Skip to contentSkip to conversation

Add support for _TFP and configurable throttle pctg

PATCH Superseded Sumit Gupta 2 participants 9 messages Aug 17, 2023 – Aug 24, 2023

Subsystems (2)

v1 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

cover Add support for _TFP and configurable throttle pctg

This patch set adds support for two features to get a finer control over the impact of Thermal Throttling on performance.

1) Patch 1: Adds support to read Thermal fast Sampling Period (_TFP) ACPI object and use it over "Thermal Sampling Period (_TSP)" for Passive cooling if both are present.

2) Patch 2: Adds support to configure the CPUFREQ reduction percentage and not always cause throttling in steps of "20%".

Both patches can be applied independently.

Jeff Brasen (1):
ACPI: thermal: Add Thermal fast Sampling Period (_TFP) support

Srikar Srimath Tirumala (1):
ACPI: processor: Add support to configure CPUFREQ reduction pctg

 drivers/acpi/Kconfig             | 15 +++++++++++++++
 drivers/acpi/processor_thermal.c | 19 ++++++++++++++++---
 drivers/acpi/thermal.c           | 21 ++++++++++++++-------
 3 files changed, 45 insertions(+), 10 deletions(-)
Signature · 1 line
2.17.1
#1 lore raw

PATCH 1/2 ACPI: thermal: Add Thermal fast Sampling Period (_TFP) support

From: Jeff Brasen <[email protected]>

Add support for Thermal fast Sampling Period (_TFP) for Passive cooling. As per UEFI spec, _TFP overrides the "Thermal Sampling Period (_TSP)" if both are present in a Thermal zone.

---
 drivers/acpi/thermal.c | 21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)
drivers/acpi/thermal.c +14 -7
index f9f6ebb08fdb..5dee3722509c 100644
@@ -111,7 +111,7 @@ struct acpi_thermal_passive {
 	unsigned long temperature;
 	unsigned long tc1;
 	unsigned long tc2;
-	unsigned long tsp;
+	unsigned long sampling_period;
 	bool valid;
 };
 
@@ -289,11 +289,18 @@ static int acpi_thermal_trips_update(struct acpi_thermal *tz, int flag)
 					tz->trips.passive.tc2 = tmp;
 
 				status = acpi_evaluate_integer(tz->device->handle,
-							       "_TSP", NULL, &tmp);
-				if (ACPI_FAILURE(status))
-					tz->trips.passive.valid = false;
-				else
-					tz->trips.passive.tsp = tmp;
+							       "_TFP", NULL, &tmp);
+				if (ACPI_FAILURE(status)) {
+					status = acpi_evaluate_integer(tz->device->handle,
+								       "_TSP", NULL, &tmp);
+					if (ACPI_FAILURE(status))
+						tz->trips.passive.valid = false;
+					else
+						tz->trips.passive.sampling_period = tmp * 100;
+
+				} else {
+					tz->trips.passive.sampling_period = tmp;
+				}
 			}
 		}
 	}
@@ -765,7 +772,7 @@ static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz)
 	if (tz->trips.passive.valid)
 		tz->thermal_zone = thermal_zone_device_register("acpitz", trips, 0, tz,
 								&acpi_thermal_zone_ops, NULL,
-								tz->trips.passive.tsp * 100,
+								tz->trips.passive.sampling_period,
 								tz->polling_frequency * 100);
 	else
 		tz->thermal_zone =
Signature · 1 line
2.17.1
#2 lore raw

Re: [Patch 1/2] ACPI: thermal: Add Thermal fast Sampling Period (_TFP) support

Quoted 5 lines · Sumit Gupta

From: Jeff Brasen <[email protected]>

Add support for Thermal fast Sampling Period (_TFP) for Passive cooling. As per UEFI spec,

You mean the ACPI spec I suppose? It would be good to give the relevant section number and title.

Quoted 47 lines

_TFP overrides the "Thermal Sampling Period (_TSP)" if both are present in a Thermal zone.

