From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
To: Rik van Riel <riel@redhat.com>
Cc: kosaki.motohiro@jp.fujitsu.com, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, lwoodman@redhat.com,
kosaki.motohiro@fujitsu.co.jp,
Tomasz Chmielewski <mangoo@wpkg.org>,
akpm@linux-foundation.org
Subject: Re: [PATCH] vmscan: do not evict inactive pages when skipping an active list scan
Date: Thu, 26 Nov 2009 11:50:17 +0900 (JST) [thread overview]
Message-ID: <20091126110340.5A62.A69D9226@jp.fujitsu.com> (raw)
In-Reply-To: <20091125133752.2683c3e4@bree.surriel.com>
> In AIM7 runs, recent kernels start swapping out anonymous pages
> well before they should. This is due to shrink_list falling
> through to shrink_inactive_list if !inactive_anon_is_low(zone, sc),
> when all we really wanted to do is pre-age some anonymous pages to
> give them extra time to be referenced while on the inactive list.
>
> The obvious fix is to make sure that shrink_list does not fall
> through to scanning/reclaiming inactive pages when we called it
> to scan one of the active lists.
>
> This change should be safe because the loop in shrink_zone ensures
> that we will still shrink the anon and file inactive lists whenever
> we should.
Good catch!
>
> Reported-by: Larry Woodman <lwoodman@redhat.com>
> Signed-off-by: Rik van Riel <riel@redhat.com>
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 777af57..ec4dfda 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1469,13 +1469,15 @@ static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
> {
> int file = is_file_lru(lru);
>
> - if (lru == LRU_ACTIVE_FILE && inactive_file_is_low(zone, sc)) {
> - shrink_active_list(nr_to_scan, zone, sc, priority, file);
> + if (lru == LRU_ACTIVE_FILE) {
> + if (inactive_file_is_low(zone, sc))
> + shrink_active_list(nr_to_scan, zone, sc, priority, file);
> return 0;
> }
>
> - if (lru == LRU_ACTIVE_ANON && inactive_anon_is_low(zone, sc)) {
> - shrink_active_list(nr_to_scan, zone, sc, priority, file);
> + if (lru == LRU_ACTIVE_ANON) {
> + if (inactive_file_is_low(zone, sc))
This inactive_file_is_low() should be inactive_anon_is_low().
cut-n-paste programming often makes similar mistake. probaby we need make
more cleanup to this function.
How about this? (this is incremental patch from you)
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
---
mm/vmscan.c | 17 ++++++++++-------
1 files changed, 10 insertions(+), 7 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index a8f61c0..80e94a2 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1467,22 +1467,25 @@ static int inactive_file_is_low(struct zone *zone, struct scan_control *sc)
return low;
}
+static int inactive_list_is_low(struct zone *zone, struct scan_control *sc, int file)
+{
+ if (file)
+ return inactive_file_is_low(zone, sc);
+ else
+ return inactive_anon_is_low(zone, sc);
+}
+
static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
struct zone *zone, struct scan_control *sc, int priority)
{
int file = is_file_lru(lru);
- if (lru == LRU_ACTIVE_FILE) {
- if (inactive_file_is_low(zone, sc))
+ if (is_active_lru(lru)) {
+ if (inactive_list_is_low(zone, sc, file))
shrink_active_list(nr_to_scan, zone, sc, priority, file);
return 0;
}
- if (lru == LRU_ACTIVE_ANON) {
- if (inactive_file_is_low(zone, sc))
- shrink_active_list(nr_to_scan, zone, sc, priority, file);
- return 0;
- }
return shrink_inactive_list(nr_to_scan, zone, sc, priority, file);
}
--
1.6.5.2
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2009-11-26 2:50 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-11-25 18:37 Rik van Riel
2009-11-25 20:35 ` Johannes Weiner
2009-11-25 20:47 ` Rik van Riel
2009-11-26 2:50 ` KOSAKI Motohiro [this message]
2009-11-26 2:57 ` Rik van Riel
2009-11-30 22:00 ` [RFC] high system time & lock contention running large mixed workload Larry Woodman
2009-12-01 10:04 ` Andrea Arcangeli
2009-12-01 12:31 ` KOSAKI Motohiro
2009-12-01 12:46 ` Andrea Arcangeli
2009-12-02 2:02 ` KOSAKI Motohiro
2009-12-02 2:04 ` Rik van Riel
2009-12-02 2:00 ` Rik van Riel
2009-12-01 12:23 ` KOSAKI Motohiro
2009-12-01 16:41 ` Larry Woodman
2009-12-02 2:20 ` Rik van Riel
2009-12-02 2:41 ` KOSAKI Motohiro
2009-12-03 22:14 ` Larry Woodman
2009-12-04 0:29 ` Rik van Riel
2009-12-04 21:26 ` Larry Woodman
2009-12-06 21:04 ` Rik van Riel
2009-12-04 0:36 ` KOSAKI Motohiro
2009-12-04 19:31 ` Larry Woodman
2009-12-02 2:55 ` [PATCH] Clear reference bit although page isn't mapped KOSAKI Motohiro
2009-12-02 3:07 ` Rik van Riel
2009-12-02 3:28 ` [PATCH] Replace page_mapping_inuse() with page_mapped() KOSAKI Motohiro
2009-12-02 4:57 ` Rik van Riel
2009-12-02 11:07 ` Johannes Weiner
2009-12-02 1:55 ` [RFC] high system time & lock contention running large mixed workload Rik van Riel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20091126110340.5A62.A69D9226@jp.fujitsu.com \
--to=kosaki.motohiro@jp.fujitsu.com \
--cc=akpm@linux-foundation.org \
--cc=kosaki.motohiro@fujitsu.co.jp \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lwoodman@redhat.com \
--cc=mangoo@wpkg.org \
--cc=riel@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox