Re: HEADS UP: Merging drm update

[Andrius V <vezhlys%gmail.com@localhost>]

Hi,

I potentially have the same issue as Ryo with my mac mini based on
Intel i5-4278U with Iris graphics 5100. After DRM upgrade system goes
blank on boot just before intelfb attachment (I guess) and likely
crash (no logs messages available it seems). Before drm upgrade it
also goes blank at the same time but restores the view soon after. I
didn't follow full thread, but I may help to investigate/test if
there's anything I can do. I am also not sure how to redirect boot
messages to serial attached through USB to see if any backtrace is
available.



On Tue, Jan 4, 2022 at 12:05 PM Ryo ONODERA <ryo%tetera.org@localhost> wrote:
>
> Hi,
>
> Taylor R Campbell <riastradh%NetBSD.org@localhost> writes:
>
> >> Date: Mon, 03 Jan 2022 22:08:26 +0900
> >> From: Ryo ONODERA <ryo%tetera.org@localhost>
> >>
> >> Ryo ONODERA <ryo%tetera.org@localhost> writes:
> >>
> >> > __engines_record_defaults
> >> > intel_gt_wait_for_idle
> >> > intel_gt_retire_requests_timeout
> >> > dma_fence_wait_timeout
> >> > i915_fence_wait (via *fence->ops->wait)
> >> > i915_request_wait
> >> >
> >> > In i915_request_wait, DRM_SPIN_TIMED_WAIT_UNTIL sets timeout=0
> >> > and i915_request_wait returns timeout=-ETIME.
> >
> > Thanks, I think there's something going wrong either with submitting
> > requests or noticing that they're submitted, but it only happens on
> > some machines.  It might be related to some of the other panics around
> > requests, like the cv_is_valid panic in signal_irq_work (56561), or it
> > might be something is mapped wrong or there's a missing cache flush /
> > bus_dmamap_sync...  Will investigate when I have a chance.
>
> Thank you.
>
> I will try to figure out my problem.
>
> --
> Ryo ONODERA // ryo%tetera.org@localhost
> PGP fingerprint = 82A2 DC91 76E0 A10A 8ABB  FD1B F404 27FA C7D1 15F3