cpufreq: Fix initialization with disabled boost

[PATCH] cpufreq: Fix initialization with disabled boost

Posted by Christian Loehle 3 months, 3 weeks ago

The boost_enabled early return in policy_set_boost() caused
the boost disabled at initialization to not actually set the
initial policy->max, therefore effectively enabling boost while
it should have been enabled.

Fixes: 27241c8b63bd ("cpufreq: Introduce policy_set_boost()")
Reported-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
 drivers/cpufreq/cpufreq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index d7426e1d8bdd..e85139bd0436 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1630,7 +1630,7 @@ static int cpufreq_online(unsigned int cpu)
 	 */
 	if (cpufreq_driver->set_boost && policy->boost_supported &&
 	    (new_policy || !cpufreq_boost_enabled())) {
-		ret = policy_set_boost(policy, cpufreq_boost_enabled());
+		ret = cpufreq_driver->set_boost(policy, cpufreq_boost_enabled());
 		if (ret) {
 			/* If the set_boost fails, the online operation is not affected */
 			pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu,
-- 
2.34.1

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by Rafael J. Wysocki 3 months, 3 weeks ago

On Mon, Jun 16, 2025 at 7:25 PM Christian Loehle
<christian.loehle@arm.com> wrote:
>
> The boost_enabled early return in policy_set_boost() caused
> the boost disabled at initialization to not actually set the
> initial policy->max, therefore effectively enabling boost while
> it should have been enabled.

Did you mean "disabled"?

It would be good to mention the failure scenario here too.

> Fixes: 27241c8b63bd ("cpufreq: Introduce policy_set_boost()")
> Reported-by: Robin Murphy <robin.murphy@arm.com>
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
> ---
>  drivers/cpufreq/cpufreq.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index d7426e1d8bdd..e85139bd0436 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -1630,7 +1630,7 @@ static int cpufreq_online(unsigned int cpu)
>          */
>         if (cpufreq_driver->set_boost && policy->boost_supported &&
>             (new_policy || !cpufreq_boost_enabled())) {
> -               ret = policy_set_boost(policy, cpufreq_boost_enabled());
> +               ret = cpufreq_driver->set_boost(policy, cpufreq_boost_enabled());
>                 if (ret) {
>                         /* If the set_boost fails, the online operation is not affected */
>                         pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu,
> --
> 2.34.1

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by Christian Loehle 3 months, 3 weeks ago

On 6/18/25 15:32, Rafael J. Wysocki wrote:
> On Mon, Jun 16, 2025 at 7:25 PM Christian Loehle
> <christian.loehle@arm.com> wrote:
>>
>> The boost_enabled early return in policy_set_boost() caused
>> the boost disabled at initialization to not actually set the
>> initial policy->max, therefore effectively enabling boost while
>> it should have been enabled.
> 
> Did you mean "disabled"?

Yup, the latter 'enabled' should be disabled.

> 
> It would be good to mention the failure scenario here too.
> 

Absolutely, let me respin this in a series that provides some context, too.

>> Fixes: 27241c8b63bd ("cpufreq: Introduce policy_set_boost()")
>> Reported-by: Robin Murphy <robin.murphy@arm.com>
>> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
>> ---
>>  drivers/cpufreq/cpufreq.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
>> index d7426e1d8bdd..e85139bd0436 100644
>> --- a/drivers/cpufreq/cpufreq.c
>> +++ b/drivers/cpufreq/cpufreq.c
>> @@ -1630,7 +1630,7 @@ static int cpufreq_online(unsigned int cpu)
>>          */
>>         if (cpufreq_driver->set_boost && policy->boost_supported &&
>>             (new_policy || !cpufreq_boost_enabled())) {
>> -               ret = policy_set_boost(policy, cpufreq_boost_enabled());
>> +               ret = cpufreq_driver->set_boost(policy, cpufreq_boost_enabled());
>>                 if (ret) {
>>                         /* If the set_boost fails, the online operation is not affected */
>>                         pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu,
>> --
>> 2.34.1

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by Dietmar Eggemann 3 months, 2 weeks ago

On 18/06/2025 16:57, Christian Loehle wrote:
> On 6/18/25 15:32, Rafael J. Wysocki wrote:
>> On Mon, Jun 16, 2025 at 7:25 PM Christian Loehle
>> <christian.loehle@arm.com> wrote:
>>>
>>> The boost_enabled early return in policy_set_boost() caused
>>> the boost disabled at initialization to not actually set the
>>> initial policy->max, therefore effectively enabling boost while
>>> it should have been enabled.
>>
>> Did you mean "disabled"?
> 
> Yup, the latter 'enabled' should be disabled.
> 
>>
>> It would be good to mention the failure scenario here too.
>>
> 
> Absolutely, let me respin this in a series that provides some context, too.

I got confused as well. Is this for a dt file with some (higher) OPPs
marked with 'turbo-mode' or not?

[...]

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by Robin Murphy 3 months, 3 weeks ago

On 2025-06-16 6:25 pm, Christian Loehle wrote:
> The boost_enabled early return in policy_set_boost() caused
> the boost disabled at initialization to not actually set the
> initial policy->max, therefore effectively enabling boost while
> it should have been enabled.
> 
> Fixes: 27241c8b63bd ("cpufreq: Introduce policy_set_boost()")

I think it's a bit older than that - I noticed this with 6.15 stable, 
prior to that refactoring, and from a poke through the history the 
underlying logic appears to date back to dd016f379ebc ("cpufreq: 
Introduce a more generic way to set default per-policy boost flag"). 
Hopefully someone can figure out the appropriate stable backport.

I can at least confirm that equivalently hacking out the "&& 
policy->boost_enabled != cpufreq_boost_enabled()" condition previously 
here does have the desired effect for me of initialising 
scaling_max_freq correctly at boot, but I'm not sure that's entirely 
correct on its own...

Thanks,
Robin.

> Reported-by: Robin Murphy <robin.murphy@arm.com>
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
> ---
>   drivers/cpufreq/cpufreq.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index d7426e1d8bdd..e85139bd0436 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -1630,7 +1630,7 @@ static int cpufreq_online(unsigned int cpu)
>   	 */
>   	if (cpufreq_driver->set_boost && policy->boost_supported &&
>   	    (new_policy || !cpufreq_boost_enabled())) {
> -		ret = policy_set_boost(policy, cpufreq_boost_enabled());
> +		ret = cpufreq_driver->set_boost(policy, cpufreq_boost_enabled());
>   		if (ret) {
>   			/* If the set_boost fails, the online operation is not affected */
>   			pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu,

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by zhenglifeng (A) 3 months, 3 weeks ago

On 2025/6/17 3:10, Robin Murphy wrote:
> On 2025-06-16 6:25 pm, Christian Loehle wrote:
>> The boost_enabled early return in policy_set_boost() caused
>> the boost disabled at initialization to not actually set the
>> initial policy->max, therefore effectively enabling boost while
>> it should have been enabled.
>>
>> Fixes: 27241c8b63bd ("cpufreq: Introduce policy_set_boost()")
> 
> I think it's a bit older than that - I noticed this with 6.15 stable, prior to that refactoring, and from a poke through the history the underlying logic appears to date back to dd016f379ebc ("cpufreq: Introduce a more generic way to set default per-policy boost flag"). Hopefully someone can figure out the appropriate stable backport.
> 
> I can at least confirm that equivalently hacking out the "&& policy->boost_enabled != cpufreq_boost_enabled()" condition previously here does have the desired effect for me of initialising scaling_max_freq correctly at boot, but I'm not sure that's entirely correct on its own...
> 
> Thanks,
> Robin.
> 
>> Reported-by: Robin Murphy <robin.murphy@arm.com>
>> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
>> ---
>>   drivers/cpufreq/cpufreq.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
>> index d7426e1d8bdd..e85139bd0436 100644
>> --- a/drivers/cpufreq/cpufreq.c
>> +++ b/drivers/cpufreq/cpufreq.c
>> @@ -1630,7 +1630,7 @@ static int cpufreq_online(unsigned int cpu)
>>        */
>>       if (cpufreq_driver->set_boost && policy->boost_supported &&
>>           (new_policy || !cpufreq_boost_enabled())) {
>> -        ret = policy_set_boost(policy, cpufreq_boost_enabled());
>> +        ret = cpufreq_driver->set_boost(policy, cpufreq_boost_enabled());
>>           if (ret) {
>>               /* If the set_boost fails, the online operation is not affected */
>>               pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu,
> 
> 

I don't quite understand what problem you've met. It semms like you guys
propose that set_boost() should be called no matter what
policy->boost_enabled is. Having more details would help to clarify things,
such as which driver you use and what you expect but not be achieved.

