mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH v3] cpufreq: CPPC: Enable boost by default
@ 2026-11-17 10:52 Pengjie Zhang
  2026-11-17 11:06 ` Christian Loehle
  0 siblings, 1 reply; 3+ messages in thread
From: Pengjie Zhang @ 2026-11-17 10:52 UTC (permalink / raw)
  To: rafael, viresh.kumar
  Cc: linuxarm, zhanjie9, prime.zeng, wanghuiqiang, xuwei5,
	zhenglifeng1, christian.loehle, yubowen8, pierre.gondois, sumitg,
	linux-pm, linux-kernel, zhangpengjie2, mojunjun4

The CPPC driver leaves boost disabled by default when Highest Performance
exceeds Nominal Performance. Userspace must explicitly enable boost
before governors can use the full advertised performance range.

For user convenience, set .boost_enabled = true in cppc_cpufreq_driver.
Since cppc_cpufreq_cpu_init() sets policy->boost_supported when
highest_perf > nominal_perf, the cpufreq core will enable boost by
default only for policies that support it.

Signed-off-by: Pengjie Zhang <zhangpengjie2@huawei.com>
---
Changes in v3:
- Drop cppc_cpufreq_default_boost_init() and set .boost_enabled = true
  statically in cppc_cpufreq_driver (Christian).
- Link to v2: https://lore.kernel.org/all/20261022075901.3436021-1-zhangpengjie2@huawei.com/

Changes in v2:
- Drop "even" from the commit message (Christian).
- Link to v1: https://lore.kernel.org/all/20261015015812.867988-1-zhangpengjie2@huawei.com/
---
 drivers/cpufreq/cppc_cpufreq.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
index 957fa7775260..1ea16e95c0e9 100644
--- a/drivers/cpufreq/cppc_cpufreq.c
+++ b/drivers/cpufreq/cppc_cpufreq.c
@@ -1059,6 +1059,7 @@ static struct cpufreq_driver cppc_cpufreq_driver = {
 	.fast_switch = cppc_cpufreq_fast_switch,
 	.init = cppc_cpufreq_cpu_init,
 	.exit = cppc_cpufreq_cpu_exit,
+	.boost_enabled = true,
 	.set_boost = cppc_cpufreq_set_boost,
 	.attr = cppc_cpufreq_attr,
 	.name = "cppc_cpufreq",
-- 
2.33.0


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v3] cpufreq: CPPC: Enable boost by default
  2026-11-17 10:52 [PATCH v3] cpufreq: CPPC: Enable boost by default Pengjie Zhang
@ 2026-11-17 11:06 ` Christian Loehle
  2026-11-17 12:21   ` Pengjie Zhang
  0 siblings, 1 reply; 3+ messages in thread
From: Christian Loehle @ 2026-11-17 11:06 UTC (permalink / raw)
  To: Pengjie Zhang, rafael, viresh.kumar
  Cc: linuxarm, zhanjie9, prime.zeng, wanghuiqiang, xuwei5,
	zhenglifeng1, yubowen8, pierre.gondois, sumitg, linux-pm,
	linux-kernel, mojunjun4

On 11/17/26 10:52, Pengjie Zhang wrote:
> The CPPC driver leaves boost disabled by default when Highest Performance
> exceeds Nominal Performance. Userspace must explicitly enable boost
> before governors can use the full advertised performance range.
> 
> For user convenience, set .boost_enabled = true in cppc_cpufreq_driver.
> Since cppc_cpufreq_cpu_init() sets policy->boost_supported when
> highest_perf > nominal_perf, the cpufreq core will enable boost by
> default only for policies that support it.
> 
> Signed-off-by: Pengjie Zhang <zhangpengjie2@huawei.com>

Reviewed-by: Christian Loehle <christian.loehle@arm.com>

That being said it doesn't apply cleanly on bleeding-edge, so you might
wanna rebase to make Rafael's life a little easier.