---
 drivers/acpi/thermal.c | 21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)
drivers/acpi/thermal.c +13 -6
index f9f6ebb08fdb..5dee3722509c 100644
@@ -111,7 +111,7 @@ struct acpi_thermal_passive {
        unsigned long temperature;
        unsigned long tc1;
        unsigned long tc2;
-       unsigned long tsp;
+       unsigned long sampling_period;
        bool valid;
 };
 
@@ -289,11 +289,18 @@ static int acpi_thermal_trips_update(struct acpi_thermal *tz, int flag)
                                        tz->trips.passive.tc2 = tmp;
 
                                status = acpi_evaluate_integer(tz->device->handle,
-                                                              "_TSP", NULL, &tmp);
-                               if (ACPI_FAILURE(status))
-                                       tz->trips.passive.valid = false;
-                               else
-                                       tz->trips.passive.tsp = tmp;
+                                                              "_TFP", NULL, &tmp);
+                               if (ACPI_FAILURE(status)) {
+                                       status = acpi_evaluate_integer(tz->device->handle,
+                                                                      "_TSP", NULL, &tmp);
+                                       if (ACPI_FAILURE(status))
+                                               tz->trips.passive.valid = false;
+                                       else
+                                               tz->trips.passive.sampling_period = tmp * 100;
+
+                               } else {
+                                       tz->trips.passive.sampling_period = tmp;
+                               }
                        }
                }
        }

drivers/acpi/thermal.c static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz)

Quoted 9 lines
drivers/acpi/thermal.c +1 -2
@@ -765,7 +772,7 @@ static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz)
        if (tz->trips.passive.valid)
                tz->thermal_zone = thermal_zone_device_register("acpitz", trips, 0, tz,
                                                                &acpi_thermal_zone_ops, NULL,
-                                                               tz->trips.passive.tsp * 100,
+                                                               tz->trips.passive.sampling_period,
                                                                tz->polling_frequency * 100);
        else
                tz->thermal_zone =
--

So this needs to be rebased on top of the current linux-next branch in linux-pm.git or on top of the acpi-thermal branch in there.

#3 lore raw

Re: [Patch 1/2] ACPI: thermal: Add Thermal fast Sampling Period (_TFP) support

Quoted 5 lines · Rafael J. Wysocki

External email: Use caution opening links or attachments

Quoted 1 line · Sumit Gupta
Quoted 8 lines
Quoted 4 lines

From: Jeff Brasen <[email protected]>

Add support for Thermal fast Sampling Period (_TFP) for Passive cooling. As per UEFI spec,

You mean the ACPI spec I suppose? It would be good to give the relevant section number and title.

Yes, It's the ACPI Spec 6.4. Section and title are "11.4.17. _TFP (Thermal fast Sampling Period)". I will add it in description in v2. Thank you for correcting.

Quoted 51 lines
Quoted 51 lines

_TFP overrides the "Thermal Sampling Period (_TSP)" if both are present in a Thermal zone.

---
  drivers/acpi/thermal.c | 21 ++++++++++++++-------
  1 file changed, 14 insertions(+), 7 deletions(-)
drivers/acpi/thermal.c +13 -7
index f9f6ebb08fdb..5dee3722509c 100644
@@ -111,7 +111,7 @@ struct acpi_thermal_passive {
         unsigned long temperature;
         unsigned long tc1;
         unsigned long tc2;
-       unsigned long tsp;
+       unsigned long sampling_period;
         bool valid;
  };
 
@@ -289,11 +289,18 @@ static int acpi_thermal_trips_update(struct acpi_thermal *tz, int flag)
                                         tz->trips.passive.tc2 = tmp;
 
                                 status = acpi_evaluate_integer(tz->device->handle,
-                                                              "_TSP", NULL, &tmp);
-                               if (ACPI_FAILURE(status))
-                                       tz->trips.passive.valid = false;
-                               else
-                                       tz->trips.passive.tsp = tmp;
+                                                              "_TFP", NULL, &tmp);
+                               if (ACPI_FAILURE(status)) {
+                                       status = acpi_evaluate_integer(tz->device->handle,
+                                                                      "_TSP", NULL, &tmp);
+                                       if (ACPI_FAILURE(status))
+                                               tz->trips.passive.valid = false;
+                                       else
+                                               tz->trips.passive.sampling_period = tmp * 100;
+
+                               } else {
+                                       tz->trips.passive.sampling_period = tmp;
+                               }
                         }
                 }
         }
@@ -765,7 +772,7 @@ static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz)
         if (tz->trips.passive.valid)
                 tz->thermal_zone = thermal_zone_device_register("acpitz", trips, 0, tz,
                                                                 &acpi_thermal_zone_ops, NULL,
-                                                               tz->trips.passive.tsp * 100,
Quoted 8 lines

drivers/acpi/thermal.c static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz)

Quoted 5 lines
drivers/acpi/thermal.c +1 -1
+                                                               tz->trips.passive.sampling_period,
                                                                 tz->polling_frequency * 100);
         else
                 tz->thermal_zone =
--

So this needs to be rebased on top of the current linux-next branch in linux-pm.git or on top of the acpi-thermal branch in there.

