Currently runtime PM is enabled for Tegra HDA driver. One of the concern
raised was about handling !PM case in HDA driver. In this case, there is
a need to manage clocks and power explicitly in the driver and reviewers
were not really convinced about this. The consensus at the end was to
enforce PM requirement to keep things simple, rather driver mentioning a
dependency on PM. This is how it is done for ARM 64-bit Tegra platforms
and the same can be done for 32-bit Tegra platforms too.
Finally the objective is to remove dependency on PM availability for all
Tegra drivers. The detailed discussion can be found in following patch,
https://patchwork.ozlabs.org/patch/1031007/
Signed-off-by: Sameer Pujar <[email protected]>
Reviewed-by: Thierry Reding <[email protected]>
Reviewed-by: Jonathan Hunter <[email protected]>
---
arch/arm/mach-tegra/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm/mach-tegra/Kconfig b/arch/arm/mach-tegra/Kconfig
index 7f3b83e..51a8fa3 100644
--- a/arch/arm/mach-tegra/Kconfig
+++ b/arch/arm/mach-tegra/Kconfig
@@ -10,6 +10,7 @@ menuconfig ARCH_TEGRA
select HAVE_ARM_SCU if SMP
select HAVE_ARM_TWD if SMP
select PINCTRL
+ select PM
select PM_OPP
select ARCH_HAS_RESET_CONTROLLER
select RESET_CONTROLLER
--
2.7.4
27.02.2019 11:20, Sameer Pujar пишет:
> Currently runtime PM is enabled for Tegra HDA driver. One of the concern
> raised was about handling !PM case in HDA driver. In this case, there is
> a need to manage clocks and power explicitly in the driver and reviewers
> were not really convinced about this. The consensus at the end was to
> enforce PM requirement to keep things simple, rather driver mentioning a
> dependency on PM. This is how it is done for ARM 64-bit Tegra platforms
> and the same can be done for 32-bit Tegra platforms too.
>
> Finally the objective is to remove dependency on PM availability for all
> Tegra drivers. The detailed discussion can be found in following patch,
> https://patchwork.ozlabs.org/patch/1031007/
>
> Signed-off-by: Sameer Pujar <[email protected]>
> Reviewed-by: Thierry Reding <[email protected]>
> Reviewed-by: Jonathan Hunter <[email protected]>
> ---
The commit message is okay to me this time, thanks! I'm still not super-happy to lose the !PM option because of its value in regards to the PM debugging, but again we can just do something better in that regards later on if will be really needed.
Reviewed-by: Dmitry Osipenko <[email protected]>
On 2/27/2019 7:01 PM, Dmitry Osipenko wrote:
> 27.02.2019 11:20, Sameer Pujar пишет:
>> Currently runtime PM is enabled for Tegra HDA driver. One of the concern
>> raised was about handling !PM case in HDA driver. In this case, there is
>> a need to manage clocks and power explicitly in the driver and reviewers
>> were not really convinced about this. The consensus at the end was to
>> enforce PM requirement to keep things simple, rather driver mentioning a
>> dependency on PM. This is how it is done for ARM 64-bit Tegra platforms
>> and the same can be done for 32-bit Tegra platforms too.
>>
>> Finally the objective is to remove dependency on PM availability for all
>> Tegra drivers. The detailed discussion can be found in following patch,
>> https://patchwork.ozlabs.org/patch/1031007/
>>
>> Signed-off-by: Sameer Pujar <[email protected]>
>> Reviewed-by: Thierry Reding <[email protected]>
>> Reviewed-by: Jonathan Hunter <[email protected]>
>> ---
> The commit message is okay to me this time, thanks! I'm still not super-happy to lose the !PM option because of its value in regards to the PM debugging, but again we can just do something better in that regards later on if will be really needed.
>
> Reviewed-by: Dmitry Osipenko <[email protected]>
Do we have any further concerns over the patch? Request maintainers for
merge otherwise.
On Wed, Feb 27, 2019 at 01:50:28PM +0530, Sameer Pujar wrote:
> Currently runtime PM is enabled for Tegra HDA driver. One of the concern
> raised was about handling !PM case in HDA driver. In this case, there is
> a need to manage clocks and power explicitly in the driver and reviewers
> were not really convinced about this. The consensus at the end was to
> enforce PM requirement to keep things simple, rather driver mentioning a
> dependency on PM. This is how it is done for ARM 64-bit Tegra platforms
> and the same can be done for 32-bit Tegra platforms too.
>
> Finally the objective is to remove dependency on PM availability for all
> Tegra drivers. The detailed discussion can be found in following patch,
> https://patchwork.ozlabs.org/patch/1031007/
>
> Signed-off-by: Sameer Pujar <[email protected]>
> Reviewed-by: Thierry Reding <[email protected]>
> Reviewed-by: Jonathan Hunter <[email protected]>
> ---
> arch/arm/mach-tegra/Kconfig | 1 +
> 1 file changed, 1 insertion(+)
Applied to for-5.2/arm/soc, thanks.
Thierry