[PATCH v7 0/7] consolidate and cleanup CPU capacity

[PATCH v7 0/7] consolidate and cleanup CPU capacity

[Vincent Guittot]

This is the 1st part of consolidating how the max compute capacity is
used in the scheduler and how we calculate the frequency for a level of
utilization.

Fix some unconsistancy when computing frequency for an utilization. There
can be a mismatch between energy model and schedutil.

Next step will be to make a difference between the original
max compute capacity of a CPU and what is currently available when
there is a capping applying forever (i.e. seconds or more).

Changes since v6:
- Rebase on tip/sched/core and fix a conflict
- Add tags

Changes since v5:
- Remove useless return in freq_inv_set_max_ratio()
- Add tags

Changes since v4:
- Capitalize the verb in subject
- Remove usless parentheses in cppc_get_dmi_max_khz()
- Use freq_ref pattern everywhere
- Fix MHz / kHz units conversion for cppc_cpufreq
- Move default definition of arch_scale_freq_ref() in
  include/linux/sched/topology.h beside arch_scale_cpu_capacity
  which faces similar default declaration behavior. This location covers
  all cases with arch and CONFIG_* which was not the case with previous
  attempts.

Changes since v3:
- Split patch 5 cpufreq/cppc
- Fix topology_init_cpu_capacity_cppc() 
- Fix init if AMU ratio
- Added some tags

Changes since v2:
- Remove the 1st patch which has been queued in tip
- Rework how to initialize the reference frequency for cppc_cpufreq and
  change topology_init_cpu_capacity_cppc() to also set capacity_ref_freq
- Add a RFC to convert AMU to use arch_scale_freq_ref and move the config
  of the AMU ratio to be done when intializing cpu capacity and
  capacity_ref_freq
- Added some tags

Changes since v1:
- Fix typos
- Added changes in cpufreq to use arch_scale_freq_ref() when calling
  arch_set_freq_scale (patch 3).
- arch_scale_freq_ref() is always defined and returns 0 (as proposed
  by Ionela) when not defined by the arch. This simplifies the code with
  the addition of patch 3.
- Simplify Energy Model which always uses arch_scale_freq_ref(). The
  latter returns 0 when not defined by arch instead of last item of the 
  perf domain. This is not a problem because the function is only defined
  for compilation purpose in this case and we don't care about the
  returned value. (patch 5)
- Added changes in cppc cpufreq to set capacity_ref_freq (patch 6)
- Added reviewed tag for patch 1 which got a minor change but not for
  others as I did some changes which could make previous reviewed tag
  no more relevant.

Vincent Guittot (7):
  topology: Add a new arch_scale_freq_reference
  cpufreq: Use the fixed and coherent frequency for scaling capacity
  cpufreq/schedutil: Use a fixed reference frequency
  energy_model: Use a fixed reference frequency
  cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}
  cpufreq/cppc: Set the frequency used for computing the capacity
  arm64/amu: Use capacity_ref_freq to set AMU ratio

 arch/arm/include/asm/topology.h   |   1 +
 arch/arm64/include/asm/topology.h |   1 +
 arch/arm64/kernel/topology.c      |  26 +++---
 arch/riscv/include/asm/topology.h |   1 +
 drivers/acpi/cppc_acpi.c          | 104 ++++++++++++++++++++++
 drivers/base/arch_topology.c      |  56 ++++++++----
 drivers/cpufreq/cppc_cpufreq.c    | 139 ++++--------------------------
 drivers/cpufreq/cpufreq.c         |   4 +-
 include/acpi/cppc_acpi.h          |   2 +
 include/linux/arch_topology.h     |   8 ++
 include/linux/cpufreq.h           |   1 +
 include/linux/energy_model.h      |   6 +-
 include/linux/sched/topology.h    |   8 ++
 kernel/sched/cpufreq_schedutil.c  |  26 +++++-
 14 files changed, 224 insertions(+), 159 deletions(-)

-- 
2.34.1

[PATCH v7 1/7] topology: Add a new arch_scale_freq_reference

[Vincent Guittot]

Create a new method to get a unique and fixed max frequency. Currently
cpuinfo.max_freq or the highest (or last) state of performance domain are
used as the max frequency when computing the frequency for a level of
utilization but:
- cpuinfo_max_freq can change at runtime. boost is one example of
  such change.
- cpuinfo.max_freq and last item of the PD can be different leading to
  different results between cpufreq and energy model.

We need to save the reference frequency that has been used when computing
the CPUs capacity and use this fixed and coherent value to convert between
frequency and CPU's capacity.

In fact, we already save the frequency that has been used when computing
the capacity of each CPU. We extend the precision to save kHz instead of
MHz currently and we modify the type to be aligned with other variables
used when converting frequency to capacity and the other way.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
---
 arch/arm/include/asm/topology.h   |  1 +
 arch/arm64/include/asm/topology.h |  1 +
 arch/riscv/include/asm/topology.h |  1 +
 drivers/base/arch_topology.c      | 29 ++++++++++++++---------------
 include/linux/arch_topology.h     |  7 +++++++
 include/linux/sched/topology.h    |  8 ++++++++
 6 files changed, 32 insertions(+), 15 deletions(-)

diff --git a/arch/arm/include/asm/topology.h b/arch/arm/include/asm/topology.h
index c7d2510e5a78..853c4f81ba4a 100644
--- a/arch/arm/include/asm/topology.h
+++ b/arch/arm/include/asm/topology.h
@@ -13,6 +13,7 @@
 #define arch_set_freq_scale topology_set_freq_scale
 #define arch_scale_freq_capacity topology_get_freq_scale
 #define arch_scale_freq_invariant topology_scale_freq_invariant
+#define arch_scale_freq_ref topology_get_freq_ref
 #endif
 
 /* Replace task scheduler's default cpu-invariant accounting */
diff --git a/arch/arm64/include/asm/topology.h b/arch/arm64/include/asm/topology.h
index 9fab663dd2de..a323b109b9c4 100644
--- a/arch/arm64/include/asm/topology.h
+++ b/arch/arm64/include/asm/topology.h
@@ -23,6 +23,7 @@ void update_freq_counters_refs(void);
 #define arch_set_freq_scale topology_set_freq_scale
 #define arch_scale_freq_capacity topology_get_freq_scale
 #define arch_scale_freq_invariant topology_scale_freq_invariant
+#define arch_scale_freq_ref topology_get_freq_ref
 
 #ifdef CONFIG_ACPI_CPPC_LIB
 #define arch_init_invariance_cppc topology_init_cpu_capacity_cppc
diff --git a/arch/riscv/include/asm/topology.h b/arch/riscv/include/asm/topology.h
index e316ab3b77f3..61183688bdd5 100644
--- a/arch/riscv/include/asm/topology.h
+++ b/arch/riscv/include/asm/topology.h
@@ -9,6 +9,7 @@
 #define arch_set_freq_scale		topology_set_freq_scale
 #define arch_scale_freq_capacity	topology_get_freq_scale
 #define arch_scale_freq_invariant	topology_scale_freq_invariant
+#define arch_scale_freq_ref		topology_get_freq_ref
 
 /* Replace task scheduler's default cpu-invariant accounting */
 #define arch_scale_cpu_capacity	topology_get_cpu_scale
diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
index b741b5ba82bd..e8d1cdf1f761 100644
--- a/drivers/base/arch_topology.c
+++ b/drivers/base/arch_topology.c
@@ -19,6 +19,7 @@
 #include <linux/init.h>
 #include <linux/rcupdate.h>
 #include <linux/sched.h>
+#include <linux/units.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/thermal_pressure.h>
@@ -26,7 +27,8 @@
 static DEFINE_PER_CPU(struct scale_freq_data __rcu *, sft_data);
 static struct cpumask scale_freq_counters_mask;
 static bool scale_freq_invariant;
-static DEFINE_PER_CPU(u32, freq_factor) = 1;
+DEFINE_PER_CPU(unsigned long, capacity_freq_ref) = 1;
+EXPORT_PER_CPU_SYMBOL_GPL(capacity_freq_ref);
 
 static bool supports_scale_freq_counters(const struct cpumask *cpus)
 {
@@ -170,9 +172,9 @@ DEFINE_PER_CPU(unsigned long, thermal_pressure);
  * operating on stale data when hot-plug is used for some CPUs. The
  * @capped_freq reflects the currently allowed max CPUs frequency due to
  * thermal capping. It might be also a boost frequency value, which is bigger
- * than the internal 'freq_factor' max frequency. In such case the pressure
- * value should simply be removed, since this is an indication that there is
- * no thermal throttling. The @capped_freq must be provided in kHz.
+ * than the internal 'capacity_freq_ref' max frequency. In such case the
+ * pressure value should simply be removed, since this is an indication that
+ * there is no thermal throttling. The @capped_freq must be provided in kHz.
  */
 void topology_update_thermal_pressure(const struct cpumask *cpus,
 				      unsigned long capped_freq)
@@ -183,10 +185,7 @@ void topology_update_thermal_pressure(const struct cpumask *cpus,
 
 	cpu = cpumask_first(cpus);
 	max_capacity = arch_scale_cpu_capacity(cpu);
-	max_freq = per_cpu(freq_factor, cpu);
-
-	/* Convert to MHz scale which is used in 'freq_factor' */
-	capped_freq /= 1000;
+	max_freq = arch_scale_freq_ref(cpu);
 
 	/*
 	 * Handle properly the boost frequencies, which should simply clean
@@ -279,13 +278,13 @@ void topology_normalize_cpu_scale(void)
 
 	capacity_scale = 1;
 	for_each_possible_cpu(cpu) {
-		capacity = raw_capacity[cpu] * per_cpu(freq_factor, cpu);
+		capacity = raw_capacity[cpu] * per_cpu(capacity_freq_ref, cpu);
 		capacity_scale = max(capacity, capacity_scale);
 	}
 
 	pr_debug("cpu_capacity: capacity_scale=%llu\n", capacity_scale);
 	for_each_possible_cpu(cpu) {
-		capacity = raw_capacity[cpu] * per_cpu(freq_factor, cpu);
+		capacity = raw_capacity[cpu] * per_cpu(capacity_freq_ref, cpu);
 		capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT,
 			capacity_scale);
 		topology_set_cpu_scale(cpu, capacity);
@@ -321,15 +320,15 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu)
 			cpu_node, raw_capacity[cpu]);
 
 		/*
-		 * Update freq_factor for calculating early boot cpu capacities.
+		 * Update capacity_freq_ref for calculating early boot cpu capacities.
 		 * For non-clk CPU DVFS mechanism, there's no way to get the
 		 * frequency value now, assuming they are running at the same
-		 * frequency (by keeping the initial freq_factor value).
+		 * frequency (by keeping the initial capacity_freq_ref value).
 		 */
 		cpu_clk = of_clk_get(cpu_node, 0);
 		if (!PTR_ERR_OR_ZERO(cpu_clk)) {
-			per_cpu(freq_factor, cpu) =
-				clk_get_rate(cpu_clk) / 1000;
+			per_cpu(capacity_freq_ref, cpu) =
+				clk_get_rate(cpu_clk) / HZ_PER_KHZ;
 			clk_put(cpu_clk);
 		}
 	} else {
@@ -411,7 +410,7 @@ init_cpu_capacity_callback(struct notifier_block *nb,
 	cpumask_andnot(cpus_to_visit, cpus_to_visit, policy->related_cpus);
 
 	for_each_cpu(cpu, policy->related_cpus)
-		per_cpu(freq_factor, cpu) = policy->cpuinfo.max_freq / 1000;
+		per_cpu(capacity_freq_ref, cpu) = policy->cpuinfo.max_freq;
 
 	if (cpumask_empty(cpus_to_visit)) {
 		topology_normalize_cpu_scale();
diff --git a/include/linux/arch_topology.h b/include/linux/arch_topology.h
index a07b510e7dc5..32c24ff4f2a8 100644
--- a/include/linux/arch_topology.h
+++ b/include/linux/arch_topology.h
@@ -27,6 +27,13 @@ static inline unsigned long topology_get_cpu_scale(int cpu)
 
 void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity);
 
+DECLARE_PER_CPU(unsigned long, capacity_freq_ref);
+
+static inline unsigned long topology_get_freq_ref(int cpu)
+{
+	return per_cpu(capacity_freq_ref, cpu);
+}
+
 DECLARE_PER_CPU(unsigned long, arch_freq_scale);
 
 static inline unsigned long topology_get_freq_scale(int cpu)
diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h
index de545ba85218..a6e04b4a21d7 100644
--- a/include/linux/sched/topology.h
+++ b/include/linux/sched/topology.h
@@ -279,6 +279,14 @@ void arch_update_thermal_pressure(const struct cpumask *cpus,
 { }
 #endif
 
+#ifndef arch_scale_freq_ref
+static __always_inline
+unsigned int arch_scale_freq_ref(int cpu)
+{
+	return 0;
+}
+#endif
+
 static inline int task_node(const struct task_struct *p)
 {
 	return cpu_to_node(task_cpu(p));
-- 
2.34.1

[PATCH v7 2/7] cpufreq: Use the fixed and coherent frequency for scaling capacity

[Vincent Guittot]

cpuinfo.max_freq can change at runtime because of boost as an example. This
implies that the value could be different from the frequency that has been
used to compute the capacity of a CPU.

