Merge branch 'pm-cpufreq'

* pm-cpufreq: (46 commits)
  intel_pstate: provide option to only use intel_pstate with HWP
  cpufreq-dt: Drop unnecessary check before cpufreq_cooling_unregister() invocation
  cpufreq: Create for_each_governor()
  cpufreq: Create for_each_policy()
  cpufreq: Drop cpufreq_disabled() check from cpufreq_cpu_{get|put}()
  cpufreq: Set cpufreq_cpu_data to NULL before putting kobject
  intel_pstate: honor user space min_perf_pct override on resume
  intel_pstate: respect cpufreq policy request
  intel_pstate: Add num_pstates to sysfs
  intel_pstate: expose turbo range to sysfs
  intel_pstate: Add support for SkyLake
  cpufreq: stats: drop unnecessary locking
  cpufreq: stats: don't update stats on false notifiers
  cpufreq: stats: don't update stats from show_trans_table()
  cpufreq: stats: time_in_state can't be NULL in cpufreq_stats_update()
  cpufreq: stats: create sysfs group once we are ready
  cpufreq: remove CPUFREQ_UPDATE_POLICY_CPU notifications
  cpufreq: stats: drop 'cpu' field of struct cpufreq_stats
  cpufreq: Remove (now) unused 'last_cpu' from struct cpufreq_policy
  cpufreq: stats: rename 'struct cpufreq_stats' objects as 'stats'
  ...

Documentation/cpu-freq/intel-pstate.txt

@@ -37,6 +37,14 @@ controlling P state selection. These files have been added to
       no_turbo: limits the driver to selecting P states below the turbo
       frequency range.
 
+      turbo_pct: displays the percentage of the total performance that
+      is supported by hardware that is in the turbo range.  This number
+      is independent of whether turbo has been disabled or not.
+
+      num_pstates: displays the number of pstates that are supported
+      by hardware.  This number is independent of whether turbo has
+      been disabled or not.
+
 For contemporary Intel processors, the frequency is controlled by the
 processor itself and the P-states exposed to software are related to
 performance levels.  The idea that frequency can be set to a single

Documentation/kernel-parameters.txt

@@ -1470,6 +1470,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 		       no_hwp
 		         Do not enable hardware P state control (HWP)
 			 if available.
+		hwp_only
+			Only load intel_pstate on systems which support
+			hardware P state control (HWP) if available.
 
 	intremap=	[X86-64, Intel-IOMMU]
 			on	enable Interrupt Remapping (default)

arch/x86/include/uapi/asm/msr-index.h

@@ -358,6 +358,7 @@
 
 #define MSR_IA32_PERF_STATUS		0x00000198
 #define MSR_IA32_PERF_CTL		0x00000199
+#define INTEL_PERF_CTL_MASK		0xffff
 #define MSR_AMD_PSTATE_DEF_BASE		0xc0010064
 #define MSR_AMD_PERF_STATUS		0xc0010063
 #define MSR_AMD_PERF_CTL		0xc0010062

drivers/cpufreq/Kconfig.x86

@@ -57,6 +57,16 @@ config X86_ACPI_CPUFREQ_CPB
 	  By enabling this option the acpi_cpufreq driver provides the old
 	  entry in addition to the new boost ones, for compatibility reasons.
 
+config X86_SFI_CPUFREQ
+	tristate "SFI Performance-States driver"
+	depends on X86_INTEL_MID && SFI
+	help
+	  This adds a CPUFreq driver for some Silvermont based Intel Atom
+	  architectures like Z34xx and Z35xx which enumerate processor
+	  performance states through SFI.
+
+	  If in doubt, say N.
+
 config ELAN_CPUFREQ
 	tristate "AMD Elan SC400 and SC410"
 	depends on MELAN

drivers/cpufreq/Makefile

@@ -41,6 +41,7 @@ obj-$(CONFIG_X86_P4_CLOCKMOD)		+= p4-clockmod.o
 obj-$(CONFIG_X86_CPUFREQ_NFORCE2)	+= cpufreq-nforce2.o
 obj-$(CONFIG_X86_INTEL_PSTATE)		+= intel_pstate.o
 obj-$(CONFIG_X86_AMD_FREQ_SENSITIVITY)	+= amd_freq_sensitivity.o
+obj-$(CONFIG_X86_SFI_CPUFREQ)		+= sfi-cpufreq.o
 
 ##################################################################################
 # ARM SoC drivers

drivers/cpufreq/cpufreq-dt.c

@@ -320,8 +320,7 @@ static int cpufreq_exit(struct cpufreq_policy *policy)
 {
 	struct private_data *priv = policy->driver_data;
 
-	if (priv->cdev)
-		cpufreq_cooling_unregister(priv->cdev);
+	cpufreq_cooling_unregister(priv->cdev);
 	dev_pm_opp_free_cpufreq_table(priv->cpu_dev, &policy->freq_table);
 	of_free_opp_table(priv->cpu_dev);
 	clk_put(policy->clk);

drivers/cpufreq/cpufreq.c

@@ -27,9 +27,21 @@
 #include <linux/mutex.h>
 #include <linux/slab.h>
 #include <linux/suspend.h>
+#include <linux/syscore_ops.h>
 #include <linux/tick.h>
 #include <trace/events/power.h>
 
+/* Macros to iterate over lists */
+/* Iterate over online CPUs policies */
+static LIST_HEAD(cpufreq_policy_list);
+#define for_each_policy(__policy)				\
+	list_for_each_entry(__policy, &cpufreq_policy_list, policy_list)
+
+/* Iterate over governors */
+static LIST_HEAD(cpufreq_governor_list);
+#define for_each_governor(__governor)				\
+	list_for_each_entry(__governor, &cpufreq_governor_list, governor_list)
+
 /**
  * The "cpufreq driver" - the arch- or hardware-dependent low
  * level driver of CPUFreq support, and its spinlock. This lock
@@ -40,7 +52,6 @@ static DEFINE_PER_CPU(struct cpufreq_policy *, cpufreq_cpu_data);
 static DEFINE_PER_CPU(struct cpufreq_policy *, cpufreq_cpu_data_fallback);
 static DEFINE_RWLOCK(cpufreq_driver_lock);
 DEFINE_MUTEX(cpufreq_governor_lock);
-static LIST_HEAD(cpufreq_policy_list);
 
 /* This one keeps track of the previously set governor of a removed CPU */
 static DEFINE_PER_CPU(char[CPUFREQ_NAME_LEN], cpufreq_cpu_governor);
@@ -62,7 +73,7 @@ static DECLARE_RWSEM(cpufreq_rwsem);
 /* internal prototypes */
 static int __cpufreq_governor(struct cpufreq_policy *policy,
 		unsigned int event);
-static unsigned int __cpufreq_get(unsigned int cpu);
+static unsigned int __cpufreq_get(struct cpufreq_policy *policy);
 static void handle_update(struct work_struct *work);
 
