| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <0a2b7d49-bd1f-442e-9247-705cd2eb0d1c@nvidia.com>
Date: Thu, 8 Jan 2026 20:08:30 +0530
From: Sumit Gupta <sumitg@...dia.com>
To: "zhenglifeng (A)" <zhenglifeng1@...wei.com>, rafael@...nel.org,
viresh.kumar@...aro.org, lenb@...nel.org, robert.moore@...el.com,
corbet@....net, pierre.gondois@....com, rdunlap@...radead.org,
ray.huang@....com, gautham.shenoy@....com, mario.limonciello@....com,
perry.yuan@....com, ionela.voinescu@....com, zhanjie9@...ilicon.com,
linux-pm@...r.kernel.org, linux-acpi@...r.kernel.org,
linux-doc@...r.kernel.org, acpica-devel@...ts.linux.dev,
linux-kernel@...r.kernel.org
Cc: linux-tegra@...r.kernel.org, treding@...dia.com, jonathanh@...dia.com,
vsethi@...dia.com, ksitaraman@...dia.com, sanjayc@...dia.com,
nhartman@...dia.com, bbasu@...dia.com, sumitg@...dia.com
Subject: Re: [PATCH v5 06/11] ACPI: CPPC: add APIs and sysfs interface for
perf_limited
On 25/12/25 17:36, zhenglifeng (A) wrote:
> External email: Use caution opening links or attachments
>
>
> On 2025/12/23 20:13, Sumit Gupta wrote:
>> Add sysfs interface to read/write the Performance Limited register.
>>
>> The Performance Limited register indicates to the OS that an
>> unpredictable event (like thermal throttling) has limited processor
>> performance. It contains two sticky bits set by the platform:
>> - Bit 0 (Desired_Excursion): Set when delivered performance is
>> constrained below desired performance. Not used when Autonomous
>> Selection is enabled.
>> - Bit 1 (Minimum_Excursion): Set when delivered performance is
>> constrained below minimum performance.
>>
>> These bits remain set until OSPM explicitly clears them. The write
>> operation accepts a bitmask of bits to clear:
>> - Write 1 to clear bit 0
>> - Write 2 to clear bit 1
>> - Write 3 to clear both bits
> It's a bit odd that users write a 1 to and then read a 0 from the sysfs
> file. I think it is better to seperate these two bits, as two sysfs files.
> Then users can write '0' or 'clear' or others into them to clear each bit.
I think its better to keep one sysfs interface per HW register.
Can change the perf_limited write interface to accept bit index
instead of bitmask.
- Write 0 to clear bit 0 (desired performance excursion)
- Write 1 to clear bit 1 (minimum performance excursion)
Thank you,
Sumit Gupta
>
>> This enables users to detect if platform throttling impacted a workload.
>> Users clear the register before execution, run the workload, then check
>> afterward - if set, hardware throttling occurred during that time window.
>>
>> The interface is exposed as:
>> /sys/devices/system/cpu/cpuX/cpufreq/perf_limited
>>
>> Signed-off-by: Sumit Gupta <sumitg@...dia.com>
>> ---
>> drivers/acpi/cppc_acpi.c | 56 ++++++++++++++++++++++++++++++++++
>> drivers/cpufreq/cppc_cpufreq.c | 31 +++++++++++++++++++
>> include/acpi/cppc_acpi.h | 15 +++++++++
>> 3 files changed, 102 insertions(+)
>>
>> diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c
>> index 9f28c20d902d..ffd866c1c00d 100644
>> --- a/drivers/acpi/cppc_acpi.c
>> +++ b/drivers/acpi/cppc_acpi.c
>> @@ -1786,6 +1786,62 @@ int cppc_set_max_perf(int cpu, u32 max_perf)
>> }
>> EXPORT_SYMBOL_GPL(cppc_set_max_perf);
>>
>> +/**
>> + * cppc_get_perf_limited - Get the Performance Limited register value.
>> + * @cpu: CPU from which to get Performance Limited register.
>> + * @perf_limited: Pointer to store the Performance Limited value.
>> + *
>> + * The returned value contains sticky status bits indicating platform-imposed
>> + * performance limitations.
>> + *
>> + * Return: 0 for success, -EIO on failure, -EOPNOTSUPP if not supported.
>> + */
>> +int cppc_get_perf_limited(int cpu, u64 *perf_limited)
>> +{
>> + return cppc_get_reg_val(cpu, PERF_LIMITED, perf_limited);
>> +}
>> +EXPORT_SYMBOL_GPL(cppc_get_perf_limited);
>> +
>> +/**
>> + * cppc_set_perf_limited() - Clear bits in the Performance Limited register.
>> + * @cpu: CPU on which to write register.
>> + * @bits_to_clear: Bitmask of bits to clear in the perf_limited register.
>> + *
>> + * The Performance Limited register contains two sticky bits set by platform:
>> + * - Bit 0 (Desired_Excursion): Set when delivered performance is constrained
>> + * below desired performance. Not used when Autonomous Selection is enabled.
>> + * - Bit 1 (Minimum_Excursion): Set when delivered performance is constrained
>> + * below minimum performance.
>> + *
>> + * These bits are sticky and remain set until OSPM explicitly clears them.
>> + * This function only allows clearing bits (the platform sets them).
>> + *
>> + * Return: 0 for success, -EINVAL for invalid bits, -EIO on register
>> + * access failure, -EOPNOTSUPP if not supported.
