|Summary:||[HSW/BDW/BXT] Aztec Ruins Vulkan version GPU hangs|
|Product:||Mesa||Reporter:||Eero Tamminen <eero.t.tamminen>|
|Component:||Drivers/DRI/i965||Assignee:||Intel 3D Bugs Mailing List <intel-3d-bugs>|
|Status:||RESOLVED MOVED||QA Contact:||Intel 3D Bugs Mailing List <intel-3d-bugs>|
|i915 platform:||i915 features:|
Description Eero Tamminen 2018-01-15 09:40:07 UTC
Aztec Ruins test *Vulkan* version in the proprietary GfxBench v5.0-GOLD2 benchmark suite GPU hangs frequently on BXT J3455 (12 EU version) and occasionally on HSW & BDW. Setup: - Mesa, drm-tip kernel & X server from git - Ubuntu 16.04 (Hangs happen also with at least few months older version of Mesa, kernel and X server, but Mesa has had earlier other issues with this use-case, so much older data isn't very useful.) Use-case example: bin/testfw_app --gfx vulkan --gl_api vulkan --width 1920 --height 1080 --fullscreen 1 --test_id vulkan_5_normal Hang seems to happen on start of the benchmark run. If that's passed without hang, rest of test didn't seem to cause hangs. This test works fine on BXT J4205 (18 EU version). GL version of the test works fine on all of these devices. Both GL & Vulkan version use DXT5 textures. This doesn't seem to be a regression, but it seems to have became somewhat more frequent on that BXT device somewhere early December. (Filed this to Bugzilla on request from Jason. As it's a not a public test-case, binary, GPU error states & other info is in the internal ticket #347.)
Comment 1 Eero Tamminen 2018-10-08 11:47:28 UTC
Aztec Ruins Vulkan version is still frequently hanging on HSW (unlike GL version), but not anymore on the other machines. Although there aren't hangs on J3455 anymore, Aztec Ruins Vulkan version still isn't working correctly there as it gets bogus results.
Comment 2 GitLab Migration User 2019-09-25 19:07:33 UTC
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/1677.