*** Bug 104406 has been marked as a duplicate of this bug. ***

Created attachment 136550 [details]
Test List

This tests has a warn on CFL QA

see test list attached (KMSwarns)

This tests has the same warn message:

igt@kms_atomic_transition@plane-all-modeset-transition
igt@kms_setmode@invalid-clone-single-crtc
igt@kms_setmode@invalid-clone-single-crtc-stealing
igt@kms_setmode@invalid-clone-exclusive-crtc
igt@kms_setmode@clone-exclusive-crtc
igt@kms_setmode@basic-clone-single-crtc
igt@kms_pipe_crc_basic@bad-nb-words-1
igt@kms_pipe_crc_basic@bad-pipe
igt@kms_pipe_crc_basic@bad-source
igt@kms_properties@get_properties-sanity

IGT-Version: 1.20-gd86d53b (x86_64) (Linux: 4.15.0-rc5-drm-tip-config-intel-turbo-ww1-commit-16432d3+ x86_64)

(kms_setmode:15076) igt-kms-WARNING: no modes for connector 49
(kms_setmode:15076) igt-kms-WARNING: no modes for connector 49
(kms_setmode:15076) igt-kms-WARNING: no modes for connector 49
(kms_setmode:15076) igt-kms-WARNING: no modes for connector 49
(kms_setmode:15076) igt-kms-WARNING: no modes for connector 49
- - - -
(kms_pipe_crc_basic:13621) igt-kms-WARNING: connector 49/eDP-1 has no modes

Could you provide a full dmesg log containing the WARN?

(In reply to Imre Deak from comment #4)
> Could you provide a full dmesg log containing the WARN?

Also, is the problem reproducible with drm-tip?

Quickly checked from GFX QA data that tests below has been pass after this time when bug is posted. All below are pass without dmesg error from CommitDate: Fri Jan 5 09:12:48 2018 +0000 onward. Just wondering if there has been glitch on build or test procedure for causing this, since assumable has been on drm-tip for a month before this execution was done.

Elizabeth and Imre, assuming that this could be marked as resolved due to not failing on last (month). Please reopen if you disagree.

Sorry for the delay. Thank you for checking Jari. I tried to reproduce without success. If we meet it again I'll reopen it.

On today's FF we get a different warn, but it seems the same issue except with DP instead of eDP:

igt@kms_flip@basic-flip-vs-wf_vblank

Stdout	
IGT-Version: 1.21-g37bd27f (x86_64) (Linux: 4.15.0-drm-intel-qa-ww6-commit-078873d+ x86_64)
Using monotonic timestamps
Beginning basic-flip-vs-wf_vblank on pipe A, connector eDP-1
  1920x1080 60 1920 1936 1952 2104 1080 1083 1097 1116 0xa 0x48 141000
Expected frametime: 16653us; measured 16653.4us +- 1.928us accuracy 0.03%

basic-flip-vs-wf_vblank on pipe A, connector eDP-1: PASSED

Beginning basic-flip-vs-wf_vblank on pipe B, connector eDP-1
  1920x1080 60 1920 1936 1952 2104 1080 1083 1097 1116 0xa 0x48 141000
Expected frametime: 16653us; measured 16653.1us +- 2.594us accuracy 0.05%

basic-flip-vs-wf_vblank on pipe B, connector eDP-1: PASSED

Beginning basic-flip-vs-wf_vblank on pipe A, connector DP-2
  1920x1080 60 1920 2008 2052 2200 1080 1084 1089 1125 0x5 0x48 148500
Expected frametime: 16667us; measured 16666.5us +- 4.619us accuracy 0.08%

basic-flip-vs-wf_vblank on pipe A, connector DP-2: PASSED

Beginning basic-flip-vs-wf_vblank on pipe B, connector DP-2
  1920x1080 60 1920 2008 2052 2200 1080 1084 1089 1125 0x5 0x48 148500
Expected frametime: 16667us; measured 16666.9us +- 8.555us accuracy 0.15%

basic-flip-vs-wf_vblank on pipe B, connector DP-2: PASSED

Subtest basic-flip-vs-wf_vblank: SUCCESS (13.552s)
Stderr	
(kms_flip:22561) igt-kms-WARNING: connector 56/DP-1 has no modes
(kms_flip:22561) igt-kms-WARNING: connector 56/DP-1 has no modes

I'll try to reproduce manually.

This is probably a different bug from the original.

The original is probably that pruned eDP modes that got fixed with:
a306343bcd7d ("drm/i915/edp: Do not do link training fallback or prune modes on EDP")
CommitDate: Thu Feb 1 07:32:54 2018 -0800

Could you provide logs and more info on this new warning?
Do you really have 1 eDP and 2 DP plugged on this CFL? how is your current setup there?

Is it related to MST? if so a better candidate for that is:
https://bugs.freedesktop.org/show_bug.cgi?id=104158

(In reply to Rodrigo Vivi from comment #10)
> This is probably a different bug from the original.
> 
> The original is probably that pruned eDP modes that got fixed with:
> a306343bcd7d ("drm/i915/edp: Do not do link training fallback or prune modes
> on EDP")
> CommitDate: Thu Feb 1 07:32:54 2018 -0800
> 
> Could you provide logs and more info on this new warning?
> Do you really have 1 eDP and 2 DP plugged on this CFL? how is your current
> setup there?
> 
> Is it related to MST? if so a better candidate for that is:
> https://bugs.freedesktop.org/show_bug.cgi?id=104158

I tried to reproduce with latest:

(02:49 AM) [gfx@CFL-1] [tests]$ : sudo -E ./kms_flip --r basic-flip-vs-wf_vblank
IGT-Version: 1.21-g37bd27f (x86_64) (Linux: 4.15.0-drm-intel-qa-ww6-commit-7e8e106+ x86_64)
Using monotonic timestamps
Beginning basic-flip-vs-wf_vblank on pipe A, connector eDP-1
  1920x1080 60 1920 1966 1996 2080 1080 1082 1086 1112 0xa 0x48 138780
Expected frametime: 16666us; measured 16665.7us +- 5.782us accuracy 0.10%
...............
basic-flip-vs-wf_vblank on pipe A, connector eDP-1: PASSED

Beginning basic-flip-vs-wf_vblank on pipe B, connector eDP-1
  1920x1080 60 1920 1966 1996 2080 1080 1082 1086 1112 0xa 0x48 138780
Expected frametime: 16666us; measured 16666.2us +- 6.605us accuracy 0.12%
................
basic-flip-vs-wf_vblank on pipe B, connector eDP-1: PASSED

Beginning basic-flip-vs-wf_vblank on pipe A, connector DP-1
  1920x1080 60 1920 2008 2052 2200 1080 1084 1089 1125 0x9 0x48 148500
Expected frametime: 16667us; measured 16667.2us +- 6.705us accuracy 0.12%
...............
basic-flip-vs-wf_vblank on pipe A, connector DP-1: PASSED

Beginning basic-flip-vs-wf_vblank on pipe B, connector DP-1
  1920x1080 60 1920 2008 2052 2200 1080 1084 1089 1125 0x9 0x48 148500
Expected frametime: 16667us; measured 16666.6us +- 5.252us accuracy 0.09%
...............
basic-flip-vs-wf_vblank on pipe B, connector DP-1: PASSED

Subtest basic-flip-vs-wf_vblank: SUCCESS (14.188s)

Without success, besides our FF from yesterday and today didn't show any warn. 
I'm closing issue again, and if the warn appears again I'll fill a new bug for DP. 
Thank you for your time.