 /**
@@ -93,7 +104,6 @@ void disable_cpufreq(void)
 {
 	off = 1;
 }
-static LIST_HEAD(cpufreq_governor_list);
 static DEFINE_MUTEX(cpufreq_governor_mutex);
 
 bool have_governor_per_policy(void)
@@ -202,7 +212,7 @@ struct cpufreq_policy *cpufreq_cpu_get(unsigned int cpu)
 	struct cpufreq_policy *policy = NULL;
 	unsigned long flags;
 
-	if (cpufreq_disabled() || (cpu >= nr_cpu_ids))
+	if (cpu >= nr_cpu_ids)
 		return NULL;
 
 	if (!down_read_trylock(&cpufreq_rwsem))
@@ -229,9 +239,6 @@ EXPORT_SYMBOL_GPL(cpufreq_cpu_get);
 
 void cpufreq_cpu_put(struct cpufreq_policy *policy)
 {
-	if (cpufreq_disabled())
-		return;
-
 	kobject_put(&policy->kobj);
 	up_read(&cpufreq_rwsem);
 }
@@ -249,12 +256,12 @@ EXPORT_SYMBOL_GPL(cpufreq_cpu_put);
  * systems as each CPU might be scaled differently. So, use the arch
  * per-CPU loops_per_jiffy value wherever possible.
  */
-#ifndef CONFIG_SMP
-static unsigned long l_p_j_ref;
-static unsigned int l_p_j_ref_freq;
-
 static void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci)
 {
+#ifndef CONFIG_SMP
+	static unsigned long l_p_j_ref;
+	static unsigned int l_p_j_ref_freq;
+
 	if (ci->flags & CPUFREQ_CONST_LOOPS)
 		return;
 
@@ -270,13 +277,8 @@ static void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci)
 		pr_debug("scaling loops_per_jiffy to %lu for frequency %u kHz\n",
 			 loops_per_jiffy, ci->new);
 	}
-}
-#else
-static inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci)
-{
-	return;
-}
 #endif
+}
 
 static void __cpufreq_notify_transition(struct cpufreq_policy *policy,
 		struct cpufreq_freqs *freqs, unsigned int state)
@@ -432,11 +434,11 @@ static ssize_t store_boost(struct kobject *kobj, struct attribute *attr,
 }
 define_one_global_rw(boost);
 
-static struct cpufreq_governor *__find_governor(const char *str_governor)
+static struct cpufreq_governor *find_governor(const char *str_governor)
 {
 	struct cpufreq_governor *t;
 
-	list_for_each_entry(t, &cpufreq_governor_list, governor_list)
+	for_each_governor(t)
 		if (!strncasecmp(str_governor, t->name, CPUFREQ_NAME_LEN))
 			return t;
 
@@ -463,12 +465,12 @@ static int cpufreq_parse_governor(char *str_governor, unsigned int *policy,
 			*policy = CPUFREQ_POLICY_POWERSAVE;
 			err = 0;
 		}
-	} else if (has_target()) {
+	} else {
 		struct cpufreq_governor *t;
 
 		mutex_lock(&cpufreq_governor_mutex);
 
-		t = __find_governor(str_governor);
+		t = find_governor(str_governor);
 
 		if (t == NULL) {
 			int ret;
@@ -478,7 +480,7 @@ static int cpufreq_parse_governor(char *str_governor, unsigned int *policy,
 			mutex_lock(&cpufreq_governor_mutex);
 
 			if (ret == 0)
-				t = __find_governor(str_governor);
+				t = find_governor(str_governor);
 		}
 
 		if (t != NULL) {
@@ -513,8 +515,7 @@ show_one(cpuinfo_transition_latency, cpuinfo.transition_latency);
 show_one(scaling_min_freq, min);
 show_one(scaling_max_freq, max);
 
-static ssize_t show_scaling_cur_freq(
-	struct cpufreq_policy *policy, char *buf)
+static ssize_t show_scaling_cur_freq(struct cpufreq_policy *policy, char *buf)
 {
 	ssize_t ret;
 
@@ -563,7 +564,7 @@ store_one(scaling_max_freq, max);
 static ssize_t show_cpuinfo_cur_freq(struct cpufreq_policy *policy,
 					char *buf)
 {
-	unsigned int cur_freq = __cpufreq_get(policy->cpu);
+	unsigned int cur_freq = __cpufreq_get(policy);
 	if (!cur_freq)
 		return sprintf(buf, "<unknown>");
 	return sprintf(buf, "%u\n", cur_freq);
@@ -639,7 +640,7 @@ static ssize_t show_scaling_available_governors(struct cpufreq_policy *policy,
 		goto out;
 	}
 
-	list_for_each_entry(t, &cpufreq_governor_list, governor_list) {
+	for_each_governor(t) {
 		if (i >= (ssize_t) ((PAGE_SIZE / sizeof(char))
 		    - (CPUFREQ_NAME_LEN + 2)))
 			goto out;
@@ -902,7 +903,7 @@ static int cpufreq_add_dev_interface(struct cpufreq_policy *policy,
 
 	/* set up files for this cpu device */
 	drv_attr = cpufreq_driver->attr;
-	while ((drv_attr) && (*drv_attr)) {
+	while (drv_attr && *drv_attr) {
 		ret = sysfs_create_file(&policy->kobj, &((*drv_attr)->attr));
 		if (ret)
 			return ret;
@@ -936,7 +937,7 @@ static void cpufreq_init_policy(struct cpufreq_policy *policy)
 	memcpy(&new_policy, policy, sizeof(*policy));
 
 	/* Update governor of new_policy to the governor used before hotplug */
-	gov = __find_governor(per_cpu(cpufreq_cpu_governor, policy->cpu));
+	gov = find_governor(per_cpu(cpufreq_cpu_governor, policy->cpu));
 	if (gov)
 		pr_debug("Restoring governor %s for cpu %d\n",
 				policy->governor->name, policy->cpu);
@@ -958,7 +959,6 @@ static void cpufreq_init_policy(struct cpufreq_policy *policy)
 	}
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
 static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy,
 				  unsigned int cpu, struct device *dev)
 {
@@ -996,7 +996,6 @@ static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy,
 
 	return sysfs_create_link(&dev->kobj, &policy->kobj, "cpufreq");
 }
-#endif
 
 static struct cpufreq_policy *cpufreq_policy_restore(unsigned int cpu)
 {
@@ -1033,6 +1032,8 @@ static struct cpufreq_policy *cpufreq_policy_alloc(void)
 	init_rwsem(&policy->rwsem);
 	spin_lock_init(&policy->transition_lock);
 	init_waitqueue_head(&policy->transition_wait);
+	init_completion(&policy->kobj_unregister);
+	INIT_WORK(&policy->update, handle_update);
 
 	return policy;
 
