[Bug 103184] [CI] igt at kms_draw_crc@draw-method-* - fail - CRC mismatch

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=103184

--- Comment #18 from Stanislav Lisovskiy <stanislav.lisovskiy at intel.com> ---
I've noticed an interesting correlation in dmesg logs for this bug: for ICL and
SKL, crc verification always fails when we get FIFO underrun in between:

 96.710052] [IGT] kms_draw_crc: starting subtest
draw-method-rgb565-mmap-cpu-xtiled
<7>[   96.710303] [drm:drm_mode_addfb2] [FB:108]
<7>[   96.710359] [drm:drm_mode_addfb2] [FB:108]
<7>[   96.754277] [drm:drm_mode_setcrtc] [CRTC:41:pipe A]
<7>[   96.754309] [drm:drm_mode_setcrtc] [CONNECTOR:71:eDP-1]
<7>[   96.854202] [drm:drm_mode_addfb2] [FB:108]
<7>[   97.055161] [drm:drm_mode_setcrtc] [CRTC:41:pipe A]
<7>[   97.055220] [drm:drm_mode_setcrtc] [CONNECTOR:71:eDP-1]
<3>[   97.116516] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] *ERROR* CPU
pipe A FIFO underrun
<7>[   97.116837] [drm:intel_fbc_underrun_work_fn [i915]] Disabling FBC due to
FIFO underrun.
<7>[   97.148943] [IGT] kms_draw_crc: exiting, ret=99

Same pattern happens for all cases, for those architectures.

I will continue investigating.

-- 
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/20180823/676b92d0/attachment.html>