From: Filipe Manana <fdmanana@suse.com>
Date: Sat, 4 Nov 2017 00:16:59 +0000
Patch-mainline: v4.15-rc2
Git-commit: e3b8a4858566a6cc25422fbfdfdd760b13b79280
References: bsc#1061589
Subject: [PATCH] Btrfs: fix reported number of inode blocks after buffered
append writes
The patch from commit a7e3b975a0f9 ("Btrfs: fix reported number of inode
blocks") introduced a regression where if we do a buffered write starting
at position equal to or greater than the file's size and then stat(2) the
file before writeback is triggered, the number of used blocks does not
change (unless there's a prealloc/unwritten extent). Example:
$ xfs_io -f -c "pwrite -S 0xab 0 64K" foobar
$ du -h foobar
0 foobar
$ sync
$ du -h foobar
64K foobar
The first version of that patch didn't had this regression and the second
version, which was the one committed, was made only to address some
performance regression detected by the intel test robots using fs_mark.
This fixes the regression by setting the new delaloc bit in the range, and
doing it at btrfs_dirty_pages() while setting the regular dealloc bit as
well, so that this way we set both bits at once avoiding navigation of the
inode's io tree twice. Doing it at btrfs_dirty_pages() is also the most
meaninful place, as we should set the new dellaloc bit when if we set the
delalloc bit, which happens only if we copied bytes into the pages at
__btrfs_buffered_write().
This was making some of LTP's du tests fail, which can be quickly run
using a command line like the following:
$ ./runltp -q -p -l /ltp.log -f commands -s du -d /mnt
Fixes: a7e3b975a0f9 ("Btrfs: fix reported number of inode blocks")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
---
fs/btrfs/ctree.h | 1
fs/btrfs/extent_io.h | 5 ++--
fs/btrfs/file.c | 43 +++++++++++++++++++++++----------------
fs/btrfs/inode.c | 9 ++++----
fs/btrfs/relocation.c | 3 +-
fs/btrfs/tests/extent-io-tests.c | 6 ++---
fs/btrfs/tests/inode-tests.c | 12 +++++-----
7 files changed, 46 insertions(+), 33 deletions(-)
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -3199,6 +3199,7 @@ int btrfs_start_delalloc_inodes(struct b
int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int delay_iput,
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);
int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
struct btrfs_root *new_root,
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -362,10 +362,11 @@ int convert_extent_bit(struct extent_io_
struct extent_state **cached_state);
static inline int set_extent_delalloc(struct extent_io_tree *tree, u64 start,
- u64 end, struct extent_state **cached_state)
+ u64 end, unsigned int extra_bits,
+ struct extent_state **cached_state)
{
return set_extent_bit(tree, start, end,
- EXTENT_DELALLOC | EXTENT_UPTODATE,
+ EXTENT_DELALLOC | EXTENT_UPTODATE | extra_bits,
NULL, cached_state, GFP_NOFS);
}
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -538,14 +538,34 @@ int btrfs_dirty_pages(struct inode *inod
u64 end_of_last_block;
u64 end_pos = pos + write_bytes;
loff_t isize = i_size_read(inode);
+ unsigned int extra_bits = 0;
start_pos = pos & ~((u64) fs_info->sectorsize - 1);
num_bytes = round_up(write_bytes + pos - start_pos,
fs_info->sectorsize);
end_of_last_block = start_pos + num_bytes - 1;
+
+ if (!btrfs_is_free_space_inode(BTRFS_I(inode))) {
+ if (start_pos >= isize &&
+ !(BTRFS_I(inode)->flags & BTRFS_INODE_PREALLOC)) {
+ /*
+ * There can't be any extents following eof in this case
+ * so just set the delalloc new bit for the range
+ * directly.
+ */
+ extra_bits |= EXTENT_DELALLOC_NEW;
+ } else {
+ err = btrfs_find_new_delalloc_bytes(BTRFS_I(inode),
+ start_pos,
+ num_bytes, cached);
+ if (err)
+ return err;
+ }
+ }
+
err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block,
- cached, 0);
+ extra_bits, cached, 0);
if (err)
return err;
@@ -1473,10 +1493,8 @@ lock_and_cleanup_extent_if_need(struct b
+ round_up(pos + write_bytes - start_pos,
fs_info->sectorsize) - 1;
- if (start_pos < inode->vfs_inode.i_size ||
- (inode->flags & BTRFS_INODE_PREALLOC)) {
+ if (start_pos < inode->vfs_inode.i_size) {
struct btrfs_ordered_extent *ordered;
- unsigned int clear_bits;
lock_extent_bits(&inode->io_tree, start_pos, last_pos,
cached_state);
@@ -1498,19 +1516,10 @@ lock_and_cleanup_extent_if_need(struct b
}
if (ordered)
btrfs_put_ordered_extent(ordered);
- ret = btrfs_find_new_delalloc_bytes(inode, start_pos,
- last_pos - start_pos + 1,
- cached_state);
- clear_bits = EXTENT_DIRTY | EXTENT_DELALLOC |
- EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG;
- if (ret)
- clear_bits |= EXTENT_DELALLOC_NEW | EXTENT_LOCKED;
- clear_extent_bit(&inode->io_tree, start_pos,
- last_pos, clear_bits,
- (clear_bits & EXTENT_LOCKED) ? 1 : 0,
- 0, cached_state, GFP_NOFS);
- if (ret)
- return ret;
+ clear_extent_bit(&inode->io_tree, start_pos, last_pos,
+ EXTENT_DIRTY | EXTENT_DELALLOC |
+ EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
+ 0, 0, cached_state, GFP_NOFS);
*lockstart = start_pos;
*lockend = last_pos;
ret = 1;
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2025,11 +2025,12 @@ static noinline int add_pending_csums(st
}
int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end,
+ unsigned int extra_bits,
struct extent_state **cached_state, int dedupe)
{
WARN_ON((end & (PAGE_SIZE - 1)) == 0);
return set_extent_delalloc(&BTRFS_I(inode)->io_tree, start, end,
- cached_state);
+ extra_bits, cached_state);
}
/* see btrfs_writepage_start_hook for details on why this is required */
@@ -2090,7 +2091,7 @@ again:
goto out;
}
- btrfs_set_extent_delalloc(inode, page_start, page_end, &cached_state,
+ btrfs_set_extent_delalloc(inode, page_start, page_end, 0, &cached_state,
0);
ClearPageChecked(page);
set_page_dirty(page);
@@ -4842,7 +4843,7 @@ again:
EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
0, 0, &cached_state, GFP_NOFS);
- ret = btrfs_set_extent_delalloc(inode, block_start, block_end,
+ ret = btrfs_set_extent_delalloc(inode, block_start, block_end, 0,
&cached_state, 0);
if (ret) {
unlock_extent_cached(io_tree, block_start, block_end,
@@ -9174,7 +9175,7 @@ again:
EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG,
0, 0, &cached_state, GFP_NOFS);
- ret = btrfs_set_extent_delalloc(inode, page_start, end,
+ ret = btrfs_set_extent_delalloc(inode, page_start, end, 0,
&cached_state, 0);
if (ret) {
unlock_extent_cached(io_tree, page_start, page_end,
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3261,7 +3261,8 @@ static int relocate_file_extent_cluster(
nr++;
}
- btrfs_set_extent_delalloc(inode, page_start, page_end, NULL, 0);
+ btrfs_set_extent_delalloc(inode, page_start, page_end, 0, NULL,
+ 0);
set_page_dirty(page);
unlock_extent(&BTRFS_I(inode)->io_tree,
--- a/fs/btrfs/tests/extent-io-tests.c
+++ b/fs/btrfs/tests/extent-io-tests.c
@@ -114,7 +114,7 @@ static int test_find_delalloc(u32 sector
* |--- delalloc ---|
* |--- search ---|
*/
- set_extent_delalloc(&tmp, 0, sectorsize - 1, NULL);
+ set_extent_delalloc(&tmp, 0, sectorsize - 1, 0, NULL);
start = 0;
end = 0;
found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
@@ -145,7 +145,7 @@ static int test_find_delalloc(u32 sector
test_msg("Couldn't find the locked page\n");
goto out_bits;
}
- set_extent_delalloc(&tmp, sectorsize, max_bytes - 1, NULL);
+ set_extent_delalloc(&tmp, sectorsize, max_bytes - 1, 0, NULL);
start = test_start;
end = 0;
found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
@@ -200,7 +200,7 @@ static int test_find_delalloc(u32 sector
*
* We are re-using our test_start from above since it works out well.
*/
- set_extent_delalloc(&tmp, max_bytes, total_dirty - 1, NULL);
+ set_extent_delalloc(&tmp, max_bytes, total_dirty - 1, 0, NULL);
start = test_start;
end = 0;
found = find_lock_delalloc_range(inode, &tmp, locked_page, &start,
--- a/fs/btrfs/tests/inode-tests.c
+++ b/fs/btrfs/tests/inode-tests.c
@@ -968,7 +968,7 @@ static int test_extent_accounting(u32 se
btrfs_test_inode_set_ops(inode);
/* [BTRFS_MAX_EXTENT_SIZE] */
- ret = btrfs_set_extent_delalloc(inode, 0, BTRFS_MAX_EXTENT_SIZE - 1,
+ 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);
@@ -984,7 +984,7 @@ static int test_extent_accounting(u32 se
/* [BTRFS_MAX_EXTENT_SIZE][sectorsize] */
ret = btrfs_set_extent_delalloc(inode, BTRFS_MAX_EXTENT_SIZE,
BTRFS_MAX_EXTENT_SIZE + sectorsize - 1,
- NULL, 0);
+ 0, NULL, 0);
if (ret) {
test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
goto out;
@@ -1018,7 +1018,7 @@ static int test_extent_accounting(u32 se
ret = btrfs_set_extent_delalloc(inode, BTRFS_MAX_EXTENT_SIZE >> 1,
(BTRFS_MAX_EXTENT_SIZE >> 1)
+ sectorsize - 1,
- NULL, 0);
+ 0, NULL, 0);
if (ret) {
test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
goto out;
@@ -1036,7 +1036,7 @@ static int test_extent_accounting(u32 se
ret = btrfs_set_extent_delalloc(inode,
BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize,
(BTRFS_MAX_EXTENT_SIZE << 1) + 3 * sectorsize - 1,
- NULL, 0);
+ 0, NULL, 0);
if (ret) {
test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
goto out;
@@ -1053,7 +1053,7 @@ static int test_extent_accounting(u32 se
*/
ret = btrfs_set_extent_delalloc(inode,
BTRFS_MAX_EXTENT_SIZE + sectorsize,
- BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, NULL, 0);
+ BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, 0, NULL, 0);
if (ret) {
test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
goto out;
@@ -1089,7 +1089,7 @@ static int test_extent_accounting(u32 se
*/
ret = btrfs_set_extent_delalloc(inode,
BTRFS_MAX_EXTENT_SIZE + sectorsize,
- BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, NULL, 0);
+ BTRFS_MAX_EXTENT_SIZE + 2 * sectorsize - 1, 0, NULL, 0);
if (ret) {
test_msg("btrfs_set_extent_delalloc returned %d\n", ret);
goto out;