@@ -1091,15 +1092,9 @@ static int update_policy_cpu(struct cpufreq_policy *policy, unsigned int cpu,
 	}
 
 	down_write(&policy->rwsem);
-
-	policy->last_cpu = policy->cpu;
 	policy->cpu = cpu;
-
 	up_write(&policy->rwsem);
 
-	blocking_notifier_call_chain(&cpufreq_policy_notifier_list,
-			CPUFREQ_UPDATE_POLICY_CPU, policy);
-
 	return 0;
 }
 
@@ -1110,41 +1105,32 @@ static int __cpufreq_add_dev(struct device *dev, struct subsys_interface *sif)
 	struct cpufreq_policy *policy;
 	unsigned long flags;
 	bool recover_policy = cpufreq_suspended;
-#ifdef CONFIG_HOTPLUG_CPU
-	struct cpufreq_policy *tpolicy;
-#endif
 
 	if (cpu_is_offline(cpu))
 		return 0;
 
 	pr_debug("adding CPU %u\n", cpu);
 
-#ifdef CONFIG_SMP
 	/* check whether a different CPU already registered this
 	 * CPU because it is in the same boat. */
-	policy = cpufreq_cpu_get(cpu);
-	if (unlikely(policy)) {
-		cpufreq_cpu_put(policy);
+	policy = cpufreq_cpu_get_raw(cpu);
+	if (unlikely(policy))
 		return 0;
-	}
-#endif
 
 	if (!down_read_trylock(&cpufreq_rwsem))
 		return 0;
 
-#ifdef CONFIG_HOTPLUG_CPU
 	/* Check if this cpu was hot-unplugged earlier and has siblings */
 	read_lock_irqsave(&cpufreq_driver_lock, flags);
-	list_for_each_entry(tpolicy, &cpufreq_policy_list, policy_list) {
-		if (cpumask_test_cpu(cpu, tpolicy->related_cpus)) {
+	for_each_policy(policy) {
+		if (cpumask_test_cpu(cpu, policy->related_cpus)) {
 			read_unlock_irqrestore(&cpufreq_driver_lock, flags);
-			ret = cpufreq_add_policy_cpu(tpolicy, cpu, dev);
+			ret = cpufreq_add_policy_cpu(policy, cpu, dev);
 			up_read(&cpufreq_rwsem);
 			return ret;
 		}
 	}
 	read_unlock_irqrestore(&cpufreq_driver_lock, flags);
-#endif
 
 	/*
 	 * Restore the saved policy when doing light-weight init and fall back
@@ -1171,9 +1157,6 @@ static int __cpufreq_add_dev(struct device *dev, struct subsys_interface *sif)
 
 	cpumask_copy(policy->cpus, cpumask_of(cpu));
 
-	init_completion(&policy->kobj_unregister);
-	INIT_WORK(&policy->update, handle_update);
-
 	/* call driver. From then on the cpufreq must be able
 	 * to accept all calls to ->verify and ->setpolicy for this CPU
 	 */
@@ -1371,11 +1354,10 @@ static int __cpufreq_remove_dev_prepare(struct device *dev,
 			pr_err("%s: Failed to stop governor\n", __func__);
 			return ret;
 		}
-	}
 
-	if (!cpufreq_driver->setpolicy)
 		strncpy(per_cpu(cpufreq_cpu_governor, cpu),
 			policy->governor->name, CPUFREQ_NAME_LEN);
+	}
 
 	down_read(&policy->rwsem);
 	cpus = cpumask_weight(policy->cpus);
@@ -1416,9 +1398,10 @@ static int __cpufreq_remove_dev_finish(struct device *dev,
 	unsigned long flags;
 	struct cpufreq_policy *policy;
 
-	read_lock_irqsave(&cpufreq_driver_lock, flags);
+	write_lock_irqsave(&cpufreq_driver_lock, flags);
 	policy = per_cpu(cpufreq_cpu_data, cpu);
-	read_unlock_irqrestore(&cpufreq_driver_lock, flags);
+	per_cpu(cpufreq_cpu_data, cpu) = NULL;
+	write_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
 	if (!policy) {
 		pr_debug("%s: No cpu_data found\n", __func__);
@@ -1473,7 +1456,6 @@ static int __cpufreq_remove_dev_finish(struct device *dev,
 		}
 	}
 
-	per_cpu(cpufreq_cpu_data, cpu) = NULL;
 	return 0;
 }
 
@@ -1510,30 +1492,23 @@ static void handle_update(struct work_struct *work)
 /**
  *	cpufreq_out_of_sync - If actual and saved CPU frequency differs, we're
  *	in deep trouble.
- *	@cpu: cpu number
- *	@old_freq: CPU frequency the kernel thinks the CPU runs at
+ *	@policy: policy managing CPUs
  *	@new_freq: CPU frequency the CPU actually runs at
  *
  *	We adjust to current frequency first, and need to clean up later.
  *	So either call to cpufreq_update_policy() or schedule handle_update()).
  */
-static void cpufreq_out_of_sync(unsigned int cpu, unsigned int old_freq,
+static void cpufreq_out_of_sync(struct cpufreq_policy *policy,
 				unsigned int new_freq)
 {
-	struct cpufreq_policy *policy;
 	struct cpufreq_freqs freqs;
-	unsigned long flags;
 
 	pr_debug("Warning: CPU frequency out of sync: cpufreq and timing core thinks of %u, is %u kHz\n",
-		 old_freq, new_freq);
+		 policy->cur, new_freq);
 
-	freqs.old = old_freq;
+	freqs.old = policy->cur;
 	freqs.new = new_freq;
 
-	read_lock_irqsave(&cpufreq_driver_lock, flags);
-	policy = per_cpu(cpufreq_cpu_data, cpu);
-	read_unlock_irqrestore(&cpufreq_driver_lock, flags);
-
 	cpufreq_freq_transition_begin(policy, &freqs);
 	cpufreq_freq_transition_end(policy, &freqs, 0);
 }
@@ -1583,22 +1558,21 @@ unsigned int cpufreq_quick_get_max(unsigned int cpu)
 }
 EXPORT_SYMBOL(cpufreq_quick_get_max);
 
-static unsigned int __cpufreq_get(unsigned int cpu)
+static unsigned int __cpufreq_get(struct cpufreq_policy *policy)
 {
-	struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu);
 	unsigned int ret_freq = 0;
 
 	if (!cpufreq_driver->get)
 		return ret_freq;
 