The new arch_scale_freq_ref() returns a fixed and coherent frequency
that can be used to compute the capacity for a given frequency.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 drivers/cpufreq/cpufreq.c | 4 ++--
 include/linux/cpufreq.h   | 1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 934d35f570b7..44db4f59c4cc 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -454,7 +454,7 @@ void cpufreq_freq_transition_end(struct cpufreq_policy *policy,
 
 	arch_set_freq_scale(policy->related_cpus,
 			    policy->cur,
-			    policy->cpuinfo.max_freq);
+			    arch_scale_freq_ref(policy->cpu));
 
 	spin_lock(&policy->transition_lock);
 	policy->transition_ongoing = false;
@@ -2174,7 +2174,7 @@ unsigned int cpufreq_driver_fast_switch(struct cpufreq_policy *policy,
 
 	policy->cur = freq;
 	arch_set_freq_scale(policy->related_cpus, freq,
-			    policy->cpuinfo.max_freq);
+			    arch_scale_freq_ref(policy->cpu));
 	cpufreq_stats_record_transition(policy, freq);
 
 	if (trace_cpu_frequency_enabled()) {
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 1c5ca92a0555..afda5f24d3dd 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -1203,6 +1203,7 @@ void arch_set_freq_scale(const struct cpumask *cpus,
 {
 }
 #endif
+
 /* the following are really really optional */
 extern struct freq_attr cpufreq_freq_attr_scaling_available_freqs;
 extern struct freq_attr cpufreq_freq_attr_scaling_boost_freqs;
-- 
2.34.1

[PATCH v7 3/7] cpufreq/schedutil: Use a fixed reference frequency

[Vincent Guittot]

cpuinfo.max_freq can change at runtime because of boost as an example. This
implies that the value could be different than the one that has been
used when computing the capacity of a CPU.

The new arch_scale_freq_ref() returns a fixed and coherent reference
frequency that can be used when computing a frequency based on utilization.

Use this arch_scale_freq_ref() when available and fallback to
policy otherwise.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 kernel/sched/cpufreq_schedutil.c | 26 ++++++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 4ee8ad70be99..95c3c097083e 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -114,6 +114,28 @@ static void sugov_deferred_update(struct sugov_policy *sg_policy)
 	}
 }
 
+/**
+ * get_capacity_ref_freq - get the reference frequency that has been used to
+ * correlate frequency and compute capacity for a given cpufreq policy. We use
+ * the CPU managing it for the arch_scale_freq_ref() call in the function.
+ * @policy: the cpufreq policy of the CPU in question.
+ *
+ * Return: the reference CPU frequency to compute a capacity.
+ */
+static __always_inline
+unsigned long get_capacity_ref_freq(struct cpufreq_policy *policy)
+{
+	unsigned int freq = arch_scale_freq_ref(policy->cpu);
+
+	if (freq)
+		return freq;
+
+	if (arch_scale_freq_invariant())
+		return policy->cpuinfo.max_freq;
+
+	return policy->cur;
+}
+
 /**
  * get_next_freq - Compute a new frequency for a given cpufreq policy.
  * @sg_policy: schedutil policy object to compute the new frequency for.
@@ -140,9 +162,9 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy,
 				  unsigned long util, unsigned long max)
 {
 	struct cpufreq_policy *policy = sg_policy->policy;
-	unsigned int freq = arch_scale_freq_invariant() ?
-				policy->cpuinfo.max_freq : policy->cur;
+	unsigned int freq;
 
+	freq = get_capacity_ref_freq(policy);
 	freq = map_util_freq(util, freq, max);
 
 	if (freq == sg_policy->cached_raw_freq && !sg_policy->need_freq_update)
-- 
2.34.1

[PATCH v7 4/7] energy_model: Use a fixed reference frequency

[Vincent Guittot]

The last item of a performance domain is not always the performance point
that has been used to compute CPU's capacity. This can lead to different
target frequency compared with other part of the system like schedutil and
would result in wrong energy estimation.

A new arch_scale_freq_ref() is available to return a fixed and coherent
frequency reference that can be used when computing the CPU's frequency
for an level of utilization. Use this function to get this reference
frequency.

Energy model is never used without defining arch_scale_freq_ref() but
can be compiled. Define a default arch_scale_freq_ref() returning 0
in such case.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
---
 include/linux/energy_model.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/energy_model.h b/include/linux/energy_model.h
index adec808b371a..88d91e087471 100644
--- a/include/linux/energy_model.h
+++ b/include/linux/energy_model.h
@@ -224,7 +224,7 @@ static inline unsigned long em_cpu_energy(struct em_perf_domain *pd,
 				unsigned long max_util, unsigned long sum_util,
 				unsigned long allowed_cpu_cap)
 {
-	unsigned long freq, scale_cpu;
+	unsigned long freq, ref_freq, scale_cpu;
 	struct em_perf_state *ps;
 	int cpu;
 
@@ -241,10 +241,10 @@ static inline unsigned long em_cpu_energy(struct em_perf_domain *pd,
 	 */
 	cpu = cpumask_first(to_cpumask(pd->cpus));
 	scale_cpu = arch_scale_cpu_capacity(cpu);
-	ps = &pd->table[pd->nr_perf_states - 1];
+	ref_freq = arch_scale_freq_ref(cpu);
 
 	max_util = min(max_util, allowed_cpu_cap);
-	freq = map_util_freq(max_util, ps->frequency, scale_cpu);
+	freq = map_util_freq(max_util, ref_freq, scale_cpu);
 
 	/*
 	 * Find the lowest performance state of the Energy Model above the
-- 
2.34.1

[PATCH v7 5/7] cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}

[Vincent Guittot]

Move and rename cppc_cpufreq_perf_to_khz and cppc_cpufreq_khz_to_perf to
use them outside cppc_cpufreq in topology_init_cpu_capacity_cppc().

Modify the interface to use struct cppc_perf_caps *caps instead of
struct cppc_cpudata *cpu_data as we only use the fields of cppc_perf_caps.

cppc_cpufreq was converting the lowest and nominal freq from MHz to kHz
before using them. We move this conversion inside cppc_perf_to_khz and
cppc_khz_to_perf to make them generic and usable outside cppc_cpufreq.

No functional change

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Tested-by: Pierre Gondois <pierre.gondois@arm.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/acpi/cppc_acpi.c       | 104 ++++++++++++++++++++++++
 drivers/cpufreq/cppc_cpufreq.c | 139 ++++-----------------------------
 include/acpi/cppc_acpi.h       |   2 +
 3 files changed, 123 insertions(+), 122 deletions(-)

diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c
index 7ff269a78c20..d155a86a8614 100644
--- a/drivers/acpi/cppc_acpi.c
+++ b/drivers/acpi/cppc_acpi.c
@@ -39,6 +39,9 @@
 #include <linux/rwsem.h>
 #include <linux/wait.h>
 #include <linux/topology.h>
+#include <linux/dmi.h>
+#include <linux/units.h>
+#include <asm/unaligned.h>
 
 #include <acpi/cppc_acpi.h>
 
@@ -1760,3 +1763,104 @@ unsigned int cppc_get_transition_latency(int cpu_num)
 	return latency_ns;
 }
 EXPORT_SYMBOL_GPL(cppc_get_transition_latency);
+
+/* Minimum struct length needed for the DMI processor entry we want */
+#define DMI_ENTRY_PROCESSOR_MIN_LENGTH	48
+
+/* Offset in the DMI processor structure for the max frequency */
+#define DMI_PROCESSOR_MAX_SPEED		0x14
+
+/* Callback function used to retrieve the max frequency from DMI */
+static void cppc_find_dmi_mhz(const struct dmi_header *dm, void *private)
+{
+	const u8 *dmi_data = (const u8 *)dm;
+	u16 *mhz = (u16 *)private;
+
+	if (dm->type == DMI_ENTRY_PROCESSOR &&
+	    dm->length >= DMI_ENTRY_PROCESSOR_MIN_LENGTH) {
+		u16 val = (u16)get_unaligned((const u16 *)
+				(dmi_data + DMI_PROCESSOR_MAX_SPEED));
+		*mhz = val > *mhz ? val : *mhz;
+	}
+}
+
+/* Look up the max frequency in DMI */
+static u64 cppc_get_dmi_max_khz(void)
+{
+	u16 mhz = 0;
+
+	dmi_walk(cppc_find_dmi_mhz, &mhz);
+
+	/*
+	 * Real stupid fallback value, just in case there is no
+	 * actual value set.
+	 */
+	mhz = mhz ? mhz : 1;
+
+	return KHZ_PER_MHZ * mhz;
+}
+
+/*
+ * If CPPC lowest_freq and nominal_freq registers are exposed then we can
+ * use them to convert perf to freq and vice versa. The conversion is
+ * extrapolated as an affine function passing by the 2 points:
+ *  - (Low perf, Low freq)
+ *  - (Nominal perf, Nominal freq)
+ */
+unsigned int cppc_perf_to_khz(struct cppc_perf_caps *caps, unsigned int perf)
+{
+	s64 retval, offset = 0;
+	static u64 max_khz;
+	u64 mul, div;
+
+	if (caps->lowest_freq && caps->nominal_freq) {
+		mul = caps->nominal_freq - caps->lowest_freq;
+		mul *= KHZ_PER_MHZ;
+		div = caps->nominal_perf - caps->lowest_perf;
+		offset = caps->nominal_freq * KHZ_PER_MHZ -
+			 div64_u64(caps->nominal_perf * mul, div);
+	} else {
+		if (!max_khz)
+			max_khz = cppc_get_dmi_max_khz();
+		mul = max_khz;
+		div = caps->highest_perf;
+	}
+
+	retval = offset + div64_u64(perf * mul, div);
+	if (retval >= 0)
+		return retval;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(cppc_perf_to_khz);
+
+unsigned int cppc_khz_to_perf(struct cppc_perf_caps *caps, unsigned int freq)
+{
+	s64 retval, offset = 0;
+	static u64 max_khz;
+	u64  mul, div;
+
+	if (caps->lowest_freq && caps->nominal_freq) {
+		mul = caps->nominal_perf - caps->lowest_perf;
+		div = caps->nominal_freq - caps->lowest_freq;
+		/*
+		 * We don't need to convert to kHz for computing offset and can
+		 * directly use nominal_freq and lowest_freq as the div64_u64
+		 * will remove the frequency unit.
+		 */
+		offset = caps->nominal_perf -
+			 div64_u64(caps->nominal_freq * mul, div);
+		/* But we need it for computing the perf level. */
+		div *= KHZ_PER_MHZ;
+	} else {
+		if (!max_khz)
+			max_khz = cppc_get_dmi_max_khz();
+		mul = caps->highest_perf;
+		div = max_khz;
+	}
+
+	retval = offset + div64_u64(freq * mul, div);
+	if (retval >= 0)
+		return retval;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(cppc_khz_to_perf);
diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
index fe08ca419b3d..64420d9cfd1e 100644
--- a/drivers/cpufreq/cppc_cpufreq.c
+++ b/drivers/cpufreq/cppc_cpufreq.c
@@ -16,7 +16,6 @@
 #include <linux/delay.h>
 #include <linux/cpu.h>
 #include <linux/cpufreq.h>
-#include <linux/dmi.h>
 #include <linux/irq_work.h>
 #include <linux/kthread.h>
 #include <linux/time.h>
@@ -27,12 +26,6 @@
 
 #include <acpi/cppc_acpi.h>
 
-/* Minimum struct length needed for the DMI processor entry we want */
-#define DMI_ENTRY_PROCESSOR_MIN_LENGTH	48
-
-/* Offset in the DMI processor structure for the max frequency */
-#define DMI_PROCESSOR_MAX_SPEED		0x14
-
 /*
  * This list contains information parsed from per CPU ACPI _CPC and _PSD
  * structures: e.g. the highest and lowest supported performance, capabilities,
@@ -291,97 +284,9 @@ static inline void cppc_freq_invariance_exit(void)
 }
 #endif /* CONFIG_ACPI_CPPC_CPUFREQ_FIE */
 
