Bug 111216 - [CI][SHARDS] igt@perf_pmu@most-busy-idle-check-all-vecs0 - fail - Failed assertion: (double)(val[i]) <= (1.0 + (tolerance)) * (double)(slept) && (double)(val[i]) >= (1.0 - (tolerance)) * (double)(slept)
Summary: [CI][SHARDS] igt@perf_pmu@most-busy-idle-check-all-vecs0 - fail - Failed asse...
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: Other All
: high normal
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
Whiteboard: ReadyForDev
Depends on:
Reported: 2019-07-25 12:24 UTC by Martin Peres
Modified: 2019-08-09 07:45 UTC (History)
1 user (show)

See Also:
i915 platform: BXT
i915 features: Perf/PMU


Note You need to log in before you can comment on or make changes to this bug.
Description Martin Peres 2019-07-25 12:24:15 UTC

Starting subtest: most-busy-idle-check-all-vecs0
(perf_pmu:6551) CRITICAL: Test assertion failure function most_busy_check_all, file ../tests/perf_pmu.c:550:
(perf_pmu:6551) CRITICAL: Failed assertion: (double)(val[i]) <= (1.0 + (tolerance)) * (double)(slept) && (double)(val[i]) >= (1.0 - (tolerance)) * (double)(slept)
(perf_pmu:6551) CRITICAL: 'val[i]' != 'slept' (568744179.000000 not within +5.000000%/-5.000000% tolerance of 500083321.000000)
Comment 1 CI Bug Log 2019-07-25 12:24:59 UTC
The CI Bug Log issue associated to this bug has been updated.

### New filters associated

* APL: igt@perf_pmu@most-busy-idle-check-all-vecs0 - fail - Failed assertion: (double)(val[i]) &lt;= (1.0 + (tolerance)) * (double)(slept) &amp;&amp; (double)(val[i]) &gt;= (1.0 - (tolerance)) * (double)(slept)
  - https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_3289/shard-apl3/igt@perf_pmu@most-busy-idle-check-all-vecs0.html
Comment 2 Chris Wilson 2019-07-25 12:32:04 UTC
Just the usual random delay [.05s] between sampling the first and second engine before the sleep.
Comment 3 Chris Wilson 2019-08-09 07:45:14 UTC
I hereby claim that

commit c7302f204490f3eb4ef839bec228315bcd3ba43f
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Aug 8 21:27:58 2019 +0100

    drm/i915: Defer final intel_wakeref_put to process context

will improve this bug! (Money back not guaranteed.)

I think it's a CPU scheduler delay, but the changes in timing from the above commit might be enough to put us in its good graces. More likely the opposite!

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.