-	ret_freq = cpufreq_driver->get(cpu);
+	ret_freq = cpufreq_driver->get(policy->cpu);
 
 	if (ret_freq && policy->cur &&
 		!(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) {
 		/* verify no discrepancy between actual and
 					saved value exists */
 		if (unlikely(ret_freq != policy->cur)) {
-			cpufreq_out_of_sync(cpu, policy->cur, ret_freq);
+			cpufreq_out_of_sync(policy, ret_freq);
 			schedule_work(&policy->update);
 		}
 	}
@@ -1619,7 +1593,7 @@ unsigned int cpufreq_get(unsigned int cpu)
 
 	if (policy) {
 		down_read(&policy->rwsem);
-		ret_freq = __cpufreq_get(cpu);
+		ret_freq = __cpufreq_get(policy);
 		up_read(&policy->rwsem);
 
 		cpufreq_cpu_put(policy);
@@ -1682,7 +1656,7 @@ void cpufreq_suspend(void)
 
 	pr_debug("%s: Suspending Governors\n", __func__);
 
-	list_for_each_entry(policy, &cpufreq_policy_list, policy_list) {
+	for_each_policy(policy) {
 		if (__cpufreq_governor(policy, CPUFREQ_GOV_STOP))
 			pr_err("%s: Failed to stop governor for policy: %p\n",
 				__func__, policy);
@@ -1716,7 +1690,7 @@ void cpufreq_resume(void)
 
 	pr_debug("%s: Resuming Governors\n", __func__);
 
-	list_for_each_entry(policy, &cpufreq_policy_list, policy_list) {
+	for_each_policy(policy) {
 		if (cpufreq_driver->resume && cpufreq_driver->resume(policy))
 			pr_err("%s: Failed to resume driver: %p\n", __func__,
 				policy);
@@ -2006,10 +1980,6 @@ int cpufreq_driver_target(struct cpufreq_policy *policy,
 }
 EXPORT_SYMBOL_GPL(cpufreq_driver_target);
 
-/*
- * when "event" is CPUFREQ_GOV_LIMITS
- */
-
 static int __cpufreq_governor(struct cpufreq_policy *policy,
 					unsigned int event)
 {
@@ -2107,7 +2077,7 @@ int cpufreq_register_governor(struct cpufreq_governor *governor)
 
 	governor->initialized = 0;
 	err = -EBUSY;
-	if (__find_governor(governor->name) == NULL) {
+	if (!find_governor(governor->name)) {
 		err = 0;
 		list_add(&governor->governor_list, &cpufreq_governor_list);
 	}
@@ -2307,8 +2277,7 @@ int cpufreq_update_policy(unsigned int cpu)
 			policy->cur = new_policy.cur;
 		} else {
 			if (policy->cur != new_policy.cur && has_target())
-				cpufreq_out_of_sync(cpu, policy->cur,
-								new_policy.cur);
+				cpufreq_out_of_sync(policy, new_policy.cur);
 		}
 	}
 
@@ -2364,7 +2333,7 @@ static int cpufreq_boost_set_sw(int state)
 	struct cpufreq_policy *policy;
 	int ret = -EINVAL;
 
-	list_for_each_entry(policy, &cpufreq_policy_list, policy_list) {
+	for_each_policy(policy) {
 		freq_table = cpufreq_frequency_get_table(policy->cpu);
 		if (freq_table) {
 			ret = cpufreq_frequency_table_cpuinfo(policy,
@@ -2454,9 +2423,6 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
 
 	pr_debug("trying to register driver %s\n", driver_data->name);
 
-	if (driver_data->setpolicy)
-		driver_data->flags |= CPUFREQ_CONST_LOOPS;
-
 	write_lock_irqsave(&cpufreq_driver_lock, flags);
 	if (cpufreq_driver) {
 		write_unlock_irqrestore(&cpufreq_driver_lock, flags);
@@ -2465,6 +2431,9 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
 	cpufreq_driver = driver_data;
 	write_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
+	if (driver_data->setpolicy)
+		driver_data->flags |= CPUFREQ_CONST_LOOPS;
+
 	if (cpufreq_boost_supported()) {
 		/*
 		 * Check if driver provides function to enable boost -
@@ -2485,23 +2454,12 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data)
 	if (ret)
 		goto err_boost_unreg;
 
-	if (!(cpufreq_driver->flags & CPUFREQ_STICKY)) {
-		int i;
-		ret = -ENODEV;
-
-		/* check for at least one working CPU */
-		for (i = 0; i < nr_cpu_ids; i++)
-			if (cpu_possible(i) && per_cpu(cpufreq_cpu_data, i)) {
-				ret = 0;
-				break;
-			}
-
+	if (!(cpufreq_driver->flags & CPUFREQ_STICKY) &&
+	    list_empty(&cpufreq_policy_list)) {
 		/* if all ->init() calls failed, unregister */
-		if (ret) {
-			pr_debug("no CPU initialized for driver %s\n",
-				 driver_data->name);
-			goto err_if_unreg;
-		}
+		pr_debug("%s: No CPU initialized for driver %s\n", __func__,
+			 driver_data->name);
+		goto err_if_unreg;
 	}
 
 	register_hotcpu_notifier(&cpufreq_cpu_notifier);
@@ -2556,6 +2514,14 @@ int cpufreq_unregister_driver(struct cpufreq_driver *driver)
 }
 EXPORT_SYMBOL_GPL(cpufreq_unregister_driver);
 
+/*
+ * Stop cpufreq at shutdown to make sure it isn't holding any locks
+ * or mutexes when secondary CPUs are halted.
+ */
+static struct syscore_ops cpufreq_syscore_ops = {
+	.shutdown = cpufreq_suspend,
+};
+
 static int __init cpufreq_core_init(void)
 {
 	if (cpufreq_disabled())
@@ -2564,6 +2530,8 @@ static int __init cpufreq_core_init(void)
 	cpufreq_global_kobject = kobject_create();
 	BUG_ON(!cpufreq_global_kobject);
 
+	register_syscore_ops(&cpufreq_syscore_ops);
+
 	return 0;
 }
 core_initcall(cpufreq_core_init);

drivers/cpufreq/cpufreq_stats.c

@@ -18,7 +18,6 @@
 static spinlock_t cpufreq_stats_lock;
 
 struct cpufreq_stats {
-	unsigned int cpu;
 	unsigned int total_trans;
 	unsigned long long last_time;
 	unsigned int max_state;
@@ -31,50 +30,33 @@ struct cpufreq_stats {
 #endif
 };
 
-static DEFINE_PER_CPU(struct cpufreq_stats *, cpufreq_stats_table);
-
-struct cpufreq_stats_attribute {
-	struct attribute attr;
-	ssize_t(*show) (struct cpufreq_stats *, char *);
-};
-
-static int cpufreq_stats_update(unsigned int cpu)
+static int cpufreq_stats_update(struct cpufreq_stats *stats)
 {
-	struct cpufreq_stats *stat;
-	unsigned long long cur_time;
+	unsigned long long cur_time = get_jiffies_64();
 
