| Summary: | [BSW] [BAT] prime_vgem | ||
|---|---|---|---|
| Product: | DRI | Reporter: | mwa <matthew.auld> |
| Component: | DRM/Intel | Assignee: | Intel GFX Bugs mailing list <intel-gfx-bugs> |
| Status: | CLOSED NOTABUG | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> |
| Severity: | blocker | ||
| Priority: | highest | CC: | intel-gfx-bugs |
| Version: | DRI git | ||
| Hardware: | x86-64 (AMD64) | ||
| OS: | Linux (All) | ||
| Whiteboard: | |||
| i915 platform: | BSW/CHT | i915 features: | GEM/Other |
|
Description
mwa
2016-08-24 09:37:15 UTC
Should not be possible... Is it sporadic? Does only that bsw seem affected? 4.7.0-rc7-CI-Nightly_684+ is odd. This test passing depends on the 4.8-rc1 merge. On latest execution (having 4.8.0-rc3) on fi-bsw-n3050 (and all other hw attached to CI) the error message has changed to: Detail Value Returncode 99 Time 0:00:01.263335 Stdout IGT-Version: 1.15-g65a9987 (x86_64) (Linux: 4.8.0-rc3-CI-CI_DRM_1585+ x86_64) Stack trace: #0 [__igt_fail_assert+0x101] #1 [__real_main729+0xf5a] #2 [<unknown>+0xf5a] Subtest basic-fence-wait-default: FAIL (1.042s) Test requirement not met in function __real_main729, file prime_vgem.c:825: Test requirement: vgem_fence_has_flag(vgem, WIP_VGEM_FENCE_NOTIMEOUT) Stderr (prime_vgem:11198) CRITICAL: Test assertion failure function test_fence_wait, file prime_vgem.c:468: (prime_vgem:11198) CRITICAL: Failed assertion: ptr[i] == 0 (prime_vgem:11198) CRITICAL: error: 0x1 != 0 Subtest basic-fence-wait-default failed. **** DEBUG **** (prime_vgem:11198) ioctl-wrappers-DEBUG: Test requirement passed: gem_has_ring(fd, ring) (prime_vgem:11198) DEBUG: Test requirement passed: !(gen == 6 && e->exec_id == I915_EXEC_BSD) (prime_vgem:11198) CRITICAL: Test assertion failure function test_fence_wait, file prime_vgem.c:468: (prime_vgem:11198) CRITICAL: Failed assertion: ptr[i] == 0 (prime_vgem:11198) CRITICAL: error: 0x1 != 0 **** END **** Environment PIGLIT_PLATFORM="mixed_glx_egl" PIGLIT_SOURCE_DIR="/opt/igt/piglit" Command /opt/igt/tests/prime_vgem --run-subtest basic-fence-wait-default dmesg Yes. fence-wait is know to fail right now as the patches to wait upon dma-buf are still in progress. Hopefully will be finished shortly. mwa, chris - do you still see the failure on basic-wait-default? Sorry - my previous comment was for wrong sub test - ignore it on this context. For couple of the last days the CI results for basic-wait-default have been pass for fi-bsw-n3050 (and other HW attached on CI). mwa, chris Detail Value Returncode 0 Time 0:00:00.228307 Stdout IGT-Version: 1.15-g3a3c0fa (x86_64) (Linux: 4.8.0-rc3-CI-CI_DRM_1588+ x86_64) Subtest basic-wait-default: SUCCESS (0.019s) Test requirement not met in function __real_main729, file prime_vgem.c:825: Test requirement: vgem_fence_has_flag(vgem, WIP_VGEM_FENCE_NOTIMEOUT) Stderr Environment PIGLIT_SOURCE_DIR="/opt/igt/piglit" PIGLIT_PLATFORM="mixed_glx_egl" Command /opt/igt/tests/prime_vgem --run-subtest basic-wait-default dmesg Looks fine now that fi-bsw-n3050 has caught up (root fs was corrupt apparently). |
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.