Bug#1043564: linux: Please Re-enable DC states for drm/i915

[Jesse Rhodes <jesse@sney.ca>]

Ah, great! If only they were all this easy.

Thanks again,

sney

On Sun, Aug 13, 2023 at 1:44 AM Salvatore Bonaccorso <carnil@debian.org> wrote:
>
> Control: tags -1 + confirmed upstream fixed-upstream pending
>
> Hi,
>
> On Sat, Aug 12, 2023 at 10:40:09PM -0400, Jesse Rhodes wrote:
> > Source: linux
> > Severity: important
> > Tags: patch
> > X-Debbugs-Cc: jesse@sney.ca
> >
> > Dear debian kernel team,
> >
> > The upstream commit "drm/i915: Disable DC states for all commits" [1] breaks my
> > Asus C204M Chromebook's ability to wake the screen after power management shuts
> > it off. This also produces a kernel stack trace, attached. It's likely that it
> > has the same effect on other Intel Gemini Lake hardware or UHD 600 video
> > devices.
> >
> > I discovered this commit by bisecting between 6.1.0-7 (6.1.20-2) and
> > 6.1.0-8 (6.1.25-1), and confirmed that the current 6.1.0-10 (6.1.38-2) does not
> > have the issue if intel_display.c is rolled back to its previous state.
> >
> > Attached is also the lspci output for the affected machine.
> >
> > I hope that "Tags: patch" is allowed as a tag when the fix is to revert a
> > specific commit, and I apologize if this was already reported and I couldn't
> > find it in the list.
> >
> > Please let me know if you need any more information, as I would be happy to
> > keep debugging!
>
> Thanks for the report. The commit was found indeed to cause issues and
> was for now reverted upstream in the 6.1.y series and released in
> 6.1.45.
>
> The fix will be contained in the next bookworm point release update,
> have just added the bugcloser for this issue as well in the current
> rebase work:
>
> https://salsa.debian.org/kernel-team/linux/-/merge_requests/803/diffs?commit_id=050b84b3c21f5a5b5e15b562da2ffc9f0edfda3a
>
> Regards,
> Salvatore