Will rebase and send v2.

Thank you,
Sumit Gupta

#4 lore raw

PATCH 2/2 ACPI: processor: Add support to configure CPUFREQ reduction pctg

In reply to Sumit Gupta

From: Srikar Srimath Tirumala <[email protected]>

Add support to configure the CPUFREQ reduction percentage and set the maximum number of throttling steps accordingly. Current implementation of processor_thermal performs software throttling in fixed steps of "20%" which can be too coarse for some platforms. Change that by adding new config to provide the reduction percentage.

---
 drivers/acpi/Kconfig             | 15 +++++++++++++++
 drivers/acpi/processor_thermal.c | 19 ++++++++++++++++---
 2 files changed, 31 insertions(+), 3 deletions(-)
drivers/acpi/Kconfig +15 -0
index 00dd309b6682..287cf58defbf 100644
@@ -254,6 +254,21 @@ config ACPI_DOCK
 config ACPI_CPU_FREQ_PSS
 	bool
 
+config ACPI_CPU_FREQ_THERM_HAS_PARAMS
+	bool "CPU frequency throttling control"
+	depends on ACPI_PROCESSOR
+
+config ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG
+	int "Minimum throttle percentage for processor_thermal cooling device"
+	depends on ACPI_CPU_FREQ_THERM_HAS_PARAMS
+	default 20
+	help
+	  The processor_thermal driver uses this config to calculate the
+	  percentage amount by which cpu frequency must be reduced for each
+	  cooling state. The config is also used to calculate the maximum number
+	  of throttling steps or cooling states supported by the driver. Value
+	  must be an unsigned integer in the range [1, 50].
+
 config ACPI_PROCESSOR_CSTATE
 	def_bool y
 	depends on ACPI_PROCESSOR
drivers/acpi/processor_thermal.c +16 -3
index b7c6287eccca..ee443cc69b73 100644
@@ -25,8 +25,20 @@
  * _any_ cpufreq driver and not only the acpi-cpufreq driver.
  */
 
-#define CPUFREQ_THERMAL_MIN_STEP 0
-#define CPUFREQ_THERMAL_MAX_STEP 3
+#define CPUFREQ_THERMAL_MIN_STEP	0
+#ifdef CONFIG_ACPI_CPU_FREQ_THERM_HAS_PARAMS
+#define CPUFREQ_THERMAL_PCTG		CONFIG_ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG
+
+/* Derive the MAX_STEP from minimum throttle percentage so that the reduction
+ * percentage does end up becoming negative. Also cap the MAX_STEP so that
+ * the CPU performance doesn't become 0.
+ */
+#define CPUFREQ_THERMAL_MAX_STEP	((100 / CPUFREQ_THERMAL_PCTG) - 1)
+
+#else
+#define CPUFREQ_THERMAL_MAX_STEP	3
+#define CPUFREQ_THERMAL_PCTG		20
+#endif
 
 static DEFINE_PER_CPU(unsigned int, cpufreq_thermal_reduction_pctg);
 
@@ -113,7 +125,8 @@ static int cpufreq_set_cur_state(unsigned int cpu, int state)
 		if (!policy)
 			return -EINVAL;
 
-		max_freq = (policy->cpuinfo.max_freq * (100 - reduction_pctg(i) * 20)) / 100;
+		max_freq = (policy->cpuinfo.max_freq *
+			    (100 - reduction_pctg(i) * CPUFREQ_THERMAL_PCTG)) / 100;
 
 		cpufreq_cpu_put(policy);
 
Signature · 1 line
2.17.1
#5 lore raw

Re: [Patch 2/2] ACPI: processor: Add support to configure CPUFREQ reduction pctg

Quoted 29 lines · Sumit Gupta

From: Srikar Srimath Tirumala <[email protected]>

Add support to configure the CPUFREQ reduction percentage and set the maximum number of throttling steps accordingly. Current implementation of processor_thermal performs software throttling in fixed steps of "20%" which can be too coarse for some platforms. Change that by adding new config to provide the reduction percentage.