> ---
> Changes in v3:
> - Drop cppc_cpufreq_default_boost_init() and set .boost_enabled = true
>   statically in cppc_cpufreq_driver (Christian).
> - Link to v2: https://lore.kernel.org/all/20261022075901.3436021-1-zhangpengjie2@huawei.com/
> 
> Changes in v2:
> - Drop "even" from the commit message (Christian).
> - Link to v1: https://lore.kernel.org/all/20261015015812.867988-1-zhangpengjie2@huawei.com/
> ---
>  drivers/cpufreq/cppc_cpufreq.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
> index 957fa7775260..1ea16e95c0e9 100644
> --- a/drivers/cpufreq/cppc_cpufreq.c
> +++ b/drivers/cpufreq/cppc_cpufreq.c
> @@ -1059,6 +1059,7 @@ static struct cpufreq_driver cppc_cpufreq_driver = {
>  	.fast_switch = cppc_cpufreq_fast_switch,
>  	.init = cppc_cpufreq_cpu_init,
>  	.exit = cppc_cpufreq_cpu_exit,
> +	.boost_enabled = true,
>  	.set_boost = cppc_cpufreq_set_boost,
>  	.attr = cppc_cpufreq_attr,
>  	.name = "cppc_cpufreq",


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v3] cpufreq: CPPC: Enable boost by default
  2026-11-17 11:06 ` Christian Loehle
@ 2026-11-17 12:21   ` Pengjie Zhang
  0 siblings, 0 replies; 3+ messages in thread
From: Pengjie Zhang @ 2026-11-17 12:21 UTC (permalink / raw)
  To: Christian Loehle, rafael, viresh.kumar
  Cc: linuxarm, zhanjie9, prime.zeng, wanghuiqiang, xuwei5,
	zhenglifeng1, yubowen8, pierre.gondois, sumitg, linux-pm,
	linux-kernel, mojunjun4


On 11/17/2026 7:06 PM, Christian Loehle wrote:
> On 11/17/26 10:52, Pengjie Zhang wrote:
>> The CPPC driver leaves boost disabled by default when Highest Performance
>> exceeds Nominal Performance. Userspace must explicitly enable boost
>> before governors can use the full advertised performance range.
>>
>> For user convenience, set .boost_enabled = true in cppc_cpufreq_driver.
>> Since cppc_cpufreq_cpu_init() sets policy->boost_supported when
>> highest_perf > nominal_perf, the cpufreq core will enable boost by
>> default only for policies that support it.
>>
>> Signed-off-by: Pengjie Zhang <zhangpengjie2@huawei.com>
> Reviewed-by: Christian Loehle <christian.loehle@arm.com>
>
> That being said it doesn't apply cleanly on bleeding-edge, so you might
> wanna rebase to make Rafael's life a little easier.
Yes, cppc_cpufreq seems to have merged a lot of content into linux-next.
>> ---
>> Changes in v3:
>> - Drop cppc_cpufreq_default_boost_init() and set .boost_enabled = true
>>    statically in cppc_cpufreq_driver (Christian).
>> - Link to v2: https://lore.kernel.org/all/20261022075901.3436021-1-zhangpengjie2@huawei.com/
>>
>> Changes in v2:
>> - Drop "even" from the commit message (Christian).
>> - Link to v1: https://lore.kernel.org/all/20261015015812.867988-1-zhangpengjie2@huawei.com/
>> ---
>>   drivers/cpufreq/cppc_cpufreq.c | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
>> index 957fa7775260..1ea16e95c0e9 100644
>> --- a/drivers/cpufreq/cppc_cpufreq.c
>> +++ b/drivers/cpufreq/cppc_cpufreq.c
>> @@ -1059,6 +1059,7 @@ static struct cpufreq_driver cppc_cpufreq_driver = {
>>   	.fast_switch = cppc_cpufreq_fast_switch,
>>   	.init = cppc_cpufreq_cpu_init,
>>   	.exit = cppc_cpufreq_cpu_exit,
>> +	.boost_enabled = true,
>>   	.set_boost = cppc_cpufreq_set_boost,
>>   	.attr = cppc_cpufreq_attr,
>>   	.name = "cppc_cpufreq",
>
>

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2026-11-17 12:21 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-11-17 10:52 [PATCH v3] cpufreq: CPPC: Enable boost by default Pengjie Zhang
2026-11-17 11:06 ` Christian Loehle
2026-11-17 12:21   ` Pengjie Zhang

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®