Comment # 13 on bug 1011250 from Takashi Iwai

(In reply to Johannes Weberhofer from comment #12)
> So far I didn't have any issues reagarding the reported problems using the
> test-kernel.
> 
> What I see in the dmsg are the following messages:
> [86000.248277] snd_hda_codec_hdmi hdaudioC0D2: HDMI: ELD buf size is 0,
> force 128
> [86000.248382] snd_hda_codec_hdmi hdaudioC0D2: HDMI: invalid ELD data byte 0
> [86000.263372] [drm:gen8_irq_handler [i915]] *ERROR* CPU pipe C FIFO underrun
> [86000.263428] [drm:gen8_irq_handler [i915]] *ERROR* CPU pipe B FIFO underrun
> [86866.901476] [drm:gen8_irq_handler [i915]] *ERROR* CPU pipe A FIFO underrun
> 
> Whenever a "underrun" message occures, one of my thre screens turns off an -
> a few seconds later - on again.

These errors are irrelevant with the issue reported here.  The first two are
about your HDMI monitor that doesn't have an audio output.  It's no real error.
 I guess you had these beforehand, too.  The latter three are about usual i915
graphics FIFO underruns that can be seen in quite many situations.

The only question is whether these are new with the test kernel.  The test
kernel contains lots of the recent fixes.  If these error didn't appear in the
original 42.2 kernel, it's possibly a regression caused by some of them.
In that case, please open another bug to track further.