[Bug 111214] [CI][BAT] igt at i915_module_load@reload - timeout - Unable to sync register 0x2f8100

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=111214

Kai Vehmanen <kai.vehmanen at intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kai.vehmanen at intel.com

--- Comment #18 from Kai Vehmanen <kai.vehmanen at intel.com> ---
(In reply to Cezary Rojewski from comment #4)
> Problem seems to exist since SKL and isn't strictly tied to ICL. The only
> reason it is visible now, is probably of yet another issue present for ICL
> platforms. We need to fix the root cause though. GFX guys must be engaged
> and comment on this.

This seems to be the case. I'm debugging this with SOF audio driver and it
seems to match this description:

"[ICL] occasional display audio codec probe failure - fail to get afg sub
nodes"
https://github.com/thesofproject/linux/issues/1184

The reproduction rate goes up when an arbitrary delay is added between display
probe and i915 audio component setup. Bug is fixed if i915 power domains are
kept up during the whole sequence. Latest analysis in comment:
https://github.com/thesofproject/linux/issues/1184#issuecomment-532765982

-- 
You are receiving this mail because:
You are the QA Contact 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/20190919/a569aafc/attachment.html>