lkml.org 
[lkml]   [2013]   [Oct]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH] ext4: fix checking on nr_to_write
Date
Commit 4e7ea81db5(ext4: restructure writeback path) introduces
another performance regression on random write:

- one more page may be mapped to ext4 extent in mpage_prepare_extent_to_map,
and will be submitted for I/O so nr_to_write will become -1 before 'done'
is set

- the worse thing is that dirty pages may still be retrieved from page
cache after nr_to_write becomes negative, so lots of small chunks can be
submitted to block device when page writeback is catching up with write path,
and performance is hurted.

On one arm A15 board(arndale) with sata 3.0 SSD(CPU: 1.5GHz dura core, RAM: 2GB),
this patch can improve below test result from 157MB/sec to 174MB/sec(>10%):

dd if=/dev/zero of=./z.img bs=8K count=512K

The above test is actually prototype of block write in bonnie++ utility.

This patch fixes check on nr_to_write in mpage_prepare_extent_to_map()
to make sure nr_to_write won't become negative.

Cc: Ted Tso <tytso@mit.edu>
Cc: Jan Kara <jack@suse.cz>
Cc: linux-ext4@vger.kernel.org
Cc: "linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
---
fs/ext4/inode.c | 20 ++++++++++----------
1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 32c04ab..6a62803 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2356,15 +2356,6 @@ static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd)
if (mpd->map.m_len == 0)
mpd->first_page = page->index;
mpd->next_page = page->index + 1;
- /* Add all dirty buffers to mpd */
- lblk = ((ext4_lblk_t)page->index) <<
- (PAGE_CACHE_SHIFT - blkbits);
- head = page_buffers(page);
- err = mpage_process_page_bufs(mpd, head, head, lblk);
- if (err <= 0)
- goto out;
- err = 0;
-
/*
* Accumulated enough dirty pages? This doesn't apply
* to WB_SYNC_ALL mode. For integrity sync we have to
@@ -2374,9 +2365,18 @@ static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd)
* of the old dirty pages.
*/
if (mpd->wbc->sync_mode == WB_SYNC_NONE &&
- mpd->next_page - mpd->first_page >=
+ mpd->next_page - mpd->first_page >
mpd->wbc->nr_to_write)
goto out;
+
+ /* Add all dirty buffers to mpd */
+ lblk = ((ext4_lblk_t)page->index) <<
+ (PAGE_CACHE_SHIFT - blkbits);
+ head = page_buffers(page);
+ err = mpage_process_page_bufs(mpd, head, head, lblk);
+ if (err <= 0)
+ goto out;
+ err = 0;
}
pagevec_release(&pvec);
cond_resched();
--
1.7.9.5


\
 
 \ /
  Last update: 2013-10-13 19:21    [W:0.081 / U:0.104 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site