Merge back earlier cpufreq material for 3.19-rc1.

Documentation/cpu-freq/intel-pstate.txt

 Intel P-state driver
 --------------------
 
-This driver implements a scaling driver with an internal governor for
-Intel Core processors.  The driver follows the same model as the
-Transmeta scaling driver (longrun.c) and implements the setpolicy()
-instead of target().  Scaling drivers that implement setpolicy() are
-assumed to implement internal governors by the cpufreq core. All the
-logic for selecting the current P state is contained within the
-driver; no external governor is used by the cpufreq core.
-
-Intel SandyBridge+ processors are supported.
-
-New sysfs files for controlling P state selection have been added to
+This driver provides an interface to control the P state selection for
+SandyBridge+ Intel processors.  The driver can operate two different
+modes based on the processor model legacy and Hardware P state (HWP)
+mode.
+
+In legacy mode the driver implements a scaling driver with an internal
+governor for Intel Core processors.  The driver follows the same model
+as the Transmeta scaling driver (longrun.c) and implements the
+setpolicy() instead of target().  Scaling drivers that implement
+setpolicy() are assumed to implement internal governors by the cpufreq
+core. All the logic for selecting the current P state is contained
+within the driver; no external governor is used by the cpufreq core.
+
+In HWP mode P state selection is implemented in the processor
+itself. The driver provides the interfaces between the cpufreq core and
+the processor to control P state selection based on user preferences
+and reporting frequency to the cpufreq core.  In this mode the
+internal governor code is disabled.
+
+In addtion to the interfaces provided by the cpufreq core for
+controlling frequency the driver provides sysfs files for
+controlling P state selection. These files have been added to
 /sys/devices/system/cpu/intel_pstate/
 
       max_perf_pct: limits the maximum P state that will be requested by
@@ -33,7 +44,9 @@ frequency is fiction for Intel Core processors. Even if the scaling
 driver selects a single P state the actual frequency the processor
 will run at is selected by the processor itself.
 
-New debugfs files have also been added to /sys/kernel/debug/pstate_snb/
+For legacy mode debugfs files have also been added to allow tuning of
+the internal governor algorythm. These files are located at
+/sys/kernel/debug/pstate_snb/ These files are NOT present in HWP mode.
 
       deadband
       d_gain_pct

Documentation/kernel-parameters.txt

@@ -1446,6 +1446,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 		       disable
 		         Do not enable intel_pstate as the default
 		         scaling driver for the supported processors
+		       no_hwp
+		         Do not enable hardware P state control (HWP)
+			 if available.
 
 	intremap=	[X86-64, Intel-IOMMU]
 			on	enable Interrupt Remapping (default)

arch/x86/include/asm/cpufeature.h

@@ -189,6 +189,11 @@
 #define X86_FEATURE_DTHERM	( 7*32+ 7) /* Digital Thermal Sensor */
 #define X86_FEATURE_HW_PSTATE	( 7*32+ 8) /* AMD HW-PState */
 #define X86_FEATURE_PROC_FEEDBACK ( 7*32+ 9) /* AMD ProcFeedbackInterface */
+#define X86_FEATURE_HWP		( 7*32+ 10) /* "hwp" Intel HWP */
+#define X86_FEATURE_HWP_NOITFY	( 7*32+ 11) /* Intel HWP_NOTIFY */
+#define X86_FEATURE_HWP_ACT_WINDOW ( 7*32+ 12) /* Intel HWP_ACT_WINDOW */
+#define X86_FEATURE_HWP_EPP	( 7*32+13) /* Intel HWP_EPP */
+#define X86_FEATURE_HWP_PKG_REQ ( 7*32+14) /* Intel HWP_PKG_REQ */
 
 /* Virtualization flags: Linux defined, word 8 */
 #define X86_FEATURE_TPR_SHADOW  ( 8*32+ 0) /* Intel TPR Shadow */

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

@@ -152,6 +152,45 @@
 #define MSR_CC6_DEMOTION_POLICY_CONFIG	0x00000668
 #define MSR_MC6_DEMOTION_POLICY_CONFIG	0x00000669
 