-/* Callback function used to retrieve the max frequency from DMI */
-static void cppc_find_dmi_mhz(const struct dmi_header *dm, void *private)
-{
-	const u8 *dmi_data = (const u8 *)dm;
-	u16 *mhz = (u16 *)private;
-
-	if (dm->type == DMI_ENTRY_PROCESSOR &&
-	    dm->length >= DMI_ENTRY_PROCESSOR_MIN_LENGTH) {
-		u16 val = (u16)get_unaligned((const u16 *)
-				(dmi_data + DMI_PROCESSOR_MAX_SPEED));
-		*mhz = val > *mhz ? val : *mhz;
-	}
-}
-
-/* Look up the max frequency in DMI */
-static u64 cppc_get_dmi_max_khz(void)
-{
-	u16 mhz = 0;
-
-	dmi_walk(cppc_find_dmi_mhz, &mhz);
-
-	/*
-	 * Real stupid fallback value, just in case there is no
-	 * actual value set.
-	 */
-	mhz = mhz ? mhz : 1;
-
-	return (1000 * mhz);
-}
-
-/*
- * If CPPC lowest_freq and nominal_freq registers are exposed then we can
- * use them to convert perf to freq and vice versa. The conversion is
- * extrapolated as an affine function passing by the 2 points:
- *  - (Low perf, Low freq)
- *  - (Nominal perf, Nominal perf)
- */
-static unsigned int cppc_cpufreq_perf_to_khz(struct cppc_cpudata *cpu_data,
-					     unsigned int perf)
-{
-	struct cppc_perf_caps *caps = &cpu_data->perf_caps;
-	s64 retval, offset = 0;
-	static u64 max_khz;
-	u64 mul, div;
-
-	if (caps->lowest_freq && caps->nominal_freq) {
-		mul = caps->nominal_freq - caps->lowest_freq;
-		div = caps->nominal_perf - caps->lowest_perf;
-		offset = caps->nominal_freq - div64_u64(caps->nominal_perf * mul, div);
-	} else {
-		if (!max_khz)
-			max_khz = cppc_get_dmi_max_khz();
-		mul = max_khz;
-		div = caps->highest_perf;
-	}
-
-	retval = offset + div64_u64(perf * mul, div);
-	if (retval >= 0)
-		return retval;
-	return 0;
-}
-
-static unsigned int cppc_cpufreq_khz_to_perf(struct cppc_cpudata *cpu_data,
-					     unsigned int freq)
-{
-	struct cppc_perf_caps *caps = &cpu_data->perf_caps;
-	s64 retval, offset = 0;
-	static u64 max_khz;
-	u64  mul, div;
-
-	if (caps->lowest_freq && caps->nominal_freq) {
-		mul = caps->nominal_perf - caps->lowest_perf;
-		div = caps->nominal_freq - caps->lowest_freq;
-		offset = caps->nominal_perf - div64_u64(caps->nominal_freq * mul, div);
-	} else {
-		if (!max_khz)
-			max_khz = cppc_get_dmi_max_khz();
-		mul = caps->highest_perf;
-		div = max_khz;
-	}
-
-	retval = offset + div64_u64(freq * mul, div);
-	if (retval >= 0)
-		return retval;
-	return 0;
-}
-
 static int cppc_cpufreq_set_target(struct cpufreq_policy *policy,
 				   unsigned int target_freq,
 				   unsigned int relation)
-
 {
 	struct cppc_cpudata *cpu_data = policy->driver_data;
 	unsigned int cpu = policy->cpu;
@@ -389,7 +294,7 @@ static int cppc_cpufreq_set_target(struct cpufreq_policy *policy,
 	u32 desired_perf;
 	int ret = 0;
 
-	desired_perf = cppc_cpufreq_khz_to_perf(cpu_data, target_freq);
+	desired_perf = cppc_khz_to_perf(&cpu_data->perf_caps, target_freq);
 	/* Return if it is exactly the same perf */
 	if (desired_perf == cpu_data->perf_ctrls.desired_perf)
 		return ret;
@@ -417,7 +322,7 @@ static unsigned int cppc_cpufreq_fast_switch(struct cpufreq_policy *policy,
 	u32 desired_perf;
 	int ret;
 
-	desired_perf = cppc_cpufreq_khz_to_perf(cpu_data, target_freq);
+	desired_perf = cppc_khz_to_perf(&cpu_data->perf_caps, target_freq);
 	cpu_data->perf_ctrls.desired_perf = desired_perf;
 	ret = cppc_set_perf(cpu, &cpu_data->perf_ctrls);
 
@@ -530,7 +435,7 @@ static int cppc_get_cpu_power(struct device *cpu_dev,
 	min_step = min_cap / CPPC_EM_CAP_STEP;
 	max_step = max_cap / CPPC_EM_CAP_STEP;
 
-	perf_prev = cppc_cpufreq_khz_to_perf(cpu_data, *KHz);
+	perf_prev = cppc_khz_to_perf(perf_caps, *KHz);
 	step = perf_prev / perf_step;
 
 	if (step > max_step)
@@ -550,8 +455,8 @@ static int cppc_get_cpu_power(struct device *cpu_dev,
 			perf = step * perf_step;
 	}
 
-	*KHz = cppc_cpufreq_perf_to_khz(cpu_data, perf);
-	perf_check = cppc_cpufreq_khz_to_perf(cpu_data, *KHz);
+	*KHz = cppc_perf_to_khz(perf_caps, perf);
+	perf_check = cppc_khz_to_perf(perf_caps, *KHz);
 	step_check = perf_check / perf_step;
 
 	/*
@@ -561,8 +466,8 @@ static int cppc_get_cpu_power(struct device *cpu_dev,
 	 */
 	while ((*KHz == prev_freq) || (step_check != step)) {
 		perf++;
-		*KHz = cppc_cpufreq_perf_to_khz(cpu_data, perf);
-		perf_check = cppc_cpufreq_khz_to_perf(cpu_data, *KHz);
+		*KHz = cppc_perf_to_khz(perf_caps, perf);
+		perf_check = cppc_khz_to_perf(perf_caps, *KHz);
 		step_check = perf_check / perf_step;
 	}
 
@@ -591,7 +496,7 @@ static int cppc_get_cpu_cost(struct device *cpu_dev, unsigned long KHz,
 	perf_caps = &cpu_data->perf_caps;
 	max_cap = arch_scale_cpu_capacity(cpu_dev->id);
 
-	perf_prev = cppc_cpufreq_khz_to_perf(cpu_data, KHz);
+	perf_prev = cppc_khz_to_perf(perf_caps, KHz);
 	perf_step = CPPC_EM_CAP_STEP * perf_caps->highest_perf / max_cap;
 	step = perf_prev / perf_step;
 
@@ -679,10 +584,6 @@ static struct cppc_cpudata *cppc_cpufreq_get_cpu_data(unsigned int cpu)
 		goto free_mask;
 	}
 
-	/* Convert the lowest and nominal freq from MHz to KHz */
-	cpu_data->perf_caps.lowest_freq *= 1000;
-	cpu_data->perf_caps.nominal_freq *= 1000;
-
 	list_add(&cpu_data->node, &cpu_data_list);
 
 	return cpu_data;
@@ -724,20 +625,16 @@ static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	 * Set min to lowest nonlinear perf to avoid any efficiency penalty (see
 	 * Section 8.4.7.1.1.5 of ACPI 6.1 spec)
 	 */
-	policy->min = cppc_cpufreq_perf_to_khz(cpu_data,
-					       caps->lowest_nonlinear_perf);
-	policy->max = cppc_cpufreq_perf_to_khz(cpu_data,
-					       caps->nominal_perf);
+	policy->min = cppc_perf_to_khz(caps, caps->lowest_nonlinear_perf);
+	policy->max = cppc_perf_to_khz(caps, caps->nominal_perf);
 
 	/*
 	 * Set cpuinfo.min_freq to Lowest to make the full range of performance
 	 * available if userspace wants to use any perf between lowest & lowest
 	 * nonlinear perf
 	 */
-	policy->cpuinfo.min_freq = cppc_cpufreq_perf_to_khz(cpu_data,
-							    caps->lowest_perf);
-	policy->cpuinfo.max_freq = cppc_cpufreq_perf_to_khz(cpu_data,
-							    caps->nominal_perf);
+	policy->cpuinfo.min_freq = cppc_perf_to_khz(caps, caps->lowest_perf);
+	policy->cpuinfo.max_freq = cppc_perf_to_khz(caps, caps->nominal_perf);
 
 	policy->transition_delay_us = cppc_cpufreq_get_transition_delay_us(cpu);
 	policy->shared_type = cpu_data->shared_type;
@@ -773,7 +670,7 @@ static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy)
 		boost_supported = true;
 
 	/* Set policy->cur to max now. The governors will adjust later. */
-	policy->cur = cppc_cpufreq_perf_to_khz(cpu_data, caps->highest_perf);
+	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);
@@ -863,7 +760,7 @@ static unsigned int cppc_cpufreq_get_rate(unsigned int cpu)
 	delivered_perf = cppc_perf_from_fbctrs(cpu_data, &fb_ctrs_t0,
 					       &fb_ctrs_t1);
 
-	return cppc_cpufreq_perf_to_khz(cpu_data, delivered_perf);
+	return cppc_perf_to_khz(&cpu_data->perf_caps, delivered_perf);
 }
 
 static int cppc_cpufreq_set_boost(struct cpufreq_policy *policy, int state)
@@ -878,11 +775,9 @@ static int cppc_cpufreq_set_boost(struct cpufreq_policy *policy, int state)
 	}
 
 	if (state)
-		policy->max = cppc_cpufreq_perf_to_khz(cpu_data,
-						       caps->highest_perf);
+		policy->max = cppc_perf_to_khz(caps, caps->highest_perf);
 	else
-		policy->max = cppc_cpufreq_perf_to_khz(cpu_data,
-						       caps->nominal_perf);
+		policy->max = cppc_perf_to_khz(caps, caps->nominal_perf);
 	policy->cpuinfo.max_freq = policy->max;
 
 	ret = freq_qos_update_request(policy->max_freq_req, policy->max);
@@ -937,7 +832,7 @@ static unsigned int hisi_cppc_cpufreq_get_rate(unsigned int cpu)
 	if (ret < 0)
 		return -EIO;
 
-	return cppc_cpufreq_perf_to_khz(cpu_data, desired_perf);
+	return cppc_perf_to_khz(&cpu_data->perf_caps, desired_perf);
 }
 
 static void cppc_check_hisi_workaround(void)
diff --git a/include/acpi/cppc_acpi.h b/include/acpi/cppc_acpi.h
index 6126c977ece0..3a0995f8bce8 100644
--- a/include/acpi/cppc_acpi.h
+++ b/include/acpi/cppc_acpi.h
@@ -144,6 +144,8 @@ extern int cppc_set_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls);
 extern int cppc_set_enable(int cpu, bool enable);
 extern int cppc_get_perf_caps(int cpu, struct cppc_perf_caps *caps);
 extern bool cppc_perf_ctrs_in_pcc(void);
+extern unsigned int cppc_perf_to_khz(struct cppc_perf_caps *caps, unsigned int perf);
+extern unsigned int cppc_khz_to_perf(struct cppc_perf_caps *caps, unsigned int freq);
 extern bool acpi_cpc_valid(void);
 extern bool cppc_allow_fast_switch(void);
 extern int acpi_get_psd_map(unsigned int cpu, struct cppc_cpudata *cpu_data);
-- 
2.34.1

[PATCH v7 6/7] cpufreq/cppc: Set the frequency used for computing the capacity

[Vincent Guittot]

Save the frequency associated to the performance that has been used when
initializing the capacity of CPUs.
Also, cppc cpufreq driver can register an artificial energy model. In such
case, it needs the frequency for this compute capacity.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
Tested-by: Pierre Gondois <pierre.gondois@arm.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/base/arch_topology.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
index e8d1cdf1f761..0a2e43728286 100644
--- a/drivers/base/arch_topology.c
+++ b/drivers/base/arch_topology.c
@@ -349,6 +349,7 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu)
 
 void topology_init_cpu_capacity_cppc(void)
 {
+	u64 capacity, capacity_scale = 0;
 	struct cppc_perf_caps perf_caps;
 	int cpu;
 
@@ -365,6 +366,10 @@ void topology_init_cpu_capacity_cppc(void)
 		    (perf_caps.highest_perf >= perf_caps.nominal_perf) &&
 		    (perf_caps.highest_perf >= perf_caps.lowest_perf)) {
 			raw_capacity[cpu] = perf_caps.highest_perf;
+			capacity_scale = max_t(u64, capacity_scale, raw_capacity[cpu]);
+
+			per_cpu(capacity_freq_ref, cpu) = cppc_perf_to_khz(&perf_caps, raw_capacity[cpu]);
+
 			pr_debug("cpu_capacity: CPU%d cpu_capacity=%u (raw).\n",
 				 cpu, raw_capacity[cpu]);
 			continue;
@@ -375,7 +380,15 @@ void topology_init_cpu_capacity_cppc(void)
 		goto exit;
 	}
 
-	topology_normalize_cpu_scale();
+	for_each_possible_cpu(cpu) {
+		capacity = raw_capacity[cpu];
+		capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT,
+				     capacity_scale);
+		topology_set_cpu_scale(cpu, capacity);
+		pr_debug("cpu_capacity: CPU%d cpu_capacity=%lu\n",
+			cpu, topology_get_cpu_scale(cpu));
+	}
+
 	schedule_work(&update_topology_flags_work);
 	pr_debug("cpu_capacity: cpu_capacity initialization done\n");
 
-- 
2.34.1

[PATCH v7 7/7] arm64/amu: Use capacity_ref_freq to set AMU ratio

[Vincent Guittot]

Use the new capacity_ref_freq to set the ratio that is used by AMU for
computing the arch_scale_freq_capacity().
This helps to keep everything aligned using the same reference for
computing CPUs capacity.

