000..983f3946ab57 --- /dev/null +++ b/fs/afs/dynroot.c @@ -0,0 +1,209 @@ +/* dir.c: AFS dynamic root handling + * + * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/dns_resolver.h> +#include "internal.h" + +const struct file_operations afs_dynroot_file_operations = { + .open = dcache_dir_open, + .release = dcache_dir_close, + .iterate_shared = dcache_readdir, + .llseek = dcache_dir_lseek, +}; + +/* + * Probe to see if a cell may exist. This prevents positive dentries from + * being created unnecessarily. + */ +static int afs_probe_cell_name(struct dentry *dentry) +{ + struct afs_cell *cell; + const char *name = dentry->d_name.name; + size_t len = dentry->d_name.len; + int ret; + + /* Names prefixed with a dot are R/W mounts. */ + if (name[0] == '.') { + if (len == 1) + return -EINVAL; + name++; + len--; + } + + cell = afs_lookup_cell_rcu(afs_d2net(dentry), name, len); + if (!IS_ERR(cell)) { + afs_put_cell(afs_d2net(dentry), cell); + return 0; + } + + ret = dns_query("afsdb", name, len, "ipv4", NULL, NULL); + if (ret == -ENODATA) + ret = -EDESTADDRREQ; + return ret; +} + +/* + * Try to auto mount the mountpoint with pseudo directory, if the autocell + * operation is setted. + */ +struct inode *afs_try_auto_mntpt(struct dentry *dentry, struct inode *dir) +{ + struct afs_vnode *vnode = AFS_FS_I(dir); + struct inode *inode; + int ret = -ENOENT; + + _enter("%p{%pd}, {%x:%u}", + dentry, dentry, vnode->fid.vid, vnode->fid.vnode); + + if (!test_bit(AFS_VNODE_AUTOCELL, &vnode->flags)) + goto out; + + ret = afs_probe_cell_name(dentry); + if (ret < 0) + goto out; + + inode = afs_iget_pseudo_dir(dir->i_sb, false); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + goto out; + } + + _leave("= %p", inode); + return inode; + +out: + _leave("= %d", ret); + return ERR_PTR(ret); +} + +/* + * Look up @cell in a dynroot directory. This is a substitution for the + * local cell name for the net namespace. + */ +static struct dentry *afs_lookup_atcell(struct dentry *dentry) +{ + struct afs_cell *cell; + struct afs_net *net = afs_d2net(dentry); + struct dentry *ret; + unsigned int seq = 0; + char *name; + int len; + + if (!net->ws_cell) + return ERR_PTR(-ENOENT); + + ret = ERR_PTR(-ENOMEM); + name = kmalloc(AFS_MAXCELLNAME + 1, GFP_KERNEL); + if (!name) + goto out_p; + + rcu_read_lock(); + do { + read_seqbegin_or_lock(&net->cells_lock, &seq); + cell = rcu_dereference_raw(net->ws_cell); + if (cell) { + len = cell->name_len; + memcpy(name, cell->name, len + 1); + } + } while (need_seqretry(&net->cells_lock, seq)); + done_seqretry(&net->cells_lock, seq); + rcu_read_unlock(); + + ret = ERR_PTR(-ENOENT); + if (!cell) + goto out_n; + + ret = lookup_one_len(name, dentry->d_parent, len); + + /* We don't want to d_add() the @cell dentry here as we don't want to + * the cached dentry to hide changes to the local cell name. + */ + +out_n: + kfree(name); +out_p: + return ret; +} + +/* + * Look up an entry in a dynroot directory. + */ +static struct dentry *afs_dynroot_lookup(struct inode *dir, struct dentry *dentry, + unsigned int flags) +{ + struct afs_vnode *vnode; + struct inode *inode; + int ret; + + vnode = AFS_FS_I(dir); + + _enter("%pd", dentry); + + ASSERTCMP(d_inode(dentry), ==, NULL); + + if (dentry->d_name.len >= AFSNAMEMAX) { + _leave(" = -ENAMETOOLONG"); + return ERR_PTR(-ENAMETOOLONG); + } + + if (dentry->d_name.len == 5 && + memcmp(dentry->d_name.name, "@cell", 5) == 0) + return afs_lookup_atcell(dentry); + + inode = afs_try_auto_mntpt(dentry, dir); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + if (ret == -ENOENT) { + d_add(dentry, NULL); + _leave(" = NULL [negative]"); + return NULL; + } + _leave(" = %d [do]", ret); + return ERR_PTR(ret); + } + + d_add(dentry, inode); + _leave(" = 0 { ino=%lu v=%u }", + d_inode(dentry)->i_ino, d_inode(dentry)->i_generation); + return NULL; +} + +const struct inode_operations afs_dynroot_inode_operations = { + .lookup = afs_dynroot_lookup, +}; + +/* + * Dirs in the dynamic root don't need revalidation. + */ +static int afs_dynroot_d_revalidate(struct dentry *dentry, unsigned int flags) +{ + return 1; +} + +/* + * Allow the VFS to enquire as to whether a dentry should be unhashed (mustn't + * sleep) + * - called from dput() when d_count is going to 0. + * - return 1 to request dentry be unhashed, 0 otherwise + */ +static int afs_dynroot_d_delete(const struct dentry *dentry) +{ + return d_really_is_positive(dentry); +} + +const struct dentry_operations afs_dynroot_dentry_operations = { + .d_revalidate = afs_dynroot_d_revalidate, + .d_delete = afs_dynroot_d_delete, + .d_release = afs_d_release, + .d_automount = afs_d_automount, +}; diff --git a/fs/afs/file.c b/fs/afs/file.c index a39192ced99e..7d4f26198573 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -30,7 +30,6 @@ static int afs_readpages(struct file *filp, struct address_space *mapping, const struct file_operations afs_file_operations = { .open = afs_open, - .flush = afs_flush, .release = afs_release, .llseek = generic_file_llseek, .read_iter = generic_file_read_iter, @@ -146,6 +145,9 @@ int afs_open(struct inode *inode, struct file *file) if (ret < 0) goto error_af; } + + if (file->f_flags & O_TRUNC) + set_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); file->private_data = af; _leave(" = 0"); @@ -170,6 +172,9 @@ int afs_release(struct inode *inode, struct file *file) _enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode); + if ((file->f_mode & FMODE_WRITE)) + return vfs_fsync(file, 0); + file->private_data = NULL; if (af->wb) afs_put_wb_key(af->wb); @@ -187,10 +192,12 @@ void afs_put_read(struct afs_read *req) { int i; - if (atomic_dec_and_test(&req->usage)) { + if (refcount_dec_and_test(&req->usage)) { for (i = 0; i < req->nr_pages; i++) if (req->pages[i]) put_page(req->pages[i]); + if (req->pages != req->array) + kfree(req->pages); kfree(req); } } @@ -231,7 +238,7 @@ int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *de ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, vnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(vnode); afs_fs_fetch_data(&fc, desc); } @@ -240,6 +247,12 @@ int afs_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_read *de ret = afs_end_vnode_operation(&fc); } + if (ret == 0) { + afs_stat_v(vnode, n_fetches); + atomic_long_add(desc->actual_len, + &afs_v2net(vnode)->n_fetch_bytes); + } + _leave(" = %d", ret); return ret; } @@ -297,10 +310,11 @@ int afs_page_filler(void *data, struct page *page) * end of the file, the server will return a short read and the * unmarshalling code will clear the unfilled space. */ - atomic_set(&req->usage, 1); + refcount_set(&req->usage, 1); req->pos = (loff_t)page->index << PAGE_SHIFT; req->len = PAGE_SIZE; req->nr_pages = 1; + req->pages = req->array; req->pages[0] = page; get_page(page); @@ -309,10 +323,6 @@ int afs_page_filler(void *data, struct page *page) ret = afs_fetch_data(vnode, key, req); afs_put_read(req); - if (ret >= 0 && S_ISDIR(inode->i_mode) && - !afs_dir_check_page(inode, page)) - ret = -EIO; - if (ret < 0) { if (ret == -ENOENT) { _debug("got NOENT from server" @@ -339,7 +349,8 @@ int afs_page_filler(void *data, struct page *page) /* send the page to the cache */ #ifdef CONFIG_AFS_FSCACHE if (PageFsCache(page) && - fscache_write_page(vnode->cache, page, GFP_KERNEL) != 0) { + fscache_write_page(vnode->cache, page, vnode->status.size, + GFP_KERNEL) != 0) { fscache_uncache_page(vnode->cache, page); BUG_ON(PageFsCache(page)); } @@ -403,7 +414,8 @@ static void afs_readpages_page_done(struct afs_call *call, struct afs_read *req) /* send the page to the cache */ #ifdef CONFIG_AFS_FSCACHE if (PageFsCache(page) && - fscache_write_page(vnode->cache, page, GFP_KERNEL) != 0) { + fscache_write_page(vnode->cache, page, vnode->status.size, + GFP_KERNEL) != 0) { fscache_uncache_page(vnode->cache, page); BUG_ON(PageFsCache(page)); } @@ -445,10 +457,11 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, if (!req) return -ENOMEM; - atomic_set(&req->usage, 1); + refcount_set(&req->usage, 1); req->page_done = afs_readpages_page_done; req->pos = first->index; req->pos <<= PAGE_SHIFT; + req->pages = req->array; /* Transfer the pages to the request. We add them in until one fails * to add to the LRU and then we stop (as that'll make a hole in the diff --git a/fs/afs/flock.c b/fs/afs/flock.c index c40ba2fe3cbe..dc62d15a964b 100644 --- a/fs/afs/flock.c +++ b/fs/afs/flock.c @@ -86,7 +86,7 @@ static int afs_set_lock(struct afs_vnode *vnode, struct key *key, ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, vnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(vnode); afs_fs_set_lock(&fc, type); } @@ -117,7 +117,7 @@ static int afs_extend_lock(struct afs_vnode *vnode, struct key *key) ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, vnode, key)) { while (afs_select_current_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(vnode); afs_fs_extend_lock(&fc); } @@ -148,7 +148,7 @@ static int afs_release_lock(struct afs_vnode *vnode, struct key *key) ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, vnode, key)) { while (afs_select_current_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(vnode); afs_fs_release_lock(&fc); } @@ -613,7 +613,7 @@ static int afs_do_getlk(struct file *file, struct file_lock *fl) posix_test_lock(file, fl); if (fl->fl_type == F_UNLCK) { /* no local locks; consult the server */ - ret = afs_fetch_status(vnode, key); + ret = afs_fetch_status(vnode, key, false); if (ret < 0) goto error; diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c index 88ec38c2d83c..b273e1d60478 100644 --- a/fs/afs/fsclient.c +++ b/fs/afs/fsclient.c @@ -16,6 +16,7 @@ #include <linux/iversion.h> #include "internal.h" #include "afs_fs.h" +#include "xdr_fs.h" static const struct afs_fid afs_zero_fid; @@ -44,109 +45,207 @@ static void xdr_decode_AFSFid(const __be32 **_bp, struct afs_fid *fid) } /* - * decode an AFSFetchStatus block + * Dump a bad file status record. */ -static void xdr_decode_AFSFetchStatus(const __be32 **_bp, - struct afs_file_status *status, - struct afs_vnode *vnode, - afs_dataversion_t *store_version) +static void xdr_dump_bad(const __be32 *bp) { - afs_dataversion_t expected_version; - const __be32 *bp = *_bp; + __be32 x[4]; + int i; + + pr_notice("AFS XDR: Bad status record\n"); + for (i = 0; i < 5 * 4 * 4; i += 16) { + memcpy(x, bp, 16); + bp += 4; + pr_notice("%03x: %08x %08x %08x %08x\n", + i, ntohl(x[0]), ntohl(x[1]), ntohl(x[2]), ntohl(x[3])); + } + + memcpy(x, bp, 4); + pr_notice("0x50: %08x\n", ntohl(x[0])); +} + +/* + * Update the core inode struct from a returned status record. + */ +void afs_update_inode_from_status(struct afs_vnode *vnode, + struct afs_file_status *status, + const afs_dataversion_t *expected_version, + u8 flags) +{ + struct timespec t; umode_t mode; + + t.tv_sec = status->mtime_client; + t.tv_nsec = 0; + vnode->vfs_inode.i_ctime = t; + vnode->vfs_inode.i_mtime = t; + vnode->vfs_inode.i_atime = t; + + if (flags & (AFS_VNODE_META_CHANGED | AFS_VNODE_NOT_YET_SET)) { + vnode->vfs_inode.i_uid = make_kuid(&init_user_ns, status->owner); + vnode->vfs_inode.i_gid = make_kgid(&init_user_ns, status->group); + set_nlink(&vnode->vfs_inode, status->nlink); + + mode = vnode->vfs_inode.i_mode; + mode &= ~S_IALLUGO; + mode |= status->mode; + barrier(); + vnode->vfs_inode.i_mode = mode; + } + + if (!(flags & AFS_VNODE_NOT_YET_SET)) { + if (expected_version && + *expected_version != status->data_version) { + _debug("vnode modified %llx on {%x:%u} [exp %llx]", + (unsigned long long) status->data_version, + vnode->fid.vid, vnode->fid.vnode, + (unsigned long long) *expected_version); + vnode->invalid_before = status->data_version; + if (vnode->status.type == AFS_FTYPE_DIR) { + if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + afs_stat_v(vnode, n_inval); + } else { + set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); + } + } else if (vnode->status.type == AFS_FTYPE_DIR) { + /* Expected directory change is handled elsewhere so + * that we can locally edit the directory and save on a + * download. + */ + if (test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + flags &= ~AFS_VNODE_DATA_CHANGED; + } + } + + if (flags & (AFS_VNODE_DATA_CHANGED | AFS_VNODE_NOT_YET_SET)) { + inode_set_iversion_raw(&vnode->vfs_inode, status->data_version); + i_size_write(&vnode->vfs_inode, status->size); + } +} + +/* + * decode an AFSFetchStatus block + */ +static int xdr_decode_AFSFetchStatus(struct afs_call *call, + const __be32 **_bp, + struct afs_file_status *status, + struct afs_vnode *vnode, + const afs_dataversion_t *expected_version, + struct afs_read *read_req) +{ + const struct afs_xdr_AFSFetchStatus *xdr = (const void *)*_bp; + bool inline_error = (call->operation_ID == afs_FS_InlineBulkStatus); u64 data_version, size; - bool changed = false; - kuid_t owner; - kgid_t group; + u32 type, abort_code; + u8 flags = 0; + int ret; if (vnode) write_seqlock(&vnode->cb_lock); -#define EXTRACT(DST) \ - do { \ - u32 x = ntohl(*bp++); \ - if (DST != x) \ - changed |= true; \ - DST = x; \ - } while (0) - - status->if_version = ntohl(*bp++); - EXTRACT(status->type); - EXTRACT(status->nlink); - size = ntohl(*bp++); - data_version = ntohl(*bp++); - EXTRACT(status->author); - owner = make_kuid(&init_user_ns, ntohl(*bp++)); - changed |= !uid_eq(owner, status->owner); - status->owner = owner; - EXTRACT(status->caller_access); /* call ticket dependent */ - EXTRACT(status->anon_access); - EXTRACT(status->mode); - bp++; /* parent.vnode */ - bp++; /* parent.unique */ - bp++; /* seg size */ - status->mtime_client = ntohl(*bp++); - status->mtime_server = ntohl(*bp++); - group = make_kgid(&init_user_ns, ntohl(*bp++)); - changed |= !gid_eq(group, status->group); - status->group = group; - bp++; /* sync counter */ - data_version |= (u64) ntohl(*bp++) << 32; - EXTRACT(status->lock_count); - size |= (u64) ntohl(*bp++) << 32; - bp++; /* spare 4 */ - *_bp = bp; + abort_code = ntohl(xdr->abort_code); + + if (xdr->if_version != htonl(AFS_FSTATUS_VERSION)) { + if (xdr->if_version == htonl(0) && + abort_code != 0 && + inline_error) { + /* The OpenAFS fileserver has a bug in FS.InlineBulkStatus + * whereby it doesn't set the interface version in the error + * case. + */ + status->abort_code = abort_code; + ret = 0; + goto out; + } - if (size != status->size) { - status->size = size; - changed |= true; + pr_warn("Unknown AFSFetchStatus version %u\n", ntohl(xdr->if_version)); + goto bad; } - status->mode &= S_IALLUGO; - _debug("vnode time %lx, %lx", - status->mtime_client, status->mtime_server); + if (abort_code != 0 && inline_error) { + status->abort_code = abort_code; + ret = 0; + goto out; + } - if (vnode) { - if (changed && !test_bit(AFS_VNODE_UNSET, &vnode->flags)) { - _debug("vnode changed"); - i_size_write(&vnode->vfs_inode, size); - vnode->vfs_inode.i_uid = status->owner; - vnode->vfs_inode.i_gid = status->group; - vnode->vfs_inode.i_generation = vnode->fid.unique; - set_nlink(&vnode->vfs_inode, status->nlink); - - mode = vnode->vfs_inode.i_mode; - mode &= ~S_IALLUGO; - mode |= status->mode; - barrier(); - vnode->vfs_inode.i_mode = mode; + type = ntohl(xdr->type); + switch (type) { + case AFS_FTYPE_FILE: + case AFS_FTYPE_DIR: + case AFS_FTYPE_SYMLINK: + if (type != status->type && + vnode && + !test_bit(AFS_VNODE_UNSET, &vnode->flags)) { + pr_warning("Vnode %x:%x:%x changed type %u to %u\n", + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + status->type, type); + goto bad; } - - vnode->vfs_inode.i_ctime.tv_sec = status->mtime_client; - vnode->vfs_inode.i_mtime = vnode->vfs_inode.i_ctime; - vnode->vfs_inode.i_atime = vnode->vfs_inode.i_ctime; - inode_set_iversion_raw(&vnode->vfs_inode, data_version); + status->type = type; + break; + default: + goto bad; } - expected_version = status->data_version; - if (store_version) - expected_version = *store_version; +#define EXTRACT_M(FIELD) \ + do { \ + u32 x = ntohl(xdr->FIELD); \ + if (status->FIELD != x) { \ + flags |= AFS_VNODE_META_CHANGED; \ + status->FIELD = x; \ + } \ + } while (0) - if (expected_version != data_version) { - status->data_version = data_version; - if (vnode && !test_bit(AFS_VNODE_UNSET, &vnode->flags)) { - _debug("vnode modified %llx on {%x:%u}", - (unsigned long long) data_version, - vnode->fid.vid, vnode->fid.vnode); - set_bit(AFS_VNODE_DIR_MODIFIED, &vnode->flags); - set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); - } - } else if (store_version) { + EXTRACT_M(nlink); + EXTRACT_M(author); + EXTRACT_M(owner); + EXTRACT_M(caller_access); /* call ticket dependent */ + EXTRACT_M(anon_access); + EXTRACT_M(mode); + EXTRACT_M(group); + + status->mtime_client = ntohl(xdr->mtime_client); + status->mtime_server = ntohl(xdr->mtime_server); + status->lock_count = ntohl(xdr->lock_count); + + size = (u64)ntohl(xdr->size_lo); + size |= (u64)ntohl(xdr->size_hi) << 32; + status->size = size; + + data_version = (u64)ntohl(xdr->data_version_lo); + data_version |= (u64)ntohl(xdr->data_version_hi) << 32; + if (data_version != status->data_version) { status->data_version = data_version; + flags |= AFS_VNODE_DATA_CHANGED; + } + + if (read_req) { + read_req->data_version = data_version; + read_req->file_size = size; + } + + *_bp = (const void *)*_bp + sizeof(*xdr); + + if (vnode) { + if (test_bit(AFS_VNODE_UNSET, &vnode->flags)) + flags |= AFS_VNODE_NOT_YET_SET; + afs_update_inode_from_status(vnode, status, expected_version, + flags); } + ret = 0; + +out: if (vnode) write_sequnlock(&vnode->cb_lock); + return ret; + +bad: + xdr_dump_bad(*_bp); + ret = afs_protocol_error(call, -EBADMSG); + goto out; } /* @@ -162,7 +261,7 @@ static void xdr_decode_AFSCallBack(struct afs_call *call, write_seqlock(&vnode->cb_lock); - if (call->cb_break == (vnode->cb_break + cbi->server->cb_s_break)) { + if (call->cb_break == afs_cb_break_sum(vnode, cbi)) { vnode->cb_version = ntohl(*bp++); cb_expiry = ntohl(*bp++); vnode->cb_type = ntohl(*bp++); @@ -274,7 +373,7 @@ static void xdr_decode_AFSFetchVolumeStatus(const __be32 **_bp, /* * deliver reply data to an FS.FetchStatus */ -static int afs_deliver_fs_fetch_status(struct afs_call *call) +static int afs_deliver_fs_fetch_status_vnode(struct afs_call *call) { struct afs_vnode *vnode = call->reply[0]; const __be32 *bp; @@ -288,7 +387,9 @@ static int afs_deliver_fs_fetch_status(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); + if (xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); xdr_decode_AFSCallBack(call, vnode, &bp); if (call->reply[1]) xdr_decode_AFSVolSync(&bp, call->reply[1]); @@ -300,17 +401,18 @@ static int afs_deliver_fs_fetch_status(struct afs_call *call) /* * FS.FetchStatus operation type */ -static const struct afs_call_type afs_RXFSFetchStatus = { - .name = "FS.FetchStatus", +static const struct afs_call_type afs_RXFSFetchStatus_vnode = { + .name = "FS.FetchStatus(vnode)", .op = afs_FS_FetchStatus, - .deliver = afs_deliver_fs_fetch_status, + .deliver = afs_deliver_fs_fetch_status_vnode, .destructor = afs_flat_call_destructor, }; /* * fetch the status information for a file */ -int afs_fs_fetch_file_status(struct afs_fs_cursor *fc, struct afs_volsync *volsync) +int afs_fs_fetch_file_status(struct afs_fs_cursor *fc, struct afs_volsync *volsync, + bool new_inode) { struct afs_vnode *vnode = fc->vnode; struct afs_call *call; @@ -320,7 +422,8 @@ int afs_fs_fetch_file_status(struct afs_fs_cursor *fc, struct afs_volsync *volsy _enter(",%x,{%x:%u},,", key_serial(fc->key), vnode->fid.vid, vnode->fid.vnode); - call = afs_alloc_flat_call(net, &afs_RXFSFetchStatus, 16, (21 + 3 + 6) * 4); + call = afs_alloc_flat_call(net, &afs_RXFSFetchStatus_vnode, + 16, (21 + 3 + 6) * 4); if (!call) { fc->ac.error = -ENOMEM; return -ENOMEM; @@ -329,6 +432,7 @@ int afs_fs_fetch_file_status(struct afs_fs_cursor *fc, struct afs_volsync *volsy call->key = fc->key; call->reply[0] = vnode; call->reply[1] = volsync; + call->expected_version = new_inode ? 1 : vnode->status.data_version; /* marshall the parameters */ bp = call->request; @@ -464,7 +568,9 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call) return ret; bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); + if (xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, + &vnode->status.data_version, req) < 0) + return afs_protocol_error(call, -EBADMSG); xdr_decode_AFSCallBack(call, vnode, &bp); if (call->reply[1]) xdr_decode_AFSVolSync(&bp, call->reply[1]); @@ -534,6 +640,7 @@ static int afs_fs_fetch_data64(struct afs_fs_cursor *fc, struct afs_read *req) call->reply[0] = vnode; call->reply[1] = NULL; /* volsync */ call->reply[2] = req; + call->expected_version = vnode->status.data_version; /* marshall the parameters */ bp = call->request; @@ -546,7 +653,7 @@ static int afs_fs_fetch_data64(struct afs_fs_cursor *fc, struct afs_read *req) bp[6] = 0; bp[7] = htonl(lower_32_bits(req->len)); - atomic_inc(&req->usage); + refcount_inc(&req->usage); call->cb_break = fc->cb_break; afs_use_fs_server(call, fc->cbi); trace_afs_make_fs_call(call, &vnode->fid); @@ -578,6 +685,7 @@ int afs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_read *req) call->reply[0] = vnode; call->reply[1] = NULL; /* volsync */ call->reply[2] = req; + call->expected_version = vnode->status.data_version; /* marshall the parameters */ bp = call->request; @@ -588,7 +696,7 @@ int afs_fs_fetch_data(struct afs_fs_cursor *fc, struct afs_read *req) bp[4] = htonl(lower_32_bits(req->pos)); bp[5] = htonl(lower_32_bits(req->len)); - atomic_inc(&req->usage); + refcount_inc(&req->usage); call->cb_break = fc->cb_break; afs_use_fs_server(call, fc->cbi); trace_afs_make_fs_call(call, &vnode->fid); @@ -613,8 +721,10 @@ static int afs_deliver_fs_create_vnode(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; xdr_decode_AFSFid(&bp, call->reply[1]); - xdr_decode_AFSFetchStatus(&bp, call->reply[2], NULL, NULL); - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); + if (xdr_decode_AFSFetchStatus(call, &bp, call->reply[2], NULL, NULL, NULL) < 0 || + xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); xdr_decode_AFSCallBack_raw(&bp, call->reply[3]); /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ @@ -645,6 +755,7 @@ static const struct afs_call_type afs_RXFSMakeDir = { int afs_fs_create(struct afs_fs_cursor *fc, const char *name, umode_t mode, + u64 current_data_version, struct afs_fid *newfid, struct afs_file_status *newstatus, struct afs_callback *newcb) @@ -672,6 +783,7 @@ int afs_fs_create(struct afs_fs_cursor *fc, call->reply[1] = newfid; call->reply[2] = newstatus; call->reply[3] = newcb; + call->expected_version = current_data_version + 1; /* marshall the parameters */ bp = call->request; @@ -715,7 +827,9 @@ static int afs_deliver_fs_remove(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); + if (xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ _leave(" = 0 [done]"); @@ -742,7 +856,8 @@ static const struct afs_call_type afs_RXFSRemoveDir = { /* * remove a file or directory */ -int afs_fs_remove(struct afs_fs_cursor *fc, const char *name, bool isdir) +int afs_fs_remove(struct afs_fs_cursor *fc, const char *name, bool isdir, + u64 current_data_version) { struct afs_vnode *vnode = fc->vnode; struct afs_call *call; @@ -764,6 +879,7 @@ int afs_fs_remove(struct afs_fs_cursor *fc, const char *name, bool isdir) call->key = fc->key; call->reply[0] = vnode; + call->expected_version = current_data_version + 1; /* marshall the parameters */ bp = call->request; @@ -801,8 +917,10 @@ static int afs_deliver_fs_link(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); - xdr_decode_AFSFetchStatus(&bp, &dvnode->status, dvnode, NULL); + if (xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, NULL, NULL) < 0 || + xdr_decode_AFSFetchStatus(call, &bp, &dvnode->status, dvnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ _leave(" = 0 [done]"); @@ -823,7 +941,7 @@ static const struct afs_call_type afs_RXFSLink = { * make a hard link */ int afs_fs_link(struct afs_fs_cursor *fc, struct afs_vnode *vnode, - const char *name) + const char *name, u64 current_data_version) { struct afs_vnode *dvnode = fc->vnode; struct afs_call *call; @@ -844,6 +962,7 @@ int afs_fs_link(struct afs_fs_cursor *fc, struct afs_vnode *vnode, call->key = fc->key; call->reply[0] = dvnode; call->reply[1] = vnode; + call->expected_version = current_data_version + 1; /* marshall the parameters */ bp = call->request; @@ -885,8 +1004,10 @@ static int afs_deliver_fs_symlink(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; xdr_decode_AFSFid(&bp, call->reply[1]); - xdr_decode_AFSFetchStatus(&bp, call->reply[2], NULL, NULL); - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); + if (xdr_decode_AFSFetchStatus(call, &bp, call->reply[2], NULL, NULL, NULL) || + xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ _leave(" = 0 [done]"); @@ -909,6 +1030,7 @@ static const struct afs_call_type afs_RXFSSymlink = { int afs_fs_symlink(struct afs_fs_cursor *fc, const char *name, const char *contents, + u64 current_data_version, struct afs_fid *newfid, struct afs_file_status *newstatus) { @@ -937,6 +1059,7 @@ int afs_fs_symlink(struct afs_fs_cursor *fc, call->reply[0] = vnode; call->reply[1] = newfid; call->reply[2] = newstatus; + call->expected_version = current_data_version + 1; /* marshall the parameters */ bp = call->request; @@ -987,10 +1110,13 @@ static int afs_deliver_fs_rename(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &orig_dvnode->status, orig_dvnode, NULL); - if (new_dvnode != orig_dvnode) - xdr_decode_AFSFetchStatus(&bp, &new_dvnode->status, new_dvnode, - NULL); + if (xdr_decode_AFSFetchStatus(call, &bp, &orig_dvnode->status, orig_dvnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); + if (new_dvnode != orig_dvnode && + xdr_decode_AFSFetchStatus(call, &bp, &new_dvnode->status, new_dvnode, + &call->expected_version_2, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ _leave(" = 0 [done]"); @@ -1013,7 +1139,9 @@ static const struct afs_call_type afs_RXFSRename = { int afs_fs_rename(struct afs_fs_cursor *fc, const char *orig_name, struct afs_vnode *new_dvnode, - const char *new_name) + const char *new_name, + u64 current_orig_data_version, + u64 current_new_data_version) { struct afs_vnode *orig_dvnode = fc->vnode; struct afs_call *call; @@ -1041,6 +1169,8 @@ int afs_fs_rename(struct afs_fs_cursor *fc, call->key = fc->key; call->reply[0] = orig_dvnode; call->reply[1] = new_dvnode; + call->expected_version = current_orig_data_version + 1; + call->expected_version_2 = current_new_data_version + 1; /* marshall the parameters */ bp = call->request; @@ -1089,8 +1219,9 @@ static int afs_deliver_fs_store_data(struct afs_call *call) /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, - &call->store_version); + if (xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ afs_pages_written_back(vnode, call); @@ -1147,7 +1278,7 @@ static int afs_fs_store_data64(struct afs_fs_cursor *fc, call->first_offset = offset; call->last_to = to; call->send_pages = true; - call->store_version = vnode->status.data_version + 1; + call->expected_version = vnode->status.data_version + 1; /* marshall the parameters */ bp = call->request; @@ -1222,7 +1353,7 @@ int afs_fs_store_data(struct afs_fs_cursor *fc, struct address_space *mapping, call->first_offset = offset; call->last_to = to; call->send_pages = true; - call->store_version = vnode->status.data_version + 1; + call->expected_version = vnode->status.data_version + 1; /* marshall the parameters */ bp = call->request; @@ -1252,7 +1383,6 @@ int afs_fs_store_data(struct afs_fs_cursor *fc, struct address_space *mapping, */ static int afs_deliver_fs_store_status(struct afs_call *call) { - afs_dataversion_t *store_version; struct afs_vnode *vnode = call->reply[0]; const __be32 *bp; int ret; @@ -1264,12 +1394,10 @@ static int afs_deliver_fs_store_status(struct afs_call *call) return ret; /* unmarshall the reply once we've received all of it */ - store_version = NULL; - if (call->operation_ID == FSSTOREDATA) - store_version = &call->store_version; - bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, store_version); + if (xdr_decode_AFSFetchStatus(call, &bp, &vnode->status, vnode, + &call->expected_version, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); /* xdr_decode_AFSVolSync(&bp, call->reply[X]); */ _leave(" = 0 [done]"); @@ -1324,7 +1452,7 @@ static int afs_fs_setattr_size64(struct afs_fs_cursor *fc, struct iattr *attr) call->key = fc->key; call->reply[0] = vnode; - call->store_version = vnode->status.data_version + 1; + call->expected_version = vnode->status.data_version + 1; /* marshall the parameters */ bp = call->request; @@ -1373,7 +1501,7 @@ static int afs_fs_setattr_size(struct afs_fs_cursor *fc, struct iattr *attr) call->key = fc->key; call->reply[0] = vnode; - call->store_version = vnode->status.data_version + 1; + call->expected_version = vnode->status.data_version + 1; /* marshall the parameters */ bp = call->request; @@ -1418,6 +1546,7 @@ int afs_fs_setattr(struct afs_fs_cursor *fc, struct iattr *attr) call->key = fc->key; call->reply[0] = vnode; + call->expected_version = vnode->status.data_version; /* marshall the parameters */ bp = call->request; @@ -1471,7 +1600,7 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("volname length: %u", call->count); if (call->count >= AFSNAMEMAX) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); call->offset = 0; call->unmarshall++; @@ -1518,7 +1647,7 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("offline msg length: %u", call->count); if (call->count >= AFSNAMEMAX) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); call->offset = 0; call->unmarshall++; @@ -1565,7 +1694,7 @@ static int afs_deliver_fs_get_volume_status(struct afs_call *call) call->count = ntohl(call->tmp); _debug("motd length: %u", call->count); if (call->count >= AFSNAMEMAX) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); call->offset = 0; call->unmarshall++; @@ -1947,3 +2076,265 @@ int afs_fs_get_capabilities(struct afs_net *net, trace_afs_make_fs_call(call, NULL); return afs_make_call(ac, call, GFP_NOFS, false); } + +/* + * Deliver reply data to an FS.FetchStatus with no vnode. + */ +static int afs_deliver_fs_fetch_status(struct afs_call *call) +{ + struct afs_file_status *status = call->reply[1]; + struct afs_callback *callback = call->reply[2]; + struct afs_volsync *volsync = call->reply[3]; + struct afs_vnode *vnode = call->reply[0]; + const __be32 *bp; + int ret; + + ret = afs_transfer_reply(call); + if (ret < 0) + return ret; + + _enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode); + + /* unmarshall the reply once we've received all of it */ + bp = call->buffer; + xdr_decode_AFSFetchStatus(call, &bp, status, vnode, + &call->expected_version, NULL); + callback[call->count].version = ntohl(bp[0]); + callback[call->count].expiry = ntohl(bp[1]); + callback[call->count].type = ntohl(bp[2]); + if (vnode) + xdr_decode_AFSCallBack(call, vnode, &bp); + else + bp += 3; + if (volsync) + xdr_decode_AFSVolSync(&bp, volsync); + + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.FetchStatus operation type + */ +static const struct afs_call_type afs_RXFSFetchStatus = { + .name = "FS.FetchStatus", + .op = afs_FS_FetchStatus, + .deliver = afs_deliver_fs_fetch_status, + .destructor = afs_flat_call_destructor, +}; + +/* + * Fetch the status information for a fid without needing a vnode handle. + */ +int afs_fs_fetch_status(struct afs_fs_cursor *fc, + struct afs_net *net, + struct afs_fid *fid, + struct afs_file_status *status, + struct afs_callback *callback, + struct afs_volsync *volsync) +{ + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%x:%u},,", + key_serial(fc->key), fid->vid, fid->vnode); + + call = afs_alloc_flat_call(net, &afs_RXFSFetchStatus, 16, (21 + 3 + 6) * 4); + if (!call) { + fc->ac.error = -ENOMEM; + return -ENOMEM; + } + + call->key = fc->key; + call->reply[0] = NULL; /* vnode for fid[0] */ + call->reply[1] = status; + call->reply[2] = callback; + call->reply[3] = volsync; + call->expected_version = 1; /* vnode->status.data_version */ + + /* marshall the parameters */ + bp = call->request; + bp[0] = htonl(FSFETCHSTATUS); + bp[1] = htonl(fid->vid); + bp[2] = htonl(fid->vnode); + bp[3] = htonl(fid->unique); + + call->cb_break = fc->cb_break; + afs_use_fs_server(call, fc->cbi); + trace_afs_make_fs_call(call, fid); + return afs_make_call(&fc->ac, call, GFP_NOFS, false); +} + +/* + * Deliver reply data to an FS.InlineBulkStatus call + */ +static int afs_deliver_fs_inline_bulk_status(struct afs_call *call) +{ + struct afs_file_status *statuses; + struct afs_callback *callbacks; + struct afs_vnode *vnode = call->reply[0]; + const __be32 *bp; + u32 tmp; + int ret; + + _enter("{%u}", call->unmarshall); + + switch (call->unmarshall) { + case 0: + call->offset = 0; + call->unmarshall++; + + /* Extract the file status count and array in two steps */ + case 1: + _debug("extract status count"); + ret = afs_extract_data(call, &call->tmp, 4, true); + if (ret < 0) + return ret; + + tmp = ntohl(call->tmp); + _debug("status count: %u/%u", tmp, call->count2); + if (tmp != call->count2) + return afs_protocol_error(call, -EBADMSG); + + call->count = 0; + call->unmarshall++; + more_counts: + call->offset = 0; + + case 2: + _debug("extract status array %u", call->count); + ret = afs_extract_data(call, call->buffer, 21 * 4, true); + if (ret < 0) + return ret; + + bp = call->buffer; + statuses = call->reply[1]; + if (xdr_decode_AFSFetchStatus(call, &bp, &statuses[call->count], + call->count == 0 ? vnode : NULL, + NULL, NULL) < 0) + return afs_protocol_error(call, -EBADMSG); + + call->count++; + if (call->count < call->count2) + goto more_counts; + + call->count = 0; + call->unmarshall++; + call->offset = 0; + + /* Extract the callback count and array in two steps */ + case 3: + _debug("extract CB count"); + ret = afs_extract_data(call, &call->tmp, 4, true); + if (ret < 0) + return ret; + + tmp = ntohl(call->tmp); + _debug("CB count: %u", tmp); + if (tmp != call->count2) + return afs_protocol_error(call, -EBADMSG); + call->count = 0; + call->unmarshall++; + more_cbs: + call->offset = 0; + + case 4: + _debug("extract CB array"); + ret = afs_extract_data(call, call->buffer, 3 * 4, true); + if (ret < 0) + return ret; + + _debug("unmarshall CB array"); + bp = call->buffer; + callbacks = call->reply[2]; + callbacks[call->count].version = ntohl(bp[0]); + callbacks[call->count].expiry = ntohl(bp[1]); + callbacks[call->count].type = ntohl(bp[2]); + statuses = call->reply[1]; + if (call->count == 0 && vnode && statuses[0].abort_code == 0) + xdr_decode_AFSCallBack(call, vnode, &bp); + call->count++; + if (call->count < call->count2) + goto more_cbs; + + call->offset = 0; + call->unmarshall++; + + case 5: + ret = afs_extract_data(call, call->buffer, 6 * 4, false); + if (ret < 0) + return ret; + + bp = call->buffer; + if (call->reply[3]) + xdr_decode_AFSVolSync(&bp, call->reply[3]); + + call->offset = 0; + call->unmarshall++; + + case 6: + break; + } + + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.InlineBulkStatus operation type + */ +static const struct afs_call_type afs_RXFSInlineBulkStatus = { + .name = "FS.InlineBulkStatus", + .op = afs_FS_InlineBulkStatus, + .deliver = afs_deliver_fs_inline_bulk_status, + .destructor = afs_flat_call_destructor, +}; + +/* + * Fetch the status information for up to 50 files + */ +int afs_fs_inline_bulk_status(struct afs_fs_cursor *fc, + struct afs_net *net, + struct afs_fid *fids, + struct afs_file_status *statuses, + struct afs_callback *callbacks, + unsigned int nr_fids, + struct afs_volsync *volsync) +{ + struct afs_call *call; + __be32 *bp; + int i; + + _enter(",%x,{%x:%u},%u", + key_serial(fc->key), fids[0].vid, fids[1].vnode, nr_fids); + + call = afs_alloc_flat_call(net, &afs_RXFSInlineBulkStatus, + (2 + nr_fids * 3) * 4, + 21 * 4); + if (!call) { + fc->ac.error = -ENOMEM; + return -ENOMEM; + } + + call->key = fc->key; + call->reply[0] = NULL; /* vnode for fid[0] */ + call->reply[1] = statuses; + call->reply[2] = callbacks; + call->reply[3] = volsync; + call->count2 = nr_fids; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSINLINEBULKSTATUS); + *bp++ = htonl(nr_fids); + for (i = 0; i < nr_fids; i++) { + *bp++ = htonl(fids[i].vid); + *bp++ = htonl(fids[i].vnode); + *bp++ = htonl(fids[i].unique); + } + + call->cb_break = fc->cb_break; + afs_use_fs_server(call, fc->cbi); + trace_afs_make_fs_call(call, &fids[0]); + return afs_make_call(&fc->ac, call, GFP_NOFS, false); +} diff --git a/fs/afs/inode.c b/fs/afs/inode.c index 6b39d0255b72..479b7fdda124 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -30,12 +30,11 @@ static const struct inode_operations afs_symlink_inode_operations = { }; /* - * map the AFS file status to the inode member variables + * Initialise an inode from the vnode status. */ -static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key) +static int afs_inode_init_from_status(struct afs_vnode *vnode, struct key *key) { struct inode *inode = AFS_VNODE_TO_I(vnode); - bool changed; _debug("FS: ft=%d lk=%d sz=%llu ver=%Lu mod=%hu", vnode->status.type, @@ -46,16 +45,21 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key) read_seqlock_excl(&vnode->cb_lock); + afs_update_inode_from_status(vnode, &vnode->status, NULL, + AFS_VNODE_NOT_YET_SET); + switch (vnode->status.type) { case AFS_FTYPE_FILE: inode->i_mode = S_IFREG | vnode->status.mode; inode->i_op = &afs_file_inode_operations; inode->i_fop = &afs_file_operations; + inode->i_mapping->a_ops = &afs_fs_aops; break; case AFS_FTYPE_DIR: inode->i_mode = S_IFDIR | vnode->status.mode; inode->i_op = &afs_dir_inode_operations; inode->i_fop = &afs_dir_file_operations; + inode->i_mapping->a_ops = &afs_dir_aops; break; case AFS_FTYPE_SYMLINK: /* Symlinks with a mode of 0644 are actually mountpoints. */ @@ -67,45 +71,31 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key) inode->i_mode = S_IFDIR | 0555; inode->i_op = &afs_mntpt_inode_operations; inode->i_fop = &afs_mntpt_file_operations; + inode->i_mapping->a_ops = &afs_fs_aops; } else { inode->i_mode = S_IFLNK | vnode->status.mode; inode->i_op = &afs_symlink_inode_operations; + inode->i_mapping->a_ops = &afs_fs_aops; } inode_nohighmem(inode); break; default: printk("kAFS: AFS vnode with undefined type\n"); read_sequnlock_excl(&vnode->cb_lock); - return -EBADMSG; + return afs_protocol_error(NULL, -EBADMSG); } - changed = (vnode->status.size != inode->i_size); - - set_nlink(inode, vnode->status.nlink); - inode->i_uid = vnode->status.owner; - inode->i_gid = vnode->status.group; - inode->i_size = vnode->status.size; - inode->i_ctime.tv_sec = vnode->status.mtime_client; - inode->i_ctime.tv_nsec = 0; - inode->i_atime = inode->i_mtime = inode->i_ctime; inode->i_blocks = 0; - inode->i_generation = vnode->fid.unique; - inode_set_iversion_raw(inode, vnode->status.data_version); - inode->i_mapping->a_ops = &afs_fs_aops; + vnode->invalid_before = vnode->status.data_version; read_sequnlock_excl(&vnode->cb_lock); - -#ifdef CONFIG_AFS_FSCACHE - if (changed) - fscache_attr_changed(vnode->cache); -#endif return 0; } /* * Fetch file status from the volume. */ -int afs_fetch_status(struct afs_vnode *vnode, struct key *key) +int afs_fetch_status(struct afs_vnode *vnode, struct key *key, bool new_inode) { struct afs_fs_cursor fc; int ret; @@ -118,8 +108,8 @@ int afs_fetch_status(struct afs_vnode *vnode, struct key *key) ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, vnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; - afs_fs_fetch_file_status(&fc, NULL); + fc.cb_break = afs_calc_vnode_cb_break(vnode); + afs_fs_fetch_file_status(&fc, NULL, new_inode); } afs_check_for_remote_deletion(&fc, fc.vnode); @@ -243,6 +233,38 @@ struct inode *afs_iget_pseudo_dir(struct super_block *sb, bool root) } /* + * Get a cache cookie for an inode. + */ +static void afs_get_inode_cache(struct afs_vnode *vnode) +{ +#ifdef CONFIG_AFS_FSCACHE + struct { + u32 vnode_id; + u32 unique; + u32 vnode_id_ext[2]; /* Allow for a 96-bit key */ + } __packed key; + struct afs_vnode_cache_aux aux; + + if (vnode->status.type == AFS_FTYPE_DIR) { + vnode->cache = NULL; + return; + } + + key.vnode_id = vnode->fid.vnode; + key.unique = vnode->fid.unique; + key.vnode_id_ext[0] = 0; + key.vnode_id_ext[1] = 0; + aux.data_version = vnode->status.data_version; + + vnode->cache = fscache_acquire_cookie(vnode->volume->cache, + &afs_vnode_cache_index_def, + &key, sizeof(key), + &aux, sizeof(aux), + vnode, vnode->status.size, true); +#endif +} + +/* * inode retrieval */ struct inode *afs_iget(struct super_block *sb, struct key *key, @@ -280,7 +302,7 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, if (!status) { /* it's a remotely extant inode */ - ret = afs_fetch_status(vnode, key); + ret = afs_fetch_status(vnode, key, true); if (ret < 0) goto bad_inode; } else { @@ -304,19 +326,12 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, vnode->cb_expires_at += ktime_get_real_seconds(); } - /* set up caching before mapping the status, as map-status reads the - * first page of symlinks to see if they're really mountpoints */ - inode->i_size = vnode->status.size; -#ifdef CONFIG_AFS_FSCACHE - vnode->cache = fscache_acquire_cookie(vnode->volume->cache, - &afs_vnode_cache_index_def, - vnode, true); -#endif - - ret = afs_inode_map_status(vnode, key); + ret = afs_inode_init_from_status(vnode, key); if (ret < 0) goto bad_inode; + afs_get_inode_cache(vnode); + /* success */ clear_bit(AFS_VNODE_UNSET, &vnode->flags); inode->i_flags |= S_NOATIME; @@ -326,10 +341,6 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, /* failure */ bad_inode: -#ifdef CONFIG_AFS_FSCACHE - fscache_relinquish_cookie(vnode->cache, 0); - vnode->cache = NULL; -#endif iget_failed(inode); _leave(" = %d [bad]", ret); return ERR_PTR(ret); @@ -343,6 +354,10 @@ void afs_zap_data(struct afs_vnode *vnode) { _enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode); +#ifdef CONFIG_AFS_FSCACHE + fscache_invalidate(vnode->cache); +#endif + /* nuke all the non-dirty pages that aren't locked, mapped or being * written back in a regular file and completely discard the pages in a * directory or symlink */ @@ -378,12 +393,18 @@ int afs_validate(struct afs_vnode *vnode, struct key *key) read_seqlock_excl(&vnode->cb_lock); if (test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) { - if (vnode->cb_s_break != vnode->cb_interest->server->cb_s_break) { + if (vnode->cb_s_break != vnode->cb_interest->server->cb_s_break || + vnode->cb_v_break != vnode->volume->cb_v_break) { vnode->cb_s_break = vnode->cb_interest->server->cb_s_break; - } else if (!test_bit(AFS_VNODE_DIR_MODIFIED, &vnode->flags) && - !test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags) && + vnode->cb_v_break = vnode->volume->cb_v_break; + valid = false; + } else if (vnode->status.type == AFS_FTYPE_DIR && + test_bit(AFS_VNODE_DIR_VALID, &vnode->flags) && vnode->cb_expires_at - 10 > now) { - valid = true; + valid = true; + } else if (!test_bit(AFS_VNODE_ZAP_DATA, &vnode->flags) && + vnode->cb_expires_at - 10 > now) { + valid = true; } } else if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { valid = true; @@ -397,7 +418,7 @@ int afs_validate(struct afs_vnode *vnode, struct key *key) if (valid) goto valid; - mutex_lock(&vnode->validate_lock); + down_write(&vnode->validate_lock); /* if the promise has expired, we need to check the server again to get * a new promise - note that if the (parent) directory's metadata was @@ -405,7 +426,7 @@ int afs_validate(struct afs_vnode *vnode, struct key *key) * access */ if (!test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) { _debug("not promised"); - ret = afs_fetch_status(vnode, key); + ret = afs_fetch_status(vnode, key, false); if (ret < 0) { if (ret == -ENOENT) { set_bit(AFS_VNODE_DELETED, &vnode->flags); @@ -426,15 +447,13 @@ int afs_validate(struct afs_vnode *vnode, struct key *key) * different */ if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) afs_zap_data(vnode); - - clear_bit(AFS_VNODE_DIR_MODIFIED, &vnode->flags); - mutex_unlock(&vnode->validate_lock); + up_write(&vnode->validate_lock); valid: _leave(" = 0"); return 0; error_unlock: - mutex_unlock(&vnode->validate_lock); + up_write(&vnode->validate_lock); _leave(" = %d", ret); return ret; } @@ -507,11 +526,17 @@ void afs_evict_inode(struct inode *inode) } #ifdef CONFIG_AFS_FSCACHE - fscache_relinquish_cookie(vnode->cache, 0); - vnode->cache = NULL; + { + struct afs_vnode_cache_aux aux; + + aux.data_version = vnode->status.data_version; + fscache_relinquish_cookie(vnode->cache, &aux, + test_bit(AFS_VNODE_DELETED, &vnode->flags)); + vnode->cache = NULL; + } #endif - afs_put_permits(vnode->permit_cache); + afs_put_permits(rcu_access_pointer(vnode->permit_cache)); _leave(""); } @@ -552,7 +577,7 @@ int afs_setattr(struct dentry *dentry, struct iattr *attr) ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, vnode, key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(vnode); afs_fs_setattr(&fc, attr); } diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 72217170b155..e3f8a46663db 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -122,7 +122,8 @@ struct afs_call { u32 operation_ID; /* operation ID for an incoming call */ u32 count; /* count for use in unmarshalling */ __be32 tmp; /* place to extract temporary data */ - afs_dataversion_t store_version; /* updated version expected from store */ + afs_dataversion_t expected_version; /* Updated version expected from store */ + afs_dataversion_t expected_version_2; /* 2nd updated version expected from store */ }; struct afs_call_type { @@ -173,11 +174,14 @@ struct afs_read { loff_t len; /* How much we're asking for */ loff_t actual_len; /* How much we're actually getting */ loff_t remain; /* Amount remaining */ - atomic_t usage; + loff_t file_size; /* File size returned by server */ + afs_dataversion_t data_version; /* Version number returned by server */ + refcount_t usage; unsigned int index; /* Which page we're reading into */ unsigned int nr_pages; void (*page_done)(struct afs_call *, struct afs_read *); - struct page *pages[]; + struct page **pages; + struct page *array[]; }; /* @@ -199,6 +203,18 @@ static inline struct afs_super_info *AFS_FS_S(struct super_block *sb) extern struct file_system_type afs_fs_type; /* + * Set of substitutes for @sys. + */ +struct afs_sysnames { +#define AFS_NR_SYSNAME 16 + char *subs[AFS_NR_SYSNAME]; + refcount_t usage; + unsigned short nr; + short error; + char blank[1]; +}; + +/* * AFS network namespace record. */ struct afs_net { @@ -245,9 +261,25 @@ struct afs_net { struct mutex lock_manager_mutex; /* Misc */ - struct proc_dir_entry *proc_afs; /* /proc/net/afs directory */ + struct proc_dir_entry *proc_afs; /* /proc/net/afs directory */ + struct afs_sysnames *sysnames; + rwlock_t sysnames_lock; + + /* Statistics counters */ + atomic_t n_lookup; /* Number of lookups done */ + atomic_t n_reval; /* Number of dentries needing revalidation */ + atomic_t n_inval; /* Number of invalidations by the server */ + atomic_t n_relpg; /* Number of invalidations by releasepage */ + atomic_t n_read_dir; /* Number of directory pages read */ + atomic_t n_dir_cr; /* Number of directory entry creation edits */ + atomic_t n_dir_rm; /* Number of directory entry removal edits */ + atomic_t n_stores; /* Number of store ops */ + atomic_long_t n_store_bytes; /* Number of bytes stored */ + atomic_long_t n_fetch_bytes; /* Number of bytes fetched */ + atomic_t n_fetches; /* Number of data fetch ops */ }; +extern const char afs_init_sysname[]; extern struct afs_net __afs_net;// Dummy AFS network namespace; TODO: replace with real netns enum afs_cell_state { @@ -363,6 +395,8 @@ struct afs_server { #define AFS_SERVER_FL_UPDATING 4 #define AFS_SERVER_FL_PROBED 5 /* The fileserver has been probed */ #define AFS_SERVER_FL_PROBING 6 /* Fileserver is being probed */ +#define AFS_SERVER_FL_NO_IBULK 7 /* Fileserver doesn't support FS.InlineBulkStatus */ +#define AFS_SERVER_FL_MAY_HAVE_CB 8 /* May have callbacks on this fileserver */ atomic_t usage; u32 addr_version; /* Address list version */ @@ -400,6 +434,7 @@ struct afs_server_list { unsigned short index; /* Server currently in use */ unsigned short vnovol_mask; /* Servers to be skipped due to VNOVOL */ unsigned int seq; /* Set to ->servers_seq when installed */ + rwlock_t lock; struct afs_server_entry servers[]; }; @@ -426,6 +461,9 @@ struct afs_volume { rwlock_t servers_lock; /* Lock for ->servers */ unsigned int servers_seq; /* Incremented each time ->servers changes */ + unsigned cb_v_break; /* Break-everything counter. */ + rwlock_t cb_break_lock; + afs_voltype_t type; /* type of volume */ short error; char type_force; /* force volume type (suppress R/O -> R/W) */ @@ -455,23 +493,25 @@ struct afs_vnode { struct afs_volume *volume; /* volume on which vnode resides */ struct afs_fid fid; /* the file identifier for this inode */ struct afs_file_status status; /* AFS status info for this file */ + afs_dataversion_t invalid_before; /* Child dentries are invalid before this */ #ifdef CONFIG_AFS_FSCACHE struct fscache_cookie *cache; /* caching cookie */ #endif - struct afs_permits *permit_cache; /* cache of permits so far obtained */ + struct afs_permits __rcu *permit_cache; /* cache of permits so far obtained */ struct mutex io_lock; /* Lock for serialising I/O on this mutex */ - struct mutex validate_lock; /* lock for validating this vnode */ + struct rw_semaphore validate_lock; /* lock for validating this vnode */ spinlock_t wb_lock; /* lock for wb_keys */ spinlock_t lock; /* waitqueue/flags lock */ unsigned long flags; #define AFS_VNODE_CB_PROMISED 0 /* Set if vnode has a callback promise */ #define AFS_VNODE_UNSET 1 /* set if vnode attributes not yet set */ -#define AFS_VNODE_DIR_MODIFIED 2 /* set if dir vnode's data modified */ +#define AFS_VNODE_DIR_VALID 2 /* Set if dir contents are valid */ #define AFS_VNODE_ZAP_DATA 3 /* set if vnode's data should be invalidated */ #define AFS_VNODE_DELETED 4 /* set if vnode deleted on server */ #define AFS_VNODE_MOUNTPOINT 5 /* set if vnode is a mountpoint symlink */ #define AFS_VNODE_AUTOCELL 6 /* set if Vnode is an auto mount point */ #define AFS_VNODE_PSEUDODIR 7 /* set if Vnode is a pseudo directory */ +#define AFS_VNODE_NEW_CONTENT 8 /* Set if file has new content (create/trunc-0) */ struct list_head wb_keys; /* List of keys available for writeback */ struct list_head pending_locks; /* locks waiting to be granted */ @@ -484,6 +524,7 @@ struct afs_vnode { /* outstanding callback notification on this file */ struct afs_cb_interest *cb_interest; /* Server on which this resides */ unsigned int cb_s_break; /* Mass break counter on ->server */ + unsigned int cb_v_break; /* Mass break counter on ->volume */ unsigned int cb_break; /* Break counter on vnode */ seqlock_t cb_lock; /* Lock for ->cb_interest, ->status, ->cb_*break */ @@ -559,6 +600,13 @@ struct afs_fs_cursor { #define AFS_FS_CURSOR_NO_VSLEEP 0x0020 /* Set to prevent sleep on VBUSY, VOFFLINE, ... */ }; +/* + * Cache auxiliary data. + */ +struct afs_vnode_cache_aux { + u64 data_version; +} __packed; + #include <trace/events/afs.h> /*****************************************************************************/ @@ -604,18 +652,31 @@ extern struct fscache_cookie_def afs_vnode_cache_index_def; */ extern void afs_init_callback_state(struct afs_server *); extern void afs_break_callback(struct afs_vnode *); -extern void afs_break_callbacks(struct afs_server *, size_t,struct afs_callback[]); +extern void afs_break_callbacks(struct afs_server *, size_t, struct afs_callback_break*); -extern int afs_register_server_cb_interest(struct afs_vnode *, struct afs_server_entry *); +extern int afs_register_server_cb_interest(struct afs_vnode *, + struct afs_server_list *, unsigned int); extern void afs_put_cb_interest(struct afs_net *, struct afs_cb_interest *); extern void afs_clear_callback_interests(struct afs_net *, struct afs_server_list *); static inline struct afs_cb_interest *afs_get_cb_interest(struct afs_cb_interest *cbi) { - refcount_inc(&cbi->usage); + if (cbi) + refcount_inc(&cbi->usage); return cbi; } +static inline unsigned int afs_calc_vnode_cb_break(struct afs_vnode *vnode) +{ + return vnode->cb_break + vnode->cb_s_break + vnode->cb_v_break; +} + +static inline unsigned int afs_cb_break_sum(struct afs_vnode *vnode, + struct afs_cb_interest *cbi) +{ + return vnode->cb_break + cbi->server->cb_s_break + vnode->volume->cb_v_break; +} + /* * cell.c */ @@ -639,11 +700,26 @@ extern bool afs_cm_incoming_call(struct afs_call *); */ extern const struct file_operations afs_dir_file_operations; extern const struct inode_operations afs_dir_inode_operations; +extern const struct address_space_operations afs_dir_aops; +extern const struct dentry_operations afs_fs_dentry_operations; + +extern void afs_d_release(struct dentry *); + +/* + * dir_edit.c + */ +extern void afs_edit_dir_add(struct afs_vnode *, struct qstr *, struct afs_fid *, + enum afs_edit_dir_reason); +extern void afs_edit_dir_remove(struct afs_vnode *, struct qstr *, enum afs_edit_dir_reason); + +/* + * dynroot.c + */ extern const struct file_operations afs_dynroot_file_operations; extern const struct inode_operations afs_dynroot_inode_operations; -extern const struct dentry_operations afs_fs_dentry_operations; +extern const struct dentry_operations afs_dynroot_dentry_operations; -extern bool afs_dir_check_page(struct inode *, struct page *); +extern struct inode *afs_try_auto_mntpt(struct dentry *, struct inode *); /* * file.c @@ -673,17 +749,23 @@ extern int afs_flock(struct file *, int, struct file_lock *); /* * fsclient.c */ -extern int afs_fs_fetch_file_status(struct afs_fs_cursor *, struct afs_volsync *); +#define AFS_VNODE_NOT_YET_SET 0x01 +#define AFS_VNODE_META_CHANGED 0x02 +#define AFS_VNODE_DATA_CHANGED 0x04 +extern void afs_update_inode_from_status(struct afs_vnode *, struct afs_file_status *, + const afs_dataversion_t *, u8); + +extern int afs_fs_fetch_file_status(struct afs_fs_cursor *, struct afs_volsync *, bool); extern int afs_fs_give_up_callbacks(struct afs_net *, struct afs_server *); extern int afs_fs_fetch_data(struct afs_fs_cursor *, struct afs_read *); -extern int afs_fs_create(struct afs_fs_cursor *, const char *, umode_t, +extern int afs_fs_create(struct afs_fs_cursor *, const char *, umode_t, u64, struct afs_fid *, struct afs_file_status *, struct afs_callback *); -extern int afs_fs_remove(struct afs_fs_cursor *, const char *, bool); -extern int afs_fs_link(struct afs_fs_cursor *, struct afs_vnode *, const char *); -extern int afs_fs_symlink(struct afs_fs_cursor *, const char *, const char *, +extern int afs_fs_remove(struct afs_fs_cursor *, const char *, bool, u64); +extern int afs_fs_link(struct afs_fs_cursor *, struct afs_vnode *, const char *, u64); +extern int afs_fs_symlink(struct afs_fs_cursor *, const char *, const char *, u64, struct afs_fid *, struct afs_file_status *); extern int afs_fs_rename(struct afs_fs_cursor *, const char *, - struct afs_vnode *, const char *); + struct afs_vnode *, const char *, u64, u64); extern int afs_fs_store_data(struct afs_fs_cursor *, struct address_space *, pgoff_t, pgoff_t, unsigned, unsigned); extern int afs_fs_setattr(struct afs_fs_cursor *, struct iattr *); @@ -695,11 +777,18 @@ extern int afs_fs_give_up_all_callbacks(struct afs_net *, struct afs_server *, struct afs_addr_cursor *, struct key *); extern int afs_fs_get_capabilities(struct afs_net *, struct afs_server *, struct afs_addr_cursor *, struct key *); +extern int afs_fs_inline_bulk_status(struct afs_fs_cursor *, struct afs_net *, + struct afs_fid *, struct afs_file_status *, + struct afs_callback *, unsigned int, + struct afs_volsync *); +extern int afs_fs_fetch_status(struct afs_fs_cursor *, struct afs_net *, + struct afs_fid *, struct afs_file_status *, + struct afs_callback *, struct afs_volsync *); /* * inode.c */ -extern int afs_fetch_status(struct afs_vnode *, struct key *); +extern int afs_fetch_status(struct afs_vnode *, struct key *, bool); extern int afs_iget5_test(struct inode *, void *); extern struct inode *afs_iget_pseudo_dir(struct super_block *, bool); extern struct inode *afs_iget(struct super_block *, struct key *, @@ -747,6 +836,13 @@ static inline void afs_put_net(struct afs_net *net) { } +static inline void __afs_stat(atomic_t *s) +{ + atomic_inc(s); +} + +#define afs_stat_v(vnode, n) __afs_stat(&afs_v2net(vnode)->n) + /* * misc.c */ @@ -774,6 +870,7 @@ extern int __net_init afs_proc_init(struct afs_net *); extern void __net_exit afs_proc_cleanup(struct afs_net *); extern int afs_proc_cell_setup(struct afs_net *, struct afs_cell *); extern void afs_proc_cell_remove(struct afs_net *, struct afs_cell *); +extern void afs_put_sysnames(struct afs_sysnames *); /* * rotate.c @@ -802,6 +899,7 @@ extern void afs_flat_call_destructor(struct afs_call *); extern void afs_send_empty_reply(struct afs_call *); extern void afs_send_simple_reply(struct afs_call *, const void *, size_t); extern int afs_extract_data(struct afs_call *, void *, size_t, bool); +extern int afs_protocol_error(struct afs_call *, int); static inline int afs_transfer_reply(struct afs_call *call) { @@ -948,7 +1046,6 @@ extern int afs_writepage(struct page *, struct writeback_control *); extern int afs_writepages(struct address_space *, struct writeback_control *); extern void afs_pages_written_back(struct afs_vnode *, struct afs_call *); extern ssize_t afs_file_write(struct kiocb *, struct iov_iter *); -extern int afs_flush(struct file *, fl_owner_t); extern int afs_fsync(struct file *, loff_t, loff_t, int); extern int afs_page_mkwrite(struct vm_fault *); extern void afs_prune_wb_keys(struct afs_vnode *); diff --git a/fs/afs/main.c b/fs/afs/main.c index 15a02a05ff40..d7560168b3bf 100644 --- a/fs/afs/main.c +++ b/fs/afs/main.c @@ -34,11 +34,42 @@ MODULE_PARM_DESC(rootcell, "root AFS cell name and VL server IP addr list"); struct workqueue_struct *afs_wq; struct afs_net __afs_net; +#if defined(CONFIG_ALPHA) +const char afs_init_sysname[] = "alpha_linux26"; +#elif defined(CONFIG_X86_64) +const char afs_init_sysname[] = "amd64_linux26"; +#elif defined(CONFIG_ARM) +const char afs_init_sysname[] = "arm_linux26"; +#elif defined(CONFIG_ARM64) +const char afs_init_sysname[] = "aarch64_linux26"; +#elif defined(CONFIG_X86_32) +const char afs_init_sysname[] = "i386_linux26"; +#elif defined(CONFIG_IA64) +const char afs_init_sysname[] = "ia64_linux26"; +#elif defined(CONFIG_PPC64) +const char afs_init_sysname[] = "ppc64_linux26"; +#elif defined(CONFIG_PPC32) +const char afs_init_sysname[] = "ppc_linux26"; +#elif defined(CONFIG_S390) +#ifdef CONFIG_64BIT +const char afs_init_sysname[] = "s390x_linux26"; +#else +const char afs_init_sysname[] = "s390_linux26"; +#endif +#elif defined(CONFIG_SPARC64) +const char afs_init_sysname[] = "sparc64_linux26"; +#elif defined(CONFIG_SPARC32) +const char afs_init_sysname[] = "sparc_linux26"; +#else +const char afs_init_sysname[] = "unknown_linux26"; +#endif + /* * Initialise an AFS network namespace record. */ static int __net_init afs_net_init(struct afs_net *net) { + struct afs_sysnames *sysnames; int ret; net->live = true; @@ -67,6 +98,16 @@ static int __net_init afs_net_init(struct afs_net *net) INIT_WORK(&net->fs_manager, afs_manage_servers); timer_setup(&net->fs_timer, afs_servers_timer, 0); + ret = -ENOMEM; + sysnames = kzalloc(sizeof(*sysnames), GFP_KERNEL); + if (!sysnames) + goto error_sysnames; + sysnames->subs[0] = (char *)&afs_init_sysname; + sysnames->nr = 1; + refcount_set(&sysnames->usage, 1); + net->sysnames = sysnames; + rwlock_init(&net->sysnames_lock); + /* Register the /proc stuff */ ret = afs_proc_init(net); if (ret < 0) @@ -92,6 +133,8 @@ error_cell_init: net->live = false; afs_proc_cleanup(net); error_proc: + afs_put_sysnames(net->sysnames); +error_sysnames: net->live = false; return ret; } @@ -106,6 +149,7 @@ static void __net_exit afs_net_exit(struct afs_net *net) afs_purge_servers(net); afs_close_socket(net); afs_proc_cleanup(net); + afs_put_sysnames(net->sysnames); } /* diff --git a/fs/afs/proc.c b/fs/afs/proc.c index 4508dd54f789..3aad32762989 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -62,7 +62,6 @@ static const struct file_operations afs_proc_rootcell_fops = { .llseek = no_llseek, }; -static int afs_proc_cell_volumes_open(struct inode *inode, struct file *file); static void *afs_proc_cell_volumes_start(struct seq_file *p, loff_t *pos); static void *afs_proc_cell_volumes_next(struct seq_file *p, void *v, loff_t *pos); @@ -76,15 +75,6 @@ static const struct seq_operations afs_proc_cell_volumes_ops = { .show = afs_proc_cell_volumes_show, }; -static const struct file_operations afs_proc_cell_volumes_fops = { - .open = afs_proc_cell_volumes_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -static int afs_proc_cell_vlservers_open(struct inode *inode, - struct file *file); static void *afs_proc_cell_vlservers_start(struct seq_file *p, loff_t *pos); static void *afs_proc_cell_vlservers_next(struct seq_file *p, void *v, loff_t *pos); @@ -98,14 +88,6 @@ static const struct seq_operations afs_proc_cell_vlservers_ops = { .show = afs_proc_cell_vlservers_show, }; -static const struct file_operations afs_proc_cell_vlservers_fops = { - .open = afs_proc_cell_vlservers_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -static int afs_proc_servers_open(struct inode *inode, struct file *file); static void *afs_proc_servers_start(struct seq_file *p, loff_t *pos); static void *afs_proc_servers_next(struct seq_file *p, void *v, loff_t *pos); @@ -119,13 +101,34 @@ static const struct seq_operations afs_proc_servers_ops = { .show = afs_proc_servers_show, }; -static const struct file_operations afs_proc_servers_fops = { - .open = afs_proc_servers_open, +static int afs_proc_sysname_open(struct inode *inode, struct file *file); +static int afs_proc_sysname_release(struct inode *inode, struct file *file); +static void *afs_proc_sysname_start(struct seq_file *p, loff_t *pos); +static void *afs_proc_sysname_next(struct seq_file *p, void *v, + loff_t *pos); +static void afs_proc_sysname_stop(struct seq_file *p, void *v); +static int afs_proc_sysname_show(struct seq_file *m, void *v); +static ssize_t afs_proc_sysname_write(struct file *file, + const char __user *buf, + size_t size, loff_t *_pos); + +static const struct seq_operations afs_proc_sysname_ops = { + .start = afs_proc_sysname_start, + .next = afs_proc_sysname_next, + .stop = afs_proc_sysname_stop, + .show = afs_proc_sysname_show, +}; + +static const struct file_operations afs_proc_sysname_fops = { + .open = afs_proc_sysname_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = afs_proc_sysname_release, + .write = afs_proc_sysname_write, }; +static int afs_proc_stats_show(struct seq_file *m, void *v); + /* * initialise the /proc/fs/afs/ directory */ @@ -139,7 +142,9 @@ int afs_proc_init(struct afs_net *net) if (!proc_create("cells", 0644, net->proc_afs, &afs_proc_cells_fops) || !proc_create("rootcell", 0644, net->proc_afs, &afs_proc_rootcell_fops) || - !proc_create("servers", 0644, net->proc_afs, &afs_proc_servers_fops)) + !proc_create_seq("servers", 0644, net->proc_afs, &afs_proc_servers_ops) || + !proc_create_single("stats", 0644, net->proc_afs, afs_proc_stats_show) || + !proc_create("sysname", 0644, net->proc_afs, &afs_proc_sysname_fops)) goto error_tree; _leave(" = 0"); @@ -166,16 +171,7 @@ void afs_proc_cleanup(struct afs_net *net) */ static int afs_proc_cells_open(struct inode *inode, struct file *file) { - struct seq_file *m; - int ret; - - ret = seq_open(file, &afs_proc_cells_ops); - if (ret < 0) - return ret; - - m = file->private_data; - m->private = PDE_DATA(inode); - return 0; + return seq_open(file, &afs_proc_cells_ops); } /* @@ -183,6 +179,7 @@ static int afs_proc_cells_open(struct inode *inode, struct file *file) * first item */ static void *afs_proc_cells_start(struct seq_file *m, loff_t *_pos) + __acquires(rcu) { struct afs_net *net = afs_seq2net(m); @@ -204,6 +201,7 @@ static void *afs_proc_cells_next(struct seq_file *m, void *v, loff_t *pos) * clean up after reading from the cells list */ static void afs_proc_cells_stop(struct seq_file *m, void *v) + __releases(rcu) { rcu_read_unlock(); } @@ -282,7 +280,8 @@ static ssize_t afs_proc_cells_write(struct file *file, const char __user *buf, goto done; } - set_bit(AFS_CELL_FL_NO_GC, &cell->flags); + if (test_and_set_bit(AFS_CELL_FL_NO_GC, &cell->flags)) + afs_put_cell(net, cell); printk("kAFS: Added new cell '%s'\n", name); } else { goto inval; @@ -304,7 +303,40 @@ inval: static ssize_t afs_proc_rootcell_read(struct file *file, char __user *buf, size_t size, loff_t *_pos) { - return 0; + struct afs_cell *cell; + struct afs_net *net = afs_proc2net(file); + unsigned int seq = 0; + char name[AFS_MAXCELLNAME + 1]; + int len; + + if (*_pos > 0) + return 0; + if (!net->ws_cell) + return 0; + + rcu_read_lock(); + do { + read_seqbegin_or_lock(&net->cells_lock, &seq); + len = 0; + cell = rcu_dereference_raw(net->ws_cell); + if (cell) { + len = cell->name_len; + memcpy(name, cell->name, len); + } + } while (need_seqretry(&net->cells_lock, seq)); + done_seqretry(&net->cells_lock, seq); + rcu_read_unlock(); + + if (!len) + return 0; + + name[len++] = '\n'; + if (len > size) + len = size; + if (copy_to_user(buf, name, len) != 0) + return -EFAULT; + *_pos = 1; + return len; } /* @@ -327,6 +359,12 @@ static ssize_t afs_proc_rootcell_write(struct file *file, if (IS_ERR(kbuf)) return PTR_ERR(kbuf); + ret = -EINVAL; + if (kbuf[0] == '.') + goto out; + if (memchr(kbuf, '/', size)) + goto out; + /* trim to first NL */ s = memchr(kbuf, '\n', size); if (s) @@ -339,6 +377,7 @@ static ssize_t afs_proc_rootcell_write(struct file *file, if (ret >= 0) ret = size; /* consume everything, always */ +out: kfree(kbuf); _leave(" = %d", ret); return ret; @@ -357,10 +396,11 @@ int afs_proc_cell_setup(struct afs_net *net, struct afs_cell *cell) if (!dir) goto error_dir; - if (!proc_create_data("vlservers", 0, dir, - &afs_proc_cell_vlservers_fops, cell) || - !proc_create_data("volumes", 0, dir, - &afs_proc_cell_volumes_fops, cell)) + if (!proc_create_seq_data("vlservers", 0, dir, + &afs_proc_cell_vlservers_ops, cell)) + goto error_tree; + if (!proc_create_seq_data("volumes", 0, dir, &afs_proc_cell_volumes_ops, + cell)) goto error_tree; _leave(" = 0"); @@ -386,35 +426,13 @@ void afs_proc_cell_remove(struct afs_net *net, struct afs_cell *cell) } /* - * open "/proc/fs/afs/<cell>/volumes" which provides a summary of extant cells - */ -static int afs_proc_cell_volumes_open(struct inode *inode, struct file *file) -{ - struct afs_cell *cell; - struct seq_file *m; - int ret; - - cell = PDE_DATA(inode); - if (!cell) - return -ENOENT; - - ret = seq_open(file, &afs_proc_cell_volumes_ops); - if (ret < 0) - return ret; - - m = file->private_data; - m->private = cell; - - return 0; -} - -/* * set up the iterator to start reading from the cells list and return the * first item */ static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos) + __acquires(cell->proc_lock) { - struct afs_cell *cell = m->private; + struct afs_cell *cell = PDE_DATA(file_inode(m->file)); _enter("cell=%p pos=%Ld", cell, *_pos); @@ -428,7 +446,7 @@ static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos) static void *afs_proc_cell_volumes_next(struct seq_file *p, void *v, loff_t *_pos) { - struct afs_cell *cell = p->private; + struct afs_cell *cell = PDE_DATA(file_inode(p->file)); _enter("cell=%p pos=%Ld", cell, *_pos); return seq_list_next(v, &cell->proc_volumes, _pos); @@ -438,8 +456,9 @@ static void *afs_proc_cell_volumes_next(struct seq_file *p, void *v, * clean up after reading from the cells list */ static void afs_proc_cell_volumes_stop(struct seq_file *p, void *v) + __releases(cell->proc_lock) { - struct afs_cell *cell = p->private; + struct afs_cell *cell = PDE_DATA(file_inode(p->file)); read_unlock(&cell->proc_lock); } @@ -455,7 +474,7 @@ static const char afs_vol_types[3][3] = { */ static int afs_proc_cell_volumes_show(struct seq_file *m, void *v) { - struct afs_cell *cell = m->private; + struct afs_cell *cell = PDE_DATA(file_inode(m->file)); struct afs_volume *vol = list_entry(v, struct afs_volume, proc_link); /* Display header on line 1 */ @@ -472,37 +491,14 @@ static int afs_proc_cell_volumes_show(struct seq_file *m, void *v) } /* - * open "/proc/fs/afs/<cell>/vlservers" which provides a list of volume - * location server - */ -static int afs_proc_cell_vlservers_open(struct inode *inode, struct file *file) -{ - struct afs_cell *cell; - struct seq_file *m; - int ret; - - cell = PDE_DATA(inode); - if (!cell) - return -ENOENT; - - ret = seq_open(file, &afs_proc_cell_vlservers_ops); - if (ret<0) - return ret; - - m = file->private_data; - m->private = cell; - - return 0; -} - -/* * set up the iterator to start reading from the cells list and return the * first item */ static void *afs_proc_cell_vlservers_start(struct seq_file *m, loff_t *_pos) + __acquires(rcu) { struct afs_addr_list *alist; - struct afs_cell *cell = m->private; + struct afs_cell *cell = PDE_DATA(file_inode(m->file)); loff_t pos = *_pos; rcu_read_lock(); @@ -527,7 +523,7 @@ static void *afs_proc_cell_vlservers_next(struct seq_file *p, void *v, loff_t *_pos) { struct afs_addr_list *alist; - struct afs_cell *cell = p->private; + struct afs_cell *cell = PDE_DATA(file_inode(p->file)); loff_t pos; alist = rcu_dereference(cell->vl_addrs); @@ -544,6 +540,7 @@ static void *afs_proc_cell_vlservers_next(struct seq_file *p, void *v, * clean up after reading from the cells list */ static void afs_proc_cell_vlservers_stop(struct seq_file *p, void *v) + __releases(rcu) { rcu_read_unlock(); } @@ -567,19 +564,11 @@ static int afs_proc_cell_vlservers_show(struct seq_file *m, void *v) } /* - * open "/proc/fs/afs/servers" which provides a summary of active - * servers - */ -static int afs_proc_servers_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &afs_proc_servers_ops); -} - -/* * Set up the iterator to start reading from the server list and return the * first item. */ static void *afs_proc_servers_start(struct seq_file *m, loff_t *_pos) + __acquires(rcu) { struct afs_net *net = afs_seq2net(m); @@ -601,6 +590,7 @@ static void *afs_proc_servers_next(struct seq_file *m, void *v, loff_t *_pos) * clean up after reading from the cells list */ static void afs_proc_servers_stop(struct seq_file *p, void *v) + __releases(rcu) { rcu_read_unlock(); } @@ -626,3 +616,229 @@ static int afs_proc_servers_show(struct seq_file *m, void *v) &alist->addrs[alist->index].transport); return 0; } + +void afs_put_sysnames(struct afs_sysnames *sysnames) +{ + int i; + + if (sysnames && refcount_dec_and_test(&sysnames->usage)) { + for (i = 0; i < sysnames->nr; i++) + if (sysnames->subs[i] != afs_init_sysname && + sysnames->subs[i] != sysnames->blank) + kfree(sysnames->subs[i]); + } +} + +/* + * Handle opening of /proc/fs/afs/sysname. If it is opened for writing, we + * assume the caller wants to change the substitution list and we allocate a + * buffer to hold the list. + */ +static int afs_proc_sysname_open(struct inode *inode, struct file *file) +{ + struct afs_sysnames *sysnames; + struct seq_file *m; + int ret; + + ret = seq_open(file, &afs_proc_sysname_ops); + if (ret < 0) + return ret; + + if (file->f_mode & FMODE_WRITE) { + sysnames = kzalloc(sizeof(*sysnames), GFP_KERNEL); + if (!sysnames) { + seq_release(inode, file); + return -ENOMEM; + } + + refcount_set(&sysnames->usage, 1); + m = file->private_data; + m->private = sysnames; + } + + return 0; +} + +/* + * Handle writes to /proc/fs/afs/sysname to set the @sys substitution. + */ +static ssize_t afs_proc_sysname_write(struct file *file, + const char __user *buf, + size_t size, loff_t *_pos) +{ + struct afs_sysnames *sysnames; + struct seq_file *m = file->private_data; + char *kbuf = NULL, *s, *p, *sub; + int ret, len; + + sysnames = m->private; + if (!sysnames) + return -EINVAL; + if (sysnames->error) + return sysnames->error; + + if (size >= PAGE_SIZE - 1) { + sysnames->error = -EINVAL; + return -EINVAL; + } + if (size == 0) + return 0; + + kbuf = memdup_user_nul(buf, size); + if (IS_ERR(kbuf)) + return PTR_ERR(kbuf); + + inode_lock(file_inode(file)); + + p = kbuf; + while ((s = strsep(&p, " \t\n"))) { + len = strlen(s); + if (len == 0) + continue; + ret = -ENAMETOOLONG; + if (len >= AFSNAMEMAX) + goto error; + + if (len >= 4 && + s[len - 4] == '@' && + s[len - 3] == 's' && + s[len - 2] == 'y' && + s[len - 1] == 's') + /* Protect against recursion */ + goto invalid; + + if (s[0] == '.' && + (len < 2 || (len == 2 && s[1] == '.'))) + goto invalid; + + if (memchr(s, '/', len)) + goto invalid; + + ret = -EFBIG; + if (sysnames->nr >= AFS_NR_SYSNAME) + goto out; + + if (strcmp(s, afs_init_sysname) == 0) { + sub = (char *)afs_init_sysname; + } else { + ret = -ENOMEM; + sub = kmemdup(s, len + 1, GFP_KERNEL); + if (!sub) + goto out; + } + + sysnames->subs[sysnames->nr] = sub; + sysnames->nr++; + } + + ret = size; /* consume everything, always */ +out: + inode_unlock(file_inode(file)); + kfree(kbuf); + return ret; + +invalid: + ret = -EINVAL; +error: + sysnames->error = ret; + goto out; +} + +static int afs_proc_sysname_release(struct inode *inode, struct file *file) +{ + struct afs_sysnames *sysnames, *kill = NULL; + struct seq_file *m = file->private_data; + struct afs_net *net = afs_seq2net(m); + + sysnames = m->private; + if (sysnames) { + if (!sysnames->error) { + kill = sysnames; + if (sysnames->nr == 0) { + sysnames->subs[0] = sysnames->blank; + sysnames->nr++; + } + write_lock(&net->sysnames_lock); + kill = net->sysnames; + net->sysnames = sysnames; + write_unlock(&net->sysnames_lock); + } + afs_put_sysnames(kill); + } + + return seq_release(inode, file); +} + +static void *afs_proc_sysname_start(struct seq_file *m, loff_t *pos) + __acquires(&net->sysnames_lock) +{ + struct afs_net *net = afs_seq2net(m); + struct afs_sysnames *names = net->sysnames; + + read_lock(&net->sysnames_lock); + + if (*pos >= names->nr) + return NULL; + return (void *)(unsigned long)(*pos + 1); +} + +static void *afs_proc_sysname_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct afs_net *net = afs_seq2net(m); + struct afs_sysnames *names = net->sysnames; + + *pos += 1; + if (*pos >= names->nr) + return NULL; + return (void *)(unsigned long)(*pos + 1); +} + +static void afs_proc_sysname_stop(struct seq_file *m, void *v) + __releases(&net->sysnames_lock) +{ + struct afs_net *net = afs_seq2net(m); + + read_unlock(&net->sysnames_lock); +} + +static int afs_proc_sysname_show(struct seq_file *m, void *v) +{ + struct afs_net *net = afs_seq2net(m); + struct afs_sysnames *sysnames = net->sysnames; + unsigned int i = (unsigned long)v - 1; + + if (i < sysnames->nr) + seq_printf(m, "%s\n", sysnames->subs[i]); + return 0; +} + +/* + * Display general per-net namespace statistics + */ +static int afs_proc_stats_show(struct seq_file *m, void *v) +{ + struct afs_net *net = afs_seq2net(m); + + seq_puts(m, "kAFS statistics\n"); + + seq_printf(m, "dir-mgmt: look=%u reval=%u inval=%u relpg=%u\n", + atomic_read(&net->n_lookup), + atomic_read(&net->n_reval), + atomic_read(&net->n_inval), + atomic_read(&net->n_relpg)); + + seq_printf(m, "dir-data: rdpg=%u\n", + atomic_read(&net->n_read_dir)); + + seq_printf(m, "dir-edit: cr=%u rm=%u\n", + atomic_read(&net->n_dir_cr), + atomic_read(&net->n_dir_rm)); + + seq_printf(m, "file-rd : n=%u nb=%lu\n", + atomic_read(&net->n_fetches), + atomic_long_read(&net->n_fetch_bytes)); + seq_printf(m, "file-wr : n=%u nb=%lu\n", + atomic_read(&net->n_stores), + atomic_long_read(&net->n_store_bytes)); + return 0; +} diff --git a/fs/afs/rotate.c b/fs/afs/rotate.c index ad1328d85526..e065bc0768e6 100644 --- a/fs/afs/rotate.c +++ b/fs/afs/rotate.c @@ -21,7 +21,7 @@ /* * Initialise a filesystem server cursor for iterating over FS servers. */ -void afs_init_fs_cursor(struct afs_fs_cursor *fc, struct afs_vnode *vnode) +static void afs_init_fs_cursor(struct afs_fs_cursor *fc, struct afs_vnode *vnode) { memset(fc, 0, sizeof(*fc)); } @@ -179,7 +179,7 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc) */ if (fc->flags & AFS_FS_CURSOR_VNOVOL) { fc->ac.error = -EREMOTEIO; - goto failed; + goto next_server; } write_lock(&vnode->volume->servers_lock); @@ -201,7 +201,7 @@ bool afs_select_fileserver(struct afs_fs_cursor *fc) */ if (vnode->volume->servers == fc->server_list) { fc->ac.error = -EREMOTEIO; - goto failed; + goto next_server; } /* Try again */ @@ -350,8 +350,8 @@ use_server: * break request before we've finished decoding the reply and * installing the vnode. */ - fc->ac.error = afs_register_server_cb_interest( - vnode, &fc->server_list->servers[fc->index]); + fc->ac.error = afs_register_server_cb_interest(vnode, fc->server_list, + fc->index); if (fc->ac.error < 0) goto failed; @@ -369,8 +369,16 @@ use_server: if (!test_bit(AFS_SERVER_FL_PROBED, &server->flags)) { fc->ac.alist = afs_get_addrlist(alist); - if (!afs_probe_fileserver(fc)) - goto failed; + if (!afs_probe_fileserver(fc)) { + switch (fc->ac.error) { + case -ENOMEM: + case -ERESTARTSYS: + case -EINTR: + goto failed; + default: + goto next_server; + } + } } if (!fc->ac.alist) diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index f7ae54b6a393..08735948f15d 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -41,6 +41,7 @@ int afs_open_socket(struct afs_net *net) { struct sockaddr_rxrpc srx; struct socket *socket; + unsigned int min_level; int ret; _enter(""); @@ -60,6 +61,12 @@ int afs_open_socket(struct afs_net *net) srx.transport.sin6.sin6_family = AF_INET6; srx.transport.sin6.sin6_port = htons(AFS_CM_PORT); + min_level = RXRPC_SECURITY_ENCRYPT; + ret = kernel_setsockopt(socket, SOL_RXRPC, RXRPC_MIN_SECURITY_LEVEL, + (void *)&min_level, sizeof(min_level)); + if (ret < 0) + goto error_2; + ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); if (ret == -EADDRINUSE) { srx.transport.sin6.sin6_port = 0; @@ -482,8 +489,12 @@ static void afs_deliver_to_call(struct afs_call *call) state = READ_ONCE(call->state); switch (ret) { case 0: - if (state == AFS_CALL_CL_PROC_REPLY) + if (state == AFS_CALL_CL_PROC_REPLY) { + if (call->cbi) + set_bit(AFS_SERVER_FL_MAY_HAVE_CB, + &call->cbi->server->flags); goto call_complete; + } ASSERTCMP(state, >, AFS_CALL_CL_PROC_REPLY); goto done; case -EINPROGRESS: @@ -493,11 +504,6 @@ static void afs_deliver_to_call(struct afs_call *call) case -ECONNABORTED: ASSERTCMP(state, ==, AFS_CALL_COMPLETE); goto done; - case -ENOTCONN: - abort_code = RX_CALL_DEAD; - rxrpc_kernel_abort_call(call->net->socket, call->rxcall, - abort_code, ret, "KNC"); - goto local_abort; case -ENOTSUPP: abort_code = RXGEN_OPCODE; rxrpc_kernel_abort_call(call->net->socket, call->rxcall, @@ -926,3 +932,12 @@ int afs_extract_data(struct afs_call *call, void *buf, size_t count, afs_set_call_complete(call, ret, remote_abort); return ret; } + +/* + * Log protocol error production. + */ +noinline int afs_protocol_error(struct afs_call *call, int error) +{ + trace_afs_protocol_error(call, error, __builtin_return_address(0)); + return error; +} diff --git a/fs/afs/security.c b/fs/afs/security.c index b88b7d45fdaa..81dfedb7879f 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -147,8 +147,7 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, break; } - if (cb_break != (vnode->cb_break + - vnode->cb_interest->server->cb_s_break)) { + if (cb_break != afs_cb_break_sum(vnode, vnode->cb_interest)) { changed = true; break; } @@ -178,18 +177,14 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, } } - if (cb_break != (vnode->cb_break + vnode->cb_interest->server->cb_s_break)) { - rcu_read_unlock(); + if (cb_break != afs_cb_break_sum(vnode, vnode->cb_interest)) goto someone_else_changed_it; - } /* We need a ref on any permits list we want to copy as we'll have to * drop the lock to do memory allocation. */ - if (permits && !refcount_inc_not_zero(&permits->usage)) { - rcu_read_unlock(); + if (permits && !refcount_inc_not_zero(&permits->usage)) goto someone_else_changed_it; - } rcu_read_unlock(); @@ -261,7 +256,7 @@ found: spin_lock(&vnode->lock); zap = rcu_access_pointer(vnode->permit_cache); - if (cb_break == (vnode->cb_break + vnode->cb_interest->server->cb_s_break) && + if (cb_break == afs_cb_break_sum(vnode, vnode->cb_interest) && zap == permits) rcu_assign_pointer(vnode->permit_cache, replacement); else @@ -278,6 +273,7 @@ someone_else_changed_it: /* Someone else changed the cache under us - don't recheck at this * time. */ + rcu_read_unlock(); return; } @@ -296,8 +292,6 @@ int afs_check_permit(struct afs_vnode *vnode, struct key *key, _enter("{%x:%u},%x", vnode->fid.vid, vnode->fid.vnode, key_serial(key)); - permits = vnode->permit_cache; - /* check the permits to see if we've got one yet */ if (key == vnode->volume->cell->anonymous_key) { _debug("anon"); @@ -327,7 +321,7 @@ int afs_check_permit(struct afs_vnode *vnode, struct key *key, */ _debug("no valid permit"); - ret = afs_fetch_status(vnode, key); + ret = afs_fetch_status(vnode, key, false); if (ret < 0) { *_access = 0; _leave(" = %d", ret); @@ -378,18 +372,14 @@ int afs_permission(struct inode *inode, int mask) mask, access, S_ISDIR(inode->i_mode) ? "dir" : "file"); if (S_ISDIR(inode->i_mode)) { - if (mask & MAY_EXEC) { - if (!(access & AFS_ACE_LOOKUP)) - goto permission_denied; - } else if (mask & MAY_READ) { + if (mask & (MAY_EXEC | MAY_READ | MAY_CHDIR)) { if (!(access & AFS_ACE_LOOKUP)) goto permission_denied; - } else if (mask & MAY_WRITE) { + } + if (mask & MAY_WRITE) { if (!(access & (AFS_ACE_DELETE | /* rmdir, unlink, rename from */ AFS_ACE_INSERT))) /* create, mkdir, symlink, rename to */ goto permission_denied; - } else { - BUG(); } } else { if (!(access & AFS_ACE_LOOKUP)) diff --git a/fs/afs/server.c b/fs/afs/server.c index a43ef77dabae..3af4625e2f8c 100644 --- a/fs/afs/server.c +++ b/fs/afs/server.c @@ -59,19 +59,14 @@ struct afs_server *afs_find_server(struct afs_net *net, alist = rcu_dereference(server->addresses); for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) { b = &alist->addrs[i].transport.sin6; - diff = (u16)a->sin6_port - (u16)b->sin6_port; + diff = ((u16 __force)a->sin6_port - + (u16 __force)b->sin6_port); if (diff == 0) diff = memcmp(&a->sin6_addr, &b->sin6_addr, sizeof(struct in6_addr)); if (diff == 0) goto found; - if (diff < 0) { - // TODO: Sort the list - //if (i == alist->nr_ipv4) - // goto not_found; - break; - } } } } else { @@ -79,23 +74,17 @@ struct afs_server *afs_find_server(struct afs_net *net, alist = rcu_dereference(server->addresses); for (i = 0; i < alist->nr_ipv4; i++) { b = &alist->addrs[i].transport.sin6; - diff = (u16)a->sin6_port - (u16)b->sin6_port; + diff = ((u16 __force)a->sin6_port - + (u16 __force)b->sin6_port); if (diff == 0) - diff = ((u32)a->sin6_addr.s6_addr32[3] - - (u32)b->sin6_addr.s6_addr32[3]); + diff = ((u32 __force)a->sin6_addr.s6_addr32[3] - + (u32 __force)b->sin6_addr.s6_addr32[3]); if (diff == 0) goto found; - if (diff < 0) { - // TODO: Sort the list - //if (i == 0) - // goto not_found; - break; - } } } } - //not_found: server = NULL; found: if (server && !atomic_inc_not_zero(&server->usage)) @@ -381,7 +370,7 @@ static void afs_server_rcu(struct rcu_head *rcu) { struct afs_server *server = container_of(rcu, struct afs_server, rcu); - afs_put_addrlist(server->addresses); + afs_put_addrlist(rcu_access_pointer(server->addresses)); kfree(server); } @@ -390,17 +379,19 @@ static void afs_server_rcu(struct rcu_head *rcu) */ static void afs_destroy_server(struct afs_net *net, struct afs_server *server) { - struct afs_addr_list *alist = server->addresses; + struct afs_addr_list *alist = rcu_access_pointer(server->addresses); struct afs_addr_cursor ac = { .alist = alist, - .addr = &alist->addrs[0], .start = alist->index, - .index = alist->index, + .index = 0, + .addr = &alist->addrs[alist->index], .error = 0, }; _enter("%p", server); - afs_fs_give_up_all_callbacks(net, server, &ac, NULL); + if (test_bit(AFS_SERVER_FL_MAY_HAVE_CB, &server->flags)) + afs_fs_give_up_all_callbacks(net, server, &ac, NULL); + call_rcu(&server->rcu, afs_server_rcu); afs_dec_servers_outstanding(net); } @@ -426,8 +417,15 @@ static void afs_gc_servers(struct afs_net *net, struct afs_server *gc_list) } write_sequnlock(&net->fs_lock); - if (deleted) + if (deleted) { + write_seqlock(&net->fs_addr_lock); + if (!hlist_unhashed(&server->addr4_link)) + hlist_del_rcu(&server->addr4_link); + if (!hlist_unhashed(&server->addr6_link)) + hlist_del_rcu(&server->addr6_link); + write_sequnlock(&net->fs_addr_lock); afs_destroy_server(net, server); + } } } diff --git a/fs/afs/server_list.c b/fs/afs/server_list.c index 0f8dc4c8f07c..8a5760aa5832 100644 --- a/fs/afs/server_list.c +++ b/fs/afs/server_list.c @@ -49,6 +49,7 @@ struct afs_server_list *afs_alloc_server_list(struct afs_cell *cell, goto error; refcount_set(&slist->usage, 1); + rwlock_init(&slist->lock); /* Make sure a records exists for each server in the list. */ for (i = 0; i < vldb->nr_servers; i++) { @@ -64,9 +65,11 @@ struct afs_server_list *afs_alloc_server_list(struct afs_cell *cell, goto error_2; } - /* Insertion-sort by server pointer */ + /* Insertion-sort by UUID */ for (j = 0; j < slist->nr_servers; j++) - if (slist->servers[j].server >= server) + if (memcmp(&slist->servers[j].server->uuid, + &server->uuid, + sizeof(server->uuid)) >= 0) break; if (j < slist->nr_servers) { if (slist->servers[j].server == server) { diff --git a/fs/afs/super.c b/fs/afs/super.c index 3623c952b6ff..9e5d7966621c 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -154,7 +154,7 @@ static int afs_show_devname(struct seq_file *m, struct dentry *root) seq_puts(m, "none"); return 0; } - + switch (volume->type) { case AFSVL_RWVOL: break; @@ -269,7 +269,7 @@ static int afs_parse_device_name(struct afs_mount_params *params, int cellnamesz; _enter(",%s", name); - + if (!name) { printk(KERN_ERR "kAFS: no volume name specified\n"); return -EINVAL; @@ -418,7 +418,10 @@ static int afs_fill_super(struct super_block *sb, if (!sb->s_root) goto error; - sb->s_d_op = &afs_fs_dentry_operations; + if (params->dyn_root) + sb->s_d_op = &afs_dynroot_dentry_operations; + else + sb->s_d_op = &afs_fs_dentry_operations; _leave(" = 0"); return 0; @@ -587,7 +590,7 @@ static void afs_i_init_once(void *_vnode) memset(vnode, 0, sizeof(*vnode)); inode_init_once(&vnode->vfs_inode); mutex_init(&vnode->io_lock); - mutex_init(&vnode->validate_lock); + init_rwsem(&vnode->validate_lock); spin_lock_init(&vnode->wb_lock); spin_lock_init(&vnode->lock); INIT_LIST_HEAD(&vnode->wb_keys); @@ -676,7 +679,7 @@ static int afs_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_bfree = 0; return 0; } - + key = afs_request_key(vnode->volume->cell); if (IS_ERR(key)) return PTR_ERR(key); @@ -685,7 +688,7 @@ static int afs_statfs(struct dentry *dentry, struct kstatfs *buf) if (afs_begin_vnode_operation(&fc, vnode, key)) { fc.flags |= AFS_FS_CURSOR_NO_VSLEEP; while (afs_select_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(vnode); afs_fs_get_volume_status(&fc, &vs); } diff --git a/fs/afs/vlclient.c b/fs/afs/vlclient.c index 5d8562f1ad4a..c3b740813fc7 100644 --- a/fs/afs/vlclient.c +++ b/fs/afs/vlclient.c @@ -23,7 +23,7 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) struct afs_uvldbentry__xdr *uvldb; struct afs_vldb_entry *entry; bool new_only = false; - u32 tmp, nr_servers; + u32 tmp, nr_servers, vlflags; int i, ret; _enter(""); @@ -55,6 +55,7 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) new_only = true; } + vlflags = ntohl(uvldb->flags); for (i = 0; i < nr_servers; i++) { struct afs_uuid__xdr *xdr; struct afs_uuid *uuid; @@ -64,12 +65,13 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) if (tmp & AFS_VLSF_DONTUSE || (new_only && !(tmp & AFS_VLSF_NEWREPSITE))) continue; - if (tmp & AFS_VLSF_RWVOL) + if (tmp & AFS_VLSF_RWVOL) { entry->fs_mask[i] |= AFS_VOL_VTM_RW; + if (vlflags & AFS_VLF_BACKEXISTS) + entry->fs_mask[i] |= AFS_VOL_VTM_BAK; + } if (tmp & AFS_VLSF_ROVOL) entry->fs_mask[i] |= AFS_VOL_VTM_RO; - if (tmp & AFS_VLSF_BACKVOL) - entry->fs_mask[i] |= AFS_VOL_VTM_BAK; if (!entry->fs_mask[i]) continue; @@ -89,15 +91,14 @@ static int afs_deliver_vl_get_entry_by_name_u(struct afs_call *call) for (i = 0; i < AFS_MAXTYPES; i++) entry->vid[i] = ntohl(uvldb->volumeId[i]); - tmp = ntohl(uvldb->flags); - if (tmp & AFS_VLF_RWEXISTS) + if (vlflags & AFS_VLF_RWEXISTS) __set_bit(AFS_VLDB_HAS_RW, &entry->flags); - if (tmp & AFS_VLF_ROEXISTS) + if (vlflags & AFS_VLF_ROEXISTS) __set_bit(AFS_VLDB_HAS_RO, &entry->flags); - if (tmp & AFS_VLF_BACKEXISTS) + if (vlflags & AFS_VLF_BACKEXISTS) __set_bit(AFS_VLDB_HAS_BAK, &entry->flags); - if (!(tmp & (AFS_VLF_RWEXISTS | AFS_VLF_ROEXISTS | AFS_VLF_BACKEXISTS))) { + if (!(vlflags & (AFS_VLF_RWEXISTS | AFS_VLF_ROEXISTS | AFS_VLF_BACKEXISTS))) { entry->error = -ENOMEDIUM; __set_bit(AFS_VLDB_QUERY_ERROR, &entry->flags); } @@ -303,7 +304,7 @@ struct afs_addr_list *afs_vl_get_addrs_u(struct afs_net *net, r->uuid.clock_seq_hi_and_reserved = htonl(u->clock_seq_hi_and_reserved); r->uuid.clock_seq_low = htonl(u->clock_seq_low); for (i = 0; i < 6; i++) - r->uuid.node[i] = ntohl(u->node[i]); + r->uuid.node[i] = htonl(u->node[i]); trace_afs_make_vl_call(call); return (struct afs_addr_list *)afs_make_call(ac, call, GFP_KERNEL, false); @@ -450,7 +451,7 @@ again: call->count2 = ntohl(*bp); /* Type or next count */ if (call->count > YFS_MAXENDPOINTS) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); alist = afs_alloc_addrlist(call->count, FS_SERVICE, AFS_FS_PORT); if (!alist) @@ -474,7 +475,7 @@ again: size = sizeof(__be32) * (1 + 4 + 1); break; default: - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); } size += sizeof(__be32); @@ -487,24 +488,24 @@ again: switch (call->count2) { case YFS_ENDPOINT_IPV4: if (ntohl(bp[0]) != sizeof(__be32) * 2) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); afs_merge_fs_addr4(alist, bp[1], ntohl(bp[2])); bp += 3; break; case YFS_ENDPOINT_IPV6: if (ntohl(bp[0]) != sizeof(__be32) * 5) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); afs_merge_fs_addr6(alist, bp + 1, ntohl(bp[5])); bp += 6; break; default: - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); } /* Got either the type of the next entry or the count of * volEndpoints if no more fsEndpoints. */ - call->count2 = htonl(*bp++); + call->count2 = ntohl(*bp++); call->offset = 0; call->count--; @@ -517,7 +518,7 @@ again: if (!call->count) goto end; if (call->count > YFS_MAXENDPOINTS) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); call->unmarshall = 3; @@ -531,7 +532,7 @@ again: return ret; bp = call->buffer; - call->count2 = htonl(*bp++); + call->count2 = ntohl(*bp++); call->offset = 0; call->unmarshall = 4; @@ -545,7 +546,7 @@ again: size = sizeof(__be32) * (1 + 4 + 1); break; default: - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); } if (call->count > 1) @@ -558,16 +559,16 @@ again: switch (call->count2) { case YFS_ENDPOINT_IPV4: if (ntohl(bp[0]) != sizeof(__be32) * 2) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); bp += 3; break; case YFS_ENDPOINT_IPV6: if (ntohl(bp[0]) != sizeof(__be32) * 5) - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); bp += 6; break; default: - return -EBADMSG; + return afs_protocol_error(call, -EBADMSG); } /* Got either the type of the next entry or the count of @@ -576,7 +577,7 @@ again: call->offset = 0; call->count--; if (call->count > 0) { - call->count2 = htonl(*bp++); + call->count2 = ntohl(*bp++); goto again; } diff --git a/fs/afs/volume.c b/fs/afs/volume.c index b517a588781f..3037bd01f617 100644 --- a/fs/afs/volume.c +++ b/fs/afs/volume.c @@ -225,7 +225,9 @@ void afs_activate_volume(struct afs_volume *volume) #ifdef CONFIG_AFS_FSCACHE volume->cache = fscache_acquire_cookie(volume->cell->cache, &afs_volume_cache_index_def, - volume, true); + &volume->vid, sizeof(volume->vid), + NULL, 0, + volume, 0, true); #endif write_lock(&volume->cell->proc_lock); @@ -245,7 +247,7 @@ void afs_deactivate_volume(struct afs_volume *volume) write_unlock(&volume->cell->proc_lock); #ifdef CONFIG_AFS_FSCACHE - fscache_relinquish_cookie(volume->cache, + fscache_relinquish_cookie(volume->cache, NULL, test_bit(AFS_VOLUME_DELETED, &volume->flags)); volume->cache = NULL; #endif diff --git a/fs/afs/write.c b/fs/afs/write.c index 9370e2feb999..8b39e6ebb40b 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -42,10 +42,11 @@ static int afs_fill_page(struct afs_vnode *vnode, struct key *key, if (!req) return -ENOMEM; - atomic_set(&req->usage, 1); + refcount_set(&req->usage, 1); req->pos = pos; req->len = len; req->nr_pages = 1; + req->pages = req->array; req->pages[0] = page; get_page(page); @@ -124,7 +125,12 @@ try_again: page->index, priv); goto flush_conflicting_write; } - if (to < f || from > t) + /* If the file is being filled locally, allow inter-write + * spaces to be merged into writes. If it's not, only write + * back what the user gives us. + */ + if (!test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags) && + (to < f || from > t)) goto flush_conflicting_write; if (from < f) f = from; @@ -345,7 +351,7 @@ found_key: ret = -ERESTARTSYS; if (afs_begin_vnode_operation(&fc, vnode, wbk->key)) { while (afs_select_fileserver(&fc)) { - fc.cb_break = vnode->cb_break + vnode->cb_s_break; + fc.cb_break = afs_calc_vnode_cb_break(vnode); afs_fs_store_data(&fc, mapping, first, last, offset, to); } @@ -355,6 +361,12 @@ found_key: } switch (ret) { + case 0: + afs_stat_v(vnode, n_stores); + atomic_long_add((last * PAGE_SIZE + to) - + (first * PAGE_SIZE + offset), + &afs_v2net(vnode)->n_store_bytes); + break; case -EACCES: case -EPERM: case -ENOKEY: @@ -412,7 +424,8 @@ static int afs_write_back_from_locked_page(struct address_space *mapping, trace_afs_page_dirty(vnode, tracepoint_string("WARN"), primary_page->index, priv); - if (start >= final_page || to < PAGE_SIZE) + if (start >= final_page || + (to < PAGE_SIZE && !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags))) goto no_more; start++; @@ -433,9 +446,10 @@ static int afs_write_back_from_locked_page(struct address_space *mapping, } for (loop = 0; loop < n; loop++) { - if (to != PAGE_SIZE) - break; page = pages[loop]; + if (to != PAGE_SIZE && + !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags)) + break; if (page->index > final_page) break; if (!trylock_page(page)) @@ -448,7 +462,8 @@ static int afs_write_back_from_locked_page(struct address_space *mapping, priv = page_private(page); f = priv & AFS_PRIV_MAX; t = priv >> AFS_PRIV_SHIFT; - if (f != 0) { + if (f != 0 && + !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags)) { unlock_page(page); break; } @@ -570,10 +585,11 @@ static int afs_writepages_region(struct address_space *mapping, _debug("wback %lx", page->index); - /* at this point we hold neither mapping->tree_lock nor lock on - * the page itself: the page may be truncated or invalidated - * (changing page->mapping to NULL), or even swizzled back from - * swapper_space to tmpfs file mapping + /* + * at this point we hold neither the i_pages lock nor the + * page lock: the page may be truncated or invalidated + * (changing page->mapping to NULL), or even swizzled + * back from swapper_space to tmpfs file mapping */ ret = lock_page_killable(page); if (ret < 0) { @@ -734,20 +750,6 @@ int afs_fsync(struct file *file, loff_t start, loff_t end, int datasync) } /* - * Flush out all outstanding writes on a file opened for writing when it is - * closed. - */ -int afs_flush(struct file *file, fl_owner_t id) -{ - _enter(""); - - if ((file->f_mode & FMODE_WRITE) == 0) - return 0; - - return vfs_fsync(file, 0); -} - -/* * notification that a previously read-only page is about to become writable * - if it returns an error, the caller will deliver a bus error signal */ diff --git a/fs/afs/xdr_fs.h b/fs/afs/xdr_fs.h new file mode 100644 index 000000000000..aa21f3068d52 --- /dev/null +++ b/fs/afs/xdr_fs.h @@ -0,0 +1,103 @@ +/* AFS fileserver XDR types + * + * Copyright (C) 2018 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#ifndef XDR_FS_H +#define XDR_FS_H + +struct afs_xdr_AFSFetchStatus { + __be32 if_version; +#define AFS_FSTATUS_VERSION 1 + __be32 type; + __be32 nlink; + __be32 size_lo; + __be32 data_version_lo; + __be32 author; + __be32 owner; + __be32 caller_access; + __be32 anon_access; + __be32 mode; + __be32 parent_vnode; + __be32 parent_unique; + __be32 seg_size; + __be32 mtime_client; + __be32 mtime_server; + __be32 group; + __be32 sync_counter; + __be32 data_version_hi; + __be32 lock_count; + __be32 size_hi; + __be32 abort_code; +} __packed; + +#define AFS_DIR_HASHTBL_SIZE 128 +#define AFS_DIR_DIRENT_SIZE 32 +#define AFS_DIR_SLOTS_PER_BLOCK 64 +#define AFS_DIR_BLOCK_SIZE 2048 +#define AFS_DIR_BLOCKS_PER_PAGE (PAGE_SIZE / AFS_DIR_BLOCK_SIZE) +#define AFS_DIR_MAX_SLOTS 65536 +#define AFS_DIR_BLOCKS_WITH_CTR 128 +#define AFS_DIR_MAX_BLOCKS 1023 +#define AFS_DIR_RESV_BLOCKS 1 +#define AFS_DIR_RESV_BLOCKS0 13 + +/* + * Directory entry structure. + */ +union afs_xdr_dirent { + struct { + u8 valid; + u8 unused[1]; + __be16 hash_next; + __be32 vnode; + __be32 unique; + u8 name[16]; + u8 overflow[4]; /* if any char of the name (inc + * NUL) reaches here, consume + * the next dirent too */ + } u; + u8 extended_name[32]; +} __packed; + +/* + * Directory block header (one at the beginning of every 2048-byte block). + */ +struct afs_xdr_dir_hdr { + __be16 npages; + __be16 magic; +#define AFS_DIR_MAGIC htons(1234) + u8 reserved; + u8 bitmap[8]; + u8 pad[19]; +} __packed; + +/* + * Directory block layout + */ +union afs_xdr_dir_block { + struct afs_xdr_dir_hdr hdr; + + struct { + struct afs_xdr_dir_hdr hdr; + u8 alloc_ctrs[AFS_DIR_MAX_BLOCKS]; + __be16 hashtable[AFS_DIR_HASHTBL_SIZE]; + } meta; + + union afs_xdr_dirent dirents[AFS_DIR_SLOTS_PER_BLOCK]; +} __packed; + +/* + * Directory layout on a linux VM page. + */ +struct afs_xdr_dir_page { + union afs_xdr_dir_block blocks[AFS_DIR_BLOCKS_PER_PAGE]; +}; + +#endif /* XDR_FS_H */ @@ -5,6 +5,7 @@ * Implements an efficient asynchronous io interface. * * Copyright 2000, 2001, 2002 Red Hat, Inc. All Rights Reserved. + * Copyright 2018 Christoph Hellwig. * * See ../COPYING for licensing terms. */ @@ -46,6 +47,8 @@ #include "internal.h" +#define KIOCB_KEY 0 + #define AIO_RING_MAGIC 0xa10a10a1 #define AIO_RING_COMPAT_FEATURES 1 #define AIO_RING_INCOMPAT_FEATURES 0 @@ -156,21 +159,29 @@ struct kioctx { unsigned id; }; -/* - * We use ki_cancel == KIOCB_CANCELLED to indicate that a kiocb has been either - * cancelled or completed (this makes a certain amount of sense because - * successful cancellation - io_cancel() - does deliver the completion to - * userspace). - * - * And since most things don't implement kiocb cancellation and we'd really like - * kiocb completion to be lockless when possible, we use ki_cancel to - * synchronize cancellation and completion - we only set it to KIOCB_CANCELLED - * with xchg() or cmpxchg(), see batch_complete_aio() and kiocb_cancel(). - */ -#define KIOCB_CANCELLED ((void *) (~0ULL)) +struct fsync_iocb { + struct work_struct work; + struct file *file; + bool datasync; +}; + +struct poll_iocb { + struct file *file; + __poll_t events; + struct wait_queue_head *head; + + union { + struct wait_queue_entry wait; + struct work_struct work; + }; +}; struct aio_kiocb { - struct kiocb common; + union { + struct kiocb rw; + struct fsync_iocb fsync; + struct poll_iocb poll; + }; struct kioctx *ki_ctx; kiocb_cancel_fn *ki_cancel; @@ -264,9 +275,6 @@ static int __init aio_setup(void) kiocb_cachep = KMEM_CACHE(aio_kiocb, SLAB_HWCACHE_ALIGN|SLAB_PANIC); kioctx_cachep = KMEM_CACHE(kioctx,SLAB_HWCACHE_ALIGN|SLAB_PANIC); - - pr_debug("sizeof(struct page) = %zu\n", sizeof(struct page)); - return 0; } __initcall(aio_setup); @@ -552,42 +560,20 @@ static int aio_setup_ring(struct kioctx *ctx, unsigned int nr_events) void kiocb_set_cancel_fn(struct kiocb *iocb, kiocb_cancel_fn *cancel) { - struct aio_kiocb *req = container_of(iocb, struct aio_kiocb, common); + struct aio_kiocb *req = container_of(iocb, struct aio_kiocb, rw); struct kioctx *ctx = req->ki_ctx; unsigned long flags; - spin_lock_irqsave(&ctx->ctx_lock, flags); - - if (!req->ki_list.next) - list_add(&req->ki_list, &ctx->active_reqs); + if (WARN_ON_ONCE(!list_empty(&req->ki_list))) + return; + spin_lock_irqsave(&ctx->ctx_lock, flags); + list_add_tail(&req->ki_list, &ctx->active_reqs); req->ki_cancel = cancel; - spin_unlock_irqrestore(&ctx->ctx_lock, flags); } EXPORT_SYMBOL(kiocb_set_cancel_fn); -static int kiocb_cancel(struct aio_kiocb *kiocb) -{ - kiocb_cancel_fn *old, *cancel; - - /* - * Don't want to set kiocb->ki_cancel = KIOCB_CANCELLED unless it - * actually has a cancel function, hence the cmpxchg() - */ - - cancel = READ_ONCE(kiocb->ki_cancel); - do { - if (!cancel || cancel == KIOCB_CANCELLED) - return -EINVAL; - - old = cancel; - cancel = cmpxchg(&kiocb->ki_cancel, old, KIOCB_CANCELLED); - } while (cancel != old); - - return cancel(&kiocb->common); -} - /* * free_ioctx() should be RCU delayed to synchronize against the RCU * protected lookup_ioctx() and also needs process context to call @@ -634,9 +620,8 @@ static void free_ioctx_users(struct percpu_ref *ref) while (!list_empty(&ctx->active_reqs)) { req = list_first_entry(&ctx->active_reqs, struct aio_kiocb, ki_list); - + req->ki_cancel(&req->rw); list_del_init(&req->ki_list); - kiocb_cancel(req); } spin_unlock_irq(&ctx->ctx_lock); @@ -1042,7 +1027,7 @@ static inline struct aio_kiocb *aio_get_req(struct kioctx *ctx) goto out_put; percpu_ref_get(&ctx->reqs); - + INIT_LIST_HEAD(&req->ki_list); req->ki_ctx = ctx; return req; out_put: @@ -1050,15 +1035,6 @@ out_put: return NULL; } -static void kiocb_free(struct aio_kiocb *req) -{ - if (req->common.ki_filp) - fput(req->common.ki_filp); - if (req->ki_eventfd != NULL) - eventfd_ctx_put(req->ki_eventfd); - kmem_cache_free(kiocb_cachep, req); -} - static struct kioctx *lookup_ioctx(unsigned long ctx_id) { struct aio_ring __user *ring = (void __user *)ctx_id; @@ -1078,8 +1054,8 @@ static struct kioctx *lookup_ioctx(unsigned long ctx_id) ctx = rcu_dereference(table->table[id]); if (ctx && ctx->user_id == ctx_id) { - percpu_ref_get(&ctx->users); - ret = ctx; + if (percpu_ref_tryget_live(&ctx->users)) + ret = ctx; } out: rcu_read_unlock(); @@ -1089,44 +1065,14 @@ out: /* aio_complete * Called when the io request on the given iocb is complete. */ -static void aio_complete(struct kiocb *kiocb, long res, long res2) +static void aio_complete(struct aio_kiocb *iocb, long res, long res2) { - struct aio_kiocb *iocb = container_of(kiocb, struct aio_kiocb, common); struct kioctx *ctx = iocb->ki_ctx; struct aio_ring *ring; struct io_event *ev_page, *event; unsigned tail, pos, head; unsigned long flags; - if (kiocb->ki_flags & IOCB_WRITE) { - struct file *file = kiocb->ki_filp; - - /* - * Tell lockdep we inherited freeze protection from submission - * thread. - */ - if (S_ISREG(file_inode(file)->i_mode)) - __sb_writers_acquired(file_inode(file)->i_sb, SB_FREEZE_WRITE); - file_end_write(file); - } - - /* - * Special case handling for sync iocbs: - * - events go directly into the iocb for fast handling - * - the sync task with the iocb in its stack holds the single iocb - * ref, no other paths have a way to get another ref - * - the sync task helpfully left a reference to itself in the iocb - */ - BUG_ON(is_sync_kiocb(kiocb)); - - if (iocb->ki_list.next) { - unsigned long flags; - - spin_lock_irqsave(&ctx->ctx_lock, flags); - list_del(&iocb->ki_list); - spin_unlock_irqrestore(&ctx->ctx_lock, flags); - } - /* * Add a completion event to the ring buffer. Must be done holding * ctx->completion_lock to prevent other code from messing with the tail @@ -1180,11 +1126,12 @@ static void aio_complete(struct kiocb *kiocb, long res, long res2) * eventfd. The eventfd_signal() function is safe to be called * from IRQ context. */ - if (iocb->ki_eventfd != NULL) + if (iocb->ki_eventfd) { eventfd_signal(iocb->ki_eventfd, 1); + eventfd_ctx_put(iocb->ki_eventfd); + } - /* everything turned out well, dispose of the aiocb. */ - kiocb_free(iocb); + kmem_cache_free(kiocb_cachep, iocb); /* * We have to order our ring_info tail store above and test @@ -1250,14 +1197,13 @@ static long aio_read_events_ring(struct kioctx *ctx, if (head == tail) break; - avail = min(avail, nr - ret); - avail = min_t(long, avail, AIO_EVENTS_PER_PAGE - - ((head + AIO_EVENTS_OFFSET) % AIO_EVENTS_PER_PAGE)); - pos = head + AIO_EVENTS_OFFSET; page = ctx->ring_pages[pos / AIO_EVENTS_PER_PAGE]; pos %= AIO_EVENTS_PER_PAGE; + avail = min(avail, nr - ret); + avail = min_t(long, avail, AIO_EVENTS_PER_PAGE - pos); + ev = kmap(page); copy_ret = copy_to_user(event + ret, ev + pos, sizeof(*ev) * avail); @@ -1328,10 +1274,6 @@ static long read_events(struct kioctx *ctx, long min_nr, long nr, wait_event_interruptible_hrtimeout(ctx->wait, aio_read_events(ctx, min_nr, nr, event, &ret), until); - - if (!ret && signal_pending(current)) - ret = -EINTR; - return ret; } @@ -1447,6 +1389,58 @@ SYSCALL_DEFINE1(io_destroy, aio_context_t, ctx) return -EINVAL; } +static void aio_remove_iocb(struct aio_kiocb *iocb) +{ + struct kioctx *ctx = iocb->ki_ctx; + unsigned long flags; + + spin_lock_irqsave(&ctx->ctx_lock, flags); + list_del(&iocb->ki_list); + spin_unlock_irqrestore(&ctx->ctx_lock, flags); +} + +static void aio_complete_rw(struct kiocb *kiocb, long res, long res2) +{ + struct aio_kiocb *iocb = container_of(kiocb, struct aio_kiocb, rw); + + if (!list_empty_careful(&iocb->ki_list)) + aio_remove_iocb(iocb); + + if (kiocb->ki_flags & IOCB_WRITE) { + struct inode *inode = file_inode(kiocb->ki_filp); + + /* + * Tell lockdep we inherited freeze protection from submission + * thread. + */ + if (S_ISREG(inode->i_mode)) + __sb_writers_acquired(inode->i_sb, SB_FREEZE_WRITE); + file_end_write(kiocb->ki_filp); + } + + fput(kiocb->ki_filp); + aio_complete(iocb, res, res2); +} + +static int aio_prep_rw(struct kiocb *req, struct iocb *iocb) +{ + int ret; + + req->ki_filp = fget(iocb->aio_fildes); + if (unlikely(!req->ki_filp)) + return -EBADF; + req->ki_complete = aio_complete_rw; + req->ki_pos = iocb->aio_offset; + req->ki_flags = iocb_flags(req->ki_filp); + if (iocb->aio_flags & IOCB_FLAG_RESFD) + req->ki_flags |= IOCB_EVENTFD; + req->ki_hint = file_write_hint(req->ki_filp); + ret = kiocb_set_rw_flags(req, iocb->aio_rw_flags); + if (unlikely(ret)) + fput(req->ki_filp); + return ret; +} + static int aio_setup_rw(int rw, struct iocb *iocb, struct iovec **iovec, bool vectored, bool compat, struct iov_iter *iter) { @@ -1466,11 +1460,11 @@ static int aio_setup_rw(int rw, struct iocb *iocb, struct iovec **iovec, return import_iovec(rw, buf, len, UIO_FASTIOV, iovec, iter); } -static inline ssize_t aio_ret(struct kiocb *req, ssize_t ret) +static inline void aio_rw_done(struct kiocb *req, ssize_t ret) { switch (ret) { case -EIOCBQUEUED: - return ret; + break; case -ERESTARTSYS: case -ERESTARTNOINTR: case -ERESTARTNOHAND: @@ -1482,85 +1476,270 @@ static inline ssize_t aio_ret(struct kiocb *req, ssize_t ret) ret = -EINTR; /*FALLTHRU*/ default: - aio_complete(req, ret, 0); - return 0; + aio_complete_rw(req, ret, 0); } } static ssize_t aio_read(struct kiocb *req, struct iocb *iocb, bool vectored, bool compat) { - struct file *file = req->ki_filp; struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs; struct iov_iter iter; + struct file *file; ssize_t ret; + ret = aio_prep_rw(req, iocb); + if (ret) + return ret; + file = req->ki_filp; + + ret = -EBADF; if (unlikely(!(file->f_mode & FMODE_READ))) - return -EBADF; + goto out_fput; + ret = -EINVAL; if (unlikely(!file->f_op->read_iter)) - return -EINVAL; + goto out_fput; ret = aio_setup_rw(READ, iocb, &iovec, vectored, compat, &iter); if (ret) - return ret; + goto out_fput; ret = rw_verify_area(READ, file, &req->ki_pos, iov_iter_count(&iter)); if (!ret) - ret = aio_ret(req, call_read_iter(file, req, &iter)); + aio_rw_done(req, call_read_iter(file, req, &iter)); kfree(iovec); +out_fput: + if (unlikely(ret)) + fput(file); return ret; } static ssize_t aio_write(struct kiocb *req, struct iocb *iocb, bool vectored, bool compat) { - struct file *file = req->ki_filp; struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs; struct iov_iter iter; + struct file *file; ssize_t ret; + ret = aio_prep_rw(req, iocb); + if (ret) + return ret; + file = req->ki_filp; + + ret = -EBADF; if (unlikely(!(file->f_mode & FMODE_WRITE))) - return -EBADF; + goto out_fput; + ret = -EINVAL; if (unlikely(!file->f_op->write_iter)) - return -EINVAL; + goto out_fput; ret = aio_setup_rw(WRITE, iocb, &iovec, vectored, compat, &iter); if (ret) - return ret; + goto out_fput; ret = rw_verify_area(WRITE, file, &req->ki_pos, iov_iter_count(&iter)); if (!ret) { - req->ki_flags |= IOCB_WRITE; - file_start_write(file); - ret = aio_ret(req, call_write_iter(file, req, &iter)); /* - * We release freeze protection in aio_complete(). Fool lockdep - * by telling it the lock got released so that it doesn't - * complain about held lock when we return to userspace. + * Open-code file_start_write here to grab freeze protection, + * which will be released by another thread in + * aio_complete_rw(). Fool lockdep by telling it the lock got + * released so that it doesn't complain about the held lock when + * we return to userspace. */ - if (S_ISREG(file_inode(file)->i_mode)) + if (S_ISREG(file_inode(file)->i_mode)) { + __sb_start_write(file_inode(file)->i_sb, SB_FREEZE_WRITE, true); __sb_writers_release(file_inode(file)->i_sb, SB_FREEZE_WRITE); + } + req->ki_flags |= IOCB_WRITE; + aio_rw_done(req, call_write_iter(file, req, &iter)); } kfree(iovec); +out_fput: + if (unlikely(ret)) + fput(file); return ret; } +static void aio_fsync_work(struct work_struct *work) +{ + struct fsync_iocb *req = container_of(work, struct fsync_iocb, work); + int ret; + + ret = vfs_fsync(req->file, req->datasync); + fput(req->file); + aio_complete(container_of(req, struct aio_kiocb, fsync), ret, 0); +} + +static int aio_fsync(struct fsync_iocb *req, struct iocb *iocb, bool datasync) +{ + if (unlikely(iocb->aio_buf || iocb->aio_offset || iocb->aio_nbytes || + iocb->aio_rw_flags)) + return -EINVAL; + req->file = fget(iocb->aio_fildes); + if (unlikely(!req->file)) + return -EBADF; + if (unlikely(!req->file->f_op->fsync)) { + fput(req->file); + return -EINVAL; + } + + req->datasync = datasync; + INIT_WORK(&req->work, aio_fsync_work); + schedule_work(&req->work); + return 0; +} + +/* need to use list_del_init so we can check if item was present */ +static inline bool __aio_poll_remove(struct poll_iocb *req) +{ + if (list_empty(&req->wait.entry)) + return false; + list_del_init(&req->wait.entry); + return true; +} + +static inline void __aio_poll_complete(struct aio_kiocb *iocb, __poll_t mask) +{ + fput(iocb->poll.file); + aio_complete(iocb, mangle_poll(mask), 0); +} + +static void aio_poll_work(struct work_struct *work) +{ + struct aio_kiocb *iocb = container_of(work, struct aio_kiocb, poll.work); + + if (!list_empty_careful(&iocb->ki_list)) + aio_remove_iocb(iocb); + __aio_poll_complete(iocb, iocb->poll.events); +} + +static int aio_poll_cancel(struct kiocb *iocb) +{ + struct aio_kiocb *aiocb = container_of(iocb, struct aio_kiocb, rw); + struct poll_iocb *req = &aiocb->poll; + struct wait_queue_head *head = req->head; + bool found = false; + + spin_lock(&head->lock); + found = __aio_poll_remove(req); + spin_unlock(&head->lock); + + if (found) { + req->events = 0; + INIT_WORK(&req->work, aio_poll_work); + schedule_work(&req->work); + } + return 0; +} + +static int aio_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync, + void *key) +{ + struct poll_iocb *req = container_of(wait, struct poll_iocb, wait); + struct aio_kiocb *iocb = container_of(req, struct aio_kiocb, poll); + struct file *file = req->file; + __poll_t mask = key_to_poll(key); + + assert_spin_locked(&req->head->lock); + + /* for instances that support it check for an event match first: */ + if (mask && !(mask & req->events)) + return 0; + + mask = file->f_op->poll_mask(file, req->events); + if (!mask) + return 0; + + __aio_poll_remove(req); + + /* + * Try completing without a context switch if we can acquire ctx_lock + * without spinning. Otherwise we need to defer to a workqueue to + * avoid a deadlock due to the lock order. + */ + if (spin_trylock(&iocb->ki_ctx->ctx_lock)) { + list_del_init(&iocb->ki_list); + spin_unlock(&iocb->ki_ctx->ctx_lock); + + __aio_poll_complete(iocb, mask); + } else { + req->events = mask; + INIT_WORK(&req->work, aio_poll_work); + schedule_work(&req->work); + } + + return 1; +} + +static ssize_t aio_poll(struct aio_kiocb *aiocb, struct iocb *iocb) +{ + struct kioctx *ctx = aiocb->ki_ctx; + struct poll_iocb *req = &aiocb->poll; + __poll_t mask; + + /* reject any unknown events outside the normal event mask. */ + if ((u16)iocb->aio_buf != iocb->aio_buf) + return -EINVAL; + /* reject fields that are not defined for poll */ + if (iocb->aio_offset || iocb->aio_nbytes || iocb->aio_rw_flags) + return -EINVAL; + + req->events = demangle_poll(iocb->aio_buf) | EPOLLERR | EPOLLHUP; + req->file = fget(iocb->aio_fildes); + if (unlikely(!req->file)) + return -EBADF; + if (!file_has_poll_mask(req->file)) + goto out_fail; + + req->head = req->file->f_op->get_poll_head(req->file, req->events); + if (!req->head) + goto out_fail; + if (IS_ERR(req->head)) { + mask = EPOLLERR; + goto done; + } + + init_waitqueue_func_entry(&req->wait, aio_poll_wake); + aiocb->ki_cancel = aio_poll_cancel; + + spin_lock_irq(&ctx->ctx_lock); + spin_lock(&req->head->lock); + mask = req->file->f_op->poll_mask(req->file, req->events); + if (!mask) { + __add_wait_queue(req->head, &req->wait); + list_add_tail(&aiocb->ki_list, &ctx->active_reqs); + } + spin_unlock(&req->head->lock); + spin_unlock_irq(&ctx->ctx_lock); +done: + if (mask) + __aio_poll_complete(aiocb, mask); + return 0; +out_fail: + fput(req->file); + return -EINVAL; /* same as no support for IOCB_CMD_POLL */ +} + static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, - struct iocb *iocb, bool compat) + bool compat) { struct aio_kiocb *req; - struct file *file; + struct iocb iocb; ssize_t ret; + if (unlikely(copy_from_user(&iocb, user_iocb, sizeof(iocb)))) + return -EFAULT; + /* enforce forwards compatibility on users */ - if (unlikely(iocb->aio_reserved2)) { + if (unlikely(iocb.aio_reserved2)) { pr_debug("EINVAL: reserve field set\n"); return -EINVAL; } /* prevent overflows */ if (unlikely( - (iocb->aio_buf != (unsigned long)iocb->aio_buf) || - (iocb->aio_nbytes != (size_t)iocb->aio_nbytes) || - ((ssize_t)iocb->aio_nbytes < 0) + (iocb.aio_buf != (unsigned long)iocb.aio_buf) || + (iocb.aio_nbytes != (size_t)iocb.aio_nbytes) || + ((ssize_t)iocb.aio_nbytes < 0) )) { pr_debug("EINVAL: overflow check\n"); return -EINVAL; @@ -1570,37 +1749,19 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, if (unlikely(!req)) return -EAGAIN; - req->common.ki_filp = file = fget(iocb->aio_fildes); - if (unlikely(!req->common.ki_filp)) { - ret = -EBADF; - goto out_put_req; - } - req->common.ki_pos = iocb->aio_offset; - req->common.ki_complete = aio_complete; - req->common.ki_flags = iocb_flags(req->common.ki_filp); - req->common.ki_hint = file_write_hint(file); - - if (iocb->aio_flags & IOCB_FLAG_RESFD) { + if (iocb.aio_flags & IOCB_FLAG_RESFD) { /* * If the IOCB_FLAG_RESFD flag of aio_flags is set, get an * instance of the file* now. The file descriptor must be * an eventfd() fd, and will be signaled for each completed * event using the eventfd_signal() function. */ - req->ki_eventfd = eventfd_ctx_fdget((int) iocb->aio_resfd); + req->ki_eventfd = eventfd_ctx_fdget((int) iocb.aio_resfd); if (IS_ERR(req->ki_eventfd)) { ret = PTR_ERR(req->ki_eventfd); req->ki_eventfd = NULL; goto out_put_req; } - - req->common.ki_flags |= IOCB_EVENTFD; - } - - ret = kiocb_set_rw_flags(&req->common, iocb->aio_rw_flags); - if (unlikely(ret)) { - pr_debug("EINVAL: aio_rw_flags\n"); - goto out_put_req; } ret = put_user(KIOCB_KEY, &user_iocb->aio_key); @@ -1610,41 +1771,67 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, } req->ki_user_iocb = user_iocb; - req->ki_user_data = iocb->aio_data; + req->ki_user_data = iocb.aio_data; - get_file(file); - switch (iocb->aio_lio_opcode) { + switch (iocb.aio_lio_opcode) { case IOCB_CMD_PREAD: - ret = aio_read(&req->common, iocb, false, compat); + ret = aio_read(&req->rw, &iocb, false, compat); break; case IOCB_CMD_PWRITE: - ret = aio_write(&req->common, iocb, false, compat); + ret = aio_write(&req->rw, &iocb, false, compat); break; case IOCB_CMD_PREADV: - ret = aio_read(&req->common, iocb, true, compat); + ret = aio_read(&req->rw, &iocb, true, compat); break; case IOCB_CMD_PWRITEV: - ret = aio_write(&req->common, iocb, true, compat); + ret = aio_write(&req->rw, &iocb, true, compat); + break; + case IOCB_CMD_FSYNC: + ret = aio_fsync(&req->fsync, &iocb, false); + break; + case IOCB_CMD_FDSYNC: + ret = aio_fsync(&req->fsync, &iocb, true); + break; + case IOCB_CMD_POLL: + ret = aio_poll(req, &iocb); break; default: - pr_debug("invalid aio operation %d\n", iocb->aio_lio_opcode); + pr_debug("invalid aio operation %d\n", iocb.aio_lio_opcode); ret = -EINVAL; break; } - fput(file); - if (ret && ret != -EIOCBQUEUED) + /* + * If ret is 0, we'd either done aio_complete() ourselves or have + * arranged for that to be done asynchronously. Anything non-zero + * means that we need to destroy req ourselves. + */ + if (ret) goto out_put_req; return 0; out_put_req: put_reqs_available(ctx, 1); percpu_ref_put(&ctx->reqs); - kiocb_free(req); + if (req->ki_eventfd) + eventfd_ctx_put(req->ki_eventfd); + kmem_cache_free(kiocb_cachep, req); return ret; } -static long do_io_submit(aio_context_t ctx_id, long nr, - struct iocb __user *__user *iocbpp, bool compat) +/* sys_io_submit: + * Queue the nr iocbs pointed to by iocbpp for processing. Returns + * the number of iocbs queued. May return -EINVAL if the aio_context + * specified by ctx_id is invalid, if nr is < 0, if the iocb at + * *iocbpp[0] is not properly initialized, if the operation specified + * is invalid for the file descriptor in the iocb. May fail with + * -EFAULT if any of the data structures point to invalid data. May + * fail with -EBADF if the file descriptor specified in the first + * iocb is invalid. May fail with -EAGAIN if insufficient resources + * are available to queue any iocbs. Will return 0 if nr is 0. Will + * fail with -ENOSYS if not implemented. + */ +SYSCALL_DEFINE3(io_submit, aio_context_t, ctx_id, long, nr, + struct iocb __user * __user *, iocbpp) { struct kioctx *ctx; long ret = 0; @@ -1654,39 +1841,25 @@ static long do_io_submit(aio_context_t ctx_id, long nr, if (unlikely(nr < 0)) return -EINVAL; - if (unlikely(nr > LONG_MAX/sizeof(*iocbpp))) - nr = LONG_MAX/sizeof(*iocbpp); - - if (unlikely(!access_ok(VERIFY_READ, iocbpp, (nr*sizeof(*iocbpp))))) - return -EFAULT; - ctx = lookup_ioctx(ctx_id); if (unlikely(!ctx)) { pr_debug("EINVAL: invalid context id\n"); return -EINVAL; } - blk_start_plug(&plug); + if (nr > ctx->nr_events) + nr = ctx->nr_events; - /* - * AKPM: should this return a partial result if some of the IOs were - * successfully submitted? - */ - for (i=0; i<nr; i++) { + blk_start_plug(&plug); + for (i = 0; i < nr; i++) { struct iocb __user *user_iocb; - struct iocb tmp; - if (unlikely(__get_user(user_iocb, iocbpp + i))) { + if (unlikely(get_user(user_iocb, iocbpp + i))) { ret = -EFAULT; break; } - if (unlikely(copy_from_user(&tmp, user_iocb, sizeof(tmp)))) { - ret = -EFAULT; - break; - } - - ret = io_submit_one(ctx, user_iocb, &tmp, compat); + ret = io_submit_one(ctx, user_iocb, false); if (ret) break; } @@ -1696,59 +1869,44 @@ static long do_io_submit(aio_context_t ctx_id, long nr, return i ? i : ret; } -/* sys_io_submit: - * Queue the nr iocbs pointed to by iocbpp for processing. Returns - * the number of iocbs queued. May return -EINVAL if the aio_context - * specified by ctx_id is invalid, if nr is < 0, if the iocb at - * *iocbpp[0] is not properly initialized, if the operation specified - * is invalid for the file descriptor in the iocb. May fail with - * -EFAULT if any of the data structures point to invalid data. May - * fail with -EBADF if the file descriptor specified in the first - * iocb is invalid. May fail with -EAGAIN if insufficient resources - * are available to queue any iocbs. Will return 0 if nr is 0. Will - * fail with -ENOSYS if not implemented. - */ -SYSCALL_DEFINE3(io_submit, aio_context_t, ctx_id, long, nr, - struct iocb __user * __user *, iocbpp) -{ - return do_io_submit(ctx_id, nr, iocbpp, 0); -} - #ifdef CONFIG_COMPAT -static inline long -copy_iocb(long nr, u32 __user *ptr32, struct iocb __user * __user *ptr64) +COMPAT_SYSCALL_DEFINE3(io_submit, compat_aio_context_t, ctx_id, + int, nr, compat_uptr_t __user *, iocbpp) { - compat_uptr_t uptr; - int i; + struct kioctx *ctx; + long ret = 0; + int i = 0; + struct blk_plug plug; - for (i = 0; i < nr; ++i) { - if (get_user(uptr, ptr32 + i)) - return -EFAULT; - if (put_user(compat_ptr(uptr), ptr64 + i)) - return -EFAULT; + if (unlikely(nr < 0)) + return -EINVAL; + + ctx = lookup_ioctx(ctx_id); + if (unlikely(!ctx)) { + pr_debug("EINVAL: invalid context id\n"); + return -EINVAL; } - return 0; -} -#define MAX_AIO_SUBMITS (PAGE_SIZE/sizeof(struct iocb *)) + if (nr > ctx->nr_events) + nr = ctx->nr_events; -COMPAT_SYSCALL_DEFINE3(io_submit, compat_aio_context_t, ctx_id, - int, nr, u32 __user *, iocb) -{ - struct iocb __user * __user *iocb64; - long ret; + blk_start_plug(&plug); + for (i = 0; i < nr; i++) { + compat_uptr_t user_iocb; - if (unlikely(nr < 0)) - return -EINVAL; + if (unlikely(get_user(user_iocb, iocbpp + i))) { + ret = -EFAULT; + break; + } - if (nr > MAX_AIO_SUBMITS) - nr = MAX_AIO_SUBMITS; + ret = io_submit_one(ctx, compat_ptr(user_iocb), true); + if (ret) + break; + } + blk_finish_plug(&plug); - iocb64 = compat_alloc_user_space(nr * sizeof(*iocb64)); - ret = copy_iocb(nr, iocb, iocb64); - if (!ret) - ret = do_io_submit(ctx_id, nr, iocb64, 1); - return ret; + percpu_ref_put(&ctx->users); + return i ? i : ret; } #endif @@ -1756,15 +1914,12 @@ COMPAT_SYSCALL_DEFINE3(io_submit, compat_aio_context_t, ctx_id, * Finds a given iocb for cancellation. */ static struct aio_kiocb * -lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb, u32 key) +lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb) { struct aio_kiocb *kiocb; assert_spin_locked(&ctx->ctx_lock); - if (key != KIOCB_KEY) - return NULL; - /* TODO: use a hash or array, this sucks. */ list_for_each_entry(kiocb, &ctx->active_reqs, ki_list) { if (kiocb->ki_user_iocb == iocb) @@ -1788,25 +1943,24 @@ SYSCALL_DEFINE3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb, { struct kioctx *ctx; struct aio_kiocb *kiocb; + int ret = -EINVAL; u32 key; - int ret; - ret = get_user(key, &iocb->aio_key); - if (unlikely(ret)) + if (unlikely(get_user(key, &iocb->aio_key))) return -EFAULT; + if (unlikely(key != KIOCB_KEY)) + return -EINVAL; ctx = lookup_ioctx(ctx_id); if (unlikely(!ctx)) return -EINVAL; spin_lock_irq(&ctx->ctx_lock); - - kiocb = lookup_kiocb(ctx, iocb, key); - if (kiocb) - ret = kiocb_cancel(kiocb); - else - ret = -EINVAL; - + kiocb = lookup_kiocb(ctx, iocb); + if (kiocb) { + ret = kiocb->ki_cancel(&kiocb->rw); + list_del_init(&kiocb->ki_list); + } spin_unlock_irq(&ctx->ctx_lock); if (!ret) { @@ -1861,13 +2015,60 @@ SYSCALL_DEFINE5(io_getevents, aio_context_t, ctx_id, struct timespec __user *, timeout) { struct timespec64 ts; + int ret; + + if (timeout && unlikely(get_timespec64(&ts, timeout))) + return -EFAULT; + + ret = do_io_getevents(ctx_id, min_nr, nr, events, timeout ? &ts : NULL); + if (!ret && signal_pending(current)) + ret = -EINTR; + return ret; +} + +SYSCALL_DEFINE6(io_pgetevents, + aio_context_t, ctx_id, + long, min_nr, + long, nr, + struct io_event __user *, events, + struct timespec __user *, timeout, + const struct __aio_sigset __user *, usig) +{ + struct __aio_sigset ksig = { NULL, }; + sigset_t ksigmask, sigsaved; + struct timespec64 ts; + int ret; + + if (timeout && unlikely(get_timespec64(&ts, timeout))) + return -EFAULT; + + if (usig && copy_from_user(&ksig, usig, sizeof(ksig))) + return -EFAULT; - if (timeout) { - if (unlikely(get_timespec64(&ts, timeout))) + if (ksig.sigmask) { + if (ksig.sigsetsize != sizeof(sigset_t)) + return -EINVAL; + if (copy_from_user(&ksigmask, ksig.sigmask, sizeof(ksigmask))) return -EFAULT; + sigdelsetmask(&ksigmask, sigmask(SIGKILL) | sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + ret = do_io_getevents(ctx_id, min_nr, nr, events, timeout ? &ts : NULL); + if (signal_pending(current)) { + if (ksig.sigmask) { + current->saved_sigmask = sigsaved; + set_restore_sigmask(); + } + + if (!ret) + ret = -ERESTARTNOHAND; + } else { + if (ksig.sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); } - return do_io_getevents(ctx_id, min_nr, nr, events, timeout ? &ts : NULL); + return ret; } #ifdef CONFIG_COMPAT @@ -1878,13 +2079,64 @@ COMPAT_SYSCALL_DEFINE5(io_getevents, compat_aio_context_t, ctx_id, struct compat_timespec __user *, timeout) { struct timespec64 t; + int ret; + + if (timeout && compat_get_timespec64(&t, timeout)) + return -EFAULT; + + ret = do_io_getevents(ctx_id, min_nr, nr, events, timeout ? &t : NULL); + if (!ret && signal_pending(current)) + ret = -EINTR; + return ret; +} + + +struct __compat_aio_sigset { + compat_sigset_t __user *sigmask; + compat_size_t sigsetsize; +}; + +COMPAT_SYSCALL_DEFINE6(io_pgetevents, + compat_aio_context_t, ctx_id, + compat_long_t, min_nr, + compat_long_t, nr, + struct io_event __user *, events, + struct compat_timespec __user *, timeout, + const struct __compat_aio_sigset __user *, usig) +{ + struct __compat_aio_sigset ksig = { NULL, }; + sigset_t ksigmask, sigsaved; + struct timespec64 t; + int ret; + + if (timeout && compat_get_timespec64(&t, timeout)) + return -EFAULT; + + if (usig && copy_from_user(&ksig, usig, sizeof(ksig))) + return -EFAULT; - if (timeout) { - if (compat_get_timespec64(&t, timeout)) + if (ksig.sigmask) { + if (ksig.sigsetsize != sizeof(compat_sigset_t)) + return -EINVAL; + if (get_compat_sigset(&ksigmask, ksig.sigmask)) return -EFAULT; + sigdelsetmask(&ksigmask, sigmask(SIGKILL) | sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + ret = do_io_getevents(ctx_id, min_nr, nr, events, timeout ? &t : NULL); + if (signal_pending(current)) { + if (ksig.sigmask) { + current->saved_sigmask = sigsaved; + set_restore_sigmask(); + } + if (!ret) + ret = -ERESTARTNOHAND; + } else { + if (ksig.sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); } - return do_io_getevents(ctx_id, min_nr, nr, events, timeout ? &t : NULL); + return ret; } #endif diff --git a/fs/attr.c b/fs/attr.c index 12ffdb6fb63c..d0b4d34878fb 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -18,6 +18,32 @@ #include <linux/evm.h> #include <linux/ima.h> +static bool chown_ok(const struct inode *inode, kuid_t uid) +{ + if (uid_eq(current_fsuid(), inode->i_uid) && + uid_eq(uid, inode->i_uid)) + return true; + if (capable_wrt_inode_uidgid(inode, CAP_CHOWN)) + return true; + if (uid_eq(inode->i_uid, INVALID_UID) && + ns_capable(inode->i_sb->s_user_ns, CAP_CHOWN)) + return true; + return false; +} + +static bool chgrp_ok(const struct inode *inode, kgid_t gid) +{ + if (uid_eq(current_fsuid(), inode->i_uid) && + (in_group_p(gid) || gid_eq(gid, inode->i_gid))) + return true; + if (capable_wrt_inode_uidgid(inode, CAP_CHOWN)) + return true; + if (gid_eq(inode->i_gid, INVALID_GID) && + ns_capable(inode->i_sb->s_user_ns, CAP_CHOWN)) + return true; + return false; +} + /** * setattr_prepare - check if attribute changes to a dentry are allowed * @dentry: dentry to check @@ -52,17 +78,11 @@ int setattr_prepare(struct dentry *dentry, struct iattr *attr) goto kill_priv; /* Make sure a caller can chown. */ - if ((ia_valid & ATTR_UID) && - (!uid_eq(current_fsuid(), inode->i_uid) || - !uid_eq(attr->ia_uid, inode->i_uid)) && - !capable_wrt_inode_uidgid(inode, CAP_CHOWN)) + if ((ia_valid & ATTR_UID) && !chown_ok(inode, attr->ia_uid)) return -EPERM; /* Make sure caller can chgrp. */ - if ((ia_valid & ATTR_GID) && - (!uid_eq(current_fsuid(), inode->i_uid) || - (!in_group_p(attr->ia_gid) && !gid_eq(attr->ia_gid, inode->i_gid))) && - !capable_wrt_inode_uidgid(inode, CAP_CHOWN)) + if ((ia_valid & ATTR_GID) && !chgrp_ok(inode, attr->ia_gid)) return -EPERM; /* Make sure a caller can chmod. */ diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 82e8f6edfb48..b12e37f27530 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -749,7 +749,7 @@ static int autofs4_dir_mkdir(struct inode *dir, autofs4_del_active(dentry); - inode = autofs4_get_inode(dir->i_sb, S_IFDIR | 0555); + inode = autofs4_get_inode(dir->i_sb, S_IFDIR | mode); if (!inode) return -ENOMEM; d_add(dentry, inode); diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index a0c57c37fa21..be9c3dc048ab 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -19,9 +19,6 @@ */ static autofs_wqt_t autofs4_next_wait_queue = 1; -/* These are the signals we allow interrupting a pending mount */ -#define SHUTDOWN_SIGS (sigmask(SIGKILL) | sigmask(SIGINT) | sigmask(SIGQUIT)) - void autofs4_catatonic_mode(struct autofs_sb_info *sbi) { struct autofs_wait_queue *wq, *nwq; @@ -486,29 +483,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, * wq->name.name is NULL iff the lock is already released * or the mount has been made catatonic. */ - if (wq->name.name) { - /* Block all but "shutdown" signals while waiting */ - unsigned long shutdown_sigs_mask; - unsigned long irqflags; - sigset_t oldset; - - spin_lock_irqsave(¤t->sighand->siglock, irqflags); - oldset = current->blocked; - shutdown_sigs_mask = SHUTDOWN_SIGS & ~oldset.sig[0]; - siginitsetinv(¤t->blocked, shutdown_sigs_mask); - recalc_sigpending(); - spin_unlock_irqrestore(¤t->sighand->siglock, irqflags); - - wait_event_interruptible(wq->queue, wq->name.name == NULL); - - spin_lock_irqsave(¤t->sighand->siglock, irqflags); - current->blocked = oldset; - recalc_sigpending(); - spin_unlock_irqrestore(¤t->sighand->siglock, irqflags); - } else { - pr_debug("skipped sleeping\n"); - } - + wait_event_killable(wq->queue, wq->name.name == NULL); status = wq->status; /* @@ -574,7 +549,7 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok kfree(wq->name.name); wq->name.name = NULL; /* Do not wait on this queue */ wq->status = status; - wake_up_interruptible(&wq->queue); + wake_up(&wq->queue); if (!--wq->wait_ctr) kfree(wq); mutex_unlock(&sbi->wq_mutex); diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index af2832aaeec5..4700b4534439 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -198,23 +198,16 @@ befs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) if (ret == BEFS_BT_NOT_FOUND) { befs_debug(sb, "<--- %s %pd not found", __func__, dentry); - d_add(dentry, NULL); - return ERR_PTR(-ENOENT); - + inode = NULL; } else if (ret != BEFS_OK || offset == 0) { befs_error(sb, "<--- %s Error", __func__); - return ERR_PTR(-ENODATA); + inode = ERR_PTR(-ENODATA); + } else { + inode = befs_iget(dir->i_sb, (ino_t) offset); } - - inode = befs_iget(dir->i_sb, (ino_t) offset); - if (IS_ERR(inode)) - return ERR_CAST(inode); - - d_add(dentry, inode); - befs_debug(sb, "<--- %s", __func__); - return NULL; + return d_splice_alias(inode, dentry); } static int diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index ee832ca5f734..f32f21c3bbc7 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -21,10 +21,9 @@ #define dprintf(x...) #endif -static int bfs_add_entry(struct inode *dir, const unsigned char *name, - int namelen, int ino); +static int bfs_add_entry(struct inode *dir, const struct qstr *child, int ino); static struct buffer_head *bfs_find_entry(struct inode *dir, - const unsigned char *name, int namelen, + const struct qstr *child, struct bfs_dirent **res_dir); static int bfs_readdir(struct file *f, struct dir_context *ctx) @@ -111,8 +110,7 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, mark_inode_dirty(inode); bfs_dump_imap("create", s); - err = bfs_add_entry(dir, dentry->d_name.name, dentry->d_name.len, - inode->i_ino); + err = bfs_add_entry(dir, &dentry->d_name, inode->i_ino); if (err) { inode_dec_link_count(inode); mutex_unlock(&info->bfs_lock); @@ -136,19 +134,14 @@ static struct dentry *bfs_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(-ENAMETOOLONG); mutex_lock(&info->bfs_lock); - bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de); + bh = bfs_find_entry(dir, &dentry->d_name, &de); if (bh) { unsigned long ino = (unsigned long)le16_to_cpu(de->ino); brelse(bh); inode = bfs_iget(dir->i_sb, ino); - if (IS_ERR(inode)) { - mutex_unlock(&info->bfs_lock); - return ERR_CAST(inode); - } } mutex_unlock(&info->bfs_lock); - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } static int bfs_link(struct dentry *old, struct inode *dir, @@ -159,8 +152,7 @@ static int bfs_link(struct dentry *old, struct inode *dir, int err; mutex_lock(&info->bfs_lock); - err = bfs_add_entry(dir, new->d_name.name, new->d_name.len, - inode->i_ino); + err = bfs_add_entry(dir, &new->d_name, inode->i_ino); if (err) { mutex_unlock(&info->bfs_lock); return err; @@ -183,7 +175,7 @@ static int bfs_unlink(struct inode *dir, struct dentry *dentry) struct bfs_sb_info *info = BFS_SB(inode->i_sb); mutex_lock(&info->bfs_lock); - bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de); + bh = bfs_find_entry(dir, &dentry->d_name, &de); if (!bh || (le16_to_cpu(de->ino) != inode->i_ino)) goto out_brelse; @@ -228,27 +220,21 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, info = BFS_SB(old_inode->i_sb); mutex_lock(&info->bfs_lock); - old_bh = bfs_find_entry(old_dir, - old_dentry->d_name.name, - old_dentry->d_name.len, &old_de); + old_bh = bfs_find_entry(old_dir, &old_dentry->d_name, &old_de); if (!old_bh || (le16_to_cpu(old_de->ino) != old_inode->i_ino)) goto end_rename; error = -EPERM; new_inode = d_inode(new_dentry); - new_bh = bfs_find_entry(new_dir, - new_dentry->d_name.name, - new_dentry->d_name.len, &new_de); + new_bh = bfs_find_entry(new_dir, &new_dentry->d_name, &new_de); if (new_bh && !new_inode) { brelse(new_bh); new_bh = NULL; } if (!new_bh) { - error = bfs_add_entry(new_dir, - new_dentry->d_name.name, - new_dentry->d_name.len, + error = bfs_add_entry(new_dir, &new_dentry->d_name, old_inode->i_ino); if (error) goto end_rename; @@ -278,9 +264,10 @@ const struct inode_operations bfs_dir_inops = { .rename = bfs_rename, }; -static int bfs_add_entry(struct inode *dir, const unsigned char *name, - int namelen, int ino) +static int bfs_add_entry(struct inode *dir, const struct qstr *child, int ino) { + const unsigned char *name = child->name; + int namelen = child->len; struct buffer_head *bh; struct bfs_dirent *de; int block, sblock, eblock, off, pos; @@ -332,12 +319,14 @@ static inline int bfs_namecmp(int len, const unsigned char *name, } static struct buffer_head *bfs_find_entry(struct inode *dir, - const unsigned char *name, int namelen, + const struct qstr *child, struct bfs_dirent **res_dir) { unsigned long block = 0, offset = 0; struct buffer_head *bh = NULL; struct bfs_dirent *de; + const unsigned char *name = child->name; + int namelen = child->len; *res_dir = NULL; if (namelen > BFS_NAMELEN) diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index ce1824f47ba6..c3deb2e35f20 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -330,6 +330,7 @@ beyond_if: #ifdef __alpha__ regs->gp = ex.a_gpvalue; #endif + finalize_exec(bprm); start_thread(regs, ex.a_entry, current->mm->start_stack); return 0; } diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index bdb201230bae..4ad6f669fe34 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -377,6 +377,11 @@ static unsigned long elf_map(struct file *filep, unsigned long addr, } else map_addr = vm_mmap(filep, addr, size, prot, type, off); + if ((type & MAP_FIXED_NOREPLACE) && + PTR_ERR((void *)map_addr) == -EEXIST) + pr_info("%d (%s): Uhuuh, elf segment at %px requested but the memory is mapped already\n", + task_pid_nr(current), current->comm, (void *)addr); + return(map_addr); } @@ -575,7 +580,7 @@ static unsigned long load_elf_interp(struct elfhdr *interp_elf_ex, elf_prot |= PROT_EXEC; vaddr = eppnt->p_vaddr; if (interp_elf_ex->e_type == ET_EXEC || load_addr_set) - elf_type |= MAP_FIXED; + elf_type |= MAP_FIXED_NOREPLACE; else if (no_base && interp_elf_ex->e_type == ET_DYN) load_addr = -vaddr; @@ -890,7 +895,7 @@ static int load_elf_binary(struct linux_binprm *bprm) the correct location in memory. */ for(i = 0, elf_ppnt = elf_phdata; i < loc->elf_ex.e_phnum; i++, elf_ppnt++) { - int elf_prot = 0, elf_flags; + int elf_prot = 0, elf_flags, elf_fixed = MAP_FIXED_NOREPLACE; unsigned long k, vaddr; unsigned long total_size = 0; @@ -922,6 +927,13 @@ static int load_elf_binary(struct linux_binprm *bprm) */ } } + + /* + * Some binaries have overlapping elf segments and then + * we have to forcefully map over an existing mapping + * e.g. over this newly established brk mapping. + */ + elf_fixed = MAP_FIXED; } if (elf_ppnt->p_flags & PF_R) @@ -939,7 +951,7 @@ static int load_elf_binary(struct linux_binprm *bprm) * the ET_DYN load_addr calculations, proceed normally. */ if (loc->elf_ex.e_type == ET_EXEC || load_addr_set) { - elf_flags |= MAP_FIXED; + elf_flags |= elf_fixed; } else if (loc->elf_ex.e_type == ET_DYN) { /* * This logic is run once for the first LOAD Program @@ -975,7 +987,7 @@ static int load_elf_binary(struct linux_binprm *bprm) load_bias = ELF_ET_DYN_BASE; if (current->flags & PF_RANDOMIZE) load_bias += arch_mmap_rnd(); - elf_flags |= MAP_FIXED; + elf_flags |= elf_fixed; } else load_bias = 0; @@ -1155,6 +1167,7 @@ static int load_elf_binary(struct linux_binprm *bprm) ELF_PLAT_INIT(regs, reloc_func_desc); #endif + finalize_exec(bprm); start_thread(regs, elf_entry, bprm->p); retval = 0; out: @@ -1234,7 +1247,7 @@ static int load_elf_library(struct file *file) (eppnt->p_filesz + ELF_PAGEOFFSET(eppnt->p_vaddr)), PROT_READ | PROT_WRITE | PROT_EXEC, - MAP_FIXED | MAP_PRIVATE | MAP_DENYWRITE, + MAP_FIXED_NOREPLACE | MAP_PRIVATE | MAP_DENYWRITE, (eppnt->p_offset - ELF_PAGEOFFSET(eppnt->p_vaddr))); if (error != ELF_PAGESTART(eppnt->p_vaddr)) diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 429326b6e2e7..d90993adeffa 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -463,6 +463,7 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm) dynaddr); #endif + finalize_exec(bprm); /* everything is now ready... get the userspace context ready to roll */ entryaddr = interp_params.entry_addr ?: exec_params.entry_addr; start_thread(regs, entryaddr, current->mm->start_stack); diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 5d6b94475f27..82a48e830018 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -994,6 +994,7 @@ static int load_flat_binary(struct linux_binprm *bprm) FLAT_PLAT_INIT(regs); #endif + finalize_exec(bprm); pr_debug("start_thread(regs=0x%p, entry=0x%lx, start_stack=0x%lx)\n", regs, start_addr, current->mm->start_stack); start_thread(regs, start_addr, current->mm->start_stack); diff --git a/fs/block_dev.c b/fs/block_dev.c index fe09ef9c21f3..bef6934b6189 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -272,7 +272,7 @@ struct blkdev_dio { struct bio bio; }; -static struct bio_set *blkdev_dio_pool __read_mostly; +static struct bio_set blkdev_dio_pool; static void blkdev_bio_end_io(struct bio *bio) { @@ -334,7 +334,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) (bdev_logical_block_size(bdev) - 1)) return -EINVAL; - bio = bio_alloc_bioset(GFP_KERNEL, nr_pages, blkdev_dio_pool); + bio = bio_alloc_bioset(GFP_KERNEL, nr_pages, &blkdev_dio_pool); bio_get(bio); /* extra ref for the completion handler */ dio = container_of(bio, struct blkdev_dio, bio); @@ -432,10 +432,7 @@ blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter) static __init int blkdev_init(void) { - blkdev_dio_pool = bioset_create(4, offsetof(struct blkdev_dio, bio), BIOSET_NEED_BVECS); - if (!blkdev_dio_pool) - return -ENOMEM; - return 0; + return bioset_init(&blkdev_dio_pool, 4, offsetof(struct blkdev_dio, bio), BIOSET_NEED_BVECS); } module_init(blkdev_init); @@ -1322,25 +1319,30 @@ static void flush_disk(struct block_device *bdev, bool kill_dirty) * check_disk_size_change - checks for disk size change and adjusts bdev size. * @disk: struct gendisk to check * @bdev: struct bdev to adjust. + * @verbose: if %true log a message about a size change if there is any * * This routine checks to see if the bdev size does not match the disk size - * and adjusts it if it differs. + * and adjusts it if it differs. When shrinking the bdev size, its all caches + * are freed. */ -void check_disk_size_change(struct gendisk *disk, struct block_device *bdev) +void check_disk_size_change(struct gendisk *disk, struct block_device *bdev, + bool verbose) { loff_t disk_size, bdev_size; disk_size = (loff_t)get_capacity(disk) << 9; bdev_size = i_size_read(bdev->bd_inode); if (disk_size != bdev_size) { - printk(KERN_INFO - "%s: detected capacity change from %lld to %lld\n", - disk->disk_name, bdev_size, disk_size); + if (verbose) { + printk(KERN_INFO + "%s: detected capacity change from %lld to %lld\n", + disk->disk_name, bdev_size, disk_size); + } i_size_write(bdev->bd_inode, disk_size); - flush_disk(bdev, false); + if (bdev_size > disk_size) + flush_disk(bdev, false); } } -EXPORT_SYMBOL(check_disk_size_change); /** * revalidate_disk - wrapper for lower-level driver's revalidate_disk call-back @@ -1362,7 +1364,7 @@ int revalidate_disk(struct gendisk *disk) return ret; mutex_lock(&bdev->bd_mutex); - check_disk_size_change(disk, bdev); + check_disk_size_change(disk, bdev, ret == 0); bdev->bd_invalidated = 0; mutex_unlock(&bdev->bd_mutex); bdput(bdev); @@ -1946,11 +1948,6 @@ static int blkdev_releasepage(struct page *page, gfp_t wait) static int blkdev_writepages(struct address_space *mapping, struct writeback_control *wbc) { - if (dax_mapping(mapping)) { - struct block_device *bdev = I_BDEV(mapping->host); - - return dax_writeback_mapping_range(mapping, bdev, wbc); - } return generic_writepages(mapping, wbc); } diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig index 167e5dc7eadd..23537bc8c827 100644 --- a/fs/btrfs/Kconfig +++ b/fs/btrfs/Kconfig @@ -1,3 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0 + config BTRFS_FS tristate "Btrfs filesystem support" select LIBCRC32C diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c index 0066d95b133f..15e1dfef56a5 100644 --- a/fs/btrfs/acl.c +++ b/fs/btrfs/acl.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Red Hat. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/fs.h> diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c index d5540749f0e5..d522494698fa 100644 --- a/fs/btrfs/async-thread.c +++ b/fs/btrfs/async-thread.c @@ -1,20 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. * Copyright (C) 2014 Fujitsu. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/kthread.h> diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h index fc957e00cef1..7861c9feba5f 100644 --- a/fs/btrfs/async-thread.h +++ b/fs/btrfs/async-thread.h @@ -1,24 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. * Copyright (C) 2014 Fujitsu. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_ASYNC_THREAD_ -#define __BTRFS_ASYNC_THREAD_ +#ifndef BTRFS_ASYNC_THREAD_H +#define BTRFS_ASYNC_THREAD_H + #include <linux/workqueue.h> struct btrfs_fs_info; @@ -85,4 +73,5 @@ void btrfs_set_work_high_priority(struct btrfs_work *work); struct btrfs_fs_info *btrfs_work_owner(const struct btrfs_work *work); struct btrfs_fs_info *btrfs_workqueue_owner(const struct __btrfs_workqueue *wq); bool btrfs_workqueue_normal_congested(const struct btrfs_workqueue *wq); + #endif diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 571024bc632e..0a8e2e29a66b 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2011 STRATO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/mm.h> diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h index 0a30028d5196..54d58988483a 100644 --- a/fs/btrfs/backref.h +++ b/fs/btrfs/backref.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2011 STRATO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_BACKREF__ -#define __BTRFS_BACKREF__ +#ifndef BTRFS_BACKREF_H +#define BTRFS_BACKREF_H #include <linux/btrfs.h> #include "ulist.h" diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index ca15be569d69..7e075343daa5 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_I__ -#define __BTRFS_I__ +#ifndef BTRFS_INODE_H +#define BTRFS_INODE_H #include <linux/hash.h> #include "extent_map.h" @@ -32,17 +19,17 @@ * ordered operations list so that we make sure to flush out any * new data the application may have written before commit. */ -#define BTRFS_INODE_ORDERED_DATA_CLOSE 0 -#define BTRFS_INODE_ORPHAN_META_RESERVED 1 -#define BTRFS_INODE_DUMMY 2 -#define BTRFS_INODE_IN_DEFRAG 3 -#define BTRFS_INODE_HAS_ORPHAN_ITEM 4 -#define BTRFS_INODE_HAS_ASYNC_EXTENT 5 -#define BTRFS_INODE_NEEDS_FULL_SYNC 6 -#define BTRFS_INODE_COPY_EVERYTHING 7 -#define BTRFS_INODE_IN_DELALLOC_LIST 8 -#define BTRFS_INODE_READDIO_NEED_LOCK 9 -#define BTRFS_INODE_HAS_PROPS 10 +enum { + BTRFS_INODE_ORDERED_DATA_CLOSE = 0, + BTRFS_INODE_DUMMY, + BTRFS_INODE_IN_DEFRAG, + BTRFS_INODE_HAS_ASYNC_EXTENT, + BTRFS_INODE_NEEDS_FULL_SYNC, + BTRFS_INODE_COPY_EVERYTHING, + BTRFS_INODE_IN_DELALLOC_LIST, + BTRFS_INODE_READDIO_NEED_LOCK, + BTRFS_INODE_HAS_PROPS, +}; /* in memory btrfs inode */ struct btrfs_inode { diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index 3baebbc021c5..dc062b195c46 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) STRATO AG 2011. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ /* diff --git a/fs/btrfs/check-integrity.h b/fs/btrfs/check-integrity.h index 2de58a99ee92..9bf4359cc44c 100644 --- a/fs/btrfs/check-integrity.h +++ b/fs/btrfs/check-integrity.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) STRATO AG 2011. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#if !defined(__BTRFS_CHECK_INTEGRITY__) -#define __BTRFS_CHECK_INTEGRITY__ +#ifndef BTRFS_CHECK_INTEGRITY_H +#define BTRFS_CHECK_INTEGRITY_H #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh); diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 562c3e633403..d3e447b45bf7 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/kernel.h> @@ -458,7 +445,7 @@ static noinline int add_ra_bio_pages(struct inode *inode, break; rcu_read_lock(); - page = radix_tree_lookup(&mapping->page_tree, pg_index); + page = radix_tree_lookup(&mapping->i_pages, pg_index); rcu_read_unlock(); if (page && !radix_tree_exceptional_entry(page)) { misses++; @@ -1003,12 +990,7 @@ static void __free_workspace(int type, struct list_head *workspace, btrfs_compress_op[idx]->free_workspace(workspace); atomic_dec(total_ws); wake: - /* - * Make sure counter is updated before we wake up waiters. - */ - smp_mb(); - if (waitqueue_active(ws_wait)) - wake_up(ws_wait); + cond_wake_up(ws_wait); } static void free_workspace(int type, struct list_head *ws) diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h index ce796557a918..ddda9b80bf20 100644 --- a/fs/btrfs/compression.h +++ b/fs/btrfs/compression.h @@ -1,23 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_COMPRESSION_ -#define __BTRFS_COMPRESSION_ +#ifndef BTRFS_COMPRESSION_H +#define BTRFS_COMPRESSION_H + +#include <linux/sizes.h> /* * We want to make sure that amount of RAM required to uncompress an extent is diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index a2c9d21176e2..4bc326df472e 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007,2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> @@ -2343,7 +2330,7 @@ static noinline void unlock_up(struct btrfs_path *path, int level, no_skips = 1; t = path->nodes[i]; - if (i >= lowest_unlock && i > skip_level && path->locks[i]) { + if (i >= lowest_unlock && i > skip_level) { btrfs_tree_unlock_rw(t, path->locks[i]); path->locks[i] = 0; if (write_lock_level && @@ -2445,14 +2432,11 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, btrfs_unlock_up_safe(p, level + 1); btrfs_set_path_blocking(p); - free_extent_buffer(tmp); if (p->reada != READA_NONE) reada_for_search(fs_info, p, level, slot, key->objectid); - btrfs_release_path(p); - ret = -EAGAIN; - tmp = read_tree_block(fs_info, blocknr, 0, parent_level - 1, + tmp = read_tree_block(fs_info, blocknr, gen, parent_level - 1, &first_key); if (!IS_ERR(tmp)) { /* @@ -2461,12 +2445,14 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, * and give up so that our caller doesn't loop forever * on our EAGAINs. */ - if (!btrfs_buffer_uptodate(tmp, 0, 0)) + if (!extent_buffer_uptodate(tmp)) ret = -EIO; free_extent_buffer(tmp); } else { ret = PTR_ERR(tmp); } + + btrfs_release_path(p); return ret; } @@ -2612,6 +2598,78 @@ int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path, return 0; } +static struct extent_buffer *btrfs_search_slot_get_root(struct btrfs_root *root, + struct btrfs_path *p, + int write_lock_level) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct extent_buffer *b; + int root_lock; + int level = 0; + + /* We try very hard to do read locks on the root */ + root_lock = BTRFS_READ_LOCK; + + if (p->search_commit_root) { + /* The commit roots are read only so we always do read locks */ + if (p->need_commit_sem) + down_read(&fs_info->commit_root_sem); + b = root->commit_root; + extent_buffer_get(b); + level = btrfs_header_level(b); + if (p->need_commit_sem) + up_read(&fs_info->commit_root_sem); + /* + * Ensure that all callers have set skip_locking when + * p->search_commit_root = 1. + */ + ASSERT(p->skip_locking == 1); + + goto out; + } + + if (p->skip_locking) { + b = btrfs_root_node(root); + level = btrfs_header_level(b); + goto out; + } + + /* + * If the level is set to maximum, we can skip trying to get the read + * lock. + */ + if (write_lock_level < BTRFS_MAX_LEVEL) { + /* + * We don't know the level of the root node until we actually + * have it read locked + */ + b = btrfs_read_lock_root_node(root); + level = btrfs_header_level(b); + if (level > write_lock_level) + goto out; + + /* Whoops, must trade for write lock */ + btrfs_tree_read_unlock(b); + free_extent_buffer(b); + } + + b = btrfs_lock_root_node(root); + root_lock = BTRFS_WRITE_LOCK; + + /* The level might have changed, check again */ + level = btrfs_header_level(b); + +out: + p->nodes[level] = b; + if (!p->skip_locking) + p->locks[level] = root_lock; + /* + * Callers are responsible for dropping b's references. + */ + return b; +} + + /* * btrfs_search_slot - look for a key in a tree and perform necessary * modifications to preserve tree invariants. @@ -2648,7 +2706,6 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root, int err; int level; int lowest_unlock = 1; - int root_lock; /* everything at write_lock_level or lower must be write locked */ int write_lock_level = 0; u8 lowest_level = 0; @@ -2686,50 +2743,7 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root, again: prev_cmp = -1; - /* - * we try very hard to do read locks on the root - */ - root_lock = BTRFS_READ_LOCK; - level = 0; - if (p->search_commit_root) { - /* - * the commit roots are read only - * so we always do read locks - */ - if (p->need_commit_sem) - down_read(&fs_info->commit_root_sem); - b = root->commit_root; - extent_buffer_get(b); - level = btrfs_header_level(b); - if (p->need_commit_sem) - up_read(&fs_info->commit_root_sem); - if (!p->skip_locking) - btrfs_tree_read_lock(b); - } else { - if (p->skip_locking) { - b = btrfs_root_node(root); - level = btrfs_header_level(b); - } else { - /* we don't know the level of the root node - * until we actually have it read locked - */ - b = btrfs_read_lock_root_node(root); - level = btrfs_header_level(b); - if (level <= write_lock_level) { - /* whoops, must trade for write lock */ - btrfs_tree_read_unlock(b); - free_extent_buffer(b); - b = btrfs_lock_root_node(root); - root_lock = BTRFS_WRITE_LOCK; - - /* the level might have changed, check again */ - level = btrfs_header_level(b); - } - } - } - p->nodes[level] = b; - if (!p->skip_locking) - p->locks[level] = root_lock; + b = btrfs_search_slot_get_root(root, p, write_lock_level); while (b) { level = btrfs_header_level(b); @@ -5427,12 +5441,24 @@ int btrfs_compare_trees(struct btrfs_root *left_root, down_read(&fs_info->commit_root_sem); left_level = btrfs_header_level(left_root->commit_root); left_root_level = left_level; - left_path->nodes[left_level] = left_root->commit_root; + left_path->nodes[left_level] = + btrfs_clone_extent_buffer(left_root->commit_root); + if (!left_path->nodes[left_level]) { + up_read(&fs_info->commit_root_sem); + ret = -ENOMEM; + goto out; + } extent_buffer_get(left_path->nodes[left_level]); right_level = btrfs_header_level(right_root->commit_root); right_root_level = right_level; - right_path->nodes[right_level] = right_root->commit_root; + right_path->nodes[right_level] = + btrfs_clone_extent_buffer(right_root->commit_root); + if (!right_path->nodes[right_level]) { + up_read(&fs_info->commit_root_sem); + ret = -ENOMEM; + goto out; + } extent_buffer_get(right_path->nodes[right_level]); up_read(&fs_info->commit_root_sem); diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 0eb55825862a..f4bf7874c24a 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_CTREE__ -#define __BTRFS_CTREE__ +#ifndef BTRFS_CTREE_H +#define BTRFS_CTREE_H #include <linux/mm.h> #include <linux/sched/signal.h> @@ -472,6 +459,25 @@ struct btrfs_block_rsv { unsigned short full; unsigned short type; unsigned short failfast; + + /* + * Qgroup equivalent for @size @reserved + * + * Unlike normal @size/@reserved for inode rsv, qgroup doesn't care + * about things like csum size nor how many tree blocks it will need to + * reserve. + * + * Qgroup cares more about net change of the extent usage. + * + * So for one newly inserted file extent, in worst case it will cause + * leaf split and level increase, nodesize for each file extent is + * already too much. + * + * In short, qgroup_size/reserved is the upper limit of possible needed + * qgroup metadata reservation. + */ + u64 qgroup_rsv_size; + u64 qgroup_rsv_reserved; }; /* @@ -727,6 +733,18 @@ struct btrfs_delayed_root; */ #define BTRFS_FS_EXCL_OP 16 +/* + * To info transaction_kthread we need an immediate commit so it doesn't + * need to wait for commit_interval + */ +#define BTRFS_FS_NEED_ASYNC_COMMIT 17 + +/* + * Indicate that balance has been set up from the ioctl and is in the main + * phase. The fs_info::balance_ctl is initialized. + */ +#define BTRFS_FS_BALANCE_RUNNING 18 + struct btrfs_fs_info { u8 fsid[BTRFS_FSID_SIZE]; u8 chunk_tree_uuid[BTRFS_UUID_SIZE]; @@ -826,7 +844,6 @@ struct btrfs_fs_info { struct mutex transaction_kthread_mutex; struct mutex cleaner_mutex; struct mutex chunk_mutex; - struct mutex volume_mutex; /* * this is taken to make sure we don't set block groups ro after @@ -992,7 +1009,6 @@ struct btrfs_fs_info { /* restriper state */ spinlock_t balance_lock; struct mutex balance_mutex; - atomic_t balance_running; atomic_t balance_pause_req; atomic_t balance_cancel_req; struct btrfs_balance_control *balance_ctl; @@ -1207,9 +1223,6 @@ struct btrfs_root { spinlock_t log_extents_lock[2]; struct list_head logged_list[2]; - spinlock_t orphan_lock; - atomic_t orphan_inodes; - struct btrfs_block_rsv *orphan_block_rsv; int orphan_cleanup_state; spinlock_t inode_lock; @@ -2752,13 +2765,9 @@ void btrfs_delalloc_release_space(struct inode *inode, void btrfs_free_reserved_data_space_noquota(struct inode *inode, u64 start, u64 len); void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans); -int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans, - struct btrfs_inode *inode); -void btrfs_orphan_release_metadata(struct btrfs_inode *inode); int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, struct btrfs_block_rsv *rsv, - int nitems, - u64 *qgroup_reserved, bool use_global_rsv); + int nitems, bool use_global_rsv); void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *rsv); void btrfs_delalloc_release_extents(struct btrfs_inode *inode, u64 num_bytes, @@ -2816,7 +2825,7 @@ void btrfs_wait_for_snapshot_creation(struct btrfs_root *root); void check_system_chunk(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, const u64 type); u64 add_new_free_space(struct btrfs_block_group_cache *block_group, - struct btrfs_fs_info *info, u64 start, u64 end); + u64 start, u64 end); /* ctree.c */ int btrfs_bin_search(struct extent_buffer *eb, const struct btrfs_key *key, @@ -3030,11 +3039,9 @@ void btrfs_update_root_times(struct btrfs_trans_handle *trans, struct btrfs_root *root); /* uuid-tree.c */ -int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u8 *uuid, u8 type, +int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type, u64 subid); -int btrfs_uuid_tree_rem(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u8 *uuid, u8 type, +int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type, u64 subid); int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info, int (*check_func)(struct btrfs_fs_info *, u8 *, u8, @@ -3151,18 +3158,6 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode, struct extent_map *em); /* inode.c */ -struct btrfs_delalloc_work { - struct inode *inode; - int delay_iput; - struct completion completion; - struct list_head list; - struct btrfs_work work; -}; - -struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode, - int delay_iput); -void btrfs_wait_and_free_delalloc_work(struct btrfs_delalloc_work *work); - struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode, struct page *page, size_t pg_offset, u64 start, u64 len, int create); @@ -3170,6 +3165,8 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, u64 *orig_start, u64 *orig_block_len, u64 *ram_bytes); +void __btrfs_del_delalloc_inode(struct btrfs_root *root, + struct btrfs_inode *inode); struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry); int btrfs_set_inode_index(struct btrfs_inode *dir, u64 *index); int btrfs_unlink_inode(struct btrfs_trans_handle *trans, @@ -3179,10 +3176,7 @@ int btrfs_unlink_inode(struct btrfs_trans_handle *trans, int btrfs_add_link(struct btrfs_trans_handle *trans, struct btrfs_inode *parent_inode, struct btrfs_inode *inode, const char *name, int name_len, int add_backref, u64 index); -int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct inode *dir, u64 objectid, - const char *name, int name_len); +int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry); int btrfs_truncate_block(struct inode *inode, loff_t from, loff_t len, int front); int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, @@ -3190,9 +3184,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, struct inode *inode, u64 new_size, u32 min_type); -int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput); -int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput, - int nr); +int btrfs_start_delalloc_inodes(struct btrfs_root *root); +int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr); int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end, unsigned int extra_bits, struct extent_state **cached_state, int dedupe); @@ -3226,10 +3219,7 @@ int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct btrfs_inode *inode); int btrfs_orphan_cleanup(struct btrfs_root *root); -void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, - struct btrfs_root *root); int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size); -void btrfs_invalidate_inodes(struct btrfs_root *root); void btrfs_add_delayed_iput(struct inode *inode); void btrfs_run_delayed_iputs(struct btrfs_fs_info *fs_info); int btrfs_prealloc_file_range(struct inode *inode, int mode, @@ -3248,14 +3238,14 @@ void btrfs_test_inode_set_ops(struct inode *inode); long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg); long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg); int btrfs_ioctl_get_supported_features(void __user *arg); -void btrfs_update_iflags(struct inode *inode); +void btrfs_sync_inode_flags_to_i_flags(struct inode *inode); int btrfs_is_empty_uuid(u8 *uuid); int btrfs_defrag_file(struct inode *inode, struct file *file, struct btrfs_ioctl_defrag_range_args *range, u64 newer_than, unsigned long max_pages); void btrfs_get_block_group_info(struct list_head *groups_list, struct btrfs_ioctl_space_info *space); -void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock, +void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info, struct btrfs_ioctl_balance_args *bargs); ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen, struct file *dst_file, u64 dst_loff); @@ -3752,4 +3742,27 @@ static inline int btrfs_is_testing(struct btrfs_fs_info *fs_info) #endif return 0; } + +static inline void cond_wake_up(struct wait_queue_head *wq) +{ + /* + * This implies a full smp_mb barrier, see comments for + * waitqueue_active why. + */ + if (wq_has_sleeper(wq)) + wake_up(wq); +} + +static inline void cond_wake_up_nomb(struct wait_queue_head *wq) +{ + /* + * Special case for conditional wakeup where the barrier required for + * waitqueue_active is implied by some of the preceding code. Eg. one + * of such atomic operations (atomic_dec_and_return, ...), or a + * unlock/lock sequence, etc. + */ + if (waitqueue_active(wq)) + wake_up(wq); +} + #endif diff --git a/fs/btrfs/dedupe.h b/fs/btrfs/dedupe.h index 83ebfe28da9e..90281a7a35a8 100644 --- a/fs/btrfs/dedupe.h +++ b/fs/btrfs/dedupe.h @@ -1,24 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2016 Fujitsu. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_DEDUPE__ -#define __BTRFS_DEDUPE__ +#ifndef BTRFS_DEDUPE_H +#define BTRFS_DEDUPE_H /* later in-band dedupe will expand this struct */ struct btrfs_dedupe_hash; + #endif diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 86ec2edc05e8..fe6caa7e698b 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1,20 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2011 Fujitsu. All rights reserved. * Written by Miao Xie <miaox@cn.fujitsu.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/slab.h> @@ -473,13 +460,10 @@ static void finish_one_item(struct btrfs_delayed_root *delayed_root) { int seq = atomic_inc_return(&delayed_root->items_seq); - /* - * atomic_dec_return implies a barrier for waitqueue_active - */ + /* atomic_dec_return implies a barrier */ if ((atomic_dec_return(&delayed_root->items) < - BTRFS_DELAYED_BACKGROUND || seq % BTRFS_DELAYED_BATCH == 0) && - waitqueue_active(&delayed_root->wait)) - wake_up(&delayed_root->wait); + BTRFS_DELAYED_BACKGROUND || seq % BTRFS_DELAYED_BATCH == 0)) + cond_wake_up_nomb(&delayed_root->wait); } static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item) @@ -569,6 +553,12 @@ static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans, dst_rsv = &fs_info->delayed_block_rsv; num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1); + + /* + * Here we migrate space rsv from transaction rsv, since have already + * reserved space when starting a transaction. So no need to reserve + * qgroup space here. + */ ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1); if (!ret) { trace_btrfs_space_reservation(fs_info, "delayed_item", @@ -590,7 +580,10 @@ static void btrfs_delayed_item_release_metadata(struct btrfs_root *root, return; rsv = &fs_info->delayed_block_rsv; - btrfs_qgroup_convert_reserved_meta(root, item->bytes_reserved); + /* + * Check btrfs_delayed_item_reserve_metadata() to see why we don't need + * to release/reserve qgroup space. + */ trace_btrfs_space_reservation(fs_info, "delayed_item", item->key.objectid, item->bytes_reserved, 0); @@ -615,9 +608,6 @@ static int btrfs_delayed_inode_reserve_metadata( num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1); - ret = btrfs_qgroup_reserve_meta_prealloc(root, num_bytes, true); - if (ret < 0) - return ret; /* * btrfs_dirty_inode will update the inode under btrfs_join_transaction * which doesn't reserve space for speed. This is a problem since we @@ -629,6 +619,10 @@ static int btrfs_delayed_inode_reserve_metadata( */ if (!src_rsv || (!trans->bytes_reserved && src_rsv->type != BTRFS_BLOCK_RSV_DELALLOC)) { + ret = btrfs_qgroup_reserve_meta_prealloc(root, + fs_info->nodesize, true); + if (ret < 0) + return ret; ret = btrfs_block_rsv_add(root, dst_rsv, num_bytes, BTRFS_RESERVE_NO_FLUSH); /* @@ -647,6 +641,8 @@ static int btrfs_delayed_inode_reserve_metadata( "delayed_inode", btrfs_ino(inode), num_bytes, 1); + } else { + btrfs_qgroup_free_meta_prealloc(root, fs_info->nodesize); } return ret; } diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h index 100a91e26b55..ca7a97f3ab6b 100644 --- a/fs/btrfs/delayed-inode.h +++ b/fs/btrfs/delayed-inode.h @@ -1,24 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2011 Fujitsu. All rights reserved. * Written by Miao Xie <miaox@cn.fujitsu.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __DELAYED_TREE_OPERATION_H -#define __DELAYED_TREE_OPERATION_H +#ifndef BTRFS_DELAYED_INODE_H +#define BTRFS_DELAYED_INODE_H #include <linux/rbtree.h> #include <linux/spinlock.h> diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 2677257c149d..03dec673d12a 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2009 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> @@ -299,10 +286,10 @@ static bool merge_ref(struct btrfs_trans_handle *trans, } void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_delayed_ref_root *delayed_refs, struct btrfs_delayed_ref_head *head) { + struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_delayed_ref_node *ref; struct rb_node *node; u64 seq = 0; @@ -336,9 +323,7 @@ again: } } -int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, - struct btrfs_delayed_ref_root *delayed_refs, - u64 seq) +int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq) { struct seq_list *elem; int ret = 0; @@ -349,10 +334,9 @@ int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, struct seq_list, list); if (seq >= elem->seq) { btrfs_debug(fs_info, - "holding back delayed_ref %#x.%x, lowest is %#x.%x (%p)", + "holding back delayed_ref %#x.%x, lowest is %#x.%x", (u32)(seq >> 32), (u32)seq, - (u32)(elem->seq >> 32), (u32)elem->seq, - delayed_refs); + (u32)(elem->seq >> 32), (u32)elem->seq); ret = 1; } } @@ -542,31 +526,20 @@ update_existing_head_ref(struct btrfs_delayed_ref_root *delayed_refs, spin_unlock(&existing->lock); } -/* - * helper function to actually insert a head node into the rbtree. - * this does all the dirty work in terms of maintaining the correct - * overall modification count. - */ -static noinline struct btrfs_delayed_ref_head * -add_delayed_ref_head(struct btrfs_fs_info *fs_info, - struct btrfs_trans_handle *trans, - struct btrfs_delayed_ref_head *head_ref, - struct btrfs_qgroup_extent_record *qrecord, - u64 bytenr, u64 num_bytes, u64 ref_root, u64 reserved, - int action, int is_data, int *qrecord_inserted_ret, - int *old_ref_mod, int *new_ref_mod) +static void init_delayed_ref_head(struct btrfs_delayed_ref_head *head_ref, + struct btrfs_qgroup_extent_record *qrecord, + u64 bytenr, u64 num_bytes, u64 ref_root, + u64 reserved, int action, bool is_data, + bool is_system) { - struct btrfs_delayed_ref_head *existing; - struct btrfs_delayed_ref_root *delayed_refs; int count_mod = 1; int must_insert_reserved = 0; - int qrecord_inserted = 0; /* If reserved is provided, it must be a data extent. */ BUG_ON(!is_data && reserved); /* - * the head node stores the sum of all the mods, so dropping a ref + * The head node stores the sum of all the mods, so dropping a ref * should drop the sum in the head node by one. */ if (action == BTRFS_UPDATE_DELAYED_HEAD) @@ -575,12 +548,11 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, count_mod = -1; /* - * BTRFS_ADD_DELAYED_EXTENT means that we need to update - * the reserved accounting when the extent is finally added, or - * if a later modification deletes the delayed ref without ever - * inserting the extent into the extent allocation tree. - * ref->must_insert_reserved is the flag used to record - * that accounting mods are required. + * BTRFS_ADD_DELAYED_EXTENT means that we need to update the reserved + * accounting when the extent is finally added, or if a later + * modification deletes the delayed ref without ever inserting the + * extent into the extent allocation tree. ref->must_insert_reserved + * is the flag used to record that accounting mods are required. * * Once we record must_insert_reserved, switch the action to * BTRFS_ADD_DELAYED_REF because other special casing is not required. @@ -590,14 +562,13 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, else must_insert_reserved = 0; - delayed_refs = &trans->transaction->delayed_refs; - refcount_set(&head_ref->refs, 1); head_ref->bytenr = bytenr; head_ref->num_bytes = num_bytes; head_ref->ref_mod = count_mod; head_ref->must_insert_reserved = must_insert_reserved; head_ref->is_data = is_data; + head_ref->is_system = is_system; head_ref->ref_tree = RB_ROOT; INIT_LIST_HEAD(&head_ref->ref_add_list); RB_CLEAR_NODE(&head_ref->href_node); @@ -608,7 +579,6 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, spin_lock_init(&head_ref->lock); mutex_init(&head_ref->mutex); - /* Record qgroup extent info if provided */ if (qrecord) { if (ref_root && reserved) { head_ref->qgroup_ref_root = ref_root; @@ -618,20 +588,44 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, qrecord->bytenr = bytenr; qrecord->num_bytes = num_bytes; qrecord->old_roots = NULL; + } +} + +/* + * helper function to actually insert a head node into the rbtree. + * this does all the dirty work in terms of maintaining the correct + * overall modification count. + */ +static noinline struct btrfs_delayed_ref_head * +add_delayed_ref_head(struct btrfs_trans_handle *trans, + struct btrfs_delayed_ref_head *head_ref, + struct btrfs_qgroup_extent_record *qrecord, + int action, int *qrecord_inserted_ret, + int *old_ref_mod, int *new_ref_mod) +{ + struct btrfs_delayed_ref_head *existing; + struct btrfs_delayed_ref_root *delayed_refs; + int qrecord_inserted = 0; + + delayed_refs = &trans->transaction->delayed_refs; - if(btrfs_qgroup_trace_extent_nolock(fs_info, + /* Record qgroup extent info if provided */ + if (qrecord) { + if (btrfs_qgroup_trace_extent_nolock(trans->fs_info, delayed_refs, qrecord)) kfree(qrecord); else qrecord_inserted = 1; } - trace_add_delayed_ref_head(fs_info, head_ref, action); + trace_add_delayed_ref_head(trans->fs_info, head_ref, action); existing = htree_insert(&delayed_refs->href_root, &head_ref->href_node); if (existing) { - WARN_ON(ref_root && reserved && existing->qgroup_ref_root + WARN_ON(qrecord && head_ref->qgroup_ref_root + && head_ref->qgroup_reserved + && existing->qgroup_ref_root && existing->qgroup_reserved); update_existing_head_ref(delayed_refs, existing, head_ref, old_ref_mod); @@ -644,8 +638,8 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, } else { if (old_ref_mod) *old_ref_mod = 0; - if (is_data && count_mod < 0) - delayed_refs->pending_csums += num_bytes; + if (head_ref->is_data && head_ref->ref_mod < 0) + delayed_refs->pending_csums += head_ref->num_bytes; delayed_refs->num_heads++; delayed_refs->num_heads_ready++; atomic_inc(&delayed_refs->num_entries); @@ -655,90 +649,48 @@ add_delayed_ref_head(struct btrfs_fs_info *fs_info, *qrecord_inserted_ret = qrecord_inserted; if (new_ref_mod) *new_ref_mod = head_ref->total_ref_mod; - return head_ref; -} - -/* - * helper to insert a delayed tree ref into the rbtree. - */ -static noinline void -add_delayed_tree_ref(struct btrfs_fs_info *fs_info, - struct btrfs_trans_handle *trans, - struct btrfs_delayed_ref_head *head_ref, - struct btrfs_delayed_ref_node *ref, u64 bytenr, - u64 num_bytes, u64 parent, u64 ref_root, int level, - int action) -{ - struct btrfs_delayed_tree_ref *full_ref; - struct btrfs_delayed_ref_root *delayed_refs; - u64 seq = 0; - int ret; - - if (action == BTRFS_ADD_DELAYED_EXTENT) - action = BTRFS_ADD_DELAYED_REF; - - if (is_fstree(ref_root)) - seq = atomic64_read(&fs_info->tree_mod_seq); - delayed_refs = &trans->transaction->delayed_refs; - /* first set the basic ref node struct up */ - refcount_set(&ref->refs, 1); - ref->bytenr = bytenr; - ref->num_bytes = num_bytes; - ref->ref_mod = 1; - ref->action = action; - ref->is_head = 0; - ref->in_tree = 1; - ref->seq = seq; - RB_CLEAR_NODE(&ref->ref_node); - INIT_LIST_HEAD(&ref->add_list); - - full_ref = btrfs_delayed_node_to_tree_ref(ref); - full_ref->parent = parent; - full_ref->root = ref_root; - if (parent) - ref->type = BTRFS_SHARED_BLOCK_REF_KEY; - else - ref->type = BTRFS_TREE_BLOCK_REF_KEY; - full_ref->level = level; - - trace_add_delayed_tree_ref(fs_info, ref, full_ref, action); - - ret = insert_delayed_ref(trans, delayed_refs, head_ref, ref); - - /* - * XXX: memory should be freed at the same level allocated. - * But bad practice is anywhere... Follow it now. Need cleanup. - */ - if (ret > 0) - kmem_cache_free(btrfs_delayed_tree_ref_cachep, full_ref); + return head_ref; } /* - * helper to insert a delayed data ref into the rbtree. + * init_delayed_ref_common - Initialize the structure which represents a + * modification to a an extent. + * + * @fs_info: Internal to the mounted filesystem mount structure. + * + * @ref: The structure which is going to be initialized. + * + * @bytenr: The logical address of the extent for which a modification is + * going to be recorded. + * + * @num_bytes: Size of the extent whose modification is being recorded. + * + * @ref_root: The id of the root where this modification has originated, this + * can be either one of the well-known metadata trees or the + * subvolume id which references this extent. + * + * @action: Can be one of BTRFS_ADD_DELAYED_REF/BTRFS_DROP_DELAYED_REF or + * BTRFS_ADD_DELAYED_EXTENT + * + * @ref_type: Holds the type of the extent which is being recorded, can be + * one of BTRFS_SHARED_BLOCK_REF_KEY/BTRFS_TREE_BLOCK_REF_KEY + * when recording a metadata extent or BTRFS_SHARED_DATA_REF_KEY/ + * BTRFS_EXTENT_DATA_REF_KEY when recording data extent */ -static noinline void -add_delayed_data_ref(struct btrfs_fs_info *fs_info, - struct btrfs_trans_handle *trans, - struct btrfs_delayed_ref_head *head_ref, - struct btrfs_delayed_ref_node *ref, u64 bytenr, - u64 num_bytes, u64 parent, u64 ref_root, u64 owner, - u64 offset, int action) +static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, + struct btrfs_delayed_ref_node *ref, + u64 bytenr, u64 num_bytes, u64 ref_root, + int action, u8 ref_type) { - struct btrfs_delayed_data_ref *full_ref; - struct btrfs_delayed_ref_root *delayed_refs; u64 seq = 0; - int ret; if (action == BTRFS_ADD_DELAYED_EXTENT) action = BTRFS_ADD_DELAYED_REF; - delayed_refs = &trans->transaction->delayed_refs; - if (is_fstree(ref_root)) seq = atomic64_read(&fs_info->tree_mod_seq); - /* first set the basic ref node struct up */ refcount_set(&ref->refs, 1); ref->bytenr = bytenr; ref->num_bytes = num_bytes; @@ -747,25 +699,9 @@ add_delayed_data_ref(struct btrfs_fs_info *fs_info, ref->is_head = 0; ref->in_tree = 1; ref->seq = seq; + ref->type = ref_type; RB_CLEAR_NODE(&ref->ref_node); INIT_LIST_HEAD(&ref->add_list); - - full_ref = btrfs_delayed_node_to_data_ref(ref); - full_ref->parent = parent; - full_ref->root = ref_root; - if (parent) - ref->type = BTRFS_SHARED_DATA_REF_KEY; - else - ref->type = BTRFS_EXTENT_DATA_REF_KEY; - - full_ref->objectid = owner; - full_ref->offset = offset; - - trace_add_delayed_data_ref(fs_info, ref, full_ref, action); - - ret = insert_delayed_ref(trans, delayed_refs, head_ref, ref); - if (ret > 0) - kmem_cache_free(btrfs_delayed_data_ref_cachep, full_ref); } /* @@ -785,12 +721,25 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info, struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_qgroup_extent_record *record = NULL; int qrecord_inserted; + bool is_system = (ref_root == BTRFS_CHUNK_TREE_OBJECTID); + int ret; + u8 ref_type; BUG_ON(extent_op && extent_op->is_data); ref = kmem_cache_alloc(btrfs_delayed_tree_ref_cachep, GFP_NOFS); if (!ref) return -ENOMEM; + if (parent) + ref_type = BTRFS_SHARED_BLOCK_REF_KEY; + else + ref_type = BTRFS_TREE_BLOCK_REF_KEY; + init_delayed_ref_common(fs_info, &ref->node, bytenr, num_bytes, + ref_root, action, ref_type); + ref->root = ref_root; + ref->parent = parent; + ref->level = level; + head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); if (!head_ref) goto free_ref; @@ -802,6 +751,8 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info, goto free_head_ref; } + init_delayed_ref_head(head_ref, record, bytenr, num_bytes, + ref_root, 0, action, false, is_system); head_ref->extent_op = extent_op; delayed_refs = &trans->transaction->delayed_refs; @@ -811,15 +762,19 @@ int btrfs_add_delayed_tree_ref(struct btrfs_fs_info *fs_info, * insert both the head node and the new ref without dropping * the spin lock */ - head_ref = add_delayed_ref_head(fs_info, trans, head_ref, record, - bytenr, num_bytes, 0, 0, action, 0, - &qrecord_inserted, old_ref_mod, - new_ref_mod); + head_ref = add_delayed_ref_head(trans, head_ref, record, + action, &qrecord_inserted, + old_ref_mod, new_ref_mod); - add_delayed_tree_ref(fs_info, trans, head_ref, &ref->node, bytenr, - num_bytes, parent, ref_root, level, action); + ret = insert_delayed_ref(trans, delayed_refs, head_ref, &ref->node); spin_unlock(&delayed_refs->lock); + trace_add_delayed_tree_ref(fs_info, &ref->node, ref, + action == BTRFS_ADD_DELAYED_EXTENT ? + BTRFS_ADD_DELAYED_REF : action); + if (ret > 0) + kmem_cache_free(btrfs_delayed_tree_ref_cachep, ref); + if (qrecord_inserted) btrfs_qgroup_trace_extent_post(fs_info, record); @@ -848,11 +803,25 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info, struct btrfs_delayed_ref_root *delayed_refs; struct btrfs_qgroup_extent_record *record = NULL; int qrecord_inserted; + int ret; + u8 ref_type; ref = kmem_cache_alloc(btrfs_delayed_data_ref_cachep, GFP_NOFS); if (!ref) return -ENOMEM; + if (parent) + ref_type = BTRFS_SHARED_DATA_REF_KEY; + else + ref_type = BTRFS_EXTENT_DATA_REF_KEY; + init_delayed_ref_common(fs_info, &ref->node, bytenr, num_bytes, + ref_root, action, ref_type); + ref->root = ref_root; + ref->parent = parent; + ref->objectid = owner; + ref->offset = offset; + + head_ref = kmem_cache_alloc(btrfs_delayed_ref_head_cachep, GFP_NOFS); if (!head_ref) { kmem_cache_free(btrfs_delayed_data_ref_cachep, ref); @@ -870,6 +839,8 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info, } } + init_delayed_ref_head(head_ref, record, bytenr, num_bytes, ref_root, + reserved, action, true, false); head_ref->extent_op = NULL; delayed_refs = &trans->transaction->delayed_refs; @@ -879,16 +850,20 @@ int btrfs_add_delayed_data_ref(struct btrfs_fs_info *fs_info, * insert both the head node and the new ref without dropping * the spin lock */ - head_ref = add_delayed_ref_head(fs_info, trans, head_ref, record, - bytenr, num_bytes, ref_root, reserved, - action, 1, &qrecord_inserted, + head_ref = add_delayed_ref_head(trans, head_ref, record, + action, &qrecord_inserted, old_ref_mod, new_ref_mod); - add_delayed_data_ref(fs_info, trans, head_ref, &ref->node, bytenr, - num_bytes, parent, ref_root, owner, offset, - action); + ret = insert_delayed_ref(trans, delayed_refs, head_ref, &ref->node); spin_unlock(&delayed_refs->lock); + trace_add_delayed_data_ref(trans->fs_info, &ref->node, ref, + action == BTRFS_ADD_DELAYED_EXTENT ? + BTRFS_ADD_DELAYED_REF : action); + if (ret > 0) + kmem_cache_free(btrfs_delayed_data_ref_cachep, ref); + + if (qrecord_inserted) return btrfs_qgroup_trace_extent_post(fs_info, record); return 0; @@ -906,14 +881,16 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info, if (!head_ref) return -ENOMEM; + init_delayed_ref_head(head_ref, NULL, bytenr, num_bytes, 0, 0, + BTRFS_UPDATE_DELAYED_HEAD, extent_op->is_data, + false); head_ref->extent_op = extent_op; delayed_refs = &trans->transaction->delayed_refs; spin_lock(&delayed_refs->lock); - add_delayed_ref_head(fs_info, trans, head_ref, NULL, bytenr, - num_bytes, 0, 0, BTRFS_UPDATE_DELAYED_HEAD, - extent_op->is_data, NULL, NULL, NULL); + add_delayed_ref_head(trans, head_ref, NULL, BTRFS_UPDATE_DELAYED_HEAD, + NULL, NULL, NULL); spin_unlock(&delayed_refs->lock); return 0; diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 9e3e5aff0937..ea1aecb6a50d 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -1,22 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __DELAYED_REF__ -#define __DELAYED_REF__ + +#ifndef BTRFS_DELAYED_REF_H +#define BTRFS_DELAYED_REF_H #include <linux/refcount.h> @@ -139,6 +127,7 @@ struct btrfs_delayed_ref_head { */ unsigned int must_insert_reserved:1; unsigned int is_data:1; + unsigned int is_system:1; unsigned int processing:1; }; @@ -262,7 +251,6 @@ int btrfs_add_delayed_extent_op(struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes, struct btrfs_delayed_extent_op *extent_op); void btrfs_merge_delayed_refs(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_delayed_ref_root *delayed_refs, struct btrfs_delayed_ref_head *head); @@ -280,9 +268,7 @@ static inline void btrfs_delayed_ref_unlock(struct btrfs_delayed_ref_head *head) struct btrfs_delayed_ref_head * btrfs_select_ref_head(struct btrfs_trans_handle *trans); -int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, - struct btrfs_delayed_ref_root *delayed_refs, - u64 seq); +int btrfs_check_delayed_seq(struct btrfs_fs_info *fs_info, u64 seq); /* * helper functions to cast a node into its container @@ -298,4 +284,5 @@ btrfs_delayed_node_to_data_ref(struct btrfs_delayed_ref_node *node) { return container_of(node, struct btrfs_delayed_data_ref, node); } + #endif diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index 0d203633bb96..e2ba0419297a 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -1,20 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) STRATO AG 2012. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ + #include <linux/sched.h> #include <linux/bio.h> #include <linux/slab.h> @@ -45,8 +33,6 @@ static void btrfs_dev_replace_update_device_in_mapping_tree( struct btrfs_device *srcdev, struct btrfs_device *tgtdev); static int btrfs_dev_replace_kthread(void *data); -static int btrfs_dev_replace_continue_on_mount(struct btrfs_fs_info *fs_info); - int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info) { @@ -191,6 +177,105 @@ out: } /* + * Initialize a new device for device replace target from a given source dev + * and path. + * + * Return 0 and new device in @device_out, otherwise return < 0 + */ +static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info, + const char *device_path, + struct btrfs_device *srcdev, + struct btrfs_device **device_out) +{ + struct btrfs_device *device; + struct block_device *bdev; + struct list_head *devices; + struct rcu_string *name; + u64 devid = BTRFS_DEV_REPLACE_DEVID; + int ret = 0; + + *device_out = NULL; + if (fs_info->fs_devices->seeding) { + btrfs_err(fs_info, "the filesystem is a seed filesystem!"); + return -EINVAL; + } + + bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL, + fs_info->bdev_holder); + if (IS_ERR(bdev)) { + btrfs_err(fs_info, "target device %s is invalid!", device_path); + return PTR_ERR(bdev); + } + + filemap_write_and_wait(bdev->bd_inode->i_mapping); + + devices = &fs_info->fs_devices->devices; + list_for_each_entry(device, devices, dev_list) { + if (device->bdev == bdev) { + btrfs_err(fs_info, + "target device is in the filesystem!"); + ret = -EEXIST; + goto error; + } + } + + + if (i_size_read(bdev->bd_inode) < + btrfs_device_get_total_bytes(srcdev)) { + btrfs_err(fs_info, + "target device is smaller than source device!"); + ret = -EINVAL; + goto error; + } + + + device = btrfs_alloc_device(NULL, &devid, NULL); + if (IS_ERR(device)) { + ret = PTR_ERR(device); + goto error; + } + + name = rcu_string_strdup(device_path, GFP_KERNEL); + if (!name) { + btrfs_free_device(device); + ret = -ENOMEM; + goto error; + } + rcu_assign_pointer(device->name, name); + + mutex_lock(&fs_info->fs_devices->device_list_mutex); + set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state); + device->generation = 0; + device->io_width = fs_info->sectorsize; + device->io_align = fs_info->sectorsize; + device->sector_size = fs_info->sectorsize; + device->total_bytes = btrfs_device_get_total_bytes(srcdev); + device->disk_total_bytes = btrfs_device_get_disk_total_bytes(srcdev); + device->bytes_used = btrfs_device_get_bytes_used(srcdev); + device->commit_total_bytes = srcdev->commit_total_bytes; + device->commit_bytes_used = device->bytes_used; + device->fs_info = fs_info; + device->bdev = bdev; + set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); + set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state); + device->mode = FMODE_EXCL; + device->dev_stats_valid = 1; + set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE); + device->fs_devices = fs_info->fs_devices; + list_add(&device->dev_list, &fs_info->fs_devices->devices); + fs_info->fs_devices->num_devices++; + fs_info->fs_devices->open_devices++; + mutex_unlock(&fs_info->fs_devices->device_list_mutex); + + *device_out = device; + return 0; + +error: + blkdev_put(bdev, FMODE_EXCL); + return ret; +} + +/* * called from commit_transaction. Writes changed device replace state to * disk. */ @@ -329,18 +414,13 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, struct btrfs_device *tgt_device = NULL; struct btrfs_device *src_device = NULL; - /* the disk copy procedure reuses the scrub code */ - mutex_lock(&fs_info->volume_mutex); ret = btrfs_find_device_by_devspec(fs_info, srcdevid, srcdev_name, &src_device); - if (ret) { - mutex_unlock(&fs_info->volume_mutex); + if (ret) return ret; - } ret = btrfs_init_dev_replace_tgtdev(fs_info, tgtdev_name, src_device, &tgt_device); - mutex_unlock(&fs_info->volume_mutex); if (ret) return ret; @@ -372,7 +452,6 @@ int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, dev_replace->cont_reading_from_srcdev_mode = read_src; WARN_ON(!src_device); dev_replace->srcdev = src_device; - WARN_ON(!tgt_device); dev_replace->tgtdev = tgt_device; btrfs_info_in_rcu(fs_info, @@ -515,7 +594,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, * flush all outstanding I/O and inode extent mappings before the * copy operation is declared as being finished */ - ret = btrfs_start_delalloc_roots(fs_info, 0, -1); + ret = btrfs_start_delalloc_roots(fs_info, -1); if (ret) { mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); return ret; @@ -530,7 +609,6 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, ret = btrfs_commit_transaction(trans); WARN_ON(ret); - mutex_lock(&uuid_mutex); /* keep away write_all_supers() during the finishing procedure */ mutex_lock(&fs_info->fs_devices->device_list_mutex); mutex_lock(&fs_info->chunk_mutex); @@ -557,7 +635,6 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, btrfs_dev_replace_write_unlock(dev_replace); mutex_unlock(&fs_info->chunk_mutex); mutex_unlock(&fs_info->fs_devices->device_list_mutex); - mutex_unlock(&uuid_mutex); btrfs_rm_dev_replace_blocked(fs_info); if (tgt_device) btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device); @@ -608,7 +685,6 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, */ mutex_unlock(&fs_info->chunk_mutex); mutex_unlock(&fs_info->fs_devices->device_list_mutex); - mutex_unlock(&uuid_mutex); /* replace the sysfs entry */ btrfs_sysfs_rm_device_link(fs_info->fs_devices, src_device); @@ -812,7 +888,17 @@ int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info) } btrfs_dev_replace_write_unlock(dev_replace); - WARN_ON(test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)); + /* + * This could collide with a paused balance, but the exclusive op logic + * should never allow both to start and pause. We don't want to allow + * dev-replace to start anyway. + */ + if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) { + btrfs_info(fs_info, + "cannot resume dev-replace, other exclusive operation running"); + return 0; + } + task = kthread_run(btrfs_dev_replace_kthread, fs_info, "btrfs-devrepl"); return PTR_ERR_OR_ZERO(task); } @@ -822,6 +908,7 @@ static int btrfs_dev_replace_kthread(void *data) struct btrfs_fs_info *fs_info = data; struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; u64 progress; + int ret; progress = btrfs_dev_replace_progress(fs_info); progress = div_u64(progress, 10); @@ -832,23 +919,14 @@ static int btrfs_dev_replace_kthread(void *data) btrfs_dev_name(dev_replace->tgtdev), (unsigned int)progress); - btrfs_dev_replace_continue_on_mount(fs_info); - clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); - - return 0; -} - -static int btrfs_dev_replace_continue_on_mount(struct btrfs_fs_info *fs_info) -{ - struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; - int ret; - ret = btrfs_scrub_dev(fs_info, dev_replace->srcdev->devid, dev_replace->committed_cursor_left, btrfs_device_get_total_bytes(dev_replace->srcdev), &dev_replace->scrub_progress, 0, 1); ret = btrfs_dev_replace_finishing(fs_info, ret); WARN_ON(ret); + + clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); return 0; } @@ -928,9 +1006,9 @@ void btrfs_dev_replace_clear_lock_blocking( ASSERT(atomic_read(&dev_replace->read_locks) > 0); ASSERT(atomic_read(&dev_replace->blocking_readers) > 0); read_lock(&dev_replace->lock); - if (atomic_dec_and_test(&dev_replace->blocking_readers) && - waitqueue_active(&dev_replace->read_lock_wq)) - wake_up(&dev_replace->read_lock_wq); + /* Barrier implied by atomic_dec_and_test */ + if (atomic_dec_and_test(&dev_replace->blocking_readers)) + cond_wake_up_nomb(&dev_replace->read_lock_wq); } void btrfs_bio_counter_inc_noblocked(struct btrfs_fs_info *fs_info) @@ -941,9 +1019,7 @@ void btrfs_bio_counter_inc_noblocked(struct btrfs_fs_info *fs_info) void btrfs_bio_counter_sub(struct btrfs_fs_info *fs_info, s64 amount) { percpu_counter_sub(&fs_info->bio_counter, amount); - - if (waitqueue_active(&fs_info->replace_wait)) - wake_up(&fs_info->replace_wait); + cond_wake_up_nomb(&fs_info->replace_wait); } void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info) diff --git a/fs/btrfs/dev-replace.h b/fs/btrfs/dev-replace.h index 8566a02ef222..b6d4206188bb 100644 --- a/fs/btrfs/dev-replace.h +++ b/fs/btrfs/dev-replace.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) STRATO AG 2012. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#if !defined(__BTRFS_DEV_REPLACE__) -#define __BTRFS_DEV_REPLACE__ +#ifndef BTRFS_DEV_REPLACE_H +#define BTRFS_DEV_REPLACE_H struct btrfs_ioctl_dev_replace_args; @@ -48,4 +35,5 @@ static inline void btrfs_dev_replace_stats_inc(atomic64_t *stat_value) { atomic64_inc(stat_value); } + #endif diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c index 29e967b2c667..39e9766d1cbd 100644 --- a/fs/btrfs/dir-item.c +++ b/fs/btrfs/dir-item.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include "ctree.h" diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 07b5e6f7df67..205092dc9390 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/fs.h> @@ -68,7 +55,6 @@ static const struct extent_io_ops btree_extent_io_ops; static void end_workqueue_fn(struct btrfs_work *work); static void free_fs_root(struct btrfs_root *root); -static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info); static void btrfs_destroy_ordered_extents(struct btrfs_root *root); static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, struct btrfs_fs_info *fs_info); @@ -429,7 +415,7 @@ static int btrfs_check_super_csum(struct btrfs_fs_info *fs_info, static int verify_level_key(struct btrfs_fs_info *fs_info, struct extent_buffer *eb, int level, - struct btrfs_key *first_key) + struct btrfs_key *first_key, u64 parent_transid) { int found_level; struct btrfs_key found_key; @@ -449,6 +435,14 @@ static int verify_level_key(struct btrfs_fs_info *fs_info, if (!first_key) return 0; + /* + * For live tree block (new tree blocks in current transaction), + * we need proper lock context to avoid race, which is impossible here. + * So we only checks tree blocks which is read from disk, whose + * generation <= fs_info->last_trans_committed. + */ + if (btrfs_header_generation(eb) > fs_info->last_trans_committed) + return 0; if (found_level) btrfs_node_key_to_cpu(eb, &found_key, 0); else @@ -459,10 +453,11 @@ static int verify_level_key(struct btrfs_fs_info *fs_info, if (ret) { WARN_ON(1); btrfs_err(fs_info, -"tree first key mismatch detected, bytenr=%llu key expected=(%llu, %u, %llu) has=(%llu, %u, %llu)", - eb->start, first_key->objectid, first_key->type, - first_key->offset, found_key.objectid, - found_key.type, found_key.offset); +"tree first key mismatch detected, bytenr=%llu parent_transid=%llu key expected=(%llu,%u,%llu) has=(%llu,%u,%llu)", + eb->start, parent_transid, first_key->objectid, + first_key->type, first_key->offset, + found_key.objectid, found_key.type, + found_key.offset); } #endif return ret; @@ -498,7 +493,7 @@ static int btree_read_extent_buffer_pages(struct btrfs_fs_info *fs_info, parent_transid, 0)) ret = -EIO; else if (verify_level_key(fs_info, eb, level, - first_key)) + first_key, parent_transid)) ret = -EUCLEAN; else break; @@ -1190,7 +1185,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, root->inode_tree = RB_ROOT; INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC); root->block_rsv = NULL; - root->orphan_block_rsv = NULL; INIT_LIST_HEAD(&root->dirty_list); INIT_LIST_HEAD(&root->root_list); @@ -1200,7 +1194,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, INIT_LIST_HEAD(&root->ordered_root); INIT_LIST_HEAD(&root->logged_list[0]); INIT_LIST_HEAD(&root->logged_list[1]); - spin_lock_init(&root->orphan_lock); spin_lock_init(&root->inode_lock); spin_lock_init(&root->delalloc_lock); spin_lock_init(&root->ordered_extent_lock); @@ -1221,7 +1214,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, atomic_set(&root->log_commit[1], 0); atomic_set(&root->log_writers, 0); atomic_set(&root->log_batch, 0); - atomic_set(&root->orphan_inodes, 0); refcount_set(&root->refs, 1); atomic_set(&root->will_be_snapshotted, 0); root->log_transid = 0; @@ -1829,6 +1821,7 @@ static int transaction_kthread(void *arg) now = get_seconds(); if (cur->state < TRANS_STATE_BLOCKED && + !test_bit(BTRFS_FS_NEED_ASYNC_COMMIT, &fs_info->flags) && (now < cur->start_time || now - cur->start_time < fs_info->commit_interval)) { spin_unlock(&fs_info->trans_lock); @@ -2168,7 +2161,6 @@ static void btrfs_init_balance(struct btrfs_fs_info *fs_info) { spin_lock_init(&fs_info->balance_lock); mutex_init(&fs_info->balance_mutex); - atomic_set(&fs_info->balance_running, 0); atomic_set(&fs_info->balance_pause_req, 0); atomic_set(&fs_info->balance_cancel_req, 0); fs_info->balance_ctl = NULL; @@ -2446,6 +2438,211 @@ out: return ret; } +/* + * Real super block validation + * NOTE: super csum type and incompat features will not be checked here. + * + * @sb: super block to check + * @mirror_num: the super block number to check its bytenr: + * 0 the primary (1st) sb + * 1, 2 2nd and 3rd backup copy + * -1 skip bytenr check + */ +static int validate_super(struct btrfs_fs_info *fs_info, + struct btrfs_super_block *sb, int mirror_num) +{ + u64 nodesize = btrfs_super_nodesize(sb); + u64 sectorsize = btrfs_super_sectorsize(sb); + int ret = 0; + + if (btrfs_super_magic(sb) != BTRFS_MAGIC) { + btrfs_err(fs_info, "no valid FS found"); + ret = -EINVAL; + } + if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP) { + btrfs_err(fs_info, "unrecognized or unsupported super flag: %llu", + btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP); + ret = -EINVAL; + } + if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) { + btrfs_err(fs_info, "tree_root level too big: %d >= %d", + btrfs_super_root_level(sb), BTRFS_MAX_LEVEL); + ret = -EINVAL; + } + if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) { + btrfs_err(fs_info, "chunk_root level too big: %d >= %d", + btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL); + ret = -EINVAL; + } + if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) { + btrfs_err(fs_info, "log_root level too big: %d >= %d", + btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL); + ret = -EINVAL; + } + + /* + * Check sectorsize and nodesize first, other check will need it. + * Check all possible sectorsize(4K, 8K, 16K, 32K, 64K) here. + */ + if (!is_power_of_2(sectorsize) || sectorsize < 4096 || + sectorsize > BTRFS_MAX_METADATA_BLOCKSIZE) { + btrfs_err(fs_info, "invalid sectorsize %llu", sectorsize); + ret = -EINVAL; + } + /* Only PAGE SIZE is supported yet */ + if (sectorsize != PAGE_SIZE) { + btrfs_err(fs_info, + "sectorsize %llu not supported yet, only support %lu", + sectorsize, PAGE_SIZE); + ret = -EINVAL; + } + if (!is_power_of_2(nodesize) || nodesize < sectorsize || + nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) { + btrfs_err(fs_info, "invalid nodesize %llu", nodesize); + ret = -EINVAL; + } + if (nodesize != le32_to_cpu(sb->__unused_leafsize)) { + btrfs_err(fs_info, "invalid leafsize %u, should be %llu", + le32_to_cpu(sb->__unused_leafsize), nodesize); + ret = -EINVAL; + } + + /* Root alignment check */ + if (!IS_ALIGNED(btrfs_super_root(sb), sectorsize)) { + btrfs_warn(fs_info, "tree_root block unaligned: %llu", + btrfs_super_root(sb)); + ret = -EINVAL; + } + if (!IS_ALIGNED(btrfs_super_chunk_root(sb), sectorsize)) { + btrfs_warn(fs_info, "chunk_root block unaligned: %llu", + btrfs_super_chunk_root(sb)); + ret = -EINVAL; + } + if (!IS_ALIGNED(btrfs_super_log_root(sb), sectorsize)) { + btrfs_warn(fs_info, "log_root block unaligned: %llu", + btrfs_super_log_root(sb)); + ret = -EINVAL; + } + + if (memcmp(fs_info->fsid, sb->dev_item.fsid, BTRFS_FSID_SIZE) != 0) { + btrfs_err(fs_info, + "dev_item UUID does not match fsid: %pU != %pU", + fs_info->fsid, sb->dev_item.fsid); + ret = -EINVAL; + } + + /* + * Hint to catch really bogus numbers, bitflips or so, more exact checks are + * done later + */ + if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) { + btrfs_err(fs_info, "bytes_used is too small %llu", + btrfs_super_bytes_used(sb)); + ret = -EINVAL; + } + if (!is_power_of_2(btrfs_super_stripesize(sb))) { + btrfs_err(fs_info, "invalid stripesize %u", + btrfs_super_stripesize(sb)); + ret = -EINVAL; + } + if (btrfs_super_num_devices(sb) > (1UL << 31)) + btrfs_warn(fs_info, "suspicious number of devices: %llu", + btrfs_super_num_devices(sb)); + if (btrfs_super_num_devices(sb) == 0) { + btrfs_err(fs_info, "number of devices is 0"); + ret = -EINVAL; + } + + if (mirror_num >= 0 && + btrfs_super_bytenr(sb) != btrfs_sb_offset(mirror_num)) { + btrfs_err(fs_info, "super offset mismatch %llu != %u", + btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET); + ret = -EINVAL; + } + + /* + * Obvious sys_chunk_array corruptions, it must hold at least one key + * and one chunk + */ + if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) { + btrfs_err(fs_info, "system chunk array too big %u > %u", + btrfs_super_sys_array_size(sb), + BTRFS_SYSTEM_CHUNK_ARRAY_SIZE); + ret = -EINVAL; + } + if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key) + + sizeof(struct btrfs_chunk)) { + btrfs_err(fs_info, "system chunk array too small %u < %zu", + btrfs_super_sys_array_size(sb), + sizeof(struct btrfs_disk_key) + + sizeof(struct btrfs_chunk)); + ret = -EINVAL; + } + + /* + * The generation is a global counter, we'll trust it more than the others + * but it's still possible that it's the one that's wrong. + */ + if (btrfs_super_generation(sb) < btrfs_super_chunk_root_generation(sb)) + btrfs_warn(fs_info, + "suspicious: generation < chunk_root_generation: %llu < %llu", + btrfs_super_generation(sb), + btrfs_super_chunk_root_generation(sb)); + if (btrfs_super_generation(sb) < btrfs_super_cache_generation(sb) + && btrfs_super_cache_generation(sb) != (u64)-1) + btrfs_warn(fs_info, + "suspicious: generation < cache_generation: %llu < %llu", + btrfs_super_generation(sb), + btrfs_super_cache_generation(sb)); + + return ret; +} + +/* + * Validation of super block at mount time. + * Some checks already done early at mount time, like csum type and incompat + * flags will be skipped. + */ +static int btrfs_validate_mount_super(struct btrfs_fs_info *fs_info) +{ + return validate_super(fs_info, fs_info->super_copy, 0); +} + +/* + * Validation of super block at write time. + * Some checks like bytenr check will be skipped as their values will be + * overwritten soon. + * Extra checks like csum type and incompat flags will be done here. + */ +static int btrfs_validate_write_super(struct btrfs_fs_info *fs_info, + struct btrfs_super_block *sb) +{ + int ret; + + ret = validate_super(fs_info, sb, -1); + if (ret < 0) + goto out; + if (btrfs_super_csum_type(sb) != BTRFS_CSUM_TYPE_CRC32) { + ret = -EUCLEAN; + btrfs_err(fs_info, "invalid csum type, has %u want %u", + btrfs_super_csum_type(sb), BTRFS_CSUM_TYPE_CRC32); + goto out; + } + if (btrfs_super_incompat_flags(sb) & ~BTRFS_FEATURE_INCOMPAT_SUPP) { + ret = -EUCLEAN; + btrfs_err(fs_info, + "invalid incompat flags, has 0x%llx valid mask 0x%llx", + btrfs_super_incompat_flags(sb), + (unsigned long long)BTRFS_FEATURE_INCOMPAT_SUPP); + goto out; + } +out: + if (ret < 0) + btrfs_err(fs_info, + "super block corruption detected before writing it to disk"); + return ret; +} + int open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_devices, char *options) @@ -2605,7 +2802,6 @@ int open_ctree(struct super_block *sb, mutex_init(&fs_info->chunk_mutex); mutex_init(&fs_info->transaction_kthread_mutex); mutex_init(&fs_info->cleaner_mutex); - mutex_init(&fs_info->volume_mutex); mutex_init(&fs_info->ro_block_group_mutex); init_rwsem(&fs_info->commit_root_sem); init_rwsem(&fs_info->cleanup_work_sem); @@ -2672,7 +2868,7 @@ int open_ctree(struct super_block *sb, memcpy(fs_info->fsid, fs_info->super_copy->fsid, BTRFS_FSID_SIZE); - ret = btrfs_check_super_valid(fs_info); + ret = btrfs_validate_mount_super(fs_info); if (ret) { btrfs_err(fs_info, "superblock contains fatal errors"); err = -EINVAL; @@ -3527,7 +3723,7 @@ int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags) for (raid_type = 0; raid_type < BTRFS_NR_RAID_TYPES; raid_type++) { if (raid_type == BTRFS_RAID_SINGLE) continue; - if (!(flags & btrfs_raid_group[raid_type])) + if (!(flags & btrfs_raid_array[raid_type].bg_flag)) continue; min_tolerated = min(min_tolerated, btrfs_raid_array[raid_type]. @@ -3607,6 +3803,14 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) flags = btrfs_super_flags(sb); btrfs_set_super_flags(sb, flags | BTRFS_HEADER_FLAG_WRITTEN); + ret = btrfs_validate_write_super(fs_info, sb); + if (ret < 0) { + mutex_unlock(&fs_info->fs_devices->device_list_mutex); + btrfs_handle_fs_error(fs_info, -EUCLEAN, + "unexpected superblock corruption detected"); + return -EUCLEAN; + } + ret = write_dev_supers(dev, sb, max_mirrors); if (ret) total_errors++; @@ -3678,8 +3882,6 @@ static void free_fs_root(struct btrfs_root *root) { iput(root->ino_cache_inode); WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree)); - btrfs_free_block_rsv(root->fs_info, root->orphan_block_rsv); - root->orphan_block_rsv = NULL; if (root->anon_dev) free_anon_bdev(root->anon_dev); if (root->subv_writers) @@ -3770,7 +3972,6 @@ int btrfs_commit_super(struct btrfs_fs_info *fs_info) void close_ctree(struct btrfs_fs_info *fs_info) { - struct btrfs_root *root = fs_info->tree_root; int ret; set_bit(BTRFS_FS_CLOSING_START, &fs_info->flags); @@ -3812,7 +4013,8 @@ void close_ctree(struct btrfs_fs_info *fs_info) btrfs_err(fs_info, "commit super ret %d", ret); } - if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) + if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state) || + test_bit(BTRFS_FS_STATE_TRANS_ABORTED, &fs_info->fs_state)) btrfs_error_commit_super(fs_info); kthread_stop(fs_info->transaction_kthread); @@ -3821,6 +4023,7 @@ void close_ctree(struct btrfs_fs_info *fs_info) set_bit(BTRFS_FS_CLOSING_DONE, &fs_info->flags); btrfs_free_qgroup_config(fs_info); + ASSERT(list_empty(&fs_info->delalloc_roots)); if (percpu_counter_sum(&fs_info->delalloc_bytes)) { btrfs_info(fs_info, "at unmount delalloc count %lld", @@ -3864,9 +4067,6 @@ void close_ctree(struct btrfs_fs_info *fs_info) btrfs_free_stripe_hash_table(fs_info); btrfs_free_ref_cache(fs_info); - __btrfs_free_block_rsv(root->orphan_block_rsv); - root->orphan_block_rsv = NULL; - while (!list_empty(&fs_info->pinned_chunks)) { struct extent_map *em; @@ -3977,166 +4177,17 @@ int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid, int level, level, first_key); } -static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info) -{ - struct btrfs_super_block *sb = fs_info->super_copy; - u64 nodesize = btrfs_super_nodesize(sb); - u64 sectorsize = btrfs_super_sectorsize(sb); - int ret = 0; - - if (btrfs_super_magic(sb) != BTRFS_MAGIC) { - btrfs_err(fs_info, "no valid FS found"); - ret = -EINVAL; - } - if (btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP) { - btrfs_err(fs_info, "unrecognized or unsupported super flag: %llu", - btrfs_super_flags(sb) & ~BTRFS_SUPER_FLAG_SUPP); - ret = -EINVAL; - } - if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) { - btrfs_err(fs_info, "tree_root level too big: %d >= %d", - btrfs_super_root_level(sb), BTRFS_MAX_LEVEL); - ret = -EINVAL; - } - if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) { - btrfs_err(fs_info, "chunk_root level too big: %d >= %d", - btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL); - ret = -EINVAL; - } - if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) { - btrfs_err(fs_info, "log_root level too big: %d >= %d", - btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL); - ret = -EINVAL; - } - - /* - * Check sectorsize and nodesize first, other check will need it. - * Check all possible sectorsize(4K, 8K, 16K, 32K, 64K) here. - */ - if (!is_power_of_2(sectorsize) || sectorsize < 4096 || - sectorsize > BTRFS_MAX_METADATA_BLOCKSIZE) { - btrfs_err(fs_info, "invalid sectorsize %llu", sectorsize); - ret = -EINVAL; - } - /* Only PAGE SIZE is supported yet */ - if (sectorsize != PAGE_SIZE) { - btrfs_err(fs_info, - "sectorsize %llu not supported yet, only support %lu", - sectorsize, PAGE_SIZE); - ret = -EINVAL; - } - if (!is_power_of_2(nodesize) || nodesize < sectorsize || - nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) { - btrfs_err(fs_info, "invalid nodesize %llu", nodesize); - ret = -EINVAL; - } - if (nodesize != le32_to_cpu(sb->__unused_leafsize)) { - btrfs_err(fs_info, "invalid leafsize %u, should be %llu", - le32_to_cpu(sb->__unused_leafsize), nodesize); - ret = -EINVAL; - } - - /* Root alignment check */ - if (!IS_ALIGNED(btrfs_super_root(sb), sectorsize)) { - btrfs_warn(fs_info, "tree_root block unaligned: %llu", - btrfs_super_root(sb)); - ret = -EINVAL; - } - if (!IS_ALIGNED(btrfs_super_chunk_root(sb), sectorsize)) { - btrfs_warn(fs_info, "chunk_root block unaligned: %llu", - btrfs_super_chunk_root(sb)); - ret = -EINVAL; - } - if (!IS_ALIGNED(btrfs_super_log_root(sb), sectorsize)) { - btrfs_warn(fs_info, "log_root block unaligned: %llu", - btrfs_super_log_root(sb)); - ret = -EINVAL; - } - - if (memcmp(fs_info->fsid, sb->dev_item.fsid, BTRFS_FSID_SIZE) != 0) { - btrfs_err(fs_info, - "dev_item UUID does not match fsid: %pU != %pU", - fs_info->fsid, sb->dev_item.fsid); - ret = -EINVAL; - } - - /* - * Hint to catch really bogus numbers, bitflips or so, more exact checks are - * done later - */ - if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) { - btrfs_err(fs_info, "bytes_used is too small %llu", - btrfs_super_bytes_used(sb)); - ret = -EINVAL; - } - if (!is_power_of_2(btrfs_super_stripesize(sb))) { - btrfs_err(fs_info, "invalid stripesize %u", - btrfs_super_stripesize(sb)); - ret = -EINVAL; - } - if (btrfs_super_num_devices(sb) > (1UL << 31)) - btrfs_warn(fs_info, "suspicious number of devices: %llu", - btrfs_super_num_devices(sb)); - if (btrfs_super_num_devices(sb) == 0) { - btrfs_err(fs_info, "number of devices is 0"); - ret = -EINVAL; - } - - if (btrfs_super_bytenr(sb) != BTRFS_SUPER_INFO_OFFSET) { - btrfs_err(fs_info, "super offset mismatch %llu != %u", - btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET); - ret = -EINVAL; - } - - /* - * Obvious sys_chunk_array corruptions, it must hold at least one key - * and one chunk - */ - if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) { - btrfs_err(fs_info, "system chunk array too big %u > %u", - btrfs_super_sys_array_size(sb), - BTRFS_SYSTEM_CHUNK_ARRAY_SIZE); - ret = -EINVAL; - } - if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key) - + sizeof(struct btrfs_chunk)) { - btrfs_err(fs_info, "system chunk array too small %u < %zu", - btrfs_super_sys_array_size(sb), - sizeof(struct btrfs_disk_key) - + sizeof(struct btrfs_chunk)); - ret = -EINVAL; - } - - /* - * The generation is a global counter, we'll trust it more than the others - * but it's still possible that it's the one that's wrong. - */ - if (btrfs_super_generation(sb) < btrfs_super_chunk_root_generation(sb)) - btrfs_warn(fs_info, - "suspicious: generation < chunk_root_generation: %llu < %llu", - btrfs_super_generation(sb), - btrfs_super_chunk_root_generation(sb)); - if (btrfs_super_generation(sb) < btrfs_super_cache_generation(sb) - && btrfs_super_cache_generation(sb) != (u64)-1) - btrfs_warn(fs_info, - "suspicious: generation < cache_generation: %llu < %llu", - btrfs_super_generation(sb), - btrfs_super_cache_generation(sb)); - - return ret; -} - static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info) { + /* cleanup FS via transaction */ + btrfs_cleanup_transaction(fs_info); + mutex_lock(&fs_info->cleaner_mutex); btrfs_run_delayed_iputs(fs_info); mutex_unlock(&fs_info->cleaner_mutex); down_write(&fs_info->cleanup_work_sem); up_write(&fs_info->cleanup_work_sem); - - /* cleanup FS via transaction */ - btrfs_cleanup_transaction(fs_info); } static void btrfs_destroy_ordered_extents(struct btrfs_root *root) @@ -4261,19 +4312,23 @@ static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root) list_splice_init(&root->delalloc_inodes, &splice); while (!list_empty(&splice)) { + struct inode *inode = NULL; btrfs_inode = list_first_entry(&splice, struct btrfs_inode, delalloc_inodes); - - list_del_init(&btrfs_inode->delalloc_inodes); - clear_bit(BTRFS_INODE_IN_DELALLOC_LIST, - &btrfs_inode->runtime_flags); + __btrfs_del_delalloc_inode(root, btrfs_inode); spin_unlock(&root->delalloc_lock); - btrfs_invalidate_inodes(btrfs_inode->root); - + /* + * Make sure we get a live inode and that it'll not disappear + * meanwhile. + */ + inode = igrab(&btrfs_inode->vfs_inode); + if (inode) { + invalidate_inode_pages2(inode->i_mapping); + iput(inode); + } spin_lock(&root->delalloc_lock); } - spin_unlock(&root->delalloc_lock); } @@ -4289,7 +4344,6 @@ static void btrfs_destroy_all_delalloc_inodes(struct btrfs_fs_info *fs_info) while (!list_empty(&splice)) { root = list_first_entry(&splice, struct btrfs_root, delalloc_root); - list_del_init(&root->delalloc_root); root = btrfs_grab_fs_root(root); BUG_ON(!root); spin_unlock(&fs_info->delalloc_root_lock); diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index 453ea9f5d4e9..1a3d277b027b 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __DISKIO__ -#define __DISKIO__ +#ifndef BTRFS_DISK_IO_H +#define BTRFS_DISK_IO_H #define BTRFS_SUPER_INFO_OFFSET SZ_64K #define BTRFS_SUPER_INFO_SIZE 4096 @@ -169,4 +156,5 @@ static inline void btrfs_set_buffer_lockdep_class(u64 objectid, { } #endif + #endif diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c index ddaccad469f8..1f3755b3a37a 100644 --- a/fs/btrfs/export.c +++ b/fs/btrfs/export.c @@ -1,4 +1,5 @@ // SPDX-License-Identifier: GPL-2.0 + #include <linux/fs.h> #include <linux/types.h> #include "ctree.h" diff --git a/fs/btrfs/export.h b/fs/btrfs/export.h index 91b3908e7c54..57488ecd7d4e 100644 --- a/fs/btrfs/export.h +++ b/fs/btrfs/export.h @@ -1,4 +1,5 @@ /* SPDX-License-Identifier: GPL-2.0 */ + #ifndef BTRFS_EXPORT_H #define BTRFS_EXPORT_H diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index e08d0d45af4f..3d9fe58c0080 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1,20 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ + #include <linux/sched.h> #include <linux/sched/signal.h> #include <linux/pagemap.h> @@ -78,10 +66,8 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans, u64 flags, u64 owner, u64 offset, struct btrfs_key *ins, int ref_mod); static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, - u64 parent, u64 root_objectid, - u64 flags, struct btrfs_disk_key *key, - int level, struct btrfs_key *ins); + struct btrfs_delayed_ref_node *node, + struct btrfs_delayed_extent_op *extent_op); static int do_chunk_alloc(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, u64 flags, int force); @@ -268,7 +254,7 @@ static int exclude_super_stripes(struct btrfs_fs_info *fs_info, for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) { bytenr = btrfs_sb_offset(i); ret = btrfs_rmap_block(fs_info, cache->key.objectid, - bytenr, 0, &logical, &nr, &stripe_len); + bytenr, &logical, &nr, &stripe_len); if (ret) return ret; @@ -355,8 +341,9 @@ static void fragment_free_space(struct btrfs_block_group_cache *block_group) * since their free space will be released as soon as the transaction commits. */ u64 add_new_free_space(struct btrfs_block_group_cache *block_group, - struct btrfs_fs_info *info, u64 start, u64 end) + u64 start, u64 end) { + struct btrfs_fs_info *info = block_group->fs_info; u64 extent_start, extent_end, size, total_added = 0; int ret; @@ -501,8 +488,7 @@ next: if (key.type == BTRFS_EXTENT_ITEM_KEY || key.type == BTRFS_METADATA_ITEM_KEY) { - total_found += add_new_free_space(block_group, - fs_info, last, + total_found += add_new_free_space(block_group, last, key.objectid); if (key.type == BTRFS_METADATA_ITEM_KEY) last = key.objectid + @@ -520,7 +506,7 @@ next: } ret = 0; - total_found += add_new_free_space(block_group, fs_info, last, + total_found += add_new_free_space(block_group, last, block_group->key.objectid + block_group->key.offset); caching_ctl->progress = (u64)-1; @@ -756,12 +742,12 @@ static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info, } static void add_pinned_bytes(struct btrfs_fs_info *fs_info, s64 num_bytes, - u64 owner, u64 root_objectid) + bool metadata, u64 root_objectid) { struct btrfs_space_info *space_info; u64 flags; - if (owner < BTRFS_FIRST_FREE_OBJECTID) { + if (metadata) { if (root_objectid == BTRFS_CHUNK_TREE_OBJECTID) flags = BTRFS_BLOCK_GROUP_SYSTEM; else @@ -2212,8 +2198,11 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, &old_ref_mod, &new_ref_mod); } - if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0) - add_pinned_bytes(fs_info, -num_bytes, owner, root_objectid); + if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0) { + bool metadata = owner < BTRFS_FIRST_FREE_OBJECTID; + + add_pinned_bytes(fs_info, -num_bytes, metadata, root_objectid); + } return ret; } @@ -2440,10 +2429,8 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, { int ret = 0; struct btrfs_delayed_tree_ref *ref; - struct btrfs_key ins; u64 parent = 0; u64 ref_root = 0; - bool skinny_metadata = btrfs_fs_incompat(fs_info, SKINNY_METADATA); ref = btrfs_delayed_node_to_tree_ref(node); trace_run_delayed_tree_ref(fs_info, node, ref, node->action); @@ -2452,15 +2439,6 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, parent = ref->parent; ref_root = ref->root; - ins.objectid = node->bytenr; - if (skinny_metadata) { - ins.offset = ref->level; - ins.type = BTRFS_METADATA_ITEM_KEY; - } else { - ins.offset = node->num_bytes; - ins.type = BTRFS_EXTENT_ITEM_KEY; - } - if (node->ref_mod != 1) { btrfs_err(fs_info, "btree block(%llu) has %d references rather than 1: action %d ref_root %llu parent %llu", @@ -2470,11 +2448,7 @@ static int run_delayed_tree_ref(struct btrfs_trans_handle *trans, } if (node->action == BTRFS_ADD_DELAYED_REF && insert_reserved) { BUG_ON(!extent_op || !extent_op->update_flags); - ret = alloc_reserved_tree_block(trans, fs_info, - parent, ref_root, - extent_op->flags_to_set, - &extent_op->key, - ref->level, &ins); + ret = alloc_reserved_tree_block(trans, node, extent_op); } else if (node->action == BTRFS_ADD_DELAYED_REF) { ret = __btrfs_inc_extent_ref(trans, fs_info, node, parent, ref_root, @@ -2606,20 +2580,26 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans, delayed_refs->num_heads--; rb_erase(&head->href_node, &delayed_refs->href_root); RB_CLEAR_NODE(&head->href_node); - spin_unlock(&delayed_refs->lock); spin_unlock(&head->lock); + spin_unlock(&delayed_refs->lock); atomic_dec(&delayed_refs->num_entries); trace_run_delayed_ref_head(fs_info, head, 0); if (head->total_ref_mod < 0) { - struct btrfs_block_group_cache *cache; + struct btrfs_space_info *space_info; + u64 flags; - cache = btrfs_lookup_block_group(fs_info, head->bytenr); - ASSERT(cache); - percpu_counter_add(&cache->space_info->total_bytes_pinned, + if (head->is_data) + flags = BTRFS_BLOCK_GROUP_DATA; + else if (head->is_system) + flags = BTRFS_BLOCK_GROUP_SYSTEM; + else + flags = BTRFS_BLOCK_GROUP_METADATA; + space_info = __find_space_info(fs_info, flags); + ASSERT(space_info); + percpu_counter_add(&space_info->total_bytes_pinned, -head->num_bytes); - btrfs_put_block_group(cache); if (head->is_data) { spin_lock(&delayed_refs->lock); @@ -2706,17 +2686,12 @@ static noinline int __btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, * insert_inline_extent_backref()). */ spin_lock(&locked_ref->lock); - btrfs_merge_delayed_refs(trans, fs_info, delayed_refs, - locked_ref); + btrfs_merge_delayed_refs(trans, delayed_refs, locked_ref); - /* - * locked_ref is the head node, so we have to go one - * node back for any delayed ref updates - */ ref = select_delayed_ref(locked_ref); if (ref && ref->seq && - btrfs_check_delayed_seq(fs_info, delayed_refs, ref->seq)) { + btrfs_check_delayed_seq(fs_info, ref->seq)) { spin_unlock(&locked_ref->lock); unselect_delayed_ref_head(delayed_refs, locked_ref); locked_ref = NULL; @@ -3148,7 +3123,11 @@ static noinline int check_delayed_ref(struct btrfs_root *root, struct rb_node *node; int ret = 0; + spin_lock(&root->fs_info->trans_lock); cur_trans = root->fs_info->running_transaction; + if (cur_trans) + refcount_inc(&cur_trans->use_count); + spin_unlock(&root->fs_info->trans_lock); if (!cur_trans) return 0; @@ -3157,6 +3136,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, head = btrfs_find_delayed_ref_head(delayed_refs, bytenr); if (!head) { spin_unlock(&delayed_refs->lock); + btrfs_put_transaction(cur_trans); return 0; } @@ -3173,6 +3153,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, mutex_lock(&head->mutex); mutex_unlock(&head->mutex); btrfs_put_delayed_ref_head(head); + btrfs_put_transaction(cur_trans); return -EAGAIN; } spin_unlock(&delayed_refs->lock); @@ -3205,6 +3186,7 @@ static noinline int check_delayed_ref(struct btrfs_root *root, } spin_unlock(&head->lock); mutex_unlock(&head->mutex); + btrfs_put_transaction(cur_trans); return ret; } @@ -3290,7 +3272,7 @@ int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset, path = btrfs_alloc_path(); if (!path) - return -ENOENT; + return -ENOMEM; do { ret = check_committed_ref(root, path, objectid, @@ -4025,8 +4007,7 @@ static const char *alloc_name(u64 flags) }; } -static int create_space_info(struct btrfs_fs_info *info, u64 flags, - struct btrfs_space_info **new) +static int create_space_info(struct btrfs_fs_info *info, u64 flags) { struct btrfs_space_info *space_info; @@ -4064,7 +4045,6 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags, return ret; } - *new = space_info; list_add_rcu(&space_info->list, &info->space_info); if (flags & BTRFS_BLOCK_GROUP_DATA) info->data_sinfo = space_info; @@ -4121,7 +4101,7 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags) * returns target flags in extended format or 0 if restripe for this * chunk_type is not in progress * - * should be called with either volume_mutex or balance_lock held + * should be called with balance_lock held */ static u64 get_restripe_target(struct btrfs_fs_info *fs_info, u64 flags) { @@ -4177,7 +4157,7 @@ static u64 btrfs_reduce_alloc_profile(struct btrfs_fs_info *fs_info, u64 flags) /* First, mask out the RAID levels which aren't possible */ for (raid_type = 0; raid_type < BTRFS_NR_RAID_TYPES; raid_type++) { if (num_devices >= btrfs_raid_array[raid_type].devs_min) - allowed |= btrfs_raid_group[raid_type]; + allowed |= btrfs_raid_array[raid_type].bg_flag; } allowed &= flags; @@ -4340,7 +4320,7 @@ commit_trans: need_commit--; if (need_commit > 0) { - btrfs_start_delalloc_roots(fs_info, 0, -1); + btrfs_start_delalloc_roots(fs_info, -1); btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1); } @@ -4642,6 +4622,7 @@ again: if (wait_for_alloc) { mutex_unlock(&fs_info->chunk_mutex); wait_for_alloc = 0; + cond_resched(); goto again; } @@ -4676,12 +4657,14 @@ again: trans->allocating_chunk = false; spin_lock(&space_info->lock); - if (ret < 0 && ret != -ENOSPC) - goto out; - if (ret) - space_info->full = 1; - else + if (ret < 0) { + if (ret == -ENOSPC) + space_info->full = 1; + else + goto out; + } else { ret = 1; + } space_info->force_alloc = CHUNK_ALLOC_NO_FORCE; out: @@ -4790,7 +4773,7 @@ static void btrfs_writeback_inodes_sb_nr(struct btrfs_fs_info *fs_info, * the filesystem is readonly(all dirty pages are written to * the disk). */ - btrfs_start_delalloc_roots(fs_info, 0, nr_items); + btrfs_start_delalloc_roots(fs_info, nr_items); if (!current->journal_info) btrfs_wait_ordered_roots(fs_info, nr_items, 0, (u64)-1); } @@ -5570,14 +5553,18 @@ again: static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *block_rsv, - struct btrfs_block_rsv *dest, u64 num_bytes) + struct btrfs_block_rsv *dest, u64 num_bytes, + u64 *qgroup_to_release_ret) { struct btrfs_space_info *space_info = block_rsv->space_info; + u64 qgroup_to_release = 0; u64 ret; spin_lock(&block_rsv->lock); - if (num_bytes == (u64)-1) + if (num_bytes == (u64)-1) { num_bytes = block_rsv->size; + qgroup_to_release = block_rsv->qgroup_rsv_size; + } block_rsv->size -= num_bytes; if (block_rsv->reserved >= block_rsv->size) { num_bytes = block_rsv->reserved - block_rsv->size; @@ -5586,6 +5573,13 @@ static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info, } else { num_bytes = 0; } + if (block_rsv->qgroup_rsv_reserved >= block_rsv->qgroup_rsv_size) { + qgroup_to_release = block_rsv->qgroup_rsv_reserved - + block_rsv->qgroup_rsv_size; + block_rsv->qgroup_rsv_reserved = block_rsv->qgroup_rsv_size; + } else { + qgroup_to_release = 0; + } spin_unlock(&block_rsv->lock); ret = num_bytes; @@ -5608,6 +5602,8 @@ static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info, space_info_add_old_bytes(fs_info, space_info, num_bytes); } + if (qgroup_to_release_ret) + *qgroup_to_release_ret = qgroup_to_release; return ret; } @@ -5749,17 +5745,21 @@ static int btrfs_inode_rsv_refill(struct btrfs_inode *inode, struct btrfs_root *root = inode->root; struct btrfs_block_rsv *block_rsv = &inode->block_rsv; u64 num_bytes = 0; + u64 qgroup_num_bytes = 0; int ret = -ENOSPC; spin_lock(&block_rsv->lock); if (block_rsv->reserved < block_rsv->size) num_bytes = block_rsv->size - block_rsv->reserved; + if (block_rsv->qgroup_rsv_reserved < block_rsv->qgroup_rsv_size) + qgroup_num_bytes = block_rsv->qgroup_rsv_size - + block_rsv->qgroup_rsv_reserved; spin_unlock(&block_rsv->lock); if (num_bytes == 0) return 0; - ret = btrfs_qgroup_reserve_meta_prealloc(root, num_bytes, true); + ret = btrfs_qgroup_reserve_meta_prealloc(root, qgroup_num_bytes, true); if (ret) return ret; ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush); @@ -5767,7 +5767,13 @@ static int btrfs_inode_rsv_refill(struct btrfs_inode *inode, block_rsv_add_bytes(block_rsv, num_bytes, 0); trace_btrfs_space_reservation(root->fs_info, "delalloc", btrfs_ino(inode), num_bytes, 1); - } + + /* Don't forget to increase qgroup_rsv_reserved */ + spin_lock(&block_rsv->lock); + block_rsv->qgroup_rsv_reserved += qgroup_num_bytes; + spin_unlock(&block_rsv->lock); + } else + btrfs_qgroup_free_meta_prealloc(root, qgroup_num_bytes); return ret; } @@ -5788,20 +5794,23 @@ static void btrfs_inode_rsv_release(struct btrfs_inode *inode, bool qgroup_free) struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv; struct btrfs_block_rsv *block_rsv = &inode->block_rsv; u64 released = 0; + u64 qgroup_to_release = 0; /* * Since we statically set the block_rsv->size we just want to say we * are releasing 0 bytes, and then we'll just get the reservation over * the size free'd. */ - released = block_rsv_release_bytes(fs_info, block_rsv, global_rsv, 0); + released = block_rsv_release_bytes(fs_info, block_rsv, global_rsv, 0, + &qgroup_to_release); if (released > 0) trace_btrfs_space_reservation(fs_info, "delalloc", btrfs_ino(inode), released, 0); if (qgroup_free) - btrfs_qgroup_free_meta_prealloc(inode->root, released); + btrfs_qgroup_free_meta_prealloc(inode->root, qgroup_to_release); else - btrfs_qgroup_convert_reserved_meta(inode->root, released); + btrfs_qgroup_convert_reserved_meta(inode->root, + qgroup_to_release); } void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, @@ -5813,7 +5822,7 @@ void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, if (global_rsv == block_rsv || block_rsv->space_info != global_rsv->space_info) global_rsv = NULL; - block_rsv_release_bytes(fs_info, block_rsv, global_rsv, num_bytes); + block_rsv_release_bytes(fs_info, block_rsv, global_rsv, num_bytes, NULL); } static void update_global_block_rsv(struct btrfs_fs_info *fs_info) @@ -5893,7 +5902,7 @@ static void init_global_block_rsv(struct btrfs_fs_info *fs_info) static void release_global_block_rsv(struct btrfs_fs_info *fs_info) { block_rsv_release_bytes(fs_info, &fs_info->global_block_rsv, NULL, - (u64)-1); + (u64)-1, NULL); WARN_ON(fs_info->trans_block_rsv.size > 0); WARN_ON(fs_info->trans_block_rsv.reserved > 0); WARN_ON(fs_info->chunk_block_rsv.size > 0); @@ -5917,48 +5926,10 @@ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans) WARN_ON_ONCE(!list_empty(&trans->new_bgs)); block_rsv_release_bytes(fs_info, &fs_info->chunk_block_rsv, NULL, - trans->chunk_bytes_reserved); + trans->chunk_bytes_reserved, NULL); trans->chunk_bytes_reserved = 0; } -/* Can only return 0 or -ENOSPC */ -int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans, - struct btrfs_inode *inode) -{ - struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); - struct btrfs_root *root = inode->root; - /* - * We always use trans->block_rsv here as we will have reserved space - * for our orphan when starting the transaction, using get_block_rsv() - * here will sometimes make us choose the wrong block rsv as we could be - * doing a reloc inode for a non refcounted root. - */ - struct btrfs_block_rsv *src_rsv = trans->block_rsv; - struct btrfs_block_rsv *dst_rsv = root->orphan_block_rsv; - - /* - * We need to hold space in order to delete our orphan item once we've - * added it, so this takes the reservation so we can release it later - * when we are truly done with the orphan item. - */ - u64 num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1); - - trace_btrfs_space_reservation(fs_info, "orphan", btrfs_ino(inode), - num_bytes, 1); - return btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1); -} - -void btrfs_orphan_release_metadata(struct btrfs_inode *inode) -{ - struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); - struct btrfs_root *root = inode->root; - u64 num_bytes = btrfs_calc_trans_metadata_size(fs_info, 1); - - trace_btrfs_space_reservation(fs_info, "orphan", btrfs_ino(inode), - num_bytes, 0); - btrfs_block_rsv_release(fs_info, root->orphan_block_rsv, num_bytes); -} - /* * btrfs_subvolume_reserve_metadata() - reserve space for subvolume operation * root: the root of the parent directory @@ -5976,7 +5947,6 @@ void btrfs_orphan_release_metadata(struct btrfs_inode *inode) int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, struct btrfs_block_rsv *rsv, int items, - u64 *qgroup_reserved, bool use_global_rsv) { u64 num_bytes; @@ -5994,8 +5964,6 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, num_bytes = 0; } - *qgroup_reserved = num_bytes; - num_bytes = btrfs_calc_trans_metadata_size(fs_info, items); rsv->space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); @@ -6005,8 +5973,8 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, if (ret == -ENOSPC && use_global_rsv) ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, 1); - if (ret && *qgroup_reserved) - btrfs_qgroup_free_meta_prealloc(root, *qgroup_reserved); + if (ret && num_bytes) + btrfs_qgroup_free_meta_prealloc(root, num_bytes); return ret; } @@ -6022,6 +5990,7 @@ static void btrfs_calculate_inode_block_rsv_size(struct btrfs_fs_info *fs_info, { struct btrfs_block_rsv *block_rsv = &inode->block_rsv; u64 reserve_size = 0; + u64 qgroup_rsv_size = 0; u64 csum_leaves; unsigned outstanding_extents; @@ -6034,9 +6003,17 @@ static void btrfs_calculate_inode_block_rsv_size(struct btrfs_fs_info *fs_info, inode->csum_bytes); reserve_size += btrfs_calc_trans_metadata_size(fs_info, csum_leaves); + /* + * For qgroup rsv, the calculation is very simple: + * account one nodesize for each outstanding extent + * + * This is overestimating in most cases. + */ + qgroup_rsv_size = outstanding_extents * fs_info->nodesize; spin_lock(&block_rsv->lock); block_rsv->size = reserve_size; + block_rsv->qgroup_rsv_size = qgroup_rsv_size; spin_unlock(&block_rsv->lock); } @@ -6317,6 +6294,7 @@ static int update_block_group(struct btrfs_trans_handle *trans, spin_lock(&info->unused_bgs_lock); if (list_empty(&cache->bg_list)) { btrfs_get_block_group(cache); + trace_btrfs_add_unused_block_group(cache); list_add_tail(&cache->bg_list, &info->unused_bgs); } @@ -6474,6 +6452,7 @@ int btrfs_exclude_logged_extents(struct btrfs_fs_info *fs_info, struct btrfs_key key; int found_type; int i; + int ret = 0; if (!btrfs_fs_incompat(fs_info, MIXED_GROUPS)) return 0; @@ -6490,10 +6469,12 @@ int btrfs_exclude_logged_extents(struct btrfs_fs_info *fs_info, continue; key.objectid = btrfs_file_extent_disk_bytenr(eb, item); key.offset = btrfs_file_extent_disk_num_bytes(eb, item); - __exclude_logged_extent(fs_info, key.objectid, key.offset); + ret = __exclude_logged_extent(fs_info, key.objectid, key.offset); + if (ret) + break; } - return 0; + return ret; } static void @@ -7085,7 +7066,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, } } - ret = add_to_free_space_tree(trans, info, bytenr, num_bytes); + ret = add_to_free_space_tree(trans, bytenr, num_bytes); if (ret) { btrfs_abort_transaction(trans, ret); goto out; @@ -7229,7 +7210,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, } out: if (pin) - add_pinned_bytes(fs_info, buf->len, btrfs_header_level(buf), + add_pinned_bytes(fs_info, buf->len, true, root->root_key.objectid); if (last_ref) { @@ -7283,8 +7264,11 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, &old_ref_mod, &new_ref_mod); } - if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0) - add_pinned_bytes(fs_info, num_bytes, owner, root_objectid); + if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0) { + bool metadata = owner < BTRFS_FIRST_FREE_OBJECTID; + + add_pinned_bytes(fs_info, num_bytes, metadata, root_objectid); + } return ret; } @@ -7336,24 +7320,6 @@ wait_block_group_cache_done(struct btrfs_block_group_cache *cache) return ret; } -static const char *btrfs_raid_type_names[BTRFS_NR_RAID_TYPES] = { - [BTRFS_RAID_RAID10] = "raid10", - [BTRFS_RAID_RAID1] = "raid1", - [BTRFS_RAID_DUP] = "dup", - [BTRFS_RAID_RAID0] = "raid0", - [BTRFS_RAID_SINGLE] = "single", - [BTRFS_RAID_RAID5] = "raid5", - [BTRFS_RAID_RAID6] = "raid6", -}; - -static const char *get_raid_name(enum btrfs_raid_types type) -{ - if (type >= BTRFS_NR_RAID_TYPES) - return NULL; - - return btrfs_raid_type_names[type]; -} - enum btrfs_loop_type { LOOP_CACHING_NOWAIT = 0, LOOP_CACHING_WAIT = 1, @@ -7625,7 +7591,7 @@ have_block_group: if (offset) { /* we have a block, we're done */ spin_unlock(&last_ptr->refill_lock); - trace_btrfs_reserve_extent_cluster(fs_info, + trace_btrfs_reserve_extent_cluster( used_block_group, search_start, num_bytes); if (used_block_group != block_group) { @@ -7698,7 +7664,7 @@ refill_cluster: if (offset) { /* we found one, proceed */ spin_unlock(&last_ptr->refill_lock); - trace_btrfs_reserve_extent_cluster(fs_info, + trace_btrfs_reserve_extent_cluster( block_group, search_start, num_bytes); goto checks; @@ -7798,8 +7764,7 @@ checks: ins->objectid = search_start; ins->offset = num_bytes; - trace_btrfs_reserve_extent(fs_info, block_group, - search_start, num_bytes); + trace_btrfs_reserve_extent(block_group, search_start, num_bytes); btrfs_release_block_group(block_group, delalloc); break; loop: @@ -8147,8 +8112,7 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans, btrfs_mark_buffer_dirty(path->nodes[0]); btrfs_free_path(path); - ret = remove_from_free_space_tree(trans, fs_info, ins->objectid, - ins->offset); + ret = remove_from_free_space_tree(trans, ins->objectid, ins->offset); if (ret) return ret; @@ -8163,37 +8127,52 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans, } static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, - u64 parent, u64 root_objectid, - u64 flags, struct btrfs_disk_key *key, - int level, struct btrfs_key *ins) + struct btrfs_delayed_ref_node *node, + struct btrfs_delayed_extent_op *extent_op) { + struct btrfs_fs_info *fs_info = trans->fs_info; int ret; struct btrfs_extent_item *extent_item; + struct btrfs_key extent_key; struct btrfs_tree_block_info *block_info; struct btrfs_extent_inline_ref *iref; struct btrfs_path *path; struct extent_buffer *leaf; + struct btrfs_delayed_tree_ref *ref; u32 size = sizeof(*extent_item) + sizeof(*iref); - u64 num_bytes = ins->offset; + u64 num_bytes; + u64 flags = extent_op->flags_to_set; bool skinny_metadata = btrfs_fs_incompat(fs_info, SKINNY_METADATA); - if (!skinny_metadata) + ref = btrfs_delayed_node_to_tree_ref(node); + + extent_key.objectid = node->bytenr; + if (skinny_metadata) { + extent_key.offset = ref->level; + extent_key.type = BTRFS_METADATA_ITEM_KEY; + num_bytes = fs_info->nodesize; + } else { + extent_key.offset = node->num_bytes; + extent_key.type = BTRFS_EXTENT_ITEM_KEY; size += sizeof(*block_info); + num_bytes = node->num_bytes; + } path = btrfs_alloc_path(); if (!path) { - btrfs_free_and_pin_reserved_extent(fs_info, ins->objectid, + btrfs_free_and_pin_reserved_extent(fs_info, + extent_key.objectid, fs_info->nodesize); return -ENOMEM; } path->leave_spinning = 1; ret = btrfs_insert_empty_item(trans, fs_info->extent_root, path, - ins, size); + &extent_key, size); if (ret) { btrfs_free_path(path); - btrfs_free_and_pin_reserved_extent(fs_info, ins->objectid, + btrfs_free_and_pin_reserved_extent(fs_info, + extent_key.objectid, fs_info->nodesize); return ret; } @@ -8208,42 +8187,41 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, if (skinny_metadata) { iref = (struct btrfs_extent_inline_ref *)(extent_item + 1); - num_bytes = fs_info->nodesize; } else { block_info = (struct btrfs_tree_block_info *)(extent_item + 1); - btrfs_set_tree_block_key(leaf, block_info, key); - btrfs_set_tree_block_level(leaf, block_info, level); + btrfs_set_tree_block_key(leaf, block_info, &extent_op->key); + btrfs_set_tree_block_level(leaf, block_info, ref->level); iref = (struct btrfs_extent_inline_ref *)(block_info + 1); } - if (parent > 0) { + if (node->type == BTRFS_SHARED_BLOCK_REF_KEY) { BUG_ON(!