linux-kernel - Re: [PATCH] bttv: fix mutex use before init

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: <AANLkTimZ=cdu--GsVWotAir-2QpuXQQBg+7UtkVvKzO=@mail.gmail.com>
Date:	Tue, 14 Dec 2010 22:13:11 +0100
From:	Torsten Kaiser <just.for.lkml@...glemail.com>
To:	Brandon Philips <brandon@...p.org>
Cc:	Dave Young <hidave.darkstar@...il.com>,
	linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
	Mauro Carvalho Chehab <mchehab@...radead.org>,
	Guennadi Liakhovetski <g.liakhovetski@....de>,
	Chris Clayton <chris2553@...glemail.com>
Subject: Re: [PATCH] bttv: fix mutex use before init

On Tue, Dec 14, 2010 at 9:56 PM, Torsten Kaiser
<just.for.lkml@...glemail.com> wrote:
> Using the card also still works, but I think I found out what was
> causing sporadic shutdown problems with 37-rc kernels: When I try to
> exit tvtime it gets stuck in an uninterruptible D state and can't be
> killed. And that seems to mess up the shutdown.
>
> But this happens independent with or without your patch and looks like
> different problem.
>
> SysRQ+W provided this stack trace, maybe someone seens an obvious bug...:
> [  274.772528]  ffff8800dec69680 0000000000000086 ffffffff81089d73
> ffff8800729d05a0
> [  274.778599]  0000000000011480 ffff8800df923fd8 0000000000011480
> ffff8800df922000
> [  274.778599]  ffff8800df923fd8 0000000000011480 ffff8800dec69680
> 0000000000011480
> [  274.778599] Call Trace:
> [  274.778599]  [<ffffffff81089d73>] ? free_pcppages_bulk+0x343/0x3b0
> [  274.778599]  [<ffffffff8156d0e1>] ? __mutex_lock_slowpath+0xe1/0x160
> [  274.778599]  [<ffffffff8156cd8a>] ? mutex_lock+0x1a/0x40
> [  274.778599]  [<ffffffff8141ab7f>] ? free_btres_lock.clone.19+0x3f/0x100
> [  274.778599]  [<ffffffff8141d311>] ? bttv_release+0x1c1/0x1e0
> [  274.778599]  [<ffffffff813fe4ba>] ? v4l2_release+0x4a/0x70
> [  274.778599]  [<ffffffff810c5291>] ? fput+0xe1/0x250
> [  274.778599]  [<ffffffff810c1d59>] ? filp_close+0x59/0x80
> [  274.778599]  [<ffffffff810c1e0b>] ? sys_close+0x8b/0xe0
> [  274.778599]  [<ffffffff8100253b>] ? system_call_fastpath+0x16/0x1b

Hmm:bttv_release() does mutex_lock(&btv->lock), then calls into
free_btres_lock(...) that also first does mutex_lock(&btv->lock);

The calls to lock btv->lock in bttv_release() where added as part of
the BKL removal, but I do not understand enough to fix this.
Can this be dropped from bttv_release() completely, or would an
unlocked version of free_btres_lock() be better?

Torsten
--
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/