+/* Hardware P state interface */
+#define MSR_PPERF			0x0000064e
+#define MSR_PERF_LIMIT_REASONS		0x0000064f
+#define MSR_PM_ENABLE			0x00000770
+#define MSR_HWP_CAPABILITIES		0x00000771
+#define MSR_HWP_REQUEST_PKG		0x00000772
+#define MSR_HWP_INTERRUPT		0x00000773
+#define MSR_HWP_REQUEST 		0x00000774
+#define MSR_HWP_STATUS			0x00000777
+
+/* CPUID.6.EAX */
+#define HWP_BASE_BIT			(1<<7)
+#define HWP_NOTIFICATIONS_BIT		(1<<8)
+#define HWP_ACTIVITY_WINDOW_BIT		(1<<9)
+#define HWP_ENERGY_PERF_PREFERENCE_BIT	(1<<10)
+#define HWP_PACKAGE_LEVEL_REQUEST_BIT	(1<<11)
+
+/* IA32_HWP_CAPABILITIES */
+#define HWP_HIGHEST_PERF(x)		(x & 0xff)
+#define HWP_GUARANTEED_PERF(x)		((x & (0xff << 8)) >>8)
+#define HWP_MOSTEFFICIENT_PERF(x)	((x & (0xff << 16)) >>16)
+#define HWP_LOWEST_PERF(x)		((x & (0xff << 24)) >>24)
+
+/* IA32_HWP_REQUEST */
+#define HWP_MIN_PERF(x) 		(x & 0xff)
+#define HWP_MAX_PERF(x) 		((x & 0xff) << 8)
+#define HWP_DESIRED_PERF(x)		((x & 0xff) << 16)
+#define HWP_ENERGY_PERF_PREFERENCE(x)	((x & 0xff) << 24)
+#define HWP_ACTIVITY_WINDOW(x)		((x & 0xff3) << 32)
+#define HWP_PACKAGE_CONTROL(x)		((x & 0x1) << 42)
+
+/* IA32_HWP_STATUS */
+#define HWP_GUARANTEED_CHANGE(x)	(x & 0x1)
+#define HWP_EXCURSION_TO_MINIMUM(x)	(x & 0x4)
+
+/* IA32_HWP_INTERRUPT */
+#define HWP_CHANGE_TO_GUARANTEED_INT(x)	(x & 0x1)
+#define HWP_EXCURSION_TO_MINIMUM_INT(x)	(x & 0x2)
+
 #define MSR_AMD64_MC0_MASK		0xc0010044
 
 #define MSR_IA32_MCx_CTL(x)		(MSR_IA32_MC0_CTL + 4*(x))
@@ -345,6 +384,8 @@
 
 #define MSR_IA32_TEMPERATURE_TARGET	0x000001a2
 
+#define MSR_MISC_PWR_MGMT		0x000001aa
+
 #define MSR_IA32_ENERGY_PERF_BIAS	0x000001b0
 #define ENERGY_PERF_BIAS_PERFORMANCE	0
 #define ENERGY_PERF_BIAS_NORMAL		6

arch/x86/kernel/cpu/scattered.c

@@ -36,6 +36,11 @@ void init_scattered_cpuid_features(struct cpuinfo_x86 *c)
 		{ X86_FEATURE_ARAT,		CR_EAX, 2, 0x00000006, 0 },
 		{ X86_FEATURE_PLN,		CR_EAX, 4, 0x00000006, 0 },
 		{ X86_FEATURE_PTS,		CR_EAX, 6, 0x00000006, 0 },
+		{ X86_FEATURE_HWP,		CR_EAX, 7, 0x00000006, 0 },
+		{ X86_FEATURE_HWP_NOITFY,	CR_EAX, 8, 0x00000006, 0 },
+		{ X86_FEATURE_HWP_ACT_WINDOW,	CR_EAX, 9, 0x00000006, 0 },
+		{ X86_FEATURE_HWP_EPP,		CR_EAX,10, 0x00000006, 0 },
+		{ X86_FEATURE_HWP_PKG_REQ,	CR_EAX,11, 0x00000006, 0 },
 		{ X86_FEATURE_APERFMPERF,	CR_ECX, 0, 0x00000006, 0 },
 		{ X86_FEATURE_EPB,		CR_ECX, 3, 0x00000006, 0 },
 		{ X86_FEATURE_HW_PSTATE,	CR_EDX, 7, 0x80000007, 0 },

drivers/cpufreq/Kconfig

@@ -63,7 +63,6 @@ config CPU_FREQ_DEFAULT_GOV_PERFORMANCE
 
 config CPU_FREQ_DEFAULT_GOV_POWERSAVE
 	bool "powersave"
-	depends on EXPERT
 	select CPU_FREQ_GOV_POWERSAVE
 	help
 	  Use the CPUFreq governor 'powersave' as default. This sets
@@ -183,6 +182,8 @@ config CPU_FREQ_GOV_CONSERVATIVE
 
 	  If in doubt, say N.
 
+comment "CPU frequency scaling drivers"
+
 config CPUFREQ_DT
 	tristate "Generic DT based cpufreq driver"
 	depends on HAVE_CLK && OF
@@ -196,19 +197,19 @@ config CPUFREQ_DT
 
 	  If in doubt, say N.
 
