Fwd: 6.1.23: 0fc6fea41c71 breaks GPD Pocket 3 modeset

Fwd: 6.1.23: 0fc6fea41c71 breaks GPD Pocket 3 modeset

[Kurt Garloff]

Hi Ville,

(sorry for the resend, had forgotten that LKML still believes
that multipart with an HTML version is evil)

While
0fc6fea41c71 drm/i915: Disable DC states for all commits
(cherry picked from commit 41b4c7fe72b6105a4b49395eea9aa40cef94288d)
does look correct to me, it does break modesetting on the
GPD Pocket 3, a i7-1195G7 laptop. I run the kernel with
fbcon=rotate:1 video=DSI-1:panel_orientation=right_side_up \
   mem_sleep_default=s2idle
No special i915 parameters.

Hardware is described here:
https://wiki.archlinux.org/title/GPD_Pocket_3

I disected this patch which was merged (backported) for 6.1.23.
I currently run 6.1.26 with it reverted and life is good.

Without reverting it, when fbcon is switched to show the splash
screen (GPD logo with Ubuntu added in on working kernels), the
screen remains black (backlight on, but nothing displayed) and
nothing happens any more. It looks like something more serious
than just blackness, as SysRq or Ctrl-Alt-Del don't do anything.
Deadlocking?

Best,

-- 

Kurt Garloff<kurt@garloff.de>
Cologne, Germany

Re: Fwd: 6.1.23: 0fc6fea41c71 breaks GPD Pocket 3 modeset

[Linux regression tracking (Thorsten Leemhuis)]

[CCing the intel drm maintainers and the lists; also CCing the
regression list, as it should be in the loop for regressions:
https://docs.kernel.org/admin-guide/reporting-regressions.html]

[TLDR: I'm adding this report to the list of tracked Linux kernel
regressions; the text you find below is based on a few templates
paragraphs you might have encountered already in similar form.
See link in footer if these mails annoy you.]

On 29.04.23 23:38, Kurt Garloff wrote:
> Hi Ville,
> 
> (sorry for the resend, had forgotten that LKML still believes
> that multipart with an HTML version is evil)
> 
> While
> 0fc6fea41c71 drm/i915: Disable DC states for all commits
> (cherry picked from commit 41b4c7fe72b6105a4b49395eea9aa40cef94288d)

/me wonders if 6.3 is affected as well, but will treat it as stable
specific regression for now

> does look correct to me, it does break modesetting on the
> GPD Pocket 3, a i7-1195G7 laptop. I run the kernel with
> fbcon=rotate:1 video=DSI-1:panel_orientation=right_side_up \
>   mem_sleep_default=s2idle
> No special i915 parameters.
> 
> Hardware is described here:
> https://wiki.archlinux.org/title/GPD_Pocket_3
> 
> I disected this patch which was merged (backported) for 6.1.23.
> I currently run 6.1.26 with it reverted and life is good.
> 
> Without reverting it, when fbcon is switched to show the splash
> screen (GPD logo with Ubuntu added in on working kernels), the
> screen remains black (backlight on, but nothing displayed) and
> nothing happens any more. It looks like something more serious
> than just blackness, as SysRq or Ctrl-Alt-Del don't do anything.
> Deadlocking?

Thanks for the report. To be sure the issue doesn't fall through the
cracks unnoticed, I'm adding it to regzbot, the Linux kernel regression
tracking bot:

#regzbot ^introduced 0fc6fea41c71
#regzbot title drm: i915: GPD Pocket 3 modeset broken
#regzbot ignore-activity

This isn't a regression? This issue or a fix for it are already
discussed somewhere else? It was fixed already? You want to clarify when
the regression started to happen? Or point out I got the title or
something else totally wrong? Then just reply and tell me -- ideally
while also telling regzbot about it, as explained by the page listed in
the footer of this mail.

Developers: When fixing the issue, remember to add 'Link:' tags pointing
to the report (the parent of this mail). See page linked in footer for
details.

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
That page also explains what to do if mails like this annoy you.

Re: [Intel-gfx] 6.1.23: 0fc6fea41c71 breaks GPD Pocket 3 modeset

[Ville Syrjälä]

On Sat, Apr 29, 2023 at 09:55:45PM +0200, Kurt Garloff wrote:
> Hi Ville,
> 
> While
> 0fc6fea41c71 drm/i915: Disable DC states for all commits
> (cherry picked from commit 41b4c7fe72b6105a4b49395eea9aa40cef94288d)
> does look correct to me, it does break modesetting on the GPD Pocket 3,
> a i7-1195G7 laptop. I run the kernel with
> fbcon=rotate:1 video=DSI-1:panel_orientation=right_side_up \
> mem_sleep_default=s2idle
> No special i915 parameters.
> Hardware is described here:
> https://wiki.archlinux.org/title/GPD_Pocket_3
> 
> I disected this patch which was merged (backported) for 6.1.23.
> I currently run 6.1.26 with it reverted.
> 
> Without reverting it, when fbcon is switched to show the splash
> screen (GPD logo with Ubuntu added in on working kernels), the
> screen remains black (backlight on, but nothing displayed) and
> nothing happens any more.

