[PATCH] thermal: hisilicon: Don't enable by default

[PATCH] thermal: hisilicon: Don't enable by default

[Lubomir Rintel]

Users are generally unlikely to have a HiSilicon thermal sensor.
Like most other thermal drivers, don't build it by default/

Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
---
 drivers/thermal/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index 02c3aa322a4a6..2062f8ec272b6 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -233,7 +233,6 @@ config HISI_THERMAL
 	depends on ARCH_HISI || COMPILE_TEST
 	depends on HAS_IOMEM
 	depends on OF
-	default y
 	help
 	  Enable this to plug hisilicon's thermal sensor driver into the Linux
 	  thermal framework. cpufreq is used as the cooling device to throttle
-- 
2.25.1

Re: [PATCH] thermal: hisilicon: Don't enable by default

[Daniel Lezcano]

On 17/03/2020 18:41, Lubomir Rintel wrote:
> Users are generally unlikely to have a HiSilicon thermal sensor.

Why ?

The thermal sensor is needed for the thermal mitigation.

> Like most other thermal drivers, don't build it by default/
>
> Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
> ---
>  drivers/thermal/Kconfig | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
> index 02c3aa322a4a6..2062f8ec272b6 100644
> --- a/drivers/thermal/Kconfig
> +++ b/drivers/thermal/Kconfig
> @@ -233,7 +233,6 @@ config HISI_THERMAL
>  	depends on ARCH_HISI || COMPILE_TEST
>  	depends on HAS_IOMEM
>  	depends on OF
> -	default y
>  	help
>  	  Enable this to plug hisilicon's thermal sensor driver into the Linux
>  	  thermal framework. cpufreq is used as the cooling device to throttle
> 


-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog

Re: [PATCH] thermal: hisilicon: Don't enable by default

[Lubomir Rintel]

On Tue, Mar 17, 2020 at 06:45:29PM +0100, Daniel Lezcano wrote:
> On 17/03/2020 18:41, Lubomir Rintel wrote:
> > Users are generally unlikely to have a HiSilicon thermal sensor.
> 
> Why ?

Because most computers don't run on HiSilicon chips.

> The thermal sensor is needed for the thermal mitigation.

If it's really needed, shouldn't ARCH_HISI depend on it?

In any case, I submitted this, because the driver enabled itself when I
turned on COMPILE_TEST, which was entirely unexpected. Unless I'm
mistaken, defaulting to off seems to be a standard practice for most
drivers, including thermal.

Would this be a better idea?

  default y if ARCH_HISI

Thanks
Lubo

> 
> > Like most other thermal drivers, don't build it by default/
> >
> > Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
> > ---
> >  drivers/thermal/Kconfig | 1 -
> >  1 file changed, 1 deletion(-)
> > 
> > diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
> > index 02c3aa322a4a6..2062f8ec272b6 100644
> > --- a/drivers/thermal/Kconfig
> > +++ b/drivers/thermal/Kconfig
> > @@ -233,7 +233,6 @@ config HISI_THERMAL
> >  	depends on ARCH_HISI || COMPILE_TEST
> >  	depends on HAS_IOMEM
> >  	depends on OF
> > -	default y
> >  	help
> >  	  Enable this to plug hisilicon's thermal sensor driver into the Linux
> >  	  thermal framework. cpufreq is used as the cooling device to throttle
> > 
> 
> 
> -- 
>  <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
> 
> Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
> <http://twitter.com/#!/linaroorg> Twitter |
> <http://www.linaro.org/linaro-blog/> Blog
> 

Re: [PATCH] thermal: hisilicon: Don't enable by default

[Daniel Lezcano]

On 17/03/2020 20:27, Lubomir Rintel wrote:
> On Tue, Mar 17, 2020 at 06:45:29PM +0100, Daniel Lezcano wrote:
>> On 17/03/2020 18:41, Lubomir Rintel wrote:
>>> Users are generally unlikely to have a HiSilicon thermal sensor.
>>
>> Why ?
> 
> Because most computers don't run on HiSilicon chips.
> 
>> The thermal sensor is needed for the thermal mitigation.
> 
> If it's really needed, shouldn't ARCH_HISI depend on it?
> 
> In any case, I submitted this, because the driver enabled itself when I
> turned on COMPILE_TEST, which was entirely unexpected.

The COMPILE_TEST option's contextual help says:

"If you are a developer and want to build everything available, say Y
       here."

> Unless I'm
> mistaken, defaulting to off seems to be a standard practice for most
> drivers, including thermal.

The COMPILE_TEST option is there in order to increase the compilation
test coverage (for example compile the drivers on a powerful x86 machine
with different option combinations).

> Would this be a better idea?
> 
>   default y if ARCH_HISI
> 
> Thanks
> Lubo
> 
>>
>>> Like most other thermal drivers, don't build it by default/
>>>
>>> Signed-off-by: Lubomir Rintel <lkundrak@v3.sk>
>>> ---
>>>  drivers/thermal/Kconfig | 1 -
>>>  1 file changed, 1 deletion(-)
>>>
>>> diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
>>> index 02c3aa322a4a6..2062f8ec272b6 100644
>>> --- a/drivers/thermal/Kconfig
>>> +++ b/drivers/thermal/Kconfig
>>> @@ -233,7 +233,6 @@ config HISI_THERMAL
>>>  	depends on ARCH_HISI || COMPILE_TEST
>>>  	depends on HAS_IOMEM
>>>  	depends on OF
>>> -	default y
>>>  	help
>>>  	  Enable this to plug hisilicon's thermal sensor driver into the Linux
>>>  	  thermal framework. cpufreq is used as the cooling device to throttle
>>>
>>
>>
>> -- 
>>  <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
>>
>> Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
>> <http://twitter.com/#!/linaroorg> Twitter |
>> <http://www.linaro.org/linaro-blog/> Blog
>>


-- 
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog