Created attachment 70792 [details] Intel gpu logs After update to xf86-video-intel-2.20.14 show GPU hung [ 1.484068] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0 [ 74.353416] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung [ 74.353429] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state [ 94.030106] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung [ 94.086769] [drm:init_ring_common] *ERROR* render ring initialization failed ctl 0001f001 head 00001000 tail 00000000 start 00001000 -- chipset: G45 -- system architecture: x86_64 -- xf86-video-intel: 2.20.14, SNA -- xserver: 1.13.0 + git fixes -- mesa: 9.0.1 -- libdrm: 2.4.40 -- kernel version: 3.6.8 -- Linux distribution: Archlinux -- Display connector: internal (LVDS) more info in atachment
Created attachment 70793 [details] Intel GPU logs from intel-gpu-tools
Can you first reproduce this and see if it bisects to any of the recent gen4 patches in particular?
(In reply to comment #2) > Can you first reproduce this and see if it bisects to any of the recent gen4 > patches in particular? Yes, I can try. But first up on Sunday.
I have a little time: [alois@lenovo xf86-video-intel]$ git bisect good 4c922eb52cadb867a0a15929e5a214c84a5992f3 is the first bad commit commit 4c922eb52cadb867a0a15929e5a214c84a5992f3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 23 14:19:59 2012 +0000 sna/gen4: Pass the mask channel explicitly rather than through a dummy mask Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> :040000 040000 4bc292e95d326851a7e1ae68bd9ecd237e6d8663 63ade7f6e8c7c911e94b109533c78d758766d73d M src
On my gm45, I tracked the hangs down to rendering to a Y-tiled surface. Can you try testing the current -intel.git?
tested, bug fixed in git. For me works!
Thanks for the report. My machine is still surviving, so I'm reasonably confident that it is stable for the time being - please do reopen if you hit another hang.
I hit another hung. Configuration is same, driver xf86-video-intel 2.20.15. (Hard to reproduce, because more as 1 hour gaming under Wine 32-bit.) [ 4505.962927] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung [ 4505.962941] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state [ 4512.839589] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung [ 4512.892929] [drm:init_ring_common] *ERROR* render ring initialization failed ctl 0001f001 head 00001000 tail 00000000 start 00001000 [ 4514.419604] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung [ 4514.420141] [drm:i915_reset] *ERROR* GPU hanging too fast, declaring wedged! [ 4514.420149] [drm:i915_reset] *ERROR* Failed to reset chip.
Created attachment 71470 [details] Intel GPU logs
Not is not an SNA hang, but dies inside Mesa. Please file a fresh bug report for mesa/i965.
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.