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)); btrfs_set_extent_inline_ref_type(leaf, iref, BTRFS_SHARED_BLOCK_REF_KEY); - btrfs_set_extent_inline_ref_offset(leaf, iref, parent); + btrfs_set_extent_inline_ref_offset(leaf, iref, ref->parent); } else { btrfs_set_extent_inline_ref_type(leaf, iref, BTRFS_TREE_BLOCK_REF_KEY); - btrfs_set_extent_inline_ref_offset(leaf, iref, root_objectid); + btrfs_set_extent_inline_ref_offset(leaf, iref, ref->root); } btrfs_mark_buffer_dirty(leaf); btrfs_free_path(path); - ret = remove_from_free_space_tree(trans, fs_info, ins->objectid, + ret = remove_from_free_space_tree(trans, extent_key.objectid, num_bytes); if (ret) return ret; - ret = update_block_group(trans, fs_info, ins->objectid, + ret = update_block_group(trans, fs_info, extent_key.objectid, fs_info->nodesize, 1); if (ret) { /* -ENOENT, logic error */ btrfs_err(fs_info, "update block group failed for %llu %llu", - ins->objectid, ins->offset); + extent_key.objectid, extent_key.offset); BUG(); } - trace_btrfs_reserved_extent_alloc(fs_info, ins->objectid, + trace_btrfs_reserved_extent_alloc(fs_info, extent_key.objectid, fs_info->nodesize); return ret; } @@ -8414,7 +8392,7 @@ static void unuse_block_rsv(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *block_rsv, u32 blocksize) { block_rsv_add_bytes(block_rsv, blocksize, 0); - block_rsv_release_bytes(fs_info, block_rsv, NULL, 0); + block_rsv_release_bytes(fs_info, block_rsv, NULL, 0, NULL); } /* @@ -10136,8 +10114,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) } else if (btrfs_block_group_used(&cache->item) == 0) { cache->last_byte_to_unpin = (u64)-1; cache->cached = BTRFS_CACHE_FINISHED; - add_new_free_space(cache, info, - found_key.objectid, + add_new_free_space(cache, found_key.objectid, found_key.objectid + found_key.offset); free_excluded_extents(info, cache); @@ -10167,6 +10144,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) /* Should always be true but just in case. */ if (list_empty(&cache->bg_list)) { btrfs_get_block_group(cache); + trace_btrfs_add_unused_block_group(cache); list_add_tail(&cache->bg_list, &info->unused_bgs); } @@ -10232,7 +10210,7 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans) key.offset); if (ret) btrfs_abort_transaction(trans, ret); - add_block_group_free_space(trans, fs_info, block_group); + add_block_group_free_space(trans, block_group); /* already aborted the transaction if it failed. */ next: list_del_init(&block_group->bg_list); @@ -10273,7 +10251,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, return ret; } - add_new_free_space(cache, fs_info, chunk_offset, chunk_offset + size); + add_new_free_space(cache, chunk_offset, chunk_offset + size); free_excluded_extents(fs_info, cache); @@ -10354,6 +10332,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, BUG_ON(!block_group); BUG_ON(!block_group->ro); + trace_btrfs_remove_block_group(block_group); /* * Free the reserved super bytes from this block group before * remove it. @@ -10611,7 +10590,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, mutex_unlock(&fs_info->chunk_mutex); - ret = remove_block_group_free_space(trans, fs_info, block_group); + ret = remove_block_group_free_space(trans, block_group); if (ret) goto out; @@ -10718,6 +10697,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) * the ro check in case balance is currently acting on * this block group. */ + trace_btrfs_skip_unused_block_group(block_group); spin_unlock(&block_group->lock); up_write(&space_info->groups_sem); goto next; @@ -10840,7 +10820,6 @@ next: int btrfs_init_space_info(struct btrfs_fs_info *fs_info) { - struct btrfs_space_info *space_info; struct btrfs_super_block *disk_super; u64 features; u64 flags; @@ -10856,21 +10835,21 @@ int btrfs_init_space_info(struct btrfs_fs_info *fs_info) mixed = 1; flags = BTRFS_BLOCK_GROUP_SYSTEM; - ret = create_space_info(fs_info, flags, &space_info); + ret = create_space_info(fs_info, flags); if (ret) goto out; if (mixed) { flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA; - ret = create_space_info(fs_info, flags, &space_info); + ret = create_space_info(fs_info, flags); } else { flags = BTRFS_BLOCK_GROUP_METADATA; - ret = create_space_info(fs_info, flags, &space_info); + ret = create_space_info(fs_info, flags); if (ret) goto out; flags = BTRFS_BLOCK_GROUP_DATA; - ret = create_space_info(fs_info, flags, &space_info); + ret = create_space_info(fs_info, flags); } out: return ret; @@ -11055,12 +11034,7 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range) void btrfs_end_write_no_snapshotting(struct btrfs_root *root) { percpu_counter_dec(&root->subv_writers->counter); - /* - * Make sure counter is updated before we wake up waiters. - */ - smp_mb(); - if (waitqueue_active(&root->subv_writers->wait)) - wake_up(&root->subv_writers->wait); + cond_wake_up(&root->subv_writers->wait); } int btrfs_start_write_no_snapshotting(struct btrfs_root *root) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 47a8fe9d22e8..51fc015c7d2c 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -1,4 +1,5 @@ // SPDX-License-Identifier: GPL-2.0 + #include <linux/bitops.h> #include <linux/slab.h> #include <linux/bio.h> @@ -25,7 +26,7 @@ static struct kmem_cache *extent_state_cache; static struct kmem_cache *extent_buffer_cache; -static struct bio_set *btrfs_bioset; +static struct bio_set btrfs_bioset; static inline bool extent_state_in_tree(const struct extent_state *state) { @@ -161,20 +162,18 @@ int __init extent_io_init(void) if (!extent_buffer_cache) goto free_state_cache; - btrfs_bioset = bioset_create(BIO_POOL_SIZE, - offsetof(struct btrfs_io_bio, bio), - BIOSET_NEED_BVECS); - if (!btrfs_bioset) + if (bioset_init(&btrfs_bioset, BIO_POOL_SIZE, + offsetof(struct btrfs_io_bio, bio), + BIOSET_NEED_BVECS)) goto free_buffer_cache; - if (bioset_integrity_create(btrfs_bioset, BIO_POOL_SIZE)) + if (bioset_integrity_create(&btrfs_bioset, BIO_POOL_SIZE)) goto free_bioset; return 0; free_bioset: - bioset_free(btrfs_bioset); - btrfs_bioset = NULL; + bioset_exit(&btrfs_bioset); free_buffer_cache: kmem_cache_destroy(extent_buffer_cache); @@ -197,8 +196,7 @@ void __cold extent_io_exit(void) rcu_barrier(); kmem_cache_destroy(extent_state_cache); kmem_cache_destroy(extent_buffer_cache); - if (btrfs_bioset) - bioset_free(btrfs_bioset); + bioset_exit(&btrfs_bioset); } void extent_io_tree_init(struct extent_io_tree *tree, @@ -2678,7 +2676,7 @@ struct bio *btrfs_bio_alloc(struct block_device *bdev, u64 first_byte) { struct bio *bio; - bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, btrfs_bioset); + bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &btrfs_bioset); bio_set_dev(bio, bdev); bio->bi_iter.bi_sector = first_byte >> 9; btrfs_io_bio_init(btrfs_io_bio(bio)); @@ -2691,7 +2689,7 @@ struct bio *btrfs_bio_clone(struct bio *bio) struct bio *new; /* Bio allocation backed by a bioset does not fail */ - new = bio_clone_fast(bio, GFP_NOFS, btrfs_bioset); + new = bio_clone_fast(bio, GFP_NOFS, &btrfs_bioset); btrfs_bio = btrfs_io_bio(new); btrfs_io_bio_init(btrfs_bio); btrfs_bio->iter = bio->bi_iter; @@ -2703,7 +2701,7 @@ struct bio *btrfs_io_bio_alloc(unsigned int nr_iovecs) struct bio *bio; /* Bio allocation backed by a bioset does not fail */ - bio = bio_alloc_bioset(GFP_NOFS, nr_iovecs, btrfs_bioset); + bio = bio_alloc_bioset(GFP_NOFS, nr_iovecs, &btrfs_bioset); btrfs_io_bio_init(btrfs_io_bio(bio)); return bio; } @@ -2714,7 +2712,7 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size) struct btrfs_io_bio *btrfs_bio; /* this will never fail when it's backed by a bioset */ - bio = bio_clone_fast(orig, GFP_NOFS, btrfs_bioset); + bio = bio_clone_fast(orig, GFP_NOFS, &btrfs_bioset); ASSERT(bio); btrfs_bio = btrfs_io_bio(bio); @@ -3963,11 +3961,11 @@ retry: done_index = page->index; /* - * At this point we hold neither mapping->tree_lock nor - * lock on the page itself: the page may be truncated or - * invalidated (changing page->mapping to NULL), or even - * swizzled back from swapper_space to tmpfs file - * mapping + * At this point we hold neither the i_pages lock nor + * the page lock: the page may be truncated or + * invalidated (changing page->mapping to NULL), + * or even swizzled back from swapper_space to + * tmpfs file mapping */ if (!trylock_page(page)) { flush_write_bio(epd); @@ -4108,14 +4106,13 @@ int extent_write_locked_range(struct inode *inode, u64 start, u64 end, return ret; } -int extent_writepages(struct extent_io_tree *tree, - struct address_space *mapping, +int extent_writepages(struct address_space *mapping, struct writeback_control *wbc) { int ret = 0; struct extent_page_data epd = { .bio = NULL, - .tree = tree, + .tree = &BTRFS_I(mapping->host)->io_tree, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; @@ -4125,9 +4122,8 @@ int extent_writepages(struct extent_io_tree *tree, return ret; } -int extent_readpages(struct extent_io_tree *tree, - struct address_space *mapping, - struct list_head *pages, unsigned nr_pages) +int extent_readpages(struct address_space *mapping, struct list_head *pages, + unsigned nr_pages) { struct bio *bio = NULL; unsigned page_idx; @@ -4135,6 +4131,7 @@ int extent_readpages(struct extent_io_tree *tree, struct page *pagepool[16]; struct page *page; struct extent_map *em_cached = NULL; + struct extent_io_tree *tree = &BTRFS_I(mapping->host)->io_tree; int nr = 0; u64 prev_em_start = (u64)-1; @@ -4201,8 +4198,7 @@ int extent_invalidatepage(struct extent_io_tree *tree, * are locked or under IO and drops the related state bits if it is safe * to drop the page. */ -static int try_release_extent_state(struct extent_map_tree *map, - struct extent_io_tree *tree, +static int try_release_extent_state(struct extent_io_tree *tree, struct page *page, gfp_t mask) { u64 start = page_offset(page); @@ -4237,13 +4233,13 @@ static int try_release_extent_state(struct extent_map_tree *map, * in the range corresponding to the page, both state records and extent * map records are removed */ -int try_release_extent_mapping(struct extent_map_tree *map, - struct extent_io_tree *tree, struct page *page, - gfp_t mask) +int try_release_extent_mapping(struct page *page, gfp_t mask) { struct extent_map *em; u64 start = page_offset(page); u64 end = start + PAGE_SIZE - 1; + struct extent_io_tree *tree = &BTRFS_I(page->mapping->host)->io_tree; + struct extent_map_tree *map = &BTRFS_I(page->mapping->host)->extent_tree; if (gfpflags_allow_blocking(mask) && page->mapping->host->i_size > SZ_16M) { @@ -4277,7 +4273,7 @@ int try_release_extent_mapping(struct extent_map_tree *map, free_extent_map(em); } } - return try_release_extent_state(map, tree, page, mask); + return try_release_extent_state(tree, page, mask); } /* @@ -5174,13 +5170,13 @@ void clear_extent_buffer_dirty(struct extent_buffer *eb) WARN_ON(!PagePrivate(page)); clear_page_dirty_for_io(page); - spin_lock_irq(&page->mapping->tree_lock); + xa_lock_irq(&page->mapping->i_pages); if (!PageDirty(page)) { - radix_tree_tag_clear(&page->mapping->page_tree, + radix_tree_tag_clear(&page->mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); } - spin_unlock_irq(&page->mapping->tree_lock); + xa_unlock_irq(&page->mapping->i_pages); ClearPageError(page); unlock_page(page); } @@ -5619,46 +5615,6 @@ void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src, } } -void le_bitmap_set(u8 *map, unsigned int start, int len) -{ - u8 *p = map + BIT_BYTE(start); - const unsigned int size = start + len; - int bits_to_set = BITS_PER_BYTE - (start % BITS_PER_BYTE); - u8 mask_to_set = BITMAP_FIRST_BYTE_MASK(start); - - while (len - bits_to_set >= 0) { - *p |= mask_to_set; - len -= bits_to_set; - bits_to_set = BITS_PER_BYTE; - mask_to_set = ~0; - p++; - } - if (len) { - mask_to_set &= BITMAP_LAST_BYTE_MASK(size); - *p |= mask_to_set; - } -} - -void le_bitmap_clear(u8 *map, unsigned int start, int len) -{ - u8 *p = map + BIT_BYTE(start); - const unsigned int size = start + len; - int bits_to_clear = BITS_PER_BYTE - (start % BITS_PER_BYTE); - u8 mask_to_clear = BITMAP_FIRST_BYTE_MASK(start); - - while (len - bits_to_clear >= 0) { - *p &= ~mask_to_clear; - len -= bits_to_clear; - bits_to_clear = BITS_PER_BYTE; - mask_to_clear = ~0; - p++; - } - if (len) { - mask_to_clear &= BITMAP_LAST_BYTE_MASK(size); - *p &= ~mask_to_clear; - } -} - /* * eb_bitmap_offset() - calculate the page and offset of the byte containing the * given bit number diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index b77d84909863..0bfd4aeb822d 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __EXTENTIO__ -#define __EXTENTIO__ + +#ifndef BTRFS_EXTENT_IO_H +#define BTRFS_EXTENT_IO_H #include <linux/rbtree.h> #include <linux/refcount.h> @@ -78,14 +79,6 @@ #define BITMAP_LAST_BYTE_MASK(nbits) \ (BYTE_MASK >> (-(nbits) & (BITS_PER_BYTE - 1))) -static inline int le_test_bit(int nr, const u8 *addr) -{ - return 1U & (addr[BIT_BYTE(nr)] >> (nr & (BITS_PER_BYTE-1))); -} - -void le_bitmap_set(u8 *map, unsigned int start, int len); -void le_bitmap_clear(u8 *map, unsigned int start, int len); - struct extent_state; struct btrfs_root; struct btrfs_inode; @@ -277,9 +270,7 @@ typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode, int create); void extent_io_tree_init(struct extent_io_tree *tree, void *private_data); -int try_release_extent_mapping(struct extent_map_tree *map, - struct extent_io_tree *tree, struct page *page, - gfp_t mask); +int try_release_extent_mapping(struct page *page, gfp_t mask); int try_release_extent_buffer(struct page *page); int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end, struct extent_state **cached); @@ -420,14 +411,12 @@ int extent_invalidatepage(struct extent_io_tree *tree, int extent_write_full_page(struct page *page, struct writeback_control *wbc); int extent_write_locked_range(struct inode *inode, u64 start, u64 end, int mode); -int extent_writepages(struct extent_io_tree *tree, - struct address_space *mapping, +int extent_writepages(struct address_space *mapping, struct writeback_control *wbc); int btree_write_cache_pages(struct address_space *mapping, struct writeback_control *wbc); -int extent_readpages(struct extent_io_tree *tree, - struct address_space *mapping, - struct list_head *pages, unsigned nr_pages); +int extent_readpages(struct address_space *mapping, struct list_head *pages, + unsigned nr_pages); int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, __u64 start, __u64 len); void set_page_extent_mapped(struct page *page); @@ -572,4 +561,5 @@ noinline u64 find_lock_delalloc_range(struct inode *inode, #endif struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info, u64 start); + #endif diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 53a0633c6ef7..6648d55e5339 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -1,4 +1,5 @@ // SPDX-License-Identifier: GPL-2.0 + #include <linux/err.h> #include <linux/slab.h> #include <linux/spinlock.h> @@ -517,6 +518,7 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, /** * btrfs_add_extent_mapping - add extent mapping into em_tree + * @fs_info - used for tracepoint * @em_tree - the extent tree into which we want to insert the extent mapping * @em_in - extent we are inserting * @start - start of the logical range btrfs_get_extent() is requesting @@ -534,7 +536,8 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, * Return 0 on success, otherwise -EEXIST. * */ -int btrfs_add_extent_mapping(struct extent_map_tree *em_tree, +int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree, struct extent_map **em_in, u64 start, u64 len) { int ret; @@ -552,7 +555,7 @@ int btrfs_add_extent_mapping(struct extent_map_tree *em_tree, existing = search_extent_mapping(em_tree, start, len); - trace_btrfs_handle_em_exist(existing, em, start, len); + trace_btrfs_handle_em_exist(fs_info, existing, em, start, len); /* * existing will always be non-NULL, since there must be diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h index f6f8ba114977..25d985e7532a 100644 --- a/fs/btrfs/extent_map.h +++ b/fs/btrfs/extent_map.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __EXTENTMAP__ -#define __EXTENTMAP__ + +#ifndef BTRFS_EXTENT_MAP_H +#define BTRFS_EXTENT_MAP_H #include <linux/rbtree.h> #include <linux/refcount.h> @@ -91,6 +92,8 @@ int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len, u64 gen void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em); struct extent_map *search_extent_mapping(struct extent_map_tree *tree, u64 start, u64 len); -int btrfs_add_extent_mapping(struct extent_map_tree *em_tree, +int btrfs_add_extent_mapping(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree, struct extent_map **em_in, u64 start, u64 len); + #endif diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index fdcb41002623..f9dd6d1836a3 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/bio.h> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index f247300170e5..f660ba1e5e58 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/fs.h> @@ -1761,7 +1748,7 @@ again: unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend, &cached_state); btrfs_delalloc_release_extents(BTRFS_I(inode), reserve_bytes, - (ret != 0)); + true); if (ret) { btrfs_drop_pages(pages, num_pages); break; diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index d0dde9e6afd7..d5f80cb300be 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Red Hat. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/pagemap.h> @@ -266,10 +253,8 @@ int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans, truncate_pagecache(inode, 0); /* - * We don't need an orphan item because truncating the free space cache - * will never be split across transactions. - * We don't need to check for -EAGAIN because we're a free space - * cache inode + * We skip the throttling logic for free space cache inodes, so we don't + * need to check for -EAGAIN. */ ret = btrfs_truncate_inode_items(trans, root, inode, 0, BTRFS_EXTENT_DATA_KEY); diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h index 79eca4cabb1c..15e30b93db0d 100644 --- a/fs/btrfs/free-space-cache.h +++ b/fs/btrfs/free-space-cache.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2009 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_FREE_SPACE_CACHE -#define __BTRFS_FREE_SPACE_CACHE +#ifndef BTRFS_FREE_SPACE_CACHE_H +#define BTRFS_FREE_SPACE_CACHE_H struct btrfs_free_space { struct rb_node offset_index; diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c index af36a6a971fe..b5950aacd697 100644 --- a/fs/btrfs/free-space-tree.c +++ b/fs/btrfs/free-space-tree.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2015 Facebook. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/kernel.h> @@ -25,7 +12,6 @@ #include "transaction.h" static int __add_block_group_free_space(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path); @@ -58,11 +44,10 @@ void set_free_space_tree_thresholds(struct btrfs_block_group_cache *cache) } static int add_new_free_space_info(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path) { - struct btrfs_root *root = fs_info->free_space_root; + struct btrfs_root *root = trans->fs_info->free_space_root; struct btrfs_free_space_info *info; struct btrfs_key key; struct extent_buffer *leaf; @@ -151,10 +136,11 @@ static inline u32 free_space_bitmap_size(u64 size, u32 sectorsize) return DIV_ROUND_UP((u32)div_u64(size, sectorsize), BITS_PER_BYTE); } -static u8 *alloc_bitmap(u32 bitmap_size) +static unsigned long *alloc_bitmap(u32 bitmap_size) { - u8 *ret; + unsigned long *ret; unsigned int nofs_flag; + u32 bitmap_rounded_size = round_up(bitmap_size, sizeof(unsigned long)); /* * GFP_NOFS doesn't work with kvmalloc(), but we really can't recurse @@ -165,21 +151,42 @@ static u8 *alloc_bitmap(u32 bitmap_size) * know that recursion is unsafe. */ nofs_flag = memalloc_nofs_save(); - ret = kvzalloc(bitmap_size, GFP_KERNEL); + ret = kvzalloc(bitmap_rounded_size, GFP_KERNEL); memalloc_nofs_restore(nofs_flag); return ret; } +static void le_bitmap_set(unsigned long *map, unsigned int start, int len) +{ + u8 *p = ((u8 *)map) + BIT_BYTE(start); + const unsigned int size = start + len; + int bits_to_set = BITS_PER_BYTE - (start % BITS_PER_BYTE); + u8 mask_to_set = BITMAP_FIRST_BYTE_MASK(start); + + while (len - bits_to_set >= 0) { + *p |= mask_to_set; + len -= bits_to_set; + bits_to_set = BITS_PER_BYTE; + mask_to_set = ~0; + p++; + } + if (len) { + mask_to_set &= BITMAP_LAST_BYTE_MASK(size); + *p |= mask_to_set; + } +} + int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path) { + struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_root *root = fs_info->free_space_root; struct btrfs_free_space_info *info; struct btrfs_key key, found_key; struct extent_buffer *leaf; - u8 *bitmap, *bitmap_cursor; + unsigned long *bitmap; + char *bitmap_cursor; u64 start, end; u64 bitmap_range, i; u32 bitmap_size, flags, expected_extent_count; @@ -268,7 +275,7 @@ int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans, goto out; } - bitmap_cursor = bitmap; + bitmap_cursor = (char *)bitmap; bitmap_range = fs_info->sectorsize * BTRFS_FREE_SPACE_BITMAP_BITS; i = start; while (i < end) { @@ -309,21 +316,18 @@ out: } int convert_free_space_to_extents(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path) { + struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_root *root = fs_info->free_space_root; struct btrfs_free_space_info *info; struct btrfs_key key, found_key; struct extent_buffer *leaf; - u8 *bitmap; + unsigned long *bitmap; u64 start, end; - /* Initialize to silence GCC. */ - u64 extent_start = 0; - u64 offset; u32 bitmap_size, flags, expected_extent_count; - int prev_bit = 0, bit, bitnr; + unsigned long nrbits, start_bit, end_bit; u32 extent_count = 0; int done = 0, nr; int ret; @@ -361,7 +365,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans, break; } else if (found_key.type == BTRFS_FREE_SPACE_BITMAP_KEY) { unsigned long ptr; - u8 *bitmap_cursor; + char *bitmap_cursor; u32 bitmap_pos, data_size; ASSERT(found_key.objectid >= start); @@ -371,7 +375,7 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans, bitmap_pos = div_u64(found_key.objectid - start, fs_info->sectorsize * BITS_PER_BYTE); - bitmap_cursor = bitmap + bitmap_pos; + bitmap_cursor = ((char *)bitmap) + bitmap_pos; data_size = free_space_bitmap_size(found_key.offset, fs_info->sectorsize); @@ -405,32 +409,16 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans, btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); - offset = start; - bitnr = 0; - while (offset < end) { - bit = !!le_test_bit(bitnr, bitmap); - if (prev_bit == 0 && bit == 1) { - extent_start = offset; - } else if (prev_bit == 1 && bit == 0) { - key.objectid = extent_start; - key.type = BTRFS_FREE_SPACE_EXTENT_KEY; - key.offset = offset - extent_start; - - ret = btrfs_insert_empty_item(trans, root, path, &key, 0); - if (ret) - goto out; - btrfs_release_path(path); + nrbits = div_u64(block_group->key.offset, block_group->fs_info->sectorsize); + start_bit = find_next_bit_le(bitmap, nrbits, 0); - extent_count++; - } - prev_bit = bit; - offset += fs_info->sectorsize; - bitnr++; - } - if (prev_bit == 1) { - key.objectid = extent_start; + while (start_bit < nrbits) { + end_bit = find_next_zero_bit_le(bitmap, nrbits, start_bit); + ASSERT(start_bit < end_bit); + + key.objectid = start + start_bit * block_group->fs_info->sectorsize; key.type = BTRFS_FREE_SPACE_EXTENT_KEY; - key.offset = end - extent_start; + key.offset = (end_bit - start_bit) * block_group->fs_info->sectorsize; ret = btrfs_insert_empty_item(trans, root, path, &key, 0); if (ret) @@ -438,6 +426,8 @@ int convert_free_space_to_extents(struct btrfs_trans_handle *trans, btrfs_release_path(path); extent_count++; + + start_bit = find_next_bit_le(bitmap, nrbits, end_bit); } if (extent_count != expected_extent_count) { @@ -459,7 +449,6 @@ out: } static int update_free_space_extent_count(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, int new_extents) @@ -472,7 +461,8 @@ static int update_free_space_extent_count(struct btrfs_trans_handle *trans, if (new_extents == 0) return 0; - info = search_free_space_info(trans, fs_info, block_group, path, 1); + info = search_free_space_info(trans, trans->fs_info, block_group, path, + 1); if (IS_ERR(info)) { ret = PTR_ERR(info); goto out; @@ -487,12 +477,10 @@ static int update_free_space_extent_count(struct btrfs_trans_handle *trans, if (!(flags & BTRFS_FREE_SPACE_USING_BITMAPS) && extent_count > block_group->bitmap_high_thresh) { - ret = convert_free_space_to_bitmaps(trans, fs_info, block_group, - path); + ret = convert_free_space_to_bitmaps(trans, block_group, path); } else if ((flags & BTRFS_FREE_SPACE_USING_BITMAPS) && extent_count < block_group->bitmap_low_thresh) { - ret = convert_free_space_to_extents(trans, fs_info, block_group, - path); + ret = convert_free_space_to_extents(trans, block_group, path); } out: @@ -589,12 +577,11 @@ static int free_space_next_bitmap(struct btrfs_trans_handle *trans, * the bitmap. */ static int modify_free_space_bitmap(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, u64 start, u64 size, int remove) { - struct btrfs_root *root = fs_info->free_space_root; + struct btrfs_root *root = block_group->fs_info->free_space_root; struct btrfs_key key; u64 end = start + size; u64 cur_start, cur_size; @@ -695,7 +682,7 @@ static int modify_free_space_bitmap(struct btrfs_trans_handle *trans, } btrfs_release_path(path); - ret = update_free_space_extent_count(trans, fs_info, block_group, path, + ret = update_free_space_extent_count(trans, block_group, path, new_extents); out: @@ -703,12 +690,11 @@ out: } static int remove_free_space_extent(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, u64 start, u64 size) { - struct btrfs_root *root = fs_info->free_space_root; + struct btrfs_root *root = trans->fs_info->free_space_root; struct btrfs_key key; u64 found_start, found_end; u64 end = start + size; @@ -782,7 +768,7 @@ static int remove_free_space_extent(struct btrfs_trans_handle *trans, } btrfs_release_path(path); - ret = update_free_space_extent_count(trans, fs_info, block_group, path, + ret = update_free_space_extent_count(trans, block_group, path, new_extents); out: @@ -790,7 +776,6 @@ out: } int __remove_from_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, u64 start, u64 size) { @@ -799,36 +784,35 @@ int __remove_from_free_space_tree(struct btrfs_trans_handle *trans, int ret; if (block_group->needs_free_space) { - ret = __add_block_group_free_space(trans, fs_info, block_group, - path); + ret = __add_block_group_free_space(trans, block_group, path); if (ret) return ret; } - info = search_free_space_info(NULL, fs_info, block_group, path, 0); + info = search_free_space_info(NULL, trans->fs_info, block_group, path, + 0); if (IS_ERR(info)) return PTR_ERR(info); flags = btrfs_free_space_flags(path->nodes[0], info); btrfs_release_path(path); if (flags & BTRFS_FREE_SPACE_USING_BITMAPS) { - return modify_free_space_bitmap(trans, fs_info, block_group, - path, start, size, 1); + return modify_free_space_bitmap(trans, block_group, path, + start, size, 1); } else { - return remove_free_space_extent(trans, fs_info, block_group, - path, start, size); + return remove_free_space_extent(trans, block_group, path, + start, size); } } int remove_from_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u64 start, u64 size) { struct btrfs_block_group_cache *block_group; struct btrfs_path *path; int ret; - if (!btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) + if (!btrfs_fs_compat_ro(trans->fs_info, FREE_SPACE_TREE)) return 0; path = btrfs_alloc_path(); @@ -837,7 +821,7 @@ int remove_from_free_space_tree(struct btrfs_trans_handle *trans, goto out; } - block_group = btrfs_lookup_block_group(fs_info, start); + block_group = btrfs_lookup_block_group(trans->fs_info, start); if (!block_group) { ASSERT(0); ret = -ENOENT; @@ -845,8 +829,8 @@ int remove_from_free_space_tree(struct btrfs_trans_handle *trans, } mutex_lock(&block_group->free_space_lock); - ret = __remove_from_free_space_tree(trans, fs_info, block_group, path, - start, size); + ret = __remove_from_free_space_tree(trans, block_group, path, start, + size); mutex_unlock(&block_group->free_space_lock); btrfs_put_block_group(block_group); @@ -858,12 +842,11 @@ out: } static int add_free_space_extent(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, u64 start, u64 size) { - struct btrfs_root *root = fs_info->free_space_root; + struct btrfs_root *root = trans->fs_info->free_space_root; struct btrfs_key key, new_key; u64 found_start, found_end; u64 end = start + size; @@ -978,7 +961,7 @@ insert: goto out; btrfs_release_path(path); - ret = update_free_space_extent_count(trans, fs_info, block_group, path, + ret = update_free_space_extent_count(trans, block_group, path, new_extents); out: @@ -986,17 +969,16 @@ out: } int __add_to_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, u64 start, u64 size) { + struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_free_space_info *info; u32 flags; int ret; if (block_group->needs_free_space) { - ret = __add_block_group_free_space(trans, fs_info, block_group, - path); + ret = __add_block_group_free_space(trans, block_group, path); if (ret) return ret; } @@ -1008,23 +990,22 @@ int __add_to_free_space_tree(struct btrfs_trans_handle *trans, btrfs_release_path(path); if (flags & BTRFS_FREE_SPACE_USING_BITMAPS) { - return modify_free_space_bitmap(trans, fs_info, block_group, - path, start, size, 0); + return modify_free_space_bitmap(trans, block_group, path, + start, size, 0); } else { - return add_free_space_extent(trans, fs_info, block_group, path, - start, size); + return add_free_space_extent(trans, block_group, path, start, + size); } } int add_to_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u64 start, u64 size) { struct btrfs_block_group_cache *block_group; struct btrfs_path *path; int ret; - if (!btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) + if (!btrfs_fs_compat_ro(trans->fs_info, FREE_SPACE_TREE)) return 0; path = btrfs_alloc_path(); @@ -1033,7 +1014,7 @@ int add_to_free_space_tree(struct btrfs_trans_handle *trans, goto out; } - block_group = btrfs_lookup_block_group(fs_info, start); + block_group = btrfs_lookup_block_group(trans->fs_info, start); if (!block_group) { ASSERT(0); ret = -ENOENT; @@ -1041,8 +1022,7 @@ int add_to_free_space_tree(struct btrfs_trans_handle *trans, } mutex_lock(&block_group->free_space_lock); - ret = __add_to_free_space_tree(trans, fs_info, block_group, path, start, - size); + ret = __add_to_free_space_tree(trans, block_group, path, start, size); mutex_unlock(&block_group->free_space_lock); btrfs_put_block_group(block_group); @@ -1059,10 +1039,9 @@ out: * through the normal add/remove hooks. */ static int populate_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group) { - struct btrfs_root *extent_root = fs_info->extent_root; + struct btrfs_root *extent_root = trans->fs_info->extent_root; struct btrfs_path *path, *path2; struct btrfs_key key; u64 start, end; @@ -1079,7 +1058,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans, return -ENOMEM; } - ret = add_new_free_space_info(trans, fs_info, block_group, path2); + ret = add_new_free_space_info(trans, block_group, path2); if (ret) goto out; @@ -1112,7 +1091,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans, break; if (start < key.objectid) { - ret = __add_to_free_space_tree(trans, fs_info, + ret = __add_to_free_space_tree(trans, block_group, path2, start, key.objectid - @@ -1122,7 +1101,7 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans, } start = key.objectid; if (key.type == BTRFS_METADATA_ITEM_KEY) - start += fs_info->nodesize; + start += trans->fs_info->nodesize; else start += key.offset; } else if (key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) { @@ -1137,8 +1116,8 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans, break; } if (start < end) { - ret = __add_to_free_space_tree(trans, fs_info, block_group, - path2, start, end - start); + ret = __add_to_free_space_tree(trans, block_group, path2, + start, end - start); if (ret) goto out_locked; } @@ -1178,7 +1157,7 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info) while (node) { block_group = rb_entry(node, struct btrfs_block_group_cache, cache_node); - ret = populate_free_space_tree(trans, fs_info, block_group); + ret = populate_free_space_tree(trans, block_group); if (ret) goto abort; node = rb_next(node); @@ -1282,7 +1261,6 @@ abort: } static int __add_block_group_free_space(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path) { @@ -1290,19 +1268,19 @@ static int __add_block_group_free_space(struct btrfs_trans_handle *trans, block_group->needs_free_space = 0; - ret = add_new_free_space_info(trans, fs_info, block_group, path); + ret = add_new_free_space_info(trans, block_group, path); if (ret) return ret; - return __add_to_free_space_tree(trans, fs_info, block_group, path, + return __add_to_free_space_tree(trans, block_group, path, block_group->key.objectid, block_group->key.offset); } int add_block_group_free_space(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group) { + struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_path *path = NULL; int ret = 0; @@ -1319,7 +1297,7 @@ int add_block_group_free_space(struct btrfs_trans_handle *trans, goto out; } - ret = __add_block_group_free_space(trans, fs_info, block_group, path); + ret = __add_block_group_free_space(trans, block_group, path); out: btrfs_free_path(path); @@ -1330,10 +1308,9 @@ out: } int remove_block_group_free_space(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group) { - struct btrfs_root *root = fs_info->free_space_root; + struct btrfs_root *root = trans->fs_info->free_space_root; struct btrfs_path *path; struct btrfs_key key, found_key; struct extent_buffer *leaf; @@ -1341,7 +1318,7 @@ int remove_block_group_free_space(struct btrfs_trans_handle *trans, int done = 0, nr; int ret; - if (!btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) + if (!btrfs_fs_compat_ro(trans->fs_info, FREE_SPACE_TREE)) return 0; if (block_group->needs_free_space) { @@ -1452,7 +1429,6 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl, extent_start = offset; } else if (prev_bit == 1 && bit == 0) { total_found += add_new_free_space(block_group, - fs_info, extent_start, offset); if (total_found > CACHING_CTL_WAKE_UP) { @@ -1466,8 +1442,8 @@ static int load_free_space_bitmaps(struct btrfs_caching_control *caching_ctl, } } if (prev_bit == 1) { - total_found += add_new_free_space(block_group, fs_info, - extent_start, end); + total_found += add_new_free_space(block_group, extent_start, + end); extent_count++; } @@ -1524,8 +1500,7 @@ static int load_free_space_extents(struct btrfs_caching_control *caching_ctl, caching_ctl->progress = key.objectid; - total_found += add_new_free_space(block_group, fs_info, - key.objectid, + total_found += add_new_free_space(block_group, key.objectid, key.objectid + key.offset); if (total_found > CACHING_CTL_WAKE_UP) { total_found = 0; diff --git a/fs/btrfs/free-space-tree.h b/fs/btrfs/free-space-tree.h index ba3787df43c3..3133651d7d70 100644 --- a/fs/btrfs/free-space-tree.h +++ b/fs/btrfs/free-space-tree.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2015 Facebook. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_FREE_SPACE_TREE -#define __BTRFS_FREE_SPACE_TREE +#ifndef BTRFS_FREE_SPACE_TREE_H +#define BTRFS_FREE_SPACE_TREE_H /* * The default size for new free space bitmap items. The last bitmap in a block @@ -32,16 +19,12 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info); int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info); int load_free_space_tree(struct btrfs_caching_control *caching_ctl); int add_block_group_free_space(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group); int remove_block_group_free_space(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group); int add_to_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u64 start, u64 size); int remove_from_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u64 start, u64 size); #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS @@ -51,19 +34,15 @@ search_free_space_info(struct btrfs_trans_handle *trans, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, int cow); int __add_to_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, u64 start, u64 size); int __remove_from_free_space_tree(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path, u64 start, u64 size); int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path); int convert_free_space_to_extents(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, struct btrfs_block_group_cache *block_group, struct btrfs_path *path); int free_space_test_bit(struct btrfs_block_group_cache *block_group, diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c index 1d5631ef2738..a8956a3c9e05 100644 --- a/fs/btrfs/inode-item.c +++ b/fs/btrfs/inode-item.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include "ctree.h" diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c index 9409dcc7020d..12fcd8897c33 100644 --- a/fs/btrfs/inode-map.c +++ b/fs/btrfs/inode-map.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/delay.h> diff --git a/fs/btrfs/inode-map.h b/fs/btrfs/inode-map.h index 6734ec92a1e9..7a962811dffe 100644 --- a/fs/btrfs/inode-map.h +++ b/fs/btrfs/inode-map.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __BTRFS_INODE_MAP -#define __BTRFS_INODE_MAP + +#ifndef BTRFS_INODE_MAP_H +#define BTRFS_INODE_MAP_H void btrfs_init_free_ino_ctl(struct btrfs_root *root); void btrfs_unpin_free_ino(struct btrfs_root *root); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1f091c2358a4..89b208201783 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/kernel.h> @@ -44,6 +31,7 @@ #include <linux/uio.h> #include <linux/magic.h> #include <linux/iversion.h> +#include <asm/unaligned.h> #include "ctree.h" #include "disk-io.h" #include "transaction.h" @@ -1030,8 +1018,10 @@ static noinline int cow_file_range(struct inode *inode, ram_size, /* ram_bytes */ BTRFS_COMPRESS_NONE, /* compress_type */ BTRFS_ORDERED_REGULAR /* type */); - if (IS_ERR(em)) + if (IS_ERR(em)) { + ret = PTR_ERR(em); goto out_reserve; + } free_extent_map(em); ret = btrfs_add_ordered_extent(inode, start, ins.objectid, @@ -1168,13 +1158,10 @@ static noinline void async_cow_submit(struct btrfs_work *work) nr_pages = (async_cow->end - async_cow->start + PAGE_SIZE) >> PAGE_SHIFT; - /* - * atomic_sub_return implies a barrier for waitqueue_active - */ + /* atomic_sub_return implies a barrier */ if (atomic_sub_return(nr_pages, &fs_info->async_delalloc_pages) < - 5 * SZ_1M && - waitqueue_active(&fs_info->async_submit_wait)) - wake_up(&fs_info->async_submit_wait); + 5 * SZ_1M) + cond_wake_up_nomb(&fs_info->async_submit_wait); if (async_cow->inode) submit_compressed_extents(async_cow->inode, async_cow); @@ -1385,6 +1372,13 @@ next_slot: btrfs_file_extent_encryption(leaf, fi) || btrfs_file_extent_other_encoding(leaf, fi)) goto out_check; + /* + * Do the same check as in btrfs_cross_ref_exist but + * without the unnecessary search. + */ + if (btrfs_file_extent_generation(leaf, fi) <= + btrfs_root_last_snapshot(&root->root_item)) + goto out_check; if (extent_type == BTRFS_FILE_EXTENT_REG && !force) goto out_check; if (btrfs_extent_readonly(fs_info, disk_bytenr)) @@ -1754,24 +1748,32 @@ static void btrfs_add_delalloc_inodes(struct btrfs_root *root, spin_unlock(&root->delalloc_lock); } -static void btrfs_del_delalloc_inode(struct btrfs_root *root, - struct btrfs_inode *inode) + +void __btrfs_del_delalloc_inode(struct btrfs_root *root, + struct btrfs_inode *inode) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); - spin_lock(&root->delalloc_lock); if (!list_empty(&inode->delalloc_inodes)) { list_del_init(&inode->delalloc_inodes); clear_bit(BTRFS_INODE_IN_DELALLOC_LIST, &inode->runtime_flags); root->nr_delalloc_inodes--; if (!root->nr_delalloc_inodes) { + ASSERT(list_empty(&root->delalloc_inodes)); spin_lock(&fs_info->delalloc_root_lock); BUG_ON(list_empty(&root->delalloc_root)); list_del_init(&root->delalloc_root); spin_unlock(&fs_info->delalloc_root_lock); } } +} + +static void btrfs_del_delalloc_inode(struct btrfs_root *root, + struct btrfs_inode *inode) +{ + spin_lock(&root->delalloc_lock); + __btrfs_del_delalloc_inode(root, inode); spin_unlock(&root->delalloc_lock); } @@ -3163,6 +3165,9 @@ out: /* once for the tree */ btrfs_put_ordered_extent(ordered_extent); + /* Try to release some metadata so we don't get an OOM but don't wait */ + btrfs_btree_balance_dirty_nodelay(fs_info); + return ret; } @@ -3305,177 +3310,31 @@ void btrfs_run_delayed_iputs(struct btrfs_fs_info *fs_info) } /* - * This is called in transaction commit time. If there are no orphan - * files in the subvolume, it removes orphan item and frees block_rsv - * structure. - */ -void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, - struct btrfs_root *root) -{ - struct btrfs_fs_info *fs_info = root->fs_info; - struct btrfs_block_rsv *block_rsv; - int ret; - - if (atomic_read(&root->orphan_inodes) || - root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE) - return; - - spin_lock(&root->orphan_lock); - if (atomic_read(&root->orphan_inodes)) { - spin_unlock(&root->orphan_lock); - return; - } - - if (root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE) { - spin_unlock(&root->orphan_lock); - return; - } - - block_rsv = root->orphan_block_rsv; - root->orphan_block_rsv = NULL; - spin_unlock(&root->orphan_lock); - - if (test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state) && - btrfs_root_refs(&root->root_item) > 0) { - ret = btrfs_del_orphan_item(trans, fs_info->tree_root, - root->root_key.objectid); - if (ret) - btrfs_abort_transaction(trans, ret); - else - clear_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, - &root->state); - } - - if (block_rsv) { - WARN_ON(block_rsv->size > 0); - btrfs_free_block_rsv(fs_info, block_rsv); - } -} - -/* - * This creates an orphan entry for the given inode in case something goes - * wrong in the middle of an unlink/truncate. - * - * NOTE: caller of this function should reserve 5 units of metadata for - * this function. + * This creates an orphan entry for the given inode in case something goes wrong + * in the middle of an unlink. */ int btrfs_orphan_add(struct btrfs_trans_handle *trans, - struct btrfs_inode *inode) + struct btrfs_inode *inode) { - struct btrfs_fs_info *fs_info = btrfs_sb(inode->vfs_inode.i_sb); - struct btrfs_root *root = inode->root; - struct btrfs_block_rsv *block_rsv = NULL; - int reserve = 0; - bool insert = false; int ret; - if (!root->orphan_block_rsv) { - block_rsv = btrfs_alloc_block_rsv(fs_info, - BTRFS_BLOCK_RSV_TEMP); - if (!block_rsv) - return -ENOMEM; - } - - if (!test_and_set_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &inode->runtime_flags)) - insert = true; - - if (!test_and_set_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &inode->runtime_flags)) - reserve = 1; - - spin_lock(&root->orphan_lock); - /* If someone has created ->orphan_block_rsv, be happy to use it. */ - if (!root->orphan_block_rsv) { - root->orphan_block_rsv = block_rsv; - } else if (block_rsv) { - btrfs_free_block_rsv(fs_info, block_rsv); - block_rsv = NULL; - } - - if (insert) - atomic_inc(&root->orphan_inodes); - spin_unlock(&root->orphan_lock); - - /* grab metadata reservation from transaction handle */ - if (reserve) { - ret = btrfs_orphan_reserve_metadata(trans, inode); - ASSERT(!ret); - if (ret) { - /* - * dec doesn't need spin_lock as ->orphan_block_rsv - * would be released only if ->orphan_inodes is - * zero. - */ - atomic_dec(&root->orphan_inodes); - clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &inode->runtime_flags); - if (insert) - clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &inode->runtime_flags); - return ret; - } - } - - /* insert an orphan item to track this unlinked/truncated file */ - if (insert) { - ret = btrfs_insert_orphan_item(trans, root, btrfs_ino(inode)); - if (ret) { - if (reserve) { - clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &inode->runtime_flags); - btrfs_orphan_release_metadata(inode); - } - /* - * btrfs_orphan_commit_root may race with us and set - * ->orphan_block_rsv to zero, in order to avoid that, - * decrease ->orphan_inodes after everything is done. - */ - atomic_dec(&root->orphan_inodes); - if (ret != -EEXIST) { - clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &inode->runtime_flags); - btrfs_abort_transaction(trans, ret); - return ret; - } - } - ret = 0; + ret = btrfs_insert_orphan_item(trans, inode->root, btrfs_ino(inode)); + if (ret && ret != -EEXIST) { + btrfs_abort_transaction(trans, ret); + return ret; } return 0; } /* - * We have done the truncate/delete so we can go ahead and remove the orphan - * item for this particular inode. + * We have done the delete so we can go ahead and remove the orphan item for + * this particular inode. */ static int btrfs_orphan_del(struct btrfs_trans_handle *trans, struct btrfs_inode *inode) { - struct btrfs_root *root = inode->root; - int delete_item = 0; - int ret = 0; - - if (test_and_clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &inode->runtime_flags)) - delete_item = 1; - - if (delete_item && trans) - ret = btrfs_del_orphan_item(trans, root, btrfs_ino(inode)); - - if (test_and_clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED, - &inode->runtime_flags)) - btrfs_orphan_release_metadata(inode); - - /* - * btrfs_orphan_commit_root may race with us and set ->orphan_block_rsv - * to zero, in order to avoid that, decrease ->orphan_inodes after - * everything is done. - */ - if (delete_item) - atomic_dec(&root->orphan_inodes); - - return ret; + return btrfs_del_orphan_item(trans, inode->root, btrfs_ino(inode)); } /* @@ -3491,7 +3350,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) struct btrfs_trans_handle *trans; struct inode *inode; u64 last_objectid = 0; - int ret = 0, nr_unlink = 0, nr_truncate = 0; + int ret = 0, nr_unlink = 0; if (cmpxchg(&root->orphan_cleanup_state, 0, ORPHAN_CLEANUP_STARTED)) return 0; @@ -3591,12 +3450,31 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) key.offset = found_key.objectid - 1; continue; } + } + /* - * Inode is already gone but the orphan item is still there, - * kill the orphan item. + * If we have an inode with links, there are a couple of + * possibilities. Old kernels (before v3.12) used to create an + * orphan item for truncate indicating that there were possibly + * extent items past i_size that needed to be deleted. In v3.12, + * truncate was changed to update i_size in sync with the extent + * items, but the (useless) orphan item was still created. Since + * v4.18, we don't create the orphan item for truncate at all. + * + * So, this item could mean that we need to do a truncate, but + * only if this filesystem was last used on a pre-v3.12 kernel + * and was not cleanly unmounted. The odds of that are quite + * slim, and it's a pain to do the truncate now, so just delete + * the orphan item. + * + * It's also possible that this orphan item was supposed to be + * deleted but wasn't. The inode number may have been reused, + * but either way, we can delete the orphan item. */ - if (ret == -ENOENT) { + if (ret == -ENOENT || inode->i_nlink) { + if (!ret) + iput(inode); trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { ret = PTR_ERR(trans); @@ -3612,42 +3490,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) continue; } - /* - * add this inode to the orphan list so btrfs_orphan_del does - * the proper thing when we hit it - */ - set_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &BTRFS_I(inode)->runtime_flags); - atomic_inc(&root->orphan_inodes); - - /* if we have links, this was a truncate, lets do that */ - if (inode->i_nlink) { - if (WARN_ON(!S_ISREG(inode->i_mode))) { - iput(inode); - continue; - } - nr_truncate++; - - /* 1 for the orphan item deletion. */ - trans = btrfs_start_transaction(root, 1); - if (IS_ERR(trans)) { - iput(inode); - ret = PTR_ERR(trans); - goto out; - } - ret = btrfs_orphan_add(trans, BTRFS_I(inode)); - btrfs_end_transaction(trans); - if (ret) { - iput(inode); - goto out; - } - - ret = btrfs_truncate(inode, false); - if (ret) - btrfs_orphan_del(NULL, BTRFS_I(inode)); - } else { - nr_unlink++; - } + nr_unlink++; /* this will do delete_inode and everything for us */ iput(inode); @@ -3659,12 +3502,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) root->orphan_cleanup_state = ORPHAN_CLEANUP_DONE; - if (root->orphan_block_rsv) - btrfs_block_rsv_release(fs_info, root->orphan_block_rsv, - (u64)-1); - - if (root->orphan_block_rsv || - test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state)) { + if (test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state)) { trans = btrfs_join_transaction(root); if (!IS_ERR(trans)) btrfs_end_transaction(trans); @@ -3672,8 +3510,6 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) if (nr_unlink) btrfs_debug(fs_info, "unlinked %d orphans", nr_unlink); - if (nr_truncate) - btrfs_debug(fs_info, "truncated %d orphans", nr_truncate); out: if (ret) @@ -3936,7 +3772,7 @@ cache_acl: break; } - btrfs_update_iflags(inode); + btrfs_sync_inode_flags_to_i_flags(inode); return 0; make_bad: @@ -4250,7 +4086,7 @@ out: return ret; } -int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, +static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *dir, u64 objectid, const char *name, int name_len) @@ -4331,6 +4167,262 @@ out: return ret; } +/* + * Helper to check if the subvolume references other subvolumes or if it's + * default. + */ +static noinline int may_destroy_subvol(struct btrfs_root *root) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct btrfs_path *path; + struct btrfs_dir_item *di; + struct btrfs_key key; + u64 dir_id; + int ret; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + /* Make sure this root isn't set as the default subvol */ + dir_id = btrfs_super_root_dir(fs_info->super_copy); + di = btrfs_lookup_dir_item(NULL, fs_info->tree_root, path, + dir_id, "default", 7, 0); + if (di && !IS_ERR(di)) { + btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key); + if (key.objectid == root->root_key.objectid) { + ret = -EPERM; + btrfs_err(fs_info, + "deleting default subvolume %llu is not allowed", + key.objectid); + goto out; + } + btrfs_release_path(path); + } + + key.objectid = root->root_key.objectid; + key.type = BTRFS_ROOT_REF_KEY; + key.offset = (u64)-1; + + ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0); + if (ret < 0) + goto out; + BUG_ON(ret == 0); + + ret = 0; + if (path->slots[0] > 0) { + path->slots[0]--; + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); + if (key.objectid == root->root_key.objectid && + key.type == BTRFS_ROOT_REF_KEY) + ret = -ENOTEMPTY; + } +out: + btrfs_free_path(path); + return ret; +} + +/* Delete all dentries for inodes belonging to the root */ +static void btrfs_prune_dentries(struct btrfs_root *root) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct rb_node *node; + struct rb_node *prev; + struct btrfs_inode *entry; + struct inode *inode; + u64 objectid = 0; + + if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) + WARN_ON(btrfs_root_refs(&root->root_item) != 0); + + spin_lock(&root->inode_lock); +again: + node = root->inode_tree.rb_node; + prev = NULL; + while (node) { + prev = node; + entry = rb_entry(node, struct btrfs_inode, rb_node); + + if (objectid < btrfs_ino(BTRFS_I(&entry->vfs_inode))) + node = node->rb_left; + else if (objectid > btrfs_ino(BTRFS_I(&entry->vfs_inode))) + node = node->rb_right; + else + break; + } + if (!node) { + while (prev) { + entry = rb_entry(prev, struct btrfs_inode, rb_node); + if (objectid <= btrfs_ino(BTRFS_I(&entry->vfs_inode))) { + node = prev; + break; + } + prev = rb_next(prev); + } + } + while (node) { + entry = rb_entry(node, struct btrfs_inode, rb_node); + objectid = btrfs_ino(BTRFS_I(&entry->vfs_inode)) + 1; + inode = igrab(&entry->vfs_inode); + if (inode) { + spin_unlock(&root->inode_lock); + if (atomic_read(&inode->i_count) > 1) + d_prune_aliases(inode); + /* + * btrfs_drop_inode will have it removed from the inode + * cache when its usage count hits zero. + */ + iput(inode); + cond_resched(); + spin_lock(&root->inode_lock); + goto again; + } + + if (cond_resched_lock(&root->inode_lock)) + goto again; + + node = rb_next(node); + } + spin_unlock(&root->inode_lock); +} + +int btrfs_delete_subvolume(struct inode *dir, struct dentry *dentry) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(dentry->d_sb); + struct btrfs_root *root = BTRFS_I(dir)->root; + struct inode *inode = d_inode(dentry); + struct btrfs_root *dest = BTRFS_I(inode)->root; + struct btrfs_trans_handle *trans; + struct btrfs_block_rsv block_rsv; + u64 root_flags; + int ret; + int err; + + /* + * Don't allow to delete a subvolume with send in progress. This is + * inside the inode lock so the error handling that has to drop the bit + * again is not run concurrently. + */ + spin_lock(&dest->root_item_lock); + root_flags = btrfs_root_flags(&dest->root_item); + if (dest->send_in_progress == 0) { + btrfs_set_root_flags(&dest->root_item, + root_flags | BTRFS_ROOT_SUBVOL_DEAD); + spin_unlock(&dest->root_item_lock); + } else { + spin_unlock(&dest->root_item_lock); + btrfs_warn(fs_info, + "attempt to delete subvolume %llu during send", + dest->root_key.objectid); + return -EPERM; + } + + down_write(&fs_info->subvol_sem); + + err = may_destroy_subvol(dest); + if (err) + goto out_up_write; + + btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP); + /* + * One for dir inode, + * two for dir entries, + * two for root ref/backref. + */ + err = btrfs_subvolume_reserve_metadata(root, &block_rsv, 5, true); + if (err) + goto out_up_write; + + trans = btrfs_start_transaction(root, 0); + if (IS_ERR(trans)) { + err = PTR_ERR(trans); + goto out_release; + } + trans->block_rsv = &block_rsv; + trans->bytes_reserved = block_rsv.size; + + btrfs_record_snapshot_destroy(trans, BTRFS_I(dir)); + + ret = btrfs_unlink_subvol(trans, root, dir, + dest->root_key.objectid, + dentry->d_name.name, + dentry->d_name.len); + if (ret) { + err = ret; + btrfs_abort_transaction(trans, ret); + goto out_end_trans; + } + + btrfs_record_root_in_trans(trans, dest); + + memset(&dest->root_item.drop_progress, 0, + sizeof(dest->root_item.drop_progress)); + dest->root_item.drop_level = 0; + btrfs_set_root_refs(&dest->root_item, 0); + + if (!test_and_set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &dest->state)) { + ret = btrfs_insert_orphan_item(trans, + fs_info->tree_root, + dest->root_key.objectid); + if (ret) { + btrfs_abort_transaction(trans, ret); + err = ret; + goto out_end_trans; + } + } + + ret = btrfs_uuid_tree_remove(trans, dest->root_item.uuid, + BTRFS_UUID_KEY_SUBVOL, + dest->root_key.objectid); + if (ret && ret != -ENOENT) { + btrfs_abort_transaction(trans, ret); + err = ret; + goto out_end_trans; + } + if (!btrfs_is_empty_uuid(dest->root_item.received_uuid)) { + ret = btrfs_uuid_tree_remove(trans, + dest->root_item.received_uuid, + BTRFS_UUID_KEY_RECEIVED_SUBVOL, + dest->root_key.objectid); + if (ret && ret != -ENOENT) { + btrfs_abort_transaction(trans, ret); + err = ret; + goto out_end_trans; + } + } + +out_end_trans: + trans->block_rsv = NULL; + trans->bytes_reserved = 0; + ret = btrfs_end_transaction(trans); + if (ret && !err) + err = ret; + inode->i_flags |= S_DEAD; +out_release: + btrfs_subvolume_release_metadata(fs_info, &block_rsv); +out_up_write: + up_write(&fs_info->subvol_sem); + if (err) { + spin_lock(&dest->root_item_lock); + root_flags = btrfs_root_flags(&dest->root_item); + btrfs_set_root_flags(&dest->root_item, + root_flags & ~BTRFS_ROOT_SUBVOL_DEAD); + spin_unlock(&dest->root_item_lock); + } else { + d_invalidate(dentry); + btrfs_prune_dentries(dest); + ASSERT(dest->send_in_progress == 0); + + /* the last ref */ + if (dest->ino_cache_inode) { + iput(dest->ino_cache_inode); + dest->ino_cache_inode = NULL; + } + } + + return err; +} + static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) { struct inode *inode = d_inode(dentry); @@ -4342,7 +4434,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) if (inode->i_size > BTRFS_EMPTY_DIR_SIZE) return -ENOTEMPTY; if (btrfs_ino(BTRFS_I(inode)) == BTRFS_FIRST_FREE_OBJECTID) - return -EPERM; + return btrfs_delete_subvolume(dir, dentry); trans = __unlink_start_trans(dir); if (IS_ERR(trans)) @@ -4454,7 +4546,6 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, int pending_del_slot = 0; int extent_type = -1; int ret; - int err = 0; u64 ino = btrfs_ino(BTRFS_I(inode)); u64 bytes_deleted = 0; bool be_nice = false; @@ -4506,22 +4597,19 @@ search_again: * up a huge file in a single leaf. Most of the time that * bytes_deleted is > 0, it will be huge by the time we get here */ - if (be_nice && bytes_deleted > SZ_32M) { - if (btrfs_should_end_transaction(trans)) { - err = -EAGAIN; - goto error; - } + if (be_nice && bytes_deleted > SZ_32M && + btrfs_should_end_transaction(trans)) { + ret = -EAGAIN; + goto out; } - path->leave_spinning = 1; ret = btrfs_search_slot(trans, root, &key, path, -1, 1); - if (ret < 0) { - err = ret; + if (ret < 0) goto out; - } if (ret > 0) { + ret = 0; /* there are no items in the tree for us to truncate, we're * done */ @@ -4632,7 +4720,7 @@ search_again: * We have to bail so the last_size is set to * just before this extent. */ - err = NEED_TRUNCATE_BLOCK; + ret = NEED_TRUNCATE_BLOCK; break; } @@ -4671,7 +4759,10 @@ delete: extent_num_bytes, 0, btrfs_header_owner(leaf), ino, extent_offset); - BUG_ON(ret); + if (ret) { + btrfs_abort_transaction(trans, ret); + break; + } if (btrfs_should_throttle_delayed_refs(trans, fs_info)) btrfs_async_run_delayed_refs(fs_info, trans->delayed_ref_updates * 2, @@ -4699,7 +4790,7 @@ delete: pending_del_nr); if (ret) { btrfs_abort_transaction(trans, ret); - goto error; + break; } pending_del_nr = 0; } @@ -4710,8 +4801,8 @@ delete: trans->delayed_ref_updates = 0; ret = btrfs_run_delayed_refs(trans, updates * 2); - if (ret && !err) - err = ret; + if (ret) + break; } } /* @@ -4719,8 +4810,8 @@ delete: * and let the transaction restart */ if (should_end) { - err = -EAGAIN; - goto error; + ret = -EAGAIN; + break; } goto search_again; } else { @@ -4728,32 +4819,37 @@ delete: } } out: - if (pending_del_nr) { - ret = btrfs_del_items(trans, root, path, pending_del_slot, + if (ret >= 0 && pending_del_nr) { + int err; + + err = btrfs_del_items(trans, root, path, pending_del_slot, pending_del_nr); - if (ret) - btrfs_abort_transaction(trans, ret); + if (err) { + btrfs_abort_transaction(trans, err); + ret = err; + } } -error: if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { ASSERT(last_size >= new_size); - if (!err && last_size > new_size) + if (!ret && last_size > new_size) last_size = new_size; btrfs_ordered_update_i_size(inode, last_size, NULL); } btrfs_free_path(path); - if (be_nice && bytes_deleted > SZ_32M) { + if (be_nice && bytes_deleted > SZ_32M && (ret >= 0 || ret == -EAGAIN)) { unsigned long updates = trans->delayed_ref_updates; + int err; + if (updates) { trans->delayed_ref_updates = 0; - ret = btrfs_run_delayed_refs(trans, updates * 2); - if (ret && !err) - err = ret; + err = btrfs_run_delayed_refs(trans, updates * 2); + if (err) + ret = err; } } - return err; + return ret; } /* @@ -5095,30 +5191,6 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) set_bit(BTRFS_INODE_ORDERED_DATA_CLOSE, &BTRFS_I(inode)->runtime_flags); - /* - * 1 for the orphan item we're going to add - * 1 for the orphan item deletion. - */ - trans = btrfs_start_transaction(root, 2); - if (IS_ERR(trans)) - return PTR_ERR(trans); - - /* - * We need to do this in case we fail at _any_ point during the - * actual truncate. Once we do the truncate_setsize we could - * invalidate pages which forces any outstanding ordered io to - * be instantly completed which will give us extents that need - * to be truncated. If we fail to get an orphan inode down we - * could have left over extents that were never meant to live, - * so we need to guarantee from this point on that everything - * will be consistent. - */ - ret = btrfs_orphan_add(trans, BTRFS_I(inode)); - btrfs_end_transaction(trans); - if (ret) - return ret; - - /* we don't support swapfiles, so vmtruncate shouldn't fail */ truncate_setsize(inode, newsize); /* Disable nonlocked read DIO to avoid the end less truncate */ @@ -5130,29 +5202,16 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) if (ret && inode->i_nlink) { int err; - /* To get a stable disk_i_size */ - err = btrfs_wait_ordered_range(inode, 0, (u64)-1); - if (err) { - btrfs_orphan_del(NULL, BTRFS_I(inode)); - return err; - } - /* - * failed to truncate, disk_i_size is only adjusted down - * as we remove extents, so it should represent the true - * size of the inode, so reset the in memory size and - * delete our orphan entry. + * Truncate failed, so fix up the in-memory size. We + * adjusted disk_i_size down as we removed extents, so + * wait for disk_i_size to be stable and then update the + * in-memory size to match. */ - trans = btrfs_join_transaction(root); - if (IS_ERR(trans)) { - btrfs_orphan_del(NULL, BTRFS_I(inode)); - return ret; - } - i_size_write(inode, BTRFS_I(inode)->disk_i_size); - err = btrfs_orphan_del(trans, BTRFS_I(inode)); + err = btrfs_wait_ordered_range(inode, 0, (u64)-1); if (err) - btrfs_abort_transaction(trans, err); - btrfs_end_transaction(trans); + return err; + i_size_write(inode, BTRFS_I(inode)->disk_i_size); } } @@ -5282,13 +5341,52 @@ static void evict_inode_truncate_pages(struct inode *inode) spin_unlock(&io_tree->lock); } +static struct btrfs_trans_handle *evict_refill_and_join(struct btrfs_root *root, + struct btrfs_block_rsv *rsv, + u64 min_size) +{ + struct btrfs_fs_info *fs_info = root->fs_info; + struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv; + int failures = 0; + + for (;;) { + struct btrfs_trans_handle *trans; + int ret; + + ret = btrfs_block_rsv_refill(root, rsv, min_size, + BTRFS_RESERVE_FLUSH_LIMIT); + + if (ret && ++failures > 2) { + btrfs_warn(fs_info, + "could not allocate space for a delete; will truncate on mount"); + return ERR_PTR(-ENOSPC); + } + + trans = btrfs_join_transaction(root); + if (IS_ERR(trans) || !ret) + return trans; + + /* + * Try to steal from the global reserve if there is space for + * it. + */ + if (!btrfs_check_space_for_delayed_refs(trans, fs_info) && + !btrfs_block_rsv_migrate(global_rsv, rsv, min_size, 0)) + return trans; + + /* If not, commit and try again. */ + ret = btrfs_commit_transaction(trans); + if (ret) + return ERR_PTR(ret); + } +} + void btrfs_evict_inode(struct inode *inode) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_block_rsv *rsv, *global_rsv; - int steal_from_global = 0; + struct btrfs_block_rsv *rsv; u64 min_size; int ret; @@ -5309,21 +5407,16 @@ void btrfs_evict_inode(struct inode *inode) btrfs_is_free_space_inode(BTRFS_I(inode)))) goto no_delete; - if (is_bad_inode(inode)) { - btrfs_orphan_del(NULL, BTRFS_I(inode)); + if (is_bad_inode(inode)) goto no_delete; - } /* do we really want it for ->i_nlink > 0 and zero btrfs_root_refs? */ if (!special_file(inode->i_mode)) btrfs_wait_ordered_range(inode, 0, (u64)-1); btrfs_free_io_failure_record(BTRFS_I(inode), 0, (u64)-1); - if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) { - BUG_ON(test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &BTRFS_I(inode)->runtime_flags)); + if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) goto no_delete; - } if (inode->i_nlink > 0) { BUG_ON(btrfs_root_refs(&root->root_item) != 0 && @@ -5332,130 +5425,63 @@ void btrfs_evict_inode(struct inode *inode) } ret = btrfs_commit_inode_delayed_inode(BTRFS_I(inode)); - if (ret) { - btrfs_orphan_del(NULL, BTRFS_I(inode)); + if (ret) goto no_delete; - } rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP); - if (!rsv) { - btrfs_orphan_del(NULL, BTRFS_I(inode)); + if (!rsv) goto no_delete; - } rsv->size = min_size; rsv->failfast = 1; - global_rsv = &fs_info->global_block_rsv; btrfs_i_size_write(BTRFS_I(inode), 0); - /* - * This is a bit simpler than btrfs_truncate since we've already - * reserved our space for our orphan item in the unlink, so we just - * need to reserve some slack space in case we add bytes and update - * inode item when doing the truncate. - */ while (1) { - ret = btrfs_block_rsv_refill(root, rsv, min_size, - BTRFS_RESERVE_FLUSH_LIMIT); - - /* - * Try and steal from the global reserve since we will - * likely not use this space anyway, we want to try as - * hard as possible to get this to work. - */ - if (ret) - steal_from_global++; - else - steal_from_global = 0; - ret = 0; - - /* - * steal_from_global == 0: we reserved stuff, hooray! - * steal_from_global == 1: we didn't reserve stuff, boo! - * steal_from_global == 2: we've committed, still not a lot of - * room but maybe we'll have room in the global reserve this - * time. - * steal_from_global == 3: abandon all hope! - */ - if (steal_from_global > 2) { - btrfs_warn(fs_info, - "Could not get space for a delete, will truncate on mount %d", - ret); - btrfs_orphan_del(NULL, BTRFS_I(inode)); - btrfs_free_block_rsv(fs_info, rsv); - goto no_delete; - } - - trans = btrfs_join_transaction(root); - if (IS_ERR(trans)) { - btrfs_orphan_del(NULL, BTRFS_I(inode)); - btrfs_free_block_rsv(fs_info, rsv); - goto no_delete; - } - - /* - * We can't just steal from the global reserve, we need to make - * sure there is room to do it, if not we need to commit and try - * again. - */ - if (steal_from_global) { - if (!btrfs_check_space_for_delayed_refs(trans, fs_info)) - ret = btrfs_block_rsv_migrate(global_rsv, rsv, - min_size, 0); - else - ret = -ENOSPC; - } - - /* - * Couldn't steal from the global reserve, we have too much - * pending stuff built up, commit the transaction and try it - * again. - */ - if (ret) { - ret = btrfs_commit_transaction(trans); - if (ret) { - btrfs_orphan_del(NULL, BTRFS_I(inode)); - btrfs_free_block_rsv(fs_info, rsv); - goto no_delete; - } - continue; - } else { - steal_from_global = 0; - } + trans = evict_refill_and_join(root, rsv, min_size); + if (IS_ERR(trans)) + goto free_rsv; trans->block_rsv = rsv; ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0); - if (ret != -ENOSPC && ret != -EAGAIN) - break; - trans->block_rsv = &fs_info->trans_block_rsv; btrfs_end_transaction(trans); - trans = NULL; btrfs_btree_balance_dirty(fs_info); + if (ret && ret != -ENOSPC && ret != -EAGAIN) + goto free_rsv; + else if (!ret) + break; } - btrfs_free_block_rsv(fs_info, rsv); - /* - * Errors here aren't a big deal, it just means we leave orphan items - * in the tree. They will be cleaned up on the next mount. + * Errors here aren't a big deal, it just means we leave orphan items in + * the tree. They will be cleaned up on the next mount. If the inode + * number gets reused, cleanup deletes the orphan item without doing + * anything, and unlink reuses the existing orphan item. + * + * If it turns out that we are dropping too many of these, we might want + * to add a mechanism for retrying these after a commit. */ - if (ret == 0) { - trans->block_rsv = root->orphan_block_rsv; + trans = evict_refill_and_join(root, rsv, min_size); + if (!IS_ERR(trans)) { + trans->block_rsv = rsv; btrfs_orphan_del(trans, BTRFS_I(inode)); - } else { - btrfs_orphan_del(NULL, BTRFS_I(inode)); + trans->block_rsv = &fs_info->trans_block_rsv; + btrfs_end_transaction(trans); } - trans->block_rsv = &fs_info->trans_block_rsv; if (!(root == fs_info->tree_root || root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)) btrfs_return_ino(root, btrfs_ino(BTRFS_I(inode))); - btrfs_end_transaction(trans); - btrfs_btree_balance_dirty(fs_info); +free_rsv: + btrfs_free_block_rsv(fs_info, rsv); no_delete: + /* + * If we didn't successfully delete, the orphan item will still be in + * the tree and we'll retry on the next mount. Again, we might also want + * to retry these periodically in the future. + */ btrfs_remove_delayed_node(BTRFS_I(inode)); clear_inode(inode); } @@ -5631,69 +5657,6 @@ static void inode_tree_del(struct inode *inode) } } -void btrfs_invalidate_inodes(struct btrfs_root *root) -{ - struct btrfs_fs_info *fs_info = root->fs_info; - struct rb_node *node; - struct rb_node *prev; - struct btrfs_inode *entry; - struct inode *inode; - u64 objectid = 0; - - if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) - WARN_ON(btrfs_root_refs(&root->root_item) != 0); - - spin_lock(&root->inode_lock); -again: - node = root->inode_tree.rb_node; - prev = NULL; - while (node) { - prev = node; - entry = rb_entry(node, struct btrfs_inode, rb_node); - - if (objectid < btrfs_ino(BTRFS_I(&entry->vfs_inode))) - node = node->rb_left; - else if (objectid > btrfs_ino(BTRFS_I(&entry->vfs_inode))) - node = node->rb_right; - else - break; - } - if (!node) { - while (prev) { - entry = rb_entry(prev, struct btrfs_inode, rb_node); - if (objectid <= btrfs_ino(BTRFS_I(&entry->vfs_inode))) { - node = prev; - break; - } - prev = rb_next(prev); - } - } - while (node) { - entry = rb_entry(node, struct btrfs_inode, rb_node); - objectid = btrfs_ino(BTRFS_I(&entry->vfs_inode)) + 1; - inode = igrab(&entry->vfs_inode); - if (inode) { - spin_unlock(&root->inode_lock); - if (atomic_read(&inode->i_count) > 1) - d_prune_aliases(inode); - /* - * btrfs_drop_inode will have it removed from - * the inode cache when its usage count - * hits zero. - */ - iput(inode); - cond_resched(); - spin_lock(&root->inode_lock); - goto again; - } - - if (cond_resched_lock(&root->inode_lock)) - goto again; - - node = rb_next(node); - } - spin_unlock(&root->inode_lock); -} static int btrfs_init_locked_inode(struct inode *inode, void *p) { @@ -5855,11 +5818,6 @@ static int btrfs_dentry_delete(const struct dentry *dentry) return 0; } -static void btrfs_dentry_release(struct dentry *dentry) -{ - kfree(dentry->d_fsdata); -} - static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { @@ -5918,11 +5876,13 @@ static int btrfs_filldir(void *addr, int entries, struct dir_context *ctx) struct dir_entry *entry = addr; char *name = (char *)(entry + 1); - ctx->pos = entry->offset; - if (!dir_emit(ctx, name, entry->name_len, entry->ino, - entry->type)) + ctx->pos = get_unaligned(&entry->offset); + if (!dir_emit(ctx, name, get_unaligned(&entry->name_len), + get_unaligned(&entry->ino), + get_unaligned(&entry->type))) return 1; - addr += sizeof(struct dir_entry) + entry->name_len; + addr += sizeof(struct dir_entry) + + get_unaligned(&entry->name_len); ctx->pos++; } return 0; @@ -6012,14 +5972,15 @@ again: } entry = addr; - entry->name_len = name_len; + put_unaligned(name_len, &entry->name_len); name_ptr = (char *)(entry + 1); read_extent_buffer(leaf, name_ptr, (unsigned long)(di + 1), name_len); - entry->type = btrfs_filetype_table[btrfs_dir_type(leaf, di)]; + put_unaligned(btrfs_filetype_table[btrfs_dir_type(leaf, di)], + &entry->type); btrfs_dir_item_key_to_cpu(leaf, di, &location); - entry->ino = location.objectid; - entry->offset = found_key.offset; + put_unaligned(location.objectid, &entry->ino); + put_unaligned(found_key.offset, &entry->offset); entries++; addr += sizeof(struct dir_entry) + name_len; total_len += sizeof(struct dir_entry) + name_len; @@ -6272,7 +6233,7 @@ static void btrfs_inherit_iflags(struct inode *inode, struct inode *dir) BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM; } - btrfs_update_iflags(inode); + btrfs_sync_inode_flags_to_i_flags(inode); } static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, @@ -6588,8 +6549,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; } else { btrfs_update_inode(trans, root, inode); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); } out_unlock: @@ -6665,8 +6625,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out_unlock: btrfs_end_transaction(trans); @@ -6709,8 +6668,9 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, * 2 items for inode and inode ref * 2 items for dir items * 1 item for parent inode + * 1 item for orphan item deletion if O_TMPFILE */ - trans = btrfs_start_transaction(root, 5); + trans = btrfs_start_transaction(root, inode->i_nlink ? 5 : 6); if (IS_ERR(trans)) { err = PTR_ERR(trans); trans = NULL; @@ -6811,12 +6771,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) if (err) goto out_fail_inode; - d_instantiate(dentry, inode); - /* - * mkdir is special. We're unlocking after we call d_instantiate - * to avoid a race with nfsd calling d_instantiate. - */ - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); drop_on_err = 0; out_fail: @@ -7092,7 +7047,7 @@ insert: err = 0; write_lock(&em_tree->lock); - err = btrfs_add_extent_mapping(em_tree, &em, start, len); + err = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); write_unlock(&em_tree->lock); out: @@ -7377,6 +7332,14 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, btrfs_file_extent_other_encoding(leaf, fi)) goto out; + /* + * Do the same check as in btrfs_cross_ref_exist but without the + * unnecessary search. + */ + if (btrfs_file_extent_generation(leaf, fi) <= + btrfs_root_last_snapshot(&root->root_item)) + goto out; + backref_offset = btrfs_file_extent_offset(leaf, fi); if (orig_start) { @@ -7577,6 +7540,125 @@ static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len, return em; } + +static int btrfs_get_blocks_direct_read(struct extent_map *em, + struct buffer_head *bh_result, + struct inode *inode, + u64 start, u64 len) +{ + if (em->block_start == EXTENT_MAP_HOLE || + test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) + return -ENOENT; + + len = min(len, em->len - (start - em->start)); + + bh_result->b_blocknr = (em->block_start + (start - em->start)) >> + inode->i_blkbits; + bh_result->b_size = len; + bh_result->b_bdev = em->bdev; + set_buffer_mapped(bh_result); + + return 0; +} + +static int btrfs_get_blocks_direct_write(struct extent_map **map, + struct buffer_head *bh_result, + struct inode *inode, + struct btrfs_dio_data *dio_data, + u64 start, u64 len) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct extent_map *em = *map; + int ret = 0; + + /* + * We don't allocate a new extent in the following cases + * + * 1) The inode is marked as NODATACOW. In this case we'll just use the + * existing extent. + * 2) The extent is marked as PREALLOC. We're good to go here and can + * just use the extent. + * + */ + if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) || + ((BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) && + em->block_start != EXTENT_MAP_HOLE)) { + int type; + u64 block_start, orig_start, orig_block_len, ram_bytes; + + if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) + type = BTRFS_ORDERED_PREALLOC; + else + type = BTRFS_ORDERED_NOCOW; + len = min(len, em->len - (start - em->start)); + block_start = em->block_start + (start - em->start); + + if (can_nocow_extent(inode, start, &len, &orig_start, + &orig_block_len, &ram_bytes) == 1 && + btrfs_inc_nocow_writers(fs_info, block_start)) { + struct extent_map *em2; + + em2 = btrfs_create_dio_extent(inode, start, len, + orig_start, block_start, + len, orig_block_len, + ram_bytes, type); + btrfs_dec_nocow_writers(fs_info, block_start); + if (type == BTRFS_ORDERED_PREALLOC) { + free_extent_map(em); + *map = em = em2; + } + + if (em2 && IS_ERR(em2)) { + ret = PTR_ERR(em2); + goto out; + } + /* + * For inode marked NODATACOW or extent marked PREALLOC, + * use the existing or preallocated extent, so does not + * need to adjust btrfs_space_info's bytes_may_use. + */ + btrfs_free_reserved_data_space_noquota(inode, start, + len); + goto skip_cow; + } + } + + /* this will cow the extent */ + len = bh_result->b_size; + free_extent_map(em); + *map = em = btrfs_new_extent_direct(inode, start, len); + if (IS_ERR(em)) { + ret = PTR_ERR(em); + goto out; + } + + len = min(len, em->len - (start - em->start)); + +skip_cow: + bh_result->b_blocknr = (em->block_start + (start - em->start)) >> + inode->i_blkbits; + bh_result->b_size = len; + bh_result->b_bdev = em->bdev; + set_buffer_mapped(bh_result); + + if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) + set_buffer_new(bh_result); + + /* + * Need to update the i_size under the extent lock so buffered + * readers will get the updated i_size when we unlock. + */ + if (!dio_data->overwrite && start + len > i_size_read(inode)) + i_size_write(inode, start + len); + + WARN_ON(dio_data->reserve < len); + dio_data->reserve -= len; + dio_data->unsubmitted_oe_range_end = start + len; + current->journal_info = dio_data; +out: + return ret; +} + static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { @@ -7645,116 +7727,36 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, goto unlock_err; } - /* Just a good old fashioned hole, return */ - if (!create && (em->block_start == EXTENT_MAP_HOLE || - test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { - free_extent_map(em); - goto unlock_err; - } - - /* - * We don't allocate a new extent in the following cases - * - * 1) The inode is marked as NODATACOW. In this case we'll just use the - * existing extent. - * 2) The extent is marked as PREALLOC. We're good to go here and can - * just use the extent. - * - */ - if (!create) { - len = min(len, em->len - (start - em->start)); - lockstart = start + len; - goto unlock; - } - - if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) || - ((BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) && - em->block_start != EXTENT_MAP_HOLE)) { - int type; - u64 block_start, orig_start, orig_block_len, ram_bytes; - - if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) - type = BTRFS_ORDERED_PREALLOC; - else - type = BTRFS_ORDERED_NOCOW; - len = min(len, em->len - (start - em->start)); - block_start = em->block_start + (start - em->start); - - if (can_nocow_extent(inode, start, &len, &orig_start, - &orig_block_len, &ram_bytes) == 1 && - btrfs_inc_nocow_writers(fs_info, block_start)) { - struct extent_map *em2; - - em2 = btrfs_create_dio_extent(inode, start, len, - orig_start, block_start, - len, orig_block_len, - ram_bytes, type); - btrfs_dec_nocow_writers(fs_info, block_start); - if (type == BTRFS_ORDERED_PREALLOC) { - free_extent_map(em); - em = em2; - } - if (em2 && IS_ERR(em2)) { - ret = PTR_ERR(em2); - goto unlock_err; - } - /* - * For inode marked NODATACOW or extent marked PREALLOC, - * use the existing or preallocated extent, so does not - * need to adjust btrfs_space_info's bytes_may_use. - */ - btrfs_free_reserved_data_space_noquota(inode, - start, len); - goto unlock; - } - } - - /* - * this will cow the extent, reset the len in case we changed - * it above - */ - len = bh_result->b_size; - free_extent_map(em); - em = btrfs_new_extent_direct(inode, start, len); - if (IS_ERR(em)) { - ret = PTR_ERR(em); - goto unlock_err; - } - len = min(len, em->len - (start - em->start)); -unlock: - bh_result->b_blocknr = (em->block_start + (start - em->start)) >> - inode->i_blkbits; - bh_result->b_size = len; - bh_result->b_bdev = em->bdev; - set_buffer_mapped(bh_result); if (create) { - if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) - set_buffer_new(bh_result); + ret = btrfs_get_blocks_direct_write(&em, bh_result, inode, + dio_data, start, len); + if (ret < 0) + goto unlock_err; + /* clear and unlock the entire range */ + clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, lockend, + unlock_bits, 1, 0, &cached_state); + } else { + ret = btrfs_get_blocks_direct_read(em, bh_result, inode, + start, len); + /* Can be negative only if we read from a hole */ + if (ret < 0) { + ret = 0; + free_extent_map(em); + goto unlock_err; + } /* - * Need to update the i_size under the extent lock so buffered - * readers will get the updated i_size when we unlock. + * We need to unlock only the end area that we aren't using. + * The rest is going to be unlocked by the endio routine. */ - if (!dio_data->overwrite && start + len > i_size_read(inode)) - i_size_write(inode, start + len); - - WARN_ON(dio_data->reserve < len); - dio_data->reserve -= len; - dio_data->unsubmitted_oe_range_end = start + len; - current->journal_info = dio_data; - } - - /* - * In the case of write we need to clear and unlock the entire range, - * in the case of read we need to unlock only the end area that we - * aren't using if there is any left over space. - */ - if (lockstart < lockend) { - clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, - lockend, unlock_bits, 1, 0, - &cached_state); - } else { - free_extent_state(cached_state); + lockstart = start + bh_result->b_size; + if (lockstart < lockend) { + clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, + lockend, unlock_bits, 1, 0, + &cached_state); + } else { + free_extent_state(cached_state); + } } free_extent_map(em); @@ -8140,7 +8142,6 @@ static void __endio_write_update_ordered(struct inode *inode, u64 ordered_offset = offset; u64 ordered_bytes = bytes; u64 last_offset; - int ret; if (btrfs_is_free_space_inode(BTRFS_I(inode))) { wq = fs_info->endio_freespace_worker; @@ -8150,32 +8151,31 @@ static void __endio_write_update_ordered(struct inode *inode, func = btrfs_endio_write_helper; } -again: - last_offset = ordered_offset; - ret = btrfs_dec_test_first_ordered_pending(inode, &ordered, - &ordered_offset, - ordered_bytes, - uptodate); - if (!ret) - goto out_test; - - btrfs_init_work(&ordered->work, func, finish_ordered_fn, NULL, NULL); - btrfs_queue_work(wq, &ordered->work); -out_test: - /* - * If btrfs_dec_test_ordered_pending does not find any ordered extent - * in the range, we can exit. - */ - if (ordered_offset == last_offset) - return; - /* - * our bio might span multiple ordered extents. If we haven't - * completed the accounting for the whole dio, go back and try again - */ - if (ordered_offset < offset + bytes) { - ordered_bytes = offset + bytes - ordered_offset; - ordered = NULL; - goto again; + while (ordered_offset < offset + bytes) { + last_offset = ordered_offset; + if (btrfs_dec_test_first_ordered_pending(inode, &ordered, + &ordered_offset, + ordered_bytes, + uptodate)) { + btrfs_init_work(&ordered->work, func, + finish_ordered_fn, + NULL, NULL); + btrfs_queue_work(wq, &ordered->work); + } + /* + * If btrfs_dec_test_ordered_pending does not find any ordered + * extent in the range, we can exit. + */ + if (ordered_offset == last_offset) + return; + /* + * Our bio might span multiple ordered extents. In this case + * we keep goin until we have accounted the whole dio. + */ + if (ordered_offset < offset + bytes) { + ordered_bytes = offset + bytes - ordered_offset; + ordered = NULL; + } } } @@ -8714,29 +8714,19 @@ static int btrfs_writepage(struct page *page, struct writeback_control *wbc) static int btrfs_writepages(struct address_space *mapping, struct writeback_control *wbc) { - struct extent_io_tree *tree; - - tree = &BTRFS_I(mapping->host)->io_tree; - return extent_writepages(tree, mapping, wbc); + return extent_writepages(mapping, wbc); } static int btrfs_readpages(struct file *file, struct address_space *mapping, struct list_head *pages, unsigned nr_pages) { - struct extent_io_tree *tree; - tree = &BTRFS_I(mapping->host)->io_tree; - return extent_readpages(tree, mapping, pages, nr_pages); + return extent_readpages(mapping, pages, nr_pages); } + static int __btrfs_releasepage(struct page *page, gfp_t gfp_flags) { - struct extent_io_tree *tree; - struct extent_map_tree *map; - int ret; - - tree = &BTRFS_I(page->mapping->host)->io_tree; - map = &BTRFS_I(page->mapping->host)->extent_tree; - ret = try_release_extent_mapping(map, tree, page, gfp_flags); + int ret = try_release_extent_mapping(page, gfp_flags); if (ret == 1) { ClearPagePrivate(page); set_page_private(page, 0); @@ -8877,8 +8867,8 @@ again: * * We are not allowed to take the i_mutex here so we have to play games to * protect against truncate races as the page could now be beyond EOF. Because - * vmtruncate() writes the inode size before removing pages, once we have the - * page lock we can determine safely if the page is beyond EOF. If it is not + * truncate_setsize() writes the inode size before removing pages, once we have + * the page lock we can determine safely if the page is beyond EOF. If it is not * beyond EOF, then the page is guaranteed safe against truncation until we * unlock the page. */ @@ -9040,8 +9030,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_block_rsv *rsv; - int ret = 0; - int err = 0; + int ret; struct btrfs_trans_handle *trans; u64 mask = fs_info->sectorsize - 1; u64 min_size = btrfs_calc_trunc_metadata_size(fs_info, 1); @@ -9054,39 +9043,31 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) } /* - * Yes ladies and gentlemen, this is indeed ugly. The fact is we have - * 3 things going on here - * - * 1) We need to reserve space for our orphan item and the space to - * delete our orphan item. Lord knows we don't want to have a dangling - * orphan item because we didn't reserve space to remove it. + * Yes ladies and gentlemen, this is indeed ugly. We have a couple of + * things going on here: * - * 2) We need to reserve space to update our inode. + * 1) We need to reserve space to update our inode. * - * 3) We need to have something to cache all the space that is going to + * 2) We need to have something to cache all the space that is going to * be free'd up by the truncate operation, but also have some slack * space reserved in case it uses space during the truncate (thank you * very much snapshotting). * - * And we need these to all be separate. The fact is we can use a lot of + * And we need these to be separate. The fact is we can use a lot of * space doing the truncate, and we have no earthly idea how much space * we will use, so we need the truncate reservation to be separate so it - * doesn't end up using space reserved for updating the inode or - * removing the orphan item. We also need to be able to stop the - * transaction and start a new one, which means we need to be able to - * update the inode several times, and we have no idea of knowing how - * many times that will be, so we can't just reserve 1 item for the - * entirety of the operation, so that has to be done separately as well. - * Then there is the orphan item, which does indeed need to be held on - * to for the whole operation, and we need nobody to touch this reserved - * space except the orphan code. + * doesn't end up using space reserved for updating the inode. We also + * need to be able to stop the transaction and start a new one, which + * means we need to be able to update the inode several times, and we + * have no idea of knowing how many times that will be, so we can't just + * reserve 1 item for the entirety of the operation, so that has to be + * done separately as well. * * So that leaves us with * - * 1) root->orphan_block_rsv - for the orphan deletion. - * 2) rsv - for the truncate reservation, which we will steal from the + * 1) rsv - for the truncate reservation, which we will steal from the * transaction reservation. - * 3) fs_info->trans_block_rsv - this will have 1 items worth left for + * 2) fs_info->trans_block_rsv - this will have 1 items worth left for * updating the inode. */ rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP); @@ -9101,7 +9082,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) */ trans = btrfs_start_transaction(root, 2); if (IS_ERR(trans)) { - err = PTR_ERR(trans); + ret = PTR_ERR(trans); goto out; } @@ -9125,23 +9106,19 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) inode->i_size, BTRFS_EXTENT_DATA_KEY); trans->block_rsv = &fs_info->trans_block_rsv; - if (ret != -ENOSPC && ret != -EAGAIN) { - err = ret; + if (ret != -ENOSPC && ret != -EAGAIN) break; - } ret = btrfs_update_inode(trans, root, inode); - if (ret) { - err = ret; + if (ret) break; - } btrfs_end_transaction(trans); btrfs_btree_balance_dirty(fs_info); trans = btrfs_start_transaction(root, 2); if (IS_ERR(trans)) { - ret = err = PTR_ERR(trans); + ret = PTR_ERR(trans); trans = NULL; break; } @@ -9174,29 +9151,23 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) btrfs_ordered_update_i_size(inode, inode->i_size, NULL); } - if (ret == 0 && inode->i_nlink > 0) { - trans->block_rsv = root->orphan_block_rsv; - ret = btrfs_orphan_del(trans, BTRFS_I(inode)); - if (ret) - err = ret; - } - if (trans) { + int ret2; + trans->block_rsv = &fs_info->trans_block_rsv; - ret = btrfs_update_inode(trans, root, inode); - if (ret && !err) - err = ret; + ret2 = btrfs_update_inode(trans, root, inode); + if (ret2 && !ret) + ret = ret2; - ret = btrfs_end_transaction(trans); + ret2 = btrfs_end_transaction(trans); + if (ret2 && !ret) + ret = ret2; btrfs_btree_balance_dirty(fs_info); } out: btrfs_free_block_rsv(fs_info, rsv); - if (ret && !err) - err = ret; - - return err; + return ret; } /* @@ -9332,13 +9303,6 @@ void btrfs_destroy_inode(struct inode *inode) if (!root) goto free; - if (test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, - &BTRFS_I(inode)->runtime_flags)) { - btrfs_info(fs_info, "inode %llu still on the orphan list", - btrfs_ino(BTRFS_I(inode))); - atomic_dec(&root->orphan_inodes); - } - while (1) { ordered = btrfs_lookup_first_ordered_extent(inode, (u64)-1); if (!ordered) @@ -9972,6 +9936,13 @@ static int btrfs_rename2(struct inode *old_dir, struct dentry *old_dentry, return btrfs_rename(old_dir, old_dentry, new_dir, new_dentry, flags); } +struct btrfs_delalloc_work { + struct inode *inode; + struct completion completion; + struct list_head list; + struct btrfs_work work; +}; + static void btrfs_run_delalloc_work(struct btrfs_work *work) { struct btrfs_delalloc_work *delalloc_work; @@ -9985,15 +9956,11 @@ static void btrfs_run_delalloc_work(struct btrfs_work *work) &BTRFS_I(inode)->runtime_flags)) filemap_flush(inode->i_mapping); - if (delalloc_work->delay_iput) - btrfs_add_delayed_iput(inode); - else - iput(inode); + iput(inode); complete(&delalloc_work->completion); } -struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode, - int delay_iput) +static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode) { struct btrfs_delalloc_work *work; @@ -10004,7 +9971,6 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode, init_completion(&work->completion); INIT_LIST_HEAD(&work->list); work->inode = inode; - work->delay_iput = delay_iput; WARN_ON_ONCE(!inode); btrfs_init_work(&work->work, btrfs_flush_delalloc_helper, btrfs_run_delalloc_work, NULL, NULL); @@ -10012,18 +9978,11 @@ struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode, return work; } -void btrfs_wait_and_free_delalloc_work(struct btrfs_delalloc_work *work) -{ - wait_for_completion(&work->completion); - kfree(work); -} - /* * some fairly slow code that needs optimization. This walks the list * of all the inodes with pending delalloc and forces them to disk. */ -static int __start_delalloc_inodes(struct btrfs_root *root, int delay_iput, - int nr) +static int start_delalloc_inodes(struct btrfs_root *root, int nr) { struct btrfs_inode *binode; struct inode *inode; @@ -10051,12 +10010,9 @@ static int __start_delalloc_inodes(struct btrfs_root *root, int delay_iput, } spin_unlock(&root->delalloc_lock); - work = btrfs_alloc_delalloc_work(inode, delay_iput); + work = btrfs_alloc_delalloc_work(inode); if (!work) { - if (delay_iput) - btrfs_add_delayed_iput(inode); - else - iput(inode); + iput(inode); ret = -ENOMEM; goto out; } @@ -10074,10 +10030,11 @@ static int __start_delalloc_inodes(struct btrfs_root *root, int delay_iput, out: list_for_each_entry_safe(work, next, &works, list) { list_del_init(&work->list); - btrfs_wait_and_free_delalloc_work(work); + wait_for_completion(&work->completion); + kfree(work); } - if (!list_empty_careful(&splice)) { + if (!list_empty(&splice)) { spin_lock(&root->delalloc_lock); list_splice_tail(&splice, &root->delalloc_inodes); spin_unlock(&root->delalloc_lock); @@ -10086,7 +10043,7 @@ out: return ret; } -int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput) +int btrfs_start_delalloc_inodes(struct btrfs_root *root) { struct btrfs_fs_info *fs_info = root->fs_info; int ret; @@ -10094,14 +10051,13 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput) if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) return -EROFS; - ret = __start_delalloc_inodes(root, delay_iput, -1); + ret = start_delalloc_inodes(root, -1); if (ret > 0) ret = 0; return ret; } -int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput, - int nr) +int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr) { struct btrfs_root *root; struct list_head splice; @@ -10124,7 +10080,7 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput, &fs_info->delalloc_roots); spin_unlock(&fs_info->delalloc_root_lock); - ret = __start_delalloc_inodes(root, delay_iput, nr); + ret = start_delalloc_inodes(root, nr); btrfs_put_fs_root(root); if (ret < 0) goto out; @@ -10139,7 +10095,7 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput, ret = 0; out: - if (!list_empty_careful(&splice)) { + if (!list_empty(&splice)) { spin_lock(&fs_info->delalloc_root_lock); list_splice_tail(&splice, &fs_info->delalloc_roots); spin_unlock(&fs_info->delalloc_root_lock); @@ -10259,8 +10215,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out_unlock: btrfs_end_transaction(trans); @@ -10678,5 +10633,4 @@ static const struct inode_operations btrfs_symlink_inode_operations = { const struct dentry_operations btrfs_dentry_operations = { .d_delete = btrfs_dentry_delete, - .d_release = btrfs_dentry_release, }; diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index b2db3988813f..d29992f7dc63 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/kernel.h> @@ -106,20 +93,22 @@ static int btrfs_clone(struct inode *src, struct inode *inode, int no_time_update); /* Mask out flags that are inappropriate for the given type of inode. */ -static unsigned int btrfs_mask_flags(umode_t mode, unsigned int flags) +static unsigned int btrfs_mask_fsflags_for_type(struct inode *inode, + unsigned int flags) { - if (S_ISDIR(mode)) + if (S_ISDIR(inode->i_mode)) return flags; - else if (S_ISREG(mode)) + else if (S_ISREG(inode->i_mode)) return flags & ~FS_DIRSYNC_FL; else return flags & (FS_NODUMP_FL | FS_NOATIME_FL); } /* - * Export inode flags to the format expected by the FS_IOC_GETFLAGS ioctl. + * Export internal inode flags to the format expected by the FS_IOC_GETFLAGS + * ioctl. */ -static unsigned int btrfs_flags_to_ioctl(unsigned int flags) +static unsigned int btrfs_inode_flags_to_fsflags(unsigned int flags) { unsigned int iflags = 0; @@ -149,20 +138,20 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags) /* * Update inode->i_flags based on the btrfs internal flags. */ -void btrfs_update_iflags(struct inode *inode) +void btrfs_sync_inode_flags_to_i_flags(struct inode *inode) { - struct btrfs_inode *ip = BTRFS_I(inode); + struct btrfs_inode *binode = BTRFS_I(inode); unsigned int new_fl = 0; - if (ip->flags & BTRFS_INODE_SYNC) + if (binode->flags & BTRFS_INODE_SYNC) new_fl |= S_SYNC; - if (ip->flags & BTRFS_INODE_IMMUTABLE) + if (binode->flags & BTRFS_INODE_IMMUTABLE) new_fl |= S_IMMUTABLE; - if (ip->flags & BTRFS_INODE_APPEND) + if (binode->flags & BTRFS_INODE_APPEND) new_fl |= S_APPEND; - if (ip->flags & BTRFS_INODE_NOATIME) + if (binode->flags & BTRFS_INODE_NOATIME) new_fl |= S_NOATIME; - if (ip->flags & BTRFS_INODE_DIRSYNC) + if (binode->flags & BTRFS_INODE_DIRSYNC) new_fl |= S_DIRSYNC; set_mask_bits(&inode->i_flags, @@ -172,15 +161,16 @@ void btrfs_update_iflags(struct inode *inode) static int btrfs_ioctl_getflags(struct file *file, void __user *arg) { - struct btrfs_inode *ip = BTRFS_I(file_inode(file)); - unsigned int flags = btrfs_flags_to_ioctl(ip->flags); + struct btrfs_inode *binode = BTRFS_I(file_inode(file)); + unsigned int flags = btrfs_inode_flags_to_fsflags(binode->flags); if (copy_to_user(arg, &flags, sizeof(flags))) return -EFAULT; return 0; } -static int check_flags(unsigned int flags) +/* Check if @flags are a supported and valid set of FS_*_FL flags */ +static int check_fsflags(unsigned int flags) { if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \ FS_NOATIME_FL | FS_NODUMP_FL | \ @@ -199,13 +189,13 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) { struct inode *inode = file_inode(file); struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_inode *ip = BTRFS_I(inode); - struct btrfs_root *root = ip->root; + struct btrfs_inode *binode = BTRFS_I(inode); + struct btrfs_root *root = binode->root; struct btrfs_trans_handle *trans; - unsigned int flags, oldflags; + unsigned int fsflags, old_fsflags; int ret; - u64 ip_oldflags; - unsigned int i_oldflags; + u64 old_flags; + unsigned int old_i_flags; umode_t mode; if (!inode_owner_or_capable(inode)) @@ -214,10 +204,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) if (btrfs_root_readonly(root)) return -EROFS; - if (copy_from_user(&flags, arg, sizeof(flags))) + if (copy_from_user(&fsflags, arg, sizeof(fsflags))) return -EFAULT; - ret = check_flags(flags); + ret = check_fsflags(fsflags); if (ret) return ret; @@ -227,44 +217,44 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) inode_lock(inode); - ip_oldflags = ip->flags; - i_oldflags = inode->i_flags; + old_flags = binode->flags; + old_i_flags = inode->i_flags; mode = inode->i_mode; - flags = btrfs_mask_flags(inode->i_mode, flags); - oldflags = btrfs_flags_to_ioctl(ip->flags); - if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) { + fsflags = btrfs_mask_fsflags_for_type(inode, fsflags); + old_fsflags = btrfs_inode_flags_to_fsflags(binode->flags); + if ((fsflags ^ old_fsflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) { if (!capable(CAP_LINUX_IMMUTABLE)) { ret = -EPERM; goto out_unlock; } } - if (flags & FS_SYNC_FL) - ip->flags |= BTRFS_INODE_SYNC; + if (fsflags & FS_SYNC_FL) + binode->flags |= BTRFS_INODE_SYNC; else - ip->flags &= ~BTRFS_INODE_SYNC; - if (flags & FS_IMMUTABLE_FL) - ip->flags |= BTRFS_INODE_IMMUTABLE; + binode->flags &= ~BTRFS_INODE_SYNC; + if (fsflags & FS_IMMUTABLE_FL) + binode->flags |= BTRFS_INODE_IMMUTABLE; else - ip->flags &= ~BTRFS_INODE_IMMUTABLE; - if (flags & FS_APPEND_FL) - ip->flags |= BTRFS_INODE_APPEND; + binode->flags &= ~BTRFS_INODE_IMMUTABLE; + if (fsflags & FS_APPEND_FL) + binode->flags |= BTRFS_INODE_APPEND; else - ip->flags &= ~BTRFS_INODE_APPEND; - if (flags & FS_NODUMP_FL) - ip->flags |= BTRFS_INODE_NODUMP; + binode->flags &= ~BTRFS_INODE_APPEND; + if (fsflags & FS_NODUMP_FL) + binode->flags |= BTRFS_INODE_NODUMP; else - ip->flags &= ~BTRFS_INODE_NODUMP; - if (flags & FS_NOATIME_FL) - ip->flags |= BTRFS_INODE_NOATIME; + binode->flags &= ~BTRFS_INODE_NODUMP; + if (fsflags & FS_NOATIME_FL) + binode->flags |= BTRFS_INODE_NOATIME; else - ip->flags &= ~BTRFS_INODE_NOATIME; - if (flags & FS_DIRSYNC_FL) - ip->flags |= BTRFS_INODE_DIRSYNC; + binode->flags &= ~BTRFS_INODE_NOATIME; + if (fsflags & FS_DIRSYNC_FL) + binode->flags |= BTRFS_INODE_DIRSYNC; else - ip->flags &= ~BTRFS_INODE_DIRSYNC; - if (flags & FS_NOCOW_FL) { + binode->flags &= ~BTRFS_INODE_DIRSYNC; + if (fsflags & FS_NOCOW_FL) { if (S_ISREG(mode)) { /* * It's safe to turn csums off here, no extents exist. @@ -272,10 +262,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) * status of the file and will not set it. */ if (inode->i_size == 0) - ip->flags |= BTRFS_INODE_NODATACOW - | BTRFS_INODE_NODATASUM; + binode->flags |= BTRFS_INODE_NODATACOW + | BTRFS_INODE_NODATASUM; } else { - ip->flags |= BTRFS_INODE_NODATACOW; + binode->flags |= BTRFS_INODE_NODATACOW; } } else { /* @@ -283,10 +273,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) */ if (S_ISREG(mode)) { if (inode->i_size == 0) - ip->flags &= ~(BTRFS_INODE_NODATACOW + binode->flags &= ~(BTRFS_INODE_NODATACOW | BTRFS_INODE_NODATASUM); } else { - ip->flags &= ~BTRFS_INODE_NODATACOW; + binode->flags &= ~BTRFS_INODE_NODATACOW; } } @@ -295,18 +285,18 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) * flag may be changed automatically if compression code won't make * things smaller. */ - if (flags & FS_NOCOMP_FL) { - ip->flags &= ~BTRFS_INODE_COMPRESS; - ip->flags |= BTRFS_INODE_NOCOMPRESS; + if (fsflags & FS_NOCOMP_FL) { + binode->flags &= ~BTRFS_INODE_COMPRESS; + binode->flags |= BTRFS_INODE_NOCOMPRESS; ret = btrfs_set_prop(inode, "btrfs.compression", NULL, 0, 0); if (ret && ret != -ENODATA) goto out_drop; - } else if (flags & FS_COMPR_FL) { + } else if (fsflags & FS_COMPR_FL) { const char *comp; - ip->flags |= BTRFS_INODE_COMPRESS; - ip->flags &= ~BTRFS_INODE_NOCOMPRESS; + binode->flags |= BTRFS_INODE_COMPRESS; + binode->flags &= ~BTRFS_INODE_NOCOMPRESS; comp = btrfs_compress_type2str(fs_info->compress_type); if (!comp || comp[0] == 0) @@ -321,7 +311,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) ret = btrfs_set_prop(inode, "btrfs.compression", NULL, 0, 0); if (ret && ret != -ENODATA) goto out_drop; - ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS); + binode->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS); } trans = btrfs_start_transaction(root, 1); @@ -330,7 +320,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) goto out_drop; } - btrfs_update_iflags(inode); + btrfs_sync_inode_flags_to_i_flags(inode); inode_inc_iversion(inode); inode->i_ctime = current_time(inode); ret = btrfs_update_inode(trans, root, inode); @@ -338,8 +328,8 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) btrfs_end_transaction(trans); out_drop: if (ret) { - ip->flags = ip_oldflags; - inode->i_flags = i_oldflags; + binode->flags = old_flags; + inode->i_flags = old_i_flags; } out_unlock: @@ -348,6 +338,148 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) return ret; } +/* + * Translate btrfs internal inode flags to xflags as expected by the + * FS_IOC_FSGETXATT ioctl. Filter only the supported ones, unknown flags are + * silently dropped. + */ +static unsigned int btrfs_inode_flags_to_xflags(unsigned int flags) +{ + unsigned int xflags = 0; + + if (flags & BTRFS_INODE_APPEND) + xflags |= FS_XFLAG_APPEND; + if (flags & BTRFS_INODE_IMMUTABLE) + xflags |= FS_XFLAG_IMMUTABLE; + if (flags & BTRFS_INODE_NOATIME) + xflags |= FS_XFLAG_NOATIME; + if (flags & BTRFS_INODE_NODUMP) + xflags |= FS_XFLAG_NODUMP; + if (flags & BTRFS_INODE_SYNC) + xflags |= FS_XFLAG_SYNC; + + return xflags; +} + +/* Check if @flags are a supported and valid set of FS_XFLAGS_* flags */ +static int check_xflags(unsigned int flags) +{ + if (flags & ~(FS_XFLAG_APPEND | FS_XFLAG_IMMUTABLE | FS_XFLAG_NOATIME | + FS_XFLAG_NODUMP | FS_XFLAG_SYNC)) + return -EOPNOTSUPP; + return 0; +} + +/* + * Set the xflags from the internal inode flags. The remaining items of fsxattr + * are zeroed. + */ +static int btrfs_ioctl_fsgetxattr(struct file *file, void __user *arg) +{ + struct btrfs_inode *binode = BTRFS_I(file_inode(file)); + struct fsxattr fa; + + memset(&fa, 0, sizeof(fa)); + fa.fsx_xflags = btrfs_inode_flags_to_xflags(binode->flags); + + if (copy_to_user(arg, &fa, sizeof(fa))) + return -EFAULT; + + return 0; +} + +static int btrfs_ioctl_fssetxattr(struct file *file, void __user *arg) +{ + struct inode *inode = file_inode(file); + struct btrfs_inode *binode = BTRFS_I(inode); + struct btrfs_root *root = binode->root; + struct btrfs_trans_handle *trans; + struct fsxattr fa; + unsigned old_flags; + unsigned old_i_flags; + int ret = 0; + + if (!inode_owner_or_capable(inode)) + return -EPERM; + + if (btrfs_root_readonly(root)) + return -EROFS; + + memset(&fa, 0, sizeof(fa)); + if (copy_from_user(&fa, arg, sizeof(fa))) + return -EFAULT; + + ret = check_xflags(fa.fsx_xflags); + if (ret) + return ret; + + if (fa.fsx_extsize != 0 || fa.fsx_projid != 0 || fa.fsx_cowextsize != 0) + return -EOPNOTSUPP; + + ret = mnt_want_write_file(file); + if (ret) + return ret; + + inode_lock(inode); + + old_flags = binode->flags; + old_i_flags = inode->i_flags; + + /* We need the capabilities to change append-only or immutable inode */ + if (((old_flags & (BTRFS_INODE_APPEND | BTRFS_INODE_IMMUTABLE)) || + (fa.fsx_xflags & (FS_XFLAG_APPEND | FS_XFLAG_IMMUTABLE))) && + !capable(CAP_LINUX_IMMUTABLE)) { + ret = -EPERM; + goto out_unlock; + } + + if (fa.fsx_xflags & FS_XFLAG_SYNC) + binode->flags |= BTRFS_INODE_SYNC; + else + binode->flags &= ~BTRFS_INODE_SYNC; + if (fa.fsx_xflags & FS_XFLAG_IMMUTABLE) + binode->flags |= BTRFS_INODE_IMMUTABLE; + else + binode->flags &= ~BTRFS_INODE_IMMUTABLE; + if (fa.fsx_xflags & FS_XFLAG_APPEND) + binode->flags |= BTRFS_INODE_APPEND; + else + binode->flags &= ~BTRFS_INODE_APPEND; + if (fa.fsx_xflags & FS_XFLAG_NODUMP) + binode->flags |= BTRFS_INODE_NODUMP; + else + binode->flags &= ~BTRFS_INODE_NODUMP; + if (fa.fsx_xflags & FS_XFLAG_NOATIME) + binode->flags |= BTRFS_INODE_NOATIME; + else + binode->flags &= ~BTRFS_INODE_NOATIME; + + /* 1 item for the inode */ + trans = btrfs_start_transaction(root, 1); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + goto out_unlock; + } + + btrfs_sync_inode_flags_to_i_flags(inode); + inode_inc_iversion(inode); + inode->i_ctime = current_time(inode); + ret = btrfs_update_inode(trans, root, inode); + + btrfs_end_transaction(trans); + +out_unlock: + if (ret) { + binode->flags = old_flags; + inode->i_flags = old_i_flags; + } + + inode_unlock(inode); + mnt_drop_write_file(file); + + return ret; +} + static int btrfs_ioctl_getversion(struct file *file, int __user *arg) { struct inode *inode = file_inode(file); @@ -437,7 +569,6 @@ static noinline int create_subvol(struct inode *dir, u64 objectid; u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID; u64 index = 0; - u64 qgroup_reserved; uuid_le new_uuid; root_item = kzalloc(sizeof(*root_item), GFP_KERNEL); @@ -462,8 +593,7 @@ static noinline int create_subvol(struct inode *dir, * The same as the snapshot creation, please see the comment * of create_snapshot(). */ - ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, - 8, &qgroup_reserved, false); + ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, 8, false); if (ret) goto fail_free; @@ -586,7 +716,7 @@ static noinline int create_subvol(struct inode *dir, btrfs_ino(BTRFS_I(dir)), index, name, namelen); BUG_ON(ret); - ret = btrfs_uuid_tree_add(trans, fs_info, root_item->uuid, + ret = btrfs_uuid_tree_add(trans, root_item->uuid, BTRFS_UUID_KEY_SUBVOL, objectid); if (ret) btrfs_abort_transaction(trans, ret); @@ -653,7 +783,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, wait_event(root->subv_writers->wait, percpu_counter_sum(&root->subv_writers->counter) == 0); - ret = btrfs_start_delalloc_inodes(root, 0); + ret = btrfs_start_delalloc_inodes(root); if (ret) goto dec_and_free; @@ -671,7 +801,6 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, */ ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root, &pending_snapshot->block_rsv, 8, - &pending_snapshot->qgroup_reserved, false); if (ret) goto dec_and_free; @@ -1470,7 +1599,6 @@ static noinline int btrfs_ioctl_resize(struct file *file, return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; } - mutex_lock(&fs_info->volume_mutex); vol_args = memdup_user(arg, sizeof(*vol_args)); if (IS_ERR(vol_args)) { ret = PTR_ERR(vol_args); @@ -1578,7 +1706,6 @@ static noinline int btrfs_ioctl_resize(struct file *file, out_free: kfree(vol_args); out: - mutex_unlock(&fs_info->volume_mutex); clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); mnt_drop_write_file(file); return ret; @@ -1845,60 +1972,6 @@ out: return ret; } -/* - * helper to check if the subvolume references other subvolumes - */ -static noinline int may_destroy_subvol(struct btrfs_root *root) -{ - struct btrfs_fs_info *fs_info = root->fs_info; - struct btrfs_path *path; - struct btrfs_dir_item *di; - struct btrfs_key key; - u64 dir_id; - int ret; - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - - /* Make sure this root isn't set as the default subvol */ - dir_id = btrfs_super_root_dir(fs_info->super_copy); - di = btrfs_lookup_dir_item(NULL, fs_info->tree_root, path, - dir_id, "default", 7, 0); - if (di && !IS_ERR(di)) { - btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key); - if (key.objectid == root->root_key.objectid) { - ret = -EPERM; - btrfs_err(fs_info, - "deleting default subvolume %llu is not allowed", - key.objectid); - goto out; - } - btrfs_release_path(path); - } - - key.objectid = root->root_key.objectid; - key.type = BTRFS_ROOT_REF_KEY; - key.offset = (u64)-1; - - ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0); - if (ret < 0) - goto out; - BUG_ON(ret == 0); - - ret = 0; - if (path->slots[0] > 0) { - path->slots[0]--; - btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); - if (key.objectid == root->root_key.objectid && - key.type == BTRFS_ROOT_REF_KEY) - ret = -ENOTEMPTY; - } -out: - btrfs_free_path(path); - return ret; -} - static noinline int key_in_sk(struct btrfs_key *key, struct btrfs_ioctl_search_key *sk) { @@ -2079,7 +2152,7 @@ static noinline int search_ioctl(struct inode *inode, root = btrfs_read_fs_root_no_name(info, &key); if (IS_ERR(root)) { btrfs_free_path(path); - return -ENOENT; + return PTR_ERR(root); } } @@ -2213,8 +2286,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, key.offset = (u64)-1; root = btrfs_read_fs_root_no_name(info, &key); if (IS_ERR(root)) { - btrfs_err(info, "could not find root %llu", tree_id); - ret = -ENOENT; + ret = PTR_ERR(root); goto out; } @@ -2269,6 +2341,165 @@ out: return ret; } +static int btrfs_search_path_in_tree_user(struct inode *inode, + struct btrfs_ioctl_ino_lookup_user_args *args) +{ + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + struct super_block *sb = inode->i_sb; + struct btrfs_key upper_limit = BTRFS_I(inode)->location; + u64 treeid = BTRFS_I(inode)->root->root_key.objectid; + u64 dirid = args->dirid; + unsigned long item_off; + unsigned long item_len; + struct btrfs_inode_ref *iref; + struct btrfs_root_ref *rref; + struct btrfs_root *root; + struct btrfs_path *path; + struct btrfs_key key, key2; + struct extent_buffer *leaf; + struct inode *temp_inode; + char *ptr; + int slot; + int len; + int total_len = 0; + int ret; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + /* + * If the bottom subvolume does not exist directly under upper_limit, + * construct the path in from the bottom up. + */ + if (dirid != upper_limit.objectid) { + ptr = &args->path[BTRFS_INO_LOOKUP_USER_PATH_MAX - 1]; + + key.objectid = treeid; + key.type = BTRFS_ROOT_ITEM_KEY; + key.offset = (u64)-1; + root = btrfs_read_fs_root_no_name(fs_info, &key); + if (IS_ERR(root)) { + ret = PTR_ERR(root); + goto out; + } + + key.objectid = dirid; + key.type = BTRFS_INODE_REF_KEY; + key.offset = (u64)-1; + while (1) { + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) { + goto out; + } else if (ret > 0) { + ret = btrfs_previous_item(root, path, dirid, + BTRFS_INODE_REF_KEY); + if (ret < 0) { + goto out; + } else if (ret > 0) { + ret = -ENOENT; + goto out; + } + } + + leaf = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(leaf, &key, slot); + + iref = btrfs_item_ptr(leaf, slot, struct btrfs_inode_ref); + len = btrfs_inode_ref_name_len(leaf, iref); + ptr -= len + 1; + total_len += len + 1; + if (ptr < args->path) { + ret = -ENAMETOOLONG; + goto out; + } + + *(ptr + len) = '/'; + read_extent_buffer(leaf, ptr, + (unsigned long)(iref + 1), len); + + /* Check the read+exec permission of this directory */ + ret = btrfs_previous_item(root, path, dirid, + BTRFS_INODE_ITEM_KEY); + if (ret < 0) { + goto out; + } else if (ret > 0) { + ret = -ENOENT; + goto out; + } + + leaf = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(leaf, &key2, slot); + if (key2.objectid != dirid) { + ret = -ENOENT; + goto out; + } + + temp_inode = btrfs_iget(sb, &key2, root, NULL); + ret = inode_permission(temp_inode, MAY_READ | MAY_EXEC); + iput(temp_inode); + if (ret) { + ret = -EACCES; + goto out; + } + + if (key.offset == upper_limit.objectid) + break; + if (key.objectid == BTRFS_FIRST_FREE_OBJECTID) { + ret = -EACCES; + goto out; + } + + btrfs_release_path(path); + key.objectid = key.offset; + key.offset = (u64)-1; + dirid = key.objectid; + } + + memmove(args->path, ptr, total_len); + args->path[total_len] = '\0'; + btrfs_release_path(path); + } + + /* Get the bottom subvolume's name from ROOT_REF */ + root = fs_info->tree_root; + key.objectid = treeid; + key.type = BTRFS_ROOT_REF_KEY; + key.offset = args->treeid; + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) { + goto out; + } else if (ret > 0) { + ret = -ENOENT; + goto out; + } + + leaf = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(leaf, &key, slot); + + item_off = btrfs_item_ptr_offset(leaf, slot); + item_len = btrfs_item_size_nr(leaf, slot); + /* Check if dirid in ROOT_REF corresponds to passed dirid */ + rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref); + if (args->dirid != btrfs_root_ref_dirid(leaf, rref)) { + ret = -EINVAL; + goto out; + } + + /* Copy subvolume's name */ + item_off += sizeof(struct btrfs_root_ref); + item_len -= sizeof(struct btrfs_root_ref); + read_extent_buffer(leaf, args->name, item_off, item_len); + args->name[item_len] = 0; + +out: + btrfs_free_path(path); + return ret; +} + static noinline int btrfs_ioctl_ino_lookup(struct file *file, void __user *argp) { @@ -2311,6 +2542,265 @@ out: return ret; } +/* + * Version of ino_lookup ioctl (unprivileged) + * + * The main differences from ino_lookup ioctl are: + * + * 1. Read + Exec permission will be checked using inode_permission() during + * path construction. -EACCES will be returned in case of failure. + * 2. Path construction will be stopped at the inode number which corresponds + * to the fd with which this ioctl is called. If constructed path does not + * exist under fd's inode, -EACCES will be returned. + * 3. The name of bottom subvolume is also searched and filled. + */ +static int btrfs_ioctl_ino_lookup_user(struct file *file, void __user *argp) +{ + struct btrfs_ioctl_ino_lookup_user_args *args; + struct inode *inode; + int ret; + + args = memdup_user(argp, sizeof(*args)); + if (IS_ERR(args)) + return PTR_ERR(args); + + inode = file_inode(file); + + if (args->dirid == BTRFS_FIRST_FREE_OBJECTID && + BTRFS_I(inode)->location.objectid != BTRFS_FIRST_FREE_OBJECTID) { + /* + * The subvolume does not exist under fd with which this is + * called + */ + kfree(args); + return -EACCES; + } + + ret = btrfs_search_path_in_tree_user(inode, args); + + if (ret == 0 && copy_to_user(argp, args, sizeof(*args))) + ret = -EFAULT; + + kfree(args); + return ret; +} + +/* Get the subvolume information in BTRFS_ROOT_ITEM and BTRFS_ROOT_BACKREF */ +static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp) +{ + struct btrfs_ioctl_get_subvol_info_args *subvol_info; + struct btrfs_fs_info *fs_info; + struct btrfs_root *root; + struct btrfs_path *path; + struct btrfs_key key; + struct btrfs_root_item *root_item; + struct btrfs_root_ref *rref; + struct extent_buffer *leaf; + unsigned long item_off; + unsigned long item_len; + struct inode *inode; + int slot; + int ret = 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + subvol_info = kzalloc(sizeof(*subvol_info), GFP_KERNEL); + if (!subvol_info) { + btrfs_free_path(path); + return -ENOMEM; + } + + inode = file_inode(file); + fs_info = BTRFS_I(inode)->root->fs_info; + + /* Get root_item of inode's subvolume */ + key.objectid = BTRFS_I(inode)->root->root_key.objectid; + key.type = BTRFS_ROOT_ITEM_KEY; + key.offset = (u64)-1; + root = btrfs_read_fs_root_no_name(fs_info, &key); + if (IS_ERR(root)) { + ret = PTR_ERR(root); + goto out; + } + root_item = &root->root_item; + + subvol_info->treeid = key.objectid; + + subvol_info->generation = btrfs_root_generation(root_item); + subvol_info->flags = btrfs_root_flags(root_item); + + memcpy(subvol_info->uuid, root_item->uuid, BTRFS_UUID_SIZE); + memcpy(subvol_info->parent_uuid, root_item->parent_uuid, + BTRFS_UUID_SIZE); + memcpy(subvol_info->received_uuid, root_item->received_uuid, + BTRFS_UUID_SIZE); + + subvol_info->ctransid = btrfs_root_ctransid(root_item); + subvol_info->ctime.sec = btrfs_stack_timespec_sec(&root_item->ctime); + subvol_info->ctime.nsec = btrfs_stack_timespec_nsec(&root_item->ctime); + + subvol_info->otransid = btrfs_root_otransid(root_item); + subvol_info->otime.sec = btrfs_stack_timespec_sec(&root_item->otime); + subvol_info->otime.nsec = btrfs_stack_timespec_nsec(&root_item->otime); + + subvol_info->stransid = btrfs_root_stransid(root_item); + subvol_info->stime.sec = btrfs_stack_timespec_sec(&root_item->stime); + subvol_info->stime.nsec = btrfs_stack_timespec_nsec(&root_item->stime); + + subvol_info->rtransid = btrfs_root_rtransid(root_item); + subvol_info->rtime.sec = btrfs_stack_timespec_sec(&root_item->rtime); + subvol_info->rtime.nsec = btrfs_stack_timespec_nsec(&root_item->rtime); + + if (key.objectid != BTRFS_FS_TREE_OBJECTID) { + /* Search root tree for ROOT_BACKREF of this subvolume */ + root = fs_info->tree_root; + + key.type = BTRFS_ROOT_BACKREF_KEY; + key.offset = 0; + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) { + goto out; + } else if (path->slots[0] >= + btrfs_header_nritems(path->nodes[0])) { + ret = btrfs_next_leaf(root, path); + if (ret < 0) { + goto out; + } else if (ret > 0) { + ret = -EUCLEAN; + goto out; + } + } + + leaf = path->nodes[0]; + slot = path->slots[0]; + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.objectid == subvol_info->treeid && + key.type == BTRFS_ROOT_BACKREF_KEY) { + subvol_info->parent_id = key.offset; + + rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref); + subvol_info->dirid = btrfs_root_ref_dirid(leaf, rref); + + item_off = btrfs_item_ptr_offset(leaf, slot) + + sizeof(struct btrfs_root_ref); + item_len = btrfs_item_size_nr(leaf, slot) + - sizeof(struct btrfs_root_ref); + read_extent_buffer(leaf, subvol_info->name, + item_off, item_len); + } else { + ret = -ENOENT; + goto out; + } + } + + if (copy_to_user(argp, subvol_info, sizeof(*subvol_info))) + ret = -EFAULT; + +out: + btrfs_free_path(path); + kzfree(subvol_info); + return ret; +} + +/* + * Return ROOT_REF information of the subvolume containing this inode + * except the subvolume name. + */ +static int btrfs_ioctl_get_subvol_rootref(struct file *file, void __user *argp) +{ + struct btrfs_ioctl_get_subvol_rootref_args *rootrefs; + struct btrfs_root_ref *rref; + struct btrfs_root *root; + struct btrfs_path *path; + struct btrfs_key key; + struct extent_buffer *leaf; + struct inode *inode; + u64 objectid; + int slot; + int ret; + u8 found; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + rootrefs = memdup_user(argp, sizeof(*rootrefs)); + if (IS_ERR(rootrefs)) { + btrfs_free_path(path); + return PTR_ERR(rootrefs); + } + + inode = file_inode(file); + root = BTRFS_I(inode)->root->fs_info->tree_root; + objectid = BTRFS_I(inode)->root->root_key.objectid; + + key.objectid = objectid; + key.type = BTRFS_ROOT_REF_KEY; + key.offset = rootrefs->min_treeid; + found = 0; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) { + goto out; + } else if (path->slots[0] >= + btrfs_header_nritems(path->nodes[0])) { + ret = btrfs_next_leaf(root, path); + if (ret < 0) { + goto out; + } else if (ret > 0) { + ret = -EUCLEAN; + goto out; + } + } + while (1) { + leaf = path->nodes[0]; + slot = path->slots[0]; + + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.objectid != objectid || key.type != BTRFS_ROOT_REF_KEY) { + ret = 0; + goto out; + } + + if (found == BTRFS_MAX_ROOTREF_BUFFER_NUM) { + ret = -EOVERFLOW; + goto out; + } + + rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref); + rootrefs->rootref[found].treeid = key.offset; + rootrefs->rootref[found].dirid = + btrfs_root_ref_dirid(leaf, rref); + found++; + + ret = btrfs_next_item(root, path); + if (ret < 0) { + goto out; + } else if (ret > 0) { + ret = -EUCLEAN; + goto out; + } + } + +out: + if (!ret || ret == -EOVERFLOW) { + rootrefs->num_items = found; + /* update min_treeid for next search */ + if (found) + rootrefs->min_treeid = + rootrefs->rootref[found - 1].treeid + 1; + if (copy_to_user(argp, rootrefs, sizeof(*rootrefs))) + ret = -EFAULT; + } + + kfree(rootrefs); + btrfs_free_path(path); + + return ret; +} + static noinline int btrfs_ioctl_snap_destroy(struct file *file, void __user *arg) { @@ -2322,12 +2812,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_root *dest = NULL; struct btrfs_ioctl_vol_args *vol_args; - struct btrfs_trans_handle *trans; - struct btrfs_block_rsv block_rsv; - u64 root_flags; - u64 qgroup_reserved; int namelen; - int ret; int err = 0; if (!S_ISDIR(dir->i_mode)) @@ -2411,133 +2896,11 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, } inode_lock(inode); - - /* - * Don't allow to delete a subvolume with send in progress. This is - * inside the i_mutex so the error handling that has to drop the bit - * again is not run concurrently. - */ - spin_lock(&dest->root_item_lock); - root_flags = btrfs_root_flags(&dest->root_item); - if (dest->send_in_progress == 0) { - btrfs_set_root_flags(&dest->root_item, - root_flags | BTRFS_ROOT_SUBVOL_DEAD); - spin_unlock(&dest->root_item_lock); - } else { - spin_unlock(&dest->root_item_lock); - btrfs_warn(fs_info, - "Attempt to delete subvolume %llu during send", - dest->root_key.objectid); - err = -EPERM; - goto out_unlock_inode; - } - - down_write(&fs_info->subvol_sem); - - err = may_destroy_subvol(dest); - if (err) - goto out_up_write; - - btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP); - /* - * One for dir inode, two for dir entries, two for root - * ref/backref. - */ - err = btrfs_subvolume_reserve_metadata(root, &block_rsv, - 5, &qgroup_reserved, true); - if (err) - goto out_up_write; - - trans = btrfs_start_transaction(root, 0); - if (IS_ERR(trans)) { - err = PTR_ERR(trans); - goto out_release; - } - trans->block_rsv = &block_rsv; - trans->bytes_reserved = block_rsv.size; - - btrfs_record_snapshot_destroy(trans, BTRFS_I(dir)); - - ret = btrfs_unlink_subvol(trans, root, dir, - dest->root_key.objectid, - dentry->d_name.name, - dentry->d_name.len); - if (ret) { - err = ret; - btrfs_abort_transaction(trans, ret); - goto out_end_trans; - } - - btrfs_record_root_in_trans(trans, dest); - - memset(&dest->root_item.drop_progress, 0, - sizeof(dest->root_item.drop_progress)); - dest->root_item.drop_level = 0; - btrfs_set_root_refs(&dest->root_item, 0); - - if (!test_and_set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &dest->state)) { - ret = btrfs_insert_orphan_item(trans, - fs_info->tree_root, - dest->root_key.objectid); - if (ret) { - btrfs_abort_transaction(trans, ret); - err = ret; - goto out_end_trans; - } - } - - ret = btrfs_uuid_tree_rem(trans, fs_info, dest->root_item.uuid, - BTRFS_UUID_KEY_SUBVOL, - dest->root_key.objectid); - if (ret && ret != -ENOENT) { - btrfs_abort_transaction(trans, ret); - err = ret; - goto out_end_trans; - } - if (!btrfs_is_empty_uuid(dest->root_item.received_uuid)) { - ret = btrfs_uuid_tree_rem(trans, fs_info, - dest->root_item.received_uuid, - BTRFS_UUID_KEY_RECEIVED_SUBVOL, - dest->root_key.objectid); - if (ret && ret != -ENOENT) { - btrfs_abort_transaction(trans, ret); - err = ret; - goto out_end_trans; - } - } - -out_end_trans: - trans->block_rsv = NULL; - trans->bytes_reserved = 0; - ret = btrfs_end_transaction(trans); - if (ret && !err) - err = ret; - inode->i_flags |= S_DEAD; -out_release: - btrfs_subvolume_release_metadata(fs_info, &block_rsv); -out_up_write: - up_write(&fs_info->subvol_sem); - if (err) { - spin_lock(&dest->root_item_lock); - root_flags = btrfs_root_flags(&dest->root_item); - btrfs_set_root_flags(&dest->root_item, - root_flags & ~BTRFS_ROOT_SUBVOL_DEAD); - spin_unlock(&dest->root_item_lock); - } -out_unlock_inode: + err = btrfs_delete_subvolume(dir, dentry); inode_unlock(inode); - if (!err) { - d_invalidate(dentry); - btrfs_invalidate_inodes(dest); + if (!err) d_delete(dentry); - ASSERT(dest->send_in_progress == 0); - /* the last ref */ - if (dest->ino_cache_inode) { - iput(dest->ino_cache_inode); - dest->ino_cache_inode = NULL; - } - } out_dput: dput(dentry); out_unlock_dir: @@ -2626,7 +2989,6 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg) if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) return BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; - mutex_lock(&fs_info->volume_mutex); vol_args = memdup_user(arg, sizeof(*vol_args)); if (IS_ERR(vol_args)) { ret = PTR_ERR(vol_args); @@ -2641,7 +3003,6 @@ static long btrfs_ioctl_add_dev(struct btrfs_fs_info *fs_info, void __user *arg) kfree(vol_args); out: - mutex_unlock(&fs_info->volume_mutex); clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); return ret; } @@ -2667,8 +3028,10 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg) } /* Check for compatibility reject unknown flags */ - if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED) - return -EOPNOTSUPP; + if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED) { + ret = -EOPNOTSUPP; + goto out; + } if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) { ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS; @@ -2967,8 +3330,6 @@ static void btrfs_cmp_data_free(struct cmp_pages *cmp) put_page(pg); } } - kfree(cmp->src_pages); - kfree(cmp->dst_pages); } static int btrfs_cmp_data_prepare(struct inode *src, u64 loff, @@ -2977,40 +3338,14 @@ static int btrfs_cmp_data_prepare(struct inode *src, u64 loff, { int ret; int num_pages = PAGE_ALIGN(len) >> PAGE_SHIFT; - struct page **src_pgarr, **dst_pgarr; - /* - * We must gather up all the pages before we initiate our - * extent locking. We use an array for the page pointers. Size - * of the array is bounded by len, which is in turn bounded by - * BTRFS_MAX_DEDUPE_LEN. - */ - src_pgarr = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL); - dst_pgarr = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL); - if (!src_pgarr || !dst_pgarr) { - kfree(src_pgarr); - kfree(dst_pgarr); - return -ENOMEM; - } cmp->num_pages = num_pages; - cmp->src_pages = src_pgarr; - cmp->dst_pages = dst_pgarr; - - /* - * If deduping ranges in the same inode, locking rules make it mandatory - * to always lock pages in ascending order to avoid deadlocks with - * concurrent tasks (such as starting writeback/delalloc). - */ - if (src == dst && dst_loff < loff) { - swap(src_pgarr, dst_pgarr); - swap(loff, dst_loff); - } - ret = gather_extent_pages(src, src_pgarr, cmp->num_pages, loff); + ret = gather_extent_pages(src, cmp->src_pages, num_pages, loff); if (ret) goto out; - ret = gather_extent_pages(dst, dst_pgarr, cmp->num_pages, dst_loff); + ret = gather_extent_pages(dst, cmp->dst_pages, num_pages, dst_loff); out: if (ret) @@ -3080,31 +3415,23 @@ static int extent_same_check_offsets(struct inode *inode, u64 off, u64 *plen, return 0; } -static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen, - struct inode *dst, u64 dst_loff) +static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 olen, + struct inode *dst, u64 dst_loff, + struct cmp_pages *cmp) { int ret; u64 len = olen; - struct cmp_pages cmp; bool same_inode = (src == dst); u64 same_lock_start = 0; u64 same_lock_len = 0; - if (len == 0) - return 0; - - if (same_inode) - inode_lock(src); - else - btrfs_double_inode_lock(src, dst); - ret = extent_same_check_offsets(src, loff, &len, olen); if (ret) - goto out_unlock; + return ret; ret = extent_same_check_offsets(dst, dst_loff, &len, olen); if (ret) - goto out_unlock; + return ret; if (same_inode) { /* @@ -3121,32 +3448,21 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen, * allow an unaligned length so long as it ends at * i_size. */ - if (len != olen) { - ret = -EINVAL; - goto out_unlock; - } + if (len != olen) + return -EINVAL; /* Check for overlapping ranges */ - if (dst_loff + len > loff && dst_loff < loff + len) { - ret = -EINVAL; - goto out_unlock; - } + if (dst_loff + len > loff && dst_loff < loff + len) + return -EINVAL; same_lock_start = min_t(u64, loff, dst_loff); same_lock_len = max_t(u64, loff, dst_loff) + len - same_lock_start; } - /* don't make the dst file partly checksummed */ - if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) != - (BTRFS_I(dst)->flags & BTRFS_INODE_NODATASUM)) { - ret = -EINVAL; - goto out_unlock; - } - again: - ret = btrfs_cmp_data_prepare(src, loff, dst, dst_loff, olen, &cmp); + ret = btrfs_cmp_data_prepare(src, loff, dst, dst_loff, olen, cmp); if (ret) - goto out_unlock; + return ret; if (same_inode) ret = lock_extent_range(src, same_lock_start, same_lock_len, @@ -3167,7 +3483,7 @@ again: * Ranges in the io trees already unlocked. Now unlock all * pages before waiting for all IO to complete. */ - btrfs_cmp_data_free(&cmp); + btrfs_cmp_data_free(cmp); if (same_inode) { btrfs_wait_ordered_range(src, same_lock_start, same_lock_len); @@ -3180,12 +3496,12 @@ again: ASSERT(ret == 0); if (WARN_ON(ret)) { /* ranges in the io trees already unlocked */ - btrfs_cmp_data_free(&cmp); + btrfs_cmp_data_free(cmp); return ret; } /* pass original length for comparison so we stay within i_size */ - ret = btrfs_cmp_data(olen, &cmp); + ret = btrfs_cmp_data(olen, cmp); if (ret == 0) ret = btrfs_clone(src, dst, loff, olen, len, dst_loff, 1); @@ -3195,18 +3511,91 @@ again: else btrfs_double_extent_unlock(src, loff, dst, dst_loff, len); - btrfs_cmp_data_free(&cmp); + btrfs_cmp_data_free(cmp); + + return ret; +} + +#define BTRFS_MAX_DEDUPE_LEN SZ_16M + +static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen, + struct inode *dst, u64 dst_loff) +{ + int ret; + struct cmp_pages cmp; + int num_pages = PAGE_ALIGN(BTRFS_MAX_DEDUPE_LEN) >> PAGE_SHIFT; + bool same_inode = (src == dst); + u64 i, tail_len, chunk_count; + + if (olen == 0) + return 0; + + if (same_inode) + inode_lock(src); + else + btrfs_double_inode_lock(src, dst); + + /* don't make the dst file partly checksummed */ + if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) != + (BTRFS_I(dst)->flags & BTRFS_INODE_NODATASUM)) { + ret = -EINVAL; + goto out_unlock; + } + + tail_len = olen % BTRFS_MAX_DEDUPE_LEN; + chunk_count = div_u64(olen, BTRFS_MAX_DEDUPE_LEN); + if (chunk_count == 0) + num_pages = PAGE_ALIGN(tail_len) >> PAGE_SHIFT; + + /* + * If deduping ranges in the same inode, locking rules make it + * mandatory to always lock pages in ascending order to avoid deadlocks + * with concurrent tasks (such as starting writeback/delalloc). + */ + if (same_inode && dst_loff < loff) + swap(loff, dst_loff); + + /* + * We must gather up all the pages before we initiate our extent + * locking. We use an array for the page pointers. Size of the array is + * bounded by len, which is in turn bounded by BTRFS_MAX_DEDUPE_LEN. + */ + cmp.src_pages = kvmalloc_array(num_pages, sizeof(struct page *), + GFP_KERNEL | __GFP_ZERO); + cmp.dst_pages = kvmalloc_array(num_pages, sizeof(struct page *), + GFP_KERNEL | __GFP_ZERO); + if (!cmp.src_pages || !cmp.dst_pages) { + ret = -ENOMEM; + goto out_free; + } + + for (i = 0; i < chunk_count; i++) { + ret = btrfs_extent_same_range(src, loff, BTRFS_MAX_DEDUPE_LEN, + dst, dst_loff, &cmp); + if (ret) + goto out_unlock; + + loff += BTRFS_MAX_DEDUPE_LEN; + dst_loff += BTRFS_MAX_DEDUPE_LEN; + } + + if (tail_len > 0) + ret = btrfs_extent_same_range(src, loff, tail_len, dst, + dst_loff, &cmp); + out_unlock: if (same_inode) inode_unlock(src); else btrfs_double_inode_unlock(src, dst); +out_free: + kvfree(cmp.src_pages); + kvfree(cmp.dst_pages); + return ret; } -#define BTRFS_MAX_DEDUPE_LEN SZ_16M - ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen, struct file *dst_file, u64 dst_loff) { @@ -3215,9 +3604,6 @@ ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen, u64 bs = BTRFS_I(src)->root->fs_info->sb->s_blocksize; ssize_t res; - if (olen > BTRFS_MAX_DEDUPE_LEN) - olen = BTRFS_MAX_DEDUPE_LEN; - if (WARN_ON_ONCE(bs < PAGE_SIZE)) { /* * Btrfs does not support blocksize < page_size. As a @@ -3839,11 +4225,6 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src, src->i_sb != inode->i_sb) return -EXDEV; - /* don't make the dst file partly checksummed */ - if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) != - (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) - return -EINVAL; - if (S_ISDIR(src->i_mode) || S_ISDIR(inode->i_mode)) return -EISDIR; @@ -3853,6 +4234,13 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src, inode_lock(src); } + /* don't make the dst file partly checksummed */ + if ((BTRFS_I(src)->flags & BTRFS_INODE_NODATASUM) != + (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) { + ret = -EINVAL; + goto out_unlock; + } + /* determine range to clone */ ret = -EINVAL; if (off + len > src->i_size || off + len < off) @@ -4020,8 +4408,8 @@ out: return ret; } -void btrfs_get_block_group_info(struct list_head *groups_list, - struct btrfs_ioctl_space_info *space) +static void get_block_group_info(struct list_head *groups_list, + struct btrfs_ioctl_space_info *space) { struct btrfs_block_group_cache *block_group; @@ -4137,8 +4525,8 @@ static long btrfs_ioctl_space_info(struct btrfs_fs_info *fs_info, down_read(&info->groups_sem); for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) { if (!list_empty(&info->block_groups[c])) { - btrfs_get_block_group_info( - &info->block_groups[c], &space); + get_block_group_info(&info->block_groups[c], + &space); memcpy(dest, &space, sizeof(space)); dest++; space_args.total_spaces++; @@ -4503,14 +4891,14 @@ out_loi: return ret; } -void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock, +void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info, struct btrfs_ioctl_balance_args *bargs) { struct btrfs_balance_control *bctl = fs_info->balance_ctl; bargs->flags = bctl->flags; - if (atomic_read(&fs_info->balance_running)) + if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) bargs->state |= BTRFS_BALANCE_STATE_RUNNING; if (atomic_read(&fs_info->balance_pause_req)) bargs->state |= BTRFS_BALANCE_STATE_PAUSE_REQ; @@ -4521,13 +4909,9 @@ void update_ioctl_balance_args(struct btrfs_fs_info *fs_info, int lock, memcpy(&bargs->meta, &bctl->meta, sizeof(bargs->meta)); memcpy(&bargs->sys, &bctl->sys, sizeof(bargs->sys)); - if (lock) { - spin_lock(&fs_info->balance_lock); - memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat)); - spin_unlock(&fs_info->balance_lock); - } else { - memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat)); - } + spin_lock(&fs_info->balance_lock); + memcpy(&bargs->stat, &bctl->stat, sizeof(bargs->stat)); + spin_unlock(&fs_info->balance_lock); } static long btrfs_ioctl_balance(struct file *file, void __user *arg) @@ -4548,7 +4932,6 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg) again: if (!test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) { - mutex_lock(&fs_info->volume_mutex); mutex_lock(&fs_info->balance_mutex); need_unlock = true; goto locked; @@ -4563,21 +4946,22 @@ again: mutex_lock(&fs_info->balance_mutex); if (fs_info->balance_ctl) { /* this is either (2) or (3) */ - if (!atomic_read(&fs_info->balance_running)) { + if (!test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) { mutex_unlock(&fs_info->balance_mutex); - if (!mutex_trylock(&fs_info->volume_mutex)) - goto again; + /* + * Lock released to allow other waiters to continue, + * we'll reexamine the status again. + */ mutex_lock(&fs_info->balance_mutex); if (fs_info->balance_ctl && - !atomic_read(&fs_info->balance_running)) { + !test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) { /* this is (3) */ need_unlock = false; goto locked; } mutex_unlock(&fs_info->balance_mutex); - mutex_unlock(&fs_info->volume_mutex); goto again; } else { /* this is (2) */ @@ -4630,7 +5014,6 @@ locked: goto out_bargs; } - bctl->fs_info = fs_info; if (arg) { memcpy(&bctl->data, &bargs->data, sizeof(bctl->data)); memcpy(&bctl->meta, &bargs->meta, sizeof(bctl->meta)); @@ -4649,14 +5032,14 @@ locked: do_balance: /* - * Ownership of bctl and filesystem flag BTRFS_FS_EXCL_OP - * goes to to btrfs_balance. bctl is freed in __cancel_balance, - * or, if restriper was paused all the way until unmount, in - * free_fs_info. The flag is cleared in __cancel_balance. + * Ownership of bctl and filesystem flag BTRFS_FS_EXCL_OP goes to + * btrfs_balance. bctl is freed in reset_balance_state, or, if + * restriper was paused all the way until unmount, in free_fs_info. + * The flag should be cleared after reset_balance_state. */ need_unlock = false; - ret = btrfs_balance(bctl, bargs); + ret = btrfs_balance(fs_info, bctl, bargs); bctl = NULL; if (arg) { @@ -4670,7 +5053,6 @@ out_bargs: kfree(bargs); out_unlock: mutex_unlock(&fs_info->balance_mutex); - mutex_unlock(&fs_info->volume_mutex); if (need_unlock) clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); out: @@ -4714,7 +5096,7 @@ static long btrfs_ioctl_balance_progress(struct btrfs_fs_info *fs_info, goto out; } - update_ioctl_balance_args(fs_info, 1, bargs); + btrfs_update_ioctl_balance_args(fs_info, bargs); if (copy_to_user(arg, bargs, sizeof(*bargs))) ret = -EFAULT; @@ -5051,8 +5433,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, BTRFS_UUID_SIZE); if (received_uuid_changed && !btrfs_is_empty_uuid(root_item->received_uuid)) { - ret = btrfs_uuid_tree_rem(trans, fs_info, - root_item->received_uuid, + ret = btrfs_uuid_tree_remove(trans, root_item->received_uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, root->root_key.objectid); if (ret && ret != -ENOENT) { @@ -5076,7 +5457,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, goto out; } if (received_uuid_changed && !btrfs_is_empty_uuid(sa->uuid)) { - ret = btrfs_uuid_tree_add(trans, fs_info, sa->uuid, + ret = btrfs_uuid_tree_add(trans, sa->uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, root->root_key.objectid); if (ret < 0 && ret != -EEXIST) { @@ -5510,7 +5891,7 @@ long btrfs_ioctl(struct file *file, unsigned int case BTRFS_IOC_SYNC: { int ret; - ret = btrfs_start_delalloc_roots(fs_info, 0, -1); + ret = btrfs_start_delalloc_roots(fs_info, -1); if (ret) return ret; ret = btrfs_sync_fs(inode->i_sb, 1); @@ -5578,6 +5959,16 @@ long btrfs_ioctl(struct file *file, unsigned int return btrfs_ioctl_get_features(file, argp); case BTRFS_IOC_SET_FEATURES: return btrfs_ioctl_set_features(file, argp); + case FS_IOC_FSGETXATTR: + return btrfs_ioctl_fsgetxattr(file, argp); + case FS_IOC_FSSETXATTR: + return btrfs_ioctl_fssetxattr(file, argp); + case BTRFS_IOC_GET_SUBVOL_INFO: + return btrfs_ioctl_get_subvol_info(file, argp); + case BTRFS_IOC_GET_SUBVOL_ROOTREF: + return btrfs_ioctl_get_subvol_rootref(file, argp); + case BTRFS_IOC_INO_LOOKUP_USER: + return btrfs_ioctl_ino_lookup_user(file, argp); } return -ENOTTY; diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 621083f8932c..1da768e5ef75 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -1,20 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ + #include <linux/sched.h> #include <linux/pagemap.h> #include <linux/spinlock.h> @@ -78,22 +66,16 @@ void btrfs_clear_lock_blocking_rw(struct extent_buffer *eb, int rw) write_lock(&eb->lock); WARN_ON(atomic_read(&eb->spinning_writers)); atomic_inc(&eb->spinning_writers); - /* - * atomic_dec_and_test implies a barrier for waitqueue_active - */ - if (atomic_dec_and_test(&eb->blocking_writers) && - waitqueue_active(&eb->write_lock_wq)) - wake_up(&eb->write_lock_wq); + /* atomic_dec_and_test implies a barrier */ + if (atomic_dec_and_test(&eb->blocking_writers)) + cond_wake_up_nomb(&eb->write_lock_wq); } else if (rw == BTRFS_READ_LOCK_BLOCKING) { BUG_ON(atomic_read(&eb->blocking_readers) == 0); read_lock(&eb->lock); atomic_inc(&eb->spinning_readers); - /* - * atomic_dec_and_test implies a barrier for waitqueue_active - */ - if (atomic_dec_and_test(&eb->blocking_readers) && - waitqueue_active(&eb->read_lock_wq)) - wake_up(&eb->read_lock_wq); + /* atomic_dec_and_test implies a barrier */ + if (atomic_dec_and_test(&eb->blocking_readers)) + cond_wake_up_nomb(&eb->read_lock_wq); } } @@ -233,12 +215,9 @@ void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb) } btrfs_assert_tree_read_locked(eb); WARN_ON(atomic_read(&eb->blocking_readers) == 0); - /* - * atomic_dec_and_test implies a barrier for waitqueue_active - */ - if (atomic_dec_and_test(&eb->blocking_readers) && - waitqueue_active(&eb->read_lock_wq)) - wake_up(&eb->read_lock_wq); + /* atomic_dec_and_test implies a barrier */ + if (atomic_dec_and_test(&eb->blocking_readers)) + cond_wake_up_nomb(&eb->read_lock_wq); atomic_dec(&eb->read_locks); } @@ -287,12 +266,9 @@ void btrfs_tree_unlock(struct extent_buffer *eb) if (blockers) { WARN_ON(atomic_read(&eb->spinning_writers)); atomic_dec(&eb->blocking_writers); - /* - * Make sure counter is updated before we wake up waiters. - */ + /* Use the lighter barrier after atomic */ smp_mb__after_atomic(); - if (waitqueue_active(&eb->write_lock_wq)) - wake_up(&eb->write_lock_wq); + cond_wake_up_nomb(&eb->write_lock_wq); } else { WARN_ON(atomic_read(&eb->spinning_writers) != 1); atomic_dec(&eb->spinning_writers); diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h index c44a9d5f5362..29135def468e 100644 --- a/fs/btrfs/locking.h +++ b/fs/btrfs/locking.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_LOCKING_ -#define __BTRFS_LOCKING_ +#ifndef BTRFS_LOCKING_H +#define BTRFS_LOCKING_H #define BTRFS_WRITE_LOCK 1 #define BTRFS_READ_LOCK 2 diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c index 1c7f7f70caf4..b6a4cc178bee 100644 --- a/fs/btrfs/lzo.c +++ b/fs/btrfs/lzo.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/kernel.h> @@ -30,6 +17,43 @@ #define LZO_LEN 4 +/* + * Btrfs LZO compression format + * + * Regular and inlined LZO compressed data extents consist of: + * + * 1. Header + * Fixed size. LZO_LEN (4) bytes long, LE32. + * Records the total size (including the header) of compressed data. + * + * 2. Segment(s) + * Variable size. Each segment includes one segment header, followd by data + * payload. + * One regular LZO compressed extent can have one or more segments. + * For inlined LZO compressed extent, only one segment is allowed. + * One segment represents at most one page of uncompressed data. + * + * 2.1 Segment header + * Fixed size. LZO_LEN (4) bytes long, LE32. + * Records the total size of the segment (not including the header). + * Segment header never crosses page boundary, thus it's possible to + * have at most 3 padding zeros at the end of the page. + * + * 2.2 Data Payload + * Variable size. Size up limit should be lzo1x_worst_compress(PAGE_SIZE) + * which is 4419 for a 4KiB page. + * + * Example: + * Page 1: + * 0 0x2 0x4 0x6 0x8 0xa 0xc 0xe 0x10 + * 0x0000 | Header | SegHdr 01 | Data payload 01 ... | + * ... + * 0x0ff0 | SegHdr N | Data payload N ... |00| + * ^^ padding zeros + * Page 2: + * 0x1000 | SegHdr N+1| Data payload N+1 ... | + */ + struct workspace { void *mem; void *buf; /* where decompressed data goes */ @@ -271,6 +295,7 @@ static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) unsigned long working_bytes; size_t in_len; size_t out_len; + const size_t max_segment_len = lzo1x_worst_compress(PAGE_SIZE); unsigned long in_offset; unsigned long in_page_bytes_left; unsigned long tot_in; @@ -284,10 +309,22 @@ static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) data_in = kmap(pages_in[0]); tot_len = read_compress_length(data_in); + /* + * Compressed data header check. + * + * The real compressed size can't exceed the maximum extent length, and + * all pages should be used (whole unused page with just the segment + * header is not possible). If this happens it means the compressed + * extent is corrupted. + */ + if (tot_len > min_t(size_t, BTRFS_MAX_COMPRESSED, srclen) || + tot_len < srclen - PAGE_SIZE) { + ret = -EUCLEAN; + goto done; + } tot_in = LZO_LEN; in_offset = LZO_LEN; - tot_len = min_t(size_t, srclen, tot_len); in_page_bytes_left = PAGE_SIZE - LZO_LEN; tot_out = 0; @@ -298,6 +335,17 @@ static int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) in_offset += LZO_LEN; tot_in += LZO_LEN; + /* + * Segment header check. + * + * The segment length must not exceed the maximum LZO + * compression size, nor the total compressed size. + */ + if (in_len > max_segment_len || tot_in + in_len > tot_len) { + ret = -EUCLEAN; + goto done; + } + tot_in += in_len; working_bytes = in_len; may_late_unmap = need_unmap = false; @@ -348,7 +396,7 @@ cont: } } - out_len = lzo1x_worst_compress(PAGE_SIZE); + out_len = max_segment_len; ret = lzo1x_decompress_safe(buf, in_len, workspace->buf, &out_len); if (need_unmap) @@ -382,15 +430,24 @@ static int lzo_decompress(struct list_head *ws, unsigned char *data_in, struct workspace *workspace = list_entry(ws, struct workspace, list); size_t in_len; size_t out_len; + size_t max_segment_len = lzo1x_worst_compress(PAGE_SIZE); int ret = 0; char *kaddr; unsigned long bytes; - BUG_ON(srclen < LZO_LEN); + if (srclen < LZO_LEN || srclen > max_segment_len + LZO_LEN * 2) + return -EUCLEAN; + in_len = read_compress_length(data_in); + if (in_len != srclen) + return -EUCLEAN; data_in += LZO_LEN; in_len = read_compress_length(data_in); + if (in_len != srclen - LZO_LEN * 2) { + ret = -EUCLEAN; + goto out; + } data_in += LZO_LEN; out_len = PAGE_SIZE; diff --git a/fs/btrfs/math.h b/fs/btrfs/math.h index 1b10a3cd1195..75246f2f56ba 100644 --- a/fs/btrfs/math.h +++ b/fs/btrfs/math.h @@ -1,25 +1,11 @@ - +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2012 Fujitsu. All rights reserved. * Written by Miao Xie <miaox@cn.fujitsu.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_MATH_H -#define __BTRFS_MATH_H +#ifndef BTRFS_MATH_H +#define BTRFS_MATH_H #include <asm/div64.h> diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 661cc3db0c7c..2e1a1694a33d 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/slab.h> @@ -356,11 +343,8 @@ int btrfs_dec_test_first_ordered_pending(struct inode *inode, if (entry->bytes_left == 0) { ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags); - /* - * Implicit memory barrier after test_and_set_bit - */ - if (waitqueue_active(&entry->wait)) - wake_up(&entry->wait); + /* test_and_set_bit implies a barrier */ + cond_wake_up_nomb(&entry->wait); } else { ret = 1; } @@ -423,11 +407,8 @@ have_entry: if (entry->bytes_left == 0) { ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags); - /* - * Implicit memory barrier after test_and_set_bit - */ - if (waitqueue_active(&entry->wait)) - wake_up(&entry->wait); + /* test_and_set_bit implies a barrier */ + cond_wake_up_nomb(&entry->wait); } else { ret = 1; } diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h index 4a1672a13ba6..3be443fb3001 100644 --- a/fs/btrfs/ordered-data.h +++ b/fs/btrfs/ordered-data.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_ORDERED_DATA__ -#define __BTRFS_ORDERED_DATA__ +#ifndef BTRFS_ORDERED_DATA_H +#define BTRFS_ORDERED_DATA_H /* one of these per inode */ struct btrfs_ordered_inode_tree { @@ -218,4 +205,5 @@ void btrfs_wait_logged_extents(struct btrfs_trans_handle *trans, void btrfs_free_logged_extents(struct btrfs_root *log, u64 transid); int __init ordered_data_init(void); void __cold ordered_data_exit(void); + #endif diff --git a/fs/btrfs/orphan.c b/fs/btrfs/orphan.c index 47767d5b8f0b..aa534108c1e2 100644 --- a/fs/btrfs/orphan.c +++ b/fs/btrfs/orphan.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Red Hat. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include "ctree.h" diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c index 4a8770485f77..a4e11cf04671 100644 --- a/fs/btrfs/print-tree.c +++ b/fs/btrfs/print-tree.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include "ctree.h" @@ -179,6 +166,25 @@ static void print_uuid_item(struct extent_buffer *l, unsigned long offset, } } +/* + * Helper to output refs and locking status of extent buffer. Useful to debug + * race condition related problems. + */ +static void print_eb_refs_lock(struct extent_buffer *eb) +{ +#ifdef CONFIG_BTRFS_DEBUG + btrfs_info(eb->fs_info, +"refs %u lock (w:%d r:%d bw:%d br:%d sw:%d sr:%d) lock_owner %u current %u", + atomic_read(&eb->refs), atomic_read(&eb->write_locks), + atomic_read(&eb->read_locks), + atomic_read(&eb->blocking_writers), + atomic_read(&eb->blocking_readers), + atomic_read(&eb->spinning_writers), + atomic_read(&eb->spinning_readers), + eb->lock_owner, current->pid); +#endif +} + void btrfs_print_leaf(struct extent_buffer *l) { struct btrfs_fs_info *fs_info; @@ -202,9 +208,11 @@ void btrfs_print_leaf(struct extent_buffer *l) fs_info = l->fs_info; nr = btrfs_header_nritems(l); - btrfs_info(fs_info, "leaf %llu total ptrs %d free space %d", - btrfs_header_bytenr(l), nr, - btrfs_leaf_free_space(fs_info, l)); + btrfs_info(fs_info, + "leaf %llu gen %llu total ptrs %d free space %d owner %llu", + btrfs_header_bytenr(l), btrfs_header_generation(l), nr, + btrfs_leaf_free_space(fs_info, l), btrfs_header_owner(l)); + print_eb_refs_lock(l); for (i = 0 ; i < nr ; i++) { item = btrfs_item_nr(i); btrfs_item_key_to_cpu(l, &key, i); @@ -338,7 +346,7 @@ void btrfs_print_leaf(struct extent_buffer *l) } } -void btrfs_print_tree(struct extent_buffer *c) +void btrfs_print_tree(struct extent_buffer *c, bool follow) { struct btrfs_fs_info *fs_info; int i; u32 nr; @@ -355,15 +363,20 @@ void btrfs_print_tree(struct extent_buffer *c) return; } btrfs_info(fs_info, - "node %llu level %d total ptrs %d free spc %u", - btrfs_header_bytenr(c), level, nr, - (u32)BTRFS_NODEPTRS_PER_BLOCK(fs_info) - nr); + "node %llu level %d gen %llu total ptrs %d free spc %u owner %llu", + btrfs_header_bytenr(c), level, btrfs_header_generation(c), + nr, (u32)BTRFS_NODEPTRS_PER_BLOCK(fs_info) - nr, + btrfs_header_owner(c)); + print_eb_refs_lock(c); for (i = 0; i < nr; i++) { btrfs_node_key_to_cpu(c, &key, i); - pr_info("\tkey %d (%llu %u %llu) block %llu\n", + pr_info("\tkey %d (%llu %u %llu) block %llu gen %llu\n", i, key.objectid, key.type, key.offset, - btrfs_node_blockptr(c, i)); + btrfs_node_blockptr(c, i), + btrfs_node_ptr_generation(c, i)); } + if (!follow) + return; for (i = 0; i < nr; i++) { struct btrfs_key first_key; struct extent_buffer *next; @@ -385,7 +398,7 @@ void btrfs_print_tree(struct extent_buffer *c) if (btrfs_header_level(next) != level - 1) BUG(); - btrfs_print_tree(next); + btrfs_print_tree(next, follow); free_extent_buffer(next); } } diff --git a/fs/btrfs/print-tree.h b/fs/btrfs/print-tree.h index 3afd508ed8c5..e6bb38fd75ad 100644 --- a/fs/btrfs/print-tree.h +++ b/fs/btrfs/print-tree.h @@ -1,23 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __PRINT_TREE_ -#define __PRINT_TREE_ +#ifndef BTRFS_PRINT_TREE_H +#define BTRFS_PRINT_TREE_H + void btrfs_print_leaf(struct extent_buffer *l); -void btrfs_print_tree(struct extent_buffer *c); +void btrfs_print_tree(struct extent_buffer *c, bool follow); + #endif diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c index 5859f7d3cf3e..dc6140013ae8 100644 --- a/fs/btrfs/props.c +++ b/fs/btrfs/props.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2014 Filipe David Borba Manana <fdmanana@gmail.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/hashtable.h> @@ -393,6 +380,7 @@ static int prop_compression_apply(struct inode *inode, const char *value, size_t len) { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); int type; if (len == 0) { @@ -403,14 +391,17 @@ static int prop_compression_apply(struct inode *inode, return 0; } - if (!strncmp("lzo", value, 3)) + if (!strncmp("lzo", value, 3)) { type = BTRFS_COMPRESS_LZO; - else if (!strncmp("zlib", value, 4)) + btrfs_set_fs_incompat(fs_info, COMPRESS_LZO); + } else if (!strncmp("zlib", value, 4)) { type = BTRFS_COMPRESS_ZLIB; - else if (!strncmp("zstd", value, len)) + } else if (!strncmp("zstd", value, len)) { type = BTRFS_COMPRESS_ZSTD; - else + btrfs_set_fs_incompat(fs_info, COMPRESS_ZSTD); + } else { return -EINVAL; + } BTRFS_I(inode)->flags &= ~BTRFS_INODE_NOCOMPRESS; BTRFS_I(inode)->flags |= BTRFS_INODE_COMPRESS; diff --git a/fs/btrfs/props.h b/fs/btrfs/props.h index 100f18829d50..618815b4f9d5 100644 --- a/fs/btrfs/props.h +++ b/fs/btrfs/props.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2014 Filipe David Borba Manana <fdmanana@gmail.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_PROPS_H -#define __BTRFS_PROPS_H +#ifndef BTRFS_PROPS_H +#define BTRFS_PROPS_H #include "ctree.h" diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index f583f13ff26e..1874a6d2e6f5 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2011 STRATO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> @@ -24,6 +11,7 @@ #include <linux/slab.h> #include <linux/workqueue.h> #include <linux/btrfs.h> +#include <linux/sizes.h> #include "ctree.h" #include "transaction.h" @@ -1894,8 +1882,8 @@ static int qgroup_update_counters(struct btrfs_fs_info *fs_info, cur_old_count = btrfs_qgroup_get_old_refcnt(qg, seq); cur_new_count = btrfs_qgroup_get_new_refcnt(qg, seq); - trace_qgroup_update_counters(fs_info, qg->qgroupid, - cur_old_count, cur_new_count); + trace_qgroup_update_counters(fs_info, qg, cur_old_count, + cur_new_count); /* Rfer update part */ if (cur_old_count == 0 && cur_new_count > 0) { @@ -2026,8 +2014,8 @@ btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans, BUG_ON(!fs_info->quota_root); - trace_btrfs_qgroup_account_extent(fs_info, bytenr, num_bytes, - nr_old_roots, nr_new_roots); + trace_btrfs_qgroup_account_extent(fs_info, trans->transid, bytenr, + num_bytes, nr_old_roots, nr_new_roots); qgroups = ulist_alloc(GFP_NOFS); if (!qgroups) { @@ -2388,8 +2376,21 @@ out: return ret; } -static bool qgroup_check_limits(const struct btrfs_qgroup *qg, u64 num_bytes) +/* + * Two limits to commit transaction in advance. + * + * For RATIO, it will be 1/RATIO of the remaining limit + * (excluding data and prealloc meta) as threshold. + * For SIZE, it will be in byte unit as threshold. + */ +#define QGROUP_PERTRANS_RATIO 32 +#define QGROUP_PERTRANS_SIZE SZ_32M +static bool qgroup_check_limits(struct btrfs_fs_info *fs_info, + const struct btrfs_qgroup *qg, u64 num_bytes) { + u64 limit; + u64 threshold; + if ((qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_RFER) && qgroup_rsv_total(qg) + (s64)qg->rfer + num_bytes > qg->max_rfer) return false; @@ -2398,6 +2399,31 @@ static bool qgroup_check_limits(const struct btrfs_qgroup *qg, u64 num_bytes) qgroup_rsv_total(qg) + (s64)qg->excl + num_bytes > qg->max_excl) return false; + /* + * Even if we passed the check, it's better to check if reservation + * for meta_pertrans is pushing us near limit. + * If there is too much pertrans reservation or it's near the limit, + * let's try commit transaction to free some, using transaction_kthread + */ + if ((qg->lim_flags & (BTRFS_QGROUP_LIMIT_MAX_RFER | + BTRFS_QGROUP_LIMIT_MAX_EXCL))) { + if (qg->lim_flags & BTRFS_QGROUP_LIMIT_MAX_EXCL) + limit = qg->max_excl; + else + limit = qg->max_rfer; + threshold = (limit - qg->rsv.values[BTRFS_QGROUP_RSV_DATA] - + qg->rsv.values[BTRFS_QGROUP_RSV_META_PREALLOC]) / + QGROUP_PERTRANS_RATIO; + threshold = min_t(u64, threshold, QGROUP_PERTRANS_SIZE); + + /* + * Use transaction_kthread to commit transaction, so we no + * longer need to bother nested transaction nor lock context. + */ + if (qg->rsv.values[BTRFS_QGROUP_RSV_META_PERTRANS] > threshold) + btrfs_commit_transaction_locksafe(fs_info); + } + return true; } @@ -2447,7 +2473,7 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes, bool enforce, qg = unode_aux_to_qgroup(unode); - if (enforce && !qgroup_check_limits(qg, num_bytes)) { + if (enforce && !qgroup_check_limits(fs_info, qg, num_bytes)) { ret = -EDQUOT; goto out; } @@ -2554,6 +2580,21 @@ out: } /* + * Check if the leaf is the last leaf. Which means all node pointers + * are at their last position. + */ +static bool is_last_leaf(struct btrfs_path *path) +{ + int i; + + for (i = 1; i < BTRFS_MAX_LEVEL && path->nodes[i]; i++) { + if (path->slots[i] != btrfs_header_nritems(path->nodes[i]) - 1) + return false; + } + return true; +} + +/* * returns < 0 on error, 0 when more leafs are to be scanned. * returns 1 when done. */ @@ -2564,8 +2605,8 @@ qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path, struct btrfs_key found; struct extent_buffer *scratch_leaf = NULL; struct ulist *roots = NULL; - struct seq_list tree_mod_seq_elem = SEQ_LIST_INIT(tree_mod_seq_elem); u64 num_bytes; + bool done; int slot; int ret; @@ -2594,12 +2635,12 @@ qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path, mutex_unlock(&fs_info->qgroup_rescan_lock); return ret; } + done = is_last_leaf(path); btrfs_item_key_to_cpu(path->nodes[0], &found, btrfs_header_nritems(path->nodes[0]) - 1); fs_info->qgroup_rescan_progress.objectid = found.objectid + 1; - btrfs_get_tree_mod_seq(fs_info, &tree_mod_seq_elem); scratch_leaf = btrfs_clone_extent_buffer(path->nodes[0]); if (!scratch_leaf) { ret = -ENOMEM; @@ -2638,8 +2679,9 @@ out: btrfs_tree_read_unlock_blocking(scratch_leaf); free_extent_buffer(scratch_leaf); } - btrfs_put_tree_mod_seq(fs_info, &tree_mod_seq_elem); + if (done && !ret) + ret = 1; return ret; } @@ -2655,6 +2697,12 @@ static void btrfs_qgroup_rescan_worker(struct btrfs_work *work) path = btrfs_alloc_path(); if (!path) goto out; + /* + * Rescan should only search for commit root, and any later difference + * should be recorded by qgroup + */ + path->search_commit_root = 1; + path->skip_locking = 1; err = 0; while (!err && !btrfs_fs_closing(fs_info)) { @@ -2734,26 +2782,36 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, { int ret = 0; - if (!init_flags && - (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) || - !(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON))) { - ret = -EINVAL; - goto err; + if (!init_flags) { + /* we're resuming qgroup rescan at mount time */ + if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN)) + btrfs_warn(fs_info, + "qgroup rescan init failed, qgroup is not enabled"); + else if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON)) + btrfs_warn(fs_info, + "qgroup rescan init failed, qgroup rescan is not queued"); + return -EINVAL; } mutex_lock(&fs_info->qgroup_rescan_lock); spin_lock(&fs_info->qgroup_lock); if (init_flags) { - if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) + if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) { + btrfs_warn(fs_info, + "qgroup rescan is already in progress"); ret = -EINPROGRESS; - else if (!(fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_ON)) + } else if (!(fs_info->qgroup_flags & + BTRFS_QGROUP_STATUS_FLAG_ON)) { + btrfs_warn(fs_info, + "qgroup rescan init failed, qgroup is not enabled"); ret = -EINVAL; + } if (ret) { spin_unlock(&fs_info->qgroup_lock); mutex_unlock(&fs_info->qgroup_rescan_lock); - goto err; + return ret; } fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_RESCAN; } @@ -2772,13 +2830,6 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, btrfs_init_work(&fs_info->qgroup_rescan_work, btrfs_qgroup_rescan_helper, btrfs_qgroup_rescan_worker, NULL, NULL); - - if (ret) { -err: - btrfs_info(fs_info, "qgroup_rescan_init failed with %d", ret); - return ret; - } - return 0; } diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h index e63e2d497a8e..d60dd06445ce 100644 --- a/fs/btrfs/qgroup.h +++ b/fs/btrfs/qgroup.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2014 Facebook. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_QGROUP__ -#define __BTRFS_QGROUP__ +#ifndef BTRFS_QGROUP_H +#define BTRFS_QGROUP_H #include "ulist.h" #include "delayed-ref.h" @@ -341,4 +328,5 @@ void btrfs_qgroup_free_meta_all_pertrans(struct btrfs_root *root); void btrfs_qgroup_convert_reserved_meta(struct btrfs_root *root, int num_bytes); void btrfs_qgroup_check_reserved_leak(struct inode *inode); -#endif /* __BTRFS_QGROUP__ */ + +#endif diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index c3a2bc8af675..5e4ad134b9ad 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -1,21 +1,9 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2012 Fusion-io All rights reserved. * Copyright (C) 2012 Intel Corp. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ + #include <linux/sched.h> #include <linux/wait.h> #include <linux/bio.h> @@ -175,6 +163,12 @@ struct btrfs_raid_bio { * bitmap to record which horizontal stripe has data */ unsigned long *dbitmap; + + /* allocated with real_stripes-many pointers for finish_*() calls */ + void **finish_pointers; + + /* allocated with stripe_npages-many bits for finish_*() calls */ + unsigned long *finish_pbitmap; }; static int __raid56_parity_recover(struct btrfs_raid_bio *rbio); @@ -993,9 +987,14 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info, int stripe_npages = DIV_ROUND_UP(stripe_len, PAGE_SIZE); void *p; - rbio = kzalloc(sizeof(*rbio) + num_pages * sizeof(struct page *) * 2 + - DIV_ROUND_UP(stripe_npages, BITS_PER_LONG) * - sizeof(long), GFP_NOFS); + rbio = kzalloc(sizeof(*rbio) + + sizeof(*rbio->stripe_pages) * num_pages + + sizeof(*rbio->bio_pages) * num_pages + + sizeof(*rbio->finish_pointers) * real_stripes + + sizeof(*rbio->dbitmap) * BITS_TO_LONGS(stripe_npages) + + sizeof(*rbio->finish_pbitmap) * + BITS_TO_LONGS(stripe_npages), + GFP_NOFS); if (!rbio) return ERR_PTR(-ENOMEM); @@ -1017,13 +1016,20 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info, atomic_set(&rbio->stripes_pending, 0); /* - * the stripe_pages and bio_pages array point to the extra + * the stripe_pages, bio_pages, etc arrays point to the extra * memory we allocated past the end of the rbio */ p = rbio + 1; - rbio->stripe_pages = p; - rbio->bio_pages = p + sizeof(struct page *) * num_pages; - rbio->dbitmap = p + sizeof(struct page *) * num_pages * 2; +#define CONSUME_ALLOC(ptr, count) do { \ + ptr = p; \ + p = (unsigned char *)p + sizeof(*(ptr)) * (count); \ + } while (0) + CONSUME_ALLOC(rbio->stripe_pages, num_pages); + CONSUME_ALLOC(rbio->bio_pages, num_pages); + CONSUME_ALLOC(rbio->finish_pointers, real_stripes); + CONSUME_ALLOC(rbio->dbitmap, BITS_TO_LONGS(stripe_npages)); + CONSUME_ALLOC(rbio->finish_pbitmap, BITS_TO_LONGS(stripe_npages)); +#undef CONSUME_ALLOC if (bbio->map_type & BTRFS_BLOCK_GROUP_RAID5) nr_data = real_stripes - 1; @@ -1192,7 +1198,7 @@ static void index_rbio_pages(struct btrfs_raid_bio *rbio) static noinline void finish_rmw(struct btrfs_raid_bio *rbio) { struct btrfs_bio *bbio = rbio->bbio; - void *pointers[rbio->real_stripes]; + void **pointers = rbio->finish_pointers; int nr_data = rbio->nr_data; int stripe; int pagenr; @@ -2362,8 +2368,8 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, int need_check) { struct btrfs_bio *bbio = rbio->bbio; - void *pointers[rbio->real_stripes]; - DECLARE_BITMAP(pbitmap, rbio->stripe_npages); + void **pointers = rbio->finish_pointers; + unsigned long *pbitmap = rbio->finish_pbitmap; int nr_data = rbio->nr_data; int stripe; int pagenr; diff --git a/fs/btrfs/raid56.h b/fs/btrfs/raid56.h index 4ee4fe346838..f5d4c13a8dbc 100644 --- a/fs/btrfs/raid56.h +++ b/fs/btrfs/raid56.h @@ -1,24 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2012 Fusion-io All rights reserved. * Copyright (C) 2012 Intel Corp. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_RAID56__ -#define __BTRFS_RAID56__ +#ifndef BTRFS_RAID56_H +#define BTRFS_RAID56_H + static inline int nr_parity_stripes(struct map_lookup *map) { if (map->type & BTRFS_BLOCK_GROUP_RAID5) @@ -65,4 +53,5 @@ void raid56_submit_missing_rbio(struct btrfs_raid_bio *rbio); int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info); void btrfs_free_stripe_hash_table(struct btrfs_fs_info *info); + #endif diff --git a/fs/btrfs/rcu-string.h b/fs/btrfs/rcu-string.h index 9e111e4576d4..a97dc74a4d3d 100644 --- a/fs/btrfs/rcu-string.h +++ b/fs/btrfs/rcu-string.h @@ -1,21 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2012 Red Hat. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ +#ifndef BTRFS_RCU_STRING_H +#define BTRFS_RCU_STRING_H + struct rcu_string { struct rcu_head rcu; char str[0]; @@ -54,3 +44,5 @@ static inline void rcu_string_free(struct rcu_string *str) struct rcu_string *__str = rcu_dereference(rcu_str); \ __str->str; \ }) + +#endif diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index a52dd12af648..40f1bcef394d 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2011 STRATO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c index 35fab67dcbe8..e5b9e596bb92 100644 --- a/fs/btrfs/ref-verify.c +++ b/fs/btrfs/ref-verify.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2014 Facebook. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> diff --git a/fs/btrfs/ref-verify.h b/fs/btrfs/ref-verify.h index 3bf02ce0e1e2..b7d2a4edfdb7 100644 --- a/fs/btrfs/ref-verify.h +++ b/fs/btrfs/ref-verify.h @@ -1,22 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2014 Facebook. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __REF_VERIFY__ -#define __REF_VERIFY__ + +#ifndef BTRFS_REF_VERIFY_H +#define BTRFS_REF_VERIFY_H #ifdef CONFIG_BTRFS_FS_REF_VERIFY int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info); @@ -59,4 +47,5 @@ static inline void btrfs_init_ref_verify(struct btrfs_fs_info *fs_info) } #endif /* CONFIG_BTRFS_FS_REF_VERIFY */ -#endif /* _REF_VERIFY__ */ + +#endif diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 4874c09f6d3c..879b76fa881a 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2009 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> @@ -1854,7 +1841,7 @@ again: old_bytenr = btrfs_node_blockptr(parent, slot); blocksize = fs_info->nodesize; old_ptr_gen = btrfs_node_ptr_generation(parent, slot); - btrfs_node_key_to_cpu(parent, &key, slot); + btrfs_node_key_to_cpu(parent, &first_key, slot); if (level <= max_level) { eb = path->nodes[level]; @@ -4312,7 +4299,7 @@ out: return inode; } -static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info) +static struct reloc_control *alloc_reloc_control(void) { struct reloc_control *rc; @@ -4357,7 +4344,7 @@ static void describe_relocation(struct btrfs_fs_info *fs_info, DESCRIBE_FLAG(RAID5, "raid5"); DESCRIBE_FLAG(RAID6, "raid6"); if (flags) - snprintf(buf, buf - bp + sizeof(buf), "|0x%llx", flags); + snprintf(bp, buf - bp + sizeof(buf), "|0x%llx", flags); #undef DESCRIBE_FLAG } @@ -4379,7 +4366,7 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start) int rw = 0; int err = 0; - rc = alloc_reloc_control(fs_info); + rc = alloc_reloc_control(); if (!rc) return -ENOMEM; @@ -4575,7 +4562,7 @@ int btrfs_recover_relocation(struct btrfs_root *root) if (list_empty(&reloc_roots)) goto out; - rc = alloc_reloc_control(fs_info); + rc = alloc_reloc_control(); if (!rc) { err = -ENOMEM; goto out; diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index aab0194efe46..6db3bda44aa5 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/err.h> diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 1a2066ac6fe7..a59005862010 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2011, 2012 STRATO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/blkdev.h> @@ -3997,6 +3984,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, spin_lock(&fs_info->unused_bgs_lock); if (list_empty(&cache->bg_list)) { btrfs_get_block_group(cache); + trace_btrfs_add_unused_block_group(cache); list_add_tail(&cache->bg_list, &fs_info->unused_bgs); } diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 1f5748c7d1c7..c47f62b19226 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2012 Alexander Block. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/bsearch.h> @@ -248,6 +235,7 @@ struct orphan_dir_info { struct rb_node node; u64 ino; u64 gen; + u64 last_dir_index_offset; }; struct name_cache_entry { @@ -2857,12 +2845,6 @@ add_orphan_dir_info(struct send_ctx *sctx, u64 dir_ino) struct rb_node *parent = NULL; struct orphan_dir_info *entry, *odi; - odi = kmalloc(sizeof(*odi), GFP_KERNEL); - if (!odi) - return ERR_PTR(-ENOMEM); - odi->ino = dir_ino; - odi->gen = 0; - while (*p) { parent = *p; entry = rb_entry(parent, struct orphan_dir_info, node); @@ -2871,11 +2853,17 @@ add_orphan_dir_info(struct send_ctx *sctx, u64 dir_ino) } else if (dir_ino > entry->ino) { p = &(*p)->rb_right; } else { - kfree(odi); return entry; } } + odi = kmalloc(sizeof(*odi), GFP_KERNEL); + if (!odi) + return ERR_PTR(-ENOMEM); + odi->ino = dir_ino; + odi->gen = 0; + odi->last_dir_index_offset = 0; + rb_link_node(&odi->node, parent, p); rb_insert_color(&odi->node, &sctx->orphan_dirs); return odi; @@ -2930,6 +2918,7 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen, struct btrfs_key found_key; struct btrfs_key loc; struct btrfs_dir_item *di; + struct orphan_dir_info *odi = NULL; /* * Don't try to rmdir the top/root subvolume dir. @@ -2944,6 +2933,11 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen, key.objectid = dir; key.type = BTRFS_DIR_INDEX_KEY; key.offset = 0; + + odi = get_orphan_dir_info(sctx, dir); + if (odi) + key.offset = odi->last_dir_index_offset; + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) goto out; @@ -2971,30 +2965,33 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen, dm = get_waiting_dir_move(sctx, loc.objectid); if (dm) { - struct orphan_dir_info *odi; - odi = add_orphan_dir_info(sctx, dir); if (IS_ERR(odi)) { ret = PTR_ERR(odi); goto out; } odi->gen = dir_gen; + odi->last_dir_index_offset = found_key.offset; dm->rmdir_ino = dir; ret = 0; goto out; } if (loc.objectid > send_progress) { - struct orphan_dir_info *odi; - - odi = get_orphan_dir_info(sctx, dir); - free_orphan_dir_info(sctx, odi); + odi = add_orphan_dir_info(sctx, dir); + if (IS_ERR(odi)) { + ret = PTR_ERR(odi); + goto out; + } + odi->gen = dir_gen; + odi->last_dir_index_offset = found_key.offset; ret = 0; goto out; } path->slots[0]++; } + free_orphan_dir_info(sctx, odi); ret = 1; @@ -3272,13 +3269,16 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) if (rmdir_ino) { struct orphan_dir_info *odi; + u64 gen; odi = get_orphan_dir_info(sctx, rmdir_ino); if (!odi) { /* already deleted */ goto finish; } - ret = can_rmdir(sctx, rmdir_ino, odi->gen, sctx->cur_ino); + gen = odi->gen; + + ret = can_rmdir(sctx, rmdir_ino, gen, sctx->cur_ino); if (ret < 0) goto out; if (!ret) @@ -3289,13 +3289,12 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) ret = -ENOMEM; goto out; } - ret = get_cur_path(sctx, rmdir_ino, odi->gen, name); + ret = get_cur_path(sctx, rmdir_ino, gen, name); if (ret < 0) goto out; ret = send_rmdir(sctx, name); if (ret < 0) goto out; - free_orphan_dir_info(sctx, odi); } finish: @@ -5249,6 +5248,10 @@ static int send_write_or_clone(struct send_ctx *sctx, len = btrfs_file_extent_num_bytes(path->nodes[0], ei); } + if (offset >= sctx->cur_inode_size) { + ret = 0; + goto out; + } if (offset + len > sctx->cur_inode_size) len = sctx->cur_inode_size - offset; if (len == 0) { @@ -6463,7 +6466,7 @@ static void btrfs_root_dec_send_in_progress(struct btrfs_root* root) */ if (root->send_in_progress < 0) btrfs_err(root->fs_info, - "send_in_progres unbalanced %d root %llu", + "send_in_progress unbalanced %d root %llu", root->send_in_progress, root->root_key.objectid); spin_unlock(&root->root_item_lock); } diff --git a/fs/btrfs/send.h b/fs/btrfs/send.h index 3aa4bc55754f..ead397f7034f 100644 --- a/fs/btrfs/send.h +++ b/fs/btrfs/send.h @@ -1,22 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2012 Alexander Block. All rights reserved. * Copyright (C) 2012 STRATO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ +#ifndef BTRFS_SEND_H +#define BTRFS_SEND_H + #include "ctree.h" #define BTRFS_SEND_STREAM_MAGIC "btrfs-stream" @@ -132,3 +122,5 @@ enum { #ifdef __KERNEL__ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg); #endif + +#endif diff --git a/fs/btrfs/struct-funcs.c b/fs/btrfs/struct-funcs.c index 5e2b92d83617..b7b4acb12833 100644 --- a/fs/btrfs/struct-funcs.c +++ b/fs/btrfs/struct-funcs.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/highmem.h> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 170baef49fae..81107ad49f3a 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/blkdev.h> @@ -336,6 +323,7 @@ enum { Opt_ssd, Opt_nossd, Opt_ssd_spread, Opt_nossd_spread, Opt_subvol, + Opt_subvol_empty, Opt_subvolid, Opt_thread_pool, Opt_treelog, Opt_notreelog, @@ -401,6 +389,7 @@ static const match_table_t tokens = { {Opt_ssd_spread, "ssd_spread"}, {Opt_nossd_spread, "nossd_spread"}, {Opt_subvol, "subvol=%s"}, + {Opt_subvol_empty, "subvol="}, {Opt_subvolid, "subvolid=%s"}, {Opt_thread_pool, "thread_pool=%u"}, {Opt_treelog, "treelog"}, @@ -474,6 +463,7 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, btrfs_set_opt(info->mount_opt, DEGRADED); break; case Opt_subvol: + case Opt_subvol_empty: case Opt_subvolid: case Opt_subvolrootid: case Opt_device: @@ -1795,10 +1785,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data) } ret = btrfs_parse_options(fs_info, data, *flags); - if (ret) { - ret = -EINVAL; + if (ret) goto restore; - } btrfs_remount_begin(fs_info, old_opts, *flags); btrfs_resize_thread_pool(fs_info, diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index ca067471cd46..4a4e960c7c66 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> @@ -223,12 +210,42 @@ static struct attribute *btrfs_supported_feature_attrs[] = { NULL }; +/* + * Features which depend on feature bits and may differ between each fs. + * + * /sys/fs/btrfs/features lists all available features of this kernel while + * /sys/fs/btrfs/UUID/features shows features of the fs which are enabled or + * can be changed online. + */ static const struct attribute_group btrfs_feature_attr_group = { .name = "features", .is_visible = btrfs_feature_visible, .attrs = btrfs_supported_feature_attrs, }; +static ssize_t rmdir_subvol_show(struct kobject *kobj, + struct kobj_attribute *ka, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "0\n"); +} +BTRFS_ATTR(static_feature, rmdir_subvol, rmdir_subvol_show); + +static struct attribute *btrfs_supported_static_feature_attrs[] = { + BTRFS_ATTR_PTR(static_feature, rmdir_subvol), + NULL +}; + +/* + * Features which only depend on kernel version. + * + * These are listed in /sys/fs/btrfs/features along with + * btrfs_feature_attr_group + */ +static const struct attribute_group btrfs_static_feature_attr_group = { + .name = "features", + .attrs = btrfs_supported_static_feature_attrs, +}; + static ssize_t btrfs_show_u64(u64 *value_ptr, spinlock_t *lock, char *buf) { u64 val; @@ -527,10 +544,11 @@ static inline struct btrfs_fs_info *to_fs_info(struct kobject *kobj) } #define NUM_FEATURE_BITS 64 -static char btrfs_unknown_feature_names[3][NUM_FEATURE_BITS][13]; -static struct btrfs_feature_attr btrfs_feature_attrs[3][NUM_FEATURE_BITS]; +#define BTRFS_FEATURE_NAME_MAX 13 +static char btrfs_unknown_feature_names[FEAT_MAX][NUM_FEATURE_BITS][BTRFS_FEATURE_NAME_MAX]; +static struct btrfs_feature_attr btrfs_feature_attrs[FEAT_MAX][NUM_FEATURE_BITS]; -static const u64 supported_feature_masks[3] = { +static const u64 supported_feature_masks[FEAT_MAX] = { [FEAT_COMPAT] = BTRFS_FEATURE_COMPAT_SUPP, [FEAT_COMPAT_RO] = BTRFS_FEATURE_COMPAT_RO_SUPP, [FEAT_INCOMPAT] = BTRFS_FEATURE_INCOMPAT_SUPP, @@ -602,7 +620,7 @@ void btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs) return; } - list_for_each_entry(fs_devs, fs_uuids, list) { + list_for_each_entry(fs_devs, fs_uuids, fs_list) { __btrfs_sysfs_remove_fsid(fs_devs); } } @@ -622,7 +640,7 @@ void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info) btrfs_sysfs_rm_device_link(fs_info->fs_devices, NULL); } -const char * const btrfs_feature_set_names[3] = { +const char * const btrfs_feature_set_names[FEAT_MAX] = { [FEAT_COMPAT] = "compat", [FEAT_COMPAT_RO] = "compat_ro", [FEAT_INCOMPAT] = "incompat", @@ -686,7 +704,7 @@ static void init_feature_attrs(void) if (fa->kobj_attr.attr.name) continue; - snprintf(name, 13, "%s:%u", + snprintf(name, BTRFS_FEATURE_NAME_MAX, "%s:%u", btrfs_feature_set_names[set], i); fa->kobj_attr.attr.name = name; @@ -913,8 +931,15 @@ int __init btrfs_init_sysfs(void) ret = sysfs_create_group(&btrfs_kset->kobj, &btrfs_feature_attr_group); if (ret) goto out2; + ret = sysfs_merge_group(&btrfs_kset->kobj, + &btrfs_static_feature_attr_group); + if (ret) + goto out_remove_group; return 0; + +out_remove_group: + sysfs_remove_group(&btrfs_kset->kobj, &btrfs_feature_attr_group); out2: debugfs_remove_recursive(btrfs_debugfs_root_dentry); out1: @@ -925,6 +950,8 @@ out1: void __cold btrfs_exit_sysfs(void) { + sysfs_unmerge_group(&btrfs_kset->kobj, + &btrfs_static_feature_attr_group); sysfs_remove_group(&btrfs_kset->kobj, &btrfs_feature_attr_group); kset_unregister(btrfs_kset); debugfs_remove_recursive(btrfs_debugfs_root_dentry); diff --git a/fs/btrfs/sysfs.h b/fs/btrfs/sysfs.h index 80457f31c29f..c6ee600aff89 100644 --- a/fs/btrfs/sysfs.h +++ b/fs/btrfs/sysfs.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _BTRFS_SYSFS_H_ -#define _BTRFS_SYSFS_H_ + +#ifndef BTRFS_SYSFS_H +#define BTRFS_SYSFS_H /* * Data exported through sysfs @@ -8,7 +9,7 @@ extern u64 btrfs_debugfs_test; enum btrfs_feature_set { - FEAT_COMPAT, + FEAT_COMPAT = 0, FEAT_COMPAT_RO, FEAT_INCOMPAT, FEAT_MAX @@ -76,7 +77,7 @@ attr_to_btrfs_feature_attr(struct attribute *attr) } char *btrfs_printable_features(enum btrfs_feature_set set, u64 flags); -extern const char * const btrfs_feature_set_names[3]; +extern const char * const btrfs_feature_set_names[FEAT_MAX]; extern struct kobj_type space_info_ktype; extern struct kobj_type btrfs_raid_ktype; int btrfs_sysfs_add_device_link(struct btrfs_fs_devices *fs_devices, @@ -90,4 +91,4 @@ void btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs); void btrfs_sysfs_feature_update(struct btrfs_fs_info *fs_info, u64 bit, enum btrfs_feature_set set); -#endif /* _BTRFS_SYSFS_H_ */ +#endif diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c index e74278170806..db72b3b6209e 100644 --- a/fs/btrfs/tests/btrfs-tests.c +++ b/fs/btrfs/tests/btrfs-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2013 Fusion IO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/fs.h> @@ -232,11 +219,13 @@ void btrfs_free_dummy_block_group(struct btrfs_block_group_cache *cache) kfree(cache); } -void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans) +void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans, + struct btrfs_fs_info *fs_info) { memset(trans, 0, sizeof(*trans)); trans->transid = 1; trans->type = __TRANS_DUMMY; + trans->fs_info = fs_info; } int btrfs_run_sanity_tests(void) diff --git a/fs/btrfs/tests/btrfs-tests.h b/fs/btrfs/tests/btrfs-tests.h index bc0615bac3cc..70ff9f9d86a1 100644 --- a/fs/btrfs/tests/btrfs-tests.h +++ b/fs/btrfs/tests/btrfs-tests.h @@ -1,28 +1,16 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2013 Fusion IO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_TESTS -#define __BTRFS_TESTS +#ifndef BTRFS_TESTS_H +#define BTRFS_TESTS_H #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS int btrfs_run_sanity_tests(void); -#define test_msg(fmt, ...) pr_info("BTRFS: selftest: " fmt, ##__VA_ARGS__) +#define test_msg(fmt, ...) pr_info("BTRFS: selftest: " fmt "\n", ##__VA_ARGS__) +#define test_err(fmt, ...) pr_err("BTRFS: selftest: " fmt "\n", ##__VA_ARGS__) struct btrfs_root; struct btrfs_trans_handle; @@ -41,7 +29,8 @@ void btrfs_free_dummy_root(struct btrfs_root *root); struct btrfs_block_group_cache * btrfs_alloc_dummy_block_group(struct btrfs_fs_info *fs_info, unsigned long length); void btrfs_free_dummy_block_group(struct btrfs_block_group_cache *cache); -void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans); +void btrfs_init_dummy_trans(struct btrfs_trans_handle *trans, + struct btrfs_fs_info *fs_info); #else static inline int btrfs_run_sanity_tests(void) { diff --git a/fs/btrfs/tests/extent-buffer-tests.c b/fs/btrfs/tests/extent-buffer-tests.c index b9142c614114..7d72eab6d32c 100644 --- a/fs/btrfs/tests/extent-buffer-tests.c +++ b/fs/btrfs/tests/extent-buffer-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2013 Fusion IO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/slab.h> @@ -39,31 +26,31 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) u32 value_len = strlen(value); int ret = 0; - test_msg("Running btrfs_split_item tests\n"); + test_msg("running btrfs_split_item tests"); fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); if (!fs_info) { - test_msg("Could not allocate fs_info\n"); + test_err("could not allocate fs_info"); return -ENOMEM; } root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(root)) { - test_msg("Could not allocate root\n"); + test_err("could not allocate root"); ret = PTR_ERR(root); goto out; } path = btrfs_alloc_path(); if (!path) { - test_msg("Could not allocate path\n"); + test_err("could not allocate path"); ret = -ENOMEM; goto out; } path->nodes[0] = eb = alloc_dummy_extent_buffer(fs_info, nodesize); if (!eb) { - test_msg("Could not allocate dummy buffer\n"); + test_err("could not allocate dummy buffer"); ret = -ENOMEM; goto out; } @@ -88,7 +75,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) */ ret = btrfs_split_item(NULL, root, path, &key, 17); if (ret) { - test_msg("Split item failed %d\n", ret); + test_err("split item failed %d", ret); goto out; } @@ -99,14 +86,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) btrfs_item_key_to_cpu(eb, &key, 0); if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY || key.offset != 0) { - test_msg("Invalid key at slot 0\n"); + test_err("invalid key at slot 0"); ret = -EINVAL; goto out; } item = btrfs_item_nr(0); if (btrfs_item_size(eb, item) != strlen(split1)) { - test_msg("Invalid len in the first split\n"); + test_err("invalid len in the first split"); ret = -EINVAL; goto out; } @@ -114,8 +101,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 0), strlen(split1)); if (memcmp(buf, split1, strlen(split1))) { - test_msg("Data in the buffer doesn't match what it should " - "in the first split have='%.*s' want '%s'\n", + test_err( +"data in the buffer doesn't match what it should in the first split have='%.*s' want '%s'", (int)strlen(split1), buf, split1); ret = -EINVAL; goto out; @@ -124,14 +111,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) btrfs_item_key_to_cpu(eb, &key, 1); if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY || key.offset != 3) { - test_msg("Invalid key at slot 1\n"); + test_err("invalid key at slot 1"); ret = -EINVAL; goto out; } item = btrfs_item_nr(1); if (btrfs_item_size(eb, item) != strlen(split2)) { - test_msg("Invalid len in the second split\n"); + test_err("invalid len in the second split"); ret = -EINVAL; goto out; } @@ -139,8 +126,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 1), strlen(split2)); if (memcmp(buf, split2, strlen(split2))) { - test_msg("Data in the buffer doesn't match what it should " - "in the second split\n"); + test_err( + "data in the buffer doesn't match what it should in the second split"); ret = -EINVAL; goto out; } @@ -149,21 +136,21 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) /* Do it again so we test memmoving the other items in the leaf */ ret = btrfs_split_item(NULL, root, path, &key, 4); if (ret) { - test_msg("Second split item failed %d\n", ret); + test_err("second split item failed %d", ret); goto out; } btrfs_item_key_to_cpu(eb, &key, 0); if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY || key.offset != 0) { - test_msg("Invalid key at slot 0\n"); + test_err("invalid key at slot 0"); ret = -EINVAL; goto out; } item = btrfs_item_nr(0); if (btrfs_item_size(eb, item) != strlen(split3)) { - test_msg("Invalid len in the first split\n"); + test_err("invalid len in the first split"); ret = -EINVAL; goto out; } @@ -171,8 +158,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 0), strlen(split3)); if (memcmp(buf, split3, strlen(split3))) { - test_msg("Data in the buffer doesn't match what it should " - "in the third split"); + test_err( + "data in the buffer doesn't match what it should in the third split"); ret = -EINVAL; goto out; } @@ -180,14 +167,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) btrfs_item_key_to_cpu(eb, &key, 1); if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY || key.offset != 1) { - test_msg("Invalid key at slot 1\n"); + test_err("invalid key at slot 1"); ret = -EINVAL; goto out; } item = btrfs_item_nr(1); if (btrfs_item_size(eb, item) != strlen(split4)) { - test_msg("Invalid len in the second split\n"); + test_err("invalid len in the second split"); ret = -EINVAL; goto out; } @@ -195,8 +182,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 1), strlen(split4)); if (memcmp(buf, split4, strlen(split4))) { - test_msg("Data in the buffer doesn't match what it should " - "in the fourth split\n"); + test_err( + "data in the buffer doesn't match what it should in the fourth split"); ret = -EINVAL; goto out; } @@ -204,14 +191,14 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) btrfs_item_key_to_cpu(eb, &key, 2); if (key.objectid != 0 || key.type != BTRFS_EXTENT_CSUM_KEY || key.offset != 3) { - test_msg("Invalid key at slot 2\n"); + test_err("invalid key at slot 2"); ret = -EINVAL; goto out; } item = btrfs_item_nr(2); if (btrfs_item_size(eb, item) != strlen(split2)) { - test_msg("Invalid len in the second split\n"); + test_err("invalid len in the second split"); ret = -EINVAL; goto out; } @@ -219,8 +206,8 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize) read_extent_buffer(eb, buf, btrfs_item_ptr_offset(eb, 2), strlen(split2)); if (memcmp(buf, split2, strlen(split2))) { - test_msg("Data in the buffer doesn't match what it should " - "in the last chunk\n"); + test_err( + "data in the buffer doesn't match what it should in the last chunk"); ret = -EINVAL; goto out; } @@ -233,6 +220,6 @@ out: int btrfs_test_extent_buffer_operations(u32 sectorsize, u32 nodesize) { - test_msg("Running extent buffer operation tests\n"); + test_msg("running extent buffer operation tests"); return test_btrfs_split_item(sectorsize, nodesize); } diff --git a/fs/btrfs/tests/extent-io-tests.c b/fs/btrfs/tests/extent-io-tests.c index 2e7f64a3b22b..d9269a531a4d 100644 --- a/fs/btrfs/tests/extent-io-tests.c +++ b/fs/btrfs/tests/extent-io-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2013 Fusion IO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/pagemap.h> @@ -59,7 +46,9 @@ static noinline int process_page_range(struct inode *inode, u64 start, u64 end, cond_resched(); loops++; if (loops > 100000) { - printk(KERN_ERR "stuck in a loop, start %Lu, end %Lu, nr_pages %lu, ret %d\n", start, end, nr_pages, ret); + printk(KERN_ERR + "stuck in a loop, start %llu, end %llu, nr_pages %lu, ret %d\n", + start, end, nr_pages, ret); break; } } @@ -79,11 +68,11 @@ static int test_find_delalloc(u32 sectorsize) u64 found; int ret = -EINVAL; - test_msg("Running find delalloc tests\n"); + test_msg("running find delalloc tests"); inode = btrfs_new_test_inode(); if (!inode) { - test_msg("Failed to allocate test inode\n"); + test_err("failed to allocate test inode"); return -ENOMEM; } @@ -97,7 +86,7 @@ static int test_find_delalloc(u32 sectorsize) for (index = 0; index < (total_dirty >> PAGE_SHIFT); index++) { page = find_or_create_page(inode->i_mapping, index, GFP_KERNEL); if (!page) { - test_msg("Failed to allocate test page\n"); + test_err("failed to allocate test page"); ret = -ENOMEM; goto out; } @@ -120,11 +109,11 @@ static int test_find_delalloc(u32 sectorsize) found = find_lock_delalloc_range(inode, &tmp, locked_page, &start, &end, max_bytes); if (!found) { - test_msg("Should have found at least one delalloc\n"); + test_err("should have found at least one delalloc"); goto out_bits; } if (start != 0 || end != (sectorsize - 1)) { - test_msg("Expected start 0 end %u, got start %llu end %llu\n", + test_err("expected start 0 end %u, got start %llu end %llu", sectorsize - 1, start, end); goto out_bits; } @@ -142,7 +131,7 @@ static int test_find_delalloc(u32 sectorsize) locked_page = find_lock_page(inode->i_mapping, test_start >> PAGE_SHIFT); if (!locked_page) { - test_msg("Couldn't find the locked page\n"); + test_err("couldn't find the locked page"); goto out_bits; } set_extent_delalloc(&tmp, sectorsize, max_bytes - 1, 0, NULL); @@ -151,17 +140,17 @@ static int test_find_delalloc(u32 sectorsize) found = find_lock_delalloc_range(inode, &tmp, locked_page, &start, &end, max_bytes); if (!found) { - test_msg("Couldn't find delalloc in our range\n"); + test_err("couldn't find delalloc in our range"); goto out_bits; } if (start != test_start || end != max_bytes - 1) { - test_msg("Expected start %Lu end %Lu, got start %Lu, end " - "%Lu\n", test_start, max_bytes - 1, start, end); + test_err("expected start %llu end %llu, got start %llu, end %llu", + test_start, max_bytes - 1, start, end); goto out_bits; } if (process_page_range(inode, start, end, PROCESS_TEST_LOCKED | PROCESS_UNLOCK)) { - test_msg("There were unlocked pages in the range\n"); + test_err("there were unlocked pages in the range"); goto out_bits; } unlock_extent(&tmp, start, end); @@ -177,7 +166,7 @@ static int test_find_delalloc(u32 sectorsize) locked_page = find_lock_page(inode->i_mapping, test_start >> PAGE_SHIFT); if (!locked_page) { - test_msg("Couldn't find the locked page\n"); + test_err("couldn't find the locked page"); goto out_bits; } start = test_start; @@ -185,11 +174,11 @@ static int test_find_delalloc(u32 sectorsize) found = find_lock_delalloc_range(inode, &tmp, locked_page, &start, &end, max_bytes); if (found) { - test_msg("Found range when we shouldn't have\n"); + test_err("found range when we shouldn't have"); goto out_bits; } if (end != (u64)-1) { - test_msg("Did not return the proper end offset\n"); + test_err("did not return the proper end offset"); goto out_bits; } @@ -206,17 +195,17 @@ static int test_find_delalloc(u32 sectorsize) found = find_lock_delalloc_range(inode, &tmp, locked_page, &start, &end, max_bytes); if (!found) { - test_msg("Didn't find our range\n"); + test_err("didn't find our range"); goto out_bits; } if (start != test_start || end != total_dirty - 1) { - test_msg("Expected start %Lu end %Lu, got start %Lu end %Lu\n", + test_err("expected start %llu end %llu, got start %llu end %llu", test_start, total_dirty - 1, start, end); goto out_bits; } if (process_page_range(inode, start, end, PROCESS_TEST_LOCKED | PROCESS_UNLOCK)) { - test_msg("Pages in range were not all locked\n"); + test_err("pages in range were not all locked"); goto out_bits; } unlock_extent(&tmp, start, end); @@ -228,7 +217,7 @@ static int test_find_delalloc(u32 sectorsize) page = find_get_page(inode->i_mapping, (max_bytes + SZ_1M) >> PAGE_SHIFT); if (!page) { - test_msg("Couldn't find our page\n"); + test_err("couldn't find our page"); goto out_bits; } ClearPageDirty(page); @@ -247,18 +236,17 @@ static int test_find_delalloc(u32 sectorsize) found = find_lock_delalloc_range(inode, &tmp, locked_page, &start, &end, max_bytes); if (!found) { - test_msg("Didn't find our range\n"); + test_err("didn't find our range"); goto out_bits; } if (start != test_start && end != test_start + PAGE_SIZE - 1) { - test_msg("Expected start %Lu end %Lu, got start %Lu end %Lu\n", - test_start, test_start + PAGE_SIZE - 1, start, - end); + test_err("expected start %llu end %llu, got start %llu end %llu", + test_start, test_start + PAGE_SIZE - 1, start, end); goto out_bits; } if (process_page_range(inode, start, end, PROCESS_TEST_LOCKED | PROCESS_UNLOCK)) { - test_msg("Pages in range were not all locked\n"); + test_err("pages in range were not all locked"); goto out_bits; } ret = 0; @@ -284,14 +272,14 @@ static int check_eb_bitmap(unsigned long *bitmap, struct extent_buffer *eb, bit = !!test_bit(i, bitmap); bit1 = !!extent_buffer_test_bit(eb, 0, i); if (bit1 != bit) { - test_msg("Bits do not match\n"); + test_err("bits do not match"); return -EINVAL; } bit1 = !!extent_buffer_test_bit(eb, i / BITS_PER_BYTE, i % BITS_PER_BYTE); if (bit1 != bit) { - test_msg("Offset bits do not match\n"); + test_err("offset bits do not match"); return -EINVAL; } } @@ -308,7 +296,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb, memset(bitmap, 0, len); memzero_extent_buffer(eb, 0, len); if (memcmp_extent_buffer(eb, bitmap, 0, len) != 0) { - test_msg("Bitmap was not zeroed\n"); + test_err("bitmap was not zeroed"); return -EINVAL; } @@ -316,7 +304,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb, extent_buffer_bitmap_set(eb, 0, 0, len * BITS_PER_BYTE); ret = check_eb_bitmap(bitmap, eb, len); if (ret) { - test_msg("Setting all bits failed\n"); + test_err("setting all bits failed"); return ret; } @@ -324,7 +312,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb, extent_buffer_bitmap_clear(eb, 0, 0, len * BITS_PER_BYTE); ret = check_eb_bitmap(bitmap, eb, len); if (ret) { - test_msg("Clearing all bits failed\n"); + test_err("clearing all bits failed"); return ret; } @@ -337,7 +325,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb, sizeof(long) * BITS_PER_BYTE); ret = check_eb_bitmap(bitmap, eb, len); if (ret) { - test_msg("Setting straddling pages failed\n"); + test_err("setting straddling pages failed"); return ret; } @@ -350,7 +338,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb, sizeof(long) * BITS_PER_BYTE); ret = check_eb_bitmap(bitmap, eb, len); if (ret) { - test_msg("Clearing straddling pages failed\n"); + test_err("clearing straddling pages failed"); return ret; } } @@ -374,7 +362,7 @@ static int __test_eb_bitmaps(unsigned long *bitmap, struct extent_buffer *eb, ret = check_eb_bitmap(bitmap, eb, len); if (ret) { - test_msg("Random bit pattern failed\n"); + test_err("random bit pattern failed"); return ret; } @@ -389,7 +377,7 @@ static int test_eb_bitmaps(u32 sectorsize, u32 nodesize) struct extent_buffer *eb; int ret; - test_msg("Running extent buffer bitmap tests\n"); + test_msg("running extent buffer bitmap tests"); /* * In ppc64, sectorsize can be 64K, thus 4 * 64K will be larger than @@ -402,13 +390,13 @@ static int test_eb_bitmaps(u32 sectorsize, u32 nodesize) bitmap = kmalloc(len, GFP_KERNEL); if (!bitmap) { - test_msg("Couldn't allocate test bitmap\n"); + test_err("couldn't allocate test bitmap"); return -ENOMEM; } eb = __alloc_dummy_extent_buffer(fs_info, 0, len); if (!eb) { - test_msg("Couldn't allocate test extent buffer\n"); + test_err("couldn't allocate test extent buffer"); kfree(bitmap); return -ENOMEM; } @@ -421,7 +409,7 @@ static int test_eb_bitmaps(u32 sectorsize, u32 nodesize) free_extent_buffer(eb); eb = __alloc_dummy_extent_buffer(NULL, nodesize / 2, len); if (!eb) { - test_msg("Couldn't allocate test extent buffer\n"); + test_err("couldn't allocate test extent buffer"); kfree(bitmap); return -ENOMEM; } @@ -437,7 +425,7 @@ int btrfs_test_extent_io(u32 sectorsize, u32 nodesize) { int ret; - test_msg("Running extent I/O tests\n"); + test_msg("running extent I/O tests"); ret = test_find_delalloc(sectorsize); if (ret) @@ -445,6 +433,6 @@ int btrfs_test_extent_io(u32 sectorsize, u32 nodesize) ret = test_eb_bitmaps(sectorsize, nodesize); out: - test_msg("Extent I/O tests finished\n"); + test_msg("extent I/O tests finished"); return ret; } diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c index c23bd00bdd92..385a5316e4bf 100644 --- a/fs/btrfs/tests/extent-map-tests.c +++ b/fs/btrfs/tests/extent-map-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2017 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/types.h> @@ -32,8 +19,8 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree) #ifdef CONFIG_BTRFS_DEBUG if (refcount_read(&em->refs) != 1) { - test_msg( -"em leak: em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx) refs %d\n", + test_err( +"em leak: em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx) refs %d", em->start, em->len, em->block_start, em->block_len, refcount_read(&em->refs)); @@ -60,7 +47,8 @@ static void free_extent_map_tree(struct extent_map_tree *em_tree) * ->add_extent_mapping(0, 16K) * -> #handle -EEXIST */ -static void test_case_1(struct extent_map_tree *em_tree) +static void test_case_1(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree) { struct extent_map *em; u64 start = 0; @@ -103,14 +91,14 @@ static void test_case_1(struct extent_map_tree *em_tree) em->len = len; em->block_start = start; em->block_len = len; - ret = btrfs_add_extent_mapping(em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); if (ret) - test_msg("case1 [%llu %llu]: ret %d\n", start, start + len, ret); + test_err("case1 [%llu %llu]: ret %d", start, start + len, ret); if (em && (em->start != 0 || extent_map_end(em) != SZ_16K || em->block_start != 0 || em->block_len != SZ_16K)) - test_msg( -"case1 [%llu %llu]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu\n", + test_err( +"case1 [%llu %llu]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu", start, start + len, ret, em->start, em->len, em->block_start, em->block_len); free_extent_map(em); @@ -125,7 +113,8 @@ out: * Reading the inline ending up with EEXIST, ie. read an inline * extent and discard page cache and read it again. */ -static void test_case_2(struct extent_map_tree *em_tree) +static void test_case_2(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree) { struct extent_map *em; int ret; @@ -166,14 +155,14 @@ static void test_case_2(struct extent_map_tree *em_tree) em->len = SZ_1K; em->block_start = EXTENT_MAP_INLINE; em->block_len = (u64)-1; - ret = btrfs_add_extent_mapping(em_tree, &em, em->start, em->len); + ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, em->start, em->len); if (ret) - test_msg("case2 [0 1K]: ret %d\n", ret); + test_err("case2 [0 1K]: ret %d", ret); if (em && (em->start != 0 || extent_map_end(em) != SZ_1K || em->block_start != EXTENT_MAP_INLINE || em->block_len != (u64)-1)) - test_msg( -"case2 [0 1K]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu\n", + test_err( +"case2 [0 1K]: ret %d return a wrong em (start %llu len %llu block_start %llu block_len %llu", ret, em->start, em->len, em->block_start, em->block_len); free_extent_map(em); @@ -182,7 +171,8 @@ out: free_extent_map_tree(em_tree); } -static void __test_case_3(struct extent_map_tree *em_tree, u64 start) +static void __test_case_3(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree, u64 start) { struct extent_map *em; u64 len = SZ_4K; @@ -211,9 +201,9 @@ static void __test_case_3(struct extent_map_tree *em_tree, u64 start) em->len = SZ_16K; em->block_start = 0; em->block_len = SZ_16K; - ret = btrfs_add_extent_mapping(em_tree, &em, start, len); + ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); if (ret) - test_msg("case3 [0x%llx 0x%llx): ret %d\n", + test_err("case3 [0x%llx 0x%llx): ret %d", start, start + len, ret); /* * Since bytes within em are contiguous, em->block_start is identical to @@ -222,8 +212,8 @@ static void __test_case_3(struct extent_map_tree *em_tree, u64 start) if (em && (start < em->start || start + len > extent_map_end(em) || em->start != em->block_start || em->len != em->block_len)) - test_msg( -"case3 [0x%llx 0x%llx): ret %d em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)\n", + test_err( +"case3 [0x%llx 0x%llx): ret %d em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)", start, start + len, ret, em->start, em->len, em->block_start, em->block_len); free_extent_map(em); @@ -248,14 +238,16 @@ out: * -> add_extent_mapping() * -> add_extent_mapping() */ -static void test_case_3(struct extent_map_tree *em_tree) +static void test_case_3(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree) { - __test_case_3(em_tree, 0); - __test_case_3(em_tree, SZ_8K); - __test_case_3(em_tree, (12 * 1024ULL)); + __test_case_3(fs_info, em_tree, 0); + __test_case_3(fs_info, em_tree, SZ_8K); + __test_case_3(fs_info, em_tree, (12 * 1024ULL)); } -static void __test_case_4(struct extent_map_tree *em_tree, u64 start) +static void __test_case_4(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree, u64 start) { struct extent_map *em; u64 len = SZ_4K; @@ -296,14 +288,14 @@ static void __test_case_4(struct extent_map_tree *em_tree, u64 start) em->len = SZ_32K; em->block_start = 0; em->block_len = SZ_32K; - ret = btrfs_add_extent_mapping(em_tree, &em, start, len); + ret = btrfs_add_extent_mapping(fs_info, em_tree, &em, start, len); if (ret) - test_msg("case4 [0x%llx 0x%llx): ret %d\n", + test_err("case4 [0x%llx 0x%llx): ret %d", start, len, ret); if (em && (start < em->start || start + len > extent_map_end(em))) - test_msg( -"case4 [0x%llx 0x%llx): ret %d, added wrong em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)\n", + test_err( +"case4 [0x%llx 0x%llx): ret %d, added wrong em (start 0x%llx len 0x%llx block_start 0x%llx block_len 0x%llx)", start, len, ret, em->start, em->len, em->block_start, em->block_len); free_extent_map(em); @@ -337,30 +329,45 @@ out: * # handle -EEXIST when adding * # [0, 32K) */ -static void test_case_4(struct extent_map_tree *em_tree) +static void test_case_4(struct btrfs_fs_info *fs_info, + struct extent_map_tree *em_tree) { - __test_case_4(em_tree, 0); - __test_case_4(em_tree, SZ_4K); + __test_case_4(fs_info, em_tree, 0); + __test_case_4(fs_info, em_tree, SZ_4K); } int btrfs_test_extent_map(void) { + struct btrfs_fs_info *fs_info = NULL; struct extent_map_tree *em_tree; - test_msg("Running extent_map tests\n"); + test_msg("running extent_map tests"); + + /* + * Note: the fs_info is not set up completely, we only need + * fs_info::fsid for the tracepoint. + */ + fs_info = btrfs_alloc_dummy_fs_info(PAGE_SIZE, PAGE_SIZE); + if (!fs_info) { + test_msg("Couldn't allocate dummy fs info"); + return -ENOMEM; + } em_tree = kzalloc(sizeof(*em_tree), GFP_KERNEL); if (!em_tree) /* Skip the test on error. */ - return 0; + goto out; extent_map_tree_init(em_tree); - test_case_1(em_tree); - test_case_2(em_tree); - test_case_3(em_tree); - test_case_4(em_tree); + test_case_1(fs_info, em_tree); + test_case_2(fs_info, em_tree); + test_case_3(fs_info, em_tree); + test_case_4(fs_info, em_tree); kfree(em_tree); +out: + btrfs_free_dummy_fs_info(fs_info); + return 0; } diff --git a/fs/btrfs/tests/free-space-tests.c b/fs/btrfs/tests/free-space-tests.c index eca6412d42bd..5c2f77e9439b 100644 --- a/fs/btrfs/tests/free-space-tests.c +++ b/fs/btrfs/tests/free-space-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2013 Fusion IO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/slab.h> @@ -33,63 +20,63 @@ static int test_extents(struct btrfs_block_group_cache *cache) { int ret = 0; - test_msg("Running extent only tests\n"); + test_msg("running extent only tests"); /* First just make sure we can remove an entire entry */ ret = btrfs_add_free_space(cache, 0, SZ_4M); if (ret) { - test_msg("Error adding initial extents %d\n", ret); + test_err("error adding initial extents %d", ret); return ret; } ret = btrfs_remove_free_space(cache, 0, SZ_4M); if (ret) { - test_msg("Error removing extent %d\n", ret); + test_err("error removing extent %d", ret); return ret; } if (test_check_exists(cache, 0, SZ_4M)) { - test_msg("Full remove left some lingering space\n"); + test_err("full remove left some lingering space"); return -1; } /* Ok edge and middle cases now */ ret = btrfs_add_free_space(cache, 0, SZ_4M); if (ret) { - test_msg("Error adding half extent %d\n", ret); + test_err("error adding half extent %d", ret); return ret; } ret = btrfs_remove_free_space(cache, 3 * SZ_1M, SZ_1M); if (ret) { - test_msg("Error removing tail end %d\n", ret); + test_err("error removing tail end %d", ret); return ret; } ret = btrfs_remove_free_space(cache, 0, SZ_1M); if (ret) { - test_msg("Error removing front end %d\n", ret); + test_err("error removing front end %d", ret); return ret; } ret = btrfs_remove_free_space(cache, SZ_2M, 4096); if (ret) { - test_msg("Error removing middle piece %d\n", ret); + test_err("error removing middle piece %d", ret); return ret; } if (test_check_exists(cache, 0, SZ_1M)) { - test_msg("Still have space at the front\n"); + test_err("still have space at the front"); return -1; } if (test_check_exists(cache, SZ_2M, 4096)) { - test_msg("Still have space in the middle\n"); + test_err("still have space in the middle"); return -1; } if (test_check_exists(cache, 3 * SZ_1M, SZ_1M)) { - test_msg("Still have space at the end\n"); + test_err("still have space at the end"); return -1; } @@ -105,34 +92,34 @@ static int test_bitmaps(struct btrfs_block_group_cache *cache, u64 next_bitmap_offset; int ret; - test_msg("Running bitmap only tests\n"); + test_msg("running bitmap only tests"); ret = test_add_free_space_entry(cache, 0, SZ_4M, 1); if (ret) { - test_msg("Couldn't create a bitmap entry %d\n", ret); + test_err("couldn't create a bitmap entry %d", ret); return ret; } ret = btrfs_remove_free_space(cache, 0, SZ_4M); if (ret) { - test_msg("Error removing bitmap full range %d\n", ret); + test_err("error removing bitmap full range %d", ret); return ret; } if (test_check_exists(cache, 0, SZ_4M)) { - test_msg("Left some space in bitmap\n"); + test_err("left some space in bitmap"); return -1; } ret = test_add_free_space_entry(cache, 0, SZ_4M, 1); if (ret) { - test_msg("Couldn't add to our bitmap entry %d\n", ret); + test_err("couldn't add to our bitmap entry %d", ret); return ret; } ret = btrfs_remove_free_space(cache, SZ_1M, SZ_2M); if (ret) { - test_msg("Couldn't remove middle chunk %d\n", ret); + test_err("couldn't remove middle chunk %d", ret); return ret; } @@ -146,19 +133,19 @@ static int test_bitmaps(struct btrfs_block_group_cache *cache, ret = test_add_free_space_entry(cache, next_bitmap_offset - SZ_2M, SZ_4M, 1); if (ret) { - test_msg("Couldn't add space that straddles two bitmaps %d\n", + test_err("couldn't add space that straddles two bitmaps %d", ret); return ret; } ret = btrfs_remove_free_space(cache, next_bitmap_offset - SZ_1M, SZ_2M); if (ret) { - test_msg("Couldn't remove overlapping space %d\n", ret); + test_err("couldn't remove overlapping space %d", ret); return ret; } if (test_check_exists(cache, next_bitmap_offset - SZ_1M, SZ_2M)) { - test_msg("Left some space when removing overlapping\n"); + test_err("left some space when removing overlapping"); return -1; } @@ -174,7 +161,7 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache, u64 bitmap_offset = (u64)(BITS_PER_BITMAP * sectorsize); int ret; - test_msg("Running bitmap and extent tests\n"); + test_msg("running bitmap and extent tests"); /* * First let's do something simple, an extent at the same offset as the @@ -183,42 +170,42 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache, */ ret = test_add_free_space_entry(cache, SZ_4M, SZ_1M, 1); if (ret) { - test_msg("Couldn't create bitmap entry %d\n", ret); + test_err("couldn't create bitmap entry %d", ret); return ret; } ret = test_add_free_space_entry(cache, 0, SZ_1M, 0); if (ret) { - test_msg("Couldn't add extent entry %d\n", ret); + test_err("couldn't add extent entry %d", ret); return ret; } ret = btrfs_remove_free_space(cache, 0, SZ_1M); if (ret) { - test_msg("Couldn't remove extent entry %d\n", ret); + test_err("couldn't remove extent entry %d", ret); return ret; } if (test_check_exists(cache, 0, SZ_1M)) { - test_msg("Left remnants after our remove\n"); + test_err("left remnants after our remove"); return -1; } /* Now to add back the extent entry and remove from the bitmap */ ret = test_add_free_space_entry(cache, 0, SZ_1M, 0); if (ret) { - test_msg("Couldn't re-add extent entry %d\n", ret); + test_err("couldn't re-add extent entry %d", ret); return ret; } ret = btrfs_remove_free_space(cache, SZ_4M, SZ_1M); if (ret) { - test_msg("Couldn't remove from bitmap %d\n", ret); + test_err("couldn't remove from bitmap %d", ret); return ret; } if (test_check_exists(cache, SZ_4M, SZ_1M)) { - test_msg("Left remnants in the bitmap\n"); + test_err("left remnants in the bitmap"); return -1; } @@ -228,18 +215,18 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache, */ ret = test_add_free_space_entry(cache, SZ_1M, SZ_4M, 1); if (ret) { - test_msg("Couldn't add to a bitmap %d\n", ret); + test_err("couldn't add to a bitmap %d", ret); return ret; } ret = btrfs_remove_free_space(cache, SZ_512K, 3 * SZ_1M); if (ret) { - test_msg("Couldn't remove overlapping space %d\n", ret); + test_err("couldn't remove overlapping space %d", ret); return ret; } if (test_check_exists(cache, SZ_512K, 3 * SZ_1M)) { - test_msg("Left over pieces after removing overlapping\n"); + test_err("left over pieces after removing overlapping"); return -1; } @@ -248,24 +235,24 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache, /* Now with the extent entry offset into the bitmap */ ret = test_add_free_space_entry(cache, SZ_4M, SZ_4M, 1); if (ret) { - test_msg("Couldn't add space to the bitmap %d\n", ret); + test_err("couldn't add space to the bitmap %d", ret); return ret; } ret = test_add_free_space_entry(cache, SZ_2M, SZ_2M, 0); if (ret) { - test_msg("Couldn't add extent to the cache %d\n", ret); + test_err("couldn't add extent to the cache %d", ret); return ret; } ret = btrfs_remove_free_space(cache, 3 * SZ_1M, SZ_4M); if (ret) { - test_msg("Problem removing overlapping space %d\n", ret); + test_err("problem removing overlapping space %d", ret); return ret; } if (test_check_exists(cache, 3 * SZ_1M, SZ_4M)) { - test_msg("Left something behind when removing space"); + test_err("left something behind when removing space"); return -1; } @@ -282,25 +269,25 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache, __btrfs_remove_free_space_cache(cache->free_space_ctl); ret = test_add_free_space_entry(cache, bitmap_offset + SZ_4M, SZ_4M, 1); if (ret) { - test_msg("Couldn't add bitmap %d\n", ret); + test_err("couldn't add bitmap %d", ret); return ret; } ret = test_add_free_space_entry(cache, bitmap_offset - SZ_1M, 5 * SZ_1M, 0); if (ret) { - test_msg("Couldn't add extent entry %d\n", ret); + test_err("couldn't add extent entry %d", ret); return ret; } ret = btrfs_remove_free_space(cache, bitmap_offset + SZ_1M, 5 * SZ_1M); if (ret) { - test_msg("Failed to free our space %d\n", ret); + test_err("failed to free our space %d", ret); return ret; } if (test_check_exists(cache, bitmap_offset + SZ_1M, 5 * SZ_1M)) { - test_msg("Left stuff over\n"); + test_err("left stuff over"); return -1; } @@ -314,19 +301,19 @@ static int test_bitmaps_and_extents(struct btrfs_block_group_cache *cache, */ ret = test_add_free_space_entry(cache, SZ_1M, SZ_2M, 1); if (ret) { - test_msg("Couldn't add bitmap entry %d\n", ret); + test_err("couldn't add bitmap entry %d", ret); return ret; } ret = test_add_free_space_entry(cache, 3 * SZ_1M, SZ_1M, 0); if (ret) { - test_msg("Couldn't add extent entry %d\n", ret); + test_err("couldn't add extent entry %d", ret); return ret; } ret = btrfs_remove_free_space(cache, SZ_1M, 3 * SZ_1M); if (ret) { - test_msg("Error removing bitmap and extent overlapping %d\n", ret); + test_err("error removing bitmap and extent overlapping %d", ret); return ret; } @@ -348,12 +335,14 @@ check_num_extents_and_bitmaps(const struct btrfs_block_group_cache *cache, const int num_bitmaps) { if (cache->free_space_ctl->free_extents != num_extents) { - test_msg("Incorrect # of extent entries in the cache: %d, expected %d\n", + test_err( + "incorrect # of extent entries in the cache: %d, expected %d", cache->free_space_ctl->free_extents, num_extents); return -EINVAL; } if (cache->free_space_ctl->total_bitmaps != num_bitmaps) { - test_msg("Incorrect # of extent entries in the cache: %d, expected %d\n", + test_err( + "incorrect # of extent entries in the cache: %d, expected %d", cache->free_space_ctl->total_bitmaps, num_bitmaps); return -EINVAL; } @@ -371,7 +360,7 @@ static int check_cache_empty(struct btrfs_block_group_cache *cache) * allocate. */ if (cache->free_space_ctl->free_space != 0) { - test_msg("Cache free space is not 0\n"); + test_err("cache free space is not 0"); return -EINVAL; } @@ -379,7 +368,7 @@ static int check_cache_empty(struct btrfs_block_group_cache *cache) offset = btrfs_find_space_for_alloc(cache, 0, 4096, 0, &max_extent_size); if (offset != 0) { - test_msg("Space allocation did not fail, returned offset: %llu", + test_err("space allocation did not fail, returned offset: %llu", offset); return -EINVAL; } @@ -415,7 +404,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, }; const struct btrfs_free_space_op *orig_free_space_ops; - test_msg("Running space stealing from bitmap to extent\n"); + test_msg("running space stealing from bitmap to extent"); /* * For this test, we want to ensure we end up with an extent entry @@ -443,7 +432,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = test_add_free_space_entry(cache, SZ_128M - SZ_256K, SZ_128K, 0); if (ret) { - test_msg("Couldn't add extent entry %d\n", ret); + test_err("couldn't add extent entry %d", ret); return ret; } @@ -451,7 +440,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, ret = test_add_free_space_entry(cache, SZ_128M + SZ_512K, SZ_128M - SZ_512K, 1); if (ret) { - test_msg("Couldn't add bitmap entry %d\n", ret); + test_err("couldn't add bitmap entry %d", ret); return ret; } @@ -470,17 +459,17 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, SZ_128M + 768 * SZ_1K, SZ_128M - 768 * SZ_1K); if (ret) { - test_msg("Failed to free part of bitmap space %d\n", ret); + test_err("failed to free part of bitmap space %d", ret); return ret; } /* Confirm that only those 2 ranges are marked as free. */ if (!test_check_exists(cache, SZ_128M - SZ_256K, SZ_128K)) { - test_msg("Free space range missing\n"); + test_err("free space range missing"); return -ENOENT; } if (!test_check_exists(cache, SZ_128M + SZ_512K, SZ_256K)) { - test_msg("Free space range missing\n"); + test_err("free space range missing"); return -ENOENT; } @@ -490,7 +479,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ if (test_check_exists(cache, SZ_128M + 768 * SZ_1K, SZ_128M - 768 * SZ_1K)) { - test_msg("Bitmap region not removed from space cache\n"); + test_err("bitmap region not removed from space cache"); return -EINVAL; } @@ -499,7 +488,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, * covered by the bitmap, isn't marked as free. */ if (test_check_exists(cache, SZ_128M + SZ_256K, SZ_256K)) { - test_msg("Invalid bitmap region marked as free\n"); + test_err("invalid bitmap region marked as free"); return -EINVAL; } @@ -508,7 +497,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, * by the bitmap too, isn't marked as free either. */ if (test_check_exists(cache, SZ_128M, SZ_256K)) { - test_msg("Invalid bitmap region marked as free\n"); + test_err("invalid bitmap region marked as free"); return -EINVAL; } @@ -519,12 +508,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = btrfs_add_free_space(cache, SZ_128M, SZ_512K); if (ret) { - test_msg("Error adding free space: %d\n", ret); + test_err("error adding free space: %d", ret); return ret; } /* Confirm the region is marked as free. */ if (!test_check_exists(cache, SZ_128M, SZ_512K)) { - test_msg("Bitmap region not marked as free\n"); + test_err("bitmap region not marked as free"); return -ENOENT; } @@ -544,7 +533,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = btrfs_add_free_space(cache, SZ_128M + SZ_16M, sectorsize); if (ret) { - test_msg("Error adding free space: %d\n", ret); + test_err("error adding free space: %d", ret); return ret; } @@ -563,12 +552,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = btrfs_add_free_space(cache, SZ_128M - SZ_128K, SZ_128K); if (ret) { - test_msg("Error adding free space: %d\n", ret); + test_err("error adding free space: %d", ret); return ret; } /* Confirm the region is marked as free. */ if (!test_check_exists(cache, SZ_128M - SZ_128K, SZ_128K)) { - test_msg("Extent region not marked as free\n"); + test_err("extent region not marked as free"); return -ENOENT; } @@ -596,12 +585,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, * allocate the whole free space at once. */ if (!test_check_exists(cache, SZ_128M - SZ_256K, SZ_1M)) { - test_msg("Expected region not marked as free\n"); + test_err("expected region not marked as free"); return -ENOENT; } if (cache->free_space_ctl->free_space != (SZ_1M + sectorsize)) { - test_msg("Cache free space is not 1Mb + %u\n", sectorsize); + test_err("cache free space is not 1Mb + %u", sectorsize); return -EINVAL; } @@ -609,7 +598,8 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, 0, SZ_1M, 0, &max_extent_size); if (offset != (SZ_128M - SZ_256K)) { - test_msg("Failed to allocate 1Mb from space cache, returned offset is: %llu\n", + test_err( + "failed to allocate 1Mb from space cache, returned offset is: %llu", offset); return -EINVAL; } @@ -623,7 +613,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, return ret; if (cache->free_space_ctl->free_space != sectorsize) { - test_msg("Cache free space is not %u\n", sectorsize); + test_err("cache free space is not %u", sectorsize); return -EINVAL; } @@ -631,7 +621,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, 0, sectorsize, 0, &max_extent_size); if (offset != (SZ_128M + SZ_16M)) { - test_msg("Failed to allocate %u, returned offset : %llu\n", + test_err("failed to allocate %u, returned offset : %llu", sectorsize, offset); return -EINVAL; } @@ -653,14 +643,14 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = test_add_free_space_entry(cache, SZ_128M + SZ_128K, SZ_128K, 0); if (ret) { - test_msg("Couldn't add extent entry %d\n", ret); + test_err("couldn't add extent entry %d", ret); return ret; } /* Bitmap entry covering free space range [0, 128Mb - 512Kb[ */ ret = test_add_free_space_entry(cache, 0, SZ_128M - SZ_512K, 1); if (ret) { - test_msg("Couldn't add bitmap entry %d\n", ret); + test_err("couldn't add bitmap entry %d", ret); return ret; } @@ -677,17 +667,17 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = btrfs_remove_free_space(cache, 0, SZ_128M - 768 * SZ_1K); if (ret) { - test_msg("Failed to free part of bitmap space %d\n", ret); + test_err("failed to free part of bitmap space %d", ret); return ret; } /* Confirm that only those 2 ranges are marked as free. */ if (!test_check_exists(cache, SZ_128M + SZ_128K, SZ_128K)) { - test_msg("Free space range missing\n"); + test_err("free space range missing"); return -ENOENT; } if (!test_check_exists(cache, SZ_128M - 768 * SZ_1K, SZ_256K)) { - test_msg("Free space range missing\n"); + test_err("free space range missing"); return -ENOENT; } @@ -696,7 +686,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, * as free anymore. */ if (test_check_exists(cache, 0, SZ_128M - 768 * SZ_1K)) { - test_msg("Bitmap region not removed from space cache\n"); + test_err("bitmap region not removed from space cache"); return -EINVAL; } @@ -705,7 +695,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, * covered by the bitmap, isn't marked as free. */ if (test_check_exists(cache, SZ_128M - SZ_512K, SZ_512K)) { - test_msg("Invalid bitmap region marked as free\n"); + test_err("invalid bitmap region marked as free"); return -EINVAL; } @@ -716,12 +706,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = btrfs_add_free_space(cache, SZ_128M - SZ_512K, SZ_512K); if (ret) { - test_msg("Error adding free space: %d\n", ret); + test_err("error adding free space: %d", ret); return ret; } /* Confirm the region is marked as free. */ if (!test_check_exists(cache, SZ_128M - SZ_512K, SZ_512K)) { - test_msg("Bitmap region not marked as free\n"); + test_err("bitmap region not marked as free"); return -ENOENT; } @@ -741,7 +731,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = btrfs_add_free_space(cache, SZ_32M, 2 * sectorsize); if (ret) { - test_msg("Error adding free space: %d\n", ret); + test_err("error adding free space: %d", ret); return ret; } @@ -752,12 +742,12 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, */ ret = btrfs_add_free_space(cache, SZ_128M, SZ_128K); if (ret) { - test_msg("Error adding free space: %d\n", ret); + test_err("error adding free space: %d", ret); return ret; } /* Confirm the region is marked as free. */ if (!test_check_exists(cache, SZ_128M, SZ_128K)) { - test_msg("Extent region not marked as free\n"); + test_err("extent region not marked as free"); return -ENOENT; } @@ -785,19 +775,20 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, * allocate the whole free space at once. */ if (!test_check_exists(cache, SZ_128M - 768 * SZ_1K, SZ_1M)) { - test_msg("Expected region not marked as free\n"); + test_err("expected region not marked as free"); return -ENOENT; } if (cache->free_space_ctl->free_space != (SZ_1M + 2 * sectorsize)) { - test_msg("Cache free space is not 1Mb + %u\n", 2 * sectorsize); + test_err("cache free space is not 1Mb + %u", 2 * sectorsize); return -EINVAL; } offset = btrfs_find_space_for_alloc(cache, 0, SZ_1M, 0, &max_extent_size); if (offset != (SZ_128M - 768 * SZ_1K)) { - test_msg("Failed to allocate 1Mb from space cache, returned offset is: %llu\n", + test_err( + "failed to allocate 1Mb from space cache, returned offset is: %llu", offset); return -EINVAL; } @@ -811,7 +802,7 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, return ret; if (cache->free_space_ctl->free_space != 2 * sectorsize) { - test_msg("Cache free space is not %u\n", 2 * sectorsize); + test_err("cache free space is not %u", 2 * sectorsize); return -EINVAL; } @@ -819,9 +810,8 @@ test_steal_space_from_bitmap_to_extent(struct btrfs_block_group_cache *cache, 0, 2 * sectorsize, 0, &max_extent_size); if (offset != SZ_32M) { - test_msg("Failed to allocate %u, offset: %llu\n", - 2 * sectorsize, - offset); + test_err("failed to allocate %u, offset: %llu", + 2 * sectorsize, offset); return -EINVAL; } @@ -842,7 +832,7 @@ int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize) struct btrfs_root *root = NULL; int ret = -ENOMEM; - test_msg("Running btrfs free space cache tests\n"); + test_msg("running btrfs free space cache tests"); fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); if (!fs_info) return -ENOMEM; @@ -856,7 +846,7 @@ int btrfs_test_free_space_cache(u32 sectorsize, u32 nodesize) cache = btrfs_alloc_dummy_block_group(fs_info, BITS_PER_BITMAP * sectorsize + PAGE_SIZE); if (!cache) { - test_msg("Couldn't run the tests\n"); + test_err("couldn't run the tests"); btrfs_free_dummy_fs_info(fs_info); return 0; } @@ -884,6 +874,6 @@ out: btrfs_free_dummy_block_group(cache); btrfs_free_dummy_root(root); btrfs_free_dummy_fs_info(fs_info); - test_msg("Free space cache tests finished\n"); + test_msg("free space cache tests finished"); return ret; } diff --git a/fs/btrfs/tests/free-space-tree-tests.c b/fs/btrfs/tests/free-space-tree-tests.c index 8444a018cca2..89346da890cf 100644 --- a/fs/btrfs/tests/free-space-tree-tests.c +++ b/fs/btrfs/tests/free-space-tree-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2015 Facebook. