[Bug 110877] [CI][DRMTIP] igt at perf_pmu@rc6 - skip - Test requirement: wait_for_rc6(fd), SKIP

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Mon Jun 24 18:12:12 UTC 2019


https://bugs.freedesktop.org/show_bug.cgi?id=110877

--- Comment #2 from Don Hiatt <don.hiatt at intel.com> ---
The 'tests/perf_pmu.c' has three testings checking for RC6 residency reporting
("rc6," "rc6-runtime-pm," and "rc6-runtime-pm-long").

In this case the IGT requirement wait_for_rc6() is failing and the test is
being skipped accordingly.

wait_for_rc6() waits for a 'RC6 evaluation interval' and then checks the sysfs
to see if we entered RC6 or not. Since this test is failing it looks like we
are not going into RC6. 

This appears to be happening on 'fi-icl-guc'. We need to determine if we are
indeed not going into RC6 or if it's just a test issue (e.g. perhaps the
timeout needs to be increased). 

We originally saw this bug on drm_tip_302 (2 weeks, 3 days old) and then again
in drm_tip_303 (1 week, 6 days old). I'm not sure if there were any runs on
'fi-icl-guc' between the first and second run, or if there is some sort of
timing issue that makes this bug difficult to reproduce.

If RC6 is broken on 'fi-icl-guc' that would be of high importance to the end
user of that machine, however if it's just and artifact of the testing
procedure then it should be fixed but it's a lower priority.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the QA Contact for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190624/43b1c76d/attachment.html>


More information about the intel-gfx-bugs mailing list