Re: [PATCH v4 8/8] cpufreq: CPPC: add autonomous mode boot parameter support

[Sumit Gupta <sumitg@nvidia.com>]

>>>
>>>
>>> On 11/5/25 12:38, Sumit Gupta wrote:
>>>> Add kernel boot parameter 'cppc_cpufreq.auto_sel_mode' to enable CPPC
>>>> autonomous performance selection at system startup. When autonomous
>>>> mode
>>>> is enabled, the hardware automatically adjusts CPU performance 
>>>> based on
>>>> workload demands using Energy Performance Preference (EPP) hints.
>>>>
>>>> This parameter allows to configure the autonomous mode on all CPUs
>>>> without requiring runtime sysfs manipulation if the 'auto_sel' 
>>>> register
>>>> is present.
>>>>
>>>> When auto_sel_mode=1:
>>>> - All CPUs are configured for autonomous operation during module init
>>>> - EPP is set to performance preference (0x0) by default
>>>> - Min/max performance bounds use defaults
>>>> - CPU frequency scaling is handled by hardware instead of OS governor
>>>>
>>>> For Documentation/:
>>>> Reviewed-by: Randy Dunlap<rdunlap@infradead.org>
>>>> Signed-off-by: Sumit Gupta<sumitg@nvidia.com>
>>>> ---
>>>>   .../admin-guide/kernel-parameters.txt         |  12 ++
>>>>   drivers/cpufreq/cppc_cpufreq.c                | 197
>>>> +++++++++++++++---
>>>>   2 files changed, 182 insertions(+), 27 deletions(-)
>>>>
>>>> diff --git a/Documentation/admin-guide/kernel-parameters.txt
>>>> b/Documentation/admin-guide/kernel-parameters.txt
>>>> index b8f8f5d74093..048f84008a7e 100644
>>>> --- a/Documentation/admin-guide/kernel-parameters.txt
>>>> +++ b/Documentation/admin-guide/kernel-parameters.txt
>>>> @@ -929,6 +929,18 @@
>>>>                       Format:
>>>> <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
>>>>
>>>> +     cppc_cpufreq.auto_sel_mode=
>>>> +                     [CPU_FREQ] Enable ACPI CPPC autonomous
>>>> performance selection.
>>>> +                     When enabled, hardware automatically adjusts
>>>> CPU frequency
>>>> +                     on all CPUs based on workload demands. In
>>>> Autonomous mode,
>>>> +                     Energy Performance Preference(EPP) hints guide
>>>> hardware
>>>> +                     toward performance(0x0) or energy efficiency
>>>> (0xff).
>>>> +                     Requires ACPI CPPC autonomous selection
>>>> register support.
>>>> +                     Format: <bool>
>>>> +                     Default: 0 (disabled)
>>>> +                     0: use cpufreq governors
>>>> +                     1: enable if supoorted by hardware
>>>> +
>>>>       cpuidle.off=1   [CPU_IDLE]
>>>>                       disable the cpuidle sub-system
>>>>
>>>> diff --git a/drivers/cpufreq/cppc_cpufreq.c
>>>> b/drivers/cpufreq/cppc_cpufreq.c
>>>> index d1b44beaddda..0a55ab011317 100644
>>>> --- a/drivers/cpufreq/cppc_cpufreq.c
>>>> +++ b/drivers/cpufreq/cppc_cpufreq.c
>>>> @@ -28,8 +28,12 @@
>>>>   #include <acpi/cppc_acpi.h>
>>>>
>>>>   static struct cpufreq_driver cppc_cpufreq_driver;
>>>> +
>>>>   static DEFINE_MUTEX(cppc_cpufreq_update_autosel_config_lock);
>>>>
>>>> +/* Autonomous Selection */
>>>> +static bool auto_sel_mode;
>>>> +
>>>>   #ifdef CONFIG_ACPI_CPPC_CPUFREQ_FIE
>>>>   static enum {
>>>>       FIE_UNSET = -1,
>>>> @@ -272,8 +276,13 @@ static int cppc_cpufreq_set_target(struct
>>>> cpufreq_policy *policy,
>>>>       freqs.old = policy->cur;
>>>>       freqs.new = target_freq;
>>>>
>>>> +     /*
>>>> +      * In autonomous selection mode, hardware handles frequency
>>>> scaling directly
>>>> +      * based on workload and EPP hints. So, skip the OS frequency
>>>> set requests.
