Merge tag 'pm-4.11-rc2' of git://git.kernel.org/pub/scm/linux/kernel/…

…git/rafael/linux-pm

Pull power management fixes from Rafael Wysocki:
 "These fix several issues in the intel_pstate driver and one issue in
  the schedutil cpufreq governor, clean up that governor a bit and hook
  up existing code for disabling cpufreq to a new kernel command line
  option.

  Specifics:

   - Three fixes for intel_pstate problems related to the passive mode
     (in which it acts as a regular cpufreq scaling driver), two for the
     handling of global P-state limits and one for the handling of the
     cpu_frequency tracepoint in that mode (Rafael Wysocki).

   - Three fixes for the handling of P-state limits in intel_pstate in
     the active mode (Rafael Wysocki).

   - Introduction of a new cpufreq.off=1 kernel command line argument
     that will disable cpufreq entirely if passed to the kernel and is
     simply hooked up to the existing code used by Xen (Len Brown).

   - Fix for the schedutil cpufreq governor to prevent it from using
     stale raw frequency values in configurations with mutiple CPUs
     sharing one policy object and a cleanup for it reducing its
     overhead slightly (Viresh Kumar)"

* tag 'pm-4.11-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  cpufreq: intel_pstate: Do not reinit performance limits in ->setpolicy
  cpufreq: intel_pstate: Fix intel_pstate_verify_policy()
  cpufreq: intel_pstate: Fix global settings in active mode
  cpufreq: Add the "cpufreq.off=1" cmdline option
  cpufreq: schedutil: Pass sg_policy to get_next_freq()
  cpufreq: schedutil: move cached_raw_freq to struct sugov_policy
  cpufreq: intel_pstate: Avoid triggering cpu_frequency tracepoint unnecessarily
  cpufreq: intel_pstate: Fix intel_cpufreq_verify_policy()
  cpufreq: intel_pstate: Do not use performance_limits in passive mode

Documentation/admin-guide/kernel-parameters.txt

@@ -653,6 +653,9 @@
 	cpuidle.off=1	[CPU_IDLE]
 			disable the cpuidle sub-system
 
+	cpufreq.off=1	[CPU_FREQ]
+			disable the cpufreq sub-system
+
 	cpu_init_udelay=N
 			[X86] Delay for N microsec between assert and de-assert
 			of APIC INIT to start processors.  This delay occurs

drivers/cpufreq/cpufreq.c

@@ -2532,4 +2532,5 @@ static int __init cpufreq_core_init(void)
 
 	return 0;
 }
+module_param(off, int, 0444);
 core_initcall(cpufreq_core_init);

drivers/cpufreq/intel_pstate.c

@@ -377,6 +377,7 @@ static void intel_pstate_set_performance_limits(struct perf_limits *limits)
 	intel_pstate_init_limits(limits);
 	limits->min_perf_pct = 100;
 	limits->min_perf = int_ext_tofp(1);
+	limits->min_sysfs_pct = 100;
 }
 
 static DEFINE_MUTEX(intel_pstate_driver_lock);
@@ -968,11 +969,20 @@ static int intel_pstate_resume(struct cpufreq_policy *policy)
 }
 
 static void intel_pstate_update_policies(void)
+	__releases(&intel_pstate_limits_lock)
+	__acquires(&intel_pstate_limits_lock)
 {
+	struct perf_limits *saved_limits = limits;
 	int cpu;
 
+	mutex_unlock(&intel_pstate_limits_lock);
+
 	for_each_possible_cpu(cpu)
 		cpufreq_update_policy(cpu);
+
+	mutex_lock(&intel_pstate_limits_lock);
+
+	limits = saved_limits;
 }
 
 /************************** debugfs begin ************************/
@@ -1180,10 +1190,10 @@ static ssize_t store_no_turbo(struct kobject *a, struct attribute *b,
 
 	limits->no_turbo = clamp_t(int, input, 0, 1);
 
-	mutex_unlock(&intel_pstate_limits_lock);
-
 	intel_pstate_update_policies();
 
+	mutex_unlock(&intel_pstate_limits_lock);
+
 	mutex_unlock(&intel_pstate_driver_lock);
 
 	return count;
@@ -1217,10 +1227,10 @@ static ssize_t store_max_perf_pct(struct kobject *a, struct attribute *b,
 				   limits->max_perf_pct);
 	limits->max_perf = div_ext_fp(limits->max_perf_pct, 100);
 
-	mutex_unlock(&intel_pstate_limits_lock);
-
 	intel_pstate_update_policies();
 
+	mutex_unlock(&intel_pstate_limits_lock);
+
 	mutex_unlock(&intel_pstate_driver_lock);
 
 	return count;
@@ -1254,10 +1264,10 @@ static ssize_t store_min_perf_pct(struct kobject *a, struct attribute *b,
 				   limits->min_perf_pct);
 	limits->min_perf = div_ext_fp(limits->min_perf_pct, 100);
 