-menu "x86 CPU frequency scaling drivers"
-depends on X86
+if X86
 source "drivers/cpufreq/Kconfig.x86"
-endmenu
+endif
 
-menu "ARM CPU frequency scaling drivers"
-depends on ARM || ARM64
+if ARM || ARM64
 source "drivers/cpufreq/Kconfig.arm"
-endmenu
+endif
 
-menu "AVR32 CPU frequency scaling drivers"
-depends on AVR32
+if PPC32 || PPC64
+source "drivers/cpufreq/Kconfig.powerpc"
+endif
 
+if AVR32
 config AVR32_AT32AP_CPUFREQ
 	bool "CPU frequency driver for AT32AP"
 	depends on PLATFORM_AT32AP
@@ -216,12 +217,9 @@ config AVR32_AT32AP_CPUFREQ
 	help
 	  This enables the CPU frequency driver for AT32AP processors.
 	  If in doubt, say N.
+endif
 
-endmenu
-
-menu "CPUFreq processor drivers"
-depends on IA64
-
+if IA64
 config IA64_ACPI_CPUFREQ
 	tristate "ACPI Processor P-States driver"
 	depends on ACPI_PROCESSOR
@@ -232,12 +230,9 @@ config IA64_ACPI_CPUFREQ
 	For details, take a look at <file:Documentation/cpu-freq/>.
 
 	If in doubt, say N.
+endif
 
-endmenu
-
-menu "MIPS CPUFreq processor drivers"
-depends on MIPS
-
+if MIPS
 config LOONGSON2_CPUFREQ
 	tristate "Loongson2 CPUFreq Driver"
 	help
@@ -250,15 +245,18 @@ config LOONGSON2_CPUFREQ
 
 	  If in doubt, say N.
 
-endmenu
+config LOONGSON1_CPUFREQ
+	tristate "Loongson1 CPUFreq Driver"
+	help
+	  This option adds a CPUFreq driver for loongson1 processors which
+	  support software configurable cpu frequency.
 
-menu "PowerPC CPU frequency scaling drivers"
-depends on PPC32 || PPC64
-source "drivers/cpufreq/Kconfig.powerpc"
-endmenu
+	  For details, take a look at <file:Documentation/cpu-freq/>.
 
-menu "SPARC CPU frequency scaling drivers"
-depends on SPARC64
+	  If in doubt, say N.
+endif
+
+if SPARC64
 config SPARC_US3_CPUFREQ
 	tristate "UltraSPARC-III CPU Frequency driver"
 	help
@@ -276,10 +274,9 @@ config SPARC_US2E_CPUFREQ
 	  For details, take a look at <file:Documentation/cpu-freq>.
 
 	  If in doubt, say N.
-endmenu
+endif
 
-menu "SH CPU Frequency scaling"
-depends on SUPERH
+if SUPERH
 config SH_CPU_FREQ
 	tristate "SuperH CPU Frequency driver"
 	help
@@ -293,7 +290,7 @@ config SH_CPU_FREQ
 	  For details, take a look at <file:Documentation/cpu-freq>.
 
 	  If unsure, say N.
-endmenu
+endif
 
 endif
 endmenu

drivers/cpufreq/Kconfig.arm

@@ -247,3 +247,11 @@ config ARM_TEGRA_CPUFREQ
 	default y
 	help
 	  This adds the CPUFreq driver support for TEGRA SOCs.
+
+config ARM_PXA2xx_CPUFREQ
+	tristate "Intel PXA2xx CPUfreq driver"
+	depends on PXA27x || PXA25x
+	help
+	  This add the CPUFreq driver support for Intel PXA2xx SOCs.
+
+	  If in doubt, say N.

drivers/cpufreq/Makefile

@@ -61,8 +61,7 @@ obj-$(CONFIG_ARM_IMX6Q_CPUFREQ)		+= imx6q-cpufreq.o
 obj-$(CONFIG_ARM_INTEGRATOR)		+= integrator-cpufreq.o
 obj-$(CONFIG_ARM_KIRKWOOD_CPUFREQ)	+= kirkwood-cpufreq.o
 obj-$(CONFIG_ARM_OMAP2PLUS_CPUFREQ)	+= omap-cpufreq.o
-obj-$(CONFIG_PXA25x)			+= pxa2xx-cpufreq.o
-obj-$(CONFIG_PXA27x)			+= pxa2xx-cpufreq.o
+obj-$(CONFIG_ARM_PXA2xx_CPUFREQ)	+= pxa2xx-cpufreq.o
 obj-$(CONFIG_PXA3xx)			+= pxa3xx-cpufreq.o
 obj-$(CONFIG_ARM_S3C24XX_CPUFREQ)	+= s3c24xx-cpufreq.o
 obj-$(CONFIG_ARM_S3C24XX_CPUFREQ_DEBUGFS) += s3c24xx-cpufreq-debugfs.o
@@ -98,6 +97,7 @@ obj-$(CONFIG_CRIS_MACH_ARTPEC3)		+= cris-artpec3-cpufreq.o
 obj-$(CONFIG_ETRAXFS)			+= cris-etraxfs-cpufreq.o
 obj-$(CONFIG_IA64_ACPI_CPUFREQ)		+= ia64-acpi-cpufreq.o
 obj-$(CONFIG_LOONGSON2_CPUFREQ)		+= loongson2_cpufreq.o
+obj-$(CONFIG_LOONGSON1_CPUFREQ)		+= ls1x-cpufreq.o
 obj-$(CONFIG_SH_CPU_FREQ)		+= sh-cpufreq.o
 obj-$(CONFIG_SPARC_US2E_CPUFREQ)	+= sparc-us2e-cpufreq.o
 obj-$(CONFIG_SPARC_US3_CPUFREQ)		+= sparc-us3-cpufreq.o

