Created attachment 121560 [details] The GPU crash dump from /sys/class/drm/card0/error While playing XCOM I experience multiple crashes of the game. I reached a point, where the crash is reproducable and unfortunately blocking any further advance in the game. In my syslog I found: Feb 6 22:45:19 speedy kernel: [14206.111415] [drm] stuck on render ring Feb 6 22:45:19 speedy kernel: [14206.121966] [drm] GPU HANG: ecode 8:0:0x84df3c04, in RenderingThread [5838], reason: Ring hung, action: reset Feb 6 22:45:19 speedy kernel: [14206.121968] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. Feb 6 22:45:19 speedy kernel: [14206.121969] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel Feb 6 22:45:19 speedy kernel: [14206.121970] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. Feb 6 22:45:19 speedy kernel: [14206.121970] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. Feb 6 22:45:19 speedy kernel: [14206.121971] [drm] GPU crash dump saved to /sys/class/drm/card0/error Feb 6 22:45:19 speedy kernel: [14206.123446] drm/i915: Resetting chip after gpu hang Feb 6 22:45:43 speedy kernel: [14230.150226] [drm] stuck on render ring Feb 6 22:45:43 speedy kernel: [14230.160639] [drm] GPU HANG: ecode 8:0:0x87d57c02, in RenderingThread [5838], reason: Ring hung, action: reset Feb 6 22:45:43 speedy kernel: [14230.163077] drm/i915: Resetting chip after gpu hang Feb 6 22:46:15 speedy kernel: [14262.151277] [drm] stuck on render ring Feb 6 22:46:15 speedy kernel: [14262.161540] [drm] GPU HANG: ecode 8:0:0x84df3c04, in RenderingThread [5838], reason: Ring hung, action: reset Feb 6 22:46:15 speedy kernel: [14262.163329] drm/i915: Resetting chip after gpu hang Feb 6 22:46:45 speedy kernel: [14292.210824] [drm] stuck on render ring Feb 6 22:46:45 speedy kernel: [14292.221358] [drm] GPU HANG: ecode 8:0:0x87d53c02, in RenderingThread [5838], reason: Ring hung, action: reset Feb 6 22:46:45 speedy kernel: [14292.222803] drm/i915: Resetting chip after gpu hang My system is a laptop without extra graphic chip, I use the integrated graphics of my Intel(R) Core(TM) i7-5500U CPU @ 2.40GHz. As requested by the log, I attached the crash dump. As the error is (unfortunately) reproduceable, I can provide any trace you may need.
Assigning to Mesa product. From this error dump, hung is happening in render ring batch with active head at 0x0beedc9c, with 0x7b000005 (3DPRIMITIVE) as IPEHR. Batch extract (around 0x0beedc9c): 0x0beedc78: 0x780c0000: 3D UNKNOWN: 3d_965 opcode = 0x780c 0x0beedc7c: 0x00000000: MI_NOOP Bad length 7 in (null), expected 6-6 0x0beedc80: 0x7b000005: 3DPRIMITIVE: fail sequential 0x0beedc84: 0x00000104: vertex count 0x0beedc88: 0x00001560: start vertex 0x0beedc8c: 0x00000000: instance count 0x0beedc90: 0x00000001: start instance 0x0beedc94: 0x00000000: index bias 0x0beedc98: 0x00000000: MI_NOOP 0x0beedc9c: 0x78150009: 3D UNKNOWN: 3d_965 opcode = 0x7815 0x0beedca0: 0x00000007: MI_NOOP
Please test a new version of Mesa (12 or 13) and mark as REOPENED if you can reproduce and RESOLVED/* if you cannot reproduce.
I retested with Mesa 12.0.3 and the error is gone. Thank you for the update, I hope the bug report helped. Finally I can finish the game.
(In reply to Tom from comment #3) > I retested with Mesa 12.0.3 and the error is gone. Thank you for the update, > I hope the bug report helped. Finally I can finish the game. Have fun! It's an awesome game (I've been playing myself). Very glad to hear the hang is fixed.
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.