From: Balbir Singh <balbir@linux.vnet.ibm.com>
To: Shaohua Li <shaohua.li@intel.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
akpm@linux-foundation.org, fengguang.wu@intel.com,
kosaki.motohiro@jp.fujitsu.com
Subject: Re: [PATCH]vmscan: handle underflow for get_scan_ratio
Date: Tue, 30 Mar 2010 17:26:56 +0530 [thread overview]
Message-ID: <661de9471003300456r6527f17au6d70bd0d2ee0a941@mail.gmail.com> (raw)
In-Reply-To: <20100330055304.GA2983@sli10-desk.sh.intel.com>
On Tue, Mar 30, 2010 at 11:23 AM, Shaohua Li <shaohua.li@intel.com> wrote:
> Commit 84b18490d1f1bc7ed5095c929f78bc002eb70f26 introduces a regression.
> With it, our tmpfs test always oom. The test has a lot of rotated anon
> pages and cause percent[0] zero. Actually the percent[0] is a very small
> value, but our calculation round it to zero. The commit makes vmscan
> completely skip anon pages and cause oops.
> An option is if percent[x] is zero in get_scan_ratio(), forces it
> to 1. See below patch.
> But the offending commit still changes behavior. Without the commit, we scan
> all pages if priority is zero, below patch doesn't fix this. Don't know if
> It's required to fix this too.
>
> Signed-off-by: Shaohua Li <shaohua.li@intel.com>
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 79c8098..d5cc34e 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1604,6 +1604,18 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc,
> /* Normalize to percentages */
> percent[0] = 100 * ap / (ap + fp + 1);
> percent[1] = 100 - percent[0];
> + /*
> + * if percent[x] is small and rounded to 0, this case doesn't mean we
> + * should skip scan. Give it at least 1% share.
> + */
> + if (percent[0] == 0) {
> + percent[0] = 1;
> + percent[1] = 99;
> + }
> + if (percent[1] == 0) {
> + percent[0] = 99;
> + percent[1] = 1;
> + }
> }
>
Can you please post the meminfo before and after the changes (diff
maybe?). Can you also please share the ap and fp data from which the
percent figures are being calculated Is your swappiness set to 60? Can
you please share the OOM/panic message as well.
Balbir
--
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>
prev parent reply other threads:[~2010-03-30 11:57 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-30 5:53 [PATCH]vmscan: handle underflow for get_scan_ratio Shaohua Li
2010-03-30 6:08 ` KOSAKI Motohiro
2010-03-30 6:32 ` Shaohua Li
2010-03-30 6:40 ` KOSAKI Motohiro
2010-03-30 6:53 ` Shaohua Li
2010-03-30 7:31 ` KOSAKI Motohiro
2010-03-30 8:13 ` Shaohua Li
2010-03-31 4:53 ` Shaohua Li
2010-03-31 5:38 ` KOSAKI Motohiro
2010-03-31 5:51 ` Wu Fengguang
2010-03-31 6:00 ` KOSAKI Motohiro
2010-03-31 6:03 ` Wu Fengguang
2010-04-01 22:16 ` Andrew Morton
2010-04-02 9:13 ` KOSAKI Motohiro
2010-04-06 1:22 ` Wu Fengguang
2010-04-06 3:36 ` Rik van Riel
2010-03-31 5:53 ` KOSAKI Motohiro
2010-04-02 6:50 ` Shaohua Li
2010-04-02 9:14 ` KOSAKI Motohiro
2010-04-02 9:24 ` Shaohua Li
2010-04-04 14:19 ` KOSAKI Motohiro
2010-04-06 1:25 ` Shaohua Li
2010-04-06 1:36 ` KOSAKI Motohiro
2010-04-06 1:50 ` Wu Fengguang
2010-04-06 2:06 ` KOSAKI Motohiro
2010-04-06 2:30 ` Wu Fengguang
2010-04-06 2:58 ` KOSAKI Motohiro
2010-04-06 3:31 ` Wu Fengguang
2010-04-06 3:40 ` Rik van Riel
2010-04-06 4:49 ` Wu Fengguang
2010-04-06 5:09 ` Shaohua Li
2010-04-04 0:48 ` Wu Fengguang
2010-04-06 1:27 ` Shaohua Li
2010-04-06 5:03 ` Wu Fengguang
2010-04-06 5:36 ` Shaohua Li
2010-04-09 6:51 ` Shaohua Li
2010-04-09 21:20 ` Andrew Morton
2010-04-09 21:25 ` Rik van Riel
2010-04-13 1:30 ` KOSAKI Motohiro
2010-04-13 2:42 ` Rik van Riel
2010-04-13 7:55 ` KOSAKI Motohiro
2010-04-13 8:55 ` KOSAKI Motohiro
2010-04-14 1:27 ` Shaohua Li
2010-04-15 3:25 ` KOSAKI Motohiro
2010-04-12 1:57 ` Shaohua Li
2010-03-31 5:41 ` Wu Fengguang
2010-03-30 10:17 ` Minchan Kim
2010-03-30 10:25 ` KOSAKI Motohiro
2010-03-30 11:56 ` Balbir Singh [this message]
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=661de9471003300456r6527f17au6d70bd0d2ee0a941@mail.gmail.com \
--to=balbir@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=fengguang.wu@intel.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=shaohua.li@intel.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;
as well as URLs for NNTP newsgroup(s).