From ec29ed5b407d618a8128f5942aade9e1758aa14b Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Wed, 23 Feb 2011 16:23:20 -0500 Subject: Btrfs: fix fiemap bugs with delalloc The Btrfs fiemap code wasn't properly returning delalloc extents, so applications that trust fiemap to decide if there are holes in the file see holes instead of delalloc. This reworks the btrfs fiemap code, adding a get_extent helper that searches for delalloc ranges and also adding a helper for extent_fiemap that skips past holes in the file. Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 126 ++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 124 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8d392ed73d57..44b926646e33 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1913,7 +1913,7 @@ static int btrfs_clean_io_failures(struct inode *inode, u64 start) private = 0; if (count_range_bits(&BTRFS_I(inode)->io_failure_tree, &private, - (u64)-1, 1, EXTENT_DIRTY)) { + (u64)-1, 1, EXTENT_DIRTY, 0)) { ret = get_state_private(&BTRFS_I(inode)->io_failure_tree, start, &private_failure); if (ret == 0) { @@ -5282,6 +5282,128 @@ out: return em; } +struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *page, + size_t pg_offset, u64 start, u64 len, + int create) +{ + struct extent_map *em; + struct extent_map *hole_em = NULL; + u64 range_start = start; + u64 end; + u64 found; + u64 found_end; + int err = 0; + + em = btrfs_get_extent(inode, page, pg_offset, start, len, create); + if (IS_ERR(em)) + return em; + if (em) { + /* + * if our em maps to a hole, there might + * actually be delalloc bytes behind it + */ + if (em->block_start != EXTENT_MAP_HOLE) + return em; + else + hole_em = em; + } + + /* check to see if we've wrapped (len == -1 or similar) */ + end = start + len; + if (end < start) + end = (u64)-1; + else + end -= 1; + + em = NULL; + + /* ok, we didn't find anything, lets look for delalloc */ + found = count_range_bits(&BTRFS_I(inode)->io_tree, &range_start, + end, len, EXTENT_DELALLOC, 1); + found_end = range_start + found; + if (found_end < range_start) + found_end = (u64)-1; + + /* + * we didn't find anything useful, return + * the original results from get_extent() + */ + if (range_start > end || found_end <= start) { + em = hole_em; + hole_em = NULL; + goto out; + } + + /* adjust the range_start to make sure it doesn't + * go backwards from the start they passed in + */ + range_start = max(start,range_start); + found = found_end - range_start; + + if (found > 0) { + u64 hole_start = start; + u64 hole_len = len; + + em = alloc_extent_map(GFP_NOFS); + if (!em) { + err = -ENOMEM; + goto out; + } + /* + * when btrfs_get_extent can't find anything it + * returns one huge hole + * + * make sure what it found really fits our range, and + * adjust to make sure it is based on the start from + * the caller + */ + if (hole_em) { + u64 calc_end = extent_map_end(hole_em); + + if (calc_end <= start || (hole_em->start > end)) { + free_extent_map(hole_em); + hole_em = NULL; + } else { + hole_start = max(hole_em->start, start); + hole_len = calc_end - hole_start; + } + } + em->bdev = NULL; + if (hole_em && range_start > hole_start) { + /* our hole starts before our delalloc, so we + * have to return just the parts of the hole + * that go until the delalloc starts + */ + em->len = min(hole_len, + range_start - hole_start); + em->start = hole_start; + em->orig_start = hole_start; + /* + * don't adjust block start at all, + * it is fixed at EXTENT_MAP_HOLE + */ + em->block_start = hole_em->block_start; + em->block_len = hole_len; + } else { + em->start = range_start; + em->len = found; + em->orig_start = range_start; + em->block_start = EXTENT_MAP_DELALLOC; + em->block_len = found; + } + } else if (hole_em) { + return hole_em; + } +out: + + free_extent_map(hole_em); + if (err) { + free_extent_map(em); + return ERR_PTR(err); + } + return em; +} + static struct extent_map *btrfs_new_extent_direct(struct inode *inode, u64 start, u64 len) { @@ -6104,7 +6226,7 @@ out: static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, __u64 start, __u64 len) { - return extent_fiemap(inode, fieinfo, start, len, btrfs_get_extent); + return extent_fiemap(inode, fieinfo, start, len, btrfs_get_extent_fiemap); } int btrfs_readpage(struct file *file, struct page *page) -- cgit v1.2.2 From b4966b7770349deb05e3dd2bd2c65d2d044abbbb Mon Sep 17 00:00:00 2001 From: Daniel J Blueman Date: Wed, 9 Mar 2011 16:46:42 +0000 Subject: btrfs: fix dip leak The btrfs DIO code leaks dip structs when dip->csums allocation fails; bio->bi_end_io isn't set at the point where the free_ordered branch is consequently taken, thus bio_endio doesn't call the function which would free it in the normal case. Fix. Signed-off-by: Daniel J Blueman Acked-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 44b926646e33..e7a8303328b2 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6058,6 +6058,7 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode, if (!skip_sum) { dip->csums = kmalloc(sizeof(u32) * bio->bi_vcnt, GFP_NOFS); if (!dip->csums) { + kfree(dip); ret = -ENOMEM; goto free_ordered; } -- cgit v1.2.2 From 7e6b6465e6efbca3985258996be9c189da96c8bf Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Fri, 18 Feb 2011 09:21:17 +0000 Subject: btrfs: fix not enough reserved space btrfs_link() will insert 3 items(inode ref, dir name item and dir index item) into the b+ tree and update 2 items(its inode, and parent's inode) in the b+ tree. So we should reserve space for these 5 items, not 3 items. Reported-by: Tsutomu Itoh Signed-off-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index e7a8303328b2..db67821ccac2 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4823,10 +4823,11 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, goto fail; /* - * 1 item for inode ref + * 2 items for inode and inode ref * 2 items for dir items + * 1 item for parent inode */ - trans = btrfs_start_transaction(root, 3); + trans = btrfs_start_transaction(root, 5); if (IS_ERR(trans)) { err = PTR_ERR(trans); goto fail; -- cgit v1.2.2 From f17b6042073e7000a90063f7edbca59a5bd1caa2 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Sat, 29 Jan 2011 18:43:30 +0530 Subject: fs: Remove i_nlink check from file system link callback Now that VFS check for inode->i_nlink == 0 and returns proper error, remove similar check from file system Signed-off-by: Aneesh Kumar K.V Signed-off-by: Al Viro --- fs/btrfs/inode.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 0efdb65953c5..c23f050f47c2 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4806,9 +4806,6 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, int err; int drop_inode = 0; - if (inode->i_nlink == 0) - return -ENOENT; - /* do not allow sys_link's with other subvols of the same device */ if (root->objectid != BTRFS_I(inode)->root->objectid) return -EPERM; -- cgit v1.2.2