-	cur_time = get_jiffies_64();
 	spin_lock(&cpufreq_stats_lock);
-	stat = per_cpu(cpufreq_stats_table, cpu);
-	if (stat->time_in_state)
-		stat->time_in_state[stat->last_index] +=
-			cur_time - stat->last_time;
-	stat->last_time = cur_time;
+	stats->time_in_state[stats->last_index] += cur_time - stats->last_time;
+	stats->last_time = cur_time;
 	spin_unlock(&cpufreq_stats_lock);
 	return 0;
 }
 
 static ssize_t show_total_trans(struct cpufreq_policy *policy, char *buf)
 {
-	struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu);
-	if (!stat)
-		return 0;
-	return sprintf(buf, "%d\n",
-			per_cpu(cpufreq_stats_table, stat->cpu)->total_trans);
+	return sprintf(buf, "%d\n", policy->stats->total_trans);
 }
 
 static ssize_t show_time_in_state(struct cpufreq_policy *policy, char *buf)
 {
+	struct cpufreq_stats *stats = policy->stats;
 	ssize_t len = 0;
 	int i;
-	struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu);
-	if (!stat)
-		return 0;
-	cpufreq_stats_update(stat->cpu);
-	for (i = 0; i < stat->state_num; i++) {
-		len += sprintf(buf + len, "%u %llu\n", stat->freq_table[i],
+
+	cpufreq_stats_update(stats);
+	for (i = 0; i < stats->state_num; i++) {
+		len += sprintf(buf + len, "%u %llu\n", stats->freq_table[i],
 			(unsigned long long)
-			jiffies_64_to_clock_t(stat->time_in_state[i]));
+			jiffies_64_to_clock_t(stats->time_in_state[i]));
 	}
 	return len;
 }
@@ -82,38 +64,35 @@ static ssize_t show_time_in_state(struct cpufreq_policy *policy, char *buf)
 #ifdef CONFIG_CPU_FREQ_STAT_DETAILS
 static ssize_t show_trans_table(struct cpufreq_policy *policy, char *buf)
 {
+	struct cpufreq_stats *stats = policy->stats;
 	ssize_t len = 0;
 	int i, j;
 
-	struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu);
-	if (!stat)
-		return 0;
-	cpufreq_stats_update(stat->cpu);
 	len += snprintf(buf + len, PAGE_SIZE - len, "   From  :    To\n");
 	len += snprintf(buf + len, PAGE_SIZE - len, "         : ");
-	for (i = 0; i < stat->state_num; i++) {
+	for (i = 0; i < stats->state_num; i++) {
 		if (len >= PAGE_SIZE)
 			break;
 		len += snprintf(buf + len, PAGE_SIZE - len, "%9u ",
-				stat->freq_table[i]);
+				stats->freq_table[i]);
 	}
 	if (len >= PAGE_SIZE)
 		return PAGE_SIZE;
 
 	len += snprintf(buf + len, PAGE_SIZE - len, "\n");
 
-	for (i = 0; i < stat->state_num; i++) {
+	for (i = 0; i < stats->state_num; i++) {
 		if (len >= PAGE_SIZE)
 			break;
 
 		len += snprintf(buf + len, PAGE_SIZE - len, "%9u: ",
-				stat->freq_table[i]);
+				stats->freq_table[i]);
 
-		for (j = 0; j < stat->state_num; j++) {
+		for (j = 0; j < stats->state_num; j++) {
 			if (len >= PAGE_SIZE)
 				break;
 			len += snprintf(buf + len, PAGE_SIZE - len, "%9u ",
-					stat->trans_table[i*stat->max_state+j]);
+					stats->trans_table[i*stats->max_state+j]);
 		}
 		if (len >= PAGE_SIZE)
 			break;
@@ -142,28 +121,29 @@ static struct attribute_group stats_attr_group = {
 	.name = "stats"
 };
 
-static int freq_table_get_index(struct cpufreq_stats *stat, unsigned int freq)
+static int freq_table_get_index(struct cpufreq_stats *stats, unsigned int freq)
 {
 	int index;
-	for (index = 0; index < stat->max_state; index++)
-		if (stat->freq_table[index] == freq)
+	for (index = 0; index < stats->max_state; index++)
+		if (stats->freq_table[index] == freq)
 			return index;
 	return -1;
 }
 
 static void __cpufreq_stats_free_table(struct cpufreq_policy *policy)
 {
-	struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu);
+	struct cpufreq_stats *stats = policy->stats;
 
-	if (!stat)
+	/* Already freed */
+	if (!stats)
 		return;
 
-	pr_debug("%s: Free stat table\n", __func__);
+	pr_debug("%s: Free stats table\n", __func__);
 
 	sysfs_remove_group(&policy->kobj, &stats_attr_group);
-	kfree(stat->time_in_state);
-	kfree(stat);
-	per_cpu(cpufreq_stats_table, policy->cpu) = NULL;
+	kfree(stats->time_in_state);
+	kfree(stats);
+	policy->stats = NULL;
 }
 
 static void cpufreq_stats_free_table(unsigned int cpu)
@@ -174,37 +154,33 @@ static void cpufreq_stats_free_table(unsigned int cpu)
 	if (!policy)
 		return;
 
-	if (cpufreq_frequency_get_table(policy->cpu))
-		__cpufreq_stats_free_table(policy);
+	__cpufreq_stats_free_table(policy);
 
 	cpufreq_cpu_put(policy);
 }
 
 static int __cpufreq_stats_create_table(struct cpufreq_policy *policy)
 {
-	unsigned int i, count = 0, ret = 0;
-	struct cpufreq_stats *stat;
+	unsigned int i = 0, count = 0, ret = -ENOMEM;
+	struct cpufreq_stats *stats;
 	unsigned int alloc_size;
 	unsigned int cpu = policy->cpu;
 	struct cpufreq_frequency_table *pos, *table;
 
+	/* We need cpufreq table for creating stats table */
 	table = cpufreq_frequency_get_table(cpu);
 	if (unlikely(!table))
 		return 0;
 
-	if (per_cpu(cpufreq_stats_table, cpu))
-		return -EBUSY;
-	stat = kzalloc(sizeof(*stat), GFP_KERNEL);
-	if ((stat) == NULL)
-		return -ENOMEM;
-
-	ret = sysfs_create_group(&policy->kobj, &stats_attr_group);
-	if (ret)
-		goto error_out;
+	/* stats already initialized */
+	if (policy->stats)
+		return -EEXIST;
 
-	stat->cpu = cpu;
-	per_cpu(cpufreq_stats_table, cpu) = stat;
+	stats = kzalloc(sizeof(*stats), GFP_KERNEL);
+	if (!stats)
+		return -ENOMEM;
 