drivers/cpufreq/cpufreq-dt.c

@@ -58,6 +58,8 @@ static int set_target(struct cpufreq_policy *policy, unsigned int index)
 	old_freq = clk_get_rate(cpu_clk) / 1000;
 
 	if (!IS_ERR(cpu_reg)) {
+		unsigned long opp_freq;
+
 		rcu_read_lock();
 		opp = dev_pm_opp_find_freq_ceil(cpu_dev, &freq_Hz);
 		if (IS_ERR(opp)) {
@@ -67,13 +69,16 @@ static int set_target(struct cpufreq_policy *policy, unsigned int index)
 			return PTR_ERR(opp);
 		}
 		volt = dev_pm_opp_get_voltage(opp);
+		opp_freq = dev_pm_opp_get_freq(opp);
 		rcu_read_unlock();
 		tol = volt * priv->voltage_tolerance / 100;
 		volt_old = regulator_get_voltage(cpu_reg);
+		dev_dbg(cpu_dev, "Found OPP: %ld kHz, %ld uV\n",
+			opp_freq / 1000, volt);
 	}
 
 	dev_dbg(cpu_dev, "%u MHz, %ld mV --> %u MHz, %ld mV\n",
-		old_freq / 1000, volt_old ? volt_old / 1000 : -1,
+		old_freq / 1000, (volt_old > 0) ? volt_old / 1000 : -1,
 		new_freq / 1000, volt ? volt / 1000 : -1);
 
 	/* scaling up?  scale voltage before frequency */
@@ -89,7 +94,7 @@ static int set_target(struct cpufreq_policy *policy, unsigned int index)
 	ret = clk_set_rate(cpu_clk, freq_exact);
 	if (ret) {
 		dev_err(cpu_dev, "failed to set clock rate: %d\n", ret);
-		if (!IS_ERR(cpu_reg))
+		if (!IS_ERR(cpu_reg) && volt_old > 0)
 			regulator_set_voltage_tol(cpu_reg, volt_old, tol);
 		return ret;
 	}
@@ -181,7 +186,6 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 {
 	struct cpufreq_dt_platform_data *pd;
 	struct cpufreq_frequency_table *freq_table;
-	struct thermal_cooling_device *cdev;
 	struct device_node *np;
 	struct private_data *priv;
 	struct device *cpu_dev;
@@ -264,20 +268,6 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 		goto out_free_priv;
 	}
 
-	/*
-	 * For now, just loading the cooling device;
-	 * thermal DT code takes care of matching them.
-	 */
-	if (of_find_property(np, "#cooling-cells", NULL)) {
-		cdev = of_cpufreq_cooling_register(np, cpu_present_mask);
-		if (IS_ERR(cdev))
-			dev_err(cpu_dev,
-				"running cpufreq without cooling device: %ld\n",
-				PTR_ERR(cdev));
-		else
-			priv->cdev = cdev;
-	}
-
 	priv->cpu_dev = cpu_dev;
 	priv->cpu_reg = cpu_reg;
 	policy->driver_data = priv;
@@ -287,7 +277,7 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 	if (ret) {
 		dev_err(cpu_dev, "%s: invalid frequency table: %d\n", __func__,
 			ret);
-		goto out_cooling_unregister;
+		goto out_free_cpufreq_table;
 	}
 
 	policy->cpuinfo.transition_latency = transition_latency;
@@ -300,8 +290,7 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 
 	return 0;
 
-out_cooling_unregister:
-	cpufreq_cooling_unregister(priv->cdev);
+out_free_cpufreq_table:
 	dev_pm_opp_free_cpufreq_table(cpu_dev, &freq_table);
 out_free_priv:
 	kfree(priv);
@@ -319,7 +308,8 @@ static int cpufreq_exit(struct cpufreq_policy *policy)
 {
 	struct private_data *priv = policy->driver_data;
 
-	cpufreq_cooling_unregister(priv->cdev);
+	if (priv->cdev)
+		cpufreq_cooling_unregister(priv->cdev);
 	dev_pm_opp_free_cpufreq_table(priv->cpu_dev, &policy->freq_table);
 	clk_put(policy->clk);
 	if (!IS_ERR(priv->cpu_reg))
@@ -329,6 +319,33 @@ static int cpufreq_exit(struct cpufreq_policy *policy)
 	return 0;
 }
 