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/types.h> @@ -45,7 +32,7 @@ static int __check_free_space_extents(struct btrfs_trans_handle *trans, info = search_free_space_info(trans, fs_info, cache, path, 0); if (IS_ERR(info)) { - test_msg("Could not find free space info\n"); + test_err("could not find free space info"); ret = PTR_ERR(info); goto out; } @@ -53,7 +40,7 @@ static int __check_free_space_extents(struct btrfs_trans_handle *trans, extent_count = btrfs_free_space_extent_count(path->nodes[0], info); if (extent_count != num_extents) { - test_msg("Extent count is wrong\n"); + test_err("extent count is wrong"); ret = -EINVAL; goto out; } @@ -112,7 +99,7 @@ out: btrfs_release_path(path); return ret; invalid: - test_msg("Free space tree is invalid\n"); + test_err("free space tree is invalid"); ret = -EINVAL; goto out; } @@ -130,7 +117,7 @@ static int check_free_space_extents(struct btrfs_trans_handle *trans, info = search_free_space_info(trans, fs_info, cache, path, 0); if (IS_ERR(info)) { - test_msg("Could not find free space info\n"); + test_err("could not find free space info"); btrfs_release_path(path); return PTR_ERR(info); } @@ -144,15 +131,15 @@ static int check_free_space_extents(struct btrfs_trans_handle *trans, /* Flip it to the other format and check that for good measure. */ if (flags & BTRFS_FREE_SPACE_USING_BITMAPS) { - ret = convert_free_space_to_extents(trans, fs_info, cache, path); + ret = convert_free_space_to_extents(trans, cache, path); if (ret) { - test_msg("Could not convert to extents\n"); + test_err("could not convert to extents"); return ret; } } else { - ret = convert_free_space_to_bitmaps(trans, fs_info, cache, path); + ret = convert_free_space_to_bitmaps(trans, cache, path); if (ret) { - test_msg("Could not convert to bitmaps\n"); + test_err("could not convert to bitmaps"); return ret; } } @@ -183,11 +170,11 @@ static int test_remove_all(struct btrfs_trans_handle *trans, const struct free_space_extent extents[] = {}; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid, cache->key.offset); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } @@ -207,10 +194,10 @@ static int test_remove_beginning(struct btrfs_trans_handle *trans, }; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid, alignment); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } @@ -230,12 +217,12 @@ static int test_remove_end(struct btrfs_trans_handle *trans, }; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid + cache->key.offset - alignment, alignment); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } @@ -256,11 +243,11 @@ static int test_remove_middle(struct btrfs_trans_handle *trans, }; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid + alignment, alignment); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } @@ -279,26 +266,26 @@ static int test_merge_left(struct btrfs_trans_handle *trans, }; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid, cache->key.offset); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, - cache->key.objectid, alignment); + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid, + alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid + alignment, alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } @@ -317,27 +304,27 @@ static int test_merge_right(struct btrfs_trans_handle *trans, }; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid, cache->key.offset); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid + 2 * alignment, alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid + alignment, alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } @@ -356,34 +343,34 @@ static int test_merge_both(struct btrfs_trans_handle *trans, }; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid, cache->key.offset); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, - cache->key.objectid, alignment); + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid, + alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid + 2 * alignment, alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid + alignment, alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } @@ -404,34 +391,34 @@ static int test_merge_none(struct btrfs_trans_handle *trans, }; int ret; - ret = __remove_from_free_space_tree(trans, fs_info, cache, path, + ret = __remove_from_free_space_tree(trans, cache, path, cache->key.objectid, cache->key.offset); if (ret) { - test_msg("Could not remove free space\n"); + test_err("could not remove free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, - cache->key.objectid, alignment); + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid, + alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid + 4 * alignment, alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } - ret = __add_to_free_space_tree(trans, fs_info, cache, path, + ret = __add_to_free_space_tree(trans, cache, path, cache->key.objectid + 2 * alignment, alignment); if (ret) { - test_msg("Could not add free space\n"); + test_err("could not add free space"); return ret; } @@ -457,14 +444,14 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize, fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); if (!fs_info) { - test_msg("Couldn't allocate dummy fs info\n"); + test_err("couldn't allocate dummy fs info"); ret = -ENOMEM; goto out; } root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(root)) { - test_msg("Couldn't allocate dummy root\n"); + test_err("couldn't allocate dummy root"); ret = PTR_ERR(root); goto out; } @@ -476,7 +463,7 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize, root->node = alloc_test_extent_buffer(root->fs_info, nodesize); if (!root->node) { - test_msg("Couldn't allocate dummy buffer\n"); + test_err("couldn't allocate dummy buffer"); ret = -ENOMEM; goto out; } @@ -486,7 +473,7 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize, cache = btrfs_alloc_dummy_block_group(fs_info, 8 * alignment); if (!cache) { - test_msg("Couldn't allocate dummy block group cache\n"); + test_err("couldn't allocate dummy block group cache"); ret = -ENOMEM; goto out; } @@ -495,26 +482,25 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize, cache->needs_free_space = 1; cache->fs_info = root->fs_info; - btrfs_init_dummy_trans(&trans); + btrfs_init_dummy_trans(&trans, root->fs_info); path = btrfs_alloc_path(); if (!path) { - test_msg("Couldn't allocate path\n"); + test_err("couldn't allocate path"); ret = -ENOMEM; goto out; } - ret = add_block_group_free_space(&trans, root->fs_info, cache); + ret = add_block_group_free_space(&trans, cache); if (ret) { - test_msg("Could not add block group free space\n"); + test_err("could not add block group free space"); goto out; } if (bitmaps) { - ret = convert_free_space_to_bitmaps(&trans, root->fs_info, - cache, path); + ret = convert_free_space_to_bitmaps(&trans, cache, path); if (ret) { - test_msg("Could not convert block group to bitmaps\n"); + test_err("could not convert block group to bitmaps"); goto out; } } @@ -523,14 +509,14 @@ static int run_test(test_func_t test_func, int bitmaps, u32 sectorsize, if (ret) goto out; - ret = remove_block_group_free_space(&trans, root->fs_info, cache); + ret = remove_block_group_free_space(&trans, cache); if (ret) { - test_msg("Could not remove block group free space\n"); + test_err("could not remove block group free space"); goto out; } if (btrfs_header_nritems(root->node) != 0) { - test_msg("Free space tree has leftover items\n"); + test_err("free space tree has leftover items"); ret = -EINVAL; goto out; } @@ -552,14 +538,16 @@ static int run_test_both_formats(test_func_t test_func, u32 sectorsize, ret = run_test(test_func, 0, sectorsize, nodesize, alignment); if (ret) { - test_msg("%pf failed with extents, sectorsize=%u, nodesize=%u, alignment=%u\n", + test_err( + "%pf failed with extents, sectorsize=%u, nodesize=%u, alignment=%u", test_func, sectorsize, nodesize, alignment); test_ret = ret; } ret = run_test(test_func, 1, sectorsize, nodesize, alignment); if (ret) { - test_msg("%pf failed with bitmaps, sectorsize=%u, nodesize=%u, alignment=%u\n", + test_err( + "%pf failed with bitmaps, sectorsize=%u, nodesize=%u, alignment=%u", test_func, sectorsize, nodesize, alignment); test_ret = ret; } @@ -590,7 +578,7 @@ int btrfs_test_free_space_tree(u32 sectorsize, u32 nodesize) */ bitmap_alignment = BTRFS_FREE_SPACE_BITMAP_BITS * PAGE_SIZE; - test_msg("Running free space tree tests\n"); + test_msg("running free space tree tests"); for (i = 0; i < ARRAY_SIZE(tests); i++) { int ret; diff --git a/fs/btrfs/tests/inode-tests.c b/fs/btrfs/tests/inode-tests.c index 13420cd19ef0..64043f028820 100644 --- a/fs/btrfs/tests/inode-tests.c +++ b/fs/btrfs/tests/inode-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2013 Fusion IO. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/types.h> @@ -241,7 +228,7 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) inode = btrfs_new_test_inode(); if (!inode) { - test_msg("Couldn't allocate inode\n"); + test_err("couldn't allocate inode"); return ret; } @@ -251,19 +238,19 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); if (!fs_info) { - test_msg("Couldn't allocate dummy fs info\n"); + test_err("couldn't allocate dummy fs info"); goto out; } root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(root)) { - test_msg("Couldn't allocate root\n"); + test_err("couldn't allocate root"); goto out; } root->node = alloc_dummy_extent_buffer(fs_info, nodesize); if (!root->node) { - test_msg("Couldn't allocate dummy buffer\n"); + test_err("couldn't allocate dummy buffer"); goto out; } @@ -281,11 +268,11 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, sectorsize, 0); if (IS_ERR(em)) { em = NULL; - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != EXTENT_MAP_HOLE) { - test_msg("Expected a hole, got %llu\n", em->block_start); + test_err("expected a hole, got %llu", em->block_start); goto out; } free_extent_map(em); @@ -300,20 +287,21 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, (u64)-1, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != EXTENT_MAP_HOLE) { - test_msg("Expected a hole, got %llu\n", em->block_start); + test_err("expected a hole, got %llu", em->block_start); goto out; } if (em->start != 0 || em->len != 5) { - test_msg("Unexpected extent wanted start 0 len 5, got start " - "%llu len %llu\n", em->start, em->len); + test_err( + "unexpected extent wanted start 0 len 5, got start %llu len %llu", + em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } offset = em->start + em->len; @@ -321,21 +309,22 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != EXTENT_MAP_INLINE) { - test_msg("Expected an inline, got %llu\n", em->block_start); + test_err("expected an inline, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != (sectorsize - 5)) { - test_msg("Unexpected extent wanted start %llu len 1, got start " - "%llu len %llu\n", offset, em->start, em->len); + test_err( + "unexpected extent wanted start %llu len 1, got start %llu len %llu", + offset, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } /* @@ -348,20 +337,21 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != EXTENT_MAP_HOLE) { - test_msg("Expected a hole, got %llu\n", em->block_start); + test_err("expected a hole, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != 4) { - test_msg("Unexpected extent wanted start %llu len 4, got start " - "%llu len %llu\n", offset, em->start, em->len); + test_err( + "unexpected extent wanted start %llu len 4, got start %llu len %llu", + offset, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } offset = em->start + em->len; @@ -370,24 +360,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) /* Regular extent */ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize - 1) { - test_msg("Unexpected extent wanted start %llu len 4095, got " - "start %llu len %llu\n", offset, em->start, em->len); + test_err( + "unexpected extent wanted start %llu len 4095, got start %llu len %llu", + offset, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -397,25 +388,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) /* The next 3 are split extents */ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -426,21 +417,21 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != EXTENT_MAP_HOLE) { - test_msg("Expected a hole, got %llu\n", em->block_start); + test_err("expected a hole, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } offset = em->start + em->len; @@ -448,31 +439,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != 2 * sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, 2 * sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } if (em->orig_start != orig_start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", + test_err("wrong orig offset, want %llu, have %llu", orig_start, em->orig_start); goto out; } disk_bytenr += (em->start - orig_start); if (em->block_start != disk_bytenr) { - test_msg("Wrong block start, want %llu, have %llu\n", + test_err("wrong block start, want %llu, have %llu", disk_bytenr, em->block_start); goto out; } @@ -482,26 +473,26 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) /* Prealloc extent */ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != prealloc_only) { - test_msg("Unexpected flags set, want %lu have %lu\n", + test_err("unexpected flags set, want %lu have %lu", prealloc_only, em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -511,26 +502,26 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) /* The next 3 are a half written prealloc extent */ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != prealloc_only) { - test_msg("Unexpected flags set, want %lu have %lu\n", + test_err("unexpected flags set, want %lu have %lu", prealloc_only, em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -541,30 +532,30 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_HOLE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } if (em->orig_start != orig_start) { - test_msg("Unexpected orig offset, wanted %llu, have %llu\n", + test_err("unexpected orig offset, wanted %llu, have %llu", orig_start, em->orig_start); goto out; } if (em->block_start != (disk_bytenr + (em->start - em->orig_start))) { - test_msg("Unexpected block start, wanted %llu, have %llu\n", + test_err("unexpected block start, wanted %llu, have %llu", disk_bytenr + (em->start - em->orig_start), em->block_start); goto out; @@ -574,31 +565,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != 2 * sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, 2 * sectorsize, em->start, em->len); goto out; } if (em->flags != prealloc_only) { - test_msg("Unexpected flags set, want %lu have %lu\n", + test_err("unexpected flags set, want %lu have %lu", prealloc_only, em->flags); goto out; } if (em->orig_start != orig_start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", orig_start, + test_err("wrong orig offset, want %llu, have %llu", orig_start, em->orig_start); goto out; } if (em->block_start != (disk_bytenr + (em->start - em->orig_start))) { - test_msg("Unexpected block start, wanted %llu, have %llu\n", + test_err("unexpected block start, wanted %llu, have %llu", disk_bytenr + (em->start - em->orig_start), em->block_start); goto out; @@ -609,31 +600,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) /* Now for the compressed extent */ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != 2 * sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u," - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, 2 * sectorsize, em->start, em->len); goto out; } if (em->flags != compressed_only) { - test_msg("Unexpected flags set, want %lu have %lu\n", + test_err("unexpected flags set, want %lu have %lu", compressed_only, em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } if (em->compress_type != BTRFS_COMPRESS_ZLIB) { - test_msg("Unexpected compress type, wanted %d, got %d\n", + test_err("unexpected compress type, wanted %d, got %d", BTRFS_COMPRESS_ZLIB, em->compress_type); goto out; } @@ -643,31 +634,31 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) /* Split compressed extent */ em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u," - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != compressed_only) { - test_msg("Unexpected flags set, want %lu have %lu\n", + test_err("unexpected flags set, want %lu have %lu", compressed_only, em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } if (em->compress_type != BTRFS_COMPRESS_ZLIB) { - test_msg("Unexpected compress type, wanted %d, got %d\n", + test_err("unexpected compress type, wanted %d, got %d", BTRFS_COMPRESS_ZLIB, em->compress_type); goto out; } @@ -678,25 +669,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -705,32 +696,32 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != disk_bytenr) { - test_msg("Block start does not match, want %llu got %llu\n", + test_err("block start does not match, want %llu got %llu", disk_bytenr, em->block_start); goto out; } if (em->start != offset || em->len != 2 * sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, 2 * sectorsize, em->start, em->len); goto out; } if (em->flags != compressed_only) { - test_msg("Unexpected flags set, want %lu have %lu\n", + test_err("unexpected flags set, want %lu have %lu", compressed_only, em->flags); goto out; } if (em->orig_start != orig_start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", + test_err("wrong orig offset, want %llu, have %llu", em->start, orig_start); goto out; } if (em->compress_type != BTRFS_COMPRESS_ZLIB) { - test_msg("Unexpected compress type, wanted %d, got %d\n", + test_err("unexpected compress type, wanted %d, got %d", BTRFS_COMPRESS_ZLIB, em->compress_type); goto out; } @@ -741,25 +732,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset + 6, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -768,11 +759,11 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, SZ_4M, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != EXTENT_MAP_HOLE) { - test_msg("Expected a hole extent, got %llu\n", em->block_start); + test_err("expected a hole extent, got %llu", em->block_start); goto out; } /* @@ -781,18 +772,18 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) * test. */ if (em->start != offset || em->len != 3 * sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, 3 * sectorsize, em->start, em->len); goto out; } if (em->flags != vacancy_only) { - test_msg("Unexpected flags set, want %lu have %lu\n", + test_err("unexpected flags set, want %lu have %lu", vacancy_only, em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -801,25 +792,25 @@ static noinline int test_btrfs_get_extent(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, offset, sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start >= EXTENT_MAP_LAST_BYTE) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != offset || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %llu len %u," - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %llu len %u, got start %llu len %llu", offset, sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, want 0 have %lu\n", em->flags); + test_err("unexpected flags set, want 0 have %lu", em->flags); goto out; } if (em->orig_start != em->start) { - test_msg("Wrong orig offset, want %llu, have %llu\n", em->start, + test_err("wrong orig offset, want %llu, have %llu", em->start, em->orig_start); goto out; } @@ -843,7 +834,7 @@ static int test_hole_first(u32 sectorsize, u32 nodesize) inode = btrfs_new_test_inode(); if (!inode) { - test_msg("Couldn't allocate inode\n"); + test_err("couldn't allocate inode"); return ret; } @@ -853,19 +844,19 @@ static int test_hole_first(u32 sectorsize, u32 nodesize) fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); if (!fs_info) { - test_msg("Couldn't allocate dummy fs info\n"); + test_err("couldn't allocate dummy fs info"); goto out; } root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(root)) { - test_msg("Couldn't allocate root\n"); + test_err("couldn't allocate root"); goto out; } root->node = alloc_dummy_extent_buffer(fs_info, nodesize); if (!root->node) { - test_msg("Couldn't allocate dummy buffer\n"); + test_err("couldn't allocate dummy buffer"); goto out; } @@ -884,21 +875,21 @@ static int test_hole_first(u32 sectorsize, u32 nodesize) sectorsize, BTRFS_FILE_EXTENT_REG, 0, 1); em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, 0, 2 * sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != EXTENT_MAP_HOLE) { - test_msg("Expected a hole, got %llu\n", em->block_start); + test_err("expected a hole, got %llu", em->block_start); goto out; } if (em->start != 0 || em->len != sectorsize) { - test_msg("Unexpected extent wanted start 0 len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start 0 len %u, got start %llu len %llu", sectorsize, em->start, em->len); goto out; } if (em->flags != vacancy_only) { - test_msg("Wrong flags, wanted %lu, have %lu\n", vacancy_only, + test_err("wrong flags, wanted %lu, have %lu", vacancy_only, em->flags); goto out; } @@ -907,21 +898,21 @@ static int test_hole_first(u32 sectorsize, u32 nodesize) em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, sectorsize, 2 * sectorsize, 0); if (IS_ERR(em)) { - test_msg("Got an error when we shouldn't have\n"); + test_err("got an error when we shouldn't have"); goto out; } if (em->block_start != sectorsize) { - test_msg("Expected a real extent, got %llu\n", em->block_start); + test_err("expected a real extent, got %llu", em->block_start); goto out; } if (em->start != sectorsize || em->len != sectorsize) { - test_msg("Unexpected extent wanted start %u len %u, " - "got start %llu len %llu\n", + test_err( + "unexpected extent wanted start %u len %u, got start %llu len %llu", sectorsize, sectorsize, em->start, em->len); goto out; } if (em->flags != 0) { - test_msg("Unexpected flags set, wanted 0 got %lu\n", + test_err("unexpected flags set, wanted 0 got %lu", em->flags); goto out; } @@ -944,19 +935,19 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) inode = btrfs_new_test_inode(); if (!inode) { - test_msg("Couldn't allocate inode\n"); + test_err("couldn't allocate inode"); return ret; } fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); if (!fs_info) { - test_msg("Couldn't allocate dummy fs info\n"); + test_err("couldn't allocate dummy fs info"); goto out; } root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(root)) { - test_msg("Couldn't allocate root\n"); + test_err("couldn't allocate root"); goto out; } @@ -967,12 +958,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) ret = btrfs_set_extent_delalloc(inode, 0, BTRFS_MAX_EXTENT_SIZE - 1, 0, NULL, 0); if (ret) { - test_msg("btrfs_set_extent_delalloc returned %d\n", ret); + test_err("btrfs_set_extent_delalloc returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 1) { ret = -EINVAL; - test_msg("Miscount, wanted 1, got %u\n", + test_err("miscount, wanted 1, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -982,12 +973,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) BTRFS_MAX_EXTENT_SIZE + sectorsize - 1, 0, NULL, 0); if (ret) { - test_msg("btrfs_set_extent_delalloc returned %d\n", ret); + test_err("btrfs_set_extent_delalloc returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 2) { ret = -EINVAL; - test_msg("Miscount, wanted 2, got %u\n", + test_err("miscount, wanted 2, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -999,12 +990,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) EXTENT_DELALLOC | EXTENT_DIRTY | EXTENT_UPTODATE, 0, 0, NULL); if (ret) { - test_msg("clear_extent_bit returned %d\n", ret); + test_err("clear_extent_bit returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 2) { ret = -EINVAL; - test_msg("Miscount, wanted 2, got %u\n", + test_err("miscount, wanted 2, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -1015,12 +1006,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) + sectorsize - 1, 0, NULL, 0); if (ret) { - test_msg("btrfs_set_extent_delalloc returned %d\n", ret); + test_err("btrfs_set_extent_delalloc returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 2) { ret = -EINVAL; - test_msg("Miscount, wanted 2, got %u\n", + test_err("miscount, wanted 2, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -1033,12 +1024,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) (BTRFS_MAX_EXTENT_SIZE << 1) + 3 * sectorsize - 1, 0, NULL, 0); if (ret) { - test_msg("btrfs_set_extent_delalloc returned %d\n", ret); + test_err("btrfs_set_extent_delalloc returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 4) { ret = -EINVAL; - test_msg("Miscount, wanted 4, got %u\n", + test_err("miscount, wanted 4, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -1050,12 +1041,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) BTRFS_MAX_EXTENT_SIZE + sectorsize, BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, 0, NULL, 0); if (ret) { - test_msg("btrfs_set_extent_delalloc returned %d\n", ret); + test_err("btrfs_set_extent_delalloc returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 3) { ret = -EINVAL; - test_msg("Miscount, wanted 3, got %u\n", + test_err("miscount, wanted 3, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -1067,12 +1058,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_UPTODATE, 0, 0, NULL); if (ret) { - test_msg("clear_extent_bit returned %d\n", ret); + test_err("clear_extent_bit returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 4) { ret = -EINVAL; - test_msg("Miscount, wanted 4, got %u\n", + test_err("miscount, wanted 4, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -1085,12 +1076,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) BTRFS_MAX_EXTENT_SIZE + sectorsize, BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, 0, NULL, 0); if (ret) { - test_msg("btrfs_set_extent_delalloc returned %d\n", ret); + test_err("btrfs_set_extent_delalloc returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents != 3) { ret = -EINVAL; - test_msg("Miscount, wanted 3, got %u\n", + test_err("miscount, wanted 3, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -1100,12 +1091,12 @@ static int test_extent_accounting(u32 sectorsize, u32 nodesize) EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_UPTODATE, 0, 0, NULL); if (ret) { - test_msg("clear_extent_bit returned %d\n", ret); + test_err("clear_extent_bit returned %d", ret); goto out; } if (BTRFS_I(inode)->outstanding_extents) { ret = -EINVAL; - test_msg("Miscount, wanted 0, got %u\n", + test_err("miscount, wanted 0, got %u", BTRFS_I(inode)->outstanding_extents); goto out; } @@ -1128,14 +1119,14 @@ int btrfs_test_inodes(u32 sectorsize, u32 nodesize) set_bit(EXTENT_FLAG_COMPRESSED, &compressed_only); set_bit(EXTENT_FLAG_PREALLOC, &prealloc_only); - test_msg("Running btrfs_get_extent tests\n"); + test_msg("running btrfs_get_extent tests"); ret = test_btrfs_get_extent(sectorsize, nodesize); if (ret) return ret; - test_msg("Running hole first btrfs_get_extent test\n"); + test_msg("running hole first btrfs_get_extent test"); ret = test_hole_first(sectorsize, nodesize); if (ret) return ret; - test_msg("Running outstanding_extents tests\n"); + test_msg("running outstanding_extents tests"); return test_extent_accounting(sectorsize, nodesize); } diff --git a/fs/btrfs/tests/qgroup-tests.c b/fs/btrfs/tests/qgroup-tests.c index 160eb2fba726..ace94db09d29 100644 --- a/fs/btrfs/tests/qgroup-tests.c +++ b/fs/btrfs/tests/qgroup-tests.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2013 Facebook. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/types.h> @@ -37,7 +24,7 @@ static int insert_normal_tree_ref(struct btrfs_root *root, u64 bytenr, u32 size = sizeof(*item) + sizeof(*iref) + sizeof(*block_info); int ret; - btrfs_init_dummy_trans(&trans); + btrfs_init_dummy_trans(&trans, NULL); ins.objectid = bytenr; ins.type = BTRFS_EXTENT_ITEM_KEY; @@ -45,14 +32,14 @@ static int insert_normal_tree_ref(struct btrfs_root *root, u64 bytenr, path = btrfs_alloc_path(); if (!path) { - test_msg("Couldn't allocate path\n"); + test_err("couldn't allocate path"); return -ENOMEM; } path->leave_spinning = 1; ret = btrfs_insert_empty_item(&trans, root, path, &ins, size); if (ret) { - test_msg("Couldn't insert ref %d\n", ret); + test_err("couldn't insert ref %d", ret); btrfs_free_path(path); return ret; } @@ -87,7 +74,7 @@ static int add_tree_ref(struct btrfs_root *root, u64 bytenr, u64 num_bytes, u64 refs; int ret; - btrfs_init_dummy_trans(&trans); + btrfs_init_dummy_trans(&trans, NULL); key.objectid = bytenr; key.type = BTRFS_EXTENT_ITEM_KEY; @@ -95,14 +82,14 @@ static int add_tree_ref(struct btrfs_root *root, u64 bytenr, u64 num_bytes, path = btrfs_alloc_path(); if (!path) { - test_msg("Couldn't allocate path\n"); + test_err("couldn't allocate path"); return -ENOMEM; } path->leave_spinning = 1; ret = btrfs_search_slot(&trans, root, &key, path, 0, 1); if (ret) { - test_msg("Couldn't find extent ref\n"); + test_err("couldn't find extent ref"); btrfs_free_path(path); return ret; } @@ -124,7 +111,7 @@ static int add_tree_ref(struct btrfs_root *root, u64 bytenr, u64 num_bytes, ret = btrfs_insert_empty_item(&trans, root, path, &key, 0); if (ret) - test_msg("Failed to insert backref\n"); + test_err("failed to insert backref"); btrfs_free_path(path); return ret; } @@ -137,7 +124,7 @@ static int remove_extent_item(struct btrfs_root *root, u64 bytenr, struct btrfs_path *path; int ret; - btrfs_init_dummy_trans(&trans); + btrfs_init_dummy_trans(&trans, NULL); key.objectid = bytenr; key.type = BTRFS_EXTENT_ITEM_KEY; @@ -145,14 +132,14 @@ static int remove_extent_item(struct btrfs_root *root, u64 bytenr, path = btrfs_alloc_path(); if (!path) { - test_msg("Couldn't allocate path\n"); + test_err("couldn't allocate path"); return -ENOMEM; } path->leave_spinning = 1; ret = btrfs_search_slot(&trans, root, &key, path, -1, 1); if (ret) { - test_msg("Didn't find our key %d\n", ret); + test_err("didn't find our key %d", ret); btrfs_free_path(path); return ret; } @@ -171,7 +158,7 @@ static int remove_extent_ref(struct btrfs_root *root, u64 bytenr, u64 refs; int ret; - btrfs_init_dummy_trans(&trans); + btrfs_init_dummy_trans(&trans, NULL); key.objectid = bytenr; key.type = BTRFS_EXTENT_ITEM_KEY; @@ -179,14 +166,14 @@ static int remove_extent_ref(struct btrfs_root *root, u64 bytenr, path = btrfs_alloc_path(); if (!path) { - test_msg("Couldn't allocate path\n"); + test_err("couldn't allocate path"); return -ENOMEM; } path->leave_spinning = 1; ret = btrfs_search_slot(&trans, root, &key, path, 0, 1); if (ret) { - test_msg("Couldn't find extent ref\n"); + test_err("couldn't find extent ref"); btrfs_free_path(path); return ret; } @@ -208,7 +195,7 @@ static int remove_extent_ref(struct btrfs_root *root, u64 bytenr, ret = btrfs_search_slot(&trans, root, &key, path, -1, 1); if (ret) { - test_msg("Couldn't find backref %d\n", ret); + test_err("couldn't find backref %d", ret); btrfs_free_path(path); return ret; } @@ -226,12 +213,12 @@ static int test_no_shared_qgroup(struct btrfs_root *root, struct ulist *new_roots = NULL; int ret; - btrfs_init_dummy_trans(&trans); + btrfs_init_dummy_trans(&trans, fs_info); - test_msg("Qgroup basic add\n"); + test_msg("qgroup basic add"); ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FS_TREE_OBJECTID); if (ret) { - test_msg("Couldn't create a qgroup %d\n", ret); + test_err("couldn't create a qgroup %d", ret); return ret; } @@ -244,7 +231,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root, false); if (ret) { ulist_free(old_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } @@ -258,20 +245,20 @@ static int test_no_shared_qgroup(struct btrfs_root *root, if (ret) { ulist_free(old_roots); ulist_free(new_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize, nodesize, old_roots, new_roots); if (ret) { - test_msg("Couldn't account space for a qgroup %d\n", ret); + test_err("couldn't account space for a qgroup %d", ret); return ret; } if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID, nodesize, nodesize)) { - test_msg("Qgroup counts didn't match expected values\n"); + test_err("qgroup counts didn't match expected values"); return -EINVAL; } old_roots = NULL; @@ -281,7 +268,7 @@ static int test_no_shared_qgroup(struct btrfs_root *root, false); if (ret) { ulist_free(old_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } @@ -294,19 +281,19 @@ static int test_no_shared_qgroup(struct btrfs_root *root, if (ret) { ulist_free(old_roots); ulist_free(new_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize, nodesize, old_roots, new_roots); if (ret) { - test_msg("Couldn't account space for a qgroup %d\n", ret); + test_err("couldn't account space for a qgroup %d", ret); return -EINVAL; } if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID, 0, 0)) { - test_msg("Qgroup counts didn't match expected values\n"); + test_err("qgroup counts didn't match expected values"); return -EINVAL; } @@ -327,9 +314,9 @@ static int test_multiple_refs(struct btrfs_root *root, struct ulist *new_roots = NULL; int ret; - btrfs_init_dummy_trans(&trans); + btrfs_init_dummy_trans(&trans, fs_info); - test_msg("Qgroup multiple refs test\n"); + test_msg("qgroup multiple refs test"); /* * We have BTRFS_FS_TREE_OBJECTID created already from the @@ -337,7 +324,7 @@ static int test_multiple_refs(struct btrfs_root *root, */ ret = btrfs_create_qgroup(NULL, fs_info, BTRFS_FIRST_FREE_OBJECTID); if (ret) { - test_msg("Couldn't create a qgroup %d\n", ret); + test_err("couldn't create a qgroup %d", ret); return ret; } @@ -345,7 +332,7 @@ static int test_multiple_refs(struct btrfs_root *root, false); if (ret) { ulist_free(old_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } @@ -359,20 +346,20 @@ static int test_multiple_refs(struct btrfs_root *root, if (ret) { ulist_free(old_roots); ulist_free(new_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize, nodesize, old_roots, new_roots); if (ret) { - test_msg("Couldn't account space for a qgroup %d\n", ret); + test_err("couldn't account space for a qgroup %d", ret); return ret; } if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID, nodesize, nodesize)) { - test_msg("Qgroup counts didn't match expected values\n"); + test_err("qgroup counts didn't match expected values"); return -EINVAL; } @@ -380,7 +367,7 @@ static int test_multiple_refs(struct btrfs_root *root, false); if (ret) { ulist_free(old_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } @@ -394,26 +381,26 @@ static int test_multiple_refs(struct btrfs_root *root, if (ret) { ulist_free(old_roots); ulist_free(new_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize, nodesize, old_roots, new_roots); if (ret) { - test_msg("Couldn't account space for a qgroup %d\n", ret); + test_err("couldn't account space for a qgroup %d", ret); return ret; } if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID, nodesize, 0)) { - test_msg("Qgroup counts didn't match expected values\n"); + test_err("qgroup counts didn't match expected values"); return -EINVAL; } if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FIRST_FREE_OBJECTID, nodesize, 0)) { - test_msg("Qgroup counts didn't match expected values\n"); + test_err("qgroup counts didn't match expected values"); return -EINVAL; } @@ -421,7 +408,7 @@ static int test_multiple_refs(struct btrfs_root *root, false); if (ret) { ulist_free(old_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } @@ -435,26 +422,26 @@ static int test_multiple_refs(struct btrfs_root *root, if (ret) { ulist_free(old_roots); ulist_free(new_roots); - test_msg("Couldn't find old roots: %d\n", ret); + test_err("couldn't find old roots: %d", ret); return ret; } ret = btrfs_qgroup_account_extent(&trans, fs_info, nodesize, nodesize, old_roots, new_roots); if (ret) { - test_msg("Couldn't account space for a qgroup %d\n", ret); + test_err("couldn't account space for a qgroup %d", ret); return ret; } if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FIRST_FREE_OBJECTID, 0, 0)) { - test_msg("Qgroup counts didn't match expected values\n"); + test_err("qgroup counts didn't match expected values"); return -EINVAL; } if (btrfs_verify_qgroup_counts(fs_info, BTRFS_FS_TREE_OBJECTID, nodesize, nodesize)) { - test_msg("Qgroup counts didn't match expected values\n"); + test_err("qgroup counts didn't match expected values"); return -EINVAL; } @@ -470,13 +457,13 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize) fs_info = btrfs_alloc_dummy_fs_info(nodesize, sectorsize); if (!fs_info) { - test_msg("Couldn't allocate dummy fs info\n"); + test_err("couldn't allocate dummy fs info"); return -ENOMEM; } root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(root)) { - test_msg("Couldn't allocate root\n"); + test_err("couldn't allocate root"); ret = PTR_ERR(root); goto out; } @@ -498,7 +485,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize) */ root->node = alloc_test_extent_buffer(root->fs_info, nodesize); if (!root->node) { - test_msg("Couldn't allocate dummy buffer\n"); + test_err("couldn't allocate dummy buffer"); ret = -ENOMEM; goto out; } @@ -508,7 +495,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize) tmp_root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(tmp_root)) { - test_msg("Couldn't allocate a fs root\n"); + test_err("couldn't allocate a fs root"); ret = PTR_ERR(tmp_root); goto out; } @@ -517,13 +504,13 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize) root->fs_info->fs_root = tmp_root; ret = btrfs_insert_fs_root(root->fs_info, tmp_root); if (ret) { - test_msg("Couldn't insert fs root %d\n", ret); + test_err("couldn't insert fs root %d", ret); goto out; } tmp_root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(tmp_root)) { - test_msg("Couldn't allocate a fs root\n"); + test_err("couldn't allocate a fs root"); ret = PTR_ERR(tmp_root); goto out; } @@ -531,11 +518,11 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize) tmp_root->root_key.objectid = BTRFS_FIRST_FREE_OBJECTID; ret = btrfs_insert_fs_root(root->fs_info, tmp_root); if (ret) { - test_msg("Couldn't insert fs root %d\n", ret); + test_err("couldn't insert fs root %d", ret); goto out; } - test_msg("Running qgroup tests\n"); + test_msg("running qgroup tests"); ret = test_no_shared_qgroup(root, sectorsize, nodesize); if (ret) goto out; diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 5c4cf0f9146b..4485eae41e88 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/fs.h> @@ -890,12 +877,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, atomic_dec(&cur_trans->num_writers); extwriter_counter_dec(cur_trans, trans->type); - /* - * Make sure counter is updated before we wake up waiters. - */ - smp_mb(); - if (waitqueue_active(&cur_trans->writer_wait)) - wake_up(&cur_trans->writer_wait); + cond_wake_up(&cur_trans->writer_wait); btrfs_put_transaction(cur_trans); if (current->journal_info == trans) @@ -1263,7 +1245,6 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans) btrfs_free_log(trans, root); btrfs_update_reloc_root(trans, root); - btrfs_orphan_commit_root(trans, root); btrfs_save_ino_cache(root, trans); @@ -1653,15 +1634,14 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, btrfs_abort_transaction(trans, ret); goto fail; } - ret = btrfs_uuid_tree_add(trans, fs_info, new_uuid.b, - BTRFS_UUID_KEY_SUBVOL, objectid); + ret = btrfs_uuid_tree_add(trans, new_uuid.b, BTRFS_UUID_KEY_SUBVOL, + objectid); if (ret) { btrfs_abort_transaction(trans, ret); goto fail; } if (!btrfs_is_empty_uuid(new_root_item->received_uuid)) { - ret = btrfs_uuid_tree_add(trans, fs_info, - new_root_item->received_uuid, + ret = btrfs_uuid_tree_add(trans, new_root_item->received_uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, objectid); if (ret && ret != -EEXIST) { @@ -2280,6 +2260,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) */ cur_trans->state = TRANS_STATE_COMPLETED; wake_up(&cur_trans->commit_wait); + clear_bit(BTRFS_FS_NEED_ASYNC_COMMIT, &fs_info->flags); spin_lock(&fs_info->trans_lock); list_del_init(&cur_trans->list); diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index b6c94ce33503..94439482a0ec 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_TRANSACTION__ -#define __BTRFS_TRANSACTION__ +#ifndef BTRFS_TRANSACTION_H +#define BTRFS_TRANSACTION_H #include <linux/refcount.h> #include "btrfs_inode.h" @@ -152,7 +139,6 @@ struct btrfs_pending_snapshot { struct btrfs_path *path; /* block reservation for the operation */ struct btrfs_block_rsv block_rsv; - u64 qgroup_reserved; /* extra metadata reservation for relocation */ int error; bool readonly; @@ -212,6 +198,20 @@ int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root); int btrfs_commit_transaction(struct btrfs_trans_handle *trans); int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans, int wait_for_unblock); + +/* + * Try to commit transaction asynchronously, so this is safe to call + * even holding a spinlock. + * + * It's done by informing transaction_kthread to commit transaction without + * waiting for commit interval. + */ +static inline void btrfs_commit_transaction_locksafe( + struct btrfs_fs_info *fs_info) +{ + set_bit(BTRFS_FS_NEED_ASYNC_COMMIT, &fs_info->flags); + wake_up_process(fs_info->transaction_kthread); +} int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans); int btrfs_should_end_transaction(struct btrfs_trans_handle *trans); void btrfs_throttle(struct btrfs_fs_info *fs_info); @@ -228,4 +228,5 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction); void btrfs_apply_pending_changes(struct btrfs_fs_info *fs_info); void btrfs_add_dropped_root(struct btrfs_trans_handle *trans, struct btrfs_root *root); + #endif diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 8871286c1a91..8d40e7dd8c30 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1,17 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) Qu Wenruo 2017. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program. */ /* diff --git a/fs/btrfs/tree-checker.h b/fs/btrfs/tree-checker.h index aba542755710..ff043275b784 100644 --- a/fs/btrfs/tree-checker.h +++ b/fs/btrfs/tree-checker.h @@ -1,21 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) Qu Wenruo 2017. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program. */ -#ifndef __BTRFS_TREE_CHECKER__ -#define __BTRFS_TREE_CHECKER__ +#ifndef BTRFS_TREE_CHECKER_H +#define BTRFS_TREE_CHECKER_H #include "ctree.h" #include "extent_io.h" diff --git a/fs/btrfs/tree-defrag.c b/fs/btrfs/tree-defrag.c index c09dbe4bd6e7..3c0987ab587d 100644 --- a/fs/btrfs/tree-defrag.c +++ b/fs/btrfs/tree-defrag.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index c91babc6aa4b..f8220ec02036 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/sched.h> @@ -235,11 +222,8 @@ int btrfs_pin_log_trans(struct btrfs_root *root) void btrfs_end_log_trans(struct btrfs_root *root) { if (atomic_dec_and_test(&root->log_writers)) { - /* - * Implicit memory barrier after atomic_dec_and_test - */ - if (waitqueue_active(&root->log_writer_wait)) - wake_up(&root->log_writer_wait); + /* atomic_dec_and_test implies a barrier */ + cond_wake_up_nomb(&root->log_writer_wait); } } @@ -2352,8 +2336,10 @@ again: nritems = btrfs_header_nritems(path->nodes[0]); if (path->slots[0] >= nritems) { ret = btrfs_next_leaf(root, path); - if (ret) + if (ret == 1) break; + else if (ret < 0) + goto out; } btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]); @@ -2456,13 +2442,41 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, if (ret) break; - /* for regular files, make sure corresponding - * orphan item exist. extents past the new EOF - * will be truncated later by orphan cleanup. + /* + * Before replaying extents, truncate the inode to its + * size. We need to do it now and not after log replay + * because before an fsync we can have prealloc extents + * added beyond the inode's i_size. If we did it after, + * through orphan cleanup for example, we would drop + * those prealloc extents just after replaying them. */ if (S_ISREG(mode)) { - ret = insert_orphan_item(wc->trans, root, - key.objectid); + struct inode *inode; + u64 from; + + inode = read_one_inode(root, key.objectid); + if (!inode) { + ret = -EIO; + break; + } + from = ALIGN(i_size_read(inode), + root->fs_info->sectorsize); + ret = btrfs_drop_extents(wc->trans, root, inode, + from, (u64)-1, 1); + /* + * If the nlink count is zero here, the iput + * will free the inode. We bump it to make + * sure it doesn't get freed until the link + * count fixup is done. + */ + if (!ret) { + if (inode->i_nlink == 0) + inc_nlink(inode); + /* Update link count and nbytes. */ + ret = btrfs_update_inode(wc->trans, + root, inode); + } + iput(inode); if (ret) break; } @@ -2971,11 +2985,8 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, mutex_lock(&log_root_tree->log_mutex); if (atomic_dec_and_test(&log_root_tree->log_writers)) { - /* - * Implicit memory barrier after atomic_dec_and_test - */ - if (waitqueue_active(&log_root_tree->log_writer_wait)) - wake_up(&log_root_tree->log_writer_wait); + /* atomic_dec_and_test implies a barrier */ + cond_wake_up_nomb(&log_root_tree->log_writer_wait); } if (ret) { @@ -3099,10 +3110,11 @@ out_wake_log_root: mutex_unlock(&log_root_tree->log_mutex); /* - * The barrier before waitqueue_active is implied by mutex_unlock + * The barrier before waitqueue_active (in cond_wake_up) is needed so + * all the updates above are seen by the woken threads. It might not be + * necessary, but proving that seems to be hard. */ - if (waitqueue_active(&log_root_tree->log_commit_wait[index2])) - wake_up(&log_root_tree->log_commit_wait[index2]); + cond_wake_up(&log_root_tree->log_commit_wait[index2]); out: mutex_lock(&root->log_mutex); btrfs_remove_all_log_ctxs(root, index1, ret); @@ -3111,10 +3123,11 @@ out: mutex_unlock(&root->log_mutex); /* - * The barrier before waitqueue_active is implied by mutex_unlock + * The barrier before waitqueue_active (in cond_wake_up) is needed so + * all the updates above are seen by the woken threads. It might not be + * necessary, but proving that seems to be hard. */ - if (waitqueue_active(&root->log_commit_wait[index1])) - wake_up(&root->log_commit_wait[index1]); + cond_wake_up(&root->log_commit_wait[index1]); return ret; } @@ -3520,8 +3533,11 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans, * from this directory and from this transaction */ ret = btrfs_next_leaf(root, path); - if (ret == 1) { - last_offset = (u64)-1; + if (ret) { + if (ret == 1) + last_offset = (u64)-1; + else + err = ret; goto done; } btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]); @@ -4300,6 +4316,110 @@ static int log_one_extent(struct btrfs_trans_handle *trans, return ret; } +/* + * Log all prealloc extents beyond the inode's i_size to make sure we do not + * lose them after doing a fast fsync and replaying the log. We scan the + * subvolume's root instead of iterating the inode's extent map tree because + * otherwise we can log incorrect extent items based on extent map conversion. + * That can happen due to the fact that extent maps are merged when they + * are not in the extent map tree's list of modified extents. + */ +static int btrfs_log_prealloc_extents(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + struct btrfs_path *path) +{ + struct btrfs_root *root = inode->root; + struct btrfs_key key; + const u64 i_size = i_size_read(&inode->vfs_inode); + const u64 ino = btrfs_ino(inode); + struct btrfs_path *dst_path = NULL; + u64 last_extent = (u64)-1; + int ins_nr = 0; + int start_slot; + int ret; + + if (!(inode->flags & BTRFS_INODE_PREALLOC)) + return 0; + + key.objectid = ino; + key.type = BTRFS_EXTENT_DATA_KEY; + key.offset = i_size; + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) + goto out; + + while (true) { + struct extent_buffer *leaf = path->nodes[0]; + int slot = path->slots[0]; + + if (slot >= btrfs_header_nritems(leaf)) { + if (ins_nr > 0) { + ret = copy_items(trans, inode, dst_path, path, + &last_extent, start_slot, + ins_nr, 1, 0); + if (ret < 0) + goto out; + ins_nr = 0; + } + ret = btrfs_next_leaf(root, path); + if (ret < 0) + goto out; + if (ret > 0) { + ret = 0; + break; + } + continue; + } + + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.objectid > ino) + break; + if (WARN_ON_ONCE(key.objectid < ino) || + key.type < BTRFS_EXTENT_DATA_KEY || + key.offset < i_size) { + path->slots[0]++; + continue; + } + if (last_extent == (u64)-1) { + last_extent = key.offset; + /* + * Avoid logging extent items logged in past fsync calls + * and leading to duplicate keys in the log tree. + */ + do { + ret = btrfs_truncate_inode_items(trans, + root->log_root, + &inode->vfs_inode, + i_size, + BTRFS_EXTENT_DATA_KEY); + } while (ret == -EAGAIN); + if (ret) + goto out; + } + if (ins_nr == 0) + start_slot = slot; + ins_nr++; + path->slots[0]++; + if (!dst_path) { + dst_path = btrfs_alloc_path(); + if (!dst_path) { + ret = -ENOMEM; + goto out; + } + } + } + if (ins_nr > 0) { + ret = copy_items(trans, inode, dst_path, path, &last_extent, + start_slot, ins_nr, 1, 0); + if (ret > 0) + ret = 0; + } +out: + btrfs_release_path(path); + btrfs_free_path(dst_path); + return ret; +} + static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_inode *inode, @@ -4342,6 +4462,11 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans, if (em->generation <= test_gen) continue; + /* We log prealloc extents beyond eof later. */ + if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) && + em->start >= i_size_read(&inode->vfs_inode)) + continue; + if (em->start < logged_start) logged_start = em->start; if ((em->start + em->len - 1) > logged_end) @@ -4398,6 +4523,9 @@ process: up_write(&inode->dio_sem); btrfs_release_path(path); + if (!ret) + ret = btrfs_log_prealloc_extents(trans, inode, path); + return ret; } @@ -4782,6 +4910,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, struct extent_map_tree *em_tree = &inode->extent_tree; u64 logged_isize = 0; bool need_log_inode_item = true; + bool xattrs_logged = false; path = btrfs_alloc_path(); if (!path) @@ -5083,6 +5212,7 @@ next_key: err = btrfs_log_all_xattrs(trans, root, inode, path, dst_path); if (err) goto out_unlock; + xattrs_logged = true; if (max_key.type >= BTRFS_EXTENT_DATA_KEY && !fast_search) { btrfs_release_path(path); btrfs_release_path(dst_path); @@ -5095,6 +5225,11 @@ log_extents: btrfs_release_path(dst_path); if (need_log_inode_item) { err = log_inode_item(trans, log, dst_path, inode); + if (!err && !xattrs_logged) { + err = btrfs_log_all_xattrs(trans, root, inode, path, + dst_path); + btrfs_release_path(path); + } if (err) goto out_unlock; } diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h index 88abc43312a1..122e68b89a5a 100644 --- a/fs/btrfs/tree-log.h +++ b/fs/btrfs/tree-log.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2008 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __TREE_LOG_ -#define __TREE_LOG_ +#ifndef BTRFS_TREE_LOG_H +#define BTRFS_TREE_LOG_H #include "ctree.h" #include "transaction.h" @@ -87,4 +74,5 @@ void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans, int btrfs_log_new_name(struct btrfs_trans_handle *trans, struct btrfs_inode *inode, struct btrfs_inode *old_dir, struct dentry *parent); + #endif diff --git a/fs/btrfs/ulist.c b/fs/btrfs/ulist.c index d8edf164f81c..3374c9e9be67 100644 --- a/fs/btrfs/ulist.c +++ b/fs/btrfs/ulist.c @@ -1,7 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2011 STRATO AG * written by Arne Jansen <sensille@gmx.net> - * Distributed under the GNU GPL license version 2. */ #include <linux/slab.h> diff --git a/fs/btrfs/ulist.h b/fs/btrfs/ulist.h index 53c913632733..02fda0a2d4ce 100644 --- a/fs/btrfs/ulist.h +++ b/fs/btrfs/ulist.h @@ -1,12 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2011 STRATO AG * written by Arne Jansen <sensille@gmx.net> - * Distributed under the GNU GPL license version 2. - * */ -#ifndef __ULIST__ -#define __ULIST__ +#ifndef BTRFS_ULIST_H +#define BTRFS_ULIST_H #include <linux/list.h> #include <linux/rbtree.h> diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c index 9916f03430bc..3b2ae342e649 100644 --- a/fs/btrfs/uuid-tree.c +++ b/fs/btrfs/uuid-tree.c @@ -1,20 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) STRATO AG 2013. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ + #include <linux/uuid.h> #include <asm/unaligned.h> #include "ctree.h" @@ -91,10 +79,10 @@ out: return ret; } -int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u8 *uuid, u8 type, +int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type, u64 subid_cpu) { + struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_root *uuid_root = fs_info->uuid_root; int ret; struct btrfs_path *path = NULL; @@ -156,10 +144,10 @@ out: return ret; } -int btrfs_uuid_tree_rem(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u8 *uuid, u8 type, +int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type, u64 subid) { + struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_root *uuid_root = fs_info->uuid_root; int ret; struct btrfs_path *path = NULL; @@ -251,7 +239,7 @@ static int btrfs_uuid_iter_rem(struct btrfs_root *uuid_root, u8 *uuid, u8 type, goto out; } - ret = btrfs_uuid_tree_rem(trans, uuid_root->fs_info, uuid, type, subid); + ret = btrfs_uuid_tree_remove(trans, uuid, type, subid); btrfs_end_transaction(trans); out: diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 93f8f17cacca..e034ad9e23b4 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1,20 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ + #include <linux/sched.h> #include <linux/bio.h> #include <linux/slab.h> @@ -52,6 +40,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { .tolerated_failures = 1, .devs_increment = 2, .ncopies = 2, + .raid_name = "raid10", + .bg_flag = BTRFS_BLOCK_GROUP_RAID10, + .mindev_error = BTRFS_ERROR_DEV_RAID10_MIN_NOT_MET, }, [BTRFS_RAID_RAID1] = { .sub_stripes = 1, @@ -61,6 +52,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { .tolerated_failures = 1, .devs_increment = 2, .ncopies = 2, + .raid_name = "raid1", + .bg_flag = BTRFS_BLOCK_GROUP_RAID1, + .mindev_error = BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET, }, [BTRFS_RAID_DUP] = { .sub_stripes = 1, @@ -70,6 +64,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { .tolerated_failures = 0, .devs_increment = 1, .ncopies = 2, + .raid_name = "dup", + .bg_flag = BTRFS_BLOCK_GROUP_DUP, + .mindev_error = 0, }, [BTRFS_RAID_RAID0] = { .sub_stripes = 1, @@ -79,6 +76,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { .tolerated_failures = 0, .devs_increment = 1, .ncopies = 1, + .raid_name = "raid0", + .bg_flag = BTRFS_BLOCK_GROUP_RAID0, + .mindev_error = 0, }, [BTRFS_RAID_SINGLE] = { .sub_stripes = 1, @@ -88,6 +88,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { .tolerated_failures = 0, .devs_increment = 1, .ncopies = 1, + .raid_name = "single", + .bg_flag = 0, + .mindev_error = 0, }, [BTRFS_RAID_RAID5] = { .sub_stripes = 1, @@ -97,6 +100,9 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { .tolerated_failures = 1, .devs_increment = 1, .ncopies = 2, + .raid_name = "raid5", + .bg_flag = BTRFS_BLOCK_GROUP_RAID5, + .mindev_error = BTRFS_ERROR_DEV_RAID5_MIN_NOT_MET, }, [BTRFS_RAID_RAID6] = { .sub_stripes = 1, @@ -106,33 +112,19 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = { .tolerated_failures = 2, .devs_increment = 1, .ncopies = 3, + .raid_name = "raid6", + .bg_flag = BTRFS_BLOCK_GROUP_RAID6, + .mindev_error = BTRFS_ERROR_DEV_RAID6_MIN_NOT_MET, }, }; -const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES] = { - [BTRFS_RAID_RAID10] = BTRFS_BLOCK_GROUP_RAID10, - [BTRFS_RAID_RAID1] = BTRFS_BLOCK_GROUP_RAID1, - [BTRFS_RAID_DUP] = BTRFS_BLOCK_GROUP_DUP, - [BTRFS_RAID_RAID0] = BTRFS_BLOCK_GROUP_RAID0, - [BTRFS_RAID_SINGLE] = 0, - [BTRFS_RAID_RAID5] = BTRFS_BLOCK_GROUP_RAID5, - [BTRFS_RAID_RAID6] = BTRFS_BLOCK_GROUP_RAID6, -}; +const char *get_raid_name(enum btrfs_raid_types type) +{ + if (type >= BTRFS_NR_RAID_TYPES) + return NULL; -/* - * Table to convert BTRFS_RAID_* to the error code if minimum number of devices - * condition is not met. Zero means there's no corresponding - * BTRFS_ERROR_DEV_*_NOT_MET value. - */ -const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES] = { - [BTRFS_RAID_RAID10] = BTRFS_ERROR_DEV_RAID10_MIN_NOT_MET, - [BTRFS_RAID_RAID1] = BTRFS_ERROR_DEV_RAID1_MIN_NOT_MET, - [BTRFS_RAID_DUP] = 0, - [BTRFS_RAID_RAID0] = 0, - [BTRFS_RAID_SINGLE] = 0, - [BTRFS_RAID_RAID5] = BTRFS_ERROR_DEV_RAID5_MIN_NOT_MET, - [BTRFS_RAID_RAID6] = BTRFS_ERROR_DEV_RAID6_MIN_NOT_MET, -}; + return btrfs_raid_array[type].raid_name; +} static int init_first_rw_device(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info); @@ -179,12 +171,6 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, * may be used to exclude some operations from running concurrently without any * modifications to the list (see write_all_supers) * - * volume_mutex - * ------------ - * coarse lock owned by a mounted filesystem; used to exclude some operations - * that cannot run in parallel and affect the higher-level properties of the - * filesystem like: device add/deleting/resize/replace, or balance - * * balance_mutex * ------------- * protects balance structures (status, state) and context accessed from @@ -209,6 +195,41 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, * device_list_mutex * chunk_mutex * balance_mutex + * + * + * Exclusive operations, BTRFS_FS_EXCL_OP + * ====================================== + * + * Maintains the exclusivity of the following operations that apply to the + * whole filesystem and cannot run in parallel. + * + * - Balance (*) + * - Device add + * - Device remove + * - Device replace (*) + * - Resize + * + * The device operations (as above) can be in one of the following states: + * + * - Running state + * - Paused state + * - Completed state + * + * Only device operations marked with (*) can go into the Paused state for the + * following reasons: + * + * - ioctl (only Balance can be Paused through ioctl) + * - filesystem remounted as read-only + * - filesystem unmounted and mounted as read-only + * - system power-cycle and filesystem mounted as read-only + * - filesystem or device errors leading to forced read-only + * + * BTRFS_FS_EXCL_OP flag is set and cleared using atomic operations. + * During the course of Paused state, the BTRFS_FS_EXCL_OP remains set. + * A device operation in Paused or Running state can be canceled or resumed + * either by ioctl (Balance only) or when remounted as read-write. + * BTRFS_FS_EXCL_OP flag is cleared when the device operation is canceled or + * completed. */ DEFINE_MUTEX(uuid_mutex); @@ -239,14 +260,14 @@ static struct btrfs_fs_devices *alloc_fs_devices(const u8 *fsid) INIT_LIST_HEAD(&fs_devs->devices); INIT_LIST_HEAD(&fs_devs->resized_devices); INIT_LIST_HEAD(&fs_devs->alloc_list); - INIT_LIST_HEAD(&fs_devs->list); + INIT_LIST_HEAD(&fs_devs->fs_list); if (fsid) memcpy(fs_devs->fsid, fsid, BTRFS_FSID_SIZE); return fs_devs; } -static void free_device(struct btrfs_device *device) +void btrfs_free_device(struct btrfs_device *device) { rcu_string_free(device->name); bio_put(device->flush_bio); @@ -261,7 +282,7 @@ static void free_fs_devices(struct btrfs_fs_devices *fs_devices) device = list_entry(fs_devices->devices.next, struct btrfs_device, dev_list); list_del(&device->dev_list); - free_device(device); + btrfs_free_device(device); } kfree(fs_devices); } @@ -285,8 +306,8 @@ void __exit btrfs_cleanup_fs_uuids(void) while (!list_empty(&fs_uuids)) { fs_devices = list_entry(fs_uuids.next, - struct btrfs_fs_devices, list); - list_del(&fs_devices->list); + struct btrfs_fs_devices, fs_list); + list_del(&fs_devices->fs_list); free_fs_devices(fs_devices); } } @@ -294,7 +315,7 @@ void __exit btrfs_cleanup_fs_uuids(void) /* * Returns a pointer to a new btrfs_device on success; ERR_PTR() on error. * Returned struct is not linked onto any lists and must be destroyed using - * free_device. + * btrfs_free_device. */ static struct btrfs_device *__alloc_device(void) { @@ -339,10 +360,9 @@ static struct btrfs_device *__alloc_device(void) static struct btrfs_device *find_device(struct btrfs_fs_devices *fs_devices, u64 devid, const u8 *uuid) { - struct list_head *head = &fs_devices->devices; struct btrfs_device *dev; - list_for_each_entry(dev, head, dev_list) { + list_for_each_entry(dev, &fs_devices->devices, dev_list) { if (dev->devid == devid && (!uuid || !memcmp(dev->uuid, uuid, BTRFS_UUID_SIZE))) { return dev; @@ -355,7 +375,7 @@ static noinline struct btrfs_fs_devices *find_fsid(u8 *fsid) { struct btrfs_fs_devices *fs_devices; - list_for_each_entry(fs_devices, &fs_uuids, list) { + list_for_each_entry(fs_devices, &fs_uuids, fs_list) { if (memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE) == 0) return fs_devices; } @@ -619,7 +639,7 @@ static void btrfs_free_stale_devices(const char *path, struct btrfs_fs_devices *fs_devs, *tmp_fs_devs; struct btrfs_device *dev, *tmp_dev; - list_for_each_entry_safe(fs_devs, tmp_fs_devs, &fs_uuids, list) { + list_for_each_entry_safe(fs_devs, tmp_fs_devs, &fs_uuids, fs_list) { if (fs_devs->opened) continue; @@ -644,13 +664,13 @@ static void btrfs_free_stale_devices(const char *path, /* delete the stale device */ if (fs_devs->num_devices == 1) { btrfs_sysfs_remove_fsid(fs_devs); - list_del(&fs_devs->list); + list_del(&fs_devs->fs_list); free_fs_devices(fs_devs); break; } else { fs_devs->num_devices--; list_del(&dev->dev_list); - free_device(dev); + btrfs_free_device(dev); } } } @@ -744,7 +764,7 @@ static noinline struct btrfs_device *device_list_add(const char *path, if (IS_ERR(fs_devices)) return ERR_CAST(fs_devices); - list_add(&fs_devices->list, &fs_uuids); + list_add(&fs_devices->fs_list, &fs_uuids); device = NULL; } else { @@ -765,7 +785,7 @@ static noinline struct btrfs_device *device_list_add(const char *path, name = rcu_string_strdup(path, GFP_NOFS); if (!name) { - free_device(device); + btrfs_free_device(device); return ERR_PTR(-ENOMEM); } rcu_assign_pointer(device->name, name); @@ -878,7 +898,7 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig) name = rcu_string_strdup(orig_dev->name->str, GFP_KERNEL); if (!name) { - free_device(device); + btrfs_free_device(device); goto error; } rcu_assign_pointer(device->name, name); @@ -950,7 +970,7 @@ again: } list_del_init(&device->dev_list); fs_devices->num_devices--; - free_device(device); + btrfs_free_device(device); } if (fs_devices->seed) { @@ -968,7 +988,7 @@ static void free_device_rcu(struct rcu_head *head) struct btrfs_device *device; device = container_of(head, struct btrfs_device, rcu); - free_device(device); + btrfs_free_device(device); } static void btrfs_close_bdev(struct btrfs_device *device) @@ -1017,7 +1037,7 @@ static void btrfs_prepare_close_one_device(struct btrfs_device *device) new_device->fs_devices = device->fs_devices; } -static int __btrfs_close_devices(struct btrfs_fs_devices *fs_devices) +static int close_fs_devices(struct btrfs_fs_devices *fs_devices) { struct btrfs_device *device, *tmp; struct list_head pending_put; @@ -1062,7 +1082,7 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices) int ret; mutex_lock(&uuid_mutex); - ret = __btrfs_close_devices(fs_devices); + ret = close_fs_devices(fs_devices); if (!fs_devices->opened) { seed_devices = fs_devices->seed; fs_devices->seed = NULL; @@ -1072,23 +1092,22 @@ int btrfs_close_devices(struct btrfs_fs_devices *fs_devices) while (seed_devices) { fs_devices = seed_devices; seed_devices = fs_devices->seed; - __btrfs_close_devices(fs_devices); + close_fs_devices(fs_devices); free_fs_devices(fs_devices); } return ret; } -static int __btrfs_open_devices(struct btrfs_fs_devices *fs_devices, +static int open_fs_devices(struct btrfs_fs_devices *fs_devices, fmode_t flags, void *holder) { - struct list_head *head = &fs_devices->devices; struct btrfs_device *device; struct btrfs_device *latest_dev = NULL; int ret = 0; flags |= FMODE_EXCL; - list_for_each_entry(device, head, dev_list) { + list_for_each_entry(device, &fs_devices->devices, dev_list) { /* Just open everything we can; ignore failures here */ if (btrfs_open_one_device(fs_devices, device, flags, holder)) continue; @@ -1127,15 +1146,16 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices, { int ret; - mutex_lock(&uuid_mutex); + mutex_lock(&fs_devices->device_list_mutex); if (fs_devices->opened) { fs_devices->opened++; ret = 0; } else { list_sort(NULL, &fs_devices->devices, devid_cmp); - ret = __btrfs_open_devices(fs_devices, flags, holder); + ret = open_fs_devices(fs_devices, flags, holder); } - mutex_unlock(&uuid_mutex); + mutex_unlock(&fs_devices->device_list_mutex); + return ret; } @@ -1213,31 +1233,29 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder, */ bytenr = btrfs_sb_offset(0); flags |= FMODE_EXCL; - mutex_lock(&uuid_mutex); bdev = blkdev_get_by_path(path, flags, holder); - if (IS_ERR(bdev)) { - ret = PTR_ERR(bdev); - goto error; - } + if (IS_ERR(bdev)) + return PTR_ERR(bdev); if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super)) { ret = -EINVAL; goto error_bdev_put; } + mutex_lock(&uuid_mutex); device = device_list_add(path, disk_super); if (IS_ERR(device)) ret = PTR_ERR(device); else *fs_devices_ret = device->fs_devices; + mutex_unlock(&uuid_mutex); btrfs_release_disk_super(page); error_bdev_put: blkdev_put(bdev, flags); -error: - mutex_unlock(&uuid_mutex); + return ret; } @@ -1869,11 +1887,11 @@ static int btrfs_check_raid_min_devices(struct btrfs_fs_info *fs_info, } while (read_seqretry(&fs_info->profiles_lock, seq)); for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) { - if (!(all_avail & btrfs_raid_group[i])) + if (!(all_avail & btrfs_raid_array[i].bg_flag)) continue; if (num_devices < btrfs_raid_array[i].devs_min) { - int ret = btrfs_raid_mindev_error[i]; + int ret = btrfs_raid_array[i].mindev_error; if (ret) return ret; @@ -1929,13 +1947,13 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, { struct btrfs_device *device; struct btrfs_fs_devices *cur_devices; + struct btrfs_fs_devices *fs_devices = fs_info->fs_devices; u64 num_devices; int ret = 0; - mutex_lock(&fs_info->volume_mutex); mutex_lock(&uuid_mutex); - num_devices = fs_info->fs_devices->num_devices; + num_devices = fs_devices->num_devices; btrfs_dev_replace_read_lock(&fs_info->dev_replace); if (btrfs_dev_replace_is_ongoing(&fs_info->dev_replace)) { WARN_ON(num_devices < 1); @@ -1998,27 +2016,32 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, * (super_copy) should hold the device list mutex. */ + /* + * In normal cases the cur_devices == fs_devices. But in case + * of deleting a seed device, the cur_devices should point to + * its own fs_devices listed under the fs_devices->seed. + */ cur_devices = device->fs_devices; - mutex_lock(&fs_info->fs_devices->device_list_mutex); + mutex_lock(&fs_devices->device_list_mutex); list_del_rcu(&device->dev_list); - device->fs_devices->num_devices--; - device->fs_devices->total_devices--; + cur_devices->num_devices--; + cur_devices->total_devices--; if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) - device->fs_devices->missing_devices--; + cur_devices->missing_devices--; btrfs_assign_next_active_device(fs_info, device, NULL); if (device->bdev) { - device->fs_devices->open_devices--; + cur_devices->open_devices--; /* remove sysfs entry */ - btrfs_sysfs_rm_device_link(fs_info->fs_devices, device); + btrfs_sysfs_rm_device_link(fs_devices, device); } num_devices = btrfs_super_num_devices(fs_info->super_copy) - 1; btrfs_set_super_num_devices(fs_info->super_copy, num_devices); - mutex_unlock(&fs_info->fs_devices->device_list_mutex); + mutex_unlock(&fs_devices->device_list_mutex); /* * at this point, the device is zero sized and detached from @@ -2032,8 +2055,6 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, call_rcu(&device->rcu, free_device_rcu); if (cur_devices->open_devices == 0) { - struct btrfs_fs_devices *fs_devices; - fs_devices = fs_info->fs_devices; while (fs_devices) { if (fs_devices->seed == cur_devices) { fs_devices->seed = cur_devices->seed; @@ -2042,20 +2063,19 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, fs_devices = fs_devices->seed; } cur_devices->seed = NULL; - __btrfs_close_devices(cur_devices); + close_fs_devices(cur_devices); free_fs_devices(cur_devices); } out: mutex_unlock(&uuid_mutex); - mutex_unlock(&fs_info->volume_mutex); return ret; error_undo: if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) { mutex_lock(&fs_info->chunk_mutex); list_add(&device->dev_alloc_list, - &fs_info->fs_devices->alloc_list); + &fs_devices->alloc_list); device->fs_devices->rw_devices++; mutex_unlock(&fs_info->chunk_mutex); } @@ -2124,7 +2144,7 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info, tmp_fs_devices = tmp_fs_devices->seed; } fs_devices->seed = NULL; - __btrfs_close_devices(fs_devices); + close_fs_devices(fs_devices); free_fs_devices(fs_devices); } } @@ -2132,23 +2152,23 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_fs_info *fs_info, void btrfs_destroy_dev_replace_tgtdev(struct btrfs_fs_info *fs_info, struct btrfs_device *tgtdev) { - mutex_lock(&uuid_mutex); + struct btrfs_fs_devices *fs_devices = fs_info->fs_devices; + WARN_ON(!tgtdev); - mutex_lock(&fs_info->fs_devices->device_list_mutex); + mutex_lock(&fs_devices->device_list_mutex); - btrfs_sysfs_rm_device_link(fs_info->fs_devices, tgtdev); + btrfs_sysfs_rm_device_link(fs_devices, tgtdev); if (tgtdev->bdev) - fs_info->fs_devices->open_devices--; + fs_devices->open_devices--; - fs_info->fs_devices->num_devices--; + fs_devices->num_devices--; btrfs_assign_next_active_device(fs_info, tgtdev, NULL); list_del_rcu(&tgtdev->dev_list); - mutex_unlock(&fs_info->fs_devices->device_list_mutex); - mutex_unlock(&uuid_mutex); + mutex_unlock(&fs_devices->device_list_mutex); /* * The update_dev_time() with in btrfs_scratch_superblocks() @@ -2200,10 +2220,6 @@ int btrfs_find_device_missing_or_by_path(struct btrfs_fs_info *fs_info, struct btrfs_device *tmp; devices = &fs_info->fs_devices->devices; - /* - * It is safe to read the devices since the volume_mutex - * is held by the caller. - */ list_for_each_entry(tmp, devices, dev_list) { if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &tmp->dev_state) && !tmp->bdev) { @@ -2271,7 +2287,7 @@ static int btrfs_prepare_sprout(struct btrfs_fs_info *fs_info) return PTR_ERR(old_devices); } - list_add(&old_devices->list, &fs_uuids); + list_add(&old_devices->fs_list, &fs_uuids); memcpy(seed_devices, fs_devices, sizeof(*seed_devices)); seed_devices->opened = 1; @@ -2582,7 +2598,7 @@ error_trans: if (trans) btrfs_end_transaction(trans); error_free_device: - free_device(device); + btrfs_free_device(device); error: blkdev_put(bdev, FMODE_EXCL); if (seeding_dev && !unlocked) { @@ -2592,99 +2608,6 @@ error: return ret; } -int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info, - const char *device_path, - struct btrfs_device *srcdev, - struct btrfs_device **device_out) -{ - struct btrfs_device *device; - struct block_device *bdev; - struct list_head *devices; - struct rcu_string *name; - u64 devid = BTRFS_DEV_REPLACE_DEVID; - int ret = 0; - - *device_out = NULL; - if (fs_info->fs_devices->seeding) { - btrfs_err(fs_info, "the filesystem is a seed filesystem!"); - return -EINVAL; - } - - bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL, - fs_info->bdev_holder); - if (IS_ERR(bdev)) { - btrfs_err(fs_info, "target device %s is invalid!", device_path); - return PTR_ERR(bdev); - } - - filemap_write_and_wait(bdev->bd_inode->i_mapping); - - devices = &fs_info->fs_devices->devices; - list_for_each_entry(device, devices, dev_list) { - if (device->bdev == bdev) { - btrfs_err(fs_info, - "target device is in the filesystem!"); - ret = -EEXIST; - goto error; - } - } - - - if (i_size_read(bdev->bd_inode) < - btrfs_device_get_total_bytes(srcdev)) { - btrfs_err(fs_info, - "target device is smaller than source device!"); - ret = -EINVAL; - goto error; - } - - - device = btrfs_alloc_device(NULL, &devid, NULL); - if (IS_ERR(device)) { - ret = PTR_ERR(device); - goto error; - } - - name = rcu_string_strdup(device_path, GFP_KERNEL); - if (!name) { - free_device(device); - ret = -ENOMEM; - goto error; - } - rcu_assign_pointer(device->name, name); - - mutex_lock(&fs_info->fs_devices->device_list_mutex); - set_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state); - device->generation = 0; - device->io_width = fs_info->sectorsize; - device->io_align = fs_info->sectorsize; - device->sector_size = fs_info->sectorsize; - device->total_bytes = btrfs_device_get_total_bytes(srcdev); - device->disk_total_bytes = btrfs_device_get_disk_total_bytes(srcdev); - device->bytes_used = btrfs_device_get_bytes_used(srcdev); - device->commit_total_bytes = srcdev->commit_total_bytes; - device->commit_bytes_used = device->bytes_used; - device->fs_info = fs_info; - device->bdev = bdev; - set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); - set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state); - device->mode = FMODE_EXCL; - device->dev_stats_valid = 1; - set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE); - device->fs_devices = fs_info->fs_devices; - list_add(&device->dev_list, &fs_info->fs_devices->devices); - fs_info->fs_devices->num_devices++; - fs_info->fs_devices->open_devices++; - mutex_unlock(&fs_info->fs_devices->device_list_mutex); - - *device_out = device; - return ret; - -error: - blkdev_put(bdev, FMODE_EXCL); - return ret; -} - static noinline int btrfs_update_device(struct btrfs_trans_handle *trans, struct btrfs_device *device) { @@ -3285,24 +3208,12 @@ static void update_balance_args(struct btrfs_balance_control *bctl) } /* - * Should be called with both balance and volume mutexes held to - * serialize other volume operations (add_dev/rm_dev/resize) with - * restriper. Same goes for unset_balance_control. + * Clear the balance status in fs_info and delete the balance item from disk. */ -static void set_balance_control(struct btrfs_balance_control *bctl) -{ - struct btrfs_fs_info *fs_info = bctl->fs_info; - - BUG_ON(fs_info->balance_ctl); - - spin_lock(&fs_info->balance_lock); - fs_info->balance_ctl = bctl; - spin_unlock(&fs_info->balance_lock); -} - -static void unset_balance_control(struct btrfs_fs_info *fs_info) +static void reset_balance_state(struct btrfs_fs_info *fs_info) { struct btrfs_balance_control *bctl = fs_info->balance_ctl; + int ret; BUG_ON(!fs_info->balance_ctl); @@ -3311,6 +3222,9 @@ static void unset_balance_control(struct btrfs_fs_info *fs_info) spin_unlock(&fs_info->balance_lock); kfree(bctl); + ret = del_balance_item(fs_info); + if (ret) + btrfs_handle_fs_error(fs_info, ret, NULL); } /* @@ -3847,18 +3761,6 @@ static inline int balance_need_close(struct btrfs_fs_info *fs_info) atomic_read(&fs_info->balance_cancel_req) == 0); } -static void __cancel_balance(struct btrfs_fs_info *fs_info) -{ - int ret; - - unset_balance_control(fs_info); - ret = del_balance_item(fs_info); - if (ret) - btrfs_handle_fs_error(fs_info, ret, NULL); - - clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); -} - /* Non-zero return value signifies invalidity */ static inline int validate_convert_profile(struct btrfs_balance_args *bctl_arg, u64 allowed) @@ -3869,12 +3771,12 @@ static inline int validate_convert_profile(struct btrfs_balance_args *bctl_arg, } /* - * Should be called with both balance and volume mutexes held + * Should be called with balance mutexe held */ -int btrfs_balance(struct btrfs_balance_control *bctl, +int btrfs_balance(struct btrfs_fs_info *fs_info, + struct btrfs_balance_control *bctl, struct btrfs_ioctl_balance_args *bargs) { - struct btrfs_fs_info *fs_info = bctl->fs_info; u64 meta_target, data_target; u64 allowed; int mixed = 0; @@ -3903,7 +3805,7 @@ int btrfs_balance(struct btrfs_balance_control *bctl, !(bctl->flags & BTRFS_BALANCE_METADATA) || memcmp(&bctl->data, &bctl->meta, sizeof(bctl->data))) { btrfs_err(fs_info, - "with mixed groups data and metadata balance options must be the same"); + "balance: mixed groups data and metadata options must be the same"); ret = -EINVAL; goto out; } @@ -3925,23 +3827,29 @@ int btrfs_balance(struct btrfs_balance_control *bctl, allowed |= (BTRFS_BLOCK_GROUP_RAID10 | BTRFS_BLOCK_GROUP_RAID6); if (validate_convert_profile(&bctl->data, allowed)) { + int index = btrfs_bg_flags_to_raid_index(bctl->data.target); + btrfs_err(fs_info, - "unable to start balance with target data profile %llu", - bctl->data.target); + "balance: invalid convert data profile %s", + get_raid_name(index)); ret = -EINVAL; goto out; } if (validate_convert_profile(&bctl->meta, allowed)) { + int index = btrfs_bg_flags_to_raid_index(bctl->meta.target); + btrfs_err(fs_info, - "unable to start balance with target metadata profile %llu", - bctl->meta.target); + "balance: invalid convert metadata profile %s", + get_raid_name(index)); ret = -EINVAL; goto out; } if (validate_convert_profile(&bctl->sys, allowed)) { + int index = btrfs_bg_flags_to_raid_index(bctl->sys.target); + btrfs_err(fs_info, - "unable to start balance with target system profile %llu", - bctl->sys.target); + "balance: invalid convert system profile %s", + get_raid_name(index)); ret = -EINVAL; goto out; } @@ -3962,10 +3870,10 @@ int btrfs_balance(struct btrfs_balance_control *bctl, !(bctl->meta.target & allowed))) { if (bctl->flags & BTRFS_BALANCE_FORCE) { btrfs_info(fs_info, - "force reducing metadata integrity"); + "balance: force reducing metadata integrity"); } else { btrfs_err(fs_info, - "balance will reduce metadata integrity, use force if you want this"); + "balance: reduces metadata integrity, use --force if you want this"); ret = -EINVAL; goto out; } @@ -3979,9 +3887,12 @@ int btrfs_balance(struct btrfs_balance_control *bctl, bctl->data.target : fs_info->avail_data_alloc_bits; if (btrfs_get_num_tolerated_disk_barrier_failures(meta_target) < btrfs_get_num_tolerated_disk_barrier_failures(data_target)) { + int meta_index = btrfs_bg_flags_to_raid_index(meta_target); + int data_index = btrfs_bg_flags_to_raid_index(data_target); + btrfs_warn(fs_info, - "metadata profile 0x%llx has lower redundancy than data profile 0x%llx", - meta_target, data_target); + "balance: metadata profile %s has lower redundancy than data profile %s", + get_raid_name(meta_index), get_raid_name(data_index)); } ret = insert_balance_item(fs_info, bctl); @@ -3990,7 +3901,10 @@ int btrfs_balance(struct btrfs_balance_control *bctl, if (!(bctl->flags & BTRFS_BALANCE_RESUME)) { BUG_ON(ret == -EEXIST); - set_balance_control(bctl); + BUG_ON(fs_info->balance_ctl); + spin_lock(&fs_info->balance_lock); + fs_info->balance_ctl = bctl; + spin_unlock(&fs_info->balance_lock); } else { BUG_ON(ret != -EEXIST); spin_lock(&fs_info->balance_lock); @@ -3998,22 +3912,24 @@ int btrfs_balance(struct btrfs_balance_control *bctl, spin_unlock(&fs_info->balance_lock); } - atomic_inc(&fs_info->balance_running); + ASSERT(!test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)); + set_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags); mutex_unlock(&fs_info->balance_mutex); ret = __btrfs_balance(fs_info); mutex_lock(&fs_info->balance_mutex); - atomic_dec(&fs_info->balance_running); + clear_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags); if (bargs) { memset(bargs, 0, sizeof(*bargs)); - update_ioctl_balance_args(fs_info, 0, bargs); + btrfs_update_ioctl_balance_args(fs_info, bargs); } if ((ret && ret != -ECANCELED && ret != -ENOSPC) || balance_need_close(fs_info)) { - __cancel_balance(fs_info); + reset_balance_state(fs_info); + clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); } wake_up(&fs_info->balance_wait_q); @@ -4021,11 +3937,11 @@ int btrfs_balance(struct btrfs_balance_control *bctl, return ret; out: if (bctl->flags & BTRFS_BALANCE_RESUME) - __cancel_balance(fs_info); - else { + reset_balance_state(fs_info); + else kfree(bctl); - clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); - } + clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); + return ret; } @@ -4034,16 +3950,12 @@ static int balance_kthread(void *data) struct btrfs_fs_info *fs_info = data; int ret = 0; - mutex_lock(&fs_info->volume_mutex); mutex_lock(&fs_info->balance_mutex); - if (fs_info->balance_ctl) { - btrfs_info(fs_info, "continuing balance"); - ret = btrfs_balance(fs_info->balance_ctl, NULL); + btrfs_info(fs_info, "balance: resuming"); + ret = btrfs_balance(fs_info, fs_info->balance_ctl, NULL); } - mutex_unlock(&fs_info->balance_mutex); - mutex_unlock(&fs_info->volume_mutex); return ret; } @@ -4052,18 +3964,27 @@ int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info) { struct task_struct *tsk; - spin_lock(&fs_info->balance_lock); + mutex_lock(&fs_info->balance_mutex); if (!fs_info->balance_ctl) { - spin_unlock(&fs_info->balance_lock); + mutex_unlock(&fs_info->balance_mutex); return 0; } - spin_unlock(&fs_info->balance_lock); + mutex_unlock(&fs_info->balance_mutex); if (btrfs_test_opt(fs_info, SKIP_BALANCE)) { - btrfs_info(fs_info, "force skipping balance"); + btrfs_info(fs_info, "balance: resume skipped"); return 0; } + /* + * A ro->rw remount sequence should continue with the paused balance + * regardless of who pauses it, system or the user as of now, so set + * the resume flag. + */ + spin_lock(&fs_info->balance_lock); + fs_info->balance_ctl->flags |= BTRFS_BALANCE_RESUME; + spin_unlock(&fs_info->balance_lock); + tsk = kthread_run(balance_kthread, fs_info, "btrfs-balance"); return PTR_ERR_OR_ZERO(tsk); } @@ -4103,7 +4024,6 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info) leaf = path->nodes[0]; item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_balance_item); - bctl->fs_info = fs_info; bctl->flags = btrfs_balance_flags(leaf, item); bctl->flags |= BTRFS_BALANCE_RESUME; @@ -4114,15 +4034,26 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info) btrfs_balance_sys(leaf, item, &disk_bargs); btrfs_disk_balance_args_to_cpu(&bctl->sys, &disk_bargs); - WARN_ON(test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)); + /* + * This should never happen, as the paused balance state is recovered + * during mount without any chance of other exclusive ops to collide. + * + * This gives the exclusive op status to balance and keeps in paused + * state until user intervention (cancel or umount). If the ownership + * cannot be assigned, show a message but do not fail. The balance + * is in a paused state and must have fs_info::balance_ctl properly + * set up. + */ + if (test_and_set_bit(BTRFS_FS_EXCL_OP, &fs_info->flags)) + btrfs_warn(fs_info, + "balance: cannot set exclusive op status, resume manually"); - mutex_lock(&fs_info->volume_mutex); mutex_lock(&fs_info->balance_mutex); - - set_balance_control(bctl); - + BUG_ON(fs_info->balance_ctl); + spin_lock(&fs_info->balance_lock); + fs_info->balance_ctl = bctl; + spin_unlock(&fs_info->balance_lock); mutex_unlock(&fs_info->balance_mutex); - mutex_unlock(&fs_info->volume_mutex); out: btrfs_free_path(path); return ret; @@ -4138,16 +4069,16 @@ int btrfs_pause_balance(struct btrfs_fs_info *fs_info) return -ENOTCONN; } - if (atomic_read(&fs_info->balance_running)) { + if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) { atomic_inc(&fs_info->balance_pause_req); mutex_unlock(&fs_info->balance_mutex); wait_event(fs_info->balance_wait_q, - atomic_read(&fs_info->balance_running) == 0); + !test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)); mutex_lock(&fs_info->balance_mutex); /* we are good with balance_ctl ripped off from under us */ - BUG_ON(atomic_read(&fs_info->balance_running)); + BUG_ON(test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)); atomic_dec(&fs_info->balance_pause_req); } else { ret = -ENOTCONN; @@ -4159,38 +4090,49 @@ int btrfs_pause_balance(struct btrfs_fs_info *fs_info) int btrfs_cancel_balance(struct btrfs_fs_info *fs_info) { - if (sb_rdonly(fs_info->sb)) - return -EROFS; - mutex_lock(&fs_info->balance_mutex); if (!fs_info->balance_ctl) { mutex_unlock(&fs_info->balance_mutex); return -ENOTCONN; } + /* + * A paused balance with the item stored on disk can be resumed at + * mount time if the mount is read-write. Otherwise it's still paused + * and we must not allow cancelling as it deletes the item. + */ + if (sb_rdonly(fs_info->sb)) { + mutex_unlock(&fs_info->balance_mutex); + return -EROFS; + } + atomic_inc(&fs_info->balance_cancel_req); /* * if we are running just wait and return, balance item is * deleted in btrfs_balance in this case */ - if (atomic_read(&fs_info->balance_running)) { + if (test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)) { mutex_unlock(&fs_info->balance_mutex); wait_event(fs_info->balance_wait_q, - atomic_read(&fs_info->balance_running) == 0); + !test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)); mutex_lock(&fs_info->balance_mutex); } else { - /* __cancel_balance needs volume_mutex */ mutex_unlock(&fs_info->balance_mutex); - mutex_lock(&fs_info->volume_mutex); + /* + * Lock released to allow other waiters to continue, we'll + * reexamine the status again. + */ mutex_lock(&fs_info->balance_mutex); - if (fs_info->balance_ctl) - __cancel_balance(fs_info); - - mutex_unlock(&fs_info->volume_mutex); + if (fs_info->balance_ctl) { + reset_balance_state(fs_info); + clear_bit(BTRFS_FS_EXCL_OP, &fs_info->flags); + btrfs_info(fs_info, "balance: canceled"); + } } - BUG_ON(fs_info->balance_ctl || atomic_read(&fs_info->balance_running)); + BUG_ON(fs_info->balance_ctl || + test_bit(BTRFS_FS_BALANCE_RUNNING, &fs_info->flags)); atomic_dec(&fs_info->balance_cancel_req); mutex_unlock(&fs_info->balance_mutex); return 0; @@ -4267,8 +4209,7 @@ static int btrfs_uuid_scan_kthread(void *data) } update_tree: if (!btrfs_is_empty_uuid(root_item.uuid)) { - ret = btrfs_uuid_tree_add(trans, fs_info, - root_item.uuid, + ret = btrfs_uuid_tree_add(trans, root_item.uuid, BTRFS_UUID_KEY_SUBVOL, key.objectid); if (ret < 0) { @@ -4279,7 +4220,7 @@ update_tree: } if (!btrfs_is_empty_uuid(root_item.received_uuid)) { - ret = btrfs_uuid_tree_add(trans, fs_info, + ret = btrfs_uuid_tree_add(trans, root_item.received_uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, key.objectid); @@ -4485,7 +4426,7 @@ int btrfs_shrink_device(struct btrfs_device *device, u64 new_size) if (!path) return -ENOMEM; - path->reada = READA_FORWARD; + path->reada = READA_BACK; mutex_lock(&fs_info->chunk_mutex); @@ -6046,9 +5987,8 @@ int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, return __btrfs_map_block(fs_info, op, logical, length, bbio_ret, 0, 1); } -int btrfs_rmap_block(struct btrfs_fs_info *fs_info, - u64 chunk_start, u64 physical, u64 devid, - u64 **logical, int *naddrs, int *stripe_len) +int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, + u64 physical, u64 **logical, int *naddrs, int *stripe_len) { struct extent_map *em; struct map_lookup *map; @@ -6080,8 +6020,6 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info, BUG_ON(!buf); /* -ENOMEM */ for (i = 0; i < map->num_stripes; i++) { - if (devid && map->stripes[i].dev->devid != devid) - continue; if (map->stripes[i].physical > physical || map->stripes[i].physical + length <= physical) continue; @@ -6413,7 +6351,7 @@ static struct btrfs_device *add_missing_dev(struct btrfs_fs_devices *fs_devices, * * Return: a pointer to a new &struct btrfs_device on success; ERR_PTR() * on error. Returned struct is not linked onto any lists and must be - * destroyed with free_device. + * destroyed with btrfs_free_device. */ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info, const u64 *devid, @@ -6436,7 +6374,7 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info, ret = find_next_devid(fs_info, &tmp); if (ret) { - free_device(dev); + btrfs_free_device(dev); return ERR_PTR(ret); } } @@ -6687,8 +6625,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info, if (IS_ERR(fs_devices)) return fs_devices; - ret = __btrfs_open_devices(fs_devices, FMODE_READ, - fs_info->bdev_holder); + ret = open_fs_devices(fs_devices, FMODE_READ, fs_info->bdev_holder); if (ret) { free_fs_devices(fs_devices); fs_devices = ERR_PTR(ret); @@ -6696,7 +6633,7 @@ static struct btrfs_fs_devices *open_seed_devices(struct btrfs_fs_info *fs_info, } if (!fs_devices->seeding) { - __btrfs_close_devices(fs_devices); + close_fs_devices(fs_devices); free_fs_devices(fs_devices); fs_devices = ERR_PTR(-EINVAL); goto out; @@ -7005,6 +6942,10 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info) if (!path) return -ENOMEM; + /* + * uuid_mutex is needed only if we are mounting a sprout FS + * otherwise we don't need it. + */ mutex_lock(&uuid_mutex); mutex_lock(&fs_info->chunk_mutex); diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index d1fcaea9fef5..5139ec8daf4c 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Oracle. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __BTRFS_VOLUMES_ -#define __BTRFS_VOLUMES_ +#ifndef BTRFS_VOLUMES_H +#define BTRFS_VOLUMES_H #include <linux/bio.h> #include <linux/sort.h> @@ -221,6 +208,7 @@ BTRFS_DEVICE_GETSET_FUNCS(bytes_used); struct btrfs_fs_devices { u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */ + struct list_head fs_list; u64 num_devices; u64 open_devices; @@ -242,7 +230,6 @@ struct btrfs_fs_devices { struct list_head resized_devices; /* devices not currently being allocated */ struct list_head alloc_list; - struct list_head list; struct btrfs_fs_devices *seed; int seeding; @@ -342,11 +329,12 @@ struct btrfs_raid_attr { int tolerated_failures; /* max tolerated fail devs */ int devs_increment; /* ndevs has to be a multiple of this */ int ncopies; /* how many copies to data has */ + int mindev_error; /* error code if min devs requisite is unmet */ + const char raid_name[8]; /* name of the raid */ + u64 bg_flag; /* block group flag of the raid */ }; extern const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES]; -extern const int btrfs_raid_mindev_error[BTRFS_NR_RAID_TYPES]; -extern const u64 btrfs_raid_group[BTRFS_NR_RAID_TYPES]; struct map_lookup { u64 type; @@ -364,8 +352,6 @@ struct map_lookup { struct btrfs_balance_args; struct btrfs_balance_progress; struct btrfs_balance_control { - struct btrfs_fs_info *fs_info; - struct btrfs_balance_args data; struct btrfs_balance_args meta; struct btrfs_balance_args sys; @@ -406,9 +392,8 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, u64 logical, u64 *length, struct btrfs_bio **bbio_ret); -int btrfs_rmap_block(struct btrfs_fs_info *fs_info, - u64 chunk_start, u64 physical, u64 devid, - u64 **logical, int *naddrs, int *stripe_len); +int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, + u64 physical, u64 **logical, int *naddrs, int *stripe_len); int btrfs_read_sys_array(struct btrfs_fs_info *fs_info); int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info); int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, @@ -434,6 +419,7 @@ int btrfs_find_device_by_devspec(struct btrfs_fs_info *fs_info, u64 devid, struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info, const u64 *devid, const u8 *uuid); +void btrfs_free_device(struct btrfs_device *device); int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, u64 devid); void __exit btrfs_cleanup_fs_uuids(void); @@ -444,11 +430,8 @@ struct btrfs_device *btrfs_find_device(struct btrfs_fs_info *fs_info, u64 devid, u8 *uuid, u8 *fsid); int btrfs_shrink_device(struct btrfs_device *device, u64 new_size); int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *path); -int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info, - const char *device_path, - struct btrfs_device *srcdev, - struct btrfs_device **device_out); -int btrfs_balance(struct btrfs_balance_control *bctl, +int btrfs_balance(struct btrfs_fs_info *fs_info, + struct btrfs_balance_control *bctl, struct btrfs_ioctl_balance_args *bargs); int btrfs_resume_balance_async(struct btrfs_fs_info *fs_info); int btrfs_recover_balance(struct btrfs_fs_info *fs_info); @@ -566,6 +549,8 @@ static inline enum btrfs_raid_types btrfs_bg_flags_to_raid_index(u64 flags) return BTRFS_RAID_SINGLE; /* BTRFS_BLOCK_GROUP_SINGLE */ } +const char *get_raid_name(enum btrfs_raid_types type); + void btrfs_update_commit_device_size(struct btrfs_fs_info *fs_info); void btrfs_update_commit_device_bytes_used(struct btrfs_transaction *trans); diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index e1e8177deb5e..ea78c3d6dcfc 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -1,19 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2007 Red Hat. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ #include <linux/init.h> @@ -32,7 +19,6 @@ #include "props.h" #include "locking.h" - int btrfs_getxattr(struct inode *inode, const char *name, void *buffer, size_t size) { diff --git a/fs/btrfs/xattr.h b/fs/btrfs/xattr.h index e215a3212a2a..471fcac6ff55 100644 --- a/fs/btrfs/xattr.h +++ b/fs/btrfs/xattr.h @@ -1,23 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright (C) 2007 Red Hat. All rights reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. */ -#ifndef __XATTR__ -#define __XATTR__ +#ifndef BTRFS_XATTR_H +#define BTRFS_XATTR_H #include <linux/xattr.h> @@ -34,4 +21,4 @@ int btrfs_xattr_security_init(struct btrfs_trans_handle *trans, struct inode *inode, struct inode *dir, const struct qstr *qstr); -#endif /* __XATTR__ */ +#endif diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index 2b52950dc2c6..970ff3e35bb3 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -1,20 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2008 Oracle. All rights reserved. * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public - * License along with this program; if not, write to the - * Free Software Foundation, Inc., 59 Temple Place - Suite 330, - * Boston, MA 021110-1307, USA. - * * Based on jffs2 zlib code: * Copyright © 2001-2007 Red Hat, Inc. * Created by David Woodhouse <dwmw2@infradead.org> diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c index 01a4eab602a3..af6ec59972f5 100644 --- a/fs/btrfs/zstd.c +++ b/fs/btrfs/zstd.c @@ -1,16 +1,10 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2016-present, Facebook, Inc. * All rights reserved. * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public - * License v2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. */ + #include <linux/bio.h> #include <linux/err.h> #include <linux/init.h> diff --git a/fs/buffer.c b/fs/buffer.c index 9a73924db22f..cabc045f483d 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -185,10 +185,9 @@ EXPORT_SYMBOL(end_buffer_write_sync); * we get exclusion from try_to_free_buffers with the blockdev mapping's * private_lock. * - * Hack idea: for the blockdev mapping, i_bufferlist_lock contention + * Hack idea: for the blockdev mapping, private_lock contention * may be quite high. This code could TryLock the page, and if that - * succeeds, there is no need to take private_lock. (But if - * private_lock is contended then so is mapping->tree_lock). + * succeeds, there is no need to take private_lock. */ static struct buffer_head * __find_get_block_slow(struct block_device *bdev, sector_t block) @@ -495,35 +494,12 @@ repeat: return err; } -static void do_thaw_one(struct super_block *sb, void *unused) +void emergency_thaw_bdev(struct super_block *sb) { while (sb->s_bdev && !thaw_bdev(sb->s_bdev, sb)) printk(KERN_WARNING "Emergency Thaw on %pg\n", sb->s_bdev); } -static void do_thaw_all(struct work_struct *work) -{ - iterate_supers(do_thaw_one, NULL); - kfree(work); - printk(KERN_WARNING "Emergency Thaw complete\n"); -} - -/** - * emergency_thaw_all -- forcibly thaw every frozen filesystem - * - * Used for emergency unfreeze of all filesystems via SysRq - */ -void emergency_thaw_all(void) -{ - struct work_struct *work; - - work = kmalloc(sizeof(*work), GFP_ATOMIC); - if (work) { - INIT_WORK(work, do_thaw_all); - schedule_work(work); - } -} - /** * sync_mapping_buffers - write out & wait upon a mapping's "associated" buffers * @mapping: the mapping which wants those buffers written @@ -594,20 +570,21 @@ EXPORT_SYMBOL(mark_buffer_dirty_inode); * * The caller must hold lock_page_memcg(). */ -static void __set_page_dirty(struct page *page, struct address_space *mapping, +void __set_page_dirty(struct page *page, struct address_space *mapping, int warn) { unsigned long flags; - spin_lock_irqsave(&mapping->tree_lock, flags); + xa_lock_irqsave(&mapping->i_pages, flags); if (page->mapping) { /* Race with truncate? */ WARN_ON_ONCE(warn && !PageUptodate(page)); account_page_dirtied(page, mapping); - radix_tree_tag_set(&mapping->page_tree, + radix_tree_tag_set(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); } - spin_unlock_irqrestore(&mapping->tree_lock, flags); + xa_unlock_irqrestore(&mapping->i_pages, flags); } +EXPORT_SYMBOL_GPL(__set_page_dirty); /* * Add a page to the dirty page list. @@ -1095,7 +1072,7 @@ __getblk_slow(struct block_device *bdev, sector_t block, * inode list. * * mark_buffer_dirty() is atomic. It takes bh->b_page->mapping->private_lock, - * mapping->tree_lock and mapping->host->i_lock. + * i_pages lock and mapping->host->i_lock. */ void mark_buffer_dirty(struct buffer_head *bh) { @@ -1511,7 +1488,7 @@ void block_invalidatepage(struct page *page, unsigned int offset, * The get_block cached value has been unconditionally invalidated, * so real IO is not possible anymore. */ - if (offset == 0) + if (length == PAGE_SIZE) try_to_release_page(page, 0); out: return; @@ -3450,120 +3427,6 @@ int bh_submit_read(struct buffer_head *bh) } EXPORT_SYMBOL(bh_submit_read); -/* - * Seek for SEEK_DATA / SEEK_HOLE within @page, starting at @lastoff. - * - * Returns the offset within the file on success, and -ENOENT otherwise. - */ -static loff_t -page_seek_hole_data(struct page *page, loff_t lastoff, int whence) -{ - loff_t offset = page_offset(page); - struct buffer_head *bh, *head; - bool seek_data = whence == SEEK_DATA; - - if (lastoff < offset) - lastoff = offset; - - bh = head = page_buffers(page); - do { - offset += bh->b_size; - if (lastoff >= offset) - continue; - - /* - * Unwritten extents that have data in the page cache covering - * them can be identified by the BH_Unwritten state flag. - * Pages with multiple buffers might have a mix of holes, data - * and unwritten extents - any buffer with valid data in it - * should have BH_Uptodate flag set on it. - */ - - if ((buffer_unwritten(bh) || buffer_uptodate(bh)) == seek_data) - return lastoff; - - lastoff = offset; - } while ((bh = bh->b_this_page) != head); - return -ENOENT; -} - -/* - * Seek for SEEK_DATA / SEEK_HOLE in the page cache. - * - * Within unwritten extents, the page cache determines which parts are holes - * and which are data: unwritten and uptodate buffer heads count as data; - * everything else counts as a hole. - * - * Returns the resulting offset on successs, and -ENOENT otherwise. - */ -loff_t -page_cache_seek_hole_data(struct inode *inode, loff_t offset, loff_t length, - int whence) -{ - pgoff_t index = offset >> PAGE_SHIFT; - pgoff_t end = DIV_ROUND_UP(offset + length, PAGE_SIZE); - loff_t lastoff = offset; - struct pagevec pvec; - - if (length <= 0) - return -ENOENT; - - pagevec_init(&pvec); - - do { - unsigned nr_pages, i; - - nr_pages = pagevec_lookup_range(&pvec, inode->i_mapping, &index, - end - 1); - if (nr_pages == 0) - break; - - for (i = 0; i < nr_pages; i++) { - struct page *page = pvec.pages[i]; - - /* - * At this point, the page may be truncated or - * invalidated (changing page->mapping to NULL), or - * even swizzled back from swapper_space to tmpfs file - * mapping. However, page->index will not change - * because we have a reference on the page. - * - * If current page offset is beyond where we've ended, - * we've found a hole. - */ - if (whence == SEEK_HOLE && - lastoff < page_offset(page)) - goto check_range; - - lock_page(page); - if (likely(page->mapping == inode->i_mapping) && - page_has_buffers(page)) { - lastoff = page_seek_hole_data(page, lastoff, whence); - if (lastoff >= 0) { - unlock_page(page); - goto check_range; - } - } - unlock_page(page); - lastoff = page_offset(page) + PAGE_SIZE; - } - pagevec_release(&pvec); - } while (index < end); - - /* When no page at lastoff and we are not done, we found a hole. */ - if (whence != SEEK_HOLE) - goto not_found; - -check_range: - if (lastoff < offset + length) - goto out; -not_found: - lastoff = -ENOENT; -out: - pagevec_release(&pvec); - return lastoff; -} - void __init buffer_init(void) { unsigned long nrpages; diff --git a/fs/cachefiles/interface.c b/fs/cachefiles/interface.c index e7f16a77a22a..222bc5d8b62c 100644 --- a/fs/cachefiles/interface.c +++ b/fs/cachefiles/interface.c @@ -32,7 +32,7 @@ static struct fscache_object *cachefiles_alloc_object( struct cachefiles_cache *cache; struct cachefiles_xattr *auxdata; unsigned keylen, auxlen; - void *buffer; + void *buffer, *p; char *key; cache = container_of(_cache, struct cachefiles_cache, cache); @@ -65,8 +65,12 @@ static struct fscache_object *cachefiles_alloc_object( if (!buffer) goto nomem_buffer; - keylen = cookie->def->get_key(cookie->netfs_data, buffer + 2, 512); - ASSERTCMP(keylen, <, 512); + keylen = cookie->key_len; + if (keylen <= sizeof(cookie->inline_key)) + p = cookie->inline_key; + else + p = cookie->key; + memcpy(buffer + 2, p, keylen); *(uint16_t *)buffer = keylen; ((char *)buffer)[keylen + 2] = 0; @@ -80,15 +84,17 @@ static struct fscache_object *cachefiles_alloc_object( /* get hold of the auxiliary data and prepend the object type */ auxdata = buffer; - auxlen = 0; - if (cookie->def->get_aux) { - auxlen = cookie->def->get_aux(cookie->netfs_data, - auxdata->data, 511); - ASSERTCMP(auxlen, <, 511); + auxlen = cookie->aux_len; + if (auxlen) { + if (auxlen <= sizeof(cookie->inline_aux)) + p = cookie->inline_aux; + else + p = cookie->aux; + memcpy(auxdata->data, p, auxlen); } auxdata->len = auxlen + 1; - auxdata->type = cookie->def->type; + auxdata->type = cookie->type; lookup_data->auxdata = auxdata; lookup_data->key = key; @@ -177,10 +183,12 @@ static void cachefiles_lookup_complete(struct fscache_object *_object) * increment the usage count on an inode object (may fail if unmounting) */ static -struct fscache_object *cachefiles_grab_object(struct fscache_object *_object) +struct fscache_object *cachefiles_grab_object(struct fscache_object *_object, + enum fscache_obj_ref_trace why) { struct cachefiles_object *object = container_of(_object, struct cachefiles_object, fscache); + int u; _enter("{OBJ%x,%d}", _object->debug_id, atomic_read(&object->usage)); @@ -188,7 +196,9 @@ struct fscache_object *cachefiles_grab_object(struct fscache_object *_object) ASSERT((atomic_read(&object->usage) & 0xffff0000) != 0x6b6b0000); #endif - atomic_inc(&object->usage); + u = atomic_inc_return(&object->usage); + trace_cachefiles_ref(object, _object->cookie, + (enum cachefiles_obj_ref_trace)why, u); return &object->fscache; } @@ -202,6 +212,7 @@ static void cachefiles_update_object(struct fscache_object *_object) struct cachefiles_cache *cache; struct fscache_cookie *cookie; const struct cred *saved_cred; + const void *aux; unsigned auxlen; _enter("{OBJ%x}", _object->debug_id); @@ -216,26 +227,29 @@ static void cachefiles_update_object(struct fscache_object *_object) } cookie = object->fscache.cookie; + auxlen = cookie->aux_len; - if (!cookie->def->get_aux) { + if (!auxlen) { fscache_unuse_cookie(_object); _leave(" [no aux]"); return; } - auxdata = kmalloc(2 + 512 + 3, cachefiles_gfp); + auxdata = kmalloc(2 + auxlen + 3, cachefiles_gfp); if (!auxdata) { fscache_unuse_cookie(_object); _leave(" [nomem]"); return; } - auxlen = cookie->def->get_aux(cookie->netfs_data, auxdata->data, 511); + aux = (auxlen <= sizeof(cookie->inline_aux)) ? + cookie->inline_aux : cookie->aux; + + memcpy(auxdata->data, aux, auxlen); fscache_unuse_cookie(_object); - ASSERTCMP(auxlen, <, 511); auxdata->len = auxlen + 1; - auxdata->type = cookie->def->type; + auxdata->type = cookie->type; cachefiles_begin_secure(cache, &saved_cred); cachefiles_update_object_xattr(object, auxdata); @@ -309,10 +323,12 @@ static void cachefiles_drop_object(struct fscache_object *_object) /* * dispose of a reference to an object */ -static void cachefiles_put_object(struct fscache_object *_object) +static void cachefiles_put_object(struct fscache_object *_object, + enum fscache_obj_ref_trace why) { struct cachefiles_object *object; struct fscache_cache *cache; + int u; ASSERT(_object); @@ -328,7 +344,11 @@ static void cachefiles_put_object(struct fscache_object *_object) ASSERTIFCMP(object->fscache.parent, object->fscache.parent->n_children, >, 0); - if (atomic_dec_and_test(&object->usage)) { + u = atomic_dec_return(&object->usage); + trace_cachefiles_ref(object, _object->cookie, + (enum cachefiles_obj_ref_trace)why, u); + ASSERTCMP(u, !=, -1); + if (u == 0) { _debug("- kill object OBJ%x", object->fscache.debug_id); ASSERT(!test_bit(CACHEFILES_OBJECT_ACTIVE, &object->flags)); @@ -421,7 +441,7 @@ static int cachefiles_attr_changed(struct fscache_object *_object) loff_t oi_size; int ret; - _object->cookie->def->get_attr(_object->cookie->netfs_data, &ni_size); + ni_size = _object->store_limit_l; _enter("{OBJ%x},[%llu]", _object->debug_id, (unsigned long long) ni_size); @@ -493,8 +513,7 @@ static void cachefiles_invalidate_object(struct fscache_operation *op) cache = container_of(object->fscache.cache, struct cachefiles_cache, cache); - op->object->cookie->def->get_attr(op->object->cookie->netfs_data, - &ni_size); + ni_size = op->object->store_limit_l; _enter("{OBJ%x},[%llu]", op->object->debug_id, (unsigned long long)ni_size); diff --git a/fs/cachefiles/internal.h b/fs/cachefiles/internal.h index bb3a02ca9da4..d2f6f996e65a 100644 --- a/fs/cachefiles/internal.h +++ b/fs/cachefiles/internal.h @@ -124,6 +124,8 @@ struct cachefiles_xattr { uint8_t data[]; }; +#include <trace/events/cachefiles.h> + /* * note change of state for daemon */ diff --git a/fs/cachefiles/main.c b/fs/cachefiles/main.c index 711f13d8c2de..f54d3f5b2e40 100644 --- a/fs/cachefiles/main.c +++ b/fs/cachefiles/main.c @@ -22,6 +22,7 @@ #include <linux/statfs.h> #include <linux/sysctl.h> #include <linux/miscdevice.h> +#define CREATE_TRACE_POINTS #include "internal.h" unsigned cachefiles_debug; diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index 3978b324cbca..ab0bbe93b398 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -30,11 +30,11 @@ */ static noinline void __cachefiles_printk_object(struct cachefiles_object *object, - const char *prefix, - u8 *keybuf) + const char *prefix) { struct fscache_cookie *cookie; - unsigned keylen, loop; + const u8 *k; + unsigned loop; pr_err("%sobject: OBJ%x\n", prefix, object->fscache.debug_id); pr_err("%sobjstate=%s fl=%lx wbusy=%x ev=%lx[%lx]\n", @@ -56,23 +56,16 @@ void __cachefiles_printk_object(struct cachefiles_object *object, object->fscache.cookie->parent, object->fscache.cookie->netfs_data, object->fscache.cookie->flags); - if (keybuf && cookie->def) - keylen = cookie->def->get_key(cookie->netfs_data, keybuf, - CACHEFILES_KEYBUF_SIZE); - else - keylen = 0; + pr_err("%skey=[%u] '", prefix, cookie->key_len); + k = (cookie->key_len <= sizeof(cookie->inline_key)) ? + cookie->inline_key : cookie->key; + for (loop = 0; loop < cookie->key_len; loop++) + pr_cont("%02x", k[loop]); + pr_cont("'\n"); } else { pr_err("%scookie=NULL\n", prefix); - keylen = 0; } spin_unlock(&object->fscache.lock); - - if (keylen) { - pr_err("%skey=[%u] '", prefix, keylen); - for (loop = 0; loop < keylen; loop++) - pr_cont("%02x", keybuf[loop]); - pr_cont("'\n"); - } } /* @@ -81,14 +74,10 @@ void __cachefiles_printk_object(struct cachefiles_object *object, static noinline void cachefiles_printk_object(struct cachefiles_object *object, struct cachefiles_object *xobject) { - u8 *keybuf; - - keybuf = kmalloc(CACHEFILES_KEYBUF_SIZE, GFP_NOIO); if (object) - __cachefiles_printk_object(object, "", keybuf); + __cachefiles_printk_object(object, ""); if (xobject) - __cachefiles_printk_object(xobject, "x", keybuf); - kfree(keybuf); + __cachefiles_printk_object(xobject, "x"); } /* @@ -120,6 +109,7 @@ static void cachefiles_mark_object_buried(struct cachefiles_cache *cache, } write_unlock(&cache->active_lock); + trace_cachefiles_mark_buried(NULL, dentry, why); _leave(" [no owner]"); return; @@ -130,6 +120,8 @@ found_dentry: object->fscache.state->name, dentry); + trace_cachefiles_mark_buried(object, dentry, why); + if (fscache_object_is_live(&object->fscache)) { pr_err("\n"); pr_err("Error: Can't preemptively bury live object\n"); @@ -158,13 +150,15 @@ static int cachefiles_mark_object_active(struct cachefiles_cache *cache, try_again: write_lock(&cache->active_lock); + dentry = object->dentry; + trace_cachefiles_mark_active(object, dentry); + if (test_and_set_bit(CACHEFILES_OBJECT_ACTIVE, &object->flags)) { pr_err("Error: Object already active\n"); cachefiles_printk_object(object, NULL); BUG(); } - dentry = object->dentry; _p = &cache->active_nodes.rb_node; while (*_p) { _parent = *_p; @@ -191,6 +185,8 @@ try_again: /* an old object from a previous incarnation is hogging the slot - we * need to wait for it to be destroyed */ wait_for_old_object: + trace_cachefiles_wait_active(object, dentry, xobject); + if (fscache_object_is_live(&xobject->fscache)) { pr_err("\n"); pr_err("Error: Unexpected object collision\n"); @@ -248,12 +244,12 @@ wait_for_old_object: ASSERT(!test_bit(CACHEFILES_OBJECT_ACTIVE, &xobject->flags)); - cache->cache.ops->put_object(&xobject->fscache); + cache->cache.ops->put_object(&xobject->fscache, cachefiles_obj_put_wait_retry); goto try_again; requeue: clear_bit(CACHEFILES_OBJECT_ACTIVE, &object->flags); - cache->cache.ops->put_object(&xobject->fscache); + cache->cache.ops->put_object(&xobject->fscache, cachefiles_obj_put_wait_timeo); _leave(" = -ETIMEDOUT"); return -ETIMEDOUT; } @@ -265,6 +261,11 @@ void cachefiles_mark_object_inactive(struct cachefiles_cache *cache, struct cachefiles_object *object, blkcnt_t i_blocks) { + struct dentry *dentry = object->dentry; + struct inode *inode = d_backing_inode(dentry); + + trace_cachefiles_mark_inactive(object, dentry, inode); + write_lock(&cache->active_lock); rb_erase(&object->active_node, &cache->active_nodes); clear_bit(CACHEFILES_OBJECT_ACTIVE, &object->flags); @@ -288,6 +289,7 @@ void cachefiles_mark_object_inactive(struct cachefiles_cache *cache, * - unlocks the directory mutex */ static int cachefiles_bury_object(struct cachefiles_cache *cache, + struct cachefiles_object *object, struct dentry *dir, struct dentry *rep, bool preemptive, @@ -312,6 +314,7 @@ static int cachefiles_bury_object(struct cachefiles_cache *cache, if (ret < 0) { cachefiles_io_error(cache, "Unlink security error"); } else { + trace_cachefiles_unlink(object, rep, why); ret = vfs_unlink(d_inode(dir), rep, NULL); if (preemptive) @@ -413,6 +416,7 @@ try_again: if (ret < 0) { cachefiles_io_error(cache, "Rename security error %d", ret); } else { + trace_cachefiles_rename(object, rep, grave, why); ret = vfs_rename(d_inode(dir), rep, d_inode(cache->graveyard), grave, NULL, 0); if (ret != 0 && ret != -ENOMEM) @@ -458,7 +462,7 @@ int cachefiles_delete_object(struct cachefiles_cache *cache, /* we need to check that our parent is _still_ our parent - it * may have been renamed */ if (dir == object->dentry->d_parent) { - ret = cachefiles_bury_object(cache, dir, + ret = cachefiles_bury_object(cache, object, dir, object->dentry, false, FSCACHE_OBJECT_WAS_RETIRED); } else { @@ -486,6 +490,7 @@ int cachefiles_walk_to_object(struct cachefiles_object *parent, { struct cachefiles_cache *cache; struct dentry *dir, *next = NULL; + struct inode *inode; struct path path; unsigned long start; const char *name; @@ -529,13 +534,17 @@ lookup_again: start = jiffies; next = lookup_one_len(name, dir, nlen); cachefiles_hist(cachefiles_lookup_histogram, start); - if (IS_ERR(next)) + if (IS_ERR(next)) { + trace_cachefiles_lookup(object, next, NULL); goto lookup_error; + } - _debug("next -> %p %s", next, d_backing_inode(next) ? "positive" : "negative"); + inode = d_backing_inode(next); + trace_cachefiles_lookup(object, next, inode); + _debug("next -> %p %s", next, inode ? "positive" : "negative"); if (!key) - object->new = !d_backing_inode(next); + object->new = !inode; /* if this element of the path doesn't exist, then the lookup phase * failed, and we can release any readers in the certain knowledge that @@ -558,9 +567,16 @@ lookup_again: start = jiffies; ret = vfs_mkdir(d_inode(dir), next, 0); cachefiles_hist(cachefiles_mkdir_histogram, start); + if (!key) + trace_cachefiles_mkdir(object, next, ret); if (ret < 0) goto create_error; + if (unlikely(d_unhashed(next))) { + dput(next); + inode_unlock(d_inode(dir)); + goto lookup_again; + } ASSERT(d_backing_inode(next)); _debug("mkdir -> %p{%p{ino=%lu}}", @@ -587,6 +603,7 @@ lookup_again: start = jiffies; ret = vfs_create(d_inode(dir), next, S_IFREG, true); cachefiles_hist(cachefiles_create_histogram, start); + trace_cachefiles_create(object, next, ret); if (ret < 0) goto create_error; @@ -629,7 +646,8 @@ lookup_again: * mutex) */ object->dentry = NULL; - ret = cachefiles_bury_object(cache, dir, next, true, + ret = cachefiles_bury_object(cache, object, dir, next, + true, FSCACHE_OBJECT_IS_STALE); dput(next); next = NULL; @@ -751,6 +769,7 @@ struct dentry *cachefiles_get_directory(struct cachefiles_cache *cache, /* search the current directory for the element name */ inode_lock(d_inode(dir)); +retry: start = jiffies; subdir = lookup_one_len(dirname, dir, strlen(dirname)); cachefiles_hist(cachefiles_lookup_histogram, start); @@ -780,6 +799,10 @@ struct dentry *cachefiles_get_directory(struct cachefiles_cache *cache, if (ret < 0) goto mkdir_error; + if (unlikely(d_unhashed(subdir))) { + dput(subdir); + goto retry; + } ASSERT(d_backing_inode(subdir)); _debug("mkdir -> %p{%p{ino=%lu}}", @@ -955,7 +978,7 @@ int cachefiles_cull(struct cachefiles_cache *cache, struct dentry *dir, /* actually remove the victim (drops the dir mutex) */ _debug("bury"); - ret = cachefiles_bury_object(cache, dir, victim, false, + ret = cachefiles_bury_object(cache, NULL, dir, victim, false, FSCACHE_OBJECT_WAS_CULLED); if (ret < 0) goto error; diff --git a/fs/cachefiles/proc.c b/fs/cachefiles/proc.c index 125b90f6c796..0ce1aa56b67f 100644 --- a/fs/cachefiles/proc.c +++ b/fs/cachefiles/proc.c @@ -85,21 +85,6 @@ static const struct seq_operations cachefiles_histogram_ops = { }; /* - * open "/proc/fs/cachefiles/XXX" which provide statistics summaries - */ -static int cachefiles_histogram_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &cachefiles_histogram_ops); -} - -static const struct file_operations cachefiles_histogram_fops = { - .open = cachefiles_histogram_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -/* * initialise the /proc/fs/cachefiles/ directory */ int __init cachefiles_proc_init(void) @@ -109,8 +94,8 @@ int __init cachefiles_proc_init(void) if (!proc_mkdir("fs/cachefiles", NULL)) goto error_dir; - if (!proc_create("fs/cachefiles/histogram", S_IFREG | 0444, NULL, - &cachefiles_histogram_fops)) + if (!proc_create_seq("fs/cachefiles/histogram", S_IFREG | 0444, NULL, + &cachefiles_histogram_ops)) goto error_histogram; _leave(" = 0"); diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c index 883bc7bb12c5..5082c8a49686 100644 --- a/fs/cachefiles/rdwr.c +++ b/fs/cachefiles/rdwr.c @@ -952,6 +952,7 @@ error: * - cache withdrawal is prevented by the caller */ void cachefiles_uncache_page(struct fscache_object *_object, struct page *page) + __releases(&object->fscache.cookie->lock) { struct cachefiles_object *object; struct cachefiles_cache *cache; diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c index d31c1a72d8a5..0a29a00aed2e 100644 --- a/fs/cachefiles/xattr.c +++ b/fs/cachefiles/xattr.c @@ -113,6 +113,7 @@ int cachefiles_set_object_xattr(struct cachefiles_object *object, /* attempt to install the cache metadata directly */ _debug("SET #%u", auxdata->len); + clear_bit(FSCACHE_COOKIE_AUX_UPDATED, &object->fscache.cookie->flags); ret = vfs_setxattr(dentry, cachefiles_xattr_cache, &auxdata->type, auxdata->len, XATTR_CREATE); @@ -141,6 +142,7 @@ int cachefiles_update_object_xattr(struct cachefiles_object *object, /* attempt to install the cache metadata directly */ _debug("SET #%u", auxdata->len); + clear_bit(FSCACHE_COOKIE_AUX_UPDATED, &object->fscache.cookie->flags); ret = vfs_setxattr(dentry, cachefiles_xattr_cache, &auxdata->type, auxdata->len, XATTR_REPLACE); @@ -180,7 +182,8 @@ int cachefiles_check_auxdata(struct cachefiles_object *object) goto error; xlen--; - validity = fscache_check_aux(&object->fscache, &auxbuf->data, xlen); + validity = fscache_check_aux(&object->fscache, &auxbuf->data, xlen, + i_size_read(d_backing_inode(dentry))); if (validity != FSCACHE_CHECKAUX_OKAY) goto error; @@ -249,7 +252,8 @@ int cachefiles_check_object_xattr(struct cachefiles_object *object, object->fscache.cookie->def->name, dlen); result = fscache_check_aux(&object->fscache, - &auxbuf->data, dlen); + &auxbuf->data, dlen, + i_size_read(d_backing_inode(dentry))); switch (result) { /* entry okay as is */ diff --git a/fs/ceph/Makefile b/fs/ceph/Makefile index 174f5709e508..a699e320393f 100644 --- a/fs/ceph/Makefile +++ b/fs/ceph/Makefile @@ -6,7 +6,7 @@ obj-$(CONFIG_CEPH_FS) += ceph.o ceph-y := super.o inode.o dir.o file.o locks.o addr.o ioctl.o \ - export.o caps.o snap.o xattr.o \ + export.o caps.o snap.o xattr.o quota.o \ mds_client.o mdsmap.o strings.o ceph_frag.o \ debugfs.o diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index b4336b42ce3b..5f7ad3d0df2e 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -15,6 +15,7 @@ #include "mds_client.h" #include "cache.h" #include <linux/ceph/osd_client.h> +#include <linux/ceph/striper.h> /* * Ceph address space ops. @@ -438,7 +439,7 @@ static int ceph_readpages(struct file *file, struct address_space *mapping, { struct inode *inode = file_inode(file); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); - struct ceph_file_info *ci = file->private_data; + struct ceph_file_info *fi = file->private_data; struct ceph_rw_context *rw_ctx; int rc = 0; int max = 0; @@ -452,7 +453,7 @@ static int ceph_readpages(struct file *file, struct address_space *mapping, if (rc == 0) goto out; - rw_ctx = ceph_find_rw_context(ci); + rw_ctx = ceph_find_rw_context(fi); max = fsc->mount_options->rsize >> PAGE_SHIFT; dout("readpages %p file %p ctx %p nr_pages %d max %d\n", inode, file, rw_ctx, nr_pages, max); @@ -800,7 +801,7 @@ static int ceph_writepages_start(struct address_space *mapping, struct ceph_osd_request *req = NULL; struct ceph_writeback_ctl ceph_wbc; bool should_loop, range_whole = false; - bool stop, done = false; + bool done = false; dout("writepages_start %p (mode=%s)\n", inode, wbc->sync_mode == WB_SYNC_NONE ? "NONE" : @@ -856,7 +857,7 @@ retry: * in that range can be associated with newer snapc. * They are not writeable until we write all dirty pages * associated with 'snapc' get written */ - if (index > 0 || wbc->sync_mode != WB_SYNC_NONE) + if (index > 0) should_loop = true; dout(" non-head snapc, range whole\n"); } @@ -864,8 +865,7 @@ retry: ceph_put_snap_context(last_snapc); last_snapc = snapc; - stop = false; - while (!stop && index <= end) { + while (!done && index <= end) { int num_ops = 0, op_idx; unsigned i, pvec_pages, max_pages, locked_pages = 0; struct page **pages = NULL, **data_pages; @@ -898,16 +898,30 @@ get_more_pages: unlock_page(page); continue; } - if (strip_unit_end && (page->index > strip_unit_end)) { - dout("end of strip unit %p\n", page); + /* only if matching snap context */ + pgsnapc = page_snap_context(page); + if (pgsnapc != snapc) { + dout("page snapc %p %lld != oldest %p %lld\n", + pgsnapc, pgsnapc->seq, snapc, snapc->seq); + if (!should_loop && + !ceph_wbc.head_snapc && + wbc->sync_mode != WB_SYNC_NONE) + should_loop = true; unlock_page(page); - break; + continue; } if (page_offset(page) >= ceph_wbc.i_size) { dout("%p page eof %llu\n", page, ceph_wbc.i_size); - /* not done if range_cyclic */ - stop = true; + if (ceph_wbc.size_stable || + page_offset(page) >= i_size_read(inode)) + mapping->a_ops->invalidatepage(page, + 0, PAGE_SIZE); + unlock_page(page); + continue; + } + if (strip_unit_end && (page->index > strip_unit_end)) { + dout("end of strip unit %p\n", page); unlock_page(page); break; } @@ -921,15 +935,6 @@ get_more_pages: wait_on_page_writeback(page); } - /* only if matching snap context */ - pgsnapc = page_snap_context(page); - if (pgsnapc != snapc) { - dout("page snapc %p %lld != oldest %p %lld\n", - pgsnapc, pgsnapc->seq, snapc, snapc->seq); - unlock_page(page); - continue; - } - if (!clear_page_dirty_for_io(page)) { dout("%p !clear_page_dirty_for_io\n", page); unlock_page(page); @@ -945,19 +950,15 @@ get_more_pages: if (locked_pages == 0) { u64 objnum; u64 objoff; + u32 xlen; /* prepare async write request */ offset = (u64)page_offset(page); - len = wsize; - - rc = ceph_calc_file_object_mapping(&ci->i_layout, - offset, len, - &objnum, &objoff, - &len); - if (rc < 0) { - unlock_page(page); - break; - } + ceph_calc_file_object_mapping(&ci->i_layout, + offset, wsize, + &objnum, &objoff, + &xlen); + len = xlen; num_ops = 1; strip_unit_end = page->index + @@ -1146,7 +1147,7 @@ new_request: * we tagged for writeback prior to entering this loop. */ if (wbc->nr_to_write <= 0 && wbc->sync_mode == WB_SYNC_NONE) - done = stop = true; + done = true; release_pvec_pages: dout("pagevec_release on %d pages (%p)\n", (int)pvec.nr, diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c index a3ab265d3215..bb524c880b1e 100644 --- a/fs/ceph/cache.c +++ b/fs/ceph/cache.c @@ -27,7 +27,6 @@ struct ceph_aux_inode { u64 version; struct timespec mtime; - loff_t size; }; struct fscache_netfs ceph_cache_netfs = { @@ -41,37 +40,18 @@ static LIST_HEAD(ceph_fscache_list); struct ceph_fscache_entry { struct list_head list; struct fscache_cookie *fscache; - struct ceph_fsid fsid; size_t uniq_len; + /* The following members must be last */ + struct ceph_fsid fsid; char uniquifier[0]; }; -static uint16_t ceph_fscache_session_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t maxbuf) -{ - const struct ceph_fs_client* fsc = cookie_netfs_data; - const char *fscache_uniq = fsc->mount_options->fscache_uniq; - uint16_t fsid_len, uniq_len; - - fsid_len = sizeof(fsc->client->fsid); - uniq_len = fscache_uniq ? strlen(fscache_uniq) : 0; - if (fsid_len + uniq_len > maxbuf) - return 0; - - memcpy(buffer, &fsc->client->fsid, fsid_len); - if (uniq_len) - memcpy(buffer + fsid_len, fscache_uniq, uniq_len); - - return fsid_len + uniq_len; -} - static const struct fscache_cookie_def ceph_fscache_fsid_object_def = { .name = "CEPH.fsid", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = ceph_fscache_session_get_key, }; -int ceph_fscache_register(void) +int __init ceph_fscache_register(void) { return fscache_register_netfs(&ceph_cache_netfs); } @@ -110,16 +90,19 @@ int ceph_fscache_register_fs(struct ceph_fs_client* fsc) goto out_unlock; } + memcpy(&ent->fsid, fsid, sizeof(*fsid)); + if (uniq_len > 0) { + memcpy(&ent->uniquifier, fscache_uniq, uniq_len); + ent->uniq_len = uniq_len; + } + fsc->fscache = fscache_acquire_cookie(ceph_cache_netfs.primary_index, &ceph_fscache_fsid_object_def, - fsc, true); + &ent->fsid, sizeof(ent->fsid) + uniq_len, + NULL, 0, + fsc, 0, true); if (fsc->fscache) { - memcpy(&ent->fsid, fsid, sizeof(*fsid)); - if (uniq_len > 0) { - memcpy(&ent->uniquifier, fscache_uniq, uniq_len); - ent->uniq_len = uniq_len; - } ent->fscache = fsc->fscache; list_add_tail(&ent->list, &ceph_fscache_list); } else { @@ -133,73 +116,32 @@ out_unlock: return err; } -static uint16_t ceph_fscache_inode_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t maxbuf) -{ - const struct ceph_inode_info* ci = cookie_netfs_data; - uint16_t klen; - - /* use ceph virtual inode (id + snapshot) */ - klen = sizeof(ci->i_vino); - if (klen > maxbuf) - return 0; - - memcpy(buffer, &ci->i_vino, klen); - return klen; -} - -static uint16_t ceph_fscache_inode_get_aux(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - struct ceph_aux_inode aux; - const struct ceph_inode_info* ci = cookie_netfs_data; - const struct inode* inode = &ci->vfs_inode; - - memset(&aux, 0, sizeof(aux)); - aux.version = ci->i_version; - aux.mtime = inode->i_mtime; - aux.size = i_size_read(inode); - - memcpy(buffer, &aux, sizeof(aux)); - - return sizeof(aux); -} - -static void ceph_fscache_inode_get_attr(const void *cookie_netfs_data, - uint64_t *size) -{ - const struct ceph_inode_info* ci = cookie_netfs_data; - *size = i_size_read(&ci->vfs_inode); -} - static enum fscache_checkaux ceph_fscache_inode_check_aux( - void *cookie_netfs_data, const void *data, uint16_t dlen) + void *cookie_netfs_data, const void *data, uint16_t dlen, + loff_t object_size) { struct ceph_aux_inode aux; struct ceph_inode_info* ci = cookie_netfs_data; struct inode* inode = &ci->vfs_inode; - if (dlen != sizeof(aux)) + if (dlen != sizeof(aux) || + i_size_read(inode) != object_size) return FSCACHE_CHECKAUX_OBSOLETE; memset(&aux, 0, sizeof(aux)); aux.version = ci->i_version; aux.mtime = inode->i_mtime; - aux.size = i_size_read(inode); if (memcmp(data, &aux, sizeof(aux)) != 0) return FSCACHE_CHECKAUX_OBSOLETE; - dout("ceph inode 0x%p cached okay", ci); + dout("ceph inode 0x%p cached okay\n", ci); return FSCACHE_CHECKAUX_OKAY; } static const struct fscache_cookie_def ceph_fscache_inode_object_def = { .name = "CEPH.inode", .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .get_key = ceph_fscache_inode_get_key, - .get_attr = ceph_fscache_inode_get_attr, - .get_aux = ceph_fscache_inode_get_aux, .check_aux = ceph_fscache_inode_check_aux, }; @@ -207,6 +149,7 @@ void ceph_fscache_register_inode_cookie(struct inode *inode) { struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); + struct ceph_aux_inode aux; /* No caching for filesystem */ if (!fsc->fscache) @@ -218,9 +161,14 @@ void ceph_fscache_register_inode_cookie(struct inode *inode) inode_lock_nested(inode, I_MUTEX_CHILD); if (!ci->fscache) { + memset(&aux, 0, sizeof(aux)); + aux.version = ci->i_version; + aux.mtime = inode->i_mtime; ci->fscache = fscache_acquire_cookie(fsc->fscache, - &ceph_fscache_inode_object_def, - ci, false); + &ceph_fscache_inode_object_def, + &ci->i_vino, sizeof(ci->i_vino), + &aux, sizeof(aux), + ci, i_size_read(inode), false); } inode_unlock(inode); } @@ -235,7 +183,7 @@ void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci) ci->fscache = NULL; fscache_uncache_all_inode_pages(cookie, &ci->vfs_inode); - fscache_relinquish_cookie(cookie, 0); + fscache_relinquish_cookie(cookie, &ci->i_vino, false); } static bool ceph_fscache_can_enable(void *data) @@ -254,11 +202,11 @@ void ceph_fscache_file_set_cookie(struct inode *inode, struct file *filp) if (inode_is_open_for_write(inode)) { dout("fscache_file_set_cookie %p %p disabling cache\n", inode, filp); - fscache_disable_cookie(ci->fscache, false); + fscache_disable_cookie(ci->fscache, &ci->i_vino, false); fscache_uncache_all_inode_pages(ci->fscache, inode); } else { - fscache_enable_cookie(ci->fscache, ceph_fscache_can_enable, - inode); + fscache_enable_cookie(ci->fscache, &ci->i_vino, i_size_read(inode), + ceph_fscache_can_enable, inode); if (fscache_cookie_enabled(ci->fscache)) { dout("fscache_file_set_cookie %p %p enabling cache\n", inode, filp); @@ -351,7 +299,8 @@ void ceph_readpage_to_fscache(struct inode *inode, struct page *page) if (!cache_valid(ci)) return; - ret = fscache_write_page(ci->fscache, page, GFP_KERNEL); + ret = fscache_write_page(ci->fscache, page, i_size_read(inode), + GFP_KERNEL); if (ret) fscache_uncache_page(ci->fscache, page); } @@ -385,7 +334,7 @@ void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc) WARN_ON_ONCE(!found); mutex_unlock(&ceph_fscache_lock); - __fscache_relinquish_cookie(fsc->fscache, 0); + __fscache_relinquish_cookie(fsc->fscache, NULL, false); } fsc->fscache = NULL; } @@ -402,7 +351,7 @@ void ceph_fscache_revalidate_cookie(struct ceph_inode_info *ci) * truncate while the caller holds CEPH_CAP_FILE_RD */ mutex_lock(&ci->i_truncate_mutex); if (!cache_valid(ci)) { - if (fscache_check_consistency(ci->fscache)) + if (fscache_check_consistency(ci->fscache, &ci->i_vino)) fscache_invalidate(ci->fscache); spin_lock(&ci->i_ceph_lock); ci->i_fscache_gen = ci->i_rdcache_gen; diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 0e5bd3e3344e..23dbfae16156 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -184,36 +184,54 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc, mdsc->caps_avail_count); spin_unlock(&mdsc->caps_list_lock); - for (i = have; i < need; i++) { -retry: + for (i = have; i < need; ) { cap = kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS); - if (!cap) { - if (!trimmed) { - for (j = 0; j < mdsc->max_sessions; j++) { - s = __ceph_lookup_mds_session(mdsc, j); - if (!s) - continue; - mutex_unlock(&mdsc->mutex); + if (cap) { + list_add(&cap->caps_item, &newcaps); + alloc++; + i++; + continue; + } - mutex_lock(&s->s_mutex); - max_caps = s->s_nr_caps - (need - i); - ceph_trim_caps(mdsc, s, max_caps); - mutex_unlock(&s->s_mutex); + if (!trimmed) { + for (j = 0; j < mdsc->max_sessions; j++) { + s = __ceph_lookup_mds_session(mdsc, j); + if (!s) + continue; + mutex_unlock(&mdsc->mutex); - ceph_put_mds_session(s); - mutex_lock(&mdsc->mutex); - } - trimmed = true; - goto retry; - } else { - pr_warn("reserve caps ctx=%p ENOMEM " - "need=%d got=%d\n", - ctx, need, have + alloc); - goto out_nomem; + mutex_lock(&s->s_mutex); + max_caps = s->s_nr_caps - (need - i); + ceph_trim_caps(mdsc, s, max_caps); + mutex_unlock(&s->s_mutex); + + ceph_put_mds_session(s); + mutex_lock(&mdsc->mutex); } + trimmed = true; + + spin_lock(&mdsc->caps_list_lock); + if (mdsc->caps_avail_count) { + int more_have; + if (mdsc->caps_avail_count >= need - i) + more_have = need - i; + else + more_have = mdsc->caps_avail_count; + + i += more_have; + have += more_have; + mdsc->caps_avail_count -= more_have; + mdsc->caps_reserve_count += more_have; + + } + spin_unlock(&mdsc->caps_list_lock); + + continue; } - list_add(&cap->caps_item, &newcaps); - alloc++; + + pr_warn("reserve caps ctx=%p ENOMEM need=%d got=%d\n", + ctx, need, have + alloc); + goto out_nomem; } BUG_ON(have + alloc != need); @@ -234,16 +252,28 @@ retry: return 0; out_nomem: + + spin_lock(&mdsc->caps_list_lock); + mdsc->caps_avail_count += have; + mdsc->caps_reserve_count -= have; + while (!list_empty(&newcaps)) { cap = list_first_entry(&newcaps, struct ceph_cap, caps_item); list_del(&cap->caps_item); - kmem_cache_free(ceph_cap_cachep, cap); + + /* Keep some preallocated caps around (ceph_min_count), to + * avoid lots of free/alloc churn. */ + if (mdsc->caps_avail_count >= + mdsc->caps_reserve_count + mdsc->caps_min_count) { + kmem_cache_free(ceph_cap_cachep, cap); + } else { + mdsc->caps_avail_count++; + mdsc->caps_total_count++; + list_add(&cap->caps_item, &mdsc->caps_list); + } } - spin_lock(&mdsc->caps_list_lock); - mdsc->caps_avail_count += have; - mdsc->caps_reserve_count -= have; BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count + mdsc->caps_reserve_count + mdsc->caps_avail_count); @@ -254,12 +284,26 @@ out_nomem: int ceph_unreserve_caps(struct ceph_mds_client *mdsc, struct ceph_cap_reservation *ctx) { + int i; + struct ceph_cap *cap; + dout("unreserve caps ctx=%p count=%d\n", ctx, ctx->count); if (ctx->count) { spin_lock(&mdsc->caps_list_lock); BUG_ON(mdsc->caps_reserve_count < ctx->count); mdsc->caps_reserve_count -= ctx->count; - mdsc->caps_avail_count += ctx->count; + if (mdsc->caps_avail_count >= + mdsc->caps_reserve_count + mdsc->caps_min_count) { + mdsc->caps_total_count -= ctx->count; + for (i = 0; i < ctx->count; i++) { + cap = list_first_entry(&mdsc->caps_list, + struct ceph_cap, caps_item); + list_del(&cap->caps_item); + kmem_cache_free(ceph_cap_cachep, cap); + } + } else { + mdsc->caps_avail_count += ctx->count; + } ctx->count = 0; dout("unreserve caps %d = %d used + %d resv + %d avail\n", mdsc->caps_total_count, mdsc->caps_use_count, @@ -285,7 +329,23 @@ struct ceph_cap *ceph_get_cap(struct ceph_mds_client *mdsc, mdsc->caps_use_count++; mdsc->caps_total_count++; spin_unlock(&mdsc->caps_list_lock); + } else { + spin_lock(&mdsc->caps_list_lock); + if (mdsc->caps_avail_count) { + BUG_ON(list_empty(&mdsc->caps_list)); + + mdsc->caps_avail_count--; + mdsc->caps_use_count++; + cap = list_first_entry(&mdsc->caps_list, + struct ceph_cap, caps_item); + list_del(&cap->caps_item); + + BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count + + mdsc->caps_reserve_count + mdsc->caps_avail_count); + } + spin_unlock(&mdsc->caps_list_lock); } + return cap; } @@ -341,6 +401,8 @@ void ceph_reservation_status(struct ceph_fs_client *fsc, { struct ceph_mds_client *mdsc = fsc->mdsc; + spin_lock(&mdsc->caps_list_lock); + if (total) *total = mdsc->caps_total_count; if (avail) @@ -351,6 +413,8 @@ void ceph_reservation_status(struct ceph_fs_client *fsc, *reserved = mdsc->caps_reserve_count; if (min) *min = mdsc->caps_min_count; + + spin_unlock(&mdsc->caps_list_lock); } /* @@ -639,9 +703,11 @@ void ceph_add_cap(struct inode *inode, } spin_lock(&realm->inodes_with_caps_lock); - ci->i_snap_realm = realm; list_add(&ci->i_snap_realm_item, &realm->inodes_with_caps); + ci->i_snap_realm = realm; + if (realm->ino == ci->i_vino.ino) + realm->inode = inode; spin_unlock(&realm->inodes_with_caps_lock); if (oldrealm) diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c index 644def813754..abdf98deeec4 100644 --- a/fs/ceph/debugfs.c +++ b/fs/ceph/debugfs.c @@ -260,7 +260,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc) goto out; fsc->debugfs_mdsmap = debugfs_create_file("mdsmap", - 0600, + 0400, fsc->client->debugfs_dir, fsc, &mdsmap_show_fops); @@ -268,7 +268,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc) goto out; fsc->debugfs_mds_sessions = debugfs_create_file("mds_sessions", - 0600, + 0400, fsc->client->debugfs_dir, fsc, &mds_sessions_show_fops); @@ -276,7 +276,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc) goto out; fsc->debugfs_mdsc = debugfs_create_file("mdsc", - 0600, + 0400, fsc->client->debugfs_dir, fsc, &mdsc_show_fops); @@ -292,7 +292,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc) goto out; fsc->debugfs_dentry_lru = debugfs_create_file("dentry_lru", - 0600, + 0400, fsc->client->debugfs_dir, fsc, &dentry_lru_show_fops); diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index f1d9c6cc0491..1a78dd6f8bf2 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -2,7 +2,6 @@ #include <linux/ceph/ceph_debug.h> #include <linux/spinlock.h> -#include <linux/fs_struct.h> #include <linux/namei.h> #include <linux/slab.h> #include <linux/sched.h> @@ -102,18 +101,18 @@ static int fpos_cmp(loff_t l, loff_t r) * regardless of what dir changes take place on the * server. */ -static int note_last_dentry(struct ceph_file_info *fi, const char *name, +static int note_last_dentry(struct ceph_dir_file_info *dfi, const char *name, int len, unsigned next_offset) { char *buf = kmalloc(len+1, GFP_KERNEL); if (!buf) return -ENOMEM; - kfree(fi->last_name); - fi->last_name = buf; - memcpy(fi->last_name, name, len); - fi->last_name[len] = 0; - fi->next_offset = next_offset; - dout("note_last_dentry '%s'\n", fi->last_name); + kfree(dfi->last_name); + dfi->last_name = buf; + memcpy(dfi->last_name, name, len); + dfi->last_name[len] = 0; + dfi->next_offset = next_offset; + dout("note_last_dentry '%s'\n", dfi->last_name); return 0; } @@ -175,7 +174,7 @@ __dcache_find_get_entry(struct dentry *parent, u64 idx, static int __dcache_readdir(struct file *file, struct dir_context *ctx, int shared_gen) { - struct ceph_file_info *fi = file->private_data; + struct ceph_dir_file_info *dfi = file->private_data; struct dentry *parent = file->f_path.dentry; struct inode *dir = d_inode(parent); struct dentry *dentry, *last = NULL; @@ -222,7 +221,7 @@ static int __dcache_readdir(struct file *file, struct dir_context *ctx, bool emit_dentry = false; dentry = __dcache_find_get_entry(parent, idx++, &cache_ctl); if (!dentry) { - fi->flags |= CEPH_F_ATEND; + dfi->file_info.flags |= CEPH_F_ATEND; err = 0; break; } @@ -273,33 +272,33 @@ out: if (last) { int ret; di = ceph_dentry(last); - ret = note_last_dentry(fi, last->d_name.name, last->d_name.len, + ret = note_last_dentry(dfi, last->d_name.name, last->d_name.len, fpos_off(di->offset) + 1); if (ret < 0) err = ret; dput(last); /* last_name no longer match cache index */ - if (fi->readdir_cache_idx >= 0) { - fi->readdir_cache_idx = -1; - fi->dir_release_count = 0; + if (dfi->readdir_cache_idx >= 0) { + dfi->readdir_cache_idx = -1; + dfi->dir_release_count = 0; } } return err; } -static bool need_send_readdir(struct ceph_file_info *fi, loff_t pos) +static bool need_send_readdir(struct ceph_dir_file_info *dfi, loff_t pos) { - if (!fi->last_readdir) + if (!dfi->last_readdir) return true; if (is_hash_order(pos)) - return !ceph_frag_contains_value(fi->frag, fpos_hash(pos)); + return !ceph_frag_contains_value(dfi->frag, fpos_hash(pos)); else - return fi->frag != fpos_frag(pos); + return dfi->frag != fpos_frag(pos); } static int ceph_readdir(struct file *file, struct dir_context *ctx) { - struct ceph_file_info *fi = file->private_data; + struct ceph_dir_file_info *dfi = file->private_data; struct inode *inode = file_inode(file); struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); @@ -310,7 +309,7 @@ static int ceph_readdir(struct file *file, struct dir_context *ctx) struct ceph_mds_reply_info_parsed *rinfo; dout("readdir %p file %p pos %llx\n", inode, file, ctx->pos); - if (fi->flags & CEPH_F_ATEND) + if (dfi->file_info.flags & CEPH_F_ATEND) return 0; /* always start with . and .. */ @@ -351,15 +350,15 @@ static int ceph_readdir(struct file *file, struct dir_context *ctx) /* proceed with a normal readdir */ more: /* do we have the correct frag content buffered? */ - if (need_send_readdir(fi, ctx->pos)) { + if (need_send_readdir(dfi, ctx->pos)) { struct ceph_mds_request *req; int op = ceph_snap(inode) == CEPH_SNAPDIR ? CEPH_MDS_OP_LSSNAP : CEPH_MDS_OP_READDIR; /* discard old result, if any */ - if (fi->last_readdir) { - ceph_mdsc_put_request(fi->last_readdir); - fi->last_readdir = NULL; + if (dfi->last_readdir) { + ceph_mdsc_put_request(dfi->last_readdir); + dfi->last_readdir = NULL; } if (is_hash_order(ctx->pos)) { @@ -373,7 +372,7 @@ more: } dout("readdir fetching %llx.%llx frag %x offset '%s'\n", - ceph_vinop(inode), frag, fi->last_name); + ceph_vinop(inode), frag, dfi->last_name); req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS); if (IS_ERR(req)) return PTR_ERR(req); @@ -389,8 +388,8 @@ more: __set_bit(CEPH_MDS_R_DIRECT_IS_HASH, &req->r_req_flags); req->r_inode_drop = CEPH_CAP_FILE_EXCL; } - if (fi->last_name) { - req->r_path2 = kstrdup(fi->last_name, GFP_KERNEL); + if (dfi->last_name) { + req->r_path2 = kstrdup(dfi->last_name, GFP_KERNEL); if (!req->r_path2) { ceph_mdsc_put_request(req); return -ENOMEM; @@ -400,10 +399,10 @@ more: cpu_to_le32(fpos_hash(ctx->pos)); } - req->r_dir_release_cnt = fi->dir_release_count; - req->r_dir_ordered_cnt = fi->dir_ordered_count; - req->r_readdir_cache_idx = fi->readdir_cache_idx; - req->r_readdir_offset = fi->next_offset; + req->r_dir_release_cnt = dfi->dir_release_count; + req->r_dir_ordered_cnt = dfi->dir_ordered_count; + req->r_readdir_cache_idx = dfi->readdir_cache_idx; + req->r_readdir_offset = dfi->next_offset; req->r_args.readdir.frag = cpu_to_le32(frag); req->r_args.readdir.flags = cpu_to_le16(CEPH_READDIR_REPLY_BITFLAGS); @@ -427,35 +426,35 @@ more: if (le32_to_cpu(rinfo->dir_dir->frag) != frag) { frag = le32_to_cpu(rinfo->dir_dir->frag); if (!rinfo->hash_order) { - fi->next_offset = req->r_readdir_offset; + dfi->next_offset = req->r_readdir_offset; /* adjust ctx->pos to beginning of frag */ ctx->pos = ceph_make_fpos(frag, - fi->next_offset, + dfi->next_offset, false); } } - fi->frag = frag; - fi->last_readdir = req; + dfi->frag = frag; + dfi->last_readdir = req; if (test_bit(CEPH_MDS_R_DID_PREPOPULATE, &req->r_req_flags)) { - fi->readdir_cache_idx = req->r_readdir_cache_idx; - if (fi->readdir_cache_idx < 0) { + dfi->readdir_cache_idx = req->r_readdir_cache_idx; + if (dfi->readdir_cache_idx < 0) { /* preclude from marking dir ordered */ - fi->dir_ordered_count = 0; + dfi->dir_ordered_count = 0; } else if (ceph_frag_is_leftmost(frag) && - fi->next_offset == 2) { + dfi->next_offset == 2) { /* note dir version at start of readdir so * we can tell if any dentries get dropped */ - fi->dir_release_count = req->r_dir_release_cnt; - fi->dir_ordered_count = req->r_dir_ordered_cnt; + dfi->dir_release_count = req->r_dir_release_cnt; + dfi->dir_ordered_count = req->r_dir_ordered_cnt; } } else { - dout("readdir !did_prepopulate"); + dout("readdir !did_prepopulate\n"); /* disable readdir cache */ - fi->readdir_cache_idx = -1; + dfi->readdir_cache_idx = -1; /* preclude from marking dir complete */ - fi->dir_release_count = 0; + dfi->dir_release_count = 0; } /* note next offset and last dentry name */ @@ -464,19 +463,19 @@ more: rinfo->dir_entries + (rinfo->dir_nr-1); unsigned next_offset = req->r_reply_info.dir_end ? 2 : (fpos_off(rde->offset) + 1); - err = note_last_dentry(fi, rde->name, rde->name_len, + err = note_last_dentry(dfi, rde->name, rde->name_len, next_offset); if (err) return err; } else if (req->r_reply_info.dir_end) { - fi->next_offset = 2; + dfi->next_offset = 2; /* keep last name */ } } - rinfo = &fi->last_readdir->r_reply_info; + rinfo = &dfi->last_readdir->r_reply_info; dout("readdir frag %x num %d pos %llx chunk first %llx\n", - fi->frag, rinfo->dir_nr, ctx->pos, + dfi->frag, rinfo->dir_nr, ctx->pos, rinfo->dir_nr ? rinfo->dir_entries[0].offset : 0LL); i = 0; @@ -520,52 +519,55 @@ more: ctx->pos++; } - ceph_mdsc_put_request(fi->last_readdir); - fi->last_readdir = NULL; + ceph_mdsc_put_request(dfi->last_readdir); + dfi->last_readdir = NULL; - if (fi->next_offset > 2) { - frag = fi->frag; + if (dfi->next_offset > 2) { + frag = dfi->frag; goto more; } /* more frags? */ - if (!ceph_frag_is_rightmost(fi->frag)) { - frag = ceph_frag_next(fi->frag); + if (!ceph_frag_is_rightmost(dfi->frag)) { + frag = ceph_frag_next(dfi->frag); if (is_hash_order(ctx->pos)) { loff_t new_pos = ceph_make_fpos(ceph_frag_value(frag), - fi->next_offset, true); + dfi->next_offset, true); if (new_pos > ctx->pos) ctx->pos = new_pos; /* keep last_name */ } else { - ctx->pos = ceph_make_fpos(frag, fi->next_offset, false); - kfree(fi->last_name); - fi->last_name = NULL; + ctx->pos = ceph_make_fpos(frag, dfi->next_offset, + false); + kfree(dfi->last_name); + dfi->last_name = NULL; } dout("readdir next frag is %x\n", frag); goto more; } - fi->flags |= CEPH_F_ATEND; + dfi->file_info.flags |= CEPH_F_ATEND; /* * if dir_release_count still matches the dir, no dentries * were released during the whole readdir, and we should have * the complete dir contents in our cache. */ - if (atomic64_read(&ci->i_release_count) == fi->dir_release_count) { + if (atomic64_read(&ci->i_release_count) == + dfi->dir_release_count) { spin_lock(&ci->i_ceph_lock); - if (fi->dir_ordered_count == atomic64_read(&ci->i_ordered_count)) { + if (dfi->dir_ordered_count == + atomic64_read(&ci->i_ordered_count)) { dout(" marking %p complete and ordered\n", inode); /* use i_size to track number of entries in * readdir cache */ - BUG_ON(fi->readdir_cache_idx < 0); - i_size_write(inode, fi->readdir_cache_idx * + BUG_ON(dfi->readdir_cache_idx < 0); + i_size_write(inode, dfi->readdir_cache_idx * sizeof(struct dentry*)); } else { dout(" marking %p complete\n", inode); } - __ceph_dir_set_complete(ci, fi->dir_release_count, - fi->dir_ordered_count); + __ceph_dir_set_complete(ci, dfi->dir_release_count, + dfi->dir_ordered_count); spin_unlock(&ci->i_ceph_lock); } @@ -573,25 +575,25 @@ more: return 0; } -static void reset_readdir(struct ceph_file_info *fi) +static void reset_readdir(struct ceph_dir_file_info *dfi) { - if (fi->last_readdir) { - ceph_mdsc_put_request(fi->last_readdir); - fi->last_readdir = NULL; + if (dfi->last_readdir) { + ceph_mdsc_put_request(dfi->last_readdir); + dfi->last_readdir = NULL; } - kfree(fi->last_name); - fi->last_name = NULL; - fi->dir_release_count = 0; - fi->readdir_cache_idx = -1; - fi->next_offset = 2; /* compensate for . and .. */ - fi->flags &= ~CEPH_F_ATEND; + kfree(dfi->last_name); + dfi->last_name = NULL; + dfi->dir_release_count = 0; + dfi->readdir_cache_idx = -1; + dfi->next_offset = 2; /* compensate for . and .. */ + dfi->file_info.flags &= ~CEPH_F_ATEND; } /* * discard buffered readdir content on seekdir(0), or seek to new frag, * or seek prior to current chunk */ -static bool need_reset_readdir(struct ceph_file_info *fi, loff_t new_pos) +static bool need_reset_readdir(struct ceph_dir_file_info *dfi, loff_t new_pos) { struct ceph_mds_reply_info_parsed *rinfo; loff_t chunk_offset; @@ -600,10 +602,10 @@ static bool need_reset_readdir(struct ceph_file_info *fi, loff_t new_pos) if (is_hash_order(new_pos)) { /* no need to reset last_name for a forward seek when * dentries are sotred in hash order */ - } else if (fi->frag != fpos_frag(new_pos)) { + } else if (dfi->frag != fpos_frag(new_pos)) { return true; } - rinfo = fi->last_readdir ? &fi->last_readdir->r_reply_info : NULL; + rinfo = dfi->last_readdir ? &dfi->last_readdir->r_reply_info : NULL; if (!rinfo || !rinfo->dir_nr) return true; chunk_offset = rinfo->dir_entries[0].offset; @@ -613,7 +615,7 @@ static bool need_reset_readdir(struct ceph_file_info *fi, loff_t new_pos) static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int whence) { - struct ceph_file_info *fi = file->private_data; + struct ceph_dir_file_info *dfi = file->private_data; struct inode *inode = file->f_mapping->host; loff_t retval; @@ -631,20 +633,20 @@ static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int whence) } if (offset >= 0) { - if (need_reset_readdir(fi, offset)) { + if (need_reset_readdir(dfi, offset)) { dout("dir_llseek dropping %p content\n", file); - reset_readdir(fi); + reset_readdir(dfi); } else if (is_hash_order(offset) && offset > file->f_pos) { /* for hash offset, we don't know if a forward seek * is within same frag */ - fi->dir_release_count = 0; - fi->readdir_cache_idx = -1; + dfi->dir_release_count = 0; + dfi->readdir_cache_idx = -1; } if (offset != file->f_pos) { file->f_pos = offset; file->f_version = 0; - fi->flags &= ~CEPH_F_ATEND; + dfi->file_info.flags &= ~CEPH_F_ATEND; } retval = offset; } @@ -825,6 +827,9 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry, if (ceph_snap(dir) != CEPH_NOSNAP) return -EROFS; + if (ceph_quota_is_max_files_exceeded(dir)) + return -EDQUOT; + err = ceph_pre_init_acls(dir, &mode, &acls); if (err < 0) return err; @@ -878,6 +883,9 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry, if (ceph_snap(dir) != CEPH_NOSNAP) return -EROFS; + if (ceph_quota_is_max_files_exceeded(dir)) + return -EDQUOT; + dout("symlink in dir %p dentry %p to '%s'\n", dir, dentry, dest); req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SYMLINK, USE_AUTH_MDS); if (IS_ERR(req)) { @@ -927,6 +935,12 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) goto out; } + if (op == CEPH_MDS_OP_MKDIR && + ceph_quota_is_max_files_exceeded(dir)) { + err = -EDQUOT; + goto out; + } + mode |= S_IFDIR; err = ceph_pre_init_acls(dir, &mode, &acls); if (err < 0) @@ -1066,6 +1080,11 @@ static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry, else return -EROFS; } + /* don't allow cross-quota renames */ + if ((old_dir != new_dir) && + (!ceph_quota_is_same_realm(old_dir, new_dir))) + return -EXDEV; + dout("rename dir %p dentry %p to dir %p dentry %p\n", old_dir, old_dentry, new_dir, new_dentry); req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS); @@ -1352,7 +1371,7 @@ static void ceph_d_prune(struct dentry *dentry) static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size, loff_t *ppos) { - struct ceph_file_info *cf = file->private_data; + struct ceph_dir_file_info *dfi = file->private_data; struct inode *inode = file_inode(file); struct ceph_inode_info *ci = ceph_inode(inode); int left; @@ -1361,12 +1380,12 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size, if (!ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT)) return -EISDIR; - if (!cf->dir_info) { - cf->dir_info = kmalloc(bufsize, GFP_KERNEL); - if (!cf->dir_info) + if (!dfi->dir_info) { + dfi->dir_info = kmalloc(bufsize, GFP_KERNEL); + if (!dfi->dir_info) return -ENOMEM; - cf->dir_info_len = - snprintf(cf->dir_info, bufsize, + dfi->dir_info_len = + snprintf(dfi->dir_info, bufsize, "entries: %20lld\n" " files: %20lld\n" " subdirs: %20lld\n" @@ -1386,10 +1405,10 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size, (long)ci->i_rctime.tv_nsec); } - if (*ppos >= cf->dir_info_len) + if (*ppos >= dfi->dir_info_len) return 0; - size = min_t(unsigned, size, cf->dir_info_len-*ppos); - left = copy_to_user(buf, cf->dir_info + *ppos, size); + size = min_t(unsigned, size, dfi->dir_info_len-*ppos); + left = copy_to_user(buf, dfi->dir_info + *ppos, size); if (left == size) return -EFAULT; *ppos += (size - left); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index b67eec3532a1..cf0e45b10121 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -30,6 +30,8 @@ static __le32 ceph_flags_sys2wire(u32 flags) break; } + flags &= ~O_ACCMODE; + #define ceph_sys2wire(a) if (flags & a) { wire_flags |= CEPH_##a; flags &= ~a; } ceph_sys2wire(O_CREAT); @@ -41,7 +43,7 @@ static __le32 ceph_flags_sys2wire(u32 flags) #undef ceph_sys2wire if (flags) - dout("unused open flags: %x", flags); + dout("unused open flags: %x\n", flags); return cpu_to_le32(wire_flags); } @@ -68,69 +70,104 @@ static __le32 ceph_flags_sys2wire(u32 flags) */ /* - * Calculate the length sum of direct io vectors that can - * be combined into one page vector. + * How many pages to get in one call to iov_iter_get_pages(). This + * determines the size of the on-stack array used as a buffer. */ -static size_t dio_get_pagev_size(const struct iov_iter *it) +#define ITER_GET_BVECS_PAGES 64 + +static ssize_t __iter_get_bvecs(struct iov_iter *iter, size_t maxsize, + struct bio_vec *bvecs) { - const struct iovec *iov = it->iov; - const struct iovec *iovend = iov + it->nr_segs; - size_t size; - - size = iov->iov_len - it->iov_offset; - /* - * An iov can be page vectored when both the current tail - * and the next base are page aligned. - */ - while (PAGE_ALIGNED((iov->iov_base + iov->iov_len)) && - (++iov < iovend && PAGE_ALIGNED((iov->iov_base)))) { - size += iov->iov_len; - } - dout("dio_get_pagevlen len = %zu\n", size); - return size; + size_t size = 0; + int bvec_idx = 0; + + if (maxsize > iov_iter_count(iter)) + maxsize = iov_iter_count(iter); + + while (size < maxsize) { + struct page *pages[ITER_GET_BVECS_PAGES]; + ssize_t bytes; + size_t start; + int idx = 0; + + bytes = iov_iter_get_pages(iter, pages, maxsize - size, + ITER_GET_BVECS_PAGES, &start); + if (bytes < 0) + return size ?: bytes; + + iov_iter_advance(iter, bytes); + size += bytes; + + for ( ; bytes; idx++, bvec_idx++) { + struct bio_vec bv = { + .bv_page = pages[idx], + .bv_len = min_t(int, bytes, PAGE_SIZE - start), + .bv_offset = start, + }; + + bvecs[bvec_idx] = bv; + bytes -= bv.bv_len; + start = 0; + } + } + + return size; } /* - * Allocate a page vector based on (@it, @nbytes). - * The return value is the tuple describing a page vector, - * that is (@pages, @page_align, @num_pages). + * iov_iter_get_pages() only considers one iov_iter segment, no matter + * what maxsize or maxpages are given. For ITER_BVEC that is a single + * page. + * + * Attempt to get up to @maxsize bytes worth of pages from @iter. + * Return the number of bytes in the created bio_vec array, or an error. */ -static struct page ** -dio_get_pages_alloc(const struct iov_iter *it, size_t nbytes, - size_t *page_align, int *num_pages) +static ssize_t iter_get_bvecs_alloc(struct iov_iter *iter, size_t maxsize, + struct bio_vec **bvecs, int *num_bvecs) { - struct iov_iter tmp_it = *it; - size_t align; - struct page **pages; - int ret = 0, idx, npages; + struct bio_vec *bv; + size_t orig_count = iov_iter_count(iter); + ssize_t bytes; + int npages; - align = (unsigned long)(it->iov->iov_base + it->iov_offset) & - (PAGE_SIZE - 1); - npages = calc_pages_for(align, nbytes); - pages = kvmalloc(sizeof(*pages) * npages, GFP_KERNEL); - if (!pages) - return ERR_PTR(-ENOMEM); + iov_iter_truncate(iter, maxsize); + npages = iov_iter_npages(iter, INT_MAX); + iov_iter_reexpand(iter, orig_count); - for (idx = 0; idx < npages; ) { - size_t start; - ret = iov_iter_get_pages(&tmp_it, pages + idx, nbytes, - npages - idx, &start); - if (ret < 0) - goto fail; + /* + * __iter_get_bvecs() may populate only part of the array -- zero it + * out. + */ + bv = kvmalloc_array(npages, sizeof(*bv), GFP_KERNEL | __GFP_ZERO); + if (!bv) + return -ENOMEM; - iov_iter_advance(&tmp_it, ret); - nbytes -= ret; - idx += (ret + start + PAGE_SIZE - 1) / PAGE_SIZE; + bytes = __iter_get_bvecs(iter, maxsize, bv); + if (bytes < 0) { + /* + * No pages were pinned -- just free the array. + */ + kvfree(bv); + return bytes; } - BUG_ON(nbytes != 0); - *num_pages = npages; - *page_align = align; - dout("dio_get_pages_alloc: got %d pages align %zu\n", npages, align); - return pages; -fail: - ceph_put_page_vector(pages, idx, false); - return ERR_PTR(ret); + *bvecs = bv; + *num_bvecs = npages; + return bytes; +} + +static void put_bvecs(struct bio_vec *bvecs, int num_bvecs, bool should_dirty) +{ + int i; + + for (i = 0; i < num_bvecs; i++) { + if (bvecs[i].bv_page) { + if (should_dirty) + set_page_dirty_lock(bvecs[i].bv_page); + put_page(bvecs[i].bv_page); + } + } + kvfree(bvecs); } /* @@ -159,13 +196,50 @@ out: return req; } +static int ceph_init_file_info(struct inode *inode, struct file *file, + int fmode, bool isdir) +{ + struct ceph_file_info *fi; + + dout("%s %p %p 0%o (%s)\n", __func__, inode, file, + inode->i_mode, isdir ? "dir" : "regular"); + BUG_ON(inode->i_fop->release != ceph_release); + + if (isdir) { + struct ceph_dir_file_info *dfi = + kmem_cache_zalloc(ceph_dir_file_cachep, GFP_KERNEL); + if (!dfi) { + ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */ + return -ENOMEM; + } + + file->private_data = dfi; + fi = &dfi->file_info; + dfi->next_offset = 2; + dfi->readdir_cache_idx = -1; + } else { + fi = kmem_cache_zalloc(ceph_file_cachep, GFP_KERNEL); + if (!fi) { + ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */ + return -ENOMEM; + } + + file->private_data = fi; + } + + fi->fmode = fmode; + spin_lock_init(&fi->rw_contexts_lock); + INIT_LIST_HEAD(&fi->rw_contexts); + + return 0; +} + /* * initialize private struct file data. * if we fail, clean up by dropping fmode reference on the ceph_inode */ static int ceph_init_file(struct inode *inode, struct file *file, int fmode) { - struct ceph_file_info *cf; int ret = 0; switch (inode->i_mode & S_IFMT) { @@ -173,22 +247,10 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode) ceph_fscache_register_inode_cookie(inode); ceph_fscache_file_set_cookie(inode, file); case S_IFDIR: - dout("init_file %p %p 0%o (regular)\n", inode, file, - inode->i_mode); - cf = kmem_cache_zalloc(ceph_file_cachep, GFP_KERNEL); - if (!cf) { - ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */ - return -ENOMEM; - } - cf->fmode = fmode; - - spin_lock_init(&cf->rw_contexts_lock); - INIT_LIST_HEAD(&cf->rw_contexts); - - cf->next_offset = 2; - cf->readdir_cache_idx = -1; - file->private_data = cf; - BUG_ON(inode->i_fop->release != ceph_release); + ret = ceph_init_file_info(inode, file, fmode, + S_ISDIR(inode->i_mode)); + if (ret) + return ret; break; case S_IFLNK: @@ -278,11 +340,11 @@ int ceph_open(struct inode *inode, struct file *file) struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; - struct ceph_file_info *cf = file->private_data; + struct ceph_file_info *fi = file->private_data; int err; int flags, fmode, wanted; - if (cf) { + if (fi) { dout("open file %p is already opened\n", file); return 0; } @@ -375,7 +437,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, struct ceph_mds_request *req; struct dentry *dn; struct ceph_acls_info acls = {}; - int mask; + int mask; int err; dout("atomic_open %p dentry %p '%pd' %s flags %d mode 0%o\n", @@ -386,6 +448,8 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, return -ENAMETOOLONG; if (flags & O_CREAT) { + if (ceph_quota_is_max_files_exceeded(dir)) + return -EDQUOT; err = ceph_pre_init_acls(dir, &mode, &acls); if (err < 0) return err; @@ -460,16 +524,27 @@ out_acl: int ceph_release(struct inode *inode, struct file *file) { struct ceph_inode_info *ci = ceph_inode(inode); - struct ceph_file_info *cf = file->private_data; - dout("release inode %p file %p\n", inode, file); - ceph_put_fmode(ci, cf->fmode); - if (cf->last_readdir) - ceph_mdsc_put_request(cf->last_readdir); - kfree(cf->last_name); - kfree(cf->dir_info); - WARN_ON(!list_empty(&cf->rw_contexts)); - kmem_cache_free(ceph_file_cachep, cf); + if (S_ISDIR(inode->i_mode)) { + struct ceph_dir_file_info *dfi = file->private_data; + dout("release inode %p dir file %p\n", inode, file); + WARN_ON(!list_empty(&dfi->file_info.rw_contexts)); + + ceph_put_fmode(ci, dfi->file_info.fmode); + + if (dfi->last_readdir) + ceph_mdsc_put_request(dfi->last_readdir); + kfree(dfi->last_name); + kfree(dfi->dir_info); + kmem_cache_free(ceph_dir_file_cachep, dfi); + } else { + struct ceph_file_info *fi = file->private_data; + dout("release inode %p regular file %p\n", inode, file); + WARN_ON(!list_empty(&fi->rw_contexts)); + + ceph_put_fmode(ci, fi->fmode); + kmem_cache_free(ceph_file_cachep, fi); + } /* wake up anyone waiting for caps on this inode */ wake_up_all(&ci->i_cap_wq); @@ -706,11 +781,12 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req) struct inode *inode = req->r_inode; struct ceph_aio_request *aio_req = req->r_priv; struct ceph_osd_data *osd_data = osd_req_op_extent_osd_data(req, 0); - int num_pages = calc_pages_for((u64)osd_data->alignment, - osd_data->length); - dout("ceph_aio_complete_req %p rc %d bytes %llu\n", - inode, rc, osd_data->length); + BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_BVECS); + BUG_ON(!osd_data->num_bvecs); + + dout("ceph_aio_complete_req %p rc %d bytes %u\n", + inode, rc, osd_data->bvec_pos.iter.bi_size); if (rc == -EOLDSNAPC) { struct ceph_aio_work *aio_work; @@ -728,9 +804,10 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req) } else if (!aio_req->write) { if (rc == -ENOENT) rc = 0; - if (rc >= 0 && osd_data->length > rc) { - int zoff = osd_data->alignment + rc; - int zlen = osd_data->length - rc; + if (rc >= 0 && osd_data->bvec_pos.iter.bi_size > rc) { + struct iov_iter i; + int zlen = osd_data->bvec_pos.iter.bi_size - rc; + /* * If read is satisfied by single OSD request, * it can pass EOF. Otherwise read is within @@ -745,13 +822,16 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req) aio_req->total_len = rc + zlen; } - if (zlen > 0) - ceph_zero_page_vector_range(zoff, zlen, - osd_data->pages); + iov_iter_bvec(&i, ITER_BVEC, osd_data->bvec_pos.bvecs, + osd_data->num_bvecs, + osd_data->bvec_pos.iter.bi_size); + iov_iter_advance(&i, rc); + iov_iter_zero(zlen, &i); } } - ceph_put_page_vector(osd_data->pages, num_pages, aio_req->should_dirty); + put_bvecs(osd_data->bvec_pos.bvecs, osd_data->num_bvecs, + aio_req->should_dirty); ceph_osdc_put_request(req); if (rc < 0) @@ -839,7 +919,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_vino vino; struct ceph_osd_request *req; - struct page **pages; + struct bio_vec *bvecs; struct ceph_aio_request *aio_req = NULL; int num_pages = 0; int flags; @@ -874,10 +954,14 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, } while (iov_iter_count(iter) > 0) { - u64 size = dio_get_pagev_size(iter); - size_t start = 0; + u64 size = iov_iter_count(iter); ssize_t len; + if (write) + size = min_t(u64, size, fsc->mount_options->wsize); + else + size = min_t(u64, size, fsc->mount_options->rsize); + vino = ceph_vino(inode); req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, vino, pos, &size, 0, @@ -893,18 +977,14 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, break; } - if (write) - size = min_t(u64, size, fsc->mount_options->wsize); - else - size = min_t(u64, size, fsc->mount_options->rsize); - - len = size; - pages = dio_get_pages_alloc(iter, len, &start, &num_pages); - if (IS_ERR(pages)) { + len = iter_get_bvecs_alloc(iter, size, &bvecs, &num_pages); + if (len < 0) { ceph_osdc_put_request(req); - ret = PTR_ERR(pages); + ret = len; break; } + if (len != size) + osd_req_op_extent_update(req, 0, len); /* * To simplify error handling, allow AIO when IO within i_size @@ -937,8 +1017,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, req->r_mtime = mtime; } - osd_req_op_extent_osd_data_pages(req, 0, pages, len, start, - false, false); + osd_req_op_extent_osd_data_bvecs(req, 0, bvecs, num_pages, len); if (aio_req) { aio_req->total_len += len; @@ -951,7 +1030,6 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, list_add_tail(&req->r_unsafe_item, &aio_req->osd_reqs); pos += len; - iov_iter_advance(iter, len); continue; } @@ -964,25 +1042,26 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, if (ret == -ENOENT) ret = 0; if (ret >= 0 && ret < len && pos + ret < size) { + struct iov_iter i; int zlen = min_t(size_t, len - ret, size - pos - ret); - ceph_zero_page_vector_range(start + ret, zlen, - pages); + + iov_iter_bvec(&i, ITER_BVEC, bvecs, num_pages, + len); + iov_iter_advance(&i, ret); + iov_iter_zero(zlen, &i); ret += zlen; } if (ret >= 0) len = ret; } - ceph_put_page_vector(pages, num_pages, should_dirty); - + put_bvecs(bvecs, num_pages, should_dirty); ceph_osdc_put_request(req); if (ret < 0) break; pos += len; - iov_iter_advance(iter, len); - if (!write && pos >= size) break; @@ -1338,6 +1417,11 @@ retry_snap: pos = iocb->ki_pos; count = iov_iter_count(from); + if (ceph_quota_is_max_bytes_exceeded(inode, pos + count)) { + err = -EDQUOT; + goto out; + } + err = file_remove_privs(file); if (err) goto out; @@ -1419,6 +1503,7 @@ retry_snap: if (written >= 0) { int dirty; + spin_lock(&ci->i_ceph_lock); ci->i_inline_version = CEPH_INLINE_NONE; dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR, @@ -1426,6 +1511,8 @@ retry_snap: spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); + if (ceph_quota_is_max_bytes_approaching(inode, iocb->ki_pos)) + ceph_check_caps(ci, CHECK_CAPS_NODELAY, NULL); } dout("aio_write %p %llx.%llx %llu~%u dropping cap refs on %s\n", @@ -1668,6 +1755,12 @@ static long ceph_fallocate(struct file *file, int mode, goto unlock; } + if (!(mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE)) && + ceph_quota_is_max_bytes_exceeded(inode, offset + length)) { + ret = -EDQUOT; + goto unlock; + } + if (ceph_osdmap_flag(osdc, CEPH_OSDMAP_FULL) && !(mode & FALLOC_FL_PUNCH_HOLE)) { ret = -ENOSPC; @@ -1716,6 +1809,9 @@ static long ceph_fallocate(struct file *file, int mode, spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); + if ((endoff > size) && + ceph_quota_is_max_bytes_approaching(inode, endoff)) + ceph_check_caps(ci, CHECK_CAPS_NODELAY, NULL); } ceph_put_cap_refs(ci, got); diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index c6ec5aa46100..ae056927080d 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -441,6 +441,9 @@ struct inode *ceph_alloc_inode(struct super_block *sb) atomic64_set(&ci->i_complete_seq[1], 0); ci->i_symlink = NULL; + ci->i_max_bytes = 0; + ci->i_max_files = 0; + memset(&ci->i_dir_layout, 0, sizeof(ci->i_dir_layout)); RCU_INIT_POINTER(ci->i_layout.pool_ns, NULL); @@ -536,6 +539,9 @@ void ceph_destroy_inode(struct inode *inode) ceph_queue_caps_release(inode); + if (__ceph_has_any_quota(ci)) + ceph_adjust_quota_realms_count(inode, false); + /* * we may still have a snap_realm reference if there are stray * caps in i_snap_caps. @@ -548,6 +554,9 @@ void ceph_destroy_inode(struct inode *inode) dout(" dropping residual ref to snap realm %p\n", realm); spin_lock(&realm->inodes_with_caps_lock); list_del_init(&ci->i_snap_realm_item); + ci->i_snap_realm = NULL; + if (realm->ino == ci->i_vino.ino) + realm->inode = NULL; spin_unlock(&realm->inodes_with_caps_lock); ceph_put_snap_realm(mdsc, realm); } @@ -660,13 +669,15 @@ void ceph_fill_file_time(struct inode *inode, int issued, CEPH_CAP_FILE_BUFFER| CEPH_CAP_AUTH_EXCL| CEPH_CAP_XATTR_EXCL)) { - if (timespec_compare(ctime, &inode->i_ctime) > 0) { + if (ci->i_version == 0 || + timespec_compare(ctime, &inode->i_ctime) > 0) { dout("ctime %ld.%09ld -> %ld.%09ld inc w/ cap\n", inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec, ctime->tv_sec, ctime->tv_nsec); inode->i_ctime = *ctime; } - if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) { + if (ci->i_version == 0 || + ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) { /* the MDS did a utimes() */ dout("mtime %ld.%09ld -> %ld.%09ld " "tw %d -> %d\n", @@ -786,10 +797,11 @@ static int fill_inode(struct inode *inode, struct page *locked_page, new_issued = ~issued & le32_to_cpu(info->cap.caps); /* update inode */ - ci->i_version = le64_to_cpu(info->version); inode->i_rdev = le32_to_cpu(info->rdev); inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1; + __ceph_update_quota(ci, iinfo->max_bytes, iinfo->max_files); + if ((new_version || (new_issued & CEPH_CAP_AUTH_SHARED)) && (issued & CEPH_CAP_AUTH_EXCL) == 0) { inode->i_mode = le32_to_cpu(info->mode); @@ -857,6 +869,9 @@ static int fill_inode(struct inode *inode, struct page *locked_page, xattr_blob = NULL; } + /* finally update i_version */ + ci->i_version = le64_to_cpu(info->version); + inode->i_mapping->a_ops = &ceph_aops; switch (inode->i_mode & S_IFMT) { @@ -1867,20 +1882,9 @@ retry: * possibly truncate them.. so write AND block! */ if (ci->i_wrbuffer_ref_head < ci->i_wrbuffer_ref) { - struct ceph_cap_snap *capsnap; - to = ci->i_truncate_size; - list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) { - // MDS should have revoked Frw caps - WARN_ON_ONCE(capsnap->writing); - if (capsnap->dirty_pages && capsnap->size > to) - to = capsnap->size; - } spin_unlock(&ci->i_ceph_lock); dout("__do_pending_vmtruncate %p flushing snaps first\n", inode); - - truncate_pagecache(inode, to); - filemap_write_and_wait_range(&inode->i_data, 0, inode->i_sb->s_maxbytes); goto retry; @@ -2152,6 +2156,10 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) if (err != 0) return err; + if ((attr->ia_valid & ATTR_SIZE) && + ceph_quota_is_max_bytes_exceeded(inode, attr->ia_size)) + return -EDQUOT; + err = __ceph_setattr(inode, attr); if (err >= 0 && (attr->ia_valid & ATTR_MODE)) diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c index 851aa69ec8f0..c90f03beb15d 100644 --- a/fs/ceph/ioctl.c +++ b/fs/ceph/ioctl.c @@ -5,7 +5,7 @@ #include "super.h" #include "mds_client.h" #include "ioctl.h" - +#include <linux/ceph/striper.h> /* * ioctls @@ -185,7 +185,7 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg) &ceph_sb_to_client(inode->i_sb)->client->osdc; struct ceph_object_locator oloc; CEPH_DEFINE_OID_ONSTACK(oid); - u64 len = 1, olen; + u32 xlen; u64 tmp; struct ceph_pg pgid; int r; @@ -195,13 +195,8 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg) return -EFAULT; down_read(&osdc->lock); - r = ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, len, - &dl.object_no, &dl.object_offset, - &olen); - if (r < 0) { - up_read(&osdc->lock); - return -EIO; - } + ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, 1, + &dl.object_no, &dl.object_offset, &xlen); dl.file_offset -= dl.object_offset; dl.object_size = ci->i_layout.object_size; dl.block_size = ci->i_layout.stripe_unit; diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index 9e66f69ee8a5..9dae2ec7e1fa 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -95,7 +95,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode, owner = secure_addr(fl->fl_owner); dout("ceph_lock_message: rule: %d, op: %d, owner: %llx, pid: %llu, " - "start: %llu, length: %llu, wait: %d, type: %d", (int)lock_type, + "start: %llu, length: %llu, wait: %d, type: %d\n", (int)lock_type, (int)operation, owner, (u64)fl->fl_pid, fl->fl_start, length, wait, fl->fl_type); @@ -132,7 +132,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode, } ceph_mdsc_put_request(req); dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, " - "length: %llu, wait: %d, type: %d, err code %d", (int)lock_type, + "length: %llu, wait: %d, type: %d, err code %d\n", (int)lock_type, (int)operation, (u64)fl->fl_pid, fl->fl_start, length, wait, fl->fl_type, err); return err; @@ -226,7 +226,7 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) if (__mandatory_lock(file->f_mapping->host) && fl->fl_type != F_UNLCK) return -ENOLCK; - dout("ceph_lock, fl_owner: %p", fl->fl_owner); + dout("ceph_lock, fl_owner: %p\n", fl->fl_owner); /* set wait bit as appropriate, then make command as Ceph expects it*/ if (IS_GETLK(cmd)) @@ -264,7 +264,7 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) err = ceph_lock_message(CEPH_LOCK_FCNTL, op, inode, lock_cmd, wait, fl); if (!err) { if (op == CEPH_MDS_OP_SETFILELOCK) { - dout("mds locked, locking locally"); + dout("mds locked, locking locally\n"); err = posix_lock_file(file, fl, NULL); if (err) { /* undo! This should only happen if @@ -272,7 +272,7 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) * deadlock. */ ceph_lock_message(CEPH_LOCK_FCNTL, op, inode, CEPH_LOCK_UNLOCK, 0, fl); - dout("got %d on posix_lock_file, undid lock", + dout("got %d on posix_lock_file, undid lock\n", err); } } @@ -294,7 +294,7 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) if (fl->fl_type & LOCK_MAND) return -EOPNOTSUPP; - dout("ceph_flock, fl_file: %p", fl->fl_file); + dout("ceph_flock, fl_file: %p\n", fl->fl_file); spin_lock(&ci->i_ceph_lock); if (ci->i_ceph_flags & CEPH_I_ERROR_FILELOCK) { @@ -329,7 +329,7 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK, inode, CEPH_LOCK_UNLOCK, 0, fl); - dout("got %d on locks_lock_file_wait, undid lock", err); + dout("got %d on locks_lock_file_wait, undid lock\n", err); } } return err; @@ -356,7 +356,7 @@ void ceph_count_locks(struct inode *inode, int *fcntl_count, int *flock_count) ++(*flock_count); spin_unlock(&ctx->flc_lock); } - dout("counted %d flock locks and %d fcntl locks", + dout("counted %d flock locks and %d fcntl locks\n", *flock_count, *fcntl_count); } @@ -384,7 +384,7 @@ static int lock_to_ceph_filelock(struct file_lock *lock, cephlock->type = CEPH_LOCK_UNLOCK; break; default: - dout("Have unknown lock type %d", lock->fl_type); + dout("Have unknown lock type %d\n", lock->fl_type); err = -EINVAL; } @@ -407,7 +407,7 @@ int ceph_encode_locks_to_buffer(struct inode *inode, int seen_flock = 0; int l = 0; - dout("encoding %d flock and %d fcntl locks", num_flock_locks, + dout("encoding %d flock and %d fcntl locks\n", num_flock_locks, num_fcntl_locks); if (!ctx) diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 2e8f90f96540..5ece2e6ad154 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -100,6 +100,26 @@ static int parse_reply_info_in(void **p, void *end, } else info->inline_version = CEPH_INLINE_NONE; + if (features & CEPH_FEATURE_MDS_QUOTA) { + u8 struct_v, struct_compat; + u32 struct_len; + + /* + * both struct_v and struct_compat are expected to be >= 1 + */ + ceph_decode_8_safe(p, end, struct_v, bad); + ceph_decode_8_safe(p, end, struct_compat, bad); + if (!struct_v || !struct_compat) + goto bad; + ceph_decode_32_safe(p, end, struct_len, bad); + ceph_decode_need(p, end, struct_len, bad); + ceph_decode_64_safe(p, end, info->max_bytes, bad); + ceph_decode_64_safe(p, end, info->max_files, bad); + } else { + info->max_bytes = 0; + info->max_files = 0; + } + info->pool_ns_len = 0; info->pool_ns_data = NULL; if (features & CEPH_FEATURE_FS_FILE_LAYOUT_V2) { @@ -384,7 +404,7 @@ static struct ceph_mds_session *get_session(struct ceph_mds_session *s) refcount_read(&s->s_ref)-1, refcount_read(&s->s_ref)); return s; } else { - dout("mdsc get_session %p 0 -- FAIL", s); + dout("mdsc get_session %p 0 -- FAIL\n", s); return NULL; } } @@ -419,9 +439,10 @@ struct ceph_mds_session *__ceph_lookup_mds_session(struct ceph_mds_client *mdsc, static bool __have_session(struct ceph_mds_client *mdsc, int mds) { - if (mds >= mdsc->max_sessions) + if (mds >= mdsc->max_sessions || !mdsc->sessions[mds]) return false; - return mdsc->sessions[mds]; + else + return true; } static int __verify_registered_session(struct ceph_mds_client *mdsc, @@ -448,6 +469,25 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc, s = kzalloc(sizeof(*s), GFP_NOFS); if (!s) return ERR_PTR(-ENOMEM); + + if (mds >= mdsc->max_sessions) { + int newmax = 1 << get_count_order(mds + 1); + struct ceph_mds_session **sa; + + dout("%s: realloc to %d\n", __func__, newmax); + sa = kcalloc(newmax, sizeof(void *), GFP_NOFS); + if (!sa) + goto fail_realloc; + if (mdsc->sessions) { + memcpy(sa, mdsc->sessions, + mdsc->max_sessions * sizeof(void *)); + kfree(mdsc->sessions); + } + mdsc->sessions = sa; + mdsc->max_sessions = newmax; + } + + dout("%s: mds%d\n", __func__, mds); s->s_mdsc = mdsc; s->s_mds = mds; s->s_state = CEPH_MDS_SESSION_NEW; @@ -476,23 +516,6 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc, INIT_LIST_HEAD(&s->s_cap_releases); INIT_LIST_HEAD(&s->s_cap_flushing); - dout("register_session mds%d\n", mds); - if (mds >= mdsc->max_sessions) { - int newmax = 1 << get_count_order(mds+1); - struct ceph_mds_session **sa; - - dout("register_session realloc to %d\n", newmax); - sa = kcalloc(newmax, sizeof(void *), GFP_NOFS); - if (!sa) - goto fail_realloc; - if (mdsc->sessions) { - memcpy(sa, mdsc->sessions, - mdsc->max_sessions * sizeof(void *)); - kfree(mdsc->sessions); - } - mdsc->sessions = sa; - mdsc->max_sessions = newmax; - } mdsc->sessions[mds] = s; atomic_inc(&mdsc->num_sessions); refcount_inc(&s->s_ref); /* one ref to sessions[], one to caller */ @@ -2531,10 +2554,10 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg) * Otherwise we just have to return an ESTALE */ if (result == -ESTALE) { - dout("got ESTALE on request %llu", req->r_tid); + dout("got ESTALE on request %llu\n", req->r_tid); req->r_resend_mds = -1; if (req->r_direct_mode != USE_AUTH_MDS) { - dout("not using auth, setting for that now"); + dout("not using auth, setting for that now\n"); req->r_direct_mode = USE_AUTH_MDS; __do_request(mdsc, req); mutex_unlock(&mdsc->mutex); @@ -2542,13 +2565,13 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg) } else { int mds = __choose_mds(mdsc, req); if (mds >= 0 && mds != req->r_session->s_mds) { - dout("but auth changed, so resending"); + dout("but auth changed, so resending\n"); __do_request(mdsc, req); mutex_unlock(&mdsc->mutex); goto out; } } - dout("have to return ESTALE on request %llu", req->r_tid); + dout("have to return ESTALE on request %llu\n", req->r_tid); } @@ -3470,13 +3493,12 @@ void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session, } /* - * drop all leases (and dentry refs) in preparation for umount + * lock unlock sessions, to wait ongoing session activities */ -static void drop_leases(struct ceph_mds_client *mdsc) +static void lock_unlock_sessions(struct ceph_mds_client *mdsc) { int i; - dout("drop_leases\n"); mutex_lock(&mdsc->mutex); for (i = 0; i < mdsc->max_sessions; i++) { struct ceph_mds_session *s = __ceph_lookup_mds_session(mdsc, i); @@ -3572,7 +3594,6 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) if (!mdsc) return -ENOMEM; mdsc->fsc = fsc; - fsc->mdsc = mdsc; mutex_init(&mdsc->mutex); mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS); if (!mdsc->mdsmap) { @@ -3580,6 +3601,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) return -ENOMEM; } + fsc->mdsc = mdsc; init_completion(&mdsc->safe_umount_waiters); init_waitqueue_head(&mdsc->session_close_wq); INIT_LIST_HEAD(&mdsc->waiting_for_map); @@ -3587,6 +3609,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) atomic_set(&mdsc->num_sessions, 0); mdsc->max_sessions = 0; mdsc->stopping = 0; + atomic64_set(&mdsc->quotarealms_count, 0); mdsc->last_snap_seq = 0; init_rwsem(&mdsc->snap_rwsem); mdsc->snap_realms = RB_ROOT; @@ -3660,7 +3683,7 @@ void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc) dout("pre_umount\n"); mdsc->stopping = 1; - drop_leases(mdsc); + lock_unlock_sessions(mdsc); ceph_flush_dirty_caps(mdsc); wait_requests(mdsc); @@ -3858,6 +3881,9 @@ void ceph_mdsc_destroy(struct ceph_fs_client *fsc) struct ceph_mds_client *mdsc = fsc->mdsc; dout("mdsc_destroy %p\n", mdsc); + if (!mdsc) + return; + /* flush out any connection work with references to us */ ceph_msgr_flush(); @@ -4077,6 +4103,9 @@ static void dispatch(struct ceph_connection *con, struct ceph_msg *msg) case CEPH_MSG_CLIENT_LEASE: handle_lease(mdsc, s, msg); break; + case CEPH_MSG_CLIENT_QUOTA: + ceph_handle_quota(mdsc, s, msg); + break; default: pr_err("received unknown message type %d %s\n", type, diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 71e3b783ee6f..2ec3b5b35067 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -49,6 +49,8 @@ struct ceph_mds_reply_info_in { char *inline_data; u32 pool_ns_len; char *pool_ns_data; + u64 max_bytes; + u64 max_files; }; struct ceph_mds_reply_dir_entry { @@ -312,6 +314,8 @@ struct ceph_mds_client { int max_sessions; /* len of s_mds_sessions */ int stopping; /* true if shutting down */ + atomic64_t quotarealms_count; /* # realms with quota */ + /* * snap_rwsem will cover cap linkage into snaprealms, and * realm snap contexts. (later, we can do per-realm snap diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c new file mode 100644 index 000000000000..242bfa5c0539 --- /dev/null +++ b/fs/ceph/quota.c @@ -0,0 +1,361 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * quota.c - CephFS quota + * + * Copyright (C) 2017-2018 SUSE + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/statfs.h> + +#include "super.h" +#include "mds_client.h" + +void ceph_adjust_quota_realms_count(struct inode *inode, bool inc) +{ + struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; + if (inc) + atomic64_inc(&mdsc->quotarealms_count); + else + atomic64_dec(&mdsc->quotarealms_count); +} + +static inline bool ceph_has_realms_with_quotas(struct inode *inode) +{ + struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; + return atomic64_read(&mdsc->quotarealms_count) > 0; +} + +void ceph_handle_quota(struct ceph_mds_client *mdsc, + struct ceph_mds_session *session, + struct ceph_msg *msg) +{ + struct super_block *sb = mdsc->fsc->sb; + struct ceph_mds_quota *h = msg->front.iov_base; + struct ceph_vino vino; + struct inode *inode; + struct ceph_inode_info *ci; + + if (msg->front.iov_len != sizeof(*h)) { + pr_err("%s corrupt message mds%d len %d\n", __func__, + session->s_mds, (int)msg->front.iov_len); + ceph_msg_dump(msg); + return; + } + + /* increment msg sequence number */ + mutex_lock(&session->s_mutex); + session->s_seq++; + mutex_unlock(&session->s_mutex); + + /* lookup inode */ + vino.ino = le64_to_cpu(h->ino); + vino.snap = CEPH_NOSNAP; + inode = ceph_find_inode(sb, vino); + if (!inode) { + pr_warn("Failed to find inode %llu\n", vino.ino); + return; + } + ci = ceph_inode(inode); + + spin_lock(&ci->i_ceph_lock); + ci->i_rbytes = le64_to_cpu(h->rbytes); + ci->i_rfiles = le64_to_cpu(h->rfiles); + ci->i_rsubdirs = le64_to_cpu(h->rsubdirs); + __ceph_update_quota(ci, le64_to_cpu(h->max_bytes), + le64_to_cpu(h->max_files)); + spin_unlock(&ci->i_ceph_lock); + + iput(inode); +} + +/* + * This function walks through the snaprealm for an inode and returns the + * ceph_snap_realm for the first snaprealm that has quotas set (either max_files + * or max_bytes). If the root is reached, return the root ceph_snap_realm + * instead. + * + * Note that the caller is responsible for calling ceph_put_snap_realm() on the + * returned realm. + */ +static struct ceph_snap_realm *get_quota_realm(struct ceph_mds_client *mdsc, + struct inode *inode) +{ + struct ceph_inode_info *ci = NULL; + struct ceph_snap_realm *realm, *next; + struct inode *in; + bool has_quota; + + if (ceph_snap(inode) != CEPH_NOSNAP) + return NULL; + + realm = ceph_inode(inode)->i_snap_realm; + if (realm) + ceph_get_snap_realm(mdsc, realm); + else + pr_err_ratelimited("get_quota_realm: ino (%llx.%llx) " + "null i_snap_realm\n", ceph_vinop(inode)); + while (realm) { + spin_lock(&realm->inodes_with_caps_lock); + in = realm->inode ? igrab(realm->inode) : NULL; + spin_unlock(&realm->inodes_with_caps_lock); + if (!in) + break; + + ci = ceph_inode(in); + has_quota = __ceph_has_any_quota(ci); + iput(in); + + next = realm->parent; + if (has_quota || !next) + return realm; + + ceph_get_snap_realm(mdsc, next); + ceph_put_snap_realm(mdsc, realm); + realm = next; + } + if (realm) + ceph_put_snap_realm(mdsc, realm); + + return NULL; +} + +bool ceph_quota_is_same_realm(struct inode *old, struct inode *new) +{ + struct ceph_mds_client *mdsc = ceph_inode_to_client(old)->mdsc; + struct ceph_snap_realm *old_realm, *new_realm; + bool is_same; + + down_read(&mdsc->snap_rwsem); + old_realm = get_quota_realm(mdsc, old); + new_realm = get_quota_realm(mdsc, new); + is_same = (old_realm == new_realm); + up_read(&mdsc->snap_rwsem); + + if (old_realm) + ceph_put_snap_realm(mdsc, old_realm); + if (new_realm) + ceph_put_snap_realm(mdsc, new_realm); + + return is_same; +} + +enum quota_check_op { + QUOTA_CHECK_MAX_FILES_OP, /* check quota max_files limit */ + QUOTA_CHECK_MAX_BYTES_OP, /* check quota max_files limit */ + QUOTA_CHECK_MAX_BYTES_APPROACHING_OP /* check if quota max_files + limit is approaching */ +}; + +/* + * check_quota_exceeded() will walk up the snaprealm hierarchy and, for each + * realm, it will execute quota check operation defined by the 'op' parameter. + * The snaprealm walk is interrupted if the quota check detects that the quota + * is exceeded or if the root inode is reached. + */ +static bool check_quota_exceeded(struct inode *inode, enum quota_check_op op, + loff_t delta) +{ + struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; + struct ceph_inode_info *ci; + struct ceph_snap_realm *realm, *next; + struct inode *in; + u64 max, rvalue; + bool exceeded = false; + + if (ceph_snap(inode) != CEPH_NOSNAP) + return false; + + down_read(&mdsc->snap_rwsem); + realm = ceph_inode(inode)->i_snap_realm; + if (realm) + ceph_get_snap_realm(mdsc, realm); + else + pr_err_ratelimited("check_quota_exceeded: ino (%llx.%llx) " + "null i_snap_realm\n", ceph_vinop(inode)); + while (realm) { + spin_lock(&realm->inodes_with_caps_lock); + in = realm->inode ? igrab(realm->inode) : NULL; + spin_unlock(&realm->inodes_with_caps_lock); + if (!in) + break; + + ci = ceph_inode(in); + spin_lock(&ci->i_ceph_lock); + if (op == QUOTA_CHECK_MAX_FILES_OP) { + max = ci->i_max_files; + rvalue = ci->i_rfiles + ci->i_rsubdirs; + } else { + max = ci->i_max_bytes; + rvalue = ci->i_rbytes; + } + spin_unlock(&ci->i_ceph_lock); + switch (op) { + case QUOTA_CHECK_MAX_FILES_OP: + exceeded = (max && (rvalue >= max)); + break; + case QUOTA_CHECK_MAX_BYTES_OP: + exceeded = (max && (rvalue + delta > max)); + break; + case QUOTA_CHECK_MAX_BYTES_APPROACHING_OP: + if (max) { + if (rvalue >= max) + exceeded = true; + else { + /* + * when we're writing more that 1/16th + * of the available space + */ + exceeded = + (((max - rvalue) >> 4) < delta); + } + } + break; + default: + /* Shouldn't happen */ + pr_warn("Invalid quota check op (%d)\n", op); + exceeded = true; /* Just break the loop */ + } + iput(in); + + next = realm->parent; + if (exceeded || !next) + break; + ceph_get_snap_realm(mdsc, next); + ceph_put_snap_realm(mdsc, realm); + realm = next; + } + ceph_put_snap_realm(mdsc, realm); + up_read(&mdsc->snap_rwsem); + + return exceeded; +} + +/* + * ceph_quota_is_max_files_exceeded - check if we can create a new file + * @inode: directory where a new file is being created + * + * This functions returns true is max_files quota allows a new file to be + * created. It is necessary to walk through the snaprealm hierarchy (until the + * FS root) to check all realms with quotas set. + */ +bool ceph_quota_is_max_files_exceeded(struct inode *inode) +{ + if (!ceph_has_realms_with_quotas(inode)) + return false; + + WARN_ON(!S_ISDIR(inode->i_mode)); + + return check_quota_exceeded(inode, QUOTA_CHECK_MAX_FILES_OP, 0); +} + +/* + * ceph_quota_is_max_bytes_exceeded - check if we can write to a file + * @inode: inode being written + * @newsize: new size if write succeeds + * + * This functions returns true is max_bytes quota allows a file size to reach + * @newsize; it returns false otherwise. + */ +bool ceph_quota_is_max_bytes_exceeded(struct inode *inode, loff_t newsize) +{ + loff_t size = i_size_read(inode); + + if (!ceph_has_realms_with_quotas(inode)) + return false; + + /* return immediately if we're decreasing file size */ + if (newsize <= size) + return false; + + return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_OP, (newsize - size)); +} + +/* + * ceph_quota_is_max_bytes_approaching - check if we're reaching max_bytes + * @inode: inode being written + * @newsize: new size if write succeeds + * + * This function returns true if the new file size @newsize will be consuming + * more than 1/16th of the available quota space; it returns false otherwise. + */ +bool ceph_quota_is_max_bytes_approaching(struct inode *inode, loff_t newsize) +{ + loff_t size = ceph_inode(inode)->i_reported_size; + + if (!ceph_has_realms_with_quotas(inode)) + return false; + + /* return immediately if we're decreasing file size */ + if (newsize <= size) + return false; + + return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_APPROACHING_OP, + (newsize - size)); +} + +/* + * ceph_quota_update_statfs - if root has quota update statfs with quota status + * @fsc: filesystem client instance + * @buf: statfs to update + * + * If the mounted filesystem root has max_bytes quota set, update the filesystem + * statistics with the quota status. + * + * This function returns true if the stats have been updated, false otherwise. + */ +bool ceph_quota_update_statfs(struct ceph_fs_client *fsc, struct kstatfs *buf) +{ + struct ceph_mds_client *mdsc = fsc->mdsc; + struct ceph_inode_info *ci; + struct ceph_snap_realm *realm; + struct inode *in; + u64 total = 0, used, free; + bool is_updated = false; + + down_read(&mdsc->snap_rwsem); + realm = get_quota_realm(mdsc, d_inode(fsc->sb->s_root)); + up_read(&mdsc->snap_rwsem); + if (!realm) + return false; + + spin_lock(&realm->inodes_with_caps_lock); + in = realm->inode ? igrab(realm->inode) : NULL; + spin_unlock(&realm->inodes_with_caps_lock); + if (in) { + ci = ceph_inode(in); + spin_lock(&ci->i_ceph_lock); + if (ci->i_max_bytes) { + total = ci->i_max_bytes >> CEPH_BLOCK_SHIFT; + used = ci->i_rbytes >> CEPH_BLOCK_SHIFT; + /* It is possible for a quota to be exceeded. + * Report 'zero' in that case + */ + free = total > used ? total - used : 0; + } + spin_unlock(&ci->i_ceph_lock); + if (total) { + buf->f_blocks = total; + buf->f_bfree = free; + buf->f_bavail = free; + is_updated = true; + } + iput(in); + } + ceph_put_snap_realm(mdsc, realm); + + return is_updated; +} + diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index 07cf95e6413d..041c27ea8de1 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -931,6 +931,8 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, list_add(&ci->i_snap_realm_item, &realm->inodes_with_caps); ci->i_snap_realm = realm; + if (realm->ino == ci->i_vino.ino) + realm->inode = inode; spin_unlock(&realm->inodes_with_caps_lock); spin_unlock(&ci->i_ceph_lock); diff --git a/fs/ceph/super.c b/fs/ceph/super.c index fb2bc9c15a23..b33082e6878f 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -76,9 +76,18 @@ static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf) */ buf->f_bsize = 1 << CEPH_BLOCK_SHIFT; buf->f_frsize = 1 << CEPH_BLOCK_SHIFT; - buf->f_blocks = le64_to_cpu(st.kb) >> (CEPH_BLOCK_SHIFT-10); - buf->f_bfree = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10); - buf->f_bavail = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10); + + /* + * By default use root quota for stats; fallback to overall filesystem + * usage if using 'noquotadf' mount option or if the root dir doesn't + * have max_bytes quota set. + */ + if (ceph_test_mount_opt(fsc, NOQUOTADF) || + !ceph_quota_update_statfs(fsc, buf)) { + buf->f_blocks = le64_to_cpu(st.kb) >> (CEPH_BLOCK_SHIFT-10); + buf->f_bfree = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10); + buf->f_bavail = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10); + } buf->f_files = le64_to_cpu(st.num_objects); buf->f_ffree = -1; @@ -151,6 +160,8 @@ enum { Opt_acl, #endif Opt_noacl, + Opt_quotadf, + Opt_noquotadf, }; static match_table_t fsopt_tokens = { @@ -187,6 +198,8 @@ static match_table_t fsopt_tokens = { {Opt_acl, "acl"}, #endif {Opt_noacl, "noacl"}, + {Opt_quotadf, "quotadf"}, + {Opt_noquotadf, "noquotadf"}, {-1, NULL} }; @@ -314,13 +327,16 @@ static int parse_fsopt_token(char *c, void *private) break; case Opt_fscache: fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE; + kfree(fsopt->fscache_uniq); + fsopt->fscache_uniq = NULL; break; case Opt_nofscache: fsopt->flags &= ~CEPH_MOUNT_OPT_FSCACHE; + kfree(fsopt->fscache_uniq); + fsopt->fscache_uniq = NULL; break; case Opt_poolperm: fsopt->flags &= ~CEPH_MOUNT_OPT_NOPOOLPERM; - printk ("pool perm"); break; case Opt_nopoolperm: fsopt->flags |= CEPH_MOUNT_OPT_NOPOOLPERM; @@ -331,6 +347,12 @@ static int parse_fsopt_token(char *c, void *private) case Opt_norequire_active_mds: fsopt->flags |= CEPH_MOUNT_OPT_MOUNTWAIT; break; + case Opt_quotadf: + fsopt->flags &= ~CEPH_MOUNT_OPT_NOQUOTADF; + break; + case Opt_noquotadf: + fsopt->flags |= CEPH_MOUNT_OPT_NOQUOTADF; + break; #ifdef CONFIG_CEPH_FS_POSIX_ACL case Opt_acl: fsopt->sb_flags |= SB_POSIXACL; @@ -513,13 +535,12 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root) if ((fsopt->flags & CEPH_MOUNT_OPT_DCACHE) == 0) seq_puts(m, ",nodcache"); if (fsopt->flags & CEPH_MOUNT_OPT_FSCACHE) { - if (fsopt->fscache_uniq) - seq_printf(m, ",fsc=%s", fsopt->fscache_uniq); - else - seq_puts(m, ",fsc"); + seq_show_option(m, "fsc", fsopt->fscache_uniq); } if (fsopt->flags & CEPH_MOUNT_OPT_NOPOOLPERM) seq_puts(m, ",nopoolperm"); + if (fsopt->flags & CEPH_MOUNT_OPT_NOQUOTADF) + seq_puts(m, ",noquotadf"); #ifdef CONFIG_CEPH_FS_POSIX_ACL if (fsopt->sb_flags & SB_POSIXACL) @@ -529,7 +550,7 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root) #endif if (fsopt->mds_namespace) - seq_printf(m, ",mds_namespace=%s", fsopt->mds_namespace); + seq_show_option(m, "mds_namespace", fsopt->mds_namespace); if (fsopt->wsize) seq_printf(m, ",wsize=%d", fsopt->wsize); if (fsopt->rsize != CEPH_MAX_READ_SIZE) @@ -679,6 +700,7 @@ struct kmem_cache *ceph_cap_cachep; struct kmem_cache *ceph_cap_flush_cachep; struct kmem_cache *ceph_dentry_cachep; struct kmem_cache *ceph_file_cachep; +struct kmem_cache *ceph_dir_file_cachep; static void ceph_inode_init_once(void *foo) { @@ -698,8 +720,7 @@ static int __init init_caches(void) if (!ceph_inode_cachep) return -ENOMEM; - ceph_cap_cachep = KMEM_CACHE(ceph_cap, - SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD); + ceph_cap_cachep = KMEM_CACHE(ceph_cap, SLAB_MEM_SPREAD); if (!ceph_cap_cachep) goto bad_cap; ceph_cap_flush_cachep = KMEM_CACHE(ceph_cap_flush, @@ -716,6 +737,10 @@ static int __init init_caches(void) if (!ceph_file_cachep) goto bad_file; + ceph_dir_file_cachep = KMEM_CACHE(ceph_dir_file_info, SLAB_MEM_SPREAD); + if (!ceph_dir_file_cachep) + goto bad_dir_file; + error = ceph_fscache_register(); if (error) goto bad_fscache; @@ -723,6 +748,8 @@ static int __init init_caches(void) return 0; bad_fscache: + kmem_cache_destroy(ceph_dir_file_cachep); +bad_dir_file: kmem_cache_destroy(ceph_file_cachep); bad_file: kmem_cache_destroy(ceph_dentry_cachep); @@ -748,6 +775,7 @@ static void destroy_caches(void) kmem_cache_destroy(ceph_cap_flush_cachep); kmem_cache_destroy(ceph_dentry_cachep); kmem_cache_destroy(ceph_file_cachep); + kmem_cache_destroy(ceph_dir_file_cachep); ceph_fscache_unregister(); } diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 1c2086e0fec2..a7077a0c989f 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -39,6 +39,7 @@ #define CEPH_MOUNT_OPT_FSCACHE (1<<10) /* use fscache */ #define CEPH_MOUNT_OPT_NOPOOLPERM (1<<11) /* no pool permission check */ #define CEPH_MOUNT_OPT_MOUNTWAIT (1<<12) /* mount waits if no mds is up */ +#define CEPH_MOUNT_OPT_NOQUOTADF (1<<13) /* no root dir quota in statfs */ #define CEPH_MOUNT_OPT_DEFAULT CEPH_MOUNT_OPT_DCACHE @@ -310,6 +311,9 @@ struct ceph_inode_info { u64 i_rbytes, i_rfiles, i_rsubdirs; u64 i_files, i_subdirs; + /* quotas */ + u64 i_max_bytes, i_max_files; + struct rb_root i_fragtree; int i_fragtree_nsplits; struct mutex i_fragtree_mutex; @@ -671,6 +675,10 @@ struct ceph_file_info { spinlock_t rw_contexts_lock; struct list_head rw_contexts; +}; + +struct ceph_dir_file_info { + struct ceph_file_info file_info; /* readdir: position within the dir */ u32 frag; @@ -748,6 +756,7 @@ struct ceph_readdir_cache_control { */ struct ceph_snap_realm { u64 ino; + struct inode *inode; atomic_t nref; struct rb_node node; @@ -1066,4 +1075,37 @@ extern int ceph_locks_to_pagelist(struct ceph_filelock *flocks, extern int ceph_fs_debugfs_init(struct ceph_fs_client *client); extern void ceph_fs_debugfs_cleanup(struct ceph_fs_client *client); +/* quota.c */ +static inline bool __ceph_has_any_quota(struct ceph_inode_info *ci) +{ + return ci->i_max_files || ci->i_max_bytes; +} + +extern void ceph_adjust_quota_realms_count(struct inode *inode, bool inc); + +static inline void __ceph_update_quota(struct ceph_inode_info *ci, + u64 max_bytes, u64 max_files) +{ + bool had_quota, has_quota; + had_quota = __ceph_has_any_quota(ci); + ci->i_max_bytes = max_bytes; + ci->i_max_files = max_files; + has_quota = __ceph_has_any_quota(ci); + + if (had_quota != has_quota) + ceph_adjust_quota_realms_count(&ci->vfs_inode, has_quota); +} + +extern void ceph_handle_quota(struct ceph_mds_client *mdsc, + struct ceph_mds_session *session, + struct ceph_msg *msg); +extern bool ceph_quota_is_max_files_exceeded(struct inode *inode); +extern bool ceph_quota_is_same_realm(struct inode *old, struct inode *new); +extern bool ceph_quota_is_max_bytes_exceeded(struct inode *inode, + loff_t newlen); +extern bool ceph_quota_is_max_bytes_approaching(struct inode *inode, + loff_t newlen); +extern bool ceph_quota_update_statfs(struct ceph_fs_client *fsc, + struct kstatfs *buf); + #endif /* _FS_CEPH_SUPER_H */ diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index e1c4e0b12b4c..315f7e63e7cc 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -224,6 +224,39 @@ static size_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val, (long)ci->i_rctime.tv_nsec); } +/* quotas */ + +static bool ceph_vxattrcb_quota_exists(struct ceph_inode_info *ci) +{ + bool ret = false; + spin_lock(&ci->i_ceph_lock); + if ((ci->i_max_files || ci->i_max_bytes) && + ci->i_vino.snap == CEPH_NOSNAP && + ci->i_snap_realm && + ci->i_snap_realm->ino == ci->i_vino.ino) + ret = true; + spin_unlock(&ci->i_ceph_lock); + return ret; +} + +static size_t ceph_vxattrcb_quota(struct ceph_inode_info *ci, char *val, + size_t size) +{ + return snprintf(val, size, "max_bytes=%llu max_files=%llu", + ci->i_max_bytes, ci->i_max_files); +} + +static size_t ceph_vxattrcb_quota_max_bytes(struct ceph_inode_info *ci, + char *val, size_t size) +{ + return snprintf(val, size, "%llu", ci->i_max_bytes); +} + +static size_t ceph_vxattrcb_quota_max_files(struct ceph_inode_info *ci, + char *val, size_t size) +{ + return snprintf(val, size, "%llu", ci->i_max_files); +} #define CEPH_XATTR_NAME(_type, _name) XATTR_CEPH_PREFIX #_type "." #_name #define CEPH_XATTR_NAME2(_type, _name, _name2) \ @@ -247,6 +280,15 @@ static size_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val, .hidden = true, \ .exists_cb = ceph_vxattrcb_layout_exists, \ } +#define XATTR_QUOTA_FIELD(_type, _name) \ + { \ + .name = CEPH_XATTR_NAME(_type, _name), \ + .name_size = sizeof(CEPH_XATTR_NAME(_type, _name)), \ + .getxattr_cb = ceph_vxattrcb_ ## _type ## _ ## _name, \ + .readonly = false, \ + .hidden = true, \ + .exists_cb = ceph_vxattrcb_quota_exists, \ + } static struct ceph_vxattr ceph_dir_vxattrs[] = { { @@ -270,6 +312,16 @@ static struct ceph_vxattr ceph_dir_vxattrs[] = { XATTR_NAME_CEPH(dir, rsubdirs), XATTR_NAME_CEPH(dir, rbytes), XATTR_NAME_CEPH(dir, rctime), + { + .name = "ceph.quota", + .name_size = sizeof("ceph.quota"), + .getxattr_cb = ceph_vxattrcb_quota, + .readonly = false, + .hidden = true, + .exists_cb = ceph_vxattrcb_quota_exists, + }, + XATTR_QUOTA_FIELD(quota, max_bytes), + XATTR_QUOTA_FIELD(quota, max_files), { .name = NULL, 0 } /* Required table terminator */ }; static size_t ceph_dir_vxattrs_name_size; /* total size of all names */ @@ -964,14 +1016,19 @@ int __ceph_setxattr(struct inode *inode, const char *name, char *newval = NULL; struct ceph_inode_xattr *xattr = NULL; int required_blob_size; + bool check_realm = false; bool lock_snap_rwsem = false; if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; vxattr = ceph_match_vxattr(inode, name); - if (vxattr && vxattr->readonly) - return -EOPNOTSUPP; + if (vxattr) { + if (vxattr->readonly) + return -EOPNOTSUPP; + if (value && !strncmp(vxattr->name, "ceph.quota", 10)) + check_realm = true; + } /* pass any unhandled ceph.* xattrs through to the MDS */ if (!strncmp(name, XATTR_CEPH_PREFIX, XATTR_CEPH_PREFIX_LEN)) @@ -1065,6 +1122,15 @@ do_sync_unlocked: err = -EBUSY; } else { err = ceph_sync_setxattr(inode, name, value, size, flags); + if (err >= 0 && check_realm) { + /* check if snaprealm was created for quota inode */ + spin_lock(&ci->i_ceph_lock); + if ((ci->i_max_files || ci->i_max_bytes) && + !(ci->i_snap_realm && + ci->i_snap_realm->ino == ci->i_vino.ino)) + err = -EOPNOTSUPP; + spin_unlock(&ci->i_ceph_lock); + } } out: ceph_free_cap_flush(prealloc_cf); diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 741749a98614..5f132d59dfc2 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -197,7 +197,7 @@ config CIFS_SMB311 config CIFS_SMB_DIRECT bool "SMB Direct support (Experimental)" - depends on CIFS=m && INFINIBAND || CIFS=y && INFINIBAND=y + depends on CIFS=m && INFINIBAND && INFINIBAND_ADDR_TRANS || CIFS=y && INFINIBAND=y && INFINIBAND_ADDR_TRANS=y help Enables SMB Direct experimental support for SMB 3.0, 3.02 and 3.1.1. SMB Direct allows transferring SMB packets over RDMA. If unsure, diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile index 7e4a1e2f0696..85817991ee68 100644 --- a/fs/cifs/Makefile +++ b/fs/cifs/Makefile @@ -1,11 +1,12 @@ # SPDX-License-Identifier: GPL-2.0 # -# Makefile for Linux CIFS VFS client +# Makefile for Linux CIFS/SMB2/SMB3 VFS client # +ccflags-y += -I$(src) # needed for trace events obj-$(CONFIG_CIFS) += cifs.o -cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \ - link.o misc.o netmisc.o smbencrypt.o transport.o asn1.o \ +cifs-y := trace.o cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o \ + inode.o link.o misc.o netmisc.o smbencrypt.o transport.o asn1.o \ cifs_unicode.o nterr.o cifsencrypt.o \ readdir.o ioctl.o sess.o export.o smb1ops.o winucase.o \ smb2ops.o smb2maperror.o smb2transport.o \ diff --git a/fs/cifs/cache.c b/fs/cifs/cache.c index 2c14020e5e1d..edf5f40898bf 100644 --- a/fs/cifs/cache.c +++ b/fs/cifs/cache.c @@ -46,67 +46,11 @@ void cifs_fscache_unregister(void) } /* - * Key layout of CIFS server cache index object - */ -struct cifs_server_key { - uint16_t family; /* address family */ - __be16 port; /* IP port */ - union { - struct in_addr ipv4_addr; - struct in6_addr ipv6_addr; - } addr[0]; -}; - -/* - * Server object keyed by {IPaddress,port,family} tuple - */ -static uint16_t cifs_server_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t maxbuf) -{ - const struct TCP_Server_Info *server = cookie_netfs_data; - const struct sockaddr *sa = (struct sockaddr *) &server->dstaddr; - const struct sockaddr_in *addr = (struct sockaddr_in *) sa; - const struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *) sa; - struct cifs_server_key *key = buffer; - uint16_t key_len = sizeof(struct cifs_server_key); - - memset(key, 0, key_len); - - /* - * Should not be a problem as sin_family/sin6_family overlays - * sa_family field - */ - switch (sa->sa_family) { - case AF_INET: - key->family = sa->sa_family; - key->port = addr->sin_port; - key->addr[0].ipv4_addr = addr->sin_addr; - key_len += sizeof(key->addr[0].ipv4_addr); - break; - - case AF_INET6: - key->family = sa->sa_family; - key->port = addr6->sin6_port; - key->addr[0].ipv6_addr = addr6->sin6_addr; - key_len += sizeof(key->addr[0].ipv6_addr); - break; - - default: - cifs_dbg(VFS, "Unknown network family '%d'\n", sa->sa_family); - key_len = 0; - break; - } - - return key_len; -} - -/* * Server object for FS-Cache */ const struct fscache_cookie_def cifs_fscache_server_index_def = { .name = "CIFS.server", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = cifs_server_get_key, }; /* @@ -116,7 +60,7 @@ struct cifs_fscache_super_auxdata { u64 resource_id; /* unique server resource id */ }; -static char *extract_sharename(const char *treename) +char *extract_sharename(const char *treename) { const char *src; char *delim, *dst; @@ -140,56 +84,11 @@ static char *extract_sharename(const char *treename) return dst; } -/* - * Superblock object currently keyed by share name - */ -static uint16_t cifs_super_get_key(const void *cookie_netfs_data, void *buffer, - uint16_t maxbuf) -{ - const struct cifs_tcon *tcon = cookie_netfs_data; - char *sharename; - uint16_t len; - - sharename = extract_sharename(tcon->treeName); - if (IS_ERR(sharename)) { - cifs_dbg(FYI, "%s: couldn't extract sharename\n", __func__); - sharename = NULL; - return 0; - } - - len = strlen(sharename); - if (len > maxbuf) - return 0; - - memcpy(buffer, sharename, len); - - kfree(sharename); - - return len; -} - -static uint16_t -cifs_fscache_super_get_aux(const void *cookie_netfs_data, void *buffer, - uint16_t maxbuf) -{ - struct cifs_fscache_super_auxdata auxdata; - const struct cifs_tcon *tcon = cookie_netfs_data; - - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.resource_id = tcon->resource_id; - - if (maxbuf > sizeof(auxdata)) - maxbuf = sizeof(auxdata); - - memcpy(buffer, &auxdata, maxbuf); - - return maxbuf; -} - static enum fscache_checkaux cifs_fscache_super_check_aux(void *cookie_netfs_data, const void *data, - uint16_t datalen) + uint16_t datalen, + loff_t object_size) { struct cifs_fscache_super_auxdata auxdata; const struct cifs_tcon *tcon = cookie_netfs_data; @@ -212,68 +111,14 @@ fscache_checkaux cifs_fscache_super_check_aux(void *cookie_netfs_data, const struct fscache_cookie_def cifs_fscache_super_index_def = { .name = "CIFS.super", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = cifs_super_get_key, - .get_aux = cifs_fscache_super_get_aux, .check_aux = cifs_fscache_super_check_aux, }; -/* - * Auxiliary data attached to CIFS inode within the cache - */ -struct cifs_fscache_inode_auxdata { - struct timespec last_write_time; - struct timespec last_change_time; - u64 eof; -}; - -static uint16_t cifs_fscache_inode_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t maxbuf) -{ - const struct cifsInodeInfo *cifsi = cookie_netfs_data; - uint16_t keylen; - - /* use the UniqueId as the key */ - keylen = sizeof(cifsi->uniqueid); - if (keylen > maxbuf) - keylen = 0; - else - memcpy(buffer, &cifsi->uniqueid, keylen); - - return keylen; -} - -static void -cifs_fscache_inode_get_attr(const void *cookie_netfs_data, uint64_t *size) -{ - const struct cifsInodeInfo *cifsi = cookie_netfs_data; - - *size = cifsi->vfs_inode.i_size; -} - -static uint16_t -cifs_fscache_inode_get_aux(const void *cookie_netfs_data, void *buffer, - uint16_t maxbuf) -{ - struct cifs_fscache_inode_auxdata auxdata; - const struct cifsInodeInfo *cifsi = cookie_netfs_data; - - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.eof = cifsi->server_eof; - auxdata.last_write_time = cifsi->vfs_inode.i_mtime; - auxdata.last_change_time = cifsi->vfs_inode.i_ctime; - - if (maxbuf > sizeof(auxdata)) - maxbuf = sizeof(auxdata); - - memcpy(buffer, &auxdata, maxbuf); - - return maxbuf; -} - static enum fscache_checkaux cifs_fscache_inode_check_aux(void *cookie_netfs_data, const void *data, - uint16_t datalen) + uint16_t datalen, + loff_t object_size) { struct cifs_fscache_inode_auxdata auxdata; struct cifsInodeInfo *cifsi = cookie_netfs_data; @@ -295,8 +140,5 @@ fscache_checkaux cifs_fscache_inode_check_aux(void *cookie_netfs_data, const struct fscache_cookie_def cifs_fscache_inode_object_def = { .name = "CIFS.uniqueid", .