diff options
| author | Dave Chinner <dchinner@redhat.com> | 2010-08-24 11:44:56 +1000 |
|---|---|---|
| committer | Dave Chinner <david@fromorbit.com> | 2010-08-24 11:44:56 +1000 |
| commit | efceab1d563153a2b1a6e7d35376241a48126989 (patch) | |
| tree | 1c3f9cf4d947a60f4fa4055ef4847af084e9db0e | |
| parent | 546a1924224078c6f582e68f890b05b387b42653 (diff) | |
| download | linux-efceab1d563153a2b1a6e7d35376241a48126989.tar.gz linux-efceab1d563153a2b1a6e7d35376241a48126989.tar.bz2 linux-efceab1d563153a2b1a6e7d35376241a48126989.zip | |
xfs: handle negative wbc->nr_to_write during sync writeback
During data integrity (WB_SYNC_ALL) writeback, wbc->nr_to_write will
go negative on inodes with more than 1024 dirty pages due to
implementation details of write_cache_pages(). Currently XFS will
abort page clustering in writeback once nr_to_write drops below
zero, and so for data integrity writeback we will do very
inefficient page at a time allocation and IO submission for inodes
with large numbers of dirty pages.
Fix this by only aborting the page clustering code when
wbc->nr_to_write is negative and the sync mode is WB_SYNC_NONE.
Cc: <stable@kernel.org>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
| -rw-r--r-- | fs/xfs/linux-2.6/xfs_aops.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 15412fe15c3a..528be1ba1402 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -852,8 +852,8 @@ xfs_convert_page( SetPageUptodate(page); if (count) { - wbc->nr_to_write--; - if (wbc->nr_to_write <= 0) + if (--wbc->nr_to_write <= 0 && + wbc->sync_mode == WB_SYNC_NONE) done = 1; } xfs_start_page_writeback(page, !page_dirty, count); |