+static void cpufreq_ready(struct cpufreq_policy *policy)
+{
+	struct private_data *priv = policy->driver_data;
+	struct device_node *np = of_node_get(priv->cpu_dev->of_node);
+
+	if (WARN_ON(!np))
+		return;
+
+	/*
+	 * For now, just loading the cooling device;
+	 * thermal DT code takes care of matching them.
+	 */
+	if (of_find_property(np, "#cooling-cells", NULL)) {
+		priv->cdev = of_cpufreq_cooling_register(np,
+							 policy->related_cpus);
+		if (IS_ERR(priv->cdev)) {
+			dev_err(priv->cpu_dev,
+				"running cpufreq without cooling device: %ld\n",
+				PTR_ERR(priv->cdev));
+
+			priv->cdev = NULL;
+		}
+	}
+
+	of_node_put(np);
+}
+
 static struct cpufreq_driver dt_cpufreq_driver = {
 	.flags = CPUFREQ_STICKY | CPUFREQ_NEED_INITIAL_FREQ_CHECK,
 	.verify = cpufreq_generic_frequency_table_verify,
@@ -336,6 +353,7 @@ static struct cpufreq_driver dt_cpufreq_driver = {
 	.get = cpufreq_generic_get,
 	.init = cpufreq_init,
 	.exit = cpufreq_exit,
+	.ready = cpufreq_ready,
 	.name = "cpufreq-dt",
 	.attr = cpufreq_generic_attr,
 };

drivers/cpufreq/cpufreq.c

@@ -535,7 +535,7 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy,
 static ssize_t store_##file_name					\
 (struct cpufreq_policy *policy, const char *buf, size_t count)		\
 {									\
-	int ret;							\
+	int ret, temp;							\
 	struct cpufreq_policy new_policy;				\
 									\
 	ret = cpufreq_get_policy(&new_policy, policy->cpu);		\
@@ -546,8 +546,10 @@ static ssize_t store_##file_name					\
 	if (ret != 1)							\
 		return -EINVAL;						\
 									\
+	temp = new_policy.object;					\
 	ret = cpufreq_set_policy(policy, &new_policy);		\
-	policy->user_policy.object = policy->object;			\
+	if (!ret)							\
+		policy->user_policy.object = temp;			\
 									\
 	return ret ? ret : count;					\
 }
@@ -898,46 +900,31 @@ static int cpufreq_add_dev_interface(struct cpufreq_policy *policy,
 	struct freq_attr **drv_attr;
 	int ret = 0;
 
-	/* prepare interface data */
-	ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq,
-				   &dev->kobj, "cpufreq");
-	if (ret)
-		return ret;
-
 	/* set up files for this cpu device */
 	drv_attr = cpufreq_driver->attr;
 	while ((drv_attr) && (*drv_attr)) {
 		ret = sysfs_create_file(&policy->kobj, &((*drv_attr)->attr));
 		if (ret)
-			goto err_out_kobj_put;
+			return ret;
 		drv_attr++;
 	}
 	if (cpufreq_driver->get) {
 		ret = sysfs_create_file(&policy->kobj, &cpuinfo_cur_freq.attr);
 		if (ret)
-			goto err_out_kobj_put;
+			return ret;
 	}
 
 	ret = sysfs_create_file(&policy->kobj, &scaling_cur_freq.attr);
 	if (ret)
-		goto err_out_kobj_put;
+		return ret;
 
 	if (cpufreq_driver->bios_limit) {
 		ret = sysfs_create_file(&policy->kobj, &bios_limit.attr);
 		if (ret)
-			goto err_out_kobj_put;
+			return ret;
 	}
 
-	ret = cpufreq_add_dev_symlink(policy);
-	if (ret)
-		goto err_out_kobj_put;
-
-	return ret;
-
-err_out_kobj_put:
-	kobject_put(&policy->kobj);
-	wait_for_completion(&policy->kobj_unregister);
-	return ret;
+	return cpufreq_add_dev_symlink(policy);
 }
 
 static void cpufreq_init_policy(struct cpufreq_policy *policy)
@@ -1196,6 +1183,8 @@ static int __cpufreq_add_dev(struct device *dev, struct subsys_interface *sif)
 		goto err_set_policy_cpu;
 	}
 
+	down_write(&policy->rwsem);
+
 	/* related cpus should atleast have policy->cpus */
 	cpumask_or(policy->related_cpus, policy->related_cpus, policy->cpus);
 