+	/* Find total allocation size */
 	cpufreq_for_each_valid_entry(pos, table)
 		count++;
 
@@ -213,32 +189,40 @@ static int __cpufreq_stats_create_table(struct cpufreq_policy *policy)
 #ifdef CONFIG_CPU_FREQ_STAT_DETAILS
 	alloc_size += count * count * sizeof(int);
 #endif
-	stat->max_state = count;
-	stat->time_in_state = kzalloc(alloc_size, GFP_KERNEL);
-	if (!stat->time_in_state) {
-		ret = -ENOMEM;
-		goto error_alloc;
-	}
-	stat->freq_table = (unsigned int *)(stat->time_in_state + count);
+
+	/* Allocate memory for time_in_state/freq_table/trans_table in one go */
+	stats->time_in_state = kzalloc(alloc_size, GFP_KERNEL);
+	if (!stats->time_in_state)
+		goto free_stat;
+
+	stats->freq_table = (unsigned int *)(stats->time_in_state + count);
 
 #ifdef CONFIG_CPU_FREQ_STAT_DETAILS
-	stat->trans_table = stat->freq_table + count;
+	stats->trans_table = stats->freq_table + count;
 #endif
-	i = 0;
+
+	stats->max_state = count;
+
+	/* Find valid-unique entries */
 	cpufreq_for_each_valid_entry(pos, table)
-		if (freq_table_get_index(stat, pos->frequency) == -1)
-			stat->freq_table[i++] = pos->frequency;
-	stat->state_num = i;
-	spin_lock(&cpufreq_stats_lock);
-	stat->last_time = get_jiffies_64();
-	stat->last_index = freq_table_get_index(stat, policy->cur);
-	spin_unlock(&cpufreq_stats_lock);
-	return 0;
-error_alloc:
-	sysfs_remove_group(&policy->kobj, &stats_attr_group);
-error_out:
-	kfree(stat);
-	per_cpu(cpufreq_stats_table, cpu) = NULL;
+		if (freq_table_get_index(stats, pos->frequency) == -1)
+			stats->freq_table[i++] = pos->frequency;
+
+	stats->state_num = i;
+	stats->last_time = get_jiffies_64();
+	stats->last_index = freq_table_get_index(stats, policy->cur);
+
+	policy->stats = stats;
+	ret = sysfs_create_group(&policy->kobj, &stats_attr_group);
+	if (!ret)
+		return 0;
+
+	/* We failed, release resources */
+	policy->stats = NULL;
+	kfree(stats->time_in_state);
+free_stat:
+	kfree(stats);
+
 	return ret;
 }
 
@@ -259,30 +243,12 @@ static void cpufreq_stats_create_table(unsigned int cpu)
 	cpufreq_cpu_put(policy);
 }
 
-static void cpufreq_stats_update_policy_cpu(struct cpufreq_policy *policy)
-{
-	struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table,
-			policy->last_cpu);
-
-	pr_debug("Updating stats_table for new_cpu %u from last_cpu %u\n",
-			policy->cpu, policy->last_cpu);
-	per_cpu(cpufreq_stats_table, policy->cpu) = per_cpu(cpufreq_stats_table,
-			policy->last_cpu);
-	per_cpu(cpufreq_stats_table, policy->last_cpu) = NULL;
-	stat->cpu = policy->cpu;
-}
-
 static int cpufreq_stat_notifier_policy(struct notifier_block *nb,
 		unsigned long val, void *data)
 {
 	int ret = 0;
 	struct cpufreq_policy *policy = data;
 
-	if (val == CPUFREQ_UPDATE_POLICY_CPU) {
-		cpufreq_stats_update_policy_cpu(policy);
-		return 0;
-	}
-
 	if (val == CPUFREQ_CREATE_POLICY)
 		ret = __cpufreq_stats_create_table(policy);
 	else if (val == CPUFREQ_REMOVE_POLICY)
@@ -295,35 +261,45 @@ static int cpufreq_stat_notifier_trans(struct notifier_block *nb,
 		unsigned long val, void *data)
 {
 	struct cpufreq_freqs *freq = data;
-	struct cpufreq_stats *stat;
+	struct cpufreq_policy *policy = cpufreq_cpu_get(freq->cpu);
+	struct cpufreq_stats *stats;
 	int old_index, new_index;
 
-	if (val != CPUFREQ_POSTCHANGE)
+	if (!policy) {
+		pr_err("%s: No policy found\n", __func__);
 		return 0;
+	}
 
-	stat = per_cpu(cpufreq_stats_table, freq->cpu);
-	if (!stat)
-		return 0;
+	if (val != CPUFREQ_POSTCHANGE)
+		goto put_policy;
 
-	old_index = stat->last_index;
-	new_index = freq_table_get_index(stat, freq->new);
+	if (!policy->stats) {
+		pr_debug("%s: No stats found\n", __func__);
+		goto put_policy;
+	}
 
-	/* We can't do stat->time_in_state[-1]= .. */
-	if (old_index == -1 || new_index == -1)
-		return 0;
+	stats = policy->stats;
+
+	old_index = stats->last_index;
+	new_index = freq_table_get_index(stats, freq->new);
 
-	cpufreq_stats_update(freq->cpu);
+	/* We can't do stats->time_in_state[-1]= .. */
+	if (old_index == -1 || new_index == -1)
+		goto put_policy;
 
 	if (old_index == new_index)
-		return 0;
+		goto put_policy;
 
-	spin_lock(&cpufreq_stats_lock);
-	stat->last_index = new_index;
+	cpufreq_stats_update(stats);
+
+	stats->last_index = new_index;
 #ifdef CONFIG_CPU_FREQ_STAT_DETAILS
-	stat->trans_table[old_index * stat->max_state + new_index]++;
+	stats->trans_table[old_index * stats->max_state + new_index]++;
 #endif
-	stat->total_trans++;
-	spin_unlock(&cpufreq_stats_lock);
+	stats->total_trans++;
+
+put_policy:
+	cpufreq_cpu_put(policy);
 	return 0;
 }
 
@@ -374,8 +350,7 @@ static void __exit cpufreq_stats_exit(void)
 }
 
 MODULE_AUTHOR("Zou Nan hai <nanhai.zou@intel.com>");
-MODULE_DESCRIPTION("'cpufreq_stats' - A driver to export cpufreq stats "
-				"through sysfs filesystem");
+MODULE_DESCRIPTION("Export cpufreq stats via sysfs");
 MODULE_LICENSE("GPL");
 
 module_init(cpufreq_stats_init);

drivers/cpufreq/intel_pstate.c

@@ -148,6 +148,8 @@ struct perf_limits {
 	int32_t min_perf;
 	int max_policy_pct;
 	int max_sysfs_pct;
+	int min_policy_pct;
+	int min_sysfs_pct;
 };
 
 static struct perf_limits limits = {
@@ -159,6 +161,8 @@ static struct perf_limits limits = {
 	.min_perf = 0,
 	.max_policy_pct = 100,
 	.max_sysfs_pct = 100,
+	.min_policy_pct = 0,
+	.min_sysfs_pct = 0,
 };
 
 static inline void pid_reset(struct _pid *pid, int setpoint, int busy,
@@ -338,6 +342,33 @@ static void __init intel_pstate_debug_expose_params(void)
 		return sprintf(buf, "%u\n", limits.object);		\
 	}
 
