linux-kernel - [REGRESSION] perf merge breaks resume from standby, bisected

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-next>] [day] [month] [year] [list]

Message-ID: <20111101173847.GA4142@amos.fritz.box>
Date:	Tue, 1 Nov 2011 18:38:47 +0100
From:	Andreas Bombe <aeb@...ian.org>
To:	Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc:	Ingo Molnar <mingo@...e.hu>, linux-kernel@...r.kernel.org
Subject: [REGRESSION] perf merge breaks resume from standby, bisected

I found that resuming from suspend to memory stopped working with the
recent merges to mainline after 3.1. I have observed different modes of
failure even with the same version. Most often I just get the blinking
caps/scroll lock LEDs on the keyboard and no video output.

Sometimes I get the normal text output from a resume ending with an oops
dump. It has been mostly NULL pointer dereferences, but I think there
was a scheduling in atomic context at one time. I have a photo of the
NULL deref on the latest kernel attached, according to gdb the IP points
to the return statement of the event_sched_out() function.

Anyway, I bisected the problem down to 144060f "perf: Add PM notifiers
to fix CPU hotplug races". The hardware is a Core 2 Duo E8500 on an Asus
P5Q, config attached.

-- 
Andreas Bombe

Download attachment "crash.jpeg" of type "image/jpeg" (222485 bytes)

View attachment "config" of type "text/plain" (85070 bytes)