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by Christian Loehle 3 months, 3 weeks ago

On 6/17/25 03:14, zhenglifeng (A) wrote:
> On 2025/6/17 3:10, Robin Murphy wrote:
>> On 2025-06-16 6:25 pm, Christian Loehle wrote:
>>> The boost_enabled early return in policy_set_boost() caused
>>> the boost disabled at initialization to not actually set the
>>> initial policy->max, therefore effectively enabling boost while
>>> it should have been enabled.
>>>
>>> Fixes: 27241c8b63bd ("cpufreq: Introduce policy_set_boost()")
>>
>> I think it's a bit older than that - I noticed this with 6.15 stable, prior to that refactoring, and from a poke through the history the underlying logic appears to date back to dd016f379ebc ("cpufreq: Introduce a more generic way to set default per-policy boost flag"). Hopefully someone can figure out the appropriate stable backport.
>>
>> I can at least confirm that equivalently hacking out the "&& policy->boost_enabled != cpufreq_boost_enabled()" condition previously here does have the desired effect for me of initialising scaling_max_freq correctly at boot, but I'm not sure that's entirely correct on its own...
>>
>> Thanks,
>> Robin.
>>
>>> Reported-by: Robin Murphy <robin.murphy@arm.com>
>>> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
>>> ---
>>>   drivers/cpufreq/cpufreq.c | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
>>> index d7426e1d8bdd..e85139bd0436 100644
>>> --- a/drivers/cpufreq/cpufreq.c
>>> +++ b/drivers/cpufreq/cpufreq.c
>>> @@ -1630,7 +1630,7 @@ static int cpufreq_online(unsigned int cpu)
>>>        */
>>>       if (cpufreq_driver->set_boost && policy->boost_supported &&
>>>           (new_policy || !cpufreq_boost_enabled())) {
>>> -        ret = policy_set_boost(policy, cpufreq_boost_enabled());
>>> +        ret = cpufreq_driver->set_boost(policy, cpufreq_boost_enabled());
>>>           if (ret) {
>>>               /* If the set_boost fails, the online operation is not affected */
>>>               pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu,
>>
>>
> 
> I don't quite understand what problem you've met. It semms like you guys
> propose that set_boost() should be called no matter what
> policy->boost_enabled is. Having more details would help to clarify things,
> such as which driver you use and what you expect but not be achieved.
> 

so calling policy_set_boost(policy, enable) is a noop here if
policy->boost_enabled == cpufreq_boost_enabled():

	if (policy->boost_enabled == enable)
		return 0;

We have policy->boost_enabled == false on boot, thus never actually
setting policy->max up ever, which leads to the following:

# cat /sys/devices/system/cpu/cpufreq/policy4/scaling_boost_frequencies
2016000 
# cat /sys/devices/system/cpu/cpufreq/policy4/scaling_max_freq
2016000 
# cat /sys/devices/system/cpu/cpufreq/boost
0
# echo 1 > /sys/devices/system/cpu/cpufreq/boost
# echo 0 > /sys/devices/system/cpu/cpufreq/boost
# cat /sys/devices/system/cpu/cpufreq/policy4/scaling_boost_frequencies
1800000

Anyway I'll bisect some more to find the actual first bad commit and
resend.

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by Robin Murphy 3 months, 3 weeks ago

On 17/06/2025 9:20 am, Christian Loehle wrote:
> On 6/17/25 03:14, zhenglifeng (A) wrote:
>> On 2025/6/17 3:10, Robin Murphy wrote:
>>> On 2025-06-16 6:25 pm, Christian Loehle wrote:
>>>> The boost_enabled early return in policy_set_boost() caused
>>>> the boost disabled at initialization to not actually set the
>>>> initial policy->max, therefore effectively enabling boost while
>>>> it should have been enabled.
>>>>
>>>> Fixes: 27241c8b63bd ("cpufreq: Introduce policy_set_boost()")
>>>
>>> I think it's a bit older than that - I noticed this with 6.15 stable, prior to that refactoring, and from a poke through the history the underlying logic appears to date back to dd016f379ebc ("cpufreq: Introduce a more generic way to set default per-policy boost flag"). Hopefully someone can figure out the appropriate stable backport.
>>>
>>> I can at least confirm that equivalently hacking out the "&& policy->boost_enabled != cpufreq_boost_enabled()" condition previously here does have the desired effect for me of initialising scaling_max_freq correctly at boot, but I'm not sure that's entirely correct on its own...
>>>
>>> Thanks,
>>> Robin.
>>>
>>>> Reported-by: Robin Murphy <robin.murphy@arm.com>
>>>> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
>>>> ---
>>>>    drivers/cpufreq/cpufreq.c | 2 +-
>>>>    1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
>>>> index d7426e1d8bdd..e85139bd0436 100644
>>>> --- a/drivers/cpufreq/cpufreq.c
>>>> +++ b/drivers/cpufreq/cpufreq.c
>>>> @@ -1630,7 +1630,7 @@ static int cpufreq_online(unsigned int cpu)
>>>>         */
>>>>        if (cpufreq_driver->set_boost && policy->boost_supported &&
>>>>            (new_policy || !cpufreq_boost_enabled())) {
>>>> -        ret = policy_set_boost(policy, cpufreq_boost_enabled());
>>>> +        ret = cpufreq_driver->set_boost(policy, cpufreq_boost_enabled());
>>>>            if (ret) {
>>>>                /* If the set_boost fails, the online operation is not affected */
>>>>                pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu,
>>>
>>>
>>
>> I don't quite understand what problem you've met. It semms like you guys
>> propose that set_boost() should be called no matter what
>> policy->boost_enabled is. Having more details would help to clarify things,
>> such as which driver you use and what you expect but not be achieved.
>>
> 
> so calling policy_set_boost(policy, enable) is a noop here if
> policy->boost_enabled == cpufreq_boost_enabled():
> 
> 	if (policy->boost_enabled == enable)
> 		return 0;
> 
> We have policy->boost_enabled == false on boot, thus never actually
> setting policy->max up ever, which leads to the following:

And for clarity, this is with the cpufreq_dt driver (at least in my case).