-	mutex_unlock(&intel_pstate_limits_lock);
-
 	intel_pstate_update_policies();
 
+	mutex_unlock(&intel_pstate_limits_lock);
+
 	mutex_unlock(&intel_pstate_driver_lock);
 
 	return count;
@@ -1874,13 +1884,11 @@ static int intel_pstate_prepare_request(struct cpudata *cpu, int pstate)
 
 	intel_pstate_get_min_max(cpu, &min_perf, &max_perf);
 	pstate = clamp_t(int, pstate, min_perf, max_perf);
-	trace_cpu_frequency(pstate * cpu->pstate.scaling, cpu->cpu);
 	return pstate;
 }
 
 static void intel_pstate_update_pstate(struct cpudata *cpu, int pstate)
 {
-	pstate = intel_pstate_prepare_request(cpu, pstate);
 	if (pstate == cpu->pstate.current_pstate)
 		return;
 
@@ -1900,6 +1908,8 @@ static inline void intel_pstate_adjust_busy_pstate(struct cpudata *cpu)
 
 	update_turbo_state();
 
+	target_pstate = intel_pstate_prepare_request(cpu, target_pstate);
+	trace_cpu_frequency(target_pstate * cpu->pstate.scaling, cpu->cpu);
 	intel_pstate_update_pstate(cpu, target_pstate);
 
 	sample = &cpu->sample;
@@ -2132,16 +2142,11 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 	mutex_lock(&intel_pstate_limits_lock);
 
 	if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) {
+		pr_debug("set performance\n");
 		if (!perf_limits) {
 			limits = &performance_limits;
 			perf_limits = limits;
 		}
-		if (policy->max >= policy->cpuinfo.max_freq &&
-		    !limits->no_turbo) {
-			pr_debug("set performance\n");
-			intel_pstate_set_performance_limits(perf_limits);
-			goto out;
-		}
 	} else {
 		pr_debug("set powersave\n");
 		if (!perf_limits) {
@@ -2152,7 +2157,7 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 	}
 
 	intel_pstate_update_perf_limits(policy, perf_limits);
- out:
+
 	if (cpu->policy == CPUFREQ_POLICY_PERFORMANCE) {
 		/*
 		 * NOHZ_FULL CPUs need this as the governor callback may not
@@ -2198,9 +2203,9 @@ static int intel_pstate_verify_policy(struct cpufreq_policy *policy)
 		unsigned int max_freq, min_freq;
 
 		max_freq = policy->cpuinfo.max_freq *
-						limits->max_sysfs_pct / 100;
+					perf_limits->max_sysfs_pct / 100;
 		min_freq = policy->cpuinfo.max_freq *
-						limits->min_sysfs_pct / 100;
+					perf_limits->min_sysfs_pct / 100;
 		cpufreq_verify_within_limits(policy, min_freq, max_freq);
 	}
 
@@ -2243,13 +2248,8 @@ static int __intel_pstate_cpu_init(struct cpufreq_policy *policy)
 
 	cpu = all_cpu_data[policy->cpu];
 
-	/*
-	 * We need sane value in the cpu->perf_limits, so inherit from global
-	 * perf_limits limits, which are seeded with values based on the
-	 * CONFIG_CPU_FREQ_DEFAULT_GOV_*, during boot up.
-	 */
 	if (per_cpu_limits)
-		memcpy(cpu->perf_limits, limits, sizeof(struct perf_limits));
+		intel_pstate_init_limits(cpu->perf_limits);
 
 	policy->min = cpu->pstate.min_pstate * cpu->pstate.scaling;
 	policy->max = cpu->pstate.turbo_pstate * cpu->pstate.scaling;
@@ -2301,23 +2301,13 @@ static struct cpufreq_driver intel_pstate = {
 static int intel_cpufreq_verify_policy(struct cpufreq_policy *policy)
 {
 	struct cpudata *cpu = all_cpu_data[policy->cpu];
-	struct perf_limits *perf_limits = limits;
 
 	update_turbo_state();
 	policy->cpuinfo.max_freq = limits->turbo_disabled ?
 			cpu->pstate.max_freq : cpu->pstate.turbo_freq;
 
 	cpufreq_verify_within_cpu_limits(policy);
 
-	if (per_cpu_limits)
-		perf_limits = cpu->perf_limits;
-
-	mutex_lock(&intel_pstate_limits_lock);
-
-	intel_pstate_update_perf_limits(policy, perf_limits);
-
-	mutex_unlock(&intel_pstate_limits_lock);
-
 	return 0;
 }
 
