diff options
Diffstat (limited to 'fs')
-rw-r--r-- | fs/btrfs/extent-tree.c | 27 | ||||
-rw-r--r-- | fs/btrfs/scrub.c | 2 | ||||
-rw-r--r-- | fs/ext3/namei.c | 2 | ||||
-rw-r--r-- | fs/ext4/extents.c | 23 | ||||
-rw-r--r-- | fs/ext4/extents_status.c | 51 | ||||
-rw-r--r-- | fs/ext4/inode.c | 7 | ||||
-rw-r--r-- | fs/ext4/namei.c | 2 | ||||
-rw-r--r-- | fs/fuse/dir.c | 51 | ||||
-rw-r--r-- | fs/nfs/nfs4xdr.c | 21 | ||||
-rw-r--r-- | fs/nfsd/vfs.c | 5 | ||||
-rw-r--r-- | fs/ocfs2/refcounttree.c | 5 | ||||
-rw-r--r-- | fs/open.c | 2 | ||||
-rw-r--r-- | fs/proc/vmcore.c | 2 | ||||
-rw-r--r-- | fs/super.c | 25 | ||||
-rw-r--r-- | fs/xfs/xfs_dinode.h | 3 | ||||
-rw-r--r-- | fs/xfs/xfs_inode.c | 31 | ||||
-rw-r--r-- | fs/xfs/xfs_log_recover.c | 13 |
17 files changed, 191 insertions, 81 deletions
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 0236de7..1204c8e 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -7466,6 +7466,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, int err = 0; int ret; int level; + bool root_dropped = false; path = btrfs_alloc_path(); if (!path) { @@ -7523,6 +7524,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, while (1) { btrfs_tree_lock(path->nodes[level]); btrfs_set_lock_blocking(path->nodes[level]); + path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; ret = btrfs_lookup_extent_info(trans, root, path->nodes[level]->start, @@ -7538,6 +7540,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, break; btrfs_tree_unlock(path->nodes[level]); + path->locks[level] = 0; WARN_ON(wc->refs[level] != 1); level--; } @@ -7552,11 +7555,6 @@ int btrfs_drop_snapshot(struct btrfs_root *root, wc->reada_count = BTRFS_NODEPTRS_PER_BLOCK(root); while (1) { - if (!for_reloc && btrfs_need_cleaner_sleep(root)) { - pr_debug("btrfs: drop snapshot early exit\n"); - err = -EAGAIN; - goto out_end_trans; - } ret = walk_down_tree(trans, root, path, wc); if (ret < 0) { @@ -7584,7 +7582,8 @@ int btrfs_drop_snapshot(struct btrfs_root *root, } BUG_ON(wc->level == 0); - if (btrfs_should_end_transaction(trans, tree_root)) { + if (btrfs_should_end_transaction(trans, tree_root) || + (!for_reloc && btrfs_need_cleaner_sleep(root))) { ret = btrfs_update_root(trans, tree_root, &root->root_key, root_item); @@ -7595,6 +7594,12 @@ int btrfs_drop_snapshot(struct btrfs_root *root, } btrfs_end_transaction_throttle(trans, tree_root); + if (!for_reloc && btrfs_need_cleaner_sleep(root)) { + pr_debug("btrfs: drop snapshot early exit\n"); + err = -EAGAIN; + goto out_free; + } + trans = btrfs_start_transaction(tree_root, 0); if (IS_ERR(trans)) { err = PTR_ERR(trans); @@ -7639,12 +7644,22 @@ int btrfs_drop_snapshot(struct btrfs_root *root, free_extent_buffer(root->commit_root); btrfs_put_fs_root(root); } + root_dropped = true; out_end_trans: btrfs_end_transaction_throttle(trans, tree_root); out_free: kfree(wc); btrfs_free_path(path); out: + /* + * So if we need to stop dropping the snapshot for whatever reason we + * need to make sure to add it back to the dead root list so that we + * keep trying to do the work later. This also cleans up roots if we + * don't have it in the radix (like when we recover after a power fail + * or unmount) so we don't leak memory. + */ + if (root_dropped == false) + btrfs_add_dead_root(root); if (err) btrfs_std_error(root->fs_info, err); return err; diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 4ba2a69..64a157b 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -2495,7 +2495,7 @@ again: ret = scrub_extent(sctx, extent_logical, extent_len, extent_physical, extent_dev, flags, generation, extent_mirror_num, - extent_physical); + extent_logical - logical + physical); if (ret) goto out; diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 998ea11..1194b1f 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1780,11 +1780,11 @@ retry: inode->i_op = &ext3_file_inode_operations; inode->i_fop = &ext3_file_operations; ext3_set_aops(inode); + d_tmpfile(dentry, inode); err = ext3_orphan_add(handle, inode); if (err) goto err_drop_inode; mark_inode_dirty(inode); - d_tmpfile(dentry, inode); unlock_new_inode(inode); } ext3_journal_stop(handle); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 7097b0f..a618738 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2835,6 +2835,9 @@ again: err = -EIO; break; } + /* Yield here to deal with large extent trees. + * Should be a no-op if we did IO above. */ + cond_resched(); if (WARN_ON(i + 1 > depth)) { err = -EIO; break; @@ -4261,8 +4264,8 @@ got_allocated_blocks: /* not a good idea to call discard here directly, * but otherwise we'd need to call it every free() */ ext4_discard_preallocations(inode); - ext4_free_blocks(handle, inode, NULL, ext4_ext_pblock(&newex), - ext4_ext_get_actual_len(&newex), fb_flags); + ext4_free_blocks(handle, inode, NULL, newblock, + EXT4_C2B(sbi, allocated_clusters), fb_flags); goto out2; } @@ -4382,8 +4385,9 @@ out2: } out3: - trace_ext4_ext_map_blocks_exit(inode, flags, map, err ? err : allocated); - + trace_ext4_ext_map_blocks_exit(inode, flags, map, + err ? err : allocated); + ext4_es_lru_add(inode); return err ? err : allocated; } @@ -4405,9 +4409,20 @@ void ext4_ext_truncate(handle_t *handle, struct inode *inode) last_block = (inode->i_size + sb->s_blocksize - 1) >> EXT4_BLOCK_SIZE_BITS(sb); +retry: err = ext4_es_remove_extent(inode, last_block, EXT_MAX_BLOCKS - last_block); + if (err == ENOMEM) { + cond_resched(); + congestion_wait(BLK_RW_ASYNC, HZ/50); + goto retry; + } + if (err) { + ext4_std_error(inode->i_sb, err); + return; + } err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCKS - 1); + ext4_std_error(inode->i_sb, err); } static void ext4_falloc_update_inode(struct inode *inode, diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c index 4b8df7f..91cb110 100644 --- a/fs/ext4/extents_status.c +++ b/fs/ext4/extents_status.c @@ -148,6 +148,8 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t end); static int __es_try_to_reclaim_extents(struct ext4_inode_info *ei, int nr_to_scan); +static int __ext4_es_shrink(struct ext4_sb_info *sbi, int nr_to_scan, + struct ext4_inode_info *locked_ei); int __init ext4_init_es(void) { @@ -665,7 +667,13 @@ int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk, err = __es_remove_extent(inode, lblk, end); if (err != 0) goto error; +retry: err = __es_insert_extent(inode, &newes); + if (err == -ENOMEM && __ext4_es_shrink(EXT4_SB(inode->i_sb), 1, + EXT4_I(inode))) + goto retry; + if (err == -ENOMEM && !ext4_es_is_delayed(&newes)) + err = 0; error: write_unlock(&EXT4_I(inode)->i_es_lock); @@ -744,8 +752,10 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk, struct extent_status orig_es; ext4_lblk_t len1, len2; ext4_fsblk_t block; - int err = 0; + int err; +retry: + err = 0; es = __es_tree_search(&tree->root, lblk); if (!es) goto out; @@ -780,6 +790,10 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk, if (err) { es->es_lblk = orig_es.es_lblk; es->es_len = orig_es.es_len; + if ((err == -ENOMEM) && + __ext4_es_shrink(EXT4_SB(inode->i_sb), 1, + EXT4_I(inode))) + goto retry; goto out; } } else { @@ -889,22 +903,14 @@ static int ext4_inode_touch_time_cmp(void *priv, struct list_head *a, return -1; } -static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) +static