---
 drivers/acpi/Kconfig             | 15 +++++++++++++++
 drivers/acpi/processor_thermal.c | 19 ++++++++++++++++---
 2 files changed, 31 insertions(+), 3 deletions(-)
drivers/acpi/Kconfig +5 -0
index 00dd309b6682..287cf58defbf 100644
@@ -254,6 +254,21 @@ config ACPI_DOCK
 config ACPI_CPU_FREQ_PSS
        bool
 
+config ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       bool "CPU frequency throttling control"
+       depends on ACPI_PROCESSOR
+
+config ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG

drivers/acpi/Kconfig config ACPI_DOCK · line 271

Quoted 10 lines
drivers/acpi/Kconfig +10 -0
+       int "Minimum throttle percentage for processor_thermal cooling device"
+       depends on ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       default 20
+       help
+         The processor_thermal driver uses this config to calculate the
+         percentage amount by which cpu frequency must be reduced for each
+         cooling state. The config is also used to calculate the maximum number
+         of throttling steps or cooling states supported by the driver. Value
+         must be an unsigned integer in the range [1, 50].
+

I don't think that the new Kconfig symbols are particularly useful. At least they don't help the distro vendors that each would need to pick up a specific value for their kernel anyway.

I also wonder how the users building their own kernels are supposed to determine the values suitable for the target systems.

Quoted 41 lines
drivers/acpi/Kconfig +0 -0
 config ACPI_PROCESSOR_CSTATE
        def_bool y
        depends on ACPI_PROCESSOR
drivers/acpi/processor_thermal.c +16 -4
index b7c6287eccca..ee443cc69b73 100644
@@ -25,8 +25,20 @@
  * _any_ cpufreq driver and not only the acpi-cpufreq driver.
  */
 
-#define CPUFREQ_THERMAL_MIN_STEP 0
-#define CPUFREQ_THERMAL_MAX_STEP 3
+#define CPUFREQ_THERMAL_MIN_STEP       0
+#ifdef CONFIG_ACPI_CPU_FREQ_THERM_HAS_PARAMS
+#define CPUFREQ_THERMAL_PCTG           CONFIG_ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG
+
+/* Derive the MAX_STEP from minimum throttle percentage so that the reduction
+ * percentage does end up becoming negative. Also cap the MAX_STEP so that
+ * the CPU performance doesn't become 0.
+ */
+#define CPUFREQ_THERMAL_MAX_STEP       ((100 / CPUFREQ_THERMAL_PCTG) - 1)
+
+#else
+#define CPUFREQ_THERMAL_MAX_STEP       3
+#define CPUFREQ_THERMAL_PCTG           20
+#endif
 
 static DEFINE_PER_CPU(unsigned int, cpufreq_thermal_reduction_pctg);
 
@@ -113,7 +125,8 @@ static int cpufreq_set_cur_state(unsigned int cpu, int state)
                if (!policy)
                        return -EINVAL;
 
-               max_freq = (policy->cpuinfo.max_freq * (100 - reduction_pctg(i) * 20)) / 100;
+               max_freq = (policy->cpuinfo.max_freq *
+                           (100 - reduction_pctg(i) * CPUFREQ_THERMAL_PCTG)) / 100;
 
                cpufreq_cpu_put(policy);
 
--
#6 lore raw

Re: [Patch 2/2] ACPI: processor: Add support to configure CPUFREQ reduction pctg

Quoted 48 lines · Rafael J. Wysocki

External email: Use caution opening links or attachments

Quoted 29 lines · Sumit Gupta

From: Srikar Srimath Tirumala <[email protected]>

Add support to configure the CPUFREQ reduction percentage and set the maximum number of throttling steps accordingly. Current implementation of processor_thermal performs software throttling in fixed steps of "20%" which can be too coarse for some platforms. Change that by adding new config to provide the reduction percentage.