Thanks,
Robin.

> # cat /sys/devices/system/cpu/cpufreq/policy4/scaling_boost_frequencies
> 2016000
> # cat /sys/devices/system/cpu/cpufreq/policy4/scaling_max_freq
> 2016000
> # cat /sys/devices/system/cpu/cpufreq/boost
> 0
> # echo 1 > /sys/devices/system/cpu/cpufreq/boost
> # echo 0 > /sys/devices/system/cpu/cpufreq/boost
> # cat /sys/devices/system/cpu/cpufreq/policy4/scaling_boost_frequencies
> 1800000
> 
> Anyway I'll bisect some more to find the actual first bad commit and
> resend.

Re: [PATCH] cpufreq: Fix initialization with disabled boost

Posted by Viresh Kumar 3 months, 3 weeks ago

On 17-06-25, 13:56, Robin Murphy wrote:
> On 17/06/2025 9:20 am, Christian Loehle wrote:
> > so calling policy_set_boost(policy, enable) is a noop here if
> > policy->boost_enabled == cpufreq_boost_enabled():
> > 
> > 	if (policy->boost_enabled == enable)
> > 		return 0;
> > 
> > We have policy->boost_enabled == false on boot, thus never actually
> > setting policy->max up ever, which leads to the following:
 
> And for clarity, this is with the cpufreq_dt driver (at least in my case).
 
> > # cat /sys/devices/system/cpu/cpufreq/policy4/scaling_boost_frequencies
> > 2016000
> > # cat /sys/devices/system/cpu/cpufreq/policy4/scaling_max_freq
> > 2016000
> > # cat /sys/devices/system/cpu/cpufreq/boost
> > 0
> > # echo 1 > /sys/devices/system/cpu/cpufreq/boost
> > # echo 0 > /sys/devices/system/cpu/cpufreq/boost
> > # cat /sys/devices/system/cpu/cpufreq/policy4/scaling_boost_frequencies
> > 1800000

Hi Christian and Robin,

I am not clear on why this happens. It would be helpful to get a few answers
here:

- Which driver are you using Christian ?
- Is this during system boot or suspend/resume ?
- At boot you see the boost state as 0, while policy->max uses it ?

I tried this with cpufreq-dt driver and this is what I see on boot:

# grep . /sys/devices/system/cpu/cpufreq/policy0/*
/sys/devices/system/cpu/cpufreq/policy0/affected_cpus:0 1 2 3
/sys/devices/system/cpu/cpufreq/policy0/boost:0
/sys/devices/system/cpu/cpufreq/policy0/cpuinfo_cur_freq:24000
/sys/devices/system/cpu/cpufreq/policy0/cpuinfo_max_freq:960000
/sys/devices/system/cpu/cpufreq/policy0/cpuinfo_min_freq:208000
/sys/devices/system/cpu/cpufreq/policy0/cpuinfo_transition_latency:900000
/sys/devices/system/cpu/cpufreq/policy0/related_cpus:0 1 2 3
/sys/devices/system/cpu/cpufreq/policy0/scaling_available_frequencies:208000 432000 729000 960000
/sys/devices/system/cpu/cpufreq/policy0/scaling_available_governors:ondemand userspace performance schedutil
/sys/devices/system/cpu/cpufreq/policy0/scaling_boost_frequencies:1200000
/sys/devices/system/cpu/cpufreq/policy0/scaling_cur_freq:208000
/sys/devices/system/cpu/cpufreq/policy0/scaling_driver:cpufreq-dt
/sys/devices/system/cpu/cpufreq/policy0/scaling_governor:schedutil
/sys/devices/system/cpu/cpufreq/policy0/scaling_max_freq:960000
/sys/devices/system/cpu/cpufreq/policy0/scaling_min_freq:208000
/sys/devices/system/cpu/cpufreq/policy0/scaling_setspeed:<unsupported>

# grep . /sys/devices/system/cpu/cpufreq/boost
0

So both global and policy boost are disabled and they don't show up in
scaling_max_freq or scaling_available_frequencies.

-- 
viresh