From: Michal Hocko <mhocko@suse.cz>
To: linux-mm@kvack.org
Cc: Ying Han <yinghan@google.com>,
Johannes Weiner <hannes@cmpxchg.org>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Rik van Riel <riel@redhat.com>, Hugh Dickins <hughd@google.com>,
Mel Gorman <mgorman@suse.de>,
Glauber Costa <glommer@parallels.com>
Subject: Re: [RFC 0/3] soft reclaim rework
Date: Tue, 9 Apr 2013 17:37:42 +0200 [thread overview]
Message-ID: <20130409153742.GL29860@dhcp22.suse.cz> (raw)
In-Reply-To: <1365509595-665-1-git-send-email-mhocko@suse.cz>
On Tue 09-04-13 14:13:12, Michal Hocko wrote:
[...]
> 2) kbuild test showed more or less the same results
> usage_in_bytes
> Base
> Group A Group B
> Median 394817536 395634688
>
> Patches applied
> median 483481600 302131200
>
> A is kept closer to the soft limit again. There is some fluctuation
> around the limit because kbuild creates a lot of short lived processes.
> Base: pgscan_kswapd_dma32 1648718 pgsteal_kswapd_dma32 1510749
> Patched: pgscan_kswapd_dma32 2042065 pgsteal_kswapd_dma32 1667745
OK, so I have patched the base version with the patch bellow which
uncovers soft reclaim scanning and reclaim and guess what:
Base: pgscan_kswapd_dma32 3710092 pgsteal_kswapd_dma32 3225191
Patched: pgscan_kswapd_dma32 1846700 pgsteal_kswapd_dma32 1442232
Base: pgscan_direct_dma32 2417683 pgsteal_direct_dma32 459702
Patched: pgscan_direct_dma32 1839331 pgsteal_direct_dma32 244338
The numbers are obviously timing dependent (wrt. previous run ~10% for
the patched kernel) but the ~1/2 half wrt. the base kernel seems real
we just haven't seen it previously because it wasn't accounted. I guess
this can be attributed to prio-0 soft reclaim behavior and a lot of
dirty pages on the LRU.
> The differences are much bigger now so it would be interesting how much
> has been scanned/reclaimed during soft reclaim in the base kernel.
---
next prev parent reply other threads:[~2013-04-09 15:37 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-09 12:13 Michal Hocko
2013-04-09 12:13 ` [RFC 1/3] memcg: integrate soft reclaim tighter with zone shrinking code Michal Hocko
2013-04-09 13:08 ` Johannes Weiner
2013-04-09 13:31 ` Michal Hocko
2013-04-09 13:57 ` Glauber Costa
2013-04-09 14:22 ` Michal Hocko
2013-04-09 16:45 ` Kamezawa Hiroyuki
2013-04-09 17:05 ` Michal Hocko
2013-04-14 0:42 ` Mel Gorman
2013-04-14 14:34 ` Michal Hocko
2013-04-14 14:55 ` Johannes Weiner
2013-04-14 15:04 ` Michal Hocko
2013-04-14 15:11 ` Michal Hocko
2013-04-14 18:03 ` Rik van Riel
2013-04-09 12:13 ` [RFC 2/3] memcg: Ignore soft limit until it is explicitly specified Michal Hocko
2013-04-09 13:24 ` Johannes Weiner
2013-04-09 13:42 ` Michal Hocko
2013-04-09 17:10 ` Kamezawa Hiroyuki
2013-04-09 17:22 ` Michal Hocko
2013-04-09 12:13 ` [RFC 3/3] vmscan, memcg: Do softlimit reclaim also for targeted reclaim Michal Hocko
2013-04-22 2:14 ` Michal Hocko
2013-04-09 15:37 ` Michal Hocko [this message]
2013-04-09 15:50 ` [RFC 0/3] soft reclaim rework Michal Hocko
2013-04-11 8:43 ` Michal Hocko
2013-04-11 9:07 ` Michal Hocko
2013-04-11 13:04 ` Michal Hocko
2013-04-17 22:52 ` Ying Han
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=20130409153742.GL29860@dhcp22.suse.cz \
--to=mhocko@suse.cz \
--cc=glommer@parallels.com \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=riel@redhat.com \
--cc=yinghan@google.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