The default value of the ratio (stored in per_cpu(arch_max_freq_scale))
ensures that arch_scale_freq_capacity() returns max capacity until it is
set to its correct value with the cpu capacity and capacity_ref_freq.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
Acked-by: Will Deacon <will@kernel.org>
---
 arch/arm64/kernel/topology.c  | 26 +++++++++++++-------------
 drivers/base/arch_topology.c  | 12 +++++++++++-
 include/linux/arch_topology.h |  1 +
 3 files changed, 25 insertions(+), 14 deletions(-)

diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
index 817d788cd866..1a2c72f3e7f8 100644
--- a/arch/arm64/kernel/topology.c
+++ b/arch/arm64/kernel/topology.c
@@ -82,7 +82,12 @@ int __init parse_acpi_topology(void)
 #undef pr_fmt
 #define pr_fmt(fmt) "AMU: " fmt
 
-static DEFINE_PER_CPU_READ_MOSTLY(unsigned long, arch_max_freq_scale);
+/*
+ * Ensure that amu_scale_freq_tick() will return SCHED_CAPACITY_SCALE until
+ * the CPU capacity and its associated frequency have been correctly
+ * initialized.
+ */
+static DEFINE_PER_CPU_READ_MOSTLY(unsigned long, arch_max_freq_scale) =  1UL << (2 * SCHED_CAPACITY_SHIFT);
 static DEFINE_PER_CPU(u64, arch_const_cycles_prev);
 static DEFINE_PER_CPU(u64, arch_core_cycles_prev);
 static cpumask_var_t amu_fie_cpus;
@@ -112,14 +117,14 @@ static inline bool freq_counters_valid(int cpu)
 	return true;
 }
 
-static int freq_inv_set_max_ratio(int cpu, u64 max_rate, u64 ref_rate)
+void freq_inv_set_max_ratio(int cpu, u64 max_rate)
 {
-	u64 ratio;
+	u64 ratio, ref_rate = arch_timer_get_rate();
 
 	if (unlikely(!max_rate || !ref_rate)) {
-		pr_debug("CPU%d: invalid maximum or reference frequency.\n",
+		WARN_ONCE(1, "CPU%d: invalid maximum or reference frequency.\n",
 			 cpu);
-		return -EINVAL;
+		return;
 	}
 
 	/*
@@ -139,12 +144,10 @@ static int freq_inv_set_max_ratio(int cpu, u64 max_rate, u64 ref_rate)
 	ratio = div64_u64(ratio, max_rate);
 	if (!ratio) {
 		WARN_ONCE(1, "Reference frequency too low.\n");
-		return -EINVAL;
+		return;
 	}
 
-	per_cpu(arch_max_freq_scale, cpu) = (unsigned long)ratio;
-
-	return 0;
+	WRITE_ONCE(per_cpu(arch_max_freq_scale, cpu), (unsigned long)ratio);
 }
 
 static void amu_scale_freq_tick(void)
@@ -195,10 +198,7 @@ static void amu_fie_setup(const struct cpumask *cpus)
 		return;
 
 	for_each_cpu(cpu, cpus) {
-		if (!freq_counters_valid(cpu) ||
-		    freq_inv_set_max_ratio(cpu,
-					   cpufreq_get_hw_max_freq(cpu) * 1000ULL,
-					   arch_timer_get_rate()))
+		if (!freq_counters_valid(cpu))
 			return;
 	}
 
diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
index 0a2e43728286..0906114963ff 100644
--- a/drivers/base/arch_topology.c
+++ b/drivers/base/arch_topology.c
@@ -344,6 +344,10 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu)
 	return !ret;
 }
 
+void __weak freq_inv_set_max_ratio(int cpu, u64 max_rate)
+{
+}
+
 #ifdef CONFIG_ACPI_CPPC_LIB
 #include <acpi/cppc_acpi.h>
 
@@ -381,6 +385,9 @@ void topology_init_cpu_capacity_cppc(void)
 	}
 
 	for_each_possible_cpu(cpu) {
+		freq_inv_set_max_ratio(cpu,
+				       per_cpu(capacity_freq_ref, cpu) * HZ_PER_KHZ);
+
 		capacity = raw_capacity[cpu];
 		capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT,
 				     capacity_scale);
@@ -422,8 +429,11 @@ init_cpu_capacity_callback(struct notifier_block *nb,
 
 	cpumask_andnot(cpus_to_visit, cpus_to_visit, policy->related_cpus);
 
-	for_each_cpu(cpu, policy->related_cpus)
+	for_each_cpu(cpu, policy->related_cpus) {
 		per_cpu(capacity_freq_ref, cpu) = policy->cpuinfo.max_freq;
+		freq_inv_set_max_ratio(cpu,
+				       per_cpu(capacity_freq_ref, cpu) * HZ_PER_KHZ);
+	}
 
 	if (cpumask_empty(cpus_to_visit)) {
 		topology_normalize_cpu_scale();
diff --git a/include/linux/arch_topology.h b/include/linux/arch_topology.h
index 32c24ff4f2a8..a63d61ca55af 100644
--- a/include/linux/arch_topology.h
+++ b/include/linux/arch_topology.h
@@ -99,6 +99,7 @@ void update_siblings_masks(unsigned int cpu);
 void remove_cpu_topology(unsigned int cpuid);
 void reset_cpu_topology(void);
 int parse_acpi_topology(void);
+void freq_inv_set_max_ratio(int cpu, u64 max_rate);
 #endif
 
 #endif /* _LINUX_ARCH_TOPOLOGY_H_ */
-- 
2.34.1

Re: [PATCH v7 0/7] consolidate and cleanup CPU capacity

[Rafael J. Wysocki]

On Mon, Dec 11, 2023 at 11:49 AM Vincent Guittot
<vincent.guittot@linaro.org> wrote:
>
> This is the 1st part of consolidating how the max compute capacity is
> used in the scheduler and how we calculate the frequency for a level of
> utilization.
>
> Fix some unconsistancy when computing frequency for an utilization. There
> can be a mismatch between energy model and schedutil.
>
> Next step will be to make a difference between the original
> max compute capacity of a CPU and what is currently available when
> there is a capping applying forever (i.e. seconds or more).
>
> Changes since v6:
> - Rebase on tip/sched/core and fix a conflict
> - Add tags
>
> Changes since v5:
> - Remove useless return in freq_inv_set_max_ratio()
> - Add tags
>
> Changes since v4:
> - Capitalize the verb in subject
> - Remove usless parentheses in cppc_get_dmi_max_khz()
> - Use freq_ref pattern everywhere
> - Fix MHz / kHz units conversion for cppc_cpufreq
> - Move default definition of arch_scale_freq_ref() in
>   include/linux/sched/topology.h beside arch_scale_cpu_capacity
>   which faces similar default declaration behavior. This location covers
>   all cases with arch and CONFIG_* which was not the case with previous
>   attempts.
>
> Changes since v3:
> - Split patch 5 cpufreq/cppc
> - Fix topology_init_cpu_capacity_cppc()
> - Fix init if AMU ratio
> - Added some tags
>
> Changes since v2:
> - Remove the 1st patch which has been queued in tip
> - Rework how to initialize the reference frequency for cppc_cpufreq and
>   change topology_init_cpu_capacity_cppc() to also set capacity_ref_freq
> - Add a RFC to convert AMU to use arch_scale_freq_ref and move the config
>   of the AMU ratio to be done when intializing cpu capacity and
>   capacity_ref_freq
> - Added some tags
>
> Changes since v1:
> - Fix typos
> - Added changes in cpufreq to use arch_scale_freq_ref() when calling
>   arch_set_freq_scale (patch 3).
> - arch_scale_freq_ref() is always defined and returns 0 (as proposed
>   by Ionela) when not defined by the arch. This simplifies the code with
>   the addition of patch 3.
> - Simplify Energy Model which always uses arch_scale_freq_ref(). The
>   latter returns 0 when not defined by arch instead of last item of the
>   perf domain. This is not a problem because the function is only defined
>   for compilation purpose in this case and we don't care about the
>   returned value. (patch 5)
> - Added changes in cppc cpufreq to set capacity_ref_freq (patch 6)
> - Added reviewed tag for patch 1 which got a minor change but not for
>   others as I did some changes which could make previous reviewed tag
>   no more relevant.
>
> Vincent Guittot (7):
>   topology: Add a new arch_scale_freq_reference
>   cpufreq: Use the fixed and coherent frequency for scaling capacity
>   cpufreq/schedutil: Use a fixed reference frequency
>   energy_model: Use a fixed reference frequency
>   cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}
>   cpufreq/cppc: Set the frequency used for computing the capacity
>   arm64/amu: Use capacity_ref_freq to set AMU ratio

This series touches multiple places, but mostly schedutil, cpufreq and
the EM, so please let me know if you want me to pick it up.

Re: [PATCH v7 0/7] consolidate and cleanup CPU capacity

[Vincent Guittot]

On Mon, 11 Dec 2023 at 20:52, Rafael J. Wysocki <rafael@kernel.org> wrote:
>
> On Mon, Dec 11, 2023 at 11:49 AM Vincent Guittot
> <vincent.guittot@linaro.org> wrote:

[..]

> >
> > Vincent Guittot (7):
> >   topology: Add a new arch_scale_freq_reference
> >   cpufreq: Use the fixed and coherent frequency for scaling capacity
> >   cpufreq/schedutil: Use a fixed reference frequency
> >   energy_model: Use a fixed reference frequency
> >   cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}
> >   cpufreq/cppc: Set the frequency used for computing the capacity
> >   arm64/amu: Use capacity_ref_freq to set AMU ratio
>
> This series touches multiple places, but mostly schedutil, cpufreq and
> the EM, so please let me know if you want me to pick it up.

This serie has been rebased on top of tip/ched/core to fix a conflict
with another change already queued in kernel/sched/cpufreq_schedutil.c

Re: [PATCH v7 0/7] consolidate and cleanup CPU capacity

[Rafael J. Wysocki]

On Tue, Dec 12, 2023 at 9:32 AM Vincent Guittot
<vincent.guittot@linaro.org> wrote:
>
> On Mon, 11 Dec 2023 at 20:52, Rafael J. Wysocki <rafael@kernel.org> wrote:
> >
> > On Mon, Dec 11, 2023 at 11:49 AM Vincent Guittot
> > <vincent.guittot@linaro.org> wrote:
>
> [..]
>
> > >
> > > Vincent Guittot (7):
> > >   topology: Add a new arch_scale_freq_reference
> > >   cpufreq: Use the fixed and coherent frequency for scaling capacity
> > >   cpufreq/schedutil: Use a fixed reference frequency
> > >   energy_model: Use a fixed reference frequency
> > >   cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}
> > >   cpufreq/cppc: Set the frequency used for computing the capacity
> > >   arm64/amu: Use capacity_ref_freq to set AMU ratio
> >
> > This series touches multiple places, but mostly schedutil, cpufreq and
> > the EM, so please let me know if you want me to pick it up.
>
> This serie has been rebased on top of tip/ched/core to fix a conflict
> with another change already queued in kernel/sched/cpufreq_schedutil.c

OK, so I'm assuming that it will go in through tip.

Thanks!

[tip: sched/core] arm64/amu: Use capacity_ref_freq() to set AMU ratio

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     1f023007f5e782bda19ad9104830c404fd622c5d
Gitweb:        https://git.kernel.org/tip/1f023007f5e782bda19ad9104830c404fd622c5d
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Mon, 11 Dec 2023 11:48:55 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 15:52:36 +01:00

arm64/amu: Use capacity_ref_freq() to set AMU ratio

Use the new capacity_ref_freq() method to set the ratio that is used by AMU for
computing the arch_scale_freq_capacity().
This helps to keep everything aligned using the same reference for
computing CPUs capacity.

The default value of the ratio (stored in per_cpu(arch_max_freq_scale))
ensures that arch_scale_freq_capacity() returns max capacity until it is
set to its correct value with the cpu capacity and capacity_ref_freq().

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20231211104855.558096-8-vincent.guittot@linaro.org
---
 arch/arm64/kernel/topology.c  | 26 +++++++++++++-------------
 drivers/base/arch_topology.c  | 12 +++++++++++-
 include/linux/arch_topology.h |  1 +
 3 files changed, 25 insertions(+), 14 deletions(-)

diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
index 817d788..1a2c72f 100644
--- a/arch/arm64/kernel/topology.c
+++ b/arch/arm64/kernel/topology.c
@@ -82,7 +82,12 @@ int __init parse_acpi_topology(void)
 #undef pr_fmt
 #define pr_fmt(fmt) "AMU: " fmt
 
-static DEFINE_PER_CPU_READ_MOSTLY(unsigned long, arch_max_freq_scale);
+/*
+ * Ensure that amu_scale_freq_tick() will return SCHED_CAPACITY_SCALE until
+ * the CPU capacity and its associated frequency have been correctly
+ * initialized.
+ */
+static DEFINE_PER_CPU_READ_MOSTLY(unsigned long, arch_max_freq_scale) =  1UL << (2 * SCHED_CAPACITY_SHIFT);
 static DEFINE_PER_CPU(u64, arch_const_cycles_prev);
 static DEFINE_PER_CPU(u64, arch_core_cycles_prev);
 static cpumask_var_t amu_fie_cpus;
@@ -112,14 +117,14 @@ static inline bool freq_counters_valid(int cpu)
 	return true;
 }
 
-static int freq_inv_set_max_ratio(int cpu, u64 max_rate, u64 ref_rate)
+void freq_inv_set_max_ratio(int cpu, u64 max_rate)
 {
-	u64 ratio;
+	u64 ratio, ref_rate = arch_timer_get_rate();
 
 	if (unlikely(!max_rate || !ref_rate)) {
-		pr_debug("CPU%d: invalid maximum or reference frequency.\n",
+		WARN_ONCE(1, "CPU%d: invalid maximum or reference frequency.\n",
 			 cpu);
-		return -EINVAL;
+		return;
 	}
 
 	/*
@@ -139,12 +144,10 @@ static int freq_inv_set_max_ratio(int cpu, u64 max_rate, u64 ref_rate)
 	ratio = div64_u64(ratio, max_rate);
 	if (!ratio) {
 		WARN_ONCE(1, "Reference frequency too low.\n");
-		return -EINVAL;
+		return;
 	}
 
-	per_cpu(arch_max_freq_scale, cpu) = (unsigned long)ratio;
-
-	return 0;
+	WRITE_ONCE(per_cpu(arch_max_freq_scale, cpu), (unsigned long)ratio);
 }
 
 static void amu_scale_freq_tick(void)
@@ -195,10 +198,7 @@ static void amu_fie_setup(const struct cpumask *cpus)
 		return;
 
 	for_each_cpu(cpu, cpus) {
-		if (!freq_counters_valid(cpu) ||
-		    freq_inv_set_max_ratio(cpu,
-					   cpufreq_get_hw_max_freq(cpu) * 1000ULL,
-					   arch_timer_get_rate()))
+		if (!freq_counters_valid(cpu))
 			return;
 	}
 
diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
index 1aa76b5..5aaa086 100644
--- a/drivers/base/arch_topology.c
+++ b/drivers/base/arch_topology.c
@@ -344,6 +344,10 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu)
 	return !ret;
 }
 