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .get_key = cifs_fscache_inode_get_key, - .get_attr = cifs_fscache_inode_get_attr, - .get_aux = cifs_fscache_inode_get_aux, .check_aux = cifs_fscache_inode_check_aux, }; diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index e35e711db68e..116146022aa1 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -42,24 +42,7 @@ cifs_dump_mem(char *label, void *data, int length) data, length, true); } -#ifdef CONFIG_CIFS_DEBUG -void cifs_vfs_err(const char *fmt, ...) -{ - struct va_format vaf; - va_list args; - - va_start(args, fmt); - - vaf.fmt = fmt; - vaf.va = &args; - - pr_err_ratelimited("CIFS VFS: %pV", &vaf); - - va_end(args); -} -#endif - -void cifs_dump_detail(void *buf) +void cifs_dump_detail(void *buf, struct TCP_Server_Info *server) { #ifdef CONFIG_CIFS_DEBUG2 struct smb_hdr *smb = (struct smb_hdr *)buf; @@ -67,7 +50,8 @@ void cifs_dump_detail(void *buf) cifs_dbg(VFS, "Cmd: %d Err: 0x%x Flags: 0x%x Flgs2: 0x%x Mid: %d Pid: %d\n", smb->Command, smb->Status.CifsError, smb->Flags, smb->Flags2, smb->Mid, smb->Pid); - cifs_dbg(VFS, "smb buf %p len %u\n", smb, smbCalcSize(smb)); + cifs_dbg(VFS, "smb buf %p len %u\n", smb, + server->ops->calc_smb_size(smb, server)); #endif /* CONFIG_CIFS_DEBUG2 */ } @@ -100,7 +84,7 @@ void cifs_dump_mids(struct TCP_Server_Info *server) cifs_dbg(VFS, "IsMult: %d IsEnd: %d\n", mid_entry->multiRsp, mid_entry->multiEnd); if (mid_entry->resp_buf) { - cifs_dump_detail(mid_entry->resp_buf); + cifs_dump_detail(mid_entry->resp_buf, server); cifs_dump_mem("existing buf: ", mid_entry->resp_buf, 62); } @@ -130,6 +114,8 @@ static void cifs_debug_tcon(struct seq_file *m, struct cifs_tcon *tcon) seq_printf(m, " type: %d ", dev_type); if (tcon->seal) seq_printf(m, " Encrypted"); + if (tcon->nocase) + seq_printf(m, " nocase"); if (tcon->unix_ext) seq_printf(m, " POSIX Extensions"); if (tcon->ses->server->ops->dump_share_caps) @@ -254,6 +240,10 @@ skip_rdma: server->credits, server->dialect); if (server->sign) seq_printf(m, " signed"); +#ifdef CONFIG_CIFS_SMB311 + if (server->posix_ext_supported) + seq_printf(m, " posix"); +#endif /* 3.1.1 */ i++; list_for_each(tmp2, &server->smb_ses_list) { ses = list_entry(tmp2, struct cifs_ses, @@ -331,18 +321,6 @@ skip_rdma: return 0; } -static int cifs_debug_data_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, cifs_debug_data_proc_show, NULL); -} - -static const struct file_operations cifs_debug_data_proc_fops = { - .open = cifs_debug_data_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - #ifdef CONFIG_CIFS_STATS static ssize_t cifs_stats_proc_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) @@ -514,35 +492,36 @@ cifs_proc_init(void) if (proc_fs_cifs == NULL) return; - proc_create("DebugData", 0, proc_fs_cifs, &cifs_debug_data_proc_fops); + proc_create_single("DebugData", 0, proc_fs_cifs, + cifs_debug_data_proc_show); #ifdef CONFIG_CIFS_STATS - proc_create("Stats", 0, proc_fs_cifs, &cifs_stats_proc_fops); + proc_create("Stats", 0644, proc_fs_cifs, &cifs_stats_proc_fops); #endif /* STATS */ - proc_create("cifsFYI", 0, proc_fs_cifs, &cifsFYI_proc_fops); - proc_create("traceSMB", 0, proc_fs_cifs, &traceSMB_proc_fops); - proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, + proc_create("cifsFYI", 0644, proc_fs_cifs, &cifsFYI_proc_fops); + proc_create("traceSMB", 0644, proc_fs_cifs, &traceSMB_proc_fops); + proc_create("LinuxExtensionsEnabled", 0644, proc_fs_cifs, &cifs_linux_ext_proc_fops); - proc_create("SecurityFlags", 0, proc_fs_cifs, + proc_create("SecurityFlags", 0644, proc_fs_cifs, &cifs_security_flags_proc_fops); - proc_create("LookupCacheEnabled", 0, proc_fs_cifs, + proc_create("LookupCacheEnabled", 0644, proc_fs_cifs, &cifs_lookup_cache_proc_fops); #ifdef CONFIG_CIFS_SMB_DIRECT - proc_create("rdma_readwrite_threshold", 0, proc_fs_cifs, + proc_create("rdma_readwrite_threshold", 0644, proc_fs_cifs, &cifs_rdma_readwrite_threshold_proc_fops); - proc_create("smbd_max_frmr_depth", 0, proc_fs_cifs, + proc_create("smbd_max_frmr_depth", 0644, proc_fs_cifs, &cifs_smbd_max_frmr_depth_proc_fops); - proc_create("smbd_keep_alive_interval", 0, proc_fs_cifs, + proc_create("smbd_keep_alive_interval", 0644, proc_fs_cifs, &cifs_smbd_keep_alive_interval_proc_fops); - proc_create("smbd_max_receive_size", 0, proc_fs_cifs, + proc_create("smbd_max_receive_size", 0644, proc_fs_cifs, &cifs_smbd_max_receive_size_proc_fops); - proc_create("smbd_max_fragmented_recv_size", 0, proc_fs_cifs, + proc_create("smbd_max_fragmented_recv_size", 0644, proc_fs_cifs, &cifs_smbd_max_fragmented_recv_size_proc_fops); - proc_create("smbd_max_send_size", 0, proc_fs_cifs, + proc_create("smbd_max_send_size", 0644, proc_fs_cifs, &cifs_smbd_max_send_size_proc_fops); - proc_create("smbd_send_credit_target", 0, proc_fs_cifs, + proc_create("smbd_send_credit_target", 0644, proc_fs_cifs, &cifs_smbd_send_credit_target_proc_fops); - proc_create("smbd_receive_credit_max", 0, proc_fs_cifs, + proc_create("smbd_receive_credit_max", 0644, proc_fs_cifs, &cifs_smbd_receive_credit_max_proc_fops); #endif } @@ -600,6 +579,8 @@ static ssize_t cifsFYI_proc_write(struct file *file, const char __user *buffer, cifsFYI = bv; else if ((c[0] > '1') && (c[0] <= '9')) cifsFYI = (int) (c[0] - '0'); /* see cifs_debug.h for meanings */ + else + return -EINVAL; return count; } diff --git a/fs/cifs/cifs_debug.h b/fs/cifs/cifs_debug.h index c611ca2339d7..f4f3f0853c6e 100644 --- a/fs/cifs/cifs_debug.h +++ b/fs/cifs/cifs_debug.h @@ -23,7 +23,7 @@ #define _H_CIFS_DEBUG void cifs_dump_mem(char *label, void *data, int length); -void cifs_dump_detail(void *); +void cifs_dump_detail(void *buf, struct TCP_Server_Info *ptcp_info); void cifs_dump_mids(struct TCP_Server_Info *); extern bool traceSMB; /* flag which enables the function below */ void dump_smb(void *, int); @@ -39,6 +39,7 @@ extern int cifsFYI; #else #define NOISY 0 #endif +#define ONCE 8 /* * debug ON @@ -46,19 +47,28 @@ extern int cifsFYI; */ #ifdef CONFIG_CIFS_DEBUG -__printf(1, 2) void cifs_vfs_err(const char *fmt, ...); - /* information message: e.g., configuration, major event */ -#define cifs_dbg(type, fmt, ...) \ -do { \ - if (type == FYI && cifsFYI & CIFS_INFO) { \ - pr_debug_ratelimited("%s: " \ - fmt, __FILE__, ##__VA_ARGS__); \ - } else if (type == VFS) { \ - cifs_vfs_err(fmt, ##__VA_ARGS__); \ - } else if (type == NOISY && type != 0) { \ - pr_debug_ratelimited(fmt, ##__VA_ARGS__); \ - } \ +#define cifs_dbg_func(ratefunc, type, fmt, ...) \ +do { \ + if ((type) & FYI && cifsFYI & CIFS_INFO) { \ + pr_debug_ ## ratefunc("%s: " \ + fmt, __FILE__, ##__VA_ARGS__); \ + } else if ((type) & VFS) { \ + pr_err_ ## ratefunc("CIFS VFS: " \ + fmt, ##__VA_ARGS__); \ + } else if ((type) & NOISY && (NOISY != 0)) { \ + pr_debug_ ## ratefunc(fmt, ##__VA_ARGS__); \ + } \ +} while (0) + +#define cifs_dbg(type, fmt, ...) \ +do { \ + if ((type) & ONCE) \ + cifs_dbg_func(once, \ + type, fmt, ##__VA_ARGS__); \ + else \ + cifs_dbg_func(ratelimited, \ + type, fmt, ##__VA_ARGS__); \ } while (0) /* diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index 350fa55a1bf7..9731d0d891e7 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -50,6 +50,7 @@ * root mountable */ #define CIFS_MOUNT_UID_FROM_ACL 0x2000000 /* try to get UID via special SID */ +#define CIFS_MOUNT_NO_HANDLE_CACHE 0x4000000 /* disable caching dir handles */ struct cifs_sb_info { struct rb_root tlink_tree; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index f715609b13f3..eb7b6573f322 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -58,13 +58,15 @@ bool traceSMB; bool enable_oplocks = true; bool linuxExtEnabled = true; bool lookupCacheEnabled = true; +bool disable_legacy_dialects; /* false by default */ unsigned int global_secflags = CIFSSEC_DEF; /* unsigned int ntlmv2_support = 0; */ unsigned int sign_CIFS_PDUs = 1; static const struct super_operations cifs_super_ops; unsigned int CIFSMaxBufSize = CIFS_MAX_MSGSIZE; module_param(CIFSMaxBufSize, uint, 0444); -MODULE_PARM_DESC(CIFSMaxBufSize, "Network buffer size (not including header). " +MODULE_PARM_DESC(CIFSMaxBufSize, "Network buffer size (not including header) " + "for CIFS requests. " "Default: 16384 Range: 8192 to 130048"); unsigned int cifs_min_rcv = CIFS_MIN_RCV_POOL; module_param(cifs_min_rcv, uint, 0444); @@ -76,11 +78,21 @@ MODULE_PARM_DESC(cifs_min_small, "Small network buffers in pool. Default: 30 " "Range: 2 to 256"); unsigned int cifs_max_pending = CIFS_MAX_REQ; module_param(cifs_max_pending, uint, 0444); -MODULE_PARM_DESC(cifs_max_pending, "Simultaneous requests to server. " +MODULE_PARM_DESC(cifs_max_pending, "Simultaneous requests to server for " + "CIFS/SMB1 dialect (N/A for SMB3) " "Default: 32767 Range: 2 to 32767."); module_param(enable_oplocks, bool, 0644); MODULE_PARM_DESC(enable_oplocks, "Enable or disable oplocks. Default: y/Y/1"); +module_param(disable_legacy_dialects, bool, 0644); +MODULE_PARM_DESC(disable_legacy_dialects, "To improve security it may be " + "helpful to restrict the ability to " + "override the default dialects (SMB2.1, " + "SMB3 and SMB3.02) on mount with old " + "dialects (CIFS/SMB1 and SMB2) since " + "vers=1.0 (CIFS/SMB1) and vers=2.0 are weaker" + " and less secure. Default: n/N/0"); + extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; extern mempool_t *cifs_mid_poolp; @@ -469,10 +481,20 @@ cifs_show_options(struct seq_file *s, struct dentry *root) seq_puts(s, ",persistenthandles"); else if (tcon->use_resilient) seq_puts(s, ",resilienthandles"); + +#ifdef CONFIG_CIFS_SMB311 + if (tcon->posix_extensions) + seq_puts(s, ",posix"); + else if (tcon->unix_ext) + seq_puts(s, ",unix"); + else + seq_puts(s, ",nounix"); +#else if (tcon->unix_ext) seq_puts(s, ",unix"); else seq_puts(s, ",nounix"); +#endif /* SMB311 */ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) seq_puts(s, ",posixpaths"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) @@ -495,6 +517,8 @@ cifs_show_options(struct seq_file *s, struct dentry *root) seq_puts(s, ",sfu"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) seq_puts(s, ",nobrl"); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_HANDLE_CACHE) + seq_puts(s, ",nohandlecache"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) seq_puts(s, ",cifsacl"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) @@ -897,6 +921,17 @@ struct file_system_type cifs_fs_type = { /* .fs_flags */ }; MODULE_ALIAS_FS("cifs"); + +static struct file_system_type smb3_fs_type = { + .owner = THIS_MODULE, + .name = "smb3", + .mount = cifs_do_mount, + .kill_sb = cifs_kill_sb, + /* .fs_flags */ +}; +MODULE_ALIAS_FS("smb3"); +MODULE_ALIAS("smb3"); + const struct inode_operations cifs_dir_inode_ops = { .create = cifs_create, .atomic_open = cifs_atomic_open, @@ -1047,6 +1082,18 @@ out: return rc; } +/* + * Directory operations under CIFS/SMB2/SMB3 are synchronous, so fsync() + * is a dummy operation. + */ +static int cifs_dir_fsync(struct file *file, loff_t start, loff_t end, int datasync) +{ + cifs_dbg(FYI, "Sync directory - name: %pD datasync: 0x%x\n", + file, datasync); + + return 0; +} + static ssize_t cifs_copy_file_range(struct file *src_file, loff_t off, struct file *dst_file, loff_t destoff, size_t len, unsigned int flags) @@ -1181,6 +1228,7 @@ const struct file_operations cifs_dir_ops = { .copy_file_range = cifs_copy_file_range, .clone_file_range = cifs_clone_file_range, .llseek = generic_file_llseek, + .fsync = cifs_dir_fsync, }; static void @@ -1422,6 +1470,12 @@ init_cifs(void) if (rc) goto out_init_cifs_idmap; + rc = register_filesystem(&smb3_fs_type); + if (rc) { + unregister_filesystem(&cifs_fs_type); + goto out_init_cifs_idmap; + } + return 0; out_init_cifs_idmap: @@ -1452,8 +1506,9 @@ out_clean_proc: static void __exit exit_cifs(void) { - cifs_dbg(NOISY, "exit_cifs\n"); + cifs_dbg(NOISY, "exit_smb3\n"); unregister_filesystem(&cifs_fs_type); + unregister_filesystem(&smb3_fs_type); cifs_dfs_release_automount_timer(); #ifdef CONFIG_CIFS_ACL exit_cifs_idmap(); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 013ba2aed8d9..5f0231803431 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -149,5 +149,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg); extern const struct export_operations cifs_export_ops; #endif /* CONFIG_CIFS_NFSD_EXPORT */ -#define CIFS_VERSION "2.11" +#define CIFS_VERSION "2.12" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 2282562e78a1..08d1cdd96701 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -176,6 +176,7 @@ struct smb_rqst { struct kvec *rq_iov; /* array of kvecs */ unsigned int rq_nvec; /* number of kvecs in array */ struct page **rq_pages; /* pointer to array of page ptrs */ + unsigned int rq_offset; /* the offset to the 1st page */ unsigned int rq_npages; /* number pages in array */ unsigned int rq_pagesz; /* page size to use */ unsigned int rq_tailsz; /* length of last page */ @@ -244,7 +245,7 @@ struct smb_version_operations { int (*map_error)(char *, bool); /* find mid corresponding to the response message */ struct mid_q_entry * (*find_mid)(struct TCP_Server_Info *, char *); - void (*dump_detail)(void *); + void (*dump_detail)(void *buf, struct TCP_Server_Info *ptcp_info); void (*clear_stats)(struct cifs_tcon *); void (*print_stats)(struct seq_file *m, struct cifs_tcon *); void (*dump_share_caps)(struct seq_file *, struct cifs_tcon *); @@ -372,7 +373,7 @@ struct smb_version_operations { int (*close_dir)(const unsigned int, struct cifs_tcon *, struct cifs_fid *); /* calculate a size of SMB message */ - unsigned int (*calc_smb_size)(void *); + unsigned int (*calc_smb_size)(void *buf, struct TCP_Server_Info *ptcpi); /* check for STATUS_PENDING and process it in a positive case */ bool (*is_status_pending)(char *, struct TCP_Server_Info *, int); /* check for STATUS_NETWORK_SESSION_EXPIRED */ @@ -417,7 +418,7 @@ struct smb_version_operations { /* create lease context buffer for CREATE request */ char * (*create_lease_buf)(u8 *, u8); /* parse lease context buffer and return oplock/epoch info */ - __u8 (*parse_lease_buf)(void *, unsigned int *); + __u8 (*parse_lease_buf)(void *buf, unsigned int *epoch, char *lkey); ssize_t (*copychunk_range)(const unsigned int, struct cifsFileInfo *src_file, struct cifsFileInfo *target_file, @@ -457,7 +458,7 @@ struct smb_version_operations { struct mid_q_entry **); enum securityEnum (*select_sectype)(struct TCP_Server_Info *, enum securityEnum); - + int (*next_header)(char *); }; struct smb_version_values { @@ -521,10 +522,12 @@ struct smb_vol { bool sfu_remap:1; /* remap seven reserved chars ala SFU */ bool posix_paths:1; /* unset to not ask for posix pathnames. */ bool no_linux_ext:1; + bool linux_ext:1; bool sfu_emul:1; bool nullauth:1; /* attempt to authenticate with null user */ bool nocase:1; /* request case insensitive filenames */ bool nobrl:1; /* disable sending byte range locks to srv */ + bool nohandlecache:1; /* disable caching dir handles if srvr probs */ bool mand_lock:1; /* send mandatory not posix byte range lock reqs */ bool seal:1; /* request transport encryption on share */ bool nodfs:1; /* Do not request DFS, even if available */ @@ -630,7 +633,7 @@ struct TCP_Server_Info { bool oplocks:1; /* enable oplocks */ unsigned int maxReq; /* Clients should submit no more */ /* than maxReq distinct unanswered SMBs to the server when using */ - /* multiplexed reads or writes */ + /* multiplexed reads or writes (for SMB1/CIFS only, not SMB2/SMB3) */ unsigned int maxBuf; /* maxBuf specifies the maximum */ /* message size the server can send or receive for non-raw SMBs */ /* maxBuf is returned by SMB NegotiateProtocol so maxBuf is only 0 */ @@ -665,6 +668,8 @@ struct TCP_Server_Info { struct delayed_work echo; /* echo ping workqueue job */ char *smallbuf; /* pointer to current "small" buffer */ char *bigbuf; /* pointer to current "big" buffer */ + /* Total size of this PDU. Only valid from cifs_demultiplex_thread */ + unsigned int pdu_size; unsigned int total_read; /* total amount of data read in this pass */ #ifdef CONFIG_CIFS_FSCACHE struct fscache_cookie *fscache; /* client index cache cookie */ @@ -676,8 +681,10 @@ struct TCP_Server_Info { unsigned int max_read; unsigned int max_write; #ifdef CONFIG_CIFS_SMB311 + __le16 cipher_type; /* save initital negprot hash */ __u8 preauth_sha_hash[SMB2_PREAUTH_HASH_SIZE]; + bool posix_ext_supported; #endif /* 3.1.1 */ struct delayed_work reconnect; /* reconnect workqueue job */ struct mutex reconnect_mutex; /* prevent simultaneous reconnects */ @@ -950,9 +957,13 @@ struct cifs_tcon { bool print:1; /* set if connection to printer share */ bool retry:1; bool nocase:1; + bool nohandlecache:1; /* if strange server resource prob can turn off */ bool seal:1; /* transport encryption for this mounted share */ bool unix_ext:1; /* if false disable Linux extensions to CIFS protocol for this mount even if server would support */ +#ifdef CONFIG_CIFS_SMB311 + bool posix_extensions; /* if true SMB3.11 posix extensions enabled */ +#endif /* CIFS_311 */ bool local_lease:1; /* check leases (only) on local system not remote */ bool broken_posix_open; /* e.g. Samba server versions < 3.3.2, 3.2.9 */ bool broken_sparse_sup; /* if server or share does not support sparse */ @@ -976,6 +987,9 @@ struct cifs_tcon { struct fscache_cookie *fscache; /* cookie for share */ #endif struct list_head pending_opens; /* list of incomplete opens */ + bool valid_root_fid:1; /* Do we have a useable root fid */ + struct mutex prfid_mutex; /* prevents reopen race after dead ses*/ + struct cifs_fid *prfid; /* handle to the directory at top of share */ /* BB add field for back pointer to sb struct(s)? */ }; @@ -1068,6 +1082,7 @@ struct cifs_open_parms { int create_options; const char *path; struct cifs_fid *fid; + umode_t mode; bool reconnect:1; }; @@ -1166,10 +1181,11 @@ struct cifs_readdata { struct smbd_mr *mr; #endif unsigned int pagesz; + unsigned int page_offset; unsigned int tailsz; unsigned int credits; unsigned int nr_pages; - struct page *pages[]; + struct page **pages; }; struct cifs_writedata; @@ -1191,10 +1207,11 @@ struct cifs_writedata { struct smbd_mr *mr; #endif unsigned int pagesz; + unsigned int page_offset; unsigned int tailsz; unsigned int credits; unsigned int nr_pages; - struct page *pages[]; + struct page **pages; }; /* @@ -1373,6 +1390,7 @@ struct mid_q_entry { mid_handle_t *handle; /* call handle mid callback */ void *callback_data; /* general purpose pointer for callback */ void *resp_buf; /* pointer to received SMB header */ + unsigned int resp_buf_size; int mid_state; /* wish this were enum but can not pass to wait_event */ unsigned int mid_flags; __le16 command; /* smb command code */ @@ -1688,16 +1706,17 @@ GLOBAL_EXTERN atomic_t smBufAllocCount; GLOBAL_EXTERN atomic_t midCount; /* Misc globals */ -GLOBAL_EXTERN bool enable_oplocks; /* enable or disable oplocks */ -GLOBAL_EXTERN bool lookupCacheEnabled; -GLOBAL_EXTERN unsigned int global_secflags; /* if on, session setup sent +extern bool enable_oplocks; /* enable or disable oplocks */ +extern bool lookupCacheEnabled; +extern unsigned int global_secflags; /* if on, session setup sent with more secure ntlmssp2 challenge/resp */ -GLOBAL_EXTERN unsigned int sign_CIFS_PDUs; /* enable smb packet signing */ -GLOBAL_EXTERN bool linuxExtEnabled;/*enable Linux/Unix CIFS extensions*/ -GLOBAL_EXTERN unsigned int CIFSMaxBufSize; /* max size not including hdr */ -GLOBAL_EXTERN unsigned int cifs_min_rcv; /* min size of big ntwrk buf pool */ -GLOBAL_EXTERN unsigned int cifs_min_small; /* min size of small buf pool */ -GLOBAL_EXTERN unsigned int cifs_max_pending; /* MAX requests at once to server*/ +extern unsigned int sign_CIFS_PDUs; /* enable smb packet signing */ +extern bool linuxExtEnabled;/*enable Linux/Unix CIFS extensions*/ +extern unsigned int CIFSMaxBufSize; /* max size not including hdr */ +extern unsigned int cifs_min_rcv; /* min size of big ntwrk buf pool */ +extern unsigned int cifs_min_small; /* min size of small buf pool */ +extern unsigned int cifs_max_pending; /* MAX requests at once to server*/ +extern bool disable_legacy_dialects; /* forbid vers=1.0 and vers=2.0 mounts */ #ifdef CONFIG_CIFS_ACL GLOBAL_EXTERN struct rb_root uidtree; diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 365a414a75e9..7933c5f9c076 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -21,6 +21,7 @@ #ifndef _CIFSPROTO_H #define _CIFSPROTO_H #include <linux/nls.h> +#include "trace.h" struct statfs; struct smb_vol; @@ -47,6 +48,7 @@ extern void _free_xid(unsigned int); cifs_dbg(FYI, "CIFS VFS: in %s as Xid: %u with uid: %d\n", \ __func__, __xid, \ from_kuid(&init_user_ns, current_fsuid())); \ + trace_smb3_enter(__xid, __func__); \ __xid; \ }) @@ -54,7 +56,11 @@ extern void _free_xid(unsigned int); do { \ _free_xid(curr_xid); \ cifs_dbg(FYI, "CIFS VFS: leaving %s (xid = %u) rc = %d\n", \ - __func__, curr_xid, (int)rc); \ + __func__, curr_xid, (int)rc); \ + if (rc) \ + trace_smb3_exit_err(curr_xid, __func__, (int)rc); \ + else \ + trace_smb3_exit_done(curr_xid, __func__); \ } while (0) extern int init_cifs_idmap(void); extern void exit_cifs_idmap(void); @@ -124,7 +130,7 @@ extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, unsigned int bytes_written); extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, bool); extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *, bool); -extern unsigned int smbCalcSize(void *buf); +extern unsigned int smbCalcSize(void *buf, struct TCP_Server_Info *server); extern int decode_negTokenInit(unsigned char *security_blob, int length, struct TCP_Server_Info *server); extern int cifs_convert_address(struct sockaddr *dst, const char *src, int len); @@ -197,7 +203,9 @@ extern void dequeue_mid(struct mid_q_entry *mid, bool malformed); extern int cifs_read_from_socket(struct TCP_Server_Info *server, char *buf, unsigned int to_read); extern int cifs_read_page_from_socket(struct TCP_Server_Info *server, - struct page *page, unsigned int to_read); + struct page *page, + unsigned int page_offset, + unsigned int to_read); extern int cifs_setup_cifs_sb(struct smb_vol *pvolume_info, struct cifs_sb_info *cifs_sb); extern int cifs_match_super(struct super_block *, void *); @@ -525,6 +533,8 @@ int cifs_async_writev(struct cifs_writedata *wdata, void cifs_writev_complete(struct work_struct *work); struct cifs_writedata *cifs_writedata_alloc(unsigned int nr_pages, work_func_t complete); +struct cifs_writedata *cifs_writedata_direct_alloc(struct page **pages, + work_func_t complete); void cifs_writedata_release(struct kref *refcount); int cifs_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 59c09a596c0a..5aca336642c0 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -106,6 +106,12 @@ cifs_mark_open_files_invalid(struct cifs_tcon *tcon) open_file->oplock_break_cancelled = true; } spin_unlock(&tcon->open_file_lock); + + mutex_lock(&tcon->prfid_mutex); + tcon->valid_root_fid = false; + memset(tcon->prfid, 0, sizeof(struct cifs_fid)); + mutex_unlock(&tcon->prfid_mutex); + /* * BB Add call to invalidate_inodes(sb) for all superblocks mounted * to this tcon. @@ -206,8 +212,10 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) mutex_unlock(&ses->session_mutex); cifs_dbg(FYI, "reconnect tcon rc = %d\n", rc); - if (rc) + if (rc) { + printk_once(KERN_WARNING "reconnect tcon failed rc = %d\n", rc); goto out; + } atomic_inc(&tconInfoReconnectCount); @@ -453,6 +461,9 @@ cifs_enable_signing(struct TCP_Server_Info *server, bool mnt_sign_required) server->sign = true; } + if (cifs_rdma_enabled(server) && server->sign) + cifs_dbg(VFS, "Signing is enabled, and RDMA read/write will be disabled"); + return 0; } @@ -1416,8 +1427,9 @@ openRetry: int cifs_discard_remaining_data(struct TCP_Server_Info *server) { - unsigned int rfclen = get_rfc1002_length(server->smallbuf); - int remaining = rfclen + 4 - server->total_read; + unsigned int rfclen = server->pdu_size; + int remaining = rfclen + server->vals->header_preamble_size - + server->total_read; while (remaining > 0) { int length; @@ -1454,7 +1466,7 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) unsigned int data_offset, data_len; struct cifs_readdata *rdata = mid->callback_data; char *buf = server->smallbuf; - unsigned int buflen = get_rfc1002_length(buf) + + unsigned int buflen = server->pdu_size + server->vals->header_preamble_size; bool use_rdma_mr = false; @@ -1940,6 +1952,7 @@ cifs_writedata_release(struct kref *refcount) if (wdata->cfile) cifsFileInfo_put(wdata->cfile); + kvfree(wdata->pages); kfree(wdata); } @@ -2063,12 +2076,22 @@ cifs_writev_complete(struct work_struct *work) struct cifs_writedata * cifs_writedata_alloc(unsigned int nr_pages, work_func_t complete) { + struct page **pages = + kzalloc(sizeof(struct page *) * nr_pages, GFP_NOFS); + if (pages) + return cifs_writedata_direct_alloc(pages, complete); + + return NULL; +} + +struct cifs_writedata * +cifs_writedata_direct_alloc(struct page **pages, work_func_t complete) +{ struct cifs_writedata *wdata; - /* writedata + number of page pointers */ - wdata = kzalloc(sizeof(*wdata) + - sizeof(struct page *) * nr_pages, GFP_NOFS); + wdata = kzalloc(sizeof(*wdata), GFP_NOFS); if (wdata != NULL) { + wdata->pages = pages; kref_init(&wdata->refcount); INIT_LIST_HEAD(&wdata->list); init_completion(&wdata->done); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 4e0808f40195..e5a2fe7f0dd4 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -61,6 +61,7 @@ #define RFC1001_PORT 139 extern mempool_t *cifs_req_poolp; +extern bool disable_legacy_dialects; /* FIXME: should these be tunable? */ #define TLINK_ERROR_EXPIRE (1 * HZ) @@ -76,9 +77,10 @@ enum { Opt_mapposix, Opt_nomapposix, Opt_mapchars, Opt_nomapchars, Opt_sfu, Opt_nosfu, Opt_nodfs, Opt_posixpaths, - Opt_noposixpaths, Opt_nounix, + Opt_noposixpaths, Opt_nounix, Opt_unix, Opt_nocase, Opt_brl, Opt_nobrl, + Opt_handlecache, Opt_nohandlecache, Opt_forcemandatorylock, Opt_setuidfromacl, Opt_setuids, Opt_nosetuids, Opt_dynperm, Opt_nodynperm, Opt_nohard, Opt_nosoft, @@ -144,10 +146,16 @@ static const match_table_t cifs_mount_option_tokens = { { Opt_noposixpaths, "noposixpaths" }, { Opt_nounix, "nounix" }, { Opt_nounix, "nolinux" }, + { Opt_nounix, "noposix" }, + { Opt_unix, "unix" }, + { Opt_unix, "linux" }, + { Opt_unix, "posix" }, { Opt_nocase, "nocase" }, { Opt_nocase, "ignorecase" }, { Opt_brl, "brl" }, { Opt_nobrl, "nobrl" }, + { Opt_handlecache, "handlecache" }, + { Opt_nohandlecache, "nohandlecache" }, { Opt_nobrl, "nolock" }, { Opt_forcemandatorylock, "forcemandatorylock" }, { Opt_forcemandatorylock, "forcemand" }, @@ -591,10 +599,11 @@ cifs_read_from_socket(struct TCP_Server_Info *server, char *buf, int cifs_read_page_from_socket(struct TCP_Server_Info *server, struct page *page, - unsigned int to_read) + unsigned int page_offset, unsigned int to_read) { struct msghdr smb_msg; - struct bio_vec bv = {.bv_page = page, .bv_len = to_read}; + struct bio_vec bv = { + .bv_page = page, .bv_len = to_read, .bv_offset = page_offset}; iov_iter_bvec(&smb_msg.msg_iter, READ | ITER_BVEC, &bv, 1, to_read); return cifs_readv_from_socket(server, &smb_msg); } @@ -772,7 +781,7 @@ standard_receive3(struct TCP_Server_Info *server, struct mid_q_entry *mid) { int length; char *buf = server->smallbuf; - unsigned int pdu_length = get_rfc1002_length(buf); + unsigned int pdu_length = server->pdu_size; /* make sure this will fit in a large buffer */ if (pdu_length > CIFSMaxBufSize + MAX_HEADER_SIZE(server) - @@ -848,6 +857,7 @@ cifs_demultiplex_thread(void *p) int length; struct TCP_Server_Info *server = p; unsigned int pdu_length; + unsigned int next_offset; char *buf = NULL; struct task_struct *task_to_wake = NULL; struct mid_q_entry *mid_entry; @@ -874,7 +884,11 @@ cifs_demultiplex_thread(void *p) length = cifs_read_from_socket(server, buf, pdu_length); if (length < 0) continue; - server->total_read = length; + + if (server->vals->header_preamble_size == 0) + server->total_read = 0; + else + server->total_read = length; /* * The right amount was read from socket - 4 bytes, @@ -885,12 +899,14 @@ cifs_demultiplex_thread(void *p) cifs_dbg(FYI, "RFC1002 header 0x%x\n", pdu_length); if (!is_smb_response(server, buf[0])) continue; +next_pdu: + server->pdu_size = pdu_length; /* make sure we have enough to get to the MID */ - if (pdu_length < HEADER_SIZE(server) - 1 - + if (server->pdu_size < HEADER_SIZE(server) - 1 - server->vals->header_preamble_size) { cifs_dbg(VFS, "SMB response too short (%u bytes)\n", - pdu_length); + server->pdu_size); cifs_reconnect(server); wake_up(&server->response_q); continue; @@ -905,6 +921,12 @@ cifs_demultiplex_thread(void *p) continue; server->total_read += length; + if (server->ops->next_header) { + next_offset = server->ops->next_header(buf); + if (next_offset) + server->pdu_size = next_offset; + } + if (server->ops->is_transform_hdr && server->ops->receive_transform && server->ops->is_transform_hdr(buf)) { @@ -927,6 +949,7 @@ cifs_demultiplex_thread(void *p) server->lstrp = jiffies; if (mid_entry != NULL) { + mid_entry->resp_buf_size = server->pdu_size; if ((mid_entry->mid_flags & MID_WAIT_CANCELLED) && mid_entry->mid_state == MID_RESPONSE_RECEIVED && server->ops->handle_cancelled_mid) @@ -946,10 +969,18 @@ cifs_demultiplex_thread(void *p) HEADER_SIZE(server)); #ifdef CONFIG_CIFS_DEBUG2 if (server->ops->dump_detail) - server->ops->dump_detail(buf); + server->ops->dump_detail(buf, server); cifs_dump_mids(server); #endif /* CIFS_DEBUG2 */ - + } + if (pdu_length > server->pdu_size) { + if (!allocate_buffers(server)) + continue; + pdu_length -= server->pdu_size; + server->total_read = 0; + server->large_buf = false; + buf = server->smallbuf; + goto next_pdu; } } /* end while !EXITING */ @@ -1141,10 +1172,18 @@ cifs_parse_smb_version(char *value, struct smb_vol *vol) switch (match_token(value, cifs_smb_version_tokens, args)) { case Smb_1: + if (disable_legacy_dialects) { + cifs_dbg(VFS, "mount with legacy dialect disabled\n"); + return 1; + } vol->ops = &smb1_operations; vol->vals = &smb1_values; break; case Smb_20: + if (disable_legacy_dialects) { + cifs_dbg(VFS, "mount with legacy dialect disabled\n"); + return 1; + } vol->ops = &smb20_operations; vol->vals = &smb20_values; break; @@ -1424,8 +1463,17 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, vol->posix_paths = 0; break; case Opt_nounix: + if (vol->linux_ext) + cifs_dbg(VFS, + "conflicting unix mount options\n"); vol->no_linux_ext = 1; break; + case Opt_unix: + if (vol->no_linux_ext) + cifs_dbg(VFS, + "conflicting unix mount options\n"); + vol->linux_ext = 1; + break; case Opt_nocase: vol->nocase = 1; break; @@ -1443,6 +1491,12 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, (S_IALLUGO & ~(S_ISUID | S_IXGRP))) vol->file_mode = S_IALLUGO; break; + case Opt_nohandlecache: + vol->nohandlecache = 1; + break; + case Opt_handlecache: + vol->nohandlecache = 0; + break; case Opt_forcemandatorylock: vol->mand_lock = 1; break; @@ -1975,14 +2029,6 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, goto cifs_parse_mount_err; } -#ifdef CONFIG_CIFS_SMB_DIRECT - if (vol->rdma && vol->sign) { - cifs_dbg(VFS, "Currently SMB direct doesn't support signing." - " This is being fixed\n"); - goto cifs_parse_mount_err; - } -#endif - #ifndef CONFIG_KEYS /* Muliuser mounts require CONFIG_KEYS support */ if (vol->multiuser) { @@ -2957,6 +3003,29 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb_vol *volume_info) } } + if (volume_info->seal) { + if (ses->server->vals->protocol_id == 0) { + cifs_dbg(VFS, + "SMB3 or later required for encryption\n"); + rc = -EOPNOTSUPP; + goto out_fail; + } else if (tcon->ses->server->capabilities & + SMB2_GLOBAL_CAP_ENCRYPTION) + tcon->seal = true; + else { + cifs_dbg(VFS, "Encryption is not supported on share\n"); + rc = -EOPNOTSUPP; + goto out_fail; + } + } + +#ifdef CONFIG_CIFS_SMB311 + if ((volume_info->linux_ext) && (ses->server->posix_ext_supported)) { + if (ses->server->vals->protocol_id == SMB311_PROT_ID) + tcon->posix_extensions = true; + } +#endif /* 311 */ + /* * BB Do we need to wrap session_mutex around this TCon call and Unix * SetFS as we do on SessSetup and reconnect? @@ -3005,22 +3074,6 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb_vol *volume_info) tcon->use_resilient = true; } - if (volume_info->seal) { - if (ses->server->vals->protocol_id == 0) { - cifs_dbg(VFS, - "SMB3 or later required for encryption\n"); - rc = -EOPNOTSUPP; - goto out_fail; - } else if (tcon->ses->server->capabilities & - SMB2_GLOBAL_CAP_ENCRYPTION) - tcon->seal = true; - else { - cifs_dbg(VFS, "Encryption is not supported on share\n"); - rc = -EOPNOTSUPP; - goto out_fail; - } - } - /* * We can have only one retry value for a connection to a share so for * resources mounted more than once to the same server share the last @@ -3028,6 +3081,7 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb_vol *volume_info) */ tcon->retry = volume_info->retry; tcon->nocase = volume_info->nocase; + tcon->nohandlecache = volume_info->nohandlecache; tcon->local_lease = volume_info->local_lease; INIT_LIST_HEAD(&tcon->pending_opens); @@ -3586,6 +3640,8 @@ int cifs_setup_cifs_sb(struct smb_vol *pvolume_info, cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_UNX_EMUL; if (pvolume_info->nobrl) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_BRL; + if (pvolume_info->nohandlecache) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_HANDLE_CACHE; if (pvolume_info->nostrictsync) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NOSSYNC; if (pvolume_info->mand_lock) @@ -3928,6 +3984,12 @@ try_mount_again: goto remote_path_check; } +#ifdef CONFIG_CIFS_SMB311 + /* if new SMB3.11 POSIX extensions are supported do not remap / and \ */ + if (tcon->posix_extensions) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_POSIX_PATHS; +#endif /* SMB3.11 */ + /* tell server which Unix caps we support */ if (cap_unix(tcon->ses)) { /* reset of caps checks mount to see if unix extensions @@ -4359,6 +4421,7 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) vol_info->UNC = master_tcon->treeName; vol_info->retry = master_tcon->retry; vol_info->nocase = master_tcon->nocase; + vol_info->nohandlecache = master_tcon->nohandlecache; vol_info->local_lease = master_tcon->local_lease; vol_info->no_linux_ext = !master_tcon->unix_ext; vol_info->sectype = master_tcon->ses->sectype; @@ -4388,8 +4451,14 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, kuid_t fsuid) goto out; } +#ifdef CONFIG_CIFS_SMB311 + /* if new SMB3.11 POSIX extensions are supported do not remap / and \ */ + if (tcon->posix_extensions) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_POSIX_PATHS; +#endif /* SMB3.11 */ if (cap_unix(ses)) reset_cifs_unix_caps(0, tcon, NULL, vol_info); + out: kfree(vol_info->username); kzfree(vol_info->password); diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 81ba6e0d88d8..ddae52bd1993 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -369,7 +369,7 @@ cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned int xid, oparms.path = full_path; oparms.fid = fid; oparms.reconnect = false; - + oparms.mode = mode; rc = server->ops->open(xid, &oparms, oplock, buf); if (rc) { cifs_dbg(FYI, "cifs_create returned 0x%x\n", rc); @@ -684,6 +684,9 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, umode_t mode, goto mknod_out; } + if (!S_ISCHR(mode) && !S_ISBLK(mode)) + goto mknod_out; + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL)) goto mknod_out; @@ -692,10 +695,8 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, umode_t mode, buf = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL); if (buf == NULL) { - kfree(full_path); rc = -ENOMEM; - free_xid(xid); - return rc; + goto mknod_out; } if (backup_cred(cifs_sb)) @@ -742,7 +743,7 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, umode_t mode, pdev->minor = cpu_to_le64(MINOR(device_number)); rc = tcon->ses->server->ops->sync_write(xid, &fid, &io_parms, &bytes_written, iov, 1); - } /* else if (S_ISFIFO) */ + } tcon->ses->server->ops->close(xid, tcon, &fid); d_drop(direntry); @@ -779,21 +780,25 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) { free_xid(xid); - return (struct dentry *)tlink; + return ERR_CAST(tlink); } pTcon = tlink_tcon(tlink); rc = check_name(direntry, pTcon); - if (rc) - goto lookup_out; + if (unlikely(rc)) { + cifs_put_tlink(tlink); + free_xid(xid); + return ERR_PTR(rc); + } /* can not grab the rename sem here since it would deadlock in the cases (beginning of sys_rename itself) in which we already have the sb rename sem */ full_path = build_path_from_dentry(direntry); if (full_path == NULL) { - rc = -ENOMEM; - goto lookup_out; + cifs_put_tlink(tlink); + free_xid(xid); + return ERR_PTR(-ENOMEM); } if (d_really_is_positive(direntry)) { @@ -812,29 +817,25 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, parent_dir_inode->i_sb, xid, NULL); } - if ((rc == 0) && (newInode != NULL)) { - d_add(direntry, newInode); + if (rc == 0) { /* since paths are not looked up by component - the parent directories are presumed to be good here */ renew_parental_timestamps(direntry); - } else if (rc == -ENOENT) { - rc = 0; cifs_set_time(direntry, jiffies); - d_add(direntry, NULL); - /* if it was once a directory (but how can we tell?) we could do - shrink_dcache_parent(direntry); */ - } else if (rc != -EACCES) { - cifs_dbg(FYI, "Unexpected lookup error %d\n", rc); - /* We special case check for Access Denied - since that - is a common return code */ + newInode = NULL; + } else { + if (rc != -EACCES) { + cifs_dbg(FYI, "Unexpected lookup error %d\n", rc); + /* We special case check for Access Denied - since that + is a common return code */ + } + newInode = ERR_PTR(rc); } - -lookup_out: kfree(full_path); cifs_put_tlink(tlink); free_xid(xid); - return ERR_PTR(rc); + return d_splice_alias(newInode, direntry); } static int diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 7cee97b93a61..87eece6fbd48 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1987,11 +1987,10 @@ wdata_prepare_pages(struct cifs_writedata *wdata, unsigned int found_pages, for (i = 0; i < found_pages; i++) { page = wdata->pages[i]; /* - * At this point we hold neither mapping->tree_lock nor - * lock on the page itself: the page may be truncated or - * invalidated (changing page->mapping to NULL), or even - * swizzled back from swapper_space to tmpfs file - * mapping + * At this point we hold neither the i_pages lock nor the + * page lock: the page may be truncated or invalidated + * (changing page->mapping to NULL), or even swizzled + * back from swapper_space to tmpfs file mapping */ if (nr_pages == 0) @@ -2881,13 +2880,13 @@ out: } static struct cifs_readdata * -cifs_readdata_alloc(unsigned int nr_pages, work_func_t complete) +cifs_readdata_direct_alloc(struct page **pages, work_func_t complete) { struct cifs_readdata *rdata; - rdata = kzalloc(sizeof(*rdata) + (sizeof(struct page *) * nr_pages), - GFP_KERNEL); + rdata = kzalloc(sizeof(*rdata), GFP_KERNEL); if (rdata != NULL) { + rdata->pages = pages; kref_init(&rdata->refcount); INIT_LIST_HEAD(&rdata->list); init_completion(&rdata->done); @@ -2897,6 +2896,22 @@ cifs_readdata_alloc(unsigned int nr_pages, work_func_t complete) return rdata; } +static struct cifs_readdata * +cifs_readdata_alloc(unsigned int nr_pages, work_func_t complete) +{ + struct page **pages = + kzalloc(sizeof(struct page *) * nr_pages, GFP_KERNEL); + struct cifs_readdata *ret = NULL; + + if (pages) { + ret = cifs_readdata_direct_alloc(pages, complete); + if (!ret) + kfree(pages); + } + + return ret; +} + void cifs_readdata_release(struct kref *refcount) { @@ -2911,6 +2926,7 @@ cifs_readdata_release(struct kref *refcount) if (rdata->cfile) cifsFileInfo_put(rdata->cfile); + kvfree(rdata->pages); kfree(rdata); } @@ -3010,12 +3026,20 @@ uncached_fill_pages(struct TCP_Server_Info *server, int result = 0; unsigned int i; unsigned int nr_pages = rdata->nr_pages; + unsigned int page_offset = rdata->page_offset; rdata->got_bytes = 0; rdata->tailsz = PAGE_SIZE; for (i = 0; i < nr_pages; i++) { struct page *page = rdata->pages[i]; size_t n; + unsigned int segment_size = rdata->pagesz; + + if (i == 0) + segment_size -= page_offset; + else + page_offset = 0; + if (len <= 0) { /* no need to hold page hostage */ @@ -3024,24 +3048,25 @@ uncached_fill_pages(struct TCP_Server_Info *server, put_page(page); continue; } + n = len; - if (len >= PAGE_SIZE) { + if (len >= segment_size) /* enough data to fill the page */ - n = PAGE_SIZE; - len -= n; - } else { - zero_user(page, len, PAGE_SIZE - len); + n = segment_size; + else rdata->tailsz = len; - len = 0; - } + len -= n; + if (iter) - result = copy_page_from_iter(page, 0, n, iter); + result = copy_page_from_iter( + page, page_offset, n, iter); #ifdef CONFIG_CIFS_SMB_DIRECT else if (rdata->mr) result = n; #endif else - result = cifs_read_page_from_socket(server, page, n); + result = cifs_read_page_from_socket( + server, page, page_offset, n); if (result < 0) break; @@ -3114,6 +3139,7 @@ cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file, rdata->bytes = cur_len; rdata->pid = pid; rdata->pagesz = PAGE_SIZE; + rdata->tailsz = PAGE_SIZE; rdata->read_into_pages = cifs_uncached_read_into_pages; rdata->copy_into_pages = cifs_uncached_copy_into_pages; rdata->credits = credits; @@ -3463,7 +3489,7 @@ cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset) * If the page is mmap'ed into a process' page tables, then we need to make * sure that it doesn't change while being written back. */ -static int +static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf) { struct page *page = vmf->page; @@ -3558,6 +3584,7 @@ readpages_fill_pages(struct TCP_Server_Info *server, u64 eof; pgoff_t eof_index; unsigned int nr_pages = rdata->nr_pages; + unsigned int page_offset = rdata->page_offset; /* determine the eof that the server (probably) has */ eof = CIFS_I(rdata->mapping->host)->server_eof; @@ -3568,13 +3595,21 @@ readpages_fill_pages(struct TCP_Server_Info *server, rdata->tailsz = PAGE_SIZE; for (i = 0; i < nr_pages; i++) { struct page *page = rdata->pages[i]; - size_t n = PAGE_SIZE; + unsigned int to_read = rdata->pagesz; + size_t n; + + if (i == 0) + to_read -= page_offset; + else + page_offset = 0; + + n = to_read; - if (len >= PAGE_SIZE) { - len -= PAGE_SIZE; + if (len >= to_read) { + len -= to_read; } else if (len > 0) { /* enough for partial page, fill and zero the rest */ - zero_user(page, len, PAGE_SIZE - len); + zero_user(page, len + page_offset, to_read - len); n = rdata->tailsz = len; len = 0; } else if (page->index > eof_index) { @@ -3606,13 +3641,15 @@ readpages_fill_pages(struct TCP_Server_Info *server, } if (iter) - result = copy_page_from_iter(page, 0, n, iter); + result = copy_page_from_iter( + page, page_offset, n, iter); #ifdef CONFIG_CIFS_SMB_DIRECT else if (rdata->mr) result = n; #endif else - result = cifs_read_page_from_socket(server, page, n); + result = cifs_read_page_from_socket( + server, page, page_offset, n); if (result < 0) break; @@ -3791,6 +3828,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, rdata->bytes = bytes; rdata->pid = pid; rdata->pagesz = PAGE_SIZE; + rdata->tailsz = PAGE_SIZE; rdata->read_into_pages = cifs_readpages_read_into_pages; rdata->copy_into_pages = cifs_readpages_copy_into_pages; rdata->credits = credits; diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c index 8d4b7bc8ae91..25d3f66b2d50 100644 --- a/fs/cifs/fscache.c +++ b/fs/cifs/fscache.c @@ -23,11 +23,63 @@ #include "cifs_debug.h" #include "cifs_fs_sb.h" +/* + * Key layout of CIFS server cache index object + */ +struct cifs_server_key { + struct { + uint16_t family; /* address family */ + __be16 port; /* IP port */ + } hdr; + union { + struct in_addr ipv4_addr; + struct in6_addr ipv6_addr; + }; +} __packed; + +/* + * Get a cookie for a server object keyed by {IPaddress,port,family} tuple + */ void cifs_fscache_get_client_cookie(struct TCP_Server_Info *server) { + const struct sockaddr *sa = (struct sockaddr *) &server->dstaddr; + const struct sockaddr_in *addr = (struct sockaddr_in *) sa; + const struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *) sa; + struct cifs_server_key key; + uint16_t key_len = sizeof(key.hdr); + + memset(&key, 0, sizeof(key)); + + /* + * Should not be a problem as sin_family/sin6_family overlays + * sa_family field + */ + key.hdr.family = sa->sa_family; + switch (sa->sa_family) { + case AF_INET: + key.hdr.port = addr->sin_port; + key.ipv4_addr = addr->sin_addr; + key_len += sizeof(key.ipv4_addr); + break; + + case AF_INET6: + key.hdr.port = addr6->sin6_port; + key.ipv6_addr = addr6->sin6_addr; + key_len += sizeof(key.ipv6_addr); + break; + + default: + cifs_dbg(VFS, "Unknown network family '%d'\n", sa->sa_family); + server->fscache = NULL; + return; + } + server->fscache = fscache_acquire_cookie(cifs_fscache_netfs.primary_index, - &cifs_fscache_server_index_def, server, true); + &cifs_fscache_server_index_def, + &key, key_len, + NULL, 0, + server, 0, true); cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", __func__, server, server->fscache); } @@ -36,17 +88,29 @@ void cifs_fscache_release_client_cookie(struct TCP_Server_Info *server) { cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", __func__, server, server->fscache); - fscache_relinquish_cookie(server->fscache, 0); + fscache_relinquish_cookie(server->fscache, NULL, false); server->fscache = NULL; } void cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) { struct TCP_Server_Info *server = tcon->ses->server; + char *sharename; + + sharename = extract_sharename(tcon->treeName); + if (IS_ERR(sharename)) { + cifs_dbg(FYI, "%s: couldn't extract sharename\n", __func__); + tcon->fscache = NULL; + return; + } tcon->fscache = fscache_acquire_cookie(server->fscache, - &cifs_fscache_super_index_def, tcon, true); + &cifs_fscache_super_index_def, + sharename, strlen(sharename), + &tcon->resource_id, sizeof(tcon->resource_id), + tcon, 0, true); + kfree(sharename); cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", __func__, server->fscache, tcon->fscache); } @@ -54,10 +118,28 @@ void cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) void cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) { cifs_dbg(FYI, "%s: (0x%p)\n", __func__, tcon->fscache); - fscache_relinquish_cookie(tcon->fscache, 0); + fscache_relinquish_cookie(tcon->fscache, &tcon->resource_id, false); tcon->fscache = NULL; } +static void cifs_fscache_acquire_inode_cookie(struct cifsInodeInfo *cifsi, + struct cifs_tcon *tcon) +{ + struct cifs_fscache_inode_auxdata auxdata; + + memset(&auxdata, 0, sizeof(auxdata)); + auxdata.eof = cifsi->server_eof; + auxdata.last_write_time = cifsi->vfs_inode.i_mtime; + auxdata.last_change_time = cifsi->vfs_inode.i_ctime; + + cifsi->fscache = + fscache_acquire_cookie(tcon->fscache, + &cifs_fscache_inode_object_def, + &cifsi->uniqueid, sizeof(cifsi->uniqueid), + &auxdata, sizeof(auxdata), + cifsi, cifsi->vfs_inode.i_size, true); +} + static void cifs_fscache_enable_inode_cookie(struct inode *inode) { struct cifsInodeInfo *cifsi = CIFS_I(inode); @@ -67,21 +149,28 @@ static void cifs_fscache_enable_inode_cookie(struct inode *inode) if (cifsi->fscache) return; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) { - cifsi->fscache = fscache_acquire_cookie(tcon->fscache, - &cifs_fscache_inode_object_def, cifsi, true); - cifs_dbg(FYI, "%s: got FH cookie (0x%p/0x%p)\n", - __func__, tcon->fscache, cifsi->fscache); - } + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE)) + return; + + cifs_fscache_acquire_inode_cookie(cifsi, tcon); + + cifs_dbg(FYI, "%s: got FH cookie (0x%p/0x%p)\n", + __func__, tcon->fscache, cifsi->fscache); } void cifs_fscache_release_inode_cookie(struct inode *inode) { + struct cifs_fscache_inode_auxdata auxdata; struct cifsInodeInfo *cifsi = CIFS_I(inode); if (cifsi->fscache) { + memset(&auxdata, 0, sizeof(auxdata)); + auxdata.eof = cifsi->server_eof; + auxdata.last_write_time = cifsi->vfs_inode.i_mtime; + auxdata.last_change_time = cifsi->vfs_inode.i_ctime; + cifs_dbg(FYI, "%s: (0x%p)\n", __func__, cifsi->fscache); - fscache_relinquish_cookie(cifsi->fscache, 0); + fscache_relinquish_cookie(cifsi->fscache, &auxdata, false); cifsi->fscache = NULL; } } @@ -93,7 +182,7 @@ static void cifs_fscache_disable_inode_cookie(struct inode *inode) if (cifsi->fscache) { cifs_dbg(FYI, "%s: (0x%p)\n", __func__, cifsi->fscache); fscache_uncache_all_inode_pages(cifsi->fscache, inode); - fscache_relinquish_cookie(cifsi->fscache, 1); + fscache_relinquish_cookie(cifsi->fscache, NULL, true); cifsi->fscache = NULL; } } @@ -110,16 +199,14 @@ void cifs_fscache_reset_inode_cookie(struct inode *inode) { struct cifsInodeInfo *cifsi = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct fscache_cookie *old = cifsi->fscache; if (cifsi->fscache) { /* retire the current fscache cache and get a new one */ - fscache_relinquish_cookie(cifsi->fscache, 1); + fscache_relinquish_cookie(cifsi->fscache, NULL, true); - cifsi->fscache = fscache_acquire_cookie( - cifs_sb_master_tcon(cifs_sb)->fscache, - &cifs_fscache_inode_object_def, - cifsi, true); + cifs_fscache_acquire_inode_cookie(cifsi, tcon); cifs_dbg(FYI, "%s: new cookie 0x%p oldcookie 0x%p\n", __func__, cifsi->fscache, old); } @@ -214,13 +301,15 @@ int __cifs_readpages_from_fscache(struct inode *inode, void __cifs_readpage_to_fscache(struct inode *inode, struct page *page) { + struct cifsInodeInfo *cifsi = CIFS_I(inode); int ret; cifs_dbg(FYI, "%s: (fsc: %p, p: %p, i: %p)\n", - __func__, CIFS_I(inode)->fscache, page, inode); - ret = fscache_write_page(CIFS_I(inode)->fscache, page, GFP_KERNEL); + __func__, cifsi->fscache, page, inode); + ret = fscache_write_page(cifsi->fscache, page, + cifsi->vfs_inode.i_size, GFP_KERNEL); if (ret != 0) - fscache_uncache_page(CIFS_I(inode)->fscache, page); + fscache_uncache_page(cifsi->fscache, page); } void __cifs_fscache_readpages_cancel(struct inode *inode, struct list_head *pages) @@ -239,4 +328,3 @@ void __cifs_fscache_invalidate_page(struct page *page, struct inode *inode) fscache_wait_on_page_write(cookie, page); fscache_uncache_page(cookie, page); } - diff --git a/fs/cifs/fscache.h b/fs/cifs/fscache.h index 24794b6cd8ec..c7e3ac251e16 100644 --- a/fs/cifs/fscache.h +++ b/fs/cifs/fscache.h @@ -27,6 +27,18 @@ #ifdef CONFIG_CIFS_FSCACHE +/* + * Auxiliary data attached to CIFS inode within the cache + */ +struct cifs_fscache_inode_auxdata { + struct timespec last_write_time; + struct timespec last_change_time; + u64 eof; +}; + +/* + * cache.c + */ extern struct fscache_netfs cifs_fscache_netfs; extern const struct fscache_cookie_def cifs_fscache_server_index_def; extern const struct fscache_cookie_def cifs_fscache_super_index_def; @@ -34,6 +46,7 @@ extern const struct fscache_cookie_def cifs_fscache_inode_object_def; extern int cifs_fscache_register(void); extern void cifs_fscache_unregister(void); +extern char *extract_sharename(const char *); /* * fscache.c diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index f856df4adae3..745fd7fe8d0e 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -710,7 +710,7 @@ cgfi_exit: /* Simple function to return a 64 bit hash of string. Rarely called */ static __u64 simple_hashstr(const char *str) { - const __u64 hash_mult = 1125899906842597L; /* a big enough prime */ + const __u64 hash_mult = 1125899906842597ULL; /* a big enough prime */ __u64 hash = 0; while (*str) @@ -746,7 +746,8 @@ cifs_get_inode_info(struct inode **inode, const char *full_path, cifs_dbg(FYI, "Getting info on %s\n", full_path); if ((data == NULL) && (*inode != NULL)) { - if (CIFS_CACHE_READ(CIFS_I(*inode))) { + if (CIFS_CACHE_READ(CIFS_I(*inode)) && + CIFS_I(*inode)->time != 0) { cifs_dbg(FYI, "No need to revalidate cached inode sizes\n"); goto cgii_exit; } @@ -1857,15 +1858,15 @@ cifs_inode_needs_reval(struct inode *inode) struct cifsInodeInfo *cifs_i = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + if (cifs_i->time == 0) + return true; + if (CIFS_CACHE_READ(cifs_i)) return false; if (!lookupCacheEnabled) return true; - if (cifs_i->time == 0) - return true; - if (!cifs_sb->actimeo) return true; @@ -2104,10 +2105,14 @@ static int cifs_truncate_page(struct address_space *mapping, loff_t from) static void cifs_setsize(struct inode *inode, loff_t offset) { + struct cifsInodeInfo *cifs_i = CIFS_I(inode); + spin_lock(&inode->i_lock); i_size_write(inode, offset); spin_unlock(&inode->i_lock); + /* Cached inode must be refreshed on truncate */ + cifs_i->time = 0; truncate_pagecache(inode, offset); } diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 460084a8eac5..aba3fc3058da 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -117,6 +117,8 @@ tconInfoAlloc(void) INIT_LIST_HEAD(&ret_buf->openFileList); INIT_LIST_HEAD(&ret_buf->tcon_list); spin_lock_init(&ret_buf->open_file_lock); + mutex_init(&ret_buf->prfid_mutex); + ret_buf->prfid = kzalloc(sizeof(struct cifs_fid), GFP_KERNEL); #ifdef CONFIG_CIFS_STATS spin_lock_init(&ret_buf->stat_lock); #endif @@ -134,6 +136,7 @@ tconInfoFree(struct cifs_tcon *buf_to_free) atomic_dec(&tconInfoAllocCount); kfree(buf_to_free->nativeFileSystem); kzfree(buf_to_free->password); + kfree(buf_to_free->prfid); kfree(buf_to_free); } @@ -145,7 +148,7 @@ cifs_buf_get(void) * SMB2 header is bigger than CIFS one - no problems to clean some * more bytes for CIFS. */ - size_t buf_size = sizeof(struct smb2_hdr); + size_t buf_size = sizeof(struct smb2_sync_hdr); /* * We could use negotiated size instead of max_msgsize - @@ -339,7 +342,7 @@ checkSMB(char *buf, unsigned int total_read, struct TCP_Server_Info *server) /* otherwise, there is enough to get to the BCC */ if (check_smb_hdr(smb)) return -EIO; - clc_len = smbCalcSize(smb); + clc_len = smbCalcSize(smb, server); if (4 + rfclen != total_read) { cifs_dbg(VFS, "Length read does not match RFC1001 length %d\n", diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index cc88f4f0325e..d7ad0dfe4e68 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -903,7 +903,7 @@ map_smb_to_linux_error(char *buf, bool logErr) * portion, the number of word parameters and the data portion of the message */ unsigned int -smbCalcSize(void *buf) +smbCalcSize(void *buf, struct TCP_Server_Info *server) { struct smb_hdr *ptr = (struct smb_hdr *)buf; return (sizeof(struct smb_hdr) + (2 * ptr->WordCount) + diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index a27fc8791551..eeab81c9452f 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -650,7 +650,8 @@ find_cifs_entry(const unsigned int xid, struct cifs_tcon *tcon, loff_t pos, char *cur_ent; char *end_of_smb = cfile->srch_inf.ntwrk_buf_start + server->ops->calc_smb_size( - cfile->srch_inf.ntwrk_buf_start); + cfile->srch_inf.ntwrk_buf_start, + server); cur_ent = cfile->srch_inf.srch_entries_start; first_entry_in_buffer = cfile->srch_inf.index_of_last_entry @@ -831,7 +832,8 @@ int cifs_readdir(struct file *file, struct dir_context *ctx) cifs_dbg(FYI, "loop through %d times filling dir for net buf %p\n", num_to_fill, cifsFile->srch_inf.ntwrk_buf_start); max_len = tcon->ses->server->ops->calc_smb_size( - cifsFile->srch_inf.ntwrk_buf_start); + cifsFile->srch_inf.ntwrk_buf_start, + tcon->ses->server); end_of_smb = cifsFile->srch_inf.ntwrk_buf_start + max_len; tmp_buf = kmalloc(UNICODE_NAME_MAX, GFP_KERNEL); diff --git a/fs/cifs/smb2glob.h b/fs/cifs/smb2glob.h index 401a5d856636..0ffa18094335 100644 --- a/fs/cifs/smb2glob.h +++ b/fs/cifs/smb2glob.h @@ -61,9 +61,4 @@ /* Maximum buffer size value we can send with 1 credit */ #define SMB2_MAX_BUFFER_SIZE 65536 -static inline struct smb2_sync_hdr *get_sync_hdr(void *buf) -{ - return &(((struct smb2_hdr *)buf)->sync_hdr); -} - #endif /* _SMB2_GLOB_H */ diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c index 1238cd3552f9..a6e786e39248 100644 --- a/fs/cifs/smb2inode.c +++ b/fs/cifs/smb2inode.c @@ -44,26 +44,38 @@ smb2_open_op_close(const unsigned int xid, struct cifs_tcon *tcon, __u32 create_options, void *data, int command) { int rc, tmprc = 0; - __le16 *utf16_path; + __le16 *utf16_path = NULL; __u8 oplock = SMB2_OPLOCK_LEVEL_NONE; struct cifs_open_parms oparms; struct cifs_fid fid; + bool use_cached_root_handle = false; - utf16_path = cifs_convert_path_to_utf16(full_path, cifs_sb); - if (!utf16_path) - return -ENOMEM; + if ((strcmp(full_path, "") == 0) && (create_options == 0) && + (desired_access == FILE_READ_ATTRIBUTES) && + (create_disposition == FILE_OPEN) && + (tcon->nohandlecache == false)) { + rc = open_shroot(xid, tcon, &fid); + if (rc == 0) + use_cached_root_handle = true; + } + + if (use_cached_root_handle == false) { + utf16_path = cifs_convert_path_to_utf16(full_path, cifs_sb); + if (!utf16_path) + return -ENOMEM; - oparms.tcon = tcon; - oparms.desired_access = desired_access; - oparms.disposition = create_disposition; - oparms.create_options = create_options; - oparms.fid = &fid; - oparms.reconnect = false; + oparms.tcon = tcon; + oparms.desired_access = desired_access; + oparms.disposition = create_disposition; + oparms.create_options = create_options; + oparms.fid = &fid; + oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL); - if (rc) { - kfree(utf16_path); - return rc; + rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL); + if (rc) { + kfree(utf16_path); + return rc; + } } switch (command) { @@ -107,7 +119,8 @@ smb2_open_op_close(const unsigned int xid, struct cifs_tcon *tcon, break; } - rc = SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid); + if (use_cached_root_handle == false) + rc = SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid); if (tmprc) rc = tmprc; kfree(utf16_path); diff --git a/fs/cifs/smb2maperror.c b/fs/cifs/smb2maperror.c index 3bfc9c990724..20a2d304c603 100644 --- a/fs/cifs/smb2maperror.c +++ b/fs/cifs/smb2maperror.c @@ -27,6 +27,7 @@ #include "smb2proto.h" #include "smb2status.h" #include "smb2glob.h" +#include "trace.h" struct status_to_posix_error { __le32 smb2_status; @@ -2450,13 +2451,16 @@ smb2_print_status(__le32 status) int map_smb2_to_linux_error(char *buf, bool log_err) { - struct smb2_sync_hdr *shdr = get_sync_hdr(buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; unsigned int i; int rc = -EIO; __le32 smb2err = shdr->Status; - if (smb2err == 0) + if (smb2err == 0) { + trace_smb3_cmd_done(shdr->TreeId, shdr->SessionId, + le16_to_cpu(shdr->Command), le64_to_cpu(shdr->MessageId)); return 0; + } /* mask facility */ if (log_err && (smb2err != STATUS_MORE_PROCESSING_REQUIRED) && @@ -2478,5 +2482,8 @@ map_smb2_to_linux_error(char *buf, bool log_err) cifs_dbg(FYI, "Mapping SMB2 status code 0x%08x to POSIX err %d\n", __le32_to_cpu(smb2err), rc); + trace_smb3_cmd_err(shdr->TreeId, shdr->SessionId, + le16_to_cpu(shdr->Command), + le64_to_cpu(shdr->MessageId), le32_to_cpu(smb2err), rc); return rc; } diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c index 5406e95f5d92..cb5728e3d87d 100644 --- a/fs/cifs/smb2misc.c +++ b/fs/cifs/smb2misc.c @@ -93,26 +93,57 @@ static const __le16 smb2_rsp_struct_sizes[NUMBER_OF_SMB2_COMMANDS] = { /* SMB2_OPLOCK_BREAK */ cpu_to_le16(24) }; +#ifdef CONFIG_CIFS_SMB311 +static __u32 get_neg_ctxt_len(struct smb2_sync_hdr *hdr, __u32 len, + __u32 non_ctxlen) +{ + __u16 neg_count; + __u32 nc_offset, size_of_pad_before_neg_ctxts; + struct smb2_negotiate_rsp *pneg_rsp = (struct smb2_negotiate_rsp *)hdr; + + /* Negotiate contexts are only valid for latest dialect SMB3.11 */ + neg_count = le16_to_cpu(pneg_rsp->NegotiateContextCount); + if ((neg_count == 0) || + (pneg_rsp->DialectRevision != cpu_to_le16(SMB311_PROT_ID))) + return 0; + + /* Make sure that negotiate contexts start after gss security blob */ + nc_offset = le32_to_cpu(pneg_rsp->NegotiateContextOffset); + if (nc_offset < non_ctxlen) { + printk_once(KERN_WARNING "invalid negotiate context offset\n"); + return 0; + } + size_of_pad_before_neg_ctxts = nc_offset - non_ctxlen; + + /* Verify that at least minimal negotiate contexts fit within frame */ + if (len < nc_offset + (neg_count * sizeof(struct smb2_neg_context))) { + printk_once(KERN_WARNING "negotiate context goes beyond end\n"); + return 0; + } + + cifs_dbg(FYI, "length of negcontexts %d pad %d\n", + len - nc_offset, size_of_pad_before_neg_ctxts); + + /* length of negcontexts including pad from end of sec blob to them */ + return (len - nc_offset) + size_of_pad_before_neg_ctxts; +} +#endif /* CIFS_SMB311 */ + int -smb2_check_message(char *buf, unsigned int length, struct TCP_Server_Info *srvr) +smb2_check_message(char *buf, unsigned int len, struct TCP_Server_Info *srvr) { - struct smb2_pdu *pdu = (struct smb2_pdu *)buf; - struct smb2_hdr *hdr = &pdu->hdr; - struct smb2_sync_hdr *shdr = get_sync_hdr(buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; + struct smb2_sync_pdu *pdu = (struct smb2_sync_pdu *)shdr; __u64 mid; - __u32 len = get_rfc1002_length(buf); __u32 clc_len; /* calculated length */ int command; - - /* BB disable following printk later */ - cifs_dbg(FYI, "%s length: 0x%x, smb_buf_length: 0x%x\n", - __func__, length, len); + int pdu_size = sizeof(struct smb2_sync_pdu); + int hdr_size = sizeof(struct smb2_sync_hdr); /* * Add function to do table lookup of StructureSize by command * ie Validate the wct via smb2_struct_sizes table above */ - if (shdr->ProtocolId == SMB2_TRANSFORM_PROTO_NUM) { struct smb2_transform_hdr *thdr = (struct smb2_transform_hdr *)buf; @@ -136,8 +167,8 @@ smb2_check_message(char *buf, unsigned int length, struct TCP_Server_Info *srvr) } mid = le64_to_cpu(shdr->MessageId); - if (length < sizeof(struct smb2_pdu)) { - if ((length >= sizeof(struct smb2_hdr)) + if (len < pdu_size) { + if ((len >= hdr_size) && (shdr->Status != 0)) { pdu->StructureSize2 = 0; /* @@ -150,8 +181,7 @@ smb2_check_message(char *buf, unsigned int length, struct TCP_Server_Info *srvr) } return 1; } - if (len > CIFSMaxBufSize + MAX_SMB2_HDR_SIZE - - srvr->vals->header_preamble_size) { + if (len > CIFSMaxBufSize + MAX_SMB2_HDR_SIZE) { cifs_dbg(VFS, "SMB length greater than maximum, mid=%llu\n", mid); return 1; @@ -190,39 +220,38 @@ smb2_check_message(char *buf, unsigned int length, struct TCP_Server_Info *srvr) } } - if (srvr->vals->header_preamble_size + len != length) { - cifs_dbg(VFS, "Total length %u RFC1002 length %zu mismatch mid %llu\n", - length, srvr->vals->header_preamble_size + len, mid); - return 1; - } - - clc_len = smb2_calc_size(hdr); + clc_len = smb2_calc_size(buf, srvr); - if (srvr->vals->header_preamble_size + len != clc_len) { - cifs_dbg(FYI, "Calculated size %u length %zu mismatch mid %llu\n", - clc_len, srvr->vals->header_preamble_size + len, mid); +#ifdef CONFIG_CIFS_SMB311 + if (shdr->Command == SMB2_NEGOTIATE) + clc_len += get_neg_ctxt_len(shdr, len, clc_len); +#endif /* SMB311 */ + if (len != clc_len) { + cifs_dbg(FYI, "Calculated size %u length %u mismatch mid %llu\n", + clc_len, len, mid); /* create failed on symlink */ if (command == SMB2_CREATE_HE && shdr->Status == STATUS_STOPPED_ON_SYMLINK) return 0; /* Windows 7 server returns 24 bytes more */ - if (clc_len + 24 - srvr->vals->header_preamble_size == len && command == SMB2_OPLOCK_BREAK_HE) + if (clc_len + 24 == len && command == SMB2_OPLOCK_BREAK_HE) return 0; /* server can return one byte more due to implied bcc[0] */ - if (clc_len == srvr->vals->header_preamble_size + len + 1) + if (clc_len == len + 1) return 0; /* * MacOS server pads after SMB2.1 write response with 3 bytes * of junk. Other servers match RFC1001 len to actual * SMB2/SMB3 frame length (header + smb2 response specific data) + * Some windows servers do too when compounding is used. * Log the server error (once), but allow it and continue * since the frame is parseable. */ - if (clc_len < srvr->vals->header_preamble_size /* RFC1001 header size */ + len) { + if (clc_len < len) { printk_once(KERN_WARNING - "SMB2 server sent bad RFC1001 len %d not %zu\n", - len, clc_len - srvr->vals->header_preamble_size); + "SMB2 server sent bad RFC1001 len %d not %d\n", + len, clc_len); return 0; } @@ -263,15 +292,14 @@ static const bool has_smb2_data_area[NUMBER_OF_SMB2_COMMANDS] = { * area and the offset to it (from the beginning of the smb are also returned. */ char * -smb2_get_data_area_len(int *off, int *len, struct smb2_hdr *hdr) +smb2_get_data_area_len(int *off, int *len, struct smb2_sync_hdr *shdr) { - struct smb2_sync_hdr *shdr = get_sync_hdr(hdr); *off = 0; *len = 0; /* error responses do not have data area */ if (shdr->Status && shdr->Status != STATUS_MORE_PROCESSING_REQUIRED && - (((struct smb2_err_rsp *)hdr)->StructureSize) == + (((struct smb2_err_rsp *)shdr)->StructureSize) == SMB2_ERROR_STRUCTURE_SIZE2) return NULL; @@ -283,42 +311,44 @@ smb2_get_data_area_len(int *off, int *len, struct smb2_hdr *hdr) switch (shdr->Command) { case SMB2_NEGOTIATE: *off = le16_to_cpu( - ((struct smb2_negotiate_rsp *)hdr)->SecurityBufferOffset); + ((struct smb2_negotiate_rsp *)shdr)->SecurityBufferOffset); *len = le16_to_cpu( - ((struct smb2_negotiate_rsp *)hdr)->SecurityBufferLength); + ((struct smb2_negotiate_rsp *)shdr)->SecurityBufferLength); break; case SMB2_SESSION_SETUP: *off = le16_to_cpu( - ((struct smb2_sess_setup_rsp *)hdr)->SecurityBufferOffset); + ((struct smb2_sess_setup_rsp *)shdr)->SecurityBufferOffset); *len = le16_to_cpu( - ((struct smb2_sess_setup_rsp *)hdr)->SecurityBufferLength); + ((struct smb2_sess_setup_rsp *)shdr)->SecurityBufferLength); break; case SMB2_CREATE: *off = le32_to_cpu( - ((struct smb2_create_rsp *)hdr)->CreateContextsOffset); + ((struct smb2_create_rsp *)shdr)->CreateContextsOffset); *len = le32_to_cpu( - ((struct smb2_create_rsp *)hdr)->CreateContextsLength); + ((struct smb2_create_rsp *)shdr)->CreateContextsLength); break; case SMB2_QUERY_INFO: *off = le16_to_cpu( - ((struct smb2_query_info_rsp *)hdr)->OutputBufferOffset); + ((struct smb2_query_info_rsp *)shdr)->OutputBufferOffset); *len = le32_to_cpu( - ((struct smb2_query_info_rsp *)hdr)->OutputBufferLength); + ((struct smb2_query_info_rsp *)shdr)->OutputBufferLength); break; case SMB2_READ: - *off = ((struct smb2_read_rsp *)hdr)->DataOffset; - *len = le32_to_cpu(((struct smb2_read_rsp *)hdr)->DataLength); + /* TODO: is this a bug ? */ + *off = ((struct smb2_read_rsp *)shdr)->DataOffset; + *len = le32_to_cpu(((struct smb2_read_rsp *)shdr)->DataLength); break; case SMB2_QUERY_DIRECTORY: *off = le16_to_cpu( - ((struct smb2_query_directory_rsp *)hdr)->OutputBufferOffset); + ((struct smb2_query_directory_rsp *)shdr)->OutputBufferOffset); *len = le32_to_cpu( - ((struct smb2_query_directory_rsp *)hdr)->OutputBufferLength); + ((struct smb2_query_directory_rsp *)shdr)->OutputBufferLength); break; case SMB2_IOCTL: *off = le32_to_cpu( - ((struct smb2_ioctl_rsp *)hdr)->OutputOffset); - *len = le32_to_cpu(((struct smb2_ioctl_rsp *)hdr)->OutputCount); + ((struct smb2_ioctl_rsp *)shdr)->OutputOffset); + *len = le32_to_cpu( + ((struct smb2_ioctl_rsp *)shdr)->OutputCount); break; case SMB2_CHANGE_NOTIFY: default: @@ -361,15 +391,14 @@ smb2_get_data_area_len(int *off, int *len, struct smb2_hdr *hdr) * portion, the number of word parameters and the data portion of the message. */ unsigned int -smb2_calc_size(void *buf) +smb2_calc_size(void *buf, struct TCP_Server_Info *srvr) { - struct smb2_pdu *pdu = (struct smb2_pdu *)buf; - struct smb2_hdr *hdr = &pdu->hdr; - struct smb2_sync_hdr *shdr = get_sync_hdr(hdr); + struct smb2_sync_pdu *pdu = (struct smb2_sync_pdu *)buf; + struct smb2_sync_hdr *shdr = &pdu->sync_hdr; int offset; /* the offset from the beginning of SMB to data area */ int data_length; /* the length of the variable length data area */ /* Structure Size has already been checked to make sure it is 64 */ - int len = 4 + le16_to_cpu(shdr->StructureSize); + int len = le16_to_cpu(shdr->StructureSize); /* * StructureSize2, ie length of fixed parameter area has already @@ -380,7 +409,7 @@ smb2_calc_size(void *buf) if (has_smb2_data_area[le16_to_cpu(shdr->Command)] == false) goto calc_size_exit; - smb2_get_data_area_len(&offset, &data_length, hdr); + smb2_get_data_area_len(&offset, &data_length, shdr); cifs_dbg(FYI, "SMB2 data length %d offset %d\n", data_length, offset); if (data_length > 0) { @@ -388,15 +417,14 @@ smb2_calc_size(void *buf) * Check to make sure that data area begins after fixed area, * Note that last byte of the fixed area is part of data area * for some commands, typically those with odd StructureSize, - * so we must add one to the calculation (and 4 to account for - * the size of the RFC1001 hdr. + * so we must add one to the calculation. */ - if (offset + 4 + 1 < len) { + if (offset + 1 < len) { cifs_dbg(VFS, "data area offset %d overlaps SMB2 header %d\n", - offset + 4 + 1, len); + offset + 1, len); data_length = 0; } else { - len = 4 + offset + data_length; + len = offset + data_length; } } calc_size_exit: @@ -423,8 +451,14 @@ cifs_convert_path_to_utf16(const char *from, struct cifs_sb_info *cifs_sb) /* Windows doesn't allow paths beginning with \ */ if (from[0] == '\\') start_of_path = from + 1; +#ifdef CONFIG_CIFS_SMB311 + /* SMB311 POSIX extensions paths do not include leading slash */ + else if (cifs_sb_master_tcon(cifs_sb)->posix_extensions) + start_of_path = from + 1; +#endif /* 311 */ else start_of_path = from; + to = cifs_strndup_to_utf16(start_of_path, PATH_MAX, &len, cifs_sb->local_nls, map_type); return to; @@ -579,7 +613,7 @@ smb2_is_valid_lease_break(char *buffer) bool smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server) { - struct smb2_oplock_break_rsp *rsp = (struct smb2_oplock_break_rsp *)buffer; + struct smb2_oplock_break *rsp = (struct smb2_oplock_break *)buffer; struct list_head *tmp, *tmp1, *tmp2; struct cifs_ses *ses; struct cifs_tcon *tcon; @@ -588,7 +622,7 @@ smb2_is_valid_oplock_break(char *buffer, struct TCP_Server_Info *server) cifs_dbg(FYI, "Checking for oplock break\n"); - if (rsp->hdr.sync_hdr.Command != SMB2_OPLOCK_BREAK) + if (rsp->sync_hdr.Command != SMB2_OPLOCK_BREAK) return false; if (rsp->StructureSize != @@ -679,7 +713,7 @@ smb2_cancelled_close_fid(struct work_struct *work) int smb2_handle_cancelled_mid(char *buffer, struct TCP_Server_Info *server) { - struct smb2_sync_hdr *sync_hdr = get_sync_hdr(buffer); + struct smb2_sync_hdr *sync_hdr = (struct smb2_sync_hdr *)buffer; struct smb2_create_rsp *rsp = (struct smb2_create_rsp *)buffer; struct cifs_tcon *tcon; struct close_cancelled_open *cancelled; diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 968b1d43a1ea..950d0ab2cc61 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -123,7 +123,7 @@ smb2_get_credits_field(struct TCP_Server_Info *server, const int optype) static unsigned int smb2_get_credits(struct mid_q_entry *mid) { - struct smb2_sync_hdr *shdr = get_sync_hdr(mid->resp_buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)mid->resp_buf; return le16_to_cpu(shdr->CreditRequest); } @@ -190,7 +190,7 @@ static struct mid_q_entry * smb2_find_mid(struct TCP_Server_Info *server, char *buf) { struct mid_q_entry *mid; - struct smb2_sync_hdr *shdr = get_sync_hdr(buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; __u64 wire_mid = le64_to_cpu(shdr->MessageId); if (shdr->ProtocolId == SMB2_TRANSFORM_PROTO_NUM) { @@ -212,15 +212,16 @@ smb2_find_mid(struct TCP_Server_Info *server, char *buf) } static void -smb2_dump_detail(void *buf) +smb2_dump_detail(void *buf, struct TCP_Server_Info *server) { #ifdef CONFIG_CIFS_DEBUG2 - struct smb2_sync_hdr *shdr = get_sync_hdr(buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; cifs_dbg(VFS, "Cmd: %d Err: 0x%x Flags: 0x%x Mid: %llu Pid: %d\n", shdr->Command, shdr->Status, shdr->Flags, shdr->MessageId, shdr->ProcessId); - cifs_dbg(VFS, "smb buf %p len %u\n", buf, smb2_calc_size(buf)); + cifs_dbg(VFS, "smb buf %p len %u\n", buf, + server->ops->calc_smb_size(buf, server)); #endif } @@ -252,9 +253,14 @@ smb2_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *volume_info) wsize = volume_info->wsize ? volume_info->wsize : CIFS_DEFAULT_IOSIZE; wsize = min_t(unsigned int, wsize, server->max_write); #ifdef CONFIG_CIFS_SMB_DIRECT - if (server->rdma) - wsize = min_t(unsigned int, + if (server->rdma) { + if (server->sign) + wsize = min_t(unsigned int, + wsize, server->smbd_conn->max_fragmented_send_size); + else + wsize = min_t(unsigned int, wsize, server->smbd_conn->max_readwrite_size); + } #endif if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU)) wsize = min_t(unsigned int, wsize, SMB2_MAX_BUFFER_SIZE); @@ -272,9 +278,14 @@ smb2_negotiate_rsize(struct cifs_tcon *tcon, struct smb_vol *volume_info) rsize = volume_info->rsize ? volume_info->rsize : CIFS_DEFAULT_IOSIZE; rsize = min_t(unsigned int, rsize, server->max_read); #ifdef CONFIG_CIFS_SMB_DIRECT - if (server->rdma) - rsize = min_t(unsigned int, + if (server->rdma) { + if (server->sign) + rsize = min_t(unsigned int, + rsize, server->smbd_conn->max_fragmented_recv_size); + else + rsize = min_t(unsigned int, rsize, server->smbd_conn->max_readwrite_size); + } #endif if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU)) @@ -312,6 +323,40 @@ SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon) } #endif /* STATS2 */ +/* + * Open the directory at the root of a share + */ +int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid) +{ + struct cifs_open_parms oparams; + int rc; + __le16 srch_path = 0; /* Null - since an open of top of share */ + u8 oplock = SMB2_OPLOCK_LEVEL_NONE; + + mutex_lock(&tcon->prfid_mutex); + if (tcon->valid_root_fid) { + cifs_dbg(FYI, "found a cached root file handle\n"); + memcpy(pfid, tcon->prfid, sizeof(struct cifs_fid)); + mutex_unlock(&tcon->prfid_mutex); + return 0; + } + + oparams.tcon = tcon; + oparams.create_options = 0; + oparams.desired_access = FILE_READ_ATTRIBUTES; + oparams.disposition = FILE_OPEN; + oparams.fid = pfid; + oparams.reconnect = false; + + rc = SMB2_open(xid, &oparams, &srch_path, &oplock, NULL, NULL); + if (rc == 0) { + memcpy(tcon->prfid, pfid, sizeof(struct cifs_fid)); + tcon->valid_root_fid = true; + } + mutex_unlock(&tcon->prfid_mutex); + return rc; +} + static void smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon) { @@ -320,6 +365,7 @@ smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon) u8 oplock = SMB2_OPLOCK_LEVEL_NONE; struct cifs_open_parms oparms; struct cifs_fid fid; + bool no_cached_open = tcon->nohandlecache; oparms.tcon = tcon; oparms.desired_access = FILE_READ_ATTRIBUTES; @@ -328,7 +374,11 @@ smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon) oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL); + if (no_cached_open) + rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL); + else + rc = open_shroot(xid, tcon, &fid); + if (rc) return; @@ -342,7 +392,8 @@ smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon) FS_DEVICE_INFORMATION); SMB2_QFS_attr(xid, tcon, fid.persistent_fid, fid.volatile_fid, FS_SECTOR_SIZE_INFORMATION); /* SMB3 specific */ - SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid); + if (no_cached_open) + SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid); return; } @@ -384,6 +435,9 @@ smb2_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon, struct cifs_open_parms oparms; struct cifs_fid fid; + if ((*full_path == 0) && tcon->valid_root_fid) + return 0; + utf16_path = cifs_convert_path_to_utf16(full_path, cifs_sb); if (!utf16_path) return -ENOMEM; @@ -579,9 +633,15 @@ smb2_query_eas(const unsigned int xid, struct cifs_tcon *tcon, SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid); + /* + * If ea_name is NULL (listxattr) and there are no EAs, return 0 as it's + * not an error. Otherwise, the specified ea_name was not found. + */ if (!rc) rc = move_smb2_ea_to_cifs(ea_data, buf_size, smb2_data, SMB2_MAX_EA_BUF, ea_name); + else if (!ea_name && rc == -ENODATA) + rc = 0; kfree(smb2_data); return rc; @@ -688,9 +748,11 @@ smb2_dump_share_caps(struct seq_file *m, struct cifs_tcon *tcon) seq_puts(m, " TRIM-support,"); seq_printf(m, "\tShare Flags: 0x%x", tcon->share_flags); + seq_printf(m, "\n\ttid: 0x%x", tcon->tid); if (tcon->perf_sector_size) seq_printf(m, "\tOptimal sector size: 0x%x", tcon->perf_sector_size); + seq_printf(m, "\tMaximal Access: 0x%x", tcon->maximal_access); } static void @@ -1241,7 +1303,7 @@ smb2_close_dir(const unsigned int xid, struct cifs_tcon *tcon, static bool smb2_is_status_pending(char *buf, struct TCP_Server_Info *server, int length) { - struct smb2_sync_hdr *shdr = get_sync_hdr(buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; if (shdr->Status != STATUS_PENDING) return false; @@ -1259,12 +1321,13 @@ smb2_is_status_pending(char *buf, struct TCP_Server_Info *server, int length) static bool smb2_is_session_expired(char *buf) { - struct smb2_sync_hdr *shdr = get_sync_hdr(buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; - if (shdr->Status != STATUS_NETWORK_SESSION_EXPIRED) + if (shdr->Status != STATUS_NETWORK_SESSION_EXPIRED && + shdr->Status != STATUS_USER_SESSION_DELETED) return false; - cifs_dbg(FYI, "Session expired\n"); + cifs_dbg(FYI, "Session expired or deleted\n"); return true; } @@ -1451,14 +1514,13 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, __u8 oplock = SMB2_OPLOCK_LEVEL_NONE; struct cifs_open_parms oparms; struct cifs_fid fid; - struct smb2_err_rsp *err_buf = NULL; + struct kvec err_iov = {NULL, 0}; + struct smb2_err_rsp *err_buf; struct smb2_symlink_err_rsp *symlink; unsigned int sub_len; unsigned int sub_offset; unsigned int print_len; unsigned int print_offset; - struct cifs_ses *ses = tcon->ses; - struct TCP_Server_Info *server = ses->server; cifs_dbg(FYI, "%s: path: %s\n", __func__, full_path); @@ -1473,15 +1535,16 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, &err_buf); + rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, &err_iov); - if (!rc || !err_buf) { + if (!rc || !err_iov.iov_base) { kfree(utf16_path); return -ENOENT; } + err_buf = err_iov.iov_base; if (le32_to_cpu(err_buf->ByteCount) < sizeof(struct smb2_symlink_err_rsp) || - get_rfc1002_length(err_buf) + server->vals->header_preamble_size < SMB2_SYMLINK_STRUCT_SIZE) { + err_iov.iov_len < SMB2_SYMLINK_STRUCT_SIZE) { kfree(utf16_path); return -ENOENT; } @@ -1494,14 +1557,13 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, print_len = le16_to_cpu(symlink->PrintNameLength); print_offset = le16_to_cpu(symlink->PrintNameOffset); - if (get_rfc1002_length(err_buf) + server->vals->header_preamble_size < - SMB2_SYMLINK_STRUCT_SIZE + sub_offset + sub_len) { + if (err_iov.iov_len < SMB2_SYMLINK_STRUCT_SIZE + sub_offset + sub_len) { kfree(utf16_path); return -ENOENT; } - if (get_rfc1002_length(err_buf) + server->vals->header_preamble_size < - SMB2_SYMLINK_STRUCT_SIZE + print_offset + print_len) { + if (err_iov.iov_len < + SMB2_SYMLINK_STRUCT_SIZE + print_offset + print_len) { kfree(utf16_path); return -ENOENT; } @@ -1575,8 +1637,11 @@ get_smb2_acl_by_path(struct cifs_sb_info *cifs_sb, oparms.create_options = 0; utf16_path = cifs_convert_path_to_utf16(path, cifs_sb); - if (!utf16_path) - return ERR_PTR(-ENOMEM); + if (!utf16_path) { + rc = -ENOMEM; + free_xid(xid); + return ERR_PTR(rc); + } oparms.tcon = tcon; oparms.desired_access = READ_CONTROL; @@ -1634,8 +1699,11 @@ set_smb2_acl(struct cifs_ntsd *pnntsd, __u32 acllen, access_flags = WRITE_DAC; utf16_path = cifs_convert_path_to_utf16(path, cifs_sb); - if (!utf16_path) - return -ENOMEM; + if (!utf16_path) { + rc = -ENOMEM; + free_xid(xid); + return rc; + } oparms.tcon = tcon; oparms.desired_access = access_flags; @@ -1695,15 +1763,21 @@ static long smb3_zero_range(struct file *file, struct cifs_tcon *tcon, /* if file not oplocked can't be sure whether asking to extend size */ if (!CIFS_CACHE_READ(cifsi)) - if (keep_size == false) - return -EOPNOTSUPP; + if (keep_size == false) { + rc = -EOPNOTSUPP; + free_xid(xid); + return rc; + } /* * Must check if file sparse since fallocate -z (zero range) assumes * non-sparse allocation */ - if (!(cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE)) - return -EOPNOTSUPP; + if (!(cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE)) { + rc = -EOPNOTSUPP; + free_xid(xid); + return rc; + } /* * need to make sure we are not asked to extend the file since the SMB3 @@ -1712,8 +1786,11 @@ static long smb3_zero_range(struct file *file, struct cifs_tcon *tcon, * which for a non sparse file would zero the newly extended range */ if (keep_size == false) - if (i_size_read(inode) < offset + len) - return -EOPNOTSUPP; + if (i_size_read(inode) < offset + len) { + rc = -EOPNOTSUPP; + free_xid(xid); + return rc; + } cifs_dbg(FYI, "offset %lld len %lld", offset, len); @@ -1746,8 +1823,11 @@ static long smb3_punch_hole(struct file *file, struct cifs_tcon *tcon, /* Need to make file sparse, if not already, before freeing range. */ /* Consider adding equivalent for compressed since it could also work */ - if (!smb2_set_sparse(xid, tcon, cfile, inode, set_sparse)) - return -EOPNOTSUPP; + if (!smb2_set_sparse(xid, tcon, cfile, inode, set_sparse)) { + rc = -EOPNOTSUPP; + free_xid(xid); + return rc; + } cifs_dbg(FYI, "offset %lld len %lld", offset, len); @@ -1778,8 +1858,10 @@ static long smb3_simple_falloc(struct file *file, struct cifs_tcon *tcon, /* if file not oplocked can't be sure whether asking to extend size */ if (!CIFS_CACHE_READ(cifsi)) - if (keep_size == false) - return -EOPNOTSUPP; + if (keep_size == false) { + free_xid(xid); + return rc; + } /* * Files are non-sparse by default so falloc may be a no-op @@ -1788,14 +1870,16 @@ static long smb3_simple_falloc(struct file *file, struct cifs_tcon *tcon, */ if ((cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) == 0) { if (keep_size == true) - return 0; + rc = 0; /* check if extending file */ else if (i_size_read(inode) >= off + len) /* not extending file and already not sparse */ - return 0; + rc = 0; /* BB: in future add else clause to extend file */ else - return -EOPNOTSUPP; + rc = -EOPNOTSUPP; + free_xid(xid); + return rc; } if ((keep_size == true) || (i_size_read(inode) >= off + len)) { @@ -1807,8 +1891,11 @@ static long smb3_simple_falloc(struct file *file, struct cifs_tcon *tcon, * ie potentially making a few extra pages at the beginning * or end of the file non-sparse via set_sparse is harmless. */ - if ((off > 8192) || (off + len + 8192 < i_size_read(inode))) - return -EOPNOTSUPP; + if ((off > 8192) || (off + len + 8192 < i_size_read(inode))) { + rc = -EOPNOTSUPP; + free_xid(xid); + return rc; + } rc = smb2_set_sparse(xid, tcon, cfile, inode, false); } @@ -2017,7 +2104,7 @@ smb3_create_lease_buf(u8 *lease_key, u8 oplock) } static __u8 -smb2_parse_lease_buf(void *buf, unsigned int *epoch) +smb2_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) { struct create_lease *lc = (struct create_lease *)buf; @@ -2028,13 +2115,16 @@ smb2_parse_lease_buf(void *buf, unsigned int *epoch) } static __u8 -smb3_parse_lease_buf(void *buf, unsigned int *epoch) +smb3_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) { struct create_lease_v2 *lc = (struct create_lease_v2 *)buf; *epoch = le16_to_cpu(lc->lcontext.Epoch); if (lc->lcontext.LeaseFlags & SMB2_LEASE_FLAG_BREAK_IN_PROGRESS) return SMB2_OPLOCK_LEVEL_NOCHANGE; + if (lease_key) + memcpy(lease_key, &lc->lcontext.LeaseKeyLow, + SMB2_LEASE_KEY_SIZE); return le32_to_cpu(lc->lcontext.LeaseState); } @@ -2052,12 +2142,11 @@ smb2_dir_needs_close(struct cifsFileInfo *cfile) } static void -fill_transform_hdr(struct TCP_Server_Info *server, - struct smb2_transform_hdr *tr_hdr, struct smb_rqst *old_rq) +fill_transform_hdr(struct smb2_transform_hdr *tr_hdr, unsigned int orig_len, + struct smb_rqst *old_rq) { struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)old_rq->rq_iov[1].iov_base; - unsigned int orig_len = get_rfc1002_length(old_rq->rq_iov[0].iov_base); memset(tr_hdr, 0, sizeof(struct smb2_transform_hdr)); tr_hdr->ProtocolId = SMB2_TRANSFORM_PROTO_NUM; @@ -2065,8 +2154,6 @@ fill_transform_hdr(struct TCP_Server_Info *server, tr_hdr->Flags = cpu_to_le16(0x01); get_random_bytes(&tr_hdr->Nonce, SMB3_AES128CMM_NONCE); memcpy(&tr_hdr->SessionId, &shdr->SessionId, 8); - inc_rfc1001_len(tr_hdr, sizeof(struct smb2_transform_hdr) - server->vals->header_preamble_size); - inc_rfc1001_len(tr_hdr, orig_len); } /* We can not use the normal sg_set_buf() as we will sometimes pass a @@ -2078,11 +2165,16 @@ static inline void smb2_sg_set_buf(struct scatterlist *sg, const void *buf, sg_set_page(sg, virt_to_page(buf), buflen, offset_in_page(buf)); } +/* Assumes: + * rqst->rq_iov[0] is rfc1002 length + * rqst->rq_iov[1] is tranform header + * rqst->rq_iov[2+] data to be encrypted/decrypted + */ static struct scatterlist * init_sg(struct smb_rqst *rqst, u8 *sign) { - unsigned int sg_len = rqst->rq_nvec + rqst->rq_npages + 1; - unsigned int assoc_data_len = sizeof(struct smb2_transform_hdr) - 24; + unsigned int sg_len = rqst->rq_nvec + rqst->rq_npages; + unsigned int assoc_data_len = sizeof(struct smb2_transform_hdr) - 20; struct scatterlist *sg; unsigned int i; unsigned int j; @@ -2092,10 +2184,10 @@ init_sg(struct smb_rqst *rqst, u8 *sign) return NULL; sg_init_table(sg, sg_len); - smb2_sg_set_buf(&sg[0], rqst->rq_iov[0].iov_base + 24, assoc_data_len); - for (i = 1; i < rqst->rq_nvec; i++) - smb2_sg_set_buf(&sg[i], rqst->rq_iov[i].iov_base, - rqst->rq_iov[i].iov_len); + smb2_sg_set_buf(&sg[0], rqst->rq_iov[1].iov_base + 20, assoc_data_len); + for (i = 1; i < rqst->rq_nvec - 1; i++) + smb2_sg_set_buf(&sg[i], rqst->rq_iov[i+1].iov_base, + rqst->rq_iov[i+1].iov_len); for (j = 0; i < sg_len - 1; i++, j++) { unsigned int len = (j < rqst->rq_npages - 1) ? rqst->rq_pagesz : rqst->rq_tailsz; @@ -2127,9 +2219,10 @@ smb2_get_enc_key(struct TCP_Server_Info *server, __u64 ses_id, int enc, u8 *key) } /* * Encrypt or decrypt @rqst message. @rqst has the following format: - * iov[0] - transform header (associate data), - * iov[1-N] and pages - data to encrypt. - * On success return encrypted data in iov[1-N] and pages, leave iov[0] + * iov[0] - rfc1002 length + * iov[1] - transform header (associate data), + * iov[2-N] and pages - data to encrypt. + * On success return encrypted data in iov[2-N] and pages, leave iov[0-1] * untouched. */ static int @@ -2137,7 +2230,7 @@ crypt_message(struct TCP_Server_Info *server, struct smb_rqst *rqst, int enc) { struct smb2_transform_hdr *tr_hdr = (struct smb2_transform_hdr *)rqst->rq_iov[0].iov_base; - unsigned int assoc_data_len = sizeof(struct smb2_transform_hdr) - 20 - server->vals->header_preamble_size; + unsigned int assoc_data_len = sizeof(struct smb2_transform_hdr) - 20; int rc = 0; struct scatterlist *sg; u8 sign[SMB2_SIGNATURE_SIZE] = {}; @@ -2224,6 +2317,10 @@ free_req: return rc; } +/* + * This is called from smb_send_rqst. At this point we have the rfc1002 + * header as the first element in the vector. + */ static int smb3_init_transform_rq(struct TCP_Server_Info *server, struct smb_rqst *new_rq, struct smb_rqst *old_rq) @@ -2232,6 +2329,7 @@ smb3_init_transform_rq(struct TCP_Server_Info *server, struct smb_rqst *new_rq, struct page **pages; struct smb2_transform_hdr *tr_hdr; unsigned int npages = old_rq->rq_npages; + unsigned int orig_len = get_rfc1002_length(old_rq->rq_iov[0].iov_base); int i; int rc = -ENOMEM; @@ -2250,24 +2348,34 @@ smb3_init_transform_rq(struct TCP_Server_Info *server, struct smb_rqst *new_rq, goto err_free_pages; } - iov = kmalloc_array(old_rq->rq_nvec, sizeof(struct kvec), GFP_KERNEL); + /* Make space for one extra iov to hold the transform header */ + iov = kmalloc_array(old_rq->rq_nvec + 1, sizeof(struct kvec), + GFP_KERNEL); if (!iov) goto err_free_pages; /* copy all iovs from the old except the 1st one (rfc1002 length) */ - memcpy(&iov[1], &old_rq->rq_iov[1], + memcpy(&iov[2], &old_rq->rq_iov[1], sizeof(struct kvec) * (old_rq->rq_nvec - 1)); + /* copy the rfc1002 iov */ + iov[0].iov_base = old_rq->rq_iov[0].iov_base; + iov[0].iov_len = old_rq->rq_iov[0].iov_len; + new_rq->rq_iov = iov; - new_rq->rq_nvec = old_rq->rq_nvec; + new_rq->rq_nvec = old_rq->rq_nvec + 1; tr_hdr = kmalloc(sizeof(struct smb2_transform_hdr), GFP_KERNEL); if (!tr_hdr) goto err_free_iov; - /* fill the 1st iov with a transform header */ - fill_transform_hdr(server, tr_hdr, old_rq); - new_rq->rq_iov[0].iov_base = tr_hdr; - new_rq->rq_iov[0].iov_len = sizeof(struct smb2_transform_hdr); + /* fill the 2nd iov with a transform header */ + fill_transform_hdr(tr_hdr, orig_len, old_rq); + new_rq->rq_iov[1].iov_base = tr_hdr; + new_rq->rq_iov[1].iov_len = sizeof(struct smb2_transform_hdr); + + /* Update rfc1002 header */ + inc_rfc1001_len(new_rq->rq_iov[0].iov_base, + sizeof(struct smb2_transform_hdr)); /* copy pages form the old */ for (i = 0; i < npages; i++) { @@ -2307,7 +2415,7 @@ smb3_free_transform_rq(struct smb_rqst *rqst) put_page(rqst->rq_pages[i]); kfree(rqst->rq_pages); /* free transform header */ - kfree(rqst->rq_iov[0].iov_base); + kfree(rqst->rq_iov[1].iov_base); kfree(rqst->rq_iov); } @@ -2324,18 +2432,19 @@ decrypt_raw_data(struct TCP_Server_Info *server, char *buf, unsigned int buf_data_size, struct page **pages, unsigned int npages, unsigned int page_data_size) { - struct kvec iov[2]; + struct kvec iov[3]; struct smb_rqst rqst = {NULL}; - struct smb2_hdr *hdr; int rc; - iov[0].iov_base = buf; - iov[0].iov_len = sizeof(struct smb2_transform_hdr); - iov[1].iov_base = buf + sizeof(struct smb2_transform_hdr); - iov[1].iov_len = buf_data_size; + iov[0].iov_base = NULL; + iov[0].iov_len = 0; + iov[1].iov_base = buf; + iov[1].iov_len = sizeof(struct smb2_transform_hdr); + iov[2].iov_base = buf + sizeof(struct smb2_transform_hdr); + iov[2].iov_len = buf_data_size; rqst.rq_iov = iov; - rqst.rq_nvec = 2; + rqst.rq_nvec = 3; rqst.rq_pages = pages; rqst.rq_npages = npages; rqst.rq_pagesz = PAGE_SIZE; @@ -2347,10 +2456,9 @@ decrypt_raw_data(struct TCP_Server_Info *server, char *buf, if (rc) return rc; - memmove(buf + server->vals->header_preamble_size, iov[1].iov_base, buf_data_size); - hdr = (struct smb2_hdr *)buf; - hdr->smb2_buf_length = cpu_to_be32(buf_data_size + page_data_size); - server->total_read = buf_data_size + page_data_size + server->vals->header_preamble_size; + memmove(buf, iov[2].iov_base, buf_data_size); + + server->total_read = buf_data_size + page_data_size; return rc; } @@ -2375,7 +2483,7 @@ read_data_into_pages(struct TCP_Server_Info *server, struct page **pages, zero_user(page, len, PAGE_SIZE - len); len = 0; } - length = cifs_read_page_from_socket(server, page, n); + length = cifs_read_page_from_socket(server, page, 0, n); if (length < 0) return length; server->total_read += length; @@ -2423,7 +2531,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid, unsigned int cur_page_idx; unsigned int pad_len; struct cifs_readdata *rdata = mid->callback_data; - struct smb2_sync_hdr *shdr = get_sync_hdr(buf); + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; struct bio_vec *bvec = NULL; struct iov_iter iter; struct kvec iov; @@ -2454,7 +2562,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid, return 0; } - data_offset = server->ops->read_data_offset(buf) + server->vals->header_preamble_size; + data_offset = server->ops->read_data_offset(buf); #ifdef CONFIG_CIFS_SMB_DIRECT use_rdma_mr = rdata->mr; #endif @@ -2550,12 +2658,11 @@ receive_encrypted_read(struct TCP_Server_Info *server, struct mid_q_entry **mid) unsigned int npages; struct page **pages; unsigned int len; - unsigned int buflen = get_rfc1002_length(buf) + server->vals->header_preamble_size; + unsigned int buflen = server->pdu_size; int rc; int i = 0; - len = min_t(unsigned int, buflen, server->vals->read_rsp_size - - server->vals->header_preamble_size + + len = min_t(unsigned int, buflen, server->vals->read_rsp_size + sizeof(struct smb2_transform_hdr)) - HEADER_SIZE(server) + 1; rc = cifs_read_from_socket(server, buf + HEADER_SIZE(server) - 1, len); @@ -2563,8 +2670,7 @@ receive_encrypted_read(struct TCP_Server_Info *server, struct mid_q_entry **mid) return rc; server->total_read += rc; - len = le32_to_cpu(tr_hdr->OriginalMessageSize) + - server->vals->header_preamble_size - + len = le32_to_cpu(tr_hdr->OriginalMessageSize) - server->vals->read_rsp_size; npages = DIV_ROUND_UP(len, PAGE_SIZE); @@ -2591,8 +2697,7 @@ receive_encrypted_read(struct TCP_Server_Info *server, struct mid_q_entry **mid) if (rc) goto free_pages; - rc = decrypt_raw_data(server, buf, server->vals->read_rsp_size - - server->vals->header_preamble_size, + rc = decrypt_raw_data(server, buf, server->vals->read_rsp_size, pages, npages, len); if (rc) goto free_pages; @@ -2624,12 +2729,12 @@ receive_encrypted_standard(struct TCP_Server_Info *server, { int length; char *buf = server->smallbuf; - unsigned int pdu_length = get_rfc1002_length(buf); + unsigned int pdu_length = server->pdu_size; unsigned int buf_size; struct mid_q_entry *mid_entry; /* switch to large buffer if too big for a small one */ - if (pdu_length + server->vals->header_preamble_size > MAX_CIFS_SMALL_BUFFER_SIZE) { + if (pdu_length > MAX_CIFS_SMALL_BUFFER_SIZE) { server->large_buf = true; memcpy(server->bigbuf, buf, server->total_read); buf = server->bigbuf; @@ -2637,13 +2742,12 @@ receive_encrypted_standard(struct TCP_Server_Info *server, /* now read the rest */ length = cifs_read_from_socket(server, buf + HEADER_SIZE(server) - 1, - pdu_length - HEADER_SIZE(server) + 1 + - server->vals->header_preamble_size); + pdu_length - HEADER_SIZE(server) + 1); if (length < 0) return length; server->total_read += length; - buf_size = pdu_length + server->vals->header_preamble_size - sizeof(struct smb2_transform_hdr); + buf_size = pdu_length - sizeof(struct smb2_transform_hdr); length = decrypt_raw_data(server, buf, buf_size, NULL, 0, 0); if (length) return length; @@ -2668,11 +2772,11 @@ static int smb3_receive_transform(struct TCP_Server_Info *server, struct mid_q_entry **mid) { char *buf = server->smallbuf; - unsigned int pdu_length = get_rfc1002_length(buf); + unsigned int pdu_length = server->pdu_size; struct smb2_transform_hdr *tr_hdr = (struct smb2_transform_hdr *)buf; unsigned int orig_len = le32_to_cpu(tr_hdr->OriginalMessageSize); - if (pdu_length + server->vals->header_preamble_size < sizeof(struct smb2_transform_hdr) + + if (pdu_length < sizeof(struct smb2_transform_hdr) + sizeof(struct smb2_sync_hdr)) { cifs_dbg(VFS, "Transform message is too small (%u)\n", pdu_length); @@ -2681,14 +2785,14 @@ smb3_receive_transform(struct TCP_Server_Info *server, struct mid_q_entry **mid) return -ECONNABORTED; } - if (pdu_length + server->vals->header_preamble_size < orig_len + sizeof(struct smb2_transform_hdr)) { + if (pdu_length < orig_len + sizeof(struct smb2_transform_hdr)) { cifs_dbg(VFS, "Transform message is broken\n"); cifs_reconnect(server); wake_up(&server->response_q); return -ECONNABORTED; } - if (pdu_length + server->vals->header_preamble_size > CIFSMaxBufSize + MAX_HEADER_SIZE(server)) + if (pdu_length > CIFSMaxBufSize + MAX_HEADER_SIZE(server)) return receive_encrypted_read(server, mid); return receive_encrypted_standard(server, mid); @@ -2699,11 +2803,23 @@ smb3_handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid) { char *buf = server->large_buf ? server->bigbuf : server->smallbuf; - return handle_read_data(server, mid, buf, get_rfc1002_length(buf) + - server->vals->header_preamble_size, + return handle_read_data(server, mid, buf, server->pdu_size, NULL, 0, 0); } +static int +smb2_next_header(char *buf) +{ + struct smb2_sync_hdr *hdr = (struct smb2_sync_hdr *)buf; + struct smb2_transform_hdr *t_hdr = (struct smb2_transform_hdr *)buf; + + if (hdr->ProtocolId == SMB2_TRANSFORM_PROTO_NUM) + return sizeof(struct smb2_transform_hdr) + + le32_to_cpu(t_hdr->OriginalMessageSize); + + return le32_to_cpu(hdr->NextCommand); +} + struct smb_version_operations smb20_operations = { .compare_fids = smb2_compare_fids, .setup_request = smb2_setup_request, @@ -2795,6 +2911,7 @@ struct smb_version_operations smb20_operations = { .get_acl_by_fid = get_smb2_acl_by_fid, .set_acl = set_smb2_acl, #endif /* CIFS_ACL */ + .next_header = smb2_next_header, }; struct smb_version_operations smb21_operations = { @@ -2889,6 +3006,7 @@ struct smb_version_operations smb21_operations = { .get_acl_by_fid = get_smb2_acl_by_fid, .set_acl = set_smb2_acl, #endif /* CIFS_ACL */ + .next_header = smb2_next_header, }; struct smb_version_operations smb30_operations = { @@ -2993,6 +3111,7 @@ struct smb_version_operations smb30_operations = { .get_acl_by_fid = get_smb2_acl_by_fid, .set_acl = set_smb2_acl, #endif /* CIFS_ACL */ + .next_header = smb2_next_header, }; #ifdef CONFIG_CIFS_SMB311 @@ -3093,6 +3212,7 @@ struct smb_version_operations smb311_operations = { .query_all_EAs = smb2_query_eas, .set_EA = smb2_set_ea, #endif /* CIFS_XATTR */ + .next_header = smb2_next_header, }; #endif /* CIFS_SMB311 */ @@ -3104,8 +3224,8 @@ struct smb_version_values smb20_values = { .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE_LOCK, .shared_lock_type = SMB2_LOCKFLAG_SHARED_LOCK, .unlock_lock_type = SMB2_LOCKFLAG_UNLOCK, - .header_size = sizeof(struct smb2_hdr), - .header_preamble_size = 4, + .header_size = sizeof(struct smb2_sync_hdr), + .header_preamble_size = 0, .max_header_size = MAX_SMB2_HDR_SIZE, .read_rsp_size = sizeof(struct smb2_read_rsp) - 1, .lock_cmd = SMB2_LOCK, @@ -3125,8 +3245,8 @@ struct smb_version_values smb21_values = { .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE_LOCK, .shared_lock_type = SMB2_LOCKFLAG_SHARED_LOCK, .unlock_lock_type = SMB2_LOCKFLAG_UNLOCK, - .header_size = sizeof(struct smb2_hdr), - .header_preamble_size = 4, + .header_size = sizeof(struct smb2_sync_hdr), + .header_preamble_size = 0, .max_header_size = MAX_SMB2_HDR_SIZE, .read_rsp_size = sizeof(struct smb2_read_rsp) - 1, .lock_cmd = SMB2_LOCK, @@ -3146,8 +3266,8 @@ struct smb_version_values smb3any_values = { .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE_LOCK, .shared_lock_type = SMB2_LOCKFLAG_SHARED_LOCK, .unlock_lock_type = SMB2_LOCKFLAG_UNLOCK, - .header_size = sizeof(struct smb2_hdr), - .header_preamble_size = 4, + .header_size = sizeof(struct smb2_sync_hdr), + .header_preamble_size = 0, .max_header_size = MAX_SMB2_HDR_SIZE, .read_rsp_size = sizeof(struct smb2_read_rsp) - 1, .lock_cmd = SMB2_LOCK, @@ -3167,8 +3287,8 @@ struct smb_version_values smbdefault_values = { .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE_LOCK, .shared_lock_type = SMB2_LOCKFLAG_SHARED_LOCK, .unlock_lock_type = SMB2_LOCKFLAG_UNLOCK, - .header_size = sizeof(struct smb2_hdr), - .header_preamble_size = 4, + .header_size = sizeof(struct smb2_sync_hdr), + .header_preamble_size = 0, .max_header_size = MAX_SMB2_HDR_SIZE, .read_rsp_size = sizeof(struct smb2_read_rsp) - 1, .lock_cmd = SMB2_LOCK, @@ -3188,8 +3308,8 @@ struct smb_version_values smb30_values = { .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE_LOCK, .shared_lock_type = SMB2_LOCKFLAG_SHARED_LOCK, .unlock_lock_type = SMB2_LOCKFLAG_UNLOCK, - .header_size = sizeof(struct smb2_hdr), - .header_preamble_size = 4, + .header_size = sizeof(struct smb2_sync_hdr), + .header_preamble_size = 0, .max_header_size = MAX_SMB2_HDR_SIZE, .read_rsp_size = sizeof(struct smb2_read_rsp) - 1, .lock_cmd = SMB2_LOCK, @@ -3209,8 +3329,8 @@ struct smb_version_values smb302_values = { .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE_LOCK, .shared_lock_type = SMB2_LOCKFLAG_SHARED_LOCK, .unlock_lock_type = SMB2_LOCKFLAG_UNLOCK, - .header_size = sizeof(struct smb2_hdr), - .header_preamble_size = 4, + .header_size = sizeof(struct smb2_sync_hdr), + .header_preamble_size = 0, .max_header_size = MAX_SMB2_HDR_SIZE, .read_rsp_size = sizeof(struct smb2_read_rsp) - 1, .lock_cmd = SMB2_LOCK, @@ -3231,8 +3351,8 @@ struct smb_version_values smb311_values = { .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE_LOCK, .shared_lock_type = SMB2_LOCKFLAG_SHARED_LOCK, .unlock_lock_type = SMB2_LOCKFLAG_UNLOCK, - .header_size = sizeof(struct smb2_hdr), - .header_preamble_size = 4, + .header_size = sizeof(struct smb2_sync_hdr), + .header_preamble_size = 0, .max_header_size = MAX_SMB2_HDR_SIZE, .read_rsp_size = sizeof(struct smb2_read_rsp) - 1, .lock_cmd = SMB2_LOCK, diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index f7741cee2a4c..281fbc1dc720 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -49,6 +49,7 @@ #include "cifspdu.h" #include "cifs_spnego.h" #include "smbdirect.h" +#include "trace.h" /* * The following table defines the expected "StructureSize" of SMB2 requests @@ -79,7 +80,7 @@ static const int smb2_req_struct_sizes[NUMBER_OF_SMB2_COMMANDS] = { /* SMB2_OPLOCK_BREAK */ 24 /* BB this is 36 for LEASE_BREAK variant */ }; -static int encryption_required(const struct cifs_tcon *tcon) +static int smb3_encryption_required(const struct cifs_tcon *tcon) { if (!tcon) return 0; @@ -145,7 +146,7 @@ smb2_hdr_assemble(struct smb2_sync_hdr *shdr, __le16 smb2_cmd, shdr->Flags |= SMB2_FLAGS_DFS_OPERATIONS; */ if (tcon->ses && tcon->ses->server && tcon->ses->server->sign && - !encryption_required(tcon)) + !smb3_encryption_required(tcon)) shdr->Flags |= SMB2_FLAGS_SIGNED; out: return; @@ -268,8 +269,11 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon) mutex_unlock(&tcon->ses->session_mutex); cifs_dbg(FYI, "reconnect tcon rc = %d\n", rc); - if (rc) + if (rc) { + /* If sess reconnected but tcon didn't, something strange ... */ + printk_once(KERN_WARNING "reconnect tcon failed rc = %d\n", rc); goto out; + } if (smb2_command != SMB2_INTERNAL_CMD) queue_delayed_work(cifsiod_wq, &server->reconnect, 0); @@ -364,6 +368,7 @@ smb2_plain_req_init(__le16 smb2_command, struct cifs_tcon *tcon, #define SMB2_PREAUTH_INTEGRITY_CAPABILITIES cpu_to_le16(1) #define SMB2_ENCRYPTION_CAPABILITIES cpu_to_le16(2) +#define SMB2_POSIX_EXTENSIONS_AVAILABLE cpu_to_le16(0x100) static void build_preauth_ctxt(struct smb2_preauth_neg_context *pneg_ctxt) @@ -380,10 +385,17 @@ static void build_encrypt_ctxt(struct smb2_encryption_neg_context *pneg_ctxt) { pneg_ctxt->ContextType = SMB2_ENCRYPTION_CAPABILITIES; - pneg_ctxt->DataLength = cpu_to_le16(6); - pneg_ctxt->CipherCount = cpu_to_le16(2); - pneg_ctxt->Ciphers[0] = SMB2_ENCRYPTION_AES128_GCM; - pneg_ctxt->Ciphers[1] = SMB2_ENCRYPTION_AES128_CCM; + pneg_ctxt->DataLength = cpu_to_le16(4); /* Cipher Count + le16 cipher */ + pneg_ctxt->CipherCount = cpu_to_le16(1); +/* pneg_ctxt->Ciphers[0] = SMB2_ENCRYPTION_AES128_GCM;*/ /* not supported yet */ + pneg_ctxt->Ciphers[0] = SMB2_ENCRYPTION_AES128_CCM; +} + +static void +build_posix_ctxt(struct smb2_posix_neg_context *pneg_ctxt) +{ + pneg_ctxt->ContextType = SMB2_POSIX_EXTENSIONS_AVAILABLE; + pneg_ctxt->DataLength = cpu_to_le16(POSIX_CTXT_DATA_LEN); } static void @@ -391,18 +403,179 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, unsigned int *total_len) { char *pneg_ctxt = (char *)req + OFFSET_OF_NEG_CONTEXT; + unsigned int ctxt_len; + *total_len += 2; /* Add 2 due to round to 8 byte boundary for 1st ctxt */ build_preauth_ctxt((struct smb2_preauth_neg_context *)pneg_ctxt); - /* Add 2 to size to round to 8 byte boundary */ + ctxt_len = DIV_ROUND_UP(sizeof(struct smb2_preauth_neg_context), 8) * 8; + *total_len += ctxt_len; + pneg_ctxt += ctxt_len; - pneg_ctxt += 2 + sizeof(struct smb2_preauth_neg_context); build_encrypt_ctxt((struct smb2_encryption_neg_context *)pneg_ctxt); + ctxt_len = DIV_ROUND_UP(sizeof(struct smb2_encryption_neg_context), 8) * 8; + *total_len += ctxt_len; + pneg_ctxt += ctxt_len; + + build_posix_ctxt((struct smb2_posix_neg_context *)pneg_ctxt); + *total_len += sizeof(struct smb2_posix_neg_context); + req->NegotiateContextOffset = cpu_to_le32(OFFSET_OF_NEG_CONTEXT); - req->NegotiateContextCount = cpu_to_le16(2); + req->NegotiateContextCount = cpu_to_le16(3); +} + +static void decode_preauth_context(struct smb2_preauth_neg_context *ctxt) +{ + unsigned int len = le16_to_cpu(ctxt->DataLength); + + /* If invalid preauth context warn but use what we requested, SHA-512 */ + if (len < MIN_PREAUTH_CTXT_DATA_LEN) { + printk_once(KERN_WARNING "server sent bad preauth context\n"); + return; + } + if (le16_to_cpu(ctxt->HashAlgorithmCount) != 1) + printk_once(KERN_WARNING "illegal SMB3 hash algorithm count\n"); + if (ctxt->HashAlgorithms != SMB2_PREAUTH_INTEGRITY_SHA512) + printk_once(KERN_WARNING "unknown SMB3 hash algorithm\n"); +} + +static int decode_encrypt_ctx(struct TCP_Server_Info *server, + struct smb2_encryption_neg_context *ctxt) +{ + unsigned int len = le16_to_cpu(ctxt->DataLength); + + cifs_dbg(FYI, "decode SMB3.11 encryption neg context of len %d\n", len); + if (len < MIN_ENCRYPT_CTXT_DATA_LEN) { + printk_once(KERN_WARNING "server sent bad crypto ctxt len\n"); + return -EINVAL; + } + + if (le16_to_cpu(ctxt->CipherCount) != 1) { + printk_once(KERN_WARNING "illegal SMB3.11 cipher count\n"); + return -EINVAL; + } + cifs_dbg(FYI, "SMB311 cipher type:%d\n", le16_to_cpu(ctxt->Ciphers[0])); + if ((ctxt->Ciphers[0] != SMB2_ENCRYPTION_AES128_CCM) && + (ctxt->Ciphers[0] != SMB2_ENCRYPTION_AES128_GCM)) { + printk_once(KERN_WARNING "invalid SMB3.11 cipher returned\n"); + return -EINVAL; + } + server->cipher_type = ctxt->Ciphers[0]; + server->capabilities |= SMB2_GLOBAL_CAP_ENCRYPTION; + return 0; +} + +static int smb311_decode_neg_context(struct smb2_negotiate_rsp *rsp, + struct TCP_Server_Info *server, + unsigned int len_of_smb) +{ + struct smb2_neg_context *pctx; + unsigned int offset = le32_to_cpu(rsp->NegotiateContextOffset); + unsigned int ctxt_cnt = le16_to_cpu(rsp->NegotiateContextCount); + unsigned int len_of_ctxts, i; + int rc = 0; + + cifs_dbg(FYI, "decoding %d negotiate contexts\n", ctxt_cnt); + if (len_of_smb <= offset) { + cifs_dbg(VFS, "Invalid response: negotiate context offset\n"); + return -EINVAL; + } + + len_of_ctxts = len_of_smb - offset; + + for (i = 0; i < ctxt_cnt; i++) { + int clen; + /* check that offset is not beyond end of SMB */ + if (len_of_ctxts == 0) + break; + + if (len_of_ctxts < sizeof(struct smb2_neg_context)) + break; + + pctx = (struct smb2_neg_context *)(offset + (char *)rsp); + clen = le16_to_cpu(pctx->DataLength); + if (clen > len_of_ctxts) + break; + + if (pctx->ContextType == SMB2_PREAUTH_INTEGRITY_CAPABILITIES) + decode_preauth_context( + (struct smb2_preauth_neg_context *)pctx); + else if (pctx->ContextType == SMB2_ENCRYPTION_CAPABILITIES) + rc = decode_encrypt_ctx(server, + (struct smb2_encryption_neg_context *)pctx); + else if (pctx->ContextType == SMB2_POSIX_EXTENSIONS_AVAILABLE) + server->posix_ext_supported = true; + else + cifs_dbg(VFS, "unknown negcontext of type %d ignored\n", + le16_to_cpu(pctx->ContextType)); + + if (rc) + break; + /* offsets must be 8 byte aligned */ + clen = (clen + 7) & ~0x7; + offset += clen + sizeof(struct smb2_neg_context); + len_of_ctxts -= clen; + } + return rc; +} + +static struct create_posix * +create_posix_buf(umode_t mode) +{ + struct create_posix *buf; + + buf = kzalloc(sizeof(struct create_posix), + GFP_KERNEL); + if (!buf) + return NULL; - *total_len += 4 + sizeof(struct smb2_preauth_neg_context) - + sizeof(struct smb2_encryption_neg_context); + buf->ccontext.DataOffset = + cpu_to_le16(offsetof(struct create_posix, Mode)); + buf->ccontext.DataLength = cpu_to_le32(4); + buf->ccontext.NameOffset = + cpu_to_le16(offsetof(struct create_posix, Name)); + buf->ccontext.NameLength = cpu_to_le16(16); + + /* SMB2_CREATE_TAG_POSIX is "0x93AD25509CB411E7B42383DE968BCD7C" */ + buf->Name[0] = 0x93; + buf->Name[1] = 0xAD; + buf->Name[2] = 0x25; + buf->Name[3] = 0x50; + buf->Name[4] = 0x9C; + buf->Name[5] = 0xB4; + buf->Name[6] = 0x11; + buf->Name[7] = 0xE7; + buf->Name[8] = 0xB4; + buf->Name[9] = 0x23; + buf->Name[10] = 0x83; + buf->Name[11] = 0xDE; + buf->Name[12] = 0x96; + buf->Name[13] = 0x8B; + buf->Name[14] = 0xCD; + buf->Name[15] = 0x7C; + buf->Mode = cpu_to_le32(mode); + cifs_dbg(FYI, "mode on posix create 0%o", mode); + return buf; +} + +static int +add_posix_context(struct kvec *iov, unsigned int *num_iovec, umode_t mode) +{ + struct smb2_create_req *req = iov[0].iov_base; + unsigned int num = *num_iovec; + + iov[num].iov_base = create_posix_buf(mode); + if (iov[num].iov_base == NULL) + return -ENOMEM; + iov[num].iov_len = sizeof(struct create_posix); + if (!req->CreateContextsOffset) + req->CreateContextsOffset = cpu_to_le32( + sizeof(struct smb2_create_req) + + iov[num - 1].iov_len); + le32_add_cpu(&req->CreateContextsLength, sizeof(struct create_posix)); + *num_iovec = num + 1; + return 0; } + #else static void assemble_neg_contexts(struct smb2_negotiate_req *req, unsigned int *total_len) @@ -593,7 +766,7 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) server->capabilities |= SMB2_NT_FIND | SMB2_LARGE_FILES; security_blob = smb2_get_data_area_len(&blob_offset, &blob_length, - &rsp->hdr); + (struct smb2_sync_hdr *)rsp); /* * See MS-SMB2 section 2.2.4: if no blob, client picks default which * for us will be @@ -616,6 +789,16 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) else if (rc == 0) rc = -EIO; } + +#ifdef CONFIG_CIFS_SMB311 + if (rsp->DialectRevision == cpu_to_le16(SMB311_PROT_ID)) { + if (rsp->NegotiateContextCount) + rc = smb311_decode_neg_context(rsp, server, + rsp_iov.iov_len); + else + cifs_dbg(VFS, "Missing expected negotiate contexts\n"); + } +#endif /* CONFIG_CIFS_SMB311 */ neg_exit: free_rsp_buf(resp_buftype, rsp); return rc; @@ -623,19 +806,14 @@ neg_exit: int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) { - int rc = 0; - struct validate_negotiate_info_req vneg_inbuf; + int rc; + struct validate_negotiate_info_req *pneg_inbuf; struct validate_negotiate_info_rsp *pneg_rsp = NULL; u32 rsplen; u32 inbuflen; /* max of 4 dialects */ cifs_dbg(FYI, "validate negotiate\n"); -#ifdef CONFIG_CIFS_SMB_DIRECT - if (tcon->ses->server->rdma) - return 0; -#endif - /* In SMB3.11 preauth integrity supersedes validate negotiate */ if (tcon->ses->server->dialect == SMB311_PROT_ID) return 0; @@ -658,63 +836,69 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) if (tcon->ses->session_flags & SMB2_SESSION_FLAG_IS_NULL) cifs_dbg(VFS, "Unexpected null user (anonymous) auth flag sent by server\n"); - vneg_inbuf.Capabilities = + pneg_inbuf = kmalloc(sizeof(*pneg_inbuf), GFP_NOFS); + if (!pneg_inbuf) + return -ENOMEM; + + pneg_inbuf->Capabilities = cpu_to_le32(tcon->ses->server->vals->req_capabilities); - memcpy(vneg_inbuf.Guid, tcon->ses->server->client_guid, + memcpy(pneg_inbuf->Guid, tcon->ses->server->client_guid, SMB2_CLIENT_GUID_SIZE); if (tcon->ses->sign) - vneg_inbuf.SecurityMode = + pneg_inbuf->SecurityMode = cpu_to_le16(SMB2_NEGOTIATE_SIGNING_REQUIRED); else if (global_secflags & CIFSSEC_MAY_SIGN) - vneg_inbuf.SecurityMode = + pneg_inbuf->SecurityMode = cpu_to_le16(SMB2_NEGOTIATE_SIGNING_ENABLED); else - vneg_inbuf.SecurityMode = 0; + pneg_inbuf->SecurityMode = 0; if (strcmp(tcon->ses->server->vals->version_string, SMB3ANY_VERSION_STRING) == 0) { - vneg_inbuf.Dialects[0] = cpu_to_le16(SMB30_PROT_ID); - vneg_inbuf.Dialects[1] = cpu_to_le16(SMB302_PROT_ID); - vneg_inbuf.DialectCount = cpu_to_le16(2); + pneg_inbuf->Dialects[0] = cpu_to_le16(SMB30_PROT_ID); + pneg_inbuf->Dialects[1] = cpu_to_le16(SMB302_PROT_ID); + pneg_inbuf->DialectCount = cpu_to_le16(2); /* structure is big enough for 3 dialects, sending only 2 */ - inbuflen = sizeof(struct validate_negotiate_info_req) - 2; + inbuflen = sizeof(*pneg_inbuf) - + sizeof(pneg_inbuf->Dialects[0]); } else if (strcmp(tcon->ses->server->vals->version_string, SMBDEFAULT_VERSION_STRING) == 0) { - vneg_inbuf.Dialects[0] = cpu_to_le16(SMB21_PROT_ID); - vneg_inbuf.Dialects[1] = cpu_to_le16(SMB30_PROT_ID); - vneg_inbuf.Dialects[2] = cpu_to_le16(SMB302_PROT_ID); - vneg_inbuf.DialectCount = cpu_to_le16(3); + pneg_inbuf->Dialects[0] = cpu_to_le16(SMB21_PROT_ID); + pneg_inbuf->Dialects[1] = cpu_to_le16(SMB30_PROT_ID); + pneg_inbuf->Dialects[2] = cpu_to_le16(SMB302_PROT_ID); + pneg_inbuf->DialectCount = cpu_to_le16(3); /* structure is big enough for 3 dialects */ - inbuflen = sizeof(struct validate_negotiate_info_req); + inbuflen = sizeof(*pneg_inbuf); } else { /* otherwise specific dialect was requested */ - vneg_inbuf.Dialects[0] = + pneg_inbuf->Dialects[0] = cpu_to_le16(tcon->ses->server->vals->protocol_id); - vneg_inbuf.DialectCount = cpu_to_le16(1); + pneg_inbuf->DialectCount = cpu_to_le16(1); /* structure is big enough for 3 dialects, sending only 1 */ - inbuflen = sizeof(struct validate_negotiate_info_req) - 4; + inbuflen = sizeof(*pneg_inbuf) - + sizeof(pneg_inbuf->Dialects[0]) * 2; } rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID, FSCTL_VALIDATE_NEGOTIATE_INFO, true /* is_fsctl */, - (char *)&vneg_inbuf, sizeof(struct validate_negotiate_info_req), - (char **)&pneg_rsp, &rsplen); + (char *)pneg_inbuf, inbuflen, (char **)&pneg_rsp, &rsplen); if (rc != 0) { cifs_dbg(VFS, "validate protocol negotiate failed: %d\n", rc); - return -EIO; + rc = -EIO; + goto out_free_inbuf; } - if (rsplen != sizeof(struct validate_negotiate_info_rsp)) { + rc = -EIO; + if (rsplen != sizeof(*pneg_rsp)) { cifs_dbg(VFS, "invalid protocol negotiate response size: %d\n", rsplen); /* relax check since Mac returns max bufsize allowed on ioctl */ - if ((rsplen > CIFSMaxBufSize) - || (rsplen < sizeof(struct validate_negotiate_info_rsp))) - goto err_rsp_free; + if (rsplen > CIFSMaxBufSize || rsplen < sizeof(*pneg_rsp)) + goto out_free_rsp; } /* check validate negotiate info response matches what we got earlier */ @@ -731,15 +915,17 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) goto vneg_out; /* validate negotiate successful */ + rc = 0; cifs_dbg(FYI, "validate negotiate info successful\n"); - kfree(pneg_rsp); - return 0; + goto out_free_rsp; vneg_out: cifs_dbg(VFS, "protocol revalidation - security settings mismatch\n"); -err_rsp_free: +out_free_rsp: kfree(pneg_rsp); - return -EIO; +out_free_inbuf: + kfree(pneg_inbuf); + return rc; } enum securityEnum @@ -944,7 +1130,7 @@ SMB2_auth_kerberos(struct SMB2_sess_data *sess_data) goto out_put_spnego_key; rsp = (struct smb2_sess_setup_rsp *)sess_data->iov[0].iov_base; - ses->Suid = rsp->hdr.sync_hdr.SessionId; + ses->Suid = rsp->sync_hdr.SessionId; ses->session_flags = le16_to_cpu(rsp->SessionFlags); @@ -1020,13 +1206,13 @@ SMB2_sess_auth_rawntlmssp_negotiate(struct SMB2_sess_data *sess_data) /* If true, rc here is expected and not an error */ if (sess_data->buf0_type != CIFS_NO_BUFFER && - rsp->hdr.sync_hdr.Status == STATUS_MORE_PROCESSING_REQUIRED) + rsp->sync_hdr.Status == STATUS_MORE_PROCESSING_REQUIRED) rc = 0; if (rc) goto out; - if (offsetof(struct smb2_sess_setup_rsp, Buffer) - 4 != + if (offsetof(struct smb2_sess_setup_rsp, Buffer) != le16_to_cpu(rsp->SecurityBufferOffset)) { cifs_dbg(VFS, "Invalid security buffer offset %d\n", le16_to_cpu(rsp->SecurityBufferOffset)); @@ -1041,7 +1227,7 @@ SMB2_sess_auth_rawntlmssp_negotiate(struct SMB2_sess_data *sess_data) cifs_dbg(FYI, "rawntlmssp session setup challenge phase\n"); - ses->Suid = rsp->hdr.sync_hdr.SessionId; + ses->Suid = rsp->sync_hdr.SessionId; ses->session_flags = le16_to_cpu(rsp->SessionFlags); out: @@ -1099,7 +1285,7 @@ SMB2_sess_auth_rawntlmssp_authenticate(struct SMB2_sess_data *sess_data) rsp = (struct smb2_sess_setup_rsp *)sess_data->iov[0].iov_base; - ses->Suid = rsp->hdr.sync_hdr.SessionId; + ses->Suid = rsp->sync_hdr.SessionId; ses->session_flags = le16_to_cpu(rsp->SessionFlags); rc = SMB2_sess_establish_session(sess_data); @@ -1166,6 +1352,7 @@ SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses, sess_data->ses = ses; sess_data->buf0_type = CIFS_NO_BUFFER; sess_data->nls_cp = (struct nls_table *) nls_cp; + sess_data->previous_session = ses->Suid; #ifdef CONFIG_CIFS_SMB311 /* @@ -1293,7 +1480,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, return rc; } - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; iov[0].iov_base = (char *)req; @@ -1309,7 +1496,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, /* 3.11 tcon req must be signed if not encrypted. See MS-SMB2 3.2.4.1.1 */ if ((ses->server->dialect == SMB311_PROT_ID) && - !encryption_required(tcon)) + !smb3_encryption_required(tcon)) req->sync_hdr.Flags |= SMB2_FLAGS_SIGNED; rc = smb2_send_recv(xid, ses, iov, 2, &resp_buftype, flags, &rsp_iov); @@ -1347,7 +1534,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, tcon->maximal_access = le32_to_cpu(rsp->MaximalAccess); tcon->tidStatus = CifsGood; tcon->need_reconnect = false; - tcon->tid = rsp->hdr.sync_hdr.TreeId; + tcon->tid = rsp->sync_hdr.TreeId; strlcpy(tcon->treeName, tree, sizeof(tcon->treeName)); if ((rsp->Capabilities & SMB2_SHARE_CAP_DFS) && @@ -1367,7 +1554,7 @@ tcon_exit: return rc; tcon_error_exit: - if (rsp && rsp->hdr.sync_hdr.Status == STATUS_BAD_NETWORK_NAME) { + if (rsp && rsp->sync_hdr.Status == STATUS_BAD_NETWORK_NAME) { cifs_dbg(VFS, "BAD_NETWORK_NAME: %s\n", tree); } goto tcon_exit; @@ -1398,7 +1585,7 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon) if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; flags |= CIFS_NO_RESP; @@ -1465,7 +1652,7 @@ create_reconnect_durable_buf(struct cifs_fid *fid) static __u8 parse_lease_state(struct TCP_Server_Info *server, struct smb2_create_rsp *rsp, - unsigned int *epoch) + unsigned int *epoch, char *lease_key) { char *data_offset; struct create_context *cc; @@ -1473,14 +1660,15 @@ parse_lease_state(struct TCP_Server_Info *server, struct smb2_create_rsp *rsp, unsigned int remaining; char *name; - data_offset = (char *)rsp + server->vals->header_preamble_size + le32_to_cpu(rsp->CreateContextsOffset); + data_offset = (char *)rsp + le32_to_cpu(rsp->CreateContextsOffset); remaining = le32_to_cpu(rsp->CreateContextsLength); cc = (struct create_context *)data_offset; while (remaining >= sizeof(struct create_context)) { name = le16_to_cpu(cc->NameOffset) + (char *)cc; if (le16_to_cpu(cc->NameLength) == 4 && strncmp(name, "RqLs", 4) == 0) - return server->ops->parse_lease_buf(cc, epoch); + return server->ops->parse_lease_buf(cc, epoch, + lease_key); next = le32_to_cpu(cc->Next); if (!next) @@ -1701,14 +1889,14 @@ alloc_path_with_tree_prefix(__le16 **out_path, int *out_size, int *out_len, int SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, __u8 *oplock, struct smb2_file_all_info *buf, - struct smb2_err_rsp **err_buf) + struct kvec *err_iov) { struct smb2_create_req *req; struct smb2_create_rsp *rsp; struct TCP_Server_Info *server; struct cifs_tcon *tcon = oparms->tcon; struct cifs_ses *ses = tcon->ses; - struct kvec iov[4]; + struct kvec iov[5]; /* make sure at least one for each open context */ struct kvec rsp_iov = {NULL, 0}; int resp_buftype; int uni_path_len; @@ -1717,7 +1905,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, int rc = 0; unsigned int n_iov = 2; __u32 file_attributes = 0; - char *dhc_buf = NULL, *lc_buf = NULL; + char *dhc_buf = NULL, *lc_buf = NULL, *pc_buf = NULL; int flags = 0; unsigned int total_len; @@ -1733,7 +1921,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; if (oparms->create_options & CREATE_OPTION_READONLY) @@ -1834,6 +2022,27 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, dhc_buf = iov[n_iov-1].iov_base; } +#ifdef CONFIG_CIFS_SMB311 + if (tcon->posix_extensions) { + if (n_iov > 2) { + struct create_context *ccontext = + (struct create_context *)iov[n_iov-1].iov_base; + ccontext->Next = + cpu_to_le32(iov[n_iov-1].iov_len); + } + + rc = add_posix_context(iov, &n_iov, oparms->mode); + if (rc) { + cifs_small_buf_release(req); + kfree(copy_path); + kfree(lc_buf); + kfree(dhc_buf); + return rc; + } + pc_buf = iov[n_iov-1].iov_base; + } +#endif /* SMB311 */ + rc = smb2_send_recv(xid, ses, iov, n_iov, &resp_buftype, flags, &rsp_iov); cifs_small_buf_release(req); @@ -1841,11 +2050,18 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, if (rc != 0) { cifs_stats_fail_inc(tcon, SMB2_CREATE_HE); - if (err_buf && rsp) - *err_buf = kmemdup(rsp, get_rfc1002_length(rsp) + 4, - GFP_KERNEL); + if (err_iov && rsp) { + *err_iov = rsp_iov; + resp_buftype = CIFS_NO_BUFFER; + rsp = NULL; + } + trace_smb3_open_err(xid, tcon->tid, ses->Suid, + oparms->create_options, oparms->desired_access, rc); goto creat_exit; - } + } else + trace_smb3_open_done(xid, rsp->PersistentFileId, tcon->tid, + ses->Suid, oparms->create_options, + oparms->desired_access); oparms->fid->persistent_fid = rsp->PersistentFileId; oparms->fid->volatile_fid = rsp->VolatileFileId; @@ -1860,13 +2076,15 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, } if (rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) - *oplock = parse_lease_state(server, rsp, &oparms->fid->epoch); + *oplock = parse_lease_state(server, rsp, &oparms->fid->epoch, + oparms->fid->lease_key); else *oplock = rsp->OplockLevel; creat_exit: kfree(copy_path); kfree(lc_buf); kfree(dhc_buf); + kfree(pc_buf); free_rsp_buf(resp_buftype, rsp); return rc; } @@ -1882,7 +2100,6 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, { struct smb2_ioctl_req *req; struct smb2_ioctl_rsp *rsp; - struct smb2_sync_hdr *shdr; struct cifs_ses *ses; struct kvec iov[2]; struct kvec rsp_iov; @@ -1913,7 +2130,7 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->CtlCode = cpu_to_le32(opcode); @@ -1976,6 +2193,10 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, cifs_small_buf_release(req); rsp = (struct smb2_ioctl_rsp *)rsp_iov.iov_base; + if (rc != 0) + trace_smb3_fsctl_err(xid, persistent_fid, tcon->tid, + ses->Suid, 0, opcode, rc); + if ((rc != 0) && (rc != -EINVAL)) { cifs_stats_fail_inc(tcon, SMB2_IOCTL_HE); goto ioctl_exit; @@ -2003,7 +2224,7 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, goto ioctl_exit; } - if (get_rfc1002_length(rsp) < le32_to_cpu(rsp->OutputOffset) + *plen) { + if (rsp_iov.iov_len < le32_to_cpu(rsp->OutputOffset) + *plen) { cifs_dbg(VFS, "Malformed ioctl resp: len %d offset %d\n", *plen, le32_to_cpu(rsp->OutputOffset)); *plen = 0; @@ -2017,8 +2238,7 @@ SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, goto ioctl_exit; } - shdr = get_sync_hdr(rsp); - memcpy(*out_data, (char *)shdr + le32_to_cpu(rsp->OutputOffset), *plen); + memcpy(*out_data, (char *)rsp + le32_to_cpu(rsp->OutputOffset), *plen); ioctl_exit: free_rsp_buf(resp_buftype, rsp); return rc; @@ -2050,8 +2270,8 @@ SMB2_set_compression(const unsigned int xid, struct cifs_tcon *tcon, } int -SMB2_close(const unsigned int xid, struct cifs_tcon *tcon, - u64 persistent_fid, u64 volatile_fid) +SMB2_close_flags(const unsigned int xid, struct cifs_tcon *tcon, + u64 persistent_fid, u64 volatile_fid, int flags) { struct smb2_close_req *req; struct smb2_close_rsp *rsp; @@ -2060,7 +2280,6 @@ SMB2_close(const unsigned int xid, struct cifs_tcon *tcon, struct kvec rsp_iov; int resp_buftype; int rc = 0; - int flags = 0; unsigned int total_len; cifs_dbg(FYI, "Close\n"); @@ -2072,7 +2291,7 @@ SMB2_close(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->PersistentFileId = persistent_fid; @@ -2087,6 +2306,8 @@ SMB2_close(const unsigned int xid, struct cifs_tcon *tcon, if (rc != 0) { cifs_stats_fail_inc(tcon, SMB2_CLOSE_HE); + trace_smb3_close_err(xid, persistent_fid, tcon->tid, ses->Suid, + rc); goto close_exit; } @@ -2097,14 +2318,20 @@ close_exit: return rc; } -static int -validate_buf(unsigned int offset, unsigned int buffer_length, - struct smb2_hdr *hdr, unsigned int min_buf_size) +int +SMB2_close(const unsigned int xid, struct cifs_tcon *tcon, + u64 persistent_fid, u64 volatile_fid) +{ + return SMB2_close_flags(xid, tcon, persistent_fid, volatile_fid, 0); +} +static int +validate_iov(unsigned int offset, unsigned int buffer_length, + struct kvec *iov, unsigned int min_buf_size) { - unsigned int smb_len = be32_to_cpu(hdr->smb2_buf_length); - char *end_of_smb = smb_len + 4 /* RFC1001 length field */ + (char *)hdr; - char *begin_of_buf = 4 /* RFC1001 len field */ + offset + (char *)hdr; + unsigned int smb_len = iov->iov_len; + char *end_of_smb = smb_len + (char *)iov->iov_base; + char *begin_of_buf = offset + (char *)iov->iov_base; char *end_of_buf = begin_of_buf + buffer_length; @@ -2134,18 +2361,17 @@ validate_buf(unsigned int offset, unsigned int buffer_length, * Caller must free buffer. */ static int -validate_and_copy_buf(unsigned int offset, unsigned int buffer_length, - struct smb2_hdr *hdr, unsigned int minbufsize, +validate_and_copy_iov(unsigned int offset, unsigned int buffer_length, + struct kvec *iov, unsigned int minbufsize, char *data) - { - char *begin_of_buf = 4 /* RFC1001 len field */ + offset + (char *)hdr; + char *begin_of_buf = offset + (char *)iov->iov_base; int rc; if (!data) return -EINVAL; - rc = validate_buf(offset, buffer_length, hdr, minbufsize); + rc = validate_iov(offset, buffer_length, iov, minbufsize); if (rc) return rc; @@ -2180,7 +2406,7 @@ query_info(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->InfoType = info_type; @@ -2206,6 +2432,8 @@ query_info(const unsigned int xid, struct cifs_tcon *tcon, if (rc) { cifs_stats_fail_inc(tcon, SMB2_QUERY_INFO_HE); + trace_smb3_query_info_err(xid, persistent_fid, tcon->tid, + ses->Suid, info_class, (__u32)info_type, rc); goto qinf_exit; } @@ -2223,9 +2451,9 @@ query_info(const unsigned int xid, struct cifs_tcon *tcon, } } - rc = validate_and_copy_buf(le16_to_cpu(rsp->OutputBufferOffset), + rc = validate_and_copy_iov(le16_to_cpu(rsp->OutputBufferOffset), le32_to_cpu(rsp->OutputBufferLength), - &rsp->hdr, min_len, *data); + &rsp_iov, min_len, *data); qinf_exit: free_rsp_buf(resp_buftype, rsp); @@ -2294,7 +2522,7 @@ smb2_echo_callback(struct mid_q_entry *mid) unsigned int credits_received = 1; if (mid->mid_state == MID_RESPONSE_RECEIVED) - credits_received = le16_to_cpu(rsp->hdr.sync_hdr.CreditRequest); + credits_received = le16_to_cpu(rsp->sync_hdr.CreditRequest); DeleteMidQEntry(mid); add_credits(server, credits_received, CIFS_ECHO_OP); @@ -2423,7 +2651,7 @@ SMB2_flush(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->PersistentFileId = persistent_fid; @@ -2435,8 +2663,11 @@ SMB2_flush(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, rc = smb2_send_recv(xid, ses, iov, 1, &resp_buftype, flags, &rsp_iov); cifs_small_buf_release(req); - if (rc != 0) + if (rc != 0) { cifs_stats_fail_inc(tcon, SMB2_FLUSH_HE); + trace_smb3_flush_err(xid, persistent_fid, tcon->tid, ses->Suid, + rc); + } free_rsp_buf(resp_buftype, rsp_iov.iov_base); return rc; @@ -2481,7 +2712,7 @@ smb2_new_read_req(void **buf, unsigned int *total_len, * If we want to do a RDMA write, fill in and append * smbd_buffer_descriptor_v1 to the end of read request */ - if (server->rdma && rdata && + if (server->rdma && rdata && !server->sign && rdata->bytes >= server->smbd_conn->rdma_readwrite_threshold) { struct smbd_buffer_descriptor_v1 *v1; @@ -2545,11 +2776,12 @@ smb2_readv_callback(struct mid_q_entry *mid) struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); struct TCP_Server_Info *server = tcon->ses->server; struct smb2_sync_hdr *shdr = - (struct smb2_sync_hdr *)rdata->iov[1].iov_base; + (struct smb2_sync_hdr *)rdata->iov[0].iov_base; unsigned int credits_received = 1; struct smb_rqst rqst = { .rq_iov = rdata->iov, .rq_nvec = 2, .rq_pages = rdata->pages, + .rq_offset = rdata->page_offset, .rq_npages = rdata->nr_pages, .rq_pagesz = rdata->pagesz, .rq_tailsz = rdata->tailsz }; @@ -2647,7 +2879,7 @@ smb2_async_readv(struct cifs_readdata *rdata) return rc; } - if (encryption_required(io_parms.tcon)) + if (smb3_encryption_required(io_parms.tcon)) flags |= CIFS_TRANSFORM_REQ; req_len = cpu_to_be32(total_len); @@ -2678,7 +2910,13 @@ smb2_async_readv(struct cifs_readdata *rdata) if (rc) { kref_put(&rdata->refcount, cifs_readdata_release); cifs_stats_fail_inc(io_parms.tcon, SMB2_READ_HE); - } + trace_smb3_read_err(rc, 0 /* xid */, io_parms.persistent_fid, + io_parms.tcon->tid, io_parms.tcon->ses->Suid, + io_parms.offset, io_parms.length); + } else + trace_smb3_read_done(0 /* xid */, io_parms.persistent_fid, + io_parms.tcon->tid, io_parms.tcon->ses->Suid, + io_parms.offset, io_parms.length); cifs_small_buf_release(buf); return rc; @@ -2691,7 +2929,6 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, int resp_buftype, rc = -EACCES; struct smb2_read_plain_req *req = NULL; struct smb2_read_rsp *rsp = NULL; - struct smb2_sync_hdr *shdr; struct kvec iov[1]; struct kvec rsp_iov; unsigned int total_len; @@ -2703,7 +2940,7 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, if (rc) return rc; - if (encryption_required(io_parms->tcon)) + if (smb3_encryption_required(io_parms->tcon)) flags |= CIFS_TRANSFORM_REQ; iov[0].iov_base = (char *)req; @@ -2719,9 +2956,15 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, cifs_stats_fail_inc(io_parms->tcon, SMB2_READ_HE); cifs_dbg(VFS, "Send error in read = %d\n", rc); } + trace_smb3_read_err(rc, xid, req->PersistentFileId, + io_parms->tcon->tid, ses->Suid, + io_parms->offset, io_parms->length); free_rsp_buf(resp_buftype, rsp_iov.iov_base); return rc == -ENODATA ? 0 : rc; - } + } else + trace_smb3_read_done(xid, req->PersistentFileId, + io_parms->tcon->tid, ses->Suid, + io_parms->offset, io_parms->length); *nbytes = le32_to_cpu(rsp->DataLength); if ((*nbytes > CIFS_MAX_MSGSIZE) || @@ -2732,10 +2975,8 @@ SMB2_read(const unsigned int xid, struct cifs_io_parms *io_parms, *nbytes = 0; } - shdr = get_sync_hdr(rsp); - if (*buf) { - memcpy(*buf, (char *)shdr + rsp->DataOffset, *nbytes); + memcpy(*buf, (char *)rsp + rsp->DataOffset, *nbytes); free_rsp_buf(resp_buftype, rsp_iov.iov_base); } else if (resp_buftype != CIFS_NO_BUFFER) { *buf = rsp_iov.iov_base; @@ -2762,7 +3003,7 @@ smb2_writev_callback(struct mid_q_entry *mid) switch (mid->mid_state) { case MID_RESPONSE_RECEIVED: - credits_received = le16_to_cpu(rsp->hdr.sync_hdr.CreditRequest); + credits_received = le16_to_cpu(rsp->sync_hdr.CreditRequest); wdata->result = smb2_check_receive(mid, tcon->ses->server, 0); if (wdata->result != 0) break; @@ -2839,7 +3080,7 @@ smb2_async_writev(struct cifs_writedata *wdata, goto async_writev_out; } - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; shdr = (struct smb2_sync_hdr *)req; @@ -2859,7 +3100,7 @@ smb2_async_writev(struct cifs_writedata *wdata, * If we want to do a server RDMA read, fill in and append * smbd_buffer_descriptor_v1 to the end of write request */ - if (server->rdma && wdata->bytes >= + if (server->rdma && !server->sign && wdata->bytes >= server->smbd_conn->rdma_readwrite_threshold) { struct smbd_buffer_descriptor_v1 *v1; @@ -2900,6 +3141,7 @@ smb2_async_writev(struct cifs_writedata *wdata, rqst.rq_iov = iov; rqst.rq_nvec = 2; rqst.rq_pages = wdata->pages; + rqst.rq_offset = wdata->page_offset; rqst.rq_npages = wdata->nr_pages; rqst.rq_pagesz = wdata->pagesz; rqst.rq_tailsz = wdata->tailsz; @@ -2937,9 +3179,15 @@ smb2_async_writev(struct cifs_writedata *wdata, wdata, flags); if (rc) { + trace_smb3_write_err(0 /* no xid */, req->PersistentFileId, + tcon->tid, tcon->ses->Suid, wdata->offset, + wdata->bytes, rc); kref_put(&wdata->refcount, release); cifs_stats_fail_inc(tcon, SMB2_WRITE_HE); - } + } else + trace_smb3_write_done(0 /* no xid */, req->PersistentFileId, + tcon->tid, tcon->ses->Suid, wdata->offset, + wdata->bytes); async_writev_out: cifs_small_buf_release(req); @@ -2977,7 +3225,7 @@ SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms, if (io_parms->tcon->ses->server == NULL) return -ECONNABORTED; - if (encryption_required(io_parms->tcon)) + if (smb3_encryption_required(io_parms->tcon)) flags |= CIFS_TRANSFORM_REQ; req->sync_hdr.ProcessId = cpu_to_le32(io_parms->pid); @@ -3003,10 +3251,19 @@ SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms, rsp = (struct smb2_write_rsp *)rsp_iov.iov_base; if (rc) { + trace_smb3_write_err(xid, req->PersistentFileId, + io_parms->tcon->tid, + io_parms->tcon->ses->Suid, + io_parms->offset, io_parms->length, rc); cifs_stats_fail_inc(io_parms->tcon, SMB2_WRITE_HE); cifs_dbg(VFS, "Send error in write = %d\n", rc); - } else + } else { *nbytes = le32_to_cpu(rsp->DataLength); + trace_smb3_write_done(xid, req->PersistentFileId, + io_parms->tcon->tid, + io_parms->tcon->ses->Suid, + io_parms->offset, *nbytes); + } free_rsp_buf(resp_buftype, rsp); return rc; @@ -3087,7 +3344,7 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; switch (srch_inf->info_level) { @@ -3138,7 +3395,7 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon, if (rc) { if (rc == -ENODATA && - rsp->hdr.sync_hdr.Status == STATUS_NO_MORE_FILES) { + rsp->sync_hdr.Status == STATUS_NO_MORE_FILES) { srch_inf->endOfSearch = true; rc = 0; } @@ -3146,8 +3403,8 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon, goto qdir_exit; } - rc = validate_buf(le16_to_cpu(rsp->OutputBufferOffset), - le32_to_cpu(rsp->OutputBufferLength), &rsp->hdr, + rc = validate_iov(le16_to_cpu(rsp->OutputBufferOffset), + le32_to_cpu(rsp->OutputBufferLength), &rsp_iov, info_buf_size); if (rc) goto qdir_exit; @@ -3161,10 +3418,9 @@ SMB2_query_directory(const unsigned int xid, struct cifs_tcon *tcon, cifs_buf_release(srch_inf->ntwrk_buf_start); } srch_inf->ntwrk_buf_start = (char *)rsp; - srch_inf->srch_entries_start = srch_inf->last_entry = 4 /* rfclen */ + - (char *)&rsp->hdr + le16_to_cpu(rsp->OutputBufferOffset); - /* 4 for rfc1002 length field */ - end_of_smb = get_rfc1002_length(rsp) + 4 + (char *)&rsp->hdr; + srch_inf->srch_entries_start = srch_inf->last_entry = + (char *)rsp + le16_to_cpu(rsp->OutputBufferOffset); + end_of_smb = rsp_iov.iov_len + (char *)rsp; srch_inf->entries_in_buffer = num_entries(srch_inf->srch_entries_start, end_of_smb, &srch_inf->last_entry, info_buf_size); @@ -3219,7 +3475,7 @@ send_set_info(const unsigned int xid, struct cifs_tcon *tcon, return rc; } - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->sync_hdr.ProcessId = cpu_to_le32(pid); @@ -3252,8 +3508,11 @@ send_set_info(const unsigned int xid, struct cifs_tcon *tcon, cifs_small_buf_release(req); rsp = (struct smb2_set_info_rsp *)rsp_iov.iov_base; - if (rc != 0) + if (rc != 0) { cifs_stats_fail_inc(tcon, SMB2_SET_INFO_HE); + trace_smb3_set_info_err(xid, persistent_fid, tcon->tid, + ses->Suid, info_class, (__u32)info_type, rc); + } free_rsp_buf(resp_buftype, rsp); kfree(iov); @@ -3400,7 +3659,7 @@ SMB2_oplock_break(const unsigned int xid, struct cifs_tcon *tcon, __u8 oplock_level) { int rc; - struct smb2_oplock_break_req *req = NULL; + struct smb2_oplock_break *req = NULL; struct cifs_ses *ses = tcon->ses; int flags = CIFS_OBREAK_OP; unsigned int total_len; @@ -3414,7 +3673,7 @@ SMB2_oplock_break(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->VolatileFid = volatile_fid; @@ -3454,7 +3713,7 @@ static int build_qfs_info_req(struct kvec *iov, struct cifs_tcon *tcon, int level, int outbuf_len, u64 persistent_fid, u64 volatile_fid) { - struct TCP_Server_Info *server = tcon->ses->server; + struct TCP_Server_Info *server; int rc; struct smb2_query_info_req *req; unsigned int total_len; @@ -3464,6 +3723,8 @@ build_qfs_info_req(struct kvec *iov, struct cifs_tcon *tcon, int level, if ((tcon->ses == NULL) || (tcon->ses->server == NULL)) return -EIO; + server = tcon->ses->server; + rc = smb2_plain_req_init(SMB2_QUERY_INFO, tcon, (void **) &req, &total_len); if (rc) @@ -3477,7 +3738,7 @@ build_qfs_info_req(struct kvec *iov, struct cifs_tcon *tcon, int level, req->InputBufferOffset = cpu_to_le16(sizeof(struct smb2_query_info_req) - 1); req->OutputBufferLength = cpu_to_le32( - outbuf_len + sizeof(struct smb2_query_info_rsp) - 1 - server->vals->header_preamble_size); + outbuf_len + sizeof(struct smb2_query_info_rsp) - 1); iov->iov_base = (char *)req; iov->iov_len = total_len; @@ -3494,7 +3755,6 @@ SMB2_QFS_info(const unsigned int xid, struct cifs_tcon *tcon, int rc = 0; int resp_buftype; struct cifs_ses *ses = tcon->ses; - struct TCP_Server_Info *server = ses->server; struct smb2_fs_full_size_info *info = NULL; int flags = 0; @@ -3504,7 +3764,7 @@ SMB2_QFS_info(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; rc = smb2_send_recv(xid, ses, &iov, 1, &resp_buftype, flags, &rsp_iov); @@ -3515,10 +3775,10 @@ SMB2_QFS_info(const unsigned int xid, struct cifs_tcon *tcon, } rsp = (struct smb2_query_info_rsp *)rsp_iov.iov_base; - info = (struct smb2_fs_full_size_info *)(server->vals->header_preamble_size + - le16_to_cpu(rsp->OutputBufferOffset) + (char *)&rsp->hdr); - rc = validate_buf(le16_to_cpu(rsp->OutputBufferOffset), - le32_to_cpu(rsp->OutputBufferLength), &rsp->hdr, + info = (struct smb2_fs_full_size_info *)( + le16_to_cpu(rsp->OutputBufferOffset) + (char *)rsp); + rc = validate_iov(le16_to_cpu(rsp->OutputBufferOffset), + le32_to_cpu(rsp->OutputBufferLength), &rsp_iov, sizeof(struct smb2_fs_full_size_info)); if (!rc) copy_fs_info_to_kstatfs(info, fsdata); @@ -3538,7 +3798,6 @@ SMB2_QFS_attr(const unsigned int xid, struct cifs_tcon *tcon, int rc = 0; int resp_buftype, max_len, min_len; struct cifs_ses *ses = tcon->ses; - struct TCP_Server_Info *server = ses->server; unsigned int rsp_len, offset; int flags = 0; @@ -3561,7 +3820,7 @@ SMB2_QFS_attr(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; rc = smb2_send_recv(xid, ses, &iov, 1, &resp_buftype, flags, &rsp_iov); @@ -3574,20 +3833,20 @@ SMB2_QFS_attr(const unsigned int xid, struct cifs_tcon *tcon, rsp_len = le32_to_cpu(rsp->OutputBufferLength); offset = le16_to_cpu(rsp->OutputBufferOffset); - rc = validate_buf(offset, rsp_len, &rsp->hdr, min_len); + rc = validate_iov(offset, rsp_len, &rsp_iov, min_len); if (rc) goto qfsattr_exit; if (level == FS_ATTRIBUTE_INFORMATION) - memcpy(&tcon->fsAttrInfo, server->vals->header_preamble_size + offset - + (char *)&rsp->hdr, min_t(unsigned int, + memcpy(&tcon->fsAttrInfo, offset + + (char *)rsp, min_t(unsigned int, rsp_len, max_len)); else if (level == FS_DEVICE_INFORMATION) - memcpy(&tcon->fsDevInfo, server->vals->header_preamble_size + offset - + (char *)&rsp->hdr, sizeof(FILE_SYSTEM_DEVICE_INFO)); + memcpy(&tcon->fsDevInfo, offset + + (char *)rsp, sizeof(FILE_SYSTEM_DEVICE_INFO)); else if (level == FS_SECTOR_SIZE_INFORMATION) { struct smb3_fs_ss_info *ss_info = (struct smb3_fs_ss_info *) - (server->vals->header_preamble_size + offset + (char *)&rsp->hdr); + (offset + (char *)rsp); tcon->ss_flags = le32_to_cpu(ss_info->Flags); tcon->perf_sector_size = le32_to_cpu(ss_info->PhysicalBytesPerSectorForPerf); @@ -3618,7 +3877,7 @@ smb2_lockv(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->sync_hdr.ProcessId = cpu_to_le32(pid); @@ -3641,6 +3900,8 @@ smb2_lockv(const unsigned int xid, struct cifs_tcon *tcon, if (rc) { cifs_dbg(FYI, "Send error in smb2_lockv = %d\n", rc); cifs_stats_fail_inc(tcon, SMB2_LOCK_HE); + trace_smb3_lock_err(xid, persist_fid, tcon->tid, + tcon->ses->Suid, rc); } return rc; @@ -3682,7 +3943,7 @@ SMB2_lease_break(const unsigned int xid, struct cifs_tcon *tcon, if (rc) return rc; - if (encryption_required(tcon)) + if (smb3_encryption_required(tcon)) flags |= CIFS_TRANSFORM_REQ; req->sync_hdr.CreditRequest = cpu_to_le16(1); diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h index 253e2c7c952f..a345560001ce 100644 --- a/fs/cifs/smb2pdu.h +++ b/fs/cifs/smb2pdu.h @@ -122,25 +122,10 @@ struct smb2_sync_pdu { __le16 StructureSize2; /* size of wct area (varies, request specific) */ } __packed; -struct smb2_hdr { - __be32 smb2_buf_length; /* big endian on wire */ - /* length is only two or three bytes - with */ - /* one or two byte type preceding it that MBZ */ - struct smb2_sync_hdr sync_hdr; -} __packed; - -struct smb2_pdu { - struct smb2_hdr hdr; - __le16 StructureSize2; /* size of wct area (varies, request specific) */ -} __packed; - #define SMB3_AES128CMM_NONCE 11 #define SMB3_AES128GCM_NONCE 12 struct smb2_transform_hdr { - __be32 smb2_buf_length; /* big endian on wire */ - /* length is only two or three bytes - with - one or two byte type preceding it that MBZ */ __le32 ProtocolId; /* 0xFD 'S' 'M' 'B' */ __u8 Signature[16]; __u8 Nonce[16]; @@ -171,7 +156,7 @@ struct smb2_transform_hdr { #define SMB2_ERROR_STRUCTURE_SIZE2 cpu_to_le16(9) struct smb2_err_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; __le16 Reserved; /* MBZ */ __le32 ByteCount; /* even if zero, at least one byte follows */ @@ -263,11 +248,19 @@ struct smb2_negotiate_req { #define SMB2_NT_FIND 0x00100000 #define SMB2_LARGE_FILES 0x00200000 +struct smb2_neg_context { + __le16 ContextType; + __le16 DataLength; + __le32 Reserved; + /* Followed by array of data */ +} __packed; + #define SMB311_SALT_SIZE 32 /* Hash Algorithm Types */ #define SMB2_PREAUTH_INTEGRITY_SHA512 cpu_to_le16(0x0001) #define SMB2_PREAUTH_HASH_SIZE 64 +#define MIN_PREAUTH_CTXT_DATA_LEN (SMB311_SALT_SIZE + 6) struct smb2_preauth_neg_context { __le16 ContextType; /* 1 */ __le16 DataLength; @@ -282,16 +275,26 @@ struct smb2_preauth_neg_context { #define SMB2_ENCRYPTION_AES128_CCM cpu_to_le16(0x0001) #define SMB2_ENCRYPTION_AES128_GCM cpu_to_le16(0x0002) +/* Min encrypt context data is one cipher so 2 bytes + 2 byte count field */ +#define MIN_ENCRYPT_CTXT_DATA_LEN 4 struct smb2_encryption_neg_context { __le16 ContextType; /* 2 */ __le16 DataLength; __le32 Reserved; __le16 CipherCount; /* AES-128-GCM and AES-128-CCM */ - __le16 Ciphers[2]; /* Ciphers[0] since only one used now */ + __le16 Ciphers[1]; /* Ciphers[0] since only one used now */ +} __packed; + +#define POSIX_CTXT_DATA_LEN 8 +struct smb2_posix_neg_context { + __le16 ContextType; /* 0x100 */ + __le16 DataLength; + __le32 Reserved; + __le64 Reserved1; /* In case needed for future (eg version or caps) */ } __packed; struct smb2_negotiate_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 65 */ __le16 SecurityMode; __le16 DialectRevision; @@ -331,7 +334,7 @@ struct smb2_sess_setup_req { #define SMB2_SESSION_FLAG_IS_NULL 0x0002 #define SMB2_SESSION_FLAG_ENCRYPT_DATA 0x0004 struct smb2_sess_setup_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 9 */ __le16 SessionFlags; __le16 SecurityBufferOffset; @@ -346,7 +349,7 @@ struct smb2_logoff_req { } __packed; struct smb2_logoff_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 4 */ __le16 Reserved; } __packed; @@ -442,7 +445,7 @@ struct smb2_tree_connect_req_extension { } __packed; struct smb2_tree_connect_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 16 */ __u8 ShareType; /* see below */ __u8 Reserved; @@ -493,7 +496,7 @@ struct smb2_tree_disconnect_req { } __packed; struct smb2_tree_disconnect_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 4 */ __le16 Reserved; } __packed; @@ -605,7 +608,9 @@ struct smb2_tree_disconnect_rsp { #define SMB2_CREATE_DURABLE_HANDLE_REQUEST_V2 "DH2Q" #define SMB2_CREATE_DURABLE_HANDLE_RECONNECT_V2 "DH2C" #define SMB2_CREATE_APP_INSTANCE_ID 0x45BCA66AEFA7F74A9008FA462E144D74 -#define SVHDX_OPEN_DEVICE_CONTEXT 0x83CE6F1AD851E0986E34401CC9BCFCE9 +#define SVHDX_OPEN_DEVICE_CONTEX 0x9CCBCF9E04C1E643980E158DA1F6EC83 +#define SMB2_CREATE_TAG_POSIX 0x93AD25509CB411E7B42383DE968BCD7C + struct smb2_create_req { struct smb2_sync_hdr sync_hdr; @@ -628,7 +633,7 @@ struct smb2_create_req { } __packed; struct smb2_create_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 89 */ __u8 OplockLevel; __u8 Reserved; @@ -717,6 +722,13 @@ struct create_durable { } Data; } __packed; +struct create_posix { + struct create_context ccontext; + __u8 Name[16]; + __le32 Mode; + __u32 Reserved; +} __packed; + /* See MS-SMB2 2.2.13.2.11 */ /* Flags */ #define SMB2_DHANDLE_FLAG_PERSISTENT 0x00000002 @@ -884,7 +896,7 @@ struct smb2_ioctl_req { } __packed; struct smb2_ioctl_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 57 */ __u16 Reserved; __le32 CtlCode; @@ -911,7 +923,7 @@ struct smb2_close_req { } __packed; struct smb2_close_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* 60 */ __le16 Flags; __le32 Reserved; @@ -934,7 +946,7 @@ struct smb2_flush_req { } __packed; struct smb2_flush_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; __le16 Reserved; } __packed; @@ -966,7 +978,7 @@ struct smb2_read_plain_req { } __packed; struct smb2_read_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 17 */ __u8 DataOffset; __u8 Reserved; @@ -997,7 +1009,7 @@ struct smb2_write_req { } __packed; struct smb2_write_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 17 */ __u8 DataOffset; __u8 Reserved; @@ -1031,7 +1043,7 @@ struct smb2_lock_req { } __packed; struct smb2_lock_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 4 */ __le16 Reserved; } __packed; @@ -1043,7 +1055,7 @@ struct smb2_echo_req { } __packed; struct smb2_echo_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 4 */ __u16 Reserved; } __packed; @@ -1069,7 +1081,7 @@ struct smb2_query_directory_req { } __packed; struct smb2_query_directory_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 9 */ __le16 OutputBufferOffset; __le32 OutputBufferLength; @@ -1118,7 +1130,7 @@ struct smb2_query_info_req { } __packed; struct smb2_query_info_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 9 */ __le16 OutputBufferOffset; __le32 OutputBufferLength; @@ -1140,12 +1152,11 @@ struct smb2_set_info_req { } __packed; struct smb2_set_info_rsp { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 2 */ } __packed; -/* oplock break without an rfc1002 header */ -struct smb2_oplock_break_req { +struct smb2_oplock_break { struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 24 */ __u8 OplockLevel; @@ -1155,21 +1166,10 @@ struct smb2_oplock_break_req { __u64 VolatileFid; } __packed; -/* oplock break with an rfc1002 header */ -struct smb2_oplock_break_rsp { - struct smb2_hdr hdr; - __le16 StructureSize; /* Must be 24 */ - __u8 OplockLevel; - __u8 Reserved; - __le32 Reserved2; - __u64 PersistentFid; - __u64 VolatileFid; -} __packed; - #define SMB2_NOTIFY_BREAK_LEASE_FLAG_ACK_REQUIRED cpu_to_le32(0x01) struct smb2_lease_break { - struct smb2_hdr hdr; + struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 44 */ __le16 Reserved; __le32 Flags; diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h index cbcce3f7e86f..908555b1c6b5 100644 --- a/fs/cifs/smb2proto.h +++ b/fs/cifs/smb2proto.h @@ -36,8 +36,9 @@ struct smb_rqst; extern int map_smb2_to_linux_error(char *buf, bool log_err); extern int smb2_check_message(char *buf, unsigned int length, struct TCP_Server_Info *server); -extern unsigned int smb2_calc_size(void *buf); -extern char *smb2_get_data_area_len(int *off, int *len, struct smb2_hdr *hdr); +extern unsigned int smb2_calc_size(void *buf, struct TCP_Server_Info *server); +extern char *smb2_get_data_area_len(int *off, int *len, + struct smb2_sync_hdr *shdr); extern __le16 *cifs_convert_path_to_utf16(const char *from, struct cifs_sb_info *cifs_sb); @@ -65,6 +66,8 @@ extern struct cifs_ses *smb2_find_smb_ses(struct TCP_Server_Info *server, extern int smb3_handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid); +extern int open_shroot(unsigned int xid, struct cifs_tcon *tcon, + struct cifs_fid *pfid); extern void move_smb2_info_to_cifs(FILE_ALL_INFO *dst, struct smb2_file_all_info *src); extern int smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon, @@ -122,13 +125,15 @@ extern int SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon); extern int SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, __u8 *oplock, struct smb2_file_all_info *buf, - struct smb2_err_rsp **err_buf); + struct kvec *err_iov); extern int SMB2_ioctl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, u64 volatile_fid, u32 opcode, bool is_fsctl, char *in_data, u32 indatalen, char **out_data, u32 *plen /* returned data len */); extern int SMB2_close(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_file_id, u64 volatile_file_id); +extern int SMB2_close_flags(const unsigned int xid, struct cifs_tcon *tcon, + u64 persistent_fid, u64 volatile_fid, int flags); extern int SMB2_flush(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_file_id, u64 volatile_file_id); extern int SMB2_query_eas(const unsigned int xid, struct cifs_tcon *tcon, diff --git a/fs/cifs/smb2transport.c b/fs/cifs/smb2transport.c index bf49cb73b9e6..2c671123a6bf 100644 --- a/fs/cifs/smb2transport.c +++ b/fs/cifs/smb2transport.c @@ -480,7 +480,7 @@ smb2_verify_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server) unsigned int rc; char server_response_sig[16]; struct smb2_sync_hdr *shdr = - (struct smb2_sync_hdr *)rqst->rq_iov[1].iov_base; + (struct smb2_sync_hdr *)rqst->rq_iov[0].iov_base; if ((shdr->Command == SMB2_NEGOTIATE) || (shdr->Command == SMB2_SESSION_SETUP) || @@ -604,15 +604,13 @@ int smb2_check_receive(struct mid_q_entry *mid, struct TCP_Server_Info *server, bool log_error) { - unsigned int len = get_rfc1002_length(mid->resp_buf); - struct kvec iov[2]; + unsigned int len = mid->resp_buf_size; + struct kvec iov[1]; struct smb_rqst rqst = { .rq_iov = iov, - .rq_nvec = 2 }; + .rq_nvec = 1 }; iov[0].iov_base = (char *)mid->resp_buf; - iov[0].iov_len = 4; - iov[1].iov_base = (char *)mid->resp_buf + 4; - iov[1].iov_len = len; + iov[0].iov_len = len; dump_smb(mid->resp_buf, min_t(u32, 80, len)); /* convert the length into a more usable form */ diff --git a/fs/cifs/smbdirect.c b/fs/cifs/smbdirect.c index 5008af546dd1..c62f7c95683c 100644 --- a/fs/cifs/smbdirect.c +++ b/fs/cifs/smbdirect.c @@ -1028,7 +1028,7 @@ static int smbd_post_send(struct smbd_connection *info, for (i = 0; i < request->num_sge; i++) { log_rdma_send(INFO, "rdma_request sge[%d] addr=%llu length=%u\n", - i, request->sge[0].addr, request->sge[0].length); + i, request->sge[i].addr, request->sge[i].length); ib_dma_sync_single_for_device( info->id->device, request->sge[i].addr, @@ -2086,7 +2086,7 @@ int smbd_send(struct smbd_connection *info, struct smb_rqst *rqst) int start, i, j; int max_iov_size = info->max_send_size - sizeof(struct smbd_data_transfer); - struct kvec iov[SMBDIRECT_MAX_SGE]; + struct kvec *iov; int rc; info->smbd_send_pending++; @@ -2096,32 +2096,20 @@ int smbd_send(struct smbd_connection *info, struct smb_rqst *rqst) } /* - * This usually means a configuration error - * We use RDMA read/write for packet size > rdma_readwrite_threshold - * as long as it's properly configured we should never get into this - * situation - */ - if (rqst->rq_nvec + rqst->rq_npages > SMBDIRECT_MAX_SGE) { - log_write(ERR, "maximum send segment %x exceeding %x\n", - rqst->rq_nvec + rqst->rq_npages, SMBDIRECT_MAX_SGE); - rc = -EINVAL; - goto done; - } - - /* - * Remove the RFC1002 length defined in MS-SMB2 section 2.1 - * It is used only for TCP transport + * Skip the RFC1002 length defined in MS-SMB2 section 2.1 + * It is used only for TCP transport in the iov[0] * In future we may want to add a transport layer under protocol * layer so this will only be issued to TCP transport */ - iov[0].iov_base = (char *)rqst->rq_iov[0].iov_base + 4; - iov[0].iov_len = rqst->rq_iov[0].iov_len - 4; - buflen += iov[0].iov_len; + + if (rqst->rq_iov[0].iov_len != 4) { + log_write(ERR, "expected the pdu length in 1st iov, but got %zu\n", rqst->rq_iov[0].iov_len); + return -EINVAL; + } + iov = &rqst->rq_iov[1]; /* total up iov array first */ - for (i = 1; i < rqst->rq_nvec; i++) { - iov[i].iov_base = rqst->rq_iov[i].iov_base; - iov[i].iov_len = rqst->rq_iov[i].iov_len; + for (i = 0; i < rqst->rq_nvec-1; i++) { buflen += iov[i].iov_len; } @@ -2139,6 +2127,10 @@ int smbd_send(struct smbd_connection *info, struct smb_rqst *rqst) goto done; } + cifs_dbg(FYI, "Sending smb (RDMA): smb_len=%u\n", buflen); + for (i = 0; i < rqst->rq_nvec-1; i++) + dump_smb(iov[i].iov_base, iov[i].iov_len); + remaining_data_length = buflen; log_write(INFO, "rqst->rq_nvec=%d rqst->rq_npages=%d rq_pagesz=%d " @@ -2194,12 +2186,14 @@ int smbd_send(struct smbd_connection *info, struct smb_rqst *rqst) goto done; } i++; + if (i == rqst->rq_nvec-1) + break; } start = i; buflen = 0; } else { i++; - if (i == rqst->rq_nvec) { + if (i == rqst->rq_nvec-1) { /* send out all remaining vecs */ remaining_data_length -= buflen; log_write(INFO, diff --git a/fs/cifs/trace.c b/fs/cifs/trace.c new file mode 100644 index 000000000000..bd4a546feec1 --- /dev/null +++ b/fs/cifs/trace.c @@ -0,0 +1,18 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2018, Microsoft Corporation. + * + * Author(s): Steve French <stfrench@microsoft.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + */ +#define CREATE_TRACE_POINTS +#include "trace.h" diff --git a/fs/cifs/trace.h b/fs/cifs/trace.h new file mode 100644 index 000000000000..61e74d455d90 --- /dev/null +++ b/fs/cifs/trace.h @@ -0,0 +1,429 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2018, Microsoft Corporation. + * + * Author(s): Steve French <stfrench@microsoft.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM cifs + +#if !defined(_CIFS_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) +#define _CIFS_TRACE_H + +#include <linux/tracepoint.h> + +/* For logging errors in read or write */ +DECLARE_EVENT_CLASS(smb3_rw_err_class, + TP_PROTO(unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + __u64 offset, + __u32 len, + int rc), + TP_ARGS(xid, fid, tid, sesid, offset, len, rc), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u64, offset) + __field(__u32, len) + __field(int, rc) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->offset = offset; + __entry->len = len; + __entry->rc = rc; + ), + TP_printk("\txid=%u sid=0x%llx tid=0x%x fid=0x%llx offset=0x%llx len=0x%x rc=%d", + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->offset, __entry->len, __entry->rc) +) + +#define DEFINE_SMB3_RW_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_rw_err_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + __u64 offset, \ + __u32 len, \ + int rc), \ + TP_ARGS(xid, fid, tid, sesid, offset, len, rc)) + +DEFINE_SMB3_RW_ERR_EVENT(write_err); +DEFINE_SMB3_RW_ERR_EVENT(read_err); + + +/* For logging successful read or write */ +DECLARE_EVENT_CLASS(smb3_rw_done_class, + TP_PROTO(unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + __u64 offset, + __u32 len), + TP_ARGS(xid, fid, tid, sesid, offset, len), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u64, offset) + __field(__u32, len) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->offset = offset; + __entry->len = len; + ), + TP_printk("xid=%u sid=0x%llx tid=0x%x fid=0x%llx offset=0x%llx len=0x%x", + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->offset, __entry->len) +) + +#define DEFINE_SMB3_RW_DONE_EVENT(name) \ +DEFINE_EVENT(smb3_rw_done_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + __u64 offset, \ + __u32 len), \ + TP_ARGS(xid, fid, tid, sesid, offset, len)) + +DEFINE_SMB3_RW_DONE_EVENT(write_done); +DEFINE_SMB3_RW_DONE_EVENT(read_done); + +/* + * For handle based calls other than read and write, and get/set info + */ +DECLARE_EVENT_CLASS(smb3_fd_err_class, + TP_PROTO(unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + int rc), + TP_ARGS(xid, fid, tid, sesid, rc), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(int, rc) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->rc = rc; + ), + TP_printk("\txid=%u sid=0x%llx tid=0x%x fid=0x%llx rc=%d", + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->rc) +) + +#define DEFINE_SMB3_FD_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_fd_err_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + int rc), \ + TP_ARGS(xid, fid, tid, sesid, rc)) + +DEFINE_SMB3_FD_ERR_EVENT(flush_err); +DEFINE_SMB3_FD_ERR_EVENT(lock_err); +DEFINE_SMB3_FD_ERR_EVENT(close_err); + +/* + * For handle based query/set info calls + */ +DECLARE_EVENT_CLASS(smb3_inf_err_class, + TP_PROTO(unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + __u8 infclass, + __u32 type, + int rc), + TP_ARGS(xid, fid, tid, sesid, infclass, type, rc), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u8, infclass) + __field(__u32, type) + __field(int, rc) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->infclass = infclass; + __entry->type = type; + __entry->rc = rc; + ), + TP_printk("xid=%u sid=0x%llx tid=0x%x fid=0x%llx class=%u type=0x%x rc=%d", + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->infclass, __entry->type, __entry->rc) +) + +#define DEFINE_SMB3_INF_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_inf_err_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + __u8 infclass, \ + __u32 type, \ + int rc), \ + TP_ARGS(xid, fid, tid, sesid, infclass, type, rc)) + +DEFINE_SMB3_INF_ERR_EVENT(query_info_err); +DEFINE_SMB3_INF_ERR_EVENT(set_info_err); +DEFINE_SMB3_INF_ERR_EVENT(fsctl_err); + +/* + * For logging SMB3 Status code and Command for responses which return errors + */ +DECLARE_EVENT_CLASS(smb3_cmd_err_class, + TP_PROTO(__u32 tid, + __u64 sesid, + __u16 cmd, + __u64 mid, + __u32 status, + int rc), + TP_ARGS(tid, sesid, cmd, mid, status, rc), + TP_STRUCT__entry( + __field(__u32, tid) + __field(__u64, sesid) + __field(__u16, cmd) + __field(__u64, mid) + __field(__u32, status) + __field(int, rc) + ), + TP_fast_assign( + __entry->tid = tid; + __entry->sesid = sesid; + __entry->cmd = cmd; + __entry->mid = mid; + __entry->status = status; + __entry->rc = rc; + ), + TP_printk("\tsid=0x%llx tid=0x%x cmd=%u mid=%llu status=0x%x rc=%d", + __entry->sesid, __entry->tid, __entry->cmd, __entry->mid, + __entry->status, __entry->rc) +) + +#define DEFINE_SMB3_CMD_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_cmd_err_class, smb3_##name, \ + TP_PROTO(__u32 tid, \ + __u64 sesid, \ + __u16 cmd, \ + __u64 mid, \ + __u32 status, \ + int rc), \ + TP_ARGS(tid, sesid, cmd, mid, status, rc)) + +DEFINE_SMB3_CMD_ERR_EVENT(cmd_err); + +DECLARE_EVENT_CLASS(smb3_cmd_done_class, + TP_PROTO(__u32 tid, + __u64 sesid, + __u16 cmd, + __u64 mid), + TP_ARGS(tid, sesid, cmd, mid), + TP_STRUCT__entry( + __field(__u32, tid) + __field(__u64, sesid) + __field(__u16, cmd) + __field(__u64, mid) + ), + TP_fast_assign( + __entry->tid = tid; + __entry->sesid = sesid; + __entry->cmd = cmd; + __entry->mid = mid; + ), + TP_printk("\tsid=0x%llx tid=0x%x cmd=%u mid=%llu", + __entry->sesid, __entry->tid, + __entry->cmd, __entry->mid) +) + +#define DEFINE_SMB3_CMD_DONE_EVENT(name) \ +DEFINE_EVENT(smb3_cmd_done_class, smb3_##name, \ + TP_PROTO(__u32 tid, \ + __u64 sesid, \ + __u16 cmd, \ + __u64 mid), \ + TP_ARGS(tid, sesid, cmd, mid)) + +DEFINE_SMB3_CMD_DONE_EVENT(cmd_done); + +DECLARE_EVENT_CLASS(smb3_exit_err_class, + TP_PROTO(unsigned int xid, + const char *func_name, + int rc), + TP_ARGS(xid, func_name, rc), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(const char *, func_name) + __field(int, rc) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->func_name = func_name; + __entry->rc = rc; + ), + TP_printk("\t%s: xid=%u rc=%d", + __entry->func_name, __entry->xid, __entry->rc) +) + +#define DEFINE_SMB3_EXIT_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_exit_err_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + const char *func_name, \ + int rc), \ + TP_ARGS(xid, func_name, rc)) + +DEFINE_SMB3_EXIT_ERR_EVENT(exit_err); + +DECLARE_EVENT_CLASS(smb3_enter_exit_class, + TP_PROTO(unsigned int xid, + const char *func_name), + TP_ARGS(xid, func_name), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(const char *, func_name) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->func_name = func_name; + ), + TP_printk("\t%s: xid=%u", + __entry->func_name, __entry->xid) +) + +#define DEFINE_SMB3_ENTER_EXIT_EVENT(name) \ +DEFINE_EVENT(smb3_enter_exit_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + const char *func_name), \ + TP_ARGS(xid, func_name)) + +DEFINE_SMB3_ENTER_EXIT_EVENT(enter); +DEFINE_SMB3_ENTER_EXIT_EVENT(exit_done); + +/* + * For smb2/smb3 open call + */ +DECLARE_EVENT_CLASS(smb3_open_err_class, + TP_PROTO(unsigned int xid, + __u32 tid, + __u64 sesid, + int create_options, + int desired_access, + int rc), + TP_ARGS(xid, tid, sesid, create_options, desired_access, rc), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u32, tid) + __field(__u64, sesid) + __field(int, create_options) + __field(int, desired_access) + __field(int, rc) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->create_options = create_options; + __entry->desired_access = desired_access; + __entry->rc = rc; + ), + TP_printk("xid=%u sid=0x%llx tid=0x%x cr_opts=0x%x des_access=0x%x rc=%d", + __entry->xid, __entry->sesid, __entry->tid, + __entry->create_options, __entry->desired_access, __entry->rc) +) + +#define DEFINE_SMB3_OPEN_ERR_EVENT(name) \ +DEFINE_EVENT(smb3_open_err_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u32 tid, \ + __u64 sesid, \ + int create_options, \ + int desired_access, \ + int rc), \ + TP_ARGS(xid, tid, sesid, create_options, desired_access, rc)) + +DEFINE_SMB3_OPEN_ERR_EVENT(open_err); + + +DECLARE_EVENT_CLASS(smb3_open_done_class, + TP_PROTO(unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + int create_options, + int desired_access), + TP_ARGS(xid, fid, tid, sesid, create_options, desired_access), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(int, create_options) + __field(int, desired_access) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->create_options = create_options; + __entry->desired_access = desired_access; + ), + TP_printk("xid=%u sid=0x%llx tid=0x%x fid=0x%llx cr_opts=0x%x des_access=0x%x", + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->create_options, __entry->desired_access) +) + +#define DEFINE_SMB3_OPEN_DONE_EVENT(name) \ +DEFINE_EVENT(smb3_open_done_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + int create_options, \ + int desired_access), \ + TP_ARGS(xid, fid, tid, sesid, create_options, desired_access)) + +DEFINE_SMB3_OPEN_DONE_EVENT(open_done); + +#endif /* _CIFS_TRACE_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace +#include <trace/define_trace.h> diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 279718dcb2ed..e7254e386b79 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -753,7 +753,7 @@ cifs_send_recv(const unsigned int xid, struct cifs_ses *ses, goto out; #ifdef CONFIG_CIFS_SMB311 - if (ses->status == CifsNew) + if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP)) smb311_update_preauth_hash(ses, rqst->rq_iov+1, rqst->rq_nvec-1); #endif @@ -790,7 +790,7 @@ cifs_send_recv(const unsigned int xid, struct cifs_ses *ses, buf = (char *)midQ->resp_buf; resp_iov->iov_base = buf; - resp_iov->iov_len = get_rfc1002_length(buf) + + resp_iov->iov_len = midQ->resp_buf_size + ses->server->vals->header_preamble_size; if (midQ->large_buf) *resp_buf_type = CIFS_LARGE_BUFFER; @@ -798,10 +798,10 @@ cifs_send_recv(const unsigned int xid, struct cifs_ses *ses, *resp_buf_type = CIFS_SMALL_BUFFER; #ifdef CONFIG_CIFS_SMB311 - if (ses->status == CifsNew) { + if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP)) { struct kvec iov = { - .iov_base = buf + 4, - .iov_len = get_rfc1002_length(buf) + .iov_base = buf, + .iov_len = midQ->resp_buf_size }; smb311_update_preauth_hash(ses, &iov, 1); } @@ -834,8 +834,11 @@ SendReceive2(const unsigned int xid, struct cifs_ses *ses, if (n_vec + 1 > CIFS_MAX_IOV_SIZE) { new_iov = kmalloc(sizeof(struct kvec) * (n_vec + 1), GFP_KERNEL); - if (!new_iov) + if (!new_iov) { + /* otherwise cifs_send_recv below sets resp_buf_type */ + *resp_buf_type = CIFS_NO_BUFFER; return -ENOMEM; + } } else new_iov = s_iov; diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 017b0ab19bc4..c4fb9ad7c808 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -492,7 +492,7 @@ static void cramfs_kill_sb(struct super_block *sb) { struct cramfs_sb_info *sbi = CRAMFS_SB(sb); - if (IS_ENABLED(CCONFIG_CRAMFS_MTD) && sb->s_mtd) { + if (IS_ENABLED(CONFIG_CRAMFS_MTD) && sb->s_mtd) { if (sbi && sbi->mtd_point_size) mtd_unpoint(sb->s_mtd, 0, sbi->mtd_point_size); kill_mtd_super(sb); @@ -808,10 +808,7 @@ static struct dentry *cramfs_lookup(struct inode *dir, struct dentry *dentry, un } out: mutex_unlock(&read_mutex); - if (IS_ERR(inode)) - return ERR_CAST(inode); - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } static int cramfs_readpage(struct file *file, struct page *page) diff --git a/fs/crypto/crypto.c b/fs/crypto/crypto.c index ce654526c0fb..243a269e6c5f 100644 --- a/fs/crypto/crypto.c +++ b/fs/crypto/crypto.c @@ -156,12 +156,8 @@ int fscrypt_do_page_crypto(const struct inode *inode, fscrypt_direction_t rw, } req = skcipher_request_alloc(tfm, gfp_flags); - if (!req) { - printk_ratelimited(KERN_ERR - "%s: crypto_request_alloc() failed\n", - __func__); + if (!req) return -ENOMEM; - } skcipher_request_set_callback( req, CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP, @@ -178,9 +174,10 @@ int fscrypt_do_page_crypto(const struct inode *inode, fscrypt_direction_t rw, res = crypto_wait_req(crypto_skcipher_encrypt(req), &wait); skcipher_request_free(req); if (res) { - printk_ratelimited(KERN_ERR - "%s: crypto_skcipher_encrypt() returned %d\n", - __func__, res); + fscrypt_err(inode->i_sb, + "%scryption failed for inode %lu, block %llu: %d", + (rw == FS_DECRYPT ? "de" : "en"), + inode->i_ino, lblk_num, res); return res; } return 0; @@ -326,7 +323,6 @@ static int fscrypt_d_revalidate(struct dentry *dentry, unsigned int flags) return 0; } - /* this should eventually be an flag in d_flags */ spin_lock(&dentry->d_lock); cached_with_key = dentry->d_flags & DCACHE_ENCRYPTED_WITH_KEY; spin_unlock(&dentry->d_lock); @@ -353,7 +349,6 @@ static int fscrypt_d_revalidate(struct dentry *dentry, unsigned int flags) const struct dentry_operations fscrypt_d_ops = { .d_revalidate = fscrypt_d_revalidate, }; -EXPORT_SYMBOL(fscrypt_d_ops); void fscrypt_restore_control_page(struct page *page) { @@ -422,13 +417,43 @@ fail: return res; } +void fscrypt_msg(struct super_block *sb, const char *level, + const char *fmt, ...) +{ + static DEFINE_RATELIMIT_STATE(rs, DEFAULT_RATELIMIT_INTERVAL, + DEFAULT_RATELIMIT_BURST); + struct va_format vaf; + va_list args; + + if (!__ratelimit(&rs)) + return; + + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + if (sb) + printk("%sfscrypt (%s): %pV\n", level, sb->s_id, &vaf); + else + printk("%sfscrypt: %pV\n", level, &vaf); + va_end(args); +} + /** * fscrypt_init() - Set up for fs encryption. */ static int __init fscrypt_init(void) { + /* + * Use an unbound workqueue to allow bios to be decrypted in parallel + * even when they happen to complete on the same CPU. This sacrifices + * locality, but it's worthwhile since decryption is CPU-intensive. + * + * Also use a high-priority workqueue to prioritize decryption work, + * which blocks reads from completing, over regular application tasks. + */ fscrypt_read_workqueue = alloc_workqueue("fscrypt_read_queue", - WQ_HIGHPRI, 0); + WQ_UNBOUND | WQ_HIGHPRI, + num_online_cpus()); if (!fscrypt_read_workqueue) goto fail; diff --git a/fs/crypto/fname.c b/fs/crypto/fname.c index e33f3d3c5ade..d7a0f682ca12 100644 --- a/fs/crypto/fname.c +++ b/fs/crypto/fname.c @@ -59,11 +59,8 @@ int fname_encrypt(struct inode *inode, const struct qstr *iname, /* Set up the encryption request */ req = skcipher_request_alloc(tfm, GFP_NOFS); - if (!req) { - printk_ratelimited(KERN_ERR - "%s: skcipher_request_alloc() failed\n", __func__); + if (!req) return -ENOMEM; - } skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP, crypto_req_done, &wait); @@ -74,8 +71,9 @@ int fname_encrypt(struct inode *inode, const struct qstr *iname, res = crypto_wait_req(crypto_skcipher_encrypt(req), &wait); skcipher_request_free(req); if (res < 0) { - printk_ratelimited(KERN_ERR - "%s: Error (error code %d)\n", __func__, res); + fscrypt_err(inode->i_sb, + "Filename encryption failed for inode %lu: %d", + inode->i_ino, res); return res; } @@ -96,23 +94,14 @@ static int fname_decrypt(struct inode *inode, struct skcipher_request *req = NULL; DECLARE_CRYPTO_WAIT(wait); struct scatterlist src_sg, dst_sg; - struct fscrypt_info *ci = inode->i_crypt_info; - struct crypto_skcipher *tfm = ci->ci_ctfm; + struct crypto_skcipher *tfm = inode->i_crypt_info->ci_ctfm; int res = 0; char iv[FS_CRYPTO_BLOCK_SIZE]; - unsigned lim; - - lim = inode->i_sb->s_cop->max_namelen(inode); - if (iname->len <= 0 || iname->len > lim) - return -EIO; /* Allocate request */ req = skcipher_request_alloc(tfm, GFP_NOFS); - if (!req) { - printk_ratelimited(KERN_ERR - "%s: crypto_request_alloc() failed\n", __func__); + if (!req) return -ENOMEM; - } skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP, crypto_req_done, &wait); @@ -127,8 +116,9 @@ static int fname_decrypt(struct inode *inode, res = crypto_wait_req(crypto_skcipher_decrypt(req), &wait); skcipher_request_free(req); if (res < 0) { - printk_ratelimited(KERN_ERR - "%s: Error (error code %d)\n", __func__, res); + fscrypt_err(inode->i_sb, + "Filename decryption failed for inode %lu: %d", + inode->i_ino, res); return res; } @@ -341,12 +331,12 @@ int fscrypt_setup_filename(struct inode *dir, const struct qstr *iname, return 0; } ret = fscrypt_get_encryption_info(dir); - if (ret && ret != -EOPNOTSUPP) + if (ret) return ret; if (dir->i_crypt_info) { if (!fscrypt_fname_encrypted_size(dir, iname->len, - dir->i_sb->s_cop->max_namelen(dir), + dir->i_sb->s_cop->max_namelen, &fname->crypto_buf.len)) return -ENAMETOOLONG; fname->crypto_buf.name = kmalloc(fname->crypto_buf.len, diff --git a/fs/crypto/fscrypt_private.h b/fs/crypto/fscrypt_private.h index ad6722bae8b7..37562394c5de 100644 --- a/fs/crypto/fscrypt_private.h +++ b/fs/crypto/fscrypt_private.h @@ -18,15 +18,7 @@ /* Encryption parameters */ #define FS_IV_SIZE 16 -#define FS_AES_128_ECB_KEY_SIZE 16 -#define FS_AES_128_CBC_KEY_SIZE 16 -#define FS_AES_128_CTS_KEY_SIZE 16 -#define FS_AES_256_GCM_KEY_SIZE 32 -#define FS_AES_256_CBC_KEY_SIZE 32 -#define FS_AES_256_CTS_KEY_SIZE 32 -#define FS_AES_256_XTS_KEY_SIZE 64 - -#define FS_KEY_DERIVATION_NONCE_SIZE 16 +#define FS_KEY_DERIVATION_NONCE_SIZE 16 /** * Encryption context for inode @@ -91,6 +83,10 @@ static inline bool fscrypt_valid_enc_modes(u32 contents_mode, filenames_mode == FS_ENCRYPTION_MODE_AES_256_CTS) return true; + if (contents_mode == FS_ENCRYPTION_MODE_SPECK128_256_XTS && + filenames_mode == FS_ENCRYPTION_MODE_SPECK128_256_CTS) + return true; + return false; } @@ -106,6 +102,15 @@ extern int fscrypt_do_page_crypto(const struct inode *inode, gfp_t gfp_flags); extern struct page *fscrypt_alloc_bounce_page(struct fscrypt_ctx *ctx, gfp_t gfp_flags); +extern const struct dentry_operations fscrypt_d_ops; + +extern void __printf(3, 4) __cold +fscrypt_msg(struct super_block *sb, const char *level, const char *fmt, ...); + +#define fscrypt_warn(sb, fmt, ...) \ + fscrypt_msg(sb, KERN_WARNING, fmt, ##__VA_ARGS__) +#define fscrypt_err(sb, fmt, ...) \ + fscrypt_msg(sb, KERN_ERR, fmt, ##__VA_ARGS__) /* fname.c */ extern int fname_encrypt(struct inode *inode, const struct qstr *iname, diff --git a/fs/crypto/hooks.c b/fs/crypto/hooks.c index bec06490fb13..926e5df20ec3 100644 --- a/fs/crypto/hooks.c +++ b/fs/crypto/hooks.c @@ -39,8 +39,9 @@ int fscrypt_file_open(struct inode *inode, struct file *filp) dir = dget_parent(file_dentry(filp)); if (IS_ENCRYPTED(d_inode(dir)) && !fscrypt_has_permitted_context(d_inode(dir), inode)) { - pr_warn_ratelimited("fscrypt: inconsistent encryption contexts: %lu/%lu", - d_inode(dir)->i_ino, inode->i_ino); + fscrypt_warn(inode->i_sb, + "inconsistent encryption contexts: %lu/%lu", + d_inode(dir)->i_ino, inode->i_ino); err = -EPERM; } dput(dir); diff --git a/fs/crypto/keyinfo.c b/fs/crypto/keyinfo.c index 05f5ee1f0705..e997ca51192f 100644 --- a/fs/crypto/keyinfo.c +++ b/fs/crypto/keyinfo.c @@ -19,17 +19,16 @@ static struct crypto_shash *essiv_hash_tfm; -/** - * derive_key_aes() - Derive a key using AES-128-ECB - * @deriving_key: Encryption key used for derivation. - * @source_key: Source key to which to apply derivation. - * @derived_raw_key: Derived raw key. +/* + * Key derivation function. This generates the derived key by encrypting the + * master key with AES-128-ECB using the inode's nonce as the AES key. * - * Return: Zero on success; non-zero otherwise. + * The master key must be at least as long as the derived key. If the master + * key is longer, then only the first 'derived_keysize' bytes are used. */ -static int derive_key_aes(u8 deriving_key[FS_AES_128_ECB_KEY_SIZE], - const struct fscrypt_key *source_key, - u8 derived_raw_key[FS_MAX_KEY_SIZE]) +static int derive_key_aes(const u8 *master_key, + const struct fscrypt_context *ctx, + u8 *derived_key, unsigned int derived_keysize) { int res = 0; struct skcipher_request *req = NULL; @@ -51,14 +50,13 @@ static int derive_key_aes(u8 deriving_key[FS_AES_128_ECB_KEY_SIZE], skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP, crypto_req_done, &wait); - res = crypto_skcipher_setkey(tfm, deriving_key, - FS_AES_128_ECB_KEY_SIZE); + res = crypto_skcipher_setkey(tfm, ctx->nonce, sizeof(ctx->nonce)); if (res < 0) goto out; - sg_init_one(&src_sg, source_key->raw, source_key->size); - sg_init_one(&dst_sg, derived_raw_key, source_key->size); - skcipher_request_set_crypt(req, &src_sg, &dst_sg, source_key->size, + sg_init_one(&src_sg, master_key, derived_keysize); + sg_init_one(&dst_sg, derived_key, derived_keysize); + skcipher_request_set_crypt(req, &src_sg, &dst_sg, derived_keysize, NULL); res = crypto_wait_req(crypto_skcipher_encrypt(req), &wait); out: @@ -67,101 +65,147 @@ out: return res; } -static int validate_user_key(struct fscrypt_info *crypt_info, - struct fscrypt_context *ctx, u8 *raw_key, - const char *prefix, int min_keysize) +/* + * Search the current task's subscribed keyrings for a "logon" key with + * description prefix:descriptor, and if found acquire a read lock on it and + * return a pointer to its validated payload in *payload_ret. + */ +static struct key * +find_and_lock_process_key(const char *prefix, + const u8 descriptor[FS_KEY_DESCRIPTOR_SIZE], + unsigned int min_keysize, + const struct fscrypt_key **payload_ret) { char *description; - struct key *keyring_key; - struct fscrypt_key *master_key; + struct key *key; const struct user_key_payload *ukp; - int res; + const struct fscrypt_key *payload; description = kasprintf(GFP_NOFS, "%s%*phN", prefix, - FS_KEY_DESCRIPTOR_SIZE, - ctx->master_key_descriptor); + FS_KEY_DESCRIPTOR_SIZE, descriptor); if (!description) - return -ENOMEM; + return ERR_PTR(-ENOMEM); - keyring_key = request_key(&key_type_logon, description, NULL); + key = request_key(&key_type_logon, description, NULL); kfree(description); - if (IS_ERR(keyring_key)) - return PTR_ERR(keyring_key); - down_read(&keyring_key->sem); - - if (keyring_key->type != &key_type_logon) { - printk_once(KERN_WARNING - "%s: key type must be logon\n", __func__); - res = -ENOKEY; - goto out; - } - ukp = user_key_payload_locked(keyring_key); - if (!ukp) { - /* key was revoked before we acquired its semaphore */ - res = -EKEYREVOKED; - goto out; + if (IS_ERR(key)) + return key; + + down_read(&key->sem); + ukp = user_key_payload_locked(key); + + if (!ukp) /* was the key revoked before we acquired its semaphore? */ + goto invalid; + + payload = (const struct fscrypt_key *)ukp->data; + + if (ukp->datalen != sizeof(struct fscrypt_key) || + payload->size < 1 || payload->size > FS_MAX_KEY_SIZE) { + fscrypt_warn(NULL, + "key with description '%s' has invalid payload", + key->description); + goto invalid; } - if (ukp->datalen != sizeof(struct fscrypt_key)) { - res = -EINVAL; - goto out; + + if (payload->size < min_keysize) { + fscrypt_warn(NULL, + "key with description '%s' is too short (got %u bytes, need %u+ bytes)", + key->description, payload->size, min_keysize); + goto invalid; } - master_key = (struct fscrypt_key *)ukp->data; - BUILD_BUG_ON(FS_AES_128_ECB_KEY_SIZE != FS_KEY_DERIVATION_NONCE_SIZE); - - if (master_key->size < min_keysize || master_key->size > FS_MAX_KEY_SIZE - || master_key->size % AES_BLOCK_SIZE != 0) { - printk_once(KERN_WARNING - "%s: key size incorrect: %d\n", - __func__, master_key->size); - res = -ENOKEY; - goto out; + + *payload_ret = payload; + return key; + +invalid: + up_read(&key->sem); + key_put(key); + return ERR_PTR(-ENOKEY); +} + +/* Find the master key, then derive the inode's actual encryption key */ +static int find_and_derive_key(const struct inode *inode, + const struct fscrypt_context *ctx, + u8 *derived_key, unsigned int derived_keysize) +{ + struct key *key; + const struct fscrypt_key *payload; + int err; + + key = find_and_lock_process_key(FS_KEY_DESC_PREFIX, + ctx->master_key_descriptor, + derived_keysize, &payload); + if (key == ERR_PTR(-ENOKEY) && inode->i_sb->s_cop->key_prefix) { + key = find_and_lock_process_key(inode->i_sb->s_cop->key_prefix, + ctx->master_key_descriptor, + derived_keysize, &payload); } - res = derive_key_aes(ctx->nonce, master_key, raw_key); -out: - up_read(&keyring_key->sem); - key_put(keyring_key); - return res; + if (IS_ERR(key)) + return PTR_ERR(key); + err = derive_key_aes(payload->raw, ctx, derived_key, derived_keysize); + up_read(&key->sem); + key_put(key); + return err; } -static const struct { +static struct fscrypt_mode { + const char *friendly_name; const char *cipher_str; int keysize; + bool logged_impl_name; } available_modes[] = { - [FS_ENCRYPTION_MODE_AES_256_XTS] = { "xts(aes)", - FS_AES_256_XTS_KEY_SIZE }, - [FS_ENCRYPTION_MODE_AES_256_CTS] = { "cts(cbc(aes))", - FS_AES_256_CTS_KEY_SIZE }, - [FS_ENCRYPTION_MODE_AES_128_CBC] = { "cbc(aes)", - FS_AES_128_CBC_KEY_SIZE }, - [FS_ENCRYPTION_MODE_AES_128_CTS] = { "cts(cbc(aes))", - FS_AES_128_CTS_KEY_SIZE }, + [FS_ENCRYPTION_MODE_AES_256_XTS] = { + .friendly_name = "AES-256-XTS", + .cipher_str = "xts(aes)", + .keysize = 64, + }, + [FS_ENCRYPTION_MODE_AES_256_CTS] = { + .friendly_name = "AES-256-CTS-CBC", + .cipher_str = "cts(cbc(aes))", + .keysize = 32, + }, + [FS_ENCRYPTION_MODE_AES_128_CBC] = { + .friendly_name = "AES-128-CBC", + .cipher_str = "cbc(aes)", + .keysize = 16, + }, + [FS_ENCRYPTION_MODE_AES_128_CTS] = { + .friendly_name = "AES-128-CTS-CBC", + .cipher_str = "cts(cbc(aes))", + .keysize = 16, + }, + [FS_ENCRYPTION_MODE_SPECK128_256_XTS] = { + .friendly_name = "Speck128/256-XTS", + .cipher_str = "xts(speck128)", + .keysize = 64, + }, + [FS_ENCRYPTION_MODE_SPECK128_256_CTS] = { + .friendly_name = "Speck128/256-CTS-CBC", + .cipher_str = "cts(cbc(speck128))", + .keysize = 32, + }, }; -static int determine_cipher_type(struct fscrypt_info *ci, struct inode *inode, - const char **cipher_str_ret, int *keysize_ret) +static struct fscrypt_mode * +select_encryption_mode(const struct fscrypt_info *ci, const struct inode *inode) { - u32 mode; - if (!fscrypt_valid_enc_modes(ci->ci_data_mode, ci->ci_filename_mode)) { - pr_warn_ratelimited("fscrypt: inode %lu uses unsupported encryption modes (contents mode %d, filenames mode %d)\n", - inode->i_ino, - ci->ci_data_mode, ci->ci_filename_mode); - return -EINVAL; + fscrypt_warn(inode->i_sb, + "inode %lu uses unsupported encryption modes (contents mode %d, filenames mode %d)", + inode->i_ino, ci->ci_data_mode, + ci->ci_filename_mode); + return ERR_PTR(-EINVAL); } - if (S_ISREG(inode->i_mode)) { - mode = ci->ci_data_mode; - } else if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) { - mode = ci->ci_filename_mode; - } else { - WARN_ONCE(1, "fscrypt: filesystem tried to load encryption info for inode %lu, which is not encryptable (file type %d)\n", - inode->i_ino, (inode->i_mode & S_IFMT)); - return -EINVAL; - } + if (S_ISREG(inode->i_mode)) + return &available_modes[ci->ci_data_mode]; + + if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) + return &available_modes[ci->ci_filename_mode]; - *cipher_str_ret = available_modes[mode].cipher_str; - *keysize_ret = available_modes[mode].keysize; - return 0; + WARN_ONCE(1, "fscrypt: filesystem tried to load encryption info for inode %lu, which is not encryptable (file type %d)\n", + inode->i_ino, (inode->i_mode & S_IFMT)); + return ERR_PTR(-EINVAL); } static void put_crypt_info(struct fscrypt_info *ci) @@ -184,8 +228,9 @@ static int derive_essiv_salt(const u8 *key, int keysize, u8 *salt) tfm = crypto_alloc_shash("sha256", 0, 0); if (IS_ERR(tfm)) { - pr_warn_ratelimited("fscrypt: error allocating SHA-256 transform: %ld\n", - PTR_ERR(tfm)); + fscrypt_warn(NULL, + "error allocating SHA-256 transform: %ld", + PTR_ERR(tfm)); return PTR_ERR(tfm); } prev_tfm = cmpxchg(&essiv_hash_tfm, NULL, tfm); @@ -245,8 +290,7 @@ int fscrypt_get_encryption_info(struct inode *inode) struct fscrypt_info *crypt_info; struct fscrypt_context ctx; struct crypto_skcipher *ctfm; - const char *cipher_str; - int keysize; + struct fscrypt_mode *mode; u8 *raw_key = NULL; int res; @@ -290,57 +334,59 @@ int fscrypt_get_encryption_info(struct inode *inode) memcpy(crypt_info->ci_master_key, ctx.master_key_descriptor, sizeof(crypt_info->ci_master_key)); - res = determine_cipher_type(crypt_info, inode, &cipher_str, &keysize); - if (res) + mode = select_encryption_mode(crypt_info, inode); + if (IS_ERR(mode)) { + res = PTR_ERR(mode); goto out; + } /* * This cannot be a stack buffer because it is passed to the scatterlist * crypto API as part of key derivation. */ res = -ENOMEM; - raw_key = kmalloc(FS_MAX_KEY_SIZE, GFP_NOFS); + raw_key = kmalloc(mode->keysize, GFP_NOFS); if (!raw_key) goto out; - res = validate_user_key(crypt_info, &ctx, raw_key, FS_KEY_DESC_PREFIX, - keysize); - if (res && inode->i_sb->s_cop->key_prefix) { - int res2 = validate_user_key(crypt_info, &ctx, raw_key, - inode->i_sb->s_cop->key_prefix, - keysize); - if (res2) { - if (res2 == -ENOKEY) - res = -ENOKEY; - goto out; - } - } else if (res) { + res = find_and_derive_key(inode, &ctx, raw_key, mode->keysize); + if (res) goto out; - } - ctfm = crypto_alloc_skcipher(cipher_str, 0, 0); - if (!ctfm || IS_ERR(ctfm)) { - res = ctfm ? PTR_ERR(ctfm) : -ENOMEM; - pr_debug("%s: error %d (inode %lu) allocating crypto tfm\n", - __func__, res, inode->i_ino); + + ctfm = crypto_alloc_skcipher(mode->cipher_str, 0, 0); + if (IS_ERR(ctfm)) { + res = PTR_ERR(ctfm); + fscrypt_warn(inode->i_sb, + "error allocating '%s' transform for inode %lu: %d", + mode->cipher_str, inode->i_ino, res); goto out; } + if (unlikely(!mode->logged_impl_name)) { + /* + * fscrypt performance can vary greatly depending on which + * crypto algorithm implementation is used. Help people debug + * performance problems by logging the ->cra_driver_name the + * first time a mode is used. Note that multiple threads can + * race here, but it doesn't really matter. + */ + mode->logged_impl_name = true; + pr_info("fscrypt: %s using implementation \"%s\"\n", + mode->friendly_name, + crypto_skcipher_alg(ctfm)->base.cra_driver_name); + } crypt_info->ci_ctfm = ctfm; - crypto_skcipher_clear_flags(ctfm, ~0); crypto_skcipher_set_flags(ctfm, CRYPTO_TFM_REQ_WEAK_KEY); - /* - * if the provided key is longer than keysize, we use the first - * keysize bytes of the derived key only - */ - res = crypto_skcipher_setkey(ctfm, raw_key, keysize); + res = crypto_skcipher_setkey(ctfm, raw_key, mode->keysize); if (res) goto out; if (S_ISREG(inode->i_mode) && crypt_info->ci_data_mode == FS_ENCRYPTION_MODE_AES_128_CBC) { - res = init_essiv_generator(crypt_info, raw_key, keysize); + res = init_essiv_generator(crypt_info, raw_key, mode->keysize); if (res) { - pr_debug("%s: error %d (inode %lu) allocating essiv tfm\n", - __func__, res, inode->i_ino); + fscrypt_warn(inode->i_sb, + "error initializing ESSIV generator for inode %lu: %d", + inode->i_ino, res); goto out; } } @@ -73,16 +73,15 @@ fs_initcall(init_dax_wait_table); #define RADIX_DAX_ZERO_PAGE (1 << (RADIX_TREE_EXCEPTIONAL_SHIFT + 2)) #define RADIX_DAX_EMPTY (1 << (RADIX_TREE_EXCEPTIONAL_SHIFT + 3)) -static unsigned long dax_radix_sector(void *entry) +static unsigned long dax_radix_pfn(void *entry) { return (unsigned long)entry >> RADIX_DAX_SHIFT; } -static void *dax_radix_locked_entry(sector_t sector, unsigned long flags) +static void *dax_radix_locked_entry(unsigned long pfn, unsigned long flags) { return (void *)(RADIX_TREE_EXCEPTIONAL_ENTRY | flags | - ((unsigned long)sector << RADIX_DAX_SHIFT) | - RADIX_DAX_ENTRY_LOCK); + (pfn << RADIX_DAX_SHIFT) | RADIX_DAX_ENTRY_LOCK); } static unsigned int dax_radix_order(void *entry) @@ -159,11 +158,9 @@ static int wake_exceptional_entry_func(wait_queue_entry_t *wait, unsigned int mo } /* - * We do not necessarily hold the mapping->tree_lock when we call this - * function so it is possible that 'entry' is no longer a valid item in the - * radix tree. This is okay because all we really need to do is to find the - * correct waitqueue where tasks might be waiting for that old 'entry' and - * wake them. + * @entry may no longer be the entry at the index in the mapping. + * The important information it's conveying is whether the entry at + * this index used to be a PMD entry. */ static void dax_wake_mapping_entry_waiter(struct address_space *mapping, pgoff_t index, void *entry, bool wake_all) @@ -175,7 +172,7 @@ static void dax_wake_mapping_entry_waiter(struct address_space *mapping, /* * Checking for locked entry and prepare_to_wait_exclusive() happens - * under mapping->tree_lock, ditto for entry handling in our callers. + * under the i_pages lock, ditto for entry handling in our callers. * So at this point all tasks that could have seen our entry locked * must be in the waitqueue and the following check will see them. */ @@ -184,41 +181,39 @@ static void dax_wake_mapping_entry_waiter(struct address_space *mapping, } /* - * Check whether the given slot is locked. The function must be called with - * mapping->tree_lock held + * Check whether the given slot is locked. Must be called with the i_pages + * lock held. */ static inline int slot_locked(struct address_space *mapping, void **slot) { unsigned long entry = (unsigned long) - radix_tree_deref_slot_protected(slot, &mapping->tree_lock); + radix_tree_deref_slot_protected(slot, &mapping->i_pages.xa_lock); return entry & RADIX_DAX_ENTRY_LOCK; } /* - * Mark the given slot is locked. The function must be called with - * mapping->tree_lock held + * Mark the given slot as locked. Must be called with the i_pages lock held. */ static inline void *lock_slot(struct address_space *mapping, void **slot) { unsigned long entry = (unsigned long) - radix_tree_deref_slot_protected(slot, &mapping->tree_lock); + radix_tree_deref_slot_protected(slot, &mapping->i_pages.xa_lock); entry |= RADIX_DAX_ENTRY_LOCK; - radix_tree_replace_slot(&mapping->page_tree, slot, (void *)entry); + radix_tree_replace_slot(&mapping->i_pages, slot, (void *)entry); return (void *)entry; } /* - * Mark the given slot is unlocked. The function must be called with - * mapping->tree_lock held + * Mark the given slot as unlocked. Must be called with the i_pages lock held. */ static inline void *unlock_slot(struct address_space *mapping, void **slot) { unsigned long entry = (unsigned long) - radix_tree_deref_slot_protected(slot, &mapping->tree_lock); + radix_tree_deref_slot_protected(slot, &mapping->i_pages.xa_lock); entry &= ~(unsigned long)RADIX_DAX_ENTRY_LOCK; - radix_tree_replace_slot(&mapping->page_tree, slot, (void *)entry); + radix_tree_replace_slot(&mapping->i_pages, slot, (void *)entry); return (void *)entry; } @@ -229,7 +224,7 @@ static inline void *unlock_slot(struct address_space *mapping, void **slot) * put_locked_mapping_entry() when he locked the entry and now wants to * unlock it. * - * The function must be called with mapping->tree_lock held. + * Must be called with the i_pages lock held. */ static void *get_unlocked_mapping_entry(struct address_space *mapping, pgoff_t index, void ***slotp) @@ -242,7 +237,7 @@ static void *get_unlocked_mapping_entry(struct address_space *mapping, ewait.wait.func = wake_exceptional_entry_func; for (;;) { - entry = __radix_tree_lookup(&mapping->page_tree, index, NULL, + entry = __radix_tree_lookup(&mapping->i_pages, index, NULL, &slot); if (!entry || WARN_ON_ONCE(!radix_tree_exceptional_entry(entry)) || @@ -255,10 +250,10 @@ static void *get_unlocked_mapping_entry(struct address_space *mapping, wq = dax_entry_waitqueue(mapping, index, entry, &ewait.key); prepare_to_wait_exclusive(wq, &ewait.wait, TASK_UNINTERRUPTIBLE); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); schedule(); finish_wait(wq, &ewait.wait); - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(&mapping->i_pages); } } @@ -267,15 +262,15 @@ static void dax_unlock_mapping_entry(struct address_space *mapping, { void *entry, **slot; - spin_lock_irq(&mapping->tree_lock); - entry = __radix_tree_lookup(&mapping->page_tree, index, NULL, &slot); + xa_lock_irq(&mapping->i_pages); + entry = __radix_tree_lookup(&mapping->i_pages, index, NULL, &slot); if (WARN_ON_ONCE(!entry || !radix_tree_exceptional_entry(entry) || !slot_locked(mapping, slot))) { - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); return; } unlock_slot(mapping, slot); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); dax_wake_mapping_entry_waiter(mapping, index, entry, false); } @@ -299,6 +294,63 @@ static void put_unlocked_mapping_entry(struct address_space *mapping, dax_wake_mapping_entry_waiter(mapping, index, entry, false); } +static unsigned long dax_entry_size(void *entry) +{ + if (dax_is_zero_entry(entry)) + return 0; + else if (dax_is_empty_entry(entry)) + return 0; + else if (dax_is_pmd_entry(entry)) + return PMD_SIZE; + else + return PAGE_SIZE; +} + +static unsigned long dax_radix_end_pfn(void *entry) +{ + return dax_radix_pfn(entry) + dax_entry_size(entry) / PAGE_SIZE; +} + +/* + * Iterate through all mapped pfns represented by an entry, i.e. skip + * 'empty' and 'zero' entries. + */ +#define for_each_mapped_pfn(entry, pfn) \ + for (pfn = dax_radix_pfn(entry); \ + pfn < dax_radix_end_pfn(entry); pfn++) + +static void dax_associate_entry(void *entry, struct address_space *mapping) +{ + unsigned long pfn; + + if (IS_ENABLED(CONFIG_FS_DAX_LIMITED)) + return; + + for_each_mapped_pfn(entry, pfn) { + struct page *page = pfn_to_page(pfn); + + WARN_ON_ONCE(page->mapping); + page->mapping = mapping; + } +} + +static void dax_disassociate_entry(void *entry, struct address_space *mapping, + bool trunc) +{ + unsigned long pfn; + + if (IS_ENABLED(CONFIG_FS_DAX_LIMITED)) + return; + + for_each_mapped_pfn(entry, pfn) { + struct page *page = pfn_to_page(pfn); + + WARN_ON_ONCE(trunc && page_ref_count(page) > 1); + WARN_ON_ONCE(page->mapping && page->mapping != mapping); + page->mapping = NULL; + } +} + /* * Find radix tree entry at given index. If it points to an exceptional entry, * return it with the radix tree entry locked. If the radix tree doesn't @@ -332,7 +384,7 @@ static void *grab_mapping_entry(struct address_space *mapping, pgoff_t index, void *entry, **slot; restart: - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(&mapping->i_pages); entry = get_unlocked_mapping_entry(mapping, index, &slot); if (WARN_ON_ONCE(entry && !radix_tree_exceptional_entry(entry))) { @@ -364,12 +416,12 @@ restart: if (pmd_downgrade) { /* * Make sure 'entry' remains valid while we drop - * mapping->tree_lock. + * the i_pages lock. */ entry = lock_slot(mapping, slot); } - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); /* * Besides huge zero pages the only other thing that gets * downgraded are empty entries which don't need to be @@ -386,26 +438,27 @@ restart: put_locked_mapping_entry(mapping, index); return ERR_PTR(err); } - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(&mapping->i_pages); if (!entry) { /* - * We needed to drop the page_tree lock while calling + * We needed to drop the i_pages lock while calling * radix_tree_preload() and we didn't have an entry to * lock. See if another thread inserted an entry at * our index during this time. */ - entry = __radix_tree_lookup(&mapping->page_tree, index, + entry = __radix_tree_lookup(&mapping->i_pages, index, NULL, &slot); if (entry) { radix_tree_preload_end(); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); goto restart; } } if (pmd_downgrade) { - radix_tree_delete(&mapping->page_tree, index); + dax_disassociate_entry(entry, mapping, false); + radix_tree_delete(&mapping->i_pages, index); mapping->nrexceptional--; dax_wake_mapping_entry_waiter(mapping, index, entry, true); @@ -413,11 +466,11 @@ restart: entry = dax_radix_locked_entry(0, size_flag | RADIX_DAX_EMPTY); - err = __radix_tree_insert(&mapping->page_tree, index, + err = __radix_tree_insert(&mapping->i_pages, index, dax_radix_order(entry), entry); radix_tree_preload_end(); if (err) { - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); /* * Our insertion of a DAX entry failed, most likely * because we were inserting a PMD entry and it @@ -430,12 +483,12 @@ restart: } /* Good, we have inserted empty locked entry into the tree. */ mapping->nrexceptional++; - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); return entry; } entry = lock_slot(mapping, slot); out_unlock: - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); return entry; } @@ -444,22 +497,23 @@ static int __dax_invalidate_mapping_entry(struct address_space *mapping, { int ret = 0; void *entry; - struct radix_tree_root *page_tree = &mapping->page_tree; + struct radix_tree_root *pages = &mapping->i_pages; - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(pages); entry = get_unlocked_mapping_entry(mapping, index, NULL); if (!entry || WARN_ON_ONCE(!radix_tree_exceptional_entry(entry))) goto out; if (!trunc && - (radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_DIRTY) || - radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_TOWRITE))) + (radix_tree_tag_get(pages, index, PAGECACHE_TAG_DIRTY) || + radix_tree_tag_get(pages, index, PAGECACHE_TAG_TOWRITE))) goto out; - radix_tree_delete(page_tree, index); + dax_disassociate_entry(entry, mapping, trunc); + radix_tree_delete(pages, index); mapping->nrexceptional--; ret = 1; out: put_unlocked_mapping_entry(mapping, index, entry); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(pages); return ret; } /* @@ -526,12 +580,13 @@ static int copy_user_dax(struct block_device *bdev, struct dax_device *dax_dev, */ static void *dax_insert_mapping_entry(struct address_space *mapping, struct vm_fault *vmf, - void *entry, sector_t sector, + void *entry, pfn_t pfn_t, unsigned long flags, bool dirty) { - struct radix_tree_root *page_tree = &mapping->page_tree; - void *new_entry; + struct radix_tree_root *pages = &mapping->i_pages; + unsigned long pfn = pfn_t_to_pfn(pfn_t); pgoff_t index = vmf->pgoff; + void *new_entry; if (dirty) __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); @@ -545,8 +600,12 @@ static void *dax_insert_mapping_entry(struct address_space *mapping, unmap_mapping_pages(mapping, vmf->pgoff, 1, false); } - spin_lock_irq(&mapping->tree_lock); - new_entry = dax_radix_locked_entry(sector, flags); + xa_lock_irq(pages); + new_entry = dax_radix_locked_entry(pfn, flags); + if (dax_entry_size(entry) != dax_entry_size(new_entry)) { + dax_disassociate_entry(entry, mapping, false); + dax_associate_entry(new_entry, mapping); + } if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry)) { /* @@ -561,17 +620,17 @@ static void *dax_insert_mapping_entry(struct address_space *mapping, void **slot; void *ret; - ret = __radix_tree_lookup(page_tree, index, &node, &slot); + ret = __radix_tree_lookup(pages, index, &node, &slot); WARN_ON_ONCE(ret != entry); - __radix_tree_replace(page_tree, node, slot, + __radix_tree_replace(pages, node, slot, new_entry, NULL); entry = new_entry; } if (dirty) - radix_tree_tag_set(page_tree, index, PAGECACHE_TAG_DIRTY); + radix_tree_tag_set(pages, index, PAGECACHE_TAG_DIRTY); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(pages); return entry; } @@ -618,7 +677,7 @@ static void dax_mapping_entry_mkclean(struct address_space *mapping, * downgrading page table protection not changing it to point * to a new page. * - * See Documentation/vm/mmu_notifier.txt + * See Documentation/vm/mmu_notifier.rst */ if (pmdp) { #ifdef CONFIG_FS_DAX_PMD @@ -657,17 +716,14 @@ unlock_pte: i_mmap_unlock_read(mapping); } -static int dax_writeback_one(struct block_device *bdev, - struct dax_device *dax_dev, struct address_space *mapping, - pgoff_t index, void *entry) +static int dax_writeback_one(struct dax_device *dax_dev, + struct address_space *mapping, pgoff_t index, void *entry) { - struct radix_tree_root *page_tree = &mapping->page_tree; - void *entry2, **slot, *kaddr; - long ret = 0, id; - sector_t sector; - pgoff_t pgoff; + struct radix_tree_root *pages = &mapping->i_pages; + void *entry2, **slot; + unsigned long pfn; + long ret = 0; size_t size; - pfn_t pfn; /* * A page got tagged dirty in DAX mapping? Something is seriously @@ -676,17 +732,17 @@ static int dax_writeback_one(struct block_device *bdev, if (WARN_ON(!radix_tree_exceptional_entry(entry))) return -EIO; - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(pages); entry2 = get_unlocked_mapping_entry(mapping, index, &slot); /* Entry got punched out / reallocated? */ if (!entry2 || WARN_ON_ONCE(!radix_tree_exceptional_entry(entry2))) goto put_unlocked; /* * Entry got reallocated elsewhere? No need to writeback. We have to - * compare sectors as we must not bail out due to difference in lockbit + * compare pfns as we must not bail out due to difference in lockbit * or entry type. */ - if (dax_radix_sector(entry2) != dax_radix_sector(entry)) + if (dax_radix_pfn(entry2) != dax_radix_pfn(entry)) goto put_unlocked; if (WARN_ON_ONCE(dax_is_empty_entry(entry) || dax_is_zero_entry(entry))) { @@ -695,7 +751,7 @@ static int dax_writeback_one(struct block_device *bdev, } /* Another fsync thread may have already written back this entry */ - if (!radix_tree_tag_get(page_tree, index, PAGECACHE_TAG_TOWRITE)) + if (!radix_tree_tag_get(pages, index, PAGECACHE_TAG_TOWRITE)) goto put_unlocked; /* Lock the entry to serialize with page faults */ entry = lock_slot(mapping, slot); @@ -703,60 +759,40 @@ static int dax_writeback_one(struct block_device *bdev, * We can clear the tag now but we have to be careful so that concurrent * dax_writeback_one() calls for the same index cannot finish before we * actually flush the caches. This is achieved as the calls will look - * at the entry only under tree_lock and once they do that they will - * see the entry locked and wait for it to unlock. + * at the entry only under the i_pages lock and once they do that + * they will see the entry locked and wait for it to unlock. */ - radix_tree_tag_clear(page_tree, index, PAGECACHE_TAG_TOWRITE); - spin_unlock_irq(&mapping->tree_lock); + radix_tree_tag_clear(pages, index, PAGECACHE_TAG_TOWRITE); + xa_unlock_irq(pages); /* * Even if dax_writeback_mapping_range() was given a wbc->range_start * in the middle of a PMD, the 'index' we are given will be aligned to - * the start index of the PMD, as will the sector we pull from - * 'entry'. This allows us to flush for PMD_SIZE and not have to - * worry about partial PMD writebacks. + * the start index of the PMD, as will the pfn we pull from 'entry'. + * This allows us to flush for PMD_SIZE and not have to worry about + * partial PMD writebacks. */ - sector = dax_radix_sector(entry); + pfn = dax_radix_pfn(entry); size = PAGE_SIZE << dax_radix_order(entry); - id = dax_read_lock(); - ret = bdev_dax_pgoff(bdev, sector, size, &pgoff); - if (ret) - goto dax_unlock; - - /* - * dax_direct_access() may sleep, so cannot hold tree_lock over - * its invocation. - */ - ret = dax_direct_access(dax_dev, pgoff, size / PAGE_SIZE, &kaddr, &pfn); - if (ret < 0) - goto dax_unlock; - - if (WARN_ON_ONCE(ret < size / PAGE_SIZE)) { - ret = -EIO; - goto dax_unlock; - } - - dax_mapping_entry_mkclean(mapping, index, pfn_t_to_pfn(pfn)); - dax_flush(dax_dev, kaddr, size); + dax_mapping_entry_mkclean(mapping, index, pfn); + dax_flush(dax_dev, page_address(pfn_to_page(pfn)), size); /* * After we have flushed the cache, we can clear the dirty tag. There * cannot be new dirty data in the pfn after the flush has completed as * the pfn mappings are writeprotected and fault waits for mapping * entry lock. */ - spin_lock_irq(&mapping->tree_lock); - radix_tree_tag_clear(page_tree, index, PAGECACHE_TAG_DIRTY); - spin_unlock_irq(&mapping->tree_lock); + xa_lock_irq(pages); + radix_tree_tag_clear(pages, index, PAGECACHE_TAG_DIRTY); + xa_unlock_irq(pages); trace_dax_writeback_one(mapping->host, index, size >> PAGE_SHIFT); - dax_unlock: - dax_read_unlock(id); put_locked_mapping_entry(mapping, index); return ret; put_unlocked: put_unlocked_mapping_entry(mapping, index, entry2); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(pages); return ret; } @@ -808,8 +844,8 @@ int dax_writeback_mapping_range(struct address_space *mapping, break; } - ret = dax_writeback_one(bdev, dax_dev, mapping, - indices[i], pvec.pages[i]); + ret = dax_writeback_one(dax_dev, mapping, indices[i], + pvec.pages[i]); if (ret < 0) { mapping_set_error(mapping, ret); goto out; @@ -877,6 +913,7 @@ static int dax_load_hole(struct address_space *mapping, void *entry, int ret = VM_FAULT_NOPAGE; struct page *zero_page; void *entry2; + pfn_t pfn; zero_page = ZERO_PAGE(0); if (unlikely(!zero_page)) { @@ -884,14 +921,15 @@ static int dax_load_hole(struct address_space *mapping, void *entry, goto out; } - entry2 = dax_insert_mapping_entry(mapping, vmf, entry, 0, + pfn = page_to_pfn_t(zero_page); + entry2 = dax_insert_mapping_entry(mapping, vmf, entry, pfn, RADIX_DAX_ZERO_PAGE, false); if (IS_ERR(entry2)) { ret = VM_FAULT_SIGBUS; goto out; } - vm_insert_mixed(vmf->vma, vaddr, page_to_pfn_t(zero_page)); + vm_insert_mixed(vmf->vma, vaddr, pfn); out: trace_dax_load_hole(inode, vmf, ret); return ret; @@ -1200,8 +1238,7 @@ static int dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp, if (error < 0) goto error_finish_iomap; - entry = dax_insert_mapping_entry(mapping, vmf, entry, - dax_iomap_sector(&iomap, pos), + entry = dax_insert_mapping_entry(mapping, vmf, entry, pfn, 0, write && !sync); if (IS_ERR(entry)) { error = PTR_ERR(entry); @@ -1280,13 +1317,15 @@ static int dax_pmd_load_hole(struct vm_fault *vmf, struct iomap *iomap, void *ret = NULL; spinlock_t *ptl; pmd_t pmd_entry; + pfn_t pfn; zero_page = mm_get_huge_zero_page(vmf->vma->vm_mm); if (unlikely(!zero_page)) goto fallback; - ret = dax_insert_mapping_entry(mapping, vmf, entry, 0, + pfn = page_to_pfn_t(zero_page); + ret = dax_insert_mapping_entry(mapping, vmf, entry, pfn, RADIX_DAX_PMD | RADIX_DAX_ZERO_PAGE, false); if (IS_ERR(ret)) goto fallback; @@ -1409,8 +1448,7 @@ static int dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, if (error < 0) goto finish_iomap; - entry = dax_insert_mapping_entry(mapping, vmf, entry, - dax_iomap_sector(&iomap, pos), + entry = dax_insert_mapping_entry(mapping, vmf, entry, pfn, RADIX_DAX_PMD, write && !sync); if (IS_ERR(entry)) goto finish_iomap; @@ -1524,21 +1562,21 @@ static int dax_insert_pfn_mkwrite(struct vm_fault *vmf, pgoff_t index = vmf->pgoff; int vmf_ret, error; - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(&mapping->i_pages); entry = get_unlocked_mapping_entry(mapping, index, &slot); /* Did we race with someone splitting entry or so? */ if (!entry || (pe_size == PE_SIZE_PTE && !dax_is_pte_entry(entry)) || (pe_size == PE_SIZE_PMD && !dax_is_pmd_entry(entry))) { put_unlocked_mapping_entry(mapping, index, entry); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); trace_dax_insert_pfn_mkwrite_no_entry(mapping->host, vmf, VM_FAULT_NOPAGE); return VM_FAULT_NOPAGE; } - radix_tree_tag_set(&mapping->page_tree, index, PAGECACHE_TAG_DIRTY); + radix_tree_tag_set(&mapping->i_pages, index, PAGECACHE_TAG_DIRTY); entry = lock_slot(mapping, slot); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); switch (pe_size) { case PE_SIZE_PTE: error = vm_insert_mixed_mkwrite(vmf->vma, vmf->address, pfn); diff --git a/fs/dcache.c b/fs/dcache.c index 593079176123..0e8e5de3c48a 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -257,11 +257,25 @@ static void __d_free(struct rcu_head *head) kmem_cache_free(dentry_cache, dentry); } +static void __d_free_external_name(struct rcu_head *head) +{ + struct external_name *name = container_of(head, struct external_name, + u.head); + + mod_node_page_state(page_pgdat(virt_to_page(name)), + NR_INDIRECTLY_RECLAIMABLE_BYTES, + -ksize(name)); + + kfree(name); +} + static void __d_free_external(struct rcu_head *head) { struct dentry *dentry = container_of(head, struct dentry, d_u.d_rcu); - kfree(external_name(dentry)); - kmem_cache_free(dentry_cache, dentry); + + __d_free_external_name(&external_name(dentry)->u.head); + + kmem_cache_free(dentry_cache, dentry); } static inline int dname_external(const struct dentry *dentry) @@ -291,7 +305,7 @@ void release_dentry_name_snapshot(struct name_snapshot *name) struct external_name *p; p = container_of(name->name, struct external_name, name[0]); if (unlikely(atomic_dec_and_test(&p->u.count))) - kfree_rcu(p, u.head); + call_rcu(&p->u.head, __d_free_external_name); } } EXPORT_SYMBOL(release_dentry_name_snapshot); @@ -566,6 +580,7 @@ static void __dentry_kill(struct dentry *dentry) spin_unlock(&dentry->d_lock); if (likely(can_free)) dentry_free(dentry); + cond_resched(); } static struct dentry *__lock_parent(struct dentry *dentry) @@ -813,30 +828,24 @@ static inline bool fast_dput(struct dentry *dentry) */ void dput(struct dentry *dentry) { - if (unlikely(!dentry)) - return; + while (dentry) { + might_sleep(); -repeat: - might_sleep(); + rcu_read_lock(); + if (likely(fast_dput(dentry))) { + rcu_read_unlock(); + return; + } - rcu_read_lock(); - if (likely(fast_dput(dentry))) { + /* Slow case: now with the dentry lock held */ rcu_read_unlock(); - return; - } - /* Slow case: now with the dentry lock held */ - rcu_read_unlock(); + if (likely(retain_dentry(dentry))) { + spin_unlock(&dentry->d_lock); + return; + } - if (likely(retain_dentry(dentry))) { - spin_unlock(&dentry->d_lock); - return; - } - - dentry = dentry_kill(dentry); - if (dentry) { - cond_resched(); - goto repeat; + dentry = dentry_kill(dentry); } } EXPORT_SYMBOL(dput); @@ -893,6 +902,35 @@ repeat: } EXPORT_SYMBOL(dget_parent); +static struct dentry * __d_find_any_alias(struct inode *inode) +{ + struct dentry *alias; + + if (hlist_empty(&inode->i_dentry)) + return NULL; + alias = hlist_entry(inode->i_dentry.first, struct dentry, d_u.d_alias); + __dget(alias); + return alias; +} + +/** + * d_find_any_alias - find any alias for a given inode + * @inode: inode to find an alias for + * + * If any aliases exist for the given inode, take and return a + * reference for one of them. If no aliases exist, return %NULL. + */ +struct dentry *d_find_any_alias(struct inode *inode) +{ + struct dentry *de; + + spin_lock(&inode->i_lock); + de = __d_find_any_alias(inode); + spin_unlock(&inode->i_lock); + return de; +} +EXPORT_SYMBOL(d_find_any_alias); + /** * d_find_alias - grab a hashed alias of inode * @inode: inode in question @@ -909,34 +947,19 @@ EXPORT_SYMBOL(dget_parent); */ static struct dentry *__d_find_alias(struct inode *inode) { - struct dentry *alias, *discon_alias; + struct dentry *alias; + + if (S_ISDIR(inode->i_mode)) + return __d_find_any_alias(inode); -again: - discon_alias = NULL; hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) { spin_lock(&alias->d_lock); - if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { - if (IS_ROOT(alias) && - (alias->d_flags & DCACHE_DISCONNECTED)) { - discon_alias = alias; - } else { - __dget_dlock(alias); - spin_unlock(&alias->d_lock); - return alias; - } - } - spin_unlock(&alias->d_lock); - } - if (discon_alias) { - alias = discon_alias; - spin_lock(&alias->d_lock); - if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { + if (!d_unhashed(alias)) { __dget_dlock(alias); spin_unlock(&alias->d_lock); return alias; } spin_unlock(&alias->d_lock); - goto again; } return NULL; } @@ -1191,7 +1214,6 @@ void shrink_dcache_sb(struct super_block *sb) this_cpu_sub(nr_dentry_unused, freed); shrink_dentry_list(&dispose); - cond_resched(); } while (list_lru_count(&sb->s_dentry_lru) > 0); } EXPORT_SYMBOL(shrink_dcache_sb); @@ -1215,13 +1237,11 @@ enum d_walk_ret { * @parent: start of walk * @data: data passed to @enter() and @finish() * @enter: callback when first entering the dentry - * @finish: callback when successfully finished the walk * - * The @enter() and @finish() callbacks are called with d_lock held. + * The @enter() callbacks are called with d_lock held. */ static void d_walk(struct dentry *parent, void *data, - enum d_walk_ret (*enter)(void *, struct dentry *), - void (*finish)(void *)) + enum d_walk_ret (*enter)(void *, struct dentry *)) { struct dentry *this_parent; struct list_head *next; @@ -1310,8 +1330,6 @@ ascend: if (need_seqretry(&rename_lock, seq)) goto rename_retry; rcu_read_unlock(); - if (finish) - finish(data); out_unlock: spin_unlock(&this_parent->d_lock); @@ -1360,7 +1378,7 @@ int path_has_submounts(const struct path *parent) struct check_mount data = { .mnt = parent->mnt, .mounted = 0 }; read_seqlock_excl(&mount_lock); - d_walk(parent->dentry, &data, path_check_mount, NULL); + d_walk(parent->dentry, &data, path_check_mount); read_sequnlock_excl(&mount_lock); return data.mounted; @@ -1468,12 +1486,16 @@ void shrink_dcache_parent(struct dentry *parent) data.start = parent; data.found = 0; - d_walk(parent, &data, select_collect, NULL); - if (!data.found) - break; + d_walk(parent, &data, select_collect); + + if (!list_empty(&data.dispose)) { + shrink_dentry_list(&data.dispose); + continue; + } - shrink_dentry_list(&data.dispose); cond_resched(); + if (!data.found) + break; } } EXPORT_SYMBOL(shrink_dcache_parent); @@ -1504,7 +1526,7 @@ static enum d_walk_ret umount_check(void *_data, struct dentry *dentry) static void do_one_tree(struct dentry *dentry) { shrink_dcache_parent(dentry); - d_walk(dentry, dentry, umount_check, NULL); + d_walk(dentry, dentry, umount_check); d_drop(dentry); dput(dentry); } @@ -1528,79 +1550,48 @@ void shrink_dcache_for_umount(struct super_block *sb) } } -struct detach_data { - struct select_data select; - struct dentry *mountpoint; -}; -static enum d_walk_ret detach_and_collect(void *_data, struct dentry *dentry) +static enum d_walk_ret find_submount(void *_data, struct dentry *dentry) { - struct detach_data *data = _data; - + struct dentry **victim = _data; if (d_mountpoint(dentry)) { __dget_dlock(dentry); - data->mountpoint = dentry; + *victim = dentry; return D_WALK_QUIT; } - - return select_collect(&data->select, dentry); -} - -static void check_and_drop(void *_data) -{ - struct detach_data *data = _data; - - if (!data->mountpoint && list_empty(&data->select.dispose)) - __d_drop(data->select.start); + return D_WALK_CONTINUE; } /** * d_invalidate - detach submounts, prune dcache, and drop * @dentry: dentry to invalidate (aka detach, prune and drop) - * - * no dcache lock. - * - * The final d_drop is done as an atomic operation relative to - * rename_lock ensuring there are no races with d_set_mounted. This - * ensures there are no unhashed dentries on the path to a mountpoint. */ void d_invalidate(struct dentry *dentry) { - /* - * If it's already been dropped, return OK. - */ + bool had_submounts = false; spin_lock(&dentry->d_lock); if (d_unhashed(dentry)) { spin_unlock(&dentry->d_lock); return; } + __d_drop(dentry); spin_unlock(&dentry->d_lock); /* Negative dentries can be dropped without further checks */ - if (!dentry->d_inode) { - d_drop(dentry); + if (!dentry->d_inode) return; - } + shrink_dcache_parent(dentry); for (;;) { - struct detach_data data; - - data.mountpoint = NULL; - INIT_LIST_HEAD(&data.select.dispose); - data.select.start = dentry; - data.select.found = 0; - - d_walk(dentry, &data, detach_and_collect, check_and_drop); - - if (!list_empty(&data.select.dispose)) - shrink_dentry_list(&data.select.dispose); - else if (!data.mountpoint) + struct dentry *victim = NULL; + d_walk(dentry, &victim, find_submount); + if (!victim) { + if (had_submounts) + shrink_dcache_parent(dentry); return; - - if (data.mountpoint) { - detach_mounts(data.mountpoint); - dput(data.mountpoint); } - cond_resched(); + had_submounts = true; + detach_mounts(victim); + dput(victim); } } EXPORT_SYMBOL(d_invalidate); @@ -1617,6 +1608,7 @@ EXPORT_SYMBOL(d_invalidate); struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) { + struct external_name *ext = NULL; struct dentry *dentry; char *dname; int err; @@ -1637,14 +1629,14 @@ struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) dname = dentry->d_iname; } else if (name->len > DNAME_INLINE_LEN-1) { size_t size = offsetof(struct external_name, name[1]); - struct external_name *p = kmalloc(size + name->len, - GFP_KERNEL_ACCOUNT); - if (!p) { + + ext = kmalloc(size + name->len, GFP_KERNEL_ACCOUNT); + if (!ext) { kmem_cache_free(dentry_cache, dentry); return NULL; } - atomic_set(&p->u.count, 1); - dname = p->name; + atomic_set(&ext->u.count, 1); + dname = ext->name; } else { dname = dentry->d_iname; } @@ -1683,6 +1675,12 @@ struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) } } + if (unlikely(ext)) { + pg_data_t *pgdat = page_pgdat(virt_to_page(ext)); + mod_node_page_state(pgdat, NR_INDIRECTLY_RECLAIMABLE_BYTES, + ksize(ext)); + } + this_cpu_inc(nr_dentry); return dentry; @@ -1879,6 +1877,28 @@ void d_instantiate(struct dentry *entry, struct inode * inode) } EXPORT_SYMBOL(d_instantiate); +/* + * This should be equivalent to d_instantiate() + unlock_new_inode(), + * with lockdep-related part of unlock_new_inode() done before + * anything else. Use that instead of open-coding d_instantiate()/ + * unlock_new_inode() combinations. + */ +void d_instantiate_new(struct dentry *entry, struct inode *inode) +{ + BUG_ON(!hlist_unhashed(&entry->d_u.d_alias)); + BUG_ON(!inode); + lockdep_annotate_inode_mutex_key(inode); + security_d_instantiate(entry, inode); + spin_lock(&inode->i_lock); + __d_instantiate(entry, inode); + WARN_ON(!(inode->i_state & I_NEW)); + inode->i_state &= ~I_NEW; + smp_mb(); + wake_up_bit(&inode->i_state, __I_NEW); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL(d_instantiate_new); + /** * d_instantiate_no_diralias - instantiate a non-aliased dentry * @entry: dentry to complete @@ -1921,35 +1941,6 @@ struct dentry *d_make_root(struct inode *root_inode) } EXPORT_SYMBOL(d_make_root); -static struct dentry * __d_find_any_alias(struct inode *inode) -{ - struct dentry *alias; - - if (hlist_empty(&inode->i_dentry)) - return NULL; - alias = hlist_entry(inode->i_dentry.first, struct dentry, d_u.d_alias); - __dget(alias); - return alias; -} - -/** - * d_find_any_alias - find any alias for a given inode - * @inode: inode to find an alias for - * - * If any aliases exist for the given inode, take and return a - * reference for one of them. If no aliases exist, return %NULL. - */ -struct dentry *d_find_any_alias(struct inode *inode) -{ - struct dentry *de; - - spin_lock(&inode->i_lock); - de = __d_find_any_alias(inode); - spin_unlock(&inode->i_lock); - return de; -} -EXPORT_SYMBOL(d_find_any_alias); - static struct dentry *__d_instantiate_anon(struct dentry *dentry, struct inode *inode, bool disconnected) @@ -2770,7 +2761,7 @@ static void copy_name(struct dentry *dentry, struct dentry *target) dentry->d_name.hash_len = target->d_name.hash_len; } if (old_name && likely(atomic_dec_and_test(&old_name->u.count))) - kfree_rcu(old_name, u.head); + call_rcu(&old_name->u.head, __d_free_external_name); } /* @@ -3092,7 +3083,7 @@ static enum d_walk_ret d_genocide_kill(void *data, struct dentry *dentry) void d_genocide(struct dentry *parent) { - d_walk(parent, parent, d_genocide_kill, NULL); + d_walk(parent, parent, d_genocide_kill); } EXPORT_SYMBOL(d_genocide); diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 1f99678ff5d3..4fce1da7db23 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -796,19 +796,13 @@ EXPORT_SYMBOL_GPL(debugfs_read_file_bool); ssize_t debugfs_write_file_bool(struct file *file, const char __user *user_buf, size_t count, loff_t *ppos) { - char buf[32]; - size_t buf_size; bool bv; int r; bool *val = file->private_data; struct dentry *dentry = F_DENTRY(file); - buf_size = min(count, (sizeof(buf)-1)); - if (copy_from_user(buf, user_buf, buf_size)) - return -EFAULT; - - buf[buf_size] = '\0'; - if (strtobool(buf, &bv) == 0) { + r = kstrtobool_from_user(user_buf, count, &bv); + if (!r) { r = debugfs_file_get(dentry); if (unlikely(r)) return r; diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 13b01351dd1c..a913b12fc7f8 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -512,7 +512,9 @@ struct dentry *debugfs_create_dir(const char *name, struct dentry *parent) if (unlikely(!inode)) return failed_creating(dentry); - inode->i_mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO; + if (!parent) + parent = debugfs_mount->mnt_root; + inode->i_mode = S_IFDIR | ((d_inode(parent)->i_mode & 0770)); inode->i_op = &simple_dir_inode_operations; inode->i_fop = &simple_dir_operations; diff --git a/fs/direct-io.c b/fs/direct-io.c index ba12ee659673..093fb54cd316 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -432,8 +432,8 @@ dio_bio_alloc(struct dio *dio, struct dio_submit *sdio, struct bio *bio; /* - * bio_alloc() is guaranteed to return a bio when called with - * __GFP_RECLAIM and we request a valid number of vectors. + * bio_alloc() is guaranteed to return a bio when allowed to sleep and + * we request a valid number of vectors. */ bio = bio_alloc(GFP_KERNEL, nr_vecs); @@ -1177,9 +1177,9 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, unsigned blkbits = i_blkbits; unsigned blocksize_mask = (1 << blkbits) - 1; ssize_t retval = -EINVAL; - size_t count = iov_iter_count(iter); + const size_t count = iov_iter_count(iter); loff_t offset = iocb->ki_pos; - loff_t end = offset + count; + const loff_t end = offset + count; struct dio *dio; struct dio_submit sdio = { 0, }; struct buffer_head map_bh = { 0, }; @@ -1200,7 +1200,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, } /* watch out for a 0 len io from a tricksy fs */ - if (iov_iter_rw(iter) == READ && !iov_iter_count(iter)) + if (iov_iter_rw(iter) == READ && !count) return 0; dio = kmem_cache_alloc(dio_cache, GFP_KERNEL); @@ -1315,8 +1315,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, dio->should_dirty = (iter->type == ITER_IOVEC); sdio.iter = iter; - sdio.final_block_in_request = - (offset + iov_iter_count(iter)) >> blkbits; + sdio.final_block_in_request = end >> blkbits; /* * In case of non-aligned buffers, we may need 2 more diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 5243989a60cc..a5e4a221435c 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -1037,6 +1037,7 @@ static void sctp_connect_to_sock(struct connection *con) int result; int addr_len; struct socket *sock; + struct timeval tv = { .tv_sec = 5, .tv_usec = 0 }; if (con->nodeid == 0) { log_print("attempt to connect sock 0 foiled"); @@ -1080,11 +1081,22 @@ static void sctp_connect_to_sock(struct connection *con) log_print("connecting to %d", con->nodeid); /* Turn off Nagle's algorithm */ - kernel_setsockopt(sock, SOL_TCP, TCP_NODELAY, (char *)&one, + kernel_setsockopt(sock, SOL_SCTP, SCTP_NODELAY, (char *)&one, sizeof(one)); + /* + * Make sock->ops->connect() function return in specified time, + * since O_NONBLOCK argument in connect() function does not work here, + * then, we should restore the default value of this attribute. + */ + kernel_setsockopt(sock, SOL_SOCKET, SO_SNDTIMEO, (char *)&tv, + sizeof(tv)); result = sock->ops->connect(sock, (struct sockaddr *)&daddr, addr_len, - O_NONBLOCK); + 0); + memset(&tv, 0, sizeof(tv)); + kernel_setsockopt(sock, SOL_SOCKET, SO_SNDTIMEO, (char *)&tv, + sizeof(tv)); + if (result == -EINPROGRESS) result = 0; if (result == 0) diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 846ca150d52e..4dd842f72846 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1997,6 +1997,16 @@ out: return rc; } +static bool is_dot_dotdot(const char *name, size_t name_size) +{ + if (name_size == 1 && name[0] == '.') + return true; + else if (name_size == 2 && name[0] == '.' && name[1] == '.') + return true; + + return false; +} + /** * ecryptfs_decode_and_decrypt_filename - converts the encoded cipher text name to decoded plaintext * @plaintext_name: The plaintext name @@ -2021,13 +2031,21 @@ int ecryptfs_decode_and_decrypt_filename(char **plaintext_name, size_t packet_size; int rc = 0; - if ((mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) - && !(mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) - && (name_size > ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE) - && (strncmp(name, ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX, - ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE) == 0)) { - const char *orig_name = name; - size_t orig_name_size = name_size; + if ((mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) && + !(mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED)) { + if (is_dot_dotdot(name, name_size)) { + rc = ecryptfs_copy_filename(plaintext_name, + plaintext_name_size, + name, name_size); + goto out; + } + + if (name_size <= ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE || + strncmp(name, ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX, + ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE)) { + rc = -EINVAL; + goto out; + } name += ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE; name_size -= ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE; @@ -2047,12 +2065,9 @@ int ecryptfs_decode_and_decrypt_filename(char **plaintext_name, decoded_name, decoded_name_size); if (rc) { - printk(KERN_INFO "%s: Could not parse tag 70 packet " - "from filename; copying through filename " - "as-is\n", __func__); - rc = ecryptfs_copy_filename(plaintext_name, - plaintext_name_size, - orig_name, orig_name_size); + ecryptfs_printk(KERN_DEBUG, + "%s: Could not parse tag 70 packet from filename\n", + __func__); goto out_free; } } else { diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index c74ed3ca3372..b76a9853325e 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -82,17 +82,28 @@ ecryptfs_filldir(struct dir_context *ctx, const char *lower_name, buf->sb, lower_name, lower_namelen); if (rc) { - printk(KERN_ERR "%s: Error attempting to decode and decrypt " - "filename [%s]; rc = [%d]\n", __func__, lower_name, - rc); - goto out; + if (rc != -EINVAL) { + ecryptfs_printk(KERN_DEBUG, + "%s: Error attempting to decode and decrypt filename [%s]; rc = [%d]\n", + __func__, lower_name, rc); + return rc; + } + + /* Mask -EINVAL errors as these are most likely due a plaintext + * filename present in the lower filesystem despite filename + * encryption being enabled. One unavoidable example would be + * the "lost+found" dentry in the root directory of an Ext4 + * filesystem. + */ + return 0; } + buf->caller->pos = buf->ctx.pos; rc = !dir_emit(buf->caller, name, name_size, ino, d_type); kfree(name); if (!rc) buf->entries_written++; -out: + return rc; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 847904aa63a9..49121e5a8de2 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -283,8 +283,7 @@ ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry, iget_failed(ecryptfs_inode); goto out; } - unlock_new_inode(ecryptfs_inode); - d_instantiate(ecryptfs_dentry, ecryptfs_inode); + d_instantiate_new(ecryptfs_dentry, ecryptfs_inode); out: return rc; } @@ -395,8 +394,7 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, mount_crypt_stat = &ecryptfs_superblock_to_private( ecryptfs_dentry->d_sb)->mount_crypt_stat; - if (mount_crypt_stat - && (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES)) { + if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) { rc = ecryptfs_encrypt_and_encode_filename( &encrypted_and_encoded_name, &len, mount_crypt_stat, name, len); diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index c89a58cfc991..e74fe84d0886 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -1880,7 +1880,7 @@ find_next_matching_auth_tok: candidate_auth_tok = &auth_tok_list_item->auth_tok; if (unlikely(ecryptfs_verbosity > 0)) { ecryptfs_printk(KERN_DEBUG, - "Considering cadidate auth tok:\n"); + "Considering candidate auth tok:\n"); ecryptfs_dump_auth_tok(candidate_auth_tok); } rc = ecryptfs_get_auth_tok_sig(&candidate_auth_tok_sig, diff --git a/fs/eventfd.c b/fs/eventfd.c index 08d3bd602f73..61c9514da5e9 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -101,14 +101,20 @@ static int eventfd_release(struct inode *inode, struct file *file) return 0; } -static __poll_t eventfd_poll(struct file *file, poll_table *wait) +static struct wait_queue_head * +eventfd_get_poll_head(struct file *file, __poll_t events) +{ + struct eventfd_ctx *ctx = file->private_data; + + return &ctx->wqh; +} + +static __poll_t eventfd_poll_mask(struct file *file, __poll_t eventmask) { struct eventfd_ctx *ctx = file->private_data; __poll_t events = 0; u64 count; - poll_wait(file, &ctx->wqh, wait); - /* * All writes to ctx->count occur within ctx->wqh.lock. This read * can be done outside ctx->wqh.lock because we know that poll_wait @@ -305,7 +311,8 @@ static const struct file_operations eventfd_fops = { .show_fdinfo = eventfd_show_fdinfo, #endif .release = eventfd_release, - .poll = eventfd_poll, + .get_poll_head = eventfd_get_poll_head, + .poll_mask = eventfd_poll_mask, .read = eventfd_read, .write = eventfd_write, .llseek = noop_llseek, diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 602ca4285b2e..67db22fe99c5 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -884,8 +884,7 @@ static __poll_t ep_item_poll(const struct epitem *epi, poll_table *pt, pt->_key = epi->event.events; if (!is_file_epoll(epi->ffd.file)) - return epi->ffd.file->f_op->poll(epi->ffd.file, pt) & - epi->event.events; + return vfs_poll(epi->ffd.file, pt) & epi->event.events; ep = epi->ffd.file->private_data; poll_wait(epi->ffd.file, &ep->poll_wait, pt); @@ -2025,7 +2024,7 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd, /* The target file descriptor must support poll */ error = -EPERM; - if (!tf.file->f_op->poll) + if (!file_can_poll(tf.file)) goto error_tgt_fput; /* Check if EPOLLWAKEUP is allowed */ diff --git a/fs/exec.c b/fs/exec.c index 7eb8d21bcab9..183059c427b9 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -257,7 +257,7 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, * to work from. */ limit = _STK_LIM / 4 * 3; - limit = min(limit, rlimit(RLIMIT_STACK) / 4); + limit = min(limit, bprm->rlim_stack.rlim_cur / 4); if (size > limit) goto fail; } @@ -411,6 +411,11 @@ static int bprm_mm_init(struct linux_binprm *bprm) if (!mm) goto err; + /* Save current stack limit for all calculations made during exec. */ + task_lock(current->group_leader); + bprm->rlim_stack = current->signal->rlim[RLIMIT_STACK]; + task_unlock(current->group_leader); + err = __bprm_mm_init(bprm); if (err) goto err; @@ -697,7 +702,7 @@ int setup_arg_pages(struct linux_binprm *bprm, #ifdef CONFIG_STACK_GROWSUP /* Limit stack size */ - stack_base = rlimit_max(RLIMIT_STACK); + stack_base = bprm->rlim_stack.rlim_max; if (stack_base > STACK_SIZE_MAX) stack_base = STACK_SIZE_MAX; @@ -770,7 +775,7 @@ int setup_arg_pages(struct linux_binprm *bprm, * Align this down to a page boundary as expand_stack * will align it up. */ - rlim_stack = rlimit(RLIMIT_STACK) & PAGE_MASK; + rlim_stack = bprm->rlim_stack.rlim_cur & PAGE_MASK; #ifdef CONFIG_STACK_GROWSUP if (stack_size + stack_expand > rlim_stack) stack_base = vma->vm_start + rlim_stack; @@ -895,13 +900,13 @@ int kernel_read_file(struct file *file, void **buf, loff_t *size, if (!S_ISREG(file_inode(file)->i_mode) || max_size < 0) return -EINVAL; - ret = security_kernel_read_file(file, id); + ret = deny_write_access(file); if (ret) return ret; - ret = deny_write_access(file); + ret = security_kernel_read_file(file, id); if (ret) - return ret; + goto out; i_size = i_size_read(file_inode(file)); if (max_size > 0 && i_size > max_size) { @@ -1341,11 +1346,11 @@ void setup_new_exec(struct linux_binprm * bprm) * RLIMIT_STACK, but after the point of no return to avoid * needing to clean up the change on failure. */ - if (current->signal->rlim[RLIMIT_STACK].rlim_cur > _STK_LIM) - current->signal->rlim[RLIMIT_STACK].rlim_cur = _STK_LIM; + if (bprm->rlim_stack.rlim_cur > _STK_LIM) + bprm->rlim_stack.rlim_cur = _STK_LIM; } - arch_pick_mmap_layout(current->mm); + arch_pick_mmap_layout(current->mm, &bprm->rlim_stack); current->sas_ss_sp = current->sas_ss_size = 0; @@ -1378,6 +1383,16 @@ void setup_new_exec(struct linux_binprm * bprm) } EXPORT_SYMBOL(setup_new_exec); +/* Runs immediately before start_thread() takes over. */ +void finalize_exec(struct linux_binprm *bprm) +{ + /* Store any stack rlimit changes before starting thread. */ + task_lock(current->group_leader); + current->signal->rlim[RLIMIT_STACK] = bprm->rlim_stack; + task_unlock(current->group_leader); +} +EXPORT_SYMBOL(finalize_exec); + /* * Prepare credentials and lock ->cred_guard_mutex. * install_exec_creds() commits the new creds and drops the lock. diff --git a/fs/exofs/ore.c b/fs/exofs/ore.c index 3c6a9c156b7a..ddbf87246898 100644 --- a/fs/exofs/ore.c +++ b/fs/exofs/ore.c @@ -790,7 +790,7 @@ int ore_create(struct ore_io_state *ios) for (i = 0; i < ios->oc->numdevs; i++) { struct osd_request *or; - or = osd_start_request(_ios_od(ios, i), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, i)); if (unlikely(!or)) { ORE_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; @@ -815,7 +815,7 @@ int ore_remove(struct ore_io_state *ios) for (i = 0; i < ios->oc->numdevs; i++) { struct osd_request *or; - or = osd_start_request(_ios_od(ios, i), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, i)); if (unlikely(!or)) { ORE_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; @@ -847,7 +847,7 @@ static int _write_mirror(struct ore_io_state *ios, int cur_comp) struct ore_per_dev_state *per_dev = &ios->per_dev[cur_comp]; struct osd_request *or; - or = osd_start_request(_ios_od(ios, dev), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, dev)); if (unlikely(!or)) { ORE_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; @@ -966,7 +966,7 @@ int _ore_read_mirror(struct ore_io_state *ios, unsigned cur_comp) return 0; /* Just an empty slot */ first_dev = per_dev->dev + first_dev % ios->layout->mirrors_p1; - or = osd_start_request(_ios_od(ios, first_dev), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, first_dev)); if (unlikely(!or)) { ORE_ERR("%s: osd_start_request failed\n", __func__); return -ENOMEM; @@ -1060,7 +1060,7 @@ static int _truncate_mirrors(struct ore_io_state *ios, unsigned cur_comp, struct ore_per_dev_state *per_dev = &ios->per_dev[cur_comp]; struct osd_request *or; - or = osd_start_request(_ios_od(ios, cur_comp), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, cur_comp)); if (unlikely(!or)) { ORE_ERR("%s: osd_start_request failed\n", __func__); return -ENOMEM; diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 179cd5c2f52a..719a3152da80 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -229,7 +229,7 @@ void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], const struct osd_obj_id *obj) static int exofs_read_kern(struct osd_dev *od, u8 *cred, struct osd_obj_id *obj, u64 offset, void *p, unsigned length) { - struct osd_request *or = osd_start_request(od, GFP_KERNEL); + struct osd_request *or = osd_start_request(od); /* struct osd_sense_info osi = {.key = 0};*/ int ret; diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index 329a5d103846..645158dc33f1 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -435,6 +435,15 @@ struct dentry *exportfs_decode_fh(struct vfsmount *mnt, struct fid *fid, if (IS_ERR_OR_NULL(result)) return ERR_PTR(-ESTALE); + /* + * If no acceptance criteria was specified by caller, a disconnected + * dentry is also accepatable. Callers may use this mode to query if + * file handle is stale or to get a reference to an inode without + * risking the high overhead caused by directory reconnect. + */ + if (!acceptable) + return result; + if (d_is_dir(result)) { /* * This request is for a directory. diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 032295e1d386..cc40802ddfa8 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -814,6 +814,7 @@ extern const struct inode_operations ext2_file_inode_operations; extern const struct file_operations ext2_file_operations; /* inode.c */ +extern void ext2_set_file_ops(struct inode *inode); extern const struct address_space_operations ext2_aops; extern const struct address_space_operations ext2_nobh_aops; extern const struct iomap_ops ext2_iomap_ops; diff --git a/fs/ext2/file.c b/fs/ext2/file.c index 09640220fda8..047c327a6b23 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -88,11 +88,11 @@ out_unlock: * The default page_lock and i_size verification done by non-DAX fault paths * is sufficient because ext2 doesn't support hole punching. */ -static int ext2_dax_fault(struct vm_fault *vmf) +static vm_fault_t ext2_dax_fault(struct vm_fault *vmf) { struct inode *inode = file_inode(vmf->vma->vm_file); struct ext2_inode_info *ei = EXT2_I(inode); - int ret; + vm_fault_t ret; if (vmf->flags & FAULT_FLAG_WRITE) { sb_start_pagefault(inode->i_sb); diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 9b2ac55ac34f..71635909df3b 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -940,9 +940,6 @@ ext2_direct_IO(struct kiocb *iocb, struct iov_iter *iter) loff_t offset = iocb->ki_pos; ssize_t ret; - if (WARN_ON_ONCE(IS_DAX(inode))) - return -EIO; - ret = blockdev_direct_IO(iocb, inode, iter, ext2_get_block); if (ret < 0 && iov_iter_rw(iter) == WRITE) ext2_write_failed(mapping, offset + count); @@ -952,17 +949,16 @@ ext2_direct_IO(struct kiocb *iocb, struct iov_iter *iter) static int ext2_writepages(struct address_space *mapping, struct writeback_control *wbc) { -#ifdef CONFIG_FS_DAX - if (dax_mapping(mapping)) { - return dax_writeback_mapping_range(mapping, - mapping->host->i_sb->s_bdev, - wbc); - } -#endif - return mpage_writepages(mapping, wbc, ext2_get_block); } +static int +ext2_dax_writepages(struct address_space *mapping, struct writeback_control *wbc) +{ + return dax_writeback_mapping_range(mapping, + mapping->host->i_sb->s_bdev, wbc); +} + const struct address_space_operations ext2_aops = { .readpage = ext2_readpage, .readpages = ext2_readpages, @@ -990,6 +986,13 @@ const struct address_space_operations ext2_nobh_aops = { .error_remove_page = generic_error_remove_page, }; +static const struct address_space_operations ext2_dax_aops = { + .writepages = ext2_dax_writepages, + .direct_IO = noop_direct_IO, + .set_page_dirty = noop_set_page_dirty, + .invalidatepage = noop_invalidatepage, +}; + /* * Probably it should be a library function... search for first non-zero word * or memcmp with zero_page, whatever is better for particular architecture. @@ -1261,21 +1264,11 @@ do_indirects: static void ext2_truncate_blocks(struct inode *inode, loff_t offset) { - /* - * XXX: it seems like a bug here that we don't allow - * IS_APPEND inode to have blocks-past-i_size trimmed off. - * review and fix this. - * - * Also would be nice to be able to handle IO errors and such, - * but that's probably too much to ask. - */ if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) return; if (ext2_inode_is_fast_symlink(inode)) return; - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) - return; dax_sem_down_write(EXT2_I(inode)); __ext2_truncate_blocks(inode, offset); @@ -1388,6 +1381,18 @@ void ext2_set_inode_flags(struct inode *inode) inode->i_flags |= S_DAX; } +void ext2_set_file_ops(struct inode *inode) +{ + inode->i_op = &ext2_file_inode_operations; + inode->i_fop = &ext2_file_operations; + if (IS_DAX(inode)) + inode->i_mapping->a_ops = &ext2_dax_aops; + else if (test_opt(inode->i_sb, NOBH)) + inode->i_mapping->a_ops = &ext2_nobh_aops; + else + inode->i_mapping->a_ops = &ext2_aops; +} + struct inode *ext2_iget (struct super_block *sb, unsigned long ino) { struct ext2_inode_info *ei; @@ -1480,14 +1485,7 @@ struct inode *ext2_iget (struct super_block *sb, unsigned long ino) ei->i_data[n] = raw_inode->i_block[n]; if (S_ISREG(inode->i_mode)) { - inode->i_op = &ext2_file_inode_operations; - if (test_opt(inode->i_sb, NOBH)) { - inode->i_mapping->a_ops = &ext2_nobh_aops; - inode->i_fop = &ext2_file_operations; - } else { - inode->i_mapping->a_ops = &ext2_aops; - inode->i_fop = &ext2_file_operations; - } + ext2_set_file_ops(inode); } else if (S_ISDIR(inode->i_mode)) { inode->i_op = &ext2_dir_inode_operations; inode->i_fop = &ext2_dir_operations; diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index e078075dc66f..152453a91877 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -41,8 +41,7 @@ static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode) { int err = ext2_add_link(dentry, inode); if (!err) { - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -107,14 +106,7 @@ static int ext2_create (struct inode * dir, struct dentry * dentry, umode_t mode if (IS_ERR(inode)) return PTR_ERR(inode); - inode->i_op = &ext2_file_inode_operations; - if (test_opt(inode->i_sb, NOBH)) { - inode->i_mapping->a_ops = &ext2_nobh_aops; - inode->i_fop = &ext2_file_operations; - } else { - inode->i_mapping->a_ops = &ext2_aops; - inode->i_fop = &ext2_file_operations; - } + ext2_set_file_ops(inode); mark_inode_dirty(inode); return ext2_add_nondir(dentry, inode); } @@ -125,14 +117,7 @@ static int ext2_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) if (IS_ERR(inode)) return PTR_ERR(inode); - inode->i_op = &ext2_file_inode_operations; - if (test_opt(inode->i_sb, NOBH)) { - inode->i_mapping->a_ops = &ext2_nobh_aops; - inode->i_fop = &ext2_file_operations; - } else { - inode->i_mapping->a_ops = &ext2_aops; - inode->i_fop = &ext2_file_operations; - } + ext2_set_file_ops(inode); mark_inode_dirty(inode); d_tmpfile(dentry, inode); unlock_new_inode(inode); @@ -269,8 +254,7 @@ static int ext2_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) if (err) goto out_fail; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out: return err; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index de1694512f1f..c09289a42dc5 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -961,8 +961,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) blocksize = BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size); if (sbi->s_mount_opt & EXT2_MOUNT_DAX) { - err = bdev_dax_supported(sb, blocksize); - if (err) { + if (!bdev_dax_supported(sb->s_bdev, blocksize)) { ext2_msg(sb, KERN_ERR, "DAX unsupported by block device. Turning off DAX."); sbi->s_mount_opt &= ~EXT2_MOUNT_DAX; diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index a33d8fb1bf2a..b00481c475cb 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -185,25 +185,15 @@ static int ext4_init_block_bitmap(struct super_block *sb, struct ext4_sb_info *sbi = EXT4_SB(sb); ext4_fsblk_t start, tmp; int flex_bg = 0; - struct ext4_group_info *grp; J_ASSERT_BH(bh, buffer_locked(bh)); /* If checksum is bad mark all blocks used to prevent allocation * essentially implementing a per-group read-only flag. */ if (!ext4_group_desc_csum_verify(sb, block_group, gdp)) { - grp = ext4_get_group_info(sb, block_group); - if (!EXT4_MB_GRP_BBITMAP_CORRUPT(grp)) - percpu_counter_sub(&sbi->s_freeclusters_counter, - grp->bb_free); - set_bit(EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT, &grp->bb_state); - if (!EXT4_MB_GRP_IBITMAP_CORRUPT(grp)) { - int count; - count = ext4_free_inodes_count(sb, gdp); - percpu_counter_sub(&sbi->s_freeinodes_counter, - count); - } - set_bit(EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT, &grp->bb_state); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT | + EXT4_GROUP_INFO_IBITMAP_CORRUPT); return -EFSBADCRC; } memset(bh->b_data, 0, sb->s_blocksize); @@ -321,6 +311,7 @@ static ext4_fsblk_t ext4_valid_block_bitmap(struct super_block *sb, struct ext4_sb_info *sbi = EXT4_SB(sb); ext4_grpblk_t offset; ext4_grpblk_t next_zero_bit; + ext4_grpblk_t max_bit = EXT4_CLUSTERS_PER_GROUP(sb); ext4_fsblk_t blk; ext4_fsblk_t group_first_block; @@ -338,7 +329,7 @@ static ext4_fsblk_t ext4_valid_block_bitmap(struct super_block *sb, /* check whether block bitmap block number is set */ blk = ext4_block_bitmap(sb, desc); offset = blk - group_first_block; - if (offset < 0 || EXT4_B2C(sbi, offset) >= sb->s_blocksize || + if (offset < 0 || EXT4_B2C(sbi, offset) >= max_bit || !ext4_test_bit(EXT4_B2C(sbi, offset), bh->b_data)) /* bad block bitmap */ return blk; @@ -346,7 +337,7 @@ static ext4_fsblk_t ext4_valid_block_bitmap(struct super_block *sb, /* check whether the inode bitmap block number is set */ blk = ext4_inode_bitmap(sb, desc); offset = blk - group_first_block; - if (offset < 0 || EXT4_B2C(sbi, offset) >= sb->s_blocksize || + if (offset < 0 || EXT4_B2C(sbi, offset) >= max_bit || !ext4_test_bit(EXT4_B2C(sbi, offset), bh->b_data)) /* bad block bitmap */ return blk; @@ -354,8 +345,8 @@ static ext4_fsblk_t ext4_valid_block_bitmap(struct super_block *sb, /* check whether the inode table block number is set */ blk = ext4_inode_table(sb, desc); offset = blk - group_first_block; - if (offset < 0 || EXT4_B2C(sbi, offset) >= sb->s_blocksize || - EXT4_B2C(sbi, offset + sbi->s_itb_per_group) >= sb->s_blocksize) + if (offset < 0 || EXT4_B2C(sbi, offset) >= max_bit || + EXT4_B2C(sbi, offset + sbi->s_itb_per_group) >= max_bit) return blk; next_zero_bit = ext4_find_next_zero_bit(bh->b_data, EXT4_B2C(sbi, offset + sbi->s_itb_per_group), @@ -374,7 +365,6 @@ static int ext4_validate_block_bitmap(struct super_block *sb, { ext4_fsblk_t blk; struct ext4_group_info *grp = ext4_get_group_info(sb, block_group); - struct ext4_sb_info *sbi = EXT4_SB(sb); if (buffer_verified(bh)) return 0; @@ -386,10 +376,8 @@ static int ext4_validate_block_bitmap(struct super_block *sb, desc, bh))) { ext4_unlock_group(sb, block_group); ext4_error(sb, "bg %u: bad block bitmap checksum", block_group); - if (!EXT4_MB_GRP_BBITMAP_CORRUPT(grp)) - percpu_counter_sub(&sbi->s_freeclusters_counter, - grp->bb_free); - set_bit(EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT, &grp->bb_state); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); return -EFSBADCRC; } blk = ext4_valid_block_bitmap(sb, desc, block_group, bh); @@ -397,10 +385,8 @@ static int ext4_validate_block_bitmap(struct super_block *sb, ext4_unlock_group(sb, block_group); ext4_error(sb, "bg %u: block %llu: invalid block bitmap", block_group, blk); - if (!EXT4_MB_GRP_BBITMAP_CORRUPT(grp)) - percpu_counter_sub(&sbi->s_freeclusters_counter, - grp->bb_free); - set_bit(EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT, &grp->bb_state); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); return -EFSCORRUPTED; } set_buffer_verified(bh); @@ -435,6 +421,8 @@ ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group) (bitmap_blk >= ext4_blocks_count(sbi->s_es))) { ext4_error(sb, "Invalid block bitmap block %llu in " "block_group %u", bitmap_blk, block_group); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); return ERR_PTR(-EFSCORRUPTED); } bh = sb_getblk(sb, bitmap_blk); @@ -513,6 +501,8 @@ int ext4_wait_block_bitmap(struct super_block *sb, ext4_group_t block_group, ext4_error(sb, "Cannot read block bitmap - " "block_group = %u, block_bitmap = %llu", block_group, (unsigned long long) bh->b_blocknr); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); return -EIO; } clear_buffer_new(bh); diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index a42e71203e53..df95412915ea 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -2390,7 +2390,7 @@ extern int ext4_init_inode_table(struct super_block *sb, extern void ext4_end_bitmap_read(struct buffer_head *bh, int uptodate); /* mballoc.c */ -extern const struct file_operations ext4_seq_mb_groups_fops; +extern const struct seq_operations ext4_mb_seq_groups_ops; extern long ext4_mb_stats; extern long ext4_mb_max_to_scan; extern int ext4_mb_init(struct super_block *); @@ -2530,6 +2530,9 @@ extern int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup); extern const char *ext4_decode_error(struct super_block *sb, int errno, char nbuf[16]); +extern void ext4_mark_group_bitmap_corrupted(struct super_block *sb, + ext4_group_t block_group, + unsigned int flags); extern __printf(4, 5) void __ext4_error(struct super_block *, const char *, unsigned int, @@ -2857,6 +2860,10 @@ struct ext4_group_info { #define EXT4_GROUP_INFO_WAS_TRIMMED_BIT 1 #define EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT 2 #define EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT 3 +#define EXT4_GROUP_INFO_BBITMAP_CORRUPT \ + (1 << EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT) +#define EXT4_GROUP_INFO_IBITMAP_CORRUPT \ + (1 << EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT) #define EXT4_MB_GRP_NEED_INIT(grp) \ (test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state))) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 0a7315961bac..c969275ce3ee 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -5329,8 +5329,9 @@ ext4_ext_shift_extents(struct inode *inode, handle_t *handle, stop = le32_to_cpu(extent->ee_block); /* - * In case of left shift, Don't start shifting extents until we make - * sure the hole is big enough to accommodate the shift. + * For left shifts, make sure the hole on the left is big enough to + * accommodate the shift. For right shifts, make sure the last extent + * won't be shifted beyond EXT_MAX_BLOCKS. */ if (SHIFT == SHIFT_LEFT) { path = ext4_find_extent(inode, start - 1, &path, @@ -5350,9 +5351,14 @@ ext4_ext_shift_extents(struct inode *inode, handle_t *handle, if ((start == ex_start && shift > ex_start) || (shift > start - ex_end)) { - ext4_ext_drop_refs(path); - kfree(path); - return -EINVAL; + ret = -EINVAL; + goto out; + } + } else { + if (shift > EXT_MAX_BLOCKS - + (stop + ext4_ext_get_actual_len(extent))) { + ret = -EINVAL; + goto out; } } diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c index 763ef185dd17..c4e6fb15101b 100644 --- a/fs/ext4/extents_status.c +++ b/fs/ext4/extents_status.c @@ -162,8 +162,7 @@ int __init ext4_init_es(void) void ext4_exit_es(void) { - if (ext4_es_cachep) - kmem_cache_destroy(ext4_es_cachep); + kmem_cache_destroy(ext4_es_cachep); } void ext4_es_init_tree(struct ext4_es_tree *tree) diff --git a/fs/ext4/file.c b/fs/ext4/file.c index fb6f023622fe..7f8023340eb8 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -277,10 +277,11 @@ out: } #ifdef CONFIG_FS_DAX -static int ext4_dax_huge_fault(struct vm_fault *vmf, +static vm_fault_t ext4_dax_huge_fault(struct vm_fault *vmf, enum page_entry_size pe_size) { - int result, error = 0; + int error = 0; + vm_fault_t result; int retries = 0; handle_t *handle = NULL; struct inode *inode = file_inode(vmf->vma->vm_file); @@ -335,7 +336,7 @@ retry: return result; } -static int ext4_dax_fault(struct vm_fault *vmf) +static vm_fault_t ext4_dax_fault(struct vm_fault *vmf) { return ext4_dax_huge_fault(vmf, PE_SIZE_PTE); } @@ -380,50 +381,64 @@ static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) return 0; } -static int ext4_file_open(struct inode * inode, struct file * filp) +static int ext4_sample_last_mounted(struct super_block *sb, + struct vfsmount *mnt) { - struct super_block *sb = inode->i_sb; - struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - struct vfsmount *mnt = filp->f_path.mnt; + struct ext4_sb_info *sbi = EXT4_SB(sb); struct path path; char buf[64], *cp; + handle_t *handle; + int err; + + if (likely(sbi->s_mount_flags & EXT4_MF_MNTDIR_SAMPLED)) + return 0; + + if (sb_rdonly(sb) || !sb_start_intwrite_trylock(sb)) + return 0; + + sbi->s_mount_flags |= EXT4_MF_MNTDIR_SAMPLED; + /* + * Sample where the filesystem has been mounted and + * store it in the superblock for sysadmin convenience + * when trying to sort through large numbers of block + * devices or filesystem images. + */ + memset(buf, 0, sizeof(buf)); + path.mnt = mnt; + path.dentry = mnt->mnt_root; + cp = d_path(&path, buf, sizeof(buf)); + err = 0; + if (IS_ERR(cp)) + goto out; + + handle = ext4_journal_start_sb(sb, EXT4_HT_MISC, 1); + err = PTR_ERR(handle); + if (IS_ERR(handle)) + goto out; + BUFFER_TRACE(sbi->s_sbh, "get_write_access"); + err = ext4_journal_get_write_access(handle, sbi->s_sbh); + if (err) + goto out_journal; + strlcpy(sbi->s_es->s_last_mounted, cp, + sizeof(sbi->s_es->s_last_mounted)); + ext4_handle_dirty_super(handle, sb); +out_journal: + ext4_journal_stop(handle); +out: + sb_end_intwrite(sb); + return err; +} + +static int ext4_file_open(struct inode * inode, struct file * filp) +{ int ret; if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; - if (unlikely(!(sbi->s_mount_flags & EXT4_MF_MNTDIR_SAMPLED) && - !sb_rdonly(sb))) { - sbi->s_mount_flags |= EXT4_MF_MNTDIR_SAMPLED; - /* - * Sample where the filesystem has been mounted and - * store it in the superblock for sysadmin convenience - * when trying to sort through large numbers of block - * devices or filesystem images. - */ - memset(buf, 0, sizeof(buf)); - path.mnt = mnt; - path.dentry = mnt->mnt_root; - cp = d_path(&path, buf, sizeof(buf)); - if (!IS_ERR(cp)) { - handle_t *handle; - int err; - - handle = ext4_journal_start_sb(sb, EXT4_HT_MISC, 1); - if (IS_ERR(handle)) - return PTR_ERR(handle); - BUFFER_TRACE(sbi->s_sbh, "get_write_access"); - err = ext4_journal_get_write_access(handle, sbi->s_sbh); - if (err) { - ext4_journal_stop(handle); - return err; - } - strlcpy(sbi->s_es->s_last_mounted, cp, - sizeof(sbi->s_es->s_last_mounted)); - ext4_handle_dirty_super(handle, sb); - ext4_journal_stop(handle); - } - } + ret = ext4_sample_last_mounted(inode->i_sb, filp->f_path.mnt); + if (ret) + return ret; ret = fscrypt_file_open(inode, filp); if (ret) diff --git a/fs/ext4/fsmap.c b/fs/ext4/fsmap.c index e871c4bf18e9..4b99e2db95b8 100644 --- a/fs/ext4/fsmap.c +++ b/fs/ext4/fsmap.c @@ -402,8 +402,8 @@ static void ext4_getfsmap_free_fixed_metadata(struct list_head *meta_list) } /* Find all the fixed metadata in the filesystem. */ -int ext4_getfsmap_find_fixed_metadata(struct super_block *sb, - struct list_head *meta_list) +static int ext4_getfsmap_find_fixed_metadata(struct super_block *sb, + struct list_head *meta_list) { struct ext4_group_desc *gdp; ext4_group_t agno; diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index df92e3ec9913..4d6e007f3569 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -83,7 +83,6 @@ static int ext4_validate_inode_bitmap(struct super_block *sb, { ext4_fsblk_t blk; struct ext4_group_info *grp = ext4_get_group_info(sb, block_group); - struct ext4_sb_info *sbi = EXT4_SB(sb); if (buffer_verified(bh)) return 0; @@ -97,14 +96,8 @@ static int ext4_validate_inode_bitmap(struct super_block *sb, ext4_unlock_group(sb, block_group); ext4_error(sb, "Corrupt inode bitmap - block_group = %u, " "inode_bitmap = %llu", block_group, blk); - grp = ext4_get_group_info(sb, block_group); - if (!EXT4_MB_GRP_IBITMAP_CORRUPT(grp)) { - int count; - count = ext4_free_inodes_count(sb, desc); - percpu_counter_sub(&sbi->s_freeinodes_counter, - count); - } - set_bit(EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT, &grp->bb_state); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_IBITMAP_CORRUPT); return -EFSBADCRC; } set_buffer_verified(bh); @@ -136,6 +129,8 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) (bitmap_blk >= ext4_blocks_count(sbi->s_es))) { ext4_error(sb, "Invalid inode bitmap blk %llu in " "block_group %u", bitmap_blk, block_group); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_IBITMAP_CORRUPT); return ERR_PTR(-EFSCORRUPTED); } bh = sb_getblk(sb, bitmap_blk); @@ -143,7 +138,7 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) ext4_error(sb, "Cannot read inode bitmap - " "block_group = %u, inode_bitmap = %llu", block_group, bitmap_blk); - return ERR_PTR(-EIO); + return ERR_PTR(-ENOMEM); } if (bitmap_uptodate(bh)) goto verify; @@ -190,6 +185,8 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) ext4_error(sb, "Cannot read inode bitmap - " "block_group = %u, inode_bitmap = %llu", block_group, bitmap_blk); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_IBITMAP_CORRUPT); return ERR_PTR(-EIO); } @@ -337,13 +334,8 @@ out: fatal = err; } else { ext4_error(sb, "bit already cleared for inode %lu", ino); - if (gdp && !EXT4_MB_GRP_IBITMAP_CORRUPT(grp)) { - int count; - count = ext4_free_inodes_count(sb, gdp); - percpu_counter_sub(&sbi->s_freeinodes_counter, - count); - } - set_bit(EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT, &grp->bb_state); + ext4_mark_group_bitmap_corrupted(sb, block_group, + EXT4_GROUP_INFO_IBITMAP_CORRUPT); } error_return: @@ -914,6 +906,8 @@ repeat_in_this_group: if (group == 0 && (ino + 1) < EXT4_FIRST_INO(sb)) { ext4_error(sb, "reserved inode found cleared - " "inode=%lu", ino + 1); + ext4_mark_group_bitmap_corrupted(sb, group, + EXT4_GROUP_INFO_IBITMAP_CORRUPT); goto next_group; } @@ -1105,6 +1099,8 @@ got: err = -EIO; ext4_error(sb, "failed to insert inode %lu: doubly allocated?", inode->i_ino); + ext4_mark_group_bitmap_corrupted(sb, group, + EXT4_GROUP_INFO_IBITMAP_CORRUPT); goto out; } inode->i_generation = prandom_u32(); @@ -1206,11 +1202,8 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); bitmap_bh = ext4_read_inode_bitmap(sb, block_group); - if (IS_ERR(bitmap_bh)) { - ext4_error(sb, "inode bitmap error %ld for orphan %lu", - ino, PTR_ERR(bitmap_bh)); + if (IS_ERR(bitmap_bh)) return (struct inode *) bitmap_bh; - } /* Having the inode bit set should be a 100% indicator that this * is a valid orphan (no e2fsck run on fs). Orphans also include diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c index c32802c956d5..bf7fa1507e81 100644 --- a/fs/ext4/indirect.c +++ b/fs/ext4/indirect.c @@ -561,10 +561,16 @@ int ext4_ind_map_blocks(handle_t *handle, struct inode *inode, unsigned epb = inode->i_sb->s_blocksize / sizeof(u32); int i; - /* Count number blocks in a subtree under 'partial' */ - count = 1; - for (i = 0; partial + i != chain + depth - 1; i++) - count *= epb; + /* + * Count number blocks in a subtree under 'partial'. At each + * level we count number of complete empty subtrees beyond + * current offset and then descend into the subtree only + * partially beyond current offset. + */ + count = 0; + for (i = partial - chain + 1; i < depth; i++) + count = count * epb + (epb - offsets[i] - 1); + count++; /* Fill in size of a hole we found */ map->m_pblk = 0; map->m_len = min_t(unsigned int, map->m_len, count); diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 70cf4c7b268a..285ed1588730 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -144,6 +144,12 @@ int ext4_find_inline_data_nolock(struct inode *inode) goto out; if (!is.s.not_found) { + if (is.s.here->e_value_inum) { + EXT4_ERROR_INODE(inode, "inline data xattr refers " + "to an external xattr inode"); + error = -EFSCORRUPTED; + goto out; + } EXT4_I(inode)->i_inline_off = (u16)((void *)is.s.here - (void *)ext4_raw_inode(&is.iloc)); EXT4_I(inode)->i_inline_size = EXT4_MIN_INLINE_DATA_SIZE + @@ -1835,8 +1841,8 @@ int ext4_inline_data_iomap(struct inode *inode, struct iomap *iomap) iomap->offset = 0; iomap->length = min_t(loff_t, ext4_get_inline_size(inode), i_size_read(inode)); - iomap->type = 0; - iomap->flags = IOMAP_F_DATA_INLINE; + iomap->type = IOMAP_INLINE; + iomap->flags = 0; out: up_read(&EXT4_I(inode)->xattr_sem); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 18aa2ef963ad..2ea07efbe016 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2716,12 +2716,6 @@ static int ext4_writepages(struct address_space *mapping, percpu_down_read(&sbi->s_journal_flag_rwsem); trace_ext4_writepages(inode, wbc); - if (dax_mapping(mapping)) { - ret = dax_writeback_mapping_range(mapping, inode->i_sb->s_bdev, - wbc); - goto out_writepages; - } - /* * No pages to write? This is mainly a kludge to avoid starting * a transaction for special inodes like journal inode on last iput() @@ -2942,6 +2936,27 @@ out_writepages: return ret; } +static int ext4_dax_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + int ret; + long nr_to_write = wbc->nr_to_write; + struct inode *inode = mapping->host; + struct ext4_sb_info *sbi = EXT4_SB(mapping->host->i_sb); + + if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) + return -EIO; + + percpu_down_read(&sbi->s_journal_flag_rwsem); + trace_ext4_writepages(inode, wbc); + + ret = dax_writeback_mapping_range(mapping, inode->i_sb->s_bdev, wbc); + trace_ext4_writepages_result(inode, wbc, ret, + nr_to_write - wbc->nr_to_write); + percpu_up_read(&sbi->s_journal_flag_rwsem); + return ret; +} + static int ext4_nonda_switch(struct super_block *sb) { s64 free_clusters, dirty_clusters; @@ -3845,10 +3860,6 @@ static ssize_t ext4_direct_IO(struct kiocb *iocb, struct iov_iter *iter) if (ext4_has_inline_data(inode)) return 0; - /* DAX uses iomap path now */ - if (WARN_ON_ONCE(IS_DAX(inode))) - return 0; - trace_ext4_direct_IO_enter(inode, offset, count, iov_iter_rw(iter)); if (iov_iter_rw(iter) == READ) ret = ext4_direct_IO_read(iocb, iter); @@ -3934,6 +3945,13 @@ static const struct address_space_operations ext4_da_aops = { .error_remove_page = generic_error_remove_page, }; +static const struct address_space_operations ext4_dax_aops = { + .writepages = ext4_dax_writepages, + .direct_IO = noop_direct_IO, + .set_page_dirty = noop_set_page_dirty, + .invalidatepage = noop_invalidatepage, +}; + void ext4_set_aops(struct inode *inode) { switch (ext4_inode_journal_mode(inode)) { @@ -3946,7 +3964,9 @@ void ext4_set_aops(struct inode *inode) default: BUG(); } - if (test_opt(inode->i_sb, DELALLOC)) + if (IS_DAX(inode)) + inode->i_mapping->a_ops = &ext4_dax_aops; + else if (test_opt(inode->i_sb, DELALLOC)) inode->i_mapping->a_ops = &ext4_da_aops; else inode->i_mapping->a_ops = &ext4_aops; @@ -4278,28 +4298,28 @@ int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length) EXT4_BLOCK_SIZE_BITS(sb); stop_block = (offset + length) >> EXT4_BLOCK_SIZE_BITS(sb); - /* If there are no blocks to remove, return now */ - if (first_block >= stop_block) - goto out_stop; + /* If there are blocks to remove, do it */ + if (stop_block > first_block) { - down_write(&EXT4_I(inode)->i_data_sem); - ext4_discard_preallocations(inode); + down_write(&EXT4_I(inode)->i_data_sem); + ext4_discard_preallocations(inode); - ret = ext4_es_remove_extent(inode, first_block, - stop_block - first_block); - if (ret) { - up_write(&EXT4_I(inode)->i_data_sem); - goto out_stop; - } + ret = ext4_es_remove_extent(inode, first_block, + stop_block - first_block); + if (ret) { + up_write(&EXT4_I(inode)->i_data_sem); + goto out_stop; + } - if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) - ret = ext4_ext_remove_space(inode, first_block, - stop_block - 1); - else - ret = ext4_ind_remove_space(handle, inode, first_block, - stop_block); + if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) + ret = ext4_ext_remove_space(inode, first_block, + stop_block - 1); + else + ret = ext4_ind_remove_space(handle, inode, first_block, + stop_block); - up_write(&EXT4_I(inode)->i_data_sem); + up_write(&EXT4_I(inode)->i_data_sem); + } if (IS_SYNC(inode)) ext4_handle_sync(handle); @@ -4681,19 +4701,21 @@ static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode, } } -static inline void ext4_iget_extra_inode(struct inode *inode, +static inline int ext4_iget_extra_inode(struct inode *inode, struct ext4_inode *raw_inode, struct ext4_inode_info *ei) { __le32 *magic = (void *)raw_inode + EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize; + if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize + sizeof(__le32) <= EXT4_INODE_SIZE(inode->i_sb) && *magic == cpu_to_le32(EXT4_XATTR_MAGIC)) { ext4_set_inode_state(inode, EXT4_STATE_XATTR); - ext4_find_inline_data_nolock(inode); + return ext4_find_inline_data_nolock(inode); } else EXT4_I(inode)->i_inline_off = 0; + return 0; } int ext4_get_projid(struct inode *inode, kprojid_t *projid) @@ -4704,6 +4726,26 @@ int ext4_get_projid(struct inode *inode, kprojid_t *projid) return 0; } +/* + * ext4 has self-managed i_version for ea inodes, it stores the lower 32bit of + * refcount in i_version, so use raw values if inode has EXT4_EA_INODE_FL flag + * set. + */ +static inline void ext4_inode_set_iversion_queried(struct inode *inode, u64 val) +{ + if (unlikely(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL)) + inode_set_iversion_raw(inode, val); + else + inode_set_iversion_queried(inode, val); +} +static inline u64 ext4_inode_peek_iversion(const struct inode *inode) +{ + if (unlikely(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL)) + return inode_peek_iversion_raw(inode); + else + return inode_peek_iversion(inode); +} + struct inode *ext4_iget(struct super_block *sb, unsigned long ino) { struct ext4_iloc iloc; @@ -4873,7 +4915,9 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ei->i_extra_isize = sizeof(struct ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE; } else { - ext4_iget_extra_inode(inode, raw_inode, ei); + ret = ext4_iget_extra_inode(inode, raw_inode, ei); + if (ret) + goto bad_inode; } } @@ -4890,7 +4934,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ivers |= (__u64)(le32_to_cpu(raw_inode->i_version_hi)) << 32; } - inode_set_iversion_queried(inode, ivers); + ext4_inode_set_iversion_queried(inode, ivers); } ret = 0; @@ -4925,6 +4969,13 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) inode->i_op = &ext4_dir_inode_operations; inode->i_fop = &ext4_dir_operations; } else if (S_ISLNK(inode->i_mode)) { + /* VFS does not allow setting these so must be corruption */ + if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) { + EXT4_ERROR_INODE(inode, + "immutable or append flags not allowed on symlinks"); + ret = -EFSCORRUPTED; + goto bad_inode; + } if (ext4_encrypted_inode(inode)) { inode->i_op = &ext4_encrypted_symlink_inode_operations; ext4_set_aops(inode); @@ -5024,12 +5075,12 @@ static int other_inode_match(struct inode * inode, unsigned long ino, if ((inode->i_ino != ino) || (inode->i_state & (I_FREEING | I_WILL_FREE | I_NEW | - I_DIRTY_SYNC | I_DIRTY_DATASYNC)) || + I_DIRTY_INODE)) || ((inode->i_state & I_DIRTY_TIME) == 0)) return 0; spin_lock(&inode->i_lock); if (((inode->i_state & (I_FREEING | I_WILL_FREE | I_NEW | - I_DIRTY_SYNC | I_DIRTY_DATASYNC)) == 0) && + I_DIRTY_INODE)) == 0) && (inode->i_state & I_DIRTY_TIME)) { struct ext4_inode_info *ei = EXT4_I(inode); @@ -5176,7 +5227,7 @@ static int ext4_do_update_inode(handle_t *handle, } if (likely(!test_opt2(inode->i_sb, HURD_COMPAT))) { - u64 ivers = inode_peek_iversion(inode); + u64 ivers = ext4_inode_peek_iversion(inode); raw_inode->i_disk_version = cpu_to_le32(ivers); if (ei->i_extra_isize) { diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 769a62708b1c..6eae2b91aafa 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -470,6 +470,8 @@ static void mb_free_blocks_double(struct inode *inode, struct ext4_buddy *e4b, "freeing block already freed " "(bit %u)", first + i); + ext4_mark_group_bitmap_corrupted(sb, e4b->bd_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); } mb_clear_bit(first + i, e4b->bd_info->bb_bitmap); } @@ -747,10 +749,8 @@ void ext4_mb_generate_buddy(struct super_block *sb, * corrupt and update bb_free using bitmap value */ grp->bb_free = free; - if (!EXT4_MB_GRP_BBITMAP_CORRUPT(grp)) - percpu_counter_sub(&sbi->s_freeclusters_counter, - grp->bb_free); - set_bit(EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT, &grp->bb_state); + ext4_mark_group_bitmap_corrupted(sb, group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); } mb_set_largest_free_order(sb, grp); @@ -1454,12 +1454,8 @@ static void mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, "freeing already freed block " "(bit %u); block bitmap corrupt.", block); - if (!EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info)) - percpu_counter_sub(&sbi->s_freeclusters_counter, - e4b->bd_info->bb_free); - /* Mark the block group as corrupt. */ - set_bit(EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT, - &e4b->bd_info->bb_state); + ext4_mark_group_bitmap_corrupted(sb, e4b->bd_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); mb_regenerate_buddy(e4b); goto done; } @@ -1956,6 +1952,8 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac, "%d free clusters as per " "group info. But bitmap says 0", free); + ext4_mark_group_bitmap_corrupted(sb, e4b->bd_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); break; } @@ -1966,6 +1964,8 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac, "%d free clusters as per " "group info. But got %d blocks", free, ex.fe_len); + ext4_mark_group_bitmap_corrupted(sb, e4b->bd_group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); /* * The number of free blocks differs. This mostly * indicate that the bitmap is corrupt. So exit @@ -2254,7 +2254,7 @@ out: static void *ext4_mb_seq_groups_start(struct seq_file *seq, loff_t *pos) { - struct super_block *sb = seq->private; + struct super_block *sb = PDE_DATA(file_inode(seq->file)); ext4_group_t group; if (*pos < 0 || *pos >= ext4_get_groups_count(sb)) @@ -2265,7 +2265,7 @@ static void *ext4_mb_seq_groups_start(struct seq_file *seq, loff_t *pos) static void *ext4_mb_seq_groups_next(struct seq_file *seq, void *v, loff_t *pos) { - struct super_block *sb = seq->private; + struct super_block *sb = PDE_DATA(file_inode(seq->file)); ext4_group_t group; ++*pos; @@ -2277,7 +2277,7 @@ static void *ext4_mb_seq_groups_next(struct seq_file *seq, void *v, loff_t *pos) static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) { - struct super_block *sb = seq->private; + struct super_block *sb = PDE_DATA(file_inode(seq->file)); ext4_group_t group = (ext4_group_t) ((unsigned long) v); int i; int err, buddy_loaded = 0; @@ -2330,34 +2330,13 @@ static void ext4_mb_seq_groups_stop(struct seq_file *seq, void *v) { } -static const struct seq_operations ext4_mb_seq_groups_ops = { +const struct seq_operations ext4_mb_seq_groups_ops = { .start = ext4_mb_seq_groups_start, .next = ext4_mb_seq_groups_next, .stop = ext4_mb_seq_groups_stop, .show = ext4_mb_seq_groups_show, }; -static int ext4_mb_seq_groups_open(struct inode *inode, struct file *file) -{ - struct super_block *sb = PDE_DATA(inode); - int rc; - - rc = seq_open(file, &ext4_mb_seq_groups_ops); - if (rc == 0) { - struct seq_file *m = file->private_data; - m->private = sb; - } - return rc; - -} - -const struct file_operations ext4_seq_mb_groups_fops = { - .open = ext4_mb_seq_groups_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - static struct kmem_cache *get_groupinfo_cache(int blocksize_bits) { int cache_index = blocksize_bits - EXT4_MIN_BLOCK_LOG_SIZE; @@ -2537,8 +2516,7 @@ static void ext4_groupinfo_destroy_slabs(void) int i; for (i = 0; i < NR_GRPINFO_CACHES; i++) { - if (ext4_groupinfo_caches[i]) - kmem_cache_destroy(ext4_groupinfo_caches[i]); + kmem_cache_destroy(ext4_groupinfo_caches[i]); ext4_groupinfo_caches[i] = NULL; } } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index b1f21e3a0763..4a09063ce1d2 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2411,8 +2411,7 @@ static int ext4_add_nondir(handle_t *handle, int err = ext4_add_entry(handle, dentry, inode); if (!err) { ext4_mark_inode_dirty(handle, inode); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } drop_nlink(inode); @@ -2651,8 +2650,7 @@ out_clear_inode: err = ext4_mark_inode_dirty(handle, dir); if (err) goto out_clear_inode; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) ext4_handle_sync(handle); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index b6bec270a8e4..d792b7689d92 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -1933,7 +1933,7 @@ retry: return 0; n_group = ext4_get_group_number(sb, n_blocks_count - 1); - if (n_group > (0xFFFFFFFFUL / EXT4_INODES_PER_GROUP(sb))) { + if (n_group >= (0xFFFFFFFFUL / EXT4_INODES_PER_GROUP(sb))) { ext4_warning(sb, "resize would cause inodes_count overflow"); return -EINVAL; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 185f7e61f4cf..00fe75a71c4b 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -763,6 +763,36 @@ __acquires(bitlock) return; } +void ext4_mark_group_bitmap_corrupted(struct super_block *sb, + ext4_group_t group, + unsigned int flags) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_group_info *grp = ext4_get_group_info(sb, group); + struct ext4_group_desc *gdp = ext4_get_group_desc(sb, group, NULL); + + if ((flags & EXT4_GROUP_INFO_BBITMAP_CORRUPT) && + !EXT4_MB_GRP_BBITMAP_CORRUPT(grp)) { + percpu_counter_sub(&sbi->s_freeclusters_counter, + grp->bb_free); + set_bit(EXT4_GROUP_INFO_BBITMAP_CORRUPT_BIT, + &grp->bb_state); + } + + if ((flags & EXT4_GROUP_INFO_IBITMAP_CORRUPT) && + !EXT4_MB_GRP_IBITMAP_CORRUPT(grp)) { + if (gdp) { + int count; + + count = ext4_free_inodes_count(sb, gdp); + percpu_counter_sub(&sbi->s_freeinodes_counter, + count); + } + set_bit(EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT, + &grp->bb_state); + } +} + void ext4_update_dynamic_rev(struct super_block *sb) { struct ext4_super_block *es = EXT4_SB(sb)->s_es; @@ -1237,19 +1267,13 @@ static bool ext4_dummy_context(struct inode *inode) return DUMMY_ENCRYPTION_ENABLED(EXT4_SB(inode->i_sb)); } -static unsigned ext4_max_namelen(struct inode *inode) -{ - return S_ISLNK(inode->i_mode) ? inode->i_sb->s_blocksize : - EXT4_NAME_LEN; -} - static const struct fscrypt_operations ext4_cryptops = { .key_prefix = "ext4:", .get_context = ext4_get_context, .set_context = ext4_set_context, .dummy_context = ext4_dummy_context, .empty_dir = ext4_empty_dir, - .max_namelen = ext4_max_namelen, + .max_namelen = EXT4_NAME_LEN, }; #endif @@ -2116,12 +2140,12 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, int read_only) { struct ext4_sb_info *sbi = EXT4_SB(sb); - int res = 0; + int err = 0; if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) { ext4_msg(sb, KERN_ERR, "revision level too high, " "forcing read-only mode"); - res = SB_RDONLY; + err = -EROFS; } if (read_only) goto done; @@ -2154,7 +2178,7 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, if (sbi->s_journal) ext4_set_feature_journal_needs_recovery(sb); - ext4_commit_super(sb, 1); + err = ext4_commit_super(sb, 1); done: if (test_opt(sb, DEBUG)) printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%u, " @@ -2166,7 +2190,7 @@ done: sbi->s_mount_opt, sbi->s_mount_opt2); cleancache_init_fs(sb); - return res; + return err; } int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup) @@ -3732,8 +3756,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) " that may contain inline data"); sbi->s_mount_opt &= ~EXT4_MOUNT_DAX; } - err = bdev_dax_supported(sb, blocksize); - if (err) { + if (!bdev_dax_supported(sb->s_bdev, blocksize)) { ext4_msg(sb, KERN_ERR, "DAX unsupported by block device. Turning off DAX."); sbi->s_mount_opt &= ~EXT4_MOUNT_DAX; @@ -4224,8 +4247,12 @@ no_journal: goto failed_mount4; } - if (ext4_setup_super(sb, es, sb_rdonly(sb))) + ret = ext4_setup_super(sb, es, sb_rdonly(sb)); + if (ret == -EROFS) { sb->s_flags |= SB_RDONLY; + ret = 0; + } else if (ret) + goto failed_mount4a; /* determine the minimum size of new large inodes, if present */ if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE && @@ -4760,11 +4787,7 @@ static int ext4_commit_super(struct super_block *sb, int sync) unlock_buffer(sbh); error = __sync_dirty_buffer(sbh, REQ_SYNC | (test_opt(sb, BARRIER) ? REQ_FUA : 0)); - if (error) - return error; - - error = buffer_write_io_error(sbh); - if (error) { + if (buffer_write_io_error(sbh)) { ext4_msg(sb, KERN_ERR, "I/O error while writing " "superblock"); clear_buffer_write_io_error(sbh); @@ -5165,8 +5188,12 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) if (sbi->s_journal) ext4_clear_journal_err(sb, es); sbi->s_mount_state = le16_to_cpu(es->s_state); - if (!ext4_setup_super(sb, es, 0)) - sb->s_flags &= ~SB_RDONLY; + + err = ext4_setup_super(sb, es, 0); + if (err) + goto restore_opts; + + sb->s_flags &= ~SB_RDONLY; if (ext4_has_feature_mmp(sb)) if (ext4_multi_mount_protect(sb, le64_to_cpu(es->s_mmp_block))) { @@ -5190,8 +5217,11 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) } ext4_setup_system_zone(sb); - if (sbi->s_journal == NULL && !(old_sb_flags & SB_RDONLY)) - ext4_commit_super(sb, 1); + if (sbi->s_journal == NULL && !(old_sb_flags & SB_RDONLY)) { + err = ext4_commit_super(sb, 1); + if (err) + goto restore_opts; + } #ifdef CONFIG_QUOTA /* Release old quota file names */ @@ -5252,7 +5282,8 @@ static int ext4_statfs_project(struct super_block *sb, dquot->dq_dqb.dqb_bsoftlimit : dquot->dq_dqb.dqb_bhardlimit) >> sb->s_blocksize_bits; if (limit && buf->f_blocks > limit) { - curblock = dquot->dq_dqb.dqb_curspace >> sb->s_blocksize_bits; + curblock = (dquot->dq_dqb.dqb_curspace + + dquot->dq_dqb.dqb_rsvspace) >> sb->s_blocksize_bits; buf->f_blocks = limit; buf->f_bfree = buf->f_bavail = (buf->f_blocks > curblock) ? @@ -5886,5 +5917,6 @@ static void __exit ext4_exit_fs(void) MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); MODULE_DESCRIPTION("Fourth Extended Filesystem"); MODULE_LICENSE("GPL"); +MODULE_SOFTDEP("pre: crc32c"); module_init(ext4_init_fs) module_exit(ext4_exit_fs) diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 9ebd26c957c2..f34da0bb8f17 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -346,39 +346,9 @@ static struct kobject *ext4_root; static struct kobject *ext4_feat; -#define PROC_FILE_SHOW_DEFN(name) \ -static int name##_open(struct inode *inode, struct file *file) \ -{ \ - return single_open(file, ext4_seq_##name##_show, PDE_DATA(inode)); \ -} \ -\ -static const struct file_operations ext4_seq_##name##_fops = { \ - .open = name##_open, \ - .read = seq_read, \ - .llseek = seq_lseek, \ - .release = single_release, \ -} - -#define PROC_FILE_LIST(name) \ - { __stringify(name), &ext4_seq_##name##_fops } - -PROC_FILE_SHOW_DEFN(es_shrinker_info); -PROC_FILE_SHOW_DEFN(options); - -static const struct ext4_proc_files { - const char *name; - const struct file_operations *fops; -} proc_files[] = { - PROC_FILE_LIST(options), - PROC_FILE_LIST(es_shrinker_info), - PROC_FILE_LIST(mb_groups), - { NULL, NULL }, -}; - int ext4_register_sysfs(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); - const struct ext4_proc_files *p; int err; init_completion(&sbi->s_kobj_unregister); @@ -392,11 +362,14 @@ int ext4_register_sysfs(struct super_block *sb) if (ext4_proc_root) sbi->s_proc = proc_mkdir(sb->s_id, ext4_proc_root); - if (sbi->s_proc) { - for (p = proc_files; p->name; p++) - proc_create_data(p->name, S_IRUGO, sbi->s_proc, - p->fops, sb); + proc_create_single_data("options", S_IRUGO, sbi->s_proc, + ext4_seq_options_show, sb); + proc_create_single_data("es_shrinker_info", S_IRUGO, + sbi->s_proc, ext4_seq_es_shrinker_info_show, + sb); + proc_create_seq_data("mb_groups", S_IRUGO, sbi->s_proc, + &ext4_mb_seq_groups_ops, sb); } return 0; } @@ -404,13 +377,9 @@ int ext4_register_sysfs(struct super_block *sb) void ext4_unregister_sysfs(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); - const struct ext4_proc_files *p; - if (sbi->s_proc) { - for (p = proc_files; p->name; p++) - remove_proc_entry(p->name, sbi->s_proc); - remove_proc_entry(sb->s_id, ext4_proc_root); - } + if (sbi->s_proc) + remove_proc_subtree(sb->s_id, ext4_proc_root); kobject_del(&sbi->s_kobj); } diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 499cb4b1fbd2..fc4ced59c565 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1688,7 +1688,7 @@ static int ext4_xattr_set_entry(struct ext4_xattr_info *i, /* No failures allowed past this point. */ - if (!s->not_found && here->e_value_offs) { + if (!s->not_found && here->e_value_size && here->e_value_offs) { /* Remove the old value. */ void *first_val = s->base + min_offs; size_t offs = le16_to_cpu(here->e_value_offs); diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c index 629001b28632..197a9d8a15ef 100644 --- a/fs/ext4/xattr_security.c +++ b/fs/ext4/xattr_security.c @@ -43,7 +43,7 @@ ext4_initxattrs(struct inode *inode, const struct xattr *xattr_array, err = ext4_xattr_set_handle(handle, inode, EXT4_XATTR_INDEX_SECURITY, xattr->name, xattr->value, - xattr->value_len, 0); + xattr->value_len, XATTR_CREATE); if (err < 0) break; } diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index db50686f5096..02237d4d91f5 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -2424,12 +2424,12 @@ void f2fs_set_page_dirty_nobuffers(struct page *page) SetPageDirty(page); spin_unlock(&mapping->private_lock); - spin_lock_irqsave(&mapping->tree_lock, flags); + xa_lock_irqsave(&mapping->i_pages, flags); WARN_ON_ONCE(!PageUptodate(page)); account_page_dirtied(page, mapping); - radix_tree_tag_set(&mapping->page_tree, + radix_tree_tag_set(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); - spin_unlock_irqrestore(&mapping->tree_lock, flags); + xa_unlock_irqrestore(&mapping->i_pages, flags); unlock_page_memcg(page); __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index fe661274ff10..8c9c2f31b253 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -732,10 +732,10 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page, if (bit_pos == NR_DENTRY_IN_BLOCK && !truncate_hole(dir, page->index, page->index + 1)) { - spin_lock_irqsave(&mapping->tree_lock, flags); - radix_tree_tag_clear(&mapping->page_tree, page_index(page), + xa_lock_irqsave(&mapping->i_pages, flags); + radix_tree_tag_clear(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); - spin_unlock_irqrestore(&mapping->tree_lock, flags); + xa_unlock_irqrestore(&mapping->i_pages, flags); clear_page_dirty_for_io(page); ClearPagePrivate(page); diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index bfb7a4a3a929..9327411fd93b 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -1015,7 +1015,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, unsigned int init_segno = segno; struct gc_inode_list gc_list = { .ilist = LIST_HEAD_INIT(gc_list.ilist), - .iroot = RADIX_TREE_INIT(GFP_NOFS), + .iroot = RADIX_TREE_INIT(gc_list.iroot, GFP_NOFS), }; trace_f2fs_gc_begin(sbi->sb, sync, background, diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index 3b77d6421218..265da200daa8 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -226,10 +226,10 @@ int f2fs_write_inline_data(struct inode *inode, struct page *page) kunmap_atomic(src_addr); set_page_dirty(dn.inode_page); - spin_lock_irqsave(&mapping->tree_lock, flags); - radix_tree_tag_clear(&mapping->page_tree, page_index(page), + xa_lock_irqsave(&mapping->i_pages, flags); + radix_tree_tag_clear(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); - spin_unlock_irqrestore(&mapping->tree_lock, flags); + xa_unlock_irqrestore(&mapping->i_pages, flags); set_inode_flag(inode, FI_APPEND_WRITE); set_inode_flag(inode, FI_DATA_EXIST); diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index d5098efe577c..75e37fd720b2 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -294,8 +294,7 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode, alloc_nid_done(sbi, ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); @@ -597,8 +596,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry, err = page_symlink(inode, disk_link.name, disk_link.len); err_out: - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); /* * Let's flush symlink data in order to avoid broken symlink as much as @@ -661,8 +659,7 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) alloc_nid_done(sbi, inode->i_ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); @@ -713,8 +710,7 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry, alloc_nid_done(sbi, inode->i_ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 9a99243054ba..f202398e20ea 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -91,11 +91,11 @@ static void clear_node_page_dirty(struct page *page) unsigned int long flags; if (PageDirty(page)) { - spin_lock_irqsave(&mapping->tree_lock, flags); - radix_tree_tag_clear(&mapping->page_tree, + xa_lock_irqsave(&mapping->i_pages, flags); + radix_tree_tag_clear(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); - spin_unlock_irqrestore(&mapping->tree_lock, flags); + xa_unlock_irqrestore(&mapping->i_pages, flags); clear_page_dirty_for_io(page); dec_page_count(F2FS_M_SB(mapping), F2FS_DIRTY_NODES); @@ -1161,7 +1161,7 @@ void ra_node_page(struct f2fs_sb_info *sbi, nid_t nid) f2fs_bug_on(sbi, check_nid_range(sbi, nid)); rcu_read_lock(); - apage = radix_tree_lookup(&NODE_MAPPING(sbi)->page_tree, nid); + apage = radix_tree_lookup(&NODE_MAPPING(sbi)->i_pages, nid); rcu_read_unlock(); if (apage) return; diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 42d564c5ccd0..970ae27f401c 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -1930,19 +1930,13 @@ static bool f2fs_dummy_context(struct inode *inode) return DUMMY_ENCRYPTION_ENABLED(F2FS_I_SB(inode)); } -static unsigned f2fs_max_namelen(struct inode *inode) -{ - return S_ISLNK(inode->i_mode) ? - inode->i_sb->s_blocksize : F2FS_NAME_LEN; -} - static const struct fscrypt_operations f2fs_cryptops = { .key_prefix = "f2fs:", .get_context = f2fs_get_context, .set_context = f2fs_set_context, .dummy_context = f2fs_dummy_context, .empty_dir = f2fs_empty_dir, - .max_namelen = f2fs_max_namelen, + .max_namelen = F2FS_NAME_LEN, }; #endif diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index f33a56d6e6dd..4b47ca6296a7 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -572,23 +572,6 @@ static int iostat_info_seq_show(struct seq_file *seq, void *offset) return 0; } -#define F2FS_PROC_FILE_DEF(_name) \ -static int _name##_open_fs(struct inode *inode, struct file *file) \ -{ \ - return single_open(file, _name##_seq_show, PDE_DATA(inode)); \ -} \ - \ -static const struct file_operations f2fs_seq_##_name##_fops = { \ - .open = _name##_open_fs, \ - .read = seq_read, \ - .llseek = seq_lseek, \ - .release = single_release, \ -}; - -F2FS_PROC_FILE_DEF(segment_info); -F2FS_PROC_FILE_DEF(segment_bits); -F2FS_PROC_FILE_DEF(iostat_info); - int __init f2fs_init_sysfs(void) { int ret; @@ -632,12 +615,12 @@ int f2fs_register_sysfs(struct f2fs_sb_info *sbi) sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root); if (sbi->s_proc) { - proc_create_data("segment_info", S_IRUGO, sbi->s_proc, - &f2fs_seq_segment_info_fops, sb); - proc_create_data("segment_bits", S_IRUGO, sbi->s_proc, - &f2fs_seq_segment_bits_fops, sb); - proc_create_data("iostat_info", S_IRUGO, sbi->s_proc, - &f2fs_seq_iostat_info_fops, sb); + proc_create_single_data("segment_info", S_IRUGO, sbi->s_proc, + segment_info_seq_show, sb); + proc_create_single_data("segment_bits", S_IRUGO, sbi->s_proc, + segment_bits_seq_show, sb); + proc_create_single_data("iostat_info", S_IRUGO, sbi->s_proc, + iostat_info_seq_show, sb); } return 0; } diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index 582ca731a6c9..484ce674e0cd 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -314,10 +314,6 @@ static int msdos_rmdir(struct inode *dir, struct dentry *dentry) int err; mutex_lock(&MSDOS_SB(sb)->s_lock); - /* - * Check whether the directory is not in use, then check - * whether it is empty. - */ err = fat_dir_empty(inode); if (err) goto out; diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index 2649759c478a..4f4362d5a04c 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -697,15 +697,6 @@ static int vfat_find(struct inode *dir, const struct qstr *qname, return fat_search_long(dir, qname->name, len, sinfo); } -/* - * (nfsd's) anonymous disconnected dentry? - * NOTE: !IS_ROOT() is not anonymous (I.e. d_splice_alias() did the job). - */ -static int vfat_d_anon_disconn(struct dentry *dentry) -{ - return IS_ROOT(dentry) && (dentry->d_flags & DCACHE_DISCONNECTED); -} - static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { @@ -738,8 +729,7 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, * Checking "alias->d_parent == dentry->d_parent" to make sure * FS is not corrupted (especially double linked dir). */ - if (alias && alias->d_parent == dentry->d_parent && - !vfat_d_anon_disconn(alias)) { + if (alias && alias->d_parent == dentry->d_parent) { /* * This inode has non anonymous-DCACHE_DISCONNECTED * dentry. This means, the user did ->lookup() by an @@ -747,7 +737,6 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, * * Switch to new one for reason of locality if possible. */ - BUG_ON(d_unhashed(alias)); if (!S_ISDIR(inode->i_mode)) d_move(alias, dentry); iput(inode); diff --git a/fs/fcntl.c b/fs/fcntl.c index d737ff082472..c42169459298 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -871,9 +871,9 @@ int fasync_remove_entry(struct file *filp, struct fasync_struct **fapp) if (fa->fa_file != filp) continue; - spin_lock_irq(&fa->fa_lock); + write_lock_irq(&fa->fa_lock); fa->fa_file = NULL; - spin_unlock_irq(&fa->fa_lock); + write_unlock_irq(&fa->fa_lock); *fp = fa->fa_next; call_rcu(&fa->fa_rcu, fasync_free_rcu); @@ -918,13 +918,13 @@ struct fasync_struct *fasync_insert_entry(int fd, struct file *filp, struct fasy if (fa->fa_file != filp) continue; - spin_lock_irq(&fa->fa_lock); + write_lock_irq(&fa->fa_lock); fa->fa_fd = fd; - spin_unlock_irq(&fa->fa_lock); + write_unlock_irq(&fa->fa_lock); goto out; } - spin_lock_init(&new->fa_lock); + rwlock_init(&new->fa_lock); new->magic = FASYNC_MAGIC; new->fa_file = filp; new->fa_fd = fd; @@ -987,14 +987,13 @@ static void kill_fasync_rcu(struct fasync_struct *fa, int sig, int band) { while (fa) { struct fown_struct *fown; - unsigned long flags; if (fa->magic != FASYNC_MAGIC) { printk(KERN_ERR "kill_fasync: bad magic number in " "fasync_struct!\n"); return; } - spin_lock_irqsave(&fa->fa_lock, flags); + read_lock(&fa->fa_lock); if (fa->fa_file) { fown = &fa->fa_file->f_owner; /* Don't send SIGURG to processes which have not set a @@ -1003,7 +1002,7 @@ static void kill_fasync_rcu(struct fasync_struct *fa, int sig, int band) if (!(sig == SIGURG && fown->signum == 0)) send_sigio(fown, fa->fa_fd, band); } - spin_unlock_irqrestore(&fa->fa_lock, flags); + read_unlock(&fa->fa_lock); fa = rcu_dereference(fa->fa_next); } } diff --git a/fs/filesystems.c b/fs/filesystems.c index f2728a4a03a1..b03f57b1105b 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -238,21 +238,9 @@ static int filesystems_proc_show(struct seq_file *m, void *v) return 0; } -static int filesystems_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, filesystems_proc_show, NULL); -} - -static const struct file_operations filesystems_proc_fops = { - .open = filesystems_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_filesystems_init(void) { - proc_create("filesystems", 0, NULL, &filesystems_proc_fops); + proc_create_single("filesystems", 0, NULL, filesystems_proc_show); return 0; } module_init(proc_filesystems_init); diff --git a/fs/freevxfs/vxfs_lookup.c b/fs/freevxfs/vxfs_lookup.c index ce4785fd81c6..a51425634f65 100644 --- a/fs/freevxfs/vxfs_lookup.c +++ b/fs/freevxfs/vxfs_lookup.c @@ -193,13 +193,9 @@ vxfs_lookup(struct inode *dip, struct dentry *dp, unsigned int flags) return ERR_PTR(-ENAMETOOLONG); ino = vxfs_inode_by_name(dip, dp); - if (ino) { + if (ino) ip = vxfs_iget(dip->i_sb, ino); - if (IS_ERR(ip)) - return ERR_CAST(ip); - } - d_add(dp, ip); - return NULL; + return d_splice_alias(ip, dp); } /** diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index d4d04fee568a..471d863958bc 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -347,9 +347,9 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) * By the time control reaches here, RCU grace period has passed * since I_WB_SWITCH assertion and all wb stat update transactions * between unlocked_inode_to_wb_begin/end() are guaranteed to be - * synchronizing against mapping->tree_lock. + * synchronizing against the i_pages lock. * - * Grabbing old_wb->list_lock, inode->i_lock and mapping->tree_lock + * Grabbing old_wb->list_lock, inode->i_lock and the i_pages lock * gives us exclusion against all wb related operations on @inode * including IO list manipulations and stat updates. */ @@ -361,7 +361,7 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) spin_lock_nested(&old_wb->list_lock, SINGLE_DEPTH_NESTING); } spin_lock(&inode->i_lock); - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(&mapping->i_pages); /* * Once I_FREEING is visible under i_lock, the eviction path owns @@ -373,22 +373,22 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) /* * Count and transfer stats. Note that PAGECACHE_TAG_DIRTY points * to possibly dirty pages while PAGECACHE_TAG_WRITEBACK points to - * pages actually under underwriteback. + * pages actually under writeback. */ - radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, 0, + radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, 0, PAGECACHE_TAG_DIRTY) { struct page *page = radix_tree_deref_slot_protected(slot, - &mapping->tree_lock); + &mapping->i_pages.xa_lock); if (likely(page) && PageDirty(page)) { dec_wb_stat(old_wb, WB_RECLAIMABLE); inc_wb_stat(new_wb, WB_RECLAIMABLE); } } - radix_tree_for_each_tagged(slot, &mapping->page_tree, &iter, 0, + radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, 0, PAGECACHE_TAG_WRITEBACK) { struct page *page = radix_tree_deref_slot_protected(slot, - &mapping->tree_lock); + &mapping->i_pages.xa_lock); if (likely(page)) { WARN_ON_ONCE(!PageWriteback(page)); dec_wb_stat(old_wb, WB_WRITEBACK); @@ -430,7 +430,7 @@ skip_switch: */ smp_store_release(&inode->i_state, inode->i_state & ~I_WB_SWITCH); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); spin_unlock(&inode->i_lock); spin_unlock(&new_wb->list_lock); spin_unlock(&old_wb->list_lock); @@ -506,8 +506,8 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id) /* * In addition to synchronizing among switchers, I_WB_SWITCH tells - * the RCU protected stat update paths to grab the mapping's - * tree_lock so that stat transfer can synchronize against them. + * the RCU protected stat update paths to grab the i_page + * lock so that stat transfer can synchronize against them. * Let's continue after I_WB_SWITCH is guaranteed to be visible. */ call_rcu(&isw->rcu_head, inode_switch_wbs_rcu_fn); @@ -745,11 +745,12 @@ int inode_congested(struct inode *inode, int cong_bits) */ if (inode && inode_to_wb_is_valid(inode)) { struct bdi_writeback *wb; - bool locked, congested; + struct wb_lock_cookie lock_cookie = {}; + bool congested; - wb = unlocked_inode_to_wb_begin(inode, &locked); + wb = unlocked_inode_to_wb_begin(inode, &lock_cookie); congested = wb_congested(wb, cong_bits); - unlocked_inode_to_wb_end(inode, locked); + unlocked_inode_to_wb_end(inode, &lock_cookie); return congested; } @@ -1343,7 +1344,7 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) dirty = inode->i_state & I_DIRTY; if (inode->i_state & I_DIRTY_TIME) { - if ((dirty & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) || + if ((dirty & I_DIRTY_INODE) || wbc->sync_mode == WB_SYNC_ALL || unlikely(inode->i_state & I_DIRTY_TIME_EXPIRED) || unlikely(time_after(jiffies, @@ -1960,7 +1961,7 @@ void wb_workfn(struct work_struct *work) } if (!list_empty(&wb->work_list)) - mod_delayed_work(bdi_wq, &wb->dwork, 0); + wb_wakeup(wb); else if (wb_has_dirty_io(wb) && dirty_writeback_interval) wb_wakeup_delayed(wb); @@ -2112,7 +2113,6 @@ static noinline void block_dump___mark_inode_dirty(struct inode *inode) */ void __mark_inode_dirty(struct inode *inode, int flags) { -#define I_DIRTY_INODE (I_DIRTY_SYNC | I_DIRTY_DATASYNC) struct super_block *sb = inode->i_sb; int dirtytime; @@ -2122,7 +2122,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) * Don't do this for I_DIRTY_PAGES - that doesn't actually * dirty the inode itself */ - if (flags & (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_TIME)) { + if (flags & (I_DIRTY_INODE | I_DIRTY_TIME)) { trace_writeback_dirty_inode_start(inode, flags); if (sb->s_op->dirty_inode) @@ -2197,7 +2197,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) if (dirtytime) inode->dirtied_time_when = jiffies; - if (inode->i_state & (I_DIRTY_INODE | I_DIRTY_PAGES)) + if (inode->i_state & I_DIRTY) dirty_list = &wb->b_dirty; else dirty_list = &wb->b_dirty_time; @@ -2221,8 +2221,6 @@ void __mark_inode_dirty(struct inode *inode, int flags) } out_unlock_inode: spin_unlock(&inode->i_lock); - -#undef I_DIRTY_INODE } EXPORT_SYMBOL(__mark_inode_dirty); diff --git a/fs/fscache/cache.c b/fs/fscache/cache.c index 56cce7fdd39e..c184c5a356ff 100644 --- a/fs/fscache/cache.c +++ b/fs/fscache/cache.c @@ -125,7 +125,7 @@ struct fscache_cache *fscache_select_cache_for_object( } /* the parent is unbacked */ - if (cookie->def->type != FSCACHE_COOKIE_TYPE_INDEX) { + if (cookie->type != FSCACHE_COOKIE_TYPE_INDEX) { /* cookie not an index and is unbacked */ spin_unlock(&cookie->lock); _leave(" = NULL [cookie ub,ni]"); diff --git a/fs/fscache/cookie.c b/fs/fscache/cookie.c index d705125665f0..97137d7ec5ee 100644 --- a/fs/fscache/cookie.c +++ b/fs/fscache/cookie.c @@ -21,12 +21,54 @@ struct kmem_cache *fscache_cookie_jar; static atomic_t fscache_object_debug_id = ATOMIC_INIT(0); -static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie); +#define fscache_cookie_hash_shift 15 +static struct hlist_bl_head fscache_cookie_hash[1 << fscache_cookie_hash_shift]; + +static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie, + loff_t object_size); static int fscache_alloc_object(struct fscache_cache *cache, struct fscache_cookie *cookie); static int fscache_attach_object(struct fscache_cookie *cookie, struct fscache_object *object); +static void fscache_print_cookie(struct fscache_cookie *cookie, char prefix) +{ + struct hlist_node *object; + const u8 *k; + unsigned loop; + + pr_err("%c-cookie c=%p [p=%p fl=%lx nc=%u na=%u]\n", + prefix, cookie, cookie->parent, cookie->flags, + atomic_read(&cookie->n_children), + atomic_read(&cookie->n_active)); + pr_err("%c-cookie d=%p n=%p\n", + prefix, cookie->def, cookie->netfs_data); + + object = READ_ONCE(cookie->backing_objects.first); + if (object) + pr_err("%c-cookie o=%p\n", + prefix, hlist_entry(object, struct fscache_object, cookie_link)); + + pr_err("%c-key=[%u] '", prefix, cookie->key_len); + k = (cookie->key_len <= sizeof(cookie->inline_key)) ? + cookie->inline_key : cookie->key; + for (loop = 0; loop < cookie->key_len; loop++) + pr_cont("%02x", k[loop]); + pr_cont("'\n"); +} + +void fscache_free_cookie(struct fscache_cookie *cookie) +{ + if (cookie) { + BUG_ON(!hlist_empty(&cookie->backing_objects)); + if (cookie->aux_len > sizeof(cookie->inline_aux)) + kfree(cookie->aux); + if (cookie->key_len > sizeof(cookie->inline_key)) + kfree(cookie->key); + kmem_cache_free(fscache_cookie_jar, cookie); + } +} + /* * initialise an cookie jar slab element prior to any use */ @@ -41,6 +83,170 @@ void fscache_cookie_init_once(void *_cookie) } /* + * Set the index key in a cookie. The cookie struct has space for a 12-byte + * key plus length and hash, but if that's not big enough, it's instead a + * pointer to a buffer containing 3 bytes of hash, 1 byte of length and then + * the key data. + */ +static int fscache_set_key(struct fscache_cookie *cookie, + const void *index_key, size_t index_key_len) +{ + unsigned long long h; + u32 *buf; + int i; + + cookie->key_len = index_key_len; + + if (index_key_len > sizeof(cookie->inline_key)) { + buf = kzalloc(index_key_len, GFP_KERNEL); + if (!buf) + return -ENOMEM; + cookie->key = buf; + } else { + buf = (u32 *)cookie->inline_key; + buf[0] = 0; + buf[1] = 0; + buf[2] = 0; + } + + memcpy(buf, index_key, index_key_len); + + /* Calculate a hash and combine this with the length in the first word + * or first half word + */ + h = (unsigned long)cookie->parent; + h += index_key_len + cookie->type; + for (i = 0; i < (index_key_len + sizeof(u32) - 1) / sizeof(u32); i++) + h += buf[i]; + + cookie->key_hash = h ^ (h >> 32); + return 0; +} + +static long fscache_compare_cookie(const struct fscache_cookie *a, + const struct fscache_cookie *b) +{ + const void *ka, *kb; + + if (a->key_hash != b->key_hash) + return (long)a->key_hash - (long)b->key_hash; + if (a->parent != b->parent) + return (long)a->parent - (long)b->parent; + if (a->key_len != b->key_len) + return (long)a->key_len - (long)b->key_len; + if (a->type != b->type) + return (long)a->type - (long)b->type; + + if (a->key_len <= sizeof(a->inline_key)) { + ka = &a->inline_key; + kb = &b->inline_key; + } else { + ka = a->key; + kb = b->key; + } + return memcmp(ka, kb, a->key_len); +} + +/* + * Allocate a cookie. + */ +struct fscache_cookie *fscache_alloc_cookie( + struct fscache_cookie *parent, + const struct fscache_cookie_def *def, + const void *index_key, size_t index_key_len, + const void *aux_data, size_t aux_data_len, + void *netfs_data, + loff_t object_size) +{ + struct fscache_cookie *cookie; + + /* allocate and initialise a cookie */ + cookie = kmem_cache_alloc(fscache_cookie_jar, GFP_KERNEL); + if (!cookie) + return NULL; + + cookie->key_len = index_key_len; + cookie->aux_len = aux_data_len; + + if (fscache_set_key(cookie, index_key, index_key_len) < 0) + goto nomem; + + if (cookie->aux_len <= sizeof(cookie->inline_aux)) { + memcpy(cookie->inline_aux, aux_data, cookie->aux_len); + } else { + cookie->aux = kmemdup(aux_data, cookie->aux_len, GFP_KERNEL); + if (!cookie->aux) + goto nomem; + } + + atomic_set(&cookie->usage, 1); + atomic_set(&cookie->n_children, 0); + + /* We keep the active count elevated until relinquishment to prevent an + * attempt to wake up every time the object operations queue quiesces. + */ + atomic_set(&cookie->n_active, 1); + + cookie->def = def; + cookie->parent = parent; + cookie->netfs_data = netfs_data; + cookie->flags = (1 << FSCACHE_COOKIE_NO_DATA_YET); + cookie->type = def->type; + + /* radix tree insertion won't use the preallocation pool unless it's + * told it may not wait */ + INIT_RADIX_TREE(&cookie->stores, GFP_NOFS & ~__GFP_DIRECT_RECLAIM); + return cookie; + +nomem: + fscache_free_cookie(cookie); + return NULL; +} + +/* + * Attempt to insert the new cookie into the hash. If there's a collision, we + * return the old cookie if it's not in use and an error otherwise. + */ +struct fscache_cookie *fscache_hash_cookie(struct fscache_cookie *candidate) +{ + struct fscache_cookie *cursor; + struct hlist_bl_head *h; + struct hlist_bl_node *p; + unsigned int bucket; + + bucket = candidate->key_hash & (ARRAY_SIZE(fscache_cookie_hash) - 1); + h = &fscache_cookie_hash[bucket]; + + hlist_bl_lock(h); + hlist_bl_for_each_entry(cursor, p, h, hash_link) { + if (fscache_compare_cookie(candidate, cursor) == 0) + goto collision; + } + + __set_bit(FSCACHE_COOKIE_ACQUIRED, &candidate->flags); + fscache_cookie_get(candidate->parent, fscache_cookie_get_acquire_parent); + atomic_inc(&candidate->parent->n_children); + hlist_bl_add_head(&candidate->hash_link, h); + hlist_bl_unlock(h); + return candidate; + +collision: + if (test_and_set_bit(FSCACHE_COOKIE_ACQUIRED, &cursor->flags)) { + trace_fscache_cookie(cursor, fscache_cookie_collision, + atomic_read(&cursor->usage)); + pr_err("Duplicate cookie detected\n"); + fscache_print_cookie(cursor, 'O'); + fscache_print_cookie(candidate, 'N'); + hlist_bl_unlock(h); + return NULL; + } + + fscache_cookie_get(cursor, fscache_cookie_get_reacquire); + hlist_bl_unlock(h); + return cursor; +} + +/* * request a cookie to represent an object (index, datafile, xattr, etc) * - parent specifies the parent object * - the top level index cookie for each netfs is stored in the fscache_netfs @@ -58,10 +264,13 @@ void fscache_cookie_init_once(void *_cookie) struct fscache_cookie *__fscache_acquire_cookie( struct fscache_cookie *parent, const struct fscache_cookie_def *def, + const void *index_key, size_t index_key_len, + const void *aux_data, size_t aux_data_len, void *netfs_data, + loff_t object_size, bool enable) { - struct fscache_cookie *cookie; + struct fscache_cookie *candidate, *cookie; BUG_ON(!def); @@ -69,6 +278,13 @@ struct fscache_cookie *__fscache_acquire_cookie( parent ? (char *) parent->def->name : "<no-parent>", def->name, netfs_data, enable); + if (!index_key || !index_key_len || index_key_len > 255 || aux_data_len > 255) + return NULL; + if (!aux_data || !aux_data_len) { + aux_data = NULL; + aux_data_len = 0; + } + fscache_stat(&fscache_n_acquires); /* if there's no parent cookie, then we don't create one here either */ @@ -79,41 +295,31 @@ struct fscache_cookie *__fscache_acquire_cookie( } /* validate the definition */ - BUG_ON(!def->get_key); BUG_ON(!def->name[0]); BUG_ON(def->type == FSCACHE_COOKIE_TYPE_INDEX && - parent->def->type != FSCACHE_COOKIE_TYPE_INDEX); + parent->type != FSCACHE_COOKIE_TYPE_INDEX); - /* allocate and initialise a cookie */ - cookie = kmem_cache_alloc(fscache_cookie_jar, GFP_KERNEL); - if (!cookie) { + candidate = fscache_alloc_cookie(parent, def, + index_key, index_key_len, + aux_data, aux_data_len, + netfs_data, object_size); + if (!candidate) { fscache_stat(&fscache_n_acquires_oom); _leave(" [ENOMEM]"); return NULL; } - atomic_set(&cookie->usage, 1); - atomic_set(&cookie->n_children, 0); - - /* We keep the active count elevated until relinquishment to prevent an - * attempt to wake up every time the object operations queue quiesces. - */ - atomic_set(&cookie->n_active, 1); - - atomic_inc(&parent->usage); - atomic_inc(&parent->n_children); + cookie = fscache_hash_cookie(candidate); + if (!cookie) { + trace_fscache_cookie(candidate, fscache_cookie_discard, 1); + goto out; + } - cookie->def = def; - cookie->parent = parent; - cookie->netfs_data = netfs_data; - cookie->flags = (1 << FSCACHE_COOKIE_NO_DATA_YET); + if (cookie == candidate) + candidate = NULL; - /* radix tree insertion won't use the preallocation pool unless it's - * told it may not wait */ - INIT_RADIX_TREE(&cookie->stores, GFP_NOFS & ~__GFP_DIRECT_RECLAIM); - - switch (cookie->def->type) { + switch (cookie->type) { case FSCACHE_COOKIE_TYPE_INDEX: fscache_stat(&fscache_n_cookie_index); break; @@ -125,16 +331,19 @@ struct fscache_cookie *__fscache_acquire_cookie( break; } + trace_fscache_acquire(cookie); + if (enable) { /* if the object is an index then we need do nothing more here * - we create indices on disk when we need them as an index * may exist in multiple caches */ - if (cookie->def->type != FSCACHE_COOKIE_TYPE_INDEX) { - if (fscache_acquire_non_index_cookie(cookie) == 0) { + if (cookie->type != FSCACHE_COOKIE_TYPE_INDEX) { + if (fscache_acquire_non_index_cookie(cookie, object_size) == 0) { set_bit(FSCACHE_COOKIE_ENABLED, &cookie->flags); } else { atomic_dec(&parent->n_children); - __fscache_cookie_put(cookie); + fscache_cookie_put(cookie, + fscache_cookie_put_acquire_nobufs); fscache_stat(&fscache_n_acquires_nobufs); _leave(" = NULL"); return NULL; @@ -145,7 +354,9 @@ struct fscache_cookie *__fscache_acquire_cookie( } fscache_stat(&fscache_n_acquires_ok); - _leave(" = %p", cookie); + +out: + fscache_free_cookie(candidate); return cookie; } EXPORT_SYMBOL(__fscache_acquire_cookie); @@ -154,24 +365,30 @@ EXPORT_SYMBOL(__fscache_acquire_cookie); * Enable a cookie to permit it to accept new operations. */ void __fscache_enable_cookie(struct fscache_cookie *cookie, + const void *aux_data, + loff_t object_size, bool (*can_enable)(void *data), void *data) { _enter("%p", cookie); + trace_fscache_enable(cookie); + wait_on_bit_lock(&cookie->flags, FSCACHE_COOKIE_ENABLEMENT_LOCK, TASK_UNINTERRUPTIBLE); + fscache_update_aux(cookie, aux_data); + if (test_bit(FSCACHE_COOKIE_ENABLED, &cookie->flags)) goto out_unlock; if (can_enable && !can_enable(data)) { /* The netfs decided it didn't want to enable after all */ - } else if (cookie->def->type != FSCACHE_COOKIE_TYPE_INDEX) { + } else if (cookie->type != FSCACHE_COOKIE_TYPE_INDEX) { /* Wait for outstanding disablement to complete */ __fscache_wait_on_invalidate(cookie); - if (fscache_acquire_non_index_cookie(cookie) == 0) + if (fscache_acquire_non_index_cookie(cookie, object_size) == 0) set_bit(FSCACHE_COOKIE_ENABLED, &cookie->flags); } else { set_bit(FSCACHE_COOKIE_ENABLED, &cookie->flags); @@ -188,11 +405,11 @@ EXPORT_SYMBOL(__fscache_enable_cookie); * - this must make sure the index chain is instantiated and instantiate the * object representation too */ -static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie) +static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie, + loff_t object_size) { struct fscache_object *object; struct fscache_cache *cache; - uint64_t i_size; int ret; _enter(""); @@ -231,9 +448,6 @@ static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie) return ret; } - /* pass on how big the object we're caching is supposed to be */ - cookie->def->get_attr(cookie->netfs_data, &i_size); - spin_lock(&cookie->lock); if (hlist_empty(&cookie->backing_objects)) { spin_unlock(&cookie->lock); @@ -243,7 +457,7 @@ static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie) object = hlist_entry(cookie->backing_objects.first, struct fscache_object, cookie_link); - fscache_set_store_limit(object, i_size); + fscache_set_store_limit(object, object_size); /* initiate the process of looking up all the objects in the chain * (done by fscache_initialise_object()) */ @@ -318,7 +532,7 @@ static int fscache_alloc_object(struct fscache_cache *cache, * attached to the cookie */ if (fscache_attach_object(cookie, object) < 0) { fscache_stat(&fscache_n_cop_put_object); - cache->ops->put_object(object); + cache->ops->put_object(object, fscache_obj_put_attach_fail); fscache_stat_d(&fscache_n_cop_put_object); } @@ -338,7 +552,7 @@ object_already_extant: error_put: fscache_stat(&fscache_n_cop_put_object); - cache->ops->put_object(object); + cache->ops->put_object(object, fscache_obj_put_alloc_fail); fscache_stat_d(&fscache_n_cop_put_object); error: _leave(" = %d", ret); @@ -398,7 +612,7 @@ static int fscache_attach_object(struct fscache_cookie *cookie, /* attach to the cookie */ object->cookie = cookie; - atomic_inc(&cookie->usage); + fscache_cookie_get(cookie, fscache_cookie_get_attach_object); hlist_add_head(&object->cookie_link, &cookie->backing_objects); fscache_objlist_add(object); @@ -426,10 +640,7 @@ void __fscache_invalidate(struct fscache_cookie *cookie) * there, and if it's doing that, it may as well just retire the * cookie. */ - ASSERTCMP(cookie->def->type, ==, FSCACHE_COOKIE_TYPE_DATAFILE); - - /* We will be updating the cookie too. */ - BUG_ON(!cookie->def->get_aux); + ASSERTCMP(cookie->type, ==, FSCACHE_COOKIE_TYPE_DATAFILE); /* If there's an object, we tell the object state machine to handle the * invalidation on our behalf, otherwise there's nothing to do. @@ -473,7 +684,7 @@ EXPORT_SYMBOL(__fscache_wait_on_invalidate); /* * update the index entries backing a cookie */ -void __fscache_update_cookie(struct fscache_cookie *cookie) +void __fscache_update_cookie(struct fscache_cookie *cookie, const void *aux_data) { struct fscache_object *object; @@ -487,10 +698,10 @@ void __fscache_update_cookie(struct fscache_cookie *cookie) _enter("{%s}", cookie->def->name); - BUG_ON(!cookie->def->get_aux); - spin_lock(&cookie->lock); + fscache_update_aux(cookie, aux_data); + if (fscache_cookie_enabled(cookie)) { /* update the index entry on disk in each cache backing this * cookie. @@ -509,13 +720,17 @@ EXPORT_SYMBOL(__fscache_update_cookie); /* * Disable a cookie to stop it from accepting new requests from the netfs. */ -void __fscache_disable_cookie(struct fscache_cookie *cookie, bool invalidate) +void __fscache_disable_cookie(struct fscache_cookie *cookie, + const void *aux_data, + bool invalidate) { struct fscache_object *object; bool awaken = false; _enter("%p,%u", cookie, invalidate); + trace_fscache_disable(cookie); + ASSERTCMP(atomic_read(&cookie->n_active), >, 0); if (atomic_read(&cookie->n_children) != 0) { @@ -526,6 +741,9 @@ void __fscache_disable_cookie(struct fscache_cookie *cookie, bool invalidate) wait_on_bit_lock(&cookie->flags, FSCACHE_COOKIE_ENABLEMENT_LOCK, TASK_UNINTERRUPTIBLE); + + fscache_update_aux(cookie, aux_data); + if (!test_and_clear_bit(FSCACHE_COOKIE_ENABLED, &cookie->flags)) goto out_unlock_enable; @@ -563,7 +781,7 @@ void __fscache_disable_cookie(struct fscache_cookie *cookie, bool invalidate) } /* Make sure any pending writes are cancelled. */ - if (cookie->def->type != FSCACHE_COOKIE_TYPE_INDEX) + if (cookie->type != FSCACHE_COOKIE_TYPE_INDEX) fscache_invalidate_writes(cookie); /* Reset the cookie state if it wasn't relinquished */ @@ -585,7 +803,9 @@ EXPORT_SYMBOL(__fscache_disable_cookie); * - all dependents of this cookie must have already been unregistered * (indices/files/pages) */ -void __fscache_relinquish_cookie(struct fscache_cookie *cookie, bool retire) +void __fscache_relinquish_cookie(struct fscache_cookie *cookie, + const void *aux_data, + bool retire) { fscache_stat(&fscache_n_relinquishes); if (retire) @@ -601,15 +821,18 @@ void __fscache_relinquish_cookie(struct fscache_cookie *cookie, bool retire) cookie, cookie->def->name, cookie->netfs_data, atomic_read(&cookie->n_active), retire); + trace_fscache_relinquish(cookie, retire); + /* No further netfs-accessing operations on this cookie permitted */ - set_bit(FSCACHE_COOKIE_RELINQUISHED, &cookie->flags); + if (test_and_set_bit(FSCACHE_COOKIE_RELINQUISHED, &cookie->flags)) + BUG(); - __fscache_disable_cookie(cookie, retire); + __fscache_disable_cookie(cookie, aux_data, retire); /* Clear pointers back to the netfs */ cookie->netfs_data = NULL; cookie->def = NULL; - BUG_ON(cookie->stores.rnode); + BUG_ON(!radix_tree_empty(&cookie->stores)); if (cookie->parent) { ASSERTCMP(atomic_read(&cookie->parent->usage), >, 0); @@ -619,35 +842,54 @@ void __fscache_relinquish_cookie(struct fscache_cookie *cookie, bool retire) /* Dispose of the netfs's link to the cookie */ ASSERTCMP(atomic_read(&cookie->usage), >, 0); - fscache_cookie_put(cookie); + fscache_cookie_put(cookie, fscache_cookie_put_relinquish); _leave(""); } EXPORT_SYMBOL(__fscache_relinquish_cookie); /* - * destroy a cookie + * Remove a cookie from the hash table. */ -void __fscache_cookie_put(struct fscache_cookie *cookie) +static void fscache_unhash_cookie(struct fscache_cookie *cookie) +{ + struct hlist_bl_head *h; + unsigned int bucket; + + bucket = cookie->key_hash & (ARRAY_SIZE(fscache_cookie_hash) - 1); + h = &fscache_cookie_hash[bucket]; + + hlist_bl_lock(h); + hlist_bl_del(&cookie->hash_link); + hlist_bl_unlock(h); +} + +/* + * Drop a reference to a cookie. + */ +void fscache_cookie_put(struct fscache_cookie *cookie, + enum fscache_cookie_trace where) { struct fscache_cookie *parent; + int usage; _enter("%p", cookie); - for (;;) { - _debug("FREE COOKIE %p", cookie); - parent = cookie->parent; - BUG_ON(!hlist_empty(&cookie->backing_objects)); - kmem_cache_free(fscache_cookie_jar, cookie); + do { + usage = atomic_dec_return(&cookie->usage); + trace_fscache_cookie(cookie, where, usage); - if (!parent) - break; + if (usage > 0) + return; + BUG_ON(usage < 0); + + parent = cookie->parent; + fscache_unhash_cookie(cookie); + fscache_free_cookie(cookie); cookie = parent; - BUG_ON(atomic_read(&cookie->usage) <= 0); - if (!atomic_dec_and_test(&cookie->usage)) - break; - } + where = fscache_cookie_put_parent; + } while (cookie); _leave(""); } @@ -657,7 +899,8 @@ void __fscache_cookie_put(struct fscache_cookie *cookie) * * NOTE: it only serves no-index type */ -int __fscache_check_consistency(struct fscache_cookie *cookie) +int __fscache_check_consistency(struct fscache_cookie *cookie, + const void *aux_data) { struct fscache_operation *op; struct fscache_object *object; @@ -666,7 +909,7 @@ int __fscache_check_consistency(struct fscache_cookie *cookie) _enter("%p,", cookie); - ASSERTCMP(cookie->def->type, ==, FSCACHE_COOKIE_TYPE_DATAFILE); + ASSERTCMP(cookie->type, ==, FSCACHE_COOKIE_TYPE_DATAFILE); if (fscache_wait_for_deferred_lookup(cookie) < 0) return -ERESTARTSYS; @@ -678,13 +921,16 @@ int __fscache_check_consistency(struct fscache_cookie *cookie) if (!op) return -ENOMEM; - fscache_operation_init(op, NULL, NULL, NULL); + fscache_operation_init(cookie, op, NULL, NULL, NULL); op->flags = FSCACHE_OP_MYTHREAD | (1 << FSCACHE_OP_WAITING) | (1 << FSCACHE_OP_UNUSE_COOKIE); + trace_fscache_page_op(cookie, NULL, op, fscache_page_op_check_consistency); spin_lock(&cookie->lock); + fscache_update_aux(cookie, aux_data); + if (!fscache_cookie_enabled(cookie) || hlist_empty(&cookie->backing_objects)) goto inconsistent; diff --git a/fs/fscache/fsdef.c b/fs/fscache/fsdef.c index 5a117df2a9ef..aa46e48d8c75 100644 --- a/fs/fscache/fsdef.c +++ b/fs/fscache/fsdef.c @@ -13,16 +13,11 @@ #include <linux/module.h> #include "internal.h" -static uint16_t fscache_fsdef_netfs_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax); - -static uint16_t fscache_fsdef_netfs_get_aux(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax); - static enum fscache_checkaux fscache_fsdef_netfs_check_aux(void *cookie_netfs_data, const void *data, - uint16_t datalen); + uint16_t datalen, + loff_t object_size); /* * The root index is owned by FS-Cache itself. @@ -60,6 +55,7 @@ struct fscache_cookie fscache_fsdef_index = { .backing_objects = HLIST_HEAD_INIT, .def = &fscache_fsdef_index_def, .flags = 1 << FSCACHE_COOKIE_ENABLED, + .type = FSCACHE_COOKIE_TYPE_INDEX, }; EXPORT_SYMBOL(fscache_fsdef_index); @@ -71,59 +67,18 @@ EXPORT_SYMBOL(fscache_fsdef_index); struct fscache_cookie_def fscache_fsdef_netfs_def = { .name = "FSDEF.netfs", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = fscache_fsdef_netfs_get_key, - .get_aux = fscache_fsdef_netfs_get_aux, .check_aux = fscache_fsdef_netfs_check_aux, }; /* - * get the key data for an FSDEF index record - this is the name of the netfs - * for which this entry is created - */ -static uint16_t fscache_fsdef_netfs_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct fscache_netfs *netfs = cookie_netfs_data; - unsigned klen; - - _enter("{%s.%u},", netfs->name, netfs->version); - - klen = strlen(netfs->name); - if (klen > bufmax) - return 0; - - memcpy(buffer, netfs->name, klen); - return klen; -} - -/* - * get the auxiliary data for an FSDEF index record - this is the index - * structure version number of the netfs for which this version is created - */ -static uint16_t fscache_fsdef_netfs_get_aux(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct fscache_netfs *netfs = cookie_netfs_data; - unsigned dlen; - - _enter("{%s.%u},", netfs->name, netfs->version); - - dlen = sizeof(uint32_t); - if (dlen > bufmax) - return 0; - - memcpy(buffer, &netfs->version, dlen); - return dlen; -} - -/* * check that the index structure version number stored in the auxiliary data * matches the one the netfs gave us */ static enum fscache_checkaux fscache_fsdef_netfs_check_aux( void *cookie_netfs_data, const void *data, - uint16_t datalen) + uint16_t datalen, + loff_t object_size) { struct fscache_netfs *netfs = cookie_netfs_data; uint32_t version; diff --git a/fs/fscache/histogram.c b/fs/fscache/histogram.c index 15a3d042247e..9a13e9e15b69 100644 --- a/fs/fscache/histogram.c +++ b/fs/fscache/histogram.c @@ -83,24 +83,9 @@ static void fscache_histogram_stop(struct seq_file *m, void *v) { } -static const struct seq_operations fscache_histogram_ops = { +const struct seq_operations fscache_histogram_ops = { .start = fscache_histogram_start, .stop = fscache_histogram_stop, .next = fscache_histogram_next, .show = fscache_histogram_show, }; - -/* - * open "/proc/fs/fscache/histogram" to provide latency data - */ -static int fscache_histogram_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &fscache_histogram_ops); -} - -const struct file_operations fscache_histogram_fops = { - .open = fscache_histogram_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; diff --git a/fs/fscache/internal.h b/fs/fscache/internal.h index 0ff4b49a0037..f83328a7f048 100644 --- a/fs/fscache/internal.h +++ b/fs/fscache/internal.h @@ -29,7 +29,9 @@ #define pr_fmt(fmt) "FS-Cache: " fmt #include <linux/fscache-cache.h> +#include <trace/events/fscache.h> #include <linux/sched.h> +#include <linux/seq_file.h> #define FSCACHE_MIN_THREADS 4 #define FSCACHE_MAX_THREADS 32 @@ -48,8 +50,16 @@ extern struct fscache_cache *fscache_select_cache_for_object( */ extern struct kmem_cache *fscache_cookie_jar; +extern void fscache_free_cookie(struct fscache_cookie *); extern void fscache_cookie_init_once(void *); -extern void __fscache_cookie_put(struct fscache_cookie *); +extern struct fscache_cookie *fscache_alloc_cookie(struct fscache_cookie *, + const struct fscache_cookie_def *, + const void *, size_t, + const void *, size_t, + void *, loff_t); +extern struct fscache_cookie *fscache_hash_cookie(struct fscache_cookie *); +extern void fscache_cookie_put(struct fscache_cookie *, + enum fscache_cookie_trace); /* * fsdef.c @@ -75,7 +85,7 @@ static inline void fscache_hist(atomic_t histogram[], unsigned long start_jif) atomic_inc(&histogram[jif]); } -extern const struct file_operations fscache_histogram_fops; +extern const struct seq_operations fscache_histogram_ops; #else #define fscache_hist(hist, start_jif) do {} while (0) @@ -285,7 +295,7 @@ static inline void fscache_stat_d(atomic_t *stat) #define __fscache_stat(stat) (stat) -extern const struct file_operations fscache_stats_fops; +int fscache_stats_show(struct seq_file *m, void *v); #else #define __fscache_stat(stat) (NULL) @@ -311,14 +321,12 @@ static inline void fscache_raise_event(struct fscache_object *object, fscache_enqueue_object(object); } -/* - * drop a reference to a cookie - */ -static inline void fscache_cookie_put(struct fscache_cookie *cookie) +static inline void fscache_cookie_get(struct fscache_cookie *cookie, + enum fscache_cookie_trace where) { - BUG_ON(atomic_read(&cookie->usage) <= 0); - if (atomic_dec_and_test(&cookie->usage)) - __fscache_cookie_put(cookie); + int usage = atomic_inc_return(&cookie->usage); + + trace_fscache_cookie(cookie, where, usage); } /* @@ -342,6 +350,27 @@ void fscache_put_context(struct fscache_cookie *cookie, void *context) cookie->def->put_context(cookie->netfs_data, context); } +/* + * Update the auxiliary data on a cookie. + */ +static inline +void fscache_update_aux(struct fscache_cookie *cookie, const void *aux_data) +{ + void *p; + + if (!aux_data) + return; + if (cookie->aux_len <= sizeof(cookie->inline_aux)) + p = cookie->inline_aux; + else + p = cookie->aux; + + if (memcmp(p, aux_data, cookie->aux_len) != 0) { + memcpy(p, aux_data, cookie->aux_len); + set_bit(FSCACHE_COOKIE_AUX_UPDATED, &cookie->flags); + } +} + /*****************************************************************************/ /* * debug tracing diff --git a/fs/fscache/main.c b/fs/fscache/main.c index 249968dcbf5c..7dce110bf17d 100644 --- a/fs/fscache/main.c +++ b/fs/fscache/main.c @@ -16,6 +16,7 @@ #include <linux/completion.h> #include <linux/slab.h> #include <linux/seq_file.h> +#define CREATE_TRACE_POINTS #include "internal.h" MODULE_DESCRIPTION("FS Cache Manager"); diff --git a/fs/fscache/netfs.c b/fs/fscache/netfs.c index a8aa00be4444..c2f605483cc5 100644 --- a/fs/fscache/netfs.c +++ b/fs/fscache/netfs.c @@ -14,69 +14,51 @@ #include <linux/slab.h> #include "internal.h" -static LIST_HEAD(fscache_netfs_list); - /* * register a network filesystem for caching */ int __fscache_register_netfs(struct fscache_netfs *netfs) { - struct fscache_netfs *ptr; - struct fscache_cookie *cookie; - int ret; + struct fscache_cookie *candidate, *cookie; _enter("{%s}", netfs->name); - INIT_LIST_HEAD(&netfs->link); - /* allocate a cookie for the primary index */ - cookie = kmem_cache_zalloc(fscache_cookie_jar, GFP_KERNEL); - - if (!cookie) { + candidate = fscache_alloc_cookie(&fscache_fsdef_index, + &fscache_fsdef_netfs_def, + netfs->name, strlen(netfs->name), + &netfs->version, sizeof(netfs->version), + netfs, 0); + if (!candidate) { _leave(" = -ENOMEM"); return -ENOMEM; } - /* initialise the primary index cookie */ - atomic_set(&cookie->usage, 1); - atomic_set(&cookie->n_children, 0); - atomic_set(&cookie->n_active, 1); - - cookie->def = &fscache_fsdef_netfs_def; - cookie->parent = &fscache_fsdef_index; - cookie->netfs_data = netfs; - cookie->flags = 1 << FSCACHE_COOKIE_ENABLED; - - spin_lock_init(&cookie->lock); - spin_lock_init(&cookie->stores_lock); - INIT_HLIST_HEAD(&cookie->backing_objects); + candidate->flags = 1 << FSCACHE_COOKIE_ENABLED; /* check the netfs type is not already present */ - down_write(&fscache_addremove_sem); - - ret = -EEXIST; - list_for_each_entry(ptr, &fscache_netfs_list, link) { - if (strcmp(ptr->name, netfs->name) == 0) - goto already_registered; + cookie = fscache_hash_cookie(candidate); + if (!cookie) + goto already_registered; + if (cookie != candidate) { + trace_fscache_cookie(candidate, fscache_cookie_discard, 1); + fscache_free_cookie(candidate); } - atomic_inc(&cookie->parent->usage); + fscache_cookie_get(cookie->parent, fscache_cookie_get_register_netfs); atomic_inc(&cookie->parent->n_children); netfs->primary_index = cookie; - list_add(&netfs->link, &fscache_netfs_list); - ret = 0; pr_notice("Netfs '%s' registered for caching\n", netfs->name); + trace_fscache_netfs(netfs); + _leave(" = 0"); + return 0; already_registered: - up_write(&fscache_addremove_sem); - - if (ret < 0) - kmem_cache_free(fscache_cookie_jar, cookie); - - _leave(" = %d", ret); - return ret; + fscache_cookie_put(candidate, fscache_cookie_put_dup_netfs); + _leave(" = -EEXIST"); + return -EEXIST; } EXPORT_SYMBOL(__fscache_register_netfs); @@ -88,15 +70,8 @@ void __fscache_unregister_netfs(struct fscache_netfs *netfs) { _enter("{%s.%u}", netfs->name, netfs->version); - down_write(&fscache_addremove_sem); - - list_del(&netfs->link); - fscache_relinquish_cookie(netfs->primary_index, 0); - - up_write(&fscache_addremove_sem); - - pr_notice("Netfs '%s' unregistered from caching\n", - netfs->name); + fscache_relinquish_cookie(netfs->primary_index, NULL, false); + pr_notice("Netfs '%s' unregistered from caching\n", netfs->name); _leave(""); } diff --git a/fs/fscache/object-list.c b/fs/fscache/object-list.c index 0438d4cd91ef..43e6e28c164f 100644 --- a/fs/fscache/object-list.c +++ b/fs/fscache/object-list.c @@ -36,8 +36,6 @@ struct fscache_objlist_data { #define FSCACHE_OBJLIST_CONFIG_NOEVENTS 0x00000800 /* show objects without no events */ #define FSCACHE_OBJLIST_CONFIG_WORK 0x00001000 /* show objects with work */ #define FSCACHE_OBJLIST_CONFIG_NOWORK 0x00002000 /* show objects without work */ - - u8 buf[512]; /* key and aux data buffer */ }; /* @@ -170,7 +168,7 @@ static int fscache_objlist_show(struct seq_file *m, void *v) struct fscache_cookie *cookie; unsigned long config = data->config; char _type[3], *type; - u8 *buf = data->buf, *p; + u8 *p; if ((unsigned long) v == 1) { seq_puts(m, "OBJECT PARENT STAT CHLDN OPS OOP IPR EX READS" @@ -254,7 +252,7 @@ static int fscache_objlist_show(struct seq_file *m, void *v) if (fscache_use_cookie(obj)) { uint16_t keylen = 0, auxlen = 0; - switch (cookie->def->type) { + switch (cookie->type) { case 0: type = "IX"; break; @@ -263,7 +261,7 @@ static int fscache_objlist_show(struct seq_file *m, void *v) break; default: snprintf(_type, sizeof(_type), "%02u", - cookie->def->type); + cookie->type); type = _type; break; } @@ -274,30 +272,30 @@ static int fscache_objlist_show(struct seq_file *m, void *v) cookie->flags, cookie->netfs_data); - if (cookie->def->get_key && - config & FSCACHE_OBJLIST_CONFIG_KEY) - keylen = cookie->def->get_key(cookie->netfs_data, - buf, 400); + if (config & FSCACHE_OBJLIST_CONFIG_KEY) + keylen = cookie->key_len; - if (cookie->def->get_aux && - config & FSCACHE_OBJLIST_CONFIG_AUX) - auxlen = cookie->def->get_aux(cookie->netfs_data, - buf + keylen, 512 - keylen); - fscache_unuse_cookie(obj); + if (config & FSCACHE_OBJLIST_CONFIG_AUX) + auxlen = cookie->aux_len; if (keylen > 0 || auxlen > 0) { seq_puts(m, " "); - for (p = buf; keylen > 0; keylen--) + p = keylen <= sizeof(cookie->inline_key) ? + cookie->inline_key : cookie->key; + for (; keylen > 0; keylen--) seq_printf(m, "%02x", *p++); if (auxlen > 0) { if (config & FSCACHE_OBJLIST_CONFIG_KEY) seq_puts(m, ", "); + p = auxlen <= sizeof(cookie->inline_aux) ? + cookie->inline_aux : cookie->aux; for (; auxlen > 0; auxlen--) seq_printf(m, "%02x", *p++); } } seq_puts(m, "\n"); + fscache_unuse_cookie(obj); } else { seq_puts(m, "<no_netfs>\n"); } diff --git a/fs/fscache/object.c b/fs/fscache/object.c index 7a182c87f378..20e0d0a4dc8c 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -138,10 +138,13 @@ static const struct fscache_transition fscache_osm_run_oob[] = { { 0, NULL } }; -static int fscache_get_object(struct fscache_object *); -static void fscache_put_object(struct fscache_object *); +static int fscache_get_object(struct fscache_object *, + enum fscache_obj_ref_trace); +static void fscache_put_object(struct fscache_object *, + enum fscache_obj_ref_trace); static bool fscache_enqueue_dependents(struct fscache_object *, int); static void fscache_dequeue_object(struct fscache_object *); +static void fscache_update_aux_data(struct fscache_object *); /* * we need to notify the parent when an op completes that we had outstanding @@ -170,6 +173,7 @@ static void fscache_object_sm_dispatcher(struct fscache_object *object) const struct fscache_transition *t; const struct fscache_state *state, *new_state; unsigned long events, event_mask; + bool oob; int event = -1; ASSERT(object != NULL); @@ -188,6 +192,7 @@ restart_masked: if (events & object->oob_event_mask) { _debug("{OBJ%x} oob %lx", object->debug_id, events & object->oob_event_mask); + oob = true; for (t = object->oob_table; t->events; t++) { if (events & t->events) { state = t->transit_to; @@ -199,6 +204,7 @@ restart_masked: } } } + oob = false; /* Wait states are just transition tables */ if (!state->work) { @@ -207,6 +213,8 @@ restart_masked: if (events & t->events) { new_state = t->transit_to; event = fls(events & t->events) - 1; + trace_fscache_osm(object, state, + true, false, event); clear_bit(event, &object->events); _debug("{OBJ%x} ev %d: %s -> %s", object->debug_id, event, @@ -226,6 +234,7 @@ restart_masked: execute_work_state: _debug("{OBJ%x} exec %s", object->debug_id, state->name); + trace_fscache_osm(object, state, false, oob, event); new_state = state->work(object, event); event = -1; if (new_state == NO_TRANSIT) { @@ -279,7 +288,7 @@ static void fscache_object_work_func(struct work_struct *work) start = jiffies; fscache_object_sm_dispatcher(object); fscache_hist(fscache_objs_histogram, start); - fscache_put_object(object); + fscache_put_object(object, fscache_obj_put_work); } /** @@ -397,7 +406,7 @@ static const struct fscache_state *fscache_initialise_object(struct fscache_obje fscache_stat(&fscache_n_cop_grab_object); success = false; if (fscache_object_is_live(parent) && - object->cache->ops->grab_object(object)) { + object->cache->ops->grab_object(object, fscache_obj_get_add_to_deps)) { list_add(&object->dep_link, &parent->dependents); success = true; } @@ -703,6 +712,11 @@ static const struct fscache_state *fscache_drop_object(struct fscache_object *ob ASSERT(cookie != NULL); ASSERT(!hlist_unhashed(&object->cookie_link)); + if (test_bit(FSCACHE_COOKIE_AUX_UPDATED, &cookie->flags)) { + _debug("final update"); + fscache_update_aux_data(object); + } + /* Make sure the cookie no longer points here and that the netfs isn't * waiting for us. */ @@ -745,7 +759,7 @@ static const struct fscache_state *fscache_drop_object(struct fscache_object *ob } /* this just shifts the object release to the work processor */ - fscache_put_object(object); + fscache_put_object(object, fscache_obj_put_drop_obj); fscache_stat(&fscache_n_object_dead); _leave(""); @@ -755,12 +769,13 @@ static const struct fscache_state *fscache_drop_object(struct fscache_object *ob /* * get a ref on an object */ -static int fscache_get_object(struct fscache_object *object) +static int fscache_get_object(struct fscache_object *object, + enum fscache_obj_ref_trace why) { int ret; fscache_stat(&fscache_n_cop_grab_object); - ret = object->cache->ops->grab_object(object) ? 0 : -EAGAIN; + ret = object->cache->ops->grab_object(object, why) ? 0 : -EAGAIN; fscache_stat_d(&fscache_n_cop_grab_object); return ret; } @@ -768,10 +783,11 @@ static int fscache_get_object(struct fscache_object *object) /* * Discard a ref on an object */ -static void fscache_put_object(struct fscache_object *object) +static void fscache_put_object(struct fscache_object *object, + enum fscache_obj_ref_trace why) { fscache_stat(&fscache_n_cop_put_object); - object->cache->ops->put_object(object); + object->cache->ops->put_object(object, why); fscache_stat_d(&fscache_n_cop_put_object); } @@ -786,7 +802,7 @@ void fscache_object_destroy(struct fscache_object *object) fscache_objlist_remove(object); /* We can get rid of the cookie now */ - fscache_cookie_put(object->cookie); + fscache_cookie_put(object->cookie, fscache_cookie_put_object); object->cookie = NULL; } EXPORT_SYMBOL(fscache_object_destroy); @@ -798,7 +814,7 @@ void fscache_enqueue_object(struct fscache_object *object) { _enter("{OBJ%x}", object->debug_id); - if (fscache_get_object(object) >= 0) { + if (fscache_get_object(object, fscache_obj_get_queue) >= 0) { wait_queue_head_t *cong_wq = &get_cpu_var(fscache_object_cong_wait); @@ -806,7 +822,7 @@ void fscache_enqueue_object(struct fscache_object *object) if (fscache_object_congested()) wake_up(cong_wq); } else - fscache_put_object(object); + fscache_put_object(object, fscache_obj_put_queue); put_cpu_var(fscache_object_cong_wait); } @@ -866,7 +882,7 @@ static bool fscache_enqueue_dependents(struct fscache_object *object, int event) list_del_init(&dep->dep_link); fscache_raise_event(dep, event); - fscache_put_object(dep); + fscache_put_object(dep, fscache_obj_put_enq_dep); if (!list_empty(&object->dependents) && need_resched()) { ret = false; @@ -906,7 +922,8 @@ static void fscache_dequeue_object(struct fscache_object *object) * and creation). */ enum fscache_checkaux fscache_check_aux(struct fscache_object *object, - const void *data, uint16_t datalen) + const void *data, uint16_t datalen, + loff_t object_size) { enum fscache_checkaux result; @@ -916,7 +933,7 @@ enum fscache_checkaux fscache_check_aux(struct fscache_object *object, } result = object->cookie->def->check_aux(object->cookie->netfs_data, - data, datalen); + data, datalen, object_size); switch (result) { /* entry okay as is */ case FSCACHE_CHECKAUX_OKAY: @@ -956,7 +973,7 @@ static const struct fscache_state *_fscache_invalidate_object(struct fscache_obj * retire the object instead. */ if (!fscache_use_cookie(object)) { - ASSERT(object->cookie->stores.rnode == NULL); + ASSERT(radix_tree_empty(&object->cookie->stores)); set_bit(FSCACHE_OBJECT_RETIRED, &object->flags); _leave(" [no cookie]"); return transit_to(KILL_OBJECT); @@ -972,11 +989,12 @@ static const struct fscache_state *_fscache_invalidate_object(struct fscache_obj if (!op) goto nomem; - fscache_operation_init(op, object->cache->ops->invalidate_object, + fscache_operation_init(cookie, op, object->cache->ops->invalidate_object, NULL, NULL); op->flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_EXCLUSIVE) | (1 << FSCACHE_OP_UNUSE_COOKIE); + trace_fscache_page_op(cookie, NULL, op, fscache_page_op_invalidate); spin_lock(&cookie->lock); if (fscache_submit_exclusive_op(object, op) < 0) @@ -1026,6 +1044,17 @@ static const struct fscache_state *fscache_invalidate_object(struct fscache_obje } /* + * Update auxiliary data. + */ +static void fscache_update_aux_data(struct fscache_object *object) +{ + fscache_stat(&fscache_n_updates_run); + fscache_stat(&fscache_n_cop_update_object); + object->cache->ops->update_object(object); + fscache_stat_d(&fscache_n_cop_update_object); +} + +/* * Asynchronously update an object. */ static const struct fscache_state *fscache_update_object(struct fscache_object *object, @@ -1033,10 +1062,7 @@ static const struct fscache_state *fscache_update_object(struct fscache_object * { _enter("{OBJ%x},%d", object->debug_id, event); - fscache_stat(&fscache_n_updates_run); - fscache_stat(&fscache_n_cop_update_object); - object->cache->ops->update_object(object); - fscache_stat_d(&fscache_n_cop_update_object); + fscache_update_aux_data(object); _leave(""); return transit_to(WAIT_FOR_CMD); diff --git a/fs/fscache/operation.c b/fs/fscache/operation.c index de67745e1cd7..e30c5975ea58 100644 --- a/fs/fscache/operation.c +++ b/fs/fscache/operation.c @@ -32,7 +32,8 @@ static void fscache_operation_dummy_cancel(struct fscache_operation *op) * Do basic initialisation of an operation. The caller must still set flags, * object and processor if needed. */ -void fscache_operation_init(struct fscache_operation *op, +void fscache_operation_init(struct fscache_cookie *cookie, + struct fscache_operation *op, fscache_operation_processor_t processor, fscache_operation_cancel_t cancel, fscache_operation_release_t release) @@ -46,6 +47,7 @@ void fscache_operation_init(struct fscache_operation *op, op->release = release; INIT_LIST_HEAD(&op->pend_link); fscache_stat(&fscache_n_op_initialised); + trace_fscache_op(cookie, op, fscache_op_init); } EXPORT_SYMBOL(fscache_operation_init); @@ -59,6 +61,8 @@ EXPORT_SYMBOL(fscache_operation_init); */ void fscache_enqueue_operation(struct fscache_operation *op) { + struct fscache_cookie *cookie = op->object->cookie; + _enter("{OBJ%x OP%x,%u}", op->object->debug_id, op->debug_id, atomic_read(&op->usage)); @@ -71,12 +75,14 @@ void fscache_enqueue_operation(struct fscache_operation *op) fscache_stat(&fscache_n_op_enqueue); switch (op->flags & FSCACHE_OP_TYPE) { case FSCACHE_OP_ASYNC: + trace_fscache_op(cookie, op, fscache_op_enqueue_async); _debug("queue async"); atomic_inc(&op->usage); if (!queue_work(fscache_op_wq, &op->work)) fscache_put_operation(op); break; case FSCACHE_OP_MYTHREAD: + trace_fscache_op(cookie, op, fscache_op_enqueue_mythread); _debug("queue for caller's attention"); break; default: @@ -101,6 +107,8 @@ static void fscache_run_op(struct fscache_object *object, wake_up_bit(&op->flags, FSCACHE_OP_WAITING); if (op->processor) fscache_enqueue_operation(op); + else + trace_fscache_op(object->cookie, op, fscache_op_run); fscache_stat(&fscache_n_op_run); } @@ -155,6 +163,8 @@ int fscache_submit_exclusive_op(struct fscache_object *object, _enter("{OBJ%x OP%x},", object->debug_id, op->debug_id); + trace_fscache_op(object->cookie, op, fscache_op_submit_ex); + ASSERTCMP(op->state, ==, FSCACHE_OP_ST_INITIALISED); ASSERTCMP(atomic_read(&op->usage), >, 0); @@ -240,6 +250,8 @@ int fscache_submit_op(struct fscache_object *object, _enter("{OBJ%x OP%x},{%u}", object->debug_id, op->debug_id, atomic_read(&op->usage)); + trace_fscache_op(object->cookie, op, fscache_op_submit); + ASSERTCMP(op->state, ==, FSCACHE_OP_ST_INITIALISED); ASSERTCMP(atomic_read(&op->usage), >, 0); @@ -357,6 +369,8 @@ int fscache_cancel_op(struct fscache_operation *op, _enter("OBJ%x OP%x}", op->object->debug_id, op->debug_id); + trace_fscache_op(object->cookie, op, fscache_op_cancel); + ASSERTCMP(op->state, >=, FSCACHE_OP_ST_PENDING); ASSERTCMP(op->state, !=, FSCACHE_OP_ST_CANCELLED); ASSERTCMP(atomic_read(&op->usage), >, 0); @@ -419,6 +433,8 @@ void fscache_cancel_all_ops(struct fscache_object *object) fscache_stat(&fscache_n_op_cancelled); list_del_init(&op->pend_link); + trace_fscache_op(object->cookie, op, fscache_op_cancel_all); + ASSERTCMP(op->state, ==, FSCACHE_OP_ST_PENDING); op->cancel(op); op->state = FSCACHE_OP_ST_CANCELLED; @@ -454,9 +470,11 @@ void fscache_op_complete(struct fscache_operation *op, bool cancelled) spin_lock(&object->lock); if (!cancelled) { + trace_fscache_op(object->cookie, op, fscache_op_completed); op->state = FSCACHE_OP_ST_COMPLETE; } else { op->cancel(op); + trace_fscache_op(object->cookie, op, fscache_op_cancelled); op->state = FSCACHE_OP_ST_CANCELLED; } @@ -488,6 +506,8 @@ void fscache_put_operation(struct fscache_operation *op) if (!atomic_dec_and_test(&op->usage)) return; + trace_fscache_op(op->object ? op->object->cookie : NULL, op, fscache_op_put); + _debug("PUT OP"); ASSERTIFCMP(op->state != FSCACHE_OP_ST_INITIALISED && op->state != FSCACHE_OP_ST_COMPLETE, @@ -563,6 +583,8 @@ void fscache_operation_gc(struct work_struct *work) spin_unlock(&cache->op_gc_list_lock); object = op->object; + trace_fscache_op(object->cookie, op, fscache_op_gc); + spin_lock(&object->lock); _debug("GC DEFERRED REL OBJ%x OP%x", @@ -601,6 +623,8 @@ void fscache_op_work_func(struct work_struct *work) _enter("{OBJ%x OP%x,%d}", op->object->debug_id, op->debug_id, atomic_read(&op->usage)); + trace_fscache_op(op->object->cookie, op, fscache_op_work); + ASSERT(op->processor != NULL); start = jiffies; op->processor(op); diff --git a/fs/fscache/page.c b/fs/fscache/page.c index 961029e04027..111349f67d98 100644 --- a/fs/fscache/page.c +++ b/fs/fscache/page.c @@ -27,6 +27,7 @@ bool __fscache_check_page_write(struct fscache_cookie *cookie, struct page *page rcu_read_lock(); val = radix_tree_lookup(&cookie->stores, page->index); rcu_read_unlock(); + trace_fscache_check_page(cookie, page, val, 0); return val != NULL; } @@ -39,6 +40,8 @@ void __fscache_wait_on_page_write(struct fscache_cookie *cookie, struct page *pa { wait_queue_head_t *wq = bit_waitqueue(&cookie->flags, 0); + trace_fscache_page(cookie, page, fscache_page_write_wait); + wait_event(*wq, !__fscache_check_page_write(cookie, page)); } EXPORT_SYMBOL(__fscache_wait_on_page_write); @@ -69,6 +72,8 @@ bool __fscache_maybe_release_page(struct fscache_cookie *cookie, _enter("%p,%p,%x", cookie, page, gfp); + trace_fscache_page(cookie, page, fscache_page_maybe_release); + try_again: rcu_read_lock(); val = radix_tree_lookup(&cookie->stores, page->index); @@ -101,6 +106,7 @@ try_again: } xpage = radix_tree_delete(&cookie->stores, page->index); + trace_fscache_page(cookie, page, fscache_page_radix_delete); spin_unlock(&cookie->stores_lock); if (xpage) { @@ -112,6 +118,7 @@ try_again: } wake_up_bit(&cookie->flags, 0); + trace_fscache_wake_cookie(cookie); if (xpage) put_page(xpage); __fscache_uncache_page(cookie, page); @@ -144,7 +151,7 @@ static void fscache_end_page_write(struct fscache_object *object, struct page *page) { struct fscache_cookie *cookie; - struct page *xpage = NULL; + struct page *xpage = NULL, *val; spin_lock(&object->lock); cookie = object->cookie; @@ -154,13 +161,24 @@ static void fscache_end_page_write(struct fscache_object *object, spin_lock(&cookie->stores_lock); radix_tree_tag_clear(&cookie->stores, page->index, FSCACHE_COOKIE_STORING_TAG); + trace_fscache_page(cookie, page, fscache_page_radix_clear_store); if (!radix_tree_tag_get(&cookie->stores, page->index, FSCACHE_COOKIE_PENDING_TAG)) { fscache_stat(&fscache_n_store_radix_deletes); xpage = radix_tree_delete(&cookie->stores, page->index); + trace_fscache_page(cookie, page, fscache_page_radix_delete); + trace_fscache_page(cookie, page, fscache_page_write_end); + + val = radix_tree_lookup(&cookie->stores, page->index); + trace_fscache_check_page(cookie, page, val, 1); + } else { + trace_fscache_page(cookie, page, fscache_page_write_end_pend); } spin_unlock(&cookie->stores_lock); wake_up_bit(&cookie->flags, 0); + trace_fscache_wake_cookie(cookie); + } else { + trace_fscache_page(cookie, page, fscache_page_write_end_noc); } spin_unlock(&object->lock); if (xpage) @@ -185,9 +203,11 @@ static void fscache_attr_changed_op(struct fscache_operation *op) fscache_stat_d(&fscache_n_cop_attr_changed); if (ret < 0) fscache_abort_object(object); + fscache_op_complete(op, ret < 0); + } else { + fscache_op_complete(op, true); } - fscache_op_complete(op, true); _leave(""); } @@ -213,7 +233,8 @@ int __fscache_attr_changed(struct fscache_cookie *cookie) return -ENOMEM; } - fscache_operation_init(op, fscache_attr_changed_op, NULL, NULL); + fscache_operation_init(cookie, op, fscache_attr_changed_op, NULL, NULL); + trace_fscache_page_op(cookie, NULL, op, fscache_page_op_attr_changed); op->flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_EXCLUSIVE) | (1 << FSCACHE_OP_UNUSE_COOKIE); @@ -297,7 +318,7 @@ static struct fscache_retrieval *fscache_alloc_retrieval( return NULL; } - fscache_operation_init(&op->op, NULL, + fscache_operation_init(cookie, &op->op, NULL, fscache_do_cancel_retrieval, fscache_release_retrieval_op); op->op.flags = FSCACHE_OP_MYTHREAD | @@ -368,6 +389,7 @@ int fscache_wait_for_operation_activation(struct fscache_object *object, fscache_stat(stat_op_waits); if (wait_on_bit(&op->flags, FSCACHE_OP_WAITING, TASK_INTERRUPTIBLE) != 0) { + trace_fscache_op(object->cookie, op, fscache_op_signal); ret = fscache_cancel_op(op, false); if (ret == 0) return -ERESTARTSYS; @@ -389,6 +411,7 @@ check_if_dead: if (unlikely(fscache_object_is_dying(object) || fscache_cache_is_broken(object))) { enum fscache_operation_state state = op->state; + trace_fscache_op(object->cookie, op, fscache_op_signal); fscache_cancel_op(op, true); if (stat_object_dead) fscache_stat(stat_object_dead); @@ -443,6 +466,7 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie, return -ENOMEM; } atomic_set(&op->n_pages, 1); + trace_fscache_page_op(cookie, page, &op->op, fscache_page_op_retr_one); spin_lock(&cookie->lock); @@ -571,6 +595,7 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie, if (!op) return -ENOMEM; atomic_set(&op->n_pages, *nr_pages); + trace_fscache_page_op(cookie, NULL, &op->op, fscache_page_op_retr_multi); spin_lock(&cookie->lock); @@ -682,6 +707,7 @@ int __fscache_alloc_page(struct fscache_cookie *cookie, if (!op) return -ENOMEM; atomic_set(&op->n_pages, 1); + trace_fscache_page_op(cookie, page, &op->op, fscache_page_op_alloc_one); spin_lock(&cookie->lock); @@ -776,15 +802,17 @@ static void fscache_write_op(struct fscache_operation *_op) _enter("{OP%x,%d}", op->op.debug_id, atomic_read(&op->op.usage)); +again: spin_lock(&object->lock); cookie = object->cookie; if (!fscache_object_is_active(object)) { - /* If we get here, then the on-disk cache object likely longer - * exists, so we should just cancel this write operation. + /* If we get here, then the on-disk cache object likely no + * longer exists, so we should just cancel this write + * operation. */ spin_unlock(&object->lock); - fscache_op_complete(&op->op, false); + fscache_op_complete(&op->op, true); _leave(" [inactive]"); return; } @@ -797,7 +825,7 @@ static void fscache_write_op(struct fscache_operation *_op) * cancel this write operation. */ spin_unlock(&object->lock); - fscache_op_complete(&op->op, false); + fscache_op_complete(&op->op, true); _leave(" [cancel] op{f=%lx s=%u} obj{s=%s f=%lx}", _op->flags, _op->state, object->state->short_name, object->flags); @@ -809,30 +837,33 @@ static void fscache_write_op(struct fscache_operation *_op) fscache_stat(&fscache_n_store_calls); /* find a page to store */ + results[0] = NULL; page = NULL; n = radix_tree_gang_lookup_tag(&cookie->stores, results, 0, 1, FSCACHE_COOKIE_PENDING_TAG); + trace_fscache_gang_lookup(cookie, &op->op, results, n, op->store_limit); if (n != 1) goto superseded; page = results[0]; _debug("gang %d [%lx]", n, page->index); - if (page->index >= op->store_limit) { - fscache_stat(&fscache_n_store_pages_over_limit); - goto superseded; - } radix_tree_tag_set(&cookie->stores, page->index, FSCACHE_COOKIE_STORING_TAG); radix_tree_tag_clear(&cookie->stores, page->index, FSCACHE_COOKIE_PENDING_TAG); + trace_fscache_page(cookie, page, fscache_page_radix_pend2store); spin_unlock(&cookie->stores_lock); spin_unlock(&object->lock); + if (page->index >= op->store_limit) + goto discard_page; + fscache_stat(&fscache_n_store_pages); fscache_stat(&fscache_n_cop_write_page); ret = object->cache->ops->write_page(op, page); fscache_stat_d(&fscache_n_cop_write_page); + trace_fscache_wrote_page(cookie, page, &op->op, ret); fscache_end_page_write(object, page); if (ret < 0) { fscache_abort_object(object); @@ -844,6 +875,12 @@ static void fscache_write_op(struct fscache_operation *_op) _leave(""); return; +discard_page: + fscache_stat(&fscache_n_store_pages_over_limit); + trace_fscache_wrote_page(cookie, page, &op->op, -ENOBUFS); + fscache_end_page_write(object, page); + goto again; + superseded: /* this writer is going away and there aren't any more things to * write */ @@ -851,7 +888,7 @@ superseded: spin_unlock(&cookie->stores_lock); clear_bit(FSCACHE_OBJECT_PENDING_WRITE, &object->flags); spin_unlock(&object->lock); - fscache_op_complete(&op->op, true); + fscache_op_complete(&op->op, false); _leave(""); } @@ -879,6 +916,8 @@ void fscache_invalidate_writes(struct fscache_cookie *cookie) for (i = n - 1; i >= 0; i--) { page = results[i]; radix_tree_delete(&cookie->stores, page->index); + trace_fscache_page(cookie, page, fscache_page_radix_delete); + trace_fscache_page(cookie, page, fscache_page_inval); } spin_unlock(&cookie->stores_lock); @@ -888,6 +927,7 @@ void fscache_invalidate_writes(struct fscache_cookie *cookie) } wake_up_bit(&cookie->flags, 0); + trace_fscache_wake_cookie(cookie); _leave(""); } @@ -923,6 +963,7 @@ void fscache_invalidate_writes(struct fscache_cookie *cookie) */ int __fscache_write_page(struct fscache_cookie *cookie, struct page *page, + loff_t object_size, gfp_t gfp) { struct fscache_storage *op; @@ -946,7 +987,7 @@ int __fscache_write_page(struct fscache_cookie *cookie, if (!op) goto nomem; - fscache_operation_init(&op->op, fscache_write_op, NULL, + fscache_operation_init(cookie, &op->op, fscache_write_op, NULL, fscache_release_write_op); op->op.flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_WAITING) | @@ -956,6 +997,8 @@ int __fscache_write_page(struct fscache_cookie *cookie, if (ret < 0) goto nomem_free; + trace_fscache_page_op(cookie, page, &op->op, fscache_page_op_write_one); + ret = -ENOBUFS; spin_lock(&cookie->lock); @@ -967,9 +1010,15 @@ int __fscache_write_page(struct fscache_cookie *cookie, if (test_bit(FSCACHE_IOERROR, &object->cache->flags)) goto nobufs; + trace_fscache_page(cookie, page, fscache_page_write); + /* add the page to the pending-storage radix tree on the backing * object */ spin_lock(&object->lock); + + if (object->store_limit_l != object_size) + fscache_set_store_limit(object, object_size); + spin_lock(&cookie->stores_lock); _debug("store limit %llx", (unsigned long long) object->store_limit); @@ -982,8 +1031,10 @@ int __fscache_write_page(struct fscache_cookie *cookie, goto nobufs_unlock_obj; } + trace_fscache_page(cookie, page, fscache_page_radix_insert); radix_tree_tag_set(&cookie->stores, page->index, FSCACHE_COOKIE_PENDING_TAG); + trace_fscache_page(cookie, page, fscache_page_radix_set_pend); get_page(page); /* we only want one writer at a time, but we do need to queue new @@ -1026,6 +1077,7 @@ already_pending: submit_failed: spin_lock(&cookie->stores_lock); radix_tree_delete(&cookie->stores, page->index); + trace_fscache_page(cookie, page, fscache_page_radix_delete); spin_unlock(&cookie->stores_lock); wake_cookie = __fscache_unuse_cookie(cookie); put_page(page); @@ -1072,6 +1124,8 @@ void __fscache_uncache_page(struct fscache_cookie *cookie, struct page *page) if (!PageFsCache(page)) goto done; + trace_fscache_page(cookie, page, fscache_page_uncache); + /* get the object */ spin_lock(&cookie->lock); @@ -1120,6 +1174,8 @@ void fscache_mark_page_cached(struct fscache_retrieval *op, struct page *page) atomic_inc(&fscache_n_marks); #endif + trace_fscache_page(cookie, page, fscache_page_cached); + _debug("- mark %p{%lx}", page, page->index); if (TestSetPageFsCache(page)) { static bool once_only; diff --git a/fs/fscache/proc.c b/fs/fscache/proc.c index 1d9e4951a597..49a8c90414bc 100644 --- a/fs/fscache/proc.c +++ b/fs/fscache/proc.c @@ -26,14 +26,14 @@ int __init fscache_proc_init(void) goto error_dir; #ifdef CONFIG_FSCACHE_STATS - if (!proc_create("fs/fscache/stats", S_IFREG | 0444, NULL, - &fscache_stats_fops)) + if (!proc_create_single("fs/fscache/stats", S_IFREG | 0444, NULL, + fscache_stats_show)) goto error_stats; #endif #ifdef CONFIG_FSCACHE_HISTOGRAM - if (!proc_create("fs/fscache/histogram", S_IFREG | 0444, NULL, - &fscache_histogram_fops)) + if (!proc_create_seq("fs/fscache/histogram", S_IFREG | 0444, NULL, + &fscache_histogram_ops)) goto error_histogram; #endif diff --git a/fs/fscache/stats.c b/fs/fscache/stats.c index 7ac6e839b065..00564a1dfd76 100644 --- a/fs/fscache/stats.c +++ b/fs/fscache/stats.c @@ -21,7 +21,6 @@ atomic_t fscache_n_op_pend; atomic_t fscache_n_op_run; atomic_t fscache_n_op_enqueue; -atomic_t fscache_n_op_requeue; atomic_t fscache_n_op_deferred_release; atomic_t fscache_n_op_initialised; atomic_t fscache_n_op_release; @@ -139,7 +138,7 @@ atomic_t fscache_n_cache_culled_objects; /* * display the general statistics */ -static int fscache_stats_show(struct seq_file *m, void *v) +int fscache_stats_show(struct seq_file *m, void *v) { seq_puts(m, "FS-Cache statistics\n"); @@ -285,18 +284,3 @@ static int fscache_stats_show(struct seq_file *m, void *v) atomic_read(&fscache_n_cache_culled_objects)); return 0; } - -/* - * open "/proc/fs/fscache/stats" allowing provision of a statistical summary - */ -static int fscache_stats_open(struct inode *inode, struct file *file) -{ - return single_open(file, fscache_stats_show, NULL); -} - -const struct file_operations fscache_stats_fops = { - .open = fscache_stats_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 624f18bbfd2b..ef309958e060 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -1080,6 +1080,9 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) sb->s_maxbytes = MAX_LFS_FILESIZE; sb->s_time_gran = 1; sb->s_export_op = &fuse_export_operations; + sb->s_iflags |= SB_I_IMA_UNVERIFIABLE_SIGNATURE; + if (sb->s_user_ns != &init_user_ns) + sb->s_iflags |= SB_I_UNTRUSTED_MOUNTER; file = fget(d.fd); err = -EINVAL; diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index f58716567972..35f5ee23566d 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -54,8 +54,7 @@ static void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page, continue; if (start >= to) break; - if (gfs2_is_jdata(ip)) - set_buffer_uptodate(bh); + set_buffer_uptodate(bh); gfs2_trans_add_data(ip->i_gl, bh); } } @@ -747,18 +746,21 @@ out: put_page(page); gfs2_trans_end(sdp); - if (pos + len > ip->i_inode.i_size) - gfs2_trim_blocks(&ip->i_inode); - goto out_trans_fail; + if (alloc_required) { + gfs2_inplace_release(ip); + if (pos + len > ip->i_inode.i_size) + gfs2_trim_blocks(&ip->i_inode); + } + goto out_qunlock; out_endtrans: gfs2_trans_end(sdp); out_trans_fail: - if (alloc_required) { + if (alloc_required) gfs2_inplace_release(ip); out_qunlock: + if (alloc_required) gfs2_quota_unlock(ip); - } out_unlock: if (&ip->i_inode == sdp->sd_rindex) { gfs2_glock_dq(&m_ip->i_gh); @@ -814,7 +816,6 @@ out: * @inode: The inode * @dibh: The buffer_head containing the on-disk inode * @pos: The file position - * @len: The length of the write * @copied: How much was actually copied by the VFS * @page: The page * @@ -824,17 +825,15 @@ out: * Returns: errno */ static int gfs2_stuffed_write_end(struct inode *inode, struct buffer_head *dibh, - loff_t pos, unsigned len, unsigned copied, + loff_t pos, unsigned copied, struct page *page) { struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_sbd *sdp = GFS2_SB(inode); - struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode); u64 to = pos + copied; void *kaddr; unsigned char *buf = dibh->b_data + sizeof(struct gfs2_dinode); - BUG_ON(pos + len > gfs2_max_stuffed_size(ip)); + BUG_ON(pos + copied > gfs2_max_stuffed_size(ip)); kaddr = kmap_atomic(page); memcpy(buf + pos, kaddr + pos, copied); @@ -850,20 +849,6 @@ static int gfs2_stuffed_write_end(struct inode *inode, struct buffer_head *dibh, i_size_write(inode, to); mark_inode_dirty(inode); } - - if (inode == sdp->sd_rindex) { - adjust_fs_space(inode); - sdp->sd_rindex_uptodate = 0; - } - - brelse(dibh); - gfs2_trans_end(sdp); - if (inode == sdp->sd_rindex) { - gfs2_glock_dq(&m_ip->i_gh); - gfs2_holder_uninit(&m_ip->i_gh); - } - gfs2_glock_dq(&ip->i_gh); - gfs2_holder_uninit(&ip->i_gh); return copied; } @@ -877,9 +862,8 @@ static int gfs2_stuffed_write_end(struct inode *inode, struct buffer_head *dibh, * @page: The page that has been written * @fsdata: The fsdata (unused in GFS2) * - * The main write_end function for GFS2. We have a separate one for - * stuffed files as they are slightly different, otherwise we just - * put our locking around the VFS provided functions. + * The main write_end function for GFS2. We just put our locking around the VFS + * provided functions. * * Returns: errno */ @@ -900,32 +884,39 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, BUG_ON(gfs2_glock_is_locked_by_me(ip->i_gl) == NULL); ret = gfs2_meta_inode_buffer(ip, &dibh); - if (unlikely(ret)) { - unlock_page(page); - put_page(page); - goto failed; - } + if (unlikely(ret)) + goto out; - if (gfs2_is_stuffed(ip)) - return gfs2_stuffed_write_end(inode, dibh, pos, len, copied, page); + if (gfs2_is_stuffed(ip)) { + ret = gfs2_stuffed_write_end(inode, dibh, pos, copied, page); + page = NULL; + goto out2; + } - if (!gfs2_is_writeback(ip)) + if (gfs2_is_jdata(ip)) gfs2_page_add_databufs(ip, page, pos & ~PAGE_MASK, len); + else + gfs2_ordered_add_inode(ip); ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata); + page = NULL; if (tr->tr_num_buf_new) __mark_inode_dirty(inode, I_DIRTY_DATASYNC); else gfs2_trans_add_meta(ip->i_gl, dibh); - +out2: if (inode == sdp->sd_rindex) { adjust_fs_space(inode); sdp->sd_rindex_uptodate = 0; } brelse(dibh); -failed: +out: + if (page) { + unlock_page(page); + put_page(page); + } gfs2_trans_end(sdp); gfs2_inplace_release(ip); if (ip->i_qadata && ip->i_qadata->qa_qd_num) diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 685c305cbeb6..ed6699705c13 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -89,10 +89,12 @@ static int gfs2_unstuffer_page(struct gfs2_inode *ip, struct buffer_head *dibh, map_bh(bh, inode->i_sb, block); set_buffer_uptodate(bh); - if (!gfs2_is_jdata(ip)) - mark_buffer_dirty(bh); - if (!gfs2_is_writeback(ip)) + if (gfs2_is_jdata(ip)) gfs2_trans_add_data(ip->i_gl, bh); + else { + mark_buffer_dirty(bh); + gfs2_ordered_add_inode(ip); + } if (release) { unlock_page(page); @@ -176,8 +178,8 @@ out: /** * find_metapath - Find path through the metadata tree * @sdp: The superblock - * @mp: The metapath to return the result in * @block: The disk block to look up + * @mp: The metapath to return the result in * @height: The pre-calculated height of the metadata tree * * This routine returns a struct metapath structure that defines a path @@ -188,8 +190,7 @@ out: * filesystem with a blocksize of 4096. * * find_metapath() would return a struct metapath structure set to: - * mp_offset = 101342453, mp_height = 3, mp_list[0] = 0, mp_list[1] = 48, - * and mp_list[2] = 165. + * mp_fheight = 3, mp_list[0] = 0, mp_list[1] = 48, and mp_list[2] = 165. * * That means that in order to get to the block containing the byte at * offset 101342453, we would load the indirect block pointed to by pointer @@ -279,6 +280,21 @@ static inline __be64 *metapointer(unsigned int height, const struct metapath *mp return p + mp->mp_list[height]; } +static inline const __be64 *metaend(unsigned int height, const struct metapath *mp) +{ + const struct buffer_head *bh = mp->mp_bh[height]; + return (const __be64 *)(bh->b_data + bh->b_size); +} + +static void clone_metapath(struct metapath *clone, struct metapath *mp) +{ + unsigned int hgt; + + *clone = *mp; + for (hgt = 0; hgt < mp->mp_aheight; hgt++) + get_bh(clone->mp_bh[hgt]); +} + static void gfs2_metapath_ra(struct gfs2_glock *gl, __be64 *start, __be64 *end) { const __be64 *t; @@ -420,20 +436,140 @@ static inline unsigned int gfs2_extent_length(void *start, unsigned int len, __b return (ptr - first); } -static inline void bmap_lock(struct gfs2_inode *ip, int create) +typedef const __be64 *(*gfs2_metadata_walker)( + struct metapath *mp, + const __be64 *start, const __be64 *end, + u64 factor, void *data); + +#define WALK_STOP ((__be64 *)0) +#define WALK_NEXT ((__be64 *)1) + +static int gfs2_walk_metadata(struct inode *inode, sector_t lblock, + u64 len, struct metapath *mp, gfs2_metadata_walker walker, + void *data) { - if (create) - down_write(&ip->i_rw_mutex); - else - down_read(&ip->i_rw_mutex); + struct metapath clone; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); + const __be64 *start, *end, *ptr; + u64 factor = 1; + unsigned int hgt; + int ret = 0; + + for (hgt = ip->i_height - 1; hgt >= mp->mp_aheight; hgt--) + factor *= sdp->sd_inptrs; + + for (;;) { + u64 step; + + /* Walk indirect block. */ + start = metapointer(hgt, mp); + end = metaend(hgt, mp); + + step = (end - start) * factor; + if (step > len) + end = start + DIV_ROUND_UP_ULL(len, factor); + + ptr = walker(mp, start, end, factor, data); + if (ptr == WALK_STOP) + break; + if (step >= len) + break; + len -= step; + if (ptr != WALK_NEXT) { + BUG_ON(!*ptr); + mp->mp_list[hgt] += ptr - start; + goto fill_up_metapath; + } + +lower_metapath: + /* Decrease height of metapath. */ + if (mp != &clone) { + clone_metapath(&clone, mp); + mp = &clone; + } + brelse(mp->mp_bh[hgt]); + mp->mp_bh[hgt] = NULL; + if (!hgt) + break; + hgt--; + factor *= sdp->sd_inptrs; + + /* Advance in metadata tree. */ + (mp->mp_list[hgt])++; + start = metapointer(hgt, mp); + end = metaend(hgt, mp); + if (start >= end) { + mp->mp_list[hgt] = 0; + if (!hgt) + break; + goto lower_metapath; + } + +fill_up_metapath: + /* Increase height of metapath. */ + if (mp != &clone) { + clone_metapath(&clone, mp); + mp = &clone; + } + ret = fillup_metapath(ip, mp, ip->i_height - 1); + if (ret < 0) + break; + hgt += ret; + for (; ret; ret--) + do_div(factor, sdp->sd_inptrs); + mp->mp_aheight = hgt + 1; + } + if (mp == &clone) + release_metapath(mp); + return ret; } -static inline void bmap_unlock(struct gfs2_inode *ip, int create) +struct gfs2_hole_walker_args { + u64 blocks; +}; + +static const __be64 *gfs2_hole_walker(struct metapath *mp, + const __be64 *start, const __be64 *end, + u64 factor, void *data) { - if (create) - up_write(&ip->i_rw_mutex); - else - up_read(&ip->i_rw_mutex); + struct gfs2_hole_walker_args *args = data; + const __be64 *ptr; + + for (ptr = start; ptr < end; ptr++) { + if (*ptr) { + args->blocks += (ptr - start) * factor; + if (mp->mp_aheight == mp->mp_fheight) + return WALK_STOP; + return ptr; /* increase height */ + } + } + args->blocks += (end - start) * factor; + return WALK_NEXT; +} + +/** + * gfs2_hole_size - figure out the size of a hole + * @inode: The inode + * @lblock: The logical starting block number + * @len: How far to look (in blocks) + * @mp: The metapath at lblock + * @iomap: The iomap to store the hole size in + * + * This function modifies @mp. + * + * Returns: errno on error + */ +static int gfs2_hole_size(struct inode *inode, sector_t lblock, u64 len, + struct metapath *mp, struct iomap *iomap) +{ + struct gfs2_hole_walker_args args = { }; + int ret = 0; + + ret = gfs2_walk_metadata(inode, lblock, len, mp, gfs2_hole_walker, &args); + if (!ret) + iomap->length = args.blocks << inode->i_blkbits; + return ret; } static inline __be64 *gfs2_indirect_init(struct metapath *mp, @@ -462,15 +598,11 @@ enum alloc_state { }; /** - * gfs2_bmap_alloc - Build a metadata tree of the requested height + * gfs2_iomap_alloc - Build a metadata tree of the requested height * @inode: The GFS2 inode - * @lblock: The logical starting block of the extent - * @bh_map: This is used to return the mapping details - * @zero_new: True if newly allocated blocks should be zeroed + * @iomap: The iomap structure + * @flags: iomap flags * @mp: The metapath, with proper height information calculated - * @maxlen: The max number of data blocks to alloc - * @dblock: Pointer to return the resulting new block - * @dblks: Pointer to return the number of blocks allocated * * In this routine we may have to alloc: * i) Indirect blocks to grow the metadata tree height @@ -483,6 +615,13 @@ enum alloc_state { * blocks are available, there will only be one request per bmap call) * and uses the state machine to initialise the blocks in order. * + * Right now, this function will allocate at most one indirect block + * worth of data -- with a default block size of 4K, that's slightly + * less than 2M. If this limitation is ever removed to allow huge + * allocations, we would probably still want to limit the iomap size we + * return to avoid stalling other tasks during huge writes; the next + * iomap iteration would then find the blocks already allocated. + * * Returns: errno on error */ @@ -497,6 +636,7 @@ static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap, unsigned dblks = 0; unsigned ptrs_per_blk; const unsigned end_of_metadata = mp->mp_fheight - 1; + int ret; enum alloc_state state; __be64 *ptr; __be64 zero_bn = 0; @@ -507,6 +647,8 @@ static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap, gfs2_trans_add_meta(ip->i_gl, dibh); + down_write(&ip->i_rw_mutex); + if (mp->mp_fheight == mp->mp_aheight) { struct buffer_head *bh; int eob; @@ -542,11 +684,10 @@ static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap, blks = dblks + iblks; i = mp->mp_aheight; do { - int error; n = blks - alloced; - error = gfs2_alloc_blocks(ip, &bn, &n, 0, NULL); - if (error) - return error; + ret = gfs2_alloc_blocks(ip, &bn, &n, 0, NULL); + if (ret) + goto out; alloced += n; if (state != ALLOC_DATA || gfs2_is_jdata(ip)) gfs2_trans_add_unrevoke(sdp, bn, n); @@ -602,7 +743,7 @@ static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap, dblks = n; ptr = metapointer(end_of_metadata, mp); iomap->addr = bn << inode->i_blkbits; - iomap->flags |= IOMAP_F_NEW; + iomap->flags |= IOMAP_F_MERGED | IOMAP_F_NEW; while (n-- > 0) *ptr++ = cpu_to_be64(bn++); break; @@ -612,64 +753,10 @@ static int gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap, iomap->length = (u64)dblks << inode->i_blkbits; ip->i_height = mp->mp_fheight; gfs2_add_inode_blocks(&ip->i_inode, alloced); - gfs2_dinode_out(ip, mp->mp_bh[0]->b_data); - return 0; -} - -/** - * hole_size - figure out the size of a hole - * @inode: The inode - * @lblock: The logical starting block number - * @mp: The metapath - * - * Returns: The hole size in bytes - * - */ -static u64 hole_size(struct inode *inode, sector_t lblock, struct metapath *mp) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_sbd *sdp = GFS2_SB(inode); - struct metapath mp_eof; - u64 factor = 1; - int hgt; - u64 holesz = 0; - const __be64 *first, *end, *ptr; - const struct buffer_head *bh; - u64 lblock_stop = (i_size_read(inode) - 1) >> inode->i_blkbits; - int zeroptrs; - bool done = false; - - /* Get another metapath, to the very last byte */ - find_metapath(sdp, lblock_stop, &mp_eof, ip->i_height); - for (hgt = ip->i_height - 1; hgt >= 0 && !done; hgt--) { - bh = mp->mp_bh[hgt]; - if (bh) { - zeroptrs = 0; - first = metapointer(hgt, mp); - end = (const __be64 *)(bh->b_data + bh->b_size); - - for (ptr = first; ptr < end; ptr++) { - if (*ptr) { - done = true; - break; - } else { - zeroptrs++; - } - } - } else { - zeroptrs = sdp->sd_inptrs; - } - if (factor * zeroptrs >= lblock_stop - lblock + 1) { - holesz = lblock_stop - lblock + 1; - break; - } - holesz += factor * zeroptrs; - - factor *= sdp->sd_inptrs; - if (hgt && (mp->mp_list[hgt - 1] < mp_eof.mp_list[hgt - 1])) - (mp->mp_list[hgt - 1])++; - } - return holesz << inode->i_blkbits; + gfs2_dinode_out(ip, dibh->b_data); +out: + up_write(&ip->i_rw_mutex); + return ret; } static void gfs2_stuffed_iomap(struct inode *inode, struct iomap *iomap) @@ -680,126 +767,136 @@ static void gfs2_stuffed_iomap(struct inode *inode, struct iomap *iomap) sizeof(struct gfs2_dinode); iomap->offset = 0; iomap->length = i_size_read(inode); - iomap->type = IOMAP_MAPPED; - iomap->flags = IOMAP_F_DATA_INLINE; + iomap->type = IOMAP_INLINE; } +#define IOMAP_F_GFS2_BOUNDARY IOMAP_F_PRIVATE + /** - * gfs2_iomap_begin - Map blocks from an inode to disk blocks + * gfs2_iomap_get - Map blocks from an inode to disk blocks * @inode: The inode * @pos: Starting position in bytes * @length: Length to map, in bytes * @flags: iomap flags * @iomap: The iomap structure + * @mp: The metapath * * Returns: errno */ -int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length, - unsigned flags, struct iomap *iomap) +static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length, + unsigned flags, struct iomap *iomap, + struct metapath *mp) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); - struct metapath mp = { .mp_aheight = 1, }; - unsigned int factor = sdp->sd_sb.sb_bsize; - const u64 *arr = sdp->sd_heightsize; __be64 *ptr; sector_t lblock; - sector_t lend; - int ret = 0; + sector_t lblock_stop; + int ret; int eob; - unsigned int len; + u64 len; struct buffer_head *bh; u8 height; - trace_gfs2_iomap_start(ip, pos, length, flags); - if (!length) { - ret = -EINVAL; - goto out; - } + if (!length) + return -EINVAL; if (gfs2_is_stuffed(ip)) { if (flags & IOMAP_REPORT) { + if (pos >= i_size_read(inode)) + return -ENOENT; gfs2_stuffed_iomap(inode, iomap); - if (pos >= iomap->length) - ret = -ENOENT; - goto out; + return 0; } BUG_ON(!(flags & IOMAP_WRITE)); } - lblock = pos >> inode->i_blkbits; - lend = (pos + length + sdp->sd_sb.sb_bsize - 1) >> inode->i_blkbits; - iomap->offset = lblock << inode->i_blkbits; - iomap->addr = IOMAP_NULL_ADDR; - iomap->type = IOMAP_HOLE; - iomap->length = (u64)(lend - lblock) << inode->i_blkbits; - iomap->flags = IOMAP_F_MERGED; - bmap_lock(ip, flags & IOMAP_WRITE); + lblock_stop = (pos + length - 1) >> inode->i_blkbits; + len = lblock_stop - lblock + 1; - /* - * Directory data blocks have a struct gfs2_meta_header header, so the - * remaining size is smaller than the filesystem block size. Logical - * block numbers for directories are in units of this remaining size! - */ - if (gfs2_is_dir(ip)) { - factor = sdp->sd_jbsize; - arr = sdp->sd_jheightsize; - } + down_read(&ip->i_rw_mutex); - ret = gfs2_meta_inode_buffer(ip, &mp.mp_bh[0]); + ret = gfs2_meta_inode_buffer(ip, &mp->mp_bh[0]); if (ret) - goto out_release; + goto unlock; height = ip->i_height; - while ((lblock + 1) * factor > arr[height]) + while ((lblock + 1) * sdp->sd_sb.sb_bsize > sdp->sd_heightsize[height]) height++; - find_metapath(sdp, lblock, &mp, height); + find_metapath(sdp, lblock, mp, height); if (height > ip->i_height || gfs2_is_stuffed(ip)) goto do_alloc; - ret = lookup_metapath(ip, &mp); + ret = lookup_metapath(ip, mp); if (ret) - goto out_release; + goto unlock; - if (mp.mp_aheight != ip->i_height) + if (mp->mp_aheight != ip->i_height) goto do_alloc; - ptr = metapointer(ip->i_height - 1, &mp); + ptr = metapointer(ip->i_height - 1, mp); if (*ptr == 0) goto do_alloc; - iomap->type = IOMAP_MAPPED; - iomap->addr = be64_to_cpu(*ptr) << inode->i_blkbits; + bh = mp->mp_bh[ip->i_height - 1]; + len = gfs2_extent_length(bh->b_data, bh->b_size, ptr, len, &eob); - bh = mp.mp_bh[ip->i_height - 1]; - len = gfs2_extent_length(bh->b_data, bh->b_size, ptr, lend - lblock, &eob); + iomap->addr = be64_to_cpu(*ptr) << inode->i_blkbits; + iomap->length = len << inode->i_blkbits; + iomap->type = IOMAP_MAPPED; + iomap->flags = IOMAP_F_MERGED; if (eob) - iomap->flags |= IOMAP_F_BOUNDARY; - iomap->length = (u64)len << inode->i_blkbits; + iomap->flags |= IOMAP_F_GFS2_BOUNDARY; -out_release: - release_metapath(&mp); - bmap_unlock(ip, flags & IOMAP_WRITE); out: - trace_gfs2_iomap_end(ip, iomap, ret); + iomap->bdev = inode->i_sb->s_bdev; +unlock: + up_read(&ip->i_rw_mutex); return ret; do_alloc: - if (flags & IOMAP_WRITE) { - ret = gfs2_iomap_alloc(inode, iomap, flags, &mp); - } else if (flags & IOMAP_REPORT) { + iomap->addr = IOMAP_NULL_ADDR; + iomap->length = len << inode->i_blkbits; + iomap->type = IOMAP_HOLE; + iomap->flags = 0; + if (flags & IOMAP_REPORT) { loff_t size = i_size_read(inode); if (pos >= size) ret = -ENOENT; - else if (height <= ip->i_height) - iomap->length = hole_size(inode, lblock, &mp); + else if (height == ip->i_height) + ret = gfs2_hole_size(inode, lblock, len, mp, iomap); else iomap->length = size - pos; } - goto out_release; + goto out; +} + +static int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length, + unsigned flags, struct iomap *iomap) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct metapath mp = { .mp_aheight = 1, }; + int ret; + + trace_gfs2_iomap_start(ip, pos, length, flags); + if (flags & IOMAP_WRITE) { + ret = gfs2_iomap_get(inode, pos, length, flags, iomap, &mp); + if (!ret && iomap->type == IOMAP_HOLE) + ret = gfs2_iomap_alloc(inode, iomap, flags, &mp); + release_metapath(&mp); + } else { + ret = gfs2_iomap_get(inode, pos, length, flags, iomap, &mp); + release_metapath(&mp); + } + trace_gfs2_iomap_end(ip, iomap, ret); + return ret; } +const struct iomap_ops gfs2_iomap_ops = { + .iomap_begin = gfs2_iomap_begin, +}; + /** * gfs2_block_map - Map one or more blocks of an inode to a disk block * @inode: The inode @@ -825,34 +922,43 @@ int gfs2_block_map(struct inode *inode, sector_t lblock, struct buffer_head *bh_map, int create) { struct gfs2_inode *ip = GFS2_I(inode); - struct iomap iomap; - int ret, flags = 0; + loff_t pos = (loff_t)lblock << inode->i_blkbits; + loff_t length = bh_map->b_size; + struct metapath mp = { .mp_aheight = 1, }; + struct iomap iomap = { }; + int ret; clear_buffer_mapped(bh_map); clear_buffer_new(bh_map); clear_buffer_boundary(bh_map); trace_gfs2_bmap(ip, bh_map, lblock, create, 1); - if (create) - flags |= IOMAP_WRITE; - ret = gfs2_iomap_begin(inode, (loff_t)lblock << inode->i_blkbits, - bh_map->b_size, flags, &iomap); - if (ret) { - if (!create && ret == -ENOENT) { - /* Return unmapped buffer beyond the end of file. */ + if (create) { + ret = gfs2_iomap_get(inode, pos, length, IOMAP_WRITE, &iomap, &mp); + if (!ret && iomap.type == IOMAP_HOLE) + ret = gfs2_iomap_alloc(inode, &iomap, IOMAP_WRITE, &mp); + release_metapath(&mp); + } else { + ret = gfs2_iomap_get(inode, pos, length, 0, &iomap, &mp); + release_metapath(&mp); + + /* Return unmapped buffer beyond the end of file. */ + if (ret == -ENOENT) { ret = 0; + goto out; } - goto out; } + if (ret) + goto out; if (iomap.length > bh_map->b_size) { iomap.length = bh_map->b_size; - iomap.flags &= ~IOMAP_F_BOUNDARY; + iomap.flags &= ~IOMAP_F_GFS2_BOUNDARY; } if (iomap.addr != IOMAP_NULL_ADDR) map_bh(bh_map, inode->i_sb, iomap.addr >> inode->i_blkbits); bh_map->b_size = iomap.length; - if (iomap.flags & IOMAP_F_BOUNDARY) + if (iomap.flags & IOMAP_F_GFS2_BOUNDARY) set_buffer_boundary(bh_map); if (iomap.flags & IOMAP_F_NEW) set_buffer_new(bh_map); @@ -945,8 +1051,10 @@ static int gfs2_block_zero_range(struct inode *inode, loff_t from, err = 0; } - if (!gfs2_is_writeback(ip)) + if (gfs2_is_jdata(ip)) gfs2_trans_add_data(ip->i_gl, bh); + else + gfs2_ordered_add_inode(ip); zero_user(page, offset, length); mark_buffer_dirty(bh); @@ -1056,6 +1164,19 @@ out: return error; } +int gfs2_iomap_get_alloc(struct inode *inode, loff_t pos, loff_t length, + struct iomap *iomap) +{ + struct metapath mp = { .mp_aheight = 1, }; + int ret; + + ret = gfs2_iomap_get(inode, pos, length, IOMAP_WRITE, iomap, &mp); + if (!ret && iomap->type == IOMAP_HOLE) + ret = gfs2_iomap_alloc(inode, iomap, IOMAP_WRITE, &mp); + release_metapath(&mp); + return ret; +} + /** * sweep_bh_for_rgrps - find an rgrp in a meta buffer and free blocks therein * @ip: inode @@ -1744,7 +1865,7 @@ do_grow_qunlock: * @newsize: the size to make the file * * The file size can grow, shrink, or stay the same size. This - * is called holding i_mutex and an exclusive glock on the inode + * is called holding i_rwsem and an exclusive glock on the inode * in question. * * Returns: errno diff --git a/fs/gfs2/bmap.h b/fs/gfs2/bmap.h index c3402fe00653..6b18fb323f0a 100644 --- a/fs/gfs2/bmap.h +++ b/fs/gfs2/bmap.h @@ -46,11 +46,13 @@ static inline void gfs2_write_calc_reserv(const struct gfs2_inode *ip, } } +extern const struct iomap_ops gfs2_iomap_ops; + extern int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page); extern int gfs2_block_map(struct inode *inode, sector_t lblock, struct buffer_head *bh, int create); -extern int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length, - unsigned flags, struct iomap *iomap); +extern int gfs2_iomap_get_alloc(struct inode *inode, loff_t pos, loff_t length, + struct iomap *iomap); extern int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen); extern int gfs2_setattr_size(struct inode *inode, u64 size); diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 4b71f021a9e2..7137db7b0119 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -733,7 +733,7 @@ static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len, struct gfs2_inode *ip = GFS2_I(inode); loff_t end = offset + len; struct buffer_head *dibh; - struct iomap iomap; + struct iomap iomap = { }; int error; error = gfs2_meta_inode_buffer(ip, &dibh); @@ -749,8 +749,8 @@ static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len, } while (offset < end) { - error = gfs2_iomap_begin(inode, offset, end - offset, - IOMAP_WRITE, &iomap); + error = gfs2_iomap_get_alloc(inode, offset, end - offset, + &iomap); if (error) goto out; offset = iomap.offset + iomap.length; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 82fb5583445c..097bd3c0f270 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1923,28 +1923,37 @@ void gfs2_glock_exit(void) static void gfs2_glock_iter_next(struct gfs2_glock_iter *gi, loff_t n) { - if (n == 0) - gi->gl = rhashtable_walk_peek(&gi->hti); - else { - gi->gl = rhashtable_walk_next(&gi->hti); - n--; + struct gfs2_glock *gl = gi->gl; + + if (gl) { + if (n == 0) + return; + if (!lockref_put_not_zero(&gl->gl_lockref)) + gfs2_glock_queue_put(gl); } for (;;) { - if (IS_ERR_OR_NULL(gi->gl)) { - if (!gi->gl) - return; - if (PTR_ERR(gi->gl) != -EAGAIN) { - gi->gl = NULL; - return; + gl = rhashtable_walk_next(&gi->hti); + if (IS_ERR_OR_NULL(gl)) { + if (gl == ERR_PTR(-EAGAIN)) { + n = 1; + continue; } - n = 0; - } else if (gi->sdp == gi->gl->gl_name.ln_sbd && - !__lockref_is_dead(&gi->gl->gl_lockref)) { - if (!n--) - break; + gl = NULL; + break; + } + if (gl->gl_name.ln_sbd != gi->sdp) + continue; + if (n <= 1) { + if (!lockref_get_not_dead(&gl->gl_lockref)) + continue; + break; + } else { + if (__lockref_is_dead(&gl->gl_lockref)) + continue; + n--; } - gi->gl = rhashtable_walk_next(&gi->hti); } + gi->gl = gl; } static void *gfs2_glock_seq_start(struct seq_file *seq, loff_t *pos) @@ -1988,7 +1997,6 @@ static void gfs2_glock_seq_stop(struct seq_file *seq, void *iter_ptr) { struct gfs2_glock_iter *gi = seq->private; - gi->gl = NULL; rhashtable_walk_stop(&gi->hti); } @@ -2076,7 +2084,8 @@ static int gfs2_glocks_release(struct inode *inode, struct file *file) struct seq_file *seq = file->private_data; struct gfs2_glock_iter *gi = seq->private; - gi->gl = NULL; + if (gi->gl) + gfs2_glock_put(gi->gl); rhashtable_walk_exit(&gi->hti); return seq_release_private(inode, file); } diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 1b6b1e3f5caf..d2ad817e089f 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -116,6 +116,7 @@ static inline struct gfs2_bitmap *rbm_bi(const struct gfs2_rbm *rbm) static inline u64 gfs2_rbm_to_block(const struct gfs2_rbm *rbm) { + BUG_ON(rbm->offset >= rbm->rgd->rd_data); return rbm->rgd->rd_data0 + (rbm_bi(rbm)->bi_start * GFS2_NBBY) + rbm->offset; } @@ -696,8 +697,6 @@ struct gfs2_sbd { u32 sd_max_dirres; /* Max blocks needed to add a directory entry */ u32 sd_max_height; /* Max height of a file's metadata tree */ u64 sd_heightsize[GFS2_MAX_META_HEIGHT + 1]; - u32 sd_max_jheight; /* Max height of journaled file's meta tree */ - u64 sd_jheightsize[GFS2_MAX_META_HEIGHT + 1]; u32 sd_max_dents_per_leaf; /* Max number of dirents in a leaf block */ struct gfs2_args sd_args; /* Mount arguments */ diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 8700eb815638..feda55f67050 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -2006,10 +2006,6 @@ static int gfs2_getattr(const struct path *path, struct kstat *stat, return 0; } -const struct iomap_ops gfs2_iomap_ops = { - .iomap_begin = gfs2_iomap_begin, -}; - static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len) { diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index 1862e310a067..20241436126d 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -14,6 +14,7 @@ #include <linux/spinlock.h> #include <linux/writeback.h> #include "incore.h" +#include "inode.h" /** * gfs2_log_lock - acquire the right to mess with the log manager @@ -50,8 +51,12 @@ static inline void gfs2_log_pointers_init(struct gfs2_sbd *sdp, static inline void gfs2_ordered_add_inode(struct gfs2_inode *ip) { - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + struct gfs2_sbd *sdp; + if (!gfs2_is_ordered(ip)) + return; + + sdp = GFS2_SB(&ip->i_inode); if (!test_bit(GIF_ORDERED, &ip->i_flags)) { spin_lock(&sdp->sd_ordered_lock); if (!test_and_set_bit(GIF_ORDERED, &ip->i_flags)) diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index e6a0a8a89ea7..c2469833b4fb 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -335,25 +335,6 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent) sdp->sd_heightsize[x] = ~0; gfs2_assert(sdp, sdp->sd_max_height <= GFS2_MAX_META_HEIGHT); - sdp->sd_jheightsize[0] = sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_dinode); - sdp->sd_jheightsize[1] = sdp->sd_jbsize * sdp->sd_diptrs; - for (x = 2;; x++) { - u64 space, d; - u32 m; - - space = sdp->sd_jheightsize[x - 1] * sdp->sd_inptrs; - d = space; - m = do_div(d, sdp->sd_inptrs); - - if (d != sdp->sd_jheightsize[x - 1] || m) - break; - sdp->sd_jheightsize[x] = space; - } - sdp->sd_max_jheight = x; - sdp->sd_jheightsize[x] = ~0; - gfs2_assert(sdp, sdp->sd_max_jheight <= GFS2_MAX_META_HEIGHT); - sdp->sd_max_dents_per_leaf = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_leaf)) / GFS2_MIN_DIRENT_SIZE; @@ -825,7 +806,7 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) goto fail_rindex; } /* - * i_mutex on quota files is special. Since this inode is hidden system + * i_rwsem on quota files is special. Since this inode is hidden system * file, we are safe to define locking ourselves. */ lockdep_set_class(&sdp->sd_quota_inode->i_rwsem, diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 7a98abd340ee..e8585dfd209f 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -735,7 +735,10 @@ static int gfs2_write_buf_to_page(struct gfs2_inode *ip, unsigned long index, if (!buffer_uptodate(bh)) goto unlock_out; } - gfs2_trans_add_data(ip->i_gl, bh); + if (gfs2_is_jdata(ip)) + gfs2_trans_add_data(ip->i_gl, bh); + else + gfs2_ordered_add_inode(ip); /* If we need to write to the next block as well */ if (to_write > (bsize - boff)) { diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 8b683917a27e..6bc5cfe710d1 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -372,8 +372,8 @@ static u32 gfs2_free_extlen(const struct gfs2_rbm *rrbm, u32 len) start = bi->bi_bh->b_data; if (bi->bi_clone) start = bi->bi_clone; - end = start + bi->bi_bh->b_size; start += bi->bi_offset; + end = start + bi->bi_len; BUG_ON(rbm.offset & 3); start += (rbm.offset / GFS2_NBBY); bytes = min_t(u32, len / GFS2_NBBY, (end - start)); diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 620be0521866..cf5c7f3080d2 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -800,7 +800,7 @@ static void gfs2_dirty_inode(struct inode *inode, int flags) int need_endtrans = 0; int ret; - if (!(flags & (I_DIRTY_DATASYNC|I_DIRTY_SYNC))) + if (!(flags & I_DIRTY_INODE)) return; if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) return; diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c index c75cacaa349b..064c9a0ef046 100644 --- a/fs/gfs2/trans.c +++ b/fs/gfs2/trans.c @@ -143,32 +143,21 @@ static struct gfs2_bufdata *gfs2_alloc_bufdata(struct gfs2_glock *gl, * @gl: The inode glock associated with the buffer * @bh: The buffer to add * - * This is used in two distinct cases: - * i) In ordered write mode - * We put the data buffer on a list so that we can ensure that it's - * synced to disk at the right time - * ii) In journaled data mode - * We need to journal the data block in the same way as metadata in - * the functions above. The difference is that here we have a tag - * which is two __be64's being the block number (as per meta data) - * and a flag which says whether the data block needs escaping or - * not. This means we need a new log entry for each 251 or so data - * blocks, which isn't an enormous overhead but twice as much as - * for normal metadata blocks. + * This is used in journaled data mode. + * We need to journal the data block in the same way as metadata in + * the functions above. The difference is that here we have a tag + * which is two __be64's being the block number (as per meta data) + * and a flag which says whether the data block needs escaping or + * not. This means we need a new log entry for each 251 or so data + * blocks, which isn't an enormous overhead but twice as much as + * for normal metadata blocks. */ void gfs2_trans_add_data(struct gfs2_glock *gl, struct buffer_head *bh) { struct gfs2_trans *tr = current->journal_info; struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; - struct address_space *mapping = bh->b_page->mapping; - struct gfs2_inode *ip = GFS2_I(mapping->host); struct gfs2_bufdata *bd; - if (!gfs2_is_jdata(ip)) { - gfs2_ordered_add_inode(ip); - return; - } - lock_buffer(bh); if (buffer_pinned(bh)) { set_bit(TR_TOUCHED, &tr->tr_flags); diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c index 05de20954659..f2bce1e0f6fb 100644 --- a/fs/gfs2/xattr.c +++ b/fs/gfs2/xattr.c @@ -308,7 +308,7 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, } ip->i_inode.i_ctime = current_time(&ip->i_inode); - __mark_inode_dirty(&ip->i_inode, I_DIRTY_SYNC | I_DIRTY_DATASYNC); + __mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC); gfs2_trans_end(sdp); @@ -768,7 +768,7 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, goto out_end_trans; ip->i_inode.i_ctime = current_time(&ip->i_inode); - __mark_inode_dirty(&ip->i_inode, I_DIRTY_SYNC | I_DIRTY_DATASYNC); + __mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC); out_end_trans: gfs2_trans_end(GFS2_SB(&ip->i_inode)); @@ -896,7 +896,7 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh, ea_set_remove_stuffed(ip, es->es_el); ip->i_inode.i_ctime = current_time(&ip->i_inode); - __mark_inode_dirty(&ip->i_inode, I_DIRTY_SYNC | I_DIRTY_DATASYNC); + __mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC); gfs2_trans_end(GFS2_SB(&ip->i_inode)); return error; @@ -1114,7 +1114,7 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el) } ip->i_inode.i_ctime = current_time(&ip->i_inode); - __mark_inode_dirty(&ip->i_inode, I_DIRTY_SYNC | I_DIRTY_DATASYNC); + __mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC); gfs2_trans_end(GFS2_SB(&ip->i_inode)); diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c index 75b254280ff6..3bf2ae0e467c 100644 --- a/fs/hfs/dir.c +++ b/fs/hfs/dir.c @@ -31,21 +31,15 @@ static struct dentry *hfs_lookup(struct inode *dir, struct dentry *dentry, hfs_cat_build_key(dir->i_sb, fd.search_key, dir->i_ino, &dentry->d_name); res = hfs_brec_read(&fd, &rec, sizeof(rec)); if (res) { - hfs_find_exit(&fd); - if (res == -ENOENT) { - /* No such entry */ - inode = NULL; - goto done; - } - return ERR_PTR(res); + if (res != -ENOENT) + inode = ERR_PTR(res); + } else { + inode = hfs_iget(dir->i_sb, &fd.search_key->cat, &rec); + if (!inode) + inode = ERR_PTR(-EACCES); } - inode = hfs_iget(dir->i_sb, &fd.search_key->cat, &rec); hfs_find_exit(&fd); - if (!inode) - return ERR_PTR(-EACCES); -done: - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } /* diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index 2538b49cc349..b3309b83371a 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -543,9 +543,9 @@ static struct dentry *hfs_file_lookup(struct inode *dir, struct dentry *dentry, igrab(dir); hlist_add_fake(&inode->i_hash); mark_inode_dirty(inode); + dont_mount(dentry); out: - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } void hfs_evict_inode(struct inode *inode) diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 15e06fb552da..b5254378f011 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -122,8 +122,7 @@ again: if (S_ISREG(inode->i_mode)) HFSPLUS_I(inode)->linkid = linkid; out: - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); fail: hfs_find_exit(&fd); return ERR_PTR(err); diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 513c357c734b..a6c0f54c48c3 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -588,6 +588,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) return 0; out_put_hidden_dir: + cancel_delayed_work_sync(&sbi->sync_work); iput(sbi->hidden_dir); out_put_root: dput(sb->s_root); diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index b9a254dcc0e7..d508c7844681 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -138,10 +138,14 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) /* * page based offset in vm_pgoff could be sufficiently large to - * overflow a (l)off_t when converted to byte offset. + * overflow a loff_t when converted to byte offset. This can + * only happen on architectures where sizeof(loff_t) == + * sizeof(unsigned long). So, only check in those instances. */ - if (vma->vm_pgoff & PGOFF_LOFFT_MAX) - return -EINVAL; + if (sizeof(unsigned long) == sizeof(loff_t)) { + if (vma->vm_pgoff & PGOFF_LOFFT_MAX) + return -EINVAL; + } /* must be huge page aligned */ if (vma->vm_pgoff & (~huge_page_mask(h) >> PAGE_SHIFT)) diff --git a/fs/inode.c b/fs/inode.c index b153aeaa61ea..3b55391072f3 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -178,6 +178,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags = 0; + mapping->wb_err = 0; atomic_set(&mapping->i_mmap_writable, 0); mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE); mapping->private_data = NULL; @@ -348,8 +349,7 @@ EXPORT_SYMBOL(inc_nlink); static void __address_space_init_once(struct address_space *mapping) { - INIT_RADIX_TREE(&mapping->page_tree, GFP_ATOMIC | __GFP_ACCOUNT); - spin_lock_init(&mapping->tree_lock); + INIT_RADIX_TREE(&mapping->i_pages, GFP_ATOMIC | __GFP_ACCOUNT); init_rwsem(&mapping->i_mmap_rwsem); INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); @@ -504,14 +504,14 @@ EXPORT_SYMBOL(__remove_inode_hash); void clear_inode(struct inode *inode) { /* - * We have to cycle tree_lock here because reclaim can be still in the + * We have to cycle the i_pages lock here because reclaim can be in the * process of removing the last page (in __delete_from_page_cache()) - * and we must not free mapping under it. + * and we must not free the mapping under it. */ - spin_lock_irq(&inode->i_data.tree_lock); + xa_lock_irq(&inode->i_data.i_pages); BUG_ON(inode->i_data.nrpages); BUG_ON(inode->i_data.nrexceptional); - spin_unlock_irq(&inode->i_data.tree_lock); + xa_unlock_irq(&inode->i_data.i_pages); BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); diff --git a/fs/ioctl.c b/fs/ioctl.c index 4823431d1c9d..b445b13fc59b 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -549,7 +549,7 @@ static int ioctl_fsfreeze(struct file *filp) { struct super_block *sb = file_inode(filp)->i_sb; - if (!capable(CAP_SYS_ADMIN)) + if (!ns_capable(sb->s_user_ns, CAP_SYS_ADMIN)) return -EPERM; /* If filesystem doesn't support freeze feature, return. */ @@ -566,7 +566,7 @@ static int ioctl_fsthaw(struct file *filp) { struct super_block *sb = file_inode(filp)->i_sb; - if (!capable(CAP_SYS_ADMIN)) + if (!ns_capable(sb->s_user_ns, CAP_SYS_ADMIN)) return -EPERM; /* Thaw */ diff --git a/fs/iomap.c b/fs/iomap.c index afd163586aa0..206539d369a8 100644 --- a/fs/iomap.c +++ b/fs/iomap.c @@ -20,6 +20,7 @@ #include <linux/mm.h> #include <linux/swap.h> #include <linux/pagemap.h> +#include <linux/pagevec.h> #include <linux/file.h> #include <linux/uio.h> #include <linux/backing-dev.h> @@ -27,6 +28,7 @@ #include <linux/task_io_accounting_ops.h> #include <linux/dax.h> #include <linux/sched/signal.h> +#include <linux/swap.h> #include "internal.h" @@ -95,6 +97,12 @@ iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags, return written ? written : ret; } +static sector_t +iomap_sector(struct iomap *iomap, loff_t pos) +{ + return (iomap->addr + pos - iomap->offset) >> SECTOR_SHIFT; +} + static void iomap_write_failed(struct inode *inode, loff_t pos, unsigned len) { @@ -352,11 +360,8 @@ static int iomap_zero(struct inode *inode, loff_t pos, unsigned offset, static int iomap_dax_zero(loff_t pos, unsigned offset, unsigned bytes, struct iomap *iomap) { - sector_t sector = (iomap->addr + - (pos & PAGE_MASK) - iomap->offset) >> 9; - - return __dax_zero_page_range(iomap->bdev, iomap->dax_dev, sector, - offset, bytes); + return __dax_zero_page_range(iomap->bdev, iomap->dax_dev, + iomap_sector(iomap, pos & PAGE_MASK), offset, bytes); } static loff_t @@ -501,10 +506,13 @@ static int iomap_to_fiemap(struct fiemap_extent_info *fi, case IOMAP_DELALLOC: flags |= FIEMAP_EXTENT_DELALLOC | FIEMAP_EXTENT_UNKNOWN; break; + case IOMAP_MAPPED: + break; case IOMAP_UNWRITTEN: flags |= FIEMAP_EXTENT_UNWRITTEN; break; - case IOMAP_MAPPED: + case IOMAP_INLINE: + flags |= FIEMAP_EXTENT_DATA_INLINE; break; } @@ -512,8 +520,6 @@ static int iomap_to_fiemap(struct fiemap_extent_info *fi, flags |= FIEMAP_EXTENT_MERGED; if (iomap->flags & IOMAP_F_SHARED) flags |= FIEMAP_EXTENT_SHARED; - if (iomap->flags & IOMAP_F_DATA_INLINE) - flags |= FIEMAP_EXTENT_DATA_INLINE; return fiemap_fill_next_extent(fi, iomap->offset, iomap->addr != IOMAP_NULL_ADDR ? iomap->addr : 0, @@ -587,6 +593,113 @@ int iomap_fiemap(struct inode *inode, struct fiemap_extent_info *fi, } EXPORT_SYMBOL_GPL(iomap_fiemap); +/* + * Seek for SEEK_DATA / SEEK_HOLE within @page, starting at @lastoff. + * Returns true if found and updates @lastoff to the offset in file. + */ +static bool +page_seek_hole_data(struct inode *inode, struct page *page, loff_t *lastoff, + int whence) +{ + const struct address_space_operations *ops = inode->i_mapping->a_ops; + unsigned int bsize = i_blocksize(inode), off; + bool seek_data = whence == SEEK_DATA; + loff_t poff = page_offset(page); + + if (WARN_ON_ONCE(*lastoff >= poff + PAGE_SIZE)) + return false; + + if (*lastoff < poff) { + /* + * Last offset smaller than the start of the page means we found + * a hole: + */ + if (whence == SEEK_HOLE) + return true; + *lastoff = poff; + } + + /* + * Just check the page unless we can and should check block ranges: + */ + if (bsize == PAGE_SIZE || !ops->is_partially_uptodate) + return PageUptodate(page) == seek_data; + + lock_page(page); + if (unlikely(page->mapping != inode->i_mapping)) + goto out_unlock_not_found; + + for (off = 0; off < PAGE_SIZE; off += bsize) { + if ((*lastoff & ~PAGE_MASK) >= off + bsize) + continue; + if (ops->is_partially_uptodate(page, off, bsize) == seek_data) { + unlock_page(page); + return true; + } + *lastoff = poff + off + bsize; + } + +out_unlock_not_found: + unlock_page(page); + return false; +} + +/* + * Seek for SEEK_DATA / SEEK_HOLE in the page cache. + * + * Within unwritten extents, the page cache determines which parts are holes + * and which are data: uptodate buffer heads count as data; everything else + * counts as a hole. + * + * Returns the resulting offset on successs, and -ENOENT otherwise. + */ +static loff_t +page_cache_seek_hole_data(struct inode *inode, loff_t offset, loff_t length, + int whence) +{ + pgoff_t index = offset >> PAGE_SHIFT; + pgoff_t end = DIV_ROUND_UP(offset + length, PAGE_SIZE); + loff_t lastoff = offset; + struct pagevec pvec; + + if (length <= 0) + return -ENOENT; + + pagevec_init(&pvec); + + do { + unsigned nr_pages, i; + + nr_pages = pagevec_lookup_range(&pvec, inode->i_mapping, &index, + end - 1); + if (nr_pages == 0) + break; + + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + + if (page_seek_hole_data(inode, page, &lastoff, whence)) + goto check_range; + lastoff = page_offset(page) + PAGE_SIZE; + } + pagevec_release(&pvec); + } while (index < end); + + /* When no page at lastoff and we are not done, we found a hole. */ + if (whence != SEEK_HOLE) + goto not_found; + +check_range: + if (lastoff < offset + length) + goto out; +not_found: + lastoff = -ENOENT; +out: + pagevec_release(&pvec); + return lastoff; +} + + static loff_t iomap_seek_hole_actor(struct inode *inode, loff_t offset, loff_t length, void *data, struct iomap *iomap) @@ -685,6 +798,8 @@ EXPORT_SYMBOL_GPL(iomap_seek_data); * Private flags for iomap_dio, must not overlap with the public ones in * iomap.h: */ +#define IOMAP_DIO_WRITE_FUA (1 << 28) +#define IOMAP_DIO_NEED_SYNC (1 << 29) #define IOMAP_DIO_WRITE (1 << 30) #define IOMAP_DIO_DIRTY (1 << 31) @@ -759,6 +874,13 @@ static ssize_t iomap_dio_complete(struct iomap_dio *dio) dio_warn_stale_pagecache(iocb->ki_filp); } + /* + * If this is a DSYNC write, make sure we push it to stable storage now + * that we've written data. + */ + if (ret > 0 && (dio->flags & IOMAP_DIO_NEED_SYNC)) + ret = generic_write_sync(iocb, ret); + inode_dio_end(file_inode(iocb->ki_filp)); kfree(dio); @@ -769,13 +891,8 @@ static void iomap_dio_complete_work(struct work_struct *work) { struct iomap_dio *dio = container_of(work, struct iomap_dio, aio.work); struct kiocb *iocb = dio->iocb; - bool is_write = (dio->flags & IOMAP_DIO_WRITE); - ssize_t ret; - ret = iomap_dio_complete(dio); - if (is_write && ret > 0) - ret = generic_write_sync(iocb, ret); - iocb->ki_complete(iocb, ret, 0); + iocb->ki_complete(iocb, iomap_dio_complete(dio), 0); } /* @@ -833,14 +950,12 @@ iomap_dio_zero(struct iomap_dio *dio, struct iomap *iomap, loff_t pos, bio = bio_alloc(GFP_KERNEL, 1); bio_set_dev(bio, iomap->bdev); - bio->bi_iter.bi_sector = - (iomap->addr + pos - iomap->offset) >> 9; + bio->bi_iter.bi_sector = iomap_sector(iomap, pos); bio->bi_private = dio; bio->bi_end_io = iomap_dio_bio_end_io; get_page(page); - if (bio_add_page(bio, page, len, 0) != len) - BUG(); + __bio_add_page(bio, page, len, 0); bio_set_op_attrs(bio, REQ_OP_WRITE, REQ_SYNC | REQ_IDLE); atomic_inc(&dio->ref); @@ -858,6 +973,7 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, struct iov_iter iter; struct bio *bio; bool need_zeroout = false; + bool use_fua = false; int nr_pages, ret; size_t copied = 0; @@ -881,8 +997,20 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, case IOMAP_MAPPED: if (iomap->flags & IOMAP_F_SHARED) dio->flags |= IOMAP_DIO_COW; - if (iomap->flags & IOMAP_F_NEW) + if (iomap->flags & IOMAP_F_NEW) { need_zeroout = true; + } else { + /* + * Use a FUA write if we need datasync semantics, this + * is a pure data IO that doesn't require any metadata + * updates and the underlying device supports FUA. This + * allows us to avoid cache flushes on IO completion. + */ + if (!(iomap->flags & (IOMAP_F_SHARED|IOMAP_F_DIRTY)) && + (dio->flags & IOMAP_DIO_WRITE_FUA) && + blk_queue_fua(bdev_get_queue(iomap->bdev))) + use_fua = true; + } break; default: WARN_ON_ONCE(1); @@ -916,8 +1044,7 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, bio = bio_alloc(GFP_KERNEL, nr_pages); bio_set_dev(bio, iomap->bdev); - bio->bi_iter.bi_sector = - (iomap->addr + pos - iomap->offset) >> 9; + bio->bi_iter.bi_sector = iomap_sector(iomap, pos); bio->bi_write_hint = dio->iocb->ki_hint; bio->bi_private = dio; bio->bi_end_io = iomap_dio_bio_end_io; @@ -930,10 +1057,14 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, n = bio->bi_iter.bi_size; if (dio->flags & IOMAP_DIO_WRITE) { - bio_set_op_attrs(bio, REQ_OP_WRITE, REQ_SYNC | REQ_IDLE); + bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_IDLE; + if (use_fua) + bio->bi_opf |= REQ_FUA; + else + dio->flags &= ~IOMAP_DIO_WRITE_FUA; task_io_account_write(n); } else { - bio_set_op_attrs(bio, REQ_OP_READ, 0); + bio->bi_opf = REQ_OP_READ; if (dio->flags & IOMAP_DIO_DIRTY) bio_set_pages_dirty(bio); } @@ -961,6 +1092,15 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, return copied; } +/* + * iomap_dio_rw() always completes O_[D]SYNC writes regardless of whether the IO + * is being issued as AIO or not. This allows us to optimise pure data writes + * to use REQ_FUA rather than requiring generic_write_sync() to issue a + * REQ_FLUSH post write. This is slightly tricky because a single request here + * can be mapped into multiple disjoint IOs and only a subset of the IOs issued + * may be pure data writes. In that case, we still need to do a full data sync + * completion. + */ ssize_t iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, const struct iomap_ops *ops, iomap_dio_end_io_t end_io) @@ -1005,8 +1145,21 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, if (iter->type == ITER_IOVEC) dio->flags |= IOMAP_DIO_DIRTY; } else { - dio->flags |= IOMAP_DIO_WRITE; flags |= IOMAP_WRITE; + dio->flags |= IOMAP_DIO_WRITE; + + /* for data sync or sync, we need sync completion processing */ + if (iocb->ki_flags & IOCB_DSYNC) + dio->flags |= IOMAP_DIO_NEED_SYNC; + + /* + * For datasync only writes, we optimistically try using FUA for + * this IO. Any non-FUA write that occurs will clear this flag, + * hence we know before completion whether a cache flush is + * necessary. + */ + if ((iocb->ki_flags & (IOCB_DSYNC | IOCB_SYNC)) == IOCB_DSYNC) + dio->flags |= IOMAP_DIO_WRITE_FUA; } if (iocb->ki_flags & IOCB_NOWAIT) { @@ -1062,6 +1215,13 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, if (ret < 0) iomap_dio_set_error(dio, ret); + /* + * If all the writes we issued were FUA, we don't need to flush the + * cache on IO completion. Clear the sync flag for this case. + */ + if (dio->flags & IOMAP_DIO_WRITE_FUA) + dio->flags &= ~IOMAP_DIO_NEED_SYNC; + if (!atomic_dec_and_test(&dio->ref)) { if (!is_sync_kiocb(iocb)) return -EIOCBQUEUED; @@ -1089,3 +1249,203 @@ out_free_dio: return ret; } EXPORT_SYMBOL_GPL(iomap_dio_rw); + +/* Swapfile activation */ + +#ifdef CONFIG_SWAP +struct iomap_swapfile_info { + struct iomap iomap; /* accumulated iomap */ + struct swap_info_struct *sis; + uint64_t lowest_ppage; /* lowest physical addr seen (pages) */ + uint64_t highest_ppage; /* highest physical addr seen (pages) */ + unsigned long nr_pages; /* number of pages collected */ + int nr_extents; /* extent count */ +}; + +/* + * Collect physical extents for this swap file. Physical extents reported to + * the swap code must be trimmed to align to a page boundary. The logical + * offset within the file is irrelevant since the swapfile code maps logical + * page numbers of the swap device to the physical page-aligned extents. + */ +static int iomap_swapfile_add_extent(struct iomap_swapfile_info *isi) +{ + struct iomap *iomap = &isi->iomap; + unsigned long nr_pages; + uint64_t first_ppage; + uint64_t first_ppage_reported; + uint64_t next_ppage; + int error; + + /* + * Round the start up and the end down so that the physical + * extent aligns to a page boundary. + */ + first_ppage = ALIGN(iomap->addr, PAGE_SIZE) >> PAGE_SHIFT; + next_ppage = ALIGN_DOWN(iomap->addr + iomap->length, PAGE_SIZE) >> + PAGE_SHIFT; + + /* Skip too-short physical extents. */ + if (first_ppage >= next_ppage) + return 0; + nr_pages = next_ppage - first_ppage; + + /* + * Calculate how much swap space we're adding; the first page contains + * the swap header and doesn't count. The mm still wants that first + * page fed to add_swap_extent, however. + */ + first_ppage_reported = first_ppage; + if (iomap->offset == 0) + first_ppage_reported++; + if (isi->lowest_ppage > first_ppage_reported) + isi->lowest_ppage = first_ppage_reported; + if (isi->highest_ppage < (next_ppage - 1)) + isi->highest_ppage = next_ppage - 1; + + /* Add extent, set up for the next call. */ + error = add_swap_extent(isi->sis, isi->nr_pages, nr_pages, first_ppage); + if (error < 0) + return error; + isi->nr_extents += error; + isi->nr_pages += nr_pages; + return 0; +} + +/* + * Accumulate iomaps for this swap file. We have to accumulate iomaps because + * swap only cares about contiguous page-aligned physical extents and makes no + * distinction between written and unwritten extents. + */ +static loff_t iomap_swapfile_activate_actor(struct inode *inode, loff_t pos, + loff_t count, void *data, struct iomap *iomap) +{ + struct iomap_swapfile_info *isi = data; + int error; + + switch (iomap->type) { + case IOMAP_MAPPED: + case IOMAP_UNWRITTEN: + /* Only real or unwritten extents. */ + break; + case IOMAP_INLINE: + /* No inline data. */ + pr_err("swapon: file is inline\n"); + return -EINVAL; + default: + pr_err("swapon: file has unallocated extents\n"); + return -EINVAL; + } + + /* No uncommitted metadata or shared blocks. */ + if (iomap->flags & IOMAP_F_DIRTY) { + pr_err("swapon: file is not committed\n"); + return -EINVAL; + } + if (iomap->flags & IOMAP_F_SHARED) { + pr_err("swapon: file has shared extents\n"); + return -EINVAL; + } + + /* Only one bdev per swap file. */ + if (iomap->bdev != isi->sis->bdev) { + pr_err("swapon: file is on multiple devices\n"); + return -EINVAL; + } + + if (isi->iomap.length == 0) { + /* No accumulated extent, so just store it. */ + memcpy(&isi->iomap, iomap, sizeof(isi->iomap)); + } else if (isi->iomap.addr + isi->iomap.length == iomap->addr) { + /* Append this to the accumulated extent. */ + isi->iomap.length += iomap->length; + } else { + /* Otherwise, add the retained iomap and store this one. */ + error = iomap_swapfile_add_extent(isi); + if (error) + return error; + memcpy(&isi->iomap, iomap, sizeof(isi->iomap)); + } + return count; +} + +/* + * Iterate a swap file's iomaps to construct physical extents that can be + * passed to the swapfile subsystem. + */ +int iomap_swapfile_activate(struct swap_info_struct *sis, + struct file *swap_file, sector_t *pagespan, + const struct iomap_ops *ops) +{ + struct iomap_swapfile_info isi = { + .sis = sis, + .lowest_ppage = (sector_t)-1ULL, + }; + struct address_space *mapping = swap_file->f_mapping; + struct inode *inode = mapping->host; + loff_t pos = 0; + loff_t len = ALIGN_DOWN(i_size_read(inode), PAGE_SIZE); + loff_t ret; + + ret = filemap_write_and_wait(inode->i_mapping); + if (ret) + return ret; + + while (len > 0) { + ret = iomap_apply(inode, pos, len, IOMAP_REPORT, + ops, &isi, iomap_swapfile_activate_actor); + if (ret <= 0) + return ret; + + pos += ret; + len -= ret; + } + + if (isi.iomap.length) { + ret = iomap_swapfile_add_extent(&isi); + if (ret) + return ret; + } + + *pagespan = 1 + isi.highest_ppage - isi.lowest_ppage; + sis->max = isi.nr_pages; + sis->pages = isi.nr_pages - 1; + sis->highest_bit = isi.nr_pages - 1; + return isi.nr_extents; +} +EXPORT_SYMBOL_GPL(iomap_swapfile_activate); +#endif /* CONFIG_SWAP */ + +static loff_t +iomap_bmap_actor(struct inode *inode, loff_t pos, loff_t length, + void *data, struct iomap *iomap) +{ + sector_t *bno = data, addr; + + if (iomap->type == IOMAP_MAPPED) { + addr = (pos - iomap->offset + iomap->addr) >> inode->i_blkbits; + if (addr > INT_MAX) + WARN(1, "would truncate bmap result\n"); + else + *bno = addr; + } + return 0; +} + +/* legacy ->bmap interface. 0 is the error return (!) */ +sector_t +iomap_bmap(struct address_space *mapping, sector_t bno, + const struct iomap_ops *ops) +{ + struct inode *inode = mapping->host; + loff_t pos = bno >> inode->i_blkbits; + unsigned blocksize = i_blocksize(inode); + + if (filemap_write_and_wait(mapping)) + return 0; + + bno = 0; + iomap_apply(inode, pos, blocksize, 0, ops, &bno, iomap_bmap_actor); + return bno; +} +EXPORT_SYMBOL_GPL(iomap_bmap); diff --git a/fs/isofs/compress.c b/fs/isofs/compress.c index 9bb2fe35799d..10205ececc27 100644 --- a/fs/isofs/compress.c +++ b/fs/isofs/compress.c @@ -20,6 +20,7 @@ #include <linux/init.h> #include <linux/bio.h> +#include <linux/slab.h> #include <linux/vmalloc.h> #include <linux/zlib.h> @@ -59,7 +60,7 @@ static loff_t zisofs_uncompress_block(struct inode *inode, loff_t block_start, >> bufshift; int haveblocks; blkcnt_t blocknum; - struct buffer_head *bhs[needblocks + 1]; + struct buffer_head **bhs; int curbh, curpage; if (block_size > deflateBound(1UL << zisofs_block_shift)) { @@ -80,7 +81,11 @@ static loff_t zisofs_uncompress_block(struct inode *inode, loff_t block_start, /* Because zlib is not thread-safe, do all the I/O at the top. */ blocknum = block_start >> bufshift; - memset(bhs, 0, (needblocks + 1) * sizeof(struct buffer_head *)); + bhs = kcalloc(needblocks + 1, sizeof(*bhs), GFP_KERNEL); + if (!bhs) { + *errp = -ENOMEM; + return 0; + } haveblocks = isofs_get_blocks(inode, blocknum, bhs, needblocks); ll_rw_block(REQ_OP_READ, 0, haveblocks, bhs); @@ -190,6 +195,7 @@ z_eio: b_eio: for (i = 0; i < haveblocks; i++) brelse(bhs[i]); + kfree(bhs); return stream.total_out; } @@ -305,7 +311,7 @@ static int zisofs_readpage(struct file *file, struct page *page) unsigned int zisofs_pages_per_cblock = PAGE_SHIFT <= zisofs_block_shift ? (1 << (zisofs_block_shift - PAGE_SHIFT)) : 0; - struct page *pages[max_t(unsigned, zisofs_pages_per_cblock, 1)]; + struct page **pages; pgoff_t index = page->index, end_index; end_index = (inode->i_size + PAGE_SIZE - 1) >> PAGE_SHIFT; @@ -330,6 +336,12 @@ static int zisofs_readpage(struct file *file, struct page *page) full_page = 0; pcount = 1; } + pages = kcalloc(max_t(unsigned int, zisofs_pages_per_cblock, 1), + sizeof(*pages), GFP_KERNEL); + if (!pages) { + unlock_page(page); + return -ENOMEM; + } pages[full_page] = page; for (i = 0; i < pcount; i++, index++) { @@ -357,6 +369,7 @@ static int zisofs_readpage(struct file *file, struct page *page) } /* At this point, err contains 0 or -EIO depending on the "critical" page */ + kfree(pages); return err; } diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index bc258a4402f6..ec3fba7d492f 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -394,7 +394,10 @@ static int parse_options(char *options, struct iso9660_options *popt) break; #ifdef CONFIG_JOLIET case Opt_iocharset: + kfree(popt->iocharset); popt->iocharset = match_strdup(&args[0]); + if (!popt->iocharset) + return 0; break; #endif case Opt_map_a: diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index dfb057900e79..8ef6b6daaa7a 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -114,7 +114,7 @@ void __jbd2_debug(int level, const char *file, const char *func, va_start(args, fmt); vaf.fmt = fmt; vaf.va = &args; - printk(KERN_DEBUG "%s: (%s, %u): %pV\n", file, func, line, &vaf); + printk(KERN_DEBUG "%s: (%s, %u): %pV", file, func, line, &vaf); va_end(args); } EXPORT_SYMBOL(__jbd2_debug); @@ -2302,8 +2302,7 @@ static void jbd2_journal_destroy_slabs(void) int i; for (i = 0; i < JBD2_MAX_SLABS; i++) { - if (jbd2_slab[i]) - kmem_cache_destroy(jbd2_slab[i]); + kmem_cache_destroy(jbd2_slab[i]); jbd2_slab[i] = NULL; } } @@ -2404,10 +2403,8 @@ static int jbd2_journal_init_journal_head_cache(void) static void jbd2_journal_destroy_journal_head_cache(void) { - if (jbd2_journal_head_cache) { - kmem_cache_destroy(jbd2_journal_head_cache); - jbd2_journal_head_cache = NULL; - } + kmem_cache_destroy(jbd2_journal_head_cache); + jbd2_journal_head_cache = NULL; } /* @@ -2665,11 +2662,10 @@ static int __init jbd2_journal_init_handle_cache(void) static void jbd2_journal_destroy_handle_cache(void) { - if (jbd2_handle_cache) - kmem_cache_destroy(jbd2_handle_cache); - if (jbd2_inode_cache) - kmem_cache_destroy(jbd2_inode_cache); - + kmem_cache_destroy(jbd2_handle_cache); + jbd2_handle_cache = NULL; + kmem_cache_destroy(jbd2_inode_cache); + jbd2_inode_cache = NULL; } /* diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 696ef15ec942..240779e4689c 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -180,14 +180,10 @@ static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal, void jbd2_journal_destroy_revoke_caches(void) { - if (jbd2_revoke_record_cache) { - kmem_cache_destroy(jbd2_revoke_record_cache); - jbd2_revoke_record_cache = NULL; - } - if (jbd2_revoke_table_cache) { - kmem_cache_destroy(jbd2_revoke_table_cache); - jbd2_revoke_table_cache = NULL; - } + kmem_cache_destroy(jbd2_revoke_record_cache); + jbd2_revoke_record_cache = NULL; + kmem_cache_destroy(jbd2_revoke_table_cache); + jbd2_revoke_table_cache = NULL; } int __init jbd2_journal_init_revoke_caches(void) diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index ac311037d7a5..51dd68e67b0f 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -49,10 +49,8 @@ int __init jbd2_journal_init_transaction_cache(void) void jbd2_journal_destroy_transaction_cache(void) { - if (transaction_cache) { - kmem_cache_destroy(transaction_cache); - transaction_cache = NULL; - } + kmem_cache_destroy(transaction_cache); + transaction_cache = NULL; } void jbd2_journal_free_transaction(transaction_t *transaction) @@ -532,6 +530,7 @@ int jbd2_journal_start_reserved(handle_t *handle, unsigned int type, */ ret = start_this_handle(journal, handle, GFP_NOFS); if (ret < 0) { + handle->h_journal = journal; jbd2_journal_free_reserved(handle); return ret; } diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 0a754f38462e..e5a6deb38e1e 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -209,8 +209,7 @@ static int jffs2_create(struct inode *dir_i, struct dentry *dentry, __func__, inode->i_ino, inode->i_mode, inode->i_nlink, f->inocache->pino_nlink, inode->i_mapping->nrpages); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -430,8 +429,7 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -575,8 +573,7 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, umode_t mode mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -747,8 +744,7 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, umode_t mode mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index 4a6cf289be24..83b8f06b4a64 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -21,14 +21,6 @@ #include <linux/pagemap.h> #include "nodelist.h" -struct erase_priv_struct { - struct jffs2_eraseblock *jeb; - struct jffs2_sb_info *c; -}; - -#ifndef __ECOS -static void jffs2_erase_callback(struct erase_info *); -#endif static void jffs2_erase_failed(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, uint32_t bad_offset); static void jffs2_erase_succeeded(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb); static void jffs2_mark_erased_block(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb); @@ -51,7 +43,7 @@ static void jffs2_erase_block(struct jffs2_sb_info *c, jffs2_dbg(1, "%s(): erase block %#08x (range %#08x-%#08x)\n", __func__, jeb->offset, jeb->offset, jeb->offset + c->sector_size); - instr = kmalloc(sizeof(struct erase_info) + sizeof(struct erase_priv_struct), GFP_KERNEL); + instr = kmalloc(sizeof(struct erase_info), GFP_KERNEL); if (!instr) { pr_warn("kmalloc for struct erase_info in jffs2_erase_block failed. Refiling block for later\n"); mutex_lock(&c->erase_free_sem); @@ -67,18 +59,15 @@ static void jffs2_erase_block(struct jffs2_sb_info *c, memset(instr, 0, sizeof(*instr)); - instr->mtd = c->mtd; instr->addr = jeb->offset; instr->len = c->sector_size; - instr->callback = jffs2_erase_callback; - instr->priv = (unsigned long)(&instr[1]); - - ((struct erase_priv_struct *)instr->priv)->jeb = jeb; - ((struct erase_priv_struct *)instr->priv)->c = c; ret = mtd_erase(c->mtd, instr); - if (!ret) + if (!ret) { + jffs2_erase_succeeded(c, jeb); + kfree(instr); return; + } bad_offset = instr->fail_addr; kfree(instr); @@ -214,22 +203,6 @@ static void jffs2_erase_failed(struct jffs2_sb_info *c, struct jffs2_eraseblock wake_up(&c->erase_wait); } -#ifndef __ECOS -static void jffs2_erase_callback(struct erase_info *instr) -{ - struct erase_priv_struct *priv = (void *)instr->priv; - - if(instr->state != MTD_ERASE_DONE) { - pr_warn("Erase at 0x%08llx finished, but state != MTD_ERASE_DONE. State is 0x%x instead.\n", - (unsigned long long)instr->addr, instr->state); - jffs2_erase_failed(priv->c, priv->jeb, instr->fail_addr); - } else { - jffs2_erase_succeeded(priv->c, priv->jeb); - } - kfree(instr); -} -#endif /* !__ECOS */ - /* Hmmm. Maybe we should accept the extra space it takes and make this a standard doubly-linked list? */ static inline void jffs2_remove_node_refs_from_ino_list(struct jffs2_sb_info *c, diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index f60dee7faf03..87bdf0f4cba1 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -342,7 +342,7 @@ static void jffs2_put_super (struct super_block *sb) static void jffs2_kill_sb(struct super_block *sb) { struct jffs2_sb_info *c = JFFS2_SB_INFO(sb); - if (!sb_rdonly(sb)) + if (c && !sb_rdonly(sb)) jffs2_stop_garbage_collect_thread(c); kill_mtd_super(sb); kfree(c); diff --git a/fs/jfs/jfs_debug.c b/fs/jfs/jfs_debug.c index a70907606025..35a5b2a81ae0 100644 --- a/fs/jfs/jfs_debug.c +++ b/fs/jfs/jfs_debug.c @@ -29,7 +29,6 @@ #ifdef PROC_FS_JFS /* see jfs_debug.h */ -static struct proc_dir_entry *base; #ifdef CONFIG_JFS_DEBUG static int jfs_loglevel_proc_show(struct seq_file *m, void *v) { @@ -66,43 +65,29 @@ static const struct file_operations jfs_loglevel_proc_fops = { }; #endif -static struct { - const char *name; - const struct file_operations *proc_fops; -} Entries[] = { -#ifdef CONFIG_JFS_STATISTICS - { "lmstats", &jfs_lmstats_proc_fops, }, - { "txstats", &jfs_txstats_proc_fops, }, - { "xtstat", &jfs_xtstat_proc_fops, }, - { "mpstat", &jfs_mpstat_proc_fops, }, -#endif -#ifdef CONFIG_JFS_DEBUG - { "TxAnchor", &jfs_txanchor_proc_fops, }, - { "loglevel", &jfs_loglevel_proc_fops } -#endif -}; -#define NPROCENT ARRAY_SIZE(Entries) - void jfs_proc_init(void) { - int i; + struct proc_dir_entry *base; - if (!(base = proc_mkdir("fs/jfs", NULL))) + base = proc_mkdir("fs/jfs", NULL); + if (!base) return; - for (i = 0; i < NPROCENT; i++) - proc_create(Entries[i].name, 0, base, Entries[i].proc_fops); +#ifdef CONFIG_JFS_STATISTICS + proc_create_single("lmstats", 0, base, jfs_lmstats_proc_show); + proc_create_single("txstats", 0, base, jfs_txstats_proc_show); + proc_create_single("xtstat", 0, base, jfs_xtstat_proc_show); + proc_create_single("mpstat", 0, base, jfs_mpstat_proc_show); +#endif +#ifdef CONFIG_JFS_DEBUG + proc_create_single("TxAnchor", 0, base, jfs_txanchor_proc_show); + proc_create("loglevel", 0, base, &jfs_loglevel_proc_fops); +#endif } void jfs_proc_clean(void) { - int i; - - if (base) { - for (i = 0; i < NPROCENT; i++) - remove_proc_entry(Entries[i].name, base); - remove_proc_entry("fs/jfs", NULL); - } + remove_proc_subtree("fs/jfs", NULL); } #endif /* PROC_FS_JFS */ diff --git a/fs/jfs/jfs_debug.h b/fs/jfs/jfs_debug.h index eafd1300a00b..0d9e35da8462 100644 --- a/fs/jfs/jfs_debug.h +++ b/fs/jfs/jfs_debug.h @@ -62,7 +62,7 @@ extern void jfs_proc_clean(void); extern int jfsloglevel; -extern const struct file_operations jfs_txanchor_proc_fops; +int jfs_txanchor_proc_show(struct seq_file *m, void *v); /* information message: e.g., configuration, major event */ #define jfs_info(fmt, arg...) do { \ @@ -105,10 +105,10 @@ extern const struct file_operations jfs_txanchor_proc_fops; * ---------- */ #ifdef CONFIG_JFS_STATISTICS -extern const struct file_operations jfs_lmstats_proc_fops; -extern const struct file_operations jfs_txstats_proc_fops; -extern const struct file_operations jfs_mpstat_proc_fops; -extern const struct file_operations jfs_xtstat_proc_fops; +int jfs_lmstats_proc_show(struct seq_file *m, void *v); +int jfs_txstats_proc_show(struct seq_file *m, void *v); +int jfs_mpstat_proc_show(struct seq_file *m, void *v); +int jfs_xtstat_proc_show(struct seq_file *m, void *v); #define INCREMENT(x) ((x)++) #define DECREMENT(x) ((x)--) diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index 0e5d412c0b01..6b68df395892 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -2493,7 +2493,7 @@ exit: } #ifdef CONFIG_JFS_STATISTICS -static int jfs_lmstats_proc_show(struct seq_file *m, void *v) +int jfs_lmstats_proc_show(struct seq_file *m, void *v) { seq_printf(m, "JFS Logmgr stats\n" @@ -2510,16 +2510,4 @@ static int jfs_lmstats_proc_show(struct seq_file *m, void *v) lmStat.partial_page); return 0; } - -static int jfs_lmstats_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, jfs_lmstats_proc_show, NULL); -} - -const struct file_operations jfs_lmstats_proc_fops = { - .open = jfs_lmstats_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; #endif /* CONFIG_JFS_STATISTICS */ diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 1a3b0cc22ad3..fa2c6824c7f2 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -815,7 +815,7 @@ void __invalidate_metapages(struct inode *ip, s64 addr, int len) } #ifdef CONFIG_JFS_STATISTICS -static int jfs_mpstat_proc_show(struct seq_file *m, void *v) +int jfs_mpstat_proc_show(struct seq_file *m, void *v) { seq_printf(m, "JFS Metapage statistics\n" @@ -828,16 +828,4 @@ static int jfs_mpstat_proc_show(struct seq_file *m, void *v) mpStat.lockwait); return 0; } - -static int jfs_mpstat_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, jfs_mpstat_proc_show, NULL); -} - -const struct file_operations jfs_mpstat_proc_fops = { - .open = jfs_mpstat_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; #endif diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index 4d973524c887..a5663cb621d8 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -2998,7 +2998,7 @@ int jfs_sync(void *arg) } #if defined(CONFIG_PROC_FS) && defined(CONFIG_JFS_DEBUG) -static int jfs_txanchor_proc_show(struct seq_file *m, void *v) +int jfs_txanchor_proc_show(struct seq_file *m, void *v) { char *freewait; char *freelockwait; @@ -3032,22 +3032,10 @@ static int jfs_txanchor_proc_show(struct seq_file *m, void *v) list_empty(&TxAnchor.unlock_queue) ? "" : "not "); return 0; } - -static int jfs_txanchor_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, jfs_txanchor_proc_show, NULL); -} - -const struct file_operations jfs_txanchor_proc_fops = { - .open = jfs_txanchor_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; #endif #if defined(CONFIG_PROC_FS) && defined(CONFIG_JFS_STATISTICS) -static int jfs_txstats_proc_show(struct seq_file *m, void *v) +int jfs_txstats_proc_show(struct seq_file *m, void *v) { seq_printf(m, "JFS TxStats\n" @@ -3072,16 +3060,4 @@ static int jfs_txstats_proc_show(struct seq_file *m, void *v) TxStat.txLockAlloc_freelock); return 0; } - -static int jfs_txstats_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, jfs_txstats_proc_show, NULL); -} - -const struct file_operations jfs_txstats_proc_fops = { - .open = jfs_txstats_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; #endif diff --git a/fs/jfs/jfs_xtree.c b/fs/jfs/jfs_xtree.c index 5cde6d2fcfca..2c200b5256a6 100644 --- a/fs/jfs/jfs_xtree.c +++ b/fs/jfs/jfs_xtree.c @@ -3874,7 +3874,7 @@ s64 xtTruncate_pmap(tid_t tid, struct inode *ip, s64 committed_size) } #ifdef CONFIG_JFS_STATISTICS -static int jfs_xtstat_proc_show(struct seq_file *m, void *v) +int jfs_xtstat_proc_show(struct seq_file *m, void *v) { seq_printf(m, "JFS Xtree statistics\n" @@ -3887,16 +3887,4 @@ static int jfs_xtstat_proc_show(struct seq_file *m, void *v) xtStat.split); return 0; } - -static int jfs_xtstat_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, jfs_xtstat_proc_show, NULL); -} - -const struct file_operations jfs_xtstat_proc_fops = { - .open = jfs_xtstat_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; #endif diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index b41596d71858..56c3fcbfe80e 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -178,8 +178,7 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -313,8 +312,7 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, umode_t mode) unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -1059,8 +1057,7 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -1447,8 +1444,7 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out1: diff --git a/fs/kernfs/file.c b/fs/kernfs/file.c index fd5ce883072e..2015d8c45e4a 100644 --- a/fs/kernfs/file.c +++ b/fs/kernfs/file.c @@ -348,11 +348,11 @@ static void kernfs_vma_open(struct vm_area_struct *vma) kernfs_put_active(of->kn); } -static int kernfs_vma_fault(struct vm_fault *vmf) +static vm_fault_t kernfs_vma_fault(struct vm_fault *vmf) { struct file *file = vmf->vma->vm_file; struct kernfs_open_file *of = kernfs_of(file); - int ret; + vm_fault_t ret; if (!of->vm_ops) return VM_FAULT_SIGBUS; @@ -368,11 +368,11 @@ static int kernfs_vma_fault(struct vm_fault *vmf) return ret; } -static int kernfs_vma_page_mkwrite(struct vm_fault *vmf) +static vm_fault_t kernfs_vma_page_mkwrite(struct vm_fault *vmf) { struct file *file = vmf->vma->vm_file; struct kernfs_open_file *of = kernfs_of(file); - int ret; + vm_fault_t ret; if (!of->vm_ops) return VM_FAULT_SIGBUS; diff --git a/fs/kernfs/mount.c b/fs/kernfs/mount.c index 26dd9a50f383..ff2716f9322e 100644 --- a/fs/kernfs/mount.c +++ b/fs/kernfs/mount.c @@ -316,6 +316,7 @@ struct dentry *kernfs_mount_ns(struct file_system_type *fs_type, int flags, info->root = root; info->ns = ns; + INIT_LIST_HEAD(&info->node); sb = sget_userns(fs_type, kernfs_test_super, kernfs_set_super, flags, &init_user_ns, info); diff --git a/fs/libfs.c b/fs/libfs.c index 7ff3cb904acd..0fb590d79f30 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -1060,6 +1060,45 @@ int noop_fsync(struct file *file, loff_t start, loff_t end, int datasync) } EXPORT_SYMBOL(noop_fsync); +int noop_set_page_dirty(struct page *page) +{ + /* + * Unlike __set_page_dirty_no_writeback that handles dirty page + * tracking in the page object, dax does all dirty tracking in + * the inode address_space in response to mkwrite faults. In the + * dax case we only need to worry about potentially dirty CPU + * caches, not dirty page cache pages to write back. + * + * This callback is defined to prevent fallback to + * __set_page_dirty_buffers() in set_page_dirty(). + */ + return 0; +} +EXPORT_SYMBOL_GPL(noop_set_page_dirty); + +void noop_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) +{ + /* + * There is no page cache to invalidate in the dax case, however + * we need this callback defined to prevent falling back to + * block_invalidatepage() in do_invalidatepage(). + */ +} +EXPORT_SYMBOL_GPL(noop_invalidatepage); + +ssize_t noop_direct_IO(struct kiocb *iocb, struct iov_iter *iter) +{ + /* + * iomap based filesystems support direct I/O without need for + * this callback. However, it still needs to be set in + * inode->a_ops so that open/fcntl know that direct I/O is + * generally supported. + */ + return -EINVAL; +} +EXPORT_SYMBOL_GPL(noop_direct_IO); + /* Because kfree isn't assignment-compatible with void(void*) ;-/ */ void kfree_link(void *p) { diff --git a/fs/locks.c b/fs/locks.c index 62bbe8b31f26..05e211be8684 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -2788,22 +2788,10 @@ static const struct seq_operations locks_seq_operations = { .show = locks_show, }; -static int locks_open(struct inode *inode, struct file *filp) -{ - return seq_open_private(filp, &locks_seq_operations, - sizeof(struct locks_iterator)); -} - -static const struct file_operations proc_locks_operations = { - .open = locks_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_private, -}; - static int __init proc_locks_init(void) { - proc_create("locks", 0, NULL, &proc_locks_operations); + proc_create_seq_private("locks", 0, NULL, &locks_seq_operations, + sizeof(struct locks_iterator), NULL); return 0; } fs_initcall(proc_locks_init); diff --git a/fs/minix/namei.c b/fs/minix/namei.c index ccf0f00030bf..1a6084d2b02e 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -28,13 +28,9 @@ static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, un return ERR_PTR(-ENAMETOOLONG); ino = minix_inode_by_name(dentry); - if (ino) { + if (ino) inode = minix_iget(dir->i_sb, ino); - if (IS_ERR(inode)) - return ERR_CAST(inode); - } - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } static int minix_mknod(struct inode * dir, struct dentry *dentry, umode_t mode, dev_t rdev) diff --git a/fs/namei.c b/fs/namei.c index a09419379f5d..6df1f61855d6 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -39,6 +39,7 @@ #include <linux/bitops.h> #include <linux/init_task.h> #include <linux/uaccess.h> +#include <linux/build_bug.h> #include "internal.h" #include "mount.h" @@ -130,6 +131,7 @@ getname_flags(const char __user *filename, int flags, int *empty) struct filename *result; char *kname; int len; + BUILD_BUG_ON(offsetof(struct filename, iname) % sizeof(long) != 0); result = audit_reusename(filename); if (result) @@ -222,9 +224,10 @@ getname_kernel(const char * filename) if (len <= EMBEDDED_NAME_MAX) { result->name = (char *)result->iname; } else if (len <= PATH_MAX) { + const size_t size = offsetof(struct filename, iname[1]); struct filename *tmp; - tmp = kmalloc(sizeof(*tmp), GFP_KERNEL); + tmp = kmalloc(size, GFP_KERNEL); if (unlikely(!tmp)) { __putname(result); return ERR_PTR(-ENOMEM); @@ -927,7 +930,8 @@ static inline int may_follow_link(struct nameidata *nd) if (nd->flags & LOOKUP_RCU) return -ECHILD; - audit_log_link_denied("follow_link", &nd->stack[0].link); + audit_inode(nd->name, nd->stack[0].link.dentry, 0); + audit_log_link_denied("follow_link"); return -EACCES; } @@ -980,20 +984,22 @@ static bool safe_hardlink_source(struct inode *inode) */ static int may_linkat(struct path *link) { - struct inode *inode; + struct inode *inode = link->dentry->d_inode; + + /* Inode writeback is not safe when the uid or gid are invalid. */ + if (!uid_valid(inode->i_uid) || !gid_valid(inode->i_gid)) + return -EOVERFLOW; if (!sysctl_protected_hardlinks) return 0; - inode = link->dentry->d_inode; - /* Source inode owner (or CAP_FOWNER) can hardlink all they like, * otherwise, it must be a safe source. */ if (safe_hardlink_source(inode) || inode_owner_or_capable(inode)) return 0; - audit_log_link_denied("linkat", link); + audit_log_link_denied("linkat"); return -EPERM; } @@ -1434,10 +1440,8 @@ static int path_parent_directory(struct path *path) static int follow_dotdot(struct nameidata *nd) { while(1) { - if (nd->path.dentry == nd->root.dentry && - nd->path.mnt == nd->root.mnt) { + if (path_equal(&nd->path, &nd->root)) break; - } if (nd->path.dentry != nd->path.mnt->mnt_root) { int ret = path_parent_directory(&nd->path); if (ret) @@ -1594,22 +1598,21 @@ static int lookup_fast(struct nameidata *nd, } /* Fast lookup failed, do it the slow way */ -static struct dentry *lookup_slow(const struct qstr *name, - struct dentry *dir, - unsigned int flags) +static struct dentry *__lookup_slow(const struct qstr *name, + struct dentry *dir, + unsigned int flags) { - struct dentry *dentry = ERR_PTR(-ENOENT), *old; + struct dentry *dentry, *old; struct inode *inode = dir->d_inode; DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); - inode_lock_shared(inode); /* Don't go there if it's already dead */ if (unlikely(IS_DEADDIR(inode))) - goto out; + return ERR_PTR(-ENOENT); again: dentry = d_alloc_parallel(dir, name, &wq); if (IS_ERR(dentry)) - goto out; + return dentry; if (unlikely(!d_in_lookup(dentry))) { if (!(flags & LOOKUP_NO_REVAL)) { int error = d_revalidate(dentry, flags); @@ -1631,11 +1634,21 @@ again: dentry = old; } } -out: - inode_unlock_shared(inode); return dentry; } +static struct dentry *lookup_slow(const struct qstr *name, + struct dentry *dir, + unsigned int flags) +{ + struct inode *inode = dir->d_inode; + struct dentry *res; + inode_lock_shared(inode); + res = __lookup_slow(name, dir, flags); + inode_unlock_shared(inode); + return res; +} + static inline int may_lookup(struct nameidata *nd) { if (nd->flags & LOOKUP_RCU) { @@ -2418,56 +2431,63 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, } EXPORT_SYMBOL(vfs_path_lookup); -/** - * lookup_one_len - filesystem helper to lookup single pathname component - * @name: pathname component to lookup - * @base: base directory to lookup from - * @len: maximum length @len should be interpreted to - * - * Note that this routine is purely a helper for filesystem usage and should - * not be called by generic code. - * - * The caller must hold base->i_mutex. - */ -struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) +static int lookup_one_len_common(const char *name, struct dentry *base, + int len, struct qstr *this) { - struct qstr this; - unsigned int c; - int err; - - WARN_ON_ONCE(!inode_is_locked(base->d_inode)); - - this.name = name; - this.len = len; - this.hash = full_name_hash(base, name, len); + this->name = name; + this->len = len; + this->hash = full_name_hash(base, name, len); if (!len) - return ERR_PTR(-EACCES); + return -EACCES; if (unlikely(name[0] == '.')) { if (len < 2 || (len == 2 && name[1] == '.')) - return ERR_PTR(-EACCES); + return -EACCES; } while (len--) { - c = *(const unsigned char *)name++; + unsigned int c = *(const unsigned char *)name++; if (c == '/' || c == '\0') - return ERR_PTR(-EACCES); + return -EACCES; } /* * See if the low-level filesystem might want * to use its own hash.. */ if (base->d_flags & DCACHE_OP_HASH) { - int err = base->d_op->d_hash(base, &this); + int err = base->d_op->d_hash(base, this); if (err < 0) - return ERR_PTR(err); + return err; } - err = inode_permission(base->d_inode, MAY_EXEC); + return inode_permission(base->d_inode, MAY_EXEC); +} + +/** + * lookup_one_len - filesystem helper to lookup single pathname component + * @name: pathname component to lookup + * @base: base directory to lookup from + * @len: maximum length @len should be interpreted to + * + * Note that this routine is purely a helper for filesystem usage and should + * not be called by generic code. + * + * The caller must hold base->i_mutex. + */ +struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) +{ + struct dentry *dentry; + struct qstr this; + int err; + + WARN_ON_ONCE(!inode_is_locked(base->d_inode)); + + err = lookup_one_len_common(name, base, len, &this); if (err) return ERR_PTR(err); - return __lookup_hash(&this, base, 0); + dentry = lookup_dcache(&this, base, 0); + return dentry ? dentry : __lookup_slow(&this, base, 0); } EXPORT_SYMBOL(lookup_one_len); @@ -2487,37 +2507,10 @@ struct dentry *lookup_one_len_unlocked(const char *name, struct dentry *base, int len) { struct qstr this; - unsigned int c; int err; struct dentry *ret; - this.name = name; - this.len = len; - this.hash = full_name_hash(base, name, len); - if (!len) - return ERR_PTR(-EACCES); - - if (unlikely(name[0] == '.')) { - if (len < 2 || (len == 2 && name[1] == '.')) - return ERR_PTR(-EACCES); - } - - while (len--) { - c = *(const unsigned char *)name++; - if (c == '/' || c == '\0') - return ERR_PTR(-EACCES); - } - /* - * See if the low-level filesystem might want - * to use its own hash.. - */ - if (base->d_flags & DCACHE_OP_HASH) { - int err = base->d_op->d_hash(base, &this); - if (err < 0) - return ERR_PTR(err); - } - - err = inode_permission(base->d_inode, MAY_EXEC); + err = lookup_one_len_common(name, base, len, &this); if (err) return ERR_PTR(err); @@ -2756,6 +2749,11 @@ static int may_delete(struct inode *dir, struct dentry *victim, bool isdir) BUG_ON(!inode); BUG_ON(victim->d_parent->d_inode != dir); + + /* Inode writeback is not safe when the uid or gid are invalid. */ + if (!uid_valid(inode->i_uid) || !gid_valid(inode->i_gid)) + return -EOVERFLOW; + audit_inode_child(dir, victim, AUDIT_TYPE_CHILD_DELETE); error = inode_permission(dir, MAY_WRITE | MAY_EXEC); @@ -3684,7 +3682,8 @@ int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) if (error) return error; - if ((S_ISCHR(mode) || S_ISBLK(mode)) && !capable(CAP_MKNOD)) + if ((S_ISCHR(mode) || S_ISBLK(mode)) && + !ns_capable(dentry->d_sb->s_user_ns, CAP_MKNOD)) return -EPERM; if (!dir->i_op->mknod) @@ -3859,11 +3858,11 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) if (error) goto out; - shrink_dcache_parent(dentry); error = dir->i_op->rmdir(dir, dentry); if (error) goto out; + shrink_dcache_parent(dentry); dentry->d_inode->i_flags |= S_DEAD; dont_mount(dentry); detach_mounts(dentry); @@ -4446,8 +4445,6 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, old_dir->i_nlink >= max_links) goto out; } - if (is_dir && !(flags & RENAME_EXCHANGE) && target) - shrink_dcache_parent(new_dentry); if (!is_dir) { error = try_break_deleg(source, delegated_inode); if (error) @@ -4464,8 +4461,10 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, goto out; if (!(flags & RENAME_EXCHANGE) && target) { - if (is_dir) + if (is_dir) { + shrink_dcache_parent(new_dentry); target->i_flags |= S_DEAD; + } dont_mount(new_dentry); detach_mounts(new_dentry); } diff --git a/fs/namespace.c b/fs/namespace.c index e398f32d7541..8ddd14806799 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1089,7 +1089,8 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root, goto out_free; } - mnt->mnt.mnt_flags = old->mnt.mnt_flags & ~(MNT_WRITE_HOLD|MNT_MARKED); + mnt->mnt.mnt_flags = old->mnt.mnt_flags; + mnt->mnt.mnt_flags &= ~(MNT_WRITE_HOLD|MNT_MARKED|MNT_INTERNAL); /* Don't allow unprivileged users to change mount flags */ if (flag & CL_UNPRIVILEGED) { mnt->mnt.mnt_flags |= MNT_LOCK_ATIME; @@ -1589,7 +1590,7 @@ static int do_umount(struct mount *mnt, int flags) * Special case for "unmounting" root ... * we just try to remount it readonly. */ - if (!capable(CAP_SYS_ADMIN)) + if (!ns_capable(sb->s_user_ns, CAP_SYS_ADMIN)) return -EPERM; down_write(&sb->s_umount); if (!sb_rdonly(sb)) @@ -2332,7 +2333,7 @@ static int do_remount(struct path *path, int ms_flags, int sb_flags, down_write(&sb->s_umount); if (ms_flags & MS_BIND) err = change_mount_flags(path->mnt, ms_flags); - else if (!capable(CAP_SYS_ADMIN)) + else if (!ns_capable(sb->s_user_ns, CAP_SYS_ADMIN)) err = -EPERM; else err = do_remount_sb(sb, sb_flags, data, 0); @@ -2814,7 +2815,7 @@ long do_mount(const char *dev_name, const char __user *dir_name, mnt_flags |= MNT_NODIRATIME; if (flags & MS_STRICTATIME) mnt_flags &= ~(MNT_RELATIME | MNT_NOATIME); - if (flags & SB_RDONLY) + if (flags & MS_RDONLY) mnt_flags |= MNT_READONLY; /* The default atime for remount is preservation */ diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 123c069429a7..a813979b5be0 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -535,35 +535,10 @@ static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char return 0; } -#define CB_SUPPORTED_ATTR0 (FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE) -#define CB_SUPPORTED_ATTR1 (FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY) -static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, __be32 **savep) +static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, size_t sz) { - __be32 bm[2]; - __be32 *p; - - bm[0] = htonl(bitmap[0] & CB_SUPPORTED_ATTR0); - bm[1] = htonl(bitmap[1] & CB_SUPPORTED_ATTR1); - if (bm[1] != 0) { - p = xdr_reserve_space(xdr, 16); - if (unlikely(p == NULL)) - return htonl(NFS4ERR_RESOURCE); - *p++ = htonl(2); - *p++ = bm[0]; - *p++ = bm[1]; - } else if (bm[0] != 0) { - p = xdr_reserve_space(xdr, 12); - if (unlikely(p == NULL)) - return htonl(NFS4ERR_RESOURCE); - *p++ = htonl(1); - *p++ = bm[0]; - } else { - p = xdr_reserve_space(xdr, 8); - if (unlikely(p == NULL)) - return htonl(NFS4ERR_RESOURCE); - *p++ = htonl(0); - } - *savep = p; + if (xdr_stream_encode_uint32_array(xdr, bitmap, sz) < 0) + return cpu_to_be32(NFS4ERR_RESOURCE); return 0; } @@ -656,9 +631,13 @@ static __be32 encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, if (unlikely(status != 0)) goto out; - status = encode_attr_bitmap(xdr, res->bitmap, &savep); + status = encode_attr_bitmap(xdr, res->bitmap, ARRAY_SIZE(res->bitmap)); if (unlikely(status != 0)) goto out; + status = cpu_to_be32(NFS4ERR_RESOURCE); + savep = xdr_reserve_space(xdr, sizeof(*savep)); + if (unlikely(!savep)) + goto out; status = encode_attr_change(xdr, res->bitmap, res->change_attr); if (unlikely(status != 0)) goto out; diff --git a/fs/nfs/client.c b/fs/nfs/client.c index b9129e2befea..bbc91d7ca1bd 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1067,7 +1067,6 @@ void nfs_clients_init(struct net *net) } #ifdef CONFIG_PROC_FS -static int nfs_server_list_open(struct inode *inode, struct file *file); static void *nfs_server_list_start(struct seq_file *p, loff_t *pos); static void *nfs_server_list_next(struct seq_file *p, void *v, loff_t *pos); static void nfs_server_list_stop(struct seq_file *p, void *v); @@ -1080,14 +1079,6 @@ static const struct seq_operations nfs_server_list_ops = { .show = nfs_server_list_show, }; -static const struct file_operations nfs_server_list_fops = { - .open = nfs_server_list_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_net, -}; - -static int nfs_volume_list_open(struct inode *inode, struct file *file); static void *nfs_volume_list_start(struct seq_file *p, loff_t *pos); static void *nfs_volume_list_next(struct seq_file *p, void *v, loff_t *pos); static void nfs_volume_list_stop(struct seq_file *p, void *v); @@ -1100,23 +1091,6 @@ static const struct seq_operations nfs_volume_list_ops = { .show = nfs_volume_list_show, }; -static const struct file_operations nfs_volume_list_fops = { - .open = nfs_volume_list_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release_net, -}; - -/* - * open "/proc/fs/nfsfs/servers" which provides a summary of servers with which - * we're dealing - */ -static int nfs_server_list_open(struct inode *inode, struct file *file) -{ - return seq_open_net(inode, file, &nfs_server_list_ops, - sizeof(struct seq_net_private)); -} - /* * set up the iterator to start reading from the server list and return the first item */ @@ -1185,15 +1159,6 @@ static int nfs_server_list_show(struct seq_file *m, void *v) } /* - * open "/proc/fs/nfsfs/volumes" which provides a summary of extant volumes - */ -static int nfs_volume_list_open(struct inode *inode, struct file *file) -{ - return seq_open_net(inode, file, &nfs_volume_list_ops, - sizeof(struct seq_net_private)); -} - -/* * set up the iterator to start reading from the volume list and return the first item */ static void *nfs_volume_list_start(struct seq_file *m, loff_t *_pos) @@ -1278,14 +1243,14 @@ int nfs_fs_proc_net_init(struct net *net) goto error_0; /* a file of servers with which we're dealing */ - p = proc_create("servers", S_IFREG|S_IRUGO, - nn->proc_nfsfs, &nfs_server_list_fops); + p = proc_create_net("servers", S_IFREG|S_IRUGO, nn->proc_nfsfs, + &nfs_server_list_ops, sizeof(struct seq_net_private)); if (!p) goto error_1; /* a file of volumes that we have mounted */ - p = proc_create("volumes", S_IFREG|S_IRUGO, - nn->proc_nfsfs, &nfs_volume_list_fops); + p = proc_create_net("volumes", S_IFREG|S_IRUGO, nn->proc_nfsfs, + &nfs_volume_list_ops, sizeof(struct seq_net_private)); if (!p) goto error_1; return 0; diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index d8b47624fee2..1819d0d0ba4b 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -19,6 +19,7 @@ #include <linux/nfs_xdr.h> #include "nfs4_fs.h" +#include "nfs4session.h" #include "delegation.h" #include "internal.h" #include "nfs4trace.h" @@ -171,11 +172,15 @@ again: * nfs_inode_reclaim_delegation - process a delegation reclaim request * @inode: inode to process * @cred: credential to use for request - * @res: new delegation state from server + * @type: delegation type + * @stateid: delegation stateid + * @pagemod_limit: write delegation "space_limit" * */ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, - struct nfs_openres *res) + fmode_t type, + const nfs4_stateid *stateid, + unsigned long pagemod_limit) { struct nfs_delegation *delegation; struct rpc_cred *oldcred = NULL; @@ -185,9 +190,9 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, if (delegation != NULL) { spin_lock(&delegation->lock); if (delegation->inode != NULL) { - nfs4_stateid_copy(&delegation->stateid, &res->delegation); - delegation->type = res->delegation_type; - delegation->pagemod_limit = res->pagemod_limit; + nfs4_stateid_copy(&delegation->stateid, stateid); + delegation->type = type; + delegation->pagemod_limit = pagemod_limit; oldcred = delegation->cred; delegation->cred = get_rpccred(cred); clear_bit(NFS_DELEGATION_NEED_RECLAIM, @@ -195,14 +200,14 @@ void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, spin_unlock(&delegation->lock); rcu_read_unlock(); put_rpccred(oldcred); - trace_nfs4_reclaim_delegation(inode, res->delegation_type); + trace_nfs4_reclaim_delegation(inode, type); return; } /* We appear to have raced with a delegation return. */ spin_unlock(&delegation->lock); } rcu_read_unlock(); - nfs_inode_set_delegation(inode, cred, res); + nfs_inode_set_delegation(inode, cred, type, stateid, pagemod_limit); } static int nfs_do_return_delegation(struct inode *inode, struct nfs_delegation *delegation, int issync) @@ -329,11 +334,16 @@ nfs_update_inplace_delegation(struct nfs_delegation *delegation, * nfs_inode_set_delegation - set up a delegation on an inode * @inode: inode to which delegation applies * @cred: cred to use for subsequent delegation processing - * @res: new delegation state from server + * @type: delegation type + * @stateid: delegation stateid + * @pagemod_limit: write delegation "space_limit" * * Returns zero on success, or a negative errno value. */ -int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res) +int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, + fmode_t type, + const nfs4_stateid *stateid, + unsigned long pagemod_limit) { struct nfs_server *server = NFS_SERVER(inode); struct nfs_client *clp = server->nfs_client; @@ -345,9 +355,9 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct delegation = kmalloc(sizeof(*delegation), GFP_NOFS); if (delegation == NULL) return -ENOMEM; - nfs4_stateid_copy(&delegation->stateid, &res->delegation); - delegation->type = res->delegation_type; - delegation->pagemod_limit = res->pagemod_limit; + nfs4_stateid_copy(&delegation->stateid, stateid); + delegation->type = type; + delegation->pagemod_limit = pagemod_limit; delegation->change_attr = inode_peek_iversion_raw(inode); delegation->cred = get_rpccred(cred); delegation->inode = inode; @@ -392,7 +402,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct rcu_assign_pointer(nfsi->delegation, delegation); delegation = NULL; - trace_nfs4_set_delegation(inode, res->delegation_type); + trace_nfs4_set_delegation(inode, type); out: spin_unlock(&clp->cl_lock); @@ -547,6 +557,22 @@ int nfs4_inode_return_delegation(struct inode *inode) return err; } +/** + * nfs4_inode_make_writeable + * @inode: pointer to inode + * + * Make the inode writeable by returning the delegation if necessary + * + * Returns zero on success, or a negative errno value. + */ +int nfs4_inode_make_writeable(struct inode *inode) +{ + if (!nfs4_has_session(NFS_SERVER(inode)->nfs_client) || + !nfs4_check_delegation(inode, FMODE_WRITE)) + return nfs4_inode_return_delegation(inode); + return 0; +} + static void nfs_mark_return_if_closed_delegation(struct nfs_server *server, struct nfs_delegation *delegation) { diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 185a09f37a89..bb1ef8c37af4 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -36,8 +36,10 @@ enum { NFS_DELEGATION_TEST_EXPIRED, }; -int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res); -void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res); +int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, + fmode_t type, const nfs4_stateid *stateid, unsigned long pagemod_limit); +void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, + fmode_t type, const nfs4_stateid *stateid, unsigned long pagemod_limit); int nfs4_inode_return_delegation(struct inode *inode); int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid); void nfs_inode_return_delegation_noreclaim(struct inode *inode); @@ -70,6 +72,7 @@ int nfs4_check_delegation(struct inode *inode, fmode_t flags); bool nfs4_delegation_flush_on_close(const struct inode *inode); void nfs_inode_find_delegation_state_and_recover(struct inode *inode, const nfs4_stateid *stateid); +int nfs4_inode_make_writeable(struct inode *inode); #endif diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 2f3f86726f5b..73f8b43d988c 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1272,7 +1272,9 @@ static void nfs_drop_nlink(struct inode *inode) /* drop the inode if we're reasonably sure this is the last link */ if (inode->i_nlink == 1) clear_nlink(inode); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR; + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME + | NFS_INO_INVALID_OTHER; spin_unlock(&inode->i_lock); } @@ -1798,12 +1800,11 @@ static int nfs_safe_remove(struct dentry *dentry) trace_nfs_remove_enter(dir, dentry); if (inode != NULL) { - NFS_PROTO(inode)->return_delegation(inode); - error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); + error = NFS_PROTO(dir)->remove(dir, dentry); if (error == 0) nfs_drop_nlink(inode); } else - error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); + error = NFS_PROTO(dir)->remove(dir, dentry); if (error == -ENOENT) nfs_dentry_handle_enoent(dentry); trace_nfs_remove_exit(dir, dentry, error); @@ -1932,8 +1933,6 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) old_dentry, dentry); trace_nfs_link_enter(inode, dir, dentry); - NFS_PROTO(inode)->return_delegation(inode); - d_drop(dentry); error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name); if (error == 0) { @@ -2023,10 +2022,6 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, } } - NFS_PROTO(old_inode)->return_delegation(old_inode); - if (new_inode != NULL) - NFS_PROTO(new_inode)->return_delegation(new_inode); - task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry, NULL); if (IS_ERR(task)) { error = PTR_ERR(task); diff --git a/fs/nfs/fscache-index.c b/fs/nfs/fscache-index.c index 0ee4b93d36ea..1c5d8d31fc0a 100644 --- a/fs/nfs/fscache-index.c +++ b/fs/nfs/fscache-index.c @@ -50,59 +50,6 @@ void nfs_fscache_unregister(void) } /* - * Layout of the key for an NFS server cache object. - */ -struct nfs_server_key { - uint16_t nfsversion; /* NFS protocol version */ - uint16_t family; /* address family */ - uint16_t port; /* IP port */ - union { - struct in_addr ipv4_addr; /* IPv4 address */ - struct in6_addr ipv6_addr; /* IPv6 address */ - } addr[0]; -}; - -/* - * Generate a key to describe a server in the main NFS index - * - We return the length of the key, or 0 if we can't generate one - */ -static uint16_t nfs_server_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct nfs_client *clp = cookie_netfs_data; - const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) &clp->cl_addr; - const struct sockaddr_in *sin = (struct sockaddr_in *) &clp->cl_addr; - struct nfs_server_key *key = buffer; - uint16_t len = sizeof(struct nfs_server_key); - - memset(key, 0, len); - key->nfsversion = clp->rpc_ops->version; - key->family = clp->cl_addr.ss_family; - - switch (clp->cl_addr.ss_family) { - case AF_INET: - key->port = sin->sin_port; - key->addr[0].ipv4_addr = sin->sin_addr; - len += sizeof(key->addr[0].ipv4_addr); - break; - - case AF_INET6: - key->port = sin6->sin6_port; - key->addr[0].ipv6_addr = sin6->sin6_addr; - len += sizeof(key->addr[0].ipv6_addr); - break; - - default: - printk(KERN_WARNING "NFS: Unknown network family '%d'\n", - clp->cl_addr.ss_family); - len = 0; - break; - } - - return len; -} - -/* * Define the server object for FS-Cache. This is used to describe a server * object to fscache_acquire_cookie(). It is keyed by the NFS protocol and * server address parameters. @@ -110,33 +57,9 @@ static uint16_t nfs_server_get_key(const void *cookie_netfs_data, const struct fscache_cookie_def nfs_fscache_server_index_def = { .name = "NFS.server", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = nfs_server_get_key, }; /* - * Generate a key to describe a superblock key in the main NFS index - */ -static uint16_t nfs_super_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct nfs_fscache_key *key; - const struct nfs_server *nfss = cookie_netfs_data; - uint16_t len; - - key = nfss->fscache_key; - len = sizeof(key->key) + key->key.uniq_len; - if (len > bufmax) { - len = 0; - } else { - memcpy(buffer, &key->key, sizeof(key->key)); - memcpy(buffer + sizeof(key->key), - key->key.uniquifier, key->key.uniq_len); - } - - return len; -} - -/* * Define the superblock object for FS-Cache. This is used to describe a * superblock object to fscache_acquire_cookie(). It is keyed by all the NFS * parameters that might cause a separate superblock. @@ -144,84 +67,9 @@ static uint16_t nfs_super_get_key(const void *cookie_netfs_data, const struct fscache_cookie_def nfs_fscache_super_index_def = { .name = "NFS.super", .type = FSCACHE_COOKIE_TYPE_INDEX, - .get_key = nfs_super_get_key, }; /* - * Definition of the auxiliary data attached to NFS inode storage objects - * within the cache. - * - * The contents of this struct are recorded in the on-disk local cache in the - * auxiliary data attached to the data storage object backing an inode. This - * permits coherency to be managed when a new inode binds to an already extant - * cache object. - */ -struct nfs_fscache_inode_auxdata { - struct timespec mtime; - struct timespec ctime; - loff_t size; - u64 change_attr; -}; - -/* - * Generate a key to describe an NFS inode in an NFS server's index - */ -static uint16_t nfs_fscache_inode_get_key(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - const struct nfs_inode *nfsi = cookie_netfs_data; - uint16_t nsize; - - /* use the inode's NFS filehandle as the key */ - nsize = nfsi->fh.size; - memcpy(buffer, nfsi->fh.data, nsize); - return nsize; -} - -/* - * Get certain file attributes from the netfs data - * - This function can be absent for an index - * - Not permitted to return an error - * - The netfs data from the cookie being used as the source is presented - */ -static void nfs_fscache_inode_get_attr(const void *cookie_netfs_data, - uint64_t *size) -{ - const struct nfs_inode *nfsi = cookie_netfs_data; - - *size = nfsi->vfs_inode.i_size; -} - -/* - * Get the auxiliary data from netfs data - * - This function can be absent if the index carries no state data - * - Should store the auxiliary data in the buffer - * - Should return the amount of amount stored - * - Not permitted to return an error - * - The netfs data from the cookie being used as the source is presented - */ -static uint16_t nfs_fscache_inode_get_aux(const void *cookie_netfs_data, - void *buffer, uint16_t bufmax) -{ - struct nfs_fscache_inode_auxdata auxdata; - const struct nfs_inode *nfsi = cookie_netfs_data; - - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.size = nfsi->vfs_inode.i_size; - auxdata.mtime = nfsi->vfs_inode.i_mtime; - auxdata.ctime = nfsi->vfs_inode.i_ctime; - - if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4) - auxdata.change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode); - - if (bufmax > sizeof(auxdata)) - bufmax = sizeof(auxdata); - - memcpy(buffer, &auxdata, bufmax); - return bufmax; -} - -/* * Consult the netfs about the state of an object * - This function can be absent if the index carries no state data * - The netfs data from the cookie being used as the target is @@ -230,7 +78,8 @@ static uint16_t nfs_fscache_inode_get_aux(const void *cookie_netfs_data, static enum fscache_checkaux nfs_fscache_inode_check_aux(void *cookie_netfs_data, const void *data, - uint16_t datalen) + uint16_t datalen, + loff_t object_size) { struct nfs_fscache_inode_auxdata auxdata; struct nfs_inode *nfsi = cookie_netfs_data; @@ -239,7 +88,6 @@ enum fscache_checkaux nfs_fscache_inode_check_aux(void *cookie_netfs_data, return FSCACHE_CHECKAUX_OBSOLETE; memset(&auxdata, 0, sizeof(auxdata)); - auxdata.size = nfsi->vfs_inode.i_size; auxdata.mtime = nfsi->vfs_inode.i_mtime; auxdata.ctime = nfsi->vfs_inode.i_ctime; @@ -288,9 +136,6 @@ static void nfs_fh_put_context(void *cookie_netfs_data, void *context) const struct fscache_cookie_def nfs_fscache_inode_object_def = { .name = "NFS.fh", .