@@ -1208,9 +1197,17 @@ static int __cpufreq_add_dev(struct device *dev, struct subsys_interface *sif)
 	if (!recover_policy) {
 		policy->user_policy.min = policy->min;
 		policy->user_policy.max = policy->max;
+
+		/* prepare interface data */
+		ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq,
+					   &dev->kobj, "cpufreq");
+		if (ret) {
+			pr_err("%s: failed to init policy->kobj: %d\n",
+			       __func__, ret);
+			goto err_init_policy_kobj;
+		}
 	}
 
-	down_write(&policy->rwsem);
 	write_lock_irqsave(&cpufreq_driver_lock, flags);
 	for_each_cpu(j, policy->cpus)
 		per_cpu(cpufreq_cpu_data, j) = policy;
@@ -1288,8 +1285,13 @@ static int __cpufreq_add_dev(struct device *dev, struct subsys_interface *sif)
 	up_write(&policy->rwsem);
 
 	kobject_uevent(&policy->kobj, KOBJ_ADD);
+
 	up_read(&cpufreq_rwsem);
 
+	/* Callback for handling stuff after policy is ready */
+	if (cpufreq_driver->ready)
+		cpufreq_driver->ready(policy);
+
 	pr_debug("initialization complete\n");
 
 	return 0;
@@ -1301,6 +1303,11 @@ static int __cpufreq_add_dev(struct device *dev, struct subsys_interface *sif)
 		per_cpu(cpufreq_cpu_data, j) = NULL;
 	write_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
+	if (!recover_policy) {
+		kobject_put(&policy->kobj);
+		wait_for_completion(&policy->kobj_unregister);
+	}
+err_init_policy_kobj:
 	up_write(&policy->rwsem);
 
 	if (cpufreq_driver->exit)

drivers/cpufreq/intel_pstate.c

@@ -137,6 +137,7 @@ struct cpu_defaults {
 
 static struct pstate_adjust_policy pid_params;
 static struct pstate_funcs pstate_funcs;
+static int hwp_active;
 
 struct perf_limits {
 	int no_turbo;
@@ -244,6 +245,34 @@ static inline void update_turbo_state(void)
 		 cpu->pstate.max_pstate == cpu->pstate.turbo_pstate);
 }
 
+#define PCT_TO_HWP(x) (x * 255 / 100)
+static void intel_pstate_hwp_set(void)
+{
+	int min, max, cpu;
+	u64 value, freq;
+
+	get_online_cpus();
+
+	for_each_online_cpu(cpu) {
+		rdmsrl_on_cpu(cpu, MSR_HWP_REQUEST, &value);
+		min = PCT_TO_HWP(limits.min_perf_pct);
+		value &= ~HWP_MIN_PERF(~0L);
+		value |= HWP_MIN_PERF(min);
+
+		max = PCT_TO_HWP(limits.max_perf_pct);
+		if (limits.no_turbo) {
+			rdmsrl( MSR_HWP_CAPABILITIES, freq);
+			max = HWP_GUARANTEED_PERF(freq);
+		}
+
+		value &= ~HWP_MAX_PERF(~0L);
+		value |= HWP_MAX_PERF(max);
+		wrmsrl_on_cpu(cpu, MSR_HWP_REQUEST, value);
+	}
+
+	put_online_cpus();
+}
+
 /************************** debugfs begin ************************/
 static int pid_param_set(void *data, u64 val)
 {
@@ -279,6 +308,8 @@ static void __init intel_pstate_debug_expose_params(void)
 	struct dentry *debugfs_parent;
 	int i = 0;
 
+	if (hwp_active)
+		return;
 	debugfs_parent = debugfs_create_dir("pstate_snb", NULL);
 	if (IS_ERR_OR_NULL(debugfs_parent))
 		return;
@@ -329,8 +360,12 @@ static ssize_t store_no_turbo(struct kobject *a, struct attribute *b,
 		pr_warn("Turbo disabled by BIOS or unavailable on processor\n");
 		return -EPERM;
 	}
+
 	limits.no_turbo = clamp_t(int, input, 0, 1);
 
+	if (hwp_active)
+		intel_pstate_hwp_set();
+
 	return count;
 }
 
@@ -348,6 +383,8 @@ static ssize_t store_max_perf_pct(struct kobject *a, struct attribute *b,
 	limits.max_perf_pct = min(limits.max_policy_pct, limits.max_sysfs_pct);
 	limits.max_perf = div_fp(int_tofp(limits.max_perf_pct), int_tofp(100));
 
+	if (hwp_active)
+		intel_pstate_hwp_set();
 	return count;
 }
 
@@ -363,6 +400,8 @@ static ssize_t store_min_perf_pct(struct kobject *a, struct attribute *b,
 	limits.min_perf_pct = clamp_t(int, input, 0 , 100);
 	limits.min_perf = div_fp(int_tofp(limits.min_perf_pct), int_tofp(100));
 
