Bug 111184

The CI Bug Log issue associated to this bug has been updated.

### New filters associated

* ICL:  igt@kms_flip@flip-vs-panning-vs-hang-interruptible - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], x_ofs, 0) == 0
  - https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_327/fi-icl-u4/igt@kms_flip@flip-vs-panning-vs-hang-interruptible.html
  - https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_328/fi-icl-u4/igt@kms_flip@flip-vs-panning-vs-hang-interruptible.html

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

{- ICL:  igt@kms_flip@flip-vs-panning-vs-hang-interruptible - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], x_ofs, 0) == 0 -}
{+ ICL:  igt@kms_flip@flip-vs-panning-vs-hang-interruptible|igt@kms_flip@basic-flip-vs-modeset - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], x_ofs, 0) == 0 +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_327/fi-icl-u4/igt@kms_flip@basic-flip-vs-modeset.html

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

{- ICL:  igt@kms_flip@flip-vs-panning-vs-hang-interruptible|igt@kms_flip@basic-flip-vs-modeset - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], x_ofs, 0) == 0 -}
{+ ICL:  igt@kms_flip@flip-vs-panning-vs-hang-interruptible|igt@kms_flip@basic-flip-vs-modeset - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], (x_ofs|0), 0) == 0 +}


  No new failures caught with the new filter

fi-icl-u4 has two USB Type-C port connected to HDMI panel and DP panel and also eDP panel is connected.

If you see the IGT assertion for set_mode() for different tests, they are occurring when the DP4 port is tried for a modeset with different pipes.
Can see the slight sporadic in nature.
But all the time link training for DP4 is failing with dmesg like
[536.165831] [drm:intel_dp_start_link_train [i915]] [CONNECTOR:223:DP-4] Link Training failed at link rate = 540000, lane count = 1 at https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_327/fi-icl-u4/igt@kms_flip@flip-vs-panning-vs-hang-interruptible.html

So I am suspecting this could be mostly related to the panel or the DP to HDMI converter used here. Still, we need to figure out that there is no specific HW at DP4 than DP3, because of which DP4 alone failing. 
And this issue is reproduced in the latest run too with the reproducing rate of 4/8 times

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

{- ICL:  igt@kms_flip@flip-vs-panning-vs-hang-interruptible|igt@kms_flip@basic-flip-vs-modeset - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], (x_ofs|0), 0) == 0 -}
{+ ICL:  igt@kms_flip@flip-vs-panning-vs-hang-*|igt@kms_flip@basic-flip-vs-modeset - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], (x_ofs|0), 0) == 0 +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_341/fi-icl-u4/igt@kms_flip@flip-vs-panning-vs-hang.html
  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_340/fi-icl-u4/igt@kms_flip@flip-vs-panning-vs-hang.html

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

{- ICL:  igt@kms_flip@flip-vs-panning-vs-hang-*|igt@kms_flip@basic-flip-vs-modeset - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], (x_ofs|0), 0) == 0 -}
{+ ICL:  igt@kms_flip@* - fail -  Failed assertion: set_mode(o, o->fb_ids[o->current_fb_id], (x_ofs|0), 0) == 0 +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_354/fi-icl-u4/igt@kms_flip@2x-flip-vs-modeset-vs-hang-interruptible.html

This is another duplicate of bug 111185 --- the failures here arise not from what the test is trying to exercise, but rather from a general failure to set the desired display mode due to link training failures.

Interestingly, the link appears less flaky in these logs (i.e., we can drive at least two lanes on DP-4 and four lanes on DP-3, just not at maximum speed), but since we're also attempting a 4k mode here, the required data rate still winds up exceeding the link's maximum bandwidth causing the modeset (and the test) to fail.

*** This bug has been marked as a duplicate of bug 111185 ***