linux-kernel - Re: [PATCH][RFC] dirty balancing for cgroups

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20080711055926.9AF4F5A03@siro.lan>
Date:	Fri, 11 Jul 2008 14:59:26 +0900 (JST)
From:	yamamoto@...inux.co.jp (YAMAMOTO Takashi)
To:	kamezawa.hiroyu@...fujitsu.com
Cc:	a.p.zijlstra@...llo.nl, linux-mm@...ck.org, menage@...gle.com,
	containers@...ts.linux-foundation.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH][RFC] dirty balancing for cgroups

> > >  - This looks simple but, could you merge this into memory resource controller ?
> > 
> > why?
> > 
> 3 points.
>  1. Is this useful if used alone ?

it can be.  why not?

>  2. memcg requires this kind of feature, basically.
> 
>  3. I wonder I need more work to make this work well under memcg.

i'm not sure if i understand these points.  can you explain a bit?

my patch penalizes heavy-writer cgroups as task_dirty_limit does
for heavy-writer tasks.  i don't think that it's necessary to be
tied to the memory subsystem because i merely want to group writers.

otoh, if you want to limit the number (or percentage or whatever) of
dirty pages in a memory cgroup, it can't be done independently from
the memory subsystem, of course.  it's another story, tho.

YAMAMOTO Takashi

> 
> If chasing page->cgroup and memcg make this patch much more complex,
> I think this style of implimentation is a choice.
> 
>  About 3. 
>     Does this works well if I changes get_dirty_limit()'s
>     determine_dirtyable_memory() calculation under memcg ?
>     But to do this seems not valid if dirty_ratio cgroup and memcg cgroup 
>     containes different set of tasks.
>  
> Thanks,
> -Kame
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/