Patchwork [1/2] ext4: fix performance regression in writeback of random writes

login
register
mail settings
Submitter Colin King
Date Oct. 21, 2013, 6:01 p.m.
Message ID <1382378488-29977-2-git-send-email-colin.king@canonical.com>
Download mbox | patch
Permalink /patch/285261/
State New
Headers show

Comments

Colin King - Oct. 21, 2013, 6:01 p.m.
From: Jan Kara <jack@suse.cz>

BugLink: http://bugs.launchpad.net/bugs/1242812

The Linux Kernel Performance project guys have reported that commit
4e7ea81db5 introduces a performance regression for the following fio
workload:

[global]
direct=0
ioengine=mmap
size=1500M
bs=4k
pre_read=1
numjobs=1
overwrite=1
loops=5
runtime=300
group_reporting
invalidate=0
directory=/mnt/
file_service_type=random:36
file_service_type=random:36

[job0]
startdelay=0
rw=randrw
filename=data0/f1:data0/f2

[job1]
startdelay=0
rw=randrw
filename=data0/f2:data0/f1
...

[job7]
startdelay=0
rw=randrw
filename=data0/f2:data0/f1

The culprit of the problem is that after the commit ext4_writepages()
are more aggressive in writing back pages. Thus we have less consecutive
dirty pages resulting in more seeking.

This increased aggressivity is caused by a bug in the condition
terminating ext4_writepages(). We start writing from the beginning of
the file even if we should have terminated ext4_writepages() because
wbc->nr_to_write <= 0.

After fixing the condition the throughput of the fio workload is about 20%
better than before writeback reorganization.

Reported-by: "Yan, Zheng" <zheng.z.yan@intel.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
(cherry picked from commit 9c12a831d73dd938a22418d70b39aed4feb4bdf2)
Signed-off-by: Colin Ian King <colin.king@canonical.com>
---
 fs/ext4/inode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Patch

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index ea4d188..f2ab232 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2560,7 +2560,7 @@  retry:
 			break;
 	}
 	blk_finish_plug(&plug);
-	if (!ret && !cycled) {
+	if (!ret && !cycled && wbc->nr_to_write > 0) {
 		cycled = 1;
 		mpd.last_page = writeback_index - 1;
 		mpd.first_page = 0;