@@ -2370,6 +2360,7 @@ static int intel_cpufreq_target(struct cpufreq_policy *policy,
 		wrmsrl_on_cpu(policy->cpu, MSR_IA32_PERF_CTL,
 			      pstate_funcs.get_val(cpu, target_pstate));
 	}
+	freqs.new = target_pstate * cpu->pstate.scaling;
 	cpufreq_freq_transition_end(policy, &freqs, false);
 
 	return 0;
@@ -2383,8 +2374,9 @@ static unsigned int intel_cpufreq_fast_switch(struct cpufreq_policy *policy,
 
 	target_freq = intel_cpufreq_turbo_update(cpu, policy, target_freq);
 	target_pstate = DIV_ROUND_UP(target_freq, cpu->pstate.scaling);
+	target_pstate = intel_pstate_prepare_request(cpu, target_pstate);
 	intel_pstate_update_pstate(cpu, target_pstate);
-	return target_freq;
+	return target_pstate * cpu->pstate.scaling;
 }
 
 static int intel_cpufreq_cpu_init(struct cpufreq_policy *policy)
@@ -2437,8 +2429,11 @@ static int intel_pstate_register_driver(void)
 
 	intel_pstate_init_limits(&powersave_limits);
 	intel_pstate_set_performance_limits(&performance_limits);
-	limits = IS_ENABLED(CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE) ?
-			&performance_limits : &powersave_limits;
+	if (IS_ENABLED(CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE) &&
+	    intel_pstate_driver == &intel_pstate)
+		limits = &performance_limits;
+	else
+		limits = &powersave_limits;
 
 	ret = cpufreq_register_driver(intel_pstate_driver);
 	if (ret) {

kernel/sched/cpufreq_schedutil.c

@@ -36,6 +36,7 @@ struct sugov_policy {
 	u64 last_freq_update_time;
 	s64 freq_update_delay_ns;
 	unsigned int next_freq;
+	unsigned int cached_raw_freq;
 
 	/* The next fields are only needed if fast switch cannot be used. */
 	struct irq_work irq_work;
@@ -52,7 +53,6 @@ struct sugov_cpu {
 	struct update_util_data update_util;
 	struct sugov_policy *sg_policy;
 
-	unsigned int cached_raw_freq;
 	unsigned long iowait_boost;
 	unsigned long iowait_boost_max;
 	u64 last_update;
@@ -116,7 +116,7 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,
 
 /**
  * get_next_freq - Compute a new frequency for a given cpufreq policy.
- * @sg_cpu: schedutil cpu object to compute the new frequency for.
+ * @sg_policy: schedutil policy object to compute the new frequency for.
  * @util: Current CPU utilization.
  * @max: CPU capacity.
  *
@@ -136,19 +136,18 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,
  * next_freq (as calculated above) is returned, subject to policy min/max and
  * cpufreq driver limitations.
  */
-static unsigned int get_next_freq(struct sugov_cpu *sg_cpu, unsigned long util,
-				  unsigned long max)
+static unsigned int get_next_freq(struct sugov_policy *sg_policy,
+				  unsigned long util, unsigned long max)
 {
-	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
 	struct cpufreq_policy *policy = sg_policy->policy;
 	unsigned int freq = arch_scale_freq_invariant() ?
 				policy->cpuinfo.max_freq : policy->cur;
 
 	freq = (freq + (freq >> 2)) * util / max;
 
-	if (freq == sg_cpu->cached_raw_freq && sg_policy->next_freq != UINT_MAX)
+	if (freq == sg_policy->cached_raw_freq && sg_policy->next_freq != UINT_MAX)
 		return sg_policy->next_freq;
-	sg_cpu->cached_raw_freq = freq;
+	sg_policy->cached_raw_freq = freq;
 	return cpufreq_driver_resolve_freq(policy, freq);
 }
 
@@ -213,7 +212,7 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
 	} else {
 		sugov_get_util(&util, &max);
 		sugov_iowait_boost(sg_cpu, &util, &max);
-		next_f = get_next_freq(sg_cpu, util, max);
+		next_f = get_next_freq(sg_policy, util, max);
 	}
 	sugov_update_commit(sg_policy, time, next_f);
 }
@@ -267,7 +266,7 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu,
 		sugov_iowait_boost(j_sg_cpu, &util, &max);
 	}
 
-	return get_next_freq(sg_cpu, util, max);
+	return get_next_freq(sg_policy, util, max);
 }
 
 static void sugov_update_shared(struct update_util_data *hook, u64 time,
@@ -580,6 +579,7 @@ static int sugov_start(struct cpufreq_policy *policy)
 	sg_policy->next_freq = UINT_MAX;
 	sg_policy->work_in_progress = false;
 	sg_policy->need_freq_update = false;
+	sg_policy->cached_raw_freq = 0;
 
 	for_each_cpu(cpu, policy->cpus) {
 		struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu);
@@ -590,7 +590,6 @@ static int sugov_start(struct cpufreq_policy *policy)
 			sg_cpu->max = 0;
 			sg_cpu->flags = SCHED_CPUFREQ_RT;
 			sg_cpu->last_update = 0;
-			sg_cpu->cached_raw_freq = 0;
 			sg_cpu->iowait_boost = 0;
 			sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq;
 			cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util,