+	if (hwp_active)
+		intel_pstate_hwp_set();
 	return count;
 }
 
@@ -395,8 +434,16 @@ static void __init intel_pstate_sysfs_expose_params(void)
 	rc = sysfs_create_group(intel_pstate_kobject, &intel_pstate_attr_group);
 	BUG_ON(rc);
 }
-
 /************************** sysfs end ************************/
+
+static void intel_pstate_hwp_enable(void)
+{
+	hwp_active++;
+	pr_info("intel_pstate HWP enabled\n");
+
+	wrmsrl( MSR_PM_ENABLE, 0x1);
+}
+
 static int byt_get_min_pstate(void)
 {
 	u64 value;
@@ -648,6 +695,14 @@ static inline void intel_pstate_sample(struct cpudata *cpu)
 	cpu->prev_mperf = mperf;
 }
 
+static inline void intel_hwp_set_sample_time(struct cpudata *cpu)
+{
+	int delay;
+
+	delay = msecs_to_jiffies(50);
+	mod_timer_pinned(&cpu->timer, jiffies + delay);
+}
+
 static inline void intel_pstate_set_sample_time(struct cpudata *cpu)
 {
 	int delay;
@@ -694,6 +749,14 @@ static inline void intel_pstate_adjust_busy_pstate(struct cpudata *cpu)
 	intel_pstate_set_pstate(cpu, cpu->pstate.current_pstate - ctl);
 }
 
+static void intel_hwp_timer_func(unsigned long __data)
+{
+	struct cpudata *cpu = (struct cpudata *) __data;
+
+	intel_pstate_sample(cpu);
+	intel_hwp_set_sample_time(cpu);
+}
+
 static void intel_pstate_timer_func(unsigned long __data)
 {
 	struct cpudata *cpu = (struct cpudata *) __data;
@@ -730,6 +793,7 @@ static const struct x86_cpu_id intel_pstate_cpu_ids[] = {
 	ICPU(0x3f, core_params),
 	ICPU(0x45, core_params),
 	ICPU(0x46, core_params),
+	ICPU(0x47, core_params),
 	ICPU(0x4c, byt_params),
 	ICPU(0x4f, core_params),
 	ICPU(0x56, core_params),
@@ -737,6 +801,11 @@ static const struct x86_cpu_id intel_pstate_cpu_ids[] = {
 };
 MODULE_DEVICE_TABLE(x86cpu, intel_pstate_cpu_ids);
 
+static const struct x86_cpu_id intel_pstate_cpu_oob_ids[] = {
+	ICPU(0x56, core_params),
+	{}
+};
+
 static int intel_pstate_init_cpu(unsigned int cpunum)
 {
 	struct cpudata *cpu;
@@ -753,9 +822,14 @@ static int intel_pstate_init_cpu(unsigned int cpunum)
 	intel_pstate_get_cpu_pstates(cpu);
 
 	init_timer_deferrable(&cpu->timer);
-	cpu->timer.function = intel_pstate_timer_func;
 	cpu->timer.data = (unsigned long)cpu;
 	cpu->timer.expires = jiffies + HZ/100;
+
+	if (!hwp_active)
+		cpu->timer.function = intel_pstate_timer_func;
+	else
+		cpu->timer.function = intel_hwp_timer_func;
+
 	intel_pstate_busy_pid_reset(cpu);
 	intel_pstate_sample(cpu);
 
@@ -792,6 +866,7 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 		limits.no_turbo = 0;
 		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_perf = div_fp(int_tofp(limits.min_perf_pct), int_tofp(100));
@@ -801,6 +876,9 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 	limits.max_perf_pct = min(limits.max_policy_pct, limits.max_sysfs_pct);
 	limits.max_perf = div_fp(int_tofp(limits.max_perf_pct), int_tofp(100));
 
+	if (hwp_active)
+		intel_pstate_hwp_set();
+
 	return 0;
 }
 
@@ -823,6 +901,9 @@ static void intel_pstate_stop_cpu(struct cpufreq_policy *policy)
 	pr_info("intel_pstate CPU %d exiting\n", cpu_num);
 
 	del_timer_sync(&all_cpu_data[cpu_num]->timer);
+	if (hwp_active)
+		return;
+
 	intel_pstate_set_pstate(cpu, cpu->pstate.min_pstate);
 }
 
