[Bug 104782] [CI] igt at kms_color@pipe-[abc]-*gamma - fail - CRC mismatch

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=104782

--- Comment #15 from Uma Shankar <uma.shankar at intel.com> ---
This issue is happening due to pipe rounding (an expected behaviour in such
scenario, confirmed with hw architect Runyan, Art as well) causing crc to
mismatch. Unfortunately the test relies on the crc.
But checked on simulation and dump the pipe bmp output and compared bit by bit
with the expected output. The data was matching perfectly. There will no
visible impact to user as the framebuffer content is perfect, so user will not
be able to see any issues. 

Future actions planned is :
  -  Use pipe writeback to get the framebuffer and check the pixel values
directly (need to implement it first)
  - Use chamelium as a workaround for the lack of pipe writeback

All this will need significant implementation on driver side and even the tests
need to re-written to align this. We will work on it separately.

--- Comment #16 from CI Bug Log <cibuglog at gmail.com> ---
A CI Bug Log filter associated to this bug has been updated:

{- BSW BXT APL SKL KBL GLK CFL WHL ICL: igt at kms_color@pipe-[abc]-*gamma* -
Failed assertion: !mismatch -}
{+ BYT BSW BXT APL SKL KBL GLK CFL WHL ICL: igt at kms_color@pipe-[abc]-*gamma* -
Failed assertion: !mismatch +}

New failures caught by the filter:

*
https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_249/fi-byt-clapper/igt@kms_color@pipe-a-legacy-gamma.html

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190409/530e3bfa/attachment-0001.html>