Merge branch 'pm-cpufreq'

Merge cpufreq updates for 5.19-rc1:

 - Fix cpufreq governor clean up code to avoid using kfree() directly
   to free kobject-based items (Kevin Hao).

 - Prepare cpufreq for powerpc's asm/prom.h cleanup (Christophe Leroy).

 - Make intel_pstate notify frequency invariance code when no_turbo is
   turned on and off (Chen Yu).

 - Add Sapphire Rapids OOB mode support to intel_pstate (Srinivas
   Pandruvada).

 - Make cpufreq avoid unnecessary frequency updates due to mismatch
   between hardware and the frequency table (Viresh Kumar).

 - Make remove_cpu_dev_symlink() clear the real_cpus mask to simplify
   code (Viresh Kumar).

 - Rearrange cpufreq_offline() and cpufreq_remove_dev() to make the
   calling convention for some driver callbacks consistent (Rafael
   Wysocki).

 - Avoid accessing half-initialized cpufreq policies from the show()
   and store() sysfs functions (Schspa Shi).

 - Rearrange cpufreq_offline() to make the calling convention for some
   driver callbacks consistent (Schspa Shi).

 - Update CPPC handling in cpufreq (Pierre Gondois):

   * Add per_cpu efficiency_class to the CPPC driver.
   * Make the CPPC driver Register EM based on efficiency class
     information.
   * Adjust _OSC for flexible address space in the ACPI platform
     initialization code and always set CPPC _OSC bits if CPPC_LIB is
     supported.
   * Assume no transition latency if no PCCT in the CPPC driver.
   * Add fast_switch and dvfs_possible_from_any_cpu support to the CPPC
     driver.

* pm-cpufreq:
  cpufreq: CPPC: Enable dvfs_possible_from_any_cpu
  cpufreq: CPPC: Enable fast_switch
  ACPI: CPPC: Assume no transition latency if no PCCT
  ACPI: bus: Set CPPC _OSC bits for all and when CPPC_LIB is supported
  ACPI: CPPC: Check _OSC for flexible address space
  cpufreq: make interface functions and lock holding state clear
  cpufreq: Abort show()/store() for half-initialized policies
  cpufreq: Rearrange locking in cpufreq_remove_dev()
  cpufreq: Split cpufreq_offline()
  cpufreq: Reorganize checks in cpufreq_offline()
  cpufreq: Clear real_cpus mask from remove_cpu_dev_symlink()
  cpufreq: intel_pstate: Support Sapphire Rapids OOB mode
  Revert "cpufreq: Fix possible race in cpufreq online error path"
  cpufreq: CPPC: Register EM based on efficiency class information
  cpufreq: CPPC: Add per_cpu efficiency_class
  cpufreq: Avoid unnecessary frequency updates due to mismatch
  cpufreq: Fix possible race in cpufreq online error path
  cpufreq: intel_pstate: Handle no_turbo in frequency invariance
  cpufreq: Prepare cleanup of powerpc's asm/prom.h
  cpufreq: governor: Use kobject release() method to free dbs_data

1 files changed, 211 insertions, 0 deletions

diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c
index 82d370ae6a4a..d092c9bb4ba3 100644
--- a/drivers/cpufreq/cppc_cpufreq.c
+++ b/drivers/cpufreq/cppc_cpufreq.c
@@ -389,6 +389,27 @@ static int cppc_cpufreq_set_target(struct cpufreq_policy *policy,
 	return ret;
 }
 
+static unsigned int cppc_cpufreq_fast_switch(struct cpufreq_policy *policy,
+					      unsigned int target_freq)
+{
+	struct cppc_cpudata *cpu_data = policy->driver_data;
+	unsigned int cpu = policy->cpu;
+	u32 desired_perf;
+	int ret;
+
+	desired_perf = cppc_cpufreq_khz_to_perf(cpu_data, target_freq);
+	cpu_data->perf_ctrls.desired_perf = desired_perf;
+	ret = cppc_set_perf(cpu, &cpu_data->perf_ctrls);
+
+	if (ret) {
+		pr_debug("Failed to set target on CPU:%d. ret:%d\n",
+			 cpu, ret);
+		return 0;
+	}
+
+	return target_freq;
+}
+
 static int cppc_verify_policy(struct cpufreq_policy_data *policy)
 {
 	cpufreq_verify_within_cpu_limits(policy);
@@ -420,12 +441,197 @@ static unsigned int cppc_cpufreq_get_transition_delay_us(unsigned int cpu)
 	return cppc_get_transition_latency(cpu) / NSEC_PER_USEC;
 }
 
