linux-kernel - Re: Linux 2.6.33-rc5

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.LFD.2.00.1001241018500.3574@localhost.localdomain>
Date:	Sun, 24 Jan 2010 10:30:22 -0800 (PST)
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Borislav Petkov <petkovbb@...glemail.com>
cc:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Alex Deucher <alexdeucher@...il.com>,
	Jerome Glisse <glisse@...edesktop.org>,
	"Eric W. Biederman" <ebiederm@...ssion.com>
Subject: Re: Linux 2.6.33-rc5

On Sun, 24 Jan 2010, Borislav Petkov wrote:
> 
> 1. After suspend, the right monitor shows very funny colors, as if it is
> on crack, in contrast to the left one which looks fine:
> 
> http://userweb.kernel.org/~bp/right_monitor.jpg
> http://userweb.kernel.org/~bp/left_monitor.jpg

Ok, at a guess, it looks like something isn't restoring the palette 
registers for the second output. You didn't say what GPU you have, but 
judging by your CC list, it's a radeon. You might want to let the Radeon 
KMS people know exactly what your setup is etc.

> 2. During suspend I get this juicy lockdep warning:

I'm pretty sure that this is not actually a new problem, it's a new 
warning for an old problem, see commit 846f99749 ("sysfs: Add lockdep 
annotations for the sysfs active reference"). So I'm not considering it a 
regression, even though I do agree that the messages look scary and very 
annoying.

I'm not at all convinced that we are going to be able to fix the problems 
that this made visible in time for 2.6.33. The sysfs-related lock chains 
easily get rather involved, although it is always possible that this 
particular one might be easily fixable (ie if radeon is able to call the 
final kref_put outside any locks - or whatever it is that ends up 
generating a dependency from struct_mutex for the sysfs active ref).

Or maybe I'm misreading the lockdep chain. Eric Biederman added to cc.

		Linus
--
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/