>>>> +      */
>>>>       cpufreq_freq_transition_begin(policy, &freqs);
>>>> -     ret = cppc_set_perf(cpu, &cpu_data->perf_ctrls);
>>>> +     if (!cpu_data->perf_caps.auto_sel)
>>>> +             ret = cppc_set_perf(cpu, &cpu_data->perf_ctrls);
>>>>       cpufreq_freq_transition_end(policy, &freqs, ret != 0);
>>>>
>>>>       if (ret)
>>>> @@ -565,6 +574,12 @@ static struct cppc_cpudata
>>>> *cppc_cpufreq_get_cpu_data(unsigned int cpu)
>>>>               goto free_mask;
>>>>       }
>>>>
>>>> +     ret = cppc_get_perf(cpu, &cpu_data->perf_ctrls);
>>>> +     if (ret) {
>>>> +             pr_debug("Err reading CPU%d perf ctrls:ret:%d\n", cpu,
>>>> ret);
>>>> +             goto free_mask;
>>>> +     }
>>>> +
>>>>       return cpu_data;
>>>>
>>>>   free_mask:
>>>> @@ -666,11 +681,81 @@ static int
>>>> cppc_cpufreq_update_autosel_val(struct cpufreq_policy *policy, bool a
>>>>       return 0;
>>>>   }
>>>>
>>>> +static int cppc_cpufreq_update_epp_val(struct cpufreq_policy
>>>> *policy, u32 epp)
>>>> +{
>>>> +     struct cppc_cpudata *cpu_data = policy->driver_data;
>>>> +     unsigned int cpu = policy->cpu;
>>>> +     int ret;
>>>> +
>>>> +     pr_debug("cpu%d, eppcurr:%u,new:%u\n", cpu,
>>>> cpu_data->perf_ctrls.energy_perf, epp);
>>>> +
>>>> + guard(mutex)(&cppc_cpufreq_update_autosel_config_lock);
>>> Do we need to take the mutex ? Or is it reserved to auto_sel ?
>>
>> Will move this to parent function.
>> Explained more in reply of the previous patch '7/8'.
>>
>>>> +
>>>> +     ret = cppc_set_epp(cpu, epp);
>>>> +     if (ret) {
>>>> +             pr_warn("failed to set energy_perf forcpu:%d (%d)\n",
>>>> cpu, ret);
>>>> +             return ret;
>>>> +     }
>>>> +     cpu_data->perf_ctrls.energy_perf = epp;
>>>> +
>>>> +     return 0;
>>>> +}
>>>> +
>>>> +/**
>>>> + * cppc_cpufreq_update_autosel_config - Update Autonomous selection
>>>> configuration
>>>> + * @policy: cpufreq policy for the CPU
>>>> + * @min_perf: minimum performance value to set
>>>> + * @max_perf: maximum performance value to set
>>>> + * @auto_sel: autonomous selection mode enable/disable (also
>>>> controls min/max perf reg updates)
>>>> + * @epp_val: energy performance preference value
>>>> + * @update_epp: whether to update EPP register
>>>> + * @update_policy: whether to update policy constraints
>>>> + *
>>>> + * Return: 0 on success, negative error code on failure
>>>> + */
>>>> +static int cppc_cpufreq_update_autosel_config(struct cpufreq_policy
>>>> *policy,
>>>> +                                           u64 min_perf, u64
>>>> max_perf, bool auto_sel,
>>>> +                                           u32 epp_val, bool
>>>> update_epp, bool update_policy)
>>>> +{
>>>> +     const unsigned int cpu = policy->cpu;
>>>> +     int ret;
>>>> +
>>>> +     /*
>>>> +      * Set min/max performance registers and update policy
>>>> constraints.
>>>> +      *   When enabling: update both registers and policy.