+static DEFINE_PER_CPU(unsigned int, efficiency_class);
+static void cppc_cpufreq_register_em(struct cpufreq_policy *policy);
+
+/* Create an artificial performance state every CPPC_EM_CAP_STEP capacity unit. */
+#define CPPC_EM_CAP_STEP	(20)
+/* Increase the cost value by CPPC_EM_COST_STEP every performance state. */
+#define CPPC_EM_COST_STEP	(1)
+/* Add a cost gap correspnding to the energy of 4 CPUs. */
+#define CPPC_EM_COST_GAP	(4 * SCHED_CAPACITY_SCALE * CPPC_EM_COST_STEP \
+				/ CPPC_EM_CAP_STEP)
+
+static unsigned int get_perf_level_count(struct cpufreq_policy *policy)
+{
+	struct cppc_perf_caps *perf_caps;
+	unsigned int min_cap, max_cap;
+	struct cppc_cpudata *cpu_data;
+	int cpu = policy->cpu;
+
+	cpu_data = policy->driver_data;
+	perf_caps = &cpu_data->perf_caps;
+	max_cap = arch_scale_cpu_capacity(cpu);
+	min_cap = div_u64(max_cap * perf_caps->lowest_perf, perf_caps->highest_perf);
+	if ((min_cap == 0) || (max_cap < min_cap))
+		return 0;
+	return 1 + max_cap / CPPC_EM_CAP_STEP - min_cap / CPPC_EM_CAP_STEP;
+}
+
+/*
+ * The cost is defined as:
+ *   cost = power * max_frequency / frequency
+ */
+static inline unsigned long compute_cost(int cpu, int step)
+{
+	return CPPC_EM_COST_GAP * per_cpu(efficiency_class, cpu) +
+			step * CPPC_EM_COST_STEP;
+}
+
+static int cppc_get_cpu_power(struct device *cpu_dev,
+		unsigned long *power, unsigned long *KHz)
+{
+	unsigned long perf_step, perf_prev, perf, perf_check;
+	unsigned int min_step, max_step, step, step_check;
+	unsigned long prev_freq = *KHz;
+	unsigned int min_cap, max_cap;
+	struct cpufreq_policy *policy;
+
+	struct cppc_perf_caps *perf_caps;
+	struct cppc_cpudata *cpu_data;
+
+	policy = cpufreq_cpu_get_raw(cpu_dev->id);
+	cpu_data = policy->driver_data;
+	perf_caps = &cpu_data->perf_caps;
+	max_cap = arch_scale_cpu_capacity(cpu_dev->id);
+	min_cap = div_u64(max_cap * perf_caps->lowest_perf,
+			perf_caps->highest_perf);
+
+	perf_step = CPPC_EM_CAP_STEP * perf_caps->highest_perf / max_cap;
+	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);
+	step = perf_prev / perf_step;
+
+	if (step > max_step)
+		return -EINVAL;
+
+	if (min_step == max_step) {
+		step = max_step;
+		perf = perf_caps->highest_perf;
+	} else if (step < min_step) {
+		step = min_step;
+		perf = perf_caps->lowest_perf;
+	} else {
+		step++;
+		if (step == max_step)
+			perf = perf_caps->highest_perf;
+		else
+			perf = step * perf_step;
+	}
+
+	*KHz = cppc_cpufreq_perf_to_khz(cpu_data, perf);
+	perf_check = cppc_cpufreq_khz_to_perf(cpu_data, *KHz);
+	step_check = perf_check / perf_step;
+
+	/*
+	 * To avoid bad integer approximation, check that new frequency value
+	 * increased and that the new frequency will be converted to the
+	 * desired step value.
+	 */
+	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);
+		step_check = perf_check / perf_step;
+	}
+
+	/*
+	 * With an artificial EM, only the cost value is used. Still the power
+	 * is populated such as 0 < power < EM_MAX_POWER. This allows to add
+	 * more sense to the artificial performance states.
+	 */
+	*power = compute_cost(cpu_dev->id, step);
+
+	return 0;
+}
+
+static int cppc_get_cpu_cost(struct device *cpu_dev, unsigned long KHz,
+		unsigned long *cost)
+{
+	unsigned long perf_step, perf_prev;
+	struct cppc_perf_caps *perf_caps;
+	struct cpufreq_policy *policy;
+	struct cppc_cpudata *cpu_data;
+	unsigned int max_cap;
+	int step;
+
+	policy = cpufreq_cpu_get_raw(cpu_dev->id);
+	cpu_data = policy->driver_data;
+	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_step = CPPC_EM_CAP_STEP * perf_caps->highest_perf / max_cap;
+	step = perf_prev / perf_step;
+
+	*cost = compute_cost(cpu_dev->id, step);
+
+	return 0;
+}
+
+static int populate_efficiency_class(void)
+{
+	struct acpi_madt_generic_interrupt *gicc;
+	DECLARE_BITMAP(used_classes, 256) = {};
+	int class, cpu, index;
+
+	for_each_possible_cpu(cpu) {
+		gicc = acpi_cpu_get_madt_gicc(cpu);
+		class = gicc->efficiency_class;
+		bitmap_set(used_classes, class, 1);
+	}
+
+	if (bitmap_weight(used_classes, 256) <= 1) {
+		pr_debug("Efficiency classes are all equal (=%d). "
+			"No EM registered", class);
+		return -EINVAL;
+	}
+
+	/*
+	 * Squeeze efficiency class values on [0:#efficiency_class-1].
+	 * Values are per spec in [0:255].
+	 */
+	index = 0;
+	for_each_set_bit(class, used_classes, 256) {
+		for_each_possible_cpu(cpu) {
+			gicc = acpi_cpu_get_madt_gicc(cpu);
+			if (gicc->efficiency_class == class)
+				per_cpu(efficiency_class, cpu) = index;
+		}
+		index++;
+	}
+	cppc_cpufreq_driver.register_em = cppc_cpufreq_register_em;
+
+	return 0;
+}
+
+static void cppc_cpufreq_register_em(struct cpufreq_policy *policy)
+{
+	struct cppc_cpudata *cpu_data;
+	struct em_data_callback em_cb =
+		EM_ADV_DATA_CB(cppc_get_cpu_power, cppc_get_cpu_cost);
+
+	cpu_data = policy->driver_data;
+	em_dev_register_perf_domain(get_cpu_device(policy->cpu),
+			get_perf_level_count(policy), &em_cb,
+			cpu_data->shared_cpu_map, 0);
+}
+
 #else
 