int __ext4_es_shrink(struct ext4_sb_info *sbi, int nr_to_scan, + struct ext4_inode_info *locked_ei) { - struct ext4_sb_info *sbi = container_of(shrink, - struct ext4_sb_info, s_es_shrinker); struct ext4_inode_info *ei; struct list_head *cur, *tmp; LIST_HEAD(skiped); - int nr_to_scan = sc->nr_to_scan; int ret, nr_shrunk = 0; - ret = percpu_counter_read_positive(&sbi->s_extent_cache_cnt); - trace_ext4_es_shrink_enter(sbi->s_sb, nr_to_scan, ret); - - if (!nr_to_scan) - return ret; - spin_lock(&sbi->s_es_lru_lock); /* @@ -933,7 +939,7 @@ static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) continue; } - if (ei->i_es_lru_nr == 0) + if (ei->i_es_lru_nr == 0 || ei == locked_ei) continue; write_lock(&ei->i_es_lock); @@ -952,6 +958,27 @@ static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) list_splice_tail(&skiped, &sbi->s_es_lru); spin_unlock(&sbi->s_es_lru_lock); + if (locked_ei && nr_shrunk == 0) + nr_shrunk = __es_try_to_reclaim_extents(ei, nr_to_scan); + + return nr_shrunk; +} + +static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) +{ + struct ext4_sb_info *sbi = container_of(shrink, + struct ext4_sb_info, s_es_shrinker); + int nr_to_scan = sc->nr_to_scan; + int ret, nr_shrunk; + + ret = percpu_counter_read_positive(&sbi->s_extent_cache_cnt); + trace_ext4_es_shrink_enter(sbi->s_sb, nr_to_scan, ret); + + if (!nr_to_scan) + return ret; + + nr_shrunk = __ext4_es_shrink(sbi, nr_to_scan, NULL); + ret = percpu_counter_read_positive(&sbi->s_extent_cache_cnt); trace_ext4_es_shrink_exit(sbi->s_sb, nr_shrunk, ret); return ret; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 98b9bff..ba33c67 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -514,10 +514,9 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode, "logical block %lu\n", inode->i_ino, flags, map->m_len, (unsigned long) map->m_lblk); - ext4_es_lru_add(inode); - /* Lookup extent status tree firstly */ if (ext4_es_lookup_extent(inode, map->m_lblk, &es)) { + ext4_es_lru_add(inode); if (ext4_es_is_written(&es) || ext4_es_is_unwritten(&es)) { map->m_pblk = ext4_es_pblock(&es) + map->m_lblk - es.es_lblk; @@ -1529,11 +1528,9 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock, "logical block %lu\n", inode->i_ino, map->m_len, (unsigned long) map->m_lblk); - ext4_es_lru_add(inode); - /* Lookup extent status tree firstly */ if (ext4_es_lookup_extent(inode, iblock, &es)) { - + ext4_es_lru_add(inode); if (ext4_es_is_hole(&es)) { retval = 0; down_read((&EXT4_I(inode)->i_data_sem)); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 234b834..35f55a0 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2316,11 +2316,11 @@ retry: inode->i_op = &ext4_file_inode_operations; inode->i_fop = &ext4_file_operations; ext4_set_aops(inode); + d_tmpfile(dentry, inode); err = ext4_orphan_add(handle, inode); if (err) goto err_drop_inode; mark_inode_dirty(inode); - d_tmpfile(dentry, inode); unlock_new_inode(inode); } if (handle) diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 0eda527..72a5d5b 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1223,30 +1223,46 @@ static int fuse_direntplus_link(struct file *file, if (name.name[1] == '.' && name.len == 2) return 0; } + + if (invalid_nodeid(o->nodeid)) + return -EIO; + if (!fuse_valid_type(o->attr.mode)) + return -EIO; + fc = get_fuse_conn(dir); name.hash = full_name_hash(name.name, name.len); dentry = d_lookup(parent, &name); - if (dentry && dentry->d_inode) { + if (dentry) { inode = dentry->d_inode; - if (get_node_id(inode) == o->nodeid) { + if (!inode) { + d_drop(dentry); + } else if (get_node_id(inode) != o->nodeid || + ((o->attr.mode ^ inode->i_mode) & S_IFMT)) { + err = d_invalidate(dentry); + if (err) + goto out; + } else if (is_bad_inode(inode)) { + err = -EIO; + goto out; + } else { struct fuse_inode *fi; fi = get_fuse_inode(inode); spin_lock(&fc->lock); fi->nlookup++; spin_unlock(&fc->lock); + fuse_change_attributes(inode, &o->attr, + entry_attr_timeout(o), + attr_version); + /* * The other branch to 'found' comes via fuse_iget() * which bumps nlookup inside */ goto found; } - err = d_invalidate(dentry); - if (err) - goto out; dput(dentry); - dentry = NULL; } dentry = d_alloc(parent, &name); @@ -1259,25 +1275,30 @@ static int fuse_direntplus_link(struct file *file, if (!inode) goto out; - alias = d_materialise_unique(dentry, inode); - err = PTR_ERR(alias); - if (IS_ERR(alias)) - goto out; + if (S_ISDIR(inode->i_mode)) { + mutex_lock(&fc->inst_mutex); + alias = fuse_d_add_directory(dentry, inode); + mutex_unlock(&fc->inst_mutex); + err = PTR_ERR(alias); + if (IS_ERR(alias)) { + iput(inode); + goto out; + } + } else { + alias = d_splice_alias(inode, dentry); + } + if (alias) { dput(dentry); dentry = alias; } found: - fuse_change_attributes(inode, &o->attr, entry_attr_timeout(o), - attr_version); - fuse_change_entry_timeout(dentry, o); err = 0; out: - if (dentry) - dput(dentry); + dput(dentry); return err; } diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 0abfb846..3850b01 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -999,6 +999,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, __be32 *p; __be32 *q; int len; + uint32_t bmval_len = 2; uint32_t bmval0 = 0; uint32_t bmval1 = 0; uint32_t bmval2 = 0; @@ -1010,7 +1011,7 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, * = 40 bytes, plus any contribution from variable-length fields * such as owner/group. */ - len = 20; + len = 8; /* Sigh */ if (iap->ia_valid & ATTR_SIZE) @@ -1040,8 +1041,6 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, } len += 4 + (XDR_QUADLEN(owner_grouplen) << 2); } - if (label) - len += 4 + 4 + 4 + (XDR_QUADLEN(label->len) << 2); if (iap->ia_valid & ATTR_ATIME_SET) len += 16; else if (iap->ia_valid & ATTR_ATIME) @@ -1050,15 +1049,22 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, len += 16; else if (iap->ia_valid & ATTR_MTIME) len += 4; + if (label) { + len += 4 + 4 + 4 + (XDR_QUADLEN(label->len) << 2); + bmval_len = 3; + } + + len += bmval_len << 2; p = reserve_space(xdr, len); /* * We write the bitmap length now, but leave the bitmap and the attribute * buffer length to be backfilled at the end of this routine. */ - *p++ = cpu_to_be32(3); + *p++ = cpu_to_be32(bmval_len); q = p; - p += 4; + /* Skip bitmap entries + attrlen */ + p += bmval_len + 1; if (iap->ia_valid & ATTR_SIZE) { bmval0 |= FATTR4_WORD0_SIZE; @@ -1112,10 +1118,11 @@ static void encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, len, ((char *)p - (char *)q) + 4); BUG(); } - len = (char *)p - (char *)q - 16; *q++ = htonl(bmval0); *q++ = htonl(bmval1); - *q++ = htonl(bmval2); + if (bmval_len == 3) + *q++ = htonl(bmval2); + len = (char *)p - (char *)(q + 1); *q = htonl(len); /* out: */ diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 8ff6a00..c827acb 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -830,9 +830,10 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, flags = O_WRONLY|O_LARGEFILE; } *filp = dentry_open(&path, flags, current_cred()); - if (IS_ERR(*filp)) + if (IS_ERR(*filp)) { host_err = PTR_ERR(*filp); - else { + *filp = NULL; + } else { host_err = ima_file_check(*filp, may_flags); if (may_flags & NFSD_MAY_64BIT_COOKIE) diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 