>>>> +      *   When disabling: update policy only.
>>>> +      * Continue even if min/max are not supported, as EPP and 
>>>> autosel
>>>> +      * might still be supported.
>>>> +      */
>>>> +     ret = cppc_cpufreq_set_min_perf(policy, min_perf, auto_sel,
>>>> update_policy);
>>>> +     if (ret && ret != -EOPNOTSUPP)
>>>> +             return ret;
>>>> +
>>>> +     ret = cppc_cpufreq_set_max_perf(policy, max_perf, auto_sel,
>>>> update_policy);
>>>> +     if (ret && ret != -EOPNOTSUPP)
>>>> +             return ret;
>>>> +
>>>> +     if (update_epp) {
>>>> +             ret = cppc_cpufreq_update_epp_val(policy, epp_val);
>>>> +             if (ret)
>>>> +                     return ret;
>>>> +     }
>>>> +
>>>> +     ret = cppc_cpufreq_update_autosel_val(policy, auto_sel);
>>>> +     if (ret)
>>>> +             return ret;
>>>> +
>>>> +     pr_debug("Updated autonomous config [%llu-%llu] for CPU%d\n",
>>>> min_perf, max_perf, cpu);
>>>> +
>>>> +     return 0;
>>>> +}
>>>> +
>>>>   static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy)
>>>>   {
>>>>       unsigned int cpu = policy->cpu;
>>>>       struct cppc_cpudata *cpu_data;
>>>>       struct cppc_perf_caps *caps;
>>>> +     u64 min_perf, max_perf;
>>>>       int ret;
>>>>
>>>>       cpu_data = cppc_cpufreq_get_cpu_data(cpu);
>>>> @@ -734,11 +819,31 @@ static int cppc_cpufreq_cpu_init(struct
>>>> cpufreq_policy *policy)
>>>>       policy->cur = cppc_perf_to_khz(caps, caps->highest_perf);
>>>>       cpu_data->perf_ctrls.desired_perf = caps->highest_perf;
>>>>
>>>> -     ret = cppc_set_perf(cpu, &cpu_data->perf_ctrls);
>>>> -     if (ret) {
>>>> -             pr_debug("Err setting perf value:%d on CPU:%d. 
>>>> ret:%d\n",
>>>> -                      caps->highest_perf, cpu, ret);
>>>> -             goto out;
>>>> +     if (cpu_data->perf_caps.auto_sel) {
>>>> +             ret = cppc_set_enable(cpu, true);
>>> The CPPC enable register is optional.
>>> However this doesn't mean CPPC is not working.
>>
>> Ya, changed this in v5.
>>
>>>> +             if (ret) {
>>>> +                     pr_err("Failed to enable CPPC on cpu%d
>>>> (%d)\n", cpu, ret);
>>>> +                     goto out;
>>>> +             }
>>>> +
>>>> +             min_perf = cpu_data->perf_ctrls.min_perf ?
>>>> +                        cpu_data->perf_ctrls.min_perf :
>>>> caps->lowest_nonlinear_perf;
>>>> +             max_perf = cpu_data->perf_ctrls.max_perf ?
>>>> +                        cpu_data->perf_ctrls.max_perf :
>>>> caps->nominal_perf;
>>>> +
>>>> +             ret = cppc_cpufreq_update_autosel_config(policy,
>>>> min_perf, max_perf, true,
>>>> + CPPC_EPP_PERFORMANCE_PREF, true, false);
>>>> +             if (ret) {
>>>> +                     cppc_set_enable(cpu, false);
>>>> +                     goto out;
>>>> +             }
>>>> +     } else {
>>>> +             ret = cppc_set_perf(cpu, &cpu_data->perf_ctrls);
>>>> +             if (ret) {
>>>> +                     pr_debug("Err setting perf value:%d on CPU:%d.
