Created attachment 145037 [details]
i915 error state for GPU hang
- SKL i5 6600K
- Ubuntu 16.04
- drm-tip git kernel (0330b51e91)
- Mesa git (5ed4e31c08d)
- Unity desktop
- 3x fullscreen FullHD HDRBloom multi-context SynMark test-case:
- Recoverable GPU hang, but all successive GL tests fail after that:
i915 0000:00:02.0: GPU HANG: ecode 9:1:0x00000000, hang on rcs0
Main difference from this test to most other tests is only few tests use multiple contexts.
I haven't seen such hangs with the i965 driver.
I wasn't able to reproduce the hang after reboot when re-running HDRBloom 10 times, so it may depend on previous tests, or is just very hard to reproduce.
I didn't see such hang when running similar test-sets month ago, so it can be a regression.
On BXT there was a hang in a different test-case.
(In reply to Eero Tamminen from comment #0)
> I wasn't able to reproduce the hang after reboot when re-running HDRBloom 10
> times, so it may depend on previous tests, or is just very hard to reproduce.
Was able to reproduce the GPU hang with Iris by running each SynMark test 3x in alphabetical order. At HdrBloom there was again GPU hang. This time tests after HdrBloom didn't fail.
Got a (non-recoverable) HdrBloom hang also with i965 when using latest Git gfx stack, on SKL GT4e (SkullCanyon), so this might not be Iris specific issue.
Created attachment 145096 [details]
SKL GT4e (recoverable) GPU hang
In the SKL GT4e / i965 case, recoverable GPU hang during HdrBloom run appears to be happening in the X server.
If same was true also of the SKL GT2 case (i915 error state didn't specify process), then that's also i965, as I was running only the benchmark itself with Iris on SKL GT2.
Last night HdrBloom test had a (non-recoverable) GPU hang (something had at least broken test automation network connection during that exact test). Moving to i965, as Iris isn't yet enabled by default and this happens (also) with i965.