From: Glauber Costa <glommer@parallels.com>
To: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: linux-mm@kvack.org, cgroups@vger.kernel.org,
Tejun Heo <tj@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Michal Hocko <mhocko@suse.cz>,
handai.szj@gmail.com, anton.vorontsov@linaro.org,
Johannes Weiner <hannes@cmpxchg.org>,
Mel Gorman <mgorman@suse.de>
Subject: Re: [PATCH v2 3/5] memcg: make it suck faster
Date: Wed, 6 Mar 2013 12:38:22 +0400 [thread overview]
Message-ID: <5137007E.7030004@parallels.com> (raw)
In-Reply-To: <513691CD.3070806@jp.fujitsu.com>
>
>> Signed-off-by: Glauber Costa <glommer@parallels.com>
>> CC: Michal Hocko <mhocko@suse.cz>
>> CC: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>> CC: Johannes Weiner <hannes@cmpxchg.org>
>> CC: Mel Gorman <mgorman@suse.de>
>> CC: Andrew Morton <akpm@linux-foundation.org>
>
> After quick look, it seems most parts are good. But I have a concern.
>
> At memcg enablement, you move the numbers from vm_stat[] to res_counters.
>
Not only to res_counters. Mostly to mem_cgroup_stat_cpu, but I do move
to res_counters as well.
> Why you need it ? It's not explained.
Because at this point, the bypass will no longer be in effect and we
need accurate figures in root cgroup about what happened so far.
If we always have root-level hierarchy, then the bypass could go on
forever. But if we have not, we'll need to rely on whatever was in there.
> And if it's necessary, uncharge will leak because page_cgroup is not marked
> as PCG_USED, pc->mem_cgroup == NULL. So, res.usage will not be decreased.
>
The same problem happen when deriving an mz from a page. Since
pc->mem_cgroup will be NULL. I am interpreting that as "root mem cgroup".
Maybe even better would be to scan page cgroup writing a magic. Then if
we see that magic we are sure it is an uninitialized pc.
> Could you fix it if you need to move numbers to res_counter ?
>
At least for the pages in LRUs, I can scan them all, and update their
page information. I am just wondering if this isn't a *very* expensive
operation. Fine that we do it once, but still, is potentially scanning
*all* pages in the system.
So I've basically decided it is better to interpret pc->mem_cgroup =
NULL as this uninitialized state. (and can change to a magic)
--
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:[~2013-03-06 8:37 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-05 13:10 [PATCH v2 0/5] bypass root memcg charges if no memcgs are possible Glauber Costa
2013-03-05 13:10 ` [PATCH v2 1/5] memcg: make nocpu_base available for non hotplug Glauber Costa
2013-03-06 0:04 ` Kamezawa Hiroyuki
2013-03-19 11:07 ` Michal Hocko
2013-03-05 13:10 ` [PATCH v2 2/5] memcg: provide root figures from system totals Glauber Costa
2013-03-06 0:27 ` Kamezawa Hiroyuki
2013-03-06 8:30 ` Glauber Costa
2013-03-06 10:45 ` Kamezawa Hiroyuki
2013-03-06 10:52 ` Glauber Costa
2013-03-06 10:59 ` Kamezawa Hiroyuki
2013-03-13 6:58 ` Sha Zhengju
2013-03-13 9:15 ` Kamezawa Hiroyuki
2013-03-13 9:59 ` Sha Zhengju
2013-03-14 0:03 ` Kamezawa Hiroyuki
2013-03-06 10:50 ` Kamezawa Hiroyuki
2013-03-19 12:46 ` Michal Hocko
2013-03-19 12:55 ` Michal Hocko
2013-03-20 7:03 ` Glauber Costa
2013-03-20 8:03 ` Michal Hocko
2013-03-20 8:08 ` Glauber Costa
2013-03-20 8:18 ` Michal Hocko
2013-03-20 8:34 ` Glauber Costa
2013-03-20 8:58 ` Michal Hocko
2013-03-20 9:30 ` Glauber Costa
2013-03-21 6:08 ` Kamezawa Hiroyuki
2013-03-20 16:40 ` Anton Vorontsov
2013-03-20 7:04 ` Glauber Costa
2013-03-05 13:10 ` [PATCH v2 3/5] memcg: make it suck faster Glauber Costa
2013-03-06 0:46 ` Kamezawa Hiroyuki
2013-03-06 8:38 ` Glauber Costa [this message]
2013-03-06 10:54 ` Kamezawa Hiroyuki
2013-03-13 8:08 ` Sha Zhengju
2013-03-20 7:13 ` Glauber Costa
2013-03-19 13:58 ` Michal Hocko
2013-03-20 7:00 ` Glauber Costa
2013-03-20 8:13 ` Michal Hocko
2013-03-05 13:10 ` [PATCH v2 4/5] memcg: do not call page_cgroup_init at system_boot Glauber Costa
2013-03-06 1:07 ` Kamezawa Hiroyuki
2013-03-06 8:22 ` Glauber Costa
2013-03-19 14:06 ` Michal Hocko
2013-03-05 13:10 ` [PATCH v2 5/5] memcg: do not walk all the way to the root for memcg Glauber Costa
2013-03-06 1:08 ` Kamezawa Hiroyuki
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=5137007E.7030004@parallels.com \
--to=glommer@parallels.com \
--cc=akpm@linux-foundation.org \
--cc=anton.vorontsov@linaro.org \
--cc=cgroups@vger.kernel.org \
--cc=handai.szj@gmail.com \
--cc=hannes@cmpxchg.org \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@suse.cz \
--cc=tj@kernel.org \
/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).