+void __weak freq_inv_set_max_ratio(int cpu, u64 max_rate)
+{
+}
+
 #ifdef CONFIG_ACPI_CPPC_LIB
 #include <acpi/cppc_acpi.h>
 
@@ -381,6 +385,9 @@ void topology_init_cpu_capacity_cppc(void)
 	}
 
 	for_each_possible_cpu(cpu) {
+		freq_inv_set_max_ratio(cpu,
+				       per_cpu(capacity_freq_ref, cpu) * HZ_PER_KHZ);
+
 		capacity = raw_capacity[cpu];
 		capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT,
 				     capacity_scale);
@@ -422,8 +429,11 @@ init_cpu_capacity_callback(struct notifier_block *nb,
 
 	cpumask_andnot(cpus_to_visit, cpus_to_visit, policy->related_cpus);
 
-	for_each_cpu(cpu, policy->related_cpus)
+	for_each_cpu(cpu, policy->related_cpus) {
 		per_cpu(capacity_freq_ref, cpu) = policy->cpuinfo.max_freq;
+		freq_inv_set_max_ratio(cpu,
+				       per_cpu(capacity_freq_ref, cpu) * HZ_PER_KHZ);
+	}
 
 	if (cpumask_empty(cpus_to_visit)) {
 		topology_normalize_cpu_scale();
diff --git a/include/linux/arch_topology.h b/include/linux/arch_topology.h
index 32c24ff..a63d61c 100644
--- a/include/linux/arch_topology.h
+++ b/include/linux/arch_topology.h
@@ -99,6 +99,7 @@ void update_siblings_masks(unsigned int cpu);
 void remove_cpu_topology(unsigned int cpuid);
 void reset_cpu_topology(void);
 int parse_acpi_topology(void);
+void freq_inv_set_max_ratio(int cpu, u64 max_rate);
 #endif
 
 #endif /* _LINUX_ARCH_TOPOLOGY_H_ */

[tip: sched/core] cpufreq/cppc: Set the frequency used for computing the capacity

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     5477fa249b56c59c3baa1b237bf083cffa64c84a
Gitweb:        https://git.kernel.org/tip/5477fa249b56c59c3baa1b237bf083cffa64c84a
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Mon, 11 Dec 2023 11:48:54 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 15:52:36 +01:00

cpufreq/cppc: Set the frequency used for computing the capacity

Save the frequency associated to the performance that has been used when
initializing the capacity of CPUs.

Also, cppc cpufreq driver can register an artificial energy model. In such
case, it needs the frequency for this compute capacity.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Pierre Gondois <pierre.gondois@arm.com>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lore.kernel.org/r/20231211104855.558096-7-vincent.guittot@linaro.org
---
 drivers/base/arch_topology.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
index 0c9ae5b..1aa76b5 100644
--- a/drivers/base/arch_topology.c
+++ b/drivers/base/arch_topology.c
@@ -349,6 +349,7 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu)
 
 void topology_init_cpu_capacity_cppc(void)
 {
+	u64 capacity, capacity_scale = 0;
 	struct cppc_perf_caps perf_caps;
 	int cpu;
 
@@ -365,6 +366,10 @@ void topology_init_cpu_capacity_cppc(void)
 		    (perf_caps.highest_perf >= perf_caps.nominal_perf) &&
 		    (perf_caps.highest_perf >= perf_caps.lowest_perf)) {
 			raw_capacity[cpu] = perf_caps.highest_perf;
+			capacity_scale = max_t(u64, capacity_scale, raw_capacity[cpu]);
+
+			per_cpu(capacity_freq_ref, cpu) = cppc_perf_to_khz(&perf_caps, raw_capacity[cpu]);
+
 			pr_debug("cpu_capacity: CPU%d cpu_capacity=%u (raw).\n",
 				 cpu, raw_capacity[cpu]);
 			continue;
@@ -375,7 +380,15 @@ void topology_init_cpu_capacity_cppc(void)
 		goto exit;
 	}
 
-	topology_normalize_cpu_scale();
+	for_each_possible_cpu(cpu) {
+		capacity = raw_capacity[cpu];
+		capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT,
+				     capacity_scale);
+		topology_set_cpu_scale(cpu, capacity);
+		pr_debug("cpu_capacity: CPU%d cpu_capacity=%lu\n",
+			cpu, topology_get_cpu_scale(cpu));
+	}
+
 	schedule_work(&update_topology_flags_work);
 	pr_debug("cpu_capacity: cpu_capacity initialization done\n");
 

[tip: sched/core] cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}()

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     50b813b147e9eb6546a1fc49d4e703e6d23691f2
Gitweb:        https://git.kernel.org/tip/50b813b147e9eb6546a1fc49d4e703e6d23691f2
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Mon, 11 Dec 2023 11:48:53 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 15:52:35 +01:00

cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}()

Move and rename cppc_cpufreq_perf_to_khz() and cppc_cpufreq_khz_to_perf() to
use them outside cppc_cpufreq in topology_init_cpu_capacity_cppc().

Modify the interface to use struct cppc_perf_caps *caps instead of
struct cppc_cpudata *cpu_data as we only use the fields of cppc_perf_caps.

cppc_cpufreq was converting the lowest and nominal freq from MHz to kHz
before using them. We move this conversion inside cppc_perf_to_khz and
cppc_khz_to_perf to make them generic and usable outside cppc_cpufreq.

No functional change

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Pierre Gondois <pierre.gondois@arm.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lore.kernel.org/r/20231211104855.558096-6-vincent.guittot@linaro.org
---
 drivers/acpi/cppc_acpi.c       | 104 ++++++++++++++++++++++++-
 drivers/cpufreq/cppc_cpufreq.c | 139 +++-----------------------------
 include/acpi/cppc_acpi.h       |   2 +-
 3 files changed, 123 insertions(+), 122 deletions(-)

diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c
index 7ff269a..d155a86 100644
--- a/drivers/acpi/cppc_acpi.c
+++ b/drivers/acpi/cppc_acpi.c
@@ -39,6 +39,9 @@
 #include <linux/rwsem.h>
 #include <linux/wait.h>
 #include <linux/topology.h>
+#include <linux/dmi.h>
+#include <linux/units.h>
+#include <asm/unaligned.h>
 
 #include <acpi/cppc_acpi.h>
 
@@ -1760,3 +1763,104 @@ unsigned int cppc_get_transition_latency(int cpu_num)
 	return latency_ns;
 }
 EXPORT_SYMBOL_GPL(cppc_get_transition_latency);
+
+/* Minimum struct length needed for the DMI processor entry we want */
+#define DMI_ENTRY_PROCESSOR_MIN_LENGTH	48
+
+/* Offset in the DMI processor structure for the max frequency */
+#define DMI_PROCESSOR_MAX_SPEED		0x14
+
+/* Callback function used to retrieve the max frequency from DMI */
+static void cppc_find_dmi_mhz(const struct dmi_header *dm, void *private)
+{
+	const u8 *dmi_data = (const u8 *)dm;
+	u16 *mhz = (u16 *)private;
+
+	if (dm->type == DMI_ENTRY_PROCESSOR &&
+	    dm->length >= DMI_ENTRY_PROCESSOR_MIN_LENGTH) {
+		u16 val = (u16)get_unaligned((const u16 *)
+				(dmi_data + DMI_PROCESSOR_MAX_SPEED));
+		*mhz = val > *mhz ? val : *mhz;
+	}
+}
+
+/* Look up the max frequency in DMI */
+static u64 cppc_get_dmi_max_khz(void)
+{
+	u16 mhz = 0;
+
+	dmi_walk(cppc_find_dmi_mhz, &mhz);
+
+	/*
+	 * Real stupid fallback value, just in case there is no
+	 * actual value set.
+	 */
+	mhz = mhz ? mhz : 1;
+
+	return KHZ_PER_MHZ * mhz;
+}
+
+/*
+ * If CPPC lowest_freq and nominal_freq registers are exposed then we can
+ * use them to convert perf to freq and vice versa. The conversion is
+ * extrapolated as an affine function passing by the 2 points:
+ *  - (Low perf, Low freq)
+ *  - (Nominal perf, Nominal freq)
+ */
+unsigned int cppc_perf_to_khz(struct cppc_perf_caps *caps, unsigned int perf)
+{
+	s64 retval, offset = 0;
+	static u64 max_khz;
+	u64 mul, div;
+
+	if (caps->lowest_freq && caps->nominal_freq) {
+		mul = caps->nominal_freq - caps->lowest_freq;
+		mul *= KHZ_PER_MHZ;
+		div = caps->nominal_perf - caps->lowest_perf;
+		offset = caps->nominal_freq * KHZ_PER_MHZ -
+			 div64_u64(caps->nominal_perf * mul, div);
+	} else {
+		if (!max_khz)
+			max_khz = cppc_get_dmi_max_khz();
+		mul = max_khz;
+		div = caps->highest_perf;
+	}
+
+	retval = offset + div64_u64(perf * mul, div);
+	if (retval >= 0)
+		return retval;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(cppc_perf_to_khz);
+
+unsigned int cppc_khz_to_perf(struct cppc_perf_caps *caps, unsigned int freq)
+{
+	s64 retval, offset = 0;
+	static u64 max_khz;
+	u64  mul, div;
+
+	if (caps->lowest_freq && caps->nominal_freq) {
+		mul = caps->nominal_perf - caps->lowest_perf;
+		div = caps->nominal_freq - caps->lowest_freq;
+		/*
+		 * We don't need to convert to kHz for computing offset and can
+		 * directly use nominal_freq and lowest_freq as the div64_u64
+		 * will remove the frequency unit.
+		 */
+		offset = caps->nominal_perf -
+			 div64_u64(caps->nominal_freq * mul, div);
+		/* But we need it for computing the perf level. */
+		div *= KHZ_PER_MHZ;
+	} else {
+		if (!max_khz)
+			max_khz = cppc_get_dmi_max_khz();
+		mul = caps->highest_perf;
+		div = max_khz;
+	}
+
+	retval = offset + div64_u64(freq * mul, div);
+	if (retval >= 0)
+		return retval;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(cppc_khz_to_perf);
diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
index fe08ca4..64420d9 100644
--- a/drivers/cpufreq/cppc_cpufreq.c
+++ b/drivers/cpufreq/cppc_cpufreq.c
@@ -16,7 +16,6 @@
 #include <linux/delay.h>
 #include <linux/cpu.h>
 #include <linux/cpufreq.h>
-#include <linux/dmi.h>
 #include <linux/irq_work.h>
 #include <linux/kthread.h>
 #include <linux/time.h>
@@ -27,12 +26,6 @@
 
 #include <acpi/cppc_acpi.h>
 
-/* Minimum struct length needed for the DMI processor entry we want */
-#define DMI_ENTRY_PROCESSOR_MIN_LENGTH	48
-
-/* Offset in the DMI processor structure for the max frequency */
-#define DMI_PROCESSOR_MAX_SPEED		0x14
-
 /*
  * This list contains information parsed from per CPU ACPI _CPC and _PSD
  * structures: e.g. the highest and lowest supported performance, capabilities,
@@ -291,97 +284,9 @@ static inline void cppc_freq_invariance_exit(void)
 }
 #endif /* CONFIG_ACPI_CPPC_CPUFREQ_FIE */
 
-/* Callback function used to retrieve the max frequency from DMI */
-static void cppc_find_dmi_mhz(const struct dmi_header *dm, void *private)
-{
-	const u8 *dmi_data = (const u8 *)dm;
-	u16 *mhz = (u16 *)private;
-
-	if (dm->type == DMI_ENTRY_PROCESSOR &&
-	    dm->length >= DMI_ENTRY_PROCESSOR_MIN_LENGTH) {
-		u16 val = (u16)get_unaligned((const u16 *)
-				(dmi_data + DMI_PROCESSOR_MAX_SPEED));
-		*mhz = val > *mhz ? val : *mhz;
-	}
-}
-
-/* Look up the max frequency in DMI */
-static u64 cppc_get_dmi_max_khz(void)
-{
-	u16 mhz = 0;
-
-	dmi_walk(cppc_find_dmi_mhz, &mhz);
-
-	/*
-	 * Real stupid fallback value, just in case there is no
-	 * actual value set.
-	 */
-	mhz = mhz ? mhz : 1;
-
-	return (1000 * mhz);
-}
-
-/*
- * If CPPC lowest_freq and nominal_freq registers are exposed then we can
- * use them to convert perf to freq and vice versa. The conversion is
- * extrapolated as an affine function passing by the 2 points:
- *  - (Low perf, Low freq)
- *  - (Nominal perf, Nominal perf)
- */
-static unsigned int cppc_cpufreq_perf_to_khz(struct cppc_cpudata *cpu_data,
-					     unsigned int perf)
-{
-	struct cppc_perf_caps *caps = &cpu_data->perf_caps;
-	s64 retval, offset = 0;
-	static u64 max_khz;
-	u64 mul, div;
-
-	if (caps->lowest_freq && caps->nominal_freq) {
-		mul = caps->nominal_freq - caps->lowest_freq;
-		div = caps->nominal_perf - caps->lowest_perf;
-		offset = caps->nominal_freq - div64_u64(caps->nominal_perf * mul, div);
-	} else {
-		if (!max_khz)
-			max_khz = cppc_get_dmi_max_khz();
-		mul = max_khz;
-		div = caps->highest_perf;
-	}
-
-	retval = offset + div64_u64(perf * mul, div);
-	if (retval >= 0)
-		return retval;
-	return 0;
-}
-
-static unsigned int cppc_cpufreq_khz_to_perf(struct cppc_cpudata *cpu_data,
-					     unsigned int freq)
-{
-	struct cppc_perf_caps *caps = &cpu_data->perf_caps;
-	s64 retval, offset = 0;
-	static u64 max_khz;
-	u64  mul, div;
-
-	if (caps->lowest_freq && caps->nominal_freq) {
-		mul = caps->nominal_perf - caps->lowest_perf;
-		div = caps->nominal_freq - caps->lowest_freq;
-		offset = caps->nominal_perf - div64_u64(caps->nominal_freq * mul, div);
-	} else {
-		if (!max_khz)
-			max_khz = cppc_get_dmi_max_khz();
-		mul = caps->highest_perf;
-		div = max_khz;
-	}
-
-	retval = offset + div64_u64(freq * mul, div);
-	if (retval >= 0)
-		return retval;
-	return 0;
-}
-
 static int cppc_cpufreq_set_target(struct cpufreq_policy *policy,
 				   unsigned int target_freq,
 				   unsigned int relation)
-
 {
 	struct cppc_cpudata *cpu_data = policy->driver_data;
 	unsigned int cpu = policy->cpu;
@@ -389,7 +294,7 @@ static int cppc_cpufreq_set_target(struct cpufreq_policy *policy,
 	u32 desired_perf;
 	int ret = 0;
 
