Re: [PATCH v2] ACPI: CPPC: mask Desired_Excursion when autonomous selection is enabled

[Pierre Gondois]

Hello Xueqin,

NIT: The patch only modifies the cppc_cpufreq.c file, so I think the 
commit line should be:

"ACPI: CPPC:" -> "cpufreq: CPPC:"

On 4/30/26 11:22, luoxueqin wrote:
> From: Xueqin luo <luoxueqin@xxxxxxxxxx>
>
> According to the ACPI specification, the Desired_Excursion field is not
> utilized when Autonomous Selection is enabled. In this mode, the bit is
> architecturally ignored and does not carry meaningful information.
>
> Currently, the kernel exposes the raw Performance Limited register
> value to userspace through the cpufreq sysfs interface. This may lead to
> misinterpretation, as userspace may assume Desired_Excursion is valid
> even when autonomous selection is active.
>
> To provide a stable and semantically correct ABI, mask out the
> Desired_Excursion bit when autonomous selection is enabled, so that
> userspace does not observe undefined or misleading values.
>
> Writes are left unchanged, as the field is architecturally ignored in
> this mode and write attempts are harmless.
>
> Signed-off-by: Xueqin Luo <luoxueqin@xxxxxxxxxx>
> ---
>
> v2: update subject
>
>   drivers/cpufreq/cppc_cpufreq.c | 40 ++++++++++++++++++++++++++++++++--
>   1 file changed, 38 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
> index 7e7f9dfb7a24..c7444a6ceb55 100644
> --- a/drivers/cpufreq/cppc_cpufreq.c
> +++ b/drivers/cpufreq/cppc_cpufreq.c
> @@ -982,8 +982,44 @@ store_energy_performance_preference_val(struct cpufreq_policy *policy,
>   	return count;
>   }
>   
> -CPPC_CPUFREQ_ATTR_RW_U64(perf_limited, cppc_get_perf_limited,
> -			 cppc_set_perf_limited)
> +static ssize_t show_perf_limited(struct cpufreq_policy *policy, char *buf)
> +{
> +	bool auto_sel_enabled = false;
> +	u64 perf_limited;
> +	int ret;
> +
> +	/* Get Autonomous Selection status */
> +	ret = cppc_get_auto_sel(policy->cpu, &auto_sel_enabled);

Would it be equivalent to used the cached value ?

cpu_data->perf_ctrls.auto_sel

------------

Also, if the perf_limited register is not supported,
we are not emitting "<unsupported>" anymore.

Maybe we can keep the CPPC_CPUFREQ_ATTR_RW_U64()
macro and replace cppc_get_perf_limited() by a wrapper
doing what you want ?

> +	if (ret) {
> +		if (ret == -EOPNOTSUPP)
> +			auto_sel_enabled = false;
> +		else
> +			return ret;
> +	}
> +
> +	/* Read the Performance Limited register */
> +	ret = cppc_get_perf_limited(policy->cpu, &perf_limited);
> +	if (ret)
> +		return ret;
> +
> +	/*
> +	 * Desired_Excursion is architecturally ignored when Autonomous
> +	 * Selection is enabled. Mask it to avoid exposing undefined
> +	 * semantics to userspace.
> +	 */
> +	if (auto_sel_enabled)
> +		perf_limited &= ~CPPC_PERF_LIMITED_DESIRED_EXCURSION;
> +
> +	return sysfs_emit(buf, "%llu\n", perf_limited);
> +
> +}
> +
> +static ssize_t store_perf_limited(struct cpufreq_policy *policy,
> +				 const char *buf, size_t count)
> +{
> +	return cppc_cpufreq_sysfs_store_u64(policy->cpu,
> +					cppc_set_perf_limited, buf, count);
> +}
>   
>   cpufreq_freq_attr_ro(freqdomain_cpus);
>   cpufreq_freq_attr_rw(auto_select);