AMU counters on certain CPPC-based platforms tend to yield inaccurate
delivered performance measurements on systems that are idle/mostly idle.
This results in an inaccurate frequency being stored by cpufreq in its
policy structure when the CPU is brought online. [1]

Consequently, if the userspace governor tries to set the frequency to a
new value, there is a possibility that it would be the erroneous value
stored earlier. In such a scenario, cpufreq would assume that the
requested frequency has already been set and return early, resulting in
the correct/new frequency request never making it to the hardware.

Since the operating frequency is liable to this sort of inconsistency,
mark the CPPC driver with CPUFREQ_NEED_UPDATE_LIMITS so that it is always
invoked when a target frequency update is requested.

[1] https://lore.kernel.org/linux-pm/20250619000925.415528-3-pmalani@google.com/

Cc: Beata Michalska <beata.michalska@arm.com>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Prashant Malani <pmalani@google.com>
---
 drivers/cpufreq/cppc_cpufreq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
index a1fd0ff22bc5..4a17162a392d 100644
--- a/drivers/cpufreq/cppc_cpufreq.c
+++ b/drivers/cpufreq/cppc_cpufreq.c
@@ -910,7 +910,7 @@ static struct freq_attr *cppc_cpufreq_attr[] = {
 };
 
 static struct cpufreq_driver cppc_cpufreq_driver = {
-	.flags = CPUFREQ_CONST_LOOPS,
+	.flags = CPUFREQ_CONST_LOOPS | CPUFREQ_NEED_UPDATE_LIMITS,
 	.verify = cppc_verify_policy,
 	.target = cppc_cpufreq_set_target,
 	.get = cppc_cpufreq_get_rate,
-- 
2.50.0.727.gbf7dc18ff4-goog

On 22-07-25, 05:55, Prashant Malani wrote:
> AMU counters on certain CPPC-based platforms tend to yield inaccurate
> delivered performance measurements on systems that are idle/mostly idle.
> This results in an inaccurate frequency being stored by cpufreq in its
> policy structure when the CPU is brought online. [1]
> 
> Consequently, if the userspace governor tries to set the frequency to a
> new value, there is a possibility that it would be the erroneous value
> stored earlier. In such a scenario, cpufreq would assume that the
> requested frequency has already been set and return early, resulting in
> the correct/new frequency request never making it to the hardware.
> 
> Since the operating frequency is liable to this sort of inconsistency,
> mark the CPPC driver with CPUFREQ_NEED_UPDATE_LIMITS so that it is always
> invoked when a target frequency update is requested.
> 
> [1] https://lore.kernel.org/linux-pm/20250619000925.415528-3-pmalani@google.com/
> 
> Cc: Beata Michalska <beata.michalska@arm.com>
> Cc: Viresh Kumar <viresh.kumar@linaro.org>
> Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> Signed-off-by: Prashant Malani <pmalani@google.com>
> ---
>  drivers/cpufreq/cppc_cpufreq.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
> index a1fd0ff22bc5..4a17162a392d 100644
> --- a/drivers/cpufreq/cppc_cpufreq.c
> +++ b/drivers/cpufreq/cppc_cpufreq.c
> @@ -910,7 +910,7 @@ static struct freq_attr *cppc_cpufreq_attr[] = {
>  };
>  
>  static struct cpufreq_driver cppc_cpufreq_driver = {
> -	.flags = CPUFREQ_CONST_LOOPS,
> +	.flags = CPUFREQ_CONST_LOOPS | CPUFREQ_NEED_UPDATE_LIMITS,
>  	.verify = cppc_verify_policy,
>  	.target = cppc_cpufreq_set_target,
>  	.get = cppc_cpufreq_get_rate,

Acked-by: Viresh Kumar <viresh.kumar@linaro.org>

-- 
viresh

On Tue, Jul 22, 2025 at 8:03 AM Viresh Kumar <viresh.kumar@linaro.org> wrote:
>
> On 22-07-25, 05:55, Prashant Malani wrote:
> > AMU counters on certain CPPC-based platforms tend to yield inaccurate
> > delivered performance measurements on systems that are idle/mostly idle.
> > This results in an inaccurate frequency being stored by cpufreq in its
> > policy structure when the CPU is brought online. [1]
> >
> > Consequently, if the userspace governor tries to set the frequency to a
> > new value, there is a possibility that it would be the erroneous value
> > stored earlier. In such a scenario, cpufreq would assume that the
> > requested frequency has already been set and return early, resulting in
> > the correct/new frequency request never making it to the hardware.
> >
> > Since the operating frequency is liable to this sort of inconsistency,
> > mark the CPPC driver with CPUFREQ_NEED_UPDATE_LIMITS so that it is always
> > invoked when a target frequency update is requested.
> >
> > [1] https://lore.kernel.org/linux-pm/20250619000925.415528-3-pmalani@google.com/
> >
> > Cc: Beata Michalska <beata.michalska@arm.com>
> > Cc: Viresh Kumar <viresh.kumar@linaro.org>
> > Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > Signed-off-by: Prashant Malani <pmalani@google.com>
> > ---
> >  drivers/cpufreq/cppc_cpufreq.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
> > index a1fd0ff22bc5..4a17162a392d 100644
> > --- a/drivers/cpufreq/cppc_cpufreq.c
> > +++ b/drivers/cpufreq/cppc_cpufreq.c
> > @@ -910,7 +910,7 @@ static struct freq_attr *cppc_cpufreq_attr[] = {
> >  };
> >
> >  static struct cpufreq_driver cppc_cpufreq_driver = {
> > -     .flags = CPUFREQ_CONST_LOOPS,
> > +     .flags = CPUFREQ_CONST_LOOPS | CPUFREQ_NEED_UPDATE_LIMITS,
> >       .verify = cppc_verify_policy,
> >       .target = cppc_cpufreq_set_target,
> >       .get = cppc_cpufreq_get_rate,
>
> Acked-by: Viresh Kumar <viresh.kumar@linaro.org>

Applied as 6.17 material, thanks!