aboutsummaryrefslogtreecommitdiffstats
path: root/mm/page-writeback.c
diff options
context:
space:
mode:
authorRichard Kennedy <richard@rsk.demon.co.uk>2009-06-30 14:41:35 -0400
committerLinus Torvalds <torvalds@linux-foundation.org>2009-06-30 21:56:01 -0400
commitd7831a0bdf06b9f722b947bb0c205ff7d77cebd8 (patch)
tree99afa4a0e30b18724eece3a819867b290728dafe /mm/page-writeback.c
parentdf279ca8966c3de83105428e3391ab17690802a9 (diff)
mm: prevent balance_dirty_pages() from doing too much work
balance_dirty_pages can overreact and move all of the dirty pages to writeback unnecessarily. balance_dirty_pages makes its decision to throttle based on the number of dirty plus writeback pages that are over the calculated limit,so it will continue to move pages even when there are plenty of pages in writeback and less than the threshold still dirty. This allows it to overshoot its limits and move all the dirty pages to writeback while waiting for the drives to catch up and empty the writeback list. A simple fio test easily demonstrates this problem. fio --name=f1 --directory=/disk1 --size=2G -rw=write --name=f2 --directory=/disk2 --size=1G --rw=write --startdelay=10 This is the simplest fix I could find, but I'm not entirely sure that it alone will be enough for all cases. But it certainly is an improvement on my desktop machine writing to 2 disks. Do we need something more for machines with large arrays where bdi_threshold * number_of_drives is greater than the dirty_ratio ? Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> 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 'mm/page-writeback.c')
-rw-r--r--mm/page-writeback.c5
1 files changed, 4 insertions, 1 deletions
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 7b0dcea4935b..7687879253b9 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -541,8 +541,11 @@ static void balance_dirty_pages(struct address_space *mapping)
541 * filesystems (i.e. NFS) in which data may have been 541 * filesystems (i.e. NFS) in which data may have been
542 * written to the server's write cache, but has not yet 542 * written to the server's write cache, but has not yet
543 * been flushed to permanent storage. 543 * been flushed to permanent storage.
544 * Only move pages to writeback if this bdi is over its
545 * threshold otherwise wait until the disk writes catch
546 * up.
544 */ 547 */
545 if (bdi_nr_reclaimable) { 548 if (bdi_nr_reclaimable > bdi_thresh) {
546 writeback_inodes(&wbc); 549 writeback_inodes(&wbc);
547 pages_written += write_chunk - wbc.nr_to_write; 550 pages_written += write_chunk - wbc.nr_to_write;
548 get_dirty_limits(&background_thresh, &dirty_thresh, 551 get_dirty_limits(&background_thresh, &dirty_thresh,