998b17e..9f6b96a 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -2965,6 +2965,11 @@ int ocfs2_duplicate_clusters_by_page(handle_t *handle, to = map_end & (PAGE_CACHE_SIZE - 1); page = find_or_create_page(mapping, page_index, GFP_NOFS); + if (!page) { + ret = -ENOMEM; + mlog_errno(ret); + break; + } /* * In case PAGE_CACHE_SIZE <= CLUSTER_SIZE, This page @@ -844,6 +844,8 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o if ((flags & O_TMPFILE_MASK) != O_TMPFILE) return -EINVAL; acc_mode = MAY_OPEN | ACC_MODE(flags); + if (!(acc_mode & MAY_WRITE)) + return -EINVAL; } else if (flags & O_PATH) { /* * If we have O_PATH in the open flag. Then we diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 2850317..a1a16eb 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -223,7 +223,7 @@ static inline char *alloc_elfnotes_buf(size_t notes_sz) * regions in the 1st kernel pointed to by PT_LOAD entries) into * virtually contiguous user-space in ELF layout. */ -#ifdef CONFIG_MMU +#if defined(CONFIG_MMU) && !defined(CONFIG_S390) static int mmap_vmcore(struct file *file, struct vm_area_struct *vma) { size_t size = vma->vm_end - vma->vm_start; @@ -336,19 +336,19 @@ EXPORT_SYMBOL(deactivate_super); * and want to turn it into a full-blown active reference. grab_super() * is called with sb_lock held and drops it. Returns 1 in case of * success, 0 if we had failed (superblock contents was already dead or - * dying when grab_super() had been called). + * dying when grab_super() had been called). Note that this is only + * called for superblocks not in rundown mode (== ones still on ->fs_supers + * of their type), so increment of ->s_count is OK here. */ static int grab_super(struct super_block *s) __releases(sb_lock) { - if (atomic_inc_not_zero(&s->s_active)) { - spin_unlock(&sb_lock); - return 1; - } - /* it's going away */ s->s_count++; spin_unlock(&sb_lock); - /* wait for it to die */ down_write(&s->s_umount); + if ((s->s_flags & MS_BORN) && atomic_inc_not_zero(&s->s_active)) { + put_super(s); + return 1; + } up_write(&s->s_umount); put_super(s); return 0; @@ -463,11 +463,6 @@ retry: destroy_super(s); s = NULL; } - down_write(&old->s_umount); - if (unlikely(!(old->s_flags & MS_BORN))) { - deactivate_locked_super(old); - goto retry; - } return old; } } @@ -660,10 +655,10 @@ restart: if (hlist_unhashed(&sb->s_instances)) continue; if (sb->s_bdev == bdev) { - if (grab_super(sb)) /* drops sb_lock */ - return sb; - else + if (!grab_super(sb)) goto restart; + up_write(&sb->s_umount); + return sb; } } spin_unlock(&sb_lock); diff --git a/fs/xfs/xfs_dinode.h b/fs/xfs/xfs_dinode.h index 07d735a..e5869b5 100644 --- a/fs/xfs/xfs_dinode.h +++ b/fs/xfs/xfs_dinode.h @@ -39,6 +39,9 @@ typedef struct xfs_timestamp { * There is a very similar struct icdinode in xfs_inode which matches the * layout of the first 96 bytes of this structure, but is kept in native * format instead of big endian. + * + * Note: di_flushiter is only used by v1/2 inodes - it's effectively a zeroed + * padding field for v3 inodes. */ typedef struct xfs_dinode { __be16 di_magic; /* inode magic # = XFS_DINODE_MAGIC */ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index b78481f..bb262c2 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -896,7 +896,6 @@ xfs_dinode_to_disk( to->di_projid_lo = cpu_to_be16(from->di_projid_lo); to->di_projid_hi = cpu_to_be16(from->di_projid_hi); memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad)); - to->di_flushiter = cpu_to_be16(from->di_flushiter); to->di_atime.t_sec = cpu_to_be32(from->di_atime.t_sec); to->di_atime.t_nsec = cpu_to_be32(from->di_atime.t_nsec); to->di_mtime.t_sec = cpu_to_be32(from->di_mtime.t_sec); @@ -924,6 +923,9 @@ xfs_dinode_to_disk( to->di_lsn = cpu_to_be64(from->di_lsn); memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2)); uuid_copy(&to->di_uuid, &from->di_uuid); + to->di_flushiter = 0; + } else { + to->di_flushiter = cpu_to_be16(from->di_flushiter); } } @@ -1029,10 +1031,14 @@ xfs_dinode_calc_crc( /* * Read the disk inode attributes into the in-core inode structure. * - * If we are initialising a new inode and we are not utilising the - * XFS_MOUNT_IKEEP inode cluster mode, we can simple build the new inode core - * with a random generation number. If we are keeping inodes around, we need to - * read the inode cluster to get the existing generation number off disk. + * For version 5 superblocks, if we are initialising a new inode and we are not + * utilising the XFS_MOUNT_IKEEP inode cluster mode, we can simple build the new + * inode core with a random generation number. If we are keeping inodes around, + * we need to read the inode cluster to get the existing generation number off + * disk. Further, if we are using version 4 superblocks (i.e. v1/v2 inode + * format) then log recovery is dependent on the di_flushiter field being + * initialised from the current on-disk value and hence we must also read the + * inode off disk. */ int xfs_iread( @@ -1054,6 +1060,7 @@ xfs_iread( /* shortcut IO on inode allocation if possible */ if ((iget_flags & XFS_IGET_CREATE) && + xfs_sb_version_hascrc(&mp->m_sb) && !(mp->m_flags & XFS_MOUNT_IKEEP)) { /* initialise the on-disk inode core */ memset(&ip->i_d, 0, sizeof(ip->i_d)); @@ -2882,12 +2889,18 @@ xfs_iflush_int( __func__, ip->i_ino, ip->i_d.di_forkoff, ip); goto corrupt_out; } + /* - * bump the flush iteration count, used to detect flushes which - * postdate a log record during recovery. This is redundant as we now - * log every change and hence this can't happen. Still, it doesn't hurt. + * Inode item log recovery for v1/v2 inodes are dependent on the + * di_flushiter count for correct sequencing. We bump the flush + * iteration count so we can detect flushes which postdate a log record + * during recovery. This is redundant as we now log every change and + * hence this can't happen but we need to still do it to ensure + * backwards compatibility with old kernels that predate logging all + * inode changes. */ - ip->i_d.di_flushiter++; + if (ip->i_d.di_version < 3) + ip->i_d.di_flushiter++; /* * Copy the dirty parts of the inode into the on-disk diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 6fcc910a..7681b19 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -2592,8 +2592,16 @@ xlog_recover_inode_pass2( goto error; } - /* Skip replay when the on disk inode is newer than the log one */ - if (dicp->di_flushiter < be16_to_cpu(dip->di_flushiter)) { + /* + * di_flushiter is only valid for v1/2 inodes. All changes for v3 inodes + * are transactional and if ordering is necessary we can determine that + * more accurately by the LSN field in the V3 inode core. Don't trust + * the inode versions we might be changing them here - use the + * superblock flag to determine whether we need to look at di_flushiter + * to skip replay when the on disk inode is newer than the log one + */ + if (!xfs_sb_version_hascrc(&mp->m_sb) && + dicp->di_flushiter < be16_to_cpu(dip->di_flushiter)) { /* * Deal with the wrap case, DI_MAX_FLUSH is less * than smaller numbers @@ -2608,6 +2616,7 @@ xlog_recover_inode_pass2( goto error; } } + /* Take the opportunity to reset the flush iteration count */ dicp->di_flushiter = 0; |