linux-kernel - Re: [PATCH] oom_kill: use rss value instead of vm size for badness

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: <alpine.DEB.2.00.0911020237440.13146@chino.kir.corp.google.com>
Date:	Mon, 2 Nov 2009 02:42:03 -0800 (PST)
From:	David Rientjes <rientjes@...gle.com>
To:	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
cc:	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	linux-mm@...ck.org, linux-kernel@...r.kernel.org,
	Andrew Morton <akpm@...ux-foundation.org>,
	Hugh Dickins <hugh.dickins@...cali.co.uk>,
	Andrea Arcangeli <aarcange@...hat.com>, vedran.furac@...il.com
Subject: Re: [PATCH] oom_kill: use rss value instead of vm size for badness

On Sun, 1 Nov 2009, KOSAKI Motohiro wrote:

> > total_vm
> > 673222 test
> > 195695 krunner
> > 168881 plasma-desktop
> > 130567 ktorrent
> > 127081 knotify4
> > 125881 icedove-bin
> > 123036 akregator
> > 121869 firefox-bin
> >
> > rss
> > 672271 test
> > 42192 Xorg
> > 30763 firefox-bin
> > 13292 icedove-bin
> > 10208 ktorrent
> > 9260 akregator
> > 8859 plasma-desktop
> > 7528 krunner
> >
> > firefox-bin seems much more preferred in this case than total_vm, but Xorg
> > still ranks very high with this patch compared to the current
> > implementation.
> 
> Hi David,
> 
> I'm very interesting your pointing out. thanks good testing.
> So, I'd like to clarify your point a bit.
> 
> following are badness list on my desktop environment (x86_64 6GB mem).
> it show Xorg have pretty small badness score. Do you know why such
> different happen?
> 

I don't know specifically what's different on your machine than Vedran's, 
my data is simply a collection of the /proc/sys/vm/oom_dump_tasks output 
from Vedran's oom log.

I guess we could add a call to badness() for the oom_dump_tasks tasklist 
dump to get a clearer picture so we know the score for each thread group 
leader.  Anything else would be speculation at this point, though.

> score    pid        comm
> ==============================
> 56382   3241    run-mozilla.sh
> 23345   3289    run-mozilla.sh
> 21461   3050    gnome-do
> 20079   2867    gnome-session
> 14016   3258    firefox
> 9212    3306    firefox
> 8468    3115    gnome-do
> 6902    3325    emacs
> 6783    3212    tomboy
> 4865    2968    python
> 4861    2948    nautilus
> 4221    1       init
> (snip about 100line)
> 548     2590    Xorg
> 

Are these scores with your rss patch or without?  If it's without the 
patch, this is understandable since Xorg didn't appear highly in Vedran's 
log either.
--
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/