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .get_key = nfs_fscache_inode_get_key, - .get_attr = nfs_fscache_inode_get_attr, - .get_aux = nfs_fscache_inode_get_aux, .check_aux = nfs_fscache_inode_check_aux, .get_context = nfs_fh_get_context, .put_context = nfs_fh_put_context, diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c index d63bea8bbfbb..b55fc7920c3b 100644 --- a/fs/nfs/fscache.c +++ b/fs/nfs/fscache.c @@ -18,6 +18,7 @@ #include <linux/in6.h> #include <linux/seq_file.h> #include <linux/slab.h> +#include <linux/iversion.h> #include "internal.h" #include "iostat.h" @@ -29,6 +30,21 @@ static struct rb_root nfs_fscache_keys = RB_ROOT; static DEFINE_SPINLOCK(nfs_fscache_keys_lock); /* + * Layout of the key for an NFS server cache object. + */ +struct nfs_server_key { + struct { + uint16_t nfsversion; /* NFS protocol version */ + uint16_t family; /* address family */ + __be16 port; /* IP port */ + } hdr; + union { + struct in_addr ipv4_addr; /* IPv4 address */ + struct in6_addr ipv6_addr; /* IPv6 address */ + }; +} __packed; + +/* * Get the per-client index cookie for an NFS client if the appropriate mount * flag was set * - We always try and get an index cookie for the client, but get filehandle @@ -36,10 +52,41 @@ static DEFINE_SPINLOCK(nfs_fscache_keys_lock); */ void nfs_fscache_get_client_cookie(struct nfs_client *clp) { + const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) &clp->cl_addr; + const struct sockaddr_in *sin = (struct sockaddr_in *) &clp->cl_addr; + struct nfs_server_key key; + uint16_t len = sizeof(key.hdr); + + memset(&key, 0, sizeof(key)); + key.hdr.nfsversion = clp->rpc_ops->version; + key.hdr.family = clp->cl_addr.ss_family; + + switch (clp->cl_addr.ss_family) { + case AF_INET: + key.hdr.port = sin->sin_port; + key.ipv4_addr = sin->sin_addr; + len += sizeof(key.ipv4_addr); + break; + + case AF_INET6: + key.hdr.port = sin6->sin6_port; + key.ipv6_addr = sin6->sin6_addr; + len += sizeof(key.ipv6_addr); + break; + + default: + printk(KERN_WARNING "NFS: Unknown network family '%d'\n", + clp->cl_addr.ss_family); + clp->fscache = NULL; + return; + } + /* create a cache index for looking up filehandles */ clp->fscache = fscache_acquire_cookie(nfs_fscache_netfs.primary_index, &nfs_fscache_server_index_def, - clp, true); + &key, len, + NULL, 0, + clp, 0, true); dfprintk(FSCACHE, "NFS: get client cookie (0x%p/0x%p)\n", clp, clp->fscache); } @@ -52,7 +99,7 @@ void nfs_fscache_release_client_cookie(struct nfs_client *clp) dfprintk(FSCACHE, "NFS: releasing client cookie (0x%p/0x%p)\n", clp, clp->fscache); - fscache_relinquish_cookie(clp->fscache, 0); + fscache_relinquish_cookie(clp->fscache, NULL, false); clp->fscache = NULL; } @@ -139,7 +186,9 @@ void nfs_fscache_get_super_cookie(struct super_block *sb, const char *uniq, int /* create a cache index for looking up filehandles */ nfss->fscache = fscache_acquire_cookie(nfss->nfs_client->fscache, &nfs_fscache_super_index_def, - nfss, true); + key, sizeof(*key) + ulen, + NULL, 0, + nfss, 0, true); dfprintk(FSCACHE, "NFS: get superblock cookie (0x%p/0x%p)\n", nfss, nfss->fscache); return; @@ -163,7 +212,7 @@ void nfs_fscache_release_super_cookie(struct super_block *sb) dfprintk(FSCACHE, "NFS: releasing superblock cookie (0x%p/0x%p)\n", nfss, nfss->fscache); - fscache_relinquish_cookie(nfss->fscache, 0); + fscache_relinquish_cookie(nfss->fscache, NULL, false); nfss->fscache = NULL; if (nfss->fscache_key) { @@ -180,14 +229,25 @@ void nfs_fscache_release_super_cookie(struct super_block *sb) */ void nfs_fscache_init_inode(struct inode *inode) { + struct nfs_fscache_inode_auxdata auxdata; struct nfs_inode *nfsi = NFS_I(inode); nfsi->fscache = NULL; if (!S_ISREG(inode->i_mode)) return; + + memset(&auxdata, 0, sizeof(auxdata)); + auxdata.mtime = nfsi->vfs_inode.i_mtime; + auxdata.ctime = nfsi->vfs_inode.i_ctime; + + if (NFS_SERVER(&nfsi->vfs_inode)->nfs_client->rpc_ops->version == 4) + auxdata.change_attr = inode_peek_iversion_raw(&nfsi->vfs_inode); + nfsi->fscache = fscache_acquire_cookie(NFS_SB(inode->i_sb)->fscache, &nfs_fscache_inode_object_def, - nfsi, false); + nfsi->fh.data, nfsi->fh.size, + &auxdata, sizeof(auxdata), + nfsi, nfsi->vfs_inode.i_size, false); } /* @@ -195,12 +255,16 @@ void nfs_fscache_init_inode(struct inode *inode) */ void nfs_fscache_clear_inode(struct inode *inode) { + struct nfs_fscache_inode_auxdata auxdata; struct nfs_inode *nfsi = NFS_I(inode); struct fscache_cookie *cookie = nfs_i_fscache(inode); dfprintk(FSCACHE, "NFS: clear cookie (0x%p/0x%p)\n", nfsi, cookie); - fscache_relinquish_cookie(cookie, false); + memset(&auxdata, 0, sizeof(auxdata)); + auxdata.mtime = nfsi->vfs_inode.i_mtime; + auxdata.ctime = nfsi->vfs_inode.i_ctime; + fscache_relinquish_cookie(cookie, &auxdata, false); nfsi->fscache = NULL; } @@ -232,20 +296,26 @@ static bool nfs_fscache_can_enable(void *data) */ void nfs_fscache_open_file(struct inode *inode, struct file *filp) { + struct nfs_fscache_inode_auxdata auxdata; struct nfs_inode *nfsi = NFS_I(inode); struct fscache_cookie *cookie = nfs_i_fscache(inode); if (!fscache_cookie_valid(cookie)) return; + memset(&auxdata, 0, sizeof(auxdata)); + auxdata.mtime = nfsi->vfs_inode.i_mtime; + auxdata.ctime = nfsi->vfs_inode.i_ctime; + if (inode_is_open_for_write(inode)) { dfprintk(FSCACHE, "NFS: nfsi 0x%p disabling cache\n", nfsi); clear_bit(NFS_INO_FSCACHE, &nfsi->flags); - fscache_disable_cookie(cookie, true); + fscache_disable_cookie(cookie, &auxdata, true); fscache_uncache_all_inode_pages(cookie, inode); } else { dfprintk(FSCACHE, "NFS: nfsi 0x%p enabling cache\n", nfsi); - fscache_enable_cookie(cookie, nfs_fscache_can_enable, inode); + fscache_enable_cookie(cookie, &auxdata, nfsi->vfs_inode.i_size, + nfs_fscache_can_enable, inode); if (fscache_cookie_enabled(cookie)) set_bit(NFS_INO_FSCACHE, &NFS_I(inode)->flags); } @@ -422,7 +492,8 @@ void __nfs_readpage_to_fscache(struct inode *inode, struct page *page, int sync) "NFS: readpage_to_fscache(fsc:%p/p:%p(i:%lx f:%lx)/%d)\n", nfs_i_fscache(inode), page, page->index, page->flags, sync); - ret = fscache_write_page(nfs_i_fscache(inode), page, GFP_KERNEL); + ret = fscache_write_page(nfs_i_fscache(inode), page, + inode->i_size, GFP_KERNEL); dfprintk(FSCACHE, "NFS: readpage_to_fscache: p:%p(i:%lu f:%lx) ret %d\n", page, page->index, page->flags, ret); diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h index d7fe3e799f2f..161ba2edb9d0 100644 --- a/fs/nfs/fscache.h +++ b/fs/nfs/fscache.h @@ -57,6 +57,21 @@ struct nfs_fscache_key { }; /* + * Definition of the auxiliary data attached to NFS inode storage objects + * within the cache. + * + * The contents of this struct are recorded in the on-disk local cache in the + * auxiliary data attached to the data storage object backing an inode. This + * permits coherency to be managed when a new inode binds to an already extant + * cache object. + */ +struct nfs_fscache_inode_auxdata { + struct timespec mtime; + struct timespec ctime; + u64 change_attr; +}; + +/* * fscache-index.c */ extern struct fscache_netfs nfs_fscache_netfs; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index d17a90c4fa37..bd15d0b57626 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -195,7 +195,10 @@ bool nfs_check_cache_invalid(struct inode *inode, unsigned long flags) static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) { struct nfs_inode *nfsi = NFS_I(inode); + bool have_delegation = nfs_have_delegated_attributes(inode); + if (have_delegation) + flags &= ~(NFS_INO_INVALID_CHANGE|NFS_INO_REVAL_PAGECACHE); if (inode->i_mapping->nrpages == 0) flags &= ~NFS_INO_INVALID_DATA; nfsi->cache_validity |= flags; @@ -447,7 +450,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st inode->i_mode = fattr->mode; if ((fattr->valid & NFS_ATTR_FATTR_MODE) == 0 && nfs_server_capable(inode, NFS_CAP_MODE)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); /* Why so? Because we want revalidate for devices/FIFOs, and * that's precisely what we have in nfs_file_inode_operations. */ @@ -493,37 +496,35 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st if (fattr->valid & NFS_ATTR_FATTR_ATIME) inode->i_atime = fattr->atime; else if (nfs_server_capable(inode, NFS_CAP_ATIME)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME); if (fattr->valid & NFS_ATTR_FATTR_MTIME) inode->i_mtime = fattr->mtime; else if (nfs_server_capable(inode, NFS_CAP_MTIME)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); if (fattr->valid & NFS_ATTR_FATTR_CTIME) inode->i_ctime = fattr->ctime; else if (nfs_server_capable(inode, NFS_CAP_CTIME)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CTIME); if (fattr->valid & NFS_ATTR_FATTR_CHANGE) inode_set_iversion_raw(inode, fattr->change_attr); else - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR - | NFS_INO_REVAL_PAGECACHE); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE); if (fattr->valid & NFS_ATTR_FATTR_SIZE) inode->i_size = nfs_size_to_loff_t(fattr->size); else - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR - | NFS_INO_REVAL_PAGECACHE); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_SIZE); if (fattr->valid & NFS_ATTR_FATTR_NLINK) set_nlink(inode, fattr->nlink); else if (nfs_server_capable(inode, NFS_CAP_NLINK)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); if (fattr->valid & NFS_ATTR_FATTR_OWNER) inode->i_uid = fattr->uid; else if (nfs_server_capable(inode, NFS_CAP_OWNER)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); if (fattr->valid & NFS_ATTR_FATTR_GROUP) inode->i_gid = fattr->gid; else if (nfs_server_capable(inode, NFS_CAP_OWNER_GROUP)) - nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED) inode->i_blocks = fattr->du.nfs2.blocks; if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) { @@ -608,11 +609,6 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr) goto out; } - /* - * Return any delegations if we're going to change ACLs - */ - if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) - NFS_PROTO(inode)->return_delegation(inode); error = NFS_PROTO(inode)->setattr(dentry, fattr, attr); if (error == 0) error = nfs_refresh_inode(inode, fattr); @@ -645,6 +641,7 @@ static int nfs_vmtruncate(struct inode * inode, loff_t offset) /* Optimisation */ if (offset == 0) NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_DATA; + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; spin_unlock(&inode->i_lock); truncate_pagecache(inode, offset); @@ -657,6 +654,7 @@ out: * nfs_setattr_update_inode - Update inode metadata after a setattr call. * @inode: pointer to struct inode * @attr: pointer to struct iattr + * @fattr: pointer to struct nfs_fattr * * Note: we do this in the *proc.c in order to ensure that * it works for things like exclusive creates too. @@ -669,6 +667,8 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr, spin_lock(&inode->i_lock); NFS_I(inode)->attr_gencount = fattr->gencount; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) { if ((attr->ia_valid & ATTR_MODE) != 0) { int mode = attr->ia_mode & S_IALLUGO; @@ -683,13 +683,12 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr, | NFS_INO_INVALID_ACL); } if ((attr->ia_valid & ATTR_SIZE) != 0) { + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC); nfs_vmtruncate(inode, attr->ia_size); } if (fattr->valid) nfs_update_inode(inode, fattr); - else - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR; spin_unlock(&inode->i_lock); } EXPORT_SYMBOL_GPL(nfs_setattr_update_inode); @@ -1303,24 +1302,20 @@ static bool nfs_file_has_buffered_writers(struct nfs_inode *nfsi) return nfs_file_has_writers(nfsi) && nfs_file_io_is_buffered(nfsi); } -static unsigned long nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) +static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr) { - unsigned long ret = 0; - if ((fattr->valid & NFS_ATTR_FATTR_PRECHANGE) && (fattr->valid & NFS_ATTR_FATTR_CHANGE) && inode_eq_iversion_raw(inode, fattr->pre_change_attr)) { inode_set_iversion_raw(inode, fattr->change_attr); if (S_ISDIR(inode->i_mode)) nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); - ret |= NFS_INO_INVALID_ATTR; } /* If we have atomic WCC data, we may update some attributes */ if ((fattr->valid & NFS_ATTR_FATTR_PRECTIME) && (fattr->valid & NFS_ATTR_FATTR_CTIME) && timespec_equal(&inode->i_ctime, &fattr->pre_ctime)) { memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); - ret |= NFS_INO_INVALID_ATTR; } if ((fattr->valid & NFS_ATTR_FATTR_PREMTIME) @@ -1329,17 +1324,13 @@ static unsigned long nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); if (S_ISDIR(inode->i_mode)) nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); - ret |= NFS_INO_INVALID_ATTR; } if ((fattr->valid & NFS_ATTR_FATTR_PRESIZE) && (fattr->valid & NFS_ATTR_FATTR_SIZE) && i_size_read(inode) == nfs_size_to_loff_t(fattr->pre_size) && !nfs_have_writebacks(inode)) { i_size_write(inode, nfs_size_to_loff_t(fattr->size)); - ret |= NFS_INO_INVALID_ATTR; } - - return ret; } /** @@ -1369,33 +1360,41 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat if (!nfs_file_has_buffered_writers(nfsi)) { /* Verify a few of the more important attributes */ if ((fattr->valid & NFS_ATTR_FATTR_CHANGE) != 0 && !inode_eq_iversion_raw(inode, fattr->change_attr)) - invalid |= NFS_INO_INVALID_ATTR | NFS_INO_REVAL_PAGECACHE; + invalid |= NFS_INO_INVALID_CHANGE + | NFS_INO_REVAL_PAGECACHE; if ((fattr->valid & NFS_ATTR_FATTR_MTIME) && !timespec_equal(&inode->i_mtime, &fattr->mtime)) - invalid |= NFS_INO_INVALID_ATTR; + invalid |= NFS_INO_INVALID_MTIME; if ((fattr->valid & NFS_ATTR_FATTR_CTIME) && !timespec_equal(&inode->i_ctime, &fattr->ctime)) - invalid |= NFS_INO_INVALID_ATTR; + invalid |= NFS_INO_INVALID_CTIME; if (fattr->valid & NFS_ATTR_FATTR_SIZE) { cur_size = i_size_read(inode); new_isize = nfs_size_to_loff_t(fattr->size); if (cur_size != new_isize) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; + invalid |= NFS_INO_INVALID_SIZE + | NFS_INO_REVAL_PAGECACHE; } } /* Have any file permissions changed? */ if ((fattr->valid & NFS_ATTR_FATTR_MODE) && (inode->i_mode & S_IALLUGO) != (fattr->mode & S_IALLUGO)) - invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER; if ((fattr->valid & NFS_ATTR_FATTR_OWNER) && !uid_eq(inode->i_uid, fattr->uid)) - invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER; if ((fattr->valid & NFS_ATTR_FATTR_GROUP) && !gid_eq(inode->i_gid, fattr->gid)) - invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER; /* Has the link count changed? */ if ((fattr->valid & NFS_ATTR_FATTR_NLINK) && inode->i_nlink != fattr->nlink) - invalid |= NFS_INO_INVALID_ATTR; + invalid |= NFS_INO_INVALID_OTHER; if ((fattr->valid & NFS_ATTR_FATTR_ATIME) && !timespec_equal(&inode->i_atime, &fattr->atime)) invalid |= NFS_INO_INVALID_ATIME; @@ -1597,10 +1596,9 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) } EXPORT_SYMBOL_GPL(nfs_refresh_inode); -static int nfs_post_op_update_inode_locked(struct inode *inode, struct nfs_fattr *fattr) +static int nfs_post_op_update_inode_locked(struct inode *inode, + struct nfs_fattr *fattr, unsigned int invalid) { - unsigned long invalid = NFS_INO_INVALID_ATTR; - if (S_ISDIR(inode->i_mode)) invalid |= NFS_INO_INVALID_DATA; nfs_set_cache_invalid(inode, invalid); @@ -1629,7 +1627,9 @@ int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr) spin_lock(&inode->i_lock); nfs_fattr_set_barrier(fattr); - status = nfs_post_op_update_inode_locked(inode, fattr); + status = nfs_post_op_update_inode_locked(inode, fattr, + NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); spin_unlock(&inode->i_lock); return status; @@ -1681,7 +1681,10 @@ int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fa fattr->valid |= NFS_ATTR_FATTR_PRESIZE; } out_noforce: - status = nfs_post_op_update_inode_locked(inode, fattr); + status = nfs_post_op_update_inode_locked(inode, fattr, + NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME + | NFS_INO_INVALID_MTIME); return status; } @@ -1789,7 +1792,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) | NFS_INO_REVAL_PAGECACHE); /* Do atomic weak cache consistency updates */ - invalid |= nfs_wcc_update_inode(inode, fattr); + nfs_wcc_update_inode(inode, fattr); if (pnfs_layoutcommit_outstanding(inode)) { nfsi->cache_validity |= save_cache_validity & NFS_INO_INVALID_ATTR; @@ -1803,17 +1806,25 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) inode->i_sb->s_id, inode->i_ino); /* Could it be a race with writeback? */ if (!have_writers) { - invalid |= NFS_INO_INVALID_ATTR + invalid |= NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_DATA | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; + /* Force revalidate of all attributes */ + save_cache_validity |= NFS_INO_INVALID_CTIME + | NFS_INO_INVALID_MTIME + | NFS_INO_INVALID_SIZE + | NFS_INO_INVALID_OTHER; if (S_ISDIR(inode->i_mode)) nfs_force_lookup_revalidate(inode); } inode_set_iversion_raw(inode, fattr->change_attr); } } else { - nfsi->cache_validity |= save_cache_validity; + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_CHANGE + | NFS_INO_REVAL_PAGECACHE + | NFS_INO_REVAL_FORCED); cache_revalidated = false; } @@ -1821,7 +1832,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); } else if (server->caps & NFS_CAP_MTIME) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ATTR + (NFS_INO_INVALID_MTIME | NFS_INO_REVAL_FORCED); cache_revalidated = false; } @@ -1830,7 +1841,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); } else if (server->caps & NFS_CAP_CTIME) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ATTR + (NFS_INO_INVALID_CTIME | NFS_INO_REVAL_FORCED); cache_revalidated = false; } @@ -1845,7 +1856,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (!nfs_have_writebacks(inode) || new_isize > cur_isize) { i_size_write(inode, new_isize); if (!have_writers) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + invalid |= NFS_INO_INVALID_DATA; } dprintk("NFS: isize change on server for file %s/%ld " "(%Ld to %Ld)\n", @@ -1856,7 +1867,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) } } else { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ATTR + (NFS_INO_INVALID_SIZE | NFS_INO_REVAL_PAGECACHE | NFS_INO_REVAL_FORCED); cache_revalidated = false; @@ -1877,55 +1888,61 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) umode_t newmode = inode->i_mode & S_IFMT; newmode |= fattr->mode & S_IALLUGO; inode->i_mode = newmode; - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER; } } else if (server->caps & NFS_CAP_MODE) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ATTR - | NFS_INO_INVALID_ACCESS + (NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); cache_revalidated = false; } if (fattr->valid & NFS_ATTR_FATTR_OWNER) { if (!uid_eq(inode->i_uid, fattr->uid)) { - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER; inode->i_uid = fattr->uid; } } else if (server->caps & NFS_CAP_OWNER) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ATTR - | NFS_INO_INVALID_ACCESS + (NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); cache_revalidated = false; } if (fattr->valid & NFS_ATTR_FATTR_GROUP) { if (!gid_eq(inode->i_gid, fattr->gid)) { - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER; inode->i_gid = fattr->gid; } } else if (server->caps & NFS_CAP_OWNER_GROUP) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ATTR - | NFS_INO_INVALID_ACCESS + (NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL + | NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); cache_revalidated = false; } if (fattr->valid & NFS_ATTR_FATTR_NLINK) { if (inode->i_nlink != fattr->nlink) { - invalid |= NFS_INO_INVALID_ATTR; + invalid |= NFS_INO_INVALID_OTHER; if (S_ISDIR(inode->i_mode)) invalid |= NFS_INO_INVALID_DATA; set_nlink(inode, fattr->nlink); } } else if (server->caps & NFS_CAP_NLINK) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ATTR + (NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); cache_revalidated = false; } @@ -1942,6 +1959,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) /* Update attrtimeo value if we're out of the unstable period */ if (invalid & NFS_INO_INVALID_ATTR) { + invalid &= ~NFS_INO_INVALID_ATTR; nfs_inc_stats(inode, NFSIOS_ATTRINVALIDATE); nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); nfsi->attrtimeo_timestamp = now; @@ -1962,10 +1980,6 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) nfsi->attr_gencount = fattr->gencount; } - /* Don't declare attrcache up to date if there were no attrs! */ - if (cache_revalidated) - invalid &= ~NFS_INO_INVALID_ATTR; - /* Don't invalidate the data if we were to blame */ if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 7327930ad970..eadf1ab31d16 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -138,8 +138,11 @@ nfs3_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, msg.rpc_cred = nfs_file_cred(sattr->ia_file); nfs_fattr_init(fattr); status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); - if (status == 0) + if (status == 0) { + if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL) + nfs_zap_acl_cache(inode); nfs_setattr_update_inode(inode, sattr, fattr); + } dprintk("NFS reply setattr: %d\n", status); return status; } @@ -383,11 +386,11 @@ out: } static int -nfs3_proc_remove(struct inode *dir, const struct qstr *name) +nfs3_proc_remove(struct inode *dir, struct dentry *dentry) { struct nfs_removeargs arg = { .fh = NFS_FH(dir), - .name = *name, + .name = dentry->d_name, }; struct nfs_removeres res; struct rpc_message msg = { @@ -397,7 +400,7 @@ nfs3_proc_remove(struct inode *dir, const struct qstr *name) }; int status = -ENOMEM; - dprintk("NFS call remove %s\n", name->name); + dprintk("NFS call remove %pd2\n", dentry); res.dir_attr = nfs_alloc_fattr(); if (res.dir_attr == NULL) goto out; @@ -411,7 +414,7 @@ out: } static void -nfs3_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) +nfs3_proc_unlink_setup(struct rpc_message *msg, struct dentry *dentry) { msg->rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE]; } @@ -433,7 +436,9 @@ nfs3_proc_unlink_done(struct rpc_task *task, struct inode *dir) } static void -nfs3_proc_rename_setup(struct rpc_message *msg, struct inode *dir) +nfs3_proc_rename_setup(struct rpc_message *msg, + struct dentry *old_dentry, + struct dentry *new_dentry) { msg->rpc_proc = &nfs3_procedures[NFS3PROC_RENAME]; } @@ -908,12 +913,6 @@ static int nfs3_have_delegation(struct inode *inode, fmode_t flags) return 0; } -static int nfs3_return_delegation(struct inode *inode) -{ - nfs_wb_all(inode); - return 0; -} - static const struct inode_operations nfs3_dir_inode_operations = { .create = nfs_create, .lookup = nfs_lookup, @@ -990,7 +989,6 @@ const struct nfs_rpc_ops nfs_v3_clientops = { .clear_acl_cache = forget_all_cached_acls, .close_context = nfs_close_context, .have_delegation = nfs3_have_delegation, - .return_delegation = nfs3_return_delegation, .alloc_client = nfs_alloc_client, .init_client = nfs_init_client, .free_client = nfs_free_client, diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 6cd33bd5da87..09ee36dd8426 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -1997,6 +1997,7 @@ int nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, struct nfs_entry old = *entry; __be32 *p; int error; + u64 new_cookie; p = xdr_inline_decode(xdr, 4); if (unlikely(p == NULL)) @@ -2019,8 +2020,7 @@ int nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, if (unlikely(error)) return error; - entry->prev_cookie = entry->cookie; - error = decode_cookie3(xdr, &entry->cookie); + error = decode_cookie3(xdr, &new_cookie); if (unlikely(error)) return error; @@ -2054,6 +2054,9 @@ int nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, zero_nfs_fh3(entry->fh); } + entry->prev_cookie = entry->cookie; + entry->cookie = new_cookie; + return 0; out_overflow: diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 47f3c273245e..b71757e85066 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1045,7 +1045,9 @@ static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo, struct nfs_inode *nfsi = NFS_I(dir); spin_lock(&dir->i_lock); - nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + nfsi->cache_validity |= NFS_INO_INVALID_CTIME + | NFS_INO_INVALID_MTIME + | NFS_INO_INVALID_DATA; if (cinfo->atomic && cinfo->before == inode_peek_iversion_raw(dir)) { nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE; nfsi->attrtimeo_timestamp = jiffies; @@ -1669,6 +1671,7 @@ static void nfs4_return_incompatible_delegation(struct inode *inode, fmode_t fmo { struct nfs_delegation *delegation; + fmode &= FMODE_READ|FMODE_WRITE; rcu_read_lock(); delegation = rcu_dereference(NFS_I(inode)->delegation); if (delegation == NULL || (delegation->type & fmode) == fmode) { @@ -1751,12 +1754,16 @@ nfs4_opendata_check_deleg(struct nfs4_opendata *data, struct nfs4_state *state) } if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0) nfs_inode_set_delegation(state->inode, - data->owner->so_cred, - &data->o_res); + data->owner->so_cred, + data->o_res.delegation_type, + &data->o_res.delegation, + data->o_res.pagemod_limit); else nfs_inode_reclaim_delegation(state->inode, - data->owner->so_cred, - &data->o_res); + data->owner->so_cred, + data->o_res.delegation_type, + &data->o_res.delegation, + data->o_res.pagemod_limit); } /* @@ -2743,27 +2750,40 @@ static int nfs41_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st * fields corresponding to attributes that were used to store the verifier. * Make sure we clobber those fields in the later setattr call */ -static inline void nfs4_exclusive_attrset(struct nfs4_opendata *opendata, +static unsigned nfs4_exclusive_attrset(struct nfs4_opendata *opendata, struct iattr *sattr, struct nfs4_label **label) { - const u32 *attrset = opendata->o_res.attrset; + const __u32 *bitmask = opendata->o_arg.server->exclcreat_bitmask; + __u32 attrset[3]; + unsigned ret; + unsigned i; - if ((attrset[1] & FATTR4_WORD1_TIME_ACCESS) && - !(sattr->ia_valid & ATTR_ATIME_SET)) - sattr->ia_valid |= ATTR_ATIME; + for (i = 0; i < ARRAY_SIZE(attrset); i++) { + attrset[i] = opendata->o_res.attrset[i]; + if (opendata->o_arg.createmode == NFS4_CREATE_EXCLUSIVE4_1) + attrset[i] &= ~bitmask[i]; + } + + ret = (opendata->o_arg.createmode == NFS4_CREATE_EXCLUSIVE) ? + sattr->ia_valid : 0; - if ((attrset[1] & FATTR4_WORD1_TIME_MODIFY) && - !(sattr->ia_valid & ATTR_MTIME_SET)) - sattr->ia_valid |= ATTR_MTIME; + if ((attrset[1] & (FATTR4_WORD1_TIME_ACCESS|FATTR4_WORD1_TIME_ACCESS_SET))) { + if (sattr->ia_valid & ATTR_ATIME_SET) + ret |= ATTR_ATIME_SET; + else + ret |= ATTR_ATIME; + } - /* Except MODE, it seems harmless of setting twice. */ - if (opendata->o_arg.createmode != NFS4_CREATE_EXCLUSIVE && - (attrset[1] & FATTR4_WORD1_MODE || - attrset[2] & FATTR4_WORD2_MODE_UMASK)) - sattr->ia_valid &= ~ATTR_MODE; + if ((attrset[1] & (FATTR4_WORD1_TIME_MODIFY|FATTR4_WORD1_TIME_MODIFY_SET))) { + if (sattr->ia_valid & ATTR_MTIME_SET) + ret |= ATTR_MTIME_SET; + else + ret |= ATTR_MTIME; + } - if (attrset[2] & FATTR4_WORD2_SECURITY_LABEL) + if (!(attrset[2] & FATTR4_WORD2_SECURITY_LABEL)) *label = NULL; + return ret; } static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata, @@ -2892,12 +2912,15 @@ static int _nfs4_do_open(struct inode *dir, if ((opendata->o_arg.open_flags & (O_CREAT|O_EXCL)) == (O_CREAT|O_EXCL) && (opendata->o_arg.createmode != NFS4_CREATE_GUARDED)) { - nfs4_exclusive_attrset(opendata, sattr, &label); + unsigned attrs = nfs4_exclusive_attrset(opendata, sattr, &label); /* * send create attributes which was not set by open * with an extra setattr. */ - if (sattr->ia_valid & NFS4_VALID_ATTRS) { + if (attrs || label) { + unsigned ia_old = sattr->ia_valid; + + sattr->ia_valid = attrs; nfs_fattr_init(opendata->o_res.f_attr); status = nfs4_do_setattr(state->inode, cred, opendata->o_res.f_attr, sattr, @@ -2907,6 +2930,7 @@ static int _nfs4_do_open(struct inode *dir, opendata->o_res.f_attr); nfs_setsecurity(state->inode, opendata->o_res.f_attr, olabel); } + sattr->ia_valid = ia_old; } } if (opened && opendata->file_created) @@ -3874,6 +3898,10 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, if (IS_ERR(label)) return PTR_ERR(label); + /* Return any delegations if we're going to change ACLs */ + if ((sattr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) + nfs4_inode_make_writeable(inode); + status = nfs4_do_setattr(inode, cred, fattr, sattr, ctx, NULL, label); if (status == 0) { nfs_setattr_update_inode(inode, sattr, fattr); @@ -4048,7 +4076,6 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry struct nfs_server *server = NFS_SERVER(inode); struct nfs4_accessargs args = { .fh = NFS_FH(inode), - .bitmask = server->cache_consistency_bitmask, .access = entry->mask, }; struct nfs4_accessres res = { @@ -4062,14 +4089,18 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry }; int status = 0; - res.fattr = nfs_alloc_fattr(); - if (res.fattr == NULL) - return -ENOMEM; + if (!nfs_have_delegated_attributes(inode)) { + res.fattr = nfs_alloc_fattr(); + if (res.fattr == NULL) + return -ENOMEM; + args.bitmask = server->cache_consistency_bitmask; + } status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); if (!status) { nfs_access_set_mask(entry, res.access); - nfs_refresh_inode(inode, res.fattr); + if (res.fattr) + nfs_refresh_inode(inode, res.fattr); } nfs_free_fattr(res.fattr); return status; @@ -4199,10 +4230,32 @@ static int _nfs4_proc_remove(struct inode *dir, const struct qstr *name) return status; } -static int nfs4_proc_remove(struct inode *dir, const struct qstr *name) +static int nfs4_proc_remove(struct inode *dir, struct dentry *dentry) +{ + struct nfs4_exception exception = { }; + struct inode *inode = d_inode(dentry); + int err; + + if (inode) { + if (inode->i_nlink == 1) + nfs4_inode_return_delegation(inode); + else + nfs4_inode_make_writeable(inode); + } + do { + err = _nfs4_proc_remove(dir, &dentry->d_name); + trace_nfs4_remove(dir, &dentry->d_name, err); + err = nfs4_handle_exception(NFS_SERVER(dir), err, + &exception); + } while (exception.retry); + return err; +} + +static int nfs4_proc_rmdir(struct inode *dir, const struct qstr *name) { struct nfs4_exception exception = { }; int err; + do { err = _nfs4_proc_remove(dir, name); trace_nfs4_remove(dir, name, err); @@ -4212,17 +4265,20 @@ static int nfs4_proc_remove(struct inode *dir, const struct qstr *name) return err; } -static void nfs4_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) +static void nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dentry) { - struct nfs_server *server = NFS_SERVER(dir); struct nfs_removeargs *args = msg->rpc_argp; struct nfs_removeres *res = msg->rpc_resp; + struct inode *inode = d_inode(dentry); - res->server = server; + res->server = NFS_SB(dentry->d_sb); msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE]; nfs4_init_sequence(&args->seq_args, &res->seq_res, 1); nfs_fattr_init(res->dir_attr); + + if (inode) + nfs4_inode_return_delegation(inode); } static void nfs4_proc_unlink_rpc_prepare(struct rpc_task *task, struct nfs_unlinkdata *data) @@ -4248,14 +4304,21 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir) return 1; } -static void nfs4_proc_rename_setup(struct rpc_message *msg, struct inode *dir) +static void nfs4_proc_rename_setup(struct rpc_message *msg, + struct dentry *old_dentry, + struct dentry *new_dentry) { - struct nfs_server *server = NFS_SERVER(dir); struct nfs_renameargs *arg = msg->rpc_argp; struct nfs_renameres *res = msg->rpc_resp; + struct inode *old_inode = d_inode(old_dentry); + struct inode *new_inode = d_inode(new_dentry); + if (old_inode) + nfs4_inode_make_writeable(old_inode); + if (new_inode) + nfs4_inode_return_delegation(new_inode); msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME]; - res->server = server; + res->server = NFS_SB(old_dentry->d_sb); nfs4_init_sequence(&arg->seq_args, &res->seq_res, 1); } @@ -4317,6 +4380,8 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct } arg.bitmask = nfs4_bitmask(server, res.label); + nfs4_inode_make_writeable(inode); + status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); if (!status) { update_changeattr(dir, &res.cinfo, res.fattr->time_start); @@ -5310,7 +5375,7 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl i = buf_to_pages_noslab(buf, buflen, arg.acl_pages); if (i < 0) return i; - nfs4_inode_return_delegation(inode); + nfs4_inode_make_writeable(inode); ret = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); /* @@ -5325,7 +5390,8 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl * so mark the attribute cache invalid. */ spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR; + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME; spin_unlock(&inode->i_lock); nfs_access_zap_cache(inode); nfs_zap_acl_cache(inode); @@ -6621,22 +6687,24 @@ static int nfs4_wake_lock_waiter(wait_queue_entry_t *wait, unsigned int mode, int flags, void *key) { int ret; - struct cb_notify_lock_args *cbnl = key; struct nfs4_lock_waiter *waiter = wait->private; - struct nfs_lowner *lowner = &cbnl->cbnl_owner, - *wowner = waiter->owner; - /* Only wake if the callback was for the same owner */ - if (lowner->clientid != wowner->clientid || - lowner->id != wowner->id || - lowner->s_dev != wowner->s_dev) - return 0; + /* NULL key means to wake up everyone */ + if (key) { + struct cb_notify_lock_args *cbnl = key; + struct nfs_lowner *lowner = &cbnl->cbnl_owner, + *wowner = waiter->owner; - /* Make sure it's for the right inode */ - if (nfs_compare_fh(NFS_FH(waiter->inode), &cbnl->cbnl_fh)) - return 0; + /* Only wake if the callback was for the same owner. */ + if (lowner->id != wowner->id || lowner->s_dev != wowner->s_dev) + return 0; - waiter->notified = true; + /* Make sure it's for the right inode */ + if (nfs_compare_fh(NFS_FH(waiter->inode), &cbnl->cbnl_fh)) + return 0; + + waiter->notified = true; + } /* override "private" so we can use default_wake_function */ wait->private = waiter->task; @@ -6673,6 +6741,7 @@ nfs4_retry_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) add_wait_queue(q, &wait); while(!signalled()) { + waiter.notified = false; status = nfs4_proc_setlk(state, cmd, request); if ((status != -EAGAIN) || IS_SETLK(cmd)) break; @@ -8414,6 +8483,8 @@ static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nf { switch(task->tk_status) { case 0: + wake_up_all(&clp->cl_lock_waitq); + /* Fallthrough */ case -NFS4ERR_COMPLETE_ALREADY: case -NFS4ERR_WRONG_CRED: /* What to do here? */ break; @@ -9593,7 +9664,7 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .link = nfs4_proc_link, .symlink = nfs4_proc_symlink, .mkdir = nfs4_proc_mkdir, - .rmdir = nfs4_proc_remove, + .rmdir = nfs4_proc_rmdir, .readdir = nfs4_proc_readdir, .mknod = nfs4_proc_mknod, .statfs = nfs4_proc_statfs, @@ -9614,7 +9685,6 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .close_context = nfs4_close_context, .open_context = nfs4_atomic_open, .have_delegation = nfs4_have_delegation, - .return_delegation = nfs4_inode_return_delegation, .alloc_client = nfs4_alloc_client, .init_client = nfs4_init_client, .free_client = nfs4_free_client, diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 91a4d4eeb235..c10a422efe6f 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -428,7 +428,6 @@ nfs4_insert_state_owner_locked(struct nfs4_state_owner *new) struct rb_node **p = &server->state_owners.rb_node, *parent = NULL; struct nfs4_state_owner *sp; - int err; while (*p != NULL) { parent = *p; @@ -445,9 +444,6 @@ nfs4_insert_state_owner_locked(struct nfs4_state_owner *new) return sp; } } - err = ida_get_new(&server->openowner_id, &new->so_seqid.owner_id); - if (err) - return ERR_PTR(err); rb_link_node(&new->so_server_node, parent, p); rb_insert_color(&new->so_server_node, &server->state_owners); return new; @@ -460,7 +456,6 @@ nfs4_remove_state_owner_locked(struct nfs4_state_owner *sp) if (!RB_EMPTY_NODE(&sp->so_server_node)) rb_erase(&sp->so_server_node, &server->state_owners); - ida_remove(&server->openowner_id, sp->so_seqid.owner_id); } static void @@ -495,6 +490,12 @@ nfs4_alloc_state_owner(struct nfs_server *server, sp = kzalloc(sizeof(*sp), gfp_flags); if (!sp) return NULL; + sp->so_seqid.owner_id = ida_simple_get(&server->openowner_id, 0, 0, + gfp_flags); + if (sp->so_seqid.owner_id < 0) { + kfree(sp); + return NULL; + } sp->so_server = server; sp->so_cred = get_rpccred(cred); spin_lock_init(&sp->so_lock); @@ -526,6 +527,7 @@ static void nfs4_free_state_owner(struct nfs4_state_owner *sp) { nfs4_destroy_seqid_counter(&sp->so_seqid); put_rpccred(sp->so_cred); + ida_simple_remove(&sp->so_server->openowner_id, sp->so_seqid.owner_id); kfree(sp); } @@ -576,13 +578,9 @@ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, new = nfs4_alloc_state_owner(server, cred, gfp_flags); if (new == NULL) goto out; - do { - if (ida_pre_get(&server->openowner_id, gfp_flags) == 0) - break; - spin_lock(&clp->cl_lock); - sp = nfs4_insert_state_owner_locked(new); - spin_unlock(&clp->cl_lock); - } while (sp == ERR_PTR(-EAGAIN)); + spin_lock(&clp->cl_lock); + sp = nfs4_insert_state_owner_locked(new); + spin_unlock(&clp->cl_lock); if (sp != new) nfs4_free_state_owner(new); out: diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 65c9c4175145..9b7392032321 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -52,7 +52,6 @@ #include <linux/nfs.h> #include <linux/nfs4.h> #include <linux/nfs_fs.h> -#include <linux/fs_struct.h> #include "nfs4_fs.h" #include "internal.h" @@ -99,6 +98,7 @@ static int nfs4_stat_to_errno(int); ((3+NFS4_FHSIZE) >> 2)) #define nfs4_fattr_bitmap_maxsz 4 #define encode_getattr_maxsz (op_encode_hdr_maxsz + nfs4_fattr_bitmap_maxsz) +#define nfstime4_maxsz (3) #define nfs4_name_maxsz (1 + ((3 + NFS4_MAXNAMLEN) >> 2)) #define nfs4_path_maxsz (1 + ((3 + NFS4_MAXPATHLEN) >> 2)) #define nfs4_owner_maxsz (1 + XDR_QUADLEN(IDMAP_NAMESZ)) @@ -113,7 +113,8 @@ static int nfs4_stat_to_errno(int); #define decode_mdsthreshold_maxsz (1 + 1 + nfs4_fattr_bitmap_maxsz + 1 + 8) /* This is based on getfattr, which uses the most attributes: */ #define nfs4_fattr_value_maxsz (1 + (1 + 2 + 2 + 4 + 2 + 1 + 1 + 2 + 2 + \ - 3 + 3 + 3 + nfs4_owner_maxsz + \ + 3*nfstime4_maxsz + \ + nfs4_owner_maxsz + \ nfs4_group_maxsz + nfs4_label_maxsz + \ decode_mdsthreshold_maxsz)) #define nfs4_fattr_maxsz (nfs4_fattr_bitmap_maxsz + \ @@ -124,7 +125,8 @@ static int nfs4_stat_to_errno(int); nfs4_owner_maxsz + \ nfs4_group_maxsz + \ nfs4_label_maxsz + \ - 4 + 4) + 1 + nfstime4_maxsz + \ + 1 + nfstime4_maxsz) #define encode_savefh_maxsz (op_encode_hdr_maxsz) #define decode_savefh_maxsz (op_decode_hdr_maxsz) #define encode_restorefh_maxsz (op_encode_hdr_maxsz) @@ -958,6 +960,35 @@ static void encode_uint64(struct xdr_stream *xdr, u64 n) WARN_ON_ONCE(xdr_stream_encode_u64(xdr, n) < 0); } +static ssize_t xdr_encode_bitmap4(struct xdr_stream *xdr, + const __u32 *bitmap, size_t len) +{ + ssize_t ret; + + /* Trim empty words */ + while (len > 0 && bitmap[len-1] == 0) + len--; + ret = xdr_stream_encode_uint32_array(xdr, bitmap, len); + if (WARN_ON_ONCE(ret < 0)) + return ret; + return len; +} + +static size_t mask_bitmap4(const __u32 *bitmap, const __u32 *mask, + __u32 *res, size_t len) +{ + size_t i; + __u32 tmp; + + while (len > 0 && (bitmap[len-1] == 0 || mask[len-1] == 0)) + len--; + for (i = len; i-- > 0;) { + tmp = bitmap[i] & mask[i]; + res[i] = tmp; + } + return len; +} + static void encode_nfs4_seqid(struct xdr_stream *xdr, const struct nfs_seqid *seqid) { @@ -1012,6 +1043,14 @@ static void encode_nfs4_verifier(struct xdr_stream *xdr, const nfs4_verifier *ve encode_opaque_fixed(xdr, verf->data, NFS4_VERIFIER_SIZE); } +static __be32 * +xdr_encode_nfstime4(__be32 *p, const struct timespec *t) +{ + p = xdr_encode_hyper(p, (__s64)t->tv_sec); + *p++ = cpu_to_be32(t->tv_nsec); + return p; +} + static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const struct nfs4_label *label, const umode_t *umask, @@ -1023,9 +1062,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, int owner_namelen = 0; int owner_grouplen = 0; __be32 *p; - unsigned i; uint32_t len = 0; - uint32_t bmval_len; uint32_t bmval[3] = { 0 }; /* @@ -1073,7 +1110,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, if (attrmask[1] & FATTR4_WORD1_TIME_ACCESS_SET) { if (iap->ia_valid & ATTR_ATIME_SET) { bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET; - len += 16; + len += 4 + (nfstime4_maxsz << 2); } else if (iap->ia_valid & ATTR_ATIME) { bmval[1] |= FATTR4_WORD1_TIME_ACCESS_SET; len += 4; @@ -1082,7 +1119,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, if (attrmask[1] & FATTR4_WORD1_TIME_MODIFY_SET) { if (iap->ia_valid & ATTR_MTIME_SET) { bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET; - len += 16; + len += 4 + (nfstime4_maxsz << 2); } else if (iap->ia_valid & ATTR_MTIME) { bmval[1] |= FATTR4_WORD1_TIME_MODIFY_SET; len += 4; @@ -1094,19 +1131,8 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, bmval[2] |= FATTR4_WORD2_SECURITY_LABEL; } - if (bmval[2] != 0) - bmval_len = 3; - else if (bmval[1] != 0) - bmval_len = 2; - else - bmval_len = 1; - - p = reserve_space(xdr, 4 + (bmval_len << 2) + 4 + len); - - *p++ = cpu_to_be32(bmval_len); - for (i = 0; i < bmval_len; i++) - *p++ = cpu_to_be32(bmval[i]); - *p++ = cpu_to_be32(len); + xdr_encode_bitmap4(xdr, bmval, ARRAY_SIZE(bmval)); + xdr_stream_encode_opaque_inline(xdr, (void **)&p, len); if (bmval[0] & FATTR4_WORD0_SIZE) p = xdr_encode_hyper(p, iap->ia_size); @@ -1119,16 +1145,14 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, if (bmval[1] & FATTR4_WORD1_TIME_ACCESS_SET) { if (iap->ia_valid & ATTR_ATIME_SET) { *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); - p = xdr_encode_hyper(p, (s64)iap->ia_atime.tv_sec); - *p++ = cpu_to_be32(iap->ia_atime.tv_nsec); + p = xdr_encode_nfstime4(p, &iap->ia_atime); } else *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); } if (bmval[1] & FATTR4_WORD1_TIME_MODIFY_SET) { if (iap->ia_valid & ATTR_MTIME_SET) { *p++ = cpu_to_be32(NFS4_SET_TO_CLIENT_TIME); - p = xdr_encode_hyper(p, (s64)iap->ia_mtime.tv_sec); - *p++ = cpu_to_be32(iap->ia_mtime.tv_nsec); + p = xdr_encode_nfstime4(p, &iap->ia_mtime); } else *p++ = cpu_to_be32(NFS4_SET_TO_SERVER_TIME); } @@ -1200,85 +1224,45 @@ static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg * create->server, create->server->attr_bitmask); } -static void encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap, struct compound_hdr *hdr) -{ - __be32 *p; - - encode_op_hdr(xdr, OP_GETATTR, decode_getattr_maxsz, hdr); - p = reserve_space(xdr, 8); - *p++ = cpu_to_be32(1); - *p = cpu_to_be32(bitmap); -} - -static void encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1, struct compound_hdr *hdr) -{ - __be32 *p; - - encode_op_hdr(xdr, OP_GETATTR, decode_getattr_maxsz, hdr); - p = reserve_space(xdr, 12); - *p++ = cpu_to_be32(2); - *p++ = cpu_to_be32(bm0); - *p = cpu_to_be32(bm1); -} - -static void -encode_getattr_three(struct xdr_stream *xdr, - uint32_t bm0, uint32_t bm1, uint32_t bm2, - struct compound_hdr *hdr) +static void encode_getattr(struct xdr_stream *xdr, + const __u32 *bitmap, const __u32 *mask, size_t len, + struct compound_hdr *hdr) { - __be32 *p; + __u32 masked_bitmap[nfs4_fattr_bitmap_maxsz]; encode_op_hdr(xdr, OP_GETATTR, decode_getattr_maxsz, hdr); - if (bm2) { - p = reserve_space(xdr, 16); - *p++ = cpu_to_be32(3); - *p++ = cpu_to_be32(bm0); - *p++ = cpu_to_be32(bm1); - *p = cpu_to_be32(bm2); - } else if (bm1) { - p = reserve_space(xdr, 12); - *p++ = cpu_to_be32(2); - *p++ = cpu_to_be32(bm0); - *p = cpu_to_be32(bm1); - } else { - p = reserve_space(xdr, 8); - *p++ = cpu_to_be32(1); - *p = cpu_to_be32(bm0); + if (mask) { + if (WARN_ON_ONCE(len > ARRAY_SIZE(masked_bitmap))) + len = ARRAY_SIZE(masked_bitmap); + len = mask_bitmap4(bitmap, mask, masked_bitmap, len); + bitmap = masked_bitmap; } + xdr_encode_bitmap4(xdr, bitmap, len); } static void encode_getfattr(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr) { - encode_getattr_three(xdr, bitmask[0] & nfs4_fattr_bitmap[0], - bitmask[1] & nfs4_fattr_bitmap[1], - bitmask[2] & nfs4_fattr_bitmap[2], - hdr); + encode_getattr(xdr, nfs4_fattr_bitmap, bitmask, + ARRAY_SIZE(nfs4_fattr_bitmap), hdr); } static void encode_getfattr_open(struct xdr_stream *xdr, const u32 *bitmask, const u32 *open_bitmap, struct compound_hdr *hdr) { - encode_getattr_three(xdr, - bitmask[0] & open_bitmap[0], - bitmask[1] & open_bitmap[1], - bitmask[2] & open_bitmap[2], - hdr); + encode_getattr(xdr, open_bitmap, bitmask, 3, hdr); } static void encode_fsinfo(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr) { - encode_getattr_three(xdr, - bitmask[0] & nfs4_fsinfo_bitmap[0], - bitmask[1] & nfs4_fsinfo_bitmap[1], - bitmask[2] & nfs4_fsinfo_bitmap[2], - hdr); + encode_getattr(xdr, nfs4_fsinfo_bitmap, bitmask, + ARRAY_SIZE(nfs4_fsinfo_bitmap), hdr); } static void encode_fs_locations(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr) { - encode_getattr_two(xdr, bitmask[0] & nfs4_fs_locations_bitmap[0], - bitmask[1] & nfs4_fs_locations_bitmap[1], hdr); + encode_getattr(xdr, nfs4_fs_locations_bitmap, bitmask, + ARRAY_SIZE(nfs4_fs_locations_bitmap), hdr); } static void encode_getfh(struct xdr_stream *xdr, struct compound_hdr *hdr) @@ -2117,7 +2101,8 @@ static void nfs4_xdr_enc_access(struct rpc_rqst *req, struct xdr_stream *xdr, encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, args->fh, &hdr); encode_access(xdr, args->access, &hdr); - encode_getfattr(xdr, args->bitmask, &hdr); + if (args->bitmask) + encode_getfattr(xdr, args->bitmask, &hdr); encode_nops(&hdr); } @@ -2559,13 +2544,17 @@ static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr, struct compound_hdr hdr = { .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; + const __u32 nfs4_acl_bitmap[1] = { + [0] = FATTR4_WORD0_ACL, + }; uint32_t replen; encode_compound_hdr(xdr, req, &hdr); encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, args->fh, &hdr); replen = hdr.replen + op_decode_hdr_maxsz; - encode_getattr_two(xdr, FATTR4_WORD0_ACL, 0, &hdr); + encode_getattr(xdr, nfs4_acl_bitmap, NULL, + ARRAY_SIZE(nfs4_acl_bitmap), &hdr); xdr_inline_pages(&req->rq_rcv_buf, replen << 2, args->acl_pages, 0, args->acl_len); @@ -2644,8 +2633,8 @@ static void nfs4_xdr_enc_pathconf(struct rpc_rqst *req, struct xdr_stream *xdr, encode_compound_hdr(xdr, req, &hdr); encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, args->fh, &hdr); - encode_getattr_one(xdr, args->bitmask[0] & nfs4_pathconf_bitmap[0], - &hdr); + encode_getattr(xdr, nfs4_pathconf_bitmap, args->bitmask, + ARRAY_SIZE(nfs4_pathconf_bitmap), &hdr); encode_nops(&hdr); } @@ -2663,8 +2652,8 @@ static void nfs4_xdr_enc_statfs(struct rpc_rqst *req, struct xdr_stream *xdr, encode_compound_hdr(xdr, req, &hdr); encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, args->fh, &hdr); - encode_getattr_two(xdr, args->bitmask[0] & nfs4_statfs_bitmap[0], - args->bitmask[1] & nfs4_statfs_bitmap[1], &hdr); + encode_getattr(xdr, nfs4_statfs_bitmap, args->bitmask, + ARRAY_SIZE(nfs4_statfs_bitmap), &hdr); encode_nops(&hdr); } @@ -2684,7 +2673,7 @@ static void nfs4_xdr_enc_server_caps(struct rpc_rqst *req, encode_compound_hdr(xdr, req, &hdr); encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, args->fhandle, &hdr); - encode_getattr_three(xdr, bitmask[0], bitmask[1], bitmask[2], &hdr); + encode_getattr(xdr, bitmask, NULL, 3, &hdr); encode_nops(&hdr); } @@ -3218,34 +3207,27 @@ static int decode_ace(struct xdr_stream *xdr, void *ace) return -EIO; } -static int decode_attr_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) +static ssize_t +decode_bitmap4(struct xdr_stream *xdr, uint32_t *bitmap, size_t sz) { - uint32_t bmlen; - __be32 *p; - - p = xdr_inline_decode(xdr, 4); - if (unlikely(!p)) - goto out_overflow; - bmlen = be32_to_cpup(p); + ssize_t ret; - bitmap[0] = bitmap[1] = bitmap[2] = 0; - p = xdr_inline_decode(xdr, (bmlen << 2)); - if (unlikely(!p)) - goto out_overflow; - if (bmlen > 0) { - bitmap[0] = be32_to_cpup(p++); - if (bmlen > 1) { - bitmap[1] = be32_to_cpup(p++); - if (bmlen > 2) - bitmap[2] = be32_to_cpup(p); - } - } - return 0; -out_overflow: + ret = xdr_stream_decode_uint32_array(xdr, bitmap, sz); + if (likely(ret >= 0)) + return ret; + if (ret == -EMSGSIZE) + return sz; print_overflow_msg(__func__, xdr); return -EIO; } +static int decode_attr_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) +{ + ssize_t ret; + ret = decode_bitmap4(xdr, bitmap, 3); + return ret < 0 ? ret : 0; +} + static int decode_attr_length(struct xdr_stream *xdr, uint32_t *attrlen, unsigned int *savep) { __be32 *p; @@ -3981,7 +3963,7 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, bitmap[1] &= ~FATTR4_WORD1_OWNER; if (owner_name != NULL) { - len = decode_nfs4_string(xdr, owner_name, GFP_NOWAIT); + len = decode_nfs4_string(xdr, owner_name, GFP_NOIO); if (len <= 0) goto out; dprintk("%s: name=%s\n", __func__, owner_name->data); @@ -4016,7 +3998,7 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, bitmap[1] &= ~FATTR4_WORD1_OWNER_GROUP; if (group_name != NULL) { - len = decode_nfs4_string(xdr, group_name, GFP_NOWAIT); + len = decode_nfs4_string(xdr, group_name, GFP_NOIO); if (len <= 0) goto out; dprintk("%s: name=%s\n", __func__, group_name->data); @@ -4156,19 +4138,25 @@ out_overflow: return -EIO; } +static __be32 * +xdr_decode_nfstime4(__be32 *p, struct timespec *t) +{ + __u64 sec; + + p = xdr_decode_hyper(p, &sec); + t-> tv_sec = (time_t)sec; + t->tv_nsec = be32_to_cpup(p++); + return p; +} + static int decode_attr_time(struct xdr_stream *xdr, struct timespec *time) { __be32 *p; - uint64_t sec; - uint32_t nsec; - p = xdr_inline_decode(xdr, 12); + p = xdr_inline_decode(xdr, nfstime4_maxsz << 2); if (unlikely(!p)) goto out_overflow; - p = xdr_decode_hyper(p, &sec); - nsec = be32_to_cpup(p); - time->tv_sec = (time_t)sec; - time->tv_nsec = (long)nsec; + xdr_decode_nfstime4(p, time); return 0; out_overflow: print_overflow_msg(__func__, xdr); @@ -5471,21 +5459,13 @@ decode_savefh(struct xdr_stream *xdr) static int decode_setattr(struct xdr_stream *xdr) { - __be32 *p; - uint32_t bmlen; int status; status = decode_op_hdr(xdr, OP_SETATTR); if (status) return status; - p = xdr_inline_decode(xdr, 4); - if (unlikely(!p)) - goto out_overflow; - bmlen = be32_to_cpup(p); - p = xdr_inline_decode(xdr, bmlen << 2); - if (likely(p)) + if (decode_bitmap4(xdr, NULL, 0) >= 0) return 0; -out_overflow: print_overflow_msg(__func__, xdr); return -EIO; } @@ -6256,7 +6236,8 @@ static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, struct xdr_stream *xdr, status = decode_access(xdr, &res->supported, &res->access); if (status != 0) goto out; - decode_getfattr(xdr, res->fattr, res->server); + if (res->fattr) + decode_getfattr(xdr, res->fattr, res->server); out: return status; } @@ -7536,6 +7517,7 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, unsigned int savep; uint32_t bitmap[3] = {0}; uint32_t len; + uint64_t new_cookie; __be32 *p = xdr_inline_decode(xdr, 4); if (unlikely(!p)) goto out_overflow; @@ -7552,8 +7534,7 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, p = xdr_inline_decode(xdr, 12); if (unlikely(!p)) goto out_overflow; - entry->prev_cookie = entry->cookie; - p = xdr_decode_hyper(p, &entry->cookie); + p = xdr_decode_hyper(p, &new_cookie); entry->len = be32_to_cpup(p); p = xdr_inline_decode(xdr, entry->len); @@ -7587,6 +7568,9 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, if (entry->fattr->valid & NFS_ATTR_FATTR_TYPE) entry->d_type = nfs_umode_to_dtype(entry->fattr->mode); + entry->prev_cookie = entry->cookie; + entry->cookie = new_cookie; + return 0; out_overflow: diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index f7fd9192d4bc..4e93d6308733 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -300,11 +300,11 @@ out: } static int -nfs_proc_remove(struct inode *dir, const struct qstr *name) +nfs_proc_remove(struct inode *dir, struct dentry *dentry) { struct nfs_removeargs arg = { .fh = NFS_FH(dir), - .name = *name, + .name = dentry->d_name, }; struct rpc_message msg = { .rpc_proc = &nfs_procedures[NFSPROC_REMOVE], @@ -312,7 +312,7 @@ nfs_proc_remove(struct inode *dir, const struct qstr *name) }; int status; - dprintk("NFS call remove %s\n", name->name); + dprintk("NFS call remove %pd2\n",dentry); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); nfs_mark_for_revalidate(dir); @@ -321,7 +321,7 @@ nfs_proc_remove(struct inode *dir, const struct qstr *name) } static void -nfs_proc_unlink_setup(struct rpc_message *msg, struct inode *dir) +nfs_proc_unlink_setup(struct rpc_message *msg, struct dentry *dentry) { msg->rpc_proc = &nfs_procedures[NFSPROC_REMOVE]; } @@ -338,7 +338,9 @@ static int nfs_proc_unlink_done(struct rpc_task *task, struct inode *dir) } static void -nfs_proc_rename_setup(struct rpc_message *msg, struct inode *dir) +nfs_proc_rename_setup(struct rpc_message *msg, + struct dentry *old_dentry, + struct dentry *new_dentry) { msg->rpc_proc = &nfs_procedures[NFSPROC_RENAME]; } @@ -671,12 +673,6 @@ static int nfs_have_delegation(struct inode *inode, fmode_t flags) return 0; } -static int nfs_return_delegation(struct inode *inode) -{ - nfs_wb_all(inode); - return 0; -} - static const struct inode_operations nfs_dir_inode_operations = { .create = nfs_create, .lookup = nfs_lookup, @@ -741,7 +737,6 @@ const struct nfs_rpc_ops nfs_v2_clientops = { .lock_check_bounds = nfs_lock_check_bounds, .close_context = nfs_close_context, .have_delegation = nfs_have_delegation, - .return_delegation = nfs_return_delegation, .alloc_client = nfs_alloc_client, .init_client = nfs_init_client, .free_client = nfs_free_client, diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 630b4a3c1a93..bf54fc9ae135 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -105,7 +105,7 @@ static void nfs_do_call_unlink(struct nfs_unlinkdata *data) data->args.fh = NFS_FH(dir); nfs_fattr_init(data->res.dir_attr); - NFS_PROTO(dir)->unlink_setup(&msg, dir); + NFS_PROTO(dir)->unlink_setup(&msg, data->dentry); task_setup_data.rpc_client = NFS_CLIENT(dir); task = rpc_run_task(&task_setup_data); @@ -386,7 +386,7 @@ nfs_async_rename(struct inode *old_dir, struct inode *new_dir, nfs_sb_active(old_dir->i_sb); - NFS_PROTO(data->old_dir)->rename_setup(&msg, old_dir); + NFS_PROTO(data->old_dir)->rename_setup(&msg, old_dentry, new_dentry); return rpc_run_task(&task_setup_data); } @@ -463,9 +463,6 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) fileid = NFS_FILEID(d_inode(dentry)); - /* Return delegation in anticipation of the rename */ - NFS_PROTO(d_inode(dentry))->return_delegation(d_inode(dentry)); - sdentry = NULL; do { int slen; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 6579f3b367bd..0193053bc139 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -231,6 +231,7 @@ static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int c if (i_size >= end) goto out; i_size_write(inode, end); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; nfs_inc_stats(inode, NFSIOS_EXTENDWRITE); out: spin_unlock(&inode->i_lock); @@ -1562,8 +1563,11 @@ static int nfs_writeback_done(struct rpc_task *task, } /* Deal with the suid/sgid bit corner case */ - if (nfs_should_remove_suid(inode)) - nfs_mark_for_revalidate(inode); + if (nfs_should_remove_suid(inode)) { + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_OTHER; + spin_unlock(&inode->i_lock); + } return 0; } diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c index 70b8bf781fce..a43dfedd69ec 100644 --- a/fs/nfsd/blocklayout.c +++ b/fs/nfsd/blocklayout.c @@ -227,7 +227,7 @@ static int nfsd4_scsi_identify_device(struct block_device *bdev, if (!buf) return -ENOMEM; - rq = blk_get_request(q, REQ_OP_SCSI_IN, GFP_KERNEL); + rq = blk_get_request(q, REQ_OP_SCSI_IN, 0); if (IS_ERR(rq)) { error = -ENOMEM; goto out_free_buf; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 2410b093a2e6..b0555d7d8200 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1201,6 +1201,28 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp, break; case S_IFDIR: host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); + if (!host_err && unlikely(d_unhashed(dchild))) { + struct dentry *d; + d = lookup_one_len(dchild->d_name.name, + dchild->d_parent, + dchild->d_name.len); + if (IS_ERR(d)) { + host_err = PTR_ERR(d); + break; + } + if (unlikely(d_is_negative(d))) { + dput(d); + err = nfserr_serverfault; + goto out; + } + dput(resfhp->fh_dentry); + resfhp->fh_dentry = dget(d); + err = fh_update(resfhp); + dput(dchild); + dchild = d; + if (err) + goto out; + } break; case S_IFCHR: case S_IFBLK: diff --git a/fs/nilfs2/btnode.c b/fs/nilfs2/btnode.c index c21e0b4454a6..dec98cab729d 100644 --- a/fs/nilfs2/btnode.c +++ b/fs/nilfs2/btnode.c @@ -193,9 +193,9 @@ retry: (unsigned long long)oldkey, (unsigned long long)newkey); - spin_lock_irq(&btnc->tree_lock); - err = radix_tree_insert(&btnc->page_tree, newkey, obh->b_page); - spin_unlock_irq(&btnc->tree_lock); + xa_lock_irq(&btnc->i_pages); + err = radix_tree_insert(&btnc->i_pages, newkey, obh->b_page); + xa_unlock_irq(&btnc->i_pages); /* * Note: page->index will not change to newkey until * nilfs_btnode_commit_change_key() will be called. @@ -251,11 +251,11 @@ void nilfs_btnode_commit_change_key(struct address_space *btnc, (unsigned long long)newkey); mark_buffer_dirty(obh); - spin_lock_irq(&btnc->tree_lock); - radix_tree_delete(&btnc->page_tree, oldkey); - radix_tree_tag_set(&btnc->page_tree, newkey, + xa_lock_irq(&btnc->i_pages); + radix_tree_delete(&btnc->i_pages, oldkey); + radix_tree_tag_set(&btnc->i_pages, newkey, PAGECACHE_TAG_DIRTY); - spin_unlock_irq(&btnc->tree_lock); + xa_unlock_irq(&btnc->i_pages); opage->index = obh->b_blocknr = newkey; unlock_page(opage); @@ -283,9 +283,9 @@ void nilfs_btnode_abort_change_key(struct address_space *btnc, return; if (nbh == NULL) { /* blocksize == pagesize */ - spin_lock_irq(&btnc->tree_lock); - radix_tree_delete(&btnc->page_tree, newkey); - spin_unlock_irq(&btnc->tree_lock); + xa_lock_irq(&btnc->i_pages); + radix_tree_delete(&btnc->i_pages, newkey); + xa_unlock_irq(&btnc->i_pages); unlock_page(ctxt->bh->b_page); } else brelse(nbh); diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c index 1a2894aa0194..dd52d3f82e8d 100644 --- a/fs/nilfs2/namei.c +++ b/fs/nilfs2/namei.c @@ -46,8 +46,7 @@ static inline int nilfs_add_nondir(struct dentry *dentry, struct inode *inode) int err = nilfs_add_link(dentry, inode); if (!err) { - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -243,8 +242,7 @@ static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) goto out_fail; nilfs_mark_inode_dirty(inode); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); out: if (!err) err = nilfs_transaction_commit(dir->i_sb); diff --git a/fs/nilfs2/page.c b/fs/nilfs2/page.c index 68241512d7c1..4cb850a6f1c2 100644 --- a/fs/nilfs2/page.c +++ b/fs/nilfs2/page.c @@ -331,15 +331,15 @@ repeat: struct page *page2; /* move the page to the destination cache */ - spin_lock_irq(&smap->tree_lock); - page2 = radix_tree_delete(&smap->page_tree, offset); + xa_lock_irq(&smap->i_pages); + page2 = radix_tree_delete(&smap->i_pages, offset); WARN_ON(page2 != page); smap->nrpages--; - spin_unlock_irq(&smap->tree_lock); + xa_unlock_irq(&smap->i_pages); - spin_lock_irq(&dmap->tree_lock); - err = radix_tree_insert(&dmap->page_tree, offset, page); + xa_lock_irq(&dmap->i_pages); + err = radix_tree_insert(&dmap->i_pages, offset, page); if (unlikely(err < 0)) { WARN_ON(err == -EEXIST); page->mapping = NULL; @@ -348,11 +348,11 @@ repeat: page->mapping = dmap; dmap->nrpages++; if (PageDirty(page)) - radix_tree_tag_set(&dmap->page_tree, + radix_tree_tag_set(&dmap->i_pages, offset, PAGECACHE_TAG_DIRTY); } - spin_unlock_irq(&dmap->tree_lock); + xa_unlock_irq(&dmap->i_pages); } unlock_page(page); } @@ -474,15 +474,15 @@ int __nilfs_clear_page_dirty(struct page *page) struct address_space *mapping = page->mapping; if (mapping) { - spin_lock_irq(&mapping->tree_lock); + xa_lock_irq(&mapping->i_pages); if (test_bit(PG_dirty, &page->flags)) { - radix_tree_tag_clear(&mapping->page_tree, + radix_tree_tag_clear(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); return clear_page_dirty_for_io(page); } - spin_unlock_irq(&mapping->tree_lock); + xa_unlock_irq(&mapping->i_pages); return 0; } return TestClearPageDirty(page); diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c index d51e1bb781cf..d94e8031fe5f 100644 --- a/fs/notify/fanotify/fanotify.c +++ b/fs/notify/fanotify/fanotify.c @@ -92,7 +92,7 @@ static bool fanotify_should_send_event(struct fsnotify_mark *inode_mark, u32 event_mask, const void *data, int data_type) { - __u32 marks_mask, marks_ignored_mask; + __u32 marks_mask = 0, marks_ignored_mask = 0; const struct path *path = data; pr_debug("%s: inode_mark=%p vfsmnt_mark=%p mask=%x data=%p" @@ -108,24 +108,20 @@ static bool fanotify_should_send_event(struct fsnotify_mark *inode_mark, !d_can_lookup(path->dentry)) return false; - if (inode_mark && vfsmnt_mark) { - marks_mask = (vfsmnt_mark->mask | inode_mark->mask); - marks_ignored_mask = (vfsmnt_mark->ignored_mask | inode_mark->ignored_mask); - } else if (inode_mark) { - /* - * if the event is for a child and this inode doesn't care about - * events on the child, don't send it! - */ - if ((event_mask & FS_EVENT_ON_CHILD) && - !(inode_mark->mask & FS_EVENT_ON_CHILD)) - return false; - marks_mask = inode_mark->mask; - marks_ignored_mask = inode_mark->ignored_mask; - } else if (vfsmnt_mark) { - marks_mask = vfsmnt_mark->mask; - marks_ignored_mask = vfsmnt_mark->ignored_mask; - } else { - BUG(); + /* + * if the event is for a child and this inode doesn't care about + * events on the child, don't send it! + */ + if (inode_mark && + (!(event_mask & FS_EVENT_ON_CHILD) || + (inode_mark->mask & FS_EVENT_ON_CHILD))) { + marks_mask |= inode_mark->mask; + marks_ignored_mask |= inode_mark->ignored_mask; + } + + if (vfsmnt_mark) { + marks_mask |= vfsmnt_mark->mask; + marks_ignored_mask |= vfsmnt_mark->ignored_mask; } if (d_is_dir(path->dentry) && diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 219b269c737e..613ec7e5a465 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -192,8 +192,9 @@ static int send_to_group(struct inode *to_tell, struct fsnotify_iter_info *iter_info) { struct fsnotify_group *group = NULL; - __u32 inode_test_mask = 0; - __u32 vfsmount_test_mask = 0; + __u32 test_mask = (mask & ~FS_EVENT_ON_CHILD); + __u32 marks_mask = 0; + __u32 marks_ignored_mask = 0; if (unlikely(!inode_mark && !vfsmount_mark)) { BUG(); @@ -213,29 +214,25 @@ static int send_to_group(struct inode *to_tell, /* does the inode mark tell us to do something? */ if (inode_mark) { group = inode_mark->group; - inode_test_mask = (mask & ~FS_EVENT_ON_CHILD); - inode_test_mask &= inode_mark->mask; - inode_test_mask &= ~inode_mark->ignored_mask; + marks_mask |= inode_mark->mask; + marks_ignored_mask |= inode_mark->ignored_mask; } /* does the vfsmount_mark tell us to do something? */ if (vfsmount_mark) { - vfsmount_test_mask = (mask & ~FS_EVENT_ON_CHILD); group = vfsmount_mark->group; - vfsmount_test_mask &= vfsmount_mark->mask; - vfsmount_test_mask &= ~vfsmount_mark->ignored_mask; - if (inode_mark) - vfsmount_test_mask &= ~inode_mark->ignored_mask; + marks_mask |= vfsmount_mark->mask; + marks_ignored_mask |= vfsmount_mark->ignored_mask; } pr_debug("%s: group=%p to_tell=%p mask=%x inode_mark=%p" - " inode_test_mask=%x vfsmount_mark=%p vfsmount_test_mask=%x" + " vfsmount_mark=%p marks_mask=%x marks_ignored_mask=%x" " data=%p data_is=%d cookie=%d\n", - __func__, group, to_tell, mask, inode_mark, - inode_test_mask, vfsmount_mark, vfsmount_test_mask, data, + __func__, group, to_tell, mask, inode_mark, vfsmount_mark, + marks_mask, marks_ignored_mask, data, data_is, cookie); - if (!inode_test_mask && !vfsmount_test_mask) + if (!(test_mask & marks_mask & ~marks_ignored_mask)) return 0; return group->ops->handle_event(group, to_tell, inode_mark, diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c index 2831f495a674..32c523cf5a2d 100644 --- a/fs/ntfs/mft.c +++ b/fs/ntfs/mft.c @@ -381,7 +381,7 @@ unm_err_out: * vfs inode dirty. This ensures that any changes to the mft record are * written out to disk. * - * NOTE: We only set I_DIRTY_SYNC and I_DIRTY_DATASYNC (and not I_DIRTY_PAGES) + * NOTE: We only set I_DIRTY_DATASYNC (and not I_DIRTY_PAGES) * on the base vfs inode, because even though file data may have been modified, * it is dirty in the inode meta data rather than the data page cache of the * inode, and thus there are no data pages that need writing out. Therefore, a @@ -407,7 +407,7 @@ void __mark_mft_record_dirty(ntfs_inode *ni) else base_ni = ni->ext.base_ntfs_ino; mutex_unlock(&ni->extent_lock); - __mark_inode_dirty(VFS_I(base_ni), I_DIRTY_SYNC | I_DIRTY_DATASYNC); + __mark_inode_dirty(VFS_I(base_ni), I_DIRTY_DATASYNC); } static const char *ntfs_please_email = "Please email " diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 9a876bb07cac..0f157bbd3e0f 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -7119,7 +7119,7 @@ int ocfs2_convert_inline_data_to_extents(struct inode *inode, goto out_commit; did_quota = 1; - data_ac->ac_resv = &OCFS2_I(inode)->ip_la_data_resv; + data_ac->ac_resv = &oi->ip_la_data_resv; ret = ocfs2_claim_clusters(handle, data_ac, 1, &bit_off, &num); diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index e8e205bf2e41..302cd7caa4a7 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -346,7 +346,7 @@ static int ocfs2_readpage(struct file *file, struct page *page) unlock = 0; out_alloc: - up_read(&OCFS2_I(inode)->ip_alloc_sem); + up_read(&oi->ip_alloc_sem); out_inode_unlock: ocfs2_inode_unlock(inode, 0); out: @@ -2213,7 +2213,7 @@ static int ocfs2_dio_wr_get_block(struct inode *inode, sector_t iblock, down_write(&oi->ip_alloc_sem); if (first_get_block) { - if (ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb))) + if (ocfs2_sparse_alloc(osb)) ret = ocfs2_zero_tail(inode, di_bh, pos); else ret = ocfs2_expand_nonsparse_inode(inode, di_bh, pos, diff --git a/fs/ocfs2/aops.h b/fs/ocfs2/aops.h index 8614ff069d99..3494a62ed749 100644 --- a/fs/ocfs2/aops.h +++ b/fs/ocfs2/aops.h @@ -78,7 +78,7 @@ static inline void ocfs2_iocb_set_rw_locked(struct kiocb *iocb, int level) /* * Using a named enum representing lock types in terms of #N bit stored in * iocb->private, which is going to be used for communication between - * ocfs2_dio_end_io() and ocfs2_file_aio_write/read(). + * ocfs2_dio_end_io() and ocfs2_file_write/read_iter(). */ enum ocfs2_iocb_lock_bits { OCFS2_IOCB_RW_LOCK = 0, diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 977763d4c27d..b048d4fa3959 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -3072,7 +3072,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, * We need to return the correct block within the * cluster which should hold our entry. */ - off = ocfs2_dx_dir_hash_idx(OCFS2_SB(dir->i_sb), + off = ocfs2_dx_dir_hash_idx(osb, &lookup->dl_hinfo); get_bh(dx_leaves[off]); lookup->dl_dx_leaf_bh = dx_leaves[off]; diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c index fd6bbbbd7d78..39831fc2fd52 100644 --- a/fs/ocfs2/dlm/dlmast.c +++ b/fs/ocfs2/dlm/dlmast.c @@ -224,14 +224,12 @@ void dlm_do_local_ast(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_lock *lock) { dlm_astlockfunc_t *fn; - struct dlm_lockstatus *lksb; mlog(0, "%s: res %.*s, lock %u:%llu, Local AST\n", dlm->name, res->lockname.len, res->lockname.name, dlm_get_lock_cookie_node(be64_to_cpu(lock->ml.cookie)), dlm_get_lock_cookie_seq(be64_to_cpu(lock->ml.cookie))); - lksb = lock->lksb; fn = lock->ast; BUG_ON(lock->ml.node != dlm->node_num); diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index e9f3705c4c9f..d06e27ec4be4 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -140,6 +140,7 @@ struct dlm_ctxt u8 node_num; u32 key; u8 joining_node; + u8 migrate_done; /* set to 1 means node has migrated all lock resources */ wait_queue_head_t dlm_join_events; unsigned long live_nodes_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long domain_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; @@ -960,13 +961,10 @@ static inline int dlm_send_proxy_ast(struct dlm_ctxt *dlm, void dlm_print_one_lock_resource(struct dlm_lock_resource *res); void __dlm_print_one_lock_resource(struct dlm_lock_resource *res); -u8 dlm_nm_this_node(struct dlm_ctxt *dlm); void dlm_kick_thread(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); void __dlm_dirty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); -int dlm_nm_init(struct dlm_ctxt *dlm); -int dlm_heartbeat_init(struct dlm_ctxt *dlm); void dlm_hb_node_down_cb(struct o2nm_node *node, int idx, void *data); void dlm_hb_node_up_cb(struct o2nm_node *node, int idx, void *data); diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index e1fea149f50b..425081be6161 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -461,6 +461,19 @@ redo_bucket: cond_resched_lock(&dlm->spinlock); num += n; } + + if (!num) { + if (dlm->reco.state & DLM_RECO_STATE_ACTIVE) { + mlog(0, "%s: perhaps there are more lock resources " + "need to be migrated after dlm recovery\n", dlm->name); + ret = -EAGAIN; + } else { + mlog(0, "%s: we won't do dlm recovery after migrating " + "all lock resources\n", dlm->name); + dlm->migrate_done = 1; + } + } + spin_unlock(&dlm->spinlock); wake_up(&dlm->dlm_thread_wq); @@ -675,20 +688,6 @@ static void dlm_leave_domain(struct dlm_ctxt *dlm) spin_unlock(&dlm->spinlock); } -int dlm_shutting_down(struct dlm_ctxt *dlm) -{ - int ret = 0; - - spin_lock(&dlm_domain_lock); - - if (dlm->dlm_state == DLM_CTXT_IN_SHUTDOWN) - ret = 1; - - spin_unlock(&dlm_domain_lock); - - return ret; -} - void dlm_unregister_domain(struct dlm_ctxt *dlm) { int leave = 0; @@ -2052,6 +2051,8 @@ static struct dlm_ctxt *dlm_alloc_ctxt(const char *domain, dlm->joining_node = DLM_LOCK_RES_OWNER_UNKNOWN; init_waitqueue_head(&dlm->dlm_join_events); + dlm->migrate_done = 0; + dlm->reco.new_master = O2NM_INVALID_NODE_NUM; dlm->reco.dead_node = O2NM_INVALID_NODE_NUM; diff --git a/fs/ocfs2/dlm/dlmdomain.h b/fs/ocfs2/dlm/dlmdomain.h index fd6122a38dbd..8a9281411c18 100644 --- a/fs/ocfs2/dlm/dlmdomain.h +++ b/fs/ocfs2/dlm/dlmdomain.h @@ -28,7 +28,30 @@ extern spinlock_t dlm_domain_lock; extern struct list_head dlm_domains; -int dlm_shutting_down(struct dlm_ctxt *dlm); +static inline int dlm_joined(struct dlm_ctxt *dlm) +{ + int ret = 0; + + spin_lock(&dlm_domain_lock); + if (dlm->dlm_state == DLM_CTXT_JOINED) + ret = 1; + spin_unlock(&dlm_domain_lock); + + return ret; +} + +static inline int dlm_shutting_down(struct dlm_ctxt *dlm) +{ + int ret = 0; + + spin_lock(&dlm_domain_lock); + if (dlm->dlm_state == DLM_CTXT_IN_SHUTDOWN) + ret = 1; + spin_unlock(&dlm_domain_lock); + + return ret; +} + void dlm_fire_domain_eviction_callbacks(struct dlm_ctxt *dlm, int node_num); diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c index 66c2a491f68d..74962315794e 100644 --- a/fs/ocfs2/dlm/dlmlock.c +++ b/fs/ocfs2/dlm/dlmlock.c @@ -77,8 +77,7 @@ int dlm_init_lock_cache(void) void dlm_destroy_lock_cache(void) { - if (dlm_lock_cache) - kmem_cache_destroy(dlm_lock_cache); + kmem_cache_destroy(dlm_lock_cache); } /* Tell us whether we can grant a new lock request. diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index a7df226f9449..aaca0949fe53 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -414,8 +414,7 @@ int dlm_init_mle_cache(void) void dlm_destroy_mle_cache(void) { - if (dlm_mle_cache) - kmem_cache_destroy(dlm_mle_cache); + kmem_cache_destroy(dlm_mle_cache); } static void dlm_mle_release(struct kref *kref) @@ -472,15 +471,11 @@ bail: void dlm_destroy_master_caches(void) { - if (dlm_lockname_cache) { - kmem_cache_destroy(dlm_lockname_cache); - dlm_lockname_cache = NULL; - } + kmem_cache_destroy(dlm_lockname_cache); + dlm_lockname_cache = NULL; - if (dlm_lockres_cache) { - kmem_cache_destroy(dlm_lockres_cache); - dlm_lockres_cache = NULL; - } + kmem_cache_destroy(dlm_lockres_cache); + dlm_lockres_cache = NULL; } static void dlm_lockres_release(struct kref *kref) @@ -2495,13 +2490,13 @@ static void dlm_deref_lockres_worker(struct dlm_work_item *item, void *data) } /* - * A migrateable resource is one that is : + * A migratable resource is one that is : * 1. locally mastered, and, * 2. zero local locks, and, * 3. one or more non-local locks, or, one or more references * Returns 1 if yes, 0 if not. */ -static int dlm_is_lockres_migrateable(struct dlm_ctxt *dlm, +static int dlm_is_lockres_migratable(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { enum dlm_lockres_list idx; @@ -2532,7 +2527,7 @@ static int dlm_is_lockres_migrateable(struct dlm_ctxt *dlm, continue; } cookie = be64_to_cpu(lock->ml.cookie); - mlog(0, "%s: Not migrateable res %.*s, lock %u:%llu on " + mlog(0, "%s: Not migratable res %.*s, lock %u:%llu on " "%s list\n", dlm->name, res->lockname.len, res->lockname.name, dlm_get_lock_cookie_node(cookie), @@ -2548,7 +2543,7 @@ static int dlm_is_lockres_migrateable(struct dlm_ctxt *dlm, return 0; } - mlog(0, "%s: res %.*s, Migrateable\n", dlm->name, res->lockname.len, + mlog(0, "%s: res %.*s, Migratable\n", dlm->name, res->lockname.len, res->lockname.name); return 1; @@ -2792,7 +2787,7 @@ int dlm_empty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) assert_spin_locked(&dlm->spinlock); spin_lock(&res->spinlock); - if (dlm_is_lockres_migrateable(dlm, res)) + if (dlm_is_lockres_migratable(dlm, res)) target = dlm_pick_migration_target(dlm, res); spin_unlock(&res->spinlock); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index ec8f75813beb..802636d50365 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -62,7 +62,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node); static int dlm_init_recovery_area(struct dlm_ctxt *dlm, u8 dead_node); static int dlm_request_all_locks(struct dlm_ctxt *dlm, u8 request_from, u8 dead_node); -static void dlm_destroy_recovery_area(struct dlm_ctxt *dlm, u8 dead_node); +static void dlm_destroy_recovery_area(struct dlm_ctxt *dlm); static inline int dlm_num_locks_in_lockres(struct dlm_lock_resource *res); static void dlm_init_migratable_lockres(struct dlm_migratable_lockres *mres, @@ -423,12 +423,11 @@ void dlm_wait_for_recovery(struct dlm_ctxt *dlm) static void dlm_begin_recovery(struct dlm_ctxt *dlm) { - spin_lock(&dlm->spinlock); + assert_spin_locked(&dlm->spinlock); BUG_ON(dlm->reco.state & DLM_RECO_STATE_ACTIVE); printk(KERN_NOTICE "o2dlm: Begin recovery on domain %s for node %u\n", dlm->name, dlm->reco.dead_node); dlm->reco.state |= DLM_RECO_STATE_ACTIVE; - spin_unlock(&dlm->spinlock); } static void dlm_end_recovery(struct dlm_ctxt *dlm) @@ -456,6 +455,13 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) spin_lock(&dlm->spinlock); + if (dlm->migrate_done) { + mlog(0, "%s: no need do recovery after migrating all " + "lock resources\n", dlm->name); + spin_unlock(&dlm->spinlock); + return 0; + } + /* check to see if the new master has died */ if (dlm->reco.new_master != O2NM_INVALID_NODE_NUM && test_bit(dlm->reco.new_master, dlm->recovery_map)) { @@ -490,12 +496,13 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) mlog(0, "%s(%d):recovery thread found node %u in the recovery map!\n", dlm->name, task_pid_nr(dlm->dlm_reco_thread_task), dlm->reco.dead_node); - spin_unlock(&dlm->spinlock); /* take write barrier */ /* (stops the list reshuffling thread, proxy ast handling) */ dlm_begin_recovery(dlm); + spin_unlock(&dlm->spinlock); + if (dlm->reco.new_master == dlm->node_num) goto master_here; @@ -739,7 +746,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node) } if (destroy) - dlm_destroy_recovery_area(dlm, dead_node); + dlm_destroy_recovery_area(dlm); return status; } @@ -764,7 +771,7 @@ static int dlm_init_recovery_area(struct dlm_ctxt *dlm, u8 dead_node) ndata = kzalloc(sizeof(*ndata), GFP_NOFS); if (!ndata) { - dlm_destroy_recovery_area(dlm, dead_node); + dlm_destroy_recovery_area(dlm); return -ENOMEM; } ndata->node_num = num; @@ -778,7 +785,7 @@ static int dlm_init_recovery_area(struct dlm_ctxt *dlm, u8 dead_node) return 0; } -static void dlm_destroy_recovery_area(struct dlm_ctxt *dlm, u8 dead_node) +static void dlm_destroy_recovery_area(struct dlm_ctxt *dlm) { struct dlm_reco_node_data *ndata, *next; LIST_HEAD(tmplist); @@ -1378,6 +1385,15 @@ int dlm_mig_lockres_handler(struct o2net_msg *msg, u32 len, void *data, if (!dlm_grab(dlm)) return -EINVAL; + if (!dlm_joined(dlm)) { + mlog(ML_ERROR, "Domain %s not joined! " + "lockres %.*s, master %u\n", + dlm->name, mres->lockname_len, + mres->lockname, mres->master); + dlm_put(dlm); + return -EINVAL; + } + BUG_ON(!(mres->flags & (DLM_MRES_RECOVERY|DLM_MRES_MIGRATION))); real_master = mres->master; @@ -1807,7 +1823,6 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm, int i, j, bad; struct dlm_lock *lock; u8 from = O2NM_MAX_NODES; - unsigned int added = 0; __be64 c; mlog(0, "running %d locks for this lockres\n", mres->num_locks); @@ -1823,7 +1838,6 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm, spin_lock(&res->spinlock); dlm_lockres_set_refmap_bit(dlm, res, from); spin_unlock(&res->spinlock); - added++; break; } BUG_ON(ml->highest_blocked != LKM_IVMODE); @@ -1911,7 +1925,6 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm, /* do not alter lock refcount. switching lists. */ list_move_tail(&lock->list, queue); spin_unlock(&res->spinlock); - added++; mlog(0, "just reordered a local lock!\n"); continue; @@ -2037,7 +2050,6 @@ skip_lvb: "setting refmap bit\n", dlm->name, res->lockname.len, res->lockname.name, ml->node); dlm_lockres_set_refmap_bit(dlm, res, ml->node); - added++; } spin_unlock(&res->spinlock); } @@ -2331,13 +2343,6 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm, __dlm_dirty_lockres(dlm, res); } -/* if this node is the recovery master, and there are no - * locks for a given lockres owned by this node that are in - * either PR or EX mode, zero out the lvb before requesting. - * - */ - - static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node) { struct dlm_lock_resource *res; diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c index 9479f99c2145..97a972efab83 100644 --- a/fs/ocfs2/dlmglue.c +++ b/fs/ocfs2/dlmglue.c @@ -1756,8 +1756,7 @@ int ocfs2_rw_lock(struct inode *inode, int write) level = write ? DLM_LOCK_EX : DLM_LOCK_PR; - status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, level, 0, - 0); + status = ocfs2_cluster_lock(osb, lockres, level, 0, 0); if (status < 0) mlog_errno(status); @@ -1796,7 +1795,7 @@ void ocfs2_rw_unlock(struct inode *inode, int write) write ? "EXMODE" : "PRMODE"); if (!ocfs2_mount_local(osb)) - ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level); + ocfs2_cluster_unlock(osb, lockres, level); } /* @@ -1816,8 +1815,7 @@ int ocfs2_open_lock(struct inode *inode) lockres = &OCFS2_I(inode)->ip_open_lockres; - status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, - DLM_LOCK_PR, 0, 0); + status = ocfs2_cluster_lock(osb, lockres, DLM_LOCK_PR, 0, 0); if (status < 0) mlog_errno(status); @@ -1854,8 +1852,7 @@ int ocfs2_try_open_lock(struct inode *inode, int write) * other nodes and the -EAGAIN will indicate to the caller that * this inode is still in use. */ - status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, - level, DLM_LKF_NOQUEUE, 0); + status = ocfs2_cluster_lock(osb, lockres, level, DLM_LKF_NOQUEUE, 0); out: return status; @@ -1876,11 +1873,9 @@ void ocfs2_open_unlock(struct inode *inode) goto out; if(lockres->l_ro_holders) - ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, - DLM_LOCK_PR); + ocfs2_cluster_unlock(osb, lockres, DLM_LOCK_PR); if(lockres->l_ex_holders) - ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, - DLM_LOCK_EX); + ocfs2_cluster_unlock(osb, lockres, DLM_LOCK_EX); out: return; @@ -2601,9 +2596,9 @@ void ocfs2_inode_unlock(struct inode *inode, (unsigned long long)OCFS2_I(inode)->ip_blkno, ex ? "EXMODE" : "PRMODE"); - if (!ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb)) && + if (!ocfs2_is_hard_readonly(osb) && !ocfs2_mount_local(osb)) - ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level); + ocfs2_cluster_unlock(osb, lockres, level); } /* @@ -3537,7 +3532,7 @@ static int ocfs2_downconvert_lock(struct ocfs2_super *osb, * On DLM_LKF_VALBLK, fsdlm behaves differently with o2cb. It always * expects DLM_LKF_VALBLK being set if the LKB has LVB, so that * we can recover correctly from node failure. Otherwise, we may get - * invalid LVB in LKB, but without DLM_SBF_VALNOTVALID being set. + * invalid LVB in LKB, but without DLM_SBF_VALNOTVALID being set. */ if (!ocfs2_is_o2cb_active() && lockres->l_ops->flags & LOCK_TYPE_USES_LVB) diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 5d1784a365a3..6ee94bc23f5b 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -101,7 +101,7 @@ static int ocfs2_file_open(struct inode *inode, struct file *file) struct ocfs2_inode_info *oi = OCFS2_I(inode); trace_ocfs2_file_open(inode, file, file->f_path.dentry, - (unsigned long long)OCFS2_I(inode)->ip_blkno, + (unsigned long long)oi->ip_blkno, file->f_path.dentry->d_name.len, file->f_path.dentry->d_name.name, mode); @@ -116,7 +116,7 @@ static int ocfs2_file_open(struct inode *inode, struct file *file) /* Check that the inode hasn't been wiped from disk by another * node. If it hasn't then we're safe as long as we hold the * spin lock until our increment of open count. */ - if (OCFS2_I(inode)->ip_flags & OCFS2_INODE_DELETED) { + if (oi->ip_flags & OCFS2_INODE_DELETED) { spin_unlock(&oi->ip_lock); status = -ENOENT; @@ -190,7 +190,7 @@ static int ocfs2_sync_file(struct file *file, loff_t start, loff_t end, bool needs_barrier = false; trace_ocfs2_sync_file(inode, file, file->f_path.dentry, - OCFS2_I(inode)->ip_blkno, + oi->ip_blkno, file->f_path.dentry->d_name.len, file->f_path.dentry->d_name.name, (unsigned long long)datasync); @@ -296,7 +296,7 @@ int ocfs2_update_inode_atime(struct inode *inode, ocfs2_journal_dirty(handle, bh); out_commit: - ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle); + ocfs2_commit_trans(osb, handle); out: return ret; } @@ -2257,7 +2257,7 @@ static ssize_t ocfs2_file_write_iter(struct kiocb *iocb, int direct_io = iocb->ki_flags & IOCB_DIRECT ? 1 : 0; int nowait = iocb->ki_flags & IOCB_NOWAIT ? 1 : 0; - trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry, + trace_ocfs2_file_write_iter(inode, file, file->f_path.dentry, (unsigned long long)OCFS2_I(inode)->ip_blkno, file->f_path.dentry->d_name.len, file->f_path.dentry->d_name.name, @@ -2405,7 +2405,7 @@ static ssize_t ocfs2_file_read_iter(struct kiocb *iocb, int direct_io = iocb->ki_flags & IOCB_DIRECT ? 1 : 0; int nowait = iocb->ki_flags & IOCB_NOWAIT ? 1 : 0; - trace_ocfs2_file_aio_read(inode, filp, filp->f_path.dentry, + trace_ocfs2_file_read_iter(inode, filp, filp->f_path.dentry, (unsigned long long)OCFS2_I(inode)->ip_blkno, filp->f_path.dentry->d_name.len, filp->f_path.dentry->d_name.name, @@ -2448,7 +2448,7 @@ static ssize_t ocfs2_file_read_iter(struct kiocb *iocb, * * Take and drop the meta data lock to update inode fields * like i_size. This allows the checks down below - * generic_file_aio_read() a chance of actually working. + * generic_file_read_iter() a chance of actually working. */ ret = ocfs2_inode_lock_atime(inode, filp->f_path.mnt, &lock_level, !nowait); @@ -2460,7 +2460,7 @@ static ssize_t ocfs2_file_read_iter(struct kiocb *iocb, ocfs2_inode_unlock(inode, lock_level); ret = generic_file_read_iter(iocb, to); - trace_generic_file_aio_read_ret(ret); + trace_generic_file_read_iter_ret(ret); /* buffered aio wouldn't have proper lock coverage today */ BUG_ON(ret == -EIOCBQUEUED && !(iocb->ki_flags & IOCB_DIRECT)); diff --git a/fs/ocfs2/filecheck.c b/fs/ocfs2/filecheck.c index 6b92cb241138..f65f2b2f594d 100644 --- a/fs/ocfs2/filecheck.c +++ b/fs/ocfs2/filecheck.c @@ -53,36 +53,6 @@ static const char * const ocfs2_filecheck_errs[] = { "UNSUPPORTED" }; -static DEFINE_SPINLOCK(ocfs2_filecheck_sysfs_lock); -static LIST_HEAD(ocfs2_filecheck_sysfs_list); - -struct ocfs2_filecheck { - struct list_head fc_head; /* File check entry list head */ - spinlock_t fc_lock; - unsigned int fc_max; /* Maximum number of entry in list */ - unsigned int fc_size; /* Current entry count in list */ - unsigned int fc_done; /* Finished entry count in list */ -}; - -struct ocfs2_filecheck_sysfs_entry { /* sysfs entry per mounting */ - struct list_head fs_list; - atomic_t fs_count; - struct super_block *fs_sb; - struct kset *fs_devicekset; - struct kset *fs_fcheckkset; - struct ocfs2_filecheck *fs_fcheck; -}; - -#define OCFS2_FILECHECK_MAXSIZE 100 -#define OCFS2_FILECHECK_MINSIZE 10 - -/* File check operation type */ -enum { - OCFS2_FILECHECK_TYPE_CHK = 0, /* Check a file(inode) */ - OCFS2_FILECHECK_TYPE_FIX, /* Fix a file(inode) */ - OCFS2_FILECHECK_TYPE_SET = 100 /* Set entry list maximum size */ -}; - struct ocfs2_filecheck_entry { struct list_head fe_list; unsigned long fe_ino; @@ -110,35 +80,84 @@ ocfs2_filecheck_error(int errno) return ocfs2_filecheck_errs[errno - OCFS2_FILECHECK_ERR_START + 1]; } -static ssize_t ocfs2_filecheck_show(struct kobject *kobj, - struct kobj_attribute *attr, - char *buf); -static ssize_t ocfs2_filecheck_store(struct kobject *kobj, - struct kobj_attribute *attr, - const char *buf, size_t count); -static struct kobj_attribute ocfs2_attr_filecheck_chk = +static ssize_t ocfs2_filecheck_attr_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf); +static ssize_t ocfs2_filecheck_attr_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count); +static struct kobj_attribute ocfs2_filecheck_attr_chk = __ATTR(check, S_IRUSR | S_IWUSR, - ocfs2_filecheck_show, - ocfs2_filecheck_store); -static struct kobj_attribute ocfs2_attr_filecheck_fix = + ocfs2_filecheck_attr_show, + ocfs2_filecheck_attr_store); +static struct kobj_attribute ocfs2_filecheck_attr_fix = __ATTR(fix, S_IRUSR | S_IWUSR, - ocfs2_filecheck_show, - ocfs2_filecheck_store); -static struct kobj_attribute ocfs2_attr_filecheck_set = + ocfs2_filecheck_attr_show, + ocfs2_filecheck_attr_store); +static struct kobj_attribute ocfs2_filecheck_attr_set = __ATTR(set, S_IRUSR | S_IWUSR, - ocfs2_filecheck_show, - ocfs2_filecheck_store); + ocfs2_filecheck_attr_show, + ocfs2_filecheck_attr_store); +static struct attribute *ocfs2_filecheck_attrs[] = { + &ocfs2_filecheck_attr_chk.attr, + &ocfs2_filecheck_attr_fix.attr, + &ocfs2_filecheck_attr_set.attr, + NULL +}; + +static void ocfs2_filecheck_release(struct kobject *kobj) +{ + struct ocfs2_filecheck_sysfs_entry *entry = container_of(kobj, + struct ocfs2_filecheck_sysfs_entry, fs_kobj); + + complete(&entry->fs_kobj_unregister); +} + +static ssize_t +ocfs2_filecheck_show(struct kobject *kobj, struct attribute *attr, char *buf) +{ + ssize_t ret = -EIO; + struct kobj_attribute *kattr = container_of(attr, + struct kobj_attribute, attr); + + kobject_get(kobj); + if (kattr->show) + ret = kattr->show(kobj, kattr, buf); + kobject_put(kobj); + return ret; +} + +static ssize_t +ocfs2_filecheck_store(struct kobject *kobj, struct attribute *attr, + const char *buf, size_t count) +{ + ssize_t ret = -EIO; + struct kobj_attribute *kattr = container_of(attr, + struct kobj_attribute, attr); + + kobject_get(kobj); + if (kattr->store) + ret = kattr->store(kobj, kattr, buf, count); + kobject_put(kobj); + return ret; +} + +static const struct sysfs_ops ocfs2_filecheck_ops = { + .show = ocfs2_filecheck_show, + .store = ocfs2_filecheck_store, +}; + +static struct kobj_type ocfs2_ktype_filecheck = { + .default_attrs = ocfs2_filecheck_attrs, + .sysfs_ops = &ocfs2_filecheck_ops, + .release = ocfs2_filecheck_release, +}; static void ocfs2_filecheck_sysfs_free(struct ocfs2_filecheck_sysfs_entry *entry) { struct ocfs2_filecheck_entry *p; - if (!atomic_dec_and_test(&entry->fs_count)) { - wait_var_event(&entry->fs_count, - !atomic_read(&entry->fs_count)); - } - spin_lock(&entry->fs_fcheck->fc_lock); while (!list_empty(&entry->fs_fcheck->fc_head)) { p = list_first_entry(&entry->fs_fcheck->fc_head, @@ -149,151 +168,48 @@ ocfs2_filecheck_sysfs_free(struct ocfs2_filecheck_sysfs_entry *entry) } spin_unlock(&entry->fs_fcheck->fc_lock); - kset_unregister(entry->fs_fcheckkset); - kset_unregister(entry->fs_devicekset); kfree(entry->fs_fcheck); - kfree(entry); -} - -static void -ocfs2_filecheck_sysfs_add(struct ocfs2_filecheck_sysfs_entry *entry) -{ - spin_lock(&ocfs2_filecheck_sysfs_lock); - list_add_tail(&entry->fs_list, &ocfs2_filecheck_sysfs_list); - spin_unlock(&ocfs2_filecheck_sysfs_lock); + entry->fs_fcheck = NULL; } -static int ocfs2_filecheck_sysfs_del(const char *devname) +int ocfs2_filecheck_create_sysfs(struct ocfs2_super *osb) { - struct ocfs2_filecheck_sysfs_entry *p; - - spin_lock(&ocfs2_filecheck_sysfs_lock); - list_for_each_entry(p, &ocfs2_filecheck_sysfs_list, fs_list) { - if (!strcmp(p->fs_sb->s_id, devname)) { - list_del(&p->fs_list); - spin_unlock(&ocfs2_filecheck_sysfs_lock); - ocfs2_filecheck_sysfs_free(p); - return 0; - } - } - spin_unlock(&ocfs2_filecheck_sysfs_lock); - return 1; -} - -static void -ocfs2_filecheck_sysfs_put(struct ocfs2_filecheck_sysfs_entry *entry) -{ - if (atomic_dec_and_test(&entry->fs_count)) - wake_up_var(&entry->fs_count); -} - -static struct ocfs2_filecheck_sysfs_entry * -ocfs2_filecheck_sysfs_get(const char *devname) -{ - struct ocfs2_filecheck_sysfs_entry *p = NULL; - - spin_lock(&ocfs2_filecheck_sysfs_lock); - list_for_each_entry(p, &ocfs2_filecheck_sysfs_list, fs_list) { - if (!strcmp(p->fs_sb->s_id, devname)) { - atomic_inc(&p->fs_count); - spin_unlock(&ocfs2_filecheck_sysfs_lock); - return p; - } - } - spin_unlock(&ocfs2_filecheck_sysfs_lock); - return NULL; -} - -int ocfs2_filecheck_create_sysfs(struct super_block *sb) -{ - int ret = 0; - struct kset *device_kset = NULL; - struct kset *fcheck_kset = NULL; - struct ocfs2_filecheck *fcheck = NULL; - struct ocfs2_filecheck_sysfs_entry *entry = NULL; - struct attribute **attrs = NULL; - struct attribute_group attrgp; - - if (!ocfs2_kset) - return -ENOMEM; - - attrs = kmalloc(sizeof(struct attribute *) * 4, GFP_NOFS); - if (!attrs) { - ret = -ENOMEM; - goto error; - } else { - attrs[0] = &ocfs2_attr_filecheck_chk.attr; - attrs[1] = &ocfs2_attr_filecheck_fix.attr; - attrs[2] = &ocfs2_attr_filecheck_set.attr; - attrs[3] = NULL; - memset(&attrgp, 0, sizeof(attrgp)); - attrgp.attrs = attrs; - } + int ret; + struct ocfs2_filecheck *fcheck; + struct ocfs2_filecheck_sysfs_entry *entry = &osb->osb_fc_ent; fcheck = kmalloc(sizeof(struct ocfs2_filecheck), GFP_NOFS); - if (!fcheck) { - ret = -ENOMEM; - goto error; - } else { - INIT_LIST_HEAD(&fcheck->fc_head); - spin_lock_init(&fcheck->fc_lock); - fcheck->fc_max = OCFS2_FILECHECK_MINSIZE; - fcheck->fc_size = 0; - fcheck->fc_done = 0; - } - - if (strlen(sb->s_id) <= 0) { - mlog(ML_ERROR, - "Cannot get device basename when create filecheck sysfs\n"); - ret = -ENODEV; - goto error; - } - - device_kset = kset_create_and_add(sb->s_id, NULL, &ocfs2_kset->kobj); - if (!device_kset) { - ret = -ENOMEM; - goto error; - } - - fcheck_kset = kset_create_and_add("filecheck", NULL, - &device_kset->kobj); - if (!fcheck_kset) { - ret = -ENOMEM; - goto error; - } - - ret = sysfs_create_group(&fcheck_kset->kobj, &attrgp); - if (ret) - goto error; + if (!fcheck) + return -ENOMEM; - entry = kmalloc(sizeof(struct ocfs2_filecheck_sysfs_entry), GFP_NOFS); - if (!entry) { - ret = -ENOMEM; - goto error; - } else { - atomic_set(&entry->fs_count, 1); - entry->fs_sb = sb; - entry->fs_devicekset = device_kset; - entry->fs_fcheckkset = fcheck_kset; - entry->fs_fcheck = fcheck; - ocfs2_filecheck_sysfs_add(entry); + INIT_LIST_HEAD(&fcheck->fc_head); + spin_lock_init(&fcheck->fc_lock); + fcheck->fc_max = OCFS2_FILECHECK_MINSIZE; + fcheck->fc_size = 0; + fcheck->fc_done = 0; + + entry->fs_kobj.kset = osb->osb_dev_kset; + init_completion(&entry->fs_kobj_unregister); + ret = kobject_init_and_add(&entry->fs_kobj, &ocfs2_ktype_filecheck, + NULL, "filecheck"); + if (ret) { + kfree(fcheck); + return ret; } - kfree(attrs); + entry->fs_fcheck = fcheck; return 0; - -error: - kfree(attrs); - kfree(entry); - kfree(fcheck); - kset_unregister(fcheck_kset); - kset_unregister(device_kset); - return ret; } -int ocfs2_filecheck_remove_sysfs(struct super_block *sb) +void ocfs2_filecheck_remove_sysfs(struct ocfs2_super *osb) { - return ocfs2_filecheck_sysfs_del(sb->s_id); + if (!osb->osb_fc_ent.fs_fcheck) + return; + + kobject_del(&osb->osb_fc_ent.fs_kobj); + kobject_put(&osb->osb_fc_ent.fs_kobj); + wait_for_completion(&osb->osb_fc_ent.fs_kobj_unregister); + ocfs2_filecheck_sysfs_free(&osb->osb_fc_ent); } static int @@ -310,7 +226,7 @@ ocfs2_filecheck_adjust_max(struct ocfs2_filecheck_sysfs_entry *ent, spin_lock(&ent->fs_fcheck->fc_lock); if (len < (ent->fs_fcheck->fc_size - ent->fs_fcheck->fc_done)) { - mlog(ML_ERROR, + mlog(ML_NOTICE, "Cannot set online file check maximum entry number " "to %u due to too many pending entries(%u)\n", len, ent->fs_fcheck->fc_size - ent->fs_fcheck->fc_done); @@ -387,7 +303,7 @@ ocfs2_filecheck_args_parse(const char *name, const char *buf, size_t count, return 0; } -static ssize_t ocfs2_filecheck_show(struct kobject *kobj, +static ssize_t ocfs2_filecheck_attr_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { @@ -395,19 +311,12 @@ static ssize_t ocfs2_filecheck_show(struct kobject *kobj, ssize_t ret = 0, total = 0, remain = PAGE_SIZE; unsigned int type; struct ocfs2_filecheck_entry *p; - struct ocfs2_filecheck_sysfs_entry *ent; + struct ocfs2_filecheck_sysfs_entry *ent = container_of(kobj, + struct ocfs2_filecheck_sysfs_entry, fs_kobj); if (ocfs2_filecheck_type_parse(attr->attr.name, &type)) return -EINVAL; - ent = ocfs2_filecheck_sysfs_get(kobj->parent->name); - if (!ent) { - mlog(ML_ERROR, - "Cannot get the corresponding entry via device basename %s\n", - kobj->name); - return -ENODEV; - } - if (type == OCFS2_FILECHECK_TYPE_SET) { spin_lock(&ent->fs_fcheck->fc_lock); total = snprintf(buf, remain, "%u\n", ent->fs_fcheck->fc_max); @@ -441,11 +350,26 @@ static ssize_t ocfs2_filecheck_show(struct kobject *kobj, spin_unlock(&ent->fs_fcheck->fc_lock); exit: - ocfs2_filecheck_sysfs_put(ent); return total; } -static int +static inline int +ocfs2_filecheck_is_dup_entry(struct ocfs2_filecheck_sysfs_entry *ent, + unsigned long ino) +{ + struct ocfs2_filecheck_entry *p; + + list_for_each_entry(p, &ent->fs_fcheck->fc_head, fe_list) { + if (!p->fe_done) { + if (p->fe_ino == ino) + return 1; + } + } + + return 0; +} + +static inline int ocfs2_filecheck_erase_entry(struct ocfs2_filecheck_sysfs_entry *ent) { struct ocfs2_filecheck_entry *p; @@ -484,21 +408,21 @@ static void ocfs2_filecheck_done_entry(struct ocfs2_filecheck_sysfs_entry *ent, struct ocfs2_filecheck_entry *entry) { - entry->fe_done = 1; spin_lock(&ent->fs_fcheck->fc_lock); + entry->fe_done = 1; ent->fs_fcheck->fc_done++; spin_unlock(&ent->fs_fcheck->fc_lock); } static unsigned int -ocfs2_filecheck_handle(struct super_block *sb, +ocfs2_filecheck_handle(struct ocfs2_super *osb, unsigned long ino, unsigned int flags) { unsigned int ret = OCFS2_FILECHECK_ERR_SUCCESS; struct inode *inode = NULL; int rc; - inode = ocfs2_iget(OCFS2_SB(sb), ino, flags, 0); + inode = ocfs2_iget(osb, ino, flags, 0); if (IS_ERR(inode)) { rc = (int)(-(long)inode); if (rc >= OCFS2_FILECHECK_ERR_START && @@ -516,11 +440,14 @@ static void ocfs2_filecheck_handle_entry(struct ocfs2_filecheck_sysfs_entry *ent, struct ocfs2_filecheck_entry *entry) { + struct ocfs2_super *osb = container_of(ent, struct ocfs2_super, + osb_fc_ent); + if (entry->fe_type == OCFS2_FILECHECK_TYPE_CHK) - entry->fe_status = ocfs2_filecheck_handle(ent->fs_sb, + entry->fe_status = ocfs2_filecheck_handle(osb, entry->fe_ino, OCFS2_FI_FLAG_FILECHECK_CHK); else if (entry->fe_type == OCFS2_FILECHECK_TYPE_FIX) - entry->fe_status = ocfs2_filecheck_handle(ent->fs_sb, + entry->fe_status = ocfs2_filecheck_handle(osb, entry->fe_ino, OCFS2_FI_FLAG_FILECHECK_FIX); else entry->fe_status = OCFS2_FILECHECK_ERR_UNSUPPORTED; @@ -528,30 +455,21 @@ ocfs2_filecheck_handle_entry(struct ocfs2_filecheck_sysfs_entry *ent, ocfs2_filecheck_done_entry(ent, entry); } -static ssize_t ocfs2_filecheck_store(struct kobject *kobj, +static ssize_t ocfs2_filecheck_attr_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t count) { + ssize_t ret = 0; struct ocfs2_filecheck_args args; struct ocfs2_filecheck_entry *entry; - struct ocfs2_filecheck_sysfs_entry *ent; - ssize_t ret = 0; + struct ocfs2_filecheck_sysfs_entry *ent = container_of(kobj, + struct ocfs2_filecheck_sysfs_entry, fs_kobj); if (count == 0) return count; - if (ocfs2_filecheck_args_parse(attr->attr.name, buf, count, &args)) { - mlog(ML_ERROR, "Invalid arguments for online file check\n"); + if (ocfs2_filecheck_args_parse(attr->attr.name, buf, count, &args)) return -EINVAL; - } - - ent = ocfs2_filecheck_sysfs_get(kobj->parent->name); - if (!ent) { - mlog(ML_ERROR, - "Cannot get the corresponding entry via device basename %s\n", - kobj->parent->name); - return -ENODEV; - } if (args.fa_type == OCFS2_FILECHECK_TYPE_SET) { ret = ocfs2_filecheck_adjust_max(ent, args.fa_len); @@ -565,13 +483,16 @@ static ssize_t ocfs2_filecheck_store(struct kobject *kobj, } spin_lock(&ent->fs_fcheck->fc_lock); - if ((ent->fs_fcheck->fc_size >= ent->fs_fcheck->fc_max) && - (ent->fs_fcheck->fc_done == 0)) { - mlog(ML_ERROR, + if (ocfs2_filecheck_is_dup_entry(ent, args.fa_ino)) { + ret = -EEXIST; + kfree(entry); + } else if ((ent->fs_fcheck->fc_size >= ent->fs_fcheck->fc_max) && + (ent->fs_fcheck->fc_done == 0)) { + mlog(ML_NOTICE, "Cannot do more file check " "since file check queue(%u) is full now\n", ent->fs_fcheck->fc_max); - ret = -EBUSY; + ret = -EAGAIN; kfree(entry); } else { if ((ent->fs_fcheck->fc_size >= ent->fs_fcheck->fc_max) && @@ -596,6 +517,5 @@ static ssize_t ocfs2_filecheck_store(struct kobject *kobj, ocfs2_filecheck_handle_entry(ent, entry); exit: - ocfs2_filecheck_sysfs_put(ent); return (!ret ? count : ret); } diff --git a/fs/ocfs2/filecheck.h b/fs/ocfs2/filecheck.h index e5cd002a2c09..6a22ee79e8d0 100644 --- a/fs/ocfs2/filecheck.h +++ b/fs/ocfs2/filecheck.h @@ -43,7 +43,32 @@ enum { #define OCFS2_FILECHECK_ERR_START OCFS2_FILECHECK_ERR_FAILED #define OCFS2_FILECHECK_ERR_END OCFS2_FILECHECK_ERR_UNSUPPORTED -int ocfs2_filecheck_create_sysfs(struct super_block *sb); -int ocfs2_filecheck_remove_sysfs(struct super_block *sb); +struct ocfs2_filecheck { + struct list_head fc_head; /* File check entry list head */ + spinlock_t fc_lock; + unsigned int fc_max; /* Maximum number of entry in list */ + unsigned int fc_size; /* Current entry count in list */ + unsigned int fc_done; /* Finished entry count in list */ +}; + +#define OCFS2_FILECHECK_MAXSIZE 100 +#define OCFS2_FILECHECK_MINSIZE 10 + +/* File check operation type */ +enum { + OCFS2_FILECHECK_TYPE_CHK = 0, /* Check a file(inode) */ + OCFS2_FILECHECK_TYPE_FIX, /* Fix a file(inode) */ + OCFS2_FILECHECK_TYPE_SET = 100 /* Set entry list maximum size */ +}; + +struct ocfs2_filecheck_sysfs_entry { /* sysfs entry per partition */ + struct kobject fs_kobj; + struct completion fs_kobj_unregister; + struct ocfs2_filecheck *fs_fcheck; +}; + + +int ocfs2_filecheck_create_sysfs(struct ocfs2_super *osb); +void ocfs2_filecheck_remove_sysfs(struct ocfs2_super *osb); #endif /* FILECHECK_H */ diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index d51b80edd972..ddc3e9470c87 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -1135,7 +1135,7 @@ static void ocfs2_clear_inode(struct inode *inode) trace_ocfs2_clear_inode((unsigned long long)oi->ip_blkno, inode->i_nlink); - mlog_bug_on_msg(OCFS2_SB(inode->i_sb) == NULL, + mlog_bug_on_msg(osb == NULL, "Inode=%lu\n", inode->i_ino); dquot_drop(inode); @@ -1150,7 +1150,7 @@ static void ocfs2_clear_inode(struct inode *inode) ocfs2_mark_lockres_freeing(osb, &oi->ip_inode_lockres); ocfs2_mark_lockres_freeing(osb, &oi->ip_open_lockres); - ocfs2_resv_discard(&OCFS2_SB(inode->i_sb)->osb_la_resmap, + ocfs2_resv_discard(&osb->osb_la_resmap, &oi->ip_la_data_resv); ocfs2_resv_init_once(&oi->ip_la_data_resv); @@ -1160,7 +1160,7 @@ static void ocfs2_clear_inode(struct inode *inode) * exception here are successfully wiped inodes - their * metadata can now be considered to be part of the system * inodes from which it came. */ - if (!(OCFS2_I(inode)->ip_flags & OCFS2_INODE_DELETED)) + if (!(oi->ip_flags & OCFS2_INODE_DELETED)) ocfs2_checkpoint_inode(inode); mlog_bug_on_msg(!list_empty(&oi->ip_io_markers), @@ -1223,7 +1223,7 @@ static void ocfs2_clear_inode(struct inode *inode) * the journal is flushed before journal shutdown. Thus it is safe to * have inodes get cleaned up after journal shutdown. */ - jbd2_journal_release_jbd_inode(OCFS2_SB(inode->i_sb)->journal->j_journal, + jbd2_journal_release_jbd_inode(osb->journal->j_journal, &oi->ip_jinode); } diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index c801eddc4bf3..8dd6f703c819 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -525,7 +525,7 @@ static int __ocfs2_mknod_locked(struct inode *dir, * these are used by the support functions here and in * callers. */ inode->i_ino = ino_from_blkno(osb->sb, fe_blkno); - OCFS2_I(inode)->ip_blkno = fe_blkno; + oi->ip_blkno = fe_blkno; spin_lock(&osb->osb_lock); inode->i_generation = osb->s_next_generation++; spin_unlock(&osb->osb_lock); @@ -1186,8 +1186,8 @@ static int ocfs2_double_lock(struct ocfs2_super *osb, } trace_ocfs2_double_lock_end( - (unsigned long long)OCFS2_I(inode1)->ip_blkno, - (unsigned long long)OCFS2_I(inode2)->ip_blkno); + (unsigned long long)oi1->ip_blkno, + (unsigned long long)oi2->ip_blkno); bail: if (status) diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 6867eef2e06b..4f86ac0027b5 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -50,6 +50,8 @@ #include "reservations.h" +#include "filecheck.h" + /* Caching of metadata buffers */ /* Most user visible OCFS2 inodes will have very few pieces of @@ -472,6 +474,12 @@ struct ocfs2_super * workqueue and schedule on our own. */ struct workqueue_struct *ocfs2_wq; + + /* sysfs directory per partition */ + struct kset *osb_dev_kset; + + /* file check related stuff */ + struct ocfs2_filecheck_sysfs_entry osb_fc_ent; }; #define OCFS2_SB(sb) ((struct ocfs2_super *)(sb)->s_fs_info) diff --git a/fs/ocfs2/ocfs2_trace.h b/fs/ocfs2/ocfs2_trace.h index e2a11aaece10..2ee76a90ba8f 100644 --- a/fs/ocfs2/ocfs2_trace.h +++ b/fs/ocfs2/ocfs2_trace.h @@ -1311,11 +1311,11 @@ DEFINE_OCFS2_FILE_OPS(ocfs2_file_release); DEFINE_OCFS2_FILE_OPS(ocfs2_sync_file); -DEFINE_OCFS2_FILE_OPS(ocfs2_file_aio_write); +DEFINE_OCFS2_FILE_OPS(ocfs2_file_write_iter); DEFINE_OCFS2_FILE_OPS(ocfs2_file_splice_write); -DEFINE_OCFS2_FILE_OPS(ocfs2_file_aio_read); +DEFINE_OCFS2_FILE_OPS(ocfs2_file_read_iter); DEFINE_OCFS2_ULL_ULL_ULL_EVENT(ocfs2_truncate_file); @@ -1467,7 +1467,7 @@ TRACE_EVENT(ocfs2_prepare_inode_for_write, __entry->saved_pos, __entry->count, __entry->wait) ); -DEFINE_OCFS2_INT_EVENT(generic_file_aio_read_ret); +DEFINE_OCFS2_INT_EVENT(generic_file_read_iter_ret); /* End of trace events for fs/ocfs2/file.c. */ diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index ab156e35ec00..7869622af22a 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -573,7 +573,7 @@ static int ocfs2_create_refcount_tree(struct inode *inode, BUG_ON(ocfs2_is_refcount_inode(inode)); trace_ocfs2_create_refcount_tree( - (unsigned long long)OCFS2_I(inode)->ip_blkno); + (unsigned long long)oi->ip_blkno); ret = ocfs2_reserve_new_metadata_blocks(osb, 1, &meta_ac); if (ret) { @@ -3359,7 +3359,7 @@ static int ocfs2_replace_cow(struct ocfs2_cow_context *context) unsigned int ext_flags; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); - if (!ocfs2_refcount_tree(OCFS2_SB(inode->i_sb))) { + if (!ocfs2_refcount_tree(osb)) { return ocfs2_error(inode->i_sb, "Inode %lu want to use refcount tree, but the feature bit is not set in the super block\n", inode->i_ino); } @@ -3707,7 +3707,7 @@ int ocfs2_add_refcount_flag(struct inode *inode, trace_ocfs2_add_refcount_flag(ref_blocks, credits); if (ref_blocks) { - ret = ocfs2_reserve_new_metadata_blocks(OCFS2_SB(inode->i_sb), + ret = ocfs2_reserve_new_metadata_blocks(osb, ref_blocks, &meta_ac); if (ret) { mlog_errno(ret); @@ -4250,10 +4250,11 @@ out: static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry, bool preserve) { - int error; + int error, had_lock; struct inode *inode = d_inode(old_dentry); struct buffer_head *old_bh = NULL; struct inode *new_orphan_inode = NULL; + struct ocfs2_lock_holder oh; if (!ocfs2_refcount_tree(OCFS2_SB(inode->i_sb))) return -EOPNOTSUPP; @@ -4295,6 +4296,14 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir, goto out; } + had_lock = ocfs2_inode_lock_tracker(new_orphan_inode, NULL, 1, + &oh); + if (had_lock < 0) { + error = had_lock; + mlog_errno(error); + goto out; + } + /* If the security isn't preserved, we need to re-initialize them. */ if (!preserve) { error = ocfs2_init_security_and_acl(dir, new_orphan_inode, @@ -4302,14 +4311,15 @@ static int ocfs2_reflink(struct dentry *old_dentry, struct inode *dir, if (error) mlog_errno(error); } -out: if (!error) { error = ocfs2_mv_orphaned_inode_to_new(dir, new_orphan_inode, new_dentry); if (error) mlog_errno(error); } + ocfs2_inode_unlock_tracker(new_orphan_inode, 1, &oh, had_lock); +out: if (new_orphan_inode) { /* * We need to open_unlock the inode no matter whether we @@ -4766,8 +4776,8 @@ static int ocfs2_reflink_inodes_lock(struct inode *s_inode, *bh2 = *bh1; trace_ocfs2_double_lock_end( - (unsigned long long)OCFS2_I(inode1)->ip_blkno, - (unsigned long long)OCFS2_I(inode2)->ip_blkno); + (unsigned long long)oi1->ip_blkno, + (unsigned long long)oi2->ip_blkno); return 0; diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index d8f5f6ce99dc..f7c972fbed6a 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -79,8 +79,6 @@ static u64 ocfs2_group_from_res(struct ocfs2_suballoc_result *res) return ocfs2_which_suballoc_group(res->sr_blkno, res->sr_bit_offset); } -static inline void ocfs2_debug_bg(struct ocfs2_group_desc *bg); -static inline void ocfs2_debug_suballoc_inode(struct ocfs2_dinode *fe); static inline u16 ocfs2_find_victim_chain(struct ocfs2_chain_list *cl); static int ocfs2_block_group_fill(handle_t *handle, struct inode *alloc_inode, @@ -387,7 +385,7 @@ static int ocfs2_block_group_fill(handle_t *handle, memset(bg, 0, sb->s_blocksize); strcpy(bg->bg_signature, OCFS2_GROUP_DESC_SIGNATURE); - bg->bg_generation = cpu_to_le32(OCFS2_SB(sb)->fs_generation); + bg->bg_generation = cpu_to_le32(osb->fs_generation); bg->bg_size = cpu_to_le16(ocfs2_group_bitmap_size(sb, 1, osb->s_feature_incompat)); bg->bg_chain = cpu_to_le16(my_chain); @@ -1521,7 +1519,7 @@ static int ocfs2_cluster_group_search(struct inode *inode, OCFS2_I(inode)->ip_clusters, max_bits); } - ret = ocfs2_block_group_find_clear_bits(OCFS2_SB(inode->i_sb), + ret = ocfs2_block_group_find_clear_bits(osb, group_bh, bits_wanted, max_bits, res); if (ret) @@ -2626,53 +2624,6 @@ int ocfs2_release_clusters(handle_t *handle, _ocfs2_clear_bit); } -static inline void ocfs2_debug_bg(struct ocfs2_group_desc *bg) -{ - printk("Block Group:\n"); - printk("bg_signature: %s\n", bg->bg_signature); - printk("bg_size: %u\n", bg->bg_size); - printk("bg_bits: %u\n", bg->bg_bits); - printk("bg_free_bits_count: %u\n", bg->bg_free_bits_count); - printk("bg_chain: %u\n", bg->bg_chain); - printk("bg_generation: %u\n", le32_to_cpu(bg->bg_generation)); - printk("bg_next_group: %llu\n", - (unsigned long long)bg->bg_next_group); - printk("bg_parent_dinode: %llu\n", - (unsigned long long)bg->bg_parent_dinode); - printk("bg_blkno: %llu\n", - (unsigned long long)bg->bg_blkno); -} - -static inline void ocfs2_debug_suballoc_inode(struct ocfs2_dinode *fe) -{ - int i; - - printk("Suballoc Inode %llu:\n", (unsigned long long)fe->i_blkno); - printk("i_signature: %s\n", fe->i_signature); - printk("i_size: %llu\n", - (unsigned long long)fe->i_size); - printk("i_clusters: %u\n", fe->i_clusters); - printk("i_generation: %u\n", - le32_to_cpu(fe->i_generation)); - printk("id1.bitmap1.i_used: %u\n", - le32_to_cpu(fe->id1.bitmap1.i_used)); - printk("id1.bitmap1.i_total: %u\n", - le32_to_cpu(fe->id1.bitmap1.i_total)); - printk("id2.i_chain.cl_cpg: %u\n", fe->id2.i_chain.cl_cpg); - printk("id2.i_chain.cl_bpc: %u\n", fe->id2.i_chain.cl_bpc); - printk("id2.i_chain.cl_count: %u\n", fe->id2.i_chain.cl_count); - printk("id2.i_chain.cl_next_free_rec: %u\n", - fe->id2.i_chain.cl_next_free_rec); - for(i = 0; i < fe->id2.i_chain.cl_next_free_rec; i++) { - printk("fe->id2.i_chain.cl_recs[%d].c_free: %u\n", i, - fe->id2.i_chain.cl_recs[i].c_free); - printk("fe->id2.i_chain.cl_recs[%d].c_total: %u\n", i, - fe->id2.i_chain.cl_recs[i].c_total); - printk("fe->id2.i_chain.cl_recs[%d].c_blkno: %llu\n", i, - (unsigned long long)fe->id2.i_chain.cl_recs[i].c_blkno); - } -} - /* * For a given allocation, determine which allocators will need to be * accessed, and lock them, reserving the appropriate number of bits. diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index ffa4952d432b..3415e0b09398 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -423,10 +423,10 @@ static int ocfs2_sync_fs(struct super_block *sb, int wait) ocfs2_schedule_truncate_log_flush(osb, 0); } - if (jbd2_journal_start_commit(OCFS2_SB(sb)->journal->j_journal, + if (jbd2_journal_start_commit(osb->journal->j_journal, &target)) { if (wait) - jbd2_log_wait_commit(OCFS2_SB(sb)->journal->j_journal, + jbd2_log_wait_commit(osb->journal->j_journal, target); } return 0; @@ -1161,6 +1161,23 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) ocfs2_complete_mount_recovery(osb); + osb->osb_dev_kset = kset_create_and_add(sb->s_id, NULL, + &ocfs2_kset->kobj); + if (!osb->osb_dev_kset) { + status = -ENOMEM; + mlog(ML_ERROR, "Unable to create device kset %s.\n", sb->s_id); + goto read_super_error; + } + + /* Create filecheck sysfs related directories/files at + * /sys/fs/ocfs2/<devname>/filecheck */ + if (ocfs2_filecheck_create_sysfs(osb)) { + status = -ENOMEM; + mlog(ML_ERROR, "Unable to create filecheck sysfs directory at " + "/sys/fs/ocfs2/%s/filecheck.\n", sb->s_id); + goto read_super_error; + } + if (ocfs2_mount_local(osb)) snprintf(nodestr, sizeof(nodestr), "local"); else @@ -1199,9 +1216,6 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) /* Start this when the mount is almost sure of being successful */ ocfs2_orphan_scan_start(osb); - /* Create filecheck sysfile /sys/fs/ocfs2/<devname>/filecheck */ - ocfs2_filecheck_create_sysfs(sb); - return status; read_super_error: @@ -1653,7 +1667,6 @@ static void ocfs2_put_super(struct super_block *sb) ocfs2_sync_blockdev(sb); ocfs2_dismount_volume(sb, 0); - ocfs2_filecheck_remove_sysfs(sb); } static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf) @@ -1768,12 +1781,9 @@ static int ocfs2_initialize_mem_caches(void) NULL); if (!ocfs2_inode_cachep || !ocfs2_dquot_cachep || !ocfs2_qf_chunk_cachep) { - if (ocfs2_inode_cachep) - kmem_cache_destroy(ocfs2_inode_cachep); - if (ocfs2_dquot_cachep) - kmem_cache_destroy(ocfs2_dquot_cachep); - if (ocfs2_qf_chunk_cachep) - kmem_cache_destroy(ocfs2_qf_chunk_cachep); + kmem_cache_destroy(ocfs2_inode_cachep); + kmem_cache_destroy(ocfs2_dquot_cachep); + kmem_cache_destroy(ocfs2_qf_chunk_cachep); return -ENOMEM; } @@ -1787,16 +1797,13 @@ static void ocfs2_free_mem_caches(void) * destroy cache. */ rcu_barrier(); - if (ocfs2_inode_cachep) - kmem_cache_destroy(ocfs2_inode_cachep); + kmem_cache_destroy(ocfs2_inode_cachep); ocfs2_inode_cachep = NULL; - if (ocfs2_dquot_cachep) - kmem_cache_destroy(ocfs2_dquot_cachep); + kmem_cache_destroy(ocfs2_dquot_cachep); ocfs2_dquot_cachep = NULL; - if (ocfs2_qf_chunk_cachep) - kmem_cache_destroy(ocfs2_qf_chunk_cachep); + kmem_cache_destroy(ocfs2_qf_chunk_cachep); ocfs2_qf_chunk_cachep = NULL; } @@ -1899,6 +1906,12 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) osb = OCFS2_SB(sb); BUG_ON(!osb); + /* Remove file check sysfs related directores/files, + * and wait for the pending file check operations */ + ocfs2_filecheck_remove_sysfs(osb); + + kset_unregister(osb->osb_dev_kset); + debugfs_remove(osb->osb_ctxt); /* Orphan scan should be stopped as early as possible */ diff --git a/fs/ocfs2/uptodate.c b/fs/ocfs2/uptodate.c index 82e17b076ce7..78f09c76ab3c 100644 --- a/fs/ocfs2/uptodate.c +++ b/fs/ocfs2/uptodate.c @@ -633,6 +633,5 @@ int __init init_ocfs2_uptodate_cache(void) void exit_ocfs2_uptodate_cache(void) { - if (ocfs2_uptodate_cachep) - kmem_cache_destroy(ocfs2_uptodate_cachep); + kmem_cache_destroy(ocfs2_uptodate_cachep); } diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c index c261c1dfd374..3a24ce3deb01 100644 --- a/fs/ocfs2/xattr.c +++ b/fs/ocfs2/xattr.c @@ -3564,7 +3564,7 @@ int ocfs2_xattr_set(struct inode *inode, .not_found = -ENODATA, }; - if (!ocfs2_supports_xattr(OCFS2_SB(inode->i_sb))) + if (!ocfs2_supports_xattr(osb)) return -EOPNOTSUPP; /* diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c index b7146526afff..4bee3a72b9f3 100644 --- a/fs/omfs/dir.c +++ b/fs/omfs/dir.c @@ -305,11 +305,10 @@ static struct dentry *omfs_lookup(struct inode *dir, struct dentry *dentry, ino_t ino = be64_to_cpu(oi->i_head.h_self); brelse(bh); inode = omfs_iget(dir->i_sb, ino); - if (IS_ERR(inode)) - return ERR_CAST(inode); + } else if (bh != ERR_PTR(-ENOENT)) { + inode = ERR_CAST(bh); } - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } /* sanity check block's self pointer */ diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 2200662a9bf1..607092f367ad 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -256,8 +256,7 @@ found: break; } - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } static int openpromfs_readdir(struct file *file, struct dir_context *ctx) diff --git a/fs/orangefs/acl.c b/fs/orangefs/acl.c index 480ea059a680..10587413b20e 100644 --- a/fs/orangefs/acl.c +++ b/fs/orangefs/acl.c @@ -9,7 +9,6 @@ #include "orangefs-kernel.h" #include "orangefs-bufmap.h" #include <linux/posix_acl_xattr.h> -#include <linux/fs_struct.h> struct posix_acl *orangefs_get_acl(struct inode *inode, int type) { diff --git a/fs/orangefs/devorangefs-req.c b/fs/orangefs/devorangefs-req.c index b03057afac2a..66369ec90020 100644 --- a/fs/orangefs/devorangefs-req.c +++ b/fs/orangefs/devorangefs-req.c @@ -463,11 +463,10 @@ static ssize_t orangefs_devreq_write_iter(struct kiocb *iocb, if (op->downcall.type != ORANGEFS_VFS_OP_READDIR) goto wakeup; - op->downcall.trailer_buf = vmalloc(op->downcall.trailer_size); + op->downcall.trailer_buf = vzalloc(op->downcall.trailer_size); if (!op->downcall.trailer_buf) goto Enomem; - memset(op->downcall.trailer_buf, 0, op->downcall.trailer_size); if (!copy_from_iter_full(op->downcall.trailer_buf, op->downcall.trailer_size, iter)) { gossip_err("%s: failed to copy trailer.\n", __func__); @@ -779,9 +778,35 @@ static long orangefs_devreq_compat_ioctl(struct file *filp, unsigned int cmd, #endif /* CONFIG_COMPAT is in .config */ +static __poll_t orangefs_devreq_poll(struct file *file, + struct poll_table_struct *poll_table) +{ + __poll_t poll_revent_mask = 0; + + poll_wait(file, &orangefs_request_list_waitq, poll_table); + + if (!list_empty(&orangefs_request_list)) + poll_revent_mask |= EPOLLIN; + return poll_revent_mask; +} + /* the assigned character device major number */ static int orangefs_dev_major; +static const struct file_operations orangefs_devreq_file_operations = { + .owner = THIS_MODULE, + .read = orangefs_devreq_read, + .write_iter = orangefs_devreq_write_iter, + .open = orangefs_devreq_open, + .release = orangefs_devreq_release, + .unlocked_ioctl = orangefs_devreq_ioctl, + +#ifdef CONFIG_COMPAT /* CONFIG_COMPAT is in .config */ + .compat_ioctl = orangefs_devreq_compat_ioctl, +#endif + .poll = orangefs_devreq_poll +}; + /* * Initialize orangefs device specific state: * Must be called at module load time only @@ -814,29 +839,3 @@ void orangefs_dev_cleanup(void) "*** /dev/%s character device unregistered ***\n", ORANGEFS_REQDEVICE_NAME); } - -static __poll_t orangefs_devreq_poll(struct file *file, - struct poll_table_struct *poll_table) -{ - __poll_t poll_revent_mask = 0; - - poll_wait(file, &orangefs_request_list_waitq, poll_table); - - if (!list_empty(&orangefs_request_list)) - poll_revent_mask |= EPOLLIN; - return poll_revent_mask; -} - -const struct file_operations orangefs_devreq_file_operations = { - .owner = THIS_MODULE, - .read = orangefs_devreq_read, - .write_iter = orangefs_devreq_write_iter, - .open = orangefs_devreq_open, - .release = orangefs_devreq_release, - .unlocked_ioctl = orangefs_devreq_ioctl, - -#ifdef CONFIG_COMPAT /* CONFIG_COMPAT is in .config */ - .compat_ioctl = orangefs_devreq_compat_ioctl, -#endif - .poll = orangefs_devreq_poll -}; diff --git a/fs/orangefs/file.c b/fs/orangefs/file.c index 0d228cd087e6..26358efbf794 100644 --- a/fs/orangefs/file.c +++ b/fs/orangefs/file.c @@ -42,70 +42,6 @@ static int flush_racache(struct inode *inode) } /* - * Copy to client-core's address space from the buffers specified - * by the iovec upto total_size bytes. - * NOTE: the iovector can either contain addresses which - * can futher be kernel-space or user-space addresses. - * or it can pointers to struct page's - */ -static int precopy_buffers(int buffer_index, - struct iov_iter *iter, - size_t total_size) -{ - int ret = 0; - /* - * copy data from application/kernel by pulling it out - * of the iovec. - */ - - - if (total_size) { - ret = orangefs_bufmap_copy_from_iovec(iter, - buffer_index, - total_size); - if (ret < 0) - gossip_err("%s: Failed to copy-in buffers. Please make sure that the pvfs2-client is running. %ld\n", - __func__, - (long)ret); - } - - if (ret < 0) - gossip_err("%s: Failed to copy-in buffers. Please make sure that the pvfs2-client is running. %ld\n", - __func__, - (long)ret); - return ret; -} - -/* - * Copy from client-core's address space to the buffers specified - * by the iovec upto total_size bytes. - * NOTE: the iovector can either contain addresses which - * can futher be kernel-space or user-space addresses. - * or it can pointers to struct page's - */ -static int postcopy_buffers(int buffer_index, - struct iov_iter *iter, - size_t total_size) -{ - int ret = 0; - /* - * copy data to application/kernel by pushing it out to - * the iovec. NOTE; target buffers can be addresses or - * struct page pointers. - */ - if (total_size) { - ret = orangefs_bufmap_copy_to_iovec(iter, - buffer_index, - total_size); - if (ret < 0) - gossip_err("%s: Failed to copy-out buffers. Please make sure that the pvfs2-client is running (%ld)\n", - __func__, - (long)ret); - } - return ret; -} - -/* * Post and wait for the I/O upcall to finish */ static ssize_t wait_for_direct_io(enum ORANGEFS_io_type type, struct inode *inode, @@ -157,14 +93,15 @@ populate_shared_memory: total_size); /* * Stage 1: copy the buffers into client-core's address space - * precopy_buffers only pertains to writes. */ - if (type == ORANGEFS_IO_WRITE) { - ret = precopy_buffers(buffer_index, - iter, - total_size); - if (ret < 0) + if (type == ORANGEFS_IO_WRITE && total_size) { + ret = orangefs_bufmap_copy_from_iovec(iter, buffer_index, + total_size); + if (ret < 0) { + gossip_err("%s: Failed to copy-in buffers. Please make sure that the pvfs2-client is running. %ld\n", + __func__, (long)ret); goto out; + } } gossip_debug(GOSSIP_FILE_DEBUG, @@ -260,14 +197,20 @@ populate_shared_memory: /* * Stage 3: Post copy buffers from client-core's address space - * postcopy_buffers only pertains to reads. */ - if (type == ORANGEFS_IO_READ) { - ret = postcopy_buffers(buffer_index, - iter, - new_op->downcall.resp.io.amt_complete); - if (ret < 0) + if (type == ORANGEFS_IO_READ && new_op->downcall.resp.io.amt_complete) { + /* + * NOTE: the iovector can either contain addresses which + * can futher be kernel-space or user-space addresses. + * or it can pointers to struct page's + */ + ret = orangefs_bufmap_copy_to_iovec(iter, buffer_index, + new_op->downcall.resp.io.amt_complete); + if (ret < 0) { + gossip_err("%s: Failed to copy-out buffers. Please make sure that the pvfs2-client is running (%ld)\n", + __func__, (long)ret); goto out; + } } gossip_debug(GOSSIP_FILE_DEBUG, "%s(%pU): Amount %s, returned by the sys-io call:%d\n", @@ -585,6 +528,28 @@ static long orangefs_ioctl(struct file *file, unsigned int cmd, unsigned long ar return ret; } +static int orangefs_fault(struct vm_fault *vmf) +{ + struct file *file = vmf->vma->vm_file; + int rc; + rc = orangefs_inode_getattr(file->f_mapping->host, 0, 1, + STATX_SIZE); + if (rc == -ESTALE) + rc = -EIO; + if (rc) { + gossip_err("%s: orangefs_inode_getattr failed, " + "rc:%d:.\n", __func__, rc); + return rc; + } + return filemap_fault(vmf); +} + +const struct vm_operations_struct orangefs_file_vm_ops = { + .fault = orangefs_fault, + .map_pages = filemap_map_pages, + .page_mkwrite = filemap_page_mkwrite, +}; + /* * Memory map a region of a file. */ @@ -596,12 +561,16 @@ static int orangefs_file_mmap(struct file *file, struct vm_area_struct *vma) (char *)file->f_path.dentry->d_name.name : (char *)"Unknown")); + if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE)) + return -EINVAL; + /* set the sequential readahead hint */ vma->vm_flags |= VM_SEQ_READ; vma->vm_flags &= ~VM_RAND_READ; - /* Use readonly mmap since we cannot support writable maps. */ - return generic_file_readonly_mmap(file, vma); + file_accessed(file); + vma->vm_ops = &orangefs_file_vm_ops; + return 0; } #define mapping_nrpages(idata) ((idata)->nrpages) diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c index fe1d705ad91f..79c61da8b1bc 100644 --- a/fs/orangefs/inode.c +++ b/fs/orangefs/inode.c @@ -138,7 +138,7 @@ static ssize_t orangefs_direct_IO(struct kiocb *iocb, } /** ORANGEFS2 implementation of address space operations */ -const struct address_space_operations orangefs_address_operations = { +static const struct address_space_operations orangefs_address_operations = { .readpage = orangefs_readpage, .readpages = orangefs_readpages, .invalidatepage = orangefs_invalidatepage, @@ -307,7 +307,7 @@ int orangefs_update_time(struct inode *inode, struct timespec *time, int flags) } /* ORANGEDS2 implementation of VFS inode operations for files */ -const struct inode_operations orangefs_file_inode_operations = { +static const struct inode_operations orangefs_file_inode_operations = { .get_acl = orangefs_get_acl, .set_acl = orangefs_set_acl, .setattr = orangefs_setattr, diff --git a/fs/orangefs/namei.c b/fs/orangefs/namei.c index 6e3134e6d98a..365cd73d9109 100644 --- a/fs/orangefs/namei.c +++ b/fs/orangefs/namei.c @@ -75,8 +75,7 @@ static int orangefs_create(struct inode *dir, get_khandle_from_ino(inode), dentry); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; @@ -111,7 +110,6 @@ static struct dentry *orangefs_lookup(struct inode *dir, struct dentry *dentry, struct orangefs_inode_s *parent = ORANGEFS_I(dir); struct orangefs_kernel_op_s *new_op; struct inode *inode; - struct dentry *res; int ret = -EINVAL; /* @@ -159,65 +157,18 @@ static struct dentry *orangefs_lookup(struct inode *dir, struct dentry *dentry, new_op->downcall.resp.lookup.refn.fs_id, ret); - if (ret < 0) { - if (ret == -ENOENT) { - /* - * if no inode was found, add a negative dentry to - * dcache anyway; if we don't, we don't hold expected - * lookup semantics and we most noticeably break - * during directory renames. - * - * however, if the operation failed or exited, do not - * add the dentry (e.g. in the case that a touch is - * issued on a file that already exists that was - * interrupted during this lookup -- no need to add - * another negative dentry for an existing file) - */ - - gossip_debug(GOSSIP_NAME_DEBUG, - "orangefs_lookup: Adding *negative* dentry " - "%p for %pd\n", - dentry, - dentry); - - d_add(dentry, NULL); - res = NULL; - goto out; - } - + if (ret >= 0) { + orangefs_set_timeout(dentry); + inode = orangefs_iget(dir->i_sb, &new_op->downcall.resp.lookup.refn); + } else if (ret == -ENOENT) { + inode = NULL; + } else { /* must be a non-recoverable error */ - res = ERR_PTR(ret); - goto out; - } - - orangefs_set_timeout(dentry); - - inode = orangefs_iget(dir->i_sb, &new_op->downcall.resp.lookup.refn); - if (IS_ERR(inode)) { - gossip_debug(GOSSIP_NAME_DEBUG, - "error %ld from iget\n", PTR_ERR(inode)); - res = ERR_CAST(inode); - goto out; + inode = ERR_PTR(ret); } - gossip_debug(GOSSIP_NAME_DEBUG, - "%s:%s:%d " - "Found good inode [%lu] with count [%d]\n", - __FILE__, - __func__, - __LINE__, - inode->i_ino, - (int)atomic_read(&inode->i_count)); - - /* update dentry/inode pair into dcache */ - res = d_splice_alias(inode, dentry); - - gossip_debug(GOSSIP_NAME_DEBUG, - "Lookup success (inode ct = %d)\n", - (int)atomic_read(&inode->i_count)); -out: op_release(new_op); - return res; + return d_splice_alias(inode, dentry); } /* return 0 on success; non-zero otherwise */ @@ -332,8 +283,7 @@ static int orangefs_symlink(struct inode *dir, "Assigned symlink inode new number of %pU\n", get_khandle_from_ino(inode)); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; @@ -402,8 +352,7 @@ static int orangefs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode "Assigned dir inode new number of %pU\n", get_khandle_from_ino(inode)); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; diff --git a/fs/orangefs/orangefs-bufmap.c b/fs/orangefs/orangefs-bufmap.c index 59f444dced9b..4f927023d095 100644 --- a/fs/orangefs/orangefs-bufmap.c +++ b/fs/orangefs/orangefs-bufmap.c @@ -71,9 +71,9 @@ static void put(struct slot_map *m, int slot) spin_lock(&m->q.lock); __clear_bit(slot, m->map); v = ++m->c; - if (unlikely(v == 1)) /* no free slots -> one free slot */ + if (v > 0) wake_up_locked(&m->q); - else if (unlikely(v == -1)) /* finished dying */ + if (unlikely(v == -1)) /* finished dying */ wake_up_all_locked(&m->q); spin_unlock(&m->q.lock); } diff --git a/fs/orangefs/orangefs-debug.h b/fs/orangefs/orangefs-debug.h index c7db56a31b92..6e079d4230d0 100644 --- a/fs/orangefs/orangefs-debug.h +++ b/fs/orangefs/orangefs-debug.h @@ -43,12 +43,6 @@ #define GOSSIP_MAX_NR 16 #define GOSSIP_MAX_DEBUG (((__u64)1 << GOSSIP_MAX_NR) - 1) -/*function prototypes*/ -__u64 ORANGEFS_kmod_eventlog_to_mask(const char *event_logging); -__u64 ORANGEFS_debug_eventlog_to_mask(const char *event_logging); -char *ORANGEFS_debug_mask_to_eventlog(__u64 mask); -char *ORANGEFS_kmod_mask_to_eventlog(__u64 mask); - /* a private internal type */ struct __keyword_mask_s { const char *keyword; diff --git a/fs/orangefs/orangefs-kernel.h b/fs/orangefs/orangefs-kernel.h index eebbaece85ef..c29bb0ebc6bb 100644 --- a/fs/orangefs/orangefs-kernel.h +++ b/fs/orangefs/orangefs-kernel.h @@ -65,11 +65,7 @@ #define ORANGEFS_REQDEVICE_NAME "pvfs2-req" #define ORANGEFS_DEVREQ_MAGIC 0x20030529 -#define ORANGEFS_LINK_MAX 0x000000FF #define ORANGEFS_PURGE_RETRY_COUNT 0x00000005 -#define ORANGEFS_MAX_NUM_OPTIONS 0x00000004 -#define ORANGEFS_MAX_MOUNT_OPT_LEN 0x00000080 -#define ORANGEFS_MAX_FSKEY_LEN 64 #define MAX_DEV_REQ_UPSIZE (2 * sizeof(__s32) + \ sizeof(__u64) + sizeof(struct orangefs_upcall_s)) @@ -113,15 +109,6 @@ extern struct posix_acl *orangefs_get_acl(struct inode *inode, int type); extern int orangefs_set_acl(struct inode *inode, struct posix_acl *acl, int type); /* - * Redefine xtvec structure so that we could move helper functions out of - * the define - */ -struct xtvec { - __kernel_off_t xtv_off; /* must be off_t */ - __kernel_size_t xtv_len; /* must be size_t */ -}; - -/* * orangefs data structures */ struct orangefs_kernel_op_s { @@ -224,39 +211,6 @@ struct orangefs_sb_info_s { struct list_head list; }; -/* - * structure that holds the state of any async I/O operation issued - * through the VFS. Needed especially to handle cancellation requests - * or even completion notification so that the VFS client-side daemon - * can free up its vfs_request slots. - */ -struct orangefs_kiocb_s { - /* the pointer to the task that initiated the AIO */ - struct task_struct *tsk; - - /* pointer to the kiocb that kicked this operation */ - struct kiocb *kiocb; - - /* buffer index that was used for the I/O */ - struct orangefs_bufmap *bufmap; - int buffer_index; - - /* orangefs kernel operation type */ - struct orangefs_kernel_op_s *op; - - /* set to indicate the type of the operation */ - int rw; - - /* file offset */ - loff_t offset; - - /* and the count in bytes */ - size_t bytes_to_be_copied; - - ssize_t bytes_copied; - int needs_cleanup; -}; - struct orangefs_stats { unsigned long cache_hits; unsigned long cache_misses; @@ -305,21 +259,6 @@ static inline struct orangefs_khandle *get_khandle_from_ino(struct inode *inode) return &(ORANGEFS_I(inode)->refn.khandle); } -static inline ino_t get_ino_from_khandle(struct inode *inode) -{ - struct orangefs_khandle *khandle; - ino_t ino; - - khandle = get_khandle_from_ino(inode); - ino = orangefs_khandle_to_ino(khandle); - return ino; -} - -static inline ino_t get_parent_ino_from_dentry(struct dentry *dentry) -{ - return get_ino_from_khandle(dentry->d_parent->d_inode); -} - static inline int is_root_handle(struct inode *inode) { gossip_debug(GOSSIP_DCACHE_DEBUG, @@ -391,7 +330,6 @@ void fsid_key_table_finalize(void); /* * defined in inode.c */ -__u32 convert_to_orangefs_mask(unsigned long lite_mask); struct inode *orangefs_new_inode(struct super_block *sb, struct inode *dir, int mode, @@ -410,17 +348,6 @@ int orangefs_update_time(struct inode *, struct timespec *, int); /* * defined in xattr.c */ -int orangefs_setxattr(struct dentry *dentry, - const char *name, - const void *value, - size_t size, - int flags); - -ssize_t orangefs_getxattr(struct dentry *dentry, - const char *name, - void *buffer, - size_t size); - ssize_t orangefs_listxattr(struct dentry *dentry, char *buffer, size_t size); /* @@ -467,8 +394,6 @@ int orangefs_inode_check_changed(struct inode *inode); int orangefs_inode_setattr(struct inode *inode, struct iattr *iattr); -int orangefs_unmount_sb(struct super_block *sb); - bool orangefs_cancel_op_in_progress(struct orangefs_kernel_op_s *op); int orangefs_normalize_to_errno(__s32 error_code); @@ -487,16 +412,11 @@ extern struct list_head *orangefs_htable_ops_in_progress; extern spinlock_t orangefs_htable_ops_in_progress_lock; extern int hash_table_size; -extern const struct address_space_operations orangefs_address_operations; -extern const struct inode_operations orangefs_file_inode_operations; extern const struct file_operations orangefs_file_operations; extern const struct inode_operations orangefs_symlink_inode_operations; extern const struct inode_operations orangefs_dir_inode_operations; extern const struct file_operations orangefs_dir_operations; extern const struct dentry_operations orangefs_dentry_operations; -extern const struct file_operations orangefs_devreq_file_operations; - -extern wait_queue_head_t orangefs_bufmap_init_waitq; /* * misc convenience macros diff --git a/fs/orangefs/protocol.h b/fs/orangefs/protocol.h index dc6e3e6269c3..61ee8d64c842 100644 --- a/fs/orangefs/protocol.h +++ b/fs/orangefs/protocol.h @@ -5,11 +5,6 @@ #include <linux/slab.h> #include <linux/ioctl.h> -/* pvfs2-config.h ***********************************************************/ -#define ORANGEFS_VERSION_MAJOR 2 -#define ORANGEFS_VERSION_MINOR 9 -#define ORANGEFS_VERSION_SUB 0 - /* khandle stuff ***********************************************************/ /* @@ -70,16 +65,6 @@ static inline void ORANGEFS_khandle_from(struct orangefs_khandle *kh, } /* pvfs2-types.h ************************************************************/ -typedef __u32 ORANGEFS_uid; -typedef __u32 ORANGEFS_gid; -typedef __s32 ORANGEFS_fs_id; -typedef __u32 ORANGEFS_permissions; -typedef __u64 ORANGEFS_time; -typedef __s64 ORANGEFS_size; -typedef __u64 ORANGEFS_flags; -typedef __u64 ORANGEFS_ds_position; -typedef __s32 ORANGEFS_error; -typedef __s64 ORANGEFS_offset; #define ORANGEFS_SUPER_MAGIC 0x20030528 @@ -145,7 +130,6 @@ typedef __s64 ORANGEFS_offset; #define ORANGEFS_APPEND_FL FS_APPEND_FL #define ORANGEFS_NOATIME_FL FS_NOATIME_FL #define ORANGEFS_MIRROR_FL 0x01000000ULL -#define ORANGEFS_O_EXECUTE (1 << 0) #define ORANGEFS_FS_ID_NULL ((__s32)0) #define ORANGEFS_ATTR_SYS_UID (1 << 0) #define ORANGEFS_ATTR_SYS_GID (1 << 1) @@ -229,35 +213,6 @@ enum orangefs_ds_type { ORANGEFS_TYPE_INTERNAL = (1 << 5) /* for the server's private use */ }; -/* - * ORANGEFS_certificate simply stores a buffer with the buffer size. - * The buffer can be converted to an OpenSSL X509 struct for use. - */ -struct ORANGEFS_certificate { - __u32 buf_size; - unsigned char *buf; -}; - -/* - * A credential identifies a user and is signed by the client/user - * private key. - */ -struct ORANGEFS_credential { - __u32 userid; /* user id */ - __u32 num_groups; /* length of group_array */ - __u32 *group_array; /* groups for which the user is a member */ - char *issuer; /* alias of the issuing server */ - __u64 timeout; /* seconds after epoch to time out */ - __u32 sig_size; /* length of the signature in bytes */ - unsigned char *signature; /* digital signature */ - struct ORANGEFS_certificate certificate; /* user certificate buffer */ -}; -#define extra_size_ORANGEFS_credential (ORANGEFS_REQ_LIMIT_GROUPS * \ - sizeof(__u32) + \ - ORANGEFS_REQ_LIMIT_ISSUER + \ - ORANGEFS_REQ_LIMIT_SIGNATURE + \ - extra_size_ORANGEFS_certificate) - /* This structure is used by the VFS-client interaction alone */ struct ORANGEFS_keyval_pair { char key[ORANGEFS_MAX_XATTR_NAMELEN]; diff --git a/fs/orangefs/super.c b/fs/orangefs/super.c index 3ae5fdba0225..10796d3fe27d 100644 --- a/fs/orangefs/super.c +++ b/fs/orangefs/super.c @@ -579,6 +579,11 @@ void orangefs_kill_sb(struct super_block *sb) /* provided sb cleanup */ kill_anon_super(sb); + if (!ORANGEFS_SB(sb)) { + mutex_lock(&orangefs_request_mutex); + mutex_unlock(&orangefs_request_mutex); + return; + } /* * issue the unmount to userspace to tell it to remove the * dynamic mount info it has for this superblock diff --git a/fs/overlayfs/Kconfig b/fs/overlayfs/Kconfig index ce6ff5a0a6e4..17032631c5cf 100644 --- a/fs/overlayfs/Kconfig +++ b/fs/overlayfs/Kconfig @@ -86,3 +86,20 @@ config OVERLAY_FS_NFS_EXPORT case basis with the "nfs_export=on" mount option. Say N unless you fully understand the consequences. + +config OVERLAY_FS_XINO_AUTO + bool "Overlayfs: auto enable inode number mapping" + default n + depends on OVERLAY_FS + help + If this config option is enabled then overlay filesystems will use + unused high bits in undelying filesystem inode numbers to map all + inodes to a unified address space. The mapped 64bit inode numbers + might not be compatible with applications that expect 32bit inodes. + + If compatibility with applications that expect 32bit inodes is not an + issue, then it is safe and recommended to say Y here. + + For more information, see Documentation/filesystems/overlayfs.txt + + If unsure, say N. diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c index d855f508fa20..8bede0742619 100644 --- a/fs/overlayfs/copy_up.c +++ b/fs/overlayfs/copy_up.c @@ -232,7 +232,7 @@ int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat) return err; } -struct ovl_fh *ovl_encode_fh(struct dentry *real, bool is_upper) +struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper) { struct ovl_fh *fh; int fh_type, fh_len, dwords; @@ -300,7 +300,7 @@ int ovl_set_origin(struct dentry *dentry, struct dentry *lower, * up and a pure upper inode. */ if (ovl_can_decode_fh(lower->d_sb)) { - fh = ovl_encode_fh(lower, false); + fh = ovl_encode_real_fh(lower, false); if (IS_ERR(fh)) return PTR_ERR(fh); } @@ -321,7 +321,7 @@ static int ovl_set_upper_fh(struct dentry *upper, struct dentry *index) const struct ovl_fh *fh; int err; - fh = ovl_encode_fh(upper, true); + fh = ovl_encode_real_fh(upper, true); if (IS_ERR(fh)) return PTR_ERR(fh); diff --git a/fs/overlayfs/export.c b/fs/overlayfs/export.c index 87bd4148f4fb..425a94672300 100644 --- a/fs/overlayfs/export.c +++ b/fs/overlayfs/export.c @@ -228,8 +228,8 @@ static int ovl_d_to_fh(struct dentry *dentry, char *buf, int buflen) goto fail; /* Encode an upper or lower file handle */ - fh = ovl_encode_fh(enc_lower ? ovl_dentry_lower(dentry) : - ovl_dentry_upper(dentry), !enc_lower); + fh = ovl_encode_real_fh(enc_lower ? ovl_dentry_lower(dentry) : + ovl_dentry_upper(dentry), !enc_lower); err = PTR_ERR(fh); if (IS_ERR(fh)) goto fail; @@ -267,8 +267,8 @@ static int ovl_dentry_to_fh(struct dentry *dentry, u32 *fid, int *max_len) return OVL_FILEID; } -static int ovl_encode_inode_fh(struct inode *inode, u32 *fid, int *max_len, - struct inode *parent) +static int ovl_encode_fh(struct inode *inode, u32 *fid, int *max_len, + struct inode *parent) { struct dentry *dentry; int type; @@ -305,15 +305,12 @@ static struct dentry *ovl_obtain_alias(struct super_block *sb, if (d_is_dir(upper ?: lower)) return ERR_PTR(-EIO); - inode = ovl_get_inode(sb, dget(upper), lower, index, !!lower); + inode = ovl_get_inode(sb, dget(upper), lowerpath, index, !!lower); if (IS_ERR(inode)) { dput(upper); return ERR_CAST(inode); } - if (index) - ovl_set_flag(OVL_INDEX, inode); - dentry = d_find_any_alias(inode); if (!dentry) { dentry = d_alloc_anon(inode->i_sb); @@ -685,7 +682,7 @@ static struct dentry *ovl_upper_fh_to_d(struct super_block *sb, if (!ofs->upper_mnt) return ERR_PTR(-EACCES); - upper = ovl_decode_fh(fh, ofs->upper_mnt); + upper = ovl_decode_real_fh(fh, ofs->upper_mnt, true); if (IS_ERR_OR_NULL(upper)) return upper; @@ -703,25 +700,39 @@ static struct dentry *ovl_lower_fh_to_d(struct super_block *sb, struct ovl_path *stack = &origin; struct dentry *dentry = NULL; struct dentry *index = NULL; - struct inode *inode = NULL; - bool is_deleted = false; + struct inode *inode; int err; - /* First lookup indexed upper by fh */ + /* First lookup overlay inode in inode cache by origin fh */ + err = ovl_check_origin_fh(ofs, fh, false, NULL, &stack); + if (err) + return ERR_PTR(err); + + if (!d_is_dir(origin.dentry) || + !(origin.dentry->d_flags & DCACHE_DISCONNECTED)) { + inode = ovl_lookup_inode(sb, origin.dentry, false); + err = PTR_ERR(inode); + if (IS_ERR(inode)) + goto out_err; + if (inode) { + dentry = d_find_any_alias(inode); + iput(inode); + if (dentry) + goto out; + } + } + + /* Then lookup indexed upper/whiteout by origin fh */ if (ofs->indexdir) { index = ovl_get_index_fh(ofs, fh); err = PTR_ERR(index); if (IS_ERR(index)) { - if (err != -ESTALE) - return ERR_PTR(err); - - /* Found a whiteout index - treat as deleted inode */ - is_deleted = true; index = NULL; + goto out_err; } } - /* Then try to get upper dir by index */ + /* Then try to get a connected upper dir by index */ if (index && d_is_dir(index)) { struct dentry *upper = ovl_index_upper(ofs, index); @@ -734,24 +745,19 @@ static struct dentry *ovl_lower_fh_to_d(struct super_block *sb, goto out; } - /* Then lookup origin by fh */ - err = ovl_check_origin_fh(ofs, fh, NULL, &stack); - if (err) { - goto out_err; - } else if (index) { - err = ovl_verify_origin(index, origin.dentry, false); + /* Otherwise, get a connected non-upper dir or disconnected non-dir */ + if (d_is_dir(origin.dentry) && + (origin.dentry->d_flags & DCACHE_DISCONNECTED)) { + dput(origin.dentry); + origin.dentry = NULL; + err = ovl_check_origin_fh(ofs, fh, true, NULL, &stack); if (err) goto out_err; - } else if (is_deleted) { - /* Lookup deleted non-dir by origin inode */ - if (!d_is_dir(origin.dentry)) - inode = ovl_lookup_inode(sb, origin.dentry, false); - err = -ESTALE; - if (!inode || atomic_read(&inode->i_count) == 1) + } + if (index) { + err = ovl_verify_origin(index, origin.dentry, false); + if (err) goto out_err; - - /* Deleted but still open? */ - index = dget(ovl_i_dentry_upper(inode)); } dentry = ovl_get_dentry(sb, NULL, &origin, index); @@ -759,7 +765,6 @@ static struct dentry *ovl_lower_fh_to_d(struct super_block *sb, out: dput(origin.dentry); dput(index); - iput(inode); return dentry; out_err: @@ -829,7 +834,7 @@ static struct dentry *ovl_get_parent(struct dentry *dentry) } const struct export_operations ovl_export_operations = { - .encode_fh = ovl_encode_inode_fh, + .encode_fh = ovl_encode_fh, .fh_to_dentry = ovl_fh_to_dentry, .fh_to_parent = ovl_fh_to_parent, .get_name = ovl_get_name, diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c index 3b1bd469accd..6e3815fb006b 100644 --- a/fs/overlayfs/inode.c +++ b/fs/overlayfs/inode.c @@ -16,13 +16,6 @@ #include "overlayfs.h" -static dev_t ovl_get_pseudo_dev(struct dentry *dentry) -{ - struct ovl_entry *oe = dentry->d_fsdata; - - return oe->lowerstack[0].layer->pseudo_dev; -} - int ovl_setattr(struct dentry *dentry, struct iattr *attr) { int err; @@ -66,6 +59,69 @@ out: return err; } +static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, + struct ovl_layer *lower_layer) +{ + bool samefs = ovl_same_sb(dentry->d_sb); + unsigned int xinobits = ovl_xino_bits(dentry->d_sb); + + if (samefs) { + /* + * When all layers are on the same fs, all real inode + * number are unique, so we use the overlay st_dev, + * which is friendly to du -x. + */ + stat->dev = dentry->d_sb->s_dev; + return 0; + } else if (xinobits) { + unsigned int shift = 64 - xinobits; + /* + * All inode numbers of underlying fs should not be using the + * high xinobits, so we use high xinobits to partition the + * overlay st_ino address space. The high bits holds the fsid + * (upper fsid is 0). This way overlay inode numbers are unique + * and all inodes use overlay st_dev. Inode numbers are also + * persistent for a given layer configuration. + */ + if (stat->ino >> shift) { + pr_warn_ratelimited("overlayfs: inode number too big (%pd2, ino=%llu, xinobits=%d)\n", + dentry, stat->ino, xinobits); + } else { + if (lower_layer) + stat->ino |= ((u64)lower_layer->fsid) << shift; + + stat->dev = dentry->d_sb->s_dev; + return 0; + } + } + + /* The inode could not be mapped to a unified st_ino address space */ + if (S_ISDIR(dentry->d_inode->i_mode)) { + /* + * Always use the overlay st_dev for directories, so 'find + * -xdev' will scan the entire overlay mount and won't cross the + * overlay mount boundaries. + * + * If not all layers are on the same fs the pair {real st_ino; + * overlay st_dev} is not unique, so use the non persistent + * overlay st_ino for directories. + */ + stat->dev = dentry->d_sb->s_dev; + stat->ino = dentry->d_inode->i_ino; + } else if (lower_layer && lower_layer->fsid) { + /* + * For non-samefs setup, if we cannot map all layers st_ino + * to a unified address space, we need to make sure that st_dev + * is unique per lower fs. Upper layer uses real st_dev and + * lower layers use the unique anonymous bdev assigned to the + * lower fs. + */ + stat->dev = lower_layer->fs->pseudo_dev; + } + + return 0; +} + int ovl_getattr(const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { @@ -75,6 +131,7 @@ int ovl_getattr(const struct path *path, struct kstat *stat, const struct cred *old_cred; bool is_dir = S_ISDIR(dentry->d_inode->i_mode); bool samefs = ovl_same_sb(dentry->d_sb); + struct ovl_layer *lower_layer = NULL; int err; type = ovl_path_real(dentry, &realpath); @@ -84,14 +141,18 @@ int ovl_getattr(const struct path *path, struct kstat *stat, goto out; /* - * For non-dir or same fs, we use st_ino of the copy up origin, if we - * know it. This guaranties constant st_dev/st_ino across copy up. + * For non-dir or same fs, we use st_ino of the copy up origin. + * This guaranties constant st_dev/st_ino across copy up. + * With xino feature and non-samefs, we use st_ino of the copy up + * origin masked with high bits that represent the layer id. * - * If filesystem supports NFS export ops, this also guaranties + * If lower filesystem supports NFS file handles, this also guaranties * persistent st_ino across mount cycle. */ - if (!is_dir || samefs) { - if (OVL_TYPE_ORIGIN(type)) { + if (!is_dir || samefs || ovl_xino_bits(dentry->d_sb)) { + if (!OVL_TYPE_UPPER(type)) { + lower_layer = ovl_layer_lower(dentry); + } else if (OVL_TYPE_ORIGIN(type)) { struct kstat lowerstat; u32 lowermask = STATX_INO | (!is_dir ? STATX_NLINK : 0); @@ -118,43 +179,17 @@ int ovl_getattr(const struct path *path, struct kstat *stat, */ if (ovl_test_flag(OVL_INDEX, d_inode(dentry)) || (!ovl_verify_lower(dentry->d_sb) && - (is_dir || lowerstat.nlink == 1))) + (is_dir || lowerstat.nlink == 1))) { stat->ino = lowerstat.ino; - - if (samefs) - WARN_ON_ONCE(stat->dev != lowerstat.dev); - else - stat->dev = ovl_get_pseudo_dev(dentry); - } - if (samefs) { - /* - * When all layers are on the same fs, all real inode - * number are unique, so we use the overlay st_dev, - * which is friendly to du -x. - */ - stat->dev = dentry->d_sb->s_dev; - } else if (!OVL_TYPE_UPPER(type)) { - /* - * For non-samefs setup, to make sure that st_dev/st_ino - * pair is unique across the system, we use a unique - * anonymous st_dev for lower layer inode. - */ - stat->dev = ovl_get_pseudo_dev(dentry); + lower_layer = ovl_layer_lower(dentry); + } } - } else { - /* - * Always use the overlay st_dev for directories, so 'find - * -xdev' will scan the entire overlay mount and won't cross the - * overlay mount boundaries. - * - * If not all layers are on the same fs the pair {real st_ino; - * overlay st_dev} is not unique, so use the non persistent - * overlay st_ino for directories. - */ - stat->dev = dentry->d_sb->s_dev; - stat->ino = dentry->d_inode->i_ino; } + err = ovl_map_dev_ino(dentry, stat, lower_layer); + if (err) + goto out; + /* * It's probably not worth it to count subdirs to get the * correct link count. nlink=1 seems to pacify 'find' and @@ -383,24 +418,18 @@ int ovl_open_maybe_copy_up(struct dentry *dentry, unsigned int file_flags) int ovl_update_time(struct inode *inode, struct timespec *ts, int flags) { - struct dentry *alias; - struct path upperpath; - - if (!(flags & S_ATIME)) - return 0; - - alias = d_find_any_alias(inode); - if (!alias) - return 0; - - ovl_path_upper(alias, &upperpath); - if (upperpath.dentry) { - touch_atime(&upperpath); - inode->i_atime = d_inode(upperpath.dentry)->i_atime; + if (flags & S_ATIME) { + struct ovl_fs *ofs = inode->i_sb->s_fs_info; + struct path upperpath = { + .mnt = ofs->upper_mnt, + .dentry = ovl_upperdentry_dereference(OVL_I(inode)), + }; + + if (upperpath.dentry) { + touch_atime(&upperpath); + inode->i_atime = d_inode(upperpath.dentry)->i_atime; + } } - - dput(alias); - return 0; } @@ -459,9 +488,27 @@ static inline void ovl_lockdep_annotate_inode_mutex_key(struct inode *inode) #endif } -static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev) +static void ovl_fill_inode(struct inode *inode, umode_t mode, dev_t rdev, + unsigned long ino, int fsid) { - inode->i_ino = get_next_ino(); + int xinobits = ovl_xino_bits(inode->i_sb); + + /* + * When NFS export is enabled and d_ino is consistent with st_ino + * (samefs or i_ino has enough bits to encode layer), set the same + * value used for d_ino to i_ino, because nfsd readdirplus compares + * d_ino values to i_ino values of child entries. When called from + * ovl_new_inode(), ino arg is 0, so i_ino will be updated to real + * upper inode i_ino on ovl_inode_init() or ovl_inode_update(). + */ + if (inode->i_sb->s_export_op && + (ovl_same_sb(inode->i_sb) || xinobits)) { + inode->i_ino = ino; + if (xinobits && fsid && !(ino >> (64 - xinobits))) + inode->i_ino |= (unsigned long)fsid << (64 - xinobits); + } else { + inode->i_ino = get_next_ino(); + } inode->i_mode = mode; inode->i_flags |= S_NOCMTIME; #ifdef CONFIG_FS_POSIX_ACL @@ -597,7 +644,7 @@ struct inode *ovl_new_inode(struct super_block *sb, umode_t mode, dev_t rdev) inode = new_inode(sb); if (inode) - ovl_fill_inode(inode, mode, rdev); + ovl_fill_inode(inode, mode, rdev, 0, 0); return inode; } @@ -703,13 +750,16 @@ static bool ovl_hash_bylower(struct super_block *sb, struct dentry *upper, } struct inode *ovl_get_inode(struct super_block *sb, struct dentry *upperdentry, - struct dentry *lowerdentry, struct dentry *index, + struct ovl_path *lowerpath, struct dentry *index, unsigned int numlower) { struct inode *realinode = upperdentry ? d_inode(upperdentry) : NULL; struct inode *inode; + struct dentry *lowerdentry = lowerpath ? lowerpath->dentry : NULL; bool bylower = ovl_hash_bylower(sb, upperdentry, lowerdentry, index); + int fsid = bylower ? lowerpath->layer->fsid : 0; bool is_dir; + unsigned long ino = 0; if (!realinode) realinode = d_inode(lowerdentry); @@ -748,18 +798,22 @@ struct inode *ovl_get_inode(struct super_block *sb, struct dentry *upperdentry, if (!is_dir) nlink = ovl_get_nlink(lowerdentry, upperdentry, nlink); set_nlink(inode, nlink); + ino = key->i_ino; } else { /* Lower hardlink that will be broken on copy up */ inode = new_inode(sb); if (!inode) goto out_nomem; } - ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev); + ovl_fill_inode(inode, realinode->i_mode, realinode->i_rdev, ino, fsid); ovl_inode_init(inode, upperdentry, lowerdentry); if (upperdentry && ovl_is_impuredir(upperdentry)) ovl_set_flag(OVL_IMPURE, inode); + if (index) + ovl_set_flag(OVL_INDEX, inode); + /* Check for non-merge dir that may have whiteouts */ if (is_dir) { if (((upperdentry && lowerdentry) || numlower > 1) || diff --git a/fs/overlayfs/namei.c b/fs/overlayfs/namei.c index 70fcfcc684cc..2dba29eadde6 100644 --- a/fs/overlayfs/namei.c +++ b/fs/overlayfs/namei.c @@ -56,6 +56,15 @@ static int ovl_check_redirect(struct dentry *dentry, struct ovl_lookup_data *d, if (s == next) goto invalid; } + /* + * One of the ancestor path elements in an absolute path + * lookup in ovl_lookup_layer() could have been opaque and + * that will stop further lookup in lower layers (d->stop=true) + * But we have found an absolute redirect in decendant path + * element and that should force continue lookup in lower + * layers (reset d->stop). + */ + d->stop = false; } else { if (strchr(buf, '/') != NULL) goto invalid; @@ -171,7 +180,8 @@ invalid: goto out; } -struct dentry *ovl_decode_fh(struct ovl_fh *fh, struct vfsmount *mnt) +struct dentry *ovl_decode_real_fh(struct ovl_fh *fh, struct vfsmount *mnt, + bool connected) { struct dentry *real; int bytes; @@ -186,7 +196,7 @@ struct dentry *ovl_decode_fh(struct ovl_fh *fh, struct vfsmount *mnt) bytes = (fh->len - offsetof(struct ovl_fh, fid)); real = exportfs_decode_fh(mnt, (struct fid *)fh->fid, bytes >> 2, (int)fh->type, - ovl_acceptable, mnt); + connected ? ovl_acceptable : NULL, mnt); if (IS_ERR(real)) { /* * Treat stale file handle to lower file as "origin unknown". @@ -220,6 +230,7 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d, { struct dentry *this; int err; + bool last_element = !post[0]; this = lookup_one_len_unlocked(name, base, namelen); if (IS_ERR(this)) { @@ -245,11 +256,23 @@ static int ovl_lookup_single(struct dentry *base, struct ovl_lookup_data *d, d->stop = true; if (d->is_dir) goto put_and_out; + + /* + * NB: handle failure to lookup non-last element when non-dir + * redirects become possible + */ + WARN_ON(!last_element); goto out; } - d->is_dir = true; - if (!d->last && ovl_is_opaquedir(this)) { - d->stop = d->opaque = true; + if (last_element) + d->is_dir = true; + if (d->last) + goto out; + + if (ovl_is_opaquedir(this)) { + d->stop = true; + if (last_element) + d->opaque = true; goto out; } err = ovl_check_redirect(this, d, prelen, post); @@ -310,14 +333,15 @@ static int ovl_lookup_layer(struct dentry *base, struct ovl_lookup_data *d, } -int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, +int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, bool connected, struct dentry *upperdentry, struct ovl_path **stackp) { struct dentry *origin = NULL; int i; for (i = 0; i < ofs->numlower; i++) { - origin = ovl_decode_fh(fh, ofs->lower_layers[i].mnt); + origin = ovl_decode_real_fh(fh, ofs->lower_layers[i].mnt, + connected); if (origin) break; } @@ -361,7 +385,7 @@ static int ovl_check_origin(struct ovl_fs *ofs, struct dentry *upperdentry, if (IS_ERR_OR_NULL(fh)) return PTR_ERR(fh); - err = ovl_check_origin_fh(ofs, fh, upperdentry, stackp); + err = ovl_check_origin_fh(ofs, fh, false, upperdentry, stackp); kfree(fh); if (err) { @@ -415,7 +439,7 @@ int ovl_verify_set_fh(struct dentry *dentry, const char *name, struct ovl_fh *fh; int err; - fh = ovl_encode_fh(real, is_upper); + fh = ovl_encode_real_fh(real, is_upper); err = PTR_ERR(fh); if (IS_ERR(fh)) goto fail; @@ -451,7 +475,7 @@ struct dentry *ovl_index_upper(struct ovl_fs *ofs, struct dentry *index) if (IS_ERR_OR_NULL(fh)) return ERR_CAST(fh); - upper = ovl_decode_fh(fh, ofs->upper_mnt); + upper = ovl_decode_real_fh(fh, ofs->upper_mnt, true); kfree(fh); if (IS_ERR_OR_NULL(upper)) @@ -558,7 +582,7 @@ int ovl_verify_index(struct ovl_fs *ofs, struct dentry *index) /* Check if non-dir index is orphan and don't warn before cleaning it */ if (!d_is_dir(index) && d_inode(index)->i_nlink == 1) { - err = ovl_check_origin_fh(ofs, fh, index, &stack); + err = ovl_check_origin_fh(ofs, fh, false, index, &stack); if (err) goto fail; @@ -619,7 +643,7 @@ int ovl_get_index_name(struct dentry *origin, struct qstr *name) struct ovl_fh *fh; int err; - fh = ovl_encode_fh(origin, false); + fh = ovl_encode_real_fh(origin, false); if (IS_ERR(fh)) return PTR_ERR(fh); @@ -815,7 +839,7 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry, .is_dir = false, .opaque = false, .stop = false, - .last = !poe->numlower, + .last = ofs->config.redirect_follow ? false : !poe->numlower, .redirect = NULL, }; @@ -873,7 +897,11 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry, for (i = 0; !d.stop && i < poe->numlower; i++) { struct ovl_path lower = poe->lowerstack[i]; - d.last = i == poe->numlower - 1; + if (!ofs->config.redirect_follow) + d.last = i == poe->numlower - 1; + else + d.last = lower.layer->idx == roe->numlower; + err = ovl_lookup_layer(lower.dentry, &d, &this); if (err) goto out_put; @@ -976,17 +1004,18 @@ struct dentry *ovl_lookup(struct inode *dir, struct dentry *dentry, upperdentry = dget(index); if (upperdentry || ctr) { - if (ctr) - origin = stack[0].dentry; - inode = ovl_get_inode(dentry->d_sb, upperdentry, origin, index, + inode = ovl_get_inode(dentry->d_sb, upperdentry, stack, index, ctr); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_free_oe; + /* + * NB: handle redirected hard links when non-dir redirects + * become possible + */ + WARN_ON(OVL_I(inode)->redirect); OVL_I(inode)->redirect = upperredirect; - if (index) - ovl_set_flag(OVL_INDEX, inode); } revert_creds(old_cred); diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h index 225ff1171147..e0b7de799f6b 100644 --- a/fs/overlayfs/overlayfs.h +++ b/fs/overlayfs/overlayfs.h @@ -202,7 +202,7 @@ void ovl_drop_write(struct dentry *dentry); struct dentry *ovl_workdir(struct dentry *dentry); const struct cred *ovl_override_creds(struct super_block *sb); struct super_block *ovl_same_sb(struct super_block *sb); -bool ovl_can_decode_fh(struct super_block *sb); +int ovl_can_decode_fh(struct super_block *sb); struct dentry *ovl_indexdir(struct super_block *sb); bool ovl_index_all(struct super_block *sb); bool ovl_verify_lower(struct super_block *sb); @@ -215,6 +215,7 @@ void ovl_path_lower(struct dentry *dentry, struct path *path); enum ovl_path_type ovl_path_real(struct dentry *dentry, struct path *path); struct dentry *ovl_dentry_upper(struct dentry *dentry); struct dentry *ovl_dentry_lower(struct dentry *dentry); +struct ovl_layer *ovl_layer_lower(struct dentry *dentry); struct dentry *ovl_dentry_real(struct dentry *dentry); struct dentry *ovl_i_dentry_upper(struct inode *inode); struct inode *ovl_inode_upper(struct inode *inode); @@ -263,11 +264,19 @@ static inline bool ovl_is_impuredir(struct dentry *dentry) return ovl_check_dir_xattr(dentry, OVL_XATTR_IMPURE); } +static inline unsigned int ovl_xino_bits(struct super_block *sb) +{ + struct ovl_fs *ofs = sb->s_fs_info; + + return ofs->xino_bits; +} + /* namei.c */ int ovl_check_fh_len(struct ovl_fh *fh, int fh_len); -struct dentry *ovl_decode_fh(struct ovl_fh *fh, struct vfsmount *mnt); -int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, +struct dentry *ovl_decode_real_fh(struct ovl_fh *fh, struct vfsmount *mnt, + bool connected); +int ovl_check_origin_fh(struct ovl_fs *ofs, struct ovl_fh *fh, bool connected, struct dentry *upperdentry, struct ovl_path **stackp); int ovl_verify_set_fh(struct dentry *dentry, const char *name, struct dentry *real, bool is_upper, bool set); @@ -329,7 +338,7 @@ struct inode *ovl_new_inode(struct super_block *sb, umode_t mode, dev_t rdev); struct inode *ovl_lookup_inode(struct super_block *sb, struct dentry *real, bool is_upper); struct inode *ovl_get_inode(struct super_block *sb, struct dentry *upperdentry, - struct dentry *lowerdentry, struct dentry *index, + struct ovl_path *lowerpath, struct dentry *index, unsigned int numlower); static inline void ovl_copyattr(struct inode *from, struct inode *to) { @@ -361,7 +370,7 @@ int ovl_copy_up(struct dentry *dentry); int ovl_copy_up_flags(struct dentry *dentry, int flags); int ovl_copy_xattr(struct dentry *old, struct dentry *new); int ovl_set_attr(struct dentry *upper, struct kstat *stat); -struct ovl_fh *ovl_encode_fh(struct dentry *real, bool is_upper); +struct ovl_fh *ovl_encode_real_fh(struct dentry *real, bool is_upper); int ovl_set_origin(struct dentry *dentry, struct dentry *lower, struct dentry *upper); diff --git a/fs/overlayfs/ovl_entry.h b/fs/overlayfs/ovl_entry.h index bfef6edcc111..41655a7d6894 100644 --- a/fs/overlayfs/ovl_entry.h +++ b/fs/overlayfs/ovl_entry.h @@ -18,13 +18,21 @@ struct ovl_config { const char *redirect_mode; bool index; bool nfs_export; + int xino; +}; + +struct ovl_sb { + struct super_block *sb; + dev_t pseudo_dev; }; struct ovl_layer { struct vfsmount *mnt; - dev_t pseudo_dev; - /* Index of this layer in fs root (upper == 0) */ + struct ovl_sb *fs; + /* Index of this layer in fs root (upper idx == 0) */ int idx; + /* One fsid per unique underlying sb (upper fsid == 0) */ + int fsid; }; struct ovl_path { @@ -35,8 +43,11 @@ struct ovl_path { /* private information held for overlayfs's superblock */ struct ovl_fs { struct vfsmount *upper_mnt; - unsigned numlower; + unsigned int numlower; + /* Number of unique lower sb that differ from upper sb */ + unsigned int numlowerfs; struct ovl_layer *lower_layers; + struct ovl_sb *lower_fs; /* workbasedir is the path at workdir= mount option */ struct dentry *workbasedir; /* workdir is the 'work' directory under workbasedir */ @@ -50,11 +61,11 @@ struct ovl_fs { const struct cred *creator_cred; bool tmpfile; bool noxattr; - /* sb common to all layers */ - struct super_block *same_sb; /* Did we take the inuse lock? */ bool upperdir_locked; bool workdir_locked; + /* Inode numbers in all layers do not use the high xino_bits */ + unsigned int xino_bits; }; /* private information held for every overlayfs dentry */ diff --git a/fs/overlayfs/readdir.c b/fs/overlayfs/readdir.c index c11f5c0906c3..ef1fe42ff7bb 100644 --- a/fs/overlayfs/readdir.c +++ b/fs/overlayfs/readdir.c @@ -120,6 +120,10 @@ static bool ovl_calc_d_ino(struct ovl_readdir_data *rdd, if (!rdd->dentry) return false; + /* Always recalc d_ino when remapping lower inode numbers */ + if (ovl_xino_bits(rdd->dentry->d_sb)) + return true; + /* Always recalc d_ino for parent */ if (strcmp(p->name, "..") == 0) return true; @@ -435,6 +439,19 @@ static struct ovl_dir_cache *ovl_cache_get(struct dentry *dentry) return cache; } +/* Map inode number to lower fs unique range */ +static u64 ovl_remap_lower_ino(u64 ino, int xinobits, int fsid, + const char *name, int namelen) +{ + if (ino >> (64 - xinobits)) { + pr_warn_ratelimited("overlayfs: d_ino too big (%.*s, ino=%llu, xinobits=%d)\n", + namelen, name, ino, xinobits); + return ino; + } + + return ino | ((u64)fsid) << (64 - xinobits); +} + /* * Set d_ino for upper entries. Non-upper entries should always report * the uppermost real inode ino and should not call this function. @@ -452,9 +469,10 @@ static int ovl_cache_update_ino(struct path *path, struct ovl_cache_entry *p) struct dentry *this = NULL; enum ovl_path_type type; u64 ino = p->real_ino; + int xinobits = ovl_xino_bits(dir->d_sb); int err = 0; - if (!ovl_same_sb(dir->d_sb)) + if (!ovl_same_sb(dir->d_sb) && !xinobits) goto out; if (p->name[0] == '.') { @@ -491,6 +509,10 @@ get: WARN_ON_ONCE(dir->d_sb->s_dev != stat.dev); ino = stat.ino; + } else if (xinobits && !OVL_TYPE_UPPER(type)) { + ino = ovl_remap_lower_ino(ino, xinobits, + ovl_layer_lower(this)->fsid, + p->name, p->len); } out: @@ -618,6 +640,8 @@ struct ovl_readdir_translate { struct ovl_dir_cache *cache; struct dir_context ctx; u64 parent_ino; + int fsid; + int xinobits; }; static int ovl_fill_real(struct dir_context *ctx, const char *name, @@ -628,14 +652,17 @@ static int ovl_fill_real(struct dir_context *ctx, const char *name, container_of(ctx, struct ovl_readdir_translate, ctx); struct dir_context *orig_ctx = rdt->orig_ctx; - if (rdt->parent_ino && strcmp(name, "..") == 0) + if (rdt->parent_ino && strcmp(name, "..") == 0) { ino = rdt->parent_ino; - else if (rdt->cache) { + } else if (rdt->cache) { struct ovl_cache_entry *p; p = ovl_cache_entry_find(&rdt->cache->root, name, namelen); if (p) ino = p->ino; + } else if (rdt->xinobits) { + ino = ovl_remap_lower_ino(ino, rdt->xinobits, rdt->fsid, + name, namelen); } return orig_ctx->actor(orig_ctx, name, namelen, offset, ino, d_type); @@ -646,11 +673,16 @@ static int ovl_iterate_real(struct file *file, struct dir_context *ctx) int err; struct ovl_dir_file *od = file->private_data; struct dentry *dir = file->f_path.dentry; + struct ovl_layer *lower_layer = ovl_layer_lower(dir); struct ovl_readdir_translate rdt = { .ctx.actor = ovl_fill_real, .orig_ctx = ctx, + .xinobits = ovl_xino_bits(dir->d_sb), }; + if (rdt.xinobits && lower_layer) + rdt.fsid = lower_layer->fsid; + if (OVL_TYPE_MERGE(ovl_path_type(dir->d_parent))) { struct kstat stat; struct path statpath = file->f_path; @@ -693,9 +725,10 @@ static int ovl_iterate(struct file *file, struct dir_context *ctx) * dir is impure then need to adjust d_ino for copied up * entries. */ - if (ovl_same_sb(dentry->d_sb) && - (ovl_test_flag(OVL_IMPURE, d_inode(dentry)) || - OVL_TYPE_MERGE(ovl_path_type(dentry->d_parent)))) { + if (ovl_xino_bits(dentry->d_sb) || + (ovl_same_sb(dentry->d_sb) && + (ovl_test_flag(OVL_IMPURE, d_inode(dentry)) || + OVL_TYPE_MERGE(ovl_path_type(dentry->d_parent))))) { return ovl_iterate_real(file, ctx); } return iterate_dir(od->realfile, ctx); diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index 7c24619ae7fc..e8551c97de51 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -17,6 +17,7 @@ #include <linux/statfs.h> #include <linux/seq_file.h> #include <linux/posix_acl_xattr.h> +#include <linux/exportfs.h> #include "overlayfs.h" MODULE_AUTHOR("Miklos Szeredi <miklos@szeredi.hu>"); @@ -50,6 +51,11 @@ module_param_named(nfs_export, ovl_nfs_export_def, bool, 0644); MODULE_PARM_DESC(ovl_nfs_export_def, "Default to on or off for the NFS export feature"); +static bool ovl_xino_auto_def = IS_ENABLED(CONFIG_OVERLAY_FS_XINO_AUTO); +module_param_named(xino_auto, ovl_xino_auto_def, bool, 0644); +MODULE_PARM_DESC(ovl_xino_auto_def, + "Auto enable xino feature"); + static void ovl_entry_stack_free(struct ovl_entry *oe) { unsigned int i; @@ -236,11 +242,12 @@ static void ovl_free_fs(struct ovl_fs *ofs) if (ofs->upperdir_locked) ovl_inuse_unlock(ofs->upper_mnt->mnt_root); mntput(ofs->upper_mnt); - for (i = 0; i < ofs->numlower; i++) { + for (i = 0; i < ofs->numlower; i++) mntput(ofs->lower_layers[i].mnt); - free_anon_bdev(ofs->lower_layers[i].pseudo_dev); - } + for (i = 0; i < ofs->numlowerfs; i++) + free_anon_bdev(ofs->lower_fs[i].pseudo_dev); kfree(ofs->lower_layers); + kfree(ofs->lower_fs); kfree(ofs->config.lowerdir); kfree(ofs->config.upperdir); @@ -325,6 +332,23 @@ static const char *ovl_redirect_mode_def(void) return ovl_redirect_dir_def ? "on" : "off"; } +enum { + OVL_XINO_OFF, + OVL_XINO_AUTO, + OVL_XINO_ON, +}; + +static const char * const ovl_xino_str[] = { + "off", + "auto", + "on", +}; + +static inline int ovl_xino_def(void) +{ + return ovl_xino_auto_def ? OVL_XINO_AUTO : OVL_XINO_OFF; +} + /** * ovl_show_options * @@ -350,6 +374,8 @@ static int ovl_show_options(struct seq_file *m, struct dentry *dentry) if (ofs->config.nfs_export != ovl_nfs_export_def) seq_printf(m, ",nfs_export=%s", ofs->config.nfs_export ? "on" : "off"); + if (ofs->config.xino != ovl_xino_def()) + seq_printf(m, ",xino=%s", ovl_xino_str[ofs->config.xino]); return 0; } @@ -384,6 +410,9 @@ enum { OPT_INDEX_OFF, OPT_NFS_EXPORT_ON, OPT_NFS_EXPORT_OFF, + OPT_XINO_ON, + OPT_XINO_OFF, + OPT_XINO_AUTO, OPT_ERR, }; @@ -397,6 +426,9 @@ static const match_table_t ovl_tokens = { {OPT_INDEX_OFF, "index=off"}, {OPT_NFS_EXPORT_ON, "nfs_export=on"}, {OPT_NFS_EXPORT_OFF, "nfs_export=off"}, + {OPT_XINO_ON, "xino=on"}, + {OPT_XINO_OFF, "xino=off"}, + {OPT_XINO_AUTO, "xino=auto"}, {OPT_ERR, NULL} }; @@ -511,6 +543,18 @@ static int ovl_parse_opt(char *opt, struct ovl_config *config) config->nfs_export = false; break; + case OPT_XINO_ON: + config->xino = OVL_XINO_ON; + break; + + case OPT_XINO_OFF: + config->xino = OVL_XINO_OFF; + break; + + case OPT_XINO_AUTO: + config->xino = OVL_XINO_AUTO; + break; + default: pr_err("overlayfs: unrecognized mount option \"%s\" or missing value\n", p); return -EINVAL; @@ -700,6 +744,7 @@ static int ovl_check_namelen(struct path *path, struct ovl_fs *ofs, static int ovl_lower_dir(const char *name, struct path *path, struct ovl_fs *ofs, int *stack_depth, bool *remote) { + int fh_type; int err; err = ovl_mount_dir_noesc(name, path); @@ -719,15 +764,19 @@ static int ovl_lower_dir(const char *name, struct path *path, * The inodes index feature and NFS export need to encode and decode * file handles, so they require that all layers support them. */ + fh_type = ovl_can_decode_fh(path->dentry->d_sb); if ((ofs->config.nfs_export || - (ofs->config.index && ofs->config.upperdir)) && - !ovl_can_decode_fh(path->dentry->d_sb)) { + (ofs->config.index && ofs->config.upperdir)) && !fh_type) { ofs->config.index = false; ofs->config.nfs_export = false; pr_warn("overlayfs: fs on '%s' does not support file handles, falling back to index=off,nfs_export=off.\n", name); } + /* Check if lower fs has 32bit inode numbers */ + if (fh_type != FILEID_INO32_GEN) + ofs->xino_bits = 0; + return 0; out_put: @@ -951,6 +1000,7 @@ static int ovl_make_workdir(struct ovl_fs *ofs, struct path *workpath) { struct vfsmount *mnt = ofs->upper_mnt; struct dentry *temp; + int fh_type; int err; err = mnt_want_write(mnt); @@ -1000,12 +1050,16 @@ static int ovl_make_workdir(struct ovl_fs *ofs, struct path *workpath) } /* Check if upper/work fs supports file handles */ - if (ofs->config.index && - !ovl_can_decode_fh(ofs->workdir->d_sb)) { + fh_type = ovl_can_decode_fh(ofs->workdir->d_sb); + if (ofs->config.index && !fh_type) { ofs->config.index = false; pr_warn("overlayfs: upper fs does not support file handles, falling back to index=off.\n"); } + /* Check if upper fs has 32bit inode numbers */ + if (fh_type != FILEID_INO32_GEN) + ofs->xino_bits = 0; + /* NFS export of r/w mount depends on index */ if (ofs->config.nfs_export && !ofs->config.index) { pr_warn("overlayfs: NFS export requires \"index=on\", falling back to nfs_export=off.\n"); @@ -1108,6 +1162,35 @@ out: return err; } +/* Get a unique fsid for the layer */ +static int ovl_get_fsid(struct ovl_fs *ofs, struct super_block *sb) +{ + unsigned int i; + dev_t dev; + int err; + + /* fsid 0 is reserved for upper fs even with non upper overlay */ + if (ofs->upper_mnt && ofs->upper_mnt->mnt_sb == sb) + return 0; + + for (i = 0; i < ofs->numlowerfs; i++) { + if (ofs->lower_fs[i].sb == sb) + return i + 1; + } + + err = get_anon_bdev(&dev); + if (err) { + pr_err("overlayfs: failed to get anonymous bdev for lowerpath\n"); + return err; + } + + ofs->lower_fs[ofs->numlowerfs].sb = sb; + ofs->lower_fs[ofs->numlowerfs].pseudo_dev = dev; + ofs->numlowerfs++; + + return ofs->numlowerfs; +} + static int ovl_get_lower_layers(struct ovl_fs *ofs, struct path *stack, unsigned int numlower) { @@ -1119,23 +1202,27 @@ static int ovl_get_lower_layers(struct ovl_fs *ofs, struct path *stack, GFP_KERNEL); if (ofs->lower_layers == NULL) goto out; + + ofs->lower_fs = kcalloc(numlower, sizeof(struct ovl_sb), + GFP_KERNEL); + if (ofs->lower_fs == NULL) + goto out; + for (i = 0; i < numlower; i++) { struct vfsmount *mnt; - dev_t dev; + int fsid; - err = get_anon_bdev(&dev); - if (err) { - pr_err("overlayfs: failed to get anonymous bdev for lowerpath\n"); + err = fsid = ovl_get_fsid(ofs, stack[i].mnt->mnt_sb); + if (err < 0) goto out; - } mnt = clone_private_mount(&stack[i]); err = PTR_ERR(mnt); if (IS_ERR(mnt)) { pr_err("overlayfs: failed to clone lowerpath\n"); - free_anon_bdev(dev); goto out; } + /* * Make lower layers R/O. That way fchmod/fchown on lower file * will fail instead of modifying lower fs. @@ -1143,16 +1230,41 @@ static int ovl_get_lower_layers(struct ovl_fs *ofs, struct path *stack, mnt->mnt_flags |= MNT_READONLY | MNT_NOATIME; ofs->lower_layers[ofs->numlower].mnt = mnt; - ofs->lower_layers[ofs->numlower].pseudo_dev = dev; ofs->lower_layers[ofs->numlower].idx = i + 1; + ofs->lower_layers[ofs->numlower].fsid = fsid; + if (fsid) { + ofs->lower_layers[ofs->numlower].fs = + &ofs->lower_fs[fsid - 1]; + } ofs->numlower++; + } + + /* + * When all layers on same fs, overlay can use real inode numbers. + * With mount option "xino=on", mounter declares that there are enough + * free high bits in underlying fs to hold the unique fsid. + * If overlayfs does encounter underlying inodes using the high xino + * bits reserved for fsid, it emits a warning and uses the original + * inode number. + */ + if (!ofs->numlowerfs || (ofs->numlowerfs == 1 && !ofs->upper_mnt)) { + ofs->xino_bits = 0; + ofs->config.xino = OVL_XINO_OFF; + } else if (ofs->config.xino == OVL_XINO_ON && !ofs->xino_bits) { + /* + * This is a roundup of number of bits needed for numlowerfs+1 + * (i.e. ilog2(numlowerfs+1 - 1) + 1). fsid 0 is reserved for + * upper fs even with non upper overlay. + */ + BUILD_BUG_ON(ilog2(OVL_MAX_STACK) > 31); + ofs->xino_bits = ilog2(ofs->numlowerfs) + 1; + } - /* Check if all lower layers are on same sb */ - if (i == 0) - ofs->same_sb = mnt->mnt_sb; - else if (ofs->same_sb != mnt->mnt_sb) - ofs->same_sb = NULL; + if (ofs->xino_bits) { + pr_info("overlayfs: \"xino\" feature enabled using %d upper inode bits.\n", + ofs->xino_bits); } + err = 0; out: return err; @@ -1263,6 +1375,7 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent) ofs->config.index = ovl_index_def; ofs->config.nfs_export = ovl_nfs_export_def; + ofs->config.xino = ovl_xino_def(); err = ovl_parse_opt((char *) data, &ofs->config); if (err) goto out_err; @@ -1276,6 +1389,10 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent) sb->s_stack_depth = 0; sb->s_maxbytes = MAX_LFS_FILESIZE; + /* Assume underlaying fs uses 32bit inodes unless proven otherwise */ + if (ofs->config.xino != OVL_XINO_OFF) + ofs->xino_bits = BITS_PER_LONG - 32; + if (ofs->config.upperdir) { if (!ofs->config.workdir) { pr_err("overlayfs: missing 'workdir'\n"); @@ -1305,8 +1422,6 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent) /* If the upper fs is nonexistent, we mark overlayfs r/o too */ if (!ofs->upper_mnt) sb->s_flags |= SB_RDONLY; - else if (ofs->upper_mnt->mnt_sb != ofs->same_sb) - ofs->same_sb = NULL; if (!(ovl_force_readonly(ofs)) && ofs->config.index) { err = ovl_get_indexdir(ofs, oe, &upperpath); diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c index 930784a26623..6f1078028c66 100644 --- a/fs/overlayfs/util.c +++ b/fs/overlayfs/util.c @@ -47,13 +47,29 @@ struct super_block *ovl_same_sb(struct super_block *sb) { struct ovl_fs *ofs = sb->s_fs_info; - return ofs->same_sb; + if (!ofs->numlowerfs) + return ofs->upper_mnt->mnt_sb; + else if (ofs->numlowerfs == 1 && !ofs->upper_mnt) + return ofs->lower_fs[0].sb; + else + return NULL; } -bool ovl_can_decode_fh(struct super_block *sb) +/* + * Check if underlying fs supports file handles and try to determine encoding + * type, in order to deduce maximum inode number used by fs. + * + * Return 0 if file handles are not supported. + * Return 1 (FILEID_INO32_GEN) if fs uses the default 32bit inode encoding. + * Return -1 if fs uses a non default encoding with unknown inode size. + */ +int ovl_can_decode_fh(struct super_block *sb) { - return (sb->s_export_op && sb->s_export_op->fh_to_dentry && - !uuid_is_null(&sb->s_uuid)); + if (!sb->s_export_op || !sb->s_export_op->fh_to_dentry || + uuid_is_null(&sb->s_uuid)) + return 0; + + return sb->s_export_op->encode_fh ? -1 : FILEID_INO32_GEN; } struct dentry *ovl_indexdir(struct super_block *sb) @@ -172,6 +188,13 @@ struct dentry *ovl_dentry_lower(struct dentry *dentry) return oe->numlower ? oe->lowerstack[0].dentry : NULL; } +struct ovl_layer *ovl_layer_lower(struct dentry *dentry) +{ + struct ovl_entry *oe = dentry->d_fsdata; + + return oe->numlower ? oe->lowerstack[0].layer : NULL; +} + struct dentry *ovl_dentry_real(struct dentry *dentry) { return ovl_dentry_upper(dentry) ?: ovl_dentry_lower(dentry); @@ -279,12 +302,16 @@ void ovl_dentry_set_redirect(struct dentry *dentry, const char *redirect) void ovl_inode_init(struct inode *inode, struct dentry *upperdentry, struct dentry *lowerdentry) { + struct inode *realinode = d_inode(upperdentry ?: lowerdentry); + if (upperdentry) OVL_I(inode)->__upperdentry = upperdentry; if (lowerdentry) OVL_I(inode)->lower = igrab(d_inode(lowerdentry)); - ovl_copyattr(d_inode(upperdentry ?: lowerdentry), inode); + ovl_copyattr(realinode, inode); + if (!inode->i_ino) + inode->i_ino = realinode->i_ino; } void ovl_inode_update(struct inode *inode, struct dentry *upperdentry) @@ -299,6 +326,8 @@ void ovl_inode_update(struct inode *inode, struct dentry *upperdentry) smp_wmb(); OVL_I(inode)->__upperdentry = upperdentry; if (inode_unhashed(inode)) { + if (!inode->i_ino) + inode->i_ino = upperinode->i_ino; inode->i_private = upperinode; __insert_inode_hash(inode, (unsigned long) upperinode); } diff --git a/fs/pipe.c b/fs/pipe.c index 39d6f431da83..bb0840e234f3 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -509,19 +509,22 @@ static long pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) } } -/* No kernel lock held - fine */ -static __poll_t -pipe_poll(struct file *filp, poll_table *wait) +static struct wait_queue_head * +pipe_get_poll_head(struct file *filp, __poll_t events) { - __poll_t mask; struct pipe_inode_info *pipe = filp->private_data; - int nrbufs; - poll_wait(filp, &pipe->wait, wait); + return &pipe->wait; +} + +/* No kernel lock held - fine */ +static __poll_t pipe_poll_mask(struct file *filp, __poll_t events) +{ + struct pipe_inode_info *pipe = filp->private_data; + int nrbufs = pipe->nrbufs; + __poll_t mask = 0; /* Reading only -- no need for acquiring the semaphore. */ - nrbufs = pipe->nrbufs; - mask = 0; if (filp->f_mode & FMODE_READ) { mask = (nrbufs > 0) ? EPOLLIN | EPOLLRDNORM : 0; if (!pipe->writers && filp->f_version != pipe->w_counter) @@ -1020,7 +1023,8 @@ const struct file_operations pipefifo_fops = { .llseek = no_llseek, .read_iter = pipe_read, .write_iter = pipe_write, - .poll = pipe_poll, + .get_poll_head = pipe_get_poll_head, + .poll_mask = pipe_poll_mask, .unlocked_ioctl = pipe_ioctl, .release = pipe_release, .fasync = pipe_fasync, diff --git a/fs/proc/array.c b/fs/proc/array.c index 598803576e4c..004077f1a7bf 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -85,6 +85,7 @@ #include <linux/delayacct.h> #include <linux/seq_file.h> #include <linux/pid_namespace.h> +#include <linux/prctl.h> #include <linux/ptrace.h> #include <linux/tracehook.h> #include <linux/string_helpers.h> @@ -95,22 +96,29 @@ #include <asm/processor.h> #include "internal.h" -static inline void task_name(struct seq_file *m, struct task_struct *p) +void proc_task_name(struct seq_file *m, struct task_struct *p, bool escape) { char *buf; size_t size; - char tcomm[sizeof(p->comm)]; + char tcomm[64]; int ret; - get_task_comm(tcomm, p); - - seq_puts(m, "Name:\t"); + if (p->flags & PF_WQ_WORKER) + wq_worker_comm(tcomm, sizeof(tcomm), p); + else + __get_task_comm(tcomm, sizeof(tcomm), p); size = seq_get_buf(m, &buf); - ret = string_escape_str(tcomm, buf, size, ESCAPE_SPACE | ESCAPE_SPECIAL, "\n\\"); - seq_commit(m, ret < size ? ret : -1); + if (escape) { + ret = string_escape_str(tcomm, buf, size, + ESCAPE_SPACE | ESCAPE_SPECIAL, "\n\\"); + if (ret >= size) + ret = -1; + } else { + ret = strscpy(buf, tcomm, size); + } - seq_putc(m, '\n'); + seq_commit(m, ret); } /* @@ -141,25 +149,12 @@ static inline const char *get_task_state(struct task_struct *tsk) return task_state_array[task_state_index(tsk)]; } -static inline int get_task_umask(struct task_struct *tsk) -{ - struct fs_struct *fs; - int umask = -ENOENT; - - task_lock(tsk); - fs = tsk->fs; - if (fs) - umask = fs->umask; - task_unlock(tsk); - return umask; -} - static inline void task_state(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *p) { struct user_namespace *user_ns = seq_user_ns(m); struct group_info *group_info; - int g, umask; + int g, umask = -1; struct task_struct *tracer; const struct cred *cred; pid_t ppid, tpid = 0, tgid, ngid; @@ -177,17 +172,18 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns, ngid = task_numa_group_id(p); cred = get_task_cred(p); - umask = get_task_umask(p); - if (umask >= 0) - seq_printf(m, "Umask:\t%#04o\n", umask); - task_lock(p); + if (p->fs) + umask = p->fs->umask; if (p->files) max_fds = files_fdtable(p->files)->max_fds; task_unlock(p); rcu_read_unlock(); - seq_printf(m, "State:\t%s", get_task_state(p)); + if (umask >= 0) + seq_printf(m, "Umask:\t%#04o\n", umask); + seq_puts(m, "State:\t"); + seq_puts(m, get_task_state(p)); seq_put_decimal_ull(m, "\nTgid:\t", tgid); seq_put_decimal_ull(m, "\nNgid:\t", ngid); @@ -313,8 +309,8 @@ static void render_cap_t(struct seq_file *m, const char *header, seq_puts(m, header); CAP_FOR_EACH_U32(__capi) { - seq_printf(m, "%08x", - a->cap[CAP_LAST_U32 - __capi]); + seq_put_hex_ll(m, NULL, + a->cap[CAP_LAST_U32 - __capi], 8); } seq_putc(m, '\n'); } @@ -347,6 +343,30 @@ static inline void task_seccomp(struct seq_file *m, struct task_struct *p) #ifdef CONFIG_SECCOMP seq_put_decimal_ull(m, "\nSeccomp:\t", p->seccomp.mode); #endif + seq_printf(m, "\nSpeculation_Store_Bypass:\t"); + switch (arch_prctl_spec_ctrl_get(p, PR_SPEC_STORE_BYPASS)) { + case -EINVAL: + seq_printf(m, "unknown"); + break; + case PR_SPEC_NOT_AFFECTED: + seq_printf(m, "not vulnerable"); + break; + case PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE: + seq_printf(m, "thread force mitigated"); + break; + case PR_SPEC_PRCTL | PR_SPEC_DISABLE: + seq_printf(m, "thread mitigated"); + break; + case PR_SPEC_PRCTL | PR_SPEC_ENABLE: + seq_printf(m, "thread vulnerable"); + break; + case PR_SPEC_DISABLE: + seq_printf(m, "globally mitigated"); + break; + default: + seq_printf(m, "vulnerable"); + break; + } seq_putc(m, '\n'); } @@ -368,7 +388,8 @@ static void task_cpus_allowed(struct seq_file *m, struct task_struct *task) static inline void task_core_dumping(struct seq_file *m, struct mm_struct *mm) { - seq_printf(m, "CoreDumping:\t%d\n", !!mm->core_state); + seq_put_decimal_ull(m, "CoreDumping:\t", !!mm->core_state); + seq_putc(m, '\n'); } int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, @@ -376,7 +397,10 @@ int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, { struct mm_struct *mm = get_task_mm(task); - task_name(m, task); + seq_puts(m, "Name:\t"); + proc_task_name(m, task, true); + seq_putc(m, '\n'); + task_state(m, ns, pid, task); if (mm) { @@ -411,7 +435,6 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, u64 cutime, cstime, utime, stime; u64 cgtime, gtime; unsigned long rsslim = 0; - char tcomm[sizeof(task->comm)]; unsigned long flags; state = *get_task_state(task); @@ -438,8 +461,6 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, } } - get_task_comm(tcomm, task); - sigemptyset(&sigign); sigemptyset(&sigcatch); cutime = cstime = utime = stime = 0; @@ -504,7 +525,11 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, /* convert nsec -> ticks */ start_time = nsec_to_clock_t(task->real_start_time); - seq_printf(m, "%d (%s) %c", pid_nr_ns(pid, ns), tcomm, state); + seq_put_decimal_ull(m, "", pid_nr_ns(pid, ns)); + seq_puts(m, " ("); + proc_task_name(m, task, false); + seq_puts(m, ") "); + seq_putc(m, state); seq_put_decimal_ll(m, " ", ppid); seq_put_decimal_ll(m, " ", pgid); seq_put_decimal_ll(m, " ", sid); @@ -684,25 +709,22 @@ out: static int children_seq_show(struct seq_file *seq, void *v) { - struct inode *inode = seq->private; - pid_t pid; - - pid = pid_nr_ns(v, inode->i_sb->s_fs_info); - seq_printf(seq, "%d ", pid); + struct inode *inode = file_inode(seq->file); + seq_printf(seq, "%d ", pid_nr_ns(v, proc_pid_ns(inode))); return 0; } static void *children_seq_start(struct seq_file *seq, loff_t *pos) { - return get_children_pid(seq->private, NULL, *pos); + return get_children_pid(file_inode(seq->file), NULL, *pos); } static void *children_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct pid *pid; - pid = get_children_pid(seq->private, v, *pos + 1); + pid = get_children_pid(file_inode(seq->file), v, *pos + 1); put_pid(v); ++*pos; @@ -723,17 +745,7 @@ static const struct seq_operations children_seq_ops = { static int children_seq_open(struct inode *inode, struct file *file) { - struct seq_file *m; - int ret; - - ret = seq_open(file, &children_seq_ops); - if (ret) - return ret; - - m = file->private_data; - m->private = inode; - - return ret; + return seq_open(file, &children_seq_ops); } const struct file_operations proc_tid_children_operations = { diff --git a/fs/proc/base.c b/fs/proc/base.c index d53246863cfb..af128b374143 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -261,7 +261,7 @@ static ssize_t proc_pid_cmdline_read(struct file *file, char __user *buf, * Inherently racy -- command line shares address space * with code and data. */ - rv = access_remote_vm(mm, arg_end - 1, &c, 1, 0); + rv = access_remote_vm(mm, arg_end - 1, &c, 1, FOLL_ANON); if (rv <= 0) goto out_free_page; @@ -279,7 +279,7 @@ static ssize_t proc_pid_cmdline_read(struct file *file, char __user *buf, int nr_read; _count = min3(count, len, PAGE_SIZE); - nr_read = access_remote_vm(mm, p, page, _count, 0); + nr_read = access_remote_vm(mm, p, page, _count, FOLL_ANON); if (nr_read < 0) rv = nr_read; if (nr_read <= 0) @@ -325,7 +325,7 @@ static ssize_t proc_pid_cmdline_read(struct file *file, char __user *buf, bool final; _count = min3(count, len, PAGE_SIZE); - nr_read = access_remote_vm(mm, p, page, _count, 0); + nr_read = access_remote_vm(mm, p, page, _count, FOLL_ANON); if (nr_read < 0) rv = nr_read; if (nr_read <= 0) @@ -388,14 +388,17 @@ static int proc_pid_wchan(struct seq_file *m, struct pid_namespace *ns, unsigned long wchan; char symname[KSYM_NAME_LEN]; - wchan = get_wchan(task); + if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) + goto print0; - if (wchan && ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS) - && !lookup_symbol_name(wchan, symname)) - seq_printf(m, "%s", symname); - else - seq_putc(m, '0'); + wchan = get_wchan(task); + if (wchan && !lookup_symbol_name(wchan, symname)) { + seq_puts(m, symname); + return 0; + } +print0: + seq_putc(m, '0'); return 0; } #endif /* CONFIG_KALLSYMS */ @@ -695,7 +698,7 @@ static bool has_pid_permissions(struct pid_namespace *pid, static int proc_pid_permission(struct inode *inode, int mask) { - struct pid_namespace *pid = inode->i_sb->s_fs_info; + struct pid_namespace *pid = proc_pid_ns(inode); struct task_struct *task; bool has_perms; @@ -730,13 +733,11 @@ static const struct inode_operations proc_def_inode_operations = { static int proc_single_show(struct seq_file *m, void *v) { struct inode *inode = m->private; - struct pid_namespace *ns; - struct pid *pid; + struct pid_namespace *ns = proc_pid_ns(inode); + struct pid *pid = proc_pid(inode); struct task_struct *task; int ret; - ns = inode->i_sb->s_fs_info; - pid = proc_pid(inode); task = get_pid_task(pid, PIDTYPE_PID); if (!task) return -ESRCH; @@ -943,7 +944,7 @@ static ssize_t environ_read(struct file *file, char __user *buf, max_len = min_t(size_t, PAGE_SIZE, count); this_len = min(max_len, this_len); - retval = access_remote_vm(mm, (env_start + src), page, this_len, 0); + retval = access_remote_vm(mm, (env_start + src), page, this_len, FOLL_ANON); if (retval <= 0) { ret = retval; @@ -1407,7 +1408,7 @@ static const struct file_operations proc_fail_nth_operations = { static int sched_show(struct seq_file *m, void *v) { struct inode *inode = m->private; - struct pid_namespace *ns = inode->i_sb->s_fs_info; + struct pid_namespace *ns = proc_pid_ns(inode); struct task_struct *p; p = get_proc_task(inode); @@ -1562,9 +1563,8 @@ static int comm_show(struct seq_file *m, void *v) if (!p) return -ESRCH; - task_lock(p); - seq_printf(m, "%s\n", p->comm); - task_unlock(p); + proc_task_name(m, p, false); + seq_putc(m, '\n'); put_task_struct(p); @@ -1690,6 +1690,12 @@ void task_dump_owner(struct task_struct *task, umode_t mode, kuid_t uid; kgid_t gid; + if (unlikely(task->flags & PF_KTHREAD)) { + *ruid = GLOBAL_ROOT_UID; + *rgid = GLOBAL_ROOT_GID; + return; + } + /* Default to the tasks effective ownership */ rcu_read_lock(); cred = __task_cred(task); @@ -1773,8 +1779,8 @@ int pid_getattr(const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); + struct pid_namespace *pid = proc_pid_ns(inode); struct task_struct *task; - struct pid_namespace *pid = path->dentry->d_sb->s_fs_info; generic_fillattr(inode, stat); @@ -1800,15 +1806,22 @@ int pid_getattr(const struct path *path, struct kstat *stat, /* dentry stuff */ /* - * Exceptional case: normally we are not allowed to unhash a busy - * directory. In this case, however, we can do it - no aliasing problems - * due to the way we treat inodes. - * + * Set <pid>/... inode ownership (can change due to setuid(), etc.) + */ +void pid_update_inode(struct task_struct *task, struct inode *inode) +{ + task_dump_owner(task, inode->i_mode, &inode->i_uid, &inode->i_gid); + + inode->i_mode &= ~(S_ISUID | S_ISGID); + security_task_to_inode(task, inode); +} + +/* * Rewrite the inode's ownerships here because the owning task may have * performed a setuid(), etc. * */ -int pid_revalidate(struct dentry *dentry, unsigned int flags) +static int pid_revalidate(struct dentry *dentry, unsigned int flags) { struct inode *inode; struct task_struct *task; @@ -1820,10 +1833,7 @@ int pid_revalidate(struct dentry *dentry, unsigned int flags) task = get_proc_task(inode); if (task) { - task_dump_owner(task, inode->i_mode, &inode->i_uid, &inode->i_gid); - - inode->i_mode &= ~(S_ISUID | S_ISGID); - security_task_to_inode(task, inode); + pid_update_inode(task, inode); put_task_struct(task); return 1; } @@ -1871,8 +1881,8 @@ bool proc_fill_cache(struct file *file, struct dir_context *ctx, struct dentry *child, *dir = file->f_path.dentry; struct qstr qname = QSTR_INIT(name, len); struct inode *inode; - unsigned type; - ino_t ino; + unsigned type = DT_UNKNOWN; + ino_t ino = 1; child = d_hash_and_lookup(dir, &qname); if (!child) { @@ -1881,22 +1891,23 @@ bool proc_fill_cache(struct file *file, struct dir_context *ctx, if (IS_ERR(child)) goto end_instantiate; if (d_in_lookup(child)) { - int err = instantiate(d_inode(dir), child, task, ptr); + struct dentry *res; + res = instantiate(child, task, ptr); d_lookup_done(child); - if (err < 0) { - dput(child); + if (IS_ERR(res)) goto end_instantiate; + if (unlikely(res)) { + dput(child); + child = res; } } } inode = d_inode(child); ino = inode->i_ino; type = inode->i_mode >> 12; +end_instantiate: dput(child); return dir_emit(ctx, name, len, ino, type); - -end_instantiate: - return dir_emit(ctx, name, len, 1, DT_UNKNOWN); } /* @@ -1910,6 +1921,8 @@ static int dname_to_vma_addr(struct dentry *dentry, unsigned long long sval, eval; unsigned int len; + if (str[0] == '0' && str[1] != '-') + return -EINVAL; len = _parse_integer(str, 16, &sval); if (len & KSTRTOX_OVERFLOW) return -EINVAL; @@ -1921,6 +1934,8 @@ static int dname_to_vma_addr(struct dentry *dentry, return -EINVAL; str++; + if (str[0] == '0' && str[1]) + return -EINVAL; len = _parse_integer(str, 16, &eval); if (len & KSTRTOX_OVERFLOW) return -EINVAL; @@ -2054,19 +2069,19 @@ static const struct inode_operations proc_map_files_link_inode_operations = { .setattr = proc_setattr, }; -static int -proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, +static struct dentry * +proc_map_files_instantiate(struct dentry *dentry, struct task_struct *task, const void *ptr) { fmode_t mode = (fmode_t)(unsigned long)ptr; struct proc_inode *ei; struct inode *inode; - inode = proc_pid_make_inode(dir->i_sb, task, S_IFLNK | + inode = proc_pid_make_inode(dentry->d_sb, task, S_IFLNK | ((mode & FMODE_READ ) ? S_IRUSR : 0) | ((mode & FMODE_WRITE) ? S_IWUSR : 0)); if (!inode) - return -ENOENT; + return ERR_PTR(-ENOENT); ei = PROC_I(inode); ei->op.proc_get_link = map_files_get_link; @@ -2075,9 +2090,7 @@ proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, inode->i_size = 64; d_set_d_op(dentry, &tid_map_files_dentry_operations); - d_add(dentry, inode); - - return 0; + return d_splice_alias(inode, dentry); } static struct dentry *proc_map_files_lookup(struct inode *dir, @@ -2086,19 +2099,19 @@ static struct dentry *proc_map_files_lookup(struct inode *dir, unsigned long vm_start, vm_end; struct vm_area_struct *vma; struct task_struct *task; - int result; + struct dentry *result; struct mm_struct *mm; - result = -ENOENT; + result = ERR_PTR(-ENOENT); task = get_proc_task(dir); if (!task) goto out; - result = -EACCES; + result = ERR_PTR(-EACCES); if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) goto out_put_task; - result = -ENOENT; + result = ERR_PTR(-ENOENT); if (dname_to_vma_addr(dentry, &vm_start, &vm_end)) goto out_put_task; @@ -2112,7 +2125,7 @@ static struct dentry *proc_map_files_lookup(struct inode *dir, goto out_no_vma; if (vma->vm_file) - result = proc_map_files_instantiate(dir, dentry, task, + result = proc_map_files_instantiate(dentry, task, (void *)(unsigned long)vma->vm_file->f_mode); out_no_vma: @@ -2121,7 +2134,7 @@ out_no_vma: out_put_task: put_task_struct(task); out: - return ERR_PTR(result); + return result; } static const struct inode_operations proc_map_files_inode_operations = { @@ -2204,6 +2217,7 @@ proc_map_files_readdir(struct file *file, struct dir_context *ctx) } } up_read(&mm->mmap_sem); + mmput(mm); for (i = 0; i < nr_files; i++) { char buf[4 * sizeof(long) + 2]; /* max: %lx-%lx\0 */ @@ -2221,7 +2235,6 @@ proc_map_files_readdir(struct file *file, struct dir_context *ctx) } if (fa) flex_array_free(fa); - mmput(mm); out_put_task: put_task_struct(task); @@ -2324,7 +2337,7 @@ static int proc_timers_open(struct inode *inode, struct file *file) return -ENOMEM; tp->pid = proc_pid(inode); - tp->ns = inode->i_sb->s_fs_info; + tp->ns = proc_pid_ns(inode); return 0; } @@ -2422,16 +2435,16 @@ static const struct file_operations proc_pid_set_timerslack_ns_operations = { .release = single_release, }; -static int proc_pident_instantiate(struct inode *dir, - struct dentry *dentry, struct task_struct *task, const void *ptr) +static struct dentry *proc_pident_instantiate(struct dentry *dentry, + struct task_struct *task, const void *ptr) { const struct pid_entry *p = ptr; struct inode *inode; struct proc_inode *ei; - inode = proc_pid_make_inode(dir->i_sb, task, p->mode); + inode = proc_pid_make_inode(dentry->d_sb, task, p->mode); if (!inode) - goto out; + return ERR_PTR(-ENOENT); ei = PROC_I(inode); if (S_ISDIR(inode->i_mode)) @@ -2441,13 +2454,9 @@ static int proc_pident_instantiate(struct inode *dir, if (p->fop) inode->i_fop = p->fop; ei->op = p->op; + pid_update_inode(task, inode); d_set_d_op(dentry, &pid_dentry_operations); - d_add(dentry, inode); - /* Close the race of the process dying before we return the dentry */ - if (pid_revalidate(dentry, 0)) - return 0; -out: - return -ENOENT; + return d_splice_alias(inode, dentry); } static struct dentry *proc_pident_lookup(struct inode *dir, @@ -2455,11 +2464,9 @@ static struct dentry *proc_pident_lookup(struct inode *dir, const struct pid_entry *ents, unsigned int nents) { - int error; struct task_struct *task = get_proc_task(dir); const struct pid_entry *p, *last; - - error = -ENOENT; + struct dentry *res = ERR_PTR(-ENOENT); if (!task) goto out_no_task; @@ -2478,11 +2485,11 @@ static struct dentry *proc_pident_lookup(struct inode *dir, if (p >= last) goto out; - error = proc_pident_instantiate(dir, dentry, task, p); + res = proc_pident_instantiate(dentry, task, p); out: put_task_struct(task); out_no_task: - return ERR_PTR(error); + return res; } static int proc_pident_readdir(struct file *file, struct dir_context *ctx, @@ -3125,38 +3132,32 @@ void proc_flush_task(struct task_struct *task) } } -static int proc_pid_instantiate(struct inode *dir, - struct dentry * dentry, +static struct dentry *proc_pid_instantiate(struct dentry * dentry, struct task_struct *task, const void *ptr) { struct inode *inode; - inode = proc_pid_make_inode(dir->i_sb, task, S_IFDIR | S_IRUGO | S_IXUGO); + inode = proc_pid_make_inode(dentry->d_sb, task, S_IFDIR | S_IRUGO | S_IXUGO); if (!inode) - goto out; + return ERR_PTR(-ENOENT); inode->i_op = &proc_tgid_base_inode_operations; inode->i_fop = &proc_tgid_base_operations; inode->i_flags|=S_IMMUTABLE; set_nlink(inode, nlink_tgid); + pid_update_inode(task, inode); d_set_d_op(dentry, &pid_dentry_operations); - - d_add(dentry, inode); - /* Close the race of the process dying before we return the dentry */ - if (pid_revalidate(dentry, 0)) - return 0; -out: - return -ENOENT; + return d_splice_alias(inode, dentry); } struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) { - int result = -ENOENT; struct task_struct *task; unsigned tgid; struct pid_namespace *ns; + struct dentry *result = ERR_PTR(-ENOENT); tgid = name_to_int(&dentry->d_name); if (tgid == ~0U) @@ -3171,10 +3172,10 @@ struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsign if (!task) goto out; - result = proc_pid_instantiate(dir, dentry, task, NULL); + result = proc_pid_instantiate(dentry, task, NULL); put_task_struct(task); out: - return ERR_PTR(result); + return result; } /* @@ -3226,7 +3227,7 @@ retry: int proc_pid_readdir(struct file *file, struct dir_context *ctx) { struct tgid_iter iter; - struct pid_namespace *ns = file_inode(file)->i_sb->s_fs_info; + struct pid_namespace *ns = proc_pid_ns(file_inode(file)); loff_t pos = ctx->pos; if (pos >= PID_MAX_LIMIT + TGID_OFFSET) @@ -3422,37 +3423,32 @@ static const struct inode_operations proc_tid_base_inode_operations = { .setattr = proc_setattr, }; -static int proc_task_instantiate(struct inode *dir, - struct dentry *dentry, struct task_struct *task, const void *ptr) +static struct dentry *proc_task_instantiate(struct dentry *dentry, + struct task_struct *task, const void *ptr) { struct inode *inode; - inode = proc_pid_make_inode(dir->i_sb, task, S_IFDIR | S_IRUGO | S_IXUGO); - + inode = proc_pid_make_inode(dentry->d_sb, task, S_IFDIR | S_IRUGO | S_IXUGO); if (!inode) - goto out; + return ERR_PTR(-ENOENT); + inode->i_op = &proc_tid_base_inode_operations; inode->i_fop = &proc_tid_base_operations; - inode->i_flags|=S_IMMUTABLE; + inode->i_flags |= S_IMMUTABLE; set_nlink(inode, nlink_tid); + pid_update_inode(task, inode); d_set_d_op(dentry, &pid_dentry_operations); - - d_add(dentry, inode); - /* Close the race of the process dying before we return the dentry */ - if (pid_revalidate(dentry, 0)) - return 0; -out: - return -ENOENT; + return d_splice_alias(inode, dentry); } static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) { - int result = -ENOENT; struct task_struct *task; struct task_struct *leader = get_proc_task(dir); unsigned tid; struct pid_namespace *ns; + struct dentry *result = ERR_PTR(-ENOENT); if (!leader) goto out_no_task; @@ -3472,13 +3468,13 @@ static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry if (!same_thread_group(leader, task)) goto out_drop_task; - result = proc_task_instantiate(dir, dentry, task, NULL); + result = proc_task_instantiate(dentry, task, NULL); out_drop_task: put_task_struct(task); out: put_task_struct(leader); out_no_task: - return ERR_PTR(result); + return result; } /* @@ -3575,7 +3571,7 @@ static int proc_task_readdir(struct file *file, struct dir_context *ctx) /* f_version caches the tgid value that the last readdir call couldn't * return. lseek aka telldir automagically resets f_version to 0. */ - ns = inode->i_sb->s_fs_info; + ns = proc_pid_ns(inode); tid = (int)file->f_version; file->f_version = 0; for (task = first_tid(proc_pid(inode), tid, ctx->pos - 2, ns); diff --git a/fs/proc/cmdline.c b/fs/proc/cmdline.c index 403cbb12a6e9..fa762c5fbcb2 100644 --- a/fs/proc/cmdline.c +++ b/fs/proc/cmdline.c @@ -6,25 +6,14 @@ static int cmdline_proc_show(struct seq_file *m, void *v) { - seq_printf(m, "%s\n", saved_command_line); + seq_puts(m, saved_command_line); + seq_putc(m, '\n'); return 0; } -static int cmdline_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, cmdline_proc_show, NULL); -} - -static const struct file_operations cmdline_proc_fops = { - .open = cmdline_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_cmdline_init(void) { - proc_create("cmdline", 0, NULL, &cmdline_proc_fops); + proc_create_single("cmdline", 0, NULL, cmdline_proc_show); return 0; } fs_initcall(proc_cmdline_init); diff --git a/fs/proc/consoles.c b/fs/proc/consoles.c index a8ac48aebd59..954caf0b7fee 100644 --- a/fs/proc/consoles.c +++ b/fs/proc/consoles.c @@ -91,21 +91,9 @@ static const struct seq_operations consoles_op = { .show = show_console_dev }; -static int consoles_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &consoles_op); -} - -static const struct file_operations proc_consoles_operations = { - .open = consoles_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - static int __init proc_consoles_init(void) { - proc_create("consoles", 0, NULL, &proc_consoles_operations); + proc_create_seq("consoles", 0, NULL, &consoles_op); return 0; } fs_initcall(proc_consoles_init); diff --git a/fs/proc/devices.c b/fs/proc/devices.c index 2c7f22b14489..37d38697eaf8 100644 --- a/fs/proc/devices.c +++ b/fs/proc/devices.c @@ -51,21 +51,9 @@ static const struct seq_operations devinfo_ops = { .show = devinfo_show }; -static int devinfo_open(struct inode *inode, struct file *filp) -{ - return seq_open(filp, &devinfo_ops); -} - -static const struct file_operations proc_devinfo_operations = { - .open = devinfo_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - static int __init proc_devices_init(void) { - proc_create("devices", 0, NULL, &proc_devinfo_operations); + proc_create_seq("devices", 0, NULL, &devinfo_ops); return 0; } fs_initcall(proc_devices_init); diff --git a/fs/proc/fd.c b/fs/proc/fd.c index 6b80cd1e419a..05b9893e9a22 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -81,9 +81,41 @@ static const struct file_operations proc_fdinfo_file_operations = { .release = single_release, }; +static bool tid_fd_mode(struct task_struct *task, unsigned fd, fmode_t *mode) +{ + struct files_struct *files = get_files_struct(task); + struct file *file; + + if (!files) + return false; + + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) + *mode = file->f_mode; + rcu_read_unlock(); + put_files_struct(files); + return !!file; +} + +static void tid_fd_update_inode(struct task_struct *task, struct inode *inode, + fmode_t f_mode) +{ + task_dump_owner(task, 0, &inode->i_uid, &inode->i_gid); + + if (S_ISLNK(inode->i_mode)) { + unsigned i_mode = S_IFLNK; + if (f_mode & FMODE_READ) + i_mode |= S_IRUSR | S_IXUSR; + if (f_mode & FMODE_WRITE) + i_mode |= S_IWUSR | S_IXUSR; + inode->i_mode = i_mode; + } + security_task_to_inode(task, inode); +} + static int tid_fd_revalidate(struct dentry *dentry, unsigned int flags) { - struct files_struct *files; struct task_struct *task; struct inode *inode; unsigned int fd; @@ -96,35 +128,11 @@ static int tid_fd_revalidate(struct dentry *dentry, unsigned int flags) fd = proc_fd(inode); if (task) { - files = get_files_struct(task); - if (files) { - struct file *file; - - rcu_read_lock(); - file = fcheck_files(files, fd); - if (file) { - unsigned f_mode = file->f_mode; - - rcu_read_unlock(); - put_files_struct(files); - - task_dump_owner(task, 0, &inode->i_uid, &inode->i_gid); - - if (S_ISLNK(inode->i_mode)) { - unsigned i_mode = S_IFLNK; - if (f_mode & FMODE_READ) - i_mode |= S_IRUSR | S_IXUSR; - if (f_mode & FMODE_WRITE) - i_mode |= S_IWUSR | S_IXUSR; - inode->i_mode = i_mode; - } - - security_task_to_inode(task, inode); - put_task_struct(task); - return 1; - } - rcu_read_unlock(); - put_files_struct(files); + fmode_t f_mode; + if (tid_fd_mode(task, fd, &f_mode)) { + tid_fd_update_inode(task, inode, f_mode); + put_task_struct(task); + return 1; } put_task_struct(task); } @@ -166,34 +174,33 @@ static int proc_fd_link(struct dentry *dentry, struct path *path) return ret; } -static int -proc_fd_instantiate(struct inode *dir, struct dentry *dentry, - struct task_struct *task, const void *ptr) +struct fd_data { + fmode_t mode; + unsigned fd; +}; + +static struct dentry *proc_fd_instantiate(struct dentry *dentry, + struct task_struct *task, const void *ptr) { - unsigned fd = (unsigned long)ptr; + const struct fd_data *data = ptr; struct proc_inode *ei; struct inode *inode; - inode = proc_pid_make_inode(dir->i_sb, task, S_IFLNK); + inode = proc_pid_make_inode(dentry->d_sb, task, S_IFLNK); if (!inode) - goto out; + return ERR_PTR(-ENOENT); ei = PROC_I(inode); - ei->fd = fd; + ei->fd = data->fd; inode->i_op = &proc_pid_link_inode_operations; inode->i_size = 64; ei->op.proc_get_link = proc_fd_link; + tid_fd_update_inode(task, inode, data->mode); d_set_d_op(dentry, &tid_fd_dentry_operations); - d_add(dentry, inode); - - /* Close the race of the process dying before we return the dentry */ - if (tid_fd_revalidate(dentry, 0)) - return 0; - out: - return -ENOENT; + return d_splice_alias(inode, dentry); } static struct dentry *proc_lookupfd_common(struct inode *dir, @@ -201,19 +208,21 @@ static struct dentry *proc_lookupfd_common(struct inode *dir, instantiate_t instantiate) { struct task_struct *task = get_proc_task(dir); - int result = -ENOENT; - unsigned fd = name_to_int(&dentry->d_name); + struct fd_data data = {.fd = name_to_int(&dentry->d_name)}; + struct dentry *result = ERR_PTR(-ENOENT); if (!task) goto out_no_task; - if (fd == ~0U) + if (data.fd == ~0U) + goto out; + if (!tid_fd_mode(task, data.fd, &data.mode)) goto out; - result = instantiate(dir, dentry, task, (void *)(unsigned long)fd); + result = instantiate(dentry, task, &data); out: put_task_struct(task); out_no_task: - return ERR_PTR(result); + return result; } static int proc_readfd_common(struct file *file, struct dir_context *ctx, @@ -236,17 +245,22 @@ static int proc_readfd_common(struct file *file, struct dir_context *ctx, for (fd = ctx->pos - 2; fd < files_fdtable(files)->max_fds; fd++, ctx->pos++) { + struct file *f; + struct fd_data data; char name[10 + 1]; int len; - if (!fcheck_files(files, fd)) + f = fcheck_files(files, fd); + if (!f) continue; + data.mode = f->f_mode; rcu_read_unlock(); + data.fd = fd; len = snprintf(name, sizeof(name), "%u", fd); if (!proc_fill_cache(file, ctx, name, len, instantiate, p, - (void *)(unsigned long)fd)) + &data)) goto out_fd_loop; cond_resched(); rcu_read_lock(); @@ -304,31 +318,25 @@ const struct inode_operations proc_fd_inode_operations = { .setattr = proc_setattr, }; -static int -proc_fdinfo_instantiate(struct inode *dir, struct dentry *dentry, - struct task_struct *task, const void *ptr) +static struct dentry *proc_fdinfo_instantiate(struct dentry *dentry, + struct task_struct *task, const void *ptr) { - unsigned fd = (unsigned long)ptr; + const struct fd_data *data = ptr; struct proc_inode *ei; struct inode *inode; - inode = proc_pid_make_inode(dir->i_sb, task, S_IFREG | S_IRUSR); + inode = proc_pid_make_inode(dentry->d_sb, task, S_IFREG | S_IRUSR); if (!inode) - goto out; + return ERR_PTR(-ENOENT); ei = PROC_I(inode); - ei->fd = fd; + ei->fd = data->fd; inode->i_fop = &proc_fdinfo_file_operations; + tid_fd_update_inode(task, inode, 0); d_set_d_op(dentry, &tid_fd_dentry_operations); - d_add(dentry, inode); - - /* Close the race of the process dying before we return the dentry */ - if (tid_fd_revalidate(dentry, 0)) - return 0; - out: - return -ENOENT; + return d_splice_alias(inode, dentry); } static struct dentry * diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 5d709fa8f3a2..7b4d9714f248 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -8,12 +8,14 @@ * Copyright (C) 1997 Theodore Ts'o */ +#include <linux/cache.h> #include <linux/errno.h> #include <linux/time.h> #include <linux/proc_fs.h> #include <linux/stat.h> #include <linux/mm.h> #include <linux/module.h> +#include <linux/namei.h> #include <linux/slab.h> #include <linux/printk.h> #include <linux/mount.h> @@ -23,11 +25,23 @@ #include <linux/spinlock.h> #include <linux/completion.h> #include <linux/uaccess.h> +#include <linux/seq_file.h> #include "internal.h" static DEFINE_RWLOCK(proc_subdir_lock); +struct kmem_cache *proc_dir_entry_cache __ro_after_init; + +void pde_free(struct proc_dir_entry *pde) +{ + if (S_ISLNK(pde->mode)) + kfree(pde->data); + if (pde->name != pde->inline_name) + kfree(pde->name); + kmem_cache_free(proc_dir_entry_cache, pde); +} + static int proc_match(const char *name, struct proc_dir_entry *de, unsigned int len) { if (len < de->namelen) @@ -40,8 +54,8 @@ static int proc_match(const char *name, struct proc_dir_entry *de, unsigned int static struct proc_dir_entry *pde_subdir_first(struct proc_dir_entry *dir) { - return rb_entry_safe(rb_first_cached(&dir->subdir), - struct proc_dir_entry, subdir_node); + return rb_entry_safe(rb_first(&dir->subdir), struct proc_dir_entry, + subdir_node); } static struct proc_dir_entry *pde_subdir_next(struct proc_dir_entry *dir) @@ -54,7 +68,7 @@ static struct proc_dir_entry *pde_subdir_find(struct proc_dir_entry *dir, const char *name, unsigned int len) { - struct rb_node *node = dir->subdir.rb_root.rb_node; + struct rb_node *node = dir->subdir.rb_node; while (node) { struct proc_dir_entry *de = rb_entry(node, @@ -75,9 +89,8 @@ static struct proc_dir_entry *pde_subdir_find(struct proc_dir_entry *dir, static bool pde_subdir_insert(struct proc_dir_entry *dir, struct proc_dir_entry *de) { - struct rb_root_cached *root = &dir->subdir; - struct rb_node **new = &root->rb_root.rb_node, *parent = NULL; - bool leftmost = true; + struct rb_root *root = &dir->subdir; + struct rb_node **new = &root->rb_node, *parent = NULL; /* Figure out where to put new node */ while (*new) { @@ -89,16 +102,15 @@ static bool pde_subdir_insert(struct proc_dir_entry *dir, parent = *new; if (result < 0) new = &(*new)->rb_left; - else if (result > 0) { + else if (result > 0) new = &(*new)->rb_right; - leftmost = false; - } else + else return false; } /* Add new node and rebalance tree. */ rb_link_node(&de->subdir_node, parent, new); - rb_insert_color_cached(&de->subdir_node, root, leftmost); + rb_insert_color(&de->subdir_node, root); return true; } @@ -207,6 +219,26 @@ void proc_free_inum(unsigned int inum) ida_simple_remove(&proc_inum_ida, inum - PROC_DYNAMIC_FIRST); } +static int proc_misc_d_revalidate(struct dentry *dentry, unsigned int flags) +{ + if (flags & LOOKUP_RCU) + return -ECHILD; + + if (atomic_read(&PDE(d_inode(dentry))->in_use) < 0) + return 0; /* revalidate */ + return 1; +} + +static int proc_misc_d_delete(const struct dentry *dentry) +{ + return atomic_read(&PDE(d_inode(dentry))->in_use) < 0; +} + +static const struct dentry_operations proc_misc_dentry_ops = { + .d_revalidate = proc_misc_d_revalidate, + .d_delete = proc_misc_d_delete, +}; + /* * Don't create negative dentries here, return -ENOENT by hand * instead. @@ -224,9 +256,8 @@ struct dentry *proc_lookup_de(struct inode *dir, struct dentry *dentry, inode = proc_get_inode(dir->i_sb, de); if (!inode) return ERR_PTR(-ENOMEM); - d_set_d_op(dentry, &simple_dentry_operations); - d_add(dentry, inode); - return NULL; + d_set_d_op(dentry, &proc_misc_dentry_ops); + return d_splice_alias(inode, dentry); } read_unlock(&proc_subdir_lock); return ERR_PTR(-ENOENT); @@ -315,13 +346,12 @@ static const struct inode_operations proc_dir_inode_operations = { .setattr = proc_notify_change, }; -static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp) +/* returns the registered entry, or frees dp and returns NULL on failure */ +struct proc_dir_entry *proc_register(struct proc_dir_entry *dir, + struct proc_dir_entry *dp) { - int ret; - - ret = proc_alloc_inum(&dp->low_ino); - if (ret) - return ret; + if (proc_alloc_inum(&dp->low_ino)) + goto out_free_entry; write_lock(&proc_subdir_lock); dp->parent = dir; @@ -329,12 +359,16 @@ static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp WARN(1, "proc_dir_entry '%s/%s' already registered\n", dir->name, dp->name); write_unlock(&proc_subdir_lock); - proc_free_inum(dp->low_ino); - return -EEXIST; + goto out_free_inum; } write_unlock(&proc_subdir_lock); - return 0; + return dp; +out_free_inum: + proc_free_inum(dp->low_ino); +out_free_entry: + pde_free(dp); + return NULL; } static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, @@ -354,6 +388,14 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, WARN(1, "name len %u\n", qstr.len); return NULL; } + if (qstr.len == 1 && fn[0] == '.') { + WARN(1, "name '.'\n"); + return NULL; + } + if (qstr.len == 2 && fn[0] == '.' && fn[1] == '.') { + WARN(1, "name '..'