Maybe I would need to create a display/Chamelium feature.

This bug used to be hit very rarely (every few months) and only on fi-kbl-7500u.

However it's now also hit on ICL and CML, every week or so.

User impact: screen not detected when plugging it in.

Added a new check to make sure the test fails because of a NULL blob: https://patchwork.freedesktop.org/series/65315/

By "not detected" I mean "appears as not detected from a user PoV". Of course DRM detects it (since it's CONNECTED). However userspace will probably choke on the EDID being unset.

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

{- KBL: igt@kms_chamelium@dp-edid-read - fail - Failed assertion: edid_blob = drmModeGetPropertyBlob(data->drm_fd, edid_blob_id) -}
{+ KBL: igt@kms_chamelium@dp-edid-read - fail - Failed assertion: edid_blob = drmModeGetPropertyBlob(data->drm_fd, edid_blob_id) +}


  No new failures caught with the new filter

commit 0bf74af39ede5fa26195e4e7da590b3d0b729469
Author: Simon Ser <simon.ser@intel.com>
Date:   Fri Aug 16 16:47:13 2019 +0300

    tests/kms_chamelium: check whether EDID blob is non-NULL before trying to read it
    
    test_edid_read sets an EDID on a port, plugs it and reads it back. Sometimes
    retrieving the blob fails. Check whether the blob is set at all before trying
    to get it. This makes it obvious it's not drmModeGetPropertyBlob failing for
    another reason.
    
    Signed-off-by: Simon Ser <simon.ser@intel.com>
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106766
    Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>

*** Bug 111437 has been marked as a duplicate of this bug. ***

*** Bug 111474 has been marked as a duplicate of this bug. ***

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

{- KBL: igt@kms_chamelium@dp-edid-read - fail - Failed assertion: edid_blob = drmModeGetPropertyBlob(data->drm_fd, edid_blob_id) -}
{+ KBL: igt@kms_chamelium@dp-edid-read - fail - Failed assertion: edid_blob = drmModeGetPropertyBlob(data->drm_fd, edid_blob_id) +}


  No new failures caught with the new filter

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

{- KBL: igt@kms_chamelium@dp-edid-read - fail - Failed assertion: edid_blob = drmModeGetPropertyBlob(data->drm_fd, edid_blob_id) -}
{+ KBL: igt@kms_chamelium@dp-edid-read - fail - Failed assertion: edid_blob_id != 0 +}


  No new failures caught with the new filter

(In reply to emersion from comment #2)
> This bug used to be hit very rarely (every few months) and only on
> fi-kbl-7500u.
> 
> However it's now also hit on ICL and CML, every week or so.
> 
> User impact: screen not detected when plugging it in.
> 
> Added a new check to make sure the test fails because of a NULL blob:
> https://patchwork.freedesktop.org/series/65315/

Based on the impact bumping the severity and priority.

ASs this was merged, resolving to get CI feedback.

This issue is used to appear once in 125.2 CI_DRM runs. Last seen on CI_DRM_6673 (1 month, 2 weeks old).
Also this issue occurred only once on IGT runs GT_5174 (2 weeks, 1 day old). So, we have a wait before we close this issue. Let's come back after a month or two and see if can close.