[Bug 103166] [CI] igt at kms_plane-*@* - Failed CRC mismatch

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=103166

--- Comment #42 from Martin Peres <martin.peres at free.fr> ---
(In reply to Daniel Vetter from comment #39)
> We need to close this one, this bug has become a mess:

It has always been, but yeah, it is high time we do something about this bug!

> 
> - no more fifo underruns it seems

\o/

> 
> At least 2 new bugs:
> 
> 1. for igt at kms_plane@pixel-format-pipe-a-planes (also b/c ofc) on icl-u3:
> 
> Crash signature is:
> 
> Testing format NV12 (0x3231564e) on A.2
> Stack trace:
>   #0 ../lib/igt_core.c:1474 __igt_fail_assert()
>   #1 ../lib/igt_debugfs.c:420 igt_assert_crc_equal()
>   #2 ../tests/kms_plane.c:550 test_format_plane()
>   #3 ../tests/kms_plane.c:582 test_pixel_formats()
>   #4 ../tests/kms_plane.c:597 run_tests_for_pipe_plane()
>   #5 ../tests/kms_plane.c:652 __real_main637()
>   #6 ../tests/kms_plane.c:637 main()
>   #7 ../csu/libc-start.c:344 __libc_start_main()
>   #8 [_start+0x2a]
> Subtest pixel-format-pipe-A-planes: FAIL (547.114s)
> Test requirement not met in function run_tests_for_pipe_plane, file
> ../tests/kms_plane.c:593:
> Test requirement: !(pipe >= data->display.n_pipes)
> 
> A.2 can also be B.2 for b-planes and C.2 for c-planes

The test got reworked, and kms_plane issues on ICL are now tracked in:
https://bugs.freedesktop.org/show_bug.cgi?id=110032

APL issues got moved to https://bugs.freedesktop.org/show_bug.cgi?id=110033

The others will get a per-platform bug as they pop up :)

(In reply to Daniel Vetter from comment #39)
> 2. bug: igt at kms_plane_multiple@atomic * on icl-u3: No idea yet what's going
> on there.

The issue has been split into this bug:
https://bugs.freedesktop.org/show_bug.cgi?id=110037

> 
> 3. All the other platforms: Not sure it's a good idea to smash them all into
> one place. We need to at least separate the bugs where we also have a fifo
> underrun from crc mistmatches for other reasons. Since my igt patch
> 
> commit 4d0d81a2176227a7432762ae095ba386f3c8aba0
> Author: Daniel Vetter <daniel.vetter at ffwll.ch>
> Date:   Tue Feb 19 10:11:06 2019 +0100
> 
>     lib/crc: Reset fifo underrun reporting for every crc capture
> 
> we should be able to tell these two cases apart with 100% accuracy.

Agreed... but not too sure how to proceed on this. I should create a view that
would allow me to mass-suppress tests in order to deal with this sort of
issues...

I also split the igt at kms_plane@plane-position-covered-pipe-[abc]-planes
failures to https://bugs.freedesktop.org/show_bug.cgi?id=110038

I will suppress some of the tests and remove them from the filter.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190313/96c33394/attachment.html>