RFI: Tablet mode support on HP Pro x360 435 G9 w/ AMD Ryzen 7 5825U

RFI: Tablet mode support on HP Pro x360 435 G9 w/ AMD Ryzen 7 5825U

[Carsten Hatger]

Dear all,

I'd like to have tablet mode support on my system, probably by means
an linux input device such as implemented in the intel platform
specific driver drivers/platform/x86/intel/vbtn.c [0]

In the end I hope GNOME eventually to rotate the systems display and
to show some virtual keyboard (upon users request), cf. for [3]

It appears there has already been a patch proposed  by the chromium
team to support device PNP0C60 [1] but not merged to [5].

Since the system of interest is a HP Probook, there is already a
driver providing virtual buttons,namely hp-wmi [6]. However, the
driver loads probes and loads successfully but doesn't provide any
additional functionality plus some non critical errors on incorrect
ACPI method calls.

I've noticed AMD has started to provide platform specific driver(s)
such as pmf [2]. To my knowledge there is no support for CEZANNE/green
sardine based systems (yet).

What would be recommended practice and subsystem/folder to provide
such capability by means of a (platform specific) driver? At least the
CID PNP0C60 seems to be held by Microsoft [4] and thus be common to
both amd and intel platforms [4]. However, HID INT33D6 is held by
Intel and HID AMDI0081 by AMD. Yet I'm not quite sure if
iio-sensor-proxy [7] needs to be involved, too.

Best regards,
Carsten Hatger

[0] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/intel/vbtn.c?h=v6.2-rc8
[1] https://lore.kernel.org/lkml/1472628817-3145-1-git-send-email-wnhuang@google.com/
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/amd/pmf?h=v6.2-rc8
[3] https://gitlab.gnome.org/GNOME/mutter/-/issues/1760
[4] https://learn.microsoft.com/en-us/windows-hardware/drivers/gpiobtn/button-implementation
[5] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/acpi/button.c?h=v6.2-rc8
[6] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/hp/hp-wmi.c?h=v6.2-rc8
[7] https://gitlab.freedesktop.org/hadess/iio-sensor-proxy/

Re: RFI: Tablet mode support on HP Pro x360 435 G9 w/ AMD Ryzen 7 5825U

[Hans de Goede]

Hi Carsten,

On 2/13/23 11:43, Carsten Hatger wrote:
> Dear all,
> 
> I'd like to have tablet mode support on my system, probably by means
> an linux input device such as implemented in the intel platform
> specific driver drivers/platform/x86/intel/vbtn.c [0]
> 
> In the end I hope GNOME eventually to rotate the systems display and
> to show some virtual keyboard (upon users request), cf. for [3]
> 
> It appears there has already been a patch proposed  by the chromium
> team to support device PNP0C60 [1] but not merged to [5].
> 
> Since the system of interest is a HP Probook, there is already a
> driver providing virtual buttons,namely hp-wmi [6]. However, the
> driver loads probes and loads successfully but doesn't provide any
> additional functionality plus some non critical errors on incorrect
> ACPI method calls.
> 
> I've noticed AMD has started to provide platform specific driver(s)
> such as pmf [2]. To my knowledge there is no support for CEZANNE/green
> sardine based systems (yet).
> 
> What would be recommended practice and subsystem/folder to provide
> such capability by means of a (platform specific) driver? At least the
> CID PNP0C60 seems to be held by Microsoft [4] and thus be common to
> both amd and intel platforms [4]. However, HID INT33D6 is held by
> Intel and HID AMDI0081 by AMD. Yet I'm not quite sure if
> iio-sensor-proxy [7] needs to be involved, too.

The first thing to do here is to figure out which (ACPI) device
is the right device to get the SW_TABLET_MODE events from on this
device.

Maybe Jorge (added to the Cc) can help with this ?

Regards,

Hans




> [0] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/intel/vbtn.c?h=v6.2-rc8
> [1] https://lore.kernel.org/lkml/1472628817-3145-1-git-send-email-wnhuang@google.com/
> [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/amd/pmf?h=v6.2-rc8
> [3] https://gitlab.gnome.org/GNOME/mutter/-/issues/1760
> [4] https://learn.microsoft.com/en-us/windows-hardware/drivers/gpiobtn/button-implementation
> [5] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/acpi/button.c?h=v6.2-rc8
> [6] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/hp/hp-wmi.c?h=v6.2-rc8
> [7] https://gitlab.freedesktop.org/hadess/iio-sensor-proxy/
> 

Re: RFI: Tablet mode support on HP Pro x360 435 G9 w/ AMD Ryzen 7 5825U

[Shyam Sundar S K]

On 2/13/2023 6:01 PM, Hans de Goede wrote:
> Hi Carsten,
> 
> On 2/13/23 11:43, Carsten Hatger wrote:
>> Dear all,
>>
>> I'd like to have tablet mode support on my system, probably by means
>> an linux input device such as implemented in the intel platform
>> specific driver drivers/platform/x86/intel/vbtn.c [0]
>>
>> In the end I hope GNOME eventually to rotate the systems display and
>> to show some virtual keyboard (upon users request), cf. for [3]
>>
>> It appears there has already been a patch proposed  by the chromium
>> team to support device PNP0C60 [1] but not merged to [5].
>>
>> Since the system of interest is a HP Probook, there is already a
>> driver providing virtual buttons,namely hp-wmi [6]. However, the
>> driver loads probes and loads successfully but doesn't provide any
>> additional functionality plus some non critical errors on incorrect
>> ACPI method calls.
>>
>> I've noticed AMD has started to provide platform specific driver(s)
>> such as pmf [2]. 

PMF is meant for power and thermal management.

To my knowledge there is no support for CEZANNE/green
>> sardine based systems (yet).
>>
>> What would be recommended practice and subsystem/folder to provide
>> such capability by means of a (platform specific) driver? At least the
>> CID PNP0C60 seems to be held by Microsoft [4] and thus be common to
>> both amd and intel platforms [4]. However, HID INT33D6 is held by
>> Intel and HID AMDI0081 by AMD. Yet I'm not quite sure if

IIRC, AMDI0081 is used by SFH driver as UMDF sensor class extension
driver[1][2], but on Linux we have implemented it as HID based driver
and is a single driver.


>> iio-sensor-proxy [7] needs to be involved, too.

you mean to say, amd_sfh driver is not switching to tablet mode?

Thanks,
Shyam

[1]
https://github.com/MicrosoftDocs/windows-driver-docs/blob/staging/windows-driver-docs-pr/sensors/overview-of-converged-sensor-driver-model.md

[2]
https://learn.microsoft.com/en-us/windows-hardware/design/device-experiences/continuum

> 
> The first thing to do here is to figure out which (ACPI) device
> is the right device to get the SW_TABLET_MODE events from on this
> device.
> 
> Maybe Jorge (added to the Cc) can help with this ?
> 
> Regards,
> 
> Hans
> 
> 
> 
> 
>> [0] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/intel/vbtn.c?h=v6.2-rc8
>> [1] https://lore.kernel.org/lkml/1472628817-3145-1-git-send-email-wnhuang@google.com/
>> [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/amd/pmf?h=v6.2-rc8
>> [3] https://gitlab.gnome.org/GNOME/mutter/-/issues/1760
>> [4] https://learn.microsoft.com/en-us/windows-hardware/drivers/gpiobtn/button-implementation
>> [5] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/acpi/button.c?h=v6.2-rc8
>> [6] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/hp/hp-wmi.c?h=v6.2-rc8
>> [7] https://gitlab.freedesktop.org/hadess/iio-sensor-proxy/
>>
> 

Re: RFI: Tablet mode support on HP Pro x360 435 G9 w/ AMD Ryzen 7 5825U

[Mario Limonciello]

On 2/13/23 08:04, Shyam Sundar S K wrote:
>
> On 2/13/2023 6:01 PM, Hans de Goede wrote:
>> Hi Carsten,
>>
>> On 2/13/23 11:43, Carsten Hatger wrote:
>>> Dear all,
>>>
>>> I'd like to have tablet mode support on my system, probably by means
>>> an linux input device such as implemented in the intel platform
>>> specific driver drivers/platform/x86/intel/vbtn.c [0]
>>>
>>> In the end I hope GNOME eventually to rotate the systems display and
>>> to show some virtual keyboard (upon users request), cf. for [3]
>>>
>>> It appears there has already been a patch proposed  by the chromium
>>> team to support device PNP0C60 [1] but not merged to [5].

Does your system support a PNP0C60 device?

Or you're just mentioning this patch existed?

It might be useful to look at an acpidump.