-	desired_perf = cppc_cpufreq_khz_to_perf(cpu_data, target_freq);
+	desired_perf = cppc_khz_to_perf(&cpu_data->perf_caps, target_freq);
 	/* Return if it is exactly the same perf */
 	if (desired_perf == cpu_data->perf_ctrls.desired_perf)
 		return ret;
@@ -417,7 +322,7 @@ static unsigned int cppc_cpufreq_fast_switch(struct cpufreq_policy *policy,
 	u32 desired_perf;
 	int ret;
 
-	desired_perf = cppc_cpufreq_khz_to_perf(cpu_data, target_freq);
+	desired_perf = cppc_khz_to_perf(&cpu_data->perf_caps, target_freq);
 	cpu_data->perf_ctrls.desired_perf = desired_perf;
 	ret = cppc_set_perf(cpu, &cpu_data->perf_ctrls);
 
@@ -530,7 +435,7 @@ static int cppc_get_cpu_power(struct device *cpu_dev,
 	min_step = min_cap / CPPC_EM_CAP_STEP;
 	max_step = max_cap / CPPC_EM_CAP_STEP;
 
-	perf_prev = cppc_cpufreq_khz_to_perf(cpu_data, *KHz);
+	perf_prev = cppc_khz_to_perf(perf_caps, *KHz);
 	step = perf_prev / perf_step;
 
 	if (step > max_step)
@@ -550,8 +455,8 @@ static int cppc_get_cpu_power(struct device *cpu_dev,
 			perf = step * perf_step;
 	}
 
-	*KHz = cppc_cpufreq_perf_to_khz(cpu_data, perf);
-	perf_check = cppc_cpufreq_khz_to_perf(cpu_data, *KHz);
+	*KHz = cppc_perf_to_khz(perf_caps, perf);
+	perf_check = cppc_khz_to_perf(perf_caps, *KHz);
 	step_check = perf_check / perf_step;
 
 	/*
@@ -561,8 +466,8 @@ static int cppc_get_cpu_power(struct device *cpu_dev,
 	 */
 	while ((*KHz == prev_freq) || (step_check != step)) {
 		perf++;
-		*KHz = cppc_cpufreq_perf_to_khz(cpu_data, perf);
-		perf_check = cppc_cpufreq_khz_to_perf(cpu_data, *KHz);
+		*KHz = cppc_perf_to_khz(perf_caps, perf);
+		perf_check = cppc_khz_to_perf(perf_caps, *KHz);
 		step_check = perf_check / perf_step;
 	}
 
@@ -591,7 +496,7 @@ static int cppc_get_cpu_cost(struct device *cpu_dev, unsigned long KHz,
 	perf_caps = &cpu_data->perf_caps;
 	max_cap = arch_scale_cpu_capacity(cpu_dev->id);
 
-	perf_prev = cppc_cpufreq_khz_to_perf(cpu_data, KHz);
+	perf_prev = cppc_khz_to_perf(perf_caps, KHz);
 	perf_step = CPPC_EM_CAP_STEP * perf_caps->highest_perf / max_cap;
 	step = perf_prev / perf_step;
 
@@ -679,10 +584,6 @@ static struct cppc_cpudata *cppc_cpufreq_get_cpu_data(unsigned int cpu)
 		goto free_mask;
 	}
 
-	/* Convert the lowest and nominal freq from MHz to KHz */
-	cpu_data->perf_caps.lowest_freq *= 1000;
-	cpu_data->perf_caps.nominal_freq *= 1000;
-
 	list_add(&cpu_data->node, &cpu_data_list);
 
 	return cpu_data;
@@ -724,20 +625,16 @@ static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	 * Set min to lowest nonlinear perf to avoid any efficiency penalty (see
 	 * Section 8.4.7.1.1.5 of ACPI 6.1 spec)
 	 */
-	policy->min = cppc_cpufreq_perf_to_khz(cpu_data,
-					       caps->lowest_nonlinear_perf);
-	policy->max = cppc_cpufreq_perf_to_khz(cpu_data,
-					       caps->nominal_perf);
+	policy->min = cppc_perf_to_khz(caps, caps->lowest_nonlinear_perf);
+	policy->max = cppc_perf_to_khz(caps, caps->nominal_perf);
 
 	/*
 	 * Set cpuinfo.min_freq to Lowest to make the full range of performance
 	 * available if userspace wants to use any perf between lowest & lowest
 	 * nonlinear perf
 	 */
-	policy->cpuinfo.min_freq = cppc_cpufreq_perf_to_khz(cpu_data,
-							    caps->lowest_perf);
-	policy->cpuinfo.max_freq = cppc_cpufreq_perf_to_khz(cpu_data,
-							    caps->nominal_perf);
+	policy->cpuinfo.min_freq = cppc_perf_to_khz(caps, caps->lowest_perf);
+	policy->cpuinfo.max_freq = cppc_perf_to_khz(caps, caps->nominal_perf);
 
 	policy->transition_delay_us = cppc_cpufreq_get_transition_delay_us(cpu);
 	policy->shared_type = cpu_data->shared_type;
@@ -773,7 +670,7 @@ static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy)
 		boost_supported = true;
 
 	/* Set policy->cur to max now. The governors will adjust later. */
-	policy->cur = cppc_cpufreq_perf_to_khz(cpu_data, caps->highest_perf);
+	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);
@@ -863,7 +760,7 @@ static unsigned int cppc_cpufreq_get_rate(unsigned int cpu)
 	delivered_perf = cppc_perf_from_fbctrs(cpu_data, &fb_ctrs_t0,
 					       &fb_ctrs_t1);
 
-	return cppc_cpufreq_perf_to_khz(cpu_data, delivered_perf);
+	return cppc_perf_to_khz(&cpu_data->perf_caps, delivered_perf);
 }
 
 static int cppc_cpufreq_set_boost(struct cpufreq_policy *policy, int state)
@@ -878,11 +775,9 @@ static int cppc_cpufreq_set_boost(struct cpufreq_policy *policy, int state)
 	}
 
 	if (state)
-		policy->max = cppc_cpufreq_perf_to_khz(cpu_data,
-						       caps->highest_perf);
+		policy->max = cppc_perf_to_khz(caps, caps->highest_perf);
 	else
-		policy->max = cppc_cpufreq_perf_to_khz(cpu_data,
-						       caps->nominal_perf);
+		policy->max = cppc_perf_to_khz(caps, caps->nominal_perf);
 	policy->cpuinfo.max_freq = policy->max;
 
 	ret = freq_qos_update_request(policy->max_freq_req, policy->max);
@@ -937,7 +832,7 @@ static unsigned int hisi_cppc_cpufreq_get_rate(unsigned int cpu)
 	if (ret < 0)
 		return -EIO;
 
-	return cppc_cpufreq_perf_to_khz(cpu_data, desired_perf);
+	return cppc_perf_to_khz(&cpu_data->perf_caps, desired_perf);
 }
 
 static void cppc_check_hisi_workaround(void)
diff --git a/include/acpi/cppc_acpi.h b/include/acpi/cppc_acpi.h
index 6126c97..3a0995f 100644
--- a/include/acpi/cppc_acpi.h
+++ b/include/acpi/cppc_acpi.h
@@ -144,6 +144,8 @@ extern int cppc_set_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls);
 extern int cppc_set_enable(int cpu, bool enable);
 extern int cppc_get_perf_caps(int cpu, struct cppc_perf_caps *caps);
 extern bool cppc_perf_ctrs_in_pcc(void);
+extern unsigned int cppc_perf_to_khz(struct cppc_perf_caps *caps, unsigned int perf);
+extern unsigned int cppc_khz_to_perf(struct cppc_perf_caps *caps, unsigned int freq);
 extern bool acpi_cpc_valid(void);
 extern bool cppc_allow_fast_switch(void);
 extern int acpi_get_psd_map(unsigned int cpu, struct cppc_cpudata *cpu_data);

[tip: sched/core] energy_model: Use a fixed reference frequency

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     15cbbd1d317e07b4e5c6aca5d4c5579539a82784
Gitweb:        https://git.kernel.org/tip/15cbbd1d317e07b4e5c6aca5d4c5579539a82784
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Mon, 11 Dec 2023 11:48:52 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 15:52:35 +01:00

energy_model: Use a fixed reference frequency

The last item of a performance domain is not always the performance point
that has been used to compute CPU's capacity. This can lead to different
target frequency compared with other part of the system like schedutil and
would result in wrong energy estimation.

A new arch_scale_freq_ref() is available to return a fixed and coherent
frequency reference that can be used when computing the CPU's frequency
for an level of utilization. Use this function to get this reference
frequency.

Energy model is never used without defining arch_scale_freq_ref() but
can be compiled. Define a default arch_scale_freq_ref() returning 0
in such case.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lore.kernel.org/r/20231211104855.558096-5-vincent.guittot@linaro.org
---
 include/linux/energy_model.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/linux/energy_model.h b/include/linux/energy_model.h
index adec808..88d91e0 100644
--- a/include/linux/energy_model.h
+++ b/include/linux/energy_model.h
@@ -224,7 +224,7 @@ static inline unsigned long em_cpu_energy(struct em_perf_domain *pd,
 				unsigned long max_util, unsigned long sum_util,
 				unsigned long allowed_cpu_cap)
 {
-	unsigned long freq, scale_cpu;
+	unsigned long freq, ref_freq, scale_cpu;
 	struct em_perf_state *ps;
 	int cpu;
 
@@ -241,10 +241,10 @@ static inline unsigned long em_cpu_energy(struct em_perf_domain *pd,
 	 */
 	cpu = cpumask_first(to_cpumask(pd->cpus));
 	scale_cpu = arch_scale_cpu_capacity(cpu);
-	ps = &pd->table[pd->nr_perf_states - 1];
+	ref_freq = arch_scale_freq_ref(cpu);
 
 	max_util = min(max_util, allowed_cpu_cap);
-	freq = map_util_freq(max_util, ps->frequency, scale_cpu);
+	freq = map_util_freq(max_util, ref_freq, scale_cpu);
 
 	/*
 	 * Find the lowest performance state of the Energy Model above the

[tip: sched/core] cpufreq/schedutil: Use a fixed reference frequency

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     b3edde44e5d4504c23a176819865cd603fd16d6c
Gitweb:        https://git.kernel.org/tip/b3edde44e5d4504c23a176819865cd603fd16d6c
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Mon, 11 Dec 2023 11:48:51 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 15:52:35 +01:00

cpufreq/schedutil: Use a fixed reference frequency

cpuinfo.max_freq can change at runtime because of boost as an example. This
implies that the value could be different than the one that has been
used when computing the capacity of a CPU.

The new arch_scale_freq_ref() returns a fixed and coherent reference
frequency that can be used when computing a frequency based on utilization.

