View previous topic :: View next topic |
Author |
Message |
kolcon Tux's lil' helper


Joined: 21 Sep 2007 Posts: 96 Location: Europe, CZ
|
Posted: Fri Jan 06, 2012 8:16 pm Post subject: [SOLVED] kernel 3.2.0 - blank screen on resume |
|
|
Hello,
after moving from gentoo-sources-3.1.5 to 3.2.0-r1 (copied over .config) I cannot wake from suspend anymore.
Or, should I say, the computer wakes fine, only I cannot see anything on the (blank) screen.
But the notebook works, I can even ssh to it.
I found out in the log file line related to this:
kernel: [drm:intel_dp_complete_link_train] *ERROR* failed to train DP, aborting
I cannot get the screen back, even when trying things like xset force dpms off (or on) or playing with
xrandr.
What would you recommmend me to try or do?
Thanks
Lubos
Last edited by kolcon on Fri Jan 27, 2012 8:29 pm; edited 1 time in total |
|
Back to top |
|
 |
Hu Administrator

Joined: 06 Mar 2007 Posts: 23193
|
Posted: Sat Jan 07, 2012 1:29 am Post subject: |
|
|
If a bug report has not yet been filed, then please file one. If you find one which has been filed, please post the link(s) here for the benefit of others who encounter the problem.
If you have time, expertise, and motivation, you could try performing a git bisect to search for the specific commit which caused this. We know that v3.1.5 works for you and v3.2 fails, so there are only 12695 commits to test, which should take about 14 iterations. If you get lucky and the break is in the drivers/gpu/drm/i915 directory, then there are only 126 commits to test, which should take about 7 iterations. In both cases, the iteration count could be higher if you get unlucky and hit a broken commit or the bisect is not able to pick an optimal midpoint. |
|
Back to top |
|
 |
kolcon Tux's lil' helper


Joined: 21 Sep 2007 Posts: 96 Location: Europe, CZ
|
Posted: Sun Jan 08, 2012 2:13 pm Post subject: |
|
|
Thanks, I have done what you suggest.
The bug is here : https://bugs.gentoo.org/show_bug.cgi?id=398149
The culprit commit is :
cd9dde44f47501394b9f0715b6a36a92aa74c0d0 is the first bad commit
commit cd9dde44f47501394b9f0715b6a36a92aa74c0d0
Author: Adam Jackson <ajax@redhat.com>
Date: Fri Oct 14 12:43:49 2011 -0400
drm/i915/dp: Fix the math in intel_dp_link_required
The previous code was confused about units, which is pretty reasonable
given that the units themselves are confusing.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
:040000 040000 3c0a0f38fc6482a401340e03fa3fbd687d11eda4 294673b782b3ecc93540f46f7721d385bd30cfac M drivers |
|
Back to top |
|
 |
kolcon Tux's lil' helper


Joined: 21 Sep 2007 Posts: 96 Location: Europe, CZ
|
Posted: Sat Jan 21, 2012 10:04 am Post subject: |
|
|
... still does not work with 3.3-rc1... |
|
Back to top |
|
 |
toralf Developer


Joined: 01 Feb 2004 Posts: 3943 Location: Hamburg
|
Posted: Sat Jan 21, 2012 10:19 am Post subject: |
|
|
YOu should post the bisect results to the linux kernel mailing list directly : linux-kernel@vger.kernel.org |
|
Back to top |
|
 |
kolcon Tux's lil' helper


Joined: 21 Sep 2007 Posts: 96 Location: Europe, CZ
|
Posted: Sat Jan 21, 2012 10:32 am Post subject: |
|
|
toralf wrote: | YOu should post the bisect results to the linux kernel mailing list directly : linux-kernel@vger.kernel.org |
OK, done |
|
Back to top |
|
 |
kolcon Tux's lil' helper


Joined: 21 Sep 2007 Posts: 96 Location: Europe, CZ
|
Posted: Fri Jan 27, 2012 8:30 pm Post subject: |
|
|
kolcon wrote: | toralf wrote: | YOu should post the bisect results to the linux kernel mailing list directly : linux-kernel@vger.kernel.org |
OK, done |
So, this issue is fixed by Keith Packard by two patches - quick one will be in 3.3
and more complex one in 3.4 |
|
Back to top |
|
 |
toralf Developer


Joined: 01 Feb 2004 Posts: 3943 Location: Hamburg
|
Posted: Sat Jan 28, 2012 9:07 am Post subject: |
|
|
Yep - in general it is a good idea to post such bugs directly to the LKML or to the appropriate Bugzilla sites ate freedesktop.org. |
|
Back to top |
|
 |
Joseph K. Guru


Joined: 07 Jun 2006 Posts: 436 Location: Sydney, Australia
|
Posted: Mon Feb 06, 2012 12:52 pm Post subject: |
|
|
Hmmm, I didn't hit this bug on my i915 laptop, but I did just hit it on my new i965 laptop. Any chance that the Gentoo devs will backport a patch to 3.2.x? |
|
Back to top |
|
 |
Hu Administrator

Joined: 06 Mar 2007 Posts: 23193
|
Posted: Tue Feb 07, 2012 3:18 am Post subject: |
|
|
Joseph K. wrote: | Any chance that the Gentoo devs will backport a patch to 3.2.x? | Useful patches are often backported if (1) the patch is identified, (2) it has no significant dependencies on other new functionality (e.g. a subsystem rewrite new in the -rc series kernel), and (3) it has no substantial negative consequences. I see that kolcon found the offending commit and reported the issue to LKML, but no one has posted a patch ID or other guidance that easily leads to the patches. I see several patches in Linus's tree written by Keith Packard, but none sound obviously related to the problem reported here.
Code: | $ git rev-parse origin/master
8597559a78e1cde158b999212bc9543682638eb1
$ git log --no-merges --format='[url=https://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commit;h=%H]%s[/url]' v3.3-rc1..origin/master --author=Packard |
Revert "drm/i915: Work around gen7 BLT ring synchronization issues."
drm/i915: Hold gt_lock across forcewake register reads
drm/i915: Hold gt_lock during reset
drm/i915: Move reset forcewake processing to gen6_do_reset |
|
Back to top |
|
 |
|