Created attachment 111350 [details] dmesg ==System Environment== -------------------------- Regression: not sure on-working platforms: BSW ==kernel== -------------------------- drm-intel-next-queued/140fd38dc4962ae3694f81900b51c567df1b6d33 ==Bug detailed description== ----------------------------- It sporadically causes <3>[ 119.754886] [drm:i915_hangcheck_elapsed [i915]] *ERROR* Hangcheck timer elapsed... blitter ring idle. Fail rate: 3/5. output: IGT-Version: 1.9-geb799b2 (x86_64) (Linux: 3.18.0-rc7_drm-intel-next-queued_140fd3_20141226+ x86_64) running dummy loop on blt dummy loop run on blt completed Subtest blt: SUCCESS (50.930s) ==Reproduce steps== ---------------------------- 1. ./gem_dummy_reloc_loop --run-subtest blt
It also happens on SKL.
It also happens on BDW. Test sub cases mixed, mixed_multi_fd and render, they all have this issue.
still happen on SKL-Y with the latest kernel : Kernel commit log: commit b4c4542ba1abfb0d3d6913504502573bf2c62b12 Author: Daniel Vetter <daniel.vetter@ffwll.ch> Date: Fri Aug 28 15:51:30 2015 +0200 drm-intel-nightly: 2015y-08m-28d-13h-50m-34s UTC integration manifest
commit 7c17d377374ddbcfb7873366559fc4ed8b296e11 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 20 15:43:35 2016 +0200 drm/i915: Use ordered seqno write interrupt generation on gen8+ execlists
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.