Created attachment 68392 [details] i915 error state System Environment: -------------------------- Arch: x86_64 Platform: Ironlake Libdrm: (master)libdrm-2.4.39-16-g14db948127e549ea9234e02d8e112de3871f8f9f Xserver: (master)xorg-server-1.13.0-49-g0a75bd640b3dc26b89d9e342999a7f4b7e98edbf Xf86_video_intel: (master)2.20.9-43-gfb5205a86da09b344dbc20598655e917c263125c Libva: (staging)86484495155e65fd8ac33ed3ede43fb42defd966 Libva_intel_driver: (staging)f557dd6ad06c31bcf787468e804c948ecc4cf39b Kernel: (drm-intel-nightly) 7735f5cccb4a57c5da5554f3e005bbd5f7325f40 Mesa: (master)87a34131c427b40a561cfef1513b446a0eeabc39 Bug detailed description: ------------------------- It happens on ironlake with mesa master branch.It doesn't happen on 9.0 branch. spec_ARB_fragment_program_kil-swizzle also causes GPU huang. The last known good commit:86de501f14f11f1e993c8703c0d69bdf1f6c7835 The last known bad commit: 87a34131c427b40a561cfef1513b446a0eeabc39 dmesg: [ 168.808646] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung [ 168.808760] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state [ 170.812582] [drm:i915_hangcheck_hung] *ERROR* Hangcheck timer elapsed... GPU hung [ 170.813149] [drm:i915_reset] *ERROR* GPU hanging too fast, declaring wedged! [ 170.813151] [drm:i915_reset] *ERROR* Failed to reset chip. output: GL_RENDERER = Mesa DRI Intel(R) Ironlake Desktop Probe at (80,80) Expected: 0.000000 1.000000 0.000000 1.000000 Observed: 0.000000 0.000000 0.000000 1.000000 Probe at (12,112) Expected: 0.000000 1.000000 0.000000 1.000000 Observed: 0.000000 0.000000 0.000000 1.000000 PIGLIT: {'result': 'fail' } Reproduce steps: ---------------------------- 1. xinit 2. ./bin/fp-kil -auto
The large range of commits covers my ARB_fp nuke series, so that was probably it. Sad to see no bisect in QA reports.
Bisect shoes:97615b2d8c7c3cea6fd3a43bcb1739a96e2046c4 is the first bad commit commit 97615b2d8c7c3cea6fd3a43bcb1739a96e2046c4 Author: Eric Anholt <eric@anholt.net> AuthorDate: Mon Aug 27 14:35:01 2012 -0700 Commit: Eric Anholt <eric@anholt.net> CommitDate: Mon Oct 8 08:50:27 2012 -0700 i965: Replace brw_wm_* with dumping code into the fs_visitor. This makes a giant pile of code newly dead. It also fixes TXB on newer chipsets, which has been totally broken (I now have a piglit test for that It passes the same set of Ian's ARB_fragment_program tests. It also impro high-settings ETQW performance by 3.2 +/- 1.9% (n=3), thanks to better optimization and having 8-wide along with 16-wide shaders. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=24355 Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
To clarify: This apparently only happens on ILK and not any other platform. I've tested IVB and GM45 personally, so I need to get the ILK up and test it.
commit 97747ac88fe9207dde83b275a93ebcee3a84dc01 Author: Eric Anholt <eric@anholt.net> Date: Wed Nov 14 11:44:57 2012 -0800 i965: Fix hangs with FP KIL instructions pre-gen6.
Verified.Fixed by commit 97747ac88fe9207dde83b275a93ebcee3a84dc01.
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.