[igt-dev] ✗ Fi.CI.IGT: failure for lib/i915/perf: Fix non-card0 processing (rev4)

[Petri Latvala]

On Thu, May 06, 2021 at 12:41:07PM +0200, Janusz Krzysztofik wrote:
> On czwartek, 6 maja 2021 10:42:55 CEST Patchwork wrote:
> > == Series Details ==
> > 
> > Series: lib/i915/perf: Fix non-card0 processing (rev4)
> > URL   : https://patchwork.freedesktop.org/series/89701/
> > State : failure
> > 
> > == Summary ==
> > 
> > CI Bug Log - changes from IGT_6080_full -> IGTPW_5786_full
> > ====================================================
> > 
> > Summary
> > -------
> > 
> >   **FAILURE**
> > 
> >   Serious unknown changes coming with IGTPW_5786_full absolutely need to be
> >   verified manually.
> >   
> >   If you think the reported changes have nothing to do with the changes
> >   introduced in IGTPW_5786_full, please notify your bug team to allow them
> >   to document this new failure mode, which will reduce false positives in 
> CI.
> > 
> >   External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_5786/
> index.html
> > 
> > Possible new issues
> > -------------------
> > 
> >   Here are the unknown changes that may have been introduced in 
> IGTPW_5786_full:
> > 
> > ### IGT changes ###
> > 
> > #### Possible regressions ####
> > 
> >   * igt at kms_big_fb@y-tiled-16bpp-rotate-180:
> >     - shard-tglb:         [PASS][1] -> [FAIL][2]
> >    [1]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_6080/shard-tglb6/
> igt at kms_big_fb@y-tiled-16bpp-rotate-180.html
> >    [2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_5786/shard-tglb5/
> igt at kms_big_fb@y-tiled-16bpp-rotate-180.html
> > 
> >   * igt at kms_cursor_crc@pipe-c-cursor-128x42-sliding:
> >     - shard-apl:          [PASS][3] -> [FAIL][4]
> >    [3]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_6080/shard-apl2/
> igt at kms_cursor_crc@pipe-c-cursor-128x42-sliding.html
> >    [4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_5786/shard-apl8/
> igt at kms_cursor_crc@pipe-c-cursor-128x42-sliding.html
> >     - shard-kbl:          [PASS][5] -> [FAIL][6] +1 similar issue
> >    [5]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_6080/shard-kbl2/
> igt at kms_cursor_crc@pipe-c-cursor-128x42-sliding.html
> >    [6]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_5786/shard-kbl2/
> igt at kms_cursor_crc@pipe-c-cursor-128x42-sliding.html
> 
> The change only affects two tests which are calling intel_perf_for_fd(): perf 
> and core_hotunplug.  In the two latter cases reported above, none of those two 
> tests were executed in the same run before the reported failure occurred so 
> there is no way for the change to be responsible for that failure.
> 
> In the first case of the above three, a perf test was indeed executed a few 
> steps before the reported failure occurred, but for me it's hard to understand 
> how that could be the reason of that failure, close to impossible.  Lakshmi, 
> can we please retry?

Retest queued.


-- 
Petri Latvala