Created attachment 77472 [details] xorg.log I'm using SNA from 2.21.5-32-g4a43aa8. And sometimes when I maximize a window with gl compositing enabled in KDE4, whole screen blinks, and for a moment I see kind of b&w lines on the whole screen, i.e. whole screen is corrupted. Once this also happened while I was moving the horizontal slide bar in FF. I tried to capture a snapshot of that by this: while :; do import -window root screenshot-`date +%s`-$RANDOM.jpg usleep 500000 done but failed. All images are OK.
This sounds like a pageflip bug, widening the audience.
(In reply to comment #1) > This sounds like a pageflip bug, widening the audience. Taking this into consideration, this is a bug introduced even recently to the kernel. Like late the past week or this week into -next.
So is this a kernel regression? Can you attempt to bisect it please?
(In reply to comment #3) > So is this a kernel regression? Can you attempt to bisect it please? It looks like that. I don't think I updated ddx when this started happening.
Can you please retest with xf86-video-intel.git, specifically commit 9dae6f9f1f169c228929185a8bd94e82afe92574 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 12 11:01:08 2013 +0100 sna: Flush the scanout cache after resizing the display
(In reply to comment #5) > Can you please retest with xf86-video-intel.git, specifically > > commit 9dae6f9f1f169c228929185a8bd94e82afe92574 > Author: Chris Wilson <chris@chris-wilson.co.uk> > Date: Fri Apr 12 11:01:08 2013 +0100 > > sna: Flush the scanout cache after resizing the display It looks like it's gone now.
(In reply to comment #6) > (In reply to comment #5) > > Can you please retest with xf86-video-intel.git, specifically > > > > commit 9dae6f9f1f169c228929185a8bd94e82afe92574 > > Author: Chris Wilson <chris@chris-wilson.co.uk> > > Date: Fri Apr 12 11:01:08 2013 +0100 > > > > sna: Flush the scanout cache after resizing the display > > It looks like it's gone now. Nah, it happened right now with 2.21.6-20-gb4871f1 :(.
Could this be connected to this one I reported to LKML couple days ago? [ 702.429181] [drm] capturing error event; look for more information in /sys/kernel/debug/dri/0/i915_error_state [ 702.430649] i915: render error detected, EIR: 0x00000010 [ 702.430651] i915: page table error [ 702.430652] i915: PGTBL_ER: 0x00000002 [ 702.430654] [drm:i915_report_and_clear_eir] *ERROR* EIR stuck: 0x00000010, masking [ 702.430661] i915: render error detected, EIR: 0x00000010 [ 702.430662] i915: page table error [ 702.430663] i915: PGTBL_ER: 0x00000002 Because 702.* is a minute ago...
This might be interesting: Jan 11 20:05:32 bellona kernel: [155976.935687] i915: render error detected, EIR: 0x00000010 Jan 11 20:05:32 bellona kernel: [155976.935702] i915: render error detected, EIR: 0x00000010 Jan 12 00:15:50 bellona kernel: [ 0.265651] i915: render error detected, EIR: 0x00000010 Jan 12 00:15:50 bellona kernel: [ 0.266025] i915: render error detected, EIR: 0x00000010 Jan 16 22:35:05 bellona kernel: [ 0.615809] render error detected, EIR: 0x00000010 Jan 16 22:35:05 bellona kernel: [ 0.616175] render error detected, EIR: 0x00000010 Jan 16 22:37:15 bellona kernel: [ 135.830818] render error detected, EIR: 0x00000010 Jan 16 22:37:15 bellona kernel: [ 135.830829] render error detected, EIR: 0x00000010 Jan 17 10:24:42 bellona kernel: [ 5841.340225] i915: render error detected, EIR: 0x00000010 Jan 17 10:24:42 bellona kernel: [ 5841.340975] i915: render error detected, EIR: 0x00000010 Jan 17 14:15:56 bellona kernel: [14397.516635] i915: render error detected, EIR: 0x00000010 Jan 17 14:15:56 bellona kernel: [14397.516648] i915: render error detected, EIR: 0x00000010 Jan 17 17:32:16 bellona kernel: [ 0.257226] i915: render error detected, EIR: 0x00000010 Jan 17 17:32:16 bellona kernel: [ 0.257597] i915: render error detected, EIR: 0x00000010 Jan 23 19:34:18 bellona kernel: [ 121.003913] i915: render error detected, EIR: 0x00000010 Jan 23 19:34:18 bellona kernel: [ 121.003923] i915: render error detected, EIR: 0x00000010 Jan 24 10:40:00 bellona kernel: [20417.572917] i915: render error detected, EIR: 0x00000010 Jan 24 10:40:00 bellona kernel: [20417.572938] i915: render error detected, EIR: 0x00000010 Feb 13 23:32:20 bellona kernel: [20758.940001] i915: render error detected, EIR: 0x00000010 Feb 13 23:32:20 bellona kernel: [20758.940001] i915: render error detected, EIR: 0x00000010 Feb 23 21:53:56 bellona kernel: [199398.644012] i915: render error detected, EIR: 0x00000010 Feb 23 21:53:56 bellona kernel: [199398.644012] i915: render error detected, EIR: 0x00000010 Apr 6 16:15:16 bellona kernel: [77452.240911] i915: render error detected, EIR: 0x00000010 Apr 6 16:15:16 bellona kernel: [77452.240925] i915: render error detected, EIR: 0x00000010 Apr 6 21:27:18 bellona kernel: [91111.418392] i915: render error detected, EIR: 0x00000010 Apr 6 21:27:18 bellona kernel: [91111.418411] i915: render error detected, EIR: 0x00000010 Apr 19 11:12:20 bellona kernel: [ 1244.433860] i915: render error detected, EIR: 0x00000010 Apr 19 11:12:20 bellona kernel: [ 1244.433871] i915: render error detected, EIR: 0x00000010 Apr 20 00:21:40 bellona kernel: [ 3824.921202] i915: render error detected, EIR: 0x00000010 Apr 20 00:21:40 bellona kernel: [ 3824.921213] i915: render error detected, EIR: 0x00000010 Apr 20 17:47:31 bellona kernel: [ 702.430649] i915: render error detected, EIR: 0x00000010 Apr 20 17:47:31 bellona kernel: [ 702.430661] i915: render error detected, EIR: 0x00000010 ... the output of: # xzgrep 'render error detected' /var/log/messages-20130* /var/log/messages|sed 's@^[^:]*:@@'
The bug can't be reproduced on G45 machine. Our test environment as following: Mesa: (9.1)f00ae9c773dc944f0d900ef5c50c3b417ad95e84 Xserver: (server-1.13-branch)xorg-server-1.13.4 Xf86_video_intel:(master)2.21.7-5-g87295b1ef85505689ce326137c2794230fb3f35f Cairo:(master)631bf299256e11a17511977f357e0353fb5615f7 Libva_intel_driver:(master)31caada2967b94705d78ab7f6d07965ad7f13d42 Kernel:(drm-intel-fixes) 3598706b52cb45ba0a9e8aa99ce5ac59140f2b8b Os: Fedora17
(In reply to comment #4) > (In reply to comment #3) > > So is this a kernel regression? Can you attempt to bisect it please? > > It looks like that. I don't think I updated ddx when this started happening. Ping for the bisect ... Or if this is not a kernel issue maybe we should shovel it back to the ddx?
(In reply to comment #11) > Ping for the bisect ... Or if this is not a kernel issue maybe we should > shovel it back to the ddx? I think this was a ddx problem. I don't see that anymore with the current git. When I switch ddx with sna to the distributed one (opensuse) I see it again.
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.