>>> Since the system of interest is a HP Probook, there is already a
>>> driver providing virtual buttons,namely hp-wmi [6]. However, the
>>> driver loads probes and loads successfully but doesn't provide any
>>> additional functionality plus some non critical errors on incorrect
>>> ACPI method calls.
>>>
>>> I've noticed AMD has started to provide platform specific driver(s)
>>> such as pmf [2].
> PMF is meant for power and thermal management.
>
> To my knowledge there is no support for CEZANNE/green
>>> sardine based systems (yet).
>>>
>>> What would be recommended practice and subsystem/folder to provide
>>> such capability by means of a (platform specific) driver? At least the
>>> CID PNP0C60 seems to be held by Microsoft [4] and thus be common to
>>> both amd and intel platforms [4]. However, HID INT33D6 is held by
>>> Intel and HID AMDI0081 by AMD. Yet I'm not quite sure if
> IIRC, AMDI0081 is used by SFH driver as UMDF sensor class extension
> driver[1][2], but on Linux we have implemented it as HID based driver
> and is a single driver.
>
>
>>> iio-sensor-proxy [7] needs to be involved, too.
> you mean to say, amd_sfh driver is not switching to tablet mode?
>
> Thanks,
> Shyam
>
> [1]
> https://github.com/MicrosoftDocs/windows-driver-docs/blob/staging/windows-driver-docs-pr/sensors/overview-of-converged-sensor-driver-model.md
>
> [2]
> https://learn.microsoft.com/en-us/windows-hardware/design/device-experiences/continuum
>
>> The first thing to do here is to figure out which (ACPI) device
>> is the right device to get the SW_TABLET_MODE events from on this
>> device.
>>
>> Maybe Jorge (added to the Cc) can help with this ?
>>
>> Regards,
>>
>> Hans
>>
>>
>>
>>
>>> [0] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/intel/vbtn.c?h=v6.2-rc8
>>> [1] https://lore.kernel.org/lkml/1472628817-3145-1-git-send-email-wnhuang@google.com/
>>> [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/amd/pmf?h=v6.2-rc8
>>> [3] https://gitlab.gnome.org/GNOME/mutter/-/issues/1760
>>> [4] https://learn.microsoft.com/en-us/windows-hardware/drivers/gpiobtn/button-implementation
>>> [5] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/acpi/button.c?h=v6.2-rc8
>>> [6] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/hp/hp-wmi.c?h=v6.2-rc8
>>> [7] https://gitlab.freedesktop.org/hadess/iio-sensor-proxy/
>>>

Re: RFI: Tablet mode support on HP Pro x360 435 G9 w/ AMD Ryzen 7 5825U

[Jorge Lopez]

Hi Hans,

I am forwarding this question to the BIOS team to identify which ACPI
device is the right device to get the SW_TABLET_MODE events.

On Mon, Feb 13, 2023 at 6:31 AM Hans de Goede <hdegoede@redhat.com> wrote:
>
> Hi Carsten,
>
> On 2/13/23 11:43, Carsten Hatger wrote:
> > Dear all,
> >
> > I'd like to have tablet mode support on my system, probably by means
> > an linux input device such as implemented in the intel platform
> > specific driver drivers/platform/x86/intel/vbtn.c [0]
> >
> > In the end I hope GNOME eventually to rotate the systems display and
> > to show some virtual keyboard (upon users request), cf. for [3]
> >
> > It appears there has already been a patch proposed  by the chromium
> > team to support device PNP0C60 [1] but not merged to [5].
> >
> > Since the system of interest is a HP Probook, there is already a
> > driver providing virtual buttons,namely hp-wmi [6]. However, the
> > driver loads probes and loads successfully but doesn't provide any
> > additional functionality plus some non critical errors on incorrect
> > ACPI method calls.
> >
> > I've noticed AMD has started to provide platform specific driver(s)
> > such as pmf [2]. To my knowledge there is no support for CEZANNE/green
> > sardine based systems (yet).
> >
> > What would be recommended practice and subsystem/folder to provide
> > such capability by means of a (platform specific) driver? At least the
> > CID PNP0C60 seems to be held by Microsoft [4] and thus be common to
> > both amd and intel platforms [4]. However, HID INT33D6 is held by
> > Intel and HID AMDI0081 by AMD. Yet I'm not quite sure if
> > iio-sensor-proxy [7] needs to be involved, too.
>
> The first thing to do here is to figure out which (ACPI) device
> is the right device to get the SW_TABLET_MODE events from on this
> device.
>
> Maybe Jorge (added to the Cc) can help with this ?
>
> Regards,
>
> Hans
>
>
>
>
> > [0] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/intel/vbtn.c?h=v6.2-rc8
> > [1] https://lore.kernel.org/lkml/1472628817-3145-1-git-send-email-wnhuang@google.com/
> > [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/amd/pmf?h=v6.2-rc8
> > [3] https://gitlab.gnome.org/GNOME/mutter/-/issues/1760
> > [4] https://learn.microsoft.com/en-us/windows-hardware/drivers/gpiobtn/button-implementation
> > [5] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/acpi/button.c?h=v6.2-rc8
> > [6] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/platform/x86/hp/hp-wmi.c?h=v6.2-rc8
> > [7] https://gitlab.freedesktop.org/hadess/iio-sensor-proxy/
> >
>