linux-kernel - post-rc4 regression (bisected): blank screen on ASUS P5KPL-AM

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: <20101220140615.GA3035@automatix>
Date:	Mon, 20 Dec 2010 14:06:16 +0000
From:	Jochen Voss <voss@...huhn.de>
To:	linux-kernel@...r.kernel.org,
	Chris Wilson <chris@...is-wilson.co.uk>,
	Dave Airlie <airlied@...hat.com>
Subject: post-rc4 regression (bisected): blank screen on ASUS P5KPL-AM

Hello,

My system, when booted with recent kernels, fails to show any output
on the monitor.  The problem occurs very early during bootup (I never
see anything after the messages from grub), the monitor indicates that
there is no video signal at all and dmesg shows the following lines:

[    1.540465] usbhid: USB HID core driver
[    2.490014] No connectors reported connected with modes
[    2.490019] [drm] Cannot find any crtc or sizes - going 1024x768
[    2.496064] Console: switching to colour frame buffer device 128x48
[    2.497729] fb0: inteldrmfb frame buffer device

The machine seems fine otherwise: I can log in remotely and logging in
from the console (by typing blindly without visual feedback) also
works.

The problem is present in the current kernel as of this morning (I see
it with commit 55ec86f848a5f872fd43f5d7206464a0af419110).  Bisection
indicates that it is caused by the following commit.

    commit 541cc966915b6756e54c20eebe60ae957afdb537
    Author: Chris Wilson <chris@...is-wilson.co.uk>
    Date:   Mon Dec 6 11:24:07 2010 +0000

	drm: Don't try and disable an encoder that was never enabled

	Prevents code that assumes that the encoder is active when
	asked to be disabled from dying a horrible death.

	Reported-by: Daniel Vetter <daniel.vetter@...ll.ch>
	Signed-off-by: Chris Wilson <chris@...is-wilson.co.uk>
	Signed-off-by: Dave Airlie <airlied@...hat.com>

I verfied that the problem disappears when booting recent kernels
(i.e. 55ec86f848a5f872fd43f5d7206464a0af419110) with the offending
commit reverted.  The screen works again and the dmesg output now
changes to

[    1.542515] usbhid: USB HID core driver
[    1.547569] Console: switching to colour frame buffer device 210x65
[    1.551558] fb0: inteldrmfb frame buffer device

Note that the one second wait which was present in the bad dmesg
has now disappeared.

System information:

    ASUS P5KPL-AM mainboard (Intel G31 chipset)
    Intel Core2 quad core CPU
    Ubuntu 10.10

Appended files:

    dmesg.bad: dmesg for 55ec86f848a5f872fd43f5d7206464a0af419110
    dmesg.good: dmesg for 55ec86... with
        541cc966915b6756e54c20eebe60ae957afdb537 reverted
    lspci: "lspci -v" output (on a good kernel)

I'd be happy to send further information or to try patches etc.

I hope this helps,
Jochen
-- 
http://seehuhn.de/

View attachment "dmesg.bad" of type "text/plain" (38871 bytes)

View attachment "dmesg.good" of type "text/plain" (38871 bytes)

View attachment "lspci" of type "text/plain" (7749 bytes)