Bug 104241 - [CI] igt@perf_pmu@busy-vecs0 - fail - Failed assertion: (double)(val) <= (1.0 + (tolerance)) * (double)(ref) && (double)(val) >= (1.0 - (tolerance)) * (double)(ref) (perf_pmu:1587) CRITICAL: 'val' != 'ref' (539089723.000000 not within 5.000000% tolerance
Summary: [CI] igt@perf_pmu@busy-vecs0 - fail - Failed assertion: (double)(val) <= (1.0...
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: DRI git
Hardware: Other All
: medium normal
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard: ReadyForDev
Keywords:
Depends on:
Blocks:
 
Reported: 2017-12-13 12:58 UTC by Marta Löfstedt
Modified: 2018-01-02 08:58 UTC (History)
1 user (show)

See Also:
i915 platform: GLK
i915 features: Perf/OA

Attachments

Description Marta Löfstedt 2017-12-13 12:58:23 UTC 
There has been a lot of bugs on the perf_pmu tests, I couldn't find any obvious to reopen or duplicate this on so here is a new bug

https://intel-gfx-ci.01.org/tree/drm-tip/IGT_4060/shard-glkb6/igt@perf_pmu@busy-vecs0.html

(perf_pmu:1587) CRITICAL: Test assertion failure function single, file perf_pmu.c:159:
(perf_pmu:1587) CRITICAL: Failed assertion: (double)(val) <= (1.0 + (tolerance)) * (double)(ref) && (double)(val) >= (1.0 - (tolerance)) * (double)(ref)
(perf_pmu:1587) CRITICAL: 'val' != 'ref' (539089723.000000 not within 5.000000% tolerance of 500000000.000000)
Subtest busy-vecs0 failed.
Comment 1 Chris Wilson 2017-12-13 13:15:57 UTC 
(In reply to Marta Löfstedt from comment #0)
> There has been a lot of bugs on the perf_pmu tests, I couldn't find any
> obvious to reopen or duplicate this on so here is a new bug

Hadn't spotted an overrun one before, so yes this does make this case a bit different.
Comment 2 Chris Wilson 2017-12-22 20:51:09 UTC 
commit c0a60eb5096a39440c9599af7b5cc34c59e2c712
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Dec 15 14:33:36 2017 +0000

    igt/perf_pmu: Tighten busy measurement
    
    Sleep for a known duration. In particular, CI once saw a measurement for
    busyness greater than the intended batch_duration!
    
    v2: Go back to starting pmu sampling outside of spinner; the GPU should
    be idle.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104241
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
    Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
    Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

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.