[igt-dev] [PATCH igt] tests/kms_frontbuffer_tracking: increase FBC wait timeout to 5s

[Lofstedt, Marta]

> -----Original Message-----
> From: igt-dev [mailto:igt-dev-bounces at lists.freedesktop.org] On Behalf Of
> Chris Wilson
> Sent: Wednesday, March 21, 2018 3:06 AM
> To: Zanoni, Paulo R <paulo.r.zanoni at intel.com>; igt-
> dev at lists.freedesktop.org
> Subject: Re: [igt-dev] [PATCH igt] tests/kms_frontbuffer_tracking: increase
> FBC wait timeout to 5s
> 
> Quoting Paulo Zanoni (2018-03-21 00:29:45)
> > From: "Lofstedt, Marta" <marta.lofstedt at intel.com>
> >
> > This is a partial revert of:
> > 64590c7b768dc8d8dd962f812d5ff5a39e7e8b54 ("kms_frontbuffer_tracking:
> > reduce the FBC wait timeout to 2s")
> >
> > The subtests: igt at kms_frontbuffer_tracking@fbc-*draw* have
> > non-consistent results, pending between fail and pass. The fails are
> > always due to "FBC disabled".  With this increase in timeout the
> > flip-flop behavior is no longer reproducible.
> 
> We haven't seen this behaviour for a few months, afaict. Currently we have
> FBC perma-disabled due to bad mode, underruns and whatnot, and flip flops
> from CRC mismatches. But no FBC timeouts waiting for enable.
> -Chris

I agree with Chris that there are multiple flip-flopping going on for the kms_frontbuffer_tracking tests, this really taps into the limitation of cibuglog, but the original issue is still happening. However, I don't think that the timeout increase is a good solution since it will increase the runtime significantly on the machines that continuously fail this test. See, the results from the drmtip runs https://intel-gfx-ci.01.org/tree/drm-tip/drmtip.html 

Also, after we started to feed data from the drmtip runs into cibuglog the impact of this issue has increased drastically. The reason is that I file all occurrences of and error message with:
"(kms_frontbuffer_tracking:2783) CRITICAL: FBC disabled", for example: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_4350/shard-glkb2/igt@kms_frontbuffer_tracking@fbc-indfb-scaledprimary.html, I file  on this bug: https://bugs.freedesktop.org/show_bug.cgi?id=101623

Error messages like this: " (kms_frontbuffer_tracking:2883) CRITICAL: Failed assertion: !fbc_wait_until_enabled()" ,
https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3912/shard-glkb4/igt@kms_frontbuffer_tracking@fbc-rgb101010-draw-blt.html
I file on https://bugs.freedesktop.org/show_bug.cgi?id=105503 for GLK-shards and https://bugs.freedesktop.org/show_bug.cgi?id=105520 for the drmtip runs.

BR,
Marta

> _______________________________________________
> igt-dev mailing list
> igt-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/igt-dev