Use this arch_scale_freq_ref() when available and fallback to
policy otherwise.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lore.kernel.org/r/20231211104855.558096-4-vincent.guittot@linaro.org
---
 kernel/sched/cpufreq_schedutil.c | 26 ++++++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 4ee8ad7..95c3c09 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -115,6 +115,28 @@ static void sugov_deferred_update(struct sugov_policy *sg_policy)
 }
 
 /**
+ * get_capacity_ref_freq - get the reference frequency that has been used to
+ * correlate frequency and compute capacity for a given cpufreq policy. We use
+ * the CPU managing it for the arch_scale_freq_ref() call in the function.
+ * @policy: the cpufreq policy of the CPU in question.
+ *
+ * Return: the reference CPU frequency to compute a capacity.
+ */
+static __always_inline
+unsigned long get_capacity_ref_freq(struct cpufreq_policy *policy)
+{
+	unsigned int freq = arch_scale_freq_ref(policy->cpu);
+
+	if (freq)
+		return freq;
+
+	if (arch_scale_freq_invariant())
+		return policy->cpuinfo.max_freq;
+
+	return policy->cur;
+}
+
+/**
  * get_next_freq - Compute a new frequency for a given cpufreq policy.
  * @sg_policy: schedutil policy object to compute the new frequency for.
  * @util: Current CPU utilization.
@@ -140,9 +162,9 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy,
 				  unsigned long util, unsigned long max)
 {
 	struct cpufreq_policy *policy = sg_policy->policy;
-	unsigned int freq = arch_scale_freq_invariant() ?
-				policy->cpuinfo.max_freq : policy->cur;
+	unsigned int freq;
 
+	freq = get_capacity_ref_freq(policy);
 	freq = map_util_freq(util, freq, max);
 
 	if (freq == sg_policy->cached_raw_freq && !sg_policy->need_freq_update)

[tip: sched/core] cpufreq: Use the fixed and coherent frequency for scaling capacity

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     599457ba15403037b489fe536266a3d5f9efaed7
Gitweb:        https://git.kernel.org/tip/599457ba15403037b489fe536266a3d5f9efaed7
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Mon, 11 Dec 2023 11:48:50 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 15:52:34 +01:00

cpufreq: Use the fixed and coherent frequency for scaling capacity

cpuinfo.max_freq can change at runtime because of boost as an example. This
implies that the value could be different from the frequency that has been
used to compute the capacity of a CPU.

The new arch_scale_freq_ref() returns a fixed and coherent frequency
that can be used to compute the capacity for a given frequency.

[ Also fix a arch_set_freq_scale()  newline style wart in <linux/cpufreq.h>. ]

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Link: https://lore.kernel.org/r/20231211104855.558096-3-vincent.guittot@linaro.org
---
 drivers/cpufreq/cpufreq.c | 4 ++--
 include/linux/cpufreq.h   | 1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 934d35f..44db4f5 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -454,7 +454,7 @@ void cpufreq_freq_transition_end(struct cpufreq_policy *policy,
 
 	arch_set_freq_scale(policy->related_cpus,
 			    policy->cur,
-			    policy->cpuinfo.max_freq);
+			    arch_scale_freq_ref(policy->cpu));
 
 	spin_lock(&policy->transition_lock);
 	policy->transition_ongoing = false;
@@ -2174,7 +2174,7 @@ unsigned int cpufreq_driver_fast_switch(struct cpufreq_policy *policy,
 
 	policy->cur = freq;
 	arch_set_freq_scale(policy->related_cpus, freq,
-			    policy->cpuinfo.max_freq);
+			    arch_scale_freq_ref(policy->cpu));
 	cpufreq_stats_record_transition(policy, freq);
 
 	if (trace_cpu_frequency_enabled()) {
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 1c5ca92..afda5f2 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -1203,6 +1203,7 @@ void arch_set_freq_scale(const struct cpumask *cpus,
 {
 }
 #endif
+
 /* the following are really really optional */
 extern struct freq_attr cpufreq_freq_attr_scaling_available_freqs;
 extern struct freq_attr cpufreq_freq_attr_scaling_boost_freqs;

[tip: sched/core] sched/topology: Add a new arch_scale_freq_ref() method

[tip-bot2 for Vincent Guittot]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     9942cb22ea458c34fa17b73d143ea32d4df1caca
Gitweb:        https://git.kernel.org/tip/9942cb22ea458c34fa17b73d143ea32d4df1caca
Author:        Vincent Guittot <vincent.guittot@linaro.org>
AuthorDate:    Mon, 11 Dec 2023 11:48:49 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 15:52:34 +01:00

sched/topology: Add a new arch_scale_freq_ref() method

Create a new method to get a unique and fixed max frequency. Currently
cpuinfo.max_freq or the highest (or last) state of performance domain are
used as the max frequency when computing the frequency for a level of
utilization, but:

  - cpuinfo_max_freq can change at runtime. boost is one example of
    such change.

  - cpuinfo.max_freq and last item of the PD can be different leading to
    different results between cpufreq and energy model.

We need to save the reference frequency that has been used when computing
the CPUs capacity and use this fixed and coherent value to convert between
frequency and CPU's capacity.

In fact, we already save the frequency that has been used when computing
the capacity of each CPU. We extend the precision to save kHz instead of
MHz currently and we modify the type to be aligned with other variables
used when converting frequency to capacity and the other way.

[ mingo: Minor edits. ]

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
Link: https://lore.kernel.org/r/20231211104855.558096-2-vincent.guittot@linaro.org
---
 arch/arm/include/asm/topology.h   |  1 +
 arch/arm64/include/asm/topology.h |  1 +
 arch/riscv/include/asm/topology.h |  1 +
 drivers/base/arch_topology.c      | 29 ++++++++++++++---------------
 include/linux/arch_topology.h     |  7 +++++++
 include/linux/sched/topology.h    |  8 ++++++++
 6 files changed, 32 insertions(+), 15 deletions(-)

diff --git a/arch/arm/include/asm/topology.h b/arch/arm/include/asm/topology.h
index c7d2510..853c4f8 100644
--- a/arch/arm/include/asm/topology.h
+++ b/arch/arm/include/asm/topology.h
@@ -13,6 +13,7 @@
 #define arch_set_freq_scale topology_set_freq_scale
 #define arch_scale_freq_capacity topology_get_freq_scale
 #define arch_scale_freq_invariant topology_scale_freq_invariant
+#define arch_scale_freq_ref topology_get_freq_ref
 #endif
 
 /* Replace task scheduler's default cpu-invariant accounting */
diff --git a/arch/arm64/include/asm/topology.h b/arch/arm64/include/asm/topology.h
index 9fab663..a323b10 100644
--- a/arch/arm64/include/asm/topology.h
+++ b/arch/arm64/include/asm/topology.h
@@ -23,6 +23,7 @@ void update_freq_counters_refs(void);
 #define arch_set_freq_scale topology_set_freq_scale
 #define arch_scale_freq_capacity topology_get_freq_scale
 #define arch_scale_freq_invariant topology_scale_freq_invariant
+#define arch_scale_freq_ref topology_get_freq_ref
 
 #ifdef CONFIG_ACPI_CPPC_LIB
 #define arch_init_invariance_cppc topology_init_cpu_capacity_cppc
diff --git a/arch/riscv/include/asm/topology.h b/arch/riscv/include/asm/topology.h
index e316ab3..6118368 100644
--- a/arch/riscv/include/asm/topology.h
+++ b/arch/riscv/include/asm/topology.h
@@ -9,6 +9,7 @@
 #define arch_set_freq_scale		topology_set_freq_scale
 #define arch_scale_freq_capacity	topology_get_freq_scale
 #define arch_scale_freq_invariant	topology_scale_freq_invariant
+#define arch_scale_freq_ref		topology_get_freq_ref
 
 /* Replace task scheduler's default cpu-invariant accounting */
 #define arch_scale_cpu_capacity	topology_get_cpu_scale
diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
index b741b5b..0c9ae5b 100644
--- a/drivers/base/arch_topology.c
+++ b/drivers/base/arch_topology.c
@@ -19,6 +19,7 @@
 #include <linux/init.h>
 #include <linux/rcupdate.h>
 #include <linux/sched.h>
+#include <linux/units.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/thermal_pressure.h>
@@ -26,7 +27,8 @@
 static DEFINE_PER_CPU(struct scale_freq_data __rcu *, sft_data);
 static struct cpumask scale_freq_counters_mask;
 static bool scale_freq_invariant;
-static DEFINE_PER_CPU(u32, freq_factor) = 1;
+DEFINE_PER_CPU(unsigned long, capacity_freq_ref) = 1;
+EXPORT_PER_CPU_SYMBOL_GPL(capacity_freq_ref);
 
 static bool supports_scale_freq_counters(const struct cpumask *cpus)
 {
@@ -170,9 +172,9 @@ DEFINE_PER_CPU(unsigned long, thermal_pressure);
  * operating on stale data when hot-plug is used for some CPUs. The
  * @capped_freq reflects the currently allowed max CPUs frequency due to
  * thermal capping. It might be also a boost frequency value, which is bigger
- * than the internal 'freq_factor' max frequency. In such case the pressure
- * value should simply be removed, since this is an indication that there is
- * no thermal throttling. The @capped_freq must be provided in kHz.
+ * than the internal 'capacity_freq_ref' max frequency. In such case the
+ * pressure value should simply be removed, since this is an indication that
+ * there is no thermal throttling. The @capped_freq must be provided in kHz.
  */
 void topology_update_thermal_pressure(const struct cpumask *cpus,
 				      unsigned long capped_freq)
@@ -183,10 +185,7 @@ void topology_update_thermal_pressure(const struct cpumask *cpus,
 
 	cpu = cpumask_first(cpus);
 	max_capacity = arch_scale_cpu_capacity(cpu);
-	max_freq = per_cpu(freq_factor, cpu);
-
-	/* Convert to MHz scale which is used in 'freq_factor' */
-	capped_freq /= 1000;
+	max_freq = arch_scale_freq_ref(cpu);
 
 	/*
 	 * Handle properly the boost frequencies, which should simply clean
@@ -279,13 +278,13 @@ void topology_normalize_cpu_scale(void)
 
 	capacity_scale = 1;
 	for_each_possible_cpu(cpu) {
-		capacity = raw_capacity[cpu] * per_cpu(freq_factor, cpu);
+		capacity = raw_capacity[cpu] * per_cpu(capacity_freq_ref, cpu);
 		capacity_scale = max(capacity, capacity_scale);
 	}
 
 	pr_debug("cpu_capacity: capacity_scale=%llu\n", capacity_scale);
 	for_each_possible_cpu(cpu) {
-		capacity = raw_capacity[cpu] * per_cpu(freq_factor, cpu);
+		capacity = raw_capacity[cpu] * per_cpu(capacity_freq_ref, cpu);
 		capacity = div64_u64(capacity << SCHED_CAPACITY_SHIFT,
 			capacity_scale);
 		topology_set_cpu_scale(cpu, capacity);
@@ -321,15 +320,15 @@ bool __init topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu)
 			cpu_node, raw_capacity[cpu]);
 
 		/*
-		 * Update freq_factor for calculating early boot cpu capacities.
+		 * Update capacity_freq_ref for calculating early boot CPU capacities.
 		 * For non-clk CPU DVFS mechanism, there's no way to get the
 		 * frequency value now, assuming they are running at the same
-		 * frequency (by keeping the initial freq_factor value).
+		 * frequency (by keeping the initial capacity_freq_ref value).
 		 */
 		cpu_clk = of_clk_get(cpu_node, 0);
 		if (!PTR_ERR_OR_ZERO(cpu_clk)) {
-			per_cpu(freq_factor, cpu) =
-				clk_get_rate(cpu_clk) / 1000;
+			per_cpu(capacity_freq_ref, cpu) =
+				clk_get_rate(cpu_clk) / HZ_PER_KHZ;
 			clk_put(cpu_clk);
 		}
 	} else {
@@ -411,7 +410,7 @@ init_cpu_capacity_callback(struct notifier_block *nb,
 	cpumask_andnot(cpus_to_visit, cpus_to_visit, policy->related_cpus);
 
 	for_each_cpu(cpu, policy->related_cpus)
-		per_cpu(freq_factor, cpu) = policy->cpuinfo.max_freq / 1000;
+		per_cpu(capacity_freq_ref, cpu) = policy->cpuinfo.max_freq;
 
 	if (cpumask_empty(cpus_to_visit)) {
 		topology_normalize_cpu_scale();
diff --git a/include/linux/arch_topology.h b/include/linux/arch_topology.h
index a07b510..32c24ff 100644
--- a/include/linux/arch_topology.h
+++ b/include/linux/arch_topology.h
@@ -27,6 +27,13 @@ static inline unsigned long topology_get_cpu_scale(int cpu)
 
 void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity);
 
+DECLARE_PER_CPU(unsigned long, capacity_freq_ref);
+
+static inline unsigned long topology_get_freq_ref(int cpu)
+{
+	return per_cpu(capacity_freq_ref, cpu);
+}
+
 DECLARE_PER_CPU(unsigned long, arch_freq_scale);
 
 static inline unsigned long topology_get_freq_scale(int cpu)
diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h
index de545ba..a6e04b4 100644
--- a/include/linux/sched/topology.h
+++ b/include/linux/sched/topology.h
@@ -279,6 +279,14 @@ void arch_update_thermal_pressure(const struct cpumask *cpus,
 { }
 #endif
 
+#ifndef arch_scale_freq_ref
+static __always_inline
+unsigned int arch_scale_freq_ref(int cpu)
+{
+	return 0;
+}
+#endif
+
 static inline int task_node(const struct task_struct *p)
 {
 	return cpu_to_node(task_cpu(p));

Re: [PATCH v7 3/7] cpufreq/schedutil: Use a fixed reference frequency

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 5821 bytes --]

On Mon, Dec 11, 2023 at 11:48:51AM +0100, Vincent Guittot wrote:
> cpuinfo.max_freq can change at runtime because of boost as an example. This
> implies that the value could be different than the one that has been
> used when computing the capacity of a CPU.

So, this seems very weird but I just finished a bisection of a failure
that showed up during the merge window with the DT kselftests on
meson-gxl-s905x-libretech-cc (Libretech Le Potato) running an arm64
defconfig.  Looking at the commit this seems like a very surprising
false result but I verified that the commit before this one runs fine
and reproduced the failure, the bisect does seem to converge well also
so it appears like there's something going on.  Does this ring any bells
for anyone?

