linux-kernel - Re: [RFC PATCH 0/4] clockevents: fix clockevent_devices list corruption after cpu hotplug

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] [day] [month] [year] [list]

Message-ID: <alpine.LFD.2.00.1001181449220.2906@localhost.localdomain>
Date:	Mon, 18 Jan 2010 14:51:14 +0100 (CET)
From:	Thomas Gleixner <tglx@...utronix.de>
To:	Xiaotian Feng <dfeng@...hat.com>
cc:	Ozan Çağlayan <ozan@...dus.org.tr>,
	damm@...l.co.jp, hsweeten@...ionengravers.com,
	akpm@...ux-foundation.org, venkatesh.pallipadi@...el.com,
	linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH 0/4] clockevents: fix clockevent_devices list corruption
 after cpu hotplug

On Mon, 18 Jan 2010, Xiaotian Feng wrote:

> On 01/17/2010 05:28 PM, Ozan Çağlayan wrote:
> > Thomas Gleixner wrote:
> > 
> > > 
> > > I'm not happy about that churn. Why don't we simply scan the
> > > clockevent_devices list for leftovers of the dead CPU ?
> > > 
> > > Untested patch below solves the same problem.
> > 
> > 
> > Hi, I have 3 bug reports about a clockevent failure while halting the system
> > (2.6.31.11 kernel).
> > It exactly pinpoints to the line 262 which got changed with this patch
> > merged to 2.6.31.10 (and also 2.6.32.3):
> > 
> > [4406.986777] kernel BUG at kernel/time/clockevents.c:262!
> > [4406.986777] invalid opcode: 0000 [#1] SMP
> > [4406.986777] last sysfs file: /sysfs/module/ip_tables/initstate
> 
> I think this one is duplicated of
> http://bugzilla.kernel.org/show_bug.cgi?id=15037
> Thomas is working on the fix. I've sent a workaround patch on
> http://patchwork.kernel.org/patch/71537/

I just applied your patch, but kept the cpuweight check. This is the
least intrusive solution for now. The logic needs an overhaul, but
thats neither rc4 nor stable material.

Thanks,

	tglx