[Bug 112135] [CI][RESUME]igt at kms_flip@flip-vs-suspend - dmesg-warn - *ERROR* Restoring old state failed with -22

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=112135

--- Comment #3 from Ville Syrjala <ville.syrjala at linux.intel.com> ---
<7> [1291.127757] [drm:intel_dp_start_link_train [i915]] [CONNECTOR:307:DP-3]
Link Training Passed at Link Rate = 810000, Lane count = 4
...
<7> [1352.732720] [drm:intel_dp_start_link_train [i915]] [CONNECTOR:307:DP-3]
Link Training failed at link rate = 810000, lane count = 4
...
<7> [1367.316904] [drm:intel_atomic_check [i915]] [CONNECTOR:307:DP-3] Limiting
display bpp to 24 instead of EDID bpp 24, requested bpp 36, max platform bpp 36
<7> [1367.316963] [drm:intel_dp_compute_config [i915]] DP link computation with
max lane count 4 max rate 540000 max bpp 24 pixel clock 967000KHz
<7> [1367.316994] [drm:intel_dp_compute_config [i915]] Force DSC en = 0
<7> [1367.317026] [drm:intel_atomic_check [i915]] Encoder config failure: -22
...
<3> [1367.318852] [drm:intel_display_resume [i915]] *ERROR* Restoring old state
failed with -22

So the problem is that we originally picked 8.1 GHz link rate (which the mode
requires due to the high dotclock, later on link training fails and we reduce
the max link rate to 5.4GHZ which is not enough for the mode we already
configured, and then the resume fails when it tries to restore it. Could result
in a black screen after waking from suspend. In fact I suspect it will cause
all displays to be black because the entire atomic commit across all pipes is
declared as a failure.

Not sure if we should maybe just reset the max link params before resume...

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee 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/20191028/c3e0d302/attachment.html>