summaryrefslogtreecommitdiffstats
path: root/fs/mpage.c
diff options
context:
space:
mode:
authorMiquel van Smoorenburg <mikevs@xs4all.net>2009-01-06 23:39:02 +0100
committerLinus Torvalds <torvalds@linux-foundation.org>2009-01-07 00:58:59 +0100
commit38c8e6180939e5619140b2e9e479cb26029ff8b1 (patch)
tree1980f3dadfa02ac6c1fc2ad7236205af54f7972a /fs/mpage.c
parentoom: print triggering task's cpuset and mems allowed (diff)
downloadlinux-38c8e6180939e5619140b2e9e479cb26029ff8b1.tar.xz
linux-38c8e6180939e5619140b2e9e479cb26029ff8b1.zip
do_mpage_readpage(): don't submit lots of small bios on boundary
While tracing I/O patterns with blktrace (a great tool) a few weeks ago I identified a minor issue in fs/mpage.c As the comment above mpage_readpages() says, a fs's get_block function will set BH_Boundary when it maps a block just before a block for which extra I/O is required. Since get_block() can map a range of pages, for all these pages the BH_Boundary flag will be set. But we only need to push what I/O we have accumulated at the last block of this range. This makes do_mpage_readpage() send out the largest possible bio instead of a bunch of page-sized ones in the BH_Boundary case. Signed-off-by: Miquel van Smoorenburg <mikevs@xs4all.net> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'fs/mpage.c')
-rw-r--r--fs/mpage.c5
1 files changed, 4 insertions, 1 deletions
diff --git a/fs/mpage.c b/fs/mpage.c
index 552b80b3facc..46e977efd50a 100644
--- a/fs/mpage.c
+++ b/fs/mpage.c
@@ -308,7 +308,10 @@ alloc_new:
goto alloc_new;
}
- if (buffer_boundary(map_bh) || (first_hole != blocks_per_page))
+ relative_block = block_in_file - *first_logical_block;
+ nblocks = map_bh->b_size >> blkbits;
+ if ((buffer_boundary(map_bh) && relative_block == nblocks) ||
+ (first_hole != blocks_per_page))
bio = mpage_bio_submit(READ, bio);
else
*last_block_in_bio = blocks[blocks_per_page - 1];