ACPM clients (PMIC, clocks, etc.) will be modeled as children of the
ACPM interface.

This necessitates two changes in this driver:
* child platform_devices from device tree need to be populated here
* child drivers will need to be able to get the ACPM handle via a
  struct device_node, not via a DT property name

This series implements these changes and is a prerequisite for v2 of
https://lore.kernel.org/all/20250323-s2mpg10-v1-0-d08943702707@linaro.org/

--- dependency note ---

To apply cleanly, patch 2 must be applied after
https://lore.kernel.org/all/20250319-acpm-fixes-v2-2-ac2c1bcf322b@linaro.org/
due to touching similar areas of the code.

I can merge that and this series into one if that is preferred instead,
please let me know if so.

--- dependency note end ---

Cheers,
Andre'

Signed-off-by: André Draszik <andre.draszik@linaro.org>
---
André Draszik (1):
      firmware: exynos-acpm: introduce devm_acpm_get_by_node()

Tudor Ambarus (1):
      firmware: exynos-acpm: populate devices from device tree data

 drivers/firmware/samsung/exynos-acpm.c             | 25 +++++++++-------------
 .../linux/firmware/samsung/exynos-acpm-protocol.h  |  6 ++++--
 2 files changed, 14 insertions(+), 17 deletions(-)
---
base-commit: cc024fd4c6640be5e2e0989cc98f2c68e31289d5
change-id: 20250327-acpm-children-70ceb0138d63

Best regards,
-- 
André Draszik <andre.draszik@linaro.org>

On Thu, 27 Mar 2025 12:54:26 +0000, André Draszik wrote:
> ACPM clients (PMIC, clocks, etc.) will be modeled as children of the
> ACPM interface.
> 
> This necessitates two changes in this driver:
> * child platform_devices from device tree need to be populated here
> * child drivers will need to be able to get the ACPM handle via a
>   struct device_node, not via a DT property name
> 
> [...]

Applied, thanks!

[1/2] firmware: exynos-acpm: populate devices from device tree data
      https://git.kernel.org/krzk/linux/c/636baba9489a634c956d6f02076af6bc1725c132
[2/2] firmware: exynos-acpm: introduce devm_acpm_get_by_node()
      https://git.kernel.org/krzk/linux/c/a8dc26a0ec43fd416ca781a8030807e65f71cfc5

Best regards,
-- 
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

On Tue, 22 Apr 2025, Krzysztof Kozlowski wrote:

> 
> On Thu, 27 Mar 2025 12:54:26 +0000, André Draszik wrote:
> > ACPM clients (PMIC, clocks, etc.) will be modeled as children of the
> > ACPM interface.
> > 
> > This necessitates two changes in this driver:
> > * child platform_devices from device tree need to be populated here
> > * child drivers will need to be able to get the ACPM handle via a
> >   struct device_node, not via a DT property name
> > 
> > [...]
> 
> Applied, thanks!
> 
> [1/2] firmware: exynos-acpm: populate devices from device tree data
>       https://git.kernel.org/krzk/linux/c/636baba9489a634c956d6f02076af6bc1725c132
> [2/2] firmware: exynos-acpm: introduce devm_acpm_get_by_node()
>       https://git.kernel.org/krzk/linux/c/a8dc26a0ec43fd416ca781a8030807e65f71cfc5

Can this be made available for other subsystems to pull from please?

Without it MFD is currently broken.

Ref: https://lore.kernel.org/all/fd11ee67a3d51aef6305c477b140d911da0b2af4.camel@linaro.org/

-- 
Lee Jones [李琼斯]

On 01/05/2025 11:09, Lee Jones wrote:
> On Tue, 22 Apr 2025, Krzysztof Kozlowski wrote:
> 
>>
>> On Thu, 27 Mar 2025 12:54:26 +0000, André Draszik wrote:
>>> ACPM clients (PMIC, clocks, etc.) will be modeled as children of the
>>> ACPM interface.
>>>
>>> This necessitates two changes in this driver:
>>> * child platform_devices from device tree need to be populated here
>>> * child drivers will need to be able to get the ACPM handle via a
>>>   struct device_node, not via a DT property name
>>>
>>> [...]
>>
>> Applied, thanks!
>>
>> [1/2] firmware: exynos-acpm: populate devices from device tree data
>>       https://git.kernel.org/krzk/linux/c/636baba9489a634c956d6f02076af6bc1725c132
>> [2/2] firmware: exynos-acpm: introduce devm_acpm_get_by_node()
>>       https://git.kernel.org/krzk/linux/c/a8dc26a0ec43fd416ca781a8030807e65f71cfc5
> 
> Can this be made available for other subsystems to pull from please?
> 
> Without it MFD is currently broken.
> 
> Ref: https://lore.kernel.org/all/fd11ee67a3d51aef6305c477b140d911da0b2af4.camel@linaro.org/
> 

Yeah, I will prepare you a stable tag.

While this wasn't clearly expressed when sending the patch? There is
some "prerequisite" note in cover letter but what does that mean? What
does the contributor expect from maintainers?


Best regards,
Krzysztof