+static ssize_t show_turbo_pct(struct kobject *kobj,
+				struct attribute *attr, char *buf)
+{
+	struct cpudata *cpu;
+	int total, no_turbo, turbo_pct;
+	uint32_t turbo_fp;
+
+	cpu = all_cpu_data[0];
+
+	total = cpu->pstate.turbo_pstate - cpu->pstate.min_pstate + 1;
+	no_turbo = cpu->pstate.max_pstate - cpu->pstate.min_pstate + 1;
+	turbo_fp = div_fp(int_tofp(no_turbo), int_tofp(total));
+	turbo_pct = 100 - fp_toint(mul_fp(turbo_fp, int_tofp(100)));
+	return sprintf(buf, "%u\n", turbo_pct);
+}
+
+static ssize_t show_num_pstates(struct kobject *kobj,
+				struct attribute *attr, char *buf)
+{
+	struct cpudata *cpu;
+	int total;
+
+	cpu = all_cpu_data[0];
+	total = cpu->pstate.turbo_pstate - cpu->pstate.min_pstate + 1;
+	return sprintf(buf, "%u\n", total);
+}
+
 static ssize_t show_no_turbo(struct kobject *kobj,
 			     struct attribute *attr, char *buf)
 {
@@ -404,7 +435,9 @@ static ssize_t store_min_perf_pct(struct kobject *a, struct attribute *b,
 	ret = sscanf(buf, "%u", &input);
 	if (ret != 1)
 		return -EINVAL;
-	limits.min_perf_pct = clamp_t(int, input, 0 , 100);
+
+	limits.min_sysfs_pct = clamp_t(int, input, 0 , 100);
+	limits.min_perf_pct = max(limits.min_policy_pct, limits.min_sysfs_pct);
 	limits.min_perf = div_fp(int_tofp(limits.min_perf_pct), int_tofp(100));
 
 	if (hwp_active)
@@ -418,11 +451,15 @@ show_one(min_perf_pct, min_perf_pct);
 define_one_global_rw(no_turbo);
 define_one_global_rw(max_perf_pct);
 define_one_global_rw(min_perf_pct);
+define_one_global_ro(turbo_pct);
+define_one_global_ro(num_pstates);
 
 static struct attribute *intel_pstate_attributes[] = {
 	&no_turbo.attr,
 	&max_perf_pct.attr,
 	&min_perf_pct.attr,
+	&turbo_pct.attr,
+	&num_pstates.attr,
 	NULL
 };
 
@@ -825,6 +862,7 @@ static const struct x86_cpu_id intel_pstate_cpu_ids[] = {
 	ICPU(0x46, core_params),
 	ICPU(0x47, core_params),
 	ICPU(0x4c, byt_params),
+	ICPU(0x4e, core_params),
 	ICPU(0x4f, core_params),
 	ICPU(0x56, core_params),
 	{}
@@ -887,7 +925,9 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 	if (!policy->cpuinfo.max_freq)
 		return -ENODEV;
 
-	if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) {
+	if (policy->policy == CPUFREQ_POLICY_PERFORMANCE &&
+	    policy->max >= policy->cpuinfo.max_freq) {
+		limits.min_policy_pct = 100;
 		limits.min_perf_pct = 100;
 		limits.min_perf = int_tofp(1);
 		limits.max_policy_pct = 100;
@@ -897,8 +937,9 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 		return 0;
 	}
 
-	limits.min_perf_pct = (policy->min * 100) / policy->cpuinfo.max_freq;
-	limits.min_perf_pct = clamp_t(int, limits.min_perf_pct, 0 , 100);
+	limits.min_policy_pct = (policy->min * 100) / policy->cpuinfo.max_freq;
+	limits.min_policy_pct = clamp_t(int, limits.min_policy_pct, 0 , 100);
+	limits.min_perf_pct = max(limits.min_policy_pct, limits.min_sysfs_pct);
 	limits.min_perf = div_fp(int_tofp(limits.min_perf_pct), int_tofp(100));
 
 	limits.max_policy_pct = (policy->max * 100) / policy->cpuinfo.max_freq;
@@ -978,6 +1019,7 @@ static struct cpufreq_driver intel_pstate_driver = {
 
 static int __initdata no_load;
 static int __initdata no_hwp;
+static int __initdata hwp_only;
 static unsigned int force_load;
 
 static int intel_pstate_msrs_not_valid(void)
@@ -1175,6 +1217,9 @@ static int __init intel_pstate_init(void)
 	if (cpu_has(c,X86_FEATURE_HWP) && !no_hwp)
 		intel_pstate_hwp_enable();
 
+	if (!hwp_active && hwp_only)
+		goto out;
+
 	rc = cpufreq_register_driver(&intel_pstate_driver);
 	if (rc)
 		goto out;
@@ -1209,6 +1254,8 @@ static int __init intel_pstate_setup(char *str)
 		no_hwp = 1;
 	if (!strcmp(str, "force"))
 		force_load = 1;
+	if (!strcmp(str, "hwp_only"))
+		hwp_only = 1;
 	return 0;
 }
 early_param("intel_pstate", intel_pstate_setup);

drivers/cpufreq/ls1x-cpufreq.c

@@ -210,7 +210,6 @@ static int ls1x_cpufreq_probe(struct platform_device *pdev)
 static struct platform_driver ls1x_cpufreq_platdrv = {
 	.driver = {
 		.name	= "ls1x-cpufreq",
-		.owner	= THIS_MODULE,
 	},
 	.probe		= ls1x_cpufreq_probe,
 	.remove		= ls1x_cpufreq_remove,

