From: Hyunhee Kim <hyunhee.kim@samsung.com>
To: 'Michal Hocko' <mhocko@suse.cz>,
'Anton Vorontsov' <anton@enomsg.org>,
linux-mm@kvack.org, akpm@linux-foundation.org, rob@landley.net,
kamezawa.hiroyu@jp.fujitsu.com, hannes@cmpxchg.org,
rientjes@google.com, kirill@shutemov.name,
'Kyungmin Park' <kyungmin.park@samsung.com>,
'Minchan Kim' <minchan@kernel.org>
Subject: [PATCH v2] vmpressure: consider "scanned < reclaimed" case when calculating a pressure level.
Date: Thu, 27 Jun 2013 15:12:10 +0900 [thread overview]
Message-ID: <009601ce72fd$427eed70$c77cc850$%kim@samsung.com> (raw)
In-Reply-To: <20130626073557.GD29127@bbox>
In vmpressure, the pressure level is calculated based on the ratio
of how many pages were scanned vs. reclaimed in a given time window.
However, there is a possibility that "scanned < reclaimed" in such a
case, when reclaiming ends by fatal signal in shrink_inactive_list.
So, with this patch, we just return "low" level when "scanned < reclaimed"
happens not to have userland miss reclaim activity.
Signed-off-by: Hyunhee Kim <hyunhee.kim@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
---
mm/vmpressure.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/mm/vmpressure.c b/mm/vmpressure.c
index 736a601..8c60cad 100644
--- a/mm/vmpressure.c
+++ b/mm/vmpressure.c
@@ -119,6 +119,14 @@ static enum vmpressure_levels vmpressure_calc_level(unsigned long scanned,
unsigned long pressure;
/*
+ * This could happen, in such a case, when reclaiming ends by fatal
+ * signal in shrink_inactive_list(). In this case, return
+ * VMPRESSURE_LOW not to have userland miss reclaim activity.
+ */
+ if (reclaimed > scanned)
+ return VMPRESSURE_LOW;
+
+ /*
* We calculate the ratio (in percents) of how many pages were
* scanned vs. reclaimed in a given time frame (window). Note that
* time is in VM reclaimer's "ticks", i.e. number of pages
--
1.7.9.5
--
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:[~2013-06-27 6:12 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-17 11:30 [PATCH v3] memcg: event control at vmpressure Hyunhee Kim
2013-06-17 13:15 ` Michal Hocko
2013-06-18 6:10 ` Hyunhee Kim
2013-06-18 8:00 ` Hyunhee Kim
2013-06-18 11:01 ` Michal Hocko
2013-06-19 11:25 ` Hyunhee Kim
2013-06-19 11:59 ` Michal Hocko
2013-06-19 11:31 ` [PATCH v4] " Hyunhee Kim
2013-06-19 12:53 ` Michal Hocko
2013-06-20 2:13 ` Hyunhee Kim
2013-06-20 2:17 ` [PATCH v5] " Hyunhee Kim
2013-06-20 12:16 ` Michal Hocko
2013-06-21 0:21 ` [PATCH v6] " Hyunhee Kim
2013-06-21 0:24 ` Hyunhee Kim
2013-06-21 1:22 ` Minchan Kim
2013-06-21 9:19 ` Michal Hocko
2013-06-21 11:02 ` Hyunhee Kim
2013-06-21 11:54 ` Hyunhee Kim
2013-06-21 12:40 ` [PATCH v7] " Hyunhee Kim
2013-06-21 16:27 ` [PATCH v6] " Minchan Kim
2013-06-21 16:44 ` Minchan Kim
2013-06-22 0:27 ` Anton Vorontsov
2013-06-22 1:28 ` Hyunhee Kim
2013-06-26 7:47 ` Minchan Kim
2013-06-21 22:35 ` Anton Vorontsov
2013-06-22 4:36 ` Hyunhee Kim
2013-06-22 4:51 ` Hyunhee Kim
2013-06-22 5:50 ` [PATCH] memcg: consider "scanned < reclaimed" case when calculating Hyunhee Kim
2013-06-22 7:34 ` [PATCH] memcg: add interface to specify thresholds of vmpressure Hyunhee Kim
2013-06-25 20:46 ` Michal Hocko
2013-06-26 7:39 ` Minchan Kim
2013-06-26 7:50 ` Kyungmin Park
2013-06-26 8:03 ` Minchan Kim
2013-06-26 7:35 ` [PATCH] memcg: consider "scanned < reclaimed" case when calculating Minchan Kim
2013-06-27 6:12 ` Hyunhee Kim [this message]
2013-06-27 9:37 ` [PATCH v2] vmpressure: consider "scanned < reclaimed" case when calculating a pressure level Michal Hocko
2013-06-27 15:35 ` Minchan Kim
2013-06-27 16:11 ` Michal Hocko
2013-06-27 18:05 ` Anton Vorontsov
2013-06-28 12:17 ` Michal Hocko
2013-06-27 23:54 ` Minchan Kim
2013-06-28 7:43 ` [PATCH v3] " Hyunhee Kim
2013-06-28 12:26 ` Michal Hocko
2013-06-28 12:24 ` [PATCH v2] " Michal Hocko
2013-06-28 13:55 ` Minchan Kim
2013-06-28 15:17 ` Michal Hocko
2013-06-27 18:33 ` Anton Vorontsov
2013-06-26 7:34 ` [PATCH v6] memcg: event control at vmpressure Minchan Kim
2013-06-26 7:31 ` Minchan Kim
2013-06-25 16:07 ` Michal Hocko
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='009601ce72fd$427eed70$c77cc850$%kim@samsung.com' \
--to=hyunhee.kim@samsung.com \
--cc=akpm@linux-foundation.org \
--cc=anton@enomsg.org \
--cc=hannes@cmpxchg.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kirill@shutemov.name \
--cc=kyungmin.park@samsung.com \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.cz \
--cc=minchan@kernel.org \
--cc=rientjes@google.com \
--cc=rob@landley.net \
/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