 static unsigned int cppc_cpufreq_get_transition_delay_us(unsigned int cpu)
 {
 	return cppc_get_transition_latency(cpu) / NSEC_PER_USEC;
 }
+static int populate_efficiency_class(void)
+{
+	return 0;
+}
+static void cppc_cpufreq_register_em(struct cpufreq_policy *policy)
+{
+}
 #endif
 
 
@@ -536,6 +742,9 @@ static int cppc_cpufreq_cpu_init(struct cpufreq_policy *policy)
 		goto out;
 	}
 
+	policy->fast_switch_possible = cppc_allow_fast_switch();
+	policy->dvfs_possible_from_any_cpu = true;
+
 	/*
 	 * If 'highest_perf' is greater than 'nominal_perf', we assume CPU Boost
 	 * is supported.
@@ -681,6 +890,7 @@ static struct cpufreq_driver cppc_cpufreq_driver = {
 	.verify = cppc_verify_policy,
 	.target = cppc_cpufreq_set_target,
 	.get = cppc_cpufreq_get_rate,
+	.fast_switch = cppc_cpufreq_fast_switch,
 	.init = cppc_cpufreq_cpu_init,
 	.exit = cppc_cpufreq_cpu_exit,
 	.set_boost = cppc_cpufreq_set_boost,
@@ -742,6 +952,7 @@ static int __init cppc_cpufreq_init(void)
 
 	cppc_check_hisi_workaround();
 	cppc_freq_invariance_init();
+	populate_efficiency_class();
 
 	ret = cpufreq_register_driver(&cppc_cpufreq_driver);
 	if (ret)