Re: [PATCH] ACPI / LPSS: Do not instiate a platform_dev for devs without a mmio resource

[Hans de Goede]

Hi,

On 15-01-18 00:36, Rafael J. Wysocki wrote:
> On Sun, Jan 14, 2018 at 9:01 PM, Hans de Goede <hdegoede@xxxxxxxxxx> wrote:
> > acpi_lpss_create_device() skips handling LPSS devices which do not have
> > a mmio resources in their resource list (typically these devices are
> > disabled by the firmware). But since the LPSS code does not bind to the
> > device, acpi_bus_attach() ends up still creating a platform device for
> > it and the regular platform_driver for the ACPI HID still tries to bind
> > to it.
> >
> > This happens e.g. on some boards which do not use the pwm-controller
> > and have an empty or invalid resource-table for it. Currently this causes
> > these error messages to get logged:
> > [    3.281966] pwm-lpss 80862288:00: invalid resource
> > [    3.287098] pwm-lpss: probe of 80862288:00 failed with error -22
> >
> > This commit stops the undesirable creation of a platform_device for
> > disabled LPSS devices by setting pnp.type.platform_id to 0. Note that
> > acpi_scan_attach_handler() also sets pnp.type.platform_id to 0 when there
> > is a matching handler for the device and that handler has no attach
> > callback, so we simply behave as a handler without an attach function
> > in this case.
> >
> > Signed-off-by: Hans de Goede <hdegoede@xxxxxxxxxx>
>
> Does this fix new behavior or is it an old issue?

The problem this addresses is likely caused by the acpi_always_present
entry for the CHT 80862288 PWM devices in drivers/acpi/x86/utils.c.

The problem is that under Windows the Intel GPU driver has hardcoded
addresses for the PWM LPSS bits (or so I believe), rather then having
it as a separate device with a separate driver, so the CHT LPSS PWM
device's _STA returns 0 on all x86 devices which ship with Windows.

On most CHT laptops / tablets it is used to control the backlight
brightness so we do need the device under Linux, where we've a
separate PWM driver and we don't want to be hardcoding stuff like
this.

So we really cannot do without the acpi_always_present entry, and thus
need this extra check for devices where the pwm is actually unused
and disabled by the firmware.

Regards,

Hans

> > ---
> >   drivers/acpi/acpi_lpss.c | 2 ++
> >   1 file changed, 2 insertions(+)
> >
> > diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c
> > index 9cfe6b71078b..166a8e582d96 100644
> > --- a/drivers/acpi/acpi_lpss.c
> > +++ b/drivers/acpi/acpi_lpss.c
> > @@ -610,6 +610,8 @@ static int acpi_lpss_create_device(struct acpi_device *adev,
> >          acpi_dev_free_resource_list(&resource_list);
> >
> >          if (!pdata->mmio_base) {
> > +               /* Avoid acpi_bus_attach() instantiating a pdev for this dev. */
> > +               adev->pnp.type.platform_id = 0;
> >                  /* Skip the device, but continue the namespace scan. */
> >                  ret = 0;
> >                  goto err_out;
> > --
> > 2.14.3
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html