Bug 105554

IIIRC when I was working on improving the gamma stuff recently (none of which has been posted yet) I was seeing some underruns when turning the fancy CHV color managagement blocks on/off. But I can't immediately recall if I figured out a way to avoid the underruns or not.

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_93/fi-bsw-kefka/igt@kms_color@pipe-b-ctm-green-to-red.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_93/fi-bsw-kefka/igt@kms_color@pipe-a-ctm-negative.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_93/fi-bsw-kefka/igt@kms_color@pipe-b-ctm-negative.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_93/fi-bsw-kefka/igt@kms_color@pipe-b-ctm-0-5.html

... and 60 other instances ...

This issue occurs only with drmtip build. Changing the priority to Medium.

A CI Bug Log filter associated to this bug has been updated:

{- BSW: igt@kms_color@pipe-[abc]-ctm-* - *ERROR* CPU pipe [ABC] FIFO underrun -}
{+ BSW: igt@kms_color@pipe-[abc]-ctm-*|igt@kms_plane_lowres@pipe-* -dmesg-warn - *ERROR* CPU pipe [ABC] FIFO underrun +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_258/fi-bsw-n3050/igt@kms_plane_lowres@pipe-c-tiling-none.html

A CI Bug Log filter associated to this bug has been updated:

{- BSW: igt@kms_color@pipe-[abc]-ctm-*|igt@kms_plane_lowres@pipe-* -dmesg-warn - *ERROR* CPU pipe [ABC] FIFO underrun -}
{+ BSW: igt@all tests -dmesg-warn - *ERROR* CPU pipe [ABC] FIFO underrun +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_289/fi-bsw-n3050/igt@kms_atomic_transition@plane-all-transition-nonblocking-fencing.html

This failure still happens almost everyday on BSW platform with drmtip. There are several tests that have this failure signature - 
igt@kms_color@*
igt@kms_busy@*
igt@kms_3d
igt@kms_hdmi_inject@inject-audio
igt@kms_draw_crc@draw-method-rgb565-blt-xtiled
.. and several other tests.

There are several known bugs with "CPU pipe FIFO underrun" warning in dmesg log with parent bug 105980

-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/intel/issues/89.