\n"); + return NULL; + } if (*parent == &proc_root && name_to_int(&qstr) != ~0U) { WARN(1, "create '/proc/%s' by hand\n", qstr.name); return NULL; @@ -363,16 +405,26 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent, return NULL; } - ent = kzalloc(sizeof(struct proc_dir_entry) + qstr.len + 1, GFP_KERNEL); + ent = kmem_cache_zalloc(proc_dir_entry_cache, GFP_KERNEL); if (!ent) goto out; + if (qstr.len + 1 <= sizeof(ent->inline_name)) { + ent->name = ent->inline_name; + } else { + ent->name = kmalloc(qstr.len + 1, GFP_KERNEL); + if (!ent->name) { + pde_free(ent); + return NULL; + } + } + memcpy(ent->name, fn, qstr.len + 1); ent->namelen = qstr.len; ent->mode = mode; ent->nlink = nlink; - ent->subdir = RB_ROOT_CACHED; - atomic_set(&ent->count, 1); + ent->subdir = RB_ROOT; + refcount_set(&ent->refcnt, 1); spin_lock_init(&ent->pde_unload_lock); INIT_LIST_HEAD(&ent->pde_openers); proc_set_user(ent, (*parent)->uid, (*parent)->gid); @@ -394,13 +446,9 @@ struct proc_dir_entry *proc_symlink(const char *name, if (ent->data) { strcpy((char*)ent->data,dest); ent->proc_iops = &proc_link_inode_operations; - if (proc_register(parent, ent) < 0) { - kfree(ent->data); - kfree(ent); - ent = NULL; - } + ent = proc_register(parent, ent); } else { - kfree(ent); + pde_free(ent); ent = NULL; } } @@ -422,11 +470,9 @@ struct proc_dir_entry *proc_mkdir_data(const char *name, umode_t mode, ent->proc_fops = &proc_dir_operations; ent->proc_iops = &proc_dir_inode_operations; parent->nlink++; - if (proc_register(parent, ent) < 0) { - kfree(ent); + ent = proc_register(parent, ent); + if (!ent) parent->nlink--; - ent = NULL; - } } return ent; } @@ -457,47 +503,47 @@ struct proc_dir_entry *proc_create_mount_point(const char *name) ent->proc_fops = NULL; ent->proc_iops = NULL; parent->nlink++; - if (proc_register(parent, ent) < 0) { - kfree(ent); + ent = proc_register(parent, ent); + if (!ent) parent->nlink--; - ent = NULL; - } } return ent; } EXPORT_SYMBOL(proc_create_mount_point); -struct proc_dir_entry *proc_create_data(const char *name, umode_t mode, - struct proc_dir_entry *parent, - const struct file_operations *proc_fops, - void *data) +struct proc_dir_entry *proc_create_reg(const char *name, umode_t mode, + struct proc_dir_entry **parent, void *data) { - struct proc_dir_entry *pde; + struct proc_dir_entry *p; + if ((mode & S_IFMT) == 0) mode |= S_IFREG; - - if (!S_ISREG(mode)) { - WARN_ON(1); /* use proc_mkdir() */ + if ((mode & S_IALLUGO) == 0) + mode |= S_IRUGO; + if (WARN_ON_ONCE(!S_ISREG(mode))) return NULL; + + p = __proc_create(parent, name, mode, 1); + if (p) { + p->proc_iops = &proc_file_inode_operations; + p->data = data; } + return p; +} + +struct proc_dir_entry *proc_create_data(const char *name, umode_t mode, + struct proc_dir_entry *parent, + const struct file_operations *proc_fops, void *data) +{ + struct proc_dir_entry *p; BUG_ON(proc_fops == NULL); - if ((mode & S_IALLUGO) == 0) - mode |= S_IRUGO; - pde = __proc_create(&parent, name, mode, 1); - if (!pde) - goto out; - pde->proc_fops = proc_fops; - pde->data = data; - pde->proc_iops = &proc_file_inode_operations; - if (proc_register(parent, pde) < 0) - goto out_free; - return pde; -out_free: - kfree(pde); -out: - return NULL; + p = proc_create_reg(name, mode, &parent, data); + if (!p) + return NULL; + p->proc_fops = proc_fops; + return proc_register(parent, p); } EXPORT_SYMBOL(proc_create_data); @@ -509,6 +555,67 @@ struct proc_dir_entry *proc_create(const char *name, umode_t mode, } EXPORT_SYMBOL(proc_create); +static int proc_seq_open(struct inode *inode, struct file *file) +{ + struct proc_dir_entry *de = PDE(inode); + + if (de->state_size) + return seq_open_private(file, de->seq_ops, de->state_size); + return seq_open(file, de->seq_ops); +} + +static const struct file_operations proc_seq_fops = { + .open = proc_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +struct proc_dir_entry *proc_create_seq_private(const char *name, umode_t mode, + struct proc_dir_entry *parent, const struct seq_operations *ops, + unsigned int state_size, void *data) +{ + struct proc_dir_entry *p; + + p = proc_create_reg(name, mode, &parent, data); + if (!p) + return NULL; + p->proc_fops = &proc_seq_fops; + p->seq_ops = ops; + p->state_size = state_size; + return proc_register(parent, p); +} +EXPORT_SYMBOL(proc_create_seq_private); + +static int proc_single_open(struct inode *inode, struct file *file) +{ + struct proc_dir_entry *de = PDE(inode); + + return single_open(file, de->single_show, de->data); +} + +static const struct file_operations proc_single_fops = { + .open = proc_single_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +struct proc_dir_entry *proc_create_single_data(const char *name, umode_t mode, + struct proc_dir_entry *parent, + int (*show)(struct seq_file *, void *), void *data) +{ + struct proc_dir_entry *p; + + p = proc_create_reg(name, mode, &parent, data); + if (!p) + return NULL; + p->proc_fops = &proc_single_fops; + p->single_show = show; + return proc_register(parent, p); +} +EXPORT_SYMBOL(proc_create_single_data); + void proc_set_size(struct proc_dir_entry *de, loff_t size) { de->size = size; @@ -522,19 +629,12 @@ void proc_set_user(struct proc_dir_entry *de, kuid_t uid, kgid_t gid) } EXPORT_SYMBOL(proc_set_user); -static void free_proc_entry(struct proc_dir_entry *de) -{ - proc_free_inum(de->low_ino); - - if (S_ISLNK(de->mode)) - kfree(de->data); - kfree(de); -} - void pde_put(struct proc_dir_entry *pde) { - if (atomic_dec_and_test(&pde->count)) - free_proc_entry(pde); + if (refcount_dec_and_test(&pde->refcnt)) { + proc_free_inum(pde->low_ino); + pde_free(pde); + } } /* @@ -555,7 +655,7 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent) de = pde_subdir_find(parent, fn, len); if (de) - rb_erase_cached(&de->subdir_node, &parent->subdir); + rb_erase(&de->subdir_node, &parent->subdir); write_unlock(&proc_subdir_lock); if (!de) { WARN(1, "name '%s'\n", name); @@ -592,13 +692,13 @@ int remove_proc_subtree(const char *name, struct proc_dir_entry *parent) write_unlock(&proc_subdir_lock); return -ENOENT; } - rb_erase_cached(&root->subdir_node, &parent->subdir); + rb_erase(&root->subdir_node, &parent->subdir); de = root; while (1) { next = pde_subdir_first(de); if (next) { - rb_erase_cached(&next->subdir_node, &de->subdir); + rb_erase(&next->subdir_node, &de->subdir); de = next; continue; } diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 6e8724958116..2cf3b74391ca 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -54,6 +54,7 @@ static void proc_evict_inode(struct inode *inode) } static struct kmem_cache *proc_inode_cachep __ro_after_init; +static struct kmem_cache *pde_opener_cache __ro_after_init; static struct inode *proc_alloc_inode(struct super_block *sb) { @@ -92,7 +93,7 @@ static void init_once(void *foo) inode_init_once(&ei->vfs_inode); } -void __init proc_init_inodecache(void) +void __init proc_init_kmemcache(void) { proc_inode_cachep = kmem_cache_create("proc_inode_cache", sizeof(struct proc_inode), @@ -100,6 +101,13 @@ void __init proc_init_inodecache(void) SLAB_MEM_SPREAD|SLAB_ACCOUNT| SLAB_PANIC), init_once); + pde_opener_cache = + kmem_cache_create("pde_opener", sizeof(struct pde_opener), 0, + SLAB_ACCOUNT|SLAB_PANIC, NULL); + proc_dir_entry_cache = kmem_cache_create_usercopy( + "proc_dir_entry", sizeof(struct proc_dir_entry), 0, SLAB_PANIC, + offsetof(struct proc_dir_entry, inline_name), + sizeof_field(struct proc_dir_entry, inline_name), NULL); } static int proc_show_options(struct seq_file *seq, struct dentry *root) @@ -138,7 +146,7 @@ static void unuse_pde(struct proc_dir_entry *pde) complete(pde->pde_unload_completion); } -/* pde is locked */ +/* pde is locked on entry, unlocked on exit */ static void close_pdeo(struct proc_dir_entry *pde, struct pde_opener *pdeo) { /* @@ -157,9 +165,10 @@ static void close_pdeo(struct proc_dir_entry *pde, struct pde_opener *pdeo) pdeo->c = &c; spin_unlock(&pde->pde_unload_lock); wait_for_completion(&c); - spin_lock(&pde->pde_unload_lock); } else { struct file *file; + struct completion *c; + pdeo->closing = true; spin_unlock(&pde->pde_unload_lock); file = pdeo->file; @@ -167,9 +176,11 @@ static void close_pdeo(struct proc_dir_entry *pde, struct pde_opener *pdeo) spin_lock(&pde->pde_unload_lock); /* After ->release. */ list_del(&pdeo->lh); - if (unlikely(pdeo->c)) - complete(pdeo->c); - kfree(pdeo); + c = pdeo->c; + spin_unlock(&pde->pde_unload_lock); + if (unlikely(c)) + complete(c); + kmem_cache_free(pde_opener_cache, pdeo); } } @@ -188,6 +199,7 @@ void proc_entry_rundown(struct proc_dir_entry *de) struct pde_opener *pdeo; pdeo = list_first_entry(&de->pde_openers, struct pde_opener, lh); close_pdeo(de, pdeo); + spin_lock(&de->pde_unload_lock); } spin_unlock(&de->pde_unload_lock); } @@ -338,31 +350,36 @@ static int proc_reg_open(struct inode *inode, struct file *file) * * Save every "struct file" with custom ->release hook. */ - pdeo = kmalloc(sizeof(struct pde_opener), GFP_KERNEL); - if (!pdeo) - return -ENOMEM; - - if (!use_pde(pde)) { - kfree(pdeo); + if (!use_pde(pde)) return -ENOENT; - } - open = pde->proc_fops->open; + release = pde->proc_fops->release; + if (release) { + pdeo = kmem_cache_alloc(pde_opener_cache, GFP_KERNEL); + if (!pdeo) { + rv = -ENOMEM; + goto out_unuse; + } + } + open = pde->proc_fops->open; if (open) rv = open(inode, file); - if (rv == 0 && release) { - /* To know what to release. */ - pdeo->file = file; - pdeo->closing = false; - pdeo->c = NULL; - spin_lock(&pde->pde_unload_lock); - list_add(&pdeo->lh, &pde->pde_openers); - spin_unlock(&pde->pde_unload_lock); - } else - kfree(pdeo); + if (release) { + if (rv == 0) { + /* To know what to release. */ + pdeo->file = file; + pdeo->closing = false; + pdeo->c = NULL; + spin_lock(&pde->pde_unload_lock); + list_add(&pdeo->lh, &pde->pde_openers); + spin_unlock(&pde->pde_unload_lock); + } else + kmem_cache_free(pde_opener_cache, pdeo); + } +out_unuse: unuse_pde(pde); return rv; } @@ -375,7 +392,7 @@ static int proc_reg_release(struct inode *inode, struct file *file) list_for_each_entry(pdeo, &pde->pde_openers, lh) { if (pdeo->file == file) { close_pdeo(pde, pdeo); - break; + return 0; } } spin_unlock(&pde->pde_unload_lock); diff --git a/fs/proc/internal.h b/fs/proc/internal.h index d697c8ab0a14..93eb1906c28d 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -11,6 +11,7 @@ #include <linux/proc_fs.h> #include <linux/proc_ns.h> +#include <linux/refcount.h> #include <linux/spinlock.h> #include <linux/atomic.h> #include <linux/binfmts.h> @@ -36,27 +37,41 @@ struct proc_dir_entry { * negative -> it's going away RSN */ atomic_t in_use; - atomic_t count; /* use count */ + refcount_t refcnt; struct list_head pde_openers; /* who did ->open, but not ->release */ /* protects ->pde_openers and all struct pde_opener instances */ spinlock_t pde_unload_lock; struct completion *pde_unload_completion; const struct inode_operations *proc_iops; const struct file_operations *proc_fops; + union { + const struct seq_operations *seq_ops; + int (*single_show)(struct seq_file *, void *); + }; void *data; + unsigned int state_size; unsigned int low_ino; nlink_t nlink; kuid_t uid; kgid_t gid; loff_t size; struct proc_dir_entry *parent; - struct rb_root_cached subdir; + struct rb_root subdir; struct rb_node subdir_node; + char *name; umode_t mode; u8 namelen; - char name[]; +#ifdef CONFIG_64BIT +#define SIZEOF_PDE_INLINE_NAME (192-155) +#else +#define SIZEOF_PDE_INLINE_NAME (128-95) +#endif + char inline_name[SIZEOF_PDE_INLINE_NAME]; } __randomize_layout; +extern struct kmem_cache *proc_dir_entry_cache; +void pde_free(struct proc_dir_entry *pde); + union proc_op { int (*proc_get_link)(struct dentry *, struct path *); int (*proc_show)(struct seq_file *m, @@ -121,6 +136,8 @@ unsigned name_to_int(const struct qstr *qstr); */ extern const struct file_operations proc_tid_children_operations; +extern void proc_task_name(struct seq_file *m, struct task_struct *p, + bool escape); extern int proc_tid_stat(struct seq_file *, struct pid_namespace *, struct pid *, struct task_struct *); extern int proc_tgid_stat(struct seq_file *, struct pid_namespace *, @@ -137,14 +154,14 @@ extern const struct dentry_operations pid_dentry_operations; extern int pid_getattr(const struct path *, struct kstat *, u32, unsigned int); extern int proc_setattr(struct dentry *, struct iattr *); extern struct inode *proc_pid_make_inode(struct super_block *, struct task_struct *, umode_t); -extern int pid_revalidate(struct dentry *, unsigned int); +extern void pid_update_inode(struct task_struct *, struct inode *); extern int pid_delete_dentry(const struct dentry *); extern int proc_pid_readdir(struct file *, struct dir_context *); extern struct dentry *proc_pid_lookup(struct inode *, struct dentry *, unsigned int); extern loff_t mem_lseek(struct file *, loff_t, int); /* Lookups */ -typedef int instantiate_t(struct inode *, struct dentry *, +typedef struct dentry *instantiate_t(struct dentry *, struct task_struct *, const void *); extern bool proc_fill_cache(struct file *, struct dir_context *, const char *, int, instantiate_t, struct task_struct *, const void *); @@ -152,6 +169,10 @@ extern bool proc_fill_cache(struct file *, struct dir_context *, const char *, i /* * generic.c */ +struct proc_dir_entry *proc_create_reg(const char *name, umode_t mode, + struct proc_dir_entry **parent, void *data); +struct proc_dir_entry *proc_register(struct proc_dir_entry *dir, + struct proc_dir_entry *dp); extern struct dentry *proc_lookup(struct inode *, struct dentry *, unsigned int); struct dentry *proc_lookup_de(struct inode *, struct dentry *, struct proc_dir_entry *); extern int proc_readdir(struct file *, struct dir_context *); @@ -159,7 +180,7 @@ int proc_readdir_de(struct file *, struct dir_context *, struct proc_dir_entry * static inline struct proc_dir_entry *pde_get(struct proc_dir_entry *pde) { - atomic_inc(&pde->count); + refcount_inc(&pde->refcnt); return pde; } extern void pde_put(struct proc_dir_entry *); @@ -177,12 +198,12 @@ struct pde_opener { struct list_head lh; bool closing; struct completion *c; -}; +} __randomize_layout; extern const struct inode_operations proc_link_inode_operations; extern const struct inode_operations proc_pid_link_inode_operations; -extern void proc_init_inodecache(void); +void proc_init_kmemcache(void); void set_proc_pid_nlink(void); extern struct inode *proc_get_inode(struct super_block *, struct proc_dir_entry *); extern int proc_fill_super(struct super_block *, void *data, int flags); diff --git a/fs/proc/interrupts.c b/fs/proc/interrupts.c index 6a6bee9c603c..cb0edc7cbf09 100644 --- a/fs/proc/interrupts.c +++ b/fs/proc/interrupts.c @@ -34,21 +34,9 @@ static const struct seq_operations int_seq_ops = { .show = show_interrupts }; -static int interrupts_open(struct inode *inode, struct file *filp) -{ - return seq_open(filp, &int_seq_ops); -} - -static const struct file_operations proc_interrupts_operations = { - .open = interrupts_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - static int __init proc_interrupts_init(void) { - proc_create("interrupts", 0, NULL, &proc_interrupts_operations); + proc_create_seq("interrupts", 0, NULL, &int_seq_ops); return 0; } fs_initcall(proc_interrupts_init); diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index d1e82761de81..e64ecb9f2720 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -209,25 +209,34 @@ kclist_add_private(unsigned long pfn, unsigned long nr_pages, void *arg) { struct list_head *head = (struct list_head *)arg; struct kcore_list *ent; + struct page *p; + + if (!pfn_valid(pfn)) + return 1; + + p = pfn_to_page(pfn); + if (!memmap_valid_within(pfn, p, page_zone(p))) + return 1; ent = kmalloc(sizeof(*ent), GFP_KERNEL); if (!ent) return -ENOMEM; - ent->addr = (unsigned long)__va((pfn << PAGE_SHIFT)); + ent->addr = (unsigned long)page_to_virt(p); ent->size = nr_pages << PAGE_SHIFT; - /* Sanity check: Can happen in 32bit arch...maybe */ - if (ent->addr < (unsigned long) __va(0)) + if (!virt_addr_valid(ent->addr)) goto free_out; /* cut not-mapped area. ....from ppc-32 code. */ if (ULONG_MAX - ent->addr < ent->size) ent->size = ULONG_MAX - ent->addr; - /* cut when vmalloc() area is higher than direct-map area */ - if (VMALLOC_START > (unsigned long)__va(0)) { - if (ent->addr > VMALLOC_START) - goto free_out; + /* + * We've already checked virt_addr_valid so we know this address + * is a valid pointer, therefore we can check against it to determine + * if we need to trim + */ + if (VMALLOC_START > ent->addr) { if (VMALLOC_START - ent->addr < ent->size) ent->size = VMALLOC_START - ent->addr; } diff --git a/fs/proc/loadavg.c b/fs/proc/loadavg.c index a000d7547479..d06694757201 100644 --- a/fs/proc/loadavg.c +++ b/fs/proc/loadavg.c @@ -24,25 +24,13 @@ static int loadavg_proc_show(struct seq_file *m, void *v) LOAD_INT(avnrun[1]), LOAD_FRAC(avnrun[1]), LOAD_INT(avnrun[2]), LOAD_FRAC(avnrun[2]), nr_running(), nr_threads, - idr_get_cursor(&task_active_pid_ns(current)->idr)); + idr_get_cursor(&task_active_pid_ns(current)->idr) - 1); return 0; } -static int loadavg_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, loadavg_proc_show, NULL); -} - -static const struct file_operations loadavg_proc_fops = { - .open = loadavg_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_loadavg_init(void) { - proc_create("loadavg", 0, NULL, &loadavg_proc_fops); + proc_create_single("loadavg", 0, NULL, loadavg_proc_show); return 0; } fs_initcall(proc_loadavg_init); diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c index 6bb20f864259..2fb04846ed11 100644 --- a/fs/proc/meminfo.c +++ b/fs/proc/meminfo.c @@ -26,20 +26,7 @@ void __attribute__((weak)) arch_report_meminfo(struct seq_file *m) static void show_val_kb(struct seq_file *m, const char *s, unsigned long num) { - char v[32]; - static const char blanks[7] = {' ', ' ', ' ', ' ',' ', ' ', ' '}; - int len; - - len = num_to_str(v, sizeof(v), num << (PAGE_SHIFT - 10)); - - seq_write(m, s, 16); - - if (len > 0) { - if (len < 8) - seq_write(m, blanks, 8 - len); - - seq_write(m, v, len); - } + seq_put_decimal_ull_width(m, s, num << (PAGE_SHIFT - 10), 8); seq_write(m, " kB\n", 4); } @@ -162,21 +149,9 @@ static int meminfo_proc_show(struct seq_file *m, void *v) return 0; } -static int meminfo_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, meminfo_proc_show, NULL); -} - -static const struct file_operations meminfo_proc_fops = { - .open = meminfo_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_meminfo_init(void) { - proc_create("meminfo", 0, NULL, &meminfo_proc_fops); + proc_create_single("meminfo", 0, NULL, meminfo_proc_show); return 0; } fs_initcall(proc_meminfo_init); diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index 59b17e509f46..dd2b35f78b09 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -87,28 +87,24 @@ static const struct inode_operations proc_ns_link_inode_operations = { .setattr = proc_setattr, }; -static int proc_ns_instantiate(struct inode *dir, - struct dentry *dentry, struct task_struct *task, const void *ptr) +static struct dentry *proc_ns_instantiate(struct dentry *dentry, + struct task_struct *task, const void *ptr) { const struct proc_ns_operations *ns_ops = ptr; struct inode *inode; struct proc_inode *ei; - inode = proc_pid_make_inode(dir->i_sb, task, S_IFLNK | S_IRWXUGO); + inode = proc_pid_make_inode(dentry->d_sb, task, S_IFLNK | S_IRWXUGO); if (!inode) - goto out; + return ERR_PTR(-ENOENT); ei = PROC_I(inode); inode->i_op = &proc_ns_link_inode_operations; ei->ns_ops = ns_ops; + pid_update_inode(task, inode); d_set_d_op(dentry, &pid_dentry_operations); - d_add(dentry, inode); - /* Close the race of the process dying before we return the dentry */ - if (pid_revalidate(dentry, 0)) - return 0; -out: - return -ENOENT; + return d_splice_alias(inode, dentry); } static int proc_ns_dir_readdir(struct file *file, struct dir_context *ctx) @@ -147,12 +143,10 @@ const struct file_operations proc_ns_dir_operations = { static struct dentry *proc_ns_dir_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { - int error; struct task_struct *task = get_proc_task(dir); const struct proc_ns_operations **entry, **last; unsigned int len = dentry->d_name.len; - - error = -ENOENT; + struct dentry *res = ERR_PTR(-ENOENT); if (!task) goto out_no_task; @@ -167,11 +161,11 @@ static struct dentry *proc_ns_dir_lookup(struct inode *dir, if (entry == last) goto out; - error = proc_ns_instantiate(dir, dentry, task, *entry); + res = proc_ns_instantiate(dentry, task, *entry); out: put_task_struct(task); out_no_task: - return ERR_PTR(error); + return res; } const struct inode_operations proc_ns_dir_inode_operations = { diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c index 75634379f82e..3b63be64e436 100644 --- a/fs/proc/nommu.c +++ b/fs/proc/nommu.c @@ -113,21 +113,9 @@ static const struct seq_operations proc_nommu_region_list_seqop = { .show = nommu_region_list_show }; -static int proc_nommu_region_list_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &proc_nommu_region_list_seqop); -} - -static const struct file_operations proc_nommu_region_list_operations = { - .open = proc_nommu_region_list_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - static int __init proc_nommu_init(void) { - proc_create("maps", S_IRUGO, NULL, &proc_nommu_region_list_operations); + proc_create_seq("maps", S_IRUGO, NULL, &proc_nommu_region_list_seqop); return 0; } diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 68c06ae7888c..7d94fa005b0d 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -38,20 +38,20 @@ static struct net *get_proc_net(const struct inode *inode) return maybe_get_net(PDE_NET(PDE(inode))); } -int seq_open_net(struct inode *ino, struct file *f, - const struct seq_operations *ops, int size) +static int seq_open_net(struct inode *inode, struct file *file) { - struct net *net; + unsigned int state_size = PDE(inode)->state_size; struct seq_net_private *p; + struct net *net; - BUG_ON(size < sizeof(*p)); + WARN_ON_ONCE(state_size < sizeof(*p)); - net = get_proc_net(ino); - if (net == NULL) + net = get_proc_net(inode); + if (!net) return -ENXIO; - p = __seq_open_private(f, ops, size); - if (p == NULL) { + p = __seq_open_private(file, PDE(inode)->seq_ops, state_size); + if (!p) { put_net(net); return -ENOMEM; } @@ -60,51 +60,83 @@ int seq_open_net(struct inode *ino, struct file *f, #endif return 0; } -EXPORT_SYMBOL_GPL(seq_open_net); -int single_open_net(struct inode *inode, struct file *file, - int (*show)(struct seq_file *, void *)) +static int seq_release_net(struct inode *ino, struct file *f) { - int err; - struct net *net; - - err = -ENXIO; - net = get_proc_net(inode); - if (net == NULL) - goto err_net; - - err = single_open(file, show, net); - if (err < 0) - goto err_open; + struct seq_file *seq = f->private_data; + put_net(seq_file_net(seq)); + seq_release_private(ino, f); return 0; +} -err_open: - put_net(net); -err_net: - return err; +static const struct file_operations proc_net_seq_fops = { + .open = seq_open_net, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_net, +}; + +struct proc_dir_entry *proc_create_net_data(const char *name, umode_t mode, + struct proc_dir_entry *parent, const struct seq_operations *ops, + unsigned int state_size, void *data) +{ + struct proc_dir_entry *p; + + p = proc_create_reg(name, mode, &parent, data); + if (!p) + return NULL; + p->proc_fops = &proc_net_seq_fops; + p->seq_ops = ops; + p->state_size = state_size; + return proc_register(parent, p); } -EXPORT_SYMBOL_GPL(single_open_net); +EXPORT_SYMBOL_GPL(proc_create_net_data); -int seq_release_net(struct inode *ino, struct file *f) +static int single_open_net(struct inode *inode, struct file *file) { - struct seq_file *seq; + struct proc_dir_entry *de = PDE(inode); + struct net *net; + int err; - seq = f->private_data; + net = get_proc_net(inode); + if (!net) + return -ENXIO; - put_net(seq_file_net(seq)); - seq_release_private(ino, f); - return 0; + err = single_open(file, de->single_show, net); + if (err) + put_net(net); + return err; } -EXPORT_SYMBOL_GPL(seq_release_net); -int single_release_net(struct inode *ino, struct file *f) +static int single_release_net(struct inode *ino, struct file *f) { struct seq_file *seq = f->private_data; put_net(seq->private); return single_release(ino, f); } -EXPORT_SYMBOL_GPL(single_release_net); + +static const struct file_operations proc_net_single_fops = { + .open = single_open_net, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release_net, +}; + +struct proc_dir_entry *proc_create_net_single(const char *name, umode_t mode, + struct proc_dir_entry *parent, + int (*show)(struct seq_file *, void *), void *data) +{ + struct proc_dir_entry *p; + + p = proc_create_reg(name, mode, &parent, data); + if (!p) + return NULL; + p->proc_fops = &proc_net_single_fops; + p->single_show = show; + return proc_register(parent, p); +} +EXPORT_SYMBOL_GPL(proc_create_net_single); static struct net *get_proc_task_net(struct inode *dir) { @@ -192,15 +224,16 @@ static __net_init int proc_net_ns_init(struct net *net) int err; err = -ENOMEM; - netd = kzalloc(sizeof(*netd) + 4, GFP_KERNEL); + netd = kmem_cache_zalloc(proc_dir_entry_cache, GFP_KERNEL); if (!netd) goto out; - netd->subdir = RB_ROOT_CACHED; + netd->subdir = RB_ROOT; netd->data = net; netd->nlink = 2; netd->namelen = 3; netd->parent = &proc_root; + netd->name = netd->inline_name; memcpy(netd->name, "net", 4); uid = make_kuid(net->user_ns, 0); @@ -223,7 +256,7 @@ static __net_init int proc_net_ns_init(struct net *net) return 0; free_net: - kfree(netd); + pde_free(netd); out: return err; } @@ -231,7 +264,7 @@ out: static __net_exit void proc_net_ns_exit(struct net *net) { remove_proc_entry("stat", net->proc_net); - kfree(net->proc_net); + pde_free(net->proc_net); } static struct pernet_operations __net_initdata proc_net_ns_ops = { diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index c41ab261397d..4d765e5e91ed 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -554,9 +554,8 @@ static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry, if (!inode) goto out; - err = NULL; d_set_d_op(dentry, &proc_sys_dentry_operations); - d_add(dentry, inode); + err = d_splice_alias(inode, dentry); out: if (h) @@ -684,6 +683,7 @@ static bool proc_sys_fill_cache(struct file *file, if (IS_ERR(child)) return false; if (d_in_lookup(child)) { + struct dentry *res; inode = proc_sys_make_inode(dir->d_sb, head, table); if (!inode) { d_lookup_done(child); @@ -691,7 +691,16 @@ static bool proc_sys_fill_cache(struct file *file, return false; } d_set_d_op(child, &proc_sys_dentry_operations); - d_add(child, inode); + res = d_splice_alias(inode, child); + d_lookup_done(child); + if (unlikely(res)) { + if (IS_ERR(res)) { + dput(child); + return false; + } + dput(child); + child = res; + } } } inode = d_inode(child); @@ -707,14 +716,14 @@ static bool proc_sys_link_fill_cache(struct file *file, struct ctl_table *table) { bool ret = true; + head = sysctl_head_grab(head); + if (IS_ERR(head)) + return false; - if (S_ISLNK(table->mode)) { - /* It is not an error if we can not follow the link ignore it */ - int err = sysctl_follow_link(&head, &table); - if (err) - goto out; - } + /* It is not an error if we can not follow the link ignore it */ + if (sysctl_follow_link(&head, &table)) + goto out; ret = proc_sys_fill_cache(file, ctx, head, table); out: @@ -1086,7 +1095,7 @@ static int sysctl_check_table_array(const char *path, struct ctl_table *table) if ((table->proc_handler == proc_douintvec) || (table->proc_handler == proc_douintvec_minmax)) { if (table->maxlen != sizeof(unsigned int)) - err |= sysctl_err(path, table, "array now allowed"); + err |= sysctl_err(path, table, "array not allowed"); } return err; diff --git a/fs/proc/proc_tty.c b/fs/proc/proc_tty.c index d0cf1c50bb6c..c69ff191e5d8 100644 --- a/fs/proc/proc_tty.c +++ b/fs/proc/proc_tty.c @@ -126,18 +126,6 @@ static const struct seq_operations tty_drivers_op = { .show = show_tty_driver }; -static int tty_drivers_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &tty_drivers_op); -} - -static const struct file_operations proc_tty_drivers_operations = { - .open = tty_drivers_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - /* * This function is called by tty_register_driver() to handle * registering the driver's /proc handler into /proc/tty/driver/<foo> @@ -147,11 +135,11 @@ void proc_tty_register_driver(struct tty_driver *driver) struct proc_dir_entry *ent; if (!driver->driver_name || driver->proc_entry || - !driver->ops->proc_fops) + !driver->ops->proc_show) return; - ent = proc_create_data(driver->driver_name, 0, proc_tty_driver, - driver->ops->proc_fops, driver); + ent = proc_create_single_data(driver->driver_name, 0, proc_tty_driver, + driver->ops->proc_show, driver); driver->proc_entry = ent; } @@ -186,6 +174,6 @@ void __init proc_tty_init(void) * entry. */ proc_tty_driver = proc_mkdir_mode("tty/driver", S_IRUSR|S_IXUSR, NULL); - proc_create("tty/ldiscs", 0, NULL, &tty_ldiscs_proc_fops); - proc_create("tty/drivers", 0, NULL, &proc_tty_drivers_operations); + proc_create_seq("tty/ldiscs", 0, NULL, &tty_ldiscs_seq_ops); + proc_create_seq("tty/drivers", 0, NULL, &tty_drivers_op); } diff --git a/fs/proc/root.c b/fs/proc/root.c index ede8e64974be..61b7340b357a 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -123,23 +123,13 @@ static struct file_system_type proc_fs_type = { void __init proc_root_init(void) { - int err; - - proc_init_inodecache(); + proc_init_kmemcache(); set_proc_pid_nlink(); - err = register_filesystem(&proc_fs_type); - if (err) - return; - proc_self_init(); proc_thread_self_init(); proc_symlink("mounts", NULL, "self/mounts"); proc_net_init(); - -#ifdef CONFIG_SYSVIPC - proc_mkdir("sysvipc", NULL); -#endif proc_mkdir("fs", NULL); proc_mkdir("driver", NULL); proc_create_mount_point("fs/nfsd"); /* somewhere for the nfsd filesystem to be mounted */ @@ -150,6 +140,8 @@ void __init proc_root_init(void) proc_tty_init(); proc_mkdir("bus", NULL); proc_sys_init(); + + register_filesystem(&proc_fs_type); } static int proc_root_getattr(const struct path *path, struct kstat *stat, @@ -207,12 +199,13 @@ struct proc_dir_entry proc_root = { .namelen = 5, .mode = S_IFDIR | S_IRUGO | S_IXUGO, .nlink = 2, - .count = ATOMIC_INIT(1), + .refcnt = REFCOUNT_INIT(1), .proc_iops = &proc_root_inode_operations, .proc_fops = &proc_root_operations, .parent = &proc_root, - .subdir = RB_ROOT_CACHED, - .name = "/proc", + .subdir = RB_ROOT, + .name = proc_root.inline_name, + .inline_name = "/proc", }; int pid_ns_prepare_proc(struct pid_namespace *ns) diff --git a/fs/proc/self.c b/fs/proc/self.c index 4d7d061696b3..127265e5c55f 100644 --- a/fs/proc/self.c +++ b/fs/proc/self.c @@ -12,7 +12,7 @@ static const char *proc_self_get_link(struct dentry *dentry, struct inode *inode, struct delayed_call *done) { - struct pid_namespace *ns = inode->i_sb->s_fs_info; + struct pid_namespace *ns = proc_pid_ns(inode); pid_t tgid = task_tgid_nr_ns(current, ns); char *name; @@ -36,7 +36,7 @@ static unsigned self_inum __ro_after_init; int proc_setup_self(struct super_block *s) { struct inode *root_inode = d_inode(s->s_root); - struct pid_namespace *ns = s->s_fs_info; + struct pid_namespace *ns = proc_pid_ns(root_inode); struct dentry *self; inode_lock(root_inode); diff --git a/fs/proc/softirqs.c b/fs/proc/softirqs.c index 24072cc06e65..12901dcf57e2 100644 --- a/fs/proc/softirqs.c +++ b/fs/proc/softirqs.c @@ -25,21 +25,9 @@ static int show_softirqs(struct seq_file *p, void *v) return 0; } -static int softirqs_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_softirqs, NULL); -} - -static const struct file_operations proc_softirqs_operations = { - .open = softirqs_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_softirqs_init(void) { - proc_create("softirqs", 0, NULL, &proc_softirqs_operations); + proc_create_single("softirqs", 0, NULL, show_softirqs); return 0; } fs_initcall(proc_softirqs_init); diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index ec6d2983a5cb..a20c6e495bb2 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -24,6 +24,8 @@ #include <asm/tlbflush.h> #include "internal.h" +#define SEQ_PUT_DEC(str, val) \ + seq_put_decimal_ull_width(m, str, (val) << (PAGE_SHIFT-10), 8) void task_mem(struct seq_file *m, struct mm_struct *mm) { unsigned long text, lib, swap, anon, file, shmem; @@ -53,39 +55,28 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) lib = (mm->exec_vm << PAGE_SHIFT) - text; swap = get_mm_counter(mm, MM_SWAPENTS); - seq_printf(m, - "VmPeak:\t%8lu kB\n" - "VmSize:\t%8lu kB\n" - "VmLck:\t%8lu kB\n" - "VmPin:\t%8lu kB\n" - "VmHWM:\t%8lu kB\n" - "VmRSS:\t%8lu kB\n" - "RssAnon:\t%8lu kB\n" - "RssFile:\t%8lu kB\n" - "RssShmem:\t%8lu kB\n" - "VmData:\t%8lu kB\n" - "VmStk:\t%8lu kB\n" - "VmExe:\t%8lu kB\n" - "VmLib:\t%8lu kB\n" - "VmPTE:\t%8lu kB\n" - "VmSwap:\t%8lu kB\n", - hiwater_vm << (PAGE_SHIFT-10), - total_vm << (PAGE_SHIFT-10), - mm->locked_vm << (PAGE_SHIFT-10), - mm->pinned_vm << (PAGE_SHIFT-10), - hiwater_rss << (PAGE_SHIFT-10), - total_rss << (PAGE_SHIFT-10), - anon << (PAGE_SHIFT-10), - file << (PAGE_SHIFT-10), - shmem << (PAGE_SHIFT-10), - mm->data_vm << (PAGE_SHIFT-10), - mm->stack_vm << (PAGE_SHIFT-10), - text >> 10, - lib >> 10, - mm_pgtables_bytes(mm) >> 10, - swap << (PAGE_SHIFT-10)); + SEQ_PUT_DEC("VmPeak:\t", hiwater_vm); + SEQ_PUT_DEC(" kB\nVmSize:\t", total_vm); + SEQ_PUT_DEC(" kB\nVmLck:\t", mm->locked_vm); + SEQ_PUT_DEC(" kB\nVmPin:\t", mm->pinned_vm); + SEQ_PUT_DEC(" kB\nVmHWM:\t", hiwater_rss); + SEQ_PUT_DEC(" kB\nVmRSS:\t", total_rss); + SEQ_PUT_DEC(" kB\nRssAnon:\t", anon); + SEQ_PUT_DEC(" kB\nRssFile:\t", file); + SEQ_PUT_DEC(" kB\nRssShmem:\t", shmem); + SEQ_PUT_DEC(" kB\nVmData:\t", mm->data_vm); + SEQ_PUT_DEC(" kB\nVmStk:\t", mm->stack_vm); + seq_put_decimal_ull_width(m, + " kB\nVmExe:\t", text >> 10, 8); + seq_put_decimal_ull_width(m, + " kB\nVmLib:\t", lib >> 10, 8); + seq_put_decimal_ull_width(m, + " kB\nVmPTE:\t", mm_pgtables_bytes(mm) >> 10, 8); + SEQ_PUT_DEC(" kB\nVmSwap:\t", swap); + seq_puts(m, " kB\n"); hugetlb_report_usage(m, mm); } +#undef SEQ_PUT_DEC unsigned long task_vsize(struct mm_struct *mm) { @@ -287,15 +278,18 @@ static void show_vma_header_prefix(struct seq_file *m, dev_t dev, unsigned long ino) { seq_setwidth(m, 25 + sizeof(void *) * 6 - 1); - seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu ", - start, - end, - flags & VM_READ ? 'r' : '-', - flags & VM_WRITE ? 'w' : '-', - flags & VM_EXEC ? 'x' : '-', - flags & VM_MAYSHARE ? 's' : 'p', - pgoff, - MAJOR(dev), MINOR(dev), ino); + seq_put_hex_ll(m, NULL, start, 8); + seq_put_hex_ll(m, "-", end, 8); + seq_putc(m, ' '); + seq_putc(m, flags & VM_READ ? 'r' : '-'); + seq_putc(m, flags & VM_WRITE ? 'w' : '-'); + seq_putc(m, flags & VM_EXEC ? 'x' : '-'); + seq_putc(m, flags & VM_MAYSHARE ? 's' : 'p'); + seq_put_hex_ll(m, " ", pgoff, 8); + seq_put_hex_ll(m, " ", MAJOR(dev), 2); + seq_put_hex_ll(m, ":", MINOR(dev), 2); + seq_put_decimal_ull(m, " ", ino); + seq_putc(m, ' '); } static void @@ -694,8 +688,9 @@ static void show_smap_vma_flags(struct seq_file *m, struct vm_area_struct *vma) if (!mnemonics[i][0]) continue; if (vma->vm_flags & (1UL << i)) { - seq_printf(m, "%c%c ", - mnemonics[i][0], mnemonics[i][1]); + seq_putc(m, mnemonics[i][0]); + seq_putc(m, mnemonics[i][1]); + seq_putc(m, ' '); } } seq_putc(m, '\n'); @@ -736,6 +731,8 @@ void __weak arch_show_smap(struct seq_file *m, struct vm_area_struct *vma) { } +#define SEQ_PUT_DEC(str, val) \ + seq_put_decimal_ull_width(m, str, (val) >> 10, 8) static int show_smap(struct seq_file *m, void *v, int is_pid) { struct proc_maps_private *priv = m->private; @@ -809,51 +806,34 @@ static int show_smap(struct seq_file *m, void *v, int is_pid) ret = SEQ_SKIP; } - if (!rollup_mode) - seq_printf(m, - "Size: %8lu kB\n" - "KernelPageSize: %8lu kB\n" - "MMUPageSize: %8lu kB\n", - (vma->vm_end - vma->vm_start) >> 10, - vma_kernel_pagesize(vma) >> 10, - vma_mmu_pagesize(vma) >> 10); - - - if (!rollup_mode || last_vma) - seq_printf(m, - "Rss: %8lu kB\n" - "Pss: %8lu kB\n" - "Shared_Clean: %8lu kB\n" - "Shared_Dirty: %8lu kB\n" - "Private_Clean: %8lu kB\n" - "Private_Dirty: %8lu kB\n" - "Referenced: %8lu kB\n" - "Anonymous: %8lu kB\n" - "LazyFree: %8lu kB\n" - "AnonHugePages: %8lu kB\n" - "ShmemPmdMapped: %8lu kB\n" - "Shared_Hugetlb: %8lu kB\n" - "Private_Hugetlb: %7lu kB\n" - "Swap: %8lu kB\n" - "SwapPss: %8lu kB\n" - "Locked: %8lu kB\n", - mss->resident >> 10, - (unsigned long)(mss->pss >> (10 + PSS_SHIFT)), - mss->shared_clean >> 10, - mss->shared_dirty >> 10, - mss->private_clean >> 10, - mss->private_dirty >> 10, - mss->referenced >> 10, - mss->anonymous >> 10, - mss->lazyfree >> 10, - mss->anonymous_thp >> 10, - mss->shmem_thp >> 10, - mss->shared_hugetlb >> 10, - mss->private_hugetlb >> 10, - mss->swap >> 10, - (unsigned long)(mss->swap_pss >> (10 + PSS_SHIFT)), - (unsigned long)(mss->pss >> (10 + PSS_SHIFT))); + if (!rollup_mode) { + SEQ_PUT_DEC("Size: ", vma->vm_end - vma->vm_start); + SEQ_PUT_DEC(" kB\nKernelPageSize: ", vma_kernel_pagesize(vma)); + SEQ_PUT_DEC(" kB\nMMUPageSize: ", vma_mmu_pagesize(vma)); + seq_puts(m, " kB\n"); + } + if (!rollup_mode || last_vma) { + SEQ_PUT_DEC("Rss: ", mss->resident); + SEQ_PUT_DEC(" kB\nPss: ", mss->pss >> PSS_SHIFT); + SEQ_PUT_DEC(" kB\nShared_Clean: ", mss->shared_clean); + SEQ_PUT_DEC(" kB\nShared_Dirty: ", mss->shared_dirty); + SEQ_PUT_DEC(" kB\nPrivate_Clean: ", mss->private_clean); + SEQ_PUT_DEC(" kB\nPrivate_Dirty: ", mss->private_dirty); + SEQ_PUT_DEC(" kB\nReferenced: ", mss->referenced); + SEQ_PUT_DEC(" kB\nAnonymous: ", mss->anonymous); + SEQ_PUT_DEC(" kB\nLazyFree: ", mss->lazyfree); + SEQ_PUT_DEC(" kB\nAnonHugePages: ", mss->anonymous_thp); + SEQ_PUT_DEC(" kB\nShmemPmdMapped: ", mss->shmem_thp); + SEQ_PUT_DEC(" kB\nShared_Hugetlb: ", mss->shared_hugetlb); + seq_put_decimal_ull_width(m, " kB\nPrivate_Hugetlb: ", + mss->private_hugetlb >> 10, 7); + SEQ_PUT_DEC(" kB\nSwap: ", mss->swap); + SEQ_PUT_DEC(" kB\nSwapPss: ", + mss->swap_pss >> PSS_SHIFT); + SEQ_PUT_DEC(" kB\nLocked: ", mss->pss >> PSS_SHIFT); + seq_puts(m, " kB\n"); + } if (!rollup_mode) { arch_show_smap(m, vma); show_smap_vma_flags(m, vma); @@ -861,6 +841,7 @@ static int show_smap(struct seq_file *m, void *v, int is_pid) m_cache_vma(m, vma); return ret; } +#undef SEQ_PUT_DEC static int show_pid_smap(struct seq_file *m, void *v) { @@ -956,7 +937,7 @@ static inline void clear_soft_dirty(struct vm_area_struct *vma, /* * The soft-dirty tracker uses #PF-s to catch writes * to pages, so write-protect the pte as well. See the - * Documentation/vm/soft-dirty.txt for full description + * Documentation/admin-guide/mm/soft-dirty.rst for full description * of how soft-dirty works. */ pte_t ptent = *pte; @@ -1329,9 +1310,11 @@ static int pagemap_pmd_range(pmd_t *pmdp, unsigned long addr, unsigned long end, #ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION else if (is_swap_pmd(pmd)) { swp_entry_t entry = pmd_to_swp_entry(pmd); + unsigned long offset = swp_offset(entry); + offset += (addr & ~PMD_MASK) >> PAGE_SHIFT; frame = swp_type(entry) | - (swp_offset(entry) << MAX_SWAPFILES_SHIFT); + (offset << MAX_SWAPFILES_SHIFT); flags |= PM_SWAP; if (pmd_swp_soft_dirty(pmd)) flags |= PM_SOFT_DIRTY; @@ -1351,6 +1334,8 @@ static int pagemap_pmd_range(pmd_t *pmdp, unsigned long addr, unsigned long end, break; if (pm->show_pfn && (flags & PM_PRESENT)) frame++; + else if (flags & PM_SWAP) + frame += (1 << MAX_SWAPFILES_SHIFT); } spin_unlock(ptl); return err; @@ -1436,7 +1421,7 @@ static int pagemap_hugetlb_range(pte_t *ptep, unsigned long hmask, * Bits 0-54 page frame number (PFN) if present * Bits 0-4 swap type if swapped * Bits 5-54 swap offset if swapped - * Bit 55 pte is soft-dirty (see Documentation/vm/soft-dirty.txt) + * Bit 55 pte is soft-dirty (see Documentation/admin-guide/mm/soft-dirty.rst) * Bit 56 page exclusively mapped * Bits 57-60 zero * Bit 61 page is file-page or shared-anon diff --git a/fs/proc/thread_self.c b/fs/proc/thread_self.c index 9d2efaca499f..b905010ca9eb 100644 --- a/fs/proc/thread_self.c +++ b/fs/proc/thread_self.c @@ -12,7 +12,7 @@ static const char *proc_thread_self_get_link(struct dentry *dentry, struct inode *inode, struct delayed_call *done) { - struct pid_namespace *ns = inode->i_sb->s_fs_info; + struct pid_namespace *ns = proc_pid_ns(inode); pid_t tgid = task_tgid_nr_ns(current, ns); pid_t pid = task_pid_nr_ns(current, ns); char *name; @@ -36,7 +36,7 @@ static unsigned thread_self_inum __ro_after_init; int proc_setup_thread_self(struct super_block *s) { struct inode *root_inode = d_inode(s->s_root); - struct pid_namespace *ns = s->s_fs_info; + struct pid_namespace *ns = proc_pid_ns(root_inode); struct dentry *thread_self; inode_lock(root_inode); diff --git a/fs/proc/uptime.c b/fs/proc/uptime.c index 95a708d83721..3bd12f955867 100644 --- a/fs/proc/uptime.c +++ b/fs/proc/uptime.c @@ -30,21 +30,9 @@ static int uptime_proc_show(struct seq_file *m, void *v) return 0; } -static int uptime_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, uptime_proc_show, NULL); -} - -static const struct file_operations uptime_proc_fops = { - .open = uptime_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_uptime_init(void) { - proc_create("uptime", 0, NULL, &uptime_proc_fops); + proc_create_single("uptime", 0, NULL, uptime_proc_show); return 0; } fs_initcall(proc_uptime_init); diff --git a/fs/proc/version.c b/fs/proc/version.c index 94901e8e700d..b449f186577f 100644 --- a/fs/proc/version.c +++ b/fs/proc/version.c @@ -15,21 +15,9 @@ static int version_proc_show(struct seq_file *m, void *v) return 0; } -static int version_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, version_proc_show, NULL); -} - -static const struct file_operations version_proc_fops = { - .open = version_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static int __init proc_version_init(void) { - proc_create("version", 0, NULL, &version_proc_fops); + proc_create_single("version", 0, NULL, version_proc_show); return 0; } fs_initcall(proc_version_init); diff --git a/fs/pstore/Kconfig b/fs/pstore/Kconfig index b42e5bd6d8ff..09c19ef91526 100644 --- a/fs/pstore/Kconfig +++ b/fs/pstore/Kconfig @@ -1,5 +1,6 @@ config PSTORE tristate "Persistent store support" + select CRYPTO if PSTORE_COMPRESS default n help This option enables generic access to platform level @@ -12,35 +13,89 @@ config PSTORE If you don't have a platform persistent store driver, say N. -choice - prompt "Choose compression algorithm" - depends on PSTORE - default PSTORE_ZLIB_COMPRESS - help - This option chooses compression algorithm. - -config PSTORE_ZLIB_COMPRESS - bool "ZLIB" - select ZLIB_DEFLATE - select ZLIB_INFLATE - help - This option enables ZLIB compression algorithm support. +config PSTORE_DEFLATE_COMPRESS + tristate "DEFLATE (ZLIB) compression" + default y + depends on PSTORE + select CRYPTO_DEFLATE + help + This option enables DEFLATE (also known as ZLIB) compression + algorithm support. config PSTORE_LZO_COMPRESS - bool "LZO" - select LZO_COMPRESS - select LZO_DECOMPRESS - help - This option enables LZO compression algorithm support. + tristate "LZO compression" + depends on PSTORE + select CRYPTO_LZO + help + This option enables LZO compression algorithm support. config PSTORE_LZ4_COMPRESS - bool "LZ4" - select LZ4_COMPRESS - select LZ4_DECOMPRESS - help - This option enables LZ4 compression algorithm support. + tristate "LZ4 compression" + depends on PSTORE + select CRYPTO_LZ4 + help + This option enables LZ4 compression algorithm support. + +config PSTORE_LZ4HC_COMPRESS + tristate "LZ4HC compression" + depends on PSTORE + select CRYPTO_LZ4HC + help + This option enables LZ4HC (high compression) mode algorithm. + +config PSTORE_842_COMPRESS + bool "842 compression" + depends on PSTORE + select CRYPTO_842 + help + This option enables 842 compression algorithm support. + +config PSTORE_COMPRESS + def_bool y + depends on PSTORE + depends on PSTORE_DEFLATE_COMPRESS || PSTORE_LZO_COMPRESS || \ + PSTORE_LZ4_COMPRESS || PSTORE_LZ4HC_COMPRESS || \ + PSTORE_842_COMPRESS + +choice + prompt "Default pstore compression algorithm" + depends on PSTORE_COMPRESS + help + This option chooses the default active compression algorithm. + This change be changed at boot with "pstore.compress=..." on + the kernel command line. + + Currently, pstore has support for 5 compression algorithms: + deflate, lzo, lz4, lz4hc and 842. + + The default compression algorithm is deflate. + + config PSTORE_DEFLATE_COMPRESS_DEFAULT + bool "deflate" if PSTORE_DEFLATE_COMPRESS + + config PSTORE_LZO_COMPRESS_DEFAULT + bool "lzo" if PSTORE_LZO_COMPRESS + + config PSTORE_LZ4_COMPRESS_DEFAULT + bool "lz4" if PSTORE_LZ4_COMPRESS + + config PSTORE_LZ4HC_COMPRESS_DEFAULT + bool "lz4hc" if PSTORE_LZ4HC_COMPRESS + + config PSTORE_842_COMPRESS_DEFAULT + bool "842" if PSTORE_842_COMPRESS + endchoice +config PSTORE_COMPRESS_DEFAULT + string + depends on PSTORE_COMPRESS + default "deflate" if PSTORE_DEFLATE_COMPRESS_DEFAULT + default "lzo" if PSTORE_LZO_COMPRESS_DEFAULT + default "lz4" if PSTORE_LZ4_COMPRESS_DEFAULT + default "lz4hc" if PSTORE_LZ4HC_COMPRESS_DEFAULT + default "842" if PSTORE_842_COMPRESS_DEFAULT + config PSTORE_CONSOLE bool "Log kernel console messages" depends on PSTORE diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c index d814723fb27d..5fcb845b9fec 100644 --- a/fs/pstore/inode.c +++ b/fs/pstore/inode.c @@ -486,6 +486,8 @@ static int __init init_pstore_fs(void) { int err; + pstore_choose_compression(); + /* Create a convenient mount point for people to access pstore */ err = sysfs_create_mount_point(fs_kobj, "pstore"); if (err) diff --git a/fs/pstore/internal.h b/fs/pstore/internal.h index c029314478fa..fb767e28aeb2 100644 --- a/fs/pstore/internal.h +++ b/fs/pstore/internal.h @@ -37,4 +37,7 @@ extern bool pstore_is_mounted(void); extern void pstore_record_init(struct pstore_record *record, struct pstore_info *psi); +/* Called during module_init() */ +extern void __init pstore_choose_compression(void); + #endif diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index c3129b131e4d..dc720573fd53 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -28,15 +28,13 @@ #include <linux/console.h> #include <linux/module.h> #include <linux/pstore.h> -#ifdef CONFIG_PSTORE_ZLIB_COMPRESS -#include <linux/zlib.h> -#endif -#ifdef CONFIG_PSTORE_LZO_COMPRESS +#if IS_ENABLED(CONFIG_PSTORE_LZO_COMPRESS) #include <linux/lzo.h> #endif -#ifdef CONFIG_PSTORE_LZ4_COMPRESS +#if IS_ENABLED(CONFIG_PSTORE_LZ4_COMPRESS) || IS_ENABLED(CONFIG_PSTORE_LZ4HC_COMPRESS) #include <linux/lz4.h> #endif +#include <linux/crypto.h> #include <linux/string.h> #include <linux/timer.h> #include <linux/slab.h> @@ -74,23 +72,18 @@ static DEFINE_SPINLOCK(pstore_lock); struct pstore_info *psinfo; static char *backend; - -/* Compression parameters */ -#ifdef CONFIG_PSTORE_ZLIB_COMPRESS -#define COMPR_LEVEL 6 -#define WINDOW_BITS 12 -#define MEM_LEVEL 4 -static struct z_stream_s stream; +static char *compress = +#ifdef CONFIG_PSTORE_COMPRESS_DEFAULT + CONFIG_PSTORE_COMPRESS_DEFAULT; #else -static unsigned char *workspace; + NULL; #endif -struct pstore_zbackend { - int (*compress)(const void *in, void *out, size_t inlen, size_t outlen); - int (*decompress)(void *in, void *out, size_t inlen, size_t outlen); - void (*allocate)(void); - void (*free)(void); +/* Compression parameters */ +static struct crypto_comp *tfm; +struct pstore_zbackend { + int (*zbufsize)(size_t size); const char *name; }; @@ -149,77 +142,12 @@ bool pstore_cannot_block_path(enum kmsg_dump_reason reason) } EXPORT_SYMBOL_GPL(pstore_cannot_block_path); -#ifdef CONFIG_PSTORE_ZLIB_COMPRESS -/* Derived from logfs_compress() */ -static int compress_zlib(const void *in, void *out, size_t inlen, size_t outlen) -{ - int err, ret; - - ret = -EIO; - err = zlib_deflateInit2(&stream, COMPR_LEVEL, Z_DEFLATED, WINDOW_BITS, - MEM_LEVEL, Z_DEFAULT_STRATEGY); - if (err != Z_OK) - goto error; - - stream.next_in = in; - stream.avail_in = inlen; - stream.total_in = 0; - stream.next_out = out; - stream.avail_out = outlen; - stream.total_out = 0; - - err = zlib_deflate(&stream, Z_FINISH); - if (err != Z_STREAM_END) - goto error; - - err = zlib_deflateEnd(&stream); - if (err != Z_OK) - goto error; - - if (stream.total_out >= stream.total_in) - goto error; - - ret = stream.total_out; -error: - return ret; -} - -/* Derived from logfs_uncompress */ -static int decompress_zlib(void *in, void *out, size_t inlen, size_t outlen) +#if IS_ENABLED(CONFIG_PSTORE_DEFLATE_COMPRESS) +static int zbufsize_deflate(size_t size) { - int err, ret; - - ret = -EIO; - err = zlib_inflateInit2(&stream, WINDOW_BITS); - if (err != Z_OK) - goto error; - - stream.next_in = in; - stream.avail_in = inlen; - stream.total_in = 0; - stream.next_out = out; - stream.avail_out = outlen; - stream.total_out = 0; - - err = zlib_inflate(&stream, Z_FINISH); - if (err != Z_STREAM_END) - goto error; - - err = zlib_inflateEnd(&stream); - if (err != Z_OK) - goto error; - - ret = stream.total_out; -error: - return ret; -} - -static void allocate_zlib(void) -{ - size_t size; size_t cmpr; - switch (psinfo->bufsize) { + switch (size) { /* buffer range for efivars */ case 1000 ... 2000: cmpr = 56; @@ -239,212 +167,131 @@ static void allocate_zlib(void) break; } - big_oops_buf_sz = (psinfo->bufsize * 100) / cmpr; - big_oops_buf = kmalloc(big_oops_buf_sz, GFP_KERNEL); - if (big_oops_buf) { - size = max(zlib_deflate_workspacesize(WINDOW_BITS, MEM_LEVEL), - zlib_inflate_workspacesize()); - stream.workspace = kmalloc(size, GFP_KERNEL); - if (!stream.workspace) { - pr_err("No memory for compression workspace; skipping compression\n"); - kfree(big_oops_buf); - big_oops_buf = NULL; - } - } else { - pr_err("No memory for uncompressed data; skipping compression\n"); - stream.workspace = NULL; - } - + return (size * 100) / cmpr; } - -static void free_zlib(void) -{ - kfree(stream.workspace); - stream.workspace = NULL; - kfree(big_oops_buf); - big_oops_buf = NULL; - big_oops_buf_sz = 0; -} - -static const struct pstore_zbackend backend_zlib = { - .compress = compress_zlib, - .decompress = decompress_zlib, - .allocate = allocate_zlib, - .free = free_zlib, - .name = "zlib", -}; #endif -#ifdef CONFIG_PSTORE_LZO_COMPRESS -static int compress_lzo(const void *in, void *out, size_t inlen, size_t outlen) +#if IS_ENABLED(CONFIG_PSTORE_LZO_COMPRESS) +static int zbufsize_lzo(size_t size) { - int ret; - - ret = lzo1x_1_compress(in, inlen, out, &outlen, workspace); - if (ret != LZO_E_OK) { - pr_err("lzo_compress error, ret = %d!\n", ret); - return -EIO; - } - - return outlen; + return lzo1x_worst_compress(size); } +#endif -static int decompress_lzo(void *in, void *out, size_t inlen, size_t outlen) +#if IS_ENABLED(CONFIG_PSTORE_LZ4_COMPRESS) || IS_ENABLED(CONFIG_PSTORE_LZ4HC_COMPRESS) +static int zbufsize_lz4(size_t size) { - int ret; - - ret = lzo1x_decompress_safe(in, inlen, out, &outlen); - if (ret != LZO_E_OK) { - pr_err("lzo_decompress error, ret = %d!\n", ret); - return -EIO; - } - - return outlen; + return LZ4_compressBound(size); } +#endif -static void allocate_lzo(void) +#if IS_ENABLED(CONFIG_PSTORE_842_COMPRESS) +static int zbufsize_842(size_t size) { - big_oops_buf_sz = lzo1x_worst_compress(psinfo->bufsize); - big_oops_buf = kmalloc(big_oops_buf_sz, GFP_KERNEL); - if (big_oops_buf) { - workspace = kmalloc(LZO1X_MEM_COMPRESS, GFP_KERNEL); - if (!workspace) { - pr_err("No memory for compression workspace; skipping compression\n"); - kfree(big_oops_buf); - big_oops_buf = NULL; - } - } else { - pr_err("No memory for uncompressed data; skipping compression\n"); - workspace = NULL; - } + return size; } +#endif -static void free_lzo(void) -{ - kfree(workspace); - kfree(big_oops_buf); - big_oops_buf = NULL; - big_oops_buf_sz = 0; -} +static const struct pstore_zbackend *zbackend __ro_after_init; -static const struct pstore_zbackend backend_lzo = { - .compress = compress_lzo, - .decompress = decompress_lzo, - .allocate = allocate_lzo, - .free = free_lzo, - .name = "lzo", -}; +static const struct pstore_zbackend zbackends[] = { +#if IS_ENABLED(CONFIG_PSTORE_DEFLATE_COMPRESS) + { + .zbufsize = zbufsize_deflate, + .name = "deflate", + }, +#endif +#if IS_ENABLED(CONFIG_PSTORE_LZO_COMPRESS) + { + .zbufsize = zbufsize_lzo, + .name = "lzo", + }, +#endif +#if IS_ENABLED(CONFIG_PSTORE_LZ4_COMPRESS) + { + .zbufsize = zbufsize_lz4, + .name = "lz4", + }, #endif +#if IS_ENABLED(CONFIG_PSTORE_LZ4HC_COMPRESS) + { + .zbufsize = zbufsize_lz4, + .name = "lz4hc", + }, +#endif +#if IS_ENABLED(CONFIG_PSTORE_842_COMPRESS) + { + .zbufsize = zbufsize_842, + .name = "842", + }, +#endif + { } +}; -#ifdef CONFIG_PSTORE_LZ4_COMPRESS -static int compress_lz4(const void *in, void *out, size_t inlen, size_t outlen) +static int pstore_compress(const void *in, void *out, + unsigned int inlen, unsigned int outlen) { int ret; - ret = LZ4_compress_default(in, out, inlen, outlen, workspace); - if (!ret) { - pr_err("LZ4_compress_default error; compression failed!\n"); - return -EIO; + ret = crypto_comp_compress(tfm, in, inlen, out, &outlen); + if (ret) { + pr_err("crypto_comp_compress failed, ret = %d!\n", ret); + return ret; } - return ret; + return outlen; } -static int decompress_lz4(void *in, void *out, size_t inlen, size_t outlen) +static int pstore_decompress(void *in, void *out, + unsigned int inlen, unsigned int outlen) { int ret; - ret = LZ4_decompress_safe(in, out, inlen, outlen); - if (ret < 0) { - /* - * LZ4_decompress_safe will return an error code - * (< 0) if decompression failed - */ - pr_err("LZ4_decompress_safe error, ret = %d!\n", ret); - return -EIO; + ret = crypto_comp_decompress(tfm, in, inlen, out, &outlen); + if (ret) { + pr_err("crypto_comp_decompress failed, ret = %d!\n", ret); + return ret; } - return ret; -} - -static void allocate_lz4(void) -{ - big_oops_buf_sz = LZ4_compressBound(psinfo->bufsize); - big_oops_buf = kmalloc(big_oops_buf_sz, GFP_KERNEL); - if (big_oops_buf) { - workspace = kmalloc(LZ4_MEM_COMPRESS, GFP_KERNEL); - if (!workspace) { - pr_err("No memory for compression workspace; skipping compression\n"); - kfree(big_oops_buf); - big_oops_buf = NULL; - } - } else { - pr_err("No memory for uncompressed data; skipping compression\n"); - workspace = NULL; - } + return outlen; } -static void free_lz4(void) +static void allocate_buf_for_compression(void) { - kfree(workspace); - kfree(big_oops_buf); - big_oops_buf = NULL; - big_oops_buf_sz = 0; -} - -static const struct pstore_zbackend backend_lz4 = { - .compress = compress_lz4, - .decompress = decompress_lz4, - .allocate = allocate_lz4, - .free = free_lz4, - .name = "lz4", -}; -#endif - -static const struct pstore_zbackend *zbackend = -#if defined(CONFIG_PSTORE_ZLIB_COMPRESS) - &backend_zlib; -#elif defined(CONFIG_PSTORE_LZO_COMPRESS) - &backend_lzo; -#elif defined(CONFIG_PSTORE_LZ4_COMPRESS) - &backend_lz4; -#else - NULL; -#endif + if (!IS_ENABLED(CONFIG_PSTORE_COMPRESS) || !zbackend) + return; -static int pstore_compress(const void *in, void *out, - size_t inlen, size_t outlen) -{ - if (zbackend) - return zbackend->compress(in, out, inlen, outlen); - else - return -EIO; -} + if (!crypto_has_comp(zbackend->name, 0, 0)) { + pr_err("No %s compression\n", zbackend->name); + return; + } -static int pstore_decompress(void *in, void *out, size_t inlen, size_t outlen) -{ - if (zbackend) - return zbackend->decompress(in, out, inlen, outlen); - else - return -EIO; -} + big_oops_buf_sz = zbackend->zbufsize(psinfo->bufsize); + if (big_oops_buf_sz <= 0) + return; -static void allocate_buf_for_compression(void) -{ - if (zbackend) { - pr_info("using %s compression\n", zbackend->name); - zbackend->allocate(); - } else { + big_oops_buf = kmalloc(big_oops_buf_sz, GFP_KERNEL); + if (!big_oops_buf) { pr_err("allocate compression buffer error!\n"); + return; + } + + tfm = crypto_alloc_comp(zbackend->name, 0, 0); + if (IS_ERR_OR_NULL(tfm)) { + kfree(big_oops_buf); + big_oops_buf = NULL; + pr_err("crypto_alloc_comp() failed!\n"); + return; } } static void free_buf_for_compression(void) { - if (zbackend) - zbackend->free(); - else - pr_err("free compression buffer error!\n"); + if (IS_ENABLED(CONFIG_PSTORE_COMPRESS) && !IS_ERR_OR_NULL(tfm)) + crypto_free_comp(tfm); + kfree(big_oops_buf); + big_oops_buf = NULL; + big_oops_buf_sz = 0; } /* @@ -901,5 +748,24 @@ static void pstore_timefunc(struct timer_list *unused) jiffies + msecs_to_jiffies(pstore_update_ms)); } +void __init pstore_choose_compression(void) +{ + const struct pstore_zbackend *step; + + if (!compress) + return; + + for (step = zbackends; step->name; step++) { + if (!strcmp(compress, step->name)) { + zbackend = step; + pr_info("using %s compression\n", zbackend->name); + return; + } + } +} + +module_param(compress, charp, 0444); +MODULE_PARM_DESC(compress, "Pstore compression to use"); + module_param(backend, charp, 0444); MODULE_PARM_DESC(backend, "Pstore backend to use"); diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c index 7125b398d312..49b2bc114868 100644 --- a/fs/pstore/ram.c +++ b/fs/pstore/ram.c @@ -938,7 +938,7 @@ static int __init ramoops_init(void) ramoops_register_dummy(); return platform_driver_register(&ramoops_driver); } -postcore_initcall(ramoops_init); +late_initcall(ramoops_init); static void __exit ramoops_exit(void) { diff --git a/fs/pstore/ram_core.c b/fs/pstore/ram_core.c index e11672aa4575..951a14edcf51 100644 --- a/fs/pstore/ram_core.c +++ b/fs/pstore/ram_core.c @@ -98,24 +98,23 @@ static void notrace persistent_ram_encode_rs8(struct persistent_ram_zone *prz, uint8_t *data, size_t len, uint8_t *ecc) { int i; - uint16_t par[prz->ecc_info.ecc_size]; /* Initialize the parity buffer */ - memset(par, 0, sizeof(par)); - encode_rs8(prz->rs_decoder, data, len, par, 0); + memset(prz->ecc_info.par, 0, + prz->ecc_info.ecc_size * sizeof(prz->ecc_info.par[0])); + encode_rs8(prz->rs_decoder, data, len, prz->ecc_info.par, 0); for (i = 0; i < prz->ecc_info.ecc_size; i++) - ecc[i] = par[i]; + ecc[i] = prz->ecc_info.par[i]; } static int persistent_ram_decode_rs8(struct persistent_ram_zone *prz, void *data, size_t len, uint8_t *ecc) { int i; - uint16_t par[prz->ecc_info.ecc_size]; for (i = 0; i < prz->ecc_info.ecc_size; i++) - par[i] = ecc[i]; - return decode_rs8(prz->rs_decoder, data, par, len, + prz->ecc_info.par[i] = ecc[i]; + return decode_rs8(prz->rs_decoder, data, prz->ecc_info.par, len, NULL, 0, NULL, 0, NULL); } @@ -228,6 +227,15 @@ static int persistent_ram_init_ecc(struct persistent_ram_zone *prz, return -EINVAL; } + /* allocate workspace instead of using stack VLA */ + prz->ecc_info.par = kmalloc_array(prz->ecc_info.ecc_size, + sizeof(*prz->ecc_info.par), + GFP_KERNEL); + if (!prz->ecc_info.par) { + pr_err("cannot allocate ECC parity workspace\n"); + return -ENOMEM; + } + prz->corrected_bytes = 0; prz->bad_blocks = 0; @@ -514,6 +522,13 @@ void persistent_ram_free(struct persistent_ram_zone *prz) } prz->vaddr = NULL; } + if (prz->rs_decoder) { + free_rs(prz->rs_decoder); + prz->rs_decoder = NULL; + } + kfree(prz->ecc_info.par); + prz->ecc_info.par = NULL; + persistent_ram_free_old(prz); kfree(prz); } diff --git a/fs/qnx4/namei.c b/fs/qnx4/namei.c index eca27878079d..8d72221735d7 100644 --- a/fs/qnx4/namei.c +++ b/fs/qnx4/namei.c @@ -114,13 +114,9 @@ struct dentry * qnx4_lookup(struct inode *dir, struct dentry *dentry, unsigned i brelse(bh); foundinode = qnx4_iget(dir->i_sb, ino); - if (IS_ERR(foundinode)) { + if (IS_ERR(foundinode)) QNX4DEBUG((KERN_ERR "qnx4: lookup->iget -> error %ld\n", PTR_ERR(foundinode))); - return ERR_CAST(foundinode); - } out: - d_add(dentry, foundinode); - - return NULL; + return d_splice_alias(foundinode, dentry); } diff --git a/fs/qnx6/namei.c b/fs/qnx6/namei.c index 72c2770830be..e2e98e653b8d 100644 --- a/fs/qnx6/namei.c +++ b/fs/qnx6/namei.c @@ -29,15 +29,11 @@ struct dentry *qnx6_lookup(struct inode *dir, struct dentry *dentry, if (ino) { foundinode = qnx6_iget(dir->i_sb, ino); qnx6_put_page(page); - if (IS_ERR(foundinode)) { + if (IS_ERR(foundinode)) pr_debug("lookup->iget -> error %ld\n", PTR_ERR(foundinode)); - return ERR_CAST(foundinode); - } } else { pr_debug("%s(): not found %s\n", __func__, name); - return NULL; } - d_add(dentry, foundinode); - return NULL; + return d_splice_alias(foundinode, dentry); } diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 020c597ef9b6..d88231e3b2be 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -2966,7 +2966,7 @@ static int __init dquot_init(void) NULL); order = 0; - dquot_hash = (struct hlist_head *)__get_free_pages(GFP_ATOMIC, order); + dquot_hash = (struct hlist_head *)__get_free_pages(GFP_KERNEL, order); if (!dquot_hash) panic("Cannot create dquot hash table"); diff --git a/fs/read_write.c b/fs/read_write.c index c4eabbfc90df..e83bd9744b5d 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -2023,7 +2023,7 @@ int vfs_dedupe_file_range(struct file *file, struct file_dedupe_range *same) ret = mnt_want_write_file(dst_file); if (ret) { info->status = ret; - goto next_loop; + goto next_fdput; } dst_off = info->dest_offset; @@ -2058,9 +2058,9 @@ int vfs_dedupe_file_range(struct file *file, struct file_dedupe_range *same) next_file: mnt_drop_write_file(dst_file); -next_loop: +next_fdput: fdput(dst_fd); - +next_loop: if (fatal_signal_pending(current)) goto out; } diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 70057359fbaf..23148c3ed675 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2643,7 +2643,7 @@ static int journal_init_dev(struct super_block *super, if (IS_ERR(journal->j_dev_bd)) { result = PTR_ERR(journal->j_dev_bd); journal->j_dev_bd = NULL; - reiserfs_warning(super, + reiserfs_warning(super, "sh-457", "journal_init_dev: Cannot open '%s': %i", jdev_name, result); return result; diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index bd39a998843d..5089dac02660 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -687,8 +687,7 @@ static int reiserfs_create(struct inode *dir, struct dentry *dentry, umode_t mod reiserfs_update_inode_transaction(inode); reiserfs_update_inode_transaction(dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: @@ -771,8 +770,7 @@ static int reiserfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode goto out_failed; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: @@ -871,8 +869,7 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode /* the above add_entry did not update dir's stat data */ reiserfs_update_sd(&th, dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: reiserfs_write_unlock(dir->i_sb); @@ -1187,8 +1184,7 @@ static int reiserfs_symlink(struct inode *parent_dir, goto out_failed; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: reiserfs_write_unlock(parent_dir->i_sb); diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index fe999157dd97..e39b3910d24d 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -389,27 +389,13 @@ static int show_journal(struct seq_file *m, void *unused) return 0; } -static int r_open(struct inode *inode, struct file *file) -{ - return single_open(file, PDE_DATA(inode), - proc_get_parent_data(inode)); -} - -static const struct file_operations r_file_operations = { - .open = r_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - static struct proc_dir_entry *proc_info_root = NULL; static const char proc_info_root_name[] = "fs/reiserfs"; static void add_file(struct super_block *sb, char *name, int (*func) (struct seq_file *, void *)) { - proc_create_data(name, 0, REISERFS_SB(sb)->procdir, - &r_file_operations, func); + proc_create_single_data(name, 0, REISERFS_SB(sb)->procdir, func, sb); } int reiserfs_proc_info_init(struct super_block *sb) diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 8f06fd1f3d69..6ccb51993a76 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -213,7 +213,7 @@ static struct dentry *romfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { unsigned long offset, maxoff; - struct inode *inode; + struct inode *inode = NULL; struct romfs_inode ri; const char *name; /* got from dentry */ int len, ret; @@ -233,7 +233,7 @@ static struct dentry *romfs_lookup(struct inode *dir, struct dentry *dentry, for (;;) { if (!offset || offset >= maxoff) - goto out0; + break; ret = romfs_dev_read(dir->i_sb, offset, &ri, sizeof(ri)); if (ret < 0) @@ -244,37 +244,19 @@ static struct dentry *romfs_lookup(struct inode *dir, struct dentry *dentry, len); if (ret < 0) goto error; - if (ret == 1) + if (ret == 1) { + /* Hard link handling */ + if ((be32_to_cpu(ri.next) & ROMFH_TYPE) == ROMFH_HRD) + offset = be32_to_cpu(ri.spec) & ROMFH_MASK; + inode = romfs_iget(dir->i_sb, offset); break; + } /* next entry */ offset = be32_to_cpu(ri.next) & ROMFH_MASK; } - /* Hard link handling */ - if ((be32_to_cpu(ri.next) & ROMFH_TYPE) == ROMFH_HRD) - offset = be32_to_cpu(ri.spec) & ROMFH_MASK; - - inode = romfs_iget(dir->i_sb, offset); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); - goto error; - } - goto outi; - - /* - * it's a bit funky, _lookup needs to return an error code - * (negative) or a NULL, both as a dentry. ENOENT should not - * be returned, instead we need to create a negative dentry by - * d_add(dentry, NULL); and return 0 as no error. - * (Although as I see, it only matters on writable file - * systems). - */ -out0: - inode = NULL; -outi: - d_add(dentry, inode); - ret = 0; + return d_splice_alias(inode, dentry); error: return ERR_PTR(ret); } diff --git a/fs/select.c b/fs/select.c index ba879c51288f..bc3cc0f98896 100644 --- a/fs/select.c +++ b/fs/select.c @@ -34,6 +34,29 @@ #include <linux/uaccess.h> +__poll_t vfs_poll(struct file *file, struct poll_table_struct *pt) +{ + if (file->f_op->poll) { + return file->f_op->poll(file, pt); + } else if (file_has_poll_mask(file)) { + unsigned int events = poll_requested_events(pt); + struct wait_queue_head *head; + + if (pt && pt->_qproc) { + head = file->f_op->get_poll_head(file, events); + if (!head) + return DEFAULT_POLLMASK; + if (IS_ERR(head)) + return EPOLLERR; + pt->_qproc(file, head, pt); + } + + return file->f_op->poll_mask(file, events); + } else { + return DEFAULT_POLLMASK; + } +} +EXPORT_SYMBOL_GPL(vfs_poll); /* * Estimate expected accuracy in ns from a timeval. @@ -233,7 +256,7 @@ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, add_wait_queue(wait_address, &entry->wait); } -int poll_schedule_timeout(struct poll_wqueues *pwq, int state, +static int poll_schedule_timeout(struct poll_wqueues *pwq, int state, ktime_t *expires, unsigned long slack) { int rc = -EINTR; @@ -258,7 +281,6 @@ int poll_schedule_timeout(struct poll_wqueues *pwq, int state, return rc; } -EXPORT_SYMBOL(poll_schedule_timeout); /** * poll_select_set_timeout - helper function to setup the timeout value @@ -503,14 +525,10 @@ static int do_select(int n, fd_set_bits *fds, struct timespec64 *end_time) continue; f = fdget(i); if (f.file) { - const struct file_operations *f_op; - f_op = f.file->f_op; - mask = DEFAULT_POLLMASK; - if (f_op->poll) { - wait_key_set(wait, in, out, - bit, busy_flag); - mask = (*f_op->poll)(f.file, wait); - } + wait_key_set(wait, in, out, bit, + busy_flag); + mask = vfs_poll(f.file, wait); + fdput(f); if ((mask & POLLIN_SET) && (in & bit)) { res_in |= bit; @@ -813,34 +831,29 @@ static inline __poll_t do_pollfd(struct pollfd *pollfd, poll_table *pwait, bool *can_busy_poll, __poll_t busy_flag) { - __poll_t mask; - int fd; - - mask = 0; - fd = pollfd->fd; - if (fd >= 0) { - struct fd f = fdget(fd); - mask = EPOLLNVAL; - if (f.file) { - /* userland u16 ->events contains POLL... bitmap */ - __poll_t filter = demangle_poll(pollfd->events) | - EPOLLERR | EPOLLHUP; - mask = DEFAULT_POLLMASK; - if (f.file->f_op->poll) { - pwait->_key = filter; - pwait->_key |= busy_flag; - mask = f.file->f_op->poll(f.file, pwait); - if (mask & busy_flag) - *can_busy_poll = true; - } - /* Mask out unneeded events. */ - mask &= filter; - fdput(f); - } - } + int fd = pollfd->fd; + __poll_t mask = 0, filter; + struct fd f; + + if (fd < 0) + goto out; + mask = EPOLLNVAL; + f = fdget(fd); + if (!f.file) + goto out; + + /* userland u16 ->events contains POLL... bitmap */ + filter = demangle_poll(pollfd->events) | EPOLLERR | EPOLLHUP; + pwait->_key = filter | busy_flag; + mask = vfs_poll(f.file, pwait); + if (mask & busy_flag) + *can_busy_poll = true; + mask &= filter; /* Mask out unneeded events. */ + fdput(f); + +out: /* ... and so does ->revents */ pollfd->revents = mangle_poll(mask); - return mask; } diff --git a/fs/seq_file.c b/fs/seq_file.c index eea09f6d8830..4cc090b50cc5 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -6,6 +6,7 @@ * initial implementation -- AV, Oct 2001. */ +#include <linux/cache.h> #include <linux/fs.h> #include <linux/export.h> #include <linux/seq_file.h> @@ -19,6 +20,8 @@ #include <linux/uaccess.h> #include <asm/page.h> +static struct kmem_cache *seq_file_cache __ro_after_init; + static void seq_set_overflow(struct seq_file *m) { m->count = m->size; @@ -26,7 +29,7 @@ static void seq_set_overflow(struct seq_file *m) static void *seq_buf_alloc(unsigned long size) { - return kvmalloc(size, GFP_KERNEL); + return kvmalloc(size, GFP_KERNEL_ACCOUNT); } /** @@ -51,7 +54,7 @@ int seq_open(struct file *file, const struct seq_operations *op) WARN_ON(file->private_data); - p = kzalloc(sizeof(*p), GFP_KERNEL); + p = kmem_cache_zalloc(seq_file_cache, GFP_KERNEL); if (!p) return -ENOMEM; @@ -366,7 +369,7 @@ int seq_release(struct inode *inode, struct file *file) { struct seq_file *m = file->private_data; kvfree(m->buf); - kfree(m); + kmem_cache_free(seq_file_cache, m); return 0; } EXPORT_SYMBOL(seq_release); @@ -563,7 +566,7 @@ static void single_stop(struct seq_file *p, void *v) int single_open(struct file *file, int (*show)(struct seq_file *, void *), void *data) { - struct seq_operations *op = kmalloc(sizeof(*op), GFP_KERNEL); + struct seq_operations *op = kmalloc(sizeof(*op), GFP_KERNEL_ACCOUNT); int res = -ENOMEM; if (op) { @@ -625,7 +628,7 @@ void *__seq_open_private(struct file *f, const struct seq_operations *ops, void *private; struct seq_file *seq; - private = kzalloc(psize, GFP_KERNEL); + private = kzalloc(psize, GFP_KERNEL_ACCOUNT); if (private == NULL) goto out; @@ -673,37 +676,40 @@ void seq_puts(struct seq_file *m, const char *s) } EXPORT_SYMBOL(seq_puts); -/* +/** * A helper routine for putting decimal numbers without rich format of printf(). * only 'unsigned long long' is supported. - * This routine will put strlen(delimiter) + number into seq_file. + * @m: seq_file identifying the buffer to which data should be written + * @delimiter: a string which is printed before the number + * @num: the number + * @width: a minimum field width + * + * This routine will put strlen(delimiter) + number into seq_filed. * This routine is very quick when you show lots of numbers. * In usual cases, it will be better to use seq_printf(). It's easier to read. */ -void seq_put_decimal_ull(struct seq_file *m, const char *delimiter, - unsigned long long num) +void seq_put_decimal_ull_width(struct seq_file *m, const char *delimiter, + unsigned long long num, unsigned int width) { int len; if (m->count + 2 >= m->size) /* we'll write 2 bytes at least */ goto overflow; - len = strlen(delimiter); - if (m->count + len >= m->size) - goto overflow; + if (delimiter && delimiter[0]) { + if (delimiter[1] == 0) + seq_putc(m, delimiter[0]); + else + seq_puts(m, delimiter); + } - memcpy(m->buf + m->count, delimiter, len); - m->count += len; + if (!width) + width = 1; - if (m->count + 1 >= m->size) + if (m->count + width >= m->size) goto overflow; - if (num < 10) { - m->buf[m->count++] = num + '0'; - return; - } - - len = num_to_str(m->buf + m->count, m->size - m->count, num); + len = num_to_str(m->buf + m->count, m->size - m->count, num, width); if (!len) goto overflow; @@ -713,8 +719,60 @@ void seq_put_decimal_ull(struct seq_file *m, const char *delimiter, overflow: seq_set_overflow(m); } + +void seq_put_decimal_ull(struct seq_file *m, const char *delimiter, + unsigned long long num) +{ + return seq_put_decimal_ull_width(m, delimiter, num, 0); +} EXPORT_SYMBOL(seq_put_decimal_ull); +/** + * seq_put_hex_ll - put a number in hexadecimal notation + * @m: seq_file identifying the buffer to which data should be written + * @delimiter: a string which is printed before the number + * @v: the number + * @width: a minimum field width + * + * seq_put_hex_ll(m, "", v, 8) is equal to seq_printf(m, "%08llx", v) + * + * This routine is very quick when you show lots of numbers. + * In usual cases, it will be better to use seq_printf(). It's easier to read. + */ +void seq_put_hex_ll(struct seq_file *m, const char *delimiter, + unsigned long long v, unsigned int width) +{ + unsigned int len; + int i; + + if (delimiter && delimiter[0]) { + if (delimiter[1] == 0) + seq_putc(m, delimiter[0]); + else + seq_puts(m, delimiter); + } + + /* If x is 0, the result of __builtin_clzll is undefined */ + if (v == 0) + len = 1; + else + len = (sizeof(v) * 8 - __builtin_clzll(v) + 3) / 4; + + if (len < width) + len = width; + + if (m->count + len > m->size) { + seq_set_overflow(m); + return; + } + + for (i = len - 1; i >= 0; i--) { + m->buf[m->count + i] = hex_asc[0xf & v]; + v = v >> 4; + } + m->count += len; +} + void seq_put_decimal_ll(struct seq_file *m, const char *delimiter, long long num) { int len; @@ -722,12 +780,12 @@ void seq_put_decimal_ll(struct seq_file *m, const char *delimiter, long long num if (m->count + 3 >= m->size) /* we'll write 2 bytes at least */ goto overflow; - len = strlen(delimiter); - if (m->count + len >= m->size) - goto overflow; - - memcpy(m->buf + m->count, delimiter, len); - m->count += len; + if (delimiter && delimiter[0]) { + if (delimiter[1] == 0) + seq_putc(m, delimiter[0]); + else + seq_puts(m, delimiter); + } if (m->count + 2 >= m->size) goto overflow; @@ -742,7 +800,7 @@ void seq_put_decimal_ll(struct seq_file *m, const char *delimiter, long long num return; } - len = num_to_str(m->buf + m->count, m->size - m->count, num); + len = num_to_str(m->buf + m->count, m->size - m->count, num, 0); if (!len) goto overflow; @@ -782,8 +840,14 @@ EXPORT_SYMBOL(seq_write); void seq_pad(struct seq_file *m, char c) { int size = m->pad_until - m->count; - if (size > 0) - seq_printf(m, "%*s", size, ""); + if (size > 0) { + if (size + m->count > m->size) { + seq_set_overflow(m); + return; + } + memset(m->buf + m->count, ' ', size); + m->count += size; + } if (c) seq_putc(m, c); } @@ -1040,3 +1104,8 @@ seq_hlist_next_percpu(void *v, struct hlist_head __percpu *head, return NULL; } EXPORT_SYMBOL(seq_hlist_next_percpu); + +void __init seq_file_init(void) +{ + seq_file_cache = KMEM_CACHE(seq_file, SLAB_ACCOUNT|SLAB_PANIC); +} diff --git a/fs/signalfd.c b/fs/signalfd.c index d2187a813376..cbb42f77a2bd 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -81,83 +81,86 @@ static __poll_t signalfd_poll(struct file *file, poll_table *wait) static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, siginfo_t const *kinfo) { - long err; + struct signalfd_siginfo new; BUILD_BUG_ON(sizeof(struct signalfd_siginfo) != 128); /* * Unused members should be zero ... */ - err = __clear_user(uinfo, sizeof(*uinfo)); + memset(&new, 0, sizeof(new)); /* * If you change siginfo_t structure, please be sure * this code is fixed accordingly. */ - err |= __put_user(kinfo->si_signo, &uinfo->ssi_signo); - err |= __put_user(kinfo->si_errno, &uinfo->ssi_errno); - err |= __put_user(kinfo->si_code, &uinfo->ssi_code); + new.ssi_signo = kinfo->si_signo; + new.ssi_errno = kinfo->si_errno; + new.ssi_code = kinfo->si_code; switch (siginfo_layout(kinfo->si_signo, kinfo->si_code)) { case SIL_KILL: - err |= __put_user(kinfo->si_pid, &uinfo->ssi_pid); - err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); + new.ssi_pid = kinfo->si_pid; + new.ssi_uid = kinfo->si_uid; break; case SIL_TIMER: - err |= __put_user(kinfo->si_tid, &uinfo->ssi_tid); - err |= __put_user(kinfo->si_overrun, &uinfo->ssi_overrun); - err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); - err |= __put_user(kinfo->si_int, &uinfo->ssi_int); + new.ssi_tid = kinfo->si_tid; + new.ssi_overrun = kinfo->si_overrun; + new.ssi_ptr = (long) kinfo->si_ptr; + new.ssi_int = kinfo->si_int; break; case SIL_POLL: - err |= __put_user(kinfo->si_band, &uinfo->ssi_band); - err |= __put_user(kinfo->si_fd, &uinfo->ssi_fd); + new.ssi_band = kinfo->si_band; + new.ssi_fd = kinfo->si_fd; break; - case SIL_FAULT: - err |= __put_user((long) kinfo->si_addr, &uinfo->ssi_addr); -#ifdef __ARCH_SI_TRAPNO - err |= __put_user(kinfo->si_trapno, &uinfo->ssi_trapno); -#endif -#ifdef BUS_MCEERR_AO + case SIL_FAULT_BNDERR: + case SIL_FAULT_PKUERR: /* - * Other callers might not initialize the si_lsb field, - * so check explicitly for the right codes here. + * Fall through to the SIL_FAULT case. Both SIL_FAULT_BNDERR + * and SIL_FAULT_PKUERR are only generated by faults that + * deliver them synchronously to userspace. In case someone + * injects one of these signals and signalfd catches it treat + * it as SIL_FAULT. */ - if (kinfo->si_signo == SIGBUS && - kinfo->si_code == BUS_MCEERR_AO) - err |= __put_user((short) kinfo->si_addr_lsb, - &uinfo->ssi_addr_lsb); + case SIL_FAULT: + new.ssi_addr = (long) kinfo->si_addr; +#ifdef __ARCH_SI_TRAPNO + new.ssi_trapno = kinfo->si_trapno; #endif -#ifdef BUS_MCEERR_AR - /* - * Other callers might not initialize the si_lsb field, - * so check explicitly for the right codes here. - */ - if (kinfo->si_signo == SIGBUS && - kinfo->si_code == BUS_MCEERR_AR) - err |= __put_user((short) kinfo->si_addr_lsb, - &uinfo->ssi_addr_lsb); + break; + case SIL_FAULT_MCEERR: + new.ssi_addr = (long) kinfo->si_addr; +#ifdef __ARCH_SI_TRAPNO + new.ssi_trapno = kinfo->si_trapno; #endif + new.ssi_addr_lsb = (short) kinfo->si_addr_lsb; break; case SIL_CHLD: - err |= __put_user(kinfo->si_pid, &uinfo->ssi_pid); - err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); - err |= __put_user(kinfo->si_status, &uinfo->ssi_status); - err |= __put_user(kinfo->si_utime, &uinfo->ssi_utime); - err |= __put_user(kinfo->si_stime, &uinfo->ssi_stime); + new.ssi_pid = kinfo->si_pid; + new.ssi_uid = kinfo->si_uid; + new.ssi_status = kinfo->si_status; + new.ssi_utime = kinfo->si_utime; + new.ssi_stime = kinfo->si_stime; break; case SIL_RT: - default: /* * This case catches also the signals queued by sigqueue(). */ - err |= __put_user(kinfo->si_pid, &uinfo->ssi_pid); - err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); - err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); - err |= __put_user(kinfo->si_int, &uinfo->ssi_int); + new.ssi_pid = kinfo->si_pid; + new.ssi_uid = kinfo->si_uid; + new.ssi_ptr = (long) kinfo->si_ptr; + new.ssi_int = kinfo->si_int; + break; + case SIL_SYS: + new.ssi_call_addr = (long) kinfo->si_call_addr; + new.ssi_syscall = kinfo->si_syscall; + new.ssi_arch = kinfo->si_arch; break; } - return err ? -EFAULT: sizeof(*uinfo); + if (copy_to_user(uinfo, &new, sizeof(struct signalfd_siginfo))) + return -EFAULT; + + return sizeof(*uinfo); } static ssize_t signalfd_dequeue(struct signalfd_ctx *ctx, siginfo_t *info, diff --git a/fs/super.c b/fs/super.c index 672538ca9831..50728d9c1a05 100644 --- a/fs/super.c +++ b/fs/super.c @@ -37,6 +37,7 @@ #include <linux/user_namespace.h> #include "internal.h" +static int thaw_super_locked(struct super_block *sb); static LIST_HEAD(super_blocks); static DEFINE_SPINLOCK(sb_lock); @@ -120,13 +121,23 @@ static unsigned long super_cache_count(struct shrinker *shrink, sb = container_of(shrink, struct super_block, s_shrink); /* - * Don't call trylock_super as it is a potential - * scalability bottleneck. The counts could get updated - * between super_cache_count and super_cache_scan anyway. - * Call to super_cache_count with shrinker_rwsem held - * ensures the safety of call to list_lru_shrink_count() and - * s_op->nr_cached_objects(). + * We don't call trylock_super() here as it is a scalability bottleneck, + * so we're exposed to partial setup state. The shrinker rwsem does not + * protect filesystem operations backing list_lru_shrink_count() or + * s_op->nr_cached_objects(). Counts can change between + * super_cache_count and super_cache_scan, so we really don't need locks + * here. + * + * However, if we are currently mounting the superblock, the underlying + * filesystem might be in a state of partial construction and hence it + * is dangerous to access it. trylock_super() uses a SB_BORN check to + * avoid this situation, so do the same here. The memory barrier is + * matched with the one in mount_fs() as we don't hold locks here. */ + if (!(sb->s_flags & SB_BORN)) + return 0; + smp_rmb(); + if (sb->s_op && sb->s_op->nr_cached_objects) total_objects = sb->s_op->nr_cached_objects(sb, sc); @@ -166,6 +177,7 @@ static void destroy_unused_super(struct super_block *s) security_sb_free(s); put_user_ns(s->s_user_ns); kfree(s->s_subtype); + free_prealloced_shrinker(&s->s_shrink); /* no delays needed */ destroy_super_work(&s->destroy_work); } @@ -251,6 +263,8 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags, s->s_shrink.count_objects = super_cache_count; s->s_shrink.batch = 1024; s->s_shrink.flags = SHRINKER_NUMA_AWARE | SHRINKER_MEMCG_AWARE; + if (prealloc_shrinker(&s->s_shrink)) + goto fail; return s; fail: @@ -517,11 +531,7 @@ retry: hlist_add_head(&s->s_instances, &type->fs_supers); spin_unlock(&sb_lock); get_filesystem(type); - err = register_shrinker(&s->s_shrink); - if (err) { - deactivate_locked_super(s); - s = ERR_PTR(err); - } + register_shrinker_prepared(&s->s_shrink); return s; } @@ -574,6 +584,28 @@ void drop_super_exclusive(struct super_block *sb) } EXPORT_SYMBOL(drop_super_exclusive); +static void __iterate_supers(void (*f)(struct super_block *)) +{ + struct super_block *sb, *p = NULL; + + spin_lock(&sb_lock); + list_for_each_entry(sb, &super_blocks, s_list) { + if (hlist_unhashed(&sb->s_instances)) + continue; + sb->s_count++; + spin_unlock(&sb_lock); + + f(sb); + + spin_lock(&sb_lock); + if (p) + __put_super(p); + p = sb; + } + if (p) + __put_super(p); + spin_unlock(&sb_lock); +} /** * iterate_supers - call function for all active superblocks * @f: function to call @@ -881,33 +913,22 @@ cancel_readonly: return retval; } -static void do_emergency_remount(struct work_struct *work) +static void do_emergency_remount_callback(struct super_block *sb) { - struct super_block *sb, *p = NULL; - - spin_lock(&sb_lock); - list_for_each_entry(sb, &super_blocks, s_list) { - if (hlist_unhashed(&sb->s_instances)) - continue; - sb->s_count++; - spin_unlock(&sb_lock); - down_write(&sb->s_umount); - if (sb->s_root && sb->s_bdev && (sb->s_flags & SB_BORN) && - !sb_rdonly(sb)) { - /* - * What lock protects sb->s_flags?? - */ - do_remount_sb(sb, SB_RDONLY, NULL, 1); - } - up_write(&sb->s_umount); - spin_lock(&sb_lock); - if (p) - __put_super(p); - p = sb; + down_write(&sb->s_umount); + if (sb->s_root && sb->s_bdev && (sb->s_flags & SB_BORN) && + !sb_rdonly(sb)) { + /* + * What lock protects sb->s_flags?? + */ + do_remount_sb(sb, SB_RDONLY, NULL, 1); } - if (p) - __put_super(p); - spin_unlock(&sb_lock); + up_write(&sb->s_umount); +} + +static void do_emergency_remount(struct work_struct *work) +{ + __iterate_supers(do_emergency_remount_callback); kfree(work); printk("Emergency Remount complete\n"); } @@ -923,6 +944,40 @@ void emergency_remount(void) } } +static void do_thaw_all_callback(struct super_block *sb) +{ + down_write(&sb->s_umount); + if (sb->s_root && sb->s_flags & SB_BORN) { + emergency_thaw_bdev(sb); + thaw_super_locked(sb); + } else { + up_write(&sb->s_umount); + } +} + +static void do_thaw_all(struct work_struct *work) +{ + __iterate_supers(do_thaw_all_callback); + kfree(work); + printk(KERN_WARNING "Emergency Thaw complete\n"); +} + +/** + * emergency_thaw_all -- forcibly thaw every frozen filesystem + * + * Used for emergency unfreeze of all filesystems via SysRq + */ +void emergency_thaw_all(void) +{ + struct work_struct *work; + + work = kmalloc(sizeof(*work), GFP_ATOMIC); + if (work) { + INIT_WORK(work, do_thaw_all); + schedule_work(work); + } +} + /* * Unnamed block devices are dummy devices used by virtual * filesystems which don't use real block-devices. -- jrs @@ -1227,6 +1282,14 @@ mount_fs(struct file_system_type *type, int flags, const char *name, void *data) sb = root->d_sb; BUG_ON(!sb); WARN_ON(!sb->s_bdi); + + /* + * Write barrier is for super_cache_count(). We place it before setting + * SB_BORN as the data dependency between the two functions is the + * superblock structure contents that we just set up, not the SB_BORN + * flag. + */ + smp_wmb(); sb->s_flags |= SB_BORN; error = security_sb_kern_mount(sb, flags, secdata); @@ -1492,11 +1555,10 @@ EXPORT_SYMBOL(freeze_super); * * Unlocks the filesystem and marks it writeable again after freeze_super(). */ -int thaw_super(struct super_block *sb) +static int thaw_super_locked(struct super_block *sb) { int error; - down_write(&sb->s_umount); if (sb->s_writers.frozen != SB_FREEZE_COMPLETE) { up_write(&sb->s_umount); return -EINVAL; @@ -1527,4 +1589,10 @@ out: deactivate_locked_super(sb); return 0; } + +int thaw_super(struct super_block *sb) +{ + down_write(&sb->s_umount); + return thaw_super_locked(sb); +} EXPORT_SYMBOL(thaw_super); diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index b428d317ae92..92682fcc41f6 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -25,7 +25,7 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, { struct dentry *root; void *ns; - bool new_sb; + bool new_sb = false; if (!(flags & SB_KERNMOUNT)) { if (!kobj_ns_current_may_mount(KOBJ_NS_TYPE_NET)) @@ -35,9 +35,9 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, ns = kobj_ns_grab_current(KOBJ_NS_TYPE_NET); root = kernfs_mount_ns(fs_type, flags, sysfs_root, SYSFS_MAGIC, &new_sb, ns); - if (IS_ERR(root) || !new_sb) + if (!new_sb) kobj_ns_drop(KOBJ_NS_TYPE_NET, ns); - else if (new_sb) + else if (!IS_ERR(root)) root->d_sb->s_iflags |= SB_I_USERNS_VISIBLE; return root; diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index 250b0755b908..4d5d20491ffd 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -51,14 +51,9 @@ static struct dentry *sysv_lookup(struct inode * dir, struct dentry * dentry, un if (dentry->d_name.len > SYSV_NAMELEN) return ERR_PTR(-ENAMETOOLONG); ino = sysv_inode_by_name(dentry); - - if (ino) { + if (ino) inode = sysv_iget(dir->i_sb, ino); - if (IS_ERR(inode)) - return ERR_CAST(inode); - } - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } static int sysv_mknod(struct inode * dir, struct dentry * dentry, umode_t mode, dev_t rdev) diff --git a/fs/timerfd.c b/fs/timerfd.c index cdad49da3ff7..d84a2bee4f82 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -226,21 +226,20 @@ static int timerfd_release(struct inode *inode, struct file *file) kfree_rcu(ctx, rcu); return 0; } - -static __poll_t timerfd_poll(struct file *file, poll_table *wait) + +static struct wait_queue_head *timerfd_get_poll_head(struct file *file, + __poll_t eventmask) { struct timerfd_ctx *ctx = file->private_data; - __poll_t events = 0; - unsigned long flags; - poll_wait(file, &ctx->wqh, wait); + return &ctx->wqh; +} - spin_lock_irqsave(&ctx->wqh.lock, flags); - if (ctx->ticks) - events |= EPOLLIN; - spin_unlock_irqrestore(&ctx->wqh.lock, flags); +static __poll_t timerfd_poll_mask(struct file *file, __poll_t eventmask) +{ + struct timerfd_ctx *ctx = file->private_data; - return events; + return ctx->ticks ? EPOLLIN : 0; } static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count, @@ -364,7 +363,8 @@ static long timerfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg static const struct file_operations timerfd_fops = { .release = timerfd_release, - .poll = timerfd_poll, + .get_poll_head = timerfd_get_poll_head, + .poll_mask = timerfd_poll_mask, .read = timerfd_read, .llseek = noop_llseek, .show_fdinfo = timerfd_show, diff --git a/fs/ubifs/crypto.c b/fs/ubifs/crypto.c index 616a688f5d8f..55c508fe8131 100644 --- a/fs/ubifs/crypto.c +++ b/fs/ubifs/crypto.c @@ -24,14 +24,6 @@ static bool ubifs_crypt_empty_dir(struct inode *inode) return ubifs_check_dir_empty(inode) == 0; } -static unsigned int ubifs_crypt_max_namelen(struct inode *inode) -{ - if (S_ISLNK(inode->i_mode)) - return UBIFS_MAX_INO_DATA; - else - return UBIFS_MAX_NLEN; -} - int ubifs_encrypt(const struct inode *inode, struct ubifs_data_node *dn, unsigned int in_len, unsigned int *out_len, int block) { @@ -89,5 +81,5 @@ const struct fscrypt_operations ubifs_crypt_operations = { .get_context = ubifs_crypt_get_context, .set_context = ubifs_crypt_set_context, .empty_dir = ubifs_crypt_empty_dir, - .max_namelen = ubifs_crypt_max_namelen, + .max_namelen = UBIFS_MAX_NLEN, }; diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 9d7fb88e172e..4e267cc21c77 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -214,7 +214,7 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry, int err; union ubifs_key key; struct inode *inode = NULL; - struct ubifs_dent_node *dent; + struct ubifs_dent_node *dent = NULL; struct ubifs_info *c = dir->i_sb->s_fs_info; struct fscrypt_name nm; @@ -229,14 +229,14 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(err); if (fname_len(&nm) > UBIFS_MAX_NLEN) { - err = -ENAMETOOLONG; - goto out_fname; + inode = ERR_PTR(-ENAMETOOLONG); + goto done; } dent = kmalloc(UBIFS_MAX_DENT_NODE_SZ, GFP_NOFS); if (!dent) { - err = -ENOMEM; - goto out_fname; + inode = ERR_PTR(-ENOMEM); + goto done; } if (nm.hash) { @@ -250,16 +250,16 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry, } if (err) { - if (err == -ENOENT) { + if (err == -ENOENT) dbg_gen("not found"); - goto done; - } - goto out_dent; + else + inode = ERR_PTR(err); + goto done; } if (dbg_check_name(c, dent, &nm)) { - err = -EINVAL; - goto out_dent; + inode = ERR_PTR(-EINVAL); + goto done; } inode = ubifs_iget(dir->i_sb, le64_to_cpu(dent->inum)); @@ -272,7 +272,7 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry, ubifs_err(c, "dead directory entry '%pd', error %d", dentry, err); ubifs_ro_mode(c, err); - goto out_dent; + goto done; } if (ubifs_crypt_is_encrypted(dir) && @@ -280,27 +280,14 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry, !fscrypt_has_permitted_context(dir, inode)) { ubifs_warn(c, "Inconsistent encryption contexts: %lu/%lu", dir->i_ino, inode->i_ino); - err = -EPERM; - goto out_inode; + iput(inode); + inode = ERR_PTR(-EPERM); } done: kfree(dent); fscrypt_free_filename(&nm); - /* - * Note, d_splice_alias() would be required instead if we supported - * NFS. - */ - d_add(dentry, inode); - return NULL; - -out_inode: - iput(inode); -out_dent: - kfree(dent); -out_fname: - fscrypt_free_filename(&nm); - return ERR_PTR(err); + return d_splice_alias(inode, dentry); } static int ubifs_create(struct inode *dir, struct dentry *dentry, umode_t mode, diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index cf348ba99238..1acb2ff505e6 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1256,7 +1256,7 @@ static int do_setattr(struct ubifs_info *c, struct inode *inode, * Inode length changed, so we have to make sure * @I_DIRTY_DATASYNC is set. */ - __mark_inode_dirty(inode, I_DIRTY_SYNC | I_DIRTY_DATASYNC); + __mark_inode_dirty(inode, I_DIRTY_DATASYNC); else mark_inode_dirty_sync(inode); mutex_unlock(&ui->ui_mutex); diff --git a/fs/ubifs/find.c b/fs/ubifs/find.c index 2dcf3d473fec..9571616b5dda 100644 --- a/fs/ubifs/find.c +++ b/fs/ubifs/find.c @@ -632,7 +632,7 @@ static int scan_for_idx_cb(struct ubifs_info *c, */ static const struct ubifs_lprops *scan_for_leb_for_idx(struct ubifs_info *c) { - struct ubifs_lprops *lprops; + const struct ubifs_lprops *lprops; struct scan_data data; int err; diff --git a/fs/ubifs/lprops.c b/fs/ubifs/lprops.c index 6c3a1abd0e22..f5a46844340c 100644 --- a/fs/ubifs/lprops.c +++ b/fs/ubifs/lprops.c @@ -244,7 +244,6 @@ static void remove_from_lpt_heap(struct ubifs_info *c, /** * lpt_heap_replace - replace lprops in a category heap. * @c: UBIFS file-system description object - * @old_lprops: LEB properties to replace * @new_lprops: LEB properties with which to replace * @cat: LEB category * @@ -254,7 +253,6 @@ static void remove_from_lpt_heap(struct ubifs_info *c, * lprops. This function does that. */ static void lpt_heap_replace(struct ubifs_info *c, - struct ubifs_lprops *old_lprops, struct ubifs_lprops *new_lprops, int cat) { struct ubifs_lpt_heap *heap; @@ -362,7 +360,7 @@ void ubifs_replace_cat(struct ubifs_info *c, struct ubifs_lprops *old_lprops, case LPROPS_DIRTY: case LPROPS_DIRTY_IDX: case LPROPS_FREE: - lpt_heap_replace(c, old_lprops, new_lprops, cat); + lpt_heap_replace(c, new_lprops, cat); break; case LPROPS_UNCAT: case LPROPS_EMPTY: diff --git a/fs/ubifs/scan.c b/fs/ubifs/scan.c index aab87340d3de..16f03d9929e5 100644 --- a/fs/ubifs/scan.c +++ b/fs/ubifs/scan.c @@ -175,7 +175,6 @@ struct ubifs_scan_leb *ubifs_start_scan(const struct ubifs_info *c, int lnum, void ubifs_end_scan(const struct ubifs_info *c, struct ubifs_scan_leb *sleb, int lnum, int offs) { - lnum = lnum; dbg_scan("stop scanning LEB %d at offset %d", lnum, offs); ubifs_assert(offs % c->min_io_size == 0); diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index b16ef162344a..6c397a389105 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -1737,8 +1737,11 @@ static void ubifs_remount_ro(struct ubifs_info *c) dbg_save_space_info(c); - for (i = 0; i < c->jhead_cnt; i++) - ubifs_wbuf_sync(&c->jheads[i].wbuf); + for (i = 0; i < c->jhead_cnt; i++) { + err = ubifs_wbuf_sync(&c->jheads[i].wbuf); + if (err) + ubifs_ro_mode(c, err); + } c->mst_node->flags &= ~cpu_to_le32(UBIFS_MST_DIRTY); c->mst_node->flags |= cpu_to_le32(UBIFS_MST_NO_ORPHS); @@ -1804,8 +1807,11 @@ static void ubifs_put_super(struct super_block *sb) int err; /* Synchronize write-buffers */ - for (i = 0; i < c->jhead_cnt; i++) - ubifs_wbuf_sync(&c->jheads[i].wbuf); + for (i = 0; i < c->jhead_cnt; i++) { + err = ubifs_wbuf_sync(&c->jheads[i].wbuf); + if (err) + ubifs_ro_mode(c, err); + } /* * We are being cleanly unmounted which means the diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 0458dd47e105..c586026508db 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -622,8 +622,7 @@ static int udf_add_nondir(struct dentry *dentry, struct inode *inode) if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } @@ -733,8 +732,7 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) inc_nlink(dir); dir->i_ctime = dir->i_mtime = current_time(dir); mark_inode_dirty(dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c index f897e55f2cd0..16a8ad21b77e 100644 --- a/fs/udf/unicode.c +++ b/fs/udf/unicode.c @@ -28,6 +28,9 @@ #include "udf_sb.h" +#define SURROGATE_MASK 0xfffff800 +#define SURROGATE_PAIR 0x0000d800 + static int udf_uni2char_utf8(wchar_t uni, unsigned char *out, int boundlen) @@ -37,6 +40,9 @@ static int udf_uni2char_utf8(wchar_t uni, if (boundlen <= 0) return -ENAMETOOLONG; + if ((uni & SURROGATE_MASK) == SURROGATE_PAIR) + return -EINVAL; + if (uni < 0x80) { out[u_len++] = (unsigned char)uni; } else if (uni < 0x800) { diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index 32545cd00ceb..d5f43ba76c59 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -39,8 +39,7 @@ static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode) { int err = ufs_add_link(dentry, inode); if (!err) { - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -193,8 +192,7 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) if (err) goto out_fail; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; out_fail: diff --git a/fs/xattr.c b/fs/xattr.c index 61cd28ba25f3..f9cb1db187b7 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -229,7 +229,7 @@ out: } EXPORT_SYMBOL_GPL(vfs_setxattr); -ssize_t +static ssize_t xattr_getsecurity(struct inode *inode, const char *name, void *value, size_t size) { @@ -254,7 +254,6 @@ out: out_noalloc: return len; } -EXPORT_SYMBOL_GPL(xattr_getsecurity); /* * vfs_getxattr_alloc - allocate memory, if necessary, before calling getxattr @@ -354,7 +353,6 @@ vfs_listxattr(struct dentry *dentry, char *list, size_t size) if (error) return error; if (inode->i_op->listxattr && (inode->i_opflags & IOP_XATTR)) { - error = -EOPNOTSUPP; error = inode->i_op->listxattr(dentry, list, size); } else { error = security_inode_listsecurity(inode, list, size); diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig index 46bcf0e649f5..457ac9f97377 100644 --- a/fs/xfs/Kconfig +++ b/fs/xfs/Kconfig @@ -85,6 +85,24 @@ config XFS_ONLINE_SCRUB If unsure, say N. +config XFS_ONLINE_REPAIR + bool "XFS online metadata repair support" + default n + depends on XFS_FS && XFS_ONLINE_SCRUB + help + If you say Y here you will be able to repair metadata on a + mounted XFS filesystem. This feature is intended to reduce + filesystem downtime by fixing minor problems before they cause the + filesystem to go down. However, it requires that the filesystem be + formatted with secondary metadata, such as reverse mappings and inode + parent pointers. + + This feature is considered EXPERIMENTAL. Use with caution! + + See the xfs_scrub man page in section 8 for additional information. + + If unsure, say N. + config XFS_WARN bool "XFS Verbose Warnings" depends on XFS_FS && !XFS_DEBUG diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 7ceb41a9786a..e8d67a443bd7 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -28,6 +28,7 @@ xfs-y += xfs_trace.o # build the libxfs code first xfs-y += $(addprefix libxfs/, \ + xfs_ag.o \ xfs_alloc.o \ xfs_alloc_btree.o \ xfs_attr.o \ @@ -163,4 +164,12 @@ xfs-y += $(addprefix scrub/, \ xfs-$(CONFIG_XFS_RT) += scrub/rtbitmap.o xfs-$(CONFIG_XFS_QUOTA) += scrub/quota.o + +# online repair +ifeq ($(CONFIG_XFS_ONLINE_REPAIR),y) +xfs-y += $(addprefix scrub/, \ + agheader_repair.o \ + repair.o \ + ) +endif endif diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c new file mode 100644 index 000000000000..9345802c99f7 --- /dev/null +++ b/fs/xfs/libxfs/xfs_ag.c @@ -0,0 +1,464 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2018 Red Hat, Inc. + * All rights reserved. + */ + +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_sb.h" +#include "xfs_mount.h" +#include "xfs_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_rmap_btree.h" +#include "xfs_alloc.h" +#include "xfs_ialloc.h" +#include "xfs_rmap.h" +#include "xfs_ag.h" + +static struct xfs_buf * +xfs_get_aghdr_buf( + struct xfs_mount *mp, + xfs_daddr_t blkno, + size_t numblks, + int flags, + const struct xfs_buf_ops *ops) +{ + struct xfs_buf *bp; + + bp = xfs_buf_get_uncached(mp->m_ddev_targp, numblks, flags); + if (!bp) + return NULL; + + xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); + bp->b_bn = blkno; + bp->b_maps[0].bm_bn = blkno; + bp->b_ops = ops; + + return bp; +} + +/* + * Generic btree root block init function + */ +static void +xfs_btroot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + xfs_btree_init_block(mp, bp, id->type, 0, 0, id->agno, 0); +} + +/* + * Alloc btree root block init functions + */ +static void +xfs_bnoroot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_alloc_rec *arec; + + xfs_btree_init_block(mp, bp, XFS_BTNUM_BNO, 0, 1, id->agno, 0); + arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); + arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); + arec->ar_blockcount = cpu_to_be32(id->agsize - + be32_to_cpu(arec->ar_startblock)); +} + +static void +xfs_cntroot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_alloc_rec *arec; + + xfs_btree_init_block(mp, bp, XFS_BTNUM_CNT, 0, 1, id->agno, 0); + arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); + arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); + arec->ar_blockcount = cpu_to_be32(id->agsize - + be32_to_cpu(arec->ar_startblock)); +} + +/* + * Reverse map root block init + */ +static void +xfs_rmaproot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); + struct xfs_rmap_rec *rrec; + + xfs_btree_init_block(mp, bp, XFS_BTNUM_RMAP, 0, 4, id->agno, 0); + + /* + * mark the AG header regions as static metadata The BNO + * btree block is the first block after the headers, so + * it's location defines the size of region the static + * metadata consumes. + * + * Note: unlike mkfs, we never have to account for log + * space when growing the data regions + */ + rrec = XFS_RMAP_REC_ADDR(block, 1); + rrec->rm_startblock = 0; + rrec->rm_blockcount = cpu_to_be32(XFS_BNO_BLOCK(mp)); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_FS); + rrec->rm_offset = 0; + + /* account freespace btree root blocks */ + rrec = XFS_RMAP_REC_ADDR(block, 2); + rrec->rm_startblock = cpu_to_be32(XFS_BNO_BLOCK(mp)); + rrec->rm_blockcount = cpu_to_be32(2); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG); + rrec->rm_offset = 0; + + /* account inode btree root blocks */ + rrec = XFS_RMAP_REC_ADDR(block, 3); + rrec->rm_startblock = cpu_to_be32(XFS_IBT_BLOCK(mp)); + rrec->rm_blockcount = cpu_to_be32(XFS_RMAP_BLOCK(mp) - + XFS_IBT_BLOCK(mp)); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_INOBT); + rrec->rm_offset = 0; + + /* account for rmap btree root */ + rrec = XFS_RMAP_REC_ADDR(block, 4); + rrec->rm_startblock = cpu_to_be32(XFS_RMAP_BLOCK(mp)); + rrec->rm_blockcount = cpu_to_be32(1); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG); + rrec->rm_offset = 0; + + /* account for refc btree root */ + if (xfs_sb_version_hasreflink(&mp->m_sb)) { + rrec = XFS_RMAP_REC_ADDR(block, 5); + rrec->rm_startblock = cpu_to_be32(xfs_refc_block(mp)); + rrec->rm_blockcount = cpu_to_be32(1); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_REFC); + rrec->rm_offset = 0; + be16_add_cpu(&block->bb_numrecs, 1); + } +} + +/* + * Initialise new secondary superblocks with the pre-grow geometry, but mark + * them as "in progress" so we know they haven't yet been activated. This will + * get cleared when the update with the new geometry information is done after + * changes to the primary are committed. This isn't strictly necessary, but we + * get it for free with the delayed buffer write lists and it means we can tell + * if a grow operation didn't complete properly after the fact. + */ +static void +xfs_sbblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp); + + xfs_sb_to_disk(dsb, &mp->m_sb); + dsb->sb_inprogress = 1; +} + +static void +xfs_agfblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_agf *agf = XFS_BUF_TO_AGF(bp); + xfs_extlen_t tmpsize; + + agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC); + agf->agf_versionnum = cpu_to_be32(XFS_AGF_VERSION); + agf->agf_seqno = cpu_to_be32(id->agno); + agf->agf_length = cpu_to_be32(id->agsize); + agf->agf_roots[XFS_BTNUM_BNOi] = cpu_to_be32(XFS_BNO_BLOCK(mp)); + agf->agf_roots[XFS_BTNUM_CNTi] = cpu_to_be32(XFS_CNT_BLOCK(mp)); + agf->agf_levels[XFS_BTNUM_BNOi] = cpu_to_be32(1); + agf->agf_levels[XFS_BTNUM_CNTi] = cpu_to_be32(1); + if (xfs_sb_version_hasrmapbt(&mp->m_sb)) { + agf->agf_roots[XFS_BTNUM_RMAPi] = + cpu_to_be32(XFS_RMAP_BLOCK(mp)); + agf->agf_levels[XFS_BTNUM_RMAPi] = cpu_to_be32(1); + agf->agf_rmap_blocks = cpu_to_be32(1); + } + + agf->agf_flfirst = cpu_to_be32(1); + agf->agf_fllast = 0; + agf->agf_flcount = 0; + tmpsize = id->agsize - mp->m_ag_prealloc_blocks; + agf->agf_freeblks = cpu_to_be32(tmpsize); + agf->agf_longest = cpu_to_be32(tmpsize); + if (xfs_sb_version_hascrc(&mp->m_sb)) + uuid_copy(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid); + if (xfs_sb_version_hasreflink(&mp->m_sb)) { + agf->agf_refcount_root = cpu_to_be32( + xfs_refc_block(mp)); + agf->agf_refcount_level = cpu_to_be32(1); + agf->agf_refcount_blocks = cpu_to_be32(1); + } +} + +static void +xfs_agflblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_agfl *agfl = XFS_BUF_TO_AGFL(bp); + __be32 *agfl_bno; + int bucket; + + if (xfs_sb_version_hascrc(&mp->m_sb)) { + agfl->agfl_magicnum = cpu_to_be32(XFS_AGFL_MAGIC); + agfl->agfl_seqno = cpu_to_be32(id->agno); + uuid_copy(&agfl->agfl_uuid, &mp->m_sb.sb_meta_uuid); + } + + agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, bp); + for (bucket = 0; bucket < xfs_agfl_size(mp); bucket++) + agfl_bno[bucket] = cpu_to_be32(NULLAGBLOCK); +} + +static void +xfs_agiblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_agi *agi = XFS_BUF_TO_AGI(bp); + int bucket; + + agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC); + agi->agi_versionnum = cpu_to_be32(XFS_AGI_VERSION); + agi->agi_seqno = cpu_to_be32(id->agno); + agi->agi_length = cpu_to_be32(id->agsize); + agi->agi_count = 0; + agi->agi_root = cpu_to_be32(XFS_IBT_BLOCK(mp)); + agi->agi_level = cpu_to_be32(1); + agi->agi_freecount = 0; + agi->agi_newino = cpu_to_be32(NULLAGINO); + agi->agi_dirino = cpu_to_be32(NULLAGINO); + if (xfs_sb_version_hascrc(&mp->m_sb)) + uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_meta_uuid); + if (xfs_sb_version_hasfinobt(&mp->m_sb)) { + agi->agi_free_root = cpu_to_be32(XFS_FIBT_BLOCK(mp)); + agi->agi_free_level = cpu_to_be32(1); + } + for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) + agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO); +} + +typedef void (*aghdr_init_work_f)(struct xfs_mount *mp, struct xfs_buf *bp, + struct aghdr_init_data *id); +static int +xfs_ag_init_hdr( + struct xfs_mount *mp, + struct aghdr_init_data *id, + aghdr_init_work_f work, + const struct xfs_buf_ops *ops) + +{ + struct xfs_buf *bp; + + bp = xfs_get_aghdr_buf(mp, id->daddr, id->numblks, 0, ops); + if (!bp) + return -ENOMEM; + + (*work)(mp, bp, id); + + xfs_buf_delwri_queue(bp, &id->buffer_list); + xfs_buf_relse(bp); + return 0; +} + +struct xfs_aghdr_grow_data { + xfs_daddr_t daddr; + size_t numblks; + const struct xfs_buf_ops *ops; + aghdr_init_work_f work; + xfs_btnum_t type; + bool need_init; +}; + +/* + * Prepare new AG headers to be written to disk. We use uncached buffers here, + * as it is assumed these new AG headers are currently beyond the currently + * valid filesystem address space. Using cached buffers would trip over EOFS + * corruption detection alogrithms in the buffer cache lookup routines. + * + * This is a non-transactional function, but the prepared buffers are added to a + * delayed write buffer list supplied by the caller so they can submit them to + * disk and wait on them as required. + */ +int +xfs_ag_init_headers( + struct xfs_mount *mp, + struct aghdr_init_data *id) + +{ + struct xfs_aghdr_grow_data aghdr_data[] = { + { /* SB */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_SB_DADDR), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_sb_buf_ops, + .work = &xfs_sbblock_init, + .need_init = true + }, + { /* AGF */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_AGF_DADDR(mp)), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_agf_buf_ops, + .work = &xfs_agfblock_init, + .need_init = true + }, + { /* AGFL */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_AGFL_DADDR(mp)), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_agfl_buf_ops, + .work = &xfs_agflblock_init, + .need_init = true + }, + { /* AGI */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_AGI_DADDR(mp)), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_agi_buf_ops, + .work = &xfs_agiblock_init, + .need_init = true + }, + { /* BNO root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_BNO_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_allocbt_buf_ops, + .work = &xfs_bnoroot_init, + .need_init = true + }, + { /* CNT root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_CNT_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_allocbt_buf_ops, + .work = &xfs_cntroot_init, + .need_init = true + }, + { /* INO root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_IBT_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_inobt_buf_ops, + .work = &xfs_btroot_init, + .type = XFS_BTNUM_INO, + .need_init = true + }, + { /* FINO root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_FIBT_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_inobt_buf_ops, + .work = &xfs_btroot_init, + .type = XFS_BTNUM_FINO, + .need_init = xfs_sb_version_hasfinobt(&mp->m_sb) + }, + { /* RMAP root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_RMAP_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_rmapbt_buf_ops, + .work = &xfs_rmaproot_init, + .need_init = xfs_sb_version_hasrmapbt(&mp->m_sb) + }, + { /* REFC root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, xfs_refc_block(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_refcountbt_buf_ops, + .work = &xfs_btroot_init, + .type = XFS_BTNUM_REFC, + .need_init = xfs_sb_version_hasreflink(&mp->m_sb) + }, + { /* NULL terminating block */ + .daddr = XFS_BUF_DADDR_NULL, + } + }; + struct xfs_aghdr_grow_data *dp; + int error = 0; + + /* Account for AG free space in new AG */ + id->nfree += id->agsize - mp->m_ag_prealloc_blocks; + for (dp = &aghdr_data[0]; dp->daddr != XFS_BUF_DADDR_NULL; dp++) { + if (!dp->need_init) + continue; + + id->daddr = dp->daddr; + id->numblks = dp->numblks; + id->type = dp->type; + error = xfs_ag_init_hdr(mp, id, dp->work, dp->ops); + if (error) + break; + } + return error; +} + +/* + * Extent the AG indicated by the @id by the length passed in + */ +int +xfs_ag_extend_space( + struct xfs_mount *mp, + struct xfs_trans *tp, + struct aghdr_init_data *id, + xfs_extlen_t len) +{ + struct xfs_owner_info oinfo; + struct xfs_buf *bp; + struct xfs_agi *agi; + struct xfs_agf *agf; + int error; + + /* + * Change the agi length. + */ + error = xfs_ialloc_read_agi(mp, tp, id->agno, &bp); + if (error) + return error; + + agi = XFS_BUF_TO_AGI(bp); + be32_add_cpu(&agi->agi_length, len); + ASSERT(id->agno == mp->m_sb.sb_agcount - 1 || + be32_to_cpu(agi->agi_length) == mp->m_sb.sb_agblocks); + xfs_ialloc_log_agi(tp, bp, XFS_AGI_LENGTH); + + /* + * Change agf length. + */ + error = xfs_alloc_read_agf(mp, tp, id->agno, 0, &bp); + if (error) + return error; + + agf = XFS_BUF_TO_AGF(bp); + be32_add_cpu(&agf->agf_length, len); + ASSERT(agf->agf_length == agi->agi_length); + xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH); + + /* + * Free the new space. + * + * XFS_RMAP_OWN_NULL is used here to tell the rmap btree that + * this doesn't actually exist in the rmap btree. + */ + xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_NULL); + error = xfs_rmap_free(tp, bp, id->agno, + be32_to_cpu(agf->agf_length) - len, + len, &oinfo); + if (error) + return error; + + return xfs_free_extent(tp, XFS_AGB_TO_FSB(mp, id->agno, + be32_to_cpu(agf->agf_length) - len), + len, &oinfo, XFS_AG_RESV_NONE); +} diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h new file mode 100644 index 000000000000..412702e23f61 --- /dev/null +++ b/fs/xfs/libxfs/xfs_ag.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2018 Red Hat, Inc. + * All rights reserved. + */ + +#ifndef __LIBXFS_AG_H +#define __LIBXFS_AG_H 1 + +struct xfs_mount; +struct xfs_trans; + +struct aghdr_init_data { + /* per ag data */ + xfs_agblock_t agno; /* ag to init */ + xfs_extlen_t agsize; /* new AG size */ + struct list_head buffer_list; /* buffer writeback list */ + xfs_rfsblock_t nfree; /* cumulative new free space */ + + /* per header data */ + xfs_daddr_t daddr; /* header location */ + size_t numblks; /* size of header */ + xfs_btnum_t type; /* type of btree root block */ +}; + +int xfs_ag_init_headers(struct xfs_mount *mp, struct aghdr_init_data *id); +int xfs_ag_extend_space(struct xfs_mount *mp, struct xfs_trans *tp, + struct aghdr_init_data *id, xfs_extlen_t len); + +#endif /* __LIBXFS_AG_H */ diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 39387bdd225d..dc9dd3805d97 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -39,6 +39,9 @@ #include "xfs_buf_item.h" #include "xfs_log.h" #include "xfs_ag_resv.h" +#include "xfs_bmap.h" + +extern kmem_zone_t *xfs_bmap_free_item_zone; struct workqueue_struct *xfs_alloc_wq; @@ -1947,7 +1950,7 @@ void xfs_alloc_compute_maxlevels( xfs_mount_t *mp) /* file system mount structure */ { - mp->m_ag_maxlevels = xfs_btree_compute_maxlevels(mp, mp->m_alloc_mnr, + mp->m_ag_maxlevels = xfs_btree_compute_maxlevels(mp->m_alloc_mnr, (mp->m_sb.sb_agblocks + 1) / 2); } @@ -1959,7 +1962,6 @@ xfs_alloc_compute_maxlevels( */ xfs_extlen_t xfs_alloc_longest_free_extent( - struct xfs_mount *mp, struct xfs_perag *pag, xfs_extlen_t need, xfs_extlen_t reserved) @@ -2038,8 +2040,7 @@ xfs_alloc_space_available( /* do we have enough contiguous free space for the allocation? */ alloc_len = args->minlen + (args->alignment - 1) + args->minalignslop; - longest = xfs_alloc_longest_free_extent(args->mp, pag, min_free, - reservation); + longest = xfs_alloc_longest_free_extent(pag, min_free, reservation); if (longest < alloc_len) return false; @@ -2062,6 +2063,30 @@ xfs_alloc_space_available( return true; } +int +xfs_free_agfl_block( + struct xfs_trans *tp, + xfs_agnumber_t agno, + xfs_agblock_t agbno, + struct xfs_buf *agbp, + struct xfs_owner_info *oinfo) +{ + int error; + struct xfs_buf *bp; + + error = xfs_free_ag_extent(tp, agbp, agno, agbno, 1, oinfo, + XFS_AG_RESV_AGFL); + if (error) + return error; + + bp = xfs_btree_get_bufs(tp->t_mountp, tp, agno, agbno, 0); + if (!bp) + return -EFSCORRUPTED; + xfs_trans_binval(tp, bp); + + return 0; +} + /* * Check the agfl fields of the agf for inconsistency or corruption. The purpose * is to detect an agfl header padding mismatch between current and early v5 @@ -2150,6 +2175,40 @@ xfs_agfl_reset( } /* + * Defer an AGFL block free. This is effectively equivalent to + * xfs_bmap_add_free() with some special handling particular to AGFL blocks. + * + * Deferring AGFL frees helps prevent log reservation overruns due to too many + * allocation operations in a transaction. AGFL frees are prone to this problem + * because for one they are always freed one at a time. Further, an immediate + * AGFL block free can cause a btree join and require another block free before + * the real allocation can proceed. Deferring the free disconnects freeing up + * the AGFL slot from freeing the block. + */ +STATIC void +xfs_defer_agfl_block( + struct xfs_mount *mp, + struct xfs_defer_ops *dfops, + xfs_agnumber_t agno, + xfs_fsblock_t agbno, + struct xfs_owner_info *oinfo) +{ + struct xfs_extent_free_item *new; /* new element */ + + ASSERT(xfs_bmap_free_item_zone != NULL); + ASSERT(oinfo != NULL); + + new = kmem_zone_alloc(xfs_bmap_free_item_zone, KM_SLEEP); + new->xefi_startblock = XFS_AGB_TO_FSB(mp, agno, agbno); + new->xefi_blockcount = 1; + new->xefi_oinfo = *oinfo; + + trace_xfs_agfl_free_defer(mp, agno, 0, agbno, 1); + + xfs_defer_add(dfops, XFS_DEFER_OPS_TYPE_AGFL_FREE, &new->xefi_list); +} + +/* * Decide whether to use this allocation group for this allocation. * If so, fix up the btree freelist's size. */ @@ -2249,21 +2308,20 @@ xfs_alloc_fix_freelist( else xfs_rmap_ag_owner(&targs.oinfo, XFS_RMAP_OWN_AG); while (!(flags & XFS_ALLOC_FLAG_NOSHRINK) && pag->pagf_flcount > need) { - struct xfs_buf *bp; - error = xfs_alloc_get_freelist(tp, agbp, &bno, 0); if (error) goto out_agbp_relse; - error = xfs_free_ag_extent(tp, agbp, args->agno, bno, 1, - &targs.oinfo, XFS_AG_RESV_AGFL); - if (error) - goto out_agbp_relse; - bp = xfs_btree_get_bufs(mp, tp, args->agno, bno, 0); - if (!bp) { - error = -EFSCORRUPTED; - goto out_agbp_relse; + + /* defer agfl frees if dfops is provided */ + if (tp->t_agfl_dfops) { + xfs_defer_agfl_block(mp, tp->t_agfl_dfops, args->agno, + bno, &targs.oinfo); + } else { + error = xfs_free_agfl_block(tp, args->agno, bno, agbp, + &targs.oinfo); + if (error) + goto out_agbp_relse; } - xfs_trans_binval(tp, bp); } targs.tp = tp; @@ -2951,18 +3009,20 @@ out: * after fixing up the freelist. */ int /* error */ -xfs_free_extent( +__xfs_free_extent( struct xfs_trans *tp, /* transaction pointer */ xfs_fsblock_t bno, /* starting block number of extent */ xfs_extlen_t len, /* length of extent */ struct xfs_owner_info *oinfo, /* extent owner */ - enum xfs_ag_resv_type type) /* block reservation type */ + enum xfs_ag_resv_type type, /* block reservation type */ + bool skip_discard) { struct xfs_mount *mp = tp->t_mountp; struct xfs_buf *agbp; xfs_agnumber_t agno = XFS_FSB_TO_AGNO(mp, bno); xfs_agblock_t agbno = XFS_FSB_TO_AGBNO(mp, bno); int error; + unsigned int busy_flags = 0; ASSERT(len != 0); ASSERT(type != XFS_AG_RESV_AGFL); @@ -2986,7 +3046,9 @@ xfs_free_extent( if (error) goto err; - xfs_extent_busy_insert(tp, agno, agbno, len, 0); + if (skip_discard) + busy_flags |= XFS_EXTENT_BUSY_SKIP_DISCARD; + xfs_extent_busy_insert(tp, agno, agbno, len, busy_flags); return 0; err: @@ -3118,3 +3180,40 @@ xfs_alloc_has_record( return xfs_btree_has_record(cur, &low, &high, exists); } + +/* + * Walk all the blocks in the AGFL. The @walk_fn can return any negative + * error code or XFS_BTREE_QUERY_RANGE_ABORT. + */ +int +xfs_agfl_walk( + struct xfs_mount *mp, + struct xfs_agf *agf, + struct xfs_buf *agflbp, + xfs_agfl_walk_fn walk_fn, + void *priv) +{ + __be32 *agfl_bno; + unsigned int i; + int error; + + agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agflbp); + i = be32_to_cpu(agf->agf_flfirst); + + /* Nothing to walk in an empty AGFL. */ + if (agf->agf_flcount == cpu_to_be32(0)) + return 0; + + /* Otherwise, walk from first to last, wrapping as needed. */ + for (;;) { + error = walk_fn(mp, be32_to_cpu(agfl_bno[i]), priv); + if (error) + return error; + if (i == be32_to_cpu(agf->agf_fllast)) + break; + if (++i == xfs_agfl_size(mp)) + i = 0; + } + + return 0; +} diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index a311a2414a6b..0747adcd57d6 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -116,9 +116,8 @@ xfs_alloc_allow_busy_reuse(int datatype) unsigned int xfs_alloc_set_aside(struct xfs_mount *mp); unsigned int xfs_alloc_ag_max_usable(struct xfs_mount *mp); -xfs_extlen_t xfs_alloc_longest_free_extent(struct xfs_mount *mp, - struct xfs_perag *pag, xfs_extlen_t need, - xfs_extlen_t reserved); +xfs_extlen_t xfs_alloc_longest_free_extent(struct xfs_perag *pag, + xfs_extlen_t need, xfs_extlen_t reserved); unsigned int xfs_alloc_min_freelist(struct xfs_mount *mp, struct xfs_perag *pag); @@ -192,12 +191,24 @@ xfs_alloc_vextent( * Free an extent. */ int /* error */ -xfs_free_extent( +__xfs_free_extent( struct xfs_trans *tp, /* transaction pointer */ xfs_fsblock_t bno, /* starting block number of extent */ xfs_extlen_t len, /* length of extent */ struct xfs_owner_info *oinfo, /* extent owner */ - enum xfs_ag_resv_type type); /* block reservation type */ + enum xfs_ag_resv_type type, /* block reservation type */ + bool skip_discard); + +static inline int +xfs_free_extent( + struct xfs_trans *tp, + xfs_fsblock_t bno, + xfs_extlen_t len, + struct xfs_owner_info *oinfo, + enum xfs_ag_resv_type type) +{ + return __xfs_free_extent(tp, bno, len, oinfo, type, false); +} int /* error */ xfs_alloc_lookup_le( @@ -224,6 +235,8 @@ int xfs_read_agf(struct xfs_mount *mp, struct xfs_trans *tp, xfs_agnumber_t agno, int flags, struct xfs_buf **bpp); int xfs_alloc_read_agfl(struct xfs_mount *mp, struct xfs_trans *tp, xfs_agnumber_t agno, struct xfs_buf **bpp); +int xfs_free_agfl_block(struct xfs_trans *, xfs_agnumber_t, xfs_agblock_t, + struct xfs_buf *, struct xfs_owner_info *); int xfs_alloc_fix_freelist(struct xfs_alloc_arg *args, int flags); int xfs_free_extent_fix_freelist(struct xfs_trans *tp, xfs_agnumber_t agno, struct xfs_buf **agbp); @@ -249,4 +262,9 @@ bool xfs_verify_fsbno(struct xfs_mount *mp, xfs_fsblock_t fsbno); int xfs_alloc_has_record(struct xfs_btree_cur *cur, xfs_agblock_t bno, xfs_extlen_t len, bool *exist); +typedef int (*xfs_agfl_walk_fn)(struct xfs_mount *mp, xfs_agblock_t bno, + void *priv); +int xfs_agfl_walk(struct xfs_mount *mp, struct xfs_agf *agf, + struct xfs_buf *agflbp, xfs_agfl_walk_fn walk_fn, void *priv); + #endif /* __XFS_ALLOC_H__ */ diff --git a/fs/xfs/libxfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c index b451649ba176..18aec7a0e599 100644 --- a/fs/xfs/libxfs/xfs_alloc_btree.c +++ b/fs/xfs/libxfs/xfs_alloc_btree.c @@ -547,3 +547,12 @@ xfs_allocbt_maxrecs( return blocklen / sizeof(xfs_alloc_rec_t); return blocklen / (sizeof(xfs_alloc_key_t) + sizeof(xfs_alloc_ptr_t)); } + +/* Calculate the freespace btree size for some records. */ +xfs_extlen_t +xfs_allocbt_calc_size( + struct xfs_mount *mp, + unsigned long long len) +{ + return xfs_btree_calc_size(mp->m_alloc_mnr, len); +} diff --git a/fs/xfs/libxfs/xfs_alloc_btree.h b/fs/xfs/libxfs/xfs_alloc_btree.h index 45e189e7e81c..2fd54728871c 100644 --- a/fs/xfs/libxfs/xfs_alloc_btree.h +++ b/fs/xfs/libxfs/xfs_alloc_btree.h @@ -61,5 +61,7 @@ extern struct xfs_btree_cur *xfs_allocbt_init_cursor(struct xfs_mount *, struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t, xfs_btnum_t); extern int xfs_allocbt_maxrecs(struct xfs_mount *, int, int); +extern xfs_extlen_t xfs_allocbt_calc_size(struct xfs_mount *mp, + unsigned long long len); #endif /* __XFS_ALLOC_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index ce4a34a2751d..c3d02a66d39d 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -236,7 +236,7 @@ xfs_attr_set( args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; args.total = xfs_attr_calc_size(&args, &local); - error = xfs_qm_dqattach(dp, 0); + error = xfs_qm_dqattach(dp); if (error) return error; @@ -427,7 +427,7 @@ xfs_attr_remove( */ args.op_flags = XFS_DA_OP_OKNOENT; - error = xfs_qm_dqattach(dp, 0); + error = xfs_qm_dqattach(dp); if (error) return error; @@ -511,7 +511,14 @@ xfs_attr_shortform_addname(xfs_da_args_t *args) if (args->flags & ATTR_CREATE) return retval; retval = xfs_attr_shortform_remove(args); - ASSERT(retval == 0); + if (retval) + return retval; + /* + * Since we have removed the old attr, clear ATTR_REPLACE so + * that the leaf format add routine won't trip over the attr + * not being around. + */ + args->flags &= ~ATTR_REPLACE; } if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX || diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 21be186067a2..83a6d3c7f872 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -620,7 +620,7 @@ xfs_attr_rmtval_remove( /* * If the "remote" value is in the cache, remove it. */ - bp = xfs_incore(mp->m_ddev_targp, dblkno, dblkcnt, XBF_TRYLOCK); + bp = xfs_buf_incore(mp->m_ddev_targp, dblkno, dblkcnt, XBF_TRYLOCK); if (bp) { xfs_buf_stale(bp); xfs_buf_relse(bp); diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 3b03d886df66..7b0e2b551e23 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -246,7 +246,7 @@ xfs_bmap_get_bp( struct xfs_btree_cur *cur, xfs_fsblock_t bno) { - struct xfs_log_item_desc *lidp; + struct xfs_log_item *lip; int i; if (!cur) @@ -260,9 +260,9 @@ xfs_bmap_get_bp( } /* Chase down all the log items to see if the bp is there */ - list_for_each_entry(lidp, &cur->bc_tp->t_items, lid_trans) { - struct xfs_buf_log_item *bip; - bip = (struct xfs_buf_log_item *)lidp->lid_item; + list_for_each_entry(lip, &cur->bc_tp->t_items, li_trans) { + struct xfs_buf_log_item *bip = (struct xfs_buf_log_item *)lip; + if (bip->bli_item.li_type == XFS_LI_BUF && XFS_BUF_ADDR(bip->bli_buf) == bno) return bip->bli_buf; @@ -312,8 +312,9 @@ xfs_check_block( xfs_warn(mp, "%s: thispa(%d) == pp(%d) %Ld", __func__, j, i, (unsigned long long)be64_to_cpu(*thispa)); - panic("%s: ptrs are equal in node\n", + xfs_err(mp, "%s: ptrs are equal in node\n", __func__); + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); } } } @@ -483,7 +484,8 @@ error0: error_norelse: xfs_warn(mp, "%s: BAD after btree leaves for %d extents", __func__, i); - panic("%s: CORRUPTED BTREE OR SOMETHING", __func__); + xfs_err(mp, "%s: CORRUPTED BTREE OR SOMETHING", __func__); + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); return; } @@ -542,12 +544,13 @@ xfs_bmap_validate_ret( * The list is maintained sorted (by block number). */ void -xfs_bmap_add_free( +__xfs_bmap_add_free( struct xfs_mount *mp, struct xfs_defer_ops *dfops, xfs_fsblock_t bno, xfs_filblks_t len, - struct xfs_owner_info *oinfo) + struct xfs_owner_info *oinfo, + bool skip_discard) { struct xfs_extent_free_item *new; /* new element */ #ifdef DEBUG @@ -574,6 +577,7 @@ xfs_bmap_add_free( new->xefi_oinfo = *oinfo; else xfs_rmap_skip_owner_update(&new->xefi_oinfo); + new->xefi_skip_discard = skip_discard; trace_xfs_bmap_free_defer(mp, XFS_FSB_TO_AGNO(mp, bno), 0, XFS_FSB_TO_AGBNO(mp, bno), len); xfs_defer_add(dfops, XFS_DEFER_OPS_TYPE_FREE, &new->xefi_list); @@ -725,12 +729,16 @@ xfs_bmap_extents_to_btree( *logflagsp = 0; if ((error = xfs_alloc_vextent(&args))) { xfs_iroot_realloc(ip, -1, whichfork); + ASSERT(ifp->if_broot == NULL); + XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); return error; } if (WARN_ON_ONCE(args.fsbno == NULLFSBLOCK)) { xfs_iroot_realloc(ip, -1, whichfork); + ASSERT(ifp->if_broot == NULL); + XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); return -ENOSPC; } @@ -1997,10 +2005,13 @@ xfs_bmap_add_extent_delay_real( ASSERT(0); } - /* add reverse mapping */ - error = xfs_rmap_map_extent(mp, bma->dfops, bma->ip, whichfork, new); - if (error) - goto done; + /* add reverse mapping unless caller opted out */ + if (!(bma->flags & XFS_BMAPI_NORMAP)) { + error = xfs_rmap_map_extent(mp, bma->dfops, bma->ip, + whichfork, new); + if (error) + goto done; + } /* convert to a btree if necessary */ if (xfs_bmap_needs_btree(bma->ip, whichfork)) { @@ -2664,7 +2675,8 @@ xfs_bmap_add_extent_hole_real( struct xfs_bmbt_irec *new, xfs_fsblock_t *first, struct xfs_defer_ops *dfops, - int *logflagsp) + int *logflagsp, + int flags) { struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); struct xfs_mount *mp = ip->i_mount; @@ -2841,10 +2853,12 @@ xfs_bmap_add_extent_hole_real( break; } - /* add reverse mapping */ - error = xfs_rmap_map_extent(mp, dfops, ip, whichfork, new); - if (error) - goto done; + /* add reverse mapping unless caller opted out */ + if (!(flags & XFS_BMAPI_NORMAP)) { + error = xfs_rmap_map_extent(mp, dfops, ip, whichfork, new); + if (error) + goto done; + } /* convert to a btree if necessary */ if (xfs_bmap_needs_btree(ip, whichfork)) { @@ -3225,7 +3239,7 @@ xfs_bmap_longest_free_extent( } } - longest = xfs_alloc_longest_free_extent(mp, pag, + longest = xfs_alloc_longest_free_extent(pag, xfs_alloc_min_freelist(mp, pag), xfs_ag_resv_needed(pag, XFS_AG_RESV_NONE)); if (*blen < longest) @@ -4119,7 +4133,8 @@ xfs_bmapi_allocate( else error = xfs_bmap_add_extent_hole_real(bma->tp, bma->ip, whichfork, &bma->icur, &bma->cur, &bma->got, - bma->firstblock, bma->dfops, &bma->logflags); + bma->firstblock, bma->dfops, &bma->logflags, + bma->flags); bma->logflags |= tmp_logflags; if (error) @@ -4505,30 +4520,37 @@ error0: return error; } -static int +int xfs_bmapi_remap( struct xfs_trans *tp, struct xfs_inode *ip, xfs_fileoff_t bno, xfs_filblks_t len, xfs_fsblock_t startblock, - struct xfs_defer_ops *dfops) + struct xfs_defer_ops *dfops, + int flags) { struct xfs_mount *mp = ip->i_mount; - struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); + struct xfs_ifork *ifp; struct xfs_btree_cur *cur = NULL; xfs_fsblock_t firstblock = NULLFSBLOCK; struct xfs_bmbt_irec got; struct xfs_iext_cursor icur; + int whichfork = xfs_bmapi_whichfork(flags); int logflags = 0, error; + ifp = XFS_IFORK_PTR(ip, whichfork); ASSERT(len > 0); ASSERT(len <= (xfs_filblks_t)MAXEXTLEN); ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT(!(flags & ~(XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC | + XFS_BMAPI_NORMAP))); + ASSERT((flags & (XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC)) != + (XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC)); if (unlikely(XFS_TEST_ERROR( - (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS && - XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE), + (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS && + XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE), mp, XFS_ERRTAG_BMAPIFORMAT))) { XFS_ERROR_REPORT("xfs_bmapi_remap", XFS_ERRLEVEL_LOW, mp); return -EFSCORRUPTED; @@ -4538,7 +4560,7 @@ xfs_bmapi_remap( return -EIO; if (!(ifp->if_flags & XFS_IFEXTENTS)) { - error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK); + error = xfs_iread_extents(tp, ip, whichfork); if (error) return error; } @@ -4553,7 +4575,7 @@ xfs_bmapi_remap( xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); if (ifp->if_flags & XFS_IFBROOT) { - cur = xfs_bmbt_init_cursor(mp, tp, ip, XFS_DATA_FORK); + cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork); cur->bc_private.b.firstblock = firstblock; cur->bc_private.b.dfops = dfops; cur->bc_private.b.flags = 0; @@ -4562,18 +4584,21 @@ xfs_bmapi_remap( got.br_startoff = bno; got.br_startblock = startblock; got.br_blockcount = len; - got.br_state = XFS_EXT_NORM; + if (flags & XFS_BMAPI_PREALLOC) + got.br_state = XFS_EXT_UNWRITTEN; + else + got.br_state = XFS_EXT_NORM; - error = xfs_bmap_add_extent_hole_real(tp, ip, XFS_DATA_FORK, &icur, - &cur, &got, &firstblock, dfops, &logflags); + error = xfs_bmap_add_extent_hole_real(tp, ip, whichfork, &icur, + &cur, &got, &firstblock, dfops, &logflags, flags); if (error) goto error0; - if (xfs_bmap_wants_extents(ip, XFS_DATA_FORK)) { + if (xfs_bmap_wants_extents(ip, whichfork)) { int tmp_logflags = 0; error = xfs_bmap_btree_to_extents(tp, ip, cur, - &tmp_logflags, XFS_DATA_FORK); + &tmp_logflags, whichfork); logflags |= tmp_logflags; } @@ -5100,9 +5125,12 @@ xfs_bmap_del_extent_real( error = xfs_refcount_decrease_extent(mp, dfops, del); if (error) goto done; - } else - xfs_bmap_add_free(mp, dfops, del->br_startblock, - del->br_blockcount, NULL); + } else { + __xfs_bmap_add_free(mp, dfops, del->br_startblock, + del->br_blockcount, NULL, + (bflags & XFS_BMAPI_NODISCARD) || + del->br_state == XFS_EXT_UNWRITTEN); + } } /* @@ -5667,7 +5695,6 @@ xfs_bmap_collapse_extents( xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb, bool *done, - xfs_fileoff_t stop_fsb, xfs_fsblock_t *firstblock, struct xfs_defer_ops *dfops) { @@ -6145,7 +6172,7 @@ xfs_bmap_finish_one( switch (type) { case XFS_BMAP_MAP: error = xfs_bmapi_remap(tp, ip, startoff, *blockcount, - startblock, dfops); + startblock, dfops, 0); *blockcount = 0; break; case XFS_BMAP_UNMAP: diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index f3be6416260b..2c233f9f1a26 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -68,6 +68,7 @@ struct xfs_extent_free_item xfs_extlen_t xefi_blockcount;/* number of blocks in extent */ struct list_head xefi_list; struct xfs_owner_info xefi_oinfo; /* extent owner */ + bool xefi_skip_discard; }; #define XFS_BMAP_MAX_NMAP 4 @@ -116,6 +117,12 @@ struct xfs_extent_free_item /* Only convert unwritten extents, don't allocate new blocks */ #define XFS_BMAPI_CONVERT_ONLY 0x800 +/* Skip online discard of freed extents */ +#define XFS_BMAPI_NODISCARD 0x1000 + +/* Do not update the rmap btree. Used for reconstructing bmbt from rmapbt. */ +#define XFS_BMAPI_NORMAP 0x2000 + #define XFS_BMAPI_FLAGS \ { XFS_BMAPI_ENTIRE, "ENTIRE" }, \ { XFS_BMAPI_METADATA, "METADATA" }, \ @@ -128,7 +135,9 @@ struct xfs_extent_free_item { XFS_BMAPI_REMAP, "REMAP" }, \ { XFS_BMAPI_COWFORK, "COWFORK" }, \ { XFS_BMAPI_DELALLOC, "DELALLOC" }, \ - { XFS_BMAPI_CONVERT_ONLY, "CONVERT_ONLY" } + { XFS_BMAPI_CONVERT_ONLY, "CONVERT_ONLY" }, \ + { XFS_BMAPI_NODISCARD, "NODISCARD" }, \ + { XFS_BMAPI_NORMAP, "NORMAP" } static inline int xfs_bmapi_aflag(int w) @@ -192,9 +201,9 @@ void xfs_trim_extent(struct xfs_bmbt_irec *irec, xfs_fileoff_t bno, void xfs_trim_extent_eof(struct xfs_bmbt_irec *, struct xfs_inode *); int xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd); void xfs_bmap_local_to_extents_empty(struct xfs_inode *ip, int whichfork); -void xfs_bmap_add_free(struct xfs_mount *mp, struct xfs_defer_ops *dfops, +void __xfs_bmap_add_free(struct xfs_mount *mp, struct xfs_defer_ops *dfops, xfs_fsblock_t bno, xfs_filblks_t len, - struct xfs_owner_info *oinfo); + struct xfs_owner_info *oinfo, bool skip_discard); void xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork); int xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip, xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork); @@ -228,7 +237,7 @@ void xfs_bmap_del_extent_cow(struct xfs_inode *ip, uint xfs_default_attroffset(struct xfs_inode *ip); int xfs_bmap_collapse_extents(struct xfs_trans *tp, struct xfs_inode *ip, xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb, - bool *done, xfs_fileoff_t stop_fsb, xfs_fsblock_t *firstblock, + bool *done, xfs_fsblock_t *firstblock, struct xfs_defer_ops *dfops); int xfs_bmap_insert_extents(struct xfs_trans *tp, struct xfs_inode *ip, xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb, @@ -240,6 +249,17 @@ int xfs_bmapi_reserve_delalloc(struct xfs_inode *ip, int whichfork, struct xfs_bmbt_irec *got, struct xfs_iext_cursor *cur, int eof); +static inline void +xfs_bmap_add_free( + struct xfs_mount *mp, + struct xfs_defer_ops *dfops, + xfs_fsblock_t bno, + xfs_filblks_t len, + struct xfs_owner_info *oinfo) +{ + __xfs_bmap_add_free(mp, dfops, bno, len, oinfo, false); +} + enum xfs_bmap_intent_type { XFS_BMAP_MAP = 1, XFS_BMAP_UNMAP, @@ -277,4 +297,8 @@ static inline int xfs_bmap_fork_to_state(int whichfork) xfs_failaddr_t xfs_bmap_validate_extent(struct xfs_inode *ip, int whichfork, struct xfs_bmbt_irec *irec); +int xfs_bmapi_remap(struct xfs_trans *tp, struct xfs_inode *ip, + xfs_fileoff_t bno, xfs_filblks_t len, xfs_fsblock_t startblock, + struct xfs_defer_ops *dfops, int flags); + #endif /* __XFS_BMAP_H__ */ diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index d89d06bea6e3..ac9d4aeedb09 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -660,3 +660,12 @@ xfs_bmbt_change_owner( xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); return error; } + +/* Calculate the bmap btree size for some records. */ +unsigned long long +xfs_bmbt_calc_size( + struct xfs_mount *mp, + unsigned long long len) +{ + return xfs_btree_calc_size(mp->m_bmap_dmnr, len); +} diff --git a/fs/xfs/libxfs/xfs_bmap_btree.h b/fs/xfs/libxfs/xfs_bmap_btree.h index e4505746ccaa..fb3cd2d9e0f8 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.h +++ b/fs/xfs/libxfs/xfs_bmap_btree.h @@ -118,4 +118,7 @@ extern int xfs_bmbt_change_owner(struct xfs_trans *tp, struct xfs_inode *ip, extern struct xfs_btree_cur *xfs_bmbt_init_cursor(struct xfs_mount *, struct xfs_trans *, struct xfs_inode *, int); +extern unsigned long long xfs_bmbt_calc_size(struct xfs_mount *mp, + unsigned long long len); + #endif /* __XFS_BMAP_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c index edc0193358a5..c825c8182b30 100644 --- a/fs/xfs/libxfs/xfs_btree.c +++ b/fs/xfs/libxfs/xfs_btree.c @@ -4531,7 +4531,6 @@ xfs_btree_sblock_verify( */ uint xfs_btree_compute_maxlevels( - struct xfs_mount *mp, uint *limits, unsigned long len) { @@ -4837,15 +4836,14 @@ xfs_btree_query_all( * Calculate the number of blocks needed to store a given number of records * in a short-format (per-AG metadata) btree. */ -xfs_extlen_t +unsigned long long xfs_btree_calc_size( - struct xfs_mount *mp, uint *limits, unsigned long long len) { int level; int maxrecs; - xfs_extlen_t rval; + unsigned long long rval; maxrecs = limits[0]; for (level = 0, rval = 0; len > 1; level++) { @@ -4921,3 +4919,24 @@ xfs_btree_has_record( *exists = false; return error; } + +/* Are there more records in this btree? */ +bool +xfs_btree_has_more_records( + struct xfs_btree_cur *cur) +{ + struct xfs_btree_block *block; + struct xfs_buf *bp; + + block = xfs_btree_get_block(cur, 0, &bp); + + /* There are still records in this block. */ + if (cur->bc_ptrs[0] < xfs_btree_get_numrecs(block)) + return true; + + /* There are more record blocks. */ + if (cur->bc_flags & XFS_BTREE_LONG_PTRS) + return block->bb_u.l.bb_rightsib != cpu_to_be64(NULLFSBLOCK); + else + return block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK); +} diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h index 58e30c0975c3..d7911efee6dc 100644 --- a/fs/xfs/libxfs/xfs_btree.h +++ b/fs/xfs/libxfs/xfs_btree.h @@ -481,10 +481,8 @@ xfs_failaddr_t xfs_btree_lblock_v5hdr_verify(struct xfs_buf *bp, xfs_failaddr_t xfs_btree_lblock_verify(struct xfs_buf *bp, unsigned int max_recs); -uint xfs_btree_compute_maxlevels(struct xfs_mount *mp, uint *limits, - unsigned long len); -xfs_extlen_t xfs_btree_calc_size(struct xfs_mount *mp, uint *limits, - unsigned long long len); +uint xfs_btree_compute_maxlevels(uint *limits, unsigned long len); +unsigned long long xfs_btree_calc_size(uint *limits, unsigned long long len); /* return codes */ #define XFS_BTREE_QUERY_RANGE_CONTINUE 0 /* keep iterating */ @@ -530,5 +528,6 @@ union xfs_btree_key *xfs_btree_high_key_from_key(struct xfs_btree_cur *cur, union xfs_btree_key *key); int xfs_btree_has_record(struct xfs_btree_cur *cur, union xfs_btree_irec *low, union xfs_btree_irec *high, bool *exists); +bool xfs_btree_has_more_records(struct xfs_btree_cur *cur); #endif /* __XFS_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c index 087fea02c389..3daf175e2535 100644 --- a/fs/xfs/libxfs/xfs_defer.c +++ b/fs/xfs/libxfs/xfs_defer.c @@ -220,7 +220,7 @@ xfs_defer_trans_abort( { struct xfs_defer_pending *dfp; - trace_xfs_defer_trans_abort(tp->t_mountp, dop); + trace_xfs_defer_trans_abort(tp->t_mountp, dop, _RET_IP_); /* Abort intent items that don't have a done item. */ list_for_each_entry(dfp, &dop->dop_pending, dfp_list) { @@ -253,7 +253,7 @@ xfs_defer_trans_roll( for (i = 0; i < XFS_DEFER_OPS_NR_BUFS && dop->dop_bufs[i]; i++) xfs_trans_dirty_buf(*tp, dop->dop_bufs[i]); - trace_xfs_defer_trans_roll((*tp)->t_mountp, dop); + trace_xfs_defer_trans_roll((*tp)->t_mountp, dop, _RET_IP_); /* Roll the transaction. */ error = xfs_trans_roll(tp); @@ -352,10 +352,21 @@ xfs_defer_finish( void *state; int error = 0; void (*cleanup_fn)(struct xfs_trans *, void *, int); + struct xfs_defer_ops *orig_dop; ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES); - trace_xfs_defer_finish((*tp)->t_mountp, dop); + trace_xfs_defer_finish((*tp)->t_mountp, dop, _RET_IP_); + + /* + * Attach dfops to the transaction during deferred ops processing. This + * explicitly causes calls into the allocator to defer AGFL block frees. + * Note that this code can go away once all dfops users attach to the + * associated tp. + */ + ASSERT(!(*tp)->t_agfl_dfops || ((*tp)->t_agfl_dfops == dop)); + orig_dop = (*tp)->t_agfl_dfops; + (*tp)->t_agfl_dfops = dop; /* Until we run out of pending work to finish... */ while (xfs_defer_has_unfinished_work(dop)) { @@ -428,10 +439,11 @@ xfs_defer_finish( } out: + (*tp)->t_agfl_dfops = orig_dop; if (error) trace_xfs_defer_finish_error((*tp)->t_mountp, dop, error); else - trace_xfs_defer_finish_done((*tp)->t_mountp, dop); + trace_xfs_defer_finish_done((*tp)->t_mountp, dop, _RET_IP_); return error; } @@ -447,7 +459,7 @@ xfs_defer_cancel( struct list_head *pwi; struct list_head *n; - trace_xfs_defer_cancel(NULL, dop); + trace_xfs_defer_cancel(NULL, dop, _RET_IP_); /* * Free the pending items. Caller should already have arranged @@ -532,5 +544,5 @@ xfs_defer_init( *fbp = NULLFSBLOCK; INIT_LIST_HEAD(&dop->dop_intake); INIT_LIST_HEAD(&dop->dop_pending); - trace_xfs_defer_init(NULL, dop); + trace_xfs_defer_init(NULL, dop, _RET_IP_); } diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h index 045beacdd37d..e70725ba1f5f 100644 --- a/fs/xfs/libxfs/xfs_defer.h +++ b/fs/xfs/libxfs/xfs_defer.h @@ -55,6 +55,7 @@ enum xfs_defer_ops_type { XFS_DEFER_OPS_TYPE_REFCOUNT, XFS_DEFER_OPS_TYPE_RMAP, XFS_DEFER_OPS_TYPE_FREE, + XFS_DEFER_OPS_TYPE_AGFL_FREE, XFS_DEFER_OPS_TYPE_MAX, }; diff --git a/fs/xfs/libxfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c index 8b7a6c3cb599..cce520becee4 100644 --- a/fs/xfs/libxfs/xfs_dquot_buf.c +++ b/fs/xfs/libxfs/xfs_dquot_buf.c @@ -41,14 +41,18 @@ xfs_calc_dquots_per_chunk( /* * Do some primitive error checking on ondisk dquot data structures. + * + * The xfs_dqblk structure /contains/ the xfs_disk_dquot structure; + * we verify them separately because at some points we have only the + * smaller xfs_disk_dquot structure available. */ + xfs_failaddr_t xfs_dquot_verify( struct xfs_mount *mp, xfs_disk_dquot_t *ddq, xfs_dqid_t id, - uint type, /* used only when IO_dorepair is true */ - uint flags) + uint type) /* used only during quotacheck */ { /* * We can encounter an uninitialized dquot buffer for 2 reasons: @@ -70,6 +74,8 @@ xfs_dquot_verify( if (ddq->d_version != XFS_DQUOT_VERSION) return __this_address; + if (type && ddq->d_flags != type) + return __this_address; if (ddq->d_flags != XFS_DQ_USER && ddq->d_flags != XFS_DQ_PROJ && ddq->d_flags != XFS_DQ_GROUP) @@ -99,33 +105,44 @@ xfs_dquot_verify( return NULL; } +xfs_failaddr_t +xfs_dqblk_verify( + struct xfs_mount *mp, + struct xfs_dqblk *dqb, + xfs_dqid_t id, + uint type) /* used only during quotacheck */ +{ + if (xfs_sb_version_hascrc(&mp->m_sb) && + !uuid_equal(&dqb->dd_uuid, &mp->m_sb.sb_meta_uuid)) + return __this_address; + + return xfs_dquot_verify(mp, &dqb->dd_diskdq, id, type); +} + /* * Do some primitive error checking on ondisk dquot data structures. */ int -xfs_dquot_repair( +xfs_dqblk_repair( struct xfs_mount *mp, - struct xfs_disk_dquot *ddq, + struct xfs_dqblk *dqb, xfs_dqid_t id, uint type) { - struct xfs_dqblk *d = (struct xfs_dqblk *)ddq; - - /* * Typically, a repair is only requested by quotacheck. */ ASSERT(id != -1); - memset(d, 0, sizeof(xfs_dqblk_t)); + memset(dqb, 0, sizeof(xfs_dqblk_t)); - d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); - d->dd_diskdq.d_version = XFS_DQUOT_VERSION; - d->dd_diskdq.d_flags = type; - d->dd_diskdq.d_id = cpu_to_be32(id); + dqb->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); + dqb->dd_diskdq.d_version = XFS_DQUOT_VERSION; + dqb->dd_diskdq.d_flags = type; + dqb->dd_diskdq.d_id = cpu_to_be32(id); if (xfs_sb_version_hascrc(&mp->m_sb)) { - uuid_copy(&d->dd_uuid, &mp->m_sb.sb_meta_uuid); - xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk), + uuid_copy(&dqb->dd_uuid, &mp->m_sb.sb_meta_uuid); + xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk), XFS_DQUOT_CRC_OFF); } @@ -135,7 +152,8 @@ xfs_dquot_repair( STATIC bool xfs_dquot_buf_verify_crc( struct xfs_mount *mp, - struct xfs_buf *bp) + struct xfs_buf *bp, + bool readahead) { struct xfs_dqblk *d = (struct xfs_dqblk *)bp->b_addr; int ndquots; @@ -156,10 +174,12 @@ xfs_dquot_buf_verify_crc( for (i = 0; i < ndquots; i++, d++) { if (!xfs_verify_cksum((char *)d, sizeof(struct xfs_dqblk), - XFS_DQUOT_CRC_OFF)) - return false; - if (!uuid_equal(&d->dd_uuid, &mp->m_sb.sb_meta_uuid)) + XFS_DQUOT_CRC_OFF)) { + if (!readahead) + xfs_buf_verifier_error(bp, -EFSBADCRC, __func__, + d, sizeof(*d), __this_address); return false; + } } return true; } @@ -167,9 +187,10 @@ xfs_dquot_buf_verify_crc( STATIC xfs_failaddr_t xfs_dquot_buf_verify( struct xfs_mount *mp, - struct xfs_buf *bp) + struct xfs_buf *bp, + bool readahead) { - struct xfs_dqblk *d = (struct xfs_dqblk *)bp->b_addr; + struct xfs_dqblk *dqb = bp->b_addr; xfs_failaddr_t fa; xfs_dqid_t id = 0; int ndquots; @@ -195,14 +216,19 @@ xfs_dquot_buf_verify( for (i = 0; i < ndquots; i++) { struct xfs_disk_dquot *ddq; - ddq = &d[i].dd_diskdq; + ddq = &dqb[i].dd_diskdq; if (i == 0) id = be32_to_cpu(ddq->d_id); - fa = xfs_dquot_verify(mp, ddq, id + i, 0, 0); - if (fa) + fa = xfs_dqblk_verify(mp, &dqb[i], id + i, 0); + if (fa) { + if (!readahead) + xfs_buf_verifier_error(bp, -EFSCORRUPTED, + __func__, &dqb[i], + sizeof(struct xfs_dqblk), fa); return fa; + } } return NULL; @@ -214,7 +240,7 @@ xfs_dquot_buf_verify_struct( { struct xfs_mount *mp = bp->b_target->bt_mount; - return xfs_dquot_buf_verify(mp, bp); + return xfs_dquot_buf_verify(mp, bp, false); } static void @@ -222,15 +248,10 @@ xfs_dquot_buf_read_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_target->bt_mount; - xfs_failaddr_t fa; - if (!xfs_dquot_buf_verify_crc(mp, bp)) - xfs_verifier_error(bp, -EFSBADCRC, __this_address); - else { - fa = xfs_dquot_buf_verify(mp, bp); - if (fa) - xfs_verifier_error(bp, -EFSCORRUPTED, __this_address); - } + if (!xfs_dquot_buf_verify_crc(mp, bp, false)) + return; + xfs_dquot_buf_verify(mp, bp, false); } /* @@ -245,8 +266,8 @@ xfs_dquot_buf_readahead_verify( { struct xfs_mount *mp = bp->b_target->bt_mount; - if (!xfs_dquot_buf_verify_crc(mp, bp) || - xfs_dquot_buf_verify(mp, bp) != NULL) { + if (!xfs_dquot_buf_verify_crc(mp, bp, true) || + xfs_dquot_buf_verify(mp, bp, true) != NULL) { xfs_buf_ioerror(bp, -EIO); bp->b_flags &= ~XBF_DONE; } @@ -262,11 +283,8 @@ xfs_dquot_buf_write_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_target->bt_mount; - xfs_failaddr_t fa; - fa = xfs_dquot_buf_verify(mp, bp); - if (fa) - xfs_verifier_error(bp, -EFSCORRUPTED, __this_address); + xfs_dquot_buf_verify(mp, bp, false); } const struct xfs_buf_ops xfs_dquot_buf_ops = { diff --git a/fs/xfs/libxfs/xfs_errortag.h b/fs/xfs/libxfs/xfs_errortag.h index bc1789d95152..d47b91625945 100644 --- a/fs/xfs/libxfs/xfs_errortag.h +++ b/fs/xfs/libxfs/xfs_errortag.h @@ -65,7 +65,8 @@ #define XFS_ERRTAG_LOG_BAD_CRC 29 #define XFS_ERRTAG_LOG_ITEM_PIN 30 #define XFS_ERRTAG_BUF_LRU_REF 31 -#define XFS_ERRTAG_MAX 32 +#define XFS_ERRTAG_FORCE_SCRUB_REPAIR 32 +#define XFS_ERRTAG_MAX 33 /* * Random factors for above tags, 1 means always, 2 means 1/2 time, etc. @@ -102,5 +103,6 @@ #define XFS_RANDOM_LOG_BAD_CRC 1 #define XFS_RANDOM_LOG_ITEM_PIN 1 #define XFS_RANDOM_BUF_LRU_REF 2 +#define XFS_RANDOM_FORCE_SCRUB_REPAIR 1 #endif /* __XFS_ERRORTAG_H_ */ diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index 42956d8d95ed..c1cb29a5f4f6 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -98,6 +98,9 @@ struct xfs_ifork; XFS_SB_VERSION2_PROJID32BIT | \ XFS_SB_VERSION2_FTYPE) +/* Maximum size of the xfs filesystem label, no terminating NULL */ +#define XFSLABEL_MAX 12 + /* * Superblock - in core version. Must match the ondisk version below. * Must be padded to 64 bit alignment. @@ -122,7 +125,7 @@ typedef struct xfs_sb { uint16_t sb_sectsize; /* volume sector size, bytes */ uint16_t sb_inodesize; /* inode size, bytes */ uint16_t sb_inopblock; /* inodes per block */ - char sb_fname[12]; /* file system name */ + char sb_fname[XFSLABEL_MAX]; /* file system name */ uint8_t sb_blocklog; /* log2 of sb_blocksize */ uint8_t sb_sectlog; /* log2 of sb_sectsize */ uint8_t sb_inodelog; /* log2 of sb_inodesize */ @@ -213,7 +216,7 @@ typedef struct xfs_dsb { __be16 sb_sectsize; /* volume sector size, bytes */ __be16 sb_inodesize; /* inode size, bytes */ __be16 sb_inopblock; /* inodes per block */ - char sb_fname[12]; /* file system name */ + char sb_fname[XFSLABEL_MAX]; /* file system name */ __u8 sb_blocklog; /* log2 of sb_blocksize */ __u8 sb_sectlog; /* log2 of sb_sectsize */ __u8 sb_inodelog; /* log2 of sb_inodesize */ diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index faf1a4edd618..dddc75e4f1f6 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -542,13 +542,20 @@ struct xfs_scrub_metadata { /* o: Metadata object looked funny but isn't corrupt. */ #define XFS_SCRUB_OFLAG_WARNING (1 << 6) +/* + * o: IFLAG_REPAIR was set but metadata object did not need fixing or + * optimization and has therefore not been altered. + */ +#define XFS_SCRUB_OFLAG_NO_REPAIR_NEEDED (1 << 7) + #define XFS_SCRUB_FLAGS_IN (XFS_SCRUB_IFLAG_REPAIR) #define XFS_SCRUB_FLAGS_OUT (XFS_SCRUB_OFLAG_CORRUPT | \ XFS_SCRUB_OFLAG_PREEN | \ XFS_SCRUB_OFLAG_XFAIL | \ XFS_SCRUB_OFLAG_XCORRUPT | \ XFS_SCRUB_OFLAG_INCOMPLETE | \ - XFS_SCRUB_OFLAG_WARNING) + XFS_SCRUB_OFLAG_WARNING | \ + XFS_SCRUB_OFLAG_NO_REPAIR_NEEDED) #define XFS_SCRUB_FLAGS_ALL (XFS_SCRUB_FLAGS_IN | XFS_SCRUB_FLAGS_OUT) /* diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 0e2cf5f0be1f..4ca4ff7a757d 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -148,7 +148,7 @@ xfs_inobt_get_rec( /* * Insert a single inobt record. Cursor must already point to desired location. */ -STATIC int +int xfs_inobt_insert_rec( struct xfs_btree_cur *cur, uint16_t holemask, @@ -2406,7 +2406,7 @@ xfs_ialloc_compute_maxlevels( uint inodes; inodes = (1LL << XFS_INO_AGINO_BITS(mp)) >> XFS_INODES_PER_CHUNK_LOG; - mp->m_in_maxlevels = xfs_btree_compute_maxlevels(mp, mp->m_inobt_mnr, + mp->m_in_maxlevels = xfs_btree_compute_maxlevels(mp->m_inobt_mnr, inodes); } diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h index c5402bb4ce0c..77fffced8bac 100644 --- a/fs/xfs/libxfs/xfs_ialloc.h +++ b/fs/xfs/libxfs/xfs_ialloc.h @@ -176,6 +176,9 @@ int xfs_ialloc_has_inode_record(struct xfs_btree_cur *cur, xfs_agino_t low, xfs_agino_t high, bool *exists); int xfs_ialloc_count_inodes(struct xfs_btree_cur *cur, xfs_agino_t *count, xfs_agino_t *freecount); +int xfs_inobt_insert_rec(struct xfs_btree_cur *cur, uint16_t holemask, + uint8_t count, int32_t freecount, xfs_inofree_t free, + int *stat); int xfs_ialloc_cluster_alignment(struct xfs_mount *mp); void xfs_ialloc_agino_range(struct xfs_mount *mp, xfs_agnumber_t agno, diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index a2dd7f4a2719..b04c55512159 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -296,7 +296,7 @@ xfs_inobt_verify( case cpu_to_be32(XFS_FIBT_MAGIC): break; default: - return NULL; + return __this_address; } /* level verification */ @@ -556,7 +556,7 @@ xfs_inobt_max_size( if (mp->m_inobt_mxr[0] == 0) return 0; - return xfs_btree_calc_size(mp, mp->m_inobt_mnr, + return xfs_btree_calc_size(mp->m_inobt_mnr, (uint64_t)mp->m_sb.sb_agblocks * mp->m_sb.sb_inopblock / XFS_INODES_PER_CHUNK); } @@ -608,3 +608,12 @@ xfs_finobt_calc_reserves( *used += tree_len; return 0; } + +/* Calculate the inobt btree size for some records. */ +xfs_extlen_t +xfs_iallocbt_calc_size( + struct xfs_mount *mp, + unsigned long long len) +{ + return xfs_btree_calc_size(mp->m_inobt_mnr, len); +} diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h index aa81e2e63f3f..4acdd5458d59 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.h +++ b/fs/xfs/libxfs/xfs_ialloc_btree.h @@ -74,5 +74,7 @@ int xfs_inobt_rec_check_count(struct xfs_mount *, int xfs_finobt_calc_reserves(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used); +extern xfs_extlen_t xfs_iallocbt_calc_size(struct xfs_mount *mp, + unsigned long long len); #endif /* __XFS_IALLOC_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c index ef68b1de006a..1201107eabc6 100644 --- a/fs/xfs/libxfs/xfs_inode_buf.c +++ b/fs/xfs/libxfs/xfs_inode_buf.c @@ -466,6 +466,8 @@ xfs_dinode_verify( return __this_address; if (di_size > XFS_DFORK_DSIZE(dip, mp)) return __this_address; + if (dip->di_nextents) + return __this_address; /* fall through */ case XFS_DINODE_FMT_EXTENTS: case XFS_DINODE_FMT_BTREE: @@ -484,12 +486,31 @@ xfs_dinode_verify( if (XFS_DFORK_Q(dip)) { switch (dip->di_aformat) { case XFS_DINODE_FMT_LOCAL: + if (dip->di_anextents) + return __this_address; + /* fall through */ case XFS_DINODE_FMT_EXTENTS: case XFS_DINODE_FMT_BTREE: break; default: return __this_address; } + } else { + /* + * If there is no fork offset, this may be a freshly-made inode + * in a new disk cluster, in which case di_aformat is zeroed. + * Otherwise, such an inode must be in EXTENTS format; this goes + * for freed inodes as well. + */ + switch (dip->di_aformat) { + case 0: + case XFS_DINODE_FMT_EXTENTS: + break; + default: + return __this_address; + } + if (dip->di_anextents) + return __this_address; } /* only version 3 or greater inodes are extensively verified here */ diff --git a/fs/xfs/libxfs/xfs_quota_defs.h b/fs/xfs/libxfs/xfs_quota_defs.h index bb1b13a9b5f4..d4af2804b178 100644 --- a/fs/xfs/libxfs/xfs_quota_defs.h +++ b/fs/xfs/libxfs/xfs_quota_defs.h @@ -107,14 +107,12 @@ typedef uint16_t xfs_qwarncnt_t; * to a single function. None of these XFS_QMOPT_* flags are meant to have * persistent values (ie. their values can and will change between versions) */ -#define XFS_QMOPT_DQALLOC 0x0000002 /* alloc dquot ondisk if needed */ #define XFS_QMOPT_UQUOTA 0x0000004 /* user dquot requested */ #define XFS_QMOPT_PQUOTA 0x0000008 /* project dquot requested */ #define XFS_QMOPT_FORCE_RES 0x0000010 /* ignore quota limits */ #define XFS_QMOPT_SBVERSION 0x0000040 /* change superblock version num */ #define XFS_QMOPT_GQUOTA 0x0002000 /* group dquot requested */ #define XFS_QMOPT_ENOSPC 0x0004000 /* enospc instead of edquot (prj) */ -#define XFS_QMOPT_DQNEXT 0x0008000 /* return next dquot >= this ID */ /* * flags to xfs_trans_mod_dquot to indicate which field needs to be @@ -152,10 +150,11 @@ typedef uint16_t xfs_qwarncnt_t; #define XFS_QMOPT_RESBLK_MASK (XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_RES_RTBLKS) extern xfs_failaddr_t xfs_dquot_verify(struct xfs_mount *mp, - struct xfs_disk_dquot *ddq, xfs_dqid_t id, uint type, - uint flags); + struct xfs_disk_dquot *ddq, xfs_dqid_t id, uint type); +extern xfs_failaddr_t xfs_dqblk_verify(struct xfs_mount *mp, + struct xfs_dqblk *dqb, xfs_dqid_t id, uint type); extern int xfs_calc_dquots_per_chunk(unsigned int nbblks); -extern int xfs_dquot_repair(struct xfs_mount *mp, struct xfs_disk_dquot *ddq, +extern int xfs_dqblk_repair(struct xfs_mount *mp, struct xfs_dqblk *dqb, xfs_dqid_t id, uint type); #endif /* __XFS_QUOTA_H__ */ diff --git a/fs/xfs/libxfs/xfs_refcount.c b/fs/xfs/libxfs/xfs_refcount.c index bee68c23d612..418d53295893 100644 --- a/fs/xfs/libxfs/xfs_refcount.c +++ b/fs/xfs/libxfs/xfs_refcount.c @@ -88,8 +88,25 @@ xfs_refcount_lookup_ge( return xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat); } +/* + * Look up the first record equal to [bno, len] in the btree + * given by cur. + */ +int +xfs_refcount_lookup_eq( + struct xfs_btree_cur *cur, + xfs_agblock_t bno, + int *stat) +{ + trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_private.a.agno, bno, + XFS_LOOKUP_LE); + cur->bc_rec.rc.rc_startblock = bno; + cur->bc_rec.rc.rc_blockcount = 0; + return xfs_btree_lookup(cur, XFS_LOOKUP_EQ, stat); +} + /* Convert on-disk record to in-core format. */ -static inline void +void xfs_refcount_btrec_to_irec( union xfs_btree_rec *rec, struct xfs_refcount_irec *irec) @@ -149,7 +166,7 @@ xfs_refcount_update( * by [bno, len, refcount]. * This either works (return 0) or gets an EFSCORRUPTED error. */ -STATIC int +int xfs_refcount_insert( struct xfs_btree_cur *cur, struct xfs_refcount_irec *irec, @@ -162,7 +179,10 @@ xfs_refcount_insert( cur->bc_rec.rc.rc_blockcount = irec->rc_blockcount; cur->bc_rec.rc.rc_refcount = irec->rc_refcount; error = xfs_btree_insert(cur, i); + if (error) + goto out_error; XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error); + out_error: if (error) trace_xfs_refcount_insert_error(cur->bc_mp, @@ -351,7 +371,6 @@ xfs_refcount_merge_center_extents( struct xfs_refcount_irec *center, struct xfs_refcount_irec *right, unsigned long long extlen, - xfs_agblock_t *agbno, xfs_extlen_t *aglen) { int error; @@ -471,7 +490,6 @@ xfs_refcount_merge_right_extent( struct xfs_btree_cur *cur, struct xfs_refcount_irec *right, struct xfs_refcount_irec *cright, - xfs_agblock_t *agbno, xfs_extlen_t *aglen) { int error; @@ -749,7 +767,7 @@ xfs_refcount_merge_extents( ulen < MAXREFCEXTLEN) { *shape_changed = true; return xfs_refcount_merge_center_extents(cur, &left, &cleft, - &right, ulen, agbno, aglen); + &right, ulen, aglen); } /* Try to merge left and cleft. */ @@ -778,7 +796,7 @@ xfs_refcount_merge_extents( ulen < MAXREFCEXTLEN) { *shape_changed = true; return xfs_refcount_merge_right_extent(cur, &right, &cright, - agbno, aglen); + aglen); } return error; @@ -1356,9 +1374,7 @@ xfs_refcount_adjust_cow_extents( struct xfs_btree_cur *cur, xfs_agblock_t agbno, xfs_extlen_t aglen, - enum xfs_refc_adjust_op adj, - struct xfs_defer_ops *dfops, - struct xfs_owner_info *oinfo) + enum xfs_refc_adjust_op adj) { struct xfs_refcount_irec ext, tmp; int error; @@ -1437,8 +1453,7 @@ xfs_refcount_adjust_cow( struct xfs_btree_cur *cur, xfs_agblock_t agbno, xfs_extlen_t aglen, - enum xfs_refc_adjust_op adj, - struct xfs_defer_ops *dfops) + enum xfs_refc_adjust_op adj) { bool shape_changed; int error; @@ -1465,8 +1480,7 @@ xfs_refcount_adjust_cow( goto out_error; /* Now that we've taken care of the ends, adjust the middle extents */ - error = xfs_refcount_adjust_cow_extents(cur, agbno, aglen, adj, - dfops, NULL); + error = xfs_refcount_adjust_cow_extents(cur, agbno, aglen, adj); if (error) goto out_error; @@ -1493,7 +1507,7 @@ __xfs_refcount_cow_alloc( /* Add refcount btree reservation */ return xfs_refcount_adjust_cow(rcur, agbno, aglen, - XFS_REFCOUNT_ADJUST_COW_ALLOC, dfops); + XFS_REFCOUNT_ADJUST_COW_ALLOC); } /* @@ -1511,7 +1525,7 @@ __xfs_refcount_cow_free( /* Remove refcount btree reservation */ return xfs_refcount_adjust_cow(rcur, agbno, aglen, - XFS_REFCOUNT_ADJUST_COW_FREE, dfops); + XFS_REFCOUNT_ADJUST_COW_FREE); } /* Record a CoW staging extent in the refcount btree. */ @@ -1568,7 +1582,7 @@ struct xfs_refcount_recovery { /* Stuff an extent on the recovery list. */ STATIC int xfs_refcount_recover_extent( - struct xfs_btree_cur *cur, + struct xfs_btree_cur *cur, union xfs_btree_rec *rec, void *priv) { diff --git a/fs/xfs/libxfs/xfs_refcount.h b/fs/xfs/libxfs/xfs_refcount.h index 2a731ac68fe4..a92ad9078bc1 100644 --- a/fs/xfs/libxfs/xfs_refcount.h +++ b/fs/xfs/libxfs/xfs_refcount.h @@ -24,6 +24,8 @@ extern int xfs_refcount_lookup_le(struct xfs_btree_cur *cur, xfs_agblock_t bno, int *stat); extern int xfs_refcount_lookup_ge(struct xfs_btree_cur *cur, xfs_agblock_t bno, int *stat); +extern int xfs_refcount_lookup_eq(struct xfs_btree_cur *cur, + xfs_agblock_t bno, int *stat); extern int xfs_refcount_get_rec(struct xfs_btree_cur *cur, struct xfs_refcount_irec *irec, int *stat); @@ -85,5 +87,10 @@ static inline xfs_fileoff_t xfs_refcount_max_unmap(int log_res) extern int xfs_refcount_has_record(struct xfs_btree_cur *cur, xfs_agblock_t bno, xfs_extlen_t len, bool *exists); +union xfs_btree_rec; +extern void xfs_refcount_btrec_to_irec(union xfs_btree_rec *rec, + struct xfs_refcount_irec *irec); +extern int xfs_refcount_insert(struct xfs_btree_cur *cur, + struct xfs_refcount_irec *irec, int *stat); #endif /* __XFS_REFCOUNT_H__ */ diff --git a/fs/xfs/libxfs/xfs_refcount_btree.c b/fs/xfs/libxfs/xfs_refcount_btree.c index 265fdcefcbae..375abfeb6267 100644 --- a/fs/xfs/libxfs/xfs_refcount_btree.c +++ b/fs/xfs/libxfs/xfs_refcount_btree.c @@ -373,7 +373,6 @@ xfs_refcountbt_init_cursor( */ int xfs_refcountbt_maxrecs( - struct xfs_mount *mp, int blocklen, bool leaf) { @@ -390,7 +389,7 @@ void xfs_refcountbt_compute_maxlevels( struct xfs_mount *mp) { - mp->m_refc_maxlevels = xfs_btree_compute_maxlevels(mp, + mp->m_refc_maxlevels = xfs_btree_compute_maxlevels( mp->m_refc_mnr, mp->m_sb.sb_agblocks); } @@ -400,7 +399,7 @@ xfs_refcountbt_calc_size( struct xfs_mount *mp, unsigned long long len) { - return xfs_btree_calc_size(mp, mp->m_refc_mnr, len); + return xfs_btree_calc_size(mp->m_refc_mnr, len); } /* diff --git a/fs/xfs/libxfs/xfs_refcount_btree.h b/fs/xfs/libxfs/xfs_refcount_btree.h index 9db008b955b7..2bc4694ef146 100644 --- a/fs/xfs/libxfs/xfs_refcount_btree.h +++ b/fs/xfs/libxfs/xfs_refcount_btree.h @@ -60,8 +60,7 @@ struct xfs_mount; extern struct xfs_btree_cur *xfs_refcountbt_init_cursor(struct xfs_mount *mp, struct xfs_trans *tp, struct xfs_buf *agbp, xfs_agnumber_t agno, struct xfs_defer_ops *dfops); -extern int xfs_refcountbt_maxrecs(struct xfs_mount *mp, int blocklen, - bool leaf); +extern int xfs_refcountbt_maxrecs(int blocklen, bool leaf); extern void xfs_refcountbt_compute_maxlevels(struct xfs_mount *mp); extern xfs_extlen_t xfs_refcountbt_calc_size(struct xfs_mount *mp, diff --git a/fs/xfs/libxfs/xfs_rmap.c b/fs/xfs/libxfs/xfs_rmap.c index 79822cf6ebe3..c0644f1be8a8 100644 --- a/fs/xfs/libxfs/xfs_rmap.c +++ b/fs/xfs/libxfs/xfs_rmap.c @@ -376,7 +376,6 @@ xfs_rmap_free_check_owner( struct xfs_mount *mp, uint64_t ltoff, struct xfs_rmap_irec *rec, - xfs_fsblock_t bno, xfs_filblks_t len, uint64_t owner, uint64_t offset, @@ -519,7 +518,7 @@ xfs_rmap_unmap( bno + len, out_error); /* Check owner information. */ - error = xfs_rmap_free_check_owner(mp, ltoff, <rec, bno, len, owner, + error = xfs_rmap_free_check_owner(mp, ltoff, <rec, len, owner, offset, flags); if (error) goto out_error; @@ -1375,6 +1374,8 @@ xfs_rmap_convert_shared( */ error = xfs_rmap_lookup_le_range(cur, bno, owner, offset, flags, &PREV, &i); + if (error) + goto done; XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done); ASSERT(PREV.rm_offset <= offset); @@ -2031,6 +2032,34 @@ out_error: return error; } +/* Insert a raw rmap into the rmapbt. */ +int +xfs_rmap_map_raw( + struct xfs_btree_cur *cur, + struct xfs_rmap_irec *rmap) +{ + struct xfs_owner_info oinfo; + + oinfo.oi_owner = rmap->rm_owner; + oinfo.oi_offset = rmap->rm_offset; + oinfo.oi_flags = 0; + if (rmap->rm_flags & XFS_RMAP_ATTR_FORK) + oinfo.oi_flags |= XFS_OWNER_INFO_ATTR_FORK; + if (rmap->rm_flags & XFS_RMAP_BMBT_BLOCK) + oinfo.oi_flags |= XFS_OWNER_INFO_BMBT_BLOCK; + + if (rmap->rm_flags || XFS_RMAP_NON_INODE_OWNER(rmap->rm_owner)) + return xfs_rmap_map(cur, rmap->rm_startblock, + rmap->rm_blockcount, + rmap->rm_flags & XFS_RMAP_UNWRITTEN, + &oinfo); + + return xfs_rmap_map_shared(cur, rmap->rm_startblock, + rmap->rm_blockcount, + rmap->rm_flags & XFS_RMAP_UNWRITTEN, + &oinfo); +} + struct xfs_rmap_query_range_info { xfs_rmap_query_range_fn fn; void *priv; @@ -2454,3 +2483,56 @@ xfs_rmap_record_exists( irec.rm_startblock + irec.rm_blockcount >= bno + len); return 0; } + +struct xfs_rmap_key_state { + uint64_t owner; + uint64_t offset; + unsigned int flags; + bool has_rmap; +}; + +/* For each rmap given, figure out if it doesn't match the key we want. */ +STATIC int +xfs_rmap_has_other_keys_helper( + struct xfs_btree_cur *cur, + struct xfs_rmap_irec *rec, + void *priv) +{ + struct xfs_rmap_key_state *rks = priv; + + if (rks->owner == rec->rm_owner && rks->offset == rec->rm_offset && + ((rks->flags & rec->rm_flags) & XFS_RMAP_KEY_FLAGS) == rks->flags) + return 0; + rks->has_rmap = true; + return XFS_BTREE_QUERY_RANGE_ABORT; +} + +/* + * Given an extent and some owner info, can we find records overlapping + * the extent whose owner info does not match the given owner? + */ +int +xfs_rmap_has_other_keys( + struct xfs_btree_cur *cur, + xfs_agblock_t bno, + xfs_extlen_t len, + struct xfs_owner_info *oinfo, + bool *has_rmap) +{ + struct xfs_rmap_irec low = {0}; + struct xfs_rmap_irec high; + struct xfs_rmap_key_state rks; + int error; + + xfs_owner_info_unpack(oinfo, &rks.owner, &rks.offset, &rks.flags); + rks.has_rmap = false; + + low.rm_startblock = bno; + memset(&high, 0xFF, sizeof(high)); + high.rm_startblock = bno + len - 1; + + error = xfs_rmap_query_range(cur, &low, &high, + xfs_rmap_has_other_keys_helper, &rks); + *has_rmap = rks.has_rmap; + return error; +} diff --git a/fs/xfs/libxfs/xfs_rmap.h b/fs/xfs/libxfs/xfs_rmap.h index 380e53be98d5..43e506f67680 100644 --- a/fs/xfs/libxfs/xfs_rmap.h +++ b/fs/xfs/libxfs/xfs_rmap.h @@ -238,5 +238,9 @@ int xfs_rmap_has_record(struct xfs_btree_cur *cur, xfs_agblock_t bno, int xfs_rmap_record_exists(struct xfs_btree_cur *cur, xfs_agblock_t bno, xfs_extlen_t len, struct xfs_owner_info *oinfo, bool *has_rmap); +int xfs_rmap_has_other_keys(struct xfs_btree_cur *cur, xfs_agblock_t bno, + xfs_extlen_t len, struct xfs_owner_info *oinfo, + bool *has_rmap); +int xfs_rmap_map_raw(struct xfs_btree_cur *cur, struct xfs_rmap_irec *rmap); #endif /* __XFS_RMAP_H__ */ diff --git a/fs/xfs/libxfs/xfs_rmap_btree.c b/fs/xfs/libxfs/xfs_rmap_btree.c index 8b0d0de1cd11..d756e0b84abf 100644 --- a/fs/xfs/libxfs/xfs_rmap_btree.c +++ b/fs/xfs/libxfs/xfs_rmap_btree.c @@ -499,7 +499,6 @@ xfs_rmapbt_init_cursor( */ int xfs_rmapbt_maxrecs( - struct xfs_mount *mp, int blocklen, int leaf) { @@ -534,7 +533,7 @@ xfs_rmapbt_compute_maxlevels( if (xfs_sb_version_hasreflink(&mp->m_sb)) mp->m_rmap_maxlevels = XFS_BTREE_MAXLEVELS; else - mp->m_rmap_maxlevels = xfs_btree_compute_maxlevels(mp, + mp->m_rmap_maxlevels = xfs_btree_compute_maxlevels( mp->m_rmap_mnr, mp->m_sb.sb_agblocks); } @@ -544,7 +543,7 @@ xfs_rmapbt_calc_size( struct xfs_mount *mp, unsigned long long len) { - return xfs_btree_calc_size(mp, mp->m_rmap_mnr, len); + return xfs_btree_calc_size(mp->m_rmap_mnr, len); } /* diff --git a/fs/xfs/libxfs/xfs_rmap_btree.h b/fs/xfs/libxfs/xfs_rmap_btree.h index 19c08e933049..d68d96eed7ea 100644 --- a/fs/xfs/libxfs/xfs_rmap_btree.h +++ b/fs/xfs/libxfs/xfs_rmap_btree.h @@ -55,7 +55,7 @@ struct xfs_mount; struct xfs_btree_cur *xfs_rmapbt_init_cursor(struct xfs_mount *mp, struct xfs_trans *tp, struct xfs_buf *bp, xfs_agnumber_t agno); -int xfs_rmapbt_maxrecs(struct xfs_mount *mp, int blocklen, int leaf); +int xfs_rmapbt_maxrecs(int blocklen, int leaf); extern void xfs_rmapbt_compute_maxlevels(struct xfs_mount *mp); extern xfs_extlen_t xfs_rmapbt_calc_size(struct xfs_mount *mp, diff --git a/fs/xfs/libxfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c index 106be2d0bb88..369eeb7a52ec 100644 --- a/fs/xfs/libxfs/xfs_rtbitmap.c +++ b/fs/xfs/libxfs/xfs_rtbitmap.c @@ -90,6 +90,9 @@ xfs_rtbuf_get( if (error) return error; + if (nmap == 0 || !xfs_bmap_is_real_extent(&map)) + return -EFSCORRUPTED; + ASSERT(map.br_startblock != NULLFSBLOCK); error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, XFS_FSB_TO_DADDR(mp, map.br_startblock), @@ -1033,14 +1036,17 @@ xfs_rtalloc_query_range( int is_free; int error = 0; - if (low_rec->ar_startblock > high_rec->ar_startblock) + if (low_rec->ar_startext > high_rec->ar_startext) return -EINVAL; - else if (low_rec->ar_startblock == high_rec->ar_startblock) + if (low_rec->ar_startext >= mp->m_sb.sb_rextents || + low_rec->ar_startext == high_rec->ar_startext) return 0; + if (high_rec->ar_startext >= mp->m_sb.sb_rextents) + high_rec->ar_startext = mp->m_sb.sb_rextents - 1; /* Iterate the bitmap, looking for discrepancies. */ - rtstart = low_rec->ar_startblock; - rem = high_rec->ar_startblock - rtstart; + rtstart = low_rec->ar_startext; + rem = high_rec->ar_startext - rtstart; while (rem) { /* Is the first block free? */ error = xfs_rtcheck_range(mp, tp, rtstart, 1, 1, &rtend, @@ -1050,13 +1056,13 @@ xfs_rtalloc_query_range( /* How long does the extent go for? */ error = xfs_rtfind_forw(mp, tp, rtstart, - high_rec->ar_startblock - 1, &rtend); + high_rec->ar_startext - 1, &rtend); if (error) break; if (is_free) { - rec.ar_startblock = rtstart; - rec.ar_blockcount = rtend - rtstart + 1; + rec.ar_startext = rtstart; + rec.ar_extcount = rtend - rtstart + 1; error = fn(tp, &rec, priv); if (error) @@ -1079,9 +1085,9 @@ xfs_rtalloc_query_all( { struct xfs_rtalloc_rec keys[2]; - keys[0].ar_startblock = 0; - keys[1].ar_startblock = tp->t_mountp->m_sb.sb_rblocks; - keys[0].ar_blockcount = keys[1].ar_blockcount = 0; + keys[0].ar_startext = 0; + keys[1].ar_startext = tp->t_mountp->m_sb.sb_rextents - 1; + keys[0].ar_extcount = keys[1].ar_extcount = 0; return xfs_rtalloc_query_range(tp, &keys[0], &keys[1], fn, priv); } diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index 53433cc024fd..d485e14313c6 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -756,15 +756,13 @@ xfs_sb_mount_common( mp->m_bmap_dmnr[0] = mp->m_bmap_dmxr[0] / 2; mp->m_bmap_dmnr[1] = mp->m_bmap_dmxr[1] / 2; - mp->m_rmap_mxr[0] = xfs_rmapbt_maxrecs(mp, sbp->sb_blocksize, 1); - mp->m_rmap_mxr[1] = xfs_rmapbt_maxrecs(mp, sbp->sb_blocksize, 0); + mp->m_rmap_mxr[0] = xfs_rmapbt_maxrecs(sbp->sb_blocksize, 1); + mp->m_rmap_mxr[1] = xfs_rmapbt_maxrecs(sbp->sb_blocksize, 0); mp->m_rmap_mnr[0] = mp->m_rmap_mxr[0] / 2; mp->m_rmap_mnr[1] = mp->m_rmap_mxr[1] / 2; - mp->m_refc_mxr[0] = xfs_refcountbt_maxrecs(mp, sbp->sb_blocksize, - true); - mp->m_refc_mxr[1] = xfs_refcountbt_maxrecs(mp, sbp->sb_blocksize, - false); + mp->m_refc_mxr[0] = xfs_refcountbt_maxrecs(sbp->sb_blocksize, true); + mp->m_refc_mxr[1] = xfs_refcountbt_maxrecs(sbp->sb_blocksize, false); mp->m_refc_mnr[0] = mp->m_refc_mxr[0] / 2; mp->m_refc_mnr[1] = mp->m_refc_mxr[1] / 2; @@ -890,6 +888,109 @@ xfs_sync_sb( return xfs_trans_commit(tp); } +/* + * Update all the secondary superblocks to match the new state of the primary. + * Because we are completely overwriting all the existing fields in the + * secondary superblock buffers, there is no need to read them in from disk. + * Just get a new buffer, stamp it and write it. + * + * The sb buffers need to be cached here so that we serialise against other + * operations that access the secondary superblocks, but we don't want to keep + * them in memory once it is written so we mark it as a one-shot buffer. + */ +int +xfs_update_secondary_sbs( + struct xfs_mount *mp) +{ + xfs_agnumber_t agno; + int saved_error = 0; + int error = 0; + LIST_HEAD (buffer_list); + + /* update secondary superblocks. */ + for (agno = 1; agno < mp->m_sb.sb_agcount; agno++) { + struct xfs_buf *bp; + + bp = xfs_buf_get(mp->m_ddev_targp, + XFS_AG_DADDR(mp, agno, XFS_SB_DADDR), + XFS_FSS_TO_BB(mp, 1), 0); + /* + * If we get an error reading or writing alternate superblocks, + * continue. xfs_repair chooses the "best" superblock based + * on most matches; if we break early, we'll leave more + * superblocks un-updated than updated, and xfs_repair may + * pick them over the properly-updated primary. + */ + if (!bp) { + xfs_warn(mp, + "error allocating secondary superblock for ag %d", + agno); + if (!saved_error) + saved_error = -ENOMEM; + continue; + } + + bp->b_ops = &xfs_sb_buf_ops; + xfs_buf_oneshot(bp); + xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); + xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb); + xfs_buf_delwri_queue(bp, &buffer_list); + xfs_buf_relse(bp); + + /* don't hold too many buffers at once */ + if (agno % 16) + continue; + + error = xfs_buf_delwri_submit(&buffer_list); + if (error) { + xfs_warn(mp, + "write error %d updating a secondary superblock near ag %d", + error, agno); + if (!saved_error) + saved_error = error; + continue; + } + } + error = xfs_buf_delwri_submit(&buffer_list); + if (error) { + xfs_warn(mp, + "write error %d updating a secondary superblock near ag %d", + error, agno); + } + + return saved_error ? saved_error : error; +} + +/* + * Same behavior as xfs_sync_sb, except that it is always synchronous and it + * also writes the superblock buffer to disk sector 0 immediately. + */ +int +xfs_sync_sb_buf( + struct xfs_mount *mp) +{ + struct xfs_trans *tp; + int error; + + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_sb, 0, 0, 0, &tp); + if (error) + return error; + + xfs_log_sb(tp); + xfs_trans_bhold(tp, mp->m_sb_bp); + xfs_trans_set_sync(tp); + error = xfs_trans_commit(tp); + if (error) + goto out; + /* + * write out the sb buffer to get the changes to disk + */ + error = xfs_bwrite(mp->m_sb_bp); +out: + xfs_buf_relse(mp->m_sb_bp); + return error; +} + int xfs_fs_geometry( struct xfs_sb *sbp, @@ -974,3 +1075,47 @@ xfs_fs_geometry( return 0; } + +/* Read a secondary superblock. */ +int +xfs_sb_read_secondary( + struct xfs_mount *mp, + struct xfs_trans *tp, + xfs_agnumber_t agno, + struct xfs_buf **bpp) +{ + struct xfs_buf *bp; + int error; + + ASSERT(agno != 0 && agno != NULLAGNUMBER); + error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, + XFS_AG_DADDR(mp, agno, XFS_SB_BLOCK(mp)), + XFS_FSS_TO_BB(mp, 1), 0, &bp, &xfs_sb_buf_ops); + if (error) + return error; + xfs_buf_set_ref(bp, XFS_SSB_REF); + *bpp = bp; + return 0; +} + +/* Get an uninitialised secondary superblock buffer. */ +int +xfs_sb_get_secondary( + struct xfs_mount *mp, + struct xfs_trans *tp, + xfs_agnumber_t agno, + struct xfs_buf **bpp) +{ + struct xfs_buf *bp; + + ASSERT(agno != 0 && agno != NULLAGNUMBER); + bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, + XFS_AG_DADDR(mp, agno, XFS_SB_BLOCK(mp)), + XFS_FSS_TO_BB(mp, 1), 0); + if (!bp) + return -ENOMEM; + bp->b_ops = &xfs_sb_buf_ops; + xfs_buf_oneshot(bp); + *bpp = bp; + return 0; +} diff --git a/fs/xfs/libxfs/xfs_sb.h b/fs/xfs/libxfs/xfs_sb.h index 63dcd2a1a657..244e0162c49e 100644 --- a/fs/xfs/libxfs/xfs_sb.h +++ b/fs/xfs/libxfs/xfs_sb.h @@ -18,6 +18,13 @@ #ifndef __XFS_SB_H__ #define __XFS_SB_H__ +struct xfs_mount; +struct xfs_sb; +struct xfs_dsb; +struct xfs_trans; +struct xfs_fsop_geom; +struct xfs_perag; + /* * perag get/put wrappers for ref counting */ @@ -29,13 +36,22 @@ extern int xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t); extern void xfs_log_sb(struct xfs_trans *tp); extern int xfs_sync_sb(struct xfs_mount *mp, bool wait); +extern int xfs_sync_sb_buf(struct xfs_mount *mp); extern void xfs_sb_mount_common(struct xfs_mount *mp, struct xfs_sb *sbp); extern void xfs_sb_from_disk(struct xfs_sb *to, struct xfs_dsb *from); extern void xfs_sb_to_disk(struct xfs_dsb *to, struct xfs_sb *from); extern void xfs_sb_quota_from_disk(struct xfs_sb *sbp); +extern int xfs_update_secondary_sbs(struct xfs_mount *mp); + #define XFS_FS_GEOM_MAX_STRUCT_VER (4) extern int xfs_fs_geometry(struct xfs_sb *sbp, struct xfs_fsop_geom *geo, int struct_version); +extern int xfs_sb_read_secondary(struct xfs_mount *mp, + struct xfs_trans *tp, xfs_agnumber_t agno, + struct xfs_buf **bpp); +extern int xfs_sb_get_secondary(struct xfs_mount *mp, + struct xfs_trans *tp, xfs_agnumber_t agno, + struct xfs_buf **bpp); #endif /* __XFS_SB_H__ */ diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h index d0b84da0cb1e..ae99c260adb1 100644 --- a/fs/xfs/libxfs/xfs_shared.h +++ b/fs/xfs/libxfs/xfs_shared.h @@ -57,21 +57,6 @@ extern const struct xfs_buf_ops xfs_sb_quiet_buf_ops; extern const struct xfs_buf_ops xfs_symlink_buf_ops; extern const struct xfs_buf_ops xfs_rtbuf_ops; -/* - * This structure is used to track log items associated with - * a transaction. It points to the log item and keeps some - * flags to track the state of the log item. It also tracks - * the amount of space needed to log the item it describes - * once we get to commit processing (see xfs_trans_commit()). - */ -struct xfs_log_item_desc { - struct xfs_log_item *lid_item; - struct list_head lid_trans; - unsigned char lid_flags; -}; - -#define XFS_LID_DIRTY 0x1 - /* log size calculation functions */ int xfs_log_calc_unit_res(struct xfs_mount *mp, int unit_bytes); int xfs_log_calc_minimum_size(struct xfs_mount *); @@ -127,6 +112,7 @@ void xfs_log_get_max_trans_res(struct xfs_mount *mp, #define XFS_ATTR_BTREE_REF 1 #define XFS_DQUOT_REF 1 #define XFS_REFC_BTREE_REF 1 +#define XFS_SSB_REF 0 /* * Flags for xfs_trans_ichgtime(). diff --git a/fs/xfs/libxfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c index 5f17641f040f..3bccdf73e141 100644 --- a/fs/xfs/libxfs/xfs_trans_resv.c +++ b/fs/xfs/libxfs/xfs_trans_resv.c @@ -734,8 +734,7 @@ xfs_calc_clear_agi_bucket_reservation( * the xfs_disk_dquot_t: sizeof(struct xfs_disk_dquot) */ STATIC uint -xfs_calc_qm_setqlim_reservation( - struct xfs_mount *mp) +xfs_calc_qm_setqlim_reservation(void) { return xfs_calc_buf_res(1, sizeof(struct xfs_disk_dquot)); } @@ -772,8 +771,7 @@ xfs_calc_qm_quotaoff_reservation( * the xfs_qoff_logitem_t: sizeof(struct xfs_qoff_logitem) * 2 */ STATIC uint -xfs_calc_qm_quotaoff_end_reservation( - struct xfs_mount *mp) +xfs_calc_qm_quotaoff_end_reservation(void) { return sizeof(struct xfs_qoff_logitem) * 2; } @@ -877,14 +875,14 @@ xfs_trans_resv_calc( * The following transactions are logged in logical format with * a default log count. */ - resp->tr_qm_setqlim.tr_logres = xfs_calc_qm_setqlim_reservation(mp); + resp->tr_qm_setqlim.tr_logres = xfs_calc_qm_setqlim_reservation(); resp->tr_qm_setqlim.tr_logcount = XFS_DEFAULT_LOG_COUNT; resp->tr_qm_quotaoff.tr_logres = xfs_calc_qm_quotaoff_reservation(mp); resp->tr_qm_quotaoff.tr_logcount = XFS_DEFAULT_LOG_COUNT; resp->tr_qm_equotaoff.tr_logres = - xfs_calc_qm_quotaoff_end_reservation(mp); + xfs_calc_qm_quotaoff_end_reservation(); resp->tr_qm_equotaoff.tr_logcount = XFS_DEFAULT_LOG_COUNT; resp->tr_sb.tr_logres = xfs_calc_sb_reservation(mp); diff --git a/fs/xfs/libxfs/xfs_types.h b/fs/xfs/libxfs/xfs_types.h index 3c560695c546..ea18449bd732 100644 --- a/fs/xfs/libxfs/xfs_types.h +++ b/fs/xfs/libxfs/xfs_types.h @@ -30,7 +30,7 @@ typedef int64_t xfs_fsize_t; /* bytes in a file */ typedef uint64_t xfs_ufsize_t; /* unsigned bytes in a file */ typedef int32_t xfs_suminfo_t; /* type of bitmap summary info */ -typedef int32_t xfs_rtword_t; /* word type for bitmap manipulations */ +typedef uint32_t xfs_rtword_t; /* word type for bitmap manipulations */ typedef int64_t xfs_lsn_t; /* log sequence number */ typedef int32_t xfs_tid_t; /* transaction identifier */ diff --git a/fs/xfs/scrub/agheader.c b/fs/xfs/scrub/agheader.c index 018aabbd9394..1f71793f7db4 100644 --- a/fs/xfs/scrub/agheader.c +++ b/fs/xfs/scrub/agheader.c @@ -38,68 +38,6 @@ #include "scrub/common.h" #include "scrub/trace.h" -/* - * Walk all the blocks in the AGFL. The fn function can return any negative - * error code or XFS_BTREE_QUERY_RANGE_ABORT. - */ -int -xfs_scrub_walk_agfl( - struct xfs_scrub_context *sc, - int (*fn)(struct xfs_scrub_context *, - xfs_agblock_t bno, void *), - void *priv) -{ - struct xfs_agf *agf; - __be32 *agfl_bno; - struct xfs_mount *mp = sc->mp; - unsigned int flfirst; - unsigned int fllast; - int i; - int error; - - agf = XFS_BUF_TO_AGF(sc->sa.agf_bp); - agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, sc->sa.agfl_bp); - flfirst = be32_to_cpu(agf->agf_flfirst); - fllast = be32_to_cpu(agf->agf_fllast); - - /* Nothing to walk in an empty AGFL. */ - if (agf->agf_flcount == cpu_to_be32(0)) - return 0; - - /* first to last is a consecutive list. */ - if (fllast >= flfirst) { - for (i = flfirst; i <= fllast; i++) { - error = fn(sc, be32_to_cpu(agfl_bno[i]), priv); - if (error) - return error; - if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) - return error; - } - - return 0; - } - - /* first to the end */ - for (i = flfirst; i < xfs_agfl_size(mp); i++) { - error = fn(sc, be32_to_cpu(agfl_bno[i]), priv); - if (error) - return error; - if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) - return error; - } - - /* the start to last. */ - for (i = 0; i <= fllast; i++) { - error = fn(sc, be32_to_cpu(agfl_bno[i]), priv); - if (error) - return error; - if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) - return error; - } - - return 0; -} - /* Superblock */ /* Cross-reference with the other btrees. */ @@ -157,9 +95,7 @@ xfs_scrub_superblock( if (agno == 0) return 0; - error = xfs_trans_read_buf(mp, sc->tp, mp->m_ddev_targp, - XFS_AGB_TO_DADDR(mp, agno, XFS_SB_BLOCK(mp)), - XFS_FSS_TO_BB(mp, 1), 0, &bp, &xfs_sb_buf_ops); + error = xfs_sb_read_secondary(mp, sc->tp, agno, &bp); /* * The superblock verifier can return several different error codes * if it thinks the superblock doesn't look right. For a mount these @@ -680,6 +616,7 @@ struct xfs_scrub_agfl_info { unsigned int sz_entries; unsigned int nr_entries; xfs_agblock_t *entries; + struct xfs_scrub_context *sc; }; /* Cross-reference with the other btrees. */ @@ -701,12 +638,12 @@ xfs_scrub_agfl_block_xref( /* Scrub an AGFL block. */ STATIC int xfs_scrub_agfl_block( - struct xfs_scrub_context *sc, + struct xfs_mount *mp, xfs_agblock_t agbno, void *priv) { - struct xfs_mount *mp = sc->mp; struct xfs_scrub_agfl_info *sai = priv; + struct xfs_scrub_context *sc = sai->sc; xfs_agnumber_t agno = sc->sa.agno; if (xfs_verify_agbno(mp, agno, agbno) && @@ -717,6 +654,9 @@ xfs_scrub_agfl_block( xfs_scrub_agfl_block_xref(sc, agbno, priv); + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + return XFS_BTREE_QUERY_RANGE_ABORT; + return 0; } @@ -796,8 +736,10 @@ xfs_scrub_agfl( goto out; } memset(&sai, 0, sizeof(sai)); + sai.sc = sc; sai.sz_entries = agflcount; - sai.entries = kmem_zalloc(sizeof(xfs_agblock_t) * agflcount, KM_NOFS); + sai.entries = kmem_zalloc(sizeof(xfs_agblock_t) * agflcount, + KM_MAYFAIL); if (!sai.entries) { error = -ENOMEM; goto out; @@ -805,7 +747,12 @@ xfs_scrub_agfl( /* Check the blocks in the AGFL. */ xfs_rmap_ag_owner(&sai.oinfo, XFS_RMAP_OWN_AG); - error = xfs_scrub_walk_agfl(sc, xfs_scrub_agfl_block, &sai); + error = xfs_agfl_walk(sc->mp, XFS_BUF_TO_AGF(sc->sa.agf_bp), + sc->sa.agfl_bp, xfs_scrub_agfl_block, &sai); + if (error == XFS_BTREE_QUERY_RANGE_ABORT) { + error = 0; + goto out_free; + } if (error) goto out_free; diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c new file mode 100644 index 000000000000..8b91e9ebe1e7 --- /dev/null +++ b/fs/xfs/scrub/agheader_repair.c @@ -0,0 +1,70 @@ +/* + * Copyright (C) 2018 Oracle. All Rights Reserved. + * + * Author: Darrick J. Wong <darrick.wong@oracle.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA. + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_btree.h" +#include "xfs_bit.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_inode.h" +#include "xfs_alloc.h" +#include "xfs_ialloc.h" +#include "xfs_rmap.h" +#include "scrub/xfs_scrub.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" + +/* Superblock */ + +/* Repair the superblock. */ +int +xfs_repair_superblock( + struct xfs_scrub_context *sc) +{ + struct xfs_mount *mp = sc->mp; + struct xfs_buf *bp; + xfs_agnumber_t agno; + int error; + + /* Don't try to repair AG 0's sb; let xfs_repair deal with it. */ + agno = sc->sm->sm_agno; + if (agno == 0) + return -EOPNOTSUPP; + + error = xfs_sb_get_secondary(mp, sc->tp, agno, &bp); + if (error) + return error; + + /* Copy AG 0's superblock to this one. */ + xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); + xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb); + + /* Write this to disk. */ + xfs_trans_buf_set_type(sc->tp, bp, XFS_BLFT_SB_BUF); + xfs_trans_log_buf(sc->tp, bp, 0, BBTOB(bp->b_length) - 1); + return error; +} diff --git a/fs/xfs/scrub/alloc.c b/fs/xfs/scrub/alloc.c index 517c079d3f68..941a0a55224e 100644 --- a/fs/xfs/scrub/alloc.c +++ b/fs/xfs/scrub/alloc.c @@ -70,7 +70,7 @@ xfs_scrub_allocbt_xref_other( pcur = &sc->sa.cnt_cur; else pcur = &sc->sa.bno_cur; - if (!*pcur) + if (!*pcur || xfs_scrub_skip_xref(sc->sm)) return; error = xfs_alloc_lookup_le(*pcur, agbno, len, &has_otherrec); @@ -172,7 +172,7 @@ xfs_scrub_xref_is_used_space( bool is_freesp; int error; - if (!sc->sa.bno_cur) + if (!sc->sa.bno_cur || xfs_scrub_skip_xref(sc->sm)) return; error = xfs_alloc_has_record(sc->sa.bno_cur, agbno, len, &is_freesp); diff --git a/fs/xfs/scrub/attr.c b/fs/xfs/scrub/attr.c index 127575f0abfb..84b6d6b66578 100644 --- a/fs/xfs/scrub/attr.c +++ b/fs/xfs/scrub/attr.c @@ -126,8 +126,9 @@ xfs_scrub_xattr_listent( if (args.valuelen != valuelen) xfs_scrub_fblock_set_corrupt(sx->sc, XFS_ATTR_FORK, args.blkno); - fail_xref: + if (sx->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + context->seen_enough = 1; return; } diff --git a/fs/xfs/scrub/bmap.c b/fs/xfs/scrub/bmap.c index 639d14b51e90..eeadb33a701c 100644 --- a/fs/xfs/scrub/bmap.c +++ b/fs/xfs/scrub/bmap.c @@ -51,7 +51,6 @@ xfs_scrub_setup_inode_bmap( struct xfs_scrub_context *sc, struct xfs_inode *ip) { - struct xfs_mount *mp = sc->mp; int error; error = xfs_scrub_get_inode(sc, ip); @@ -75,7 +74,7 @@ xfs_scrub_setup_inode_bmap( } /* Got the inode, lock it and we're ready to go. */ - error = xfs_scrub_trans_alloc(sc->sm, mp, &sc->tp); + error = xfs_scrub_trans_alloc(sc, 0); if (error) goto out; sc->ilock_flags |= XFS_ILOCK_EXCL; @@ -175,7 +174,7 @@ xfs_scrub_bmap_xref_rmap( unsigned long long rmap_end; uint64_t owner; - if (!info->sc->sa.rmap_cur) + if (!info->sc->sa.rmap_cur || xfs_scrub_skip_xref(info->sc->sm)) return; if (info->whichfork == XFS_COW_FORK) @@ -684,7 +683,8 @@ xfs_scrub_bmap( info.lastoff = 0; ifp = XFS_IFORK_PTR(ip, whichfork); for_each_xfs_iext(ifp, &icur, &irec) { - if (xfs_scrub_should_terminate(sc, &error)) + if (xfs_scrub_should_terminate(sc, &error) || + (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) break; if (isnullstartblock(irec.br_startblock)) continue; diff --git a/fs/xfs/scrub/btree.c b/fs/xfs/scrub/btree.c index 54218168c8f9..2d29dceaa00e 100644 --- a/fs/xfs/scrub/btree.c +++ b/fs/xfs/scrub/btree.c @@ -442,7 +442,7 @@ xfs_scrub_btree_check_owner( */ if (cur->bc_btnum == XFS_BTNUM_BNO || cur->bc_btnum == XFS_BTNUM_RMAP) { co = kmem_alloc(sizeof(struct check_owner), - KM_MAYFAIL | KM_NOFS); + KM_MAYFAIL); if (!co) return -ENOMEM; co->level = level; @@ -455,6 +455,44 @@ xfs_scrub_btree_check_owner( } /* + * Check that this btree block has at least minrecs records or is one of the + * special blocks that don't require that. + */ +STATIC void +xfs_scrub_btree_check_minrecs( + struct xfs_scrub_btree *bs, + int level, + struct xfs_btree_block *block) +{ + unsigned int numrecs; + int ok_level; + + numrecs = be16_to_cpu(block->bb_numrecs); + + /* More records than minrecs means the block is ok. */ + if (numrecs >= bs->cur->bc_ops->get_minrecs(bs->cur, level)) + return; + + /* + * Certain btree blocks /can/ have fewer than minrecs records. Any + * level greater than or equal to the level of the highest dedicated + * btree block are allowed to violate this constraint. + * + * For a btree rooted in a block, the btree root can have fewer than + * minrecs records. If the btree is rooted in an inode and does not + * store records in the root, the direct children of the root and the + * root itself can have fewer than minrecs records. + */ + ok_level = bs->cur->bc_nlevels - 1; + if (bs->cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) + ok_level--; + if (level >= ok_level) + return; + + xfs_scrub_btree_set_corrupt(bs->sc, bs->cur, level); +} + +/* * Grab and scrub a btree block given a btree pointer. Returns block * and buffer pointers (if applicable) if they're ok to use. */ @@ -491,6 +529,8 @@ xfs_scrub_btree_get_block( if (*pbp) xfs_scrub_buffer_recheck(bs->sc, *pbp); + xfs_scrub_btree_check_minrecs(bs, level, *pblock); + /* * Check the block's owner; this function absorbs error codes * for us. diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index 8ed91d5c868d..41198a5f872c 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -44,11 +44,14 @@ #include "xfs_rmap_btree.h" #include "xfs_log.h" #include "xfs_trans_priv.h" +#include "xfs_attr.h" +#include "xfs_reflink.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" #include "scrub/btree.h" +#include "scrub/repair.h" /* Common code for the metadata scrubbers. */ @@ -539,6 +542,10 @@ xfs_scrub_ag_free( xfs_trans_brelse(sc->tp, sa->agi_bp); sa->agi_bp = NULL; } + if (sa->pag) { + xfs_perag_put(sa->pag); + sa->pag = NULL; + } sa->agno = NULLAGNUMBER; } @@ -566,15 +573,53 @@ xfs_scrub_ag_init( return xfs_scrub_ag_btcur_init(sc, sa); } +/* + * Grab the per-ag structure if we haven't already gotten it. Teardown of the + * xfs_scrub_ag will release it for us. + */ +void +xfs_scrub_perag_get( + struct xfs_mount *mp, + struct xfs_scrub_ag *sa) +{ + if (!sa->pag) + sa->pag = xfs_perag_get(mp, sa->agno); +} + /* Per-scrubber setup functions */ +/* + * Grab an empty transaction so that we can re-grab locked buffers if + * one of our btrees turns out to be cyclic. + * + * If we're going to repair something, we need to ask for the largest possible + * log reservation so that we can handle the worst case scenario for metadata + * updates while rebuilding a metadata item. We also need to reserve as many + * blocks in the head transaction as we think we're going to need to rebuild + * the metadata object. + */ +int +xfs_scrub_trans_alloc( + struct xfs_scrub_context *sc, + uint resblks) +{ + if (sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) + return xfs_trans_alloc(sc->mp, &M_RES(sc->mp)->tr_itruncate, + resblks, 0, 0, &sc->tp); + + return xfs_trans_alloc_empty(sc->mp, &sc->tp); +} + /* Set us up with a transaction and an empty context. */ int xfs_scrub_setup_fs( struct xfs_scrub_context *sc, struct xfs_inode *ip) { - return xfs_scrub_trans_alloc(sc->sm, sc->mp, &sc->tp); + uint resblks; + + resblks = xfs_repair_calc_ag_resblks(sc); + return xfs_scrub_trans_alloc(sc, resblks); } /* Set us up with AG headers and btree cursors. */ @@ -695,7 +740,6 @@ xfs_scrub_setup_inode_contents( struct xfs_inode *ip, unsigned int resblks) { - struct xfs_mount *mp = sc->mp; int error; error = xfs_scrub_get_inode(sc, ip); @@ -705,7 +749,7 @@ xfs_scrub_setup_inode_contents( /* Got the inode, lock it and we're ready to go. */ sc->ilock_flags = XFS_IOLOCK_EXCL | XFS_MMAPLOCK_EXCL; xfs_ilock(sc->ip, sc->ilock_flags); - error = xfs_scrub_trans_alloc(sc->sm, mp, &sc->tp); + error = xfs_scrub_trans_alloc(sc, resblks); if (error) goto out; sc->ilock_flags |= XFS_ILOCK_EXCL; @@ -727,6 +771,10 @@ xfs_scrub_should_check_xref( int *error, struct xfs_btree_cur **curpp) { + /* No point in xref if we already know we're corrupt. */ + if (xfs_scrub_skip_xref(sc->sm)) + return false; + if (*error == 0) return true; @@ -773,3 +821,80 @@ xfs_scrub_buffer_recheck( sc->sm->sm_flags |= XFS_SCRUB_OFLAG_CORRUPT; trace_xfs_scrub_block_error(sc, bp->b_bn, fa); } + +/* + * Scrub the attr/data forks of a metadata inode. The metadata inode must be + * pointed to by sc->ip and the ILOCK must be held. + */ +int +xfs_scrub_metadata_inode_forks( + struct xfs_scrub_context *sc) +{ + __u32 smtype; + bool shared; + int error; + + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + return 0; + + /* Metadata inodes don't live on the rt device. */ + if (sc->ip->i_d.di_flags & XFS_DIFLAG_REALTIME) { + xfs_scrub_ino_set_corrupt(sc, sc->ip->i_ino); + return 0; + } + + /* They should never participate in reflink. */ + if (xfs_is_reflink_inode(sc->ip)) { + xfs_scrub_ino_set_corrupt(sc, sc->ip->i_ino); + return 0; + } + + /* They also should never have extended attributes. */ + if (xfs_inode_hasattr(sc->ip)) { + xfs_scrub_ino_set_corrupt(sc, sc->ip->i_ino); + return 0; + } + + /* Invoke the data fork scrubber. */ + smtype = sc->sm->sm_type; + sc->sm->sm_type = XFS_SCRUB_TYPE_BMBTD; + error = xfs_scrub_bmap_data(sc); + sc->sm->sm_type = smtype; + if (error || (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) + return error; + + /* Look for incorrect shared blocks. */ + if (xfs_sb_version_hasreflink(&sc->mp->m_sb)) { + error = xfs_reflink_inode_has_shared_extents(sc->tp, sc->ip, + &shared); + if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, 0, + &error)) + return error; + if (shared) + xfs_scrub_ino_set_corrupt(sc, sc->ip->i_ino); + } + + return error; +} + +/* + * Try to lock an inode in violation of the usual locking order rules. For + * example, trying to get the IOLOCK while in transaction context, or just + * plain breaking AG-order or inode-order inode locking rules. Either way, + * the only way to avoid an ABBA deadlock is to use trylock and back off if + * we can't. + */ +int +xfs_scrub_ilock_inverted( + struct xfs_inode *ip, + uint lock_mode) +{ + int i; + + for (i = 0; i < 20; i++) { + if (xfs_ilock_nowait(ip, lock_mode)) + return 0; + delay(1); + } + return -EDEADLOCK; +} diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h index deaf60400981..76bb2d1d808c 100644 --- a/fs/xfs/scrub/common.h +++ b/fs/xfs/scrub/common.h @@ -38,19 +38,7 @@ xfs_scrub_should_terminate( return false; } -/* - * Grab an empty transaction so that we can re-grab locked buffers if - * one of our btrees turns out to be cyclic. - */ -static inline int -xfs_scrub_trans_alloc( - struct xfs_scrub_metadata *sm, - struct xfs_mount *mp, - struct xfs_trans **tpp) -{ - return xfs_trans_alloc_empty(mp, tpp); -} - +int xfs_scrub_trans_alloc(struct xfs_scrub_context *sc, uint resblks); bool xfs_scrub_process_error(struct xfs_scrub_context *sc, xfs_agnumber_t agno, xfs_agblock_t bno, int *error); bool xfs_scrub_fblock_process_error(struct xfs_scrub_context *sc, int whichfork, @@ -135,16 +123,13 @@ xfs_scrub_setup_quota(struct xfs_scrub_context *sc, struct xfs_inode *ip) void xfs_scrub_ag_free(struct xfs_scrub_context *sc, struct xfs_scrub_ag *sa); int xfs_scrub_ag_init(struct xfs_scrub_context *sc, xfs_agnumber_t agno, struct xfs_scrub_ag *sa); +void xfs_scrub_perag_get(struct xfs_mount *mp, struct xfs_scrub_ag *sa); int xfs_scrub_ag_read_headers(struct xfs_scrub_context *sc, xfs_agnumber_t agno, struct xfs_buf **agi, struct xfs_buf **agf, struct xfs_buf **agfl); void xfs_scrub_ag_btcur_free(struct xfs_scrub_ag *sa); int xfs_scrub_ag_btcur_init(struct xfs_scrub_context *sc, struct xfs_scrub_ag *sa); -int xfs_scrub_walk_agfl(struct xfs_scrub_context *sc, - int (*fn)(struct xfs_scrub_context *, xfs_agblock_t bno, - void *), - void *priv); int xfs_scrub_count_rmap_ownedby_ag(struct xfs_scrub_context *sc, struct xfs_btree_cur *cur, struct xfs_owner_info *oinfo, @@ -157,4 +142,17 @@ int xfs_scrub_setup_inode_contents(struct xfs_scrub_context *sc, struct xfs_inode *ip, unsigned int resblks); void xfs_scrub_buffer_recheck(struct xfs_scrub_context *sc, struct xfs_buf *bp); +/* + * Don't bother cross-referencing if we already found corruption or cross + * referencing discrepancies. + */ +static inline bool xfs_scrub_skip_xref(struct xfs_scrub_metadata *sm) +{ + return sm->sm_flags & (XFS_SCRUB_OFLAG_CORRUPT | + XFS_SCRUB_OFLAG_XCORRUPT); +} + +int xfs_scrub_metadata_inode_forks(struct xfs_scrub_context *sc); +int xfs_scrub_ilock_inverted(struct xfs_inode *ip, uint lock_mode); + #endif /* __XFS_SCRUB_COMMON_H__ */ diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c index 38f29806eb54..1a4309b3e786 100644 --- a/fs/xfs/scrub/dir.c +++ b/fs/xfs/scrub/dir.c @@ -172,7 +172,7 @@ xfs_scrub_dir_actor( error = xfs_dir_lookup(sdc->sc->tp, ip, &xname, &lookup_ino, NULL); if (!xfs_scrub_fblock_process_error(sdc->sc, XFS_DATA_FORK, offset, &error)) - goto fail_xref; + goto out; if (lookup_ino != ino) { xfs_scrub_fblock_set_corrupt(sdc->sc, XFS_DATA_FORK, offset); goto out; @@ -183,8 +183,13 @@ xfs_scrub_dir_actor( if (error) goto out; out: - return error; -fail_xref: + /* + * A negative error code returned here is supposed to cause the + * dir_emit caller (xfs_readdir) to abort the directory iteration + * and return zero to xfs_scrub_directory. + */ + if (error == 0 && sdc->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + return -EFSCORRUPTED; return error; } @@ -240,6 +245,9 @@ xfs_scrub_dir_rec( } xfs_scrub_buffer_recheck(ds->sc, bp); + if (ds->sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out_relse; + dent = (struct xfs_dir2_data_entry *)(((char *)bp->b_addr) + off); /* Make sure we got a real directory entry. */ @@ -357,6 +365,9 @@ xfs_scrub_directory_data_bestfree( /* XXX: Check xfs_dir3_data_hdr.pad is zero once we start setting it. */ + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out_buf; + /* Do the bestfrees correspond to actual free space? */ bf = d_ops->data_bestfree_p(bp->b_addr); smallest_bestfree = UINT_MAX; @@ -413,14 +424,18 @@ xfs_scrub_directory_data_bestfree( /* Spot check this free entry */ tag = be16_to_cpu(*xfs_dir2_data_unused_tag_p(dup)); - if (tag != ((char *)dup - (char *)bp->b_addr)) + if (tag != ((char *)dup - (char *)bp->b_addr)) { xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk); + goto out_buf; + } /* * Either this entry is a bestfree or it's smaller than * any of the bestfrees. */ xfs_scrub_directory_check_free_entry(sc, lblk, bf, dup); + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out_buf; /* Move on. */ newlen = be16_to_cpu(dup->length); @@ -546,6 +561,8 @@ xfs_scrub_directory_leaf1_bestfree( } if (leafhdr.stale != stale) xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk); + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out; /* Check all the bestfree entries. */ for (i = 0; i < bestcount; i++, bestp++) { @@ -556,9 +573,11 @@ xfs_scrub_directory_leaf1_bestfree( i * args->geo->fsbcount, -1, &dbp); if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error)) - continue; + break; xfs_scrub_directory_check_freesp(sc, lblk, dbp, best); xfs_trans_brelse(sc->tp, dbp); + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out; } out: return error; @@ -607,7 +626,7 @@ xfs_scrub_directory_free_bestfree( -1, &dbp); if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, lblk, &error)) - continue; + break; xfs_scrub_directory_check_freesp(sc, lblk, dbp, best); xfs_trans_brelse(sc->tp, dbp); } @@ -656,7 +675,7 @@ xfs_scrub_directory_blocks( /* Iterate all the data extents in the directory... */ found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &icur, &got); - while (found) { + while (found && !(sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) { /* Block directories only have a single block at offset 0. */ if (is_block && (got.br_startoff > 0 || @@ -719,7 +738,7 @@ xfs_scrub_directory_blocks( /* Scan for free blocks */ lblk = free_lblk; found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &icur, &got); - while (found) { + while (found && !(sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) { /* * Dirs can't have blocks mapped above 2^32. * Single-block dirs shouldn't even be here. diff --git a/fs/xfs/scrub/ialloc.c b/fs/xfs/scrub/ialloc.c index 106ca4bd753f..00a834d3b56d 100644 --- a/fs/xfs/scrub/ialloc.c +++ b/fs/xfs/scrub/ialloc.c @@ -387,7 +387,8 @@ xfs_scrub_iallocbt_xref_rmap_btreeblks( int error; if (!sc->sa.ino_cur || !sc->sa.rmap_cur || - (xfs_sb_version_hasfinobt(&sc->mp->m_sb) && !sc->sa.fino_cur)) + (xfs_sb_version_hasfinobt(&sc->mp->m_sb) && !sc->sa.fino_cur) || + xfs_scrub_skip_xref(sc->sm)) return; /* Check that we saw as many inobt blocks as the rmap says. */ @@ -424,7 +425,7 @@ xfs_scrub_iallocbt_xref_rmap_inodes( xfs_filblks_t blocks; int error; - if (!sc->sa.rmap_cur) + if (!sc->sa.rmap_cur || xfs_scrub_skip_xref(sc->sm)) return; /* Check that we saw as many inode blocks as the rmap knows about. */ @@ -496,7 +497,7 @@ xfs_scrub_xref_inode_check( bool has_inodes; int error; - if (!(*icur)) + if (!(*icur) || xfs_scrub_skip_xref(sc->sm)) return; error = xfs_ialloc_has_inodes_at_extent(*icur, agbno, len, &has_inodes); diff --git a/fs/xfs/scrub/inode.c b/fs/xfs/scrub/inode.c index df14930e4fc5..0c696f7018de 100644 --- a/fs/xfs/scrub/inode.c +++ b/fs/xfs/scrub/inode.c @@ -55,7 +55,6 @@ xfs_scrub_setup_inode( struct xfs_scrub_context *sc, struct xfs_inode *ip) { - struct xfs_mount *mp = sc->mp; int error; /* @@ -68,7 +67,7 @@ xfs_scrub_setup_inode( break; case -EFSCORRUPTED: case -EFSBADCRC: - return xfs_scrub_trans_alloc(sc->sm, mp, &sc->tp); + return xfs_scrub_trans_alloc(sc, 0); default: return error; } @@ -76,7 +75,7 @@ xfs_scrub_setup_inode( /* Got the inode, lock it and we're ready to go. */ sc->ilock_flags = XFS_IOLOCK_EXCL | XFS_MMAPLOCK_EXCL; xfs_ilock(sc->ip, sc->ilock_flags); - error = xfs_scrub_trans_alloc(sc->sm, mp, &sc->tp); + error = xfs_scrub_trans_alloc(sc, 0); if (error) goto out; sc->ilock_flags |= XFS_ILOCK_EXCL; @@ -449,7 +448,7 @@ xfs_scrub_inode_xref_finobt( int has_record; int error; - if (!sc->sa.fino_cur) + if (!sc->sa.fino_cur || xfs_scrub_skip_xref(sc->sm)) return; agino = XFS_INO_TO_AGINO(sc->mp, ino); @@ -492,6 +491,9 @@ xfs_scrub_inode_xref_bmap( xfs_filblks_t acount; int error; + if (xfs_scrub_skip_xref(sc->sm)) + return; + /* Walk all the extents to check nextents/naextents/nblocks. */ error = xfs_bmap_count_blocks(sc->tp, sc->ip, XFS_DATA_FORK, &nextents, &count); diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c index 1fb88c18d455..77c6b22c6bfd 100644 --- a/fs/xfs/scrub/parent.c +++ b/fs/xfs/scrub/parent.c @@ -147,6 +147,9 @@ xfs_scrub_parent_validate( *try_again = false; + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out; + /* '..' must not point to ourselves. */ if (sc->ip->i_ino == dnum) { xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); @@ -211,7 +214,9 @@ xfs_scrub_parent_validate( */ xfs_iunlock(sc->ip, sc->ilock_flags); sc->ilock_flags = 0; - xfs_ilock(dp, XFS_IOLOCK_SHARED); + error = xfs_scrub_ilock_inverted(dp, XFS_IOLOCK_SHARED); + if (error) + goto out_rele; /* Go looking for our dentry. */ error = xfs_scrub_parent_count_parent_dentries(sc, dp, &nlink); @@ -220,8 +225,10 @@ xfs_scrub_parent_validate( /* Drop the parent lock, relock this inode. */ xfs_iunlock(dp, XFS_IOLOCK_SHARED); + error = xfs_scrub_ilock_inverted(sc->ip, XFS_IOLOCK_EXCL); + if (error) + goto out_rele; sc->ilock_flags = XFS_IOLOCK_EXCL; - xfs_ilock(sc->ip, sc->ilock_flags); /* * If we're an unlinked directory, the parent /won't/ have a link @@ -323,5 +330,13 @@ xfs_scrub_parent( if (try_again && tries == 20) xfs_scrub_set_incomplete(sc); out: + /* + * If we failed to lock the parent inode even after a retry, just mark + * this scrub incomplete and return. + */ + if (sc->try_harder && error == -EDEADLOCK) { + error = 0; + xfs_scrub_set_incomplete(sc); + } return error; } diff --git a/fs/xfs/scrub/quota.c b/fs/xfs/scrub/quota.c index 6ba465e6c885..15ae4d23d6ac 100644 --- a/fs/xfs/scrub/quota.c +++ b/fs/xfs/scrub/quota.c @@ -66,25 +66,43 @@ xfs_scrub_setup_quota( struct xfs_inode *ip) { uint dqtype; + int error; + + if (!XFS_IS_QUOTA_RUNNING(sc->mp) || !XFS_IS_QUOTA_ON(sc->mp)) + return -ENOENT; dqtype = xfs_scrub_quota_to_dqtype(sc); if (dqtype == 0) return -EINVAL; + sc->has_quotaofflock = true; + mutex_lock(&sc->mp->m_quotainfo->qi_quotaofflock); if (!xfs_this_quota_on(sc->mp, dqtype)) return -ENOENT; + error = xfs_scrub_setup_fs(sc, ip); + if (error) + return error; + sc->ip = xfs_quota_inode(sc->mp, dqtype); + xfs_ilock(sc->ip, XFS_ILOCK_EXCL); + sc->ilock_flags = XFS_ILOCK_EXCL; return 0; } /* Quotas. */ +struct xfs_scrub_quota_info { + struct xfs_scrub_context *sc; + xfs_dqid_t last_id; +}; + /* Scrub the fields in an individual quota item. */ -STATIC void +STATIC int xfs_scrub_quota_item( - struct xfs_scrub_context *sc, - uint dqtype, struct xfs_dquot *dq, - xfs_dqid_t id) + uint dqtype, + void *priv) { + struct xfs_scrub_quota_info *sqi = priv; + struct xfs_scrub_context *sc = sqi->sc; struct xfs_mount *mp = sc->mp; struct xfs_disk_dquot *d = &dq->q_core; struct xfs_quotainfo *qi = mp->m_quotainfo; @@ -99,17 +117,18 @@ xfs_scrub_quota_item( unsigned long long icount; unsigned long long rcount; xfs_ino_t fs_icount; - - offset = id / qi->qi_dqperchunk; + xfs_dqid_t id = be32_to_cpu(d->d_id); /* - * We fed $id and DQNEXT into the xfs_qm_dqget call, which means - * that the actual dquot we got must either have the same id or - * the next higher id. + * Except for the root dquot, the actual dquot we got must either have + * the same or higher id as we saw before. */ - if (id > be32_to_cpu(d->d_id)) + offset = id / qi->qi_dqperchunk; + if (id && id <= sqi->last_id) xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, offset); + sqi->last_id = id; + /* Did we get the dquot type we wanted? */ if (dqtype != (d->d_flags & XFS_DQ_ALLTYPES)) xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, offset); @@ -183,115 +202,85 @@ xfs_scrub_quota_item( xfs_scrub_fblock_set_warning(sc, XFS_DATA_FORK, offset); if (id != 0 && rhard != 0 && rcount > rhard) xfs_scrub_fblock_set_warning(sc, XFS_DATA_FORK, offset); + + return 0; } -/* Scrub all of a quota type's items. */ -int -xfs_scrub_quota( +/* Check the quota's data fork. */ +STATIC int +xfs_scrub_quota_data_fork( struct xfs_scrub_context *sc) { struct xfs_bmbt_irec irec = { 0 }; - struct xfs_mount *mp = sc->mp; - struct xfs_inode *ip; - struct xfs_quotainfo *qi = mp->m_quotainfo; - struct xfs_dquot *dq; + struct xfs_iext_cursor icur; + struct xfs_quotainfo *qi = sc->mp->m_quotainfo; + struct xfs_ifork *ifp; xfs_fileoff_t max_dqid_off; - xfs_fileoff_t off = 0; - xfs_dqid_t id = 0; - uint dqtype; - int nimaps; int error = 0; - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return -ENOENT; - - mutex_lock(&qi->qi_quotaofflock); - dqtype = xfs_scrub_quota_to_dqtype(sc); - if (!xfs_this_quota_on(sc->mp, dqtype)) { - error = -ENOENT; - goto out_unlock_quota; - } - - /* Attach to the quota inode and set sc->ip so that reporting works. */ - ip = xfs_quota_inode(sc->mp, dqtype); - sc->ip = ip; + /* Invoke the fork scrubber. */ + error = xfs_scrub_metadata_inode_forks(sc); + if (error || (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) + return error; - /* Look for problem extents. */ - xfs_ilock(ip, XFS_ILOCK_EXCL); - if (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) { - xfs_scrub_ino_set_corrupt(sc, sc->ip->i_ino); - goto out_unlock_inode; - } + /* Check for data fork problems that apply only to quota files. */ max_dqid_off = ((xfs_dqid_t)-1) / qi->qi_dqperchunk; - while (1) { + ifp = XFS_IFORK_PTR(sc->ip, XFS_DATA_FORK); + for_each_xfs_iext(ifp, &icur, &irec) { if (xfs_scrub_should_terminate(sc, &error)) break; - - off = irec.br_startoff + irec.br_blockcount; - nimaps = 1; - error = xfs_bmapi_read(ip, off, -1, &irec, &nimaps, - XFS_BMAPI_ENTIRE); - if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, off, - &error)) - goto out_unlock_inode; - if (!nimaps) - break; - if (irec.br_startblock == HOLESTARTBLOCK) - continue; - - /* Check the extent record doesn't point to crap. */ - if (irec.br_startblock + irec.br_blockcount <= - irec.br_startblock) - xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, - irec.br_startoff); - if (!xfs_verify_fsbno(mp, irec.br_startblock) || - !xfs_verify_fsbno(mp, irec.br_startblock + - irec.br_blockcount - 1)) - xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, - irec.br_startoff); - /* - * Unwritten extents or blocks mapped above the highest + * delalloc extents or blocks mapped above the highest * quota id shouldn't happen. */ if (isnullstartblock(irec.br_startblock) || irec.br_startoff > max_dqid_off || - irec.br_startoff + irec.br_blockcount > max_dqid_off + 1) - xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, off); + irec.br_startoff + irec.br_blockcount - 1 > max_dqid_off) { + xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, + irec.br_startoff); + break; + } } - xfs_iunlock(ip, XFS_ILOCK_EXCL); - if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) - goto out; - /* Check all the quota items. */ - while (id < ((xfs_dqid_t)-1ULL)) { - if (xfs_scrub_should_terminate(sc, &error)) - break; + return error; +} - error = xfs_qm_dqget(mp, NULL, id, dqtype, XFS_QMOPT_DQNEXT, - &dq); - if (error == -ENOENT) - break; - if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, - id * qi->qi_dqperchunk, &error)) - break; +/* Scrub all of a quota type's items. */ +int +xfs_scrub_quota( + struct xfs_scrub_context *sc) +{ + struct xfs_scrub_quota_info sqi; + struct xfs_mount *mp = sc->mp; + struct xfs_quotainfo *qi = mp->m_quotainfo; + uint dqtype; + int error = 0; - xfs_scrub_quota_item(sc, dqtype, dq, id); + dqtype = xfs_scrub_quota_to_dqtype(sc); - id = be32_to_cpu(dq->q_core.d_id) + 1; - xfs_qm_dqput(dq); - if (!id) - break; - } + /* Look for problem extents. */ + error = xfs_scrub_quota_data_fork(sc); + if (error) + goto out; + if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT) + goto out; + + /* + * Check all the quota items. Now that we've checked the quota inode + * data fork we have to drop ILOCK_EXCL to use the regular dquot + * functions. + */ + xfs_iunlock(sc->ip, sc->ilock_flags); + sc->ilock_flags = 0; + sqi.sc = sc; + sqi.last_id = 0; + error = xfs_qm_dqiterate(mp, dqtype, xfs_scrub_quota_item, &sqi); + sc->ilock_flags = XFS_ILOCK_EXCL; + xfs_ilock(sc->ip, sc->ilock_flags); + if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, + sqi.last_id * qi->qi_dqperchunk, &error)) + goto out; out: - /* We set sc->ip earlier, so make sure we clear it now. */ - sc->ip = NULL; -out_unlock_quota: - mutex_unlock(&qi->qi_quotaofflock); return error; - -out_unlock_inode: - xfs_iunlock(ip, XFS_ILOCK_EXCL); - goto out; } diff --git a/fs/xfs/scrub/refcount.c b/fs/xfs/scrub/refcount.c index 400f1561cd3d..324a5f159145 100644 --- a/fs/xfs/scrub/refcount.c +++ b/fs/xfs/scrub/refcount.c @@ -150,7 +150,7 @@ xfs_scrub_refcountbt_rmap_check( * so we don't need insertion sort here. */ frag = kmem_alloc(sizeof(struct xfs_scrub_refcnt_frag), - KM_MAYFAIL | KM_NOFS); + KM_MAYFAIL); if (!frag) return -ENOMEM; memcpy(&frag->rm, rec, sizeof(frag->rm)); @@ -310,7 +310,7 @@ xfs_scrub_refcountbt_xref_rmap( struct xfs_scrub_refcnt_frag *n; int error; - if (!sc->sa.rmap_cur) + if (!sc->sa.rmap_cur || xfs_scrub_skip_xref(sc->sm)) return; /* Cross-reference with the rmapbt to confirm the refcount. */ @@ -404,7 +404,7 @@ xfs_scrub_refcount_xref_rmap( xfs_filblks_t blocks; int error; - if (!sc->sa.rmap_cur) + if (!sc->sa.rmap_cur || xfs_scrub_skip_xref(sc->sm)) return; /* Check that we saw as many refcbt blocks as the rmap knows about. */ @@ -460,7 +460,7 @@ xfs_scrub_xref_is_cow_staging( int has_refcount; int error; - if (!sc->sa.refc_cur) + if (!sc->sa.refc_cur || xfs_scrub_skip_xref(sc->sm)) return; /* Find the CoW staging extent. */ @@ -504,7 +504,7 @@ xfs_scrub_xref_is_not_shared( bool shared; int error; - if (!sc->sa.refc_cur) + if (!sc->sa.refc_cur || xfs_scrub_skip_xref(sc->sm)) return; error = xfs_refcount_has_record(sc->sa.refc_cur, agbno, len, &shared); diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c new file mode 100644 index 000000000000..e3e8fba1c99c --- /dev/null +++ b/fs/xfs/scrub/repair.c @@ -0,0 +1,1089 @@ +/* + * Copyright (C) 2018 Oracle. All Rights Reserved. + * + * Author: Darrick J. Wong <darrick.wong@oracle.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA. + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_mount.h" +#include "xfs_defer.h" +#include "xfs_btree.h" +#include "xfs_bit.h" +#include "xfs_log_format.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_inode.h" +#include "xfs_icache.h" +#include "xfs_alloc.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc.h" +#include "xfs_ialloc_btree.h" +#include "xfs_rmap.h" +#include "xfs_rmap_btree.h" +#include "xfs_refcount.h" +#include "xfs_refcount_btree.h" +#include "xfs_extent_busy.h" +#include "xfs_ag_resv.h" +#include "xfs_trans_space.h" +#include "xfs_quota.h" +#include "scrub/xfs_scrub.h" +#include "scrub/scrub.h" +#include "scrub/common.h" +#include "scrub/trace.h" +#include "scrub/repair.h" + +/* + * Attempt to repair some metadata, if the metadata is corrupt and userspace + * told us to fix it. This function returns -EAGAIN to mean "re-run scrub", + * and will set *fixed to true if it thinks it repaired anything. + */ +int +xfs_repair_attempt( + struct xfs_inode *ip, + struct xfs_scrub_context *sc, + bool *fixed) +{ + int error = 0; + + trace_xfs_repair_attempt(ip, sc->sm, error); + + xfs_scrub_ag_btcur_free(&sc->sa); + + /* Repair whatever's broken. */ + ASSERT(sc->ops->repair); + error = sc->ops->repair(sc); + trace_xfs_repair_done(ip, sc->sm, error); + switch (error) { + case 0: + /* + * Repair succeeded. Commit the fixes and perform a second + * scrub so that we can tell userspace if we fixed the problem. + */ + sc->sm->sm_flags &= ~XFS_SCRUB_FLAGS_OUT; + *fixed = true; + return -EAGAIN; + case -EDEADLOCK: + case -EAGAIN: + /* Tell the caller to try again having grabbed all the locks. */ + if (!sc->try_harder) { + sc->try_harder = true; + return -EAGAIN; + } + /* + * We tried harder but still couldn't grab all the resources + * we needed to fix it. The corruption has not been fixed, + * so report back to userspace. + */ + return -EFSCORRUPTED; + default: + return error; + } +} + +/* + * Complain about unfixable problems in the filesystem. We don't log + * corruptions when IFLAG_REPAIR wasn't set on the assumption that the driver + * program is xfs_scrub, which will call back with IFLAG_REPAIR set if the + * administrator isn't running xfs_scrub in no-repairs mode. + * + * Use this helper function because _ratelimited silently declares a static + * structure to track rate limiting information. + */ +void +xfs_repair_failure( + struct xfs_mount *mp) +{ + xfs_alert_ratelimited(mp, +"Corruption not fixed during online repair. Unmount and run xfs_repair."); +} + +/* + * Repair probe -- userspace uses this to probe if we're willing to repair a + * given mountpoint. + */ +int +xfs_repair_probe( + struct xfs_scrub_context *sc) +{ + int error = 0; + + if (xfs_scrub_should_terminate(sc, &error)) + return error; + + return 0; +} + +/* + * Roll a transaction, keeping the AG headers locked and reinitializing + * the btree cursors. + */ +int +xfs_repair_roll_ag_trans( + struct xfs_scrub_context *sc) +{ + int error; + + /* Keep the AG header buffers locked so we can keep going. */ + xfs_trans_bhold(sc->tp, sc->sa.agi_bp); + xfs_trans_bhold(sc->tp, sc->sa.agf_bp); + xfs_trans_bhold(sc->tp, sc->sa.agfl_bp); + + /* Roll the transaction. */ + error = xfs_trans_roll(&sc->tp); + if (error) + goto out_release; + + /* Join AG headers to the new transaction. */ + xfs_trans_bjoin(sc->tp, sc->sa.agi_bp); + xfs_trans_bjoin(sc->tp, sc->sa.agf_bp); + xfs_trans_bjoin(sc->tp, sc->sa.agfl_bp); + + return 0; + +out_release: + /* + * Rolling failed, so release the hold on the buffers. The + * buffers will be released during teardown on our way out + * of the kernel. + */ + xfs_trans_bhold_release(sc->tp, sc->sa.agi_bp); + xfs_trans_bhold_release(sc->tp, sc->sa.agf_bp); + xfs_trans_bhold_release(sc->tp, sc->sa.agfl_bp); + + return error; +} + +/* + * Does the given AG have enough space to rebuild a btree? Neither AG + * reservation can be critical, and we must have enough space (factoring + * in AG reservations) to construct a whole btree. + */ +bool +xfs_repair_ag_has_space( + struct xfs_perag *pag, + xfs_extlen_t nr_blocks, + enum xfs_ag_resv_type type) +{ + return !xfs_ag_resv_critical(pag, XFS_AG_RESV_RMAPBT) && + !xfs_ag_resv_critical(pag, XFS_AG_RESV_METADATA) && + pag->pagf_freeblks > xfs_ag_resv_needed(pag, type) + nr_blocks; +} + +/* + * Figure out how many blocks to reserve for an AG repair. We calculate the + * worst case estimate for the number of blocks we'd need to rebuild one of + * any type of per-AG btree. + */ +xfs_extlen_t +xfs_repair_calc_ag_resblks( + struct xfs_scrub_context *sc) +{ + struct xfs_mount *mp = sc->mp; + struct xfs_scrub_metadata *sm = sc->sm; + struct xfs_perag *pag; + struct xfs_buf *bp; + xfs_agino_t icount = 0; + xfs_extlen_t aglen = 0; + xfs_extlen_t usedlen; + xfs_extlen_t freelen; + xfs_extlen_t bnobt_sz; + xfs_extlen_t inobt_sz; + xfs_extlen_t rmapbt_sz; + xfs_extlen_t refcbt_sz; + int error; + + if (!(sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR)) + return 0; + + /* Use in-core counters if possible. */ + pag = xfs_perag_get(mp, sm->sm_agno); + if (pag->pagi_init) + icount = pag->pagi_count; + + /* + * Otherwise try to get the actual counters from disk; if not, make + * some worst case assumptions. + */ + if (icount == 0) { + error = xfs_ialloc_read_agi(mp, NULL, sm->sm_agno, &bp); + if (error) { + icount = mp->m_sb.sb_agblocks / mp->m_sb.sb_inopblock; + } else { + icount = pag->pagi_count; + xfs_buf_relse(bp); + } + } + + /* Now grab the block counters from the AGF. */ + error = xfs_alloc_read_agf(mp, NULL, sm->sm_agno, 0, &bp); + if (error) { + aglen = mp->m_sb.sb_agblocks; + freelen = aglen; + usedlen = aglen; + } else { + aglen = be32_to_cpu(XFS_BUF_TO_AGF(bp)->agf_length); + freelen = pag->pagf_freeblks; + usedlen = aglen - freelen; + xfs_buf_relse(bp); + } + xfs_perag_put(pag); + + trace_xfs_repair_calc_ag_resblks(mp, sm->sm_agno, icount, aglen, + freelen, usedlen); + + /* + * Figure out how many blocks we'd need worst case to rebuild + * each type of btree. Note that we can only rebuild the + * bnobt/cntbt or inobt/finobt as pairs. + */ + bnobt_sz = 2 * xfs_allocbt_calc_size(mp, freelen); + if (xfs_sb_version_hassparseinodes(&mp->m_sb)) + inobt_sz = xfs_iallocbt_calc_size(mp, icount / + XFS_INODES_PER_HOLEMASK_BIT); + else + inobt_sz = xfs_iallocbt_calc_size(mp, icount / + XFS_INODES_PER_CHUNK); + if (xfs_sb_version_hasfinobt(&mp->m_sb)) + inobt_sz *= 2; + if (xfs_sb_version_hasreflink(&mp->m_sb)) + refcbt_sz = xfs_refcountbt_calc_size(mp, usedlen); + else + refcbt_sz = 0; + if (xfs_sb_version_hasrmapbt(&mp->m_sb)) { + /* + * Guess how many blocks we need to rebuild the rmapbt. + * For non-reflink filesystems we can't have more records than + * used blocks. However, with reflink it's possible to have + * more than one rmap record per AG block. We don't know how + * many rmaps there could be in the AG, so we start off with + * what we hope is an generous over-estimation. + */ + if (xfs_sb_version_hasreflink(&mp->m_sb)) + rmapbt_sz = xfs_rmapbt_calc_size(mp, + (unsigned long long)aglen * 2); + else + rmapbt_sz = xfs_rmapbt_calc_size(mp, usedlen); + } else { + rmapbt_sz = 0; + } + + trace_xfs_repair_calc_ag_resblks_btsize(mp, sm->sm_agno, bnobt_sz, + inobt_sz, rmapbt_sz, refcbt_sz); + + return max(max(bnobt_sz, inobt_sz), max(rmapbt_sz, refcbt_sz)); +} + +/* Allocate a block in an AG. */ +int +xfs_repair_alloc_ag_block( + struct xfs_scrub_context *sc, + struct xfs_owner_info *oinfo, + xfs_fsblock_t *fsbno, + enum xfs_ag_resv_type resv) +{ + struct xfs_alloc_arg args = {0}; + xfs_agblock_t bno; + int error; + + switch (resv) { + case XFS_AG_RESV_AGFL: + case XFS_AG_RESV_RMAPBT: + error = xfs_alloc_get_freelist(sc->tp, sc->sa.agf_bp, &bno, 1); + if (error) + return error; + if (bno == NULLAGBLOCK) + return -ENOSPC; + xfs_extent_busy_reuse(sc->mp, sc->sa.agno, bno, + 1, false); + *fsbno = XFS_AGB_TO_FSB(sc->mp, sc->sa.agno, bno); + if (resv == XFS_AG_RESV_RMAPBT) + xfs_ag_resv_rmapbt_alloc(sc->mp, sc->sa.agno); + return 0; + default: + break; + } + + args.tp = sc->tp; + args.mp = sc->mp; + args.oinfo = *oinfo; + args.fsbno = XFS_AGB_TO_FSB(args.mp, sc->sa.agno, 0); + args.minlen = 1; + args.maxlen = 1; + args.prod = 1; + args.type = XFS_ALLOCTYPE_THIS_AG; + args.resv = resv; + + error = xfs_alloc_vextent(&args); + if (error) + return error; + if (args.fsbno == NULLFSBLOCK) + return -ENOSPC; + ASSERT(args.len == 1); + *fsbno = args.fsbno; + + return 0; +} + +/* Initialize a new AG btree root block with zero entries. */ +int +xfs_repair_init_btblock( + struct xfs_scrub_context *sc, + xfs_fsblock_t fsb, + struct xfs_buf **bpp, + xfs_btnum_t btnum, + const struct xfs_buf_ops *ops) +{ + struct xfs_trans *tp = sc->tp; + struct xfs_mount *mp = sc->mp; + struct xfs_buf *bp; + + trace_xfs_repair_init_btblock(mp, XFS_FSB_TO_AGNO(mp, fsb), + XFS_FSB_TO_AGBNO(mp, fsb), btnum); + + ASSERT(XFS_FSB_TO_AGNO(mp, fsb) == sc->sa.agno); + bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, XFS_FSB_TO_DADDR(mp, fsb), + XFS_FSB_TO_BB(mp, 1), 0); + xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); + xfs_btree_init_block(mp, bp, btnum, 0, 0, sc->sa.agno, 0); + xfs_trans_buf_set_type(tp, bp, XFS_BLFT_BTREE_BUF); + xfs_trans_log_buf(tp, bp, 0, bp->b_length); + bp->b_ops = ops; + *bpp = bp; + + return 0; +} + +/* + * Reconstructing per-AG Btrees + * + * When a space btree is corrupt, we don't bother trying to fix it. Instead, + * we scan secondary space metadata to derive the records that should be in + * the damaged btree, initialize a fresh btree root, and insert the records. + * Note that for rebuilding the rmapbt we scan all the primary data to + * generate the new records. + * + * However, that leaves the matter of removing all the metadata describing the + * old broken structure. For primary metadata we use the rmap data to collect + * every extent with a matching rmap owner (exlist); we then iterate all other + * metadata structures with the same rmap owner to collect the extents that + * cannot be removed (sublist). We then subtract sublist from exlist to + * derive the blocks that were used by the old btree. These blocks can be + * reaped. + * + * For rmapbt reconstructions we must use different tactics for extent + * collection. First we iterate all primary metadata (this excludes the old + * rmapbt, obviously) to generate new rmap records. The gaps in the rmap + * records are collected as exlist. The bnobt records are collected as + * sublist. As with the other btrees we subtract sublist from exlist, and the + * result (since the rmapbt lives in the free space) are the blocks from the + * old rmapbt. + */ + +/* Collect a dead btree extent for later disposal. */ +int +xfs_repair_collect_btree_extent( + struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *exlist, + xfs_fsblock_t fsbno, + xfs_extlen_t len) +{ + struct xfs_repair_extent *rex; + + trace_xfs_repair_collect_btree_extent(sc->mp, + XFS_FSB_TO_AGNO(sc->mp, fsbno), + XFS_FSB_TO_AGBNO(sc->mp, fsbno), len); + + rex = kmem_alloc(sizeof(struct xfs_repair_extent), KM_MAYFAIL); + if (!rex) + return -ENOMEM; + + INIT_LIST_HEAD(&rex->list); + rex->fsbno = fsbno; + rex->len = len; + list_add_tail(&rex->list, &exlist->list); + + return 0; +} + +/* + * An error happened during the rebuild so the transaction will be cancelled. + * The fs will shut down, and the administrator has to unmount and run repair. + * Therefore, free all the memory associated with the list so we can die. + */ +void +xfs_repair_cancel_btree_extents( + struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *exlist) +{ + struct xfs_repair_extent *rex; + struct xfs_repair_extent *n; + + for_each_xfs_repair_extent_safe(rex, n, exlist) { + list_del(&rex->list); + kmem_free(rex); + } +} + +/* Compare two btree extents. */ +static int +xfs_repair_btree_extent_cmp( + void *priv, + struct list_head *a, + struct list_head *b) +{ + struct xfs_repair_extent *ap; + struct xfs_repair_extent *bp; + + ap = container_of(a, struct xfs_repair_extent, list); + bp = container_of(b, struct xfs_repair_extent, list); + + if (ap->fsbno > bp->fsbno) + return 1; + if (ap->fsbno < bp->fsbno) + return -1; + return 0; +} + +/* + * Remove all the blocks mentioned in @sublist from the extents in @exlist. + * + * The intent is that callers will iterate the rmapbt for all of its records + * for a given owner to generate @exlist; and iterate all the blocks of the + * metadata structures that are not being rebuilt and have the same rmapbt + * owner to generate @sublist. This routine subtracts all the extents + * mentioned in sublist from all the extents linked in @exlist, which leaves + * @exlist as the list of blocks that are not accounted for, which we assume + * are the dead blocks of the old metadata structure. The blocks mentioned in + * @exlist can be reaped. + */ +#define LEFT_ALIGNED (1 << 0) +#define RIGHT_ALIGNED (1 << 1) +int +xfs_repair_subtract_extents( + struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *exlist, + struct xfs_repair_extent_list *sublist) +{ + struct list_head *lp; + struct xfs_repair_extent *ex; + struct xfs_repair_extent *newex; + struct xfs_repair_extent *subex; + xfs_fsblock_t sub_fsb; + xfs_extlen_t sub_len; + int state; + int error = 0; + + if (list_empty(&exlist->list) || list_empty(&sublist->list)) + return 0; + ASSERT(!list_empty(&sublist->list)); + + list_sort(NULL, &exlist->list, xfs_repair_btree_extent_cmp); + list_sort(NULL, &sublist->list, xfs_repair_btree_extent_cmp); + + /* + * Now that we've sorted both lists, we iterate exlist once, rolling + * forward through sublist and/or exlist as necessary until we find an + * overlap or reach the end of either list. We do not reset lp to the + * head of exlist nor do we reset subex to the head of sublist. The + * list traversal is similar to merge sort, but we're deleting + * instead. In this manner we avoid O(n^2) operations. + */ + subex = list_first_entry(&sublist->list, struct xfs_repair_extent, + list); + lp = exlist->list.next; + while (lp != &exlist->list) { + ex = list_entry(lp, struct xfs_repair_extent, list); + + /* + * Advance subex and/or ex until we find a pair that + * intersect or we run out of extents. + */ + while (subex->fsbno + subex->len <= ex->fsbno) { + if (list_is_last(&subex->list, &sublist->list)) + goto out; + subex = list_next_entry(subex, list); + } + if (subex->fsbno >= ex->fsbno + ex->len) { + lp = lp->next; + continue; + } + + /* trim subex to fit the extent we have */ + sub_fsb = subex->fsbno; + sub_len = subex->len; + if (subex->fsbno < ex->fsbno) { + sub_len -= ex->fsbno - subex->fsbno; + sub_fsb = ex->fsbno; + } + if (sub_len > ex->len) + sub_len = ex->len; + + state = 0; + if (sub_fsb == ex->fsbno) + state |= LEFT_ALIGNED; + if (sub_fsb + sub_len == ex->fsbno + ex->len) + state |= RIGHT_ALIGNED; + switch (state) { + case LEFT_ALIGNED: + /* Coincides with only the left. */ + ex->fsbno += sub_len; + ex->len -= sub_len; + break; + case RIGHT_ALIGNED: + /* Coincides with only the right. */ + ex->len -= sub_len; + lp = lp->next; + break; + case LEFT_ALIGNED | RIGHT_ALIGNED: + /* Total overlap, just delete ex. */ + lp = lp->next; + list_del(&ex->list); + kmem_free(ex); + break; + case 0: + /* + * Deleting from the middle: add the new right extent + * and then shrink the left extent. + */ + newex = kmem_alloc(sizeof(struct xfs_repair_extent), + KM_MAYFAIL); + if (!newex) { + error = -ENOMEM; + goto out; + } + INIT_LIST_HEAD(&newex->list); + newex->fsbno = sub_fsb + sub_len; + newex->len = ex->fsbno + ex->len - newex->fsbno; + list_add(&newex->list, &ex->list); + ex->len = sub_fsb - ex->fsbno; + lp = lp->next; + break; + default: + ASSERT(0); + break; + } + } + +out: + return error; +} +#undef LEFT_ALIGNED +#undef RIGHT_ALIGNED + +/* + * Disposal of Blocks from Old per-AG Btrees + * + * Now that we've constructed a new btree to replace the damaged one, we want + * to dispose of the blocks that (we think) the old btree was using. + * Previously, we used the rmapbt to collect the extents (exlist) with the + * rmap owner corresponding to the tree we rebuilt, collected extents for any + * blocks with the same rmap owner that are owned by another data structure + * (sublist), and subtracted sublist from exlist. In theory the extents + * remaining in exlist are the old btree's blocks. + * + * Unfortunately, it's possible that the btree was crosslinked with other + * blocks on disk. The rmap data can tell us if there are multiple owners, so + * if the rmapbt says there is an owner of this block other than @oinfo, then + * the block is crosslinked. Remove the reverse mapping and continue. + * + * If there is one rmap record, we can free the block, which removes the + * reverse mapping but doesn't add the block to the free space. Our repair + * strategy is to hope the other metadata objects crosslinked on this block + * will be rebuilt (atop different blocks), thereby removing all the cross + * links. + * + * If there are no rmap records at all, we also free the block. If the btree + * being rebuilt lives in the free space (bnobt/cntbt/rmapbt) then there isn't + * supposed to be a rmap record and everything is ok. For other btrees there + * had to have been an rmap entry for the block to have ended up on @exlist, + * so if it's gone now there's something wrong and the fs will shut down. + * + * Note: If there are multiple rmap records with only the same rmap owner as + * the btree we're trying to rebuild and the block is indeed owned by another + * data structure with the same rmap owner, then the block will be in sublist + * and therefore doesn't need disposal. If there are multiple rmap records + * with only the same rmap owner but the block is not owned by something with + * the same rmap owner, the block will be freed. + * + * The caller is responsible for locking the AG headers for the entire rebuild + * operation so that nothing else can sneak in and change the AG state while + * we're not looking. We also assume that the caller already invalidated any + * buffers associated with @exlist. + */ + +/* + * Invalidate buffers for per-AG btree blocks we're dumping. This function + * is not intended for use with file data repairs; we have bunmapi for that. + */ +int +xfs_repair_invalidate_blocks( + struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *exlist) +{ + struct xfs_repair_extent *rex; + struct xfs_repair_extent *n; + struct xfs_buf *bp; + xfs_fsblock_t fsbno; + xfs_agblock_t i; + + /* + * For each block in each extent, see if there's an incore buffer for + * exactly that block; if so, invalidate it. The buffer cache only + * lets us look for one buffer at a time, so we have to look one block + * at a time. Avoid invalidating AG headers and post-EOFS blocks + * because we never own those; and if we can't TRYLOCK the buffer we + * assume it's owned by someone else. + */ + for_each_xfs_repair_extent_safe(rex, n, exlist) { + for (fsbno = rex->fsbno, i = rex->len; i > 0; fsbno++, i--) { + /* Skip AG headers and post-EOFS blocks */ + if (!xfs_verify_fsbno(sc->mp, fsbno)) + continue; + bp = xfs_buf_incore(sc->mp->m_ddev_targp, + XFS_FSB_TO_DADDR(sc->mp, fsbno), + XFS_FSB_TO_BB(sc->mp, 1), XBF_TRYLOCK); + if (bp) { + xfs_trans_bjoin(sc->tp, bp); + xfs_trans_binval(sc->tp, bp); + } + } + } + + return 0; +} + +/* Ensure the freelist is the correct size. */ +int +xfs_repair_fix_freelist( + struct xfs_scrub_context *sc, + bool can_shrink) +{ + struct xfs_alloc_arg args = {0}; + + args.mp = sc->mp; + args.tp = sc->tp; + args.agno = sc->sa.agno; + args.alignment = 1; + args.pag = sc->sa.pag; + + return xfs_alloc_fix_freelist(&args, + can_shrink ? 