The test is timed out by the kselftest wrapper:

   kselftest: Running tests in dt
   TAP version 13
   1..1
   # timeout set to 45
   # selftests: dt: test_unprobed_devices.sh
   # TAP version 13
   #
   not ok 1 selftests: dt: test_unprobed_devices.sh # TIMEOUT 45 seconds

with no obvious effect on the system, a successful run takes less than
20 seconds so it shouldn't be that the timing is marginal.  I'm guessing
that reading the one of the files under /sys/devices hung.  No other
boards in my lab are affected and I'm not immediately seeing anything
remarkable about this board.  Full log from a bad run at:

    https://lava.sirena.org.uk/scheduler/job/493329

and a good run:

    https://lava.sirena.org.uk/scheduler/job/492453

Bisect log (the skipped commits didn't build):

git bisect start
# status: waiting for both good and bad commits
# bad: [4fbbed7872677b0a28ba8237169968171a61efbd] Merge tag 'timers-core-2024-01-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
git bisect bad 4fbbed7872677b0a28ba8237169968171a61efbd
# status: waiting for good commit(s), bad commit known
# good: [0dd3ee31125508cd67f7e7172247f05b7fd1753a] Linux 6.7
git bisect good 0dd3ee31125508cd67f7e7172247f05b7fd1753a
# bad: [ba5afb9a84df2e6b26a1b6389b98849cd16ea757] fs: rework listmount() implementation
git bisect bad ba5afb9a84df2e6b26a1b6389b98849cd16ea757
# bad: [de927f6c0b07d9e698416c5b287c521b07694cac] Merge tag 's390-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
git bisect bad de927f6c0b07d9e698416c5b287c521b07694cac
# bad: [35f11a3710cdcbbc5090d14017a6295454e0cc73] Merge tag 'mtd/for-6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux
git bisect bad 35f11a3710cdcbbc5090d14017a6295454e0cc73
# bad: [d30e51aa7b1f6fa7dd78d4598d1e4c047fcc3fb9] Merge tag 'slab-for-6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab
git bisect bad d30e51aa7b1f6fa7dd78d4598d1e4c047fcc3fb9
# skip: [968b80332432172dbbb773e749a43bdc846d1a13] Merge tag 'powerpc-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
git bisect skip 968b80332432172dbbb773e749a43bdc846d1a13
# good: [372a34e66fb7f95124fadae9c600b231c35696a7] fs: replace f_rcuhead with f_task_work
git bisect good 372a34e66fb7f95124fadae9c600b231c35696a7
# good: [ae24db43b3b427eb290b58d55179c32f0a7539d1] powerpc/ftrace: Remove nops after the call to ftrace_stub
git bisect good ae24db43b3b427eb290b58d55179c32f0a7539d1
# skip: [3cf1d6a5fbf3f724d12b01635319924239d42c00] Merge tag 'm68k-for-v6.8-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k
git bisect skip 3cf1d6a5fbf3f724d12b01635319924239d42c00
# good: [2a33e2ddc6ebf9b5468091aded8a38f57de9a580] splice: remove permission hook from do_splice_direct()
git bisect good 2a33e2ddc6ebf9b5468091aded8a38f57de9a580
# good: [9b7e9e2f5d5c3d079ec46bc71b114012e362ea6e] fs: factor out backing_file_splice_{read,write}() helpers
git bisect good 9b7e9e2f5d5c3d079ec46bc71b114012e362ea6e
# good: [12c1b632d970c0138b4c5c65a1065e7d0604d272] fs: reformat idmapped mounts entry
git bisect good 12c1b632d970c0138b4c5c65a1065e7d0604d272
# good: [d23627a7688fabff0096a7beaff1a93de76afaad] EDAC/igen6: Add Intel Raptor Lake-P SoCs support
git bisect good d23627a7688fabff0096a7beaff1a93de76afaad
# skip: [ab5f3fcb7c72094684760e0cd8954d8d570b5e83] Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
git bisect skip ab5f3fcb7c72094684760e0cd8954d8d570b5e83
# good: [eb183b2cd0a6549992eca3c4ada0b1bc1d9340f5] Revert "perf/arm_dmc620: Remove duplicate format attribute #defines"
git bisect good eb183b2cd0a6549992eca3c4ada0b1bc1d9340f5
# good: [1ab33c03145d0f6c345823fc2da935d9a1a9e9fc] asm-generic: make sparse happy with odd-sized put_unaligned_*()
git bisect good 1ab33c03145d0f6c345823fc2da935d9a1a9e9fc
# good: [794c68b20408bb6899f90314e36e256924cc85a1] x86/CPU/AMD: Get rid of amd_erratum_1485[]
git bisect good 794c68b20408bb6899f90314e36e256924cc85a1
# bad: [11137d384996bb05cf33c8163db271e1bac3f4bf] sched/fair: Simplify util_est
git bisect bad 11137d384996bb05cf33c8163db271e1bac3f4bf
# good: [9c0b4bb7f6303c9c4e2e34984c46f5a86478f84d] sched/cpufreq: Rework schedutil governor performance estimation
git bisect good 9c0b4bb7f6303c9c4e2e34984c46f5a86478f84d
# good: [599457ba15403037b489fe536266a3d5f9efaed7] cpufreq: Use the fixed and coherent frequency for scaling capacity
git bisect good 599457ba15403037b489fe536266a3d5f9efaed7
# bad: [50b813b147e9eb6546a1fc49d4e703e6d23691f2] cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}()
git bisect bad 50b813b147e9eb6546a1fc49d4e703e6d23691f2
# bad: [15cbbd1d317e07b4e5c6aca5d4c5579539a82784] energy_model: Use a fixed reference frequency
git bisect bad 15cbbd1d317e07b4e5c6aca5d4c5579539a82784
# bad: [b3edde44e5d4504c23a176819865cd603fd16d6c] cpufreq/schedutil: Use a fixed reference frequency
git bisect bad b3edde44e5d4504c23a176819865cd603fd16d6c
# first bad commit: [b3edde44e5d4504c23a176819865cd603fd16d6c] cpufreq/schedutil: Use a fixed reference frequency

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH v7 3/7] cpufreq/schedutil: Use a fixed reference frequency

[Vincent Guittot]

Hi Mark,

Could you tried this fix:
https://lore.kernel.org/lkml/20240117190545.596057-1-vincent.guittot@linaro.org/

Regards,
Vincent

On Tue, 23 Jan 2024 at 02:55, Mark Brown <broonie@kernel.org> wrote:
>
> On Mon, Dec 11, 2023 at 11:48:51AM +0100, Vincent Guittot wrote:
> > cpuinfo.max_freq can change at runtime because of boost as an example. This
> > implies that the value could be different than the one that has been
> > used when computing the capacity of a CPU.
>
> So, this seems very weird but I just finished a bisection of a failure
> that showed up during the merge window with the DT kselftests on
> meson-gxl-s905x-libretech-cc (Libretech Le Potato) running an arm64
> defconfig.  Looking at the commit this seems like a very surprising
> false result but I verified that the commit before this one runs fine
> and reproduced the failure, the bisect does seem to converge well also
> so it appears like there's something going on.  Does this ring any bells
> for anyone?
>
> The test is timed out by the kselftest wrapper:
>
>    kselftest: Running tests in dt
>    TAP version 13
>    1..1
>    # timeout set to 45
>    # selftests: dt: test_unprobed_devices.sh
>    # TAP version 13
>    #
>    not ok 1 selftests: dt: test_unprobed_devices.sh # TIMEOUT 45 seconds
>
> with no obvious effect on the system, a successful run takes less than
> 20 seconds so it shouldn't be that the timing is marginal.  I'm guessing
> that reading the one of the files under /sys/devices hung.  No other
> boards in my lab are affected and I'm not immediately seeing anything
> remarkable about this board.  Full log from a bad run at:
>
>     https://lava.sirena.org.uk/scheduler/job/493329
>
> and a good run:
>
>     https://lava.sirena.org.uk/scheduler/job/492453
>
> Bisect log (the skipped commits didn't build):
>
> git bisect start
> # status: waiting for both good and bad commits
> # bad: [4fbbed7872677b0a28ba8237169968171a61efbd] Merge tag 'timers-core-2024-01-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
> git bisect bad 4fbbed7872677b0a28ba8237169968171a61efbd
> # status: waiting for good commit(s), bad commit known
> # good: [0dd3ee31125508cd67f7e7172247f05b7fd1753a] Linux 6.7
> git bisect good 0dd3ee31125508cd67f7e7172247f05b7fd1753a
> # bad: [ba5afb9a84df2e6b26a1b6389b98849cd16ea757] fs: rework listmount() implementation
> git bisect bad ba5afb9a84df2e6b26a1b6389b98849cd16ea757
> # bad: [de927f6c0b07d9e698416c5b287c521b07694cac] Merge tag 's390-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
> git bisect bad de927f6c0b07d9e698416c5b287c521b07694cac
> # bad: [35f11a3710cdcbbc5090d14017a6295454e0cc73] Merge tag 'mtd/for-6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux
> git bisect bad 35f11a3710cdcbbc5090d14017a6295454e0cc73
> # bad: [d30e51aa7b1f6fa7dd78d4598d1e4c047fcc3fb9] Merge tag 'slab-for-6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab
> git bisect bad d30e51aa7b1f6fa7dd78d4598d1e4c047fcc3fb9
> # skip: [968b80332432172dbbb773e749a43bdc846d1a13] Merge tag 'powerpc-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
> git bisect skip 968b80332432172dbbb773e749a43bdc846d1a13
> # good: [372a34e66fb7f95124fadae9c600b231c35696a7] fs: replace f_rcuhead with f_task_work
> git bisect good 372a34e66fb7f95124fadae9c600b231c35696a7
> # good: [ae24db43b3b427eb290b58d55179c32f0a7539d1] powerpc/ftrace: Remove nops after the call to ftrace_stub
> git bisect good ae24db43b3b427eb290b58d55179c32f0a7539d1
> # skip: [3cf1d6a5fbf3f724d12b01635319924239d42c00] Merge tag 'm68k-for-v6.8-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k
> git bisect skip 3cf1d6a5fbf3f724d12b01635319924239d42c00
> # good: [2a33e2ddc6ebf9b5468091aded8a38f57de9a580] splice: remove permission hook from do_splice_direct()
> git bisect good 2a33e2ddc6ebf9b5468091aded8a38f57de9a580
> # good: [9b7e9e2f5d5c3d079ec46bc71b114012e362ea6e] fs: factor out backing_file_splice_{read,write}() helpers
> git bisect good 9b7e9e2f5d5c3d079ec46bc71b114012e362ea6e
> # good: [12c1b632d970c0138b4c5c65a1065e7d0604d272] fs: reformat idmapped mounts entry
> git bisect good 12c1b632d970c0138b4c5c65a1065e7d0604d272
> # good: [d23627a7688fabff0096a7beaff1a93de76afaad] EDAC/igen6: Add Intel Raptor Lake-P SoCs support
> git bisect good d23627a7688fabff0096a7beaff1a93de76afaad
> # skip: [ab5f3fcb7c72094684760e0cd8954d8d570b5e83] Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
> git bisect skip ab5f3fcb7c72094684760e0cd8954d8d570b5e83
> # good: [eb183b2cd0a6549992eca3c4ada0b1bc1d9340f5] Revert "perf/arm_dmc620: Remove duplicate format attribute #defines"
> git bisect good eb183b2cd0a6549992eca3c4ada0b1bc1d9340f5
> # good: [1ab33c03145d0f6c345823fc2da935d9a1a9e9fc] asm-generic: make sparse happy with odd-sized put_unaligned_*()
> git bisect good 1ab33c03145d0f6c345823fc2da935d9a1a9e9fc
> # good: [794c68b20408bb6899f90314e36e256924cc85a1] x86/CPU/AMD: Get rid of amd_erratum_1485[]
> git bisect good 794c68b20408bb6899f90314e36e256924cc85a1
> # bad: [11137d384996bb05cf33c8163db271e1bac3f4bf] sched/fair: Simplify util_est
> git bisect bad 11137d384996bb05cf33c8163db271e1bac3f4bf
> # good: [9c0b4bb7f6303c9c4e2e34984c46f5a86478f84d] sched/cpufreq: Rework schedutil governor performance estimation
> git bisect good 9c0b4bb7f6303c9c4e2e34984c46f5a86478f84d
> # good: [599457ba15403037b489fe536266a3d5f9efaed7] cpufreq: Use the fixed and coherent frequency for scaling capacity
> git bisect good 599457ba15403037b489fe536266a3d5f9efaed7
> # bad: [50b813b147e9eb6546a1fc49d4e703e6d23691f2] cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf}()
> git bisect bad 50b813b147e9eb6546a1fc49d4e703e6d23691f2
> # bad: [15cbbd1d317e07b4e5c6aca5d4c5579539a82784] energy_model: Use a fixed reference frequency
> git bisect bad 15cbbd1d317e07b4e5c6aca5d4c5579539a82784
> # bad: [b3edde44e5d4504c23a176819865cd603fd16d6c] cpufreq/schedutil: Use a fixed reference frequency
> git bisect bad b3edde44e5d4504c23a176819865cd603fd16d6c
> # first bad commit: [b3edde44e5d4504c23a176819865cd603fd16d6c] cpufreq/schedutil: Use a fixed reference frequency

Re: [PATCH v7 3/7] cpufreq/schedutil: Use a fixed reference frequency

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 215 bytes --]

On Tue, Jan 23, 2024 at 08:24:00AM +0100, Vincent Guittot wrote:

> Could you tried this fix:
> https://lore.kernel.org/lkml/20240117190545.596057-1-vincent.guittot@linaro.org/

That seems to fix the issue, thanks.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]