>> + */
>> +int cppc_set_perf_limited(int cpu, u64 bits_to_clear)
>> +{
>> + u64 current_val, new_val;
>> + int ret;
>> +
>> + /* Only bits 0 and 1 are valid */
>> + if (bits_to_clear & ~CPPC_PERF_LIMITED_MASK)
>> + return -EINVAL;
>> +
>> + if (!bits_to_clear)
>> + return 0;
>> +
>> + ret = cppc_get_perf_limited(cpu, ¤t_val);
>> + if (ret)
>> + return ret;
>> +
>> + /* Clear the specified bits */
>> + new_val = current_val & ~bits_to_clear;
>> +
>> + return cppc_set_reg_val(cpu, PERF_LIMITED, new_val);
>> +}
>> +EXPORT_SYMBOL_GPL(cppc_set_perf_limited);
>> +
>> /**
>> * cppc_set_enable - Set to enable CPPC on the processor by writing the
>> * Continuous Performance Control package EnableRegister field.
>> diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
>> index 1e282dfabc76..1f8825006940 100644
>> --- a/drivers/cpufreq/cppc_cpufreq.c
>> +++ b/drivers/cpufreq/cppc_cpufreq.c
>> @@ -1052,12 +1052,42 @@ static ssize_t store_max_perf(struct cpufreq_policy *policy, const char *buf,
>> return count;
>> }
>>
>> +/**
>> + * show_perf_limited - Show Performance Limited register status
>> + * @policy: cpufreq policy
>> + * @buf: buffer to write the value to
>> + *
>> + * Read the Performance Limited register to check if platform throttling
>> + * (thermal/power/current limits) occurred.
>> + */
>> +static ssize_t show_perf_limited(struct cpufreq_policy *policy, char *buf)
>> +{
>> + return cppc_cpufreq_sysfs_show_u64(policy->cpu,
>> + cppc_get_perf_limited, buf);
>> +}
>> +
>> +/**
>> + * store_perf_limited - Clear Performance Limited register bits
>> + * @policy: cpufreq policy
>> + * @buf: buffer containing the bitmask of bits to clear
>> + * @count: number of bytes in buf
>> + *
>> + * Write 1 to clear bit 0, 2 to clear bit 1, or 3 to clear both.
>> + */
>> +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);
>> cpufreq_freq_attr_rw(auto_act_window);
>> cpufreq_freq_attr_rw(energy_performance_preference_val);
>> cpufreq_freq_attr_rw(min_perf);
>> cpufreq_freq_attr_rw(max_perf);
>> +cpufreq_freq_attr_rw(perf_limited);
>>
>> static struct freq_attr *cppc_cpufreq_attr[] = {
>> &freqdomain_cpus,
>> @@ -1066,6 +1096,7 @@ static struct freq_attr *cppc_cpufreq_attr[] = {
>> &energy_performance_preference_val,
>> &min_perf,
>> &max_perf,
>> + &perf_limited,
>> NULL,
>> };
>>
>> diff --git a/include/acpi/cppc_acpi.h b/include/acpi/cppc_acpi.h
>> index a49b50bddaf9..57e04326a4b6 100644
>> --- a/include/acpi/cppc_acpi.h
>> +++ b/include/acpi/cppc_acpi.h
>> @@ -42,6 +42,11 @@
>> #define CPPC_EPP_PERFORMANCE_PREF 0x00
>> #define CPPC_EPP_ENERGY_EFFICIENCY_PREF 0xFF
>>
>> +#define CPPC_PERF_LIMITED_DESIRED_EXCURSION BIT(0)
>> +#define CPPC_PERF_LIMITED_MINIMUM_EXCURSION BIT(1)
>> +#define CPPC_PERF_LIMITED_MASK (CPPC_PERF_LIMITED_DESIRED_EXCURSION | \
>> + CPPC_PERF_LIMITED_MINIMUM_EXCURSION)
>> +
>> /* Each register has the folowing format. */
>> struct cpc_reg {
>> u8 descriptor;
>> @@ -177,6 +182,8 @@ extern int cppc_get_min_perf(int cpu, u64 *min_perf);
>> extern int cppc_set_min_perf(int cpu, u32 min_perf);
>> extern int cppc_get_max_perf(int cpu, u64 *max_perf);
>> extern int cppc_set_max_perf(int cpu, u32 max_perf);
>> +extern int cppc_get_perf_limited(int cpu, u64 *perf_limited);
>> +extern int cppc_set_perf_limited(int cpu, u64 perf_limited);
>> extern int amd_get_highest_perf(unsigned int cpu, u32 *highest_perf);
>> extern int amd_get_boost_ratio_numerator(unsigned int cpu, u64 *numerator);
>> extern int amd_detect_prefcore(bool *detected);
>> @@ -285,6 +292,14 @@ static inline int cppc_set_max_perf(int cpu, u32 max_perf)
>> {
>> return -EOPNOTSUPP;
>> }
>> +static inline int cppc_get_perf_limited(int cpu, u64 *perf_limited)
>> +{
>> + return -EOPNOTSUPP;
>> +}
>> +static inline int cppc_set_perf_limited(int cpu, u64 perf_limited)
>> +{
>> + return -EOPNOTSUPP;
>> +}
>> static inline int amd_get_highest_perf(unsigned int cpu, u32 *highest_perf)
>> {
>> return -ENODEV;
Powered by blists - more mailing lists