Please file a bug at
https://gitlab.freedesktop.org/drm/intel/issues/new

Boot both kernels (revert vs. no revert), passing
'drm.debug=0xe log_buf_len=4M' to the kernel cmdline,
and attach the resulting dmesg from each to the bug.

Also would be good if you try to reproduce on the
latest drm-tip (git://anongit.freedesktop.org/drm-tip drm-tip)
as well.

-- 
Ville Syrjälä
Intel

Re: 6.1.23: 0fc6fea41c71 breaks GPD Pocket 3 modeset

[Ville Syrjälä]

On Sat, Apr 29, 2023 at 09:55:45PM +0200, Kurt Garloff wrote:
> Hi Ville,
> 
> While
> 0fc6fea41c71 drm/i915: Disable DC states for all commits
> (cherry picked from commit 41b4c7fe72b6105a4b49395eea9aa40cef94288d)
> does look correct to me, it does break modesetting on the GPD Pocket 3,
> a i7-1195G7 laptop. I run the kernel with
> fbcon=rotate:1 video=DSI-1:panel_orientation=right_side_up \
> mem_sleep_default=s2idle
> No special i915 parameters.
> Hardware is described here:
> https://wiki.archlinux.org/title/GPD_Pocket_3
> 
> I disected this patch which was merged (backported) for 6.1.23.
> I currently run 6.1.26 with it reverted.
> 
> Without reverting it, when fbcon is switched to show the splash
> screen (GPD logo with Ubuntu added in on working kernels), the
> screen remains black (backlight on, but nothing displayed) and
> nothing happens any more.

Please file a bug at
https://gitlab.freedesktop.org/drm/intel/issues/new

Boot both kernels (revert vs. no revert), passing
'drm.debug=0xe log_buf_len=4M' to the kernel cmdline,
and attach the resulting dmesg from each to the bug.

Also would be good if you try to reproduce on the
latest drm-tip (git://anongit.freedesktop.org/drm-tip drm-tip)
as well.

-- 
Ville Syrjälä
Intel

Re: 6.1.23: 0fc6fea41c71 breaks GPD Pocket 3 modeset

[Kurt Garloff]

Hi Ville,

thanks for your response!

On 02/05/2023 15:13, Ville Syrjälä wrote:
> On Sat, Apr 29, 2023 at 09:55:45PM +0200, Kurt Garloff wrote:
>> Hi Ville,
>>
>> While
>> 0fc6fea41c71 drm/i915: Disable DC states for all commits
>> (cherry picked from commit 41b4c7fe72b6105a4b49395eea9aa40cef94288d)
>> does look correct to me, it does break modesetting on the GPD Pocket 3,
>> a i7-1195G7 laptop. I run the kernel with
>> fbcon=rotate:1 video=DSI-1:panel_orientation=right_side_up \
>> mem_sleep_default=s2idle
Note that the mem_sleep_default parm has been dropped meanwhile.
It used to be required as the laptop would show a black screen
after resuming from S3 until 6.1.5 or so.
>> No special i915 parameters.
>> Hardware is described here:
>> https://wiki.archlinux.org/title/GPD_Pocket_3
>>
>> I disected this patch which was merged (backported) for 6.1.23.
>> I currently run 6.1.26 with it reverted.
>>
>> Without reverting it, when fbcon is switched to show the splash
>> screen (GPD logo with Ubuntu added in on working kernels), the
>> screen remains black (backlight on, but nothing displayed) and
>> nothing happens any more.
> Please file a bug at
> https://gitlab.freedesktop.org/drm/intel/issues/new
Find a bug report at

https://gitlab.freedesktop.org/drm/intel/-/issues/8419

> Boot both kernels (revert vs. no revert), passing
> 'drm.debug=0xe log_buf_len=4M' to the kernel cmdline,
> and attach the resulting dmesg from each to the bug.

Getting a boot log out of the crashing laptop required
me to find a cable and get netconsole working.
(If you know better ways, I'm open to learn.)
Thus the delay ...

> Also would be good if you try to reproduce on the
> latest drm-tip (git://anongit.freedesktop.org/drm-tip drm-tip)
> as well.

I'll test drm-tip as requested and see what happens.
I'll attach the info to the bug report, let nme know
if you want eMail in addition.

Best,

-- 

Kurt Garloff <kurt@garloff.de>
Cologne, Germany