linux-kernel - Re: [stable] [tip:x86/urgent] x86: Add memory modify constraints to xchg() and cmpxchg()

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]

Date:	Thu, 9 Sep 2010 21:53:01 +0200
From:	Tomáš Janoušek <tomi@...i.cz>
To:	Greg KH <greg@...ah.com>
Cc:	Jeremy Fitzhardinge <jeremy@...p.org>, mingo@...hat.com,
	"H. Peter Anvin" <hpa@...or.com>, linux-kernel@...r.kernel.org,
	torvalds@...ux-foundation.org, glommer@...hat.com,
	alan@...rguk.ukuu.org.uk, zamsden@...hat.com, stable@...nel.org,
	mtosatti@...hat.com, gregkh@...e.de, peter@...frader.org,
	tglx@...utronix.de, avi@...hat.com,
	linux-tip-commits@...r.kernel.org
Subject: Re: [stable] [tip:x86/urgent] x86: Add memory modify constraints
	to xchg() and cmpxchg()

Hello,

2.6.32.21 does not boot on my HP Compaq nx7300. I bisected it to this patch:

On Mon, Aug 02, 2010 at 04:59:50PM -0700, Greg KH wrote:
> On Mon, Aug 02, 2010 at 04:51:27PM -0700, Jeremy Fitzhardinge wrote:
> >  On 07/27/2010 10:33 PM, tip-bot for H. Peter Anvin wrote:
> > >Commit-ID:  113fc5a6e8c2288619ff7e8187a6f556b7e0d372
> > >Gitweb:     http://git.kernel.org/tip/113fc5a6e8c2288619ff7e8187a6f556b7e0d372
> > >Author:     H. Peter Anvin<hpa@...or.com>
> > >AuthorDate: Tue, 27 Jul 2010 17:01:49 -0700
> > >Committer:  H. Peter Anvin<hpa@...or.com>
> > >CommitDate: Tue, 27 Jul 2010 17:14:02 -0700
> > >
> > >x86: Add memory modify constraints to xchg() and cmpxchg()
> > 
> > Here's the 2.6.32 version for stable.
> 
> Thanks, I'll queue it up when it hits Linus's tree.

Here's a video of the crash: http://store.lisk.in/tmp/09092010019.mp4
With boot_delay=30, it sometimes just reboots and sometimes prints the trace.
Without it, it always prints the trace. The crash occurs during the
initialization of intel-agp (which explains why I can't reproduce it in qemu).
More details in the video.

The system boots and works just fine with 2.6.32.21 with the patch reverted.
v2.6.36-rc3-185-gd56557a works fine as well, perhaps there were other fixes
that didn't get to stable (e.g. 4532b30).

I'm compiling the kernel with the latest gcc in Debian (-dumpversion says
4.4.5, dpkg reports 4:4.4.4-2). Shall I post dmesg/lspci as well?

Since I have an easy workaround and other things to do, I won't spend time
trying to backport those other fixes myself, but if you're going to fix this,
I will help with testing.

Regards,
-- 
Tomáš Janoušek, a.k.a. Liskni_si, http://work.lisk.in/
--
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/