>>>> ret:%d\n",
>>>> +                              caps->highest_perf, cpu, ret);
>>>> +                     goto out;
>>>> +             }
>>>>       }
>>>>
>>>>       cppc_cpufreq_cpu_fie_init(policy);
>>>> @@ -910,7 +1015,6 @@ static int
>>>> cppc_cpufreq_update_auto_select(struct cpufreq_policy *policy, bool e
>>>>       struct cppc_perf_caps *caps = &cpu_data->perf_caps;
>>>>       u64 min_perf = caps->lowest_nonlinear_perf;
>>>>       u64 max_perf = caps->nominal_perf;
>>>> -     int ret;
>>>>
>>>>       if (enable) {
>>>>               if (cpu_data->perf_ctrls.min_perf)
>>>> @@ -919,26 +1023,8 @@ static int
>>>> cppc_cpufreq_update_auto_select(struct cpufreq_policy *policy, bool e
>>>>                       max_perf = cpu_data->perf_ctrls.max_perf;
>>>>       }
>>>>
>>>> -     /*
>>>> -      * Set min/max performance registers and update policy
>>>> constraints.
>>>> -      *   When enabling: update both registers and policy.
>>>> -      *   When disabling: update policy only.
>>>> -      * Continue even if min/max are not supported, as EPP and 
>>>> autosel
>>>> -      * might still be supported.
>>>> -      */
>>>> -     ret = cppc_cpufreq_set_min_perf(policy, min_perf, enable, true);
>>>> -     if (ret && ret != -EOPNOTSUPP)
>>>> -             return ret;
>>>> -
>>>> -     ret = cppc_cpufreq_set_max_perf(policy, max_perf, enable, true);
>>>> -     if (ret && ret != -EOPNOTSUPP)
>>>> -             return ret;
>>>> -
>>>> -     ret = cppc_cpufreq_update_autosel_val(policy, enable);
>>>> -     if (ret)
>>>> -             return ret;
>>>> -
>>>> -     return 0;
>>>> +     return cppc_cpufreq_update_autosel_config(policy, min_perf,
>>>> max_perf, enable,
>>>> +                                               0, false, true);
>>>>   }
>>>>
>>>>   static ssize_t store_auto_select(struct cpufreq_policy *policy,
>>>> const char *buf, size_t count)
>>>> @@ -1146,13 +1232,61 @@ static struct cpufreq_driver
>>>> cppc_cpufreq_driver = {
>>>>       .name = "cppc_cpufreq",
>>>>   };
>>>>
>>>> +static int cppc_cpufreq_set_epp_autosel_allcpus(bool auto_sel, u64
>>>> epp)
>>>> +{
>>>> +     int cpu, ret;
>>>> +
>>>> +     for_each_present_cpu(cpu) {
>>>> +             ret = cppc_set_epp(cpu, epp);
>>> Isn't the EPP optional ?
>>
>> Moving this to cppc_cpufreq_cpu_init in v5. Will add handling for
>> EOPNOTSUPP.
>>
>>
>>> If autonomous selection is available but not EPP, we will bail out.
>>
>> I couldn't find in spec that EPP is mandatory when auto_select is
>> enabled.
>
> I was thinking about the case where the platform:
> - supports auto_sel
> - doesn't support EPP
> Then won't this function return an error code and not set auto_sel even
> though
> we could have enabled it (without setting the EPP value) ?
>

Ya, right. For that will handle EOPNOTSUPP after reading EPP.

Thank you,
Sumit Gupta


>
>>
>>
>>>> +             if (ret) {
>>>> +                     pr_warn("Failed to set EPP on CPU%d (%d)\n",
>>>> cpu, ret);
>>>> +                     goto disable_all;
>>>> +             }
>>>> +
>>>> +             ret = cppc_set_auto_sel(cpu, auto_sel);
>>>
>>> Also, it is possible that a platform only supports autonomous 
>>> selection.
>>> In this case, writing to auto_sel will fail, but auto_sel is still
>>> relevant.
>>
>> I am not sure if we will have such platform which only supports
>> Autonomous
>> mode and has auto_sel as read only. Will add handling for EOPNOTSUPP
>> if we
>> have such cases as the cppc_get_reg_val() will returns this error.
>>
>> Thank you,
>> Sumit Gupta
>>
>> ....
>>