---
  drivers/acpi/Kconfig             | 15 +++++++++++++++
  drivers/acpi/processor_thermal.c | 19 ++++++++++++++++---
  2 files changed, 31 insertions(+), 3 deletions(-)
drivers/acpi/Kconfig +5 -0
index 00dd309b6682..287cf58defbf 100644
@@ -254,6 +254,21 @@ config ACPI_DOCK
  config ACPI_CPU_FREQ_PSS
         bool
 
+config ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       bool "CPU frequency throttling control"
+       depends on ACPI_PROCESSOR
+
+config ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG

drivers/acpi/Kconfig config ACPI_DOCK · line 271

Quoted 10 lines
drivers/acpi/Kconfig +10 -0
+       int "Minimum throttle percentage for processor_thermal cooling device"
+       depends on ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       default 20
+       help
+         The processor_thermal driver uses this config to calculate the
+         percentage amount by which cpu frequency must be reduced for each
+         cooling state. The config is also used to calculate the maximum number
+         of throttling steps or cooling states supported by the driver. Value
+         must be an unsigned integer in the range [1, 50].
+

I don't think that the new Kconfig symbols are particularly useful. At least they don't help the distro vendors that each would need to pick up a specific value for their kernel anyway.

Quoted 3 lines

I also wonder how the users building their own kernels are supposed to determine the values suitable for the target systems.

We observed some perf gain after reducing the throttle percentage. Currently, kept the default to '20%' as before.

Based on need, a vendor can overwrite the default value with macro 'CONFIG_ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG'. Otherwise, the behavior will remain same.

Thank you,
Sumit Gupta

Quoted 41 lines
Quoted 41 lines
drivers/acpi/Kconfig +0 -0
  config ACPI_PROCESSOR_CSTATE
         def_bool y
         depends on ACPI_PROCESSOR
drivers/acpi/processor_thermal.c +16 -4
index b7c6287eccca..ee443cc69b73 100644
@@ -25,8 +25,20 @@
   * _any_ cpufreq driver and not only the acpi-cpufreq driver.
   */
 
-#define CPUFREQ_THERMAL_MIN_STEP 0
-#define CPUFREQ_THERMAL_MAX_STEP 3
+#define CPUFREQ_THERMAL_MIN_STEP       0
+#ifdef CONFIG_ACPI_CPU_FREQ_THERM_HAS_PARAMS
+#define CPUFREQ_THERMAL_PCTG           CONFIG_ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG
+
+/* Derive the MAX_STEP from minimum throttle percentage so that the reduction
+ * percentage does end up becoming negative. Also cap the MAX_STEP so that
+ * the CPU performance doesn't become 0.
+ */
+#define CPUFREQ_THERMAL_MAX_STEP       ((100 / CPUFREQ_THERMAL_PCTG) - 1)
+
+#else
+#define CPUFREQ_THERMAL_MAX_STEP       3
+#define CPUFREQ_THERMAL_PCTG           20
+#endif
 
  static DEFINE_PER_CPU(unsigned int, cpufreq_thermal_reduction_pctg);
 
@@ -113,7 +125,8 @@ static int cpufreq_set_cur_state(unsigned int cpu, int state)
                 if (!policy)
                         return -EINVAL;
 
-               max_freq = (policy->cpuinfo.max_freq * (100 - reduction_pctg(i) * 20)) / 100;
+               max_freq = (policy->cpuinfo.max_freq *
+                           (100 - reduction_pctg(i) * CPUFREQ_THERMAL_PCTG)) / 100;
 
                 cpufreq_cpu_put(policy);
 
--
#7 lore raw

Re: [Patch 2/2] ACPI: processor: Add support to configure CPUFREQ reduction pctg

Quoted 52 lines · Sumit Gupta
Quoted 48 lines · Rafael J. Wysocki

External email: Use caution opening links or attachments

Quoted 29 lines · Sumit Gupta

From: Srikar Srimath Tirumala <[email protected]>

Add support to configure the CPUFREQ reduction percentage and set the maximum number of throttling steps accordingly. Current implementation of processor_thermal performs software throttling in fixed steps of "20%" which can be too coarse for some platforms. Change that by adding new config to provide the reduction percentage.

