Bug 14840 - radeon 60 wakeups after resume
Summary: radeon 60 wakeups after resume
Status: RESOLVED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: General (show other bugs)
Version: XOrg git
Hardware: x86 (IA32) Linux (All)
: medium minor
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords: regression
Depends on:
Blocks:
 
Reported: 2008-03-05 13:02 UTC by Jose Rodriguez
Modified: 2008-05-14 15:12 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments

Description Jose Rodriguez 2008-03-05 13:02:38 UTC
Using xserver-xorg-video-ati from Debian's experimental repository (6.8.1~git20080302.a4398ac3-1) causes 60 wakeups/sec after resuming from hibernating or suspending. This doesn't happen here with the driver currently in unstable (6.8.0-1).
Comment 1 Alex Deucher 2008-03-05 16:00:11 UTC
Any chance you could use git bisect to track down the problematic commit?
Comment 2 Jose Rodriguez 2008-03-06 12:42:31 UTC
On 06/03/2008, bugzilla-daemon@freedesktop.org <
bugzilla-daemon@freedesktop.org> wrote:
>
> http://bugs.freedesktop.org/show_bug.cgi?id=14840
>
> --- Comment #1 from Alex Deucher <agd5f@yahoo.com>  2008-03-05 16:00:11
> PST ---
> Any chance you could use git bisect to track down the problematic commit?
>


This is what I got:

dd8ee1b444f4b973a1e0fadca5f943f2162b5e94 is first bad commit
commit dd8ee1b444f4b973a1e0fadca5f943f2162b5e94
Author: Alex Deucher <alex@cube.(none)>
Date:   Sat Mar 1 16:23:51 2008 -0500

    RADEON: memmap rework 1

    Don't restore memmap regs on every mode switch.
Comment 3 Alex Deucher 2008-03-07 11:28:16 UTC
does this patch help:
http://www.botchco.com/alex/xorg/entervt.diff
Comment 4 Jose Rodriguez 2008-03-08 03:38:31 UTC
OK, so it works well regarding the vblank stuff, but I definitely now have some screen corruption at random times. I can't be much more specific, it happens when I open some windows but I don't know of any regular pattern. Basically some nasty colored stripes appear, which change when switching workspaces and moving stuff around the screen.

Just to double check that I did the right thing: after git-bisecting I did a git-bisect reset and then applied the patch. Is it right like this?
Comment 5 Alex Deucher 2008-03-08 09:00:48 UTC
(In reply to comment #4)
> Just to double check that I did the right thing: after git-bisecting I did a
> git-bisect reset and then applied the patch. Is it right like this?
> 

yes, that is correct.
Comment 6 Jose Rodriguez 2008-04-07 13:23:23 UTC
I built the 2D driver from commit 8b144830fe9b4a0cee4745023de5e7d387070f60 (25 Mar) and the suspend issue is fixed. No screen corruption and interrupts back to normal when resuming.

The problem is that they go up to 60 again after I fire Tuxracer or Openarena. It doesn't happen with other two 3D applications, though (Gtkradiant and VMD). Suspending is now the cure instead of the problem. I can't remember whether this happened before.
Comment 7 Jose Rodriguez 2008-05-09 16:04:03 UTC
All the bits build from git two days ago solve the problem, except that when closing Openarena the screen resolution gets mangled (something like 800x600 or whatever I use in the game).
Comment 8 Jose Rodriguez 2008-05-14 15:12:57 UTC
Rebuilt everything today and it works fine. Issue with OA may have to do with some mess I have with two versions of it mixed.

Thanks

Jose


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.