Summary: | [IGT] drv_selftest subtest mock_evict/live_contexts incompletes | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | DRI | Reporter: | Elizabeth <elizabethx.de.la.torre.mena> | ||||||
Component: | DRM/Intel | Assignee: | Intel GFX Bugs mailing list <intel-gfx-bugs> | ||||||
Status: | CLOSED FIXED | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> | ||||||
Severity: | normal | ||||||||
Priority: | medium | CC: | intel-gfx-bugs | ||||||
Version: | unspecified | ||||||||
Hardware: | Other | ||||||||
OS: | All | ||||||||
Whiteboard: | ReadyForDev | ||||||||
i915 platform: | CFL, KBL, SKL | i915 features: | |||||||
Attachments: |
|
Description
Elizabeth
2017-10-24 20:25:47 UTC
(In reply to Elizabeth from comment #0) > Created attachment 135023 [details] > dmesg_mock_evict > > We are hitting same problem as described in bug 102973 with latest kernel: No, that is not the same bug. > $uname -a > Linux SKL-5-NUC6i7KYB 4.14.0-rc6-drm-intel-qa-ww43-commit-5c82a37+ #1 SMP > Tue Oct 24 07:34:21 CDT 2017 x86_64 x86_64 x86_64 GNU/Linux > > With drv_selftest@mock_evict and drv_selftest@live_uncore. mock_evict and live_uncore are two very different tests. Created attachment 135024 [details]
dmesg_live_contexts
drv_selftests@live_contexts also has the same behavior, once the test start running, the test gets stuck forever, though dmesg error is different for this test, it sent a warn:
WARNING: CPU: 6 PID: 72 at drivers/gpu/drm/i915/i915_gem.c:4553 __i915_gem_free_objects+0x2ad/0x2c0 [i915]
(In reply to Chris Wilson from comment #1) > (In reply to Elizabeth from comment #0) > ... > mock_evict and live_uncore are two very different tests. live_uncore re-tested. Platform "dies", no display output nor ssh, after command: $sudo -E ./drv_selftests --r live_uncore until power reset. Opening a new bug for this test. Reference to: https://patchwork.freedesktop.org/series/32576/ commit 20ccd4d3f689ac14dce8632d76769be0ac952060 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 24 23:08:55 2017 +0100 drm/i915: Use same test for eviction and submitting kernel context During evict, we wish to idle the GPU if we see that the GGTT is full. However, our test for idle in i915_gem_evict_something() and in i915_gem_switch_to_kernel_context() do not match leading to disappointment - we never believe that we are idle and keep trying to flush the GGTT ad infinitum. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103438 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20171024220855.30155-2-chris@chris-wilson.co.uk Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Fixes the first issue of mock_evict. The trace for live_contexts needs kasan and a confirmation that you are carrying the core-for-CI fixups? Also since it is a separate bug, please refile it. (03:04 AM) [gfx@SKL-5-NUC6i7KYB] [~]$ : uname -a Linux SKL-5-NUC6i7KYB 4.14.0-rc6-drm-tip-ww43-commit-619a850+ #1 SMP Wed Oct 25 09:14:59 CDT 2017 x86_64 x86_64 x86_64 GNU/Linux (03:05 AM) [gfx@SKL-5-NUC6i7KYB] [~]$ : sudo -E ./intel-graphics/intel-gpu-tools/tests/drv_selftest --r mock_evict --d IGT-Version: 1.20-ge7742ee (x86_64) (Linux: 4.14.0-rc6-drm-tip-ww43-commit-619a850+ x86_64) (drv_selftest:1985) igt-kmod-DEBUG: Test requirement passed: err == 0 || err == -ENOENT (drv_selftest:1985) igt-kmod-DEBUG: Test requirement passed: igt_kselftest_begin(&tst) == 0 (drv_selftest:1985) igt-core-DEBUG: Starting subtest: mock_evict Subtest mock_evict: SUCCESS (0.274s) (drv_selftest:1985) igt-kmod-DEBUG: Test requirement passed: !igt_list_empty(&tests) (drv_selftest:1985) igt-kmod-DEBUG: Test requirement passed: err == 0 || err == -ENOENT (drv_selftest:1985) igt-kmod-DEBUG: Test requirement passed: igt_kselftest_begin(&tst) == 0 (drv_selftest:1985) igt-kmod-DEBUG: Test requirement passed: !igt_list_empty(&tests) (drv_selftest:1985) igt-core-DEBUG: Exiting with status code 0 Verified. Closing old verified. |
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.