Re: [PATCH 0/3] Add new SFH interfaces

[Hans de Goede <hdegoede@redhat.com>]

Hi all,

On 1/4/24 09:29, Shyam Sundar S K wrote:
> Hi Jiri,
> 
> On 1/2/2024 3:53 PM, Jiri Kosina wrote:
>> On Wed, 20 Dec 2023, Basavaraj Natikar wrote:
>>
>>> This series adds new interfaces to export User presence information and
>>> Ambient light to other drivers within the kernel.
>>
>> Hi,
>>
>> thanks for the patches. I'd like this to go in together with the actual 
>> users of it on the PMF side. Does that code already exist?
> 
> I am yet to submit the patches that uses the new SFH interfaces
> defined in this series.
> 
> The suggestion from Hans is to have the SFH changes landed
> independently, so that Hans can take the PMF changes alone later
> during the rc's as fix. IMO, we can have this series merged first alone.
> 
> But I am OK to have both PMF and SFH changes together.
> 
> Hans, what is your feedback for Jiri's question?

Jiri, originally the drivers/hid/amd-sfh-hid/ changes
were posted as part of this series:

https://lore.kernel.org/platform-driver-x86/20231204101548.1458499-1-Shyam-sundar.S-k@amd.com/

specifically in these patches:

https://lore.kernel.org/platform-driver-x86/20231204101548.1458499-14-Shyam-sundar.S-k@amd.com/
https://lore.kernel.org/platform-driver-x86/20231204101548.1458499-15-Shyam-sundar.S-k@amd.com/
https://lore.kernel.org/platform-driver-x86/20231204101548.1458499-16-Shyam-sundar.S-k@amd.com/

Where the last 2 patches introduce both the HID changes and the drivers/platform/x86/amd/pmf/...
consumer of the HID changes in one go.

I have asked Shyam to split out the HID changes:
https://lore.kernel.org/platform-driver-x86/ad064333-48a4-4cfa-9428-69e8a7c44667@redhat.com/

"""
But this patch also needs to have its HID parts split out
from the rest and the HID patches need to be merged
separately.

Since the merge-window is getting close I would like
to propose the following:

1. Send a v7 addressing review remarks consisting of
only patches 1-12, assuming all review remarks are
fixed in v7 I can merge that then.

2. Send a 3 patch HID patch series separately:

[PATCH 1/3] HID: amd_sfh: rename float_to_int() to amd_sfh_float_to_int()
[PATCH 2/3] HID: amd_sfh: adding the HID hpd bits + amd_get_sfh_info() function
[PATCH 3/3] HID: amd_sfh: adding the HID als bits

3. Once v7 of patches 1-12 + hopefully the 3 patch HID series have
landed in 6.8-rc1 then I would be willing to take a patch using
amd_get_sfh_info() to implement hpd + als support in the spc code
as a fix for 6.8-rc# .
"""

I have merged patches 1-12 of the AMD PMF series already.
So assuming there are no other objections against the HID
changes it would be great if these can still make it into
6.8-rc1 then Shyam can re-send the AMD PMF changes which
consume the new HID API as a separate patch and then I'll
send that to Linus once the HID changes have been merged
by Linus.

Note I understand if the timing does not work out
in that case the HID changes + the AMD PMF patches which
consume the new HID API can wait for the next cycle.

If this is going to wait for the next cycle, then please
send me a pull-request for an immutable branch with
the HID changes in there so that I can merge that
into pdx86/for-next before merging the AMD PMF changes.

Regards,

Hans