0 : XFS_ALLOC_FLAG_NOSHRINK); +} + +/* + * Put a block back on the AGFL. + */ +STATIC int +xfs_repair_put_freelist( + struct xfs_scrub_context *sc, + xfs_agblock_t agbno) +{ + struct xfs_owner_info oinfo; + int error; + + /* Make sure there's space on the freelist. */ + error = xfs_repair_fix_freelist(sc, true); + if (error) + return error; + + /* + * Since we're "freeing" a lost block onto the AGFL, we have to + * create an rmap for the block prior to merging it or else other + * parts will break. + */ + xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_AG); + error = xfs_rmap_alloc(sc->tp, sc->sa.agf_bp, sc->sa.agno, agbno, 1, + &oinfo); + if (error) + return error; + + /* Put the block on the AGFL. */ + error = xfs_alloc_put_freelist(sc->tp, sc->sa.agf_bp, sc->sa.agfl_bp, + agbno, 0); + if (error) + return error; + xfs_extent_busy_insert(sc->tp, sc->sa.agno, agbno, 1, + XFS_EXTENT_BUSY_SKIP_DISCARD); + + return 0; +} + +/* Dispose of a single metadata block. */ +STATIC int +xfs_repair_dispose_btree_block( + struct xfs_scrub_context *sc, + xfs_fsblock_t fsbno, + struct xfs_owner_info *oinfo, + enum xfs_ag_resv_type resv) +{ + struct xfs_btree_cur *cur; + struct xfs_buf *agf_bp = NULL; + xfs_agnumber_t agno; + xfs_agblock_t agbno; + bool has_other_rmap; + int error; + + agno = XFS_FSB_TO_AGNO(sc->mp, fsbno); + agbno = XFS_FSB_TO_AGBNO(sc->mp, fsbno); + + /* + * If we are repairing per-inode metadata, we need to read in the AGF + * buffer. Otherwise, we're repairing a per-AG structure, so reuse + * the AGF buffer that the setup functions already grabbed. + */ + if (sc->ip) { + error = xfs_alloc_read_agf(sc->mp, sc->tp, agno, 0, &agf_bp); + if (error) + return error; + if (!agf_bp) + return -ENOMEM; + } else { + agf_bp = sc->sa.agf_bp; + } + cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno); + + /* Can we find any other rmappings? */ + error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap); + if (error) + goto out_cur; + xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); + + /* + * If there are other rmappings, this block is cross linked and must + * not be freed. Remove the reverse mapping and move on. Otherwise, + * we were the only owner of the block, so free the extent, which will + * also remove the rmap. + * + * XXX: XFS doesn't support detecting the case where a single block + * metadata structure is crosslinked with a multi-block structure + * because the buffer cache doesn't detect aliasing problems, so we + * can't fix 100% of crosslinking problems (yet). The verifiers will + * blow on writeout, the filesystem will shut down, and the admin gets + * to run xfs_repair. + */ + if (has_other_rmap) + error = xfs_rmap_free(sc->tp, agf_bp, agno, agbno, 1, oinfo); + else if (resv == XFS_AG_RESV_AGFL) + error = xfs_repair_put_freelist(sc, agbno); + else + error = xfs_free_extent(sc->tp, fsbno, 1, oinfo, resv); + if (agf_bp != sc->sa.agf_bp) + xfs_trans_brelse(sc->tp, agf_bp); + if (error) + return error; + + if (sc->ip) + return xfs_trans_roll_inode(&sc->tp, sc->ip); + return xfs_repair_roll_ag_trans(sc); + +out_cur: + xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); + if (agf_bp != sc->sa.agf_bp) + xfs_trans_brelse(sc->tp, agf_bp); + return error; +} + +/* Dispose of btree blocks from an old per-AG btree. */ +int +xfs_repair_reap_btree_extents( + struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *exlist, + struct xfs_owner_info *oinfo, + enum xfs_ag_resv_type type) +{ + struct xfs_repair_extent *rex; + struct xfs_repair_extent *n; + int error = 0; + + ASSERT(xfs_sb_version_hasrmapbt(&sc->mp->m_sb)); + + /* Dispose of every block from the old btree. */ + for_each_xfs_repair_extent_safe(rex, n, exlist) { + ASSERT(sc->ip != NULL || + XFS_FSB_TO_AGNO(sc->mp, rex->fsbno) == sc->sa.agno); + + trace_xfs_repair_dispose_btree_extent(sc->mp, + XFS_FSB_TO_AGNO(sc->mp, rex->fsbno), + XFS_FSB_TO_AGBNO(sc->mp, rex->fsbno), rex->len); + + for (; rex->len > 0; rex->len--, rex->fsbno++) { + error = xfs_repair_dispose_btree_block(sc, rex->fsbno, + oinfo, type); + if (error) + goto out; + } + list_del(&rex->list); + kmem_free(rex); + } + +out: + xfs_repair_cancel_btree_extents(sc, exlist); + return error; +} + +/* + * Finding per-AG Btree Roots for AGF/AGI Reconstruction + * + * If the AGF or AGI become slightly corrupted, it may be necessary to rebuild + * the AG headers by using the rmap data to rummage through the AG looking for + * btree roots. This is not guaranteed to work if the AG is heavily damaged + * or the rmap data are corrupt. + * + * Callers of xfs_repair_find_ag_btree_roots must lock the AGF and AGFL + * buffers if the AGF is being rebuilt; or the AGF and AGI buffers if the + * AGI is being rebuilt. It must maintain these locks until it's safe for + * other threads to change the btrees' shapes. The caller provides + * information about the btrees to look for by passing in an array of + * xfs_repair_find_ag_btree with the (rmap owner, buf_ops, magic) fields set. + * The (root, height) fields will be set on return if anything is found. The + * last element of the array should have a NULL buf_ops to mark the end of the + * array. + * + * For every rmapbt record matching any of the rmap owners in btree_info, + * read each block referenced by the rmap record. If the block is a btree + * block from this filesystem matching any of the magic numbers and has a + * level higher than what we've already seen, remember the block and the + * height of the tree required to have such a block. When the call completes, + * we return the highest block we've found for each btree description; those + * should be the roots. + */ + +struct xfs_repair_findroot { + struct xfs_scrub_context *sc; + struct xfs_buf *agfl_bp; + struct xfs_agf *agf; + struct xfs_repair_find_ag_btree *btree_info; +}; + +/* See if our block is in the AGFL. */ +STATIC int +xfs_repair_findroot_agfl_walk( + struct xfs_mount *mp, + xfs_agblock_t bno, + void *priv) +{ + xfs_agblock_t *agbno = priv; + + return (*agbno == bno) ? XFS_BTREE_QUERY_RANGE_ABORT : 0; +} + +/* Does this block match the btree information passed in? */ +STATIC int +xfs_repair_findroot_block( + struct xfs_repair_findroot *ri, + struct xfs_repair_find_ag_btree *fab, + uint64_t owner, + xfs_agblock_t agbno, + bool *found_it) +{ + struct xfs_mount *mp = ri->sc->mp; + struct xfs_buf *bp; + struct xfs_btree_block *btblock; + xfs_daddr_t daddr; + int error; + + daddr = XFS_AGB_TO_DADDR(mp, ri->sc->sa.agno, agbno); + + /* + * Blocks in the AGFL have stale contents that might just happen to + * have a matching magic and uuid. We don't want to pull these blocks + * in as part of a tree root, so we have to filter out the AGFL stuff + * here. If the AGFL looks insane we'll just refuse to repair. + */ + if (owner == XFS_RMAP_OWN_AG) { + error = xfs_agfl_walk(mp, ri->agf, ri->agfl_bp, + xfs_repair_findroot_agfl_walk, &agbno); + if (error == XFS_BTREE_QUERY_RANGE_ABORT) + return 0; + if (error) + return error; + } + + error = xfs_trans_read_buf(mp, ri->sc->tp, mp->m_ddev_targp, daddr, + mp->m_bsize, 0, &bp, NULL); + if (error) + return error; + + /* + * Does this look like a block matching our fs and higher than any + * other block we've found so far? If so, reattach buffer verifiers + * so the AIL won't complain if the buffer is also dirty. + */ + btblock = XFS_BUF_TO_BLOCK(bp); + if (be32_to_cpu(btblock->bb_magic) != fab->magic) + goto out; + if (xfs_sb_version_hascrc(&mp->m_sb) && + !uuid_equal(&btblock->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid)) + goto out; + bp->b_ops = fab->buf_ops; + + /* Ignore this block if it's lower in the tree than we've seen. */ + if (fab->root != NULLAGBLOCK && + xfs_btree_get_level(btblock) < fab->height) + goto out; + + /* Make sure we pass the verifiers. */ + bp->b_ops->verify_read(bp); + if (bp->b_error) + goto out; + fab->root = agbno; + fab->height = xfs_btree_get_level(btblock) + 1; + *found_it = true; + + trace_xfs_repair_findroot_block(mp, ri->sc->sa.agno, agbno, + be32_to_cpu(btblock->bb_magic), fab->height - 1); +out: + xfs_trans_brelse(ri->sc->tp, bp); + return error; +} + +/* + * Do any of the blocks in this rmap record match one of the btrees we're + * looking for? + */ +STATIC int +xfs_repair_findroot_rmap( + struct xfs_btree_cur *cur, + struct xfs_rmap_irec *rec, + void *priv) +{ + struct xfs_repair_findroot *ri = priv; + struct xfs_repair_find_ag_btree *fab; + xfs_agblock_t b; + bool found_it; + int error = 0; + + /* Ignore anything that isn't AG metadata. */ + if (!XFS_RMAP_NON_INODE_OWNER(rec->rm_owner)) + return 0; + + /* Otherwise scan each block + btree type. */ + for (b = 0; b < rec->rm_blockcount; b++) { + found_it = false; + for (fab = ri->btree_info; fab->buf_ops; fab++) { + if (rec->rm_owner != fab->rmap_owner) + continue; + error = xfs_repair_findroot_block(ri, fab, + rec->rm_owner, rec->rm_startblock + b, + &found_it); + if (error) + return error; + if (found_it) + break; + } + } + + return 0; +} + +/* Find the roots of the per-AG btrees described in btree_info. */ +int +xfs_repair_find_ag_btree_roots( + struct xfs_scrub_context *sc, + struct xfs_buf *agf_bp, + struct xfs_repair_find_ag_btree *btree_info, + struct xfs_buf *agfl_bp) +{ + struct xfs_mount *mp = sc->mp; + struct xfs_repair_findroot ri; + struct xfs_repair_find_ag_btree *fab; + struct xfs_btree_cur *cur; + int error; + + ASSERT(xfs_buf_islocked(agf_bp)); + ASSERT(agfl_bp == NULL || xfs_buf_islocked(agfl_bp)); + + ri.sc = sc; + ri.btree_info = btree_info; + ri.agf = XFS_BUF_TO_AGF(agf_bp); + ri.agfl_bp = agfl_bp; + for (fab = btree_info; fab->buf_ops; fab++) { + ASSERT(agfl_bp || fab->rmap_owner != XFS_RMAP_OWN_AG); + ASSERT(XFS_RMAP_NON_INODE_OWNER(fab->rmap_owner)); + fab->root = NULLAGBLOCK; + fab->height = 0; + } + + cur = xfs_rmapbt_init_cursor(mp, sc->tp, agf_bp, sc->sa.agno); + error = xfs_rmap_query_all(cur, xfs_repair_findroot_rmap, &ri); + xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); + + return error; +} + +/* Force a quotacheck the next time we mount. */ +void +xfs_repair_force_quotacheck( + struct xfs_scrub_context *sc, + uint dqtype) +{ + uint flag; + + flag = xfs_quota_chkd_flag(dqtype); + if (!(flag & sc->mp->m_qflags)) + return; + + sc->mp->m_qflags &= ~flag; + spin_lock(&sc->mp->m_sb_lock); + sc->mp->m_sb.sb_qflags &= ~flag; + spin_unlock(&sc->mp->m_sb_lock); + xfs_log_sb(sc->tp); +} + +/* + * Attach dquots to this inode, or schedule quotacheck to fix them. + * + * This function ensures that the appropriate dquots are attached to an inode. + * We cannot allow the dquot code to allocate an on-disk dquot block here + * because we're already in transaction context with the inode locked. The + * on-disk dquot should already exist anyway. If the quota code signals + * corruption or missing quota information, schedule quotacheck, which will + * repair corruptions in the quota metadata. + */ +int +xfs_repair_ino_dqattach( + struct xfs_scrub_context *sc) +{ + int error; + + error = xfs_qm_dqattach_locked(sc->ip, false); + switch (error) { + case -EFSBADCRC: + case -EFSCORRUPTED: + case -ENOENT: + xfs_err_ratelimited(sc->mp, +"inode %llu repair encountered quota error %d, quotacheck forced.", + (unsigned long long)sc->ip->i_ino, error); + if (XFS_IS_UQUOTA_ON(sc->mp) && !sc->ip->i_udquot) + xfs_repair_force_quotacheck(sc, XFS_DQ_USER); + if (XFS_IS_GQUOTA_ON(sc->mp) && !sc->ip->i_gdquot) + xfs_repair_force_quotacheck(sc, XFS_DQ_GROUP); + if (XFS_IS_PQUOTA_ON(sc->mp) && !sc->ip->i_pdquot) + xfs_repair_force_quotacheck(sc, XFS_DQ_PROJ); + /* fall through */ + case -ESRCH: + error = 0; + break; + default: + break; + } + + return error; +} diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h new file mode 100644 index 000000000000..f2b0895294db --- /dev/null +++ b/fs/xfs/scrub/repair.h @@ -0,0 +1,132 @@ +/* + * Copyright (C) 2018 Oracle. All Rights Reserved. + * + * Author: Darrick J. Wong <darrick.wong@oracle.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA. + */ +#ifndef __XFS_SCRUB_REPAIR_H__ +#define __XFS_SCRUB_REPAIR_H__ + +static inline int xfs_repair_notsupported(struct xfs_scrub_context *sc) +{ + return -EOPNOTSUPP; +} + +#ifdef CONFIG_XFS_ONLINE_REPAIR + +/* Repair helpers */ + +int xfs_repair_attempt(struct xfs_inode *ip, struct xfs_scrub_context *sc, + bool *fixed); +void xfs_repair_failure(struct xfs_mount *mp); +int xfs_repair_roll_ag_trans(struct xfs_scrub_context *sc); +bool xfs_repair_ag_has_space(struct xfs_perag *pag, xfs_extlen_t nr_blocks, + enum xfs_ag_resv_type type); +xfs_extlen_t xfs_repair_calc_ag_resblks(struct xfs_scrub_context *sc); +int xfs_repair_alloc_ag_block(struct xfs_scrub_context *sc, + struct xfs_owner_info *oinfo, xfs_fsblock_t *fsbno, + enum xfs_ag_resv_type resv); +int xfs_repair_init_btblock(struct xfs_scrub_context *sc, xfs_fsblock_t fsb, + struct xfs_buf **bpp, xfs_btnum_t btnum, + const struct xfs_buf_ops *ops); + +struct xfs_repair_extent { + struct list_head list; + xfs_fsblock_t fsbno; + xfs_extlen_t len; +}; + +struct xfs_repair_extent_list { + struct list_head list; +}; + +static inline void +xfs_repair_init_extent_list( + struct xfs_repair_extent_list *exlist) +{ + INIT_LIST_HEAD(&exlist->list); +} + +#define for_each_xfs_repair_extent_safe(rbe, n, exlist) \ + list_for_each_entry_safe((rbe), (n), &(exlist)->list, list) +int xfs_repair_collect_btree_extent(struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *btlist, xfs_fsblock_t fsbno, + xfs_extlen_t len); +void xfs_repair_cancel_btree_extents(struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *btlist); +int xfs_repair_subtract_extents(struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *exlist, + struct xfs_repair_extent_list *sublist); +int xfs_repair_fix_freelist(struct xfs_scrub_context *sc, bool can_shrink); +int xfs_repair_invalidate_blocks(struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *btlist); +int xfs_repair_reap_btree_extents(struct xfs_scrub_context *sc, + struct xfs_repair_extent_list *exlist, + struct xfs_owner_info *oinfo, enum xfs_ag_resv_type type); + +struct xfs_repair_find_ag_btree { + /* in: rmap owner of the btree we're looking for */ + uint64_t rmap_owner; + + /* in: buffer ops */ + const struct xfs_buf_ops *buf_ops; + + /* in: magic number of the btree */ + uint32_t magic; + + /* out: the highest btree block found and the tree height */ + xfs_agblock_t root; + unsigned int height; +}; + +int xfs_repair_find_ag_btree_roots(struct xfs_scrub_context *sc, + struct xfs_buf *agf_bp, + struct xfs_repair_find_ag_btree *btree_info, + struct xfs_buf *agfl_bp); +void xfs_repair_force_quotacheck(struct xfs_scrub_context *sc, uint dqtype); +int xfs_repair_ino_dqattach(struct xfs_scrub_context *sc); + +/* Metadata repairers */ + +int xfs_repair_probe(struct xfs_scrub_context *sc); +int xfs_repair_superblock(struct xfs_scrub_context *sc); + +#else + +static inline int xfs_repair_attempt( + struct xfs_inode *ip, + struct xfs_scrub_context *sc, + bool *fixed) +{ + return -EOPNOTSUPP; +} + +static inline void xfs_repair_failure(struct xfs_mount *mp) {} + +static inline xfs_extlen_t +xfs_repair_calc_ag_resblks( + struct xfs_scrub_context *sc) +{ + ASSERT(!(sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR)); + return 0; +} + +#define xfs_repair_probe xfs_repair_notsupported +#define xfs_repair_superblock xfs_repair_notsupported + +#endif /* CONFIG_XFS_ONLINE_REPAIR */ + +#endif /* __XFS_SCRUB_REPAIR_H__ */ diff --git a/fs/xfs/scrub/rmap.c b/fs/xfs/scrub/rmap.c index 8f2a7c3ff455..b376a9a77c04 100644 --- a/fs/xfs/scrub/rmap.c +++ b/fs/xfs/scrub/rmap.c @@ -66,7 +66,7 @@ xfs_scrub_rmapbt_xref_refc( bool is_unwritten; int error; - if (!sc->sa.refc_cur) + if (!sc->sa.refc_cur || xfs_scrub_skip_xref(sc->sm)) return; non_inode = XFS_RMAP_NON_INODE_OWNER(irec->rm_owner); @@ -207,7 +207,7 @@ xfs_scrub_xref_check_owner( bool has_rmap; int error; - if (!sc->sa.rmap_cur) + if (!sc->sa.rmap_cur || xfs_scrub_skip_xref(sc->sm)) return; error = xfs_rmap_record_exists(sc->sa.rmap_cur, bno, len, oinfo, @@ -250,7 +250,7 @@ xfs_scrub_xref_has_no_owner( bool has_rmap; int error; - if (!sc->sa.rmap_cur) + if (!sc->sa.rmap_cur || xfs_scrub_skip_xref(sc->sm)) return; error = xfs_rmap_has_record(sc->sa.rmap_cur, bno, len, &has_rmap); diff --git a/fs/xfs/scrub/rtbitmap.c b/fs/xfs/scrub/rtbitmap.c index 39c41dfe08ee..40f462a11ea5 100644 --- a/fs/xfs/scrub/rtbitmap.c +++ b/fs/xfs/scrub/rtbitmap.c @@ -66,11 +66,15 @@ xfs_scrub_rtbitmap_rec( void *priv) { struct xfs_scrub_context *sc = priv; + xfs_rtblock_t startblock; + xfs_rtblock_t blockcount; - if (rec->ar_startblock + rec->ar_blockcount <= rec->ar_startblock || - !xfs_verify_rtbno(sc->mp, rec->ar_startblock) || - !xfs_verify_rtbno(sc->mp, rec->ar_startblock + - rec->ar_blockcount - 1)) + startblock = rec->ar_startext * tp->t_mountp->m_sb.sb_rextsize; + blockcount = rec->ar_extcount * tp->t_mountp->m_sb.sb_rextsize; + + if (startblock + blockcount <= startblock || + !xfs_verify_rtbno(sc->mp, startblock) || + !xfs_verify_rtbno(sc->mp, startblock + blockcount - 1)) xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); return 0; } @@ -82,6 +86,11 @@ xfs_scrub_rtbitmap( { int error; + /* Invoke the fork scrubber. */ + error = xfs_scrub_metadata_inode_forks(sc); + if (error || (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) + return error; + error = xfs_rtalloc_query_all(sc->tp, xfs_scrub_rtbitmap_rec, sc); if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, 0, &error)) goto out; @@ -95,8 +104,35 @@ int xfs_scrub_rtsummary( struct xfs_scrub_context *sc) { + struct xfs_inode *rsumip = sc->mp->m_rsumip; + struct xfs_inode *old_ip = sc->ip; + uint old_ilock_flags = sc->ilock_flags; + int error = 0; + + /* + * We ILOCK'd the rt bitmap ip in the setup routine, now lock the + * rt summary ip in compliance with the rt inode locking rules. + * + * Since we switch sc->ip to rsumip we have to save the old ilock + * flags so that we don't mix up the inode state that @sc tracks. + */ + sc->ip = rsumip; + sc->ilock_flags = XFS_ILOCK_EXCL | XFS_ILOCK_RTSUM; + xfs_ilock(sc->ip, sc->ilock_flags); + + /* Invoke the fork scrubber. */ + error = xfs_scrub_metadata_inode_forks(sc); + if (error || (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)) + goto out; + /* XXX: implement this some day */ - return -ENOENT; + xfs_scrub_set_incomplete(sc); +out: + /* Switch back to the rtbitmap inode and lock flags. */ + xfs_iunlock(sc->ip, sc->ilock_flags); + sc->ilock_flags = old_ilock_flags; + sc->ip = old_ip; + return error; } @@ -107,11 +143,23 @@ xfs_scrub_xref_is_used_rt_space( xfs_rtblock_t fsbno, xfs_extlen_t len) { + xfs_rtblock_t startext; + xfs_rtblock_t endext; + xfs_rtblock_t extcount; bool is_free; int error; + if (xfs_scrub_skip_xref(sc->sm)) + return; + + startext = fsbno; + endext = fsbno + len - 1; + do_div(startext, sc->mp->m_sb.sb_rextsize); + if (do_div(endext, sc->mp->m_sb.sb_rextsize)) + endext++; + extcount = endext - startext; xfs_ilock(sc->mp->m_rbmip, XFS_ILOCK_SHARED | XFS_ILOCK_RTBITMAP); - error = xfs_rtalloc_extent_is_free(sc->mp, sc->tp, fsbno, len, + error = xfs_rtalloc_extent_is_free(sc->mp, sc->tp, startext, extcount, &is_free); if (!xfs_scrub_should_check_xref(sc, &error, NULL)) goto out_unlock; diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c index 26c75967a072..36db098ba583 100644 --- a/fs/xfs/scrub/scrub.c +++ b/fs/xfs/scrub/scrub.c @@ -42,11 +42,18 @@ #include "xfs_refcount_btree.h" #include "xfs_rmap.h" #include "xfs_rmap_btree.h" +#include "xfs_quota.h" +#include "xfs_qm.h" +#include "xfs_errortag.h" +#include "xfs_error.h" +#include "xfs_log.h" +#include "xfs_trans_priv.h" #include "scrub/xfs_scrub.h" #include "scrub/scrub.h" #include "scrub/common.h" #include "scrub/trace.h" #include "scrub/btree.h" +#include "scrub/repair.h" /* * Online Scrub and Repair @@ -120,6 +127,24 @@ * XCORRUPT flag; btree query function errors are noted by setting the * XFAIL flag and deleting the cursor to prevent further attempts to * cross-reference with a defective btree. + * + * If a piece of metadata proves corrupt or suboptimal, the userspace + * program can ask the kernel to apply some tender loving care (TLC) to + * the metadata object by setting the REPAIR flag and re-calling the + * scrub ioctl. "Corruption" is defined by metadata violating the + * on-disk specification; operations cannot continue if the violation is + * left untreated. It is possible for XFS to continue if an object is + * "suboptimal", however performance may be degraded. Repairs are + * usually performed by rebuilding the metadata entirely out of + * redundant metadata. Optimizing, on the other hand, can sometimes be + * done without rebuilding entire structures. + * + * Generally speaking, the repair code has the following code structure: + * Lock -> scrub -> repair -> commit -> re-lock -> re-scrub -> unlock. + * The first check helps us figure out if we need to rebuild or simply + * optimize the structure so that the rebuild knows what to do. The + * second check evaluates the completeness of the repair; that is what + * is reported to userspace. */ /* @@ -155,7 +180,10 @@ xfs_scrub_teardown( { xfs_scrub_ag_free(sc, &sc->sa); if (sc->tp) { - xfs_trans_cancel(sc->tp); + if (error == 0 && (sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR)) + error = xfs_trans_commit(sc->tp); + else + xfs_trans_cancel(sc->tp); sc->tp = NULL; } if (sc->ip) { @@ -166,6 +194,8 @@ xfs_scrub_teardown( iput(VFS_I(sc->ip)); sc->ip = NULL; } + if (sc->has_quotaofflock) + mutex_unlock(&sc->mp->m_quotainfo->qi_quotaofflock); if (sc->buf) { kmem_free(sc->buf); sc->buf = NULL; @@ -180,126 +210,150 @@ static const struct xfs_scrub_meta_ops meta_scrub_ops[] = { .type = ST_NONE, .setup = xfs_scrub_setup_fs, .scrub = xfs_scrub_probe, + .repair = xfs_repair_probe, }, [XFS_SCRUB_TYPE_SB] = { /* superblock */ .type = ST_PERAG, .setup = xfs_scrub_setup_fs, .scrub = xfs_scrub_superblock, + .repair = xfs_repair_superblock, }, [XFS_SCRUB_TYPE_AGF] = { /* agf */ .type = ST_PERAG, .setup = xfs_scrub_setup_fs, .scrub = xfs_scrub_agf, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_AGFL]= { /* agfl */ .type = ST_PERAG, .setup = xfs_scrub_setup_fs, .scrub = xfs_scrub_agfl, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_AGI] = { /* agi */ .type = ST_PERAG, .setup = xfs_scrub_setup_fs, .scrub = xfs_scrub_agi, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_BNOBT] = { /* bnobt */ .type = ST_PERAG, .setup = xfs_scrub_setup_ag_allocbt, .scrub = xfs_scrub_bnobt, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_CNTBT] = { /* cntbt */ .type = ST_PERAG, .setup = xfs_scrub_setup_ag_allocbt, .scrub = xfs_scrub_cntbt, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_INOBT] = { /* inobt */ .type = ST_PERAG, .setup = xfs_scrub_setup_ag_iallocbt, .scrub = xfs_scrub_inobt, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_FINOBT] = { /* finobt */ .type = ST_PERAG, .setup = xfs_scrub_setup_ag_iallocbt, .scrub = xfs_scrub_finobt, .has = xfs_sb_version_hasfinobt, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_RMAPBT] = { /* rmapbt */ .type = ST_PERAG, .setup = xfs_scrub_setup_ag_rmapbt, .scrub = xfs_scrub_rmapbt, .has = xfs_sb_version_hasrmapbt, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_REFCNTBT] = { /* refcountbt */ .type = ST_PERAG, .setup = xfs_scrub_setup_ag_refcountbt, .scrub = xfs_scrub_refcountbt, .has = xfs_sb_version_hasreflink, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_INODE] = { /* inode record */ .type = ST_INODE, .setup = xfs_scrub_setup_inode, .scrub = xfs_scrub_inode, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_BMBTD] = { /* inode data fork */ .type = ST_INODE, .setup = xfs_scrub_setup_inode_bmap, .scrub = xfs_scrub_bmap_data, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_BMBTA] = { /* inode attr fork */ .type = ST_INODE, .setup = xfs_scrub_setup_inode_bmap, .scrub = xfs_scrub_bmap_attr, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_BMBTC] = { /* inode CoW fork */ .type = ST_INODE, .setup = xfs_scrub_setup_inode_bmap, .scrub = xfs_scrub_bmap_cow, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_DIR] = { /* directory */ .type = ST_INODE, .setup = xfs_scrub_setup_directory, .scrub = xfs_scrub_directory, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_XATTR] = { /* extended attributes */ .type = ST_INODE, .setup = xfs_scrub_setup_xattr, .scrub = xfs_scrub_xattr, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_SYMLINK] = { /* symbolic link */ .type = ST_INODE, .setup = xfs_scrub_setup_symlink, .scrub = xfs_scrub_symlink, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_PARENT] = { /* parent pointers */ .type = ST_INODE, .setup = xfs_scrub_setup_parent, .scrub = xfs_scrub_parent, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_RTBITMAP] = { /* realtime bitmap */ .type = ST_FS, .setup = xfs_scrub_setup_rt, .scrub = xfs_scrub_rtbitmap, .has = xfs_sb_version_hasrealtime, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_RTSUM] = { /* realtime summary */ .type = ST_FS, .setup = xfs_scrub_setup_rt, .scrub = xfs_scrub_rtsummary, .has = xfs_sb_version_hasrealtime, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_UQUOTA] = { /* user quota */ .type = ST_FS, .setup = xfs_scrub_setup_quota, .scrub = xfs_scrub_quota, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_GQUOTA] = { /* group quota */ .type = ST_FS, .setup = xfs_scrub_setup_quota, .scrub = xfs_scrub_quota, + .repair = xfs_repair_notsupported, }, [XFS_SCRUB_TYPE_PQUOTA] = { /* project quota */ .type = ST_FS, .setup = xfs_scrub_setup_quota, .scrub = xfs_scrub_quota, + .repair = xfs_repair_notsupported, }, }; @@ -379,15 +433,54 @@ xfs_scrub_validate_inputs( if (!xfs_sb_version_hasextflgbit(&mp->m_sb)) goto out; - /* We don't know how to repair anything yet. */ - if (sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) - goto out; + /* + * We only want to repair read-write v5+ filesystems. Defer the check + * for ops->repair until after our scrub confirms that we need to + * perform repairs so that we avoid failing due to not supporting + * repairing an object that doesn't need repairs. + */ + if (sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) { + error = -EOPNOTSUPP; + if (!xfs_sb_version_hascrc(&mp->m_sb)) + goto out; + + error = -EROFS; + if (mp->m_flags & XFS_MOUNT_RDONLY) + goto out; + } error = 0; out: return error; } +#ifdef CONFIG_XFS_ONLINE_REPAIR +static inline void xfs_scrub_postmortem(struct xfs_scrub_context *sc) +{ + /* + * Userspace asked us to repair something, we repaired it, rescanned + * it, and the rescan says it's still broken. Scream about this in + * the system logs. + */ + if ((sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) && + (sc->sm->sm_flags & (XFS_SCRUB_OFLAG_CORRUPT | + XFS_SCRUB_OFLAG_XCORRUPT))) + xfs_repair_failure(sc->mp); +} +#else +static inline void xfs_scrub_postmortem(struct xfs_scrub_context *sc) +{ + /* + * Userspace asked us to scrub something, it's broken, and we have no + * way of fixing it. Scream in the logs. + */ + if (sc->sm->sm_flags & (XFS_SCRUB_OFLAG_CORRUPT | + XFS_SCRUB_OFLAG_XCORRUPT)) + xfs_alert_ratelimited(sc->mp, + "Corruption detected during scrub."); +} +#endif /* CONFIG_XFS_ONLINE_REPAIR */ + /* Dispatch metadata scrubbing. */ int xfs_scrub_metadata( @@ -397,6 +490,7 @@ xfs_scrub_metadata( struct xfs_scrub_context sc; struct xfs_mount *mp = ip->i_mount; bool try_harder = false; + bool already_fixed = false; int error = 0; BUILD_BUG_ON(sizeof(meta_scrub_ops) != @@ -446,10 +540,44 @@ retry_op: } else if (error) goto out_teardown; - if (sc.sm->sm_flags & (XFS_SCRUB_OFLAG_CORRUPT | - XFS_SCRUB_OFLAG_XCORRUPT)) - xfs_alert_ratelimited(mp, "Corruption detected during scrub."); + if ((sc.sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) && !already_fixed) { + bool needs_fix; + + /* Let debug users force us into the repair routines. */ + if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_FORCE_SCRUB_REPAIR)) + sc.sm->sm_flags |= XFS_SCRUB_OFLAG_CORRUPT; + + needs_fix = (sc.sm->sm_flags & (XFS_SCRUB_OFLAG_CORRUPT | + XFS_SCRUB_OFLAG_XCORRUPT | + XFS_SCRUB_OFLAG_PREEN)); + /* + * If userspace asked for a repair but it wasn't necessary, + * report that back to userspace. + */ + if (!needs_fix) { + sc.sm->sm_flags |= XFS_SCRUB_OFLAG_NO_REPAIR_NEEDED; + goto out_nofix; + } + + /* + * If it's broken, userspace wants us to fix it, and we haven't + * already tried to fix it, then attempt a repair. + */ + error = xfs_repair_attempt(ip, &sc, &already_fixed); + if (error == -EAGAIN) { + if (sc.try_harder) + try_harder = true; + error = xfs_scrub_teardown(&sc, ip, 0); + if (error) { + xfs_repair_failure(mp); + goto out; + } + goto retry_op; + } + } +out_nofix: + xfs_scrub_postmortem(&sc); out_teardown: error = xfs_scrub_teardown(&sc, ip, error); out: diff --git a/fs/xfs/scrub/scrub.h b/fs/xfs/scrub/scrub.h index 0d92af86f67a..636424d5e2ee 100644 --- a/fs/xfs/scrub/scrub.h +++ b/fs/xfs/scrub/scrub.h @@ -38,6 +38,9 @@ struct xfs_scrub_meta_ops { /* Examine metadata for errors. */ int (*scrub)(struct xfs_scrub_context *); + /* Repair or optimize the metadata. */ + int (*repair)(struct xfs_scrub_context *); + /* Decide if we even have this piece of metadata. */ bool (*has)(struct xfs_sb *); @@ -48,6 +51,7 @@ struct xfs_scrub_meta_ops { /* Buffer pointers and btree cursors for an entire AG. */ struct xfs_scrub_ag { xfs_agnumber_t agno; + struct xfs_perag *pag; /* AG btree roots */ struct xfs_buf *agf_bp; @@ -73,6 +77,7 @@ struct xfs_scrub_context { void *buf; uint ilock_flags; bool try_harder; + bool has_quotaofflock; /* State tracking for single-AG operations. */ struct xfs_scrub_ag sa; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index 5d2b1c241be5..794d56bb1af8 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -69,6 +69,8 @@ DEFINE_EVENT(xfs_scrub_class, name, \ DEFINE_SCRUB_EVENT(xfs_scrub_start); DEFINE_SCRUB_EVENT(xfs_scrub_done); DEFINE_SCRUB_EVENT(xfs_scrub_deadlock_retry); +DEFINE_SCRUB_EVENT(xfs_repair_attempt); +DEFINE_SCRUB_EVENT(xfs_repair_done); TRACE_EVENT(xfs_scrub_op_error, TP_PROTO(struct xfs_scrub_context *sc, xfs_agnumber_t agno, @@ -492,6 +494,262 @@ TRACE_EVENT(xfs_scrub_xref_error, __entry->ret_ip) ); +/* repair tracepoints */ +#if IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) + +DECLARE_EVENT_CLASS(xfs_repair_extent_class, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agblock_t agbno, xfs_extlen_t len), + TP_ARGS(mp, agno, agbno, len), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, len) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->len = len; + ), + TP_printk("dev %d:%d agno %u agbno %u len %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->len) +); +#define DEFINE_REPAIR_EXTENT_EVENT(name) \ +DEFINE_EVENT(xfs_repair_extent_class, name, \ + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \ + xfs_agblock_t agbno, xfs_extlen_t len), \ + TP_ARGS(mp, agno, agbno, len)) +DEFINE_REPAIR_EXTENT_EVENT(xfs_repair_dispose_btree_extent); +DEFINE_REPAIR_EXTENT_EVENT(xfs_repair_collect_btree_extent); +DEFINE_REPAIR_EXTENT_EVENT(xfs_repair_agfl_insert); + +DECLARE_EVENT_CLASS(xfs_repair_rmap_class, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agblock_t agbno, xfs_extlen_t len, + uint64_t owner, uint64_t offset, unsigned int flags), + TP_ARGS(mp, agno, agbno, len, owner, offset, flags), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, len) + __field(uint64_t, owner) + __field(uint64_t, offset) + __field(unsigned int, flags) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->len = len; + __entry->owner = owner; + __entry->offset = offset; + __entry->flags = flags; + ), + TP_printk("dev %d:%d agno %u agbno %u len %u owner %lld offset %llu flags 0x%x", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->len, + __entry->owner, + __entry->offset, + __entry->flags) +); +#define DEFINE_REPAIR_RMAP_EVENT(name) \ +DEFINE_EVENT(xfs_repair_rmap_class, name, \ + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \ + xfs_agblock_t agbno, xfs_extlen_t len, \ + uint64_t owner, uint64_t offset, unsigned int flags), \ + TP_ARGS(mp, agno, agbno, len, owner, offset, flags)) +DEFINE_REPAIR_RMAP_EVENT(xfs_repair_alloc_extent_fn); +DEFINE_REPAIR_RMAP_EVENT(xfs_repair_ialloc_extent_fn); +DEFINE_REPAIR_RMAP_EVENT(xfs_repair_rmap_extent_fn); +DEFINE_REPAIR_RMAP_EVENT(xfs_repair_bmap_extent_fn); + +TRACE_EVENT(xfs_repair_refcount_extent_fn, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + struct xfs_refcount_irec *irec), + TP_ARGS(mp, agno, irec), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, startblock) + __field(xfs_extlen_t, blockcount) + __field(xfs_nlink_t, refcount) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->startblock = irec->rc_startblock; + __entry->blockcount = irec->rc_blockcount; + __entry->refcount = irec->rc_refcount; + ), + TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->startblock, + __entry->blockcount, + __entry->refcount) +) + +TRACE_EVENT(xfs_repair_init_btblock, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno, + xfs_btnum_t btnum), + TP_ARGS(mp, agno, agbno, btnum), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(uint32_t, btnum) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->btnum = btnum; + ), + TP_printk("dev %d:%d agno %u agbno %u btnum %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->btnum) +) +TRACE_EVENT(xfs_repair_findroot_block, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno, + uint32_t magic, uint16_t level), + TP_ARGS(mp, agno, agbno, magic, level), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(uint32_t, magic) + __field(uint16_t, level) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->magic = magic; + __entry->level = level; + ), + TP_printk("dev %d:%d agno %u agbno %u magic 0x%x level %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->magic, + __entry->level) +) +TRACE_EVENT(xfs_repair_calc_ag_resblks, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agino_t icount, xfs_agblock_t aglen, xfs_agblock_t freelen, + xfs_agblock_t usedlen), + TP_ARGS(mp, agno, icount, aglen, freelen, usedlen), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agino_t, icount) + __field(xfs_agblock_t, aglen) + __field(xfs_agblock_t, freelen) + __field(xfs_agblock_t, usedlen) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->icount = icount; + __entry->aglen = aglen; + __entry->freelen = freelen; + __entry->usedlen = usedlen; + ), + TP_printk("dev %d:%d agno %d icount %u aglen %u freelen %u usedlen %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->icount, + __entry->aglen, + __entry->freelen, + __entry->usedlen) +) +TRACE_EVENT(xfs_repair_calc_ag_resblks_btsize, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agblock_t bnobt_sz, xfs_agblock_t inobt_sz, + xfs_agblock_t rmapbt_sz, xfs_agblock_t refcbt_sz), + TP_ARGS(mp, agno, bnobt_sz, inobt_sz, rmapbt_sz, refcbt_sz), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, bnobt_sz) + __field(xfs_agblock_t, inobt_sz) + __field(xfs_agblock_t, rmapbt_sz) + __field(xfs_agblock_t, refcbt_sz) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->bnobt_sz = bnobt_sz; + __entry->inobt_sz = inobt_sz; + __entry->rmapbt_sz = rmapbt_sz; + __entry->refcbt_sz = refcbt_sz; + ), + TP_printk("dev %d:%d agno %d bno %u ino %u rmap %u refcount %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->bnobt_sz, + __entry->inobt_sz, + __entry->rmapbt_sz, + __entry->refcbt_sz) +) +TRACE_EVENT(xfs_repair_reset_counters, + TP_PROTO(struct xfs_mount *mp), + TP_ARGS(mp), + TP_STRUCT__entry( + __field(dev_t, dev) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + ), + TP_printk("dev %d:%d", + MAJOR(__entry->dev), MINOR(__entry->dev)) +) + +TRACE_EVENT(xfs_repair_ialloc_insert, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agino_t startino, uint16_t holemask, uint8_t count, + uint8_t freecount, uint64_t freemask), + TP_ARGS(mp, agno, startino, holemask, count, freecount, freemask), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agino_t, startino) + __field(uint16_t, holemask) + __field(uint8_t, count) + __field(uint8_t, freecount) + __field(uint64_t, freemask) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->startino = startino; + __entry->holemask = holemask; + __entry->count = count; + __entry->freecount = freecount; + __entry->freemask = freemask; + ), + TP_printk("dev %d:%d agno %d startino %u holemask 0x%x count %u freecount %u freemask 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->startino, + __entry->holemask, + __entry->count, + __entry->freecount, + __entry->freemask) +) + +#endif /* IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR) */ + #endif /* _TRACE_XFS_SCRUB_TRACE_H */ #undef TRACE_INCLUDE_PATH diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 19eadc807056..ca6903726689 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -594,7 +594,7 @@ xfs_alloc_ioend( struct xfs_ioend *ioend; struct bio *bio; - bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, xfs_ioend_bioset); + bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &xfs_ioend_bioset); xfs_init_bio_from_bh(bio, bh); ioend = container_of(bio, struct xfs_ioend, io_inline_bio); @@ -1195,16 +1195,22 @@ xfs_vm_writepages( int ret; xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); - if (dax_mapping(mapping)) - return dax_writeback_mapping_range(mapping, - xfs_find_bdev_for_inode(mapping->host), wbc); - ret = write_cache_pages(mapping, wbc, xfs_do_writepage, &wpc); if (wpc.ioend) ret = xfs_submit_ioend(wbc, wpc.ioend, ret); return ret; } +STATIC int +xfs_dax_writepages( + struct address_space *mapping, + struct writeback_control *wbc) +{ + xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); + return dax_writeback_mapping_range(mapping, + xfs_find_bdev_for_inode(mapping->host), wbc); +} + /* * Called to move a page into cleanable state - and from there * to be released. The page should already be clean. We always @@ -1367,30 +1373,18 @@ out_unlock: return error; } -STATIC ssize_t -xfs_vm_direct_IO( - struct kiocb *iocb, - struct iov_iter *iter) -{ - /* - * We just need the method present so that open/fcntl allow direct I/O. - */ - return -EINVAL; -} - STATIC sector_t xfs_vm_bmap( struct address_space *mapping, sector_t block) { - struct inode *inode = (struct inode *)mapping->host; - struct xfs_inode *ip = XFS_I(inode); + struct xfs_inode *ip = XFS_I(mapping->host); - trace_xfs_vm_bmap(XFS_I(inode)); + trace_xfs_vm_bmap(ip); /* * The swap code (ab-)uses ->bmap to get a block mapping and then - * bypasseѕ the file system for actual I/O. We really can't allow + * bypasses the file system for actual I/O. We really can't allow * that on reflinks inodes, so we have to skip out here. And yes, * 0 is the magic code for a bmap error. * @@ -1399,9 +1393,7 @@ xfs_vm_bmap( */ if (xfs_is_reflink_inode(ip) || XFS_IS_REALTIME_INODE(ip)) return 0; - - filemap_write_and_wait(mapping); - return generic_block_bmap(mapping, block, xfs_get_blocks); + return iomap_bmap(mapping, block, &xfs_iomap_ops); } STATIC int @@ -1472,25 +1464,24 @@ xfs_vm_set_page_dirty( newly_dirty = !TestSetPageDirty(page); spin_unlock(&mapping->private_lock); - if (newly_dirty) { - /* sigh - __set_page_dirty() is static, so copy it here, too */ - unsigned long flags; - - spin_lock_irqsave(&mapping->tree_lock, flags); - if (page->mapping) { /* Race with truncate? */ - WARN_ON_ONCE(!PageUptodate(page)); - account_page_dirtied(page, mapping); - radix_tree_tag_set(&mapping->page_tree, - page_index(page), PAGECACHE_TAG_DIRTY); - } - spin_unlock_irqrestore(&mapping->tree_lock, flags); - } + if (newly_dirty) + __set_page_dirty(page, mapping, 1); unlock_page_memcg(page); if (newly_dirty) __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); return newly_dirty; } +static int +xfs_iomap_swapfile_activate( + struct swap_info_struct *sis, + struct file *swap_file, + sector_t *span) +{ + sis->bdev = xfs_find_bdev_for_inode(file_inode(swap_file)); + return iomap_swapfile_activate(sis, swap_file, span, &xfs_iomap_ops); +} + const struct address_space_operations xfs_address_space_operations = { .readpage = xfs_vm_readpage, .readpages = xfs_vm_readpages, @@ -1500,8 +1491,17 @@ const struct address_space_operations xfs_address_space_operations = { .releasepage = xfs_vm_releasepage, .invalidatepage = xfs_vm_invalidatepage, .bmap = xfs_vm_bmap, - .direct_IO = xfs_vm_direct_IO, + .direct_IO = noop_direct_IO, .migratepage = buffer_migrate_page, .is_partially_uptodate = block_is_partially_uptodate, .error_remove_page = generic_error_remove_page, + .swap_activate = xfs_iomap_swapfile_activate, +}; + +const struct address_space_operations xfs_dax_aops = { + .writepages = xfs_dax_writepages, + .direct_IO = noop_direct_IO, + .set_page_dirty = noop_set_page_dirty, + .invalidatepage = noop_invalidatepage, + .swap_activate = xfs_iomap_swapfile_activate, }; diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h index 88c85ea63da0..694c85b03813 100644 --- a/fs/xfs/xfs_aops.h +++ b/fs/xfs/xfs_aops.h @@ -18,7 +18,7 @@ #ifndef __XFS_AOPS_H__ #define __XFS_AOPS_H__ -extern struct bio_set *xfs_ioend_bioset; +extern struct bio_set xfs_ioend_bioset; /* * Types of I/O for bmap clustering and I/O completion tracking. @@ -54,6 +54,7 @@ struct xfs_ioend { }; extern const struct address_space_operations xfs_address_space_operations; +extern const struct address_space_operations xfs_dax_aops; int xfs_setfilesize(struct xfs_inode *ip, xfs_off_t offset, size_t size); diff --git a/fs/xfs/xfs_bmap_item.c b/fs/xfs/xfs_bmap_item.c index e5fb008d75e8..618bb71535c8 100644 --- a/fs/xfs/xfs_bmap_item.c +++ b/fs/xfs/xfs_bmap_item.c @@ -53,6 +53,25 @@ xfs_bui_item_free( kmem_zone_free(xfs_bui_zone, buip); } +/* + * Freeing the BUI requires that we remove it from the AIL if it has already + * been placed there. However, the BUI may not yet have been placed in the AIL + * when called by xfs_bui_release() from BUD processing due to the ordering of + * committed vs unpin operations in bulk insert operations. Hence the reference + * count to ensure only the last caller frees the BUI. + */ +void +xfs_bui_release( + struct xfs_bui_log_item *buip) +{ + ASSERT(atomic_read(&buip->bui_refcount) > 0); + if (atomic_dec_and_test(&buip->bui_refcount)) { + xfs_trans_ail_remove(&buip->bui_item, SHUTDOWN_LOG_IO_ERROR); + xfs_bui_item_free(buip); + } +} + + STATIC void xfs_bui_item_size( struct xfs_log_item *lip, @@ -141,8 +160,8 @@ STATIC void xfs_bui_item_unlock( struct xfs_log_item *lip) { - if (lip->li_flags & XFS_LI_ABORTED) - xfs_bui_item_free(BUI_ITEM(lip)); + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) + xfs_bui_release(BUI_ITEM(lip)); } /* @@ -206,24 +225,6 @@ xfs_bui_init( return buip; } -/* - * Freeing the BUI requires that we remove it from the AIL if it has already - * been placed there. However, the BUI may not yet have been placed in the AIL - * when called by xfs_bui_release() from BUD processing due to the ordering of - * committed vs unpin operations in bulk insert operations. Hence the reference - * count to ensure only the last caller frees the BUI. - */ -void -xfs_bui_release( - struct xfs_bui_log_item *buip) -{ - ASSERT(atomic_read(&buip->bui_refcount) > 0); - if (atomic_dec_and_test(&buip->bui_refcount)) { - xfs_trans_ail_remove(&buip->bui_item, SHUTDOWN_LOG_IO_ERROR); - xfs_bui_item_free(buip); - } -} - static inline struct xfs_bud_log_item *BUD_ITEM(struct xfs_log_item *lip) { return container_of(lip, struct xfs_bud_log_item, bud_item); @@ -304,7 +305,7 @@ xfs_bud_item_unlock( { struct xfs_bud_log_item *budp = BUD_ITEM(lip); - if (lip->li_flags & XFS_LI_ABORTED) { + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) { xfs_bui_release(budp->bud_buip); kmem_zone_free(xfs_bud_zone, budp); } diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 05dee8fdd895..06badcbadeb4 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -848,7 +848,7 @@ xfs_free_eofblocks( /* * Attach the dquots to the inode up front. */ - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return error; @@ -871,8 +871,8 @@ xfs_free_eofblocks( * contents of the file are flushed to disk then the files * may be full of holes (ie NULL files bug). */ - error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, - XFS_ISIZE(ip)); + error = xfs_itruncate_extents_flags(&tp, ip, XFS_DATA_FORK, + XFS_ISIZE(ip), XFS_BMAPI_NODISCARD); if (error) { /* * If we get an error at this point we simply don't @@ -918,7 +918,7 @@ xfs_alloc_file_space( if (XFS_FORCED_SHUTDOWN(mp)) return -EIO; - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return error; @@ -1169,7 +1169,7 @@ xfs_free_file_space( trace_xfs_free_file_space(ip); - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return error; @@ -1326,7 +1326,6 @@ xfs_collapse_file_space( int error; struct xfs_defer_ops dfops; xfs_fsblock_t first_block; - xfs_fileoff_t stop_fsb = XFS_B_TO_FSB(mp, VFS_I(ip)->i_size); xfs_fileoff_t next_fsb = XFS_B_TO_FSB(mp, offset + len); xfs_fileoff_t shift_fsb = XFS_B_TO_FSB(mp, len); uint resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); @@ -1361,7 +1360,7 @@ xfs_collapse_file_space( xfs_defer_init(&dfops, &first_block); error = xfs_bmap_collapse_extents(tp, ip, &next_fsb, shift_fsb, - &done, stop_fsb, &first_block, &dfops); + &done, &first_block, &dfops); if (error) goto out_bmap_cancel; diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index ac669a10c62f..5179ab9e3d6a 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -549,17 +549,31 @@ xfs_buf_hash_destroy( } /* - * Look up, and creates if absent, a lockable buffer for - * a given range of an inode. The buffer is returned - * locked. No I/O is implied by this call. + * Look up a buffer in the buffer cache and return it referenced and locked + * in @found_bp. + * + * If @new_bp is supplied and we have a lookup miss, insert @new_bp into the + * cache. + * + * If XBF_TRYLOCK is set in @flags, only try to lock the buffer and return + * -EAGAIN if we fail to lock it. + * + * Return values are: + * -EFSCORRUPTED if have been supplied with an invalid address + * -EAGAIN on trylock failure + * -ENOENT if we fail to find a match and @new_bp was NULL + * 0, with @found_bp: + * - @new_bp if we inserted it into the cache + * - the buffer we found and locked. */ -xfs_buf_t * -_xfs_buf_find( +static int +xfs_buf_find( struct xfs_buftarg *btp, struct xfs_buf_map *map, int nmaps, xfs_buf_flags_t flags, - xfs_buf_t *new_bp) + struct xfs_buf *new_bp, + struct xfs_buf **found_bp) { struct xfs_perag *pag; xfs_buf_t *bp; @@ -567,6 +581,8 @@ _xfs_buf_find( xfs_daddr_t eofs; int i; + *found_bp = NULL; + for (i = 0; i < nmaps; i++) cmap.bm_len += map[i].bm_len; @@ -580,16 +596,11 @@ _xfs_buf_find( */ eofs = XFS_FSB_TO_BB(btp->bt_mount, btp->bt_mount->m_sb.sb_dblocks); if (cmap.bm_bn < 0 || cmap.bm_bn >= eofs) { - /* - * XXX (dgc): we should really be returning -EFSCORRUPTED here, - * but none of the higher level infrastructure supports - * returning a specific error on buffer lookup failures. - */ xfs_alert(btp->bt_mount, "%s: daddr 0x%llx out of range, EOFS 0x%llx", __func__, cmap.bm_bn, eofs); WARN_ON(1); - return NULL; + return -EFSCORRUPTED; } pag = xfs_perag_get(btp->bt_mount, @@ -604,19 +615,20 @@ _xfs_buf_find( } /* No match found */ - if (new_bp) { - /* the buffer keeps the perag reference until it is freed */ - new_bp->b_pag = pag; - rhashtable_insert_fast(&pag->pag_buf_hash, - &new_bp->b_rhash_head, - xfs_buf_hash_params); - spin_unlock(&pag->pag_buf_lock); - } else { + if (!new_bp) { XFS_STATS_INC(btp->bt_mount, xb_miss_locked); spin_unlock(&pag->pag_buf_lock); xfs_perag_put(pag); + return -ENOENT; } - return new_bp; + + /* the buffer keeps the perag reference until it is freed */ + new_bp->b_pag = pag; + rhashtable_insert_fast(&pag->pag_buf_hash, &new_bp->b_rhash_head, + xfs_buf_hash_params); + spin_unlock(&pag->pag_buf_lock); + *found_bp = new_bp; + return 0; found: spin_unlock(&pag->pag_buf_lock); @@ -626,7 +638,7 @@ found: if (flags & XBF_TRYLOCK) { xfs_buf_rele(bp); XFS_STATS_INC(btp->bt_mount, xb_busy_locked); - return NULL; + return -EAGAIN; } xfs_buf_lock(bp); XFS_STATS_INC(btp->bt_mount, xb_get_locked_waited); @@ -646,6 +658,24 @@ found: trace_xfs_buf_find(bp, flags, _RET_IP_); XFS_STATS_INC(btp->bt_mount, xb_get_locked); + *found_bp = bp; + return 0; +} + +struct xfs_buf * +xfs_buf_incore( + struct xfs_buftarg *target, + xfs_daddr_t blkno, + size_t numblks, + xfs_buf_flags_t flags) +{ + struct xfs_buf *bp; + int error; + DEFINE_SINGLE_BUF_MAP(map, blkno, numblks); + + error = xfs_buf_find(target, &map, 1, flags, NULL, &bp); + if (error) + return NULL; return bp; } @@ -665,9 +695,27 @@ xfs_buf_get_map( struct xfs_buf *new_bp; int error = 0; - bp = _xfs_buf_find(target, map, nmaps, flags, NULL); - if (likely(bp)) + error = xfs_buf_find(target, map, nmaps, flags, NULL, &bp); + + switch (error) { + case 0: + /* cache hit */ goto found; + case -EAGAIN: + /* cache hit, trylock failure, caller handles failure */ + ASSERT(flags & XBF_TRYLOCK); + return NULL; + case -ENOENT: + /* cache miss, go for insert */ + break; + case -EFSCORRUPTED: + default: + /* + * None of the higher layers understand failure types + * yet, so return NULL to signal a fatal lookup error. + */ + return NULL; + } new_bp = _xfs_buf_alloc(target, map, nmaps, flags); if (unlikely(!new_bp)) @@ -679,8 +727,8 @@ xfs_buf_get_map( return NULL; } - bp = _xfs_buf_find(target, map, nmaps, flags, new_bp); - if (!bp) { + error = xfs_buf_find(target, map, nmaps, flags, new_bp, &bp); + if (error) { xfs_buf_free(new_bp); return NULL; } @@ -1754,7 +1802,6 @@ xfs_buftarg_shrink_count( void xfs_free_buftarg( - struct xfs_mount *mp, struct xfs_buftarg *btp) { unregister_shrinker(&btp->bt_shrinker); diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index 2f4c91452861..f5f2b71c2fde 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -218,20 +218,9 @@ typedef struct xfs_buf { } xfs_buf_t; /* Finding and Reading Buffers */ -struct xfs_buf *_xfs_buf_find(struct xfs_buftarg *target, - struct xfs_buf_map *map, int nmaps, - xfs_buf_flags_t flags, struct xfs_buf *new_bp); - -static inline struct xfs_buf * -xfs_incore( - struct xfs_buftarg *target, - xfs_daddr_t blkno, - size_t numblks, - xfs_buf_flags_t flags) -{ - DEFINE_SINGLE_BUF_MAP(map, blkno, numblks); - return _xfs_buf_find(target, &map, 1, flags, NULL); -} +struct xfs_buf *xfs_buf_incore(struct xfs_buftarg *target, + xfs_daddr_t blkno, size_t numblks, + xfs_buf_flags_t flags); struct xfs_buf *_xfs_buf_alloc(struct xfs_buftarg *target, struct xfs_buf_map *map, int nmaps, @@ -358,6 +347,18 @@ extern void xfs_buf_terminate(void); void xfs_buf_set_ref(struct xfs_buf *bp, int lru_ref); +/* + * If the buffer is already on the LRU, do nothing. Otherwise set the buffer + * up with a reference count of 0 so it will be tossed from the cache when + * released. + */ +static inline void xfs_buf_oneshot(struct xfs_buf *bp) +{ + if (!list_empty(&bp->b_lru) || atomic_read(&bp->b_lru_ref) > 1) + return; + atomic_set(&bp->b_lru_ref, 0); +} + static inline int xfs_buf_ispinned(struct xfs_buf *bp) { return atomic_read(&bp->b_pin_count); @@ -388,7 +389,7 @@ xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset) */ extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *, struct block_device *, struct dax_device *); -extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *); +extern void xfs_free_buftarg(struct xfs_buftarg *); extern void xfs_wait_buftarg(xfs_buftarg_t *); extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int); diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 82ad270e390e..c2311379d1c3 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -438,7 +438,7 @@ xfs_buf_item_unpin( * xfs_trans_uncommit() will try to reference the * buffer which we no longer have a hold on. */ - if (lip->li_desc) + if (!list_empty(&lip->li_trans)) xfs_trans_del_item(lip); /* @@ -568,13 +568,15 @@ xfs_buf_item_unlock( { struct xfs_buf_log_item *bip = BUF_ITEM(lip); struct xfs_buf *bp = bip->bli_buf; - bool aborted = !!(lip->li_flags & XFS_LI_ABORTED); + bool aborted; bool hold = !!(bip->bli_flags & XFS_BLI_HOLD); bool dirty = !!(bip->bli_flags & XFS_BLI_DIRTY); #if defined(DEBUG) || defined(XFS_WARN) bool ordered = !!(bip->bli_flags & XFS_BLI_ORDERED); #endif + aborted = test_bit(XFS_LI_ABORTED, &lip->li_flags); + /* Clear the buffer's association with this transaction. */ bp->b_transp = NULL; @@ -743,8 +745,10 @@ xfs_buf_item_init( * nothing to do here so return. */ ASSERT(bp->b_target->bt_mount == mp); - if (bip != NULL) { + if (bip) { ASSERT(bip->bli_item.li_type == XFS_LI_BUF); + ASSERT(!bp->b_transp); + ASSERT(bip->bli_buf == bp); return 0; } diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index b2cde5426182..7b68e6c9a474 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -50,19 +50,19 @@ xfs_trim_extents( pag = xfs_perag_get(mp, agno); - error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp); - if (error || !agbp) - goto out_put_perag; - - cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT); - /* * Force out the log. This means any transactions that might have freed - * space before we took the AGF buffer lock are now on disk, and the + * space before we take the AGF buffer lock are now on disk, and the * volatile disk cache is flushed. */ xfs_log_force(mp, XFS_LOG_SYNC); + error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp); + if (error || !agbp) + goto out_put_perag; + + cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT); + /* * Look up the longest btree in the AGF and start with it. */ diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index a7daef9e16bf..2567391489bd 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -288,49 +288,43 @@ xfs_dquot_set_prealloc_limits(struct xfs_dquot *dqp) } /* - * Allocate a block and fill it with dquots. - * This is called when the bmapi finds a hole. + * Ensure that the given in-core dquot has a buffer on disk backing it, and + * return the buffer. This is called when the bmapi finds a hole. */ STATIC int -xfs_qm_dqalloc( - xfs_trans_t **tpp, - xfs_mount_t *mp, - xfs_dquot_t *dqp, - xfs_inode_t *quotip, - xfs_fileoff_t offset_fsb, - xfs_buf_t **O_bpp) +xfs_dquot_disk_alloc( + struct xfs_trans **tpp, + struct xfs_dquot *dqp, + struct xfs_buf **bpp) { - xfs_fsblock_t firstblock; - struct xfs_defer_ops dfops; - xfs_bmbt_irec_t map; - int nmaps, error; - xfs_buf_t *bp; - xfs_trans_t *tp = *tpp; - - ASSERT(tp != NULL); + struct xfs_bmbt_irec map; + struct xfs_defer_ops dfops; + struct xfs_mount *mp = (*tpp)->t_mountp; + struct xfs_buf *bp; + struct xfs_inode *quotip = xfs_quota_inode(mp, dqp->dq_flags); + xfs_fsblock_t firstblock; + int nmaps = 1; + int error; trace_xfs_dqalloc(dqp); - /* - * Initialize the bmap freelist prior to calling bmapi code. - */ xfs_defer_init(&dfops, &firstblock); xfs_ilock(quotip, XFS_ILOCK_EXCL); - /* - * Return if this type of quotas is turned off while we didn't - * have an inode lock - */ if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) { + /* + * Return if this type of quotas is turned off while we didn't + * have an inode lock + */ xfs_iunlock(quotip, XFS_ILOCK_EXCL); return -ESRCH; } - xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL); - nmaps = 1; - error = xfs_bmapi_write(tp, quotip, offset_fsb, - XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, - &firstblock, XFS_QM_DQALLOC_SPACE_RES(mp), - &map, &nmaps, &dfops); + /* Create the block mapping. */ + xfs_trans_ijoin(*tpp, quotip, XFS_ILOCK_EXCL); + error = xfs_bmapi_write(*tpp, quotip, dqp->q_fileoffset, + XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, + &firstblock, XFS_QM_DQALLOC_SPACE_RES(mp), + &map, &nmaps, &dfops); if (error) goto error0; ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB); @@ -344,10 +338,8 @@ xfs_qm_dqalloc( dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); /* now we can just get the buffer (there's nothing to read yet) */ - bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, - dqp->q_blkno, - mp->m_quotainfo->qi_dqchunklen, - 0); + bp = xfs_trans_get_buf(*tpp, mp->m_ddev_targp, dqp->q_blkno, + mp->m_quotainfo->qi_dqchunklen, 0); if (!bp) { error = -ENOMEM; goto error1; @@ -358,37 +350,45 @@ xfs_qm_dqalloc( * Make a chunk of dquots out of this buffer and log * the entire thing. */ - xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id), + xfs_qm_init_dquot_blk(*tpp, mp, be32_to_cpu(dqp->q_core.d_id), dqp->dq_flags & XFS_DQ_ALLTYPES, bp); + xfs_buf_set_ref(bp, XFS_DQUOT_REF); /* - * xfs_defer_finish() may commit the current transaction and - * start a second transaction if the freelist is not empty. + * Hold the buffer and join it to the dfops so that we'll still own + * the buffer when we return to the caller. The buffer disposal on + * error must be paid attention to very carefully, as it has been + * broken since commit efa092f3d4c6 "[XFS] Fixes a bug in the quota + * code when allocating a new dquot record" in 2005, and the later + * conversion to xfs_defer_ops in commit 310a75a3c6c747 failed to keep + * the buffer locked across the _defer_finish call. We can now do + * this correctly with xfs_defer_bjoin. * - * Since we still want to modify this buffer, we need to - * ensure that the buffer is not released on commit of - * the first transaction and ensure the buffer is added to the - * second transaction. + * Above, we allocated a disk block for the dquot information and + * used get_buf to initialize the dquot. If the _defer_bjoin fails, + * the buffer is still locked to *tpp, so we must _bhold_release and + * then _trans_brelse the buffer. If the _defer_finish fails, the old + * transaction is gone but the new buffer is not joined or held to any + * transaction, so we must _buf_relse it. * - * If there is only one transaction then don't stop the buffer - * from being released when it commits later on. + * If everything succeeds, the caller of this function is returned a + * buffer that is locked and held to the transaction. The caller + * is responsible for unlocking any buffer passed back, either + * manually or by committing the transaction. */ - - xfs_trans_bhold(tp, bp); - + xfs_trans_bhold(*tpp, bp); + error = xfs_defer_bjoin(&dfops, bp); + if (error) { + xfs_trans_bhold_release(*tpp, bp); + xfs_trans_brelse(*tpp, bp); + goto error1; + } error = xfs_defer_finish(tpp, &dfops); - if (error) + if (error) { + xfs_buf_relse(bp); goto error1; - - /* Transaction was committed? */ - if (*tpp != tp) { - tp = *tpp; - xfs_trans_bjoin(tp, bp); - } else { - xfs_trans_bhold_release(tp, bp); } - - *O_bpp = bp; + *bpp = bp; return 0; error1: @@ -398,32 +398,24 @@ error0: } /* - * Maps a dquot to the buffer containing its on-disk version. - * This returns a ptr to the buffer containing the on-disk dquot - * in the bpp param, and a ptr to the on-disk dquot within that buffer + * Read in the in-core dquot's on-disk metadata and return the buffer. + * Returns ENOENT to signal a hole. */ STATIC int -xfs_qm_dqtobp( - xfs_trans_t **tpp, - xfs_dquot_t *dqp, - xfs_disk_dquot_t **O_ddpp, - xfs_buf_t **O_bpp, - uint flags) +xfs_dquot_disk_read( + struct xfs_mount *mp, + struct xfs_dquot *dqp, + struct xfs_buf **bpp) { struct xfs_bmbt_irec map; - int nmaps = 1, error; struct xfs_buf *bp; - struct xfs_inode *quotip; - struct xfs_mount *mp = dqp->q_mount; - xfs_dqid_t id = be32_to_cpu(dqp->q_core.d_id); - struct xfs_trans *tp = (tpp ? *tpp : NULL); + struct xfs_inode *quotip = xfs_quota_inode(mp, dqp->dq_flags); uint lock_mode; - - quotip = xfs_quota_inode(dqp->q_mount, dqp->dq_flags); - dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk; + int nmaps = 1; + int error; lock_mode = xfs_ilock_data_map_shared(quotip); - if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) { + if (!xfs_this_quota_on(mp, dqp->dq_flags)) { /* * Return if this type of quotas is turned off while we * didn't have the quota inode lock. @@ -436,81 +428,48 @@ xfs_qm_dqtobp( * Find the block map; no allocations yet */ error = xfs_bmapi_read(quotip, dqp->q_fileoffset, - XFS_DQUOT_CLUSTER_SIZE_FSB, &map, &nmaps, 0); - + XFS_DQUOT_CLUSTER_SIZE_FSB, &map, &nmaps, 0); xfs_iunlock(quotip, lock_mode); if (error) return error; ASSERT(nmaps == 1); - ASSERT(map.br_blockcount == 1); + ASSERT(map.br_blockcount >= 1); + ASSERT(map.br_startblock != DELAYSTARTBLOCK); + if (map.br_startblock == HOLESTARTBLOCK) + return -ENOENT; + + trace_xfs_dqtobp_read(dqp); /* - * Offset of dquot in the (fixed sized) dquot chunk. + * store the blkno etc so that we don't have to do the + * mapping all the time */ - dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) * - sizeof(xfs_dqblk_t); - - ASSERT(map.br_startblock != DELAYSTARTBLOCK); - if (map.br_startblock == HOLESTARTBLOCK) { - /* - * We don't allocate unless we're asked to - */ - if (!(flags & XFS_QMOPT_DQALLOC)) - return -ENOENT; - - ASSERT(tp); - error = xfs_qm_dqalloc(tpp, mp, dqp, quotip, - dqp->q_fileoffset, &bp); - if (error) - return error; - tp = *tpp; - } else { - trace_xfs_dqtobp_read(dqp); + dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); - /* - * store the blkno etc so that we don't have to do the - * mapping all the time - */ - dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); - - error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, - dqp->q_blkno, - mp->m_quotainfo->qi_dqchunklen, - 0, &bp, &xfs_dquot_buf_ops); - if (error) { - ASSERT(bp == NULL); - return error; - } + error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno, + mp->m_quotainfo->qi_dqchunklen, 0, &bp, + &xfs_dquot_buf_ops); + if (error) { + ASSERT(bp == NULL); + return error; } ASSERT(xfs_buf_islocked(bp)); - *O_bpp = bp; - *O_ddpp = bp->b_addr + dqp->q_bufoffset; + xfs_buf_set_ref(bp, XFS_DQUOT_REF); + *bpp = bp; return 0; } - -/* - * Read in the ondisk dquot using dqtobp() then copy it to an incore version, - * and release the buffer immediately. - * - * If XFS_QMOPT_DQALLOC is set, allocate a dquot on disk if it needed. - */ -int -xfs_qm_dqread( +/* Allocate and initialize everything we need for an incore dquot. */ +STATIC struct xfs_dquot * +xfs_dquot_alloc( struct xfs_mount *mp, xfs_dqid_t id, - uint type, - uint flags, - struct xfs_dquot **O_dqpp) + uint type) { struct xfs_dquot *dqp; - struct xfs_disk_dquot *ddqp; - struct xfs_buf *bp; - struct xfs_trans *tp = NULL; - int error; dqp = kmem_zone_zalloc(xfs_qm_dqzone, KM_SLEEP); @@ -520,6 +479,12 @@ xfs_qm_dqread( INIT_LIST_HEAD(&dqp->q_lru); mutex_init(&dqp->q_qlock); init_waitqueue_head(&dqp->q_pinwait); + dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk; + /* + * Offset of dquot in the (fixed sized) dquot chunk. + */ + dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) * + sizeof(xfs_dqblk_t); /* * Because we want to use a counting completion, complete @@ -548,35 +513,22 @@ xfs_qm_dqread( break; } - XFS_STATS_INC(mp, xs_qm_dquot); - - trace_xfs_dqread(dqp); + xfs_qm_dquot_logitem_init(dqp); - if (flags & XFS_QMOPT_DQALLOC) { - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_dqalloc, - XFS_QM_DQALLOC_SPACE_RES(mp), 0, 0, &tp); - if (error) - goto error0; - } + XFS_STATS_INC(mp, xs_qm_dquot); + return dqp; +} - /* - * get a pointer to the on-disk dquot and the buffer containing it - * dqp already knows its own type (GROUP/USER). - */ - error = xfs_qm_dqtobp(&tp, dqp, &ddqp, &bp, flags); - if (error) { - /* - * This can happen if quotas got turned off (ESRCH), - * or if the dquot didn't exist on disk and we ask to - * allocate (ENOENT). - */ - trace_xfs_dqread_fail(dqp); - goto error1; - } +/* Copy the in-core quota fields in from the on-disk buffer. */ +STATIC void +xfs_dquot_from_disk( + struct xfs_dquot *dqp, + struct xfs_buf *bp) +{ + struct xfs_disk_dquot *ddqp = bp->b_addr + dqp->q_bufoffset; /* copy everything from disk dquot to the incore dquot */ memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t)); - xfs_qm_dquot_logitem_init(dqp); /* * Reservation counters are defined as reservation plus current usage @@ -588,40 +540,90 @@ xfs_qm_dqread( /* initialize the dquot speculative prealloc thresholds */ xfs_dquot_set_prealloc_limits(dqp); +} - /* Mark the buf so that this will stay incore a little longer */ - xfs_buf_set_ref(bp, XFS_DQUOT_REF); +/* Allocate and initialize the dquot buffer for this in-core dquot. */ +static int +xfs_qm_dqread_alloc( + struct xfs_mount *mp, + struct xfs_dquot *dqp, + struct xfs_buf **bpp) +{ + struct xfs_trans *tp; + struct xfs_buf *bp; + int error; + + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_dqalloc, + XFS_QM_DQALLOC_SPACE_RES(mp), 0, 0, &tp); + if (error) + goto err; + + error = xfs_dquot_disk_alloc(&tp, dqp, &bp); + if (error) + goto err_cancel; + + error = xfs_trans_commit(tp); + if (error) { + /* + * Buffer was held to the transaction, so we have to unlock it + * manually here because we're not passing it back. + */ + xfs_buf_relse(bp); + goto err; + } + *bpp = bp; + return 0; + +err_cancel: + xfs_trans_cancel(tp); +err: + return error; +} + +/* + * Read in the ondisk dquot using dqtobp() then copy it to an incore version, + * and release the buffer immediately. If @can_alloc is true, fill any + * holes in the on-disk metadata. + */ +static int +xfs_qm_dqread( + struct xfs_mount *mp, + xfs_dqid_t id, + uint type, + bool can_alloc, + struct xfs_dquot **dqpp) +{ + struct xfs_dquot *dqp; + struct xfs_buf *bp; + int error; + + dqp = xfs_dquot_alloc(mp, id, type); + trace_xfs_dqread(dqp); + + /* Try to read the buffer, allocating if necessary. */ + error = xfs_dquot_disk_read(mp, dqp, &bp); + if (error == -ENOENT && can_alloc) + error = xfs_qm_dqread_alloc(mp, dqp, &bp); + if (error) + goto err; /* - * We got the buffer with a xfs_trans_read_buf() (in dqtobp()) - * So we need to release with xfs_trans_brelse(). - * The strategy here is identical to that of inodes; we lock - * the dquot in xfs_qm_dqget() before making it accessible to - * others. This is because dquots, like inodes, need a good level of - * concurrency, and we don't want to take locks on the entire buffers - * for dquot accesses. - * Note also that the dquot buffer may even be dirty at this point, if - * this particular dquot was repaired. We still aren't afraid to - * brelse it because we have the changes incore. + * At this point we should have a clean locked buffer. Copy the data + * to the incore dquot and release the buffer since the incore dquot + * has its own locking protocol so we needn't tie up the buffer any + * further. */ ASSERT(xfs_buf_islocked(bp)); - xfs_trans_brelse(tp, bp); + xfs_dquot_from_disk(dqp, bp); - if (tp) { - error = xfs_trans_commit(tp); - if (error) - goto error0; - } - - *O_dqpp = dqp; + xfs_buf_relse(bp); + *dqpp = dqp; return error; -error1: - if (tp) - xfs_trans_cancel(tp); -error0: +err: + trace_xfs_dqread_fail(dqp); xfs_qm_dqdestroy(dqp); - *O_dqpp = NULL; + *dqpp = NULL; return error; } @@ -679,77 +681,230 @@ xfs_dq_get_next_id( } /* - * Given the file system, inode OR id, and type (UDQUOT/GDQUOT), return a - * a locked dquot, doing an allocation (if requested) as needed. - * When both an inode and an id are given, the inode's id takes precedence. - * That is, if the id changes while we don't hold the ilock inside this - * function, the new dquot is returned, not necessarily the one requested - * in the id argument. + * Look up the dquot in the in-core cache. If found, the dquot is returned + * locked and ready to go. + */ +static struct xfs_dquot * +xfs_qm_dqget_cache_lookup( + struct xfs_mount *mp, + struct xfs_quotainfo *qi, + struct radix_tree_root *tree, + xfs_dqid_t id) +{ + struct xfs_dquot *dqp; + +restart: + mutex_lock(&qi->qi_tree_lock); + dqp = radix_tree_lookup(tree, id); + if (!dqp) { + mutex_unlock(&qi->qi_tree_lock); + XFS_STATS_INC(mp, xs_qm_dqcachemisses); + return NULL; + } + + xfs_dqlock(dqp); + if (dqp->dq_flags & XFS_DQ_FREEING) { + xfs_dqunlock(dqp); + mutex_unlock(&qi->qi_tree_lock); + trace_xfs_dqget_freeing(dqp); + delay(1); + goto restart; + } + + dqp->q_nrefs++; + mutex_unlock(&qi->qi_tree_lock); + + trace_xfs_dqget_hit(dqp); + XFS_STATS_INC(mp, xs_qm_dqcachehits); + return dqp; +} + +/* + * Try to insert a new dquot into the in-core cache. If an error occurs the + * caller should throw away the dquot and start over. Otherwise, the dquot + * is returned locked (and held by the cache) as if there had been a cache + * hit. + */ +static int +xfs_qm_dqget_cache_insert( + struct xfs_mount *mp, + struct xfs_quotainfo *qi, + struct radix_tree_root *tree, + xfs_dqid_t id, + struct xfs_dquot *dqp) +{ + int error; + + mutex_lock(&qi->qi_tree_lock); + error = radix_tree_insert(tree, id, dqp); + if (unlikely(error)) { + /* Duplicate found! Caller must try again. */ + WARN_ON(error != -EEXIST); + mutex_unlock(&qi->qi_tree_lock); + trace_xfs_dqget_dup(dqp); + return error; + } + + /* Return a locked dquot to the caller, with a reference taken. */ + xfs_dqlock(dqp); + dqp->q_nrefs = 1; + + qi->qi_dquots++; + mutex_unlock(&qi->qi_tree_lock); + + return 0; +} + +/* Check our input parameters. */ +static int +xfs_qm_dqget_checks( + struct xfs_mount *mp, + uint type) +{ + if (WARN_ON_ONCE(!XFS_IS_QUOTA_RUNNING(mp))) + return -ESRCH; + + switch (type) { + case XFS_DQ_USER: + if (!XFS_IS_UQUOTA_ON(mp)) + return -ESRCH; + return 0; + case XFS_DQ_GROUP: + if (!XFS_IS_GQUOTA_ON(mp)) + return -ESRCH; + return 0; + case XFS_DQ_PROJ: + if (!XFS_IS_PQUOTA_ON(mp)) + return -ESRCH; + return 0; + default: + WARN_ON_ONCE(0); + return -EINVAL; + } +} + +/* + * Given the file system, id, and type (UDQUOT/GDQUOT), return a a locked + * dquot, doing an allocation (if requested) as needed. */ int xfs_qm_dqget( - xfs_mount_t *mp, - xfs_inode_t *ip, /* locked inode (optional) */ - xfs_dqid_t id, /* uid/projid/gid depending on type */ - uint type, /* XFS_DQ_USER/XFS_DQ_PROJ/XFS_DQ_GROUP */ - uint flags, /* DQALLOC, DQSUSER, DQREPAIR, DOWARN */ - xfs_dquot_t **O_dqpp) /* OUT : locked incore dquot */ + struct xfs_mount *mp, + xfs_dqid_t id, + uint type, + bool can_alloc, + struct xfs_dquot **O_dqpp) { struct xfs_quotainfo *qi = mp->m_quotainfo; - struct radix_tree_root *tree = xfs_dquot_tree(qi, type); + struct radix_tree_root *tree = xfs_dquot_tree(qi, type); struct xfs_dquot *dqp; int error; - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) || - (! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) || - (! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) { - return -ESRCH; + error = xfs_qm_dqget_checks(mp, type); + if (error) + return error; + +restart: + dqp = xfs_qm_dqget_cache_lookup(mp, qi, tree, id); + if (dqp) { + *O_dqpp = dqp; + return 0; } - ASSERT(type == XFS_DQ_USER || - type == XFS_DQ_PROJ || - type == XFS_DQ_GROUP); - if (ip) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT(xfs_inode_dquot(ip, type) == NULL); + error = xfs_qm_dqread(mp, id, type, can_alloc, &dqp); + if (error) + return error; + + error = xfs_qm_dqget_cache_insert(mp, qi, tree, id, dqp); + if (error) { + /* + * Duplicate found. Just throw away the new dquot and start + * over. + */ + xfs_qm_dqdestroy(dqp); + XFS_STATS_INC(mp, xs_qm_dquot_dups); + goto restart; } -restart: - mutex_lock(&qi->qi_tree_lock); - dqp = radix_tree_lookup(tree, id); - if (dqp) { - xfs_dqlock(dqp); - if (dqp->dq_flags & XFS_DQ_FREEING) { - xfs_dqunlock(dqp); - mutex_unlock(&qi->qi_tree_lock); - trace_xfs_dqget_freeing(dqp); - delay(1); - goto restart; - } + trace_xfs_dqget_miss(dqp); + *O_dqpp = dqp; + return 0; +} - /* uninit / unused quota found in radix tree, keep looking */ - if (flags & XFS_QMOPT_DQNEXT) { - if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) { - xfs_dqunlock(dqp); - mutex_unlock(&qi->qi_tree_lock); - error = xfs_dq_get_next_id(mp, type, &id); - if (error) - return error; - goto restart; - } - } +/* + * Given a dquot id and type, read and initialize a dquot from the on-disk + * metadata. This function is only for use during quota initialization so + * it ignores the dquot cache assuming that the dquot shrinker isn't set up. + * The caller is responsible for _qm_dqdestroy'ing the returned dquot. + */ +int +xfs_qm_dqget_uncached( + struct xfs_mount *mp, + xfs_dqid_t id, + uint type, + struct xfs_dquot **dqpp) +{ + int error; - dqp->q_nrefs++; - mutex_unlock(&qi->qi_tree_lock); + error = xfs_qm_dqget_checks(mp, type); + if (error) + return error; + + return xfs_qm_dqread(mp, id, type, 0, dqpp); +} + +/* Return the quota id for a given inode and type. */ +xfs_dqid_t +xfs_qm_id_for_quotatype( + struct xfs_inode *ip, + uint type) +{ + switch (type) { + case XFS_DQ_USER: + return ip->i_d.di_uid; + case XFS_DQ_GROUP: + return ip->i_d.di_gid; + case XFS_DQ_PROJ: + return xfs_get_projid(ip); + } + ASSERT(0); + return 0; +} + +/* + * Return the dquot for a given inode and type. If @can_alloc is true, then + * allocate blocks if needed. The inode's ILOCK must be held and it must not + * have already had an inode attached. + */ +int +xfs_qm_dqget_inode( + struct xfs_inode *ip, + uint type, + bool can_alloc, + struct xfs_dquot **O_dqpp) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_quotainfo *qi = mp->m_quotainfo; + struct radix_tree_root *tree = xfs_dquot_tree(qi, type); + struct xfs_dquot *dqp; + xfs_dqid_t id; + int error; - trace_xfs_dqget_hit(dqp); - XFS_STATS_INC(mp, xs_qm_dqcachehits); + error = xfs_qm_dqget_checks(mp, type); + if (error) + return error; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT(xfs_inode_dquot(ip, type) == NULL); + + id = xfs_qm_id_for_quotatype(ip, type); + +restart: + dqp = xfs_qm_dqget_cache_lookup(mp, qi, tree, id); + if (dqp) { *O_dqpp = dqp; return 0; } - mutex_unlock(&qi->qi_tree_lock); - XFS_STATS_INC(mp, xs_qm_dqcachemisses); /* * Dquot cache miss. We don't want to keep the inode lock across @@ -758,87 +913,81 @@ restart: * lock here means dealing with a chown that can happen before * we re-acquire the lock. */ - if (ip) - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - error = xfs_qm_dqread(mp, id, type, flags, &dqp); - - if (ip) - xfs_ilock(ip, XFS_ILOCK_EXCL); - - /* If we are asked to find next active id, keep looking */ - if (error == -ENOENT && (flags & XFS_QMOPT_DQNEXT)) { - error = xfs_dq_get_next_id(mp, type, &id); - if (!error) - goto restart; - } - + xfs_iunlock(ip, XFS_ILOCK_EXCL); + error = xfs_qm_dqread(mp, id, type, can_alloc, &dqp); + xfs_ilock(ip, XFS_ILOCK_EXCL); if (error) return error; - if (ip) { - /* - * A dquot could be attached to this inode by now, since - * we had dropped the ilock. - */ - if (xfs_this_quota_on(mp, type)) { - struct xfs_dquot *dqp1; - - dqp1 = xfs_inode_dquot(ip, type); - if (dqp1) { - xfs_qm_dqdestroy(dqp); - dqp = dqp1; - xfs_dqlock(dqp); - goto dqret; - } - } else { - /* inode stays locked on return */ + /* + * A dquot could be attached to this inode by now, since we had + * dropped the ilock. + */ + if (xfs_this_quota_on(mp, type)) { + struct xfs_dquot *dqp1; + + dqp1 = xfs_inode_dquot(ip, type); + if (dqp1) { xfs_qm_dqdestroy(dqp); - return -ESRCH; + dqp = dqp1; + xfs_dqlock(dqp); + goto dqret; } + } else { + /* inode stays locked on return */ + xfs_qm_dqdestroy(dqp); + return -ESRCH; } - mutex_lock(&qi->qi_tree_lock); - error = radix_tree_insert(tree, id, dqp); - if (unlikely(error)) { - WARN_ON(error != -EEXIST); - + error = xfs_qm_dqget_cache_insert(mp, qi, tree, id, dqp); + if (error) { /* * Duplicate found. Just throw away the new dquot and start * over. */ - mutex_unlock(&qi->qi_tree_lock); - trace_xfs_dqget_dup(dqp); xfs_qm_dqdestroy(dqp); XFS_STATS_INC(mp, xs_qm_dquot_dups); goto restart; } - /* - * We return a locked dquot to the caller, with a reference taken - */ - xfs_dqlock(dqp); - dqp->q_nrefs = 1; +dqret: + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + trace_xfs_dqget_miss(dqp); + *O_dqpp = dqp; + return 0; +} - qi->qi_dquots++; - mutex_unlock(&qi->qi_tree_lock); +/* + * Starting at @id and progressing upwards, look for an initialized incore + * dquot, lock it, and return it. + */ +int +xfs_qm_dqget_next( + struct xfs_mount *mp, + xfs_dqid_t id, + uint type, + struct xfs_dquot **dqpp) +{ + struct xfs_dquot *dqp; + int error = 0; - /* If we are asked to find next active id, keep looking */ - if (flags & XFS_QMOPT_DQNEXT) { - if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) { - xfs_qm_dqput(dqp); - error = xfs_dq_get_next_id(mp, type, &id); - if (error) - return error; - goto restart; + *dqpp = NULL; + for (; !error; error = xfs_dq_get_next_id(mp, type, &id)) { + error = xfs_qm_dqget(mp, id, type, false, &dqp); + if (error == -ENOENT) + continue; + else if (error != 0) + break; + + if (!XFS_IS_DQUOT_UNINITIALIZED(dqp)) { + *dqpp = dqp; + return 0; } + + xfs_qm_dqput(dqp); } - dqret: - ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL)); - trace_xfs_dqget_miss(dqp); - *O_dqpp = dqp; - return 0; + return error; } /* @@ -913,9 +1062,9 @@ xfs_qm_dqflush_done( * since it's cheaper, and then we recheck while * holding the lock before removing the dquot from the AIL. */ - if ((lip->li_flags & XFS_LI_IN_AIL) && + if (test_bit(XFS_LI_IN_AIL, &lip->li_flags) && ((lip->li_lsn == qip->qli_flush_lsn) || - (lip->li_flags & XFS_LI_FAILED))) { + test_bit(XFS_LI_FAILED, &lip->li_flags))) { /* xfs_trans_ail_delete() drops the AIL lock. */ spin_lock(&ailp->ail_lock); @@ -926,8 +1075,7 @@ xfs_qm_dqflush_done( * Clear the failed state since we are about to drop the * flush lock */ - if (lip->li_flags & XFS_LI_FAILED) - xfs_clear_li_failed(lip); + xfs_clear_li_failed(lip); spin_unlock(&ailp->ail_lock); } } @@ -953,6 +1101,7 @@ xfs_qm_dqflush( { struct xfs_mount *mp = dqp->q_mount; struct xfs_buf *bp; + struct xfs_dqblk *dqb; struct xfs_disk_dquot *ddqp; xfs_failaddr_t fa; int error; @@ -996,12 +1145,13 @@ xfs_qm_dqflush( /* * Calculate the location of the dquot inside the buffer. */ - ddqp = bp->b_addr + dqp->q_bufoffset; + dqb = bp->b_addr + dqp->q_bufoffset; + ddqp = &dqb->dd_diskdq; /* - * A simple sanity check in case we got a corrupted dquot.. + * A simple sanity check in case we got a corrupted dquot. */ - fa = xfs_dquot_verify(mp, &dqp->q_core, be32_to_cpu(ddqp->d_id), 0, 0); + fa = xfs_dqblk_verify(mp, dqb, be32_to_cpu(ddqp->d_id), 0); if (fa) { xfs_alert(mp, "corrupt dquot ID 0x%x in memory at %pS", be32_to_cpu(ddqp->d_id), fa); @@ -1032,8 +1182,6 @@ xfs_qm_dqflush( * of a dquot without an up-to-date CRC getting to disk. */ if (xfs_sb_version_hascrc(&mp->m_sb)) { - struct xfs_dqblk *dqb = (struct xfs_dqblk *)ddqp; - dqb->dd_lsn = cpu_to_be64(dqp->q_logitem.qli_item.li_lsn); xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk), XFS_DQUOT_CRC_OFF); @@ -1119,3 +1267,35 @@ xfs_qm_exit(void) kmem_zone_destroy(xfs_qm_dqtrxzone); kmem_zone_destroy(xfs_qm_dqzone); } + +/* + * Iterate every dquot of a particular type. The caller must ensure that the + * particular quota type is active. iter_fn can return negative error codes, + * or XFS_BTREE_QUERY_RANGE_ABORT to indicate that it wants to stop iterating. + */ +int +xfs_qm_dqiterate( + struct xfs_mount *mp, + uint dqtype, + xfs_qm_dqiterate_fn iter_fn, + void *priv) +{ + struct xfs_dquot *dq; + xfs_dqid_t id = 0; + int error; + + do { + error = xfs_qm_dqget_next(mp, id, dqtype, &dq); + if (error == -ENOENT) + return 0; + if (error) + return error; + + error = iter_fn(dq, dqtype, priv); + id = be32_to_cpu(dq->q_core.d_id); + xfs_qm_dqput(dq); + id++; + } while (error == 0 && id != 0); + + return error; +} diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h index 2f536f33cd26..bdd6bd921528 100644 --- a/fs/xfs/xfs_dquot.h +++ b/fs/xfs/xfs_dquot.h @@ -160,8 +160,6 @@ static inline bool xfs_dquot_lowsp(struct xfs_dquot *dqp) #define XFS_QM_ISPDQ(dqp) ((dqp)->dq_flags & XFS_DQ_PROJ) #define XFS_QM_ISGDQ(dqp) ((dqp)->dq_flags & XFS_DQ_GROUP) -extern int xfs_qm_dqread(struct xfs_mount *, xfs_dqid_t, uint, - uint, struct xfs_dquot **); extern void xfs_qm_dqdestroy(xfs_dquot_t *); extern int xfs_qm_dqflush(struct xfs_dquot *, struct xfs_buf **); extern void xfs_qm_dqunpin_wait(xfs_dquot_t *); @@ -169,8 +167,19 @@ extern void xfs_qm_adjust_dqtimers(xfs_mount_t *, xfs_disk_dquot_t *); extern void xfs_qm_adjust_dqlimits(struct xfs_mount *, struct xfs_dquot *); -extern int xfs_qm_dqget(xfs_mount_t *, xfs_inode_t *, - xfs_dqid_t, uint, uint, xfs_dquot_t **); +extern xfs_dqid_t xfs_qm_id_for_quotatype(struct xfs_inode *ip, + uint type); +extern int xfs_qm_dqget(struct xfs_mount *mp, xfs_dqid_t id, + uint type, bool can_alloc, + struct xfs_dquot **dqpp); +extern int xfs_qm_dqget_inode(struct xfs_inode *ip, uint type, + bool can_alloc, + struct xfs_dquot **dqpp); +extern int xfs_qm_dqget_next(struct xfs_mount *mp, xfs_dqid_t id, + uint type, struct xfs_dquot **dqpp); +extern int xfs_qm_dqget_uncached(struct xfs_mount *mp, + xfs_dqid_t id, uint type, + struct xfs_dquot **dqpp); extern void xfs_qm_dqput(xfs_dquot_t *); extern void xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *); @@ -185,4 +194,9 @@ static inline struct xfs_dquot *xfs_qm_dqhold(struct xfs_dquot *dqp) return dqp; } +typedef int (*xfs_qm_dqiterate_fn)(struct xfs_dquot *dq, uint dqtype, + void *priv); +int xfs_qm_dqiterate(struct xfs_mount *mp, uint dqtype, + xfs_qm_dqiterate_fn iter_fn, void *priv); + #endif /* __XFS_DQUOT_H__ */ diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c index 4b331e354da7..8eb7415474d6 100644 --- a/fs/xfs/xfs_dquot_item.c +++ b/fs/xfs/xfs_dquot_item.c @@ -173,7 +173,7 @@ xfs_qm_dquot_logitem_push( * The buffer containing this item failed to be written back * previously. Resubmit the buffer for IO */ - if (lip->li_flags & XFS_LI_FAILED) { + if (test_bit(XFS_LI_FAILED, &lip->li_flags)) { if (!xfs_buf_trylock(bp)) return XFS_ITEM_LOCKED; @@ -209,10 +209,7 @@ xfs_qm_dquot_logitem_push( spin_unlock(&lip->li_ailp->ail_lock); error = xfs_qm_dqflush(dqp, &bp); - if (error) { - xfs_warn(dqp->q_mount, "%s: push error %d on dqp "PTR_FMT, - __func__, error, dqp); - } else { + if (!error) { if (!xfs_buf_delwri_queue(bp, buffer_list)) rval = XFS_ITEM_FLUSHING; xfs_buf_relse(bp); diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index a63f5083f497..7975634cb8fe 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -61,6 +61,7 @@ static unsigned int xfs_errortag_random_default[] = { XFS_RANDOM_LOG_BAD_CRC, XFS_RANDOM_LOG_ITEM_PIN, XFS_RANDOM_BUF_LRU_REF, + XFS_RANDOM_FORCE_SCRUB_REPAIR, }; struct xfs_errortag_attr { @@ -167,6 +168,7 @@ XFS_ERRORTAG_ATTR_RW(drop_writes, XFS_ERRTAG_DROP_WRITES); XFS_ERRORTAG_ATTR_RW(log_bad_crc, XFS_ERRTAG_LOG_BAD_CRC); XFS_ERRORTAG_ATTR_RW(log_item_pin, XFS_ERRTAG_LOG_ITEM_PIN); XFS_ERRORTAG_ATTR_RW(buf_lru_ref, XFS_ERRTAG_BUF_LRU_REF); +XFS_ERRORTAG_ATTR_RW(force_repair, XFS_ERRTAG_FORCE_SCRUB_REPAIR); static struct attribute *xfs_errortag_attrs[] = { XFS_ERRORTAG_ATTR_LIST(noerror), @@ -201,6 +203,7 @@ static struct attribute *xfs_errortag_attrs[] = { XFS_ERRORTAG_ATTR_LIST(log_bad_crc), XFS_ERRORTAG_ATTR_LIST(log_item_pin), XFS_ERRORTAG_ATTR_LIST(buf_lru_ref), + XFS_ERRORTAG_ATTR_LIST(force_repair), NULL, }; diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index 64da90655e95..a889b550979a 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -51,6 +51,24 @@ xfs_efi_item_free( } /* + * Freeing the efi requires that we remove it from the AIL if it has already + * been placed there. However, the EFI may not yet have been placed in the AIL + * when called by xfs_efi_release() from EFD processing due to the ordering of + * committed vs unpin operations in bulk insert operations. Hence the reference + * count to ensure only the last caller frees the EFI. + */ +void +xfs_efi_release( + struct xfs_efi_log_item *efip) +{ + ASSERT(atomic_read(&efip->efi_refcount) > 0); + if (atomic_dec_and_test(&efip->efi_refcount)) { + xfs_trans_ail_remove(&efip->efi_item, SHUTDOWN_LOG_IO_ERROR); + xfs_efi_item_free(efip); + } +} + +/* * This returns the number of iovecs needed to log the given efi item. * We only need 1 iovec for an efi item. It just logs the efi_log_format * structure. @@ -150,8 +168,8 @@ STATIC void xfs_efi_item_unlock( struct xfs_log_item *lip) { - if (lip->li_flags & XFS_LI_ABORTED) - xfs_efi_item_free(EFI_ITEM(lip)); + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) + xfs_efi_release(EFI_ITEM(lip)); } /* @@ -279,24 +297,6 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt) return -EFSCORRUPTED; } -/* - * Freeing the efi requires that we remove it from the AIL if it has already - * been placed there. However, the EFI may not yet have been placed in the AIL - * when called by xfs_efi_release() from EFD processing due to the ordering of - * committed vs unpin operations in bulk insert operations. Hence the reference - * count to ensure only the last caller frees the EFI. - */ -void -xfs_efi_release( - struct xfs_efi_log_item *efip) -{ - ASSERT(atomic_read(&efip->efi_refcount) > 0); - if (atomic_dec_and_test(&efip->efi_refcount)) { - xfs_trans_ail_remove(&efip->efi_item, SHUTDOWN_LOG_IO_ERROR); - xfs_efi_item_free(efip); - } -} - static inline struct xfs_efd_log_item *EFD_ITEM(struct xfs_log_item *lip) { return container_of(lip, struct xfs_efd_log_item, efd_item); @@ -402,7 +402,7 @@ xfs_efd_item_unlock( { struct xfs_efd_log_item *efdp = EFD_ITEM(lip); - if (lip->li_flags & XFS_LI_ABORTED) { + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) { xfs_efi_release(efdp->efd_efip); xfs_efd_item_free(efdp); } @@ -542,7 +542,7 @@ xfs_efi_recover( for (i = 0; i < efip->efi_format.efi_nextents; i++) { extp = &efip->efi_format.efi_extents[i]; error = xfs_trans_free_extent(tp, efdp, extp->ext_start, - extp->ext_len, &oinfo); + extp->ext_len, &oinfo, false); if (error) goto abort_error; diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 299aee4b7b0b..0e3fb8978344 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -414,6 +414,12 @@ xfs_dio_write_end_io( if (size <= 0) return size; + /* + * Capture amount written on completion as we can't reliably account + * for it on submission. + */ + XFS_STATS_ADD(ip->i_mount, xs_write_bytes, size); + if (flags & IOMAP_DIO_COW) { error = xfs_reflink_end_cow(ip, offset, size); if (error) @@ -599,7 +605,16 @@ xfs_file_dax_write( } out: xfs_iunlock(ip, iolock); - return error ? error : ret; + if (error) + return error; + + if (ret > 0) { + XFS_STATS_ADD(ip->i_mount, xs_write_bytes, ret); + + /* Handle various SYNC-type writes */ + ret = generic_write_sync(iocb, ret); + } + return ret; } STATIC ssize_t @@ -669,6 +684,12 @@ write_retry: out: if (iolock) xfs_iunlock(ip, iolock); + + if (ret > 0) { + XFS_STATS_ADD(ip->i_mount, xs_write_bytes, ret); + /* Handle various SYNC-type writes */ + ret = generic_write_sync(iocb, ret); + } return ret; } @@ -693,8 +714,9 @@ xfs_file_write_iter( return -EIO; if (IS_DAX(inode)) - ret = xfs_file_dax_write(iocb, from); - else if (iocb->ki_flags & IOCB_DIRECT) { + return xfs_file_dax_write(iocb, from); + + if (iocb->ki_flags & IOCB_DIRECT) { /* * Allow a directio write to fall back to a buffered * write *only* in the case that we're doing a reflink @@ -702,20 +724,11 @@ xfs_file_write_iter( * allow an operation to fall back to buffered mode. */ ret = xfs_file_dio_aio_write(iocb, from); - if (ret == -EREMCHG) - goto buffered; - } else { -buffered: - ret = xfs_file_buffered_aio_write(iocb, from); + if (ret != -EREMCHG) + return ret; } - if (ret > 0) { - XFS_STATS_ADD(ip->i_mount, xs_write_bytes, ret); - - /* Handle various SYNC-type writes */ - ret = generic_write_sync(iocb, ret); - } - return ret; + return xfs_file_buffered_aio_write(iocb, from); } #define XFS_FALLOC_FL_SUPPORTED \ @@ -778,22 +791,26 @@ xfs_file_fallocate( if (error) goto out_unlock; } else if (mode & FALLOC_FL_INSERT_RANGE) { - unsigned int blksize_mask = i_blocksize(inode) - 1; + unsigned int blksize_mask = i_blocksize(inode) - 1; + loff_t isize = i_size_read(inode); - new_size = i_size_read(inode) + len; if (offset & blksize_mask || len & blksize_mask) { error = -EINVAL; goto out_unlock; } - /* check the new inode size does not wrap through zero */ - if (new_size > inode->i_sb->s_maxbytes) { + /* + * New inode size must not exceed ->s_maxbytes, accounting for + * possible signed overflow. + */ + if (inode->i_sb->s_maxbytes - isize < len) { error = -EFBIG; goto out_unlock; } + new_size = isize + len; /* Offset should be less than i_size */ - if (offset >= i_size_read(inode)) { + if (offset >= isize) { error = -EINVAL; goto out_unlock; } @@ -876,8 +893,18 @@ xfs_file_dedupe_range( struct file *dst_file, u64 dst_loff) { + struct inode *srci = file_inode(src_file); + u64 max_dedupe; int error; + /* + * Since we have to read all these pages in to compare them, cut + * it off at MAX_RW_COUNT/2 rounded down to the nearest block. + * That means we won't do more than MAX_RW_COUNT IO per request. + */ + max_dedupe = (MAX_RW_COUNT >> 1) & ~(i_blocksize(srci) - 1); + if (len > max_dedupe) + len = max_dedupe; error = xfs_reflink_remap_range(src_file, loff, dst_file, dst_loff, len, true); if (error) @@ -993,7 +1020,7 @@ xfs_file_llseek( * page_lock (MM) * i_lock (XFS - extent map serialisation) */ -static int +static vm_fault_t __xfs_filemap_fault( struct vm_fault *vmf, enum page_entry_size pe_size, @@ -1001,7 +1028,7 @@ __xfs_filemap_fault( { struct inode *inode = file_inode(vmf->vma->vm_file); struct xfs_inode *ip = XFS_I(inode); - int ret; + vm_fault_t ret; trace_xfs_filemap_fault(ip, pe_size, write_fault); @@ -1030,7 +1057,7 @@ __xfs_filemap_fault( return ret; } -static int +static vm_fault_t xfs_filemap_fault( struct vm_fault *vmf) { @@ -1040,7 +1067,7 @@ xfs_filemap_fault( (vmf->flags & FAULT_FLAG_WRITE)); } -static int +static vm_fault_t xfs_filemap_huge_fault( struct vm_fault *vmf, enum page_entry_size pe_size) @@ -1053,7 +1080,7 @@ xfs_filemap_huge_fault( (vmf->flags & FAULT_FLAG_WRITE)); } -static int +static vm_fault_t xfs_filemap_page_mkwrite( struct vm_fault *vmf) { @@ -1065,7 +1092,7 @@ xfs_filemap_page_mkwrite( * on write faults. In reality, it needs to serialise against truncate and * prepare memory for writing so handle is as standard write fault. */ -static int +static vm_fault_t xfs_filemap_pfn_mkwrite( struct vm_fault *vmf) { diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 043ca3808ea2..3f8722e51dbe 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -34,7 +34,6 @@ struct xfs_fstrm_item { struct xfs_mru_cache_elem mru; - struct xfs_inode *ip; xfs_agnumber_t ag; /* AG in use for this directory */ }; @@ -122,14 +121,15 @@ xfs_filestream_put_ag( static void xfs_fstrm_free_func( + void *data, struct xfs_mru_cache_elem *mru) { + struct xfs_mount *mp = data; struct xfs_fstrm_item *item = container_of(mru, struct xfs_fstrm_item, mru); - xfs_filestream_put_ag(item->ip->i_mount, item->ag); - - trace_xfs_filestream_free(item->ip, item->ag); + xfs_filestream_put_ag(mp, item->ag); + trace_xfs_filestream_free(mp, mru->key, item->ag); kmem_free(item); } @@ -165,7 +165,7 @@ xfs_filestream_pick_ag( trylock = XFS_ALLOC_FLAG_TRYLOCK; for (nscan = 0; 1; nscan++) { - trace_xfs_filestream_scan(ip, ag); + trace_xfs_filestream_scan(mp, ip->i_ino, ag); pag = xfs_perag_get(mp, ag); @@ -198,7 +198,7 @@ xfs_filestream_pick_ag( goto next_ag; } - longest = xfs_alloc_longest_free_extent(mp, pag, + longest = xfs_alloc_longest_free_extent(pag, xfs_alloc_min_freelist(mp, pag), xfs_ag_resv_needed(pag, XFS_AG_RESV_NONE)); if (((minlen && longest >= minlen) || @@ -265,7 +265,6 @@ next_ag: goto out_put_ag; item->ag = *agp; - item->ip = ip; err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru); if (err) { @@ -333,7 +332,7 @@ xfs_filestream_lookup_ag( ag = container_of(mru, struct xfs_fstrm_item, mru)->ag; xfs_mru_cache_done(mp->m_filestream); - trace_xfs_filestream_lookup(ip, ag); + trace_xfs_filestream_lookup(mp, ip->i_ino, ag); goto out; } @@ -399,7 +398,7 @@ xfs_filestream_new_ag( * Only free the item here so we skip over the old AG earlier. */ if (mru) - xfs_fstrm_free_func(mru); + xfs_fstrm_free_func(mp, mru); IRELE(pip); exit: @@ -426,8 +425,8 @@ xfs_filestream_mount( * timer tunable to within about 10 percent. This requires at least 10 * groups. */ - return xfs_mru_cache_create(&mp->m_filestream, xfs_fstrm_centisecs * 10, - 10, xfs_fstrm_free_func); + return xfs_mru_cache_create(&mp->m_filestream, mp, + xfs_fstrm_centisecs * 10, 10, xfs_fstrm_free_func); } void diff --git a/fs/xfs/xfs_fsmap.c b/fs/xfs/xfs_fsmap.c index 43cfc07996a4..0299febece9c 100644 --- a/fs/xfs/xfs_fsmap.c +++ b/fs/xfs/xfs_fsmap.c @@ -465,10 +465,9 @@ xfs_getfsmap_rtdev_rtbitmap_helper( struct xfs_rmap_irec irec; xfs_daddr_t rec_daddr; - rec_daddr = XFS_FSB_TO_BB(mp, rec->ar_startblock); - - irec.rm_startblock = rec->ar_startblock; - irec.rm_blockcount = rec->ar_blockcount; + irec.rm_startblock = rec->ar_startext * mp->m_sb.sb_rextsize; + rec_daddr = XFS_FSB_TO_BB(mp, irec.rm_startblock); + irec.rm_blockcount = rec->ar_extcount * mp->m_sb.sb_rextsize; irec.rm_owner = XFS_RMAP_OWN_NULL; /* "free" */ irec.rm_offset = 0; irec.rm_flags = 0; @@ -534,8 +533,11 @@ xfs_getfsmap_rtdev_rtbitmap_query( xfs_ilock(tp->t_mountp->m_rbmip, XFS_ILOCK_SHARED); - alow.ar_startblock = info->low.rm_startblock; - ahigh.ar_startblock = info->high.rm_startblock; + alow.ar_startext = info->low.rm_startblock; + ahigh.ar_startext = info->high.rm_startblock; + do_div(alow.ar_startext, tp->t_mountp->m_sb.sb_rextsize); + if (do_div(ahigh.ar_startext, tp->t_mountp->m_sb.sb_rextsize)) + ahigh.ar_startext++; error = xfs_rtalloc_query_range(tp, &alow, &ahigh, xfs_getfsmap_rtdev_rtbitmap_helper, info); if (error) diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 523792768080..bc7ef18da243 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -24,85 +24,42 @@ #include "xfs_sb.h" #include "xfs_mount.h" #include "xfs_defer.h" -#include "xfs_da_format.h" -#include "xfs_da_btree.h" -#include "xfs_inode.h" #include "xfs_trans.h" -#include "xfs_inode_item.h" #include "xfs_error.h" #include "xfs_btree.h" -#include "xfs_alloc_btree.h" #include "xfs_alloc.h" -#include "xfs_rmap_btree.h" -#include "xfs_ialloc.h" #include "xfs_fsops.h" -#include "xfs_itable.h" #include "xfs_trans_space.h" #include "xfs_rtalloc.h" #include "xfs_trace.h" #include "xfs_log.h" -#include "xfs_filestream.h" -#include "xfs_rmap.h" +#include "xfs_ag.h" #include "xfs_ag_resv.h" /* - * File system operations + * growfs operations */ - -static struct xfs_buf * -xfs_growfs_get_hdr_buf( - struct xfs_mount *mp, - xfs_daddr_t blkno, - size_t numblks, - int flags, - const struct xfs_buf_ops *ops) -{ - struct xfs_buf *bp; - - bp = xfs_buf_get_uncached(mp->m_ddev_targp, numblks, flags); - if (!bp) - return NULL; - - xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); - bp->b_bn = blkno; - bp->b_maps[0].bm_bn = blkno; - bp->b_ops = ops; - - return bp; -} - static int xfs_growfs_data_private( xfs_mount_t *mp, /* mount point for filesystem */ xfs_growfs_data_t *in) /* growfs data input struct */ { - xfs_agf_t *agf; - struct xfs_agfl *agfl; - xfs_agi_t *agi; - xfs_agnumber_t agno; - xfs_extlen_t agsize; - xfs_extlen_t tmpsize; - xfs_alloc_rec_t *arec; xfs_buf_t *bp; - int bucket; - int dpct; - int error, saved_error = 0; + int error; xfs_agnumber_t nagcount; xfs_agnumber_t nagimax = 0; xfs_rfsblock_t nb, nb_mod; xfs_rfsblock_t new; - xfs_rfsblock_t nfree; xfs_agnumber_t oagcount; - int pct; xfs_trans_t *tp; + LIST_HEAD (buffer_list); + struct aghdr_init_data id = {}; nb = in->newblocks; - pct = in->imaxpct; - if (nb < mp->m_sb.sb_dblocks || pct < 0 || pct > 100) + if (nb < mp->m_sb.sb_dblocks) return -EINVAL; if ((error = xfs_sb_validate_fsb_count(&mp->m_sb, nb))) return error; - dpct = pct - mp->m_sb.sb_imax_pct; error = xfs_buf_read_uncached(mp->m_ddev_targp, XFS_FSB_TO_BB(mp, nb) - XFS_FSS_TO_BB(mp, 1), XFS_FSS_TO_BB(mp, 1), 0, &bp, NULL); @@ -135,376 +92,45 @@ xfs_growfs_data_private( return error; /* - * Write new AG headers to disk. Non-transactional, but written - * synchronously so they are completed prior to the growfs transaction - * being logged. + * Write new AG headers to disk. Non-transactional, but need to be + * written and completed prior to the growfs transaction being logged. + * To do this, we use a delayed write buffer list and wait for + * submission and IO completion of the list as a whole. This allows the + * IO subsystem to merge all the AG headers in a single AG into a single + * IO and hide most of the latency of the IO from us. + * + * This also means that if we get an error whilst building the buffer + * list to write, we can cancel the entire list without having written + * anything. */ - nfree = 0; - for (agno = nagcount - 1; agno >= oagcount; agno--, new -= agsize) { - __be32 *agfl_bno; - - /* - * AG freespace header block - */ - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)), - XFS_FSS_TO_BB(mp, 1), 0, - &xfs_agf_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - agf = XFS_BUF_TO_AGF(bp); - agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC); - agf->agf_versionnum = cpu_to_be32(XFS_AGF_VERSION); - agf->agf_seqno = cpu_to_be32(agno); - if (agno == nagcount - 1) - agsize = - nb - - (agno * (xfs_rfsblock_t)mp->m_sb.sb_agblocks); + INIT_LIST_HEAD(&id.buffer_list); + for (id.agno = nagcount - 1; + id.agno >= oagcount; + id.agno--, new -= id.agsize) { + + if (id.agno == nagcount - 1) + id.agsize = nb - + (id.agno * (xfs_rfsblock_t)mp->m_sb.sb_agblocks); else - agsize = mp->m_sb.sb_agblocks; - agf->agf_length = cpu_to_be32(agsize); - agf->agf_roots[XFS_BTNUM_BNOi] = cpu_to_be32(XFS_BNO_BLOCK(mp)); - agf->agf_roots[XFS_BTNUM_CNTi] = cpu_to_be32(XFS_CNT_BLOCK(mp)); - agf->agf_levels[XFS_BTNUM_BNOi] = cpu_to_be32(1); - agf->agf_levels[XFS_BTNUM_CNTi] = cpu_to_be32(1); - if (xfs_sb_version_hasrmapbt(&mp->m_sb)) { - agf->agf_roots[XFS_BTNUM_RMAPi] = - cpu_to_be32(XFS_RMAP_BLOCK(mp)); - agf->agf_levels[XFS_BTNUM_RMAPi] = cpu_to_be32(1); - agf->agf_rmap_blocks = cpu_to_be32(1); - } - - agf->agf_flfirst = cpu_to_be32(1); - agf->agf_fllast = 0; - agf->agf_flcount = 0; - tmpsize = agsize - mp->m_ag_prealloc_blocks; - agf->agf_freeblks = cpu_to_be32(tmpsize); - agf->agf_longest = cpu_to_be32(tmpsize); - if (xfs_sb_version_hascrc(&mp->m_sb)) - uuid_copy(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid); - if (xfs_sb_version_hasreflink(&mp->m_sb)) { - agf->agf_refcount_root = cpu_to_be32( - xfs_refc_block(mp)); - agf->agf_refcount_level = cpu_to_be32(1); - agf->agf_refcount_blocks = cpu_to_be32(1); - } - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - - /* - * AG freelist header block - */ - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AG_DADDR(mp, agno, XFS_AGFL_DADDR(mp)), - XFS_FSS_TO_BB(mp, 1), 0, - &xfs_agfl_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - agfl = XFS_BUF_TO_AGFL(bp); - if (xfs_sb_version_hascrc(&mp->m_sb)) { - agfl->agfl_magicnum = cpu_to_be32(XFS_AGFL_MAGIC); - agfl->agfl_seqno = cpu_to_be32(agno); - uuid_copy(&agfl->agfl_uuid, &mp->m_sb.sb_meta_uuid); - } - - agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, bp); - for (bucket = 0; bucket < xfs_agfl_size(mp); bucket++) - agfl_bno[bucket] = cpu_to_be32(NULLAGBLOCK); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - - /* - * AG inode header block - */ - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)), - XFS_FSS_TO_BB(mp, 1), 0, - &xfs_agi_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - agi = XFS_BUF_TO_AGI(bp); - agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC); - agi->agi_versionnum = cpu_to_be32(XFS_AGI_VERSION); - agi->agi_seqno = cpu_to_be32(agno); - agi->agi_length = cpu_to_be32(agsize); - agi->agi_count = 0; - agi->agi_root = cpu_to_be32(XFS_IBT_BLOCK(mp)); - agi->agi_level = cpu_to_be32(1); - agi->agi_freecount = 0; - agi->agi_newino = cpu_to_be32(NULLAGINO); - agi->agi_dirino = cpu_to_be32(NULLAGINO); - if (xfs_sb_version_hascrc(&mp->m_sb)) - uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_meta_uuid); - if (xfs_sb_version_hasfinobt(&mp->m_sb)) { - agi->agi_free_root = cpu_to_be32(XFS_FIBT_BLOCK(mp)); - agi->agi_free_level = cpu_to_be32(1); - } - for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) - agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; + id.agsize = mp->m_sb.sb_agblocks; - /* - * BNO btree root block - */ - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)), - BTOBB(mp->m_sb.sb_blocksize), 0, - &xfs_allocbt_buf_ops); - - if (!bp) { - error = -ENOMEM; - goto error0; - } - - xfs_btree_init_block(mp, bp, XFS_BTNUM_BNO, 0, 1, agno, 0); - - arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); - arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); - arec->ar_blockcount = cpu_to_be32( - agsize - be32_to_cpu(arec->ar_startblock)); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - - /* - * CNT btree root block - */ - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)), - BTOBB(mp->m_sb.sb_blocksize), 0, - &xfs_allocbt_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - xfs_btree_init_block(mp, bp, XFS_BTNUM_CNT, 0, 1, agno, 0); - - arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); - arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); - arec->ar_blockcount = cpu_to_be32( - agsize - be32_to_cpu(arec->ar_startblock)); - nfree += be32_to_cpu(arec->ar_blockcount); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - - /* RMAP btree root block */ - if (xfs_sb_version_hasrmapbt(&mp->m_sb)) { - struct xfs_rmap_rec *rrec; - struct xfs_btree_block *block; - - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AGB_TO_DADDR(mp, agno, XFS_RMAP_BLOCK(mp)), - BTOBB(mp->m_sb.sb_blocksize), 0, - &xfs_rmapbt_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - xfs_btree_init_block(mp, bp, XFS_BTNUM_RMAP, 0, 0, - agno, 0); - block = XFS_BUF_TO_BLOCK(bp); - - - /* - * mark the AG header regions as static metadata The BNO - * btree block is the first block after the headers, so - * it's location defines the size of region the static - * metadata consumes. - * - * Note: unlike mkfs, we never have to account for log - * space when growing the data regions - */ - rrec = XFS_RMAP_REC_ADDR(block, 1); - rrec->rm_startblock = 0; - rrec->rm_blockcount = cpu_to_be32(XFS_BNO_BLOCK(mp)); - rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_FS); - rrec->rm_offset = 0; - be16_add_cpu(&block->bb_numrecs, 1); - - /* account freespace btree root blocks */ - rrec = XFS_RMAP_REC_ADDR(block, 2); - rrec->rm_startblock = cpu_to_be32(XFS_BNO_BLOCK(mp)); - rrec->rm_blockcount = cpu_to_be32(2); - rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG); - rrec->rm_offset = 0; - be16_add_cpu(&block->bb_numrecs, 1); - - /* account inode btree root blocks */ - rrec = XFS_RMAP_REC_ADDR(block, 3); - rrec->rm_startblock = cpu_to_be32(XFS_IBT_BLOCK(mp)); - rrec->rm_blockcount = cpu_to_be32(XFS_RMAP_BLOCK(mp) - - XFS_IBT_BLOCK(mp)); - rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_INOBT); - rrec->rm_offset = 0; - be16_add_cpu(&block->bb_numrecs, 1); - - /* account for rmap btree root */ - rrec = XFS_RMAP_REC_ADDR(block, 4); - rrec->rm_startblock = cpu_to_be32(XFS_RMAP_BLOCK(mp)); - rrec->rm_blockcount = cpu_to_be32(1); - rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG); - rrec->rm_offset = 0; - be16_add_cpu(&block->bb_numrecs, 1); - - /* account for refc btree root */ - if (xfs_sb_version_hasreflink(&mp->m_sb)) { - rrec = XFS_RMAP_REC_ADDR(block, 5); - rrec->rm_startblock = cpu_to_be32( - xfs_refc_block(mp)); - rrec->rm_blockcount = cpu_to_be32(1); - rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_REFC); - rrec->rm_offset = 0; - be16_add_cpu(&block->bb_numrecs, 1); - } - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - } - - /* - * INO btree root block - */ - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)), - BTOBB(mp->m_sb.sb_blocksize), 0, - &xfs_inobt_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - xfs_btree_init_block(mp, bp, XFS_BTNUM_INO , 0, 0, agno, 0); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - - /* - * FINO btree root block - */ - if (xfs_sb_version_hasfinobt(&mp->m_sb)) { - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AGB_TO_DADDR(mp, agno, XFS_FIBT_BLOCK(mp)), - BTOBB(mp->m_sb.sb_blocksize), 0, - &xfs_inobt_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - xfs_btree_init_block(mp, bp, XFS_BTNUM_FINO, - 0, 0, agno, 0); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - } - - /* - * refcount btree root block - */ - if (xfs_sb_version_hasreflink(&mp->m_sb)) { - bp = xfs_growfs_get_hdr_buf(mp, - XFS_AGB_TO_DADDR(mp, agno, xfs_refc_block(mp)), - BTOBB(mp->m_sb.sb_blocksize), 0, - &xfs_refcountbt_buf_ops); - if (!bp) { - error = -ENOMEM; - goto error0; - } - - xfs_btree_init_block(mp, bp, XFS_BTNUM_REFC, - 0, 0, agno, 0); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) - goto error0; - } - } - xfs_trans_agblocks_delta(tp, nfree); - /* - * There are new blocks in the old last a.g. - */ - if (new) { - struct xfs_owner_info oinfo; - - /* - * Change the agi length. - */ - error = xfs_ialloc_read_agi(mp, tp, agno, &bp); - if (error) { - goto error0; - } - ASSERT(bp); - agi = XFS_BUF_TO_AGI(bp); - be32_add_cpu(&agi->agi_length, new); - ASSERT(nagcount == oagcount || - be32_to_cpu(agi->agi_length) == mp->m_sb.sb_agblocks); - xfs_ialloc_log_agi(tp, bp, XFS_AGI_LENGTH); - /* - * Change agf length. - */ - error = xfs_alloc_read_agf(mp, tp, agno, 0, &bp); + error = xfs_ag_init_headers(mp, &id); if (error) { - goto error0; + xfs_buf_delwri_cancel(&id.buffer_list); + goto out_trans_cancel; } - ASSERT(bp); - agf = XFS_BUF_TO_AGF(bp); - be32_add_cpu(&agf->agf_length, new); - ASSERT(be32_to_cpu(agf->agf_length) == - be32_to_cpu(agi->agi_length)); + } + error = xfs_buf_delwri_submit(&id.buffer_list); + if (error) + goto out_trans_cancel; - xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH); + xfs_trans_agblocks_delta(tp, id.nfree); - /* - * Free the new space. - * - * XFS_RMAP_OWN_NULL is used here to tell the rmap btree that - * this doesn't actually exist in the rmap btree. - */ - xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_NULL); - error = xfs_rmap_free(tp, bp, agno, - be32_to_cpu(agf->agf_length) - new, - new, &oinfo); - if (error) - goto error0; - error = xfs_free_extent(tp, - XFS_AGB_TO_FSB(mp, agno, - be32_to_cpu(agf->agf_length) - new), - new, &oinfo, XFS_AG_RESV_NONE); + /* If there are new blocks in the old last AG, extend it. */ + if (new) { + error = xfs_ag_extend_space(mp, tp, &id, new); if (error) - goto error0; + goto out_trans_cancel; } /* @@ -517,10 +143,8 @@ xfs_growfs_data_private( if (nb > mp->m_sb.sb_dblocks) xfs_trans_mod_sb(tp, XFS_TRANS_SB_DBLOCKS, nb - mp->m_sb.sb_dblocks); - if (nfree) - xfs_trans_mod_sb(tp, XFS_TRANS_SB_FDBLOCKS, nfree); - if (dpct) - xfs_trans_mod_sb(tp, XFS_TRANS_SB_IMAXPCT, dpct); + if (id.nfree) + xfs_trans_mod_sb(tp, XFS_TRANS_SB_FDBLOCKS, id.nfree); xfs_trans_set_sync(tp); error = xfs_trans_commit(tp); if (error) @@ -529,12 +153,6 @@ xfs_growfs_data_private( /* New allocation groups fully initialized, so update mount struct */ if (nagimax) mp->m_maxagi = nagimax; - if (mp->m_sb.sb_imax_pct) { - uint64_t icount = mp->m_sb.sb_dblocks * mp->m_sb.sb_imax_pct; - do_div(icount, 100); - mp->m_maxicount = icount << mp->m_sb.sb_inopblog; - } else - mp->m_maxicount = 0; xfs_set_low_space_thresholds(mp); mp->m_alloc_set_aside = xfs_alloc_set_aside(mp); @@ -545,73 +163,24 @@ xfs_growfs_data_private( if (new) { struct xfs_perag *pag; - pag = xfs_perag_get(mp, agno); + pag = xfs_perag_get(mp, id.agno); error = xfs_ag_resv_free(pag); xfs_perag_put(pag); if (error) - goto out; + return error; } - /* Reserve AG metadata blocks. */ + /* + * Reserve AG metadata blocks. ENOSPC here does not mean there was a + * growfs failure, just that there still isn't space for new user data + * after the grow has been run. + */ error = xfs_fs_reserve_ag_blocks(mp); - if (error && error != -ENOSPC) - goto out; - - /* update secondary superblocks. */ - for (agno = 1; agno < nagcount; agno++) { + if (error == -ENOSPC) error = 0; - /* - * new secondary superblocks need to be zeroed, not read from - * disk as the contents of the new area we are growing into is - * completely unknown. - */ - if (agno < oagcount) { - error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, - XFS_AGB_TO_DADDR(mp, agno, XFS_SB_BLOCK(mp)), - XFS_FSS_TO_BB(mp, 1), 0, &bp, - &xfs_sb_buf_ops); - } else { - bp = xfs_trans_get_buf(NULL, mp->m_ddev_targp, - XFS_AGB_TO_DADDR(mp, agno, XFS_SB_BLOCK(mp)), - XFS_FSS_TO_BB(mp, 1), 0); - if (bp) { - bp->b_ops = &xfs_sb_buf_ops; - xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); - } else - error = -ENOMEM; - } - - /* - * If we get an error reading or writing alternate superblocks, - * continue. xfs_repair chooses the "best" superblock based - * on most matches; if we break early, we'll leave more - * superblocks un-updated than updated, and xfs_repair may - * pick them over the properly-updated primary. - */ - if (error) { - xfs_warn(mp, - "error %d reading secondary superblock for ag %d", - error, agno); - saved_error = error; - continue; - } - xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb); - - error = xfs_bwrite(bp); - xfs_buf_relse(bp); - if (error) { - xfs_warn(mp, - "write error %d updating secondary superblock for ag %d", - error, agno); - saved_error = error; - continue; - } - } - - out: - return saved_error ? saved_error : error; + return error; - error0: +out_trans_cancel: xfs_trans_cancel(tp); return error; } @@ -638,25 +207,71 @@ xfs_growfs_log_private( return -ENOSYS; } +static int +xfs_growfs_imaxpct( + struct xfs_mount *mp, + __u32 imaxpct) +{ + struct xfs_trans *tp; + int dpct; + int error; + + if (imaxpct > 100) + return -EINVAL; + + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_growdata, + XFS_GROWFS_SPACE_RES(mp), 0, XFS_TRANS_RESERVE, &tp); + if (error) + return error; + + dpct = imaxpct - mp->m_sb.sb_imax_pct; + xfs_trans_mod_sb(tp, XFS_TRANS_SB_IMAXPCT, dpct); + xfs_trans_set_sync(tp); + return xfs_trans_commit(tp); +} + /* * protected versions of growfs function acquire and release locks on the mount * point - exported through ioctls: XFS_IOC_FSGROWFSDATA, XFS_IOC_FSGROWFSLOG, * XFS_IOC_FSGROWFSRT */ - - int xfs_growfs_data( - xfs_mount_t *mp, - xfs_growfs_data_t *in) + struct xfs_mount *mp, + struct xfs_growfs_data *in) { - int error; + int error = 0; if (!capable(CAP_SYS_ADMIN)) return -EPERM; if (!mutex_trylock(&mp->m_growlock)) return -EWOULDBLOCK; - error = xfs_growfs_data_private(mp, in); + + /* update imaxpct separately to the physical grow of the filesystem */ + if (in->imaxpct != mp->m_sb.sb_imax_pct) { + error = xfs_growfs_imaxpct(mp, in->imaxpct); + if (error) + goto out_error; + } + + if (in->newblocks != mp->m_sb.sb_dblocks) { + error = xfs_growfs_data_private(mp, in); + if (error) + goto out_error; + } + + /* Post growfs calculations needed to reflect new state in operations */ + if (mp->m_sb.sb_imax_pct) { + uint64_t icount = mp->m_sb.sb_dblocks * mp->m_sb.sb_imax_pct; + do_div(icount, 100); + mp->m_maxicount = icount << mp->m_sb.sb_inopblog; + } else + mp->m_maxicount = 0; + + /* Update secondary superblocks now the physical grow has completed */ + error = xfs_update_secondary_sbs(mp); + +out_error: /* * Increment the generation unconditionally, the error could be from * updating the secondary superblocks, in which case the new size diff --git a/fs/xfs/xfs_globals.c b/fs/xfs/xfs_globals.c index 3e1cc3001bcb..fdde17a2333c 100644 --- a/fs/xfs/xfs_globals.c +++ b/fs/xfs/xfs_globals.c @@ -47,6 +47,7 @@ xfs_param_t xfs_params = { struct xfs_globals xfs_globals = { .log_recovery_delay = 0, /* no delay by default */ + .mount_delay = 0, /* no delay by default */ #ifdef XFS_ASSERT_FATAL .bug_on_assert = true, /* assert failures BUG() */ #else diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 9a18f69f6e96..164350d91efc 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -107,7 +107,8 @@ xfs_inode_free_callback( xfs_idestroy_fork(ip, XFS_COW_FORK); if (ip->i_itemp) { - ASSERT(!(ip->i_itemp->ili_item.li_flags & XFS_LI_IN_AIL)); + ASSERT(!test_bit(XFS_LI_IN_AIL, + &ip->i_itemp->ili_item.li_flags)); xfs_inode_item_destroy(ip); ip->i_itemp = NULL; } @@ -309,6 +310,46 @@ xfs_reinit_inode( } /* + * If we are allocating a new inode, then check what was returned is + * actually a free, empty inode. If we are not allocating an inode, + * then check we didn't find a free inode. + * + * Returns: + * 0 if the inode free state matches the lookup context + * -ENOENT if the inode is free and we are not allocating + * -EFSCORRUPTED if there is any state mismatch at all + */ +static int +xfs_iget_check_free_state( + struct xfs_inode *ip, + int flags) +{ + if (flags & XFS_IGET_CREATE) { + /* should be a free inode */ + if (VFS_I(ip)->i_mode != 0) { + xfs_warn(ip->i_mount, +"Corruption detected! Free inode 0x%llx not marked free! (mode 0x%x)", + ip->i_ino, VFS_I(ip)->i_mode); + return -EFSCORRUPTED; + } + + if (ip->i_d.di_nblocks != 0) { + xfs_warn(ip->i_mount, +"Corruption detected! Free inode 0x%llx has blocks allocated!", + ip->i_ino); + return -EFSCORRUPTED; + } + return 0; + } + + /* should be an allocated inode */ + if (VFS_I(ip)->i_mode == 0) + return -ENOENT; + + return 0; +} + +/* * Check the validity of the inode we just found it the cache */ static int @@ -357,12 +398,12 @@ xfs_iget_cache_hit( } /* - * If lookup is racing with unlink return an error immediately. + * Check the inode free state is valid. This also detects lookup + * racing with unlinks. */ - if (VFS_I(ip)->i_mode == 0 && !(flags & XFS_IGET_CREATE)) { - error = -ENOENT; + error = xfs_iget_check_free_state(ip, flags); + if (error) goto out_error; - } /* * If IRECLAIMABLE is set, we've torn down the VFS inode already. @@ -485,29 +526,12 @@ xfs_iget_cache_miss( /* - * If we are allocating a new inode, then check what was returned is - * actually a free, empty inode. If we are not allocating an inode, - * the check we didn't find a free inode. + * Check the inode free state is valid. This also detects lookup + * racing with unlinks. */ - if (flags & XFS_IGET_CREATE) { - if (VFS_I(ip)->i_mode != 0) { - xfs_warn(mp, -"Corruption detected! Free inode 0x%llx not marked free on disk", - ino); - error = -EFSCORRUPTED; - goto out_destroy; - } - if (ip->i_d.di_nblocks != 0) { - xfs_warn(mp, -"Corruption detected! Free inode 0x%llx has blocks allocated!", - ino); - error = -EFSCORRUPTED; - goto out_destroy; - } - } else if (VFS_I(ip)->i_mode == 0) { - error = -ENOENT; + error = xfs_iget_check_free_state(ip, flags); + if (error) goto out_destroy; - } /* * Preload the radix tree so we can insert safely under the @@ -1802,3 +1826,21 @@ xfs_inode_clear_cowblocks_tag( return __xfs_inode_clear_blocks_tag(ip, trace_xfs_perag_clear_cowblocks, XFS_ICI_COWBLOCKS_TAG); } + +/* Disable post-EOF and CoW block auto-reclamation. */ +void +xfs_icache_disable_reclaim( + struct xfs_mount *mp) +{ + cancel_delayed_work_sync(&mp->m_eofblocks_work); + cancel_delayed_work_sync(&mp->m_cowblocks_work); +} + +/* Enable post-EOF and CoW block auto-reclamation. */ +void +xfs_icache_enable_reclaim( + struct xfs_mount *mp) +{ + xfs_queue_eofblocks(mp); + xfs_queue_cowblocks(mp); +} diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h index d4a77588eca1..d69a0f5a6a73 100644 --- a/fs/xfs/xfs_icache.h +++ b/fs/xfs/xfs_icache.h @@ -131,4 +131,7 @@ xfs_fs_eofblocks_from_user( int xfs_icache_inode_is_allocated(struct xfs_mount *mp, struct xfs_trans *tp, xfs_ino_t ino, bool *inuse); +void xfs_icache_disable_reclaim(struct xfs_mount *mp); +void xfs_icache_enable_reclaim(struct xfs_mount *mp); + #endif diff --git a/fs/xfs/xfs_icreate_item.c b/fs/xfs/xfs_icreate_item.c index 865ad1373e5e..5da9599156ed 100644 --- a/fs/xfs/xfs_icreate_item.c +++ b/fs/xfs/xfs_icreate_item.c @@ -91,7 +91,7 @@ xfs_icreate_item_unlock( { struct xfs_icreate_item *icp = ICR_ITEM(lip); - if (icp->ic_item.li_flags & XFS_LI_ABORTED) + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) kmem_zone_free(xfs_icreate_zone, icp); return; } @@ -184,5 +184,5 @@ xfs_icreate_log( xfs_trans_add_item(tp, &icp->ic_item); tp->t_flags |= XFS_TRANS_DIRTY; - icp->ic_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &icp->ic_item.li_flags); } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 3e3aab3888fa..05207a64dd53 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -498,7 +498,7 @@ again: if (!try_lock) { for (j = (i - 1); j >= 0 && !try_lock; j--) { lp = (xfs_log_item_t *)ips[j]->i_itemp; - if (lp && (lp->li_flags & XFS_LI_IN_AIL)) + if (lp && test_bit(XFS_LI_IN_AIL, &lp->li_flags)) try_lock++; } } @@ -598,7 +598,7 @@ xfs_lock_two_inodes( * and try again. */ lp = (xfs_log_item_t *)ip0->i_itemp; - if (lp && (lp->li_flags & XFS_LI_IN_AIL)) { + if (lp && test_bit(XFS_LI_IN_AIL, &lp->li_flags)) { if (!xfs_ilock_nowait(ip1, xfs_lock_inumorder(ip1_mode, 1))) { xfs_iunlock(ip0, ip0_mode); if ((++attempts % 5) == 0) @@ -791,6 +791,18 @@ xfs_ialloc( ASSERT(*ialloc_context == NULL); /* + * Protect against obviously corrupt allocation btree records. Later + * xfs_iget checks will catch re-allocation of other active in-memory + * and on-disk inodes. If we don't catch reallocating the parent inode + * here we will deadlock in xfs_iget() so we have to do these checks + * first. + */ + if ((pip && ino == pip->i_ino) || !xfs_verify_dir_ino(mp, ino)) { + xfs_alert(mp, "Allocated a known in-use inode 0x%llx!", ino); + return -EFSCORRUPTED; + } + + /* * Get the in-core inode with the lock held exclusively. * This is because we're setting fields here we need * to prevent others from looking at until we're done. @@ -972,10 +984,8 @@ xfs_dir_ialloc( xfs_nlink_t nlink, dev_t rdev, prid_t prid, /* project id */ - xfs_inode_t **ipp, /* pointer to inode; it will be + xfs_inode_t **ipp) /* pointer to inode; it will be locked. */ - int *committed) - { xfs_trans_t *tp; xfs_inode_t *ip; @@ -1050,8 +1060,6 @@ xfs_dir_ialloc( } code = xfs_trans_roll(&tp); - if (committed != NULL) - *committed = 1; /* * Re-attach the quota info that we detached from prev trx. @@ -1088,9 +1096,6 @@ xfs_dir_ialloc( } ASSERT(!ialloc_context && ip); - } else { - if (committed != NULL) - *committed = 0; } *ipp = ip; @@ -1203,6 +1208,7 @@ xfs_create( unlock_dp_on_error = true; xfs_defer_init(&dfops, &first_block); + tp->t_agfl_dfops = &dfops; /* * Reserve disk quota and the inode. @@ -1217,8 +1223,7 @@ xfs_create( * entry pointing to them, but a directory also the "." entry * pointing to itself. */ - error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, prid, &ip, - NULL); + error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, prid, &ip); if (error) goto out_trans_cancel; @@ -1309,7 +1314,6 @@ xfs_create( int xfs_create_tmpfile( struct xfs_inode *dp, - struct dentry *dentry, umode_t mode, struct xfs_inode **ipp) { @@ -1351,7 +1355,7 @@ xfs_create_tmpfile( if (error) goto out_trans_cancel; - error = xfs_dir_ialloc(&tp, dp, mode, 1, 0, prid, &ip, NULL); + error = xfs_dir_ialloc(&tp, dp, mode, 1, 0, prid, &ip); if (error) goto out_trans_cancel; @@ -1420,11 +1424,11 @@ xfs_link( if (XFS_FORCED_SHUTDOWN(mp)) return -EIO; - error = xfs_qm_dqattach(sip, 0); + error = xfs_qm_dqattach(sip); if (error) goto std_return; - error = xfs_qm_dqattach(tdp, 0); + error = xfs_qm_dqattach(tdp); if (error) goto std_return; @@ -1460,6 +1464,7 @@ xfs_link( } xfs_defer_init(&dfops, &first_block); + tp->t_agfl_dfops = &dfops; /* * Handle initial link state of O_TMPFILE inode @@ -1543,11 +1548,12 @@ xfs_itruncate_clear_reflink_flags( * dirty on error so that transactions can be easily aborted if possible. */ int -xfs_itruncate_extents( +xfs_itruncate_extents_flags( struct xfs_trans **tpp, struct xfs_inode *ip, int whichfork, - xfs_fsize_t new_size) + xfs_fsize_t new_size, + int flags) { struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp = *tpp; @@ -1570,6 +1576,8 @@ xfs_itruncate_extents( trace_xfs_itruncate_extents_start(ip, new_size); + flags |= xfs_bmapi_aflag(whichfork); + /* * Since it is possible for space to become allocated beyond * the end of the file (in a crash where the space is allocated @@ -1588,12 +1596,9 @@ xfs_itruncate_extents( unmap_len = last_block - first_unmap_block + 1; while (!done) { xfs_defer_init(&dfops, &first_block); - error = xfs_bunmapi(tp, ip, - first_unmap_block, unmap_len, - xfs_bmapi_aflag(whichfork), - XFS_ITRUNC_MAX_EXTENTS, - &first_block, &dfops, - &done); + error = xfs_bunmapi(tp, ip, first_unmap_block, unmap_len, flags, + XFS_ITRUNC_MAX_EXTENTS, &first_block, + &dfops, &done); if (error) goto out_bmap_cancel; @@ -1611,13 +1616,15 @@ xfs_itruncate_extents( goto out; } - /* Remove all pending CoW reservations. */ - error = xfs_reflink_cancel_cow_blocks(ip, &tp, first_unmap_block, - last_block, true); - if (error) - goto out; + if (whichfork == XFS_DATA_FORK) { + /* Remove all pending CoW reservations. */ + error = xfs_reflink_cancel_cow_blocks(ip, &tp, + first_unmap_block, last_block, true); + if (error) + goto out; - xfs_itruncate_clear_reflink_flags(ip); + xfs_itruncate_clear_reflink_flags(ip); + } /* * Always re-log the inode so that our permanent transaction can keep @@ -1818,6 +1825,7 @@ xfs_inactive_ifree( xfs_trans_ijoin(tp, ip, 0); xfs_defer_init(&dfops, &first_block); + tp->t_agfl_dfops = &dfops; error = xfs_ifree(tp, ip, &dfops); if (error) { /* @@ -1918,7 +1926,7 @@ xfs_inactive( ip->i_d.di_nextents > 0 || ip->i_delayed_blks > 0)) truncate = 1; - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return; @@ -2581,11 +2589,11 @@ xfs_remove( if (XFS_FORCED_SHUTDOWN(mp)) return -EIO; - error = xfs_qm_dqattach(dp, 0); + error = xfs_qm_dqattach(dp); if (error) goto std_return; - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) goto std_return; @@ -2654,6 +2662,7 @@ xfs_remove( goto out_trans_cancel; xfs_defer_init(&dfops, &first_block); + tp->t_agfl_dfops = &dfops; error = xfs_dir_removename(tp, dp, name, ip->i_ino, &first_block, &dfops, resblks); if (error) { @@ -2903,7 +2912,7 @@ xfs_rename_alloc_whiteout( struct xfs_inode *tmpfile; int error; - error = xfs_create_tmpfile(dp, NULL, S_IFCHR | WHITEOUT_MODE, &tmpfile); + error = xfs_create_tmpfile(dp, S_IFCHR | WHITEOUT_MODE, &tmpfile); if (error) return error; @@ -3021,6 +3030,7 @@ xfs_rename( } xfs_defer_init(&dfops, &first_block); + tp->t_agfl_dfops = &dfops; /* RENAME_EXCHANGE is unique from here on. */ if (flags & RENAME_EXCHANGE) diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 132d8aa2afc4..00fee6824745 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -393,8 +393,8 @@ int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode **ipp, struct xfs_name *ci_name); int xfs_create(struct xfs_inode *dp, struct xfs_name *name, umode_t mode, dev_t rdev, struct xfs_inode **ipp); -int xfs_create_tmpfile(struct xfs_inode *dp, struct dentry *dentry, - umode_t mode, struct xfs_inode **ipp); +int xfs_create_tmpfile(struct xfs_inode *dp, umode_t mode, + struct xfs_inode **ipp); int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode *ip); int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, @@ -415,8 +415,8 @@ uint xfs_ilock_attr_map_shared(struct xfs_inode *); uint xfs_ip2xflags(struct xfs_inode *); int xfs_ifree(struct xfs_trans *, xfs_inode_t *, struct xfs_defer_ops *); -int xfs_itruncate_extents(struct xfs_trans **, struct xfs_inode *, - int, xfs_fsize_t); +int xfs_itruncate_extents_flags(struct xfs_trans **, + struct xfs_inode *, int, xfs_fsize_t, int); void xfs_iext_realloc(xfs_inode_t *, int, int); void xfs_iunpin_wait(xfs_inode_t *); @@ -431,7 +431,17 @@ xfs_extlen_t xfs_get_cowextsz_hint(struct xfs_inode *ip); int xfs_dir_ialloc(struct xfs_trans **, struct xfs_inode *, umode_t, xfs_nlink_t, dev_t, prid_t, - struct xfs_inode **, int *); + struct xfs_inode **); + +static inline int +xfs_itruncate_extents( + struct xfs_trans **tpp, + struct xfs_inode *ip, + int whichfork, + xfs_fsize_t new_size) +{ + return xfs_itruncate_extents_flags(tpp, ip, whichfork, new_size, 0); +} /* from xfs_file.c */ enum xfs_prealloc_flags { diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 34b91b789702..3e5b8574818e 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -518,7 +518,7 @@ xfs_inode_item_push( * The buffer containing this item failed to be written back * previously. Resubmit the buffer for IO. */ - if (lip->li_flags & XFS_LI_FAILED) { + if (test_bit(XFS_LI_FAILED, &lip->li_flags)) { if (!xfs_buf_trylock(bp)) return XFS_ITEM_LOCKED; @@ -729,14 +729,14 @@ xfs_iflush_done( */ iip = INODE_ITEM(blip); if ((iip->ili_logged && blip->li_lsn == iip->ili_flush_lsn) || - (blip->li_flags & XFS_LI_FAILED)) + test_bit(XFS_LI_FAILED, &blip->li_flags)) need_ail++; } /* make sure we capture the state of the initial inode. */ iip = INODE_ITEM(lip); if ((iip->ili_logged && lip->li_lsn == iip->ili_flush_lsn) || - lip->li_flags & XFS_LI_FAILED) + test_bit(XFS_LI_FAILED, &lip->li_flags)) need_ail++; /* @@ -803,7 +803,7 @@ xfs_iflush_abort( xfs_inode_log_item_t *iip = ip->i_itemp; if (iip) { - if (iip->ili_item.li_flags & XFS_LI_IN_AIL) { + if (test_bit(XFS_LI_IN_AIL, &iip->ili_item.li_flags)) { xfs_trans_ail_remove(&iip->ili_item, stale ? SHUTDOWN_LOG_IO_ERROR : SHUTDOWN_CORRUPT_INCORE); diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 89fb1eb80aae..5dd9e22b4a4c 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -1103,7 +1103,8 @@ xfs_ioctl_setattr_dax_invalidate( if (fa->fsx_xflags & FS_XFLAG_DAX) { if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) return -EINVAL; - if (bdev_dax_supported(sb, sb->s_blocksize) < 0) + if (!bdev_dax_supported(xfs_find_bdev_for_inode(VFS_I(ip)), + sb->s_blocksize)) return -EINVAL; } @@ -1811,6 +1812,88 @@ xfs_ioc_swapext( return error; } +static int +xfs_ioc_getlabel( + struct xfs_mount *mp, + char __user *user_label) +{ + struct xfs_sb *sbp = &mp->m_sb; + char label[XFSLABEL_MAX + 1]; + + /* Paranoia */ + BUILD_BUG_ON(sizeof(sbp->sb_fname) > FSLABEL_MAX); + + spin_lock(&mp->m_sb_lock); + strncpy(label, sbp->sb_fname, sizeof(sbp->sb_fname)); + spin_unlock(&mp->m_sb_lock); + + /* xfs on-disk label is 12 chars, be sure we send a null to user */ + label[XFSLABEL_MAX] = '\0'; + if (copy_to_user(user_label, label, sizeof(sbp->sb_fname))) + return -EFAULT; + return 0; +} + +static int +xfs_ioc_setlabel( + struct file *filp, + struct xfs_mount *mp, + char __user *newlabel) +{ + struct xfs_sb *sbp = &mp->m_sb; + char label[XFSLABEL_MAX + 1]; + size_t len; + int error; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + /* + * The generic ioctl allows up to FSLABEL_MAX chars, but XFS is much + * smaller, at 12 bytes. We copy one more to be sure we find the + * (required) NULL character to test the incoming label length. + * NB: The on disk label doesn't need to be null terminated. + */ + if (copy_from_user(label, newlabel, XFSLABEL_MAX + 1)) + return -EFAULT; + len = strnlen(label, XFSLABEL_MAX + 1); + if (len > sizeof(sbp->sb_fname)) + return -EINVAL; + + error = mnt_want_write_file(filp); + if (error) + return error; + + spin_lock(&mp->m_sb_lock); + memset(sbp->sb_fname, 0, sizeof(sbp->sb_fname)); + strncpy(sbp->sb_fname, label, sizeof(sbp->sb_fname)); + spin_unlock(&mp->m_sb_lock); + + /* + * Now we do several things to satisfy userspace. + * In addition to normal logging of the primary superblock, we also + * immediately write these changes to sector zero for the primary, then + * update all backup supers (as xfs_db does for a label change), then + * invalidate the block device page cache. This is so that any prior + * buffered reads from userspace (i.e. from blkid) are invalidated, + * and userspace will see the newly-written label. + */ + error = xfs_sync_sb_buf(mp); + if (error) + goto out; + /* + * growfs also updates backup supers so lock against that. + */ + mutex_lock(&mp->m_growlock); + error = xfs_update_secondary_sbs(mp); + mutex_unlock(&mp->m_growlock); + + invalidate_bdev(mp->m_ddev_targp->bt_bdev); + +out: + mnt_drop_write_file(filp); + return error; +} + /* * Note: some of the ioctl's return positive numbers as a * byte count indicating success, such as readlink_by_handle. @@ -1834,6 +1917,10 @@ xfs_file_ioctl( switch (cmd) { case FITRIM: return xfs_ioc_trim(mp, arg); + case FS_IOC_GETFSLABEL: + return xfs_ioc_getlabel(mp, arg); + case FS_IOC_SETFSLABEL: + return xfs_ioc_setlabel(filp, mp, arg); case XFS_IOC_ALLOCSP: case XFS_IOC_FREESP: case XFS_IOC_RESVSP: diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 046469fcc1b8..c6ce6f9335b6 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -224,7 +224,7 @@ xfs_iomap_write_direct( * necessary and move on to transaction setup. */ xfs_iunlock(ip, lockmode); - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return error; @@ -576,7 +576,7 @@ xfs_file_iomap_begin_delay( goto done; } - error = xfs_qm_dqattach_locked(ip, 0); + error = xfs_qm_dqattach_locked(ip, false); if (error) goto out_unlock; @@ -692,7 +692,7 @@ xfs_iomap_write_allocate( /* * Make sure that the dquots are there. */ - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return error; @@ -946,8 +946,11 @@ error_on_bmapi_transaction: return error; } -static inline bool imap_needs_alloc(struct inode *inode, - struct xfs_bmbt_irec *imap, int nimaps) +static inline bool +imap_needs_alloc( + struct inode *inode, + struct xfs_bmbt_irec *imap, + int nimaps) { return !nimaps || imap->br_startblock == HOLESTARTBLOCK || @@ -955,31 +958,58 @@ static inline bool imap_needs_alloc(struct inode *inode, (IS_DAX(inode) && imap->br_state == XFS_EXT_UNWRITTEN); } -static inline bool needs_cow_for_zeroing(struct xfs_bmbt_irec *imap, int nimaps) +static inline bool +needs_cow_for_zeroing( + struct xfs_bmbt_irec *imap, + int nimaps) { return nimaps && imap->br_startblock != HOLESTARTBLOCK && imap->br_state != XFS_EXT_UNWRITTEN; } -static inline bool need_excl_ilock(struct xfs_inode *ip, unsigned flags) +static int +xfs_ilock_for_iomap( + struct xfs_inode *ip, + unsigned flags, + unsigned *lockmode) { + unsigned mode = XFS_ILOCK_SHARED; + /* * COW writes may allocate delalloc space or convert unwritten COW * extents, so we need to make sure to take the lock exclusively here. */ - if (xfs_is_reflink_inode(ip) && (flags & (IOMAP_WRITE | IOMAP_ZERO))) - return true; + if (xfs_is_reflink_inode(ip) && (flags & (IOMAP_WRITE | IOMAP_ZERO))) { + /* + * FIXME: It could still overwrite on unshared extents and not + * need allocation. + */ + if (flags & IOMAP_NOWAIT) + return -EAGAIN; + mode = XFS_ILOCK_EXCL; + } /* - * Extents not yet cached requires exclusive access, don't block. - * This is an opencoded xfs_ilock_data_map_shared() to cater for the + * Extents not yet cached requires exclusive access, don't block. This + * is an opencoded xfs_ilock_data_map_shared() call but with * non-blocking behaviour. */ - if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE && - !(ip->i_df.if_flags & XFS_IFEXTENTS)) - return true; - return false; + if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) { + if (flags & IOMAP_NOWAIT) + return -EAGAIN; + mode = XFS_ILOCK_EXCL; + } + + if (flags & IOMAP_NOWAIT) { + if (!xfs_ilock_nowait(ip, mode)) + return -EAGAIN; + } else { + xfs_ilock(ip, mode); + } + + *lockmode = mode; + return 0; } static int @@ -1007,19 +1037,15 @@ xfs_file_iomap_begin( return xfs_file_iomap_begin_delay(inode, offset, length, iomap); } - if (need_excl_ilock(ip, flags)) - lockmode = XFS_ILOCK_EXCL; - else - lockmode = XFS_ILOCK_SHARED; - - if (flags & IOMAP_NOWAIT) { - if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) - return -EAGAIN; - if (!xfs_ilock_nowait(ip, lockmode)) - return -EAGAIN; - } else { - xfs_ilock(ip, lockmode); - } + /* + * Lock the inode in the manner required for the specified operation and + * check for as many conditions that would result in blocking as + * possible. This removes most of the non-blocking checks from the + * mapping code below. + */ + error = xfs_ilock_for_iomap(ip, flags, &lockmode); + if (error) + return error; ASSERT(offset <= mp->m_super->s_maxbytes); if (offset > mp->m_super->s_maxbytes - length) @@ -1040,19 +1066,21 @@ xfs_file_iomap_begin( goto out_unlock; } - if (xfs_is_reflink_inode(ip) && - ((flags & IOMAP_WRITE) || - ((flags & IOMAP_ZERO) && needs_cow_for_zeroing(&imap, nimaps)))) { + /* Non-modifying mapping requested, so we are done */ + if (!(flags & (IOMAP_WRITE | IOMAP_ZERO))) + goto out_found; + + /* + * Break shared extents if necessary. Checks for non-blocking IO have + * been done up front, so we don't need to do them here. + */ + if (xfs_is_reflink_inode(ip)) { + /* if zeroing doesn't need COW allocation, then we are done. */ + if ((flags & IOMAP_ZERO) && + !needs_cow_for_zeroing(&imap, nimaps)) + goto out_found; + if (flags & IOMAP_DIRECT) { - /* - * A reflinked inode will result in CoW alloc. - * FIXME: It could still overwrite on unshared extents - * and not need allocation. - */ - if (flags & IOMAP_NOWAIT) { - error = -EAGAIN; - goto out_unlock; - } /* may drop and re-acquire the ilock */ error = xfs_reflink_allocate_cow(ip, &imap, &shared, &lockmode); @@ -1068,46 +1096,45 @@ xfs_file_iomap_begin( length = XFS_FSB_TO_B(mp, end_fsb) - offset; } - if ((flags & IOMAP_WRITE) && imap_needs_alloc(inode, &imap, nimaps)) { - /* - * If nowait is set bail since we are going to make - * allocations. - */ - if (flags & IOMAP_NOWAIT) { - error = -EAGAIN; - goto out_unlock; - } - /* - * We cap the maximum length we map here to MAX_WRITEBACK_PAGES - * pages to keep the chunks of work done where somewhat symmetric - * with the work writeback does. This is a completely arbitrary - * number pulled out of thin air as a best guess for initial - * testing. - * - * Note that the values needs to be less than 32-bits wide until - * the lower level functions are updated. - */ - length = min_t(loff_t, length, 1024 * PAGE_SIZE); - /* - * xfs_iomap_write_direct() expects the shared lock. It - * is unlocked on return. - */ - if (lockmode == XFS_ILOCK_EXCL) - xfs_ilock_demote(ip, lockmode); - error = xfs_iomap_write_direct(ip, offset, length, &imap, - nimaps); - if (error) - return error; + /* Don't need to allocate over holes when doing zeroing operations. */ + if (flags & IOMAP_ZERO) + goto out_found; - iomap->flags = IOMAP_F_NEW; - trace_xfs_iomap_alloc(ip, offset, length, 0, &imap); - } else { - ASSERT(nimaps); + if (!imap_needs_alloc(inode, &imap, nimaps)) + goto out_found; - xfs_iunlock(ip, lockmode); - trace_xfs_iomap_found(ip, offset, length, 0, &imap); + /* If nowait is set bail since we are going to make allocations. */ + if (flags & IOMAP_NOWAIT) { + error = -EAGAIN; + goto out_unlock; } + /* + * We cap the maximum length we map to a sane size to keep the chunks + * of work done where somewhat symmetric with the work writeback does. + * This is a completely arbitrary number pulled out of thin air as a + * best guess for initial testing. + * + * Note that the values needs to be less than 32-bits wide until the + * lower level functions are updated. + */ + length = min_t(loff_t, length, 1024 * PAGE_SIZE); + + /* + * xfs_iomap_write_direct() expects the shared lock. It is unlocked on + * return. + */ + if (lockmode == XFS_ILOCK_EXCL) + xfs_ilock_demote(ip, lockmode); + error = xfs_iomap_write_direct(ip, offset, length, &imap, + nimaps); + if (error) + return error; + + iomap->flags = IOMAP_F_NEW; + trace_xfs_iomap_alloc(ip, offset, length, 0, &imap); + +out_finish: if (xfs_ipincount(ip) && (ip->i_itemp->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP)) iomap->flags |= IOMAP_F_DIRTY; @@ -1117,6 +1144,13 @@ xfs_file_iomap_begin( if (shared) iomap->flags |= IOMAP_F_SHARED; return 0; + +out_found: + ASSERT(nimaps); + xfs_iunlock(ip, lockmode); + trace_xfs_iomap_found(ip, offset, length, 0, &imap); + goto out_finish; + out_unlock: xfs_iunlock(ip, lockmode); return error; diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index e0307fbff911..b0eb49bb4918 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -177,7 +177,7 @@ xfs_generic_create( if (!tmpfile) { error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip); } else { - error = xfs_create_tmpfile(XFS_I(dir), dentry, mode, &ip); + error = xfs_create_tmpfile(XFS_I(dir), mode, &ip); } if (unlikely(error)) goto out_free_acl; @@ -260,6 +260,7 @@ xfs_vn_lookup( struct dentry *dentry, unsigned int flags) { + struct inode *inode; struct xfs_inode *cip; struct xfs_name name; int error; @@ -269,14 +270,13 @@ xfs_vn_lookup( xfs_dentry_to_name(&name, dentry); error = xfs_lookup(XFS_I(dir), &name, &cip, NULL); - if (unlikely(error)) { - if (unlikely(error != -ENOENT)) - return ERR_PTR(error); - d_add(dentry, NULL); - return NULL; - } - - return d_splice_alias(VFS_I(cip), dentry); + if (likely(!error)) + inode = VFS_I(cip); + else if (likely(error == -ENOENT)) + inode = NULL; + else + inode = ERR_PTR(error); + return d_splice_alias(inode, dentry); } STATIC struct dentry * @@ -855,7 +855,7 @@ xfs_setattr_size( /* * Make sure that the dquots are attached to the inode. */ - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return error; @@ -1195,6 +1195,30 @@ static const struct inode_operations xfs_inline_symlink_inode_operations = { .update_time = xfs_vn_update_time, }; +/* Figure out if this file actually supports DAX. */ +static bool +xfs_inode_supports_dax( + struct xfs_inode *ip) +{ + struct xfs_mount *mp = ip->i_mount; + + /* Only supported on non-reflinked files. */ + if (!S_ISREG(VFS_I(ip)->i_mode) || xfs_is_reflink_inode(ip)) + return false; + + /* DAX mount option or DAX iflag must be set. */ + if (!(mp->m_flags & XFS_MOUNT_DAX) && + !(ip->i_d.di_flags2 & XFS_DIFLAG2_DAX)) + return false; + + /* Block size must match page size */ + if (mp->m_sb.sb_blocksize != PAGE_SIZE) + return false; + + /* Device has to support DAX too. */ + return xfs_find_daxdev_for_inode(VFS_I(ip)) != NULL; +} + STATIC void xfs_diflags_to_iflags( struct inode *inode, @@ -1213,11 +1237,7 @@ xfs_diflags_to_iflags( inode->i_flags |= S_SYNC; if (flags & XFS_DIFLAG_NOATIME) inode->i_flags |= S_NOATIME; - if (S_ISREG(inode->i_mode) && - ip->i_mount->m_sb.sb_blocksize == PAGE_SIZE && - !xfs_is_reflink_inode(ip) && - (ip->i_mount->m_flags & XFS_MOUNT_DAX || - ip->i_d.di_flags2 & XFS_DIFLAG2_DAX)) + if (xfs_inode_supports_dax(ip)) inode->i_flags |= S_DAX; } @@ -1285,7 +1305,10 @@ xfs_setup_iops( case S_IFREG: inode->i_op = &xfs_inode_operations; inode->i_fop = &xfs_file_operations; - inode->i_mapping->a_ops = &xfs_address_space_operations; + if (IS_DAX(inode)) + inode->i_mapping->a_ops = &xfs_dax_aops; + else + inode->i_mapping->a_ops = &xfs_address_space_operations; break; case S_IFDIR: if (xfs_sb_version_hasasciici(&XFS_M(inode->i_sb)->m_sb)) diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index b9c9c848146b..c21039f27e39 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -560,7 +560,6 @@ xfs_log_done( */ int xfs_log_notify( - struct xfs_mount *mp, struct xlog_in_core *iclog, xfs_log_callback_t *cb) { @@ -1048,6 +1047,7 @@ xfs_log_item_init( INIT_LIST_HEAD(&item->li_ail); INIT_LIST_HEAD(&item->li_cil); INIT_LIST_HEAD(&item->li_bio_list); + INIT_LIST_HEAD(&item->li_trans); } /* @@ -2111,10 +2111,10 @@ xlog_print_tic_res( */ void xlog_print_trans( - struct xfs_trans *tp) + struct xfs_trans *tp) { - struct xfs_mount *mp = tp->t_mountp; - struct xfs_log_item_desc *lidp; + struct xfs_mount *mp = tp->t_mountp; + struct xfs_log_item *lip; /* dump core transaction and ticket info */ xfs_warn(mp, "transaction summary:"); @@ -2125,15 +2125,14 @@ xlog_print_trans( xlog_print_tic_res(mp, tp->t_ticket); /* dump each log item */ - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - struct xfs_log_item *lip = lidp->lid_item; + list_for_each_entry(lip, &tp->t_items, li_trans) { struct xfs_log_vec *lv = lip->li_lv; struct xfs_log_iovec *vec; int i; xfs_warn(mp, "log item: "); xfs_warn(mp, " type = 0x%x", lip->li_type); - xfs_warn(mp, " flags = 0x%x", lip->li_flags); + xfs_warn(mp, " flags = 0x%lx", lip->li_flags); if (!lv) continue; xfs_warn(mp, " niovecs = %d", lv->lv_niovecs); diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 7e2d62922a16..fa8ad31d587f 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -141,8 +141,7 @@ int xfs_log_mount_cancel(struct xfs_mount *); xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp); xfs_lsn_t xlog_assign_tail_lsn_locked(struct xfs_mount *mp); void xfs_log_space_wake(struct xfs_mount *mp); -int xfs_log_notify(struct xfs_mount *mp, - struct xlog_in_core *iclog, +int xfs_log_notify(struct xlog_in_core *iclog, struct xfs_log_callback *callback_entry); int xfs_log_release_iclog(struct xfs_mount *mp, struct xlog_in_core *iclog); diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index cb376ac8a595..c15687724728 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -141,10 +141,9 @@ xlog_cil_alloc_shadow_bufs( struct xlog *log, struct xfs_trans *tp) { - struct xfs_log_item_desc *lidp; + struct xfs_log_item *lip; - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - struct xfs_log_item *lip = lidp->lid_item; + list_for_each_entry(lip, &tp->t_items, li_trans) { struct xfs_log_vec *lv; int niovecs = 0; int nbytes = 0; @@ -152,7 +151,7 @@ xlog_cil_alloc_shadow_bufs( bool ordered = false; /* Skip items which aren't dirty in this transaction. */ - if (!(lidp->lid_flags & XFS_LID_DIRTY)) + if (!test_bit(XFS_LI_DIRTY, &lip->li_flags)) continue; /* get number of vecs and size of data to be stored */ @@ -317,7 +316,7 @@ xlog_cil_insert_format_items( int *diff_len, int *diff_iovecs) { - struct xfs_log_item_desc *lidp; + struct xfs_log_item *lip; /* Bail out if we didn't find a log item. */ @@ -326,15 +325,14 @@ xlog_cil_insert_format_items( return; } - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - struct xfs_log_item *lip = lidp->lid_item; + list_for_each_entry(lip, &tp->t_items, li_trans) { struct xfs_log_vec *lv; struct xfs_log_vec *old_lv = NULL; struct xfs_log_vec *shadow; bool ordered = false; /* Skip items which aren't dirty in this transaction. */ - if (!(lidp->lid_flags & XFS_LID_DIRTY)) + if (!test_bit(XFS_LI_DIRTY, &lip->li_flags)) continue; /* @@ -406,7 +404,7 @@ xlog_cil_insert_items( { struct xfs_cil *cil = log->l_cilp; struct xfs_cil_ctx *ctx = cil->xc_ctx; - struct xfs_log_item_desc *lidp; + struct xfs_log_item *lip; int len = 0; int diff_iovecs = 0; int iclog_space; @@ -479,11 +477,10 @@ xlog_cil_insert_items( * We do this here so we only need to take the CIL lock once during * the transaction commit. */ - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - struct xfs_log_item *lip = lidp->lid_item; + list_for_each_entry(lip, &tp->t_items, li_trans) { /* Skip items which aren't dirty in this transaction. */ - if (!(lidp->lid_flags & XFS_LID_DIRTY)) + if (!test_bit(XFS_LI_DIRTY, &lip->li_flags)) continue; /* @@ -848,7 +845,7 @@ restart: /* attach all the transactions w/ busy extents to iclog */ ctx->log_cb.cb_func = xlog_cil_committed; ctx->log_cb.cb_arg = ctx; - error = xfs_log_notify(log->l_mp, commit_iclog, &ctx->log_cb); + error = xfs_log_notify(commit_iclog, &ctx->log_cb); if (error) goto out_abort; @@ -1013,6 +1010,7 @@ xfs_log_commit_cil( *commit_lsn = xc_commit_lsn; xfs_log_done(mp, tp->t_ticket, NULL, regrant); + tp->t_ticket = NULL; xfs_trans_unreserve_and_mod_sb(tp); /* diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 2b2383f1895e..06a09cb948b5 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2702,7 +2702,7 @@ xlog_recover_do_reg_buffer( goto next; } fa = xfs_dquot_verify(mp, item->ri_buf[i].i_addr, - -1, 0, 0); + -1, 0); if (fa) { xfs_alert(mp, "dquot corrupt at %pS trying to replay into block 0x%llx", @@ -3348,7 +3348,7 @@ xlog_recover_dquot_pass2( */ dq_f = item->ri_buf[0].i_addr; ASSERT(dq_f); - fa = xfs_dquot_verify(mp, recddq, dq_f->qlf_id, 0, 0); + fa = xfs_dquot_verify(mp, recddq, dq_f->qlf_id, 0); if (fa) { xfs_alert(mp, "corrupt dquot ID 0x%x in log at %pS", dq_f->qlf_id, fa); diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index a901b86772f8..73ed8fec0328 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1072,9 +1072,7 @@ xfs_unmountfs( uint64_t resblks; int error; - cancel_delayed_work_sync(&mp->m_eofblocks_work); - cancel_delayed_work_sync(&mp->m_cowblocks_work); - + xfs_icache_disable_reclaim(mp); xfs_fs_unreserve_ag_blocks(mp); xfs_qm_unmount_quotas(mp); xfs_rtunmount_inodes(mp); diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c index f8a674d7f092..70eea7ae2876 100644 --- a/fs/xfs/xfs_mru_cache.c +++ b/fs/xfs/xfs_mru_cache.c @@ -112,6 +112,7 @@ struct xfs_mru_cache { xfs_mru_cache_free_func_t free_func; /* Function pointer for freeing. */ struct delayed_work work; /* Workqueue data for reaping. */ unsigned int queued; /* work has been queued */ + void *data; }; static struct workqueue_struct *xfs_mru_reap_wq; @@ -259,7 +260,7 @@ _xfs_mru_cache_clear_reap_list( list_for_each_entry_safe(elem, next, &tmp, list_node) { list_del_init(&elem->list_node); - mru->free_func(elem); + mru->free_func(mru->data, elem); } spin_lock(&mru->lock); @@ -326,6 +327,7 @@ xfs_mru_cache_uninit(void) int xfs_mru_cache_create( struct xfs_mru_cache **mrup, + void *data, unsigned int lifetime_ms, unsigned int grp_count, xfs_mru_cache_free_func_t free_func) @@ -369,7 +371,7 @@ xfs_mru_cache_create( mru->grp_time = grp_time; mru->free_func = free_func; - + mru->data = data; *mrup = mru; exit: @@ -492,7 +494,7 @@ xfs_mru_cache_delete( elem = xfs_mru_cache_remove(mru, key); if (elem) - mru->free_func(elem); + mru->free_func(mru->data, elem); } /* diff --git a/fs/xfs/xfs_mru_cache.h b/fs/xfs/xfs_mru_cache.h index fb5245ba5ff7..b3f3fbdfcc47 100644 --- a/fs/xfs/xfs_mru_cache.h +++ b/fs/xfs/xfs_mru_cache.h @@ -26,13 +26,13 @@ struct xfs_mru_cache_elem { }; /* Function pointer type for callback to free a client's data pointer. */ -typedef void (*xfs_mru_cache_free_func_t)(struct xfs_mru_cache_elem *elem); +typedef void (*xfs_mru_cache_free_func_t)(void *, struct xfs_mru_cache_elem *); int xfs_mru_cache_init(void); void xfs_mru_cache_uninit(void); -int xfs_mru_cache_create(struct xfs_mru_cache **mrup, unsigned int lifetime_ms, - unsigned int grp_count, - xfs_mru_cache_free_func_t free_func); +int xfs_mru_cache_create(struct xfs_mru_cache **mrup, void *data, + unsigned int lifetime_ms, unsigned int grp_count, + xfs_mru_cache_free_func_t free_func); void xfs_mru_cache_destroy(struct xfs_mru_cache *mru); int xfs_mru_cache_insert(struct xfs_mru_cache *mru, unsigned long key, struct xfs_mru_cache_elem *elem); diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index 5b848f4b637f..c3e014bfc848 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -161,10 +161,7 @@ xfs_qm_dqpurge( * to purge this dquot anyway, so we go ahead regardless. */ error = xfs_qm_dqflush(dqp, &bp); - if (error) { - xfs_warn(mp, "%s: dquot "PTR_FMT" flush failed", - __func__, dqp); - } else { + if (!error) { error = xfs_bwrite(bp); xfs_buf_relse(bp); } @@ -173,7 +170,7 @@ xfs_qm_dqpurge( ASSERT(atomic_read(&dqp->q_pincount) == 0); ASSERT(XFS_FORCED_SHUTDOWN(mp) || - !(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL)); + !test_bit(XFS_LI_IN_AIL, &dqp->q_logitem.qli_item.li_flags)); xfs_dqfunlock(dqp); xfs_dqunlock(dqp); @@ -265,7 +262,7 @@ xfs_qm_dqattach_one( xfs_inode_t *ip, xfs_dqid_t id, uint type, - uint doalloc, + bool doalloc, xfs_dquot_t **IO_idqpp) { xfs_dquot_t *dqp; @@ -291,7 +288,7 @@ xfs_qm_dqattach_one( * exist on disk and we didn't ask it to allocate; ESRCH if quotas got * turned off suddenly. */ - error = xfs_qm_dqget(ip->i_mount, ip, id, type, doalloc, &dqp); + error = xfs_qm_dqget_inode(ip, type, doalloc, &dqp); if (error) return error; @@ -326,14 +323,14 @@ xfs_qm_need_dqattach( /* * Given a locked inode, attach dquot(s) to it, taking U/G/P-QUOTAON * into account. - * If XFS_QMOPT_DQALLOC, the dquot(s) will be allocated if needed. + * If @doalloc is true, the dquot(s) will be allocated if needed. * Inode may get unlocked and relocked in here, and the caller must deal with * the consequences. */ int xfs_qm_dqattach_locked( xfs_inode_t *ip, - uint flags) + bool doalloc) { xfs_mount_t *mp = ip->i_mount; int error = 0; @@ -345,8 +342,7 @@ xfs_qm_dqattach_locked( if (XFS_IS_UQUOTA_ON(mp) && !ip->i_udquot) { error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER, - flags & XFS_QMOPT_DQALLOC, - &ip->i_udquot); + doalloc, &ip->i_udquot); if (error) goto done; ASSERT(ip->i_udquot); @@ -354,8 +350,7 @@ xfs_qm_dqattach_locked( if (XFS_IS_GQUOTA_ON(mp) && !ip->i_gdquot) { error = xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP, - flags & XFS_QMOPT_DQALLOC, - &ip->i_gdquot); + doalloc, &ip->i_gdquot); if (error) goto done; ASSERT(ip->i_gdquot); @@ -363,8 +358,7 @@ xfs_qm_dqattach_locked( if (XFS_IS_PQUOTA_ON(mp) && !ip->i_pdquot) { error = xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ, - flags & XFS_QMOPT_DQALLOC, - &ip->i_pdquot); + doalloc, &ip->i_pdquot); if (error) goto done; ASSERT(ip->i_pdquot); @@ -381,8 +375,7 @@ done: int xfs_qm_dqattach( - struct xfs_inode *ip, - uint flags) + struct xfs_inode *ip) { int error; @@ -390,7 +383,7 @@ xfs_qm_dqattach( return 0; xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_qm_dqattach_locked(ip, flags); + error = xfs_qm_dqattach_locked(ip, false); xfs_iunlock(ip, XFS_ILOCK_EXCL); return error; @@ -479,11 +472,8 @@ xfs_qm_dquot_isolate( spin_unlock(lru_lock); error = xfs_qm_dqflush(dqp, &bp); - if (error) { - xfs_warn(dqp->q_mount, "%s: dquot "PTR_FMT" flush failed", - __func__, dqp); + if (error) goto out_unlock_dirty; - } xfs_buf_delwri_queue(bp, &isol->buffers); xfs_buf_relse(bp); @@ -571,27 +561,88 @@ xfs_qm_set_defquota( { xfs_dquot_t *dqp; struct xfs_def_quota *defq; + struct xfs_disk_dquot *ddqp; int error; - error = xfs_qm_dqread(mp, 0, type, 0, &dqp); + error = xfs_qm_dqget_uncached(mp, 0, type, &dqp); + if (error) + return; - if (!error) { - xfs_disk_dquot_t *ddqp = &dqp->q_core; + ddqp = &dqp->q_core; + defq = xfs_get_defquota(dqp, qinf); - defq = xfs_get_defquota(dqp, qinf); + /* + * Timers and warnings have been already set, let's just set the + * default limits for this quota type + */ + defq->bhardlimit = be64_to_cpu(ddqp->d_blk_hardlimit); + defq->bsoftlimit = be64_to_cpu(ddqp->d_blk_softlimit); + defq->ihardlimit = be64_to_cpu(ddqp->d_ino_hardlimit); + defq->isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit); + defq->rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); + defq->rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit); + xfs_qm_dqdestroy(dqp); +} - /* - * Timers and warnings have been already set, let's just set the - * default limits for this quota type - */ - defq->bhardlimit = be64_to_cpu(ddqp->d_blk_hardlimit); - defq->bsoftlimit = be64_to_cpu(ddqp->d_blk_softlimit); - defq->ihardlimit = be64_to_cpu(ddqp->d_ino_hardlimit); - defq->isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit); - defq->rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); - defq->rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit); - xfs_qm_dqdestroy(dqp); - } +/* Initialize quota time limits from the root dquot. */ +static void +xfs_qm_init_timelimits( + struct xfs_mount *mp, + struct xfs_quotainfo *qinf) +{ + struct xfs_disk_dquot *ddqp; + struct xfs_dquot *dqp; + uint type; + int error; + + qinf->qi_btimelimit = XFS_QM_BTIMELIMIT; + qinf->qi_itimelimit = XFS_QM_ITIMELIMIT; + qinf->qi_rtbtimelimit = XFS_QM_RTBTIMELIMIT; + qinf->qi_bwarnlimit = XFS_QM_BWARNLIMIT; + qinf->qi_iwarnlimit = XFS_QM_IWARNLIMIT; + qinf->qi_rtbwarnlimit = XFS_QM_RTBWARNLIMIT; + + /* + * We try to get the limits from the superuser's limits fields. + * This is quite hacky, but it is standard quota practice. + * + * Since we may not have done a quotacheck by this point, just read + * the dquot without attaching it to any hashtables or lists. + * + * Timers and warnings are globally set by the first timer found in + * user/group/proj quota types, otherwise a default value is used. + * This should be split into different fields per quota type. + */ + if (XFS_IS_UQUOTA_RUNNING(mp)) + type = XFS_DQ_USER; + else if (XFS_IS_GQUOTA_RUNNING(mp)) + type = XFS_DQ_GROUP; + else + type = XFS_DQ_PROJ; + error = xfs_qm_dqget_uncached(mp, 0, type, &dqp); + if (error) + return; + + ddqp = &dqp->q_core; + /* + * The warnings and timers set the grace period given to + * a user or group before he or she can not perform any + * more writing. If it is zero, a default is used. + */ + if (ddqp->d_btimer) + qinf->qi_btimelimit = be32_to_cpu(ddqp->d_btimer); + if (ddqp->d_itimer) + qinf->qi_itimelimit = be32_to_cpu(ddqp->d_itimer); + if (ddqp->d_rtbtimer) + qinf->qi_rtbtimelimit = be32_to_cpu(ddqp->d_rtbtimer); + if (ddqp->d_bwarns) + qinf->qi_bwarnlimit = be16_to_cpu(ddqp->d_bwarns); + if (ddqp->d_iwarns) + qinf->qi_iwarnlimit = be16_to_cpu(ddqp->d_iwarns); + if (ddqp->d_rtbwarns) + qinf->qi_rtbwarnlimit = be16_to_cpu(ddqp->d_rtbwarns); + + xfs_qm_dqdestroy(dqp); } /* @@ -600,11 +651,10 @@ xfs_qm_set_defquota( */ STATIC int xfs_qm_init_quotainfo( - xfs_mount_t *mp) + struct xfs_mount *mp) { - xfs_quotainfo_t *qinf; - int error; - xfs_dquot_t *dqp; + struct xfs_quotainfo *qinf; + int error; ASSERT(XFS_IS_QUOTA_RUNNING(mp)); @@ -636,52 +686,7 @@ xfs_qm_init_quotainfo( mp->m_qflags |= (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_CHKD); - /* - * We try to get the limits from the superuser's limits fields. - * This is quite hacky, but it is standard quota practice. - * - * Since we may not have done a quotacheck by this point, just read - * the dquot without attaching it to any hashtables or lists. - * - * Timers and warnings are globally set by the first timer found in - * user/group/proj quota types, otherwise a default value is used. - * This should be split into different fields per quota type. - */ - error = xfs_qm_dqread(mp, 0, - XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER : - (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP : - XFS_DQ_PROJ), - 0, &dqp); - - if (!error) { - xfs_disk_dquot_t *ddqp = &dqp->q_core; - - /* - * The warnings and timers set the grace period given to - * a user or group before he or she can not perform any - * more writing. If it is zero, a default is used. - */ - qinf->qi_btimelimit = ddqp->d_btimer ? - be32_to_cpu(ddqp->d_btimer) : XFS_QM_BTIMELIMIT; - qinf->qi_itimelimit = ddqp->d_itimer ? - be32_to_cpu(ddqp->d_itimer) : XFS_QM_ITIMELIMIT; - qinf->qi_rtbtimelimit = ddqp->d_rtbtimer ? - be32_to_cpu(ddqp->d_rtbtimer) : XFS_QM_RTBTIMELIMIT; - qinf->qi_bwarnlimit = ddqp->d_bwarns ? - be16_to_cpu(ddqp->d_bwarns) : XFS_QM_BWARNLIMIT; - qinf->qi_iwarnlimit = ddqp->d_iwarns ? - be16_to_cpu(ddqp->d_iwarns) : XFS_QM_IWARNLIMIT; - qinf->qi_rtbwarnlimit = ddqp->d_rtbwarns ? - be16_to_cpu(ddqp->d_rtbwarns) : XFS_QM_RTBWARNLIMIT; - xfs_qm_dqdestroy(dqp); - } else { - qinf->qi_btimelimit = XFS_QM_BTIMELIMIT; - qinf->qi_itimelimit = XFS_QM_ITIMELIMIT; - qinf->qi_rtbtimelimit = XFS_QM_RTBTIMELIMIT; - qinf->qi_bwarnlimit = XFS_QM_BWARNLIMIT; - qinf->qi_iwarnlimit = XFS_QM_IWARNLIMIT; - qinf->qi_rtbwarnlimit = XFS_QM_RTBWARNLIMIT; - } + xfs_qm_init_timelimits(mp, qinf); if (XFS_IS_UQUOTA_RUNNING(mp)) xfs_qm_set_defquota(mp, XFS_DQ_USER, qinf); @@ -748,7 +753,6 @@ xfs_qm_qino_alloc( { xfs_trans_t *tp; int error; - int committed; bool need_alloc = true; *ip = NULL; @@ -788,8 +792,7 @@ xfs_qm_qino_alloc( return error; if (need_alloc) { - error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, ip, - &committed); + error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, ip); if (error) { xfs_trans_cancel(tp); return error; @@ -867,9 +870,9 @@ xfs_qm_reset_dqcounts( * find uninitialised dquot blks. See comment in * xfs_dquot_verify. */ - fa = xfs_dquot_verify(mp, ddq, id + j, type, 0); + fa = xfs_dqblk_verify(mp, &dqb[j], id + j, type); if (fa) - xfs_dquot_repair(mp, ddq, id + j, type); + xfs_dqblk_repair(mp, &dqb[j], id + j, type); /* * Reset type in case we are reusing group quota file for @@ -895,7 +898,7 @@ xfs_qm_reset_dqcounts( } STATIC int -xfs_qm_dqiter_bufs( +xfs_qm_reset_dqcounts_all( struct xfs_mount *mp, xfs_dqid_t firstid, xfs_fsblock_t bno, @@ -963,11 +966,11 @@ xfs_qm_dqiter_bufs( } /* - * Iterate over all allocated USR/GRP/PRJ dquots in the system, calling a - * caller supplied function for every chunk of dquots that we find. + * Iterate over all allocated dquot blocks in this quota inode, zeroing all + * counters for every chunk of dquots that we find. */ STATIC int -xfs_qm_dqiterate( +xfs_qm_reset_dqcounts_buf( struct xfs_mount *mp, struct xfs_inode *qip, uint flags, @@ -1043,7 +1046,7 @@ xfs_qm_dqiterate( * Iterate thru all the blks in the extent and * reset the counters of all the dquots inside them. */ - error = xfs_qm_dqiter_bufs(mp, firstid, + error = xfs_qm_reset_dqcounts_all(mp, firstid, map[i].br_startblock, map[i].br_blockcount, flags, buffer_list); @@ -1068,16 +1071,17 @@ out: STATIC int xfs_qm_quotacheck_dqadjust( struct xfs_inode *ip, - xfs_dqid_t id, uint type, xfs_qcnt_t nblks, xfs_qcnt_t rtblks) { struct xfs_mount *mp = ip->i_mount; struct xfs_dquot *dqp; + xfs_dqid_t id; int error; - error = xfs_qm_dqget(mp, ip, id, type, XFS_QMOPT_DQALLOC, &dqp); + id = xfs_qm_id_for_quotatype(ip, type); + error = xfs_qm_dqget(mp, id, type, true, &dqp); if (error) { /* * Shouldn't be able to turn off quotas here. @@ -1150,13 +1154,10 @@ xfs_qm_dqusage_adjust( } /* - * We don't _need_ to take the ilock EXCL. However, the xfs_qm_dqget - * interface expects the inode to be exclusively locked because that's - * the case in all other instances. It's OK that we do this because - * quotacheck is done only at mount time. + * We don't _need_ to take the ilock EXCL here because quotacheck runs + * at mount time and therefore nobody will be racing chown/chproj. */ - error = xfs_iget(mp, NULL, ino, XFS_IGET_DONTCACHE, XFS_ILOCK_EXCL, - &ip); + error = xfs_iget(mp, NULL, ino, XFS_IGET_DONTCACHE, 0, &ip); if (error) { *res = BULKSTAT_RV_NOTHING; return error; @@ -1191,33 +1192,31 @@ xfs_qm_dqusage_adjust( * and quotaoffs don't race. (Quotachecks happen at mount time only). */ if (XFS_IS_UQUOTA_ON(mp)) { - error = xfs_qm_quotacheck_dqadjust(ip, ip->i_d.di_uid, - XFS_DQ_USER, nblks, rtblks); + error = xfs_qm_quotacheck_dqadjust(ip, XFS_DQ_USER, nblks, + rtblks); if (error) goto error0; } if (XFS_IS_GQUOTA_ON(mp)) { - error = xfs_qm_quotacheck_dqadjust(ip, ip->i_d.di_gid, - XFS_DQ_GROUP, nblks, rtblks); + error = xfs_qm_quotacheck_dqadjust(ip, XFS_DQ_GROUP, nblks, + rtblks); if (error) goto error0; } if (XFS_IS_PQUOTA_ON(mp)) { - error = xfs_qm_quotacheck_dqadjust(ip, xfs_get_projid(ip), - XFS_DQ_PROJ, nblks, rtblks); + error = xfs_qm_quotacheck_dqadjust(ip, XFS_DQ_PROJ, nblks, + rtblks); if (error) goto error0; } - xfs_iunlock(ip, XFS_ILOCK_EXCL); IRELE(ip); *res = BULKSTAT_RV_DIDONE; return 0; error0: - xfs_iunlock(ip, XFS_ILOCK_EXCL); IRELE(ip); *res = BULKSTAT_RV_GIVEUP; return error; @@ -1249,9 +1248,8 @@ xfs_qm_flush_one( */ if (!xfs_dqflock_nowait(dqp)) { /* buf is pinned in-core by delwri list */ - DEFINE_SINGLE_BUF_MAP(map, dqp->q_blkno, - mp->m_quotainfo->qi_dqchunklen); - bp = _xfs_buf_find(mp->m_ddev_targp, &map, 1, 0, NULL); + bp = xfs_buf_incore(mp->m_ddev_targp, dqp->q_blkno, + mp->m_quotainfo->qi_dqchunklen, 0); if (!bp) { error = -EINVAL; goto out_unlock; @@ -1309,7 +1307,7 @@ xfs_qm_quotacheck( * We don't log our changes till later. */ if (uip) { - error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA, + error = xfs_qm_reset_dqcounts_buf(mp, uip, XFS_QMOPT_UQUOTA, &buffer_list); if (error) goto error_return; @@ -1317,7 +1315,7 @@ xfs_qm_quotacheck( } if (gip) { - error = xfs_qm_dqiterate(mp, gip, XFS_QMOPT_GQUOTA, + error = xfs_qm_reset_dqcounts_buf(mp, gip, XFS_QMOPT_GQUOTA, &buffer_list); if (error) goto error_return; @@ -1325,7 +1323,7 @@ xfs_qm_quotacheck( } if (pip) { - error = xfs_qm_dqiterate(mp, pip, XFS_QMOPT_PQUOTA, + error = xfs_qm_reset_dqcounts_buf(mp, pip, XFS_QMOPT_PQUOTA, &buffer_list); if (error) goto error_return; @@ -1677,7 +1675,7 @@ xfs_qm_vop_dqalloc( * if necessary. The dquot(s) will not be locked. */ if (XFS_NOT_DQATTACHED(mp, ip)) { - error = xfs_qm_dqattach_locked(ip, XFS_QMOPT_DQALLOC); + error = xfs_qm_dqattach_locked(ip, true); if (error) { xfs_iunlock(ip, lockflags); return error; @@ -1696,10 +1694,7 @@ xfs_qm_vop_dqalloc( * holding ilock. */ xfs_iunlock(ip, lockflags); - error = xfs_qm_dqget(mp, NULL, uid, - XFS_DQ_USER, - XFS_QMOPT_DQALLOC, - &uq); + error = xfs_qm_dqget(mp, uid, XFS_DQ_USER, true, &uq); if (error) { ASSERT(error != -ENOENT); return error; @@ -1722,10 +1717,7 @@ xfs_qm_vop_dqalloc( if ((flags & XFS_QMOPT_GQUOTA) && XFS_IS_GQUOTA_ON(mp)) { if (ip->i_d.di_gid != gid) { xfs_iunlock(ip, lockflags); - error = xfs_qm_dqget(mp, NULL, gid, - XFS_DQ_GROUP, - XFS_QMOPT_DQALLOC, - &gq); + error = xfs_qm_dqget(mp, gid, XFS_DQ_GROUP, true, &gq); if (error) { ASSERT(error != -ENOENT); goto error_rele; @@ -1741,10 +1733,8 @@ xfs_qm_vop_dqalloc( if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) { if (xfs_get_projid(ip) != prid) { xfs_iunlock(ip, lockflags); - error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)prid, - XFS_DQ_PROJ, - XFS_QMOPT_DQALLOC, - &pq); + error = xfs_qm_dqget(mp, (xfs_dqid_t)prid, XFS_DQ_PROJ, + true, &pq); if (error) { ASSERT(error != -ENOENT); goto error_rele; @@ -1935,7 +1925,7 @@ xfs_qm_vop_rename_dqattach( */ if (i == 0 || ip != i_tab[i-1]) { if (XFS_NOT_DQATTACHED(mp, ip)) { - error = xfs_qm_dqattach(ip, 0); + error = xfs_qm_dqattach(ip); if (error) return error; } diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h index 2975a822e9f0..e3129b280423 100644 --- a/fs/xfs/xfs_qm.h +++ b/fs/xfs/xfs_qm.h @@ -170,8 +170,10 @@ extern void xfs_qm_dqrele_all_inodes(struct xfs_mount *, uint); /* quota ops */ extern int xfs_qm_scall_trunc_qfiles(struct xfs_mount *, uint); -extern int xfs_qm_scall_getquota(struct xfs_mount *, xfs_dqid_t *, - uint, struct qc_dqblk *, uint); +extern int xfs_qm_scall_getquota(struct xfs_mount *, xfs_dqid_t, + uint, struct qc_dqblk *); +extern int xfs_qm_scall_getquota_next(struct xfs_mount *, + xfs_dqid_t *, uint, struct qc_dqblk *); extern int xfs_qm_scall_setqlim(struct xfs_mount *, xfs_dqid_t, uint, struct qc_dqblk *); extern int xfs_qm_scall_quotaon(struct xfs_mount *, uint); diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c index 2be6d2735ca9..36b89e2c5eb9 100644 --- a/fs/xfs/xfs_qm_bhv.c +++ b/fs/xfs/xfs_qm_bhv.c @@ -72,7 +72,7 @@ xfs_qm_statvfs( xfs_mount_t *mp = ip->i_mount; xfs_dquot_t *dqp; - if (!xfs_qm_dqget(mp, NULL, xfs_get_projid(ip), XFS_DQ_PROJ, 0, &dqp)) { + if (!xfs_qm_dqget(mp, xfs_get_projid(ip), XFS_DQ_PROJ, false, &dqp)) { xfs_fill_statvfs_from_dquot(statp, dqp); xfs_qm_dqput(dqp); } diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c index 9cb5c381b01c..3e05d300b14e 100644 --- a/fs/xfs/xfs_qm_syscalls.c +++ b/fs/xfs/xfs_qm_syscalls.c @@ -425,7 +425,7 @@ xfs_qm_scall_setqlim( * a reference to the dquot, so it's safe to do this unlock/lock without * it being reclaimed in the mean time. */ - error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp); + error = xfs_qm_dqget(mp, id, type, true, &dqp); if (error) { ASSERT(error != -ENOENT); goto out_unlock; @@ -622,39 +622,14 @@ out: return error; } - -int -xfs_qm_scall_getquota( +/* Fill out the quota context. */ +static void +xfs_qm_scall_getquota_fill_qc( struct xfs_mount *mp, - xfs_dqid_t *id, uint type, - struct qc_dqblk *dst, - uint dqget_flags) + const struct xfs_dquot *dqp, + struct qc_dqblk *dst) { - struct xfs_dquot *dqp; - int error; - - /* - * Try to get the dquot. We don't want it allocated on disk, so - * we aren't passing the XFS_QMOPT_DOALLOC flag. If it doesn't - * exist, we'll get ENOENT back. - */ - error = xfs_qm_dqget(mp, NULL, *id, type, dqget_flags, &dqp); - if (error) - return error; - - /* - * If everything's NULL, this dquot doesn't quite exist as far as - * our utility programs are concerned. - */ - if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) { - error = -ENOENT; - goto out_put; - } - - /* Fill in the ID we actually read from disk */ - *id = be32_to_cpu(dqp->q_core.d_id); - memset(dst, 0, sizeof(*dst)); dst->d_spc_hardlimit = XFS_FSB_TO_B(mp, be64_to_cpu(dqp->q_core.d_blk_hardlimit)); @@ -696,7 +671,7 @@ xfs_qm_scall_getquota( if (((XFS_IS_UQUOTA_ENFORCED(mp) && type == XFS_DQ_USER) || (XFS_IS_GQUOTA_ENFORCED(mp) && type == XFS_DQ_GROUP) || (XFS_IS_PQUOTA_ENFORCED(mp) && type == XFS_DQ_PROJ)) && - *id != 0) { + dqp->q_core.d_id != 0) { if ((dst->d_space > dst->d_spc_softlimit) && (dst->d_spc_softlimit > 0)) { ASSERT(dst->d_spc_timer != 0); @@ -707,11 +682,69 @@ xfs_qm_scall_getquota( } } #endif +} + +/* Return the quota information for the dquot matching id. */ +int +xfs_qm_scall_getquota( + struct xfs_mount *mp, + xfs_dqid_t id, + uint type, + struct qc_dqblk *dst) +{ + struct xfs_dquot *dqp; + int error; + + /* + * Try to get the dquot. We don't want it allocated on disk, so don't + * set doalloc. If it doesn't exist, we'll get ENOENT back. + */ + error = xfs_qm_dqget(mp, id, type, false, &dqp); + if (error) + return error; + + /* + * If everything's NULL, this dquot doesn't quite exist as far as + * our utility programs are concerned. + */ + if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) { + error = -ENOENT; + goto out_put; + } + + xfs_qm_scall_getquota_fill_qc(mp, type, dqp, dst); + out_put: xfs_qm_dqput(dqp); return error; } +/* + * Return the quota information for the first initialized dquot whose id + * is at least as high as id. + */ +int +xfs_qm_scall_getquota_next( + struct xfs_mount *mp, + xfs_dqid_t *id, + uint type, + struct qc_dqblk *dst) +{ + struct xfs_dquot *dqp; + int error; + + error = xfs_qm_dqget_next(mp, *id, type, &dqp); + if (error) + return error; + + /* Fill in the ID we actually read from disk */ + *id = be32_to_cpu(dqp->q_core.d_id); + + xfs_qm_scall_getquota_fill_qc(mp, type, dqp, dst); + + xfs_qm_dqput(dqp); + return error; +} STATIC int xfs_dqrele_inode( diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index ce6506adab7b..3edf52b14919 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -48,6 +48,22 @@ struct xfs_trans; (XFS_IS_PQUOTA_ON(mp) && \ (mp->m_sb.sb_qflags & XFS_PQUOTA_CHKD) == 0)) +static inline uint +xfs_quota_chkd_flag( + uint dqtype) +{ + switch (dqtype) { + case XFS_DQ_USER: + return XFS_UQUOTA_CHKD; + case XFS_DQ_GROUP: + return XFS_GQUOTA_CHKD; + case XFS_DQ_PROJ: + return XFS_PQUOTA_CHKD; + default: + return 0; + } +} + /* * The structure kept inside the xfs_trans_t keep track of dquot changes * within a transaction and apply them later. @@ -90,8 +106,8 @@ extern struct xfs_dquot *xfs_qm_vop_chown(struct xfs_trans *, extern int xfs_qm_vop_chown_reserve(struct xfs_trans *, struct xfs_inode *, struct xfs_dquot *, struct xfs_dquot *, struct xfs_dquot *, uint); -extern int xfs_qm_dqattach(struct xfs_inode *, uint); -extern int xfs_qm_dqattach_locked(struct xfs_inode *, uint); +extern int xfs_qm_dqattach(struct xfs_inode *); +extern int xfs_qm_dqattach_locked(struct xfs_inode *ip, bool doalloc); extern void xfs_qm_dqdetach(struct xfs_inode *); extern void xfs_qm_dqrele(struct xfs_dquot *); extern void xfs_qm_statvfs(struct xfs_inode *, struct kstatfs *); @@ -132,7 +148,7 @@ static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp, #define xfs_qm_vop_rename_dqattach(it) (0) #define xfs_qm_vop_chown(tp, ip, old, new) (NULL) #define xfs_qm_vop_chown_reserve(tp, ip, u, g, p, fl) (0) -#define xfs_qm_dqattach(ip, fl) (0) +#define xfs_qm_dqattach(ip) (0) #define xfs_qm_dqattach_locked(ip, fl) (0) #define xfs_qm_dqdetach(ip) #define xfs_qm_dqrele(d) diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c index a65108594a07..c93fc913dffb 100644 --- a/fs/xfs/xfs_quotaops.c +++ b/fs/xfs/xfs_quotaops.c @@ -239,8 +239,7 @@ xfs_fs_get_dqblk( return -ESRCH; id = from_kqid(&init_user_ns, qid); - return xfs_qm_scall_getquota(mp, &id, - xfs_quota_type(qid.type), qdq, 0); + return xfs_qm_scall_getquota(mp, id, xfs_quota_type(qid.type), qdq); } /* Return quota info for active quota >= this qid */ @@ -260,9 +259,8 @@ xfs_fs_get_nextdqblk( return -ESRCH; id = from_kqid(&init_user_ns, *qid); - ret = xfs_qm_scall_getquota(mp, &id, - xfs_quota_type(qid->type), qdq, - XFS_QMOPT_DQNEXT); + ret = xfs_qm_scall_getquota_next(mp, &id, xfs_quota_type(qid->type), + qdq); if (ret) return ret; diff --git a/fs/xfs/xfs_refcount_item.c b/fs/xfs/xfs_refcount_item.c index 7a39f40645f7..e5866b714d5f 100644 --- a/fs/xfs/xfs_refcount_item.c +++ b/fs/xfs/xfs_refcount_item.c @@ -52,6 +52,25 @@ xfs_cui_item_free( kmem_zone_free(xfs_cui_zone, cuip); } +/* + * Freeing the CUI requires that we remove it from the AIL if it has already + * been placed there. However, the CUI may not yet have been placed in the AIL + * when called by xfs_cui_release() from CUD processing due to the ordering of + * committed vs unpin operations in bulk insert operations. Hence the reference + * count to ensure only the last caller frees the CUI. + */ +void +xfs_cui_release( + struct xfs_cui_log_item *cuip) +{ + ASSERT(atomic_read(&cuip->cui_refcount) > 0); + if (atomic_dec_and_test(&cuip->cui_refcount)) { + xfs_trans_ail_remove(&cuip->cui_item, SHUTDOWN_LOG_IO_ERROR); + xfs_cui_item_free(cuip); + } +} + + STATIC void xfs_cui_item_size( struct xfs_log_item *lip, @@ -140,8 +159,8 @@ STATIC void xfs_cui_item_unlock( struct xfs_log_item *lip) { - if (lip->li_flags & XFS_LI_ABORTED) - xfs_cui_item_free(CUI_ITEM(lip)); + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) + xfs_cui_release(CUI_ITEM(lip)); } /* @@ -211,24 +230,6 @@ xfs_cui_init( return cuip; } -/* - * Freeing the CUI requires that we remove it from the AIL if it has already - * been placed there. However, the CUI may not yet have been placed in the AIL - * when called by xfs_cui_release() from CUD processing due to the ordering of - * committed vs unpin operations in bulk insert operations. Hence the reference - * count to ensure only the last caller frees the CUI. - */ -void -xfs_cui_release( - struct xfs_cui_log_item *cuip) -{ - ASSERT(atomic_read(&cuip->cui_refcount) > 0); - if (atomic_dec_and_test(&cuip->cui_refcount)) { - xfs_trans_ail_remove(&cuip->cui_item, SHUTDOWN_LOG_IO_ERROR); - xfs_cui_item_free(cuip); - } -} - static inline struct xfs_cud_log_item *CUD_ITEM(struct xfs_log_item *lip) { return container_of(lip, struct xfs_cud_log_item, cud_item); @@ -309,7 +310,7 @@ xfs_cud_item_unlock( { struct xfs_cud_log_item *cudp = CUD_ITEM(lip); - if (lip->li_flags & XFS_LI_ABORTED) { + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) { xfs_cui_release(cudp->cud_cuip); kmem_zone_free(xfs_cud_zone, cudp); } diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index cdbd342a5249..713e857d9ffa 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -305,7 +305,7 @@ xfs_reflink_reserve_cow( * Fork all the shared blocks from our write offset until the end of * the extent. */ - error = xfs_qm_dqattach_locked(ip, 0); + error = xfs_qm_dqattach_locked(ip, false); if (error) return error; @@ -431,7 +431,7 @@ retry: if (error) return error; - error = xfs_qm_dqattach_locked(ip, 0); + error = xfs_qm_dqattach_locked(ip, false); if (error) goto out; goto retry; @@ -552,6 +552,9 @@ xfs_reflink_trim_irec_to_next_cow( * * If cancel_real is true this function cancels all COW fork extents for the * inode; if cancel_real is false, real extents are not cleared. + * + * Caller must have already joined the inode to the current transaction. The + * inode will be joined to the transaction returned to the caller. */ int xfs_reflink_cancel_cow_blocks( @@ -592,7 +595,6 @@ xfs_reflink_cancel_cow_blocks( if (error) break; } else if (del.br_state == XFS_EXT_UNWRITTEN || cancel_real) { - xfs_trans_ijoin(*tpp, ip, 0); xfs_defer_init(&dfops, &firstfsb); /* Free the CoW orphan record. */ @@ -1359,7 +1361,7 @@ xfs_reflink_remap_range( goto out_unlock; /* Attach dquots to dest inode before changing block map */ - ret = xfs_qm_dqattach(dest, 0); + ret = xfs_qm_dqattach(dest); if (ret) goto out_unlock; @@ -1551,7 +1553,12 @@ next: return 0; } -/* Clear the inode reflink flag if there are no shared extents. */ +/* + * Clear the inode reflink flag if there are no shared extents. + * + * The caller is responsible for joining the inode to the transaction passed in. + * The inode will be joined to the transaction that is returned to the caller. + */ int xfs_reflink_clear_inode_flag( struct xfs_inode *ip, @@ -1578,7 +1585,6 @@ xfs_reflink_clear_inode_flag( trace_xfs_reflink_unset_inode_flag(ip); ip->i_d.di_flags2 &= ~XFS_DIFLAG2_REFLINK; xfs_inode_clear_cowblocks_tag(ip); - xfs_trans_ijoin(*tpp, ip, 0); xfs_trans_log_inode(*tpp, ip, XFS_ILOG_CORE); return error; diff --git a/fs/xfs/xfs_rmap_item.c b/fs/xfs/xfs_rmap_item.c index 49d3124863a8..e5b5b3e7ef82 100644 --- a/fs/xfs/xfs_rmap_item.c +++ b/fs/xfs/xfs_rmap_item.c @@ -52,6 +52,24 @@ xfs_rui_item_free( kmem_zone_free(xfs_rui_zone, ruip); } +/* + * Freeing the RUI requires that we remove it from the AIL if it has already + * been placed there. However, the RUI may not yet have been placed in the AIL + * when called by xfs_rui_release() from RUD processing due to the ordering of + * committed vs unpin operations in bulk insert operations. Hence the reference + * count to ensure only the last caller frees the RUI. + */ +void +xfs_rui_release( + struct xfs_rui_log_item *ruip) +{ + ASSERT(atomic_read(&ruip->rui_refcount) > 0); + if (atomic_dec_and_test(&ruip->rui_refcount)) { + xfs_trans_ail_remove(&ruip->rui_item, SHUTDOWN_LOG_IO_ERROR); + xfs_rui_item_free(ruip); + } +} + STATIC void xfs_rui_item_size( struct xfs_log_item *lip, @@ -140,8 +158,8 @@ STATIC void xfs_rui_item_unlock( struct xfs_log_item *lip) { - if (lip->li_flags & XFS_LI_ABORTED) - xfs_rui_item_free(RUI_ITEM(lip)); + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) + xfs_rui_release(RUI_ITEM(lip)); } /* @@ -233,24 +251,6 @@ xfs_rui_copy_format( return 0; } -/* - * Freeing the RUI requires that we remove it from the AIL if it has already - * been placed there. However, the RUI may not yet have been placed in the AIL - * when called by xfs_rui_release() from RUD processing due to the ordering of - * committed vs unpin operations in bulk insert operations. Hence the reference - * count to ensure only the last caller frees the RUI. - */ -void -xfs_rui_release( - struct xfs_rui_log_item *ruip) -{ - ASSERT(atomic_read(&ruip->rui_refcount) > 0); - if (atomic_dec_and_test(&ruip->rui_refcount)) { - xfs_trans_ail_remove(&ruip->rui_item, SHUTDOWN_LOG_IO_ERROR); - xfs_rui_item_free(ruip); - } -} - static inline struct xfs_rud_log_item *RUD_ITEM(struct xfs_log_item *lip) { return container_of(lip, struct xfs_rud_log_item, rud_item); @@ -331,7 +331,7 @@ xfs_rud_item_unlock( { struct xfs_rud_log_item *rudp = RUD_ITEM(lip); - if (lip->li_flags & XFS_LI_ABORTED) { + if (test_bit(XFS_LI_ABORTED, &lip->li_flags)) { xfs_rui_release(rudp->rud_ruip); kmem_zone_free(xfs_rud_zone, rudp); } diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h index dfee3c991155..52632ab727f7 100644 --- a/fs/xfs/xfs_rtalloc.h +++ b/fs/xfs/xfs_rtalloc.h @@ -23,9 +23,14 @@ struct xfs_mount; struct xfs_trans; +/* + * XXX: Most of the realtime allocation functions deal in units of realtime + * extents, not realtime blocks. This looks funny when paired with the type + * name and screams for a larger cleanup. + */ struct xfs_rtalloc_rec { - xfs_rtblock_t ar_startblock; - xfs_rtblock_t ar_blockcount; + xfs_rtblock_t ar_startext; + xfs_rtblock_t ar_extcount; }; typedef int (*xfs_rtalloc_query_range_fn)( diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c index 056e12b421eb..1cc79907b377 100644 --- a/fs/xfs/xfs_stats.c +++ b/fs/xfs/xfs_stats.c @@ -113,6 +113,7 @@ void xfs_stats_clearall(struct xfsstats __percpu *stats) } } +#ifdef CONFIG_PROC_FS /* legacy quota interfaces */ #ifdef CONFIG_XFS_QUOTA static int xqm_proc_show(struct seq_file *m, void *v) @@ -124,18 +125,6 @@ static int xqm_proc_show(struct seq_file *m, void *v) return 0; } -static int xqm_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, xqm_proc_show, NULL); -} - -static const struct file_operations xqm_proc_fops = { - .open = xqm_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - /* legacy quota stats interface no 2 */ static int xqmstat_proc_show(struct seq_file *m, void *v) { @@ -147,22 +136,8 @@ static int xqmstat_proc_show(struct seq_file *m, void *v) seq_putc(m, '\n'); return 0; } - -static int xqmstat_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, xqmstat_proc_show, NULL); -} - -static const struct file_operations xqmstat_proc_fops = { - .owner = THIS_MODULE, - .open = xqmstat_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; #endif /* CONFIG_XFS_QUOTA */ -#ifdef CONFIG_PROC_FS int xfs_init_procfs(void) { @@ -174,11 +149,9 @@ xfs_init_procfs(void) goto out; #ifdef CONFIG_XFS_QUOTA - if (!proc_create("fs/xfs/xqmstat", 0, NULL, - &xqmstat_proc_fops)) + if (!proc_create_single("fs/xfs/xqmstat", 0, NULL, xqmstat_proc_show)) goto out; - if (!proc_create("fs/xfs/xqm", 0, NULL, - &xqm_proc_fops)) + if (!proc_create_single("fs/xfs/xqm", 0, NULL, xqm_proc_show)) goto out; #endif return 0; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 612c1d5348b3..ed67389f4948 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -63,7 +63,7 @@ #include <linux/parser.h> static const struct super_operations xfs_super_operations; -struct bio_set *xfs_ioend_bioset; +struct bio_set xfs_ioend_bioset; static struct kset *xfs_kset; /* top-level xfs sysfs dir */ #ifdef DEBUG @@ -722,7 +722,7 @@ xfs_close_devices( struct block_device *logdev = mp->m_logdev_targp->bt_bdev; struct dax_device *dax_logdev = mp->m_logdev_targp->bt_daxdev; - xfs_free_buftarg(mp, mp->m_logdev_targp); + xfs_free_buftarg(mp->m_logdev_targp); xfs_blkdev_put(logdev); fs_put_dax(dax_logdev); } @@ -730,11 +730,11 @@ xfs_close_devices( struct block_device *rtdev = mp->m_rtdev_targp->bt_bdev; struct dax_device *dax_rtdev = mp->m_rtdev_targp->bt_daxdev; - xfs_free_buftarg(mp, mp->m_rtdev_targp); + xfs_free_buftarg(mp->m_rtdev_targp); xfs_blkdev_put(rtdev); fs_put_dax(dax_rtdev); } - xfs_free_buftarg(mp, mp->m_ddev_targp); + xfs_free_buftarg(mp->m_ddev_targp); fs_put_dax(dax_ddev); } @@ -808,9 +808,9 @@ xfs_open_devices( out_free_rtdev_targ: if (mp->m_rtdev_targp) - xfs_free_buftarg(mp, mp->m_rtdev_targp); + xfs_free_buftarg(mp->m_rtdev_targp); out_free_ddev_targ: - xfs_free_buftarg(mp, mp->m_ddev_targp); + xfs_free_buftarg(mp->m_ddev_targp); out_close_rtdev: xfs_blkdev_put(rtdev); fs_put_dax(dax_rtdev); @@ -1247,7 +1247,6 @@ xfs_quiesce_attr( STATIC int xfs_test_remount_options( struct super_block *sb, - struct xfs_mount *mp, char *options) { int error = 0; @@ -1278,7 +1277,7 @@ xfs_fs_remount( int error; /* First, check for complete junk; i.e. invalid options */ - error = xfs_test_remount_options(sb, mp, options); + error = xfs_test_remount_options(sb, options); if (error) return error; @@ -1373,7 +1372,6 @@ xfs_fs_remount( */ xfs_restore_resvblks(mp); xfs_log_work_queue(mp); - xfs_queue_eofblocks(mp); /* Recover any CoW blocks that never got remapped. */ error = xfs_reflink_recover_cow(mp); @@ -1383,7 +1381,7 @@ xfs_fs_remount( xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); return error; } - xfs_queue_cowblocks(mp); + xfs_icache_enable_reclaim(mp); /* Create the per-AG metadata reservation pool .*/ error = xfs_fs_reserve_ag_blocks(mp); @@ -1393,8 +1391,13 @@ xfs_fs_remount( /* rw -> ro */ if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & SB_RDONLY)) { + /* + * Cancel background eofb scanning so it cannot race with the + * final log force+buftarg wait and deadlock the remount. + */ + xfs_icache_disable_reclaim(mp); + /* Get rid of any leftover CoW reservations... */ - cancel_delayed_work_sync(&mp->m_cowblocks_work); error = xfs_icache_free_cowblocks(mp, NULL); if (error) { xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); @@ -1417,12 +1420,6 @@ xfs_fs_remount( */ xfs_save_resvblks(mp); - /* - * Cancel background eofb scanning so it cannot race with the - * final log force+buftarg wait and deadlock the remount. - */ - cancel_delayed_work_sync(&mp->m_eofblocks_work); - xfs_quiesce_attr(mp); mp->m_flags |= XFS_MOUNT_RDONLY; } @@ -1442,6 +1439,7 @@ xfs_fs_freeze( { struct xfs_mount *mp = XFS_M(sb); + xfs_icache_disable_reclaim(mp); xfs_save_resvblks(mp); xfs_quiesce_attr(mp); return xfs_sync_sb(mp, true); @@ -1455,6 +1453,7 @@ xfs_fs_unfreeze( xfs_restore_resvblks(mp); xfs_log_work_queue(mp); + xfs_icache_enable_reclaim(mp); return 0; } @@ -1636,6 +1635,17 @@ xfs_fs_fill_super( #endif sb->s_op = &xfs_super_operations; + /* + * Delay mount work if the debug hook is set. This is debug + * instrumention to coordinate simulation of xfs mount failures with + * VFS superblock operations + */ + if (xfs_globals.mount_delay) { + xfs_notice(mp, "Delaying mount for %d seconds.", + xfs_globals.mount_delay); + msleep(xfs_globals.mount_delay * 1000); + } + if (silent) flags |= XFS_MFSI_QUIET; @@ -1691,11 +1701,17 @@ xfs_fs_fill_super( sb->s_flags |= SB_I_VERSION; if (mp->m_flags & XFS_MOUNT_DAX) { + bool rtdev_is_dax = false, datadev_is_dax; + xfs_warn(mp, "DAX enabled. Warning: EXPERIMENTAL, use at your own risk"); - error = bdev_dax_supported(sb, sb->s_blocksize); - if (error) { + datadev_is_dax = bdev_dax_supported(mp->m_ddev_targp->bt_bdev, + sb->s_blocksize); + if (mp->m_rtdev_targp) + rtdev_is_dax = bdev_dax_supported( + mp->m_rtdev_targp->bt_bdev, sb->s_blocksize); + if (!rtdev_is_dax && !datadev_is_dax) { xfs_alert(mp, "DAX unsupported by block device. Turning off DAX."); mp->m_flags &= ~XFS_MOUNT_DAX; @@ -1762,6 +1778,7 @@ xfs_fs_fill_super( out_close_devices: xfs_close_devices(mp); out_free_fsname: + sb->s_fs_info = NULL; xfs_free_fsname(mp); kfree(mp); out: @@ -1779,6 +1796,10 @@ xfs_fs_put_super( { struct xfs_mount *mp = XFS_M(sb); + /* if ->fill_super failed, we have no mount to tear down */ + if (!sb->s_fs_info) + return; + xfs_notice(mp, "Unmounting Filesystem"); xfs_filestream_unmount(mp); xfs_unmountfs(mp); @@ -1788,6 +1809,8 @@ xfs_fs_put_super( xfs_destroy_percpu_counters(mp); xfs_destroy_mount_workqueues(mp); xfs_close_devices(mp); + + sb->s_fs_info = NULL; xfs_free_fsname(mp); kfree(mp); } @@ -1807,6 +1830,9 @@ xfs_fs_nr_cached_objects( struct super_block *sb, struct shrink_control *sc) { + /* Paranoia: catch incorrect calls during mount setup or teardown */ + if (WARN_ON_ONCE(!sb->s_fs_info)) + return 0; return xfs_reclaim_inodes_count(XFS_M(sb)); } @@ -1846,10 +1872,9 @@ MODULE_ALIAS_FS("xfs"); STATIC int __init xfs_init_zones(void) { - xfs_ioend_bioset = bioset_create(4 * MAX_BUF_PER_PAGE, + if (bioset_init(&xfs_ioend_bioset, 4 * MAX_BUF_PER_PAGE, offsetof(struct xfs_ioend, io_inline_bio), - BIOSET_NEED_BVECS); - if (!xfs_ioend_bioset) + BIOSET_NEED_BVECS)) goto out; xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t), @@ -1881,11 +1906,6 @@ xfs_init_zones(void) if (!xfs_trans_zone) goto out_destroy_ifork_zone; - xfs_log_item_desc_zone = - kmem_zone_init(sizeof(struct xfs_log_item_desc), - "xfs_log_item_desc"); - if (!xfs_log_item_desc_zone) - goto out_destroy_trans_zone; /* * The size of the zone allocated buf log item is the maximum @@ -1895,7 +1915,7 @@ xfs_init_zones(void) xfs_buf_item_zone = kmem_zone_init(sizeof(struct xfs_buf_log_item), "xfs_buf_item"); if (!xfs_buf_item_zone) - goto out_destroy_log_item_desc_zone; + goto out_destroy_trans_zone; xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) + ((XFS_EFD_MAX_FAST_EXTENTS - 1) * @@ -1983,8 +2003,6 @@ xfs_init_zones(void) kmem_zone_destroy(xfs_efd_zone); out_destroy_buf_item_zone: kmem_zone_destroy(xfs_buf_item_zone); - out_destroy_log_item_desc_zone: - kmem_zone_destroy(xfs_log_item_desc_zone); out_destroy_trans_zone: kmem_zone_destroy(xfs_trans_zone); out_destroy_ifork_zone: @@ -1998,7 +2016,7 @@ xfs_init_zones(void) out_destroy_log_ticket_zone: kmem_zone_destroy(xfs_log_ticket_zone); out_free_ioend_bioset: - bioset_free(xfs_ioend_bioset); + bioset_exit(&xfs_ioend_bioset); out: return -ENOMEM; } @@ -2023,14 +2041,13 @@ xfs_destroy_zones(void) kmem_zone_destroy(xfs_efi_zone); kmem_zone_destroy(xfs_efd_zone); kmem_zone_destroy(xfs_buf_item_zone); - kmem_zone_destroy(xfs_log_item_desc_zone); kmem_zone_destroy(xfs_trans_zone); kmem_zone_destroy(xfs_ifork_zone); kmem_zone_destroy(xfs_da_state_zone); kmem_zone_destroy(xfs_btree_cur_zone); kmem_zone_destroy(xfs_bmap_free_item_zone); kmem_zone_destroy(xfs_log_ticket_zone); - bioset_free(xfs_ioend_bioset); + bioset_exit(&xfs_ioend_bioset); } STATIC int __init diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 2e9e793a8f9d..aed03da637d4 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -259,12 +259,13 @@ xfs_symlink( * bmapi or the directory create code. */ xfs_defer_init(&dfops, &first_block); + tp->t_agfl_dfops = &dfops; /* * Allocate an inode for the symlink. */ error = xfs_dir_ialloc(&tp, dp, S_IFLNK | (mode & ~S_IFMT), 1, 0, - prid, &ip, NULL); + prid, &ip); if (error) goto out_trans_cancel; @@ -488,16 +489,11 @@ xfs_inactive_symlink_rmt( error = xfs_defer_finish(&tp, &dfops); if (error) goto error_bmap_cancel; - /* - * The first xact was committed, so add the inode to the new one. - * Mark it dirty so it will be logged and moved forward in the log as - * part of every commit. - */ - xfs_trans_ijoin(tp, ip, 0); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + /* * Commit the transaction containing extent freeing and EFDs. */ + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); error = xfs_trans_commit(tp); if (error) { ASSERT(XFS_FORCED_SHUTDOWN(mp)); diff --git a/fs/xfs/xfs_sysctl.h b/fs/xfs/xfs_sysctl.h index 82afee005140..b53a33e69932 100644 --- a/fs/xfs/xfs_sysctl.h +++ b/fs/xfs/xfs_sysctl.h @@ -95,6 +95,7 @@ extern xfs_param_t xfs_params; struct xfs_globals { int log_recovery_delay; /* log recovery delay (secs) */ + int mount_delay; /* mount setup delay (secs) */ bool bug_on_assert; /* BUG() the kernel on assert failure */ }; extern struct xfs_globals xfs_globals; diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c index 8b2ccc234f36..2d5cd2529f8e 100644 --- a/fs/xfs/xfs_sysfs.c +++ b/fs/xfs/xfs_sysfs.c @@ -165,9 +165,40 @@ log_recovery_delay_show( } XFS_SYSFS_ATTR_RW(log_recovery_delay); +STATIC ssize_t +mount_delay_store( + struct kobject *kobject, + const char *buf, + size_t count) +{ + int ret; + int val; + + ret = kstrtoint(buf, 0, &val); + if (ret) + return ret; + + if (val < 0 || val > 60) + return -EINVAL; + + xfs_globals.mount_delay = val; + + return count; +} + +STATIC ssize_t +mount_delay_show( + struct kobject *kobject, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%d\n", xfs_globals.mount_delay); +} +XFS_SYSFS_ATTR_RW(mount_delay); + static struct attribute *xfs_dbg_attrs[] = { ATTR_LIST(bug_on_assert), ATTR_LIST(log_recovery_delay), + ATTR_LIST(mount_delay), NULL, }; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index a982c0b623d0..9d4c4ca24fe6 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -441,8 +441,7 @@ DECLARE_EVENT_CLASS(xfs_buf_item_class, __field(unsigned, bli_recur) __field(int, bli_refcount) __field(unsigned, bli_flags) - __field(void *, li_desc) - __field(unsigned, li_flags) + __field(unsigned long, li_flags) ), TP_fast_assign( __entry->dev = bip->bli_buf->b_target->bt_dev; @@ -455,12 +454,11 @@ DECLARE_EVENT_CLASS(xfs_buf_item_class, __entry->buf_hold = atomic_read(&bip->bli_buf->b_hold); __entry->buf_pincount = atomic_read(&bip->bli_buf->b_pin_count); __entry->buf_lockval = bip->bli_buf->b_sema.count; - __entry->li_desc = bip->bli_item.li_desc; __entry->li_flags = bip->bli_item.li_flags; ), TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " "lock %d flags %s recur %d refcount %d bliflags %s " - "lidesc %p liflags %s", + "liflags %s", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->buf_bno, __entry->buf_len, @@ -471,7 +469,6 @@ DECLARE_EVENT_CLASS(xfs_buf_item_class, __entry->bli_recur, __entry->bli_refcount, __print_flags(__entry->bli_flags, "|", XFS_BLI_FLAGS), - __entry->li_desc, __print_flags(__entry->li_flags, "|", XFS_LI_FLAGS)) ) @@ -506,8 +503,8 @@ DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold_release); DEFINE_BUF_ITEM_EVENT(xfs_trans_binval); DECLARE_EVENT_CLASS(xfs_filestream_class, - TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno), - TP_ARGS(ip, agno), + TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, xfs_agnumber_t agno), + TP_ARGS(mp, ino, agno), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_ino_t, ino) @@ -515,10 +512,10 @@ DECLARE_EVENT_CLASS(xfs_filestream_class, __field(int, streams) ), TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; + __entry->dev = mp->m_super->s_dev; + __entry->ino = ino; __entry->agno = agno; - __entry->streams = xfs_filestream_peek_ag(ip->i_mount, agno); + __entry->streams = xfs_filestream_peek_ag(mp, agno); ), TP_printk("dev %d:%d ino 0x%llx agno %u streams %d", MAJOR(__entry->dev), MINOR(__entry->dev), @@ -528,8 +525,8 @@ DECLARE_EVENT_CLASS(xfs_filestream_class, ) #define DEFINE_FILESTREAM_EVENT(name) \ DEFINE_EVENT(xfs_filestream_class, name, \ - TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno), \ - TP_ARGS(ip, agno)) + TP_PROTO(struct xfs_mount *mp, xfs_ino_t ino, xfs_agnumber_t agno), \ + TP_ARGS(mp, ino, agno)) DEFINE_FILESTREAM_EVENT(xfs_filestream_free); DEFINE_FILESTREAM_EVENT(xfs_filestream_lookup); DEFINE_FILESTREAM_EVENT(xfs_filestream_scan); @@ -1018,7 +1015,7 @@ DECLARE_EVENT_CLASS(xfs_log_item_class, __field(dev_t, dev) __field(void *, lip) __field(uint, type) - __field(uint, flags) + __field(unsigned long, flags) __field(xfs_lsn_t, lsn) ), TP_fast_assign( @@ -1070,7 +1067,7 @@ DECLARE_EVENT_CLASS(xfs_ail_class, __field(dev_t, dev) __field(void *, lip) __field(uint, type) - __field(uint, flags) + __field(unsigned long, flags) __field(xfs_lsn_t, old_lsn) __field(xfs_lsn_t, new_lsn) ), @@ -1750,6 +1747,7 @@ DECLARE_EVENT_CLASS(xfs_attr_class, __field(int, namelen) __field(int, valuelen) __field(xfs_dahash_t, hashval) + __field(int, flags) __field(int, op_flags) ), TP_fast_assign( @@ -1760,10 +1758,11 @@ DECLARE_EVENT_CLASS(xfs_attr_class, __entry->namelen = args->namelen; __entry->valuelen = args->valuelen; __entry->hashval = args->hashval; + __entry->flags = args->flags; __entry->op_flags = args->op_flags; ), TP_printk("dev %d:%d ino 0x%llx name %.*s namelen %d valuelen %d " - "hashval 0x%x op_flags %s", + "hashval 0x%x flags %s op_flags %s", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->namelen, @@ -1771,6 +1770,7 @@ DECLARE_EVENT_CLASS(xfs_attr_class, __entry->namelen, __entry->valuelen, __entry->hashval, + __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS), __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS)) ) @@ -2243,30 +2243,35 @@ struct xfs_defer_pending; struct xfs_defer_ops; DECLARE_EVENT_CLASS(xfs_defer_class, - TP_PROTO(struct xfs_mount *mp, struct xfs_defer_ops *dop), - TP_ARGS(mp, dop), + TP_PROTO(struct xfs_mount *mp, struct xfs_defer_ops *dop, + unsigned long caller_ip), + TP_ARGS(mp, dop, caller_ip), TP_STRUCT__entry( __field(dev_t, dev) __field(void *, dop) __field(char, committed) __field(char, low) + __field(unsigned long, caller_ip) ), TP_fast_assign( __entry->dev = mp ? mp->m_super->s_dev : 0; __entry->dop = dop; __entry->committed = dop->dop_committed; __entry->low = dop->dop_low; + __entry->caller_ip = caller_ip; ), - TP_printk("dev %d:%d ops %p committed %d low %d", + TP_printk("dev %d:%d ops %p committed %d low %d, caller %pS", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->dop, __entry->committed, - __entry->low) + __entry->low, + (char *)__entry->caller_ip) ) #define DEFINE_DEFER_EVENT(name) \ DEFINE_EVENT(xfs_defer_class, name, \ - TP_PROTO(struct xfs_mount *mp, struct xfs_defer_ops *dop), \ - TP_ARGS(mp, dop)) + TP_PROTO(struct xfs_mount *mp, struct xfs_defer_ops *dop, \ + unsigned long caller_ip), \ + TP_ARGS(mp, dop, caller_ip)) DECLARE_EVENT_CLASS(xfs_defer_error_class, TP_PROTO(struct xfs_mount *mp, struct xfs_defer_ops *dop, int error), @@ -2433,6 +2438,8 @@ DEFINE_DEFER_PENDING_EVENT(xfs_defer_pending_abort); #define DEFINE_BMAP_FREE_DEFERRED_EVENT DEFINE_PHYS_EXTENT_DEFERRED_EVENT DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_bmap_free_defer); DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_bmap_free_deferred); +DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_agfl_free_defer); +DEFINE_BMAP_FREE_DEFERRED_EVENT(xfs_agfl_free_deferred); /* rmap tracepoints */ DECLARE_EVENT_CLASS(xfs_rmap_class, @@ -3346,6 +3353,43 @@ TRACE_EVENT(xfs_trans_resv_calc, __entry->logflags) ); +DECLARE_EVENT_CLASS(xfs_trans_class, + TP_PROTO(struct xfs_trans *tp, unsigned long caller_ip), + TP_ARGS(tp, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(uint32_t, tid) + __field(uint32_t, flags) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = tp->t_mountp->m_super->s_dev; + __entry->tid = 0; + if (tp->t_ticket) + __entry->tid = tp->t_ticket->t_tid; + __entry->flags = tp->t_flags; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d trans %x flags 0x%x caller %pS", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->tid, + __entry->flags, + (char *)__entry->caller_ip) +) + +#define DEFINE_TRANS_EVENT(name) \ +DEFINE_EVENT(xfs_trans_class, name, \ + TP_PROTO(struct xfs_trans *tp, unsigned long caller_ip), \ + TP_ARGS(tp, caller_ip)) +DEFINE_TRANS_EVENT(xfs_trans_alloc); +DEFINE_TRANS_EVENT(xfs_trans_cancel); +DEFINE_TRANS_EVENT(xfs_trans_commit); +DEFINE_TRANS_EVENT(xfs_trans_dup); +DEFINE_TRANS_EVENT(xfs_trans_free); +DEFINE_TRANS_EVENT(xfs_trans_roll); +DEFINE_TRANS_EVENT(xfs_trans_add_item); +DEFINE_TRANS_EVENT(xfs_trans_free_items); + #endif /* _TRACE_XFS_H */ #undef TRACE_INCLUDE_PATH diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index d6d8f9d129a7..fc7ba75b8b69 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -31,9 +31,9 @@ #include "xfs_log.h" #include "xfs_trace.h" #include "xfs_error.h" +#include "xfs_defer.h" kmem_zone_t *xfs_trans_zone; -kmem_zone_t *xfs_log_item_desc_zone; #if defined(CONFIG_TRACEPOINTS) static void @@ -79,6 +79,7 @@ xfs_trans_free( xfs_extent_busy_sort(&tp->t_busy); xfs_extent_busy_clear(tp->t_mountp, &tp->t_busy, false); + trace_xfs_trans_free(tp, _RET_IP_); atomic_dec(&tp->t_mountp->m_active_trans); if (!(tp->t_flags & XFS_TRANS_NO_WRITECOUNT)) sb_end_intwrite(tp->t_mountp->m_super); @@ -94,11 +95,13 @@ xfs_trans_free( * blocks. Locks and log items, however, are no inherited. They must * be added to the new transaction explicitly. */ -STATIC xfs_trans_t * +STATIC struct xfs_trans * xfs_trans_dup( - xfs_trans_t *tp) + struct xfs_trans *tp) { - xfs_trans_t *ntp; + struct xfs_trans *ntp; + + trace_xfs_trans_dup(tp, _RET_IP_); ntp = kmem_zone_zalloc(xfs_trans_zone, KM_SLEEP); @@ -127,6 +130,7 @@ xfs_trans_dup( ntp->t_rtx_res = tp->t_rtx_res - tp->t_rtx_res_used; tp->t_rtx_res = tp->t_rtx_res_used; ntp->t_pflags = tp->t_pflags; + ntp->t_agfl_dfops = tp->t_agfl_dfops; xfs_trans_dup_dqinfo(tp, ntp); @@ -283,6 +287,8 @@ xfs_trans_alloc( return error; } + trace_xfs_trans_alloc(tp, _RET_IP_); + *tpp = tp; return 0; } @@ -727,73 +733,52 @@ out: return; } -/* - * Add the given log item to the transaction's list of log items. - * - * The log item will now point to its new descriptor with its li_desc field. - */ +/* Add the given log item to the transaction's list of log items. */ void xfs_trans_add_item( struct xfs_trans *tp, struct xfs_log_item *lip) { - struct xfs_log_item_desc *lidp; - ASSERT(lip->li_mountp == tp->t_mountp); ASSERT(lip->li_ailp == tp->t_mountp->m_ail); + ASSERT(list_empty(&lip->li_trans)); + ASSERT(!test_bit(XFS_LI_DIRTY, &lip->li_flags)); - lidp = kmem_zone_zalloc(xfs_log_item_desc_zone, KM_SLEEP | KM_NOFS); - - lidp->lid_item = lip; - lidp->lid_flags = 0; - list_add_tail(&lidp->lid_trans, &tp->t_items); - - lip->li_desc = lidp; -} - -STATIC void -xfs_trans_free_item_desc( - struct xfs_log_item_desc *lidp) -{ - list_del_init(&lidp->lid_trans); - kmem_zone_free(xfs_log_item_desc_zone, lidp); + list_add_tail(&lip->li_trans, &tp->t_items); + trace_xfs_trans_add_item(tp, _RET_IP_); } /* - * Unlink and free the given descriptor. + * Unlink the log item from the transaction. the log item is no longer + * considered dirty in this transaction, as the linked transaction has + * finished, either by abort or commit completion. */ void xfs_trans_del_item( struct xfs_log_item *lip) { - xfs_trans_free_item_desc(lip->li_desc); - lip->li_desc = NULL; + clear_bit(XFS_LI_DIRTY, &lip->li_flags); + list_del_init(&lip->li_trans); } -/* - * Unlock all of the items of a transaction and free all the descriptors - * of that transaction. - */ +/* Detach and unlock all of the items in a transaction */ void xfs_trans_free_items( struct xfs_trans *tp, xfs_lsn_t commit_lsn, bool abort) { - struct xfs_log_item_desc *lidp, *next; - - list_for_each_entry_safe(lidp, next, &tp->t_items, lid_trans) { - struct xfs_log_item *lip = lidp->lid_item; + struct xfs_log_item *lip, *next; - lip->li_desc = NULL; + trace_xfs_trans_free_items(tp, _RET_IP_); + list_for_each_entry_safe(lip, next, &tp->t_items, li_trans) { + xfs_trans_del_item(lip); if (commit_lsn != NULLCOMMITLSN) lip->li_ops->iop_committing(lip, commit_lsn); if (abort) - lip->li_flags |= XFS_LI_ABORTED; + set_bit(XFS_LI_ABORTED, &lip->li_flags); lip->li_ops->iop_unlock(lip); - - xfs_trans_free_item_desc(lidp); } } @@ -861,7 +846,7 @@ xfs_trans_committed_bulk( xfs_lsn_t item_lsn; if (aborted) - lip->li_flags |= XFS_LI_ABORTED; + set_bit(XFS_LI_ABORTED, &lip->li_flags); item_lsn = lip->li_ops->iop_committed(lip, commit_lsn); /* item_lsn of -1 means the item needs no further processing */ @@ -936,6 +921,11 @@ __xfs_trans_commit( int error = 0; int sync = tp->t_flags & XFS_TRANS_SYNC; + ASSERT(!tp->t_agfl_dfops || + !xfs_defer_has_unfinished_work(tp->t_agfl_dfops) || regrant); + + trace_xfs_trans_commit(tp, _RET_IP_); + /* * If there is nothing to be logged by the transaction, * then unlock all of the items associated with the @@ -991,6 +981,7 @@ out_unreserve: commit_lsn = xfs_log_done(mp, tp->t_ticket, NULL, regrant); if (commit_lsn == -1 && !error) error = -EIO; + tp->t_ticket = NULL; } current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); xfs_trans_free_items(tp, NULLCOMMITLSN, !!error); @@ -1022,6 +1013,8 @@ xfs_trans_cancel( struct xfs_mount *mp = tp->t_mountp; bool dirty = (tp->t_flags & XFS_TRANS_DIRTY); + trace_xfs_trans_cancel(tp, _RET_IP_); + /* * See if the caller is relying on us to shut down the * filesystem. This happens in paths where we detect @@ -1033,17 +1026,19 @@ xfs_trans_cancel( } #ifdef DEBUG if (!dirty && !XFS_FORCED_SHUTDOWN(mp)) { - struct xfs_log_item_desc *lidp; + struct xfs_log_item *lip; - list_for_each_entry(lidp, &tp->t_items, lid_trans) - ASSERT(!(lidp->lid_item->li_type == XFS_LI_EFD)); + list_for_each_entry(lip, &tp->t_items, li_trans) + ASSERT(!(lip->li_type == XFS_LI_EFD)); } #endif xfs_trans_unreserve_and_mod_sb(tp); xfs_trans_unreserve_and_mod_dquots(tp); - if (tp->t_ticket) + if (tp->t_ticket) { xfs_log_done(mp, tp->t_ticket, NULL, false); + tp->t_ticket = NULL; + } /* mark this thread as no longer being in a transaction */ current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); @@ -1067,6 +1062,8 @@ xfs_trans_roll( struct xfs_trans_res tres; int error; + trace_xfs_trans_roll(trans, _RET_IP_); + /* * Copy the critical parameters from one trans to the next. */ diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 9d542dfe0052..29706b8b3bd4 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -27,7 +27,6 @@ struct xfs_efi_log_item; struct xfs_inode; struct xfs_item_ops; struct xfs_log_iovec; -struct xfs_log_item_desc; struct xfs_mount; struct xfs_trans; struct xfs_trans_res; @@ -43,12 +42,12 @@ struct xfs_bud_log_item; typedef struct xfs_log_item { struct list_head li_ail; /* AIL pointers */ + struct list_head li_trans; /* transaction list */ xfs_lsn_t li_lsn; /* last on-disk lsn */ - struct xfs_log_item_desc *li_desc; /* ptr to current desc*/ struct xfs_mount *li_mountp; /* ptr to fs mount */ struct xfs_ail *li_ailp; /* ptr to AIL */ uint li_type; /* item type */ - uint li_flags; /* misc flags */ + unsigned long li_flags; /* misc flags */ struct xfs_buf *li_buf; /* real buffer pointer */ struct list_head li_bio_list; /* buffer item list */ void (*li_cb)(struct xfs_buf *, @@ -64,14 +63,21 @@ typedef struct xfs_log_item { xfs_lsn_t li_seq; /* CIL commit seq */ } xfs_log_item_t; -#define XFS_LI_IN_AIL 0x1 -#define XFS_LI_ABORTED 0x2 -#define XFS_LI_FAILED 0x4 +/* + * li_flags use the (set/test/clear)_bit atomic interfaces because updates can + * race with each other and we don't want to have to use the AIL lock to + * serialise all updates. + */ +#define XFS_LI_IN_AIL 0 +#define XFS_LI_ABORTED 1 +#define XFS_LI_FAILED 2 +#define XFS_LI_DIRTY 3 /* log item dirty in transaction */ #define XFS_LI_FLAGS \ - { XFS_LI_IN_AIL, "IN_AIL" }, \ - { XFS_LI_ABORTED, "ABORTED" }, \ - { XFS_LI_FAILED, "FAILED" } + { (1 << XFS_LI_IN_AIL), "IN_AIL" }, \ + { (1 << XFS_LI_ABORTED), "ABORTED" }, \ + { (1 << XFS_LI_FAILED), "FAILED" }, \ + { (1 << XFS_LI_DIRTY), "DIRTY" } struct xfs_item_ops { void (*iop_size)(xfs_log_item_t *, int *, int *); @@ -111,6 +117,7 @@ typedef struct xfs_trans { struct xlog_ticket *t_ticket; /* log mgr ticket */ struct xfs_mount *t_mountp; /* ptr to fs mount struct */ struct xfs_dquot_acct *t_dqinfo; /* acctg info for dquots */ + struct xfs_defer_ops *t_agfl_dfops; /* optional agfl fixup dfops */ unsigned int t_flags; /* misc flags */ int64_t t_icount_delta; /* superblock icount change */ int64_t t_ifree_delta; /* superblock ifree change */ @@ -228,7 +235,8 @@ struct xfs_efd_log_item *xfs_trans_get_efd(struct xfs_trans *, uint); int xfs_trans_free_extent(struct xfs_trans *, struct xfs_efd_log_item *, xfs_fsblock_t, - xfs_extlen_t, struct xfs_owner_info *); + xfs_extlen_t, struct xfs_owner_info *, + bool); int xfs_trans_commit(struct xfs_trans *); int xfs_trans_roll(struct xfs_trans **); int xfs_trans_roll_inode(struct xfs_trans **, struct xfs_inode *); @@ -242,7 +250,6 @@ void xfs_trans_buf_copy_type(struct xfs_buf *dst_bp, struct xfs_buf *src_bp); extern kmem_zone_t *xfs_trans_zone; -extern kmem_zone_t *xfs_log_item_desc_zone; /* rmap updates */ enum xfs_rmap_intent_type; diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index d4a2445215e6..41e280ef1483 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -32,30 +32,51 @@ #ifdef DEBUG /* * Check that the list is sorted as it should be. + * + * Called with the ail lock held, but we don't want to assert fail with it + * held otherwise we'll lock everything up and won't be able to debug the + * cause. Hence we sample and check the state under the AIL lock and return if + * everything is fine, otherwise we drop the lock and run the ASSERT checks. + * Asserts may not be fatal, so pick the lock back up and continue onwards. */ STATIC void xfs_ail_check( - struct xfs_ail *ailp, - xfs_log_item_t *lip) + struct xfs_ail *ailp, + struct xfs_log_item *lip) { - xfs_log_item_t *prev_lip; + struct xfs_log_item *prev_lip; + struct xfs_log_item *next_lip; + xfs_lsn_t prev_lsn = NULLCOMMITLSN; + xfs_lsn_t next_lsn = NULLCOMMITLSN; + xfs_lsn_t lsn; + bool in_ail; + if (list_empty(&ailp->ail_head)) return; /* - * Check the next and previous entries are valid. + * Sample then check the next and previous entries are valid. */ - ASSERT((lip->li_flags & XFS_LI_IN_AIL) != 0); - prev_lip = list_entry(lip->li_ail.prev, xfs_log_item_t, li_ail); + in_ail = test_bit(XFS_LI_IN_AIL, &lip->li_flags); + prev_lip = list_entry(lip->li_ail.prev, struct xfs_log_item, li_ail); if (&prev_lip->li_ail != &ailp->ail_head) - ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) <= 0); - - prev_lip = list_entry(lip->li_ail.next, xfs_log_item_t, li_ail); - if (&prev_lip->li_ail != &ailp->ail_head) - ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) >= 0); + prev_lsn = prev_lip->li_lsn; + next_lip = list_entry(lip->li_ail.next, struct xfs_log_item, li_ail); + if (&next_lip->li_ail != &ailp->ail_head) + next_lsn = next_lip->li_lsn; + lsn = lip->li_lsn; + if (in_ail && + (prev_lsn == NULLCOMMITLSN || XFS_LSN_CMP(prev_lsn, lsn) <= 0) && + (next_lsn == NULLCOMMITLSN || XFS_LSN_CMP(next_lsn, lsn) >= 0)) + return; + spin_unlock(&ailp->ail_lock); + ASSERT(in_ail); + ASSERT(prev_lsn == NULLCOMMITLSN || XFS_LSN_CMP(prev_lsn, lsn) <= 0); + ASSERT(next_lsn == NULLCOMMITLSN || XFS_LSN_CMP(next_lsn, lsn) >= 0); + spin_lock(&ailp->ail_lock); } #else /* !DEBUG */ #define xfs_ail_check(a,l) @@ -684,7 +705,7 @@ xfs_trans_ail_update_bulk( for (i = 0; i < nr_items; i++) { struct xfs_log_item *lip = log_items[i]; - if (lip->li_flags & XFS_LI_IN_AIL) { + if (test_and_set_bit(XFS_LI_IN_AIL, &lip->li_flags)) { /* check if we really need to move the item */ if (XFS_LSN_CMP(lsn, lip->li_lsn) <= 0) continue; @@ -694,7 +715,6 @@ xfs_trans_ail_update_bulk( if (mlip == lip) mlip_changed = 1; } else { - lip->li_flags |= XFS_LI_IN_AIL; trace_xfs_ail_insert(lip, 0, lsn); } lip->li_lsn = lsn; @@ -725,7 +745,7 @@ xfs_ail_delete_one( trace_xfs_ail_delete(lip, mlip->li_lsn, lip->li_lsn); xfs_ail_delete(ailp, lip); xfs_clear_li_failed(lip); - lip->li_flags &= ~XFS_LI_IN_AIL; + clear_bit(XFS_LI_IN_AIL, &lip->li_flags); lip->li_lsn = 0; return mlip == lip; @@ -761,7 +781,7 @@ xfs_trans_ail_delete( struct xfs_mount *mp = ailp->ail_mount; bool mlip_changed; - if (!(lip->li_flags & XFS_LI_IN_AIL)) { + if (!test_bit(XFS_LI_IN_AIL, &lip->li_flags)) { spin_unlock(&ailp->ail_lock); if (!XFS_FORCED_SHUTDOWN(mp)) { xfs_alert_tag(mp, XFS_PTAG_AILDELETE, diff --git a/fs/xfs/xfs_trans_bmap.c b/fs/xfs/xfs_trans_bmap.c index 14543d93cd4b..230a21df4b12 100644 --- a/fs/xfs/xfs_trans_bmap.c +++ b/fs/xfs/xfs_trans_bmap.c @@ -79,7 +79,7 @@ xfs_trans_log_finish_bmap_update( * 2.) shuts down the filesystem */ tp->t_flags |= XFS_TRANS_DIRTY; - budp->bud_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &budp->bud_item.li_flags); return error; } @@ -158,7 +158,7 @@ xfs_bmap_update_log_item( bmap = container_of(item, struct xfs_bmap_intent, bi_list); tp->t_flags |= XFS_TRANS_DIRTY; - buip->bui_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &buip->bui_item.li_flags); /* * atomic_inc_return gives us the value after the increment; diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c index a5d9dfc45d98..a8ddb4eed279 100644 --- a/fs/xfs/xfs_trans_buf.c +++ b/fs/xfs/xfs_trans_buf.c @@ -40,7 +40,7 @@ xfs_trans_buf_item_match( struct xfs_buf_map *map, int nmaps) { - struct xfs_log_item_desc *lidp; + struct xfs_log_item *lip; struct xfs_buf_log_item *blip; int len = 0; int i; @@ -48,8 +48,8 @@ xfs_trans_buf_item_match( for (i = 0; i < nmaps; i++) len += map[i].bm_len; - list_for_each_entry(lidp, &tp->t_items, lid_trans) { - blip = (struct xfs_buf_log_item *)lidp->lid_item; + list_for_each_entry(lip, &tp->t_items, li_trans) { + blip = (struct xfs_buf_log_item *)lip; if (blip->bli_item.li_type == XFS_LI_BUF && blip->bli_buf->b_target == target && XFS_BUF_ADDR(blip->bli_buf) == map[0].bm_bn && @@ -100,14 +100,10 @@ _xfs_trans_bjoin( atomic_inc(&bip->bli_refcount); /* - * Get a log_item_desc to point at the new item. + * Attach the item to the transaction so we can find it in + * xfs_trans_get_buf() and friends. */ xfs_trans_add_item(tp, &bip->bli_item); - - /* - * Initialize b_fsprivate2 so we can find it with incore_match() - * in xfs_trans_get_buf() and friends above. - */ bp->b_transp = tp; } @@ -391,7 +387,7 @@ xfs_trans_brelse( * If the buffer is dirty within this transaction, we can't * release it until we commit. */ - if (bip->bli_item.li_desc->lid_flags & XFS_LID_DIRTY) + if (test_bit(XFS_LI_DIRTY, &bip->bli_item.li_flags)) return; /* @@ -442,7 +438,7 @@ xfs_trans_brelse( ASSERT(bp->b_pincount == 0); ***/ ASSERT(atomic_read(&bip->bli_refcount) == 0); - ASSERT(!(bip->bli_item.li_flags & XFS_LI_IN_AIL)); + ASSERT(!test_bit(XFS_LI_IN_AIL, &bip->bli_item.li_flags)); ASSERT(!(bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF)); xfs_buf_item_relse(bp); } @@ -542,7 +538,7 @@ xfs_trans_dirty_buf( bip->bli_flags |= XFS_BLI_DIRTY | XFS_BLI_LOGGED; tp->t_flags |= XFS_TRANS_DIRTY; - bip->bli_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &bip->bli_item.li_flags); } /* @@ -626,7 +622,7 @@ xfs_trans_binval( ASSERT(!(bip->__bli_format.blf_flags & XFS_BLF_INODE_BUF)); ASSERT(!(bip->__bli_format.blf_flags & XFS_BLFT_MASK)); ASSERT(bip->__bli_format.blf_flags & XFS_BLF_CANCEL); - ASSERT(bip->bli_item.li_desc->lid_flags & XFS_LID_DIRTY); + ASSERT(test_bit(XFS_LI_DIRTY, &bip->bli_item.li_flags)); ASSERT(tp->t_flags & XFS_TRANS_DIRTY); return; } @@ -642,7 +638,7 @@ xfs_trans_binval( memset(bip->bli_formats[i].blf_data_map, 0, (bip->bli_formats[i].blf_map_size * sizeof(uint))); } - bip->bli_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &bip->bli_item.li_flags); tp->t_flags |= XFS_TRANS_DIRTY; } diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index c3d547211d16..c381c02cca45 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -77,7 +77,7 @@ xfs_trans_log_dquot( ASSERT(XFS_DQ_IS_LOCKED(dqp)); tp->t_flags |= XFS_TRANS_DIRTY; - dqp->q_logitem.qli_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &dqp->q_logitem.qli_item.li_flags); } /* @@ -879,7 +879,7 @@ xfs_trans_log_quotaoff_item( xfs_qoff_logitem_t *qlp) { tp->t_flags |= XFS_TRANS_DIRTY; - qlp->qql_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &qlp->qql_item.li_flags); } STATIC void diff --git a/fs/xfs/xfs_trans_extfree.c b/fs/xfs/xfs_trans_extfree.c index ab438647592a..2f44a08bdf65 100644 --- a/fs/xfs/xfs_trans_extfree.c +++ b/fs/xfs/xfs_trans_extfree.c @@ -68,7 +68,8 @@ xfs_trans_free_extent( struct xfs_efd_log_item *efdp, xfs_fsblock_t start_block, xfs_extlen_t ext_len, - struct xfs_owner_info *oinfo) + struct xfs_owner_info *oinfo, + bool skip_discard) { struct xfs_mount *mp = tp->t_mountp; uint next_extent; @@ -79,9 +80,8 @@ xfs_trans_free_extent( trace_xfs_bmap_free_deferred(tp->t_mountp, agno, 0, agbno, ext_len); - error = xfs_free_extent(tp, start_block, ext_len, oinfo, - XFS_AG_RESV_NONE); - + error = __xfs_free_extent(tp, start_block, ext_len, + oinfo, XFS_AG_RESV_NONE, skip_discard); /* * Mark the transaction dirty, even on error. This ensures the * transaction is aborted, which: @@ -90,7 +90,7 @@ xfs_trans_free_extent( * 2.) shuts down the filesystem */ tp->t_flags |= XFS_TRANS_DIRTY; - efdp->efd_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &efdp->efd_item.li_flags); next_extent = efdp->efd_next_extent; ASSERT(next_extent < efdp->efd_format.efd_nextents); @@ -155,7 +155,7 @@ xfs_extent_free_log_item( free = container_of(item, struct xfs_extent_free_item, xefi_list); tp->t_flags |= XFS_TRANS_DIRTY; - efip->efi_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &efip->efi_item.li_flags); /* * atomic_inc_return gives us the value after the increment; @@ -195,7 +195,7 @@ xfs_extent_free_finish_item( error = xfs_trans_free_extent(tp, done_item, free->xefi_startblock, free->xefi_blockcount, - &free->xefi_oinfo); + &free->xefi_oinfo, free->xefi_skip_discard); kmem_free(free); return error; } @@ -231,9 +231,79 @@ static const struct xfs_defer_op_type xfs_extent_free_defer_type = { .cancel_item = xfs_extent_free_cancel_item, }; +/* + * AGFL blocks are accounted differently in the reserve pools and are not + * inserted into the busy extent list. + */ +STATIC int +xfs_agfl_free_finish_item( + struct xfs_trans *tp, + struct xfs_defer_ops *dop, + struct list_head *item, + void *done_item, + void **state) +{ + struct xfs_mount *mp = tp->t_mountp; + struct xfs_efd_log_item *efdp = done_item; + struct xfs_extent_free_item *free; + struct xfs_extent *extp; + struct xfs_buf *agbp; + int error; + xfs_agnumber_t agno; + xfs_agblock_t agbno; + uint next_extent; + + free = container_of(item, struct xfs_extent_free_item, xefi_list); + ASSERT(free->xefi_blockcount == 1); + agno = XFS_FSB_TO_AGNO(mp, free->xefi_startblock); + agbno = XFS_FSB_TO_AGBNO(mp, free->xefi_startblock); + + trace_xfs_agfl_free_deferred(mp, agno, 0, agbno, free->xefi_blockcount); + + error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp); + if (!error) + error = xfs_free_agfl_block(tp, agno, agbno, agbp, + &free->xefi_oinfo); + + /* + * Mark the transaction dirty, even on error. This ensures the + * transaction is aborted, which: + * + * 1.) releases the EFI and frees the EFD + * 2.) shuts down the filesystem + */ + tp->t_flags |= XFS_TRANS_DIRTY; + set_bit(XFS_LI_DIRTY, &efdp->efd_item.li_flags); + + next_extent = efdp->efd_next_extent; + ASSERT(next_extent < efdp->efd_format.efd_nextents); + extp = &(efdp->efd_format.efd_extents[next_extent]); + extp->ext_start = free->xefi_startblock; + extp->ext_len = free->xefi_blockcount; + efdp->efd_next_extent++; + + kmem_free(free); + return error; +} + + +/* sub-type with special handling for AGFL deferred frees */ +static const struct xfs_defer_op_type xfs_agfl_free_defer_type = { + .type = XFS_DEFER_OPS_TYPE_AGFL_FREE, + .max_items = XFS_EFI_MAX_FAST_EXTENTS, + .diff_items = xfs_extent_free_diff_items, + .create_intent = xfs_extent_free_create_intent, + .abort_intent = xfs_extent_free_abort_intent, + .log_item = xfs_extent_free_log_item, + .create_done = xfs_extent_free_create_done, + .finish_item = xfs_agfl_free_finish_item, + .cancel_item = xfs_extent_free_cancel_item, +}; + /* Register the deferred op type. */ void xfs_extent_free_init_defer_op(void) { xfs_defer_init_op_type(&xfs_extent_free_defer_type); + xfs_defer_init_op_type(&xfs_agfl_free_defer_type); } diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c index 07cea592dc01..f7bd7960a90f 100644 --- a/fs/xfs/xfs_trans_inode.c +++ b/fs/xfs/xfs_trans_inode.c @@ -133,14 +133,13 @@ xfs_trans_log_inode( * set however, then go ahead and bump the i_version counter * unconditionally. */ - if (!(ip->i_itemp->ili_item.li_desc->lid_flags & XFS_LID_DIRTY) && + if (!test_and_set_bit(XFS_LI_DIRTY, &ip->i_itemp->ili_item.li_flags) && IS_I_VERSION(VFS_I(ip))) { if (inode_maybe_inc_iversion(VFS_I(ip), flags & XFS_ILOG_CORE)) flags |= XFS_ILOG_CORE; } tp->t_flags |= XFS_TRANS_DIRTY; - ip->i_itemp->ili_item.li_desc->lid_flags |= XFS_LID_DIRTY; /* * Always OR in the bits from the ili_last_fields field. diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index be24b0c8a332..9717ae74b36d 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -19,7 +19,6 @@ #define __XFS_TRANS_PRIV_H__ struct xfs_log_item; -struct xfs_log_item_desc; struct xfs_mount; struct xfs_trans; struct xfs_ail; @@ -119,7 +118,7 @@ xfs_trans_ail_remove( spin_lock(&ailp->ail_lock); /* xfs_trans_ail_delete() drops the AIL lock */ - if (lip->li_flags & XFS_LI_IN_AIL) + if (test_bit(XFS_LI_IN_AIL, &lip->li_flags)) xfs_trans_ail_delete(ailp, lip, shutdown_type); else spin_unlock(&ailp->ail_lock); @@ -171,11 +170,10 @@ xfs_clear_li_failed( { struct xfs_buf *bp = lip->li_buf; - ASSERT(lip->li_flags & XFS_LI_IN_AIL); + ASSERT(test_bit(XFS_LI_IN_AIL, &lip->li_flags)); lockdep_assert_held(&lip->li_ailp->ail_lock); - if (lip->li_flags & XFS_LI_FAILED) { - lip->li_flags &= ~XFS_LI_FAILED; + if (test_and_clear_bit(XFS_LI_FAILED, &lip->li_flags)) { lip->li_buf = NULL; xfs_buf_rele(bp); } @@ -188,9 +186,8 @@ xfs_set_li_failed( { lockdep_assert_held(&lip->li_ailp->ail_lock); - if (!(lip->li_flags & XFS_LI_FAILED)) { + if (!test_and_set_bit(XFS_LI_FAILED, &lip->li_flags)) { xfs_buf_hold(bp); - lip->li_flags |= XFS_LI_FAILED; lip->li_buf = bp; } } diff --git a/fs/xfs/xfs_trans_refcount.c b/fs/xfs/xfs_trans_refcount.c index 94c1877af834..c7f8e82f5bda 100644 --- a/fs/xfs/xfs_trans_refcount.c +++ b/fs/xfs/xfs_trans_refcount.c @@ -77,7 +77,7 @@ xfs_trans_log_finish_refcount_update( * 2.) shuts down the filesystem */ tp->t_flags |= XFS_TRANS_DIRTY; - cudp->cud_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &cudp->cud_item.li_flags); return error; } @@ -154,7 +154,7 @@ xfs_refcount_update_log_item( refc = container_of(item, struct xfs_refcount_intent, ri_list); tp->t_flags |= XFS_TRANS_DIRTY; - cuip->cui_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &cuip->cui_item.li_flags); /* * atomic_inc_return gives us the value after the increment; diff --git a/fs/xfs/xfs_trans_rmap.c b/fs/xfs/xfs_trans_rmap.c index 9b577beb43d7..5831ca0c270b 100644 --- a/fs/xfs/xfs_trans_rmap.c +++ b/fs/xfs/xfs_trans_rmap.c @@ -117,7 +117,7 @@ xfs_trans_log_finish_rmap_update( * 2.) shuts down the filesystem */ tp->t_flags |= XFS_TRANS_DIRTY; - rudp->rud_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &rudp->rud_item.li_flags); return error; } @@ -175,7 +175,7 @@ xfs_rmap_update_log_item( rmap = container_of(item, struct xfs_rmap_intent, ri_list); tp->t_flags |= XFS_TRANS_DIRTY; - ruip->rui_item.li_desc->lid_flags |= XFS_LID_DIRTY; + set_bit(XFS_LI_DIRTY, &ruip->rui_item.li_flags); /* * atomic_inc_return gives us the value after the increment; |