drivers/cpufreq/sfi-cpufreq.c

+/*
+ *  SFI Performance States Driver
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License.
+ *
+ *  This program is distributed in the hope that it will be useful, but
+ *  WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ *  General Public License for more details.
+ *
+ *  Author: Vishwesh M Rudramuni <vishwesh.m.rudramuni@intel.com>
+ *  Author: Srinidhi Kasagar <srinidhi.kasagar@intel.com>
+ */
+
+#include <linux/cpufreq.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sfi.h>
+#include <linux/slab.h>
+#include <linux/smp.h>
+
+#include <asm/msr.h>
+
+struct cpufreq_frequency_table *freq_table;
+static struct sfi_freq_table_entry *sfi_cpufreq_array;
+static int num_freq_table_entries;
+
+static int sfi_parse_freq(struct sfi_table_header *table)
+{
+	struct sfi_table_simple *sb;
+	struct sfi_freq_table_entry *pentry;
+	int totallen;
+
+	sb = (struct sfi_table_simple *)table;
+	num_freq_table_entries = SFI_GET_NUM_ENTRIES(sb,
+			struct sfi_freq_table_entry);
+	if (num_freq_table_entries <= 1) {
+		pr_err("No p-states discovered\n");
+		return -ENODEV;
+	}
+
+	pentry = (struct sfi_freq_table_entry *)sb->pentry;
+	totallen = num_freq_table_entries * sizeof(*pentry);
+
+	sfi_cpufreq_array = kzalloc(totallen, GFP_KERNEL);
+	if (!sfi_cpufreq_array)
+		return -ENOMEM;
+
+	memcpy(sfi_cpufreq_array, pentry, totallen);
+
+	return 0;
+}
+
+static int sfi_cpufreq_target(struct cpufreq_policy *policy, unsigned int index)
+{
+	unsigned int next_perf_state = 0; /* Index into perf table */
+	u32 lo, hi;
+
+	next_perf_state = policy->freq_table[index].driver_data;
+
+	rdmsr_on_cpu(policy->cpu, MSR_IA32_PERF_CTL, &lo, &hi);
+	lo = (lo & ~INTEL_PERF_CTL_MASK) |
+		((u32) sfi_cpufreq_array[next_perf_state].ctrl_val &
+		INTEL_PERF_CTL_MASK);
+	wrmsr_on_cpu(policy->cpu, MSR_IA32_PERF_CTL, lo, hi);
+
+	return 0;
+}
+
+static int sfi_cpufreq_cpu_init(struct cpufreq_policy *policy)
+{
+	policy->shared_type = CPUFREQ_SHARED_TYPE_HW;
+	policy->cpuinfo.transition_latency = 100000;	/* 100us */
+
+	return cpufreq_table_validate_and_show(policy, freq_table);
+}
+
+static struct cpufreq_driver sfi_cpufreq_driver = {
+	.flags		= CPUFREQ_CONST_LOOPS,
+	.verify		= cpufreq_generic_frequency_table_verify,
+	.target_index	= sfi_cpufreq_target,
+	.init		= sfi_cpufreq_cpu_init,
+	.name		= "sfi-cpufreq",
+	.attr		= cpufreq_generic_attr,
+};
+
+static int __init sfi_cpufreq_init(void)
+{
+	int ret, i;
+
+	/* parse the freq table from SFI */
+	ret = sfi_table_parse(SFI_SIG_FREQ, NULL, NULL, sfi_parse_freq);
+	if (ret)
+		return ret;
+
+	freq_table = kzalloc(sizeof(*freq_table) *
+			(num_freq_table_entries + 1), GFP_KERNEL);
+	if (!freq_table) {
+		ret = -ENOMEM;
+		goto err_free_array;
+	}
+
+	for (i = 0; i < num_freq_table_entries; i++) {
+		freq_table[i].driver_data = i;
+		freq_table[i].frequency = sfi_cpufreq_array[i].freq_mhz * 1000;
+	}
+	freq_table[i].frequency = CPUFREQ_TABLE_END;
+
+	ret = cpufreq_register_driver(&sfi_cpufreq_driver);
+	if (ret)
+		goto err_free_tbl;
+
+	return ret;
+
+err_free_tbl:
+	kfree(freq_table);
+err_free_array:
+	kfree(sfi_cpufreq_array);
+	return ret;
+}
+late_initcall(sfi_cpufreq_init);
+
+static void __exit sfi_cpufreq_exit(void)
+{
+	cpufreq_unregister_driver(&sfi_cpufreq_driver);
+	kfree(freq_table);
+	kfree(sfi_cpufreq_array);
+}
+module_exit(sfi_cpufreq_exit);
+
+MODULE_AUTHOR("Vishwesh M Rudramuni <vishwesh.m.rudramuni@intel.com>");
+MODULE_DESCRIPTION("SFI Performance-States Driver");
+MODULE_LICENSE("GPL");

drivers/sfi/sfi_core.c

@@ -161,7 +161,7 @@ static int sfi_verify_table(struct sfi_table_header *table)
  * Check for common case that we can re-use mapping to SYST,
  * which requires syst_pa, syst_va to be initialized.
  */
-struct sfi_table_header *sfi_map_table(u64 pa)
+static struct sfi_table_header *sfi_map_table(u64 pa)
 {
 	struct sfi_table_header *th;
 	u32 length;
@@ -189,7 +189,7 @@ struct sfi_table_header *sfi_map_table(u64 pa)
  * Undoes effect of sfi_map_table() by unmapping table
  * if it did not completely fit on same page as SYST.
  */
-void sfi_unmap_table(struct sfi_table_header *th)
+static void sfi_unmap_table(struct sfi_table_header *th)
 {
 	if (!TABLE_ON_PAGE(syst_va, th, th->len))
 		sfi_unmap_memory(th, TABLE_ON_PAGE(th, th, th->len) ?

include/linux/cpufreq.h

@@ -66,8 +66,6 @@ struct cpufreq_policy {
 	unsigned int		shared_type; /* ACPI: ANY or ALL affected CPUs
 						should set cpufreq */
 	unsigned int		cpu;    /* cpu nr of CPU managing this policy */
-	unsigned int		last_cpu; /* cpu nr of previous CPU that managed
-					   * this policy */
 	struct clk		*clk;
 	struct cpufreq_cpuinfo	cpuinfo;/* see above */
 
@@ -113,6 +111,9 @@ struct cpufreq_policy {
 	wait_queue_head_t	transition_wait;
 	struct task_struct	*transition_task; /* Task which is doing the transition */
 
+	/* cpufreq-stats */
+	struct cpufreq_stats	*stats;
+
 	/* For cpufreq driver's internal use */
 	void			*driver_data;
 };
@@ -367,9 +368,8 @@ static inline void cpufreq_resume(void) {}
 #define CPUFREQ_INCOMPATIBLE		(1)
 #define CPUFREQ_NOTIFY			(2)
 #define CPUFREQ_START			(3)
-#define CPUFREQ_UPDATE_POLICY_CPU	(4)
-#define CPUFREQ_CREATE_POLICY		(5)
-#define CPUFREQ_REMOVE_POLICY		(6)
+#define CPUFREQ_CREATE_POLICY		(4)
+#define CPUFREQ_REMOVE_POLICY		(5)
 
 #ifdef CONFIG_CPU_FREQ
 int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list);