---
  drivers/acpi/Kconfig             | 15 +++++++++++++++
  drivers/acpi/processor_thermal.c | 19 ++++++++++++++++---
  2 files changed, 31 insertions(+), 3 deletions(-)
drivers/acpi/Kconfig +5 -0
index 00dd309b6682..287cf58defbf 100644
@@ -254,6 +254,21 @@ config ACPI_DOCK
  config ACPI_CPU_FREQ_PSS
         bool
 
+config ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       bool "CPU frequency throttling control"
+       depends on ACPI_PROCESSOR
+
+config ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG

drivers/acpi/Kconfig config ACPI_DOCK · line 271

Quoted 10 lines
drivers/acpi/Kconfig +10 -0
+       int "Minimum throttle percentage for processor_thermal cooling device"
+       depends on ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       default 20
+       help
+         The processor_thermal driver uses this config to calculate the
+         percentage amount by which cpu frequency must be reduced for each
+         cooling state. The config is also used to calculate the maximum number
+         of throttling steps or cooling states supported by the driver. Value
+         must be an unsigned integer in the range [1, 50].
+

I don't think that the new Kconfig symbols are particularly useful. At least they don't help the distro vendors that each would need to pick up a specific value for their kernel anyway.

Quoted 6 lines
Quoted 3 lines

I also wonder how the users building their own kernels are supposed to determine the values suitable for the target systems.

We observed some perf gain after reducing the throttle percentage. Currently, kept the default to '20%' as before.

So you should add this information to the patch changelog, ideally along with the description of the hardware configuration in which the improvement has been observed.

Quoted 3 lines

Based on need, a vendor can overwrite the default value with macro 'CONFIG_ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG'. Otherwise, the behavior will remain same.

Yes, that's how it works.

What I'm saying is that the way it works does not appear to be particularly useful.

For example, how exactly is a distribution supposed to guess the "right" value for their general-purpose kernel?

#8 lore raw

Re: [Patch 2/2] ACPI: processor: Add support to configure CPUFREQ reduction pctg

Quoted 47 lines
Quoted 47 lines
Quoted 46 lines
Quoted 28 lines

Add support to configure the CPUFREQ reduction percentage and set the maximum number of throttling steps accordingly. Current implementation of processor_thermal performs software throttling in fixed steps of "20%" which can be too coarse for some platforms. Change that by adding new config to provide the reduction percentage.

---
   drivers/acpi/Kconfig             | 15 +++++++++++++++
   drivers/acpi/processor_thermal.c | 19 ++++++++++++++++---
   2 files changed, 31 insertions(+), 3 deletions(-)
drivers/acpi/Kconfig +5 -0
index 00dd309b6682..287cf58defbf 100644
@@ -254,6 +254,21 @@ config ACPI_DOCK
   config ACPI_CPU_FREQ_PSS
          bool
 
+config ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       bool "CPU frequency throttling control"
+       depends on ACPI_PROCESSOR
+
+config ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG

drivers/acpi/Kconfig config ACPI_DOCK · line 271

Quoted 10 lines
drivers/acpi/Kconfig +10 -0
+       int "Minimum throttle percentage for processor_thermal cooling device"
+       depends on ACPI_CPU_FREQ_THERM_HAS_PARAMS
+       default 20
+       help
+         The processor_thermal driver uses this config to calculate the
+         percentage amount by which cpu frequency must be reduced for each
+         cooling state. The config is also used to calculate the maximum number
+         of throttling steps or cooling states supported by the driver. Value
+         must be an unsigned integer in the range [1, 50].
+

I don't think that the new Kconfig symbols are particularly useful. At least they don't help the distro vendors that each would need to pick up a specific value for their kernel anyway.

I also wonder how the users building their own kernels are supposed to determine the values suitable for the target systems.

Quoted 7 lines
Quoted 2 lines

We observed some perf gain after reducing the throttle percentage. Currently, kept the default to '20%' as before.

So you should add this information to the patch changelog, ideally along with the description of the hardware configuration in which the improvement has been observed.

Sure, will add in v2.

Quoted 11 lines
Quoted 3 lines

Based on need, a vendor can overwrite the default value with macro 'CONFIG_ACPI_CPU_FREQ_THERM_MIN_THROT_PCTG'. Otherwise, the behavior will remain same.

Yes, that's how it works.

What I'm saying is that the way it works does not appear to be particularly useful.

For example, how exactly is a distribution supposed to guess the "right" value for their general-purpose kernel?

We tested on Tegra241 (Grace) SoC with "5%" throttle percentage.
Didn't change the default value as behavior could be different on other chips.
An alternate way could be to overwrite the default value for the specific SoC using "arm_smccc_get_soc_id_version()" check. But not sure if such change in the generic "processor_thermal" driver is Okay? Please suggest if the above sounds fine or any better way?

Thank you,
Sumit Gupta