Re: [PATCH] ACPI: video: Invoke _PS0 at boot for ACPI video

[Kai-Heng Feng]

On Wed, Jul 5, 2023 at 12:59 AM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
>
> On Tue, Jul 4, 2023 at 9:46 AM Kai-Heng Feng
> <kai.heng.feng@xxxxxxxxxxxxx> wrote:
> >
> > Screen brightness can only be changed once on some HP laptops.
> >
> > Vendor identified the root cause as Linux doesn't invoke _PS0 at boot
> > for all ACPI devices:
>
> This part of the changelog is confusing, because the evaluation of
> _PS0 is not a separate operation.  _PS0 gets evaluated when devices
> undergo transitions from low-power states to D0.

But not at boot.

>
> >     Scope (\_SB.PC00.GFX0)
> >     {
> >         Scope (DD1F)
> >         {
> >             Method (_PS0, 0, Serialized)  // _PS0: Power State 0
> >             {
> >                 If (CondRefOf (\_SB.PC00.LPCB.EC0.SSBC))
> >                 {
> >                     \_SB.PC00.LPCB.EC0.SSBC ()
> >                 }
> >             }
> >             ...
> >         }
> >         ...
> >     }
> >
> > _PS0 doesn't get invoked for all ACPI devices because of commit
> > 7cd8407d53ef ("ACPI / PM: Do not execute _PS0 for devices without _PSC
> > during initialization").
>
> And yes, Linux doesn't put all of the ACPI devices into D0 during
> initialization, but the above commit has a little to do with that.

Devices without _PSC now doesn't have _PS0 evaluated at boot time. I
don't quite understand why it's not related to this commit?

>
> > For now explicitly call _PS0 for ACPI video to workaround the issue.
>
> This is not what the patch is doing.

To be specific, it's for the child device nodes under ACPI GFX.

Kai-Heng

>
> > Signed-off-by: Kai-Heng Feng <kai.heng.feng@xxxxxxxxxxxxx>
> > ---
> >  drivers/acpi/acpi_video.c | 2 ++
> >  1 file changed, 2 insertions(+)
> >
> > diff --git a/drivers/acpi/acpi_video.c b/drivers/acpi/acpi_video.c
> > index 62f4364e4460..793259bd18c8 100644
> > --- a/drivers/acpi/acpi_video.c
> > +++ b/drivers/acpi/acpi_video.c
> > @@ -2027,6 +2027,8 @@ static int acpi_video_bus_add(struct acpi_device *device)
> >         if (error)
> >                 goto err_put_video;
> >
> > +       acpi_device_fix_up_power_extended(device);
> > +
>
> I would like to know what Hans thinks about this.
>
> >         pr_info("%s [%s] (multi-head: %s  rom: %s  post: %s)\n",
> >                ACPI_VIDEO_DEVICE_NAME, acpi_device_bid(device),
> >                video->flags.multihead ? "yes" : "no",
> > --