@@ -866,6 +947,7 @@ static struct cpufreq_driver intel_pstate_driver = {
 };
 
 static int __initdata no_load;
+static int __initdata no_hwp;
 
 static int intel_pstate_msrs_not_valid(void)
 {
@@ -959,6 +1041,15 @@ static bool intel_pstate_platform_pwr_mgmt_exists(void)
 {
 	struct acpi_table_header hdr;
 	struct hw_vendor_info *v_info;
+	const struct x86_cpu_id *id;
+	u64 misc_pwr;
+
+	id = x86_match_cpu(intel_pstate_cpu_oob_ids);
+	if (id) {
+		rdmsrl(MSR_MISC_PWR_MGMT, misc_pwr);
+		if ( misc_pwr & (1 << 8))
+			return true;
+	}
 
 	if (acpi_disabled ||
 	    ACPI_FAILURE(acpi_get_table_header(ACPI_SIG_FADT, 0, &hdr)))
@@ -982,6 +1073,7 @@ static int __init intel_pstate_init(void)
 	int cpu, rc = 0;
 	const struct x86_cpu_id *id;
 	struct cpu_defaults *cpu_info;
+	struct cpuinfo_x86 *c = &boot_cpu_data;
 
 	if (no_load)
 		return -ENODEV;
@@ -1011,6 +1103,9 @@ static int __init intel_pstate_init(void)
 	if (!all_cpu_data)
 		return -ENOMEM;
 
+	if (cpu_has(c,X86_FEATURE_HWP) && !no_hwp)
+		intel_pstate_hwp_enable();
+
 	rc = cpufreq_register_driver(&intel_pstate_driver);
 	if (rc)
 		goto out;
@@ -1041,6 +1136,8 @@ static int __init intel_pstate_setup(char *str)
 
 	if (!strcmp(str, "disable"))
 		no_load = 1;
+	if (!strcmp(str, "no_hwp"))
+		no_hwp = 1;
 	return 0;
 }
 early_param("intel_pstate", intel_pstate_setup);

drivers/cpufreq/ls1x-cpufreq.c

+/*
+ * CPU Frequency Scaling for Loongson 1 SoC
+ *
+ * Copyright (C) 2014 Zhang, Keguang <keguang.zhang@gmail.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/cpu.h>
+#include <linux/cpufreq.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include <asm/mach-loongson1/cpufreq.h>
+#include <asm/mach-loongson1/loongson1.h>
+
+static struct {
+	struct device *dev;
+	struct clk *clk;	/* CPU clk */
+	struct clk *mux_clk;	/* MUX of CPU clk */
+	struct clk *pll_clk;	/* PLL clk */
+	struct clk *osc_clk;	/* OSC clk */
+	unsigned int max_freq;
+	unsigned int min_freq;
+} ls1x_cpufreq;
+
+static int ls1x_cpufreq_notifier(struct notifier_block *nb,
+				 unsigned long val, void *data)
+{
+	if (val == CPUFREQ_POSTCHANGE)
+		current_cpu_data.udelay_val = loops_per_jiffy;
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block ls1x_cpufreq_notifier_block = {
+	.notifier_call = ls1x_cpufreq_notifier
+};
+
+static int ls1x_cpufreq_target(struct cpufreq_policy *policy,
+			       unsigned int index)
+{
+	unsigned int old_freq, new_freq;
+
+	old_freq = policy->cur;
+	new_freq = policy->freq_table[index].frequency;
+
+	/*
+	 * The procedure of reconfiguring CPU clk is as below.
+	 *
+	 *  - Reparent CPU clk to OSC clk
+	 *  - Reset CPU clock (very important)
+	 *  - Reconfigure CPU DIV
+	 *  - Reparent CPU clk back to CPU DIV clk
+	 */
+
+	dev_dbg(ls1x_cpufreq.dev, "%u KHz --> %u KHz\n", old_freq, new_freq);
+	clk_set_parent(policy->clk, ls1x_cpufreq.osc_clk);
+	__raw_writel(__raw_readl(LS1X_CLK_PLL_DIV) | RST_CPU_EN | RST_CPU,
+		     LS1X_CLK_PLL_DIV);
+	__raw_writel(__raw_readl(LS1X_CLK_PLL_DIV) & ~(RST_CPU_EN | RST_CPU),
+		     LS1X_CLK_PLL_DIV);
+	clk_set_rate(ls1x_cpufreq.mux_clk, new_freq * 1000);
+	clk_set_parent(policy->clk, ls1x_cpufreq.mux_clk);
+
+	return 0;
+}
+
+static int ls1x_cpufreq_init(struct cpufreq_policy *policy)
+{
+	struct cpufreq_frequency_table *freq_tbl;
+	unsigned int pll_freq, freq;
+	int steps, i, ret;
+
+	pll_freq = clk_get_rate(ls1x_cpufreq.pll_clk) / 1000;
+
+	steps = 1 << DIV_CPU_WIDTH;
+	freq_tbl = kzalloc(sizeof(*freq_tbl) * steps, GFP_KERNEL);
+	if (!freq_tbl) {
+		dev_err(ls1x_cpufreq.dev,
+			"failed to alloc cpufreq_frequency_table\n");
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	for (i = 0; i < (steps - 1); i++) {
+		freq = pll_freq / (i + 1);
+		if ((freq < ls1x_cpufreq.min_freq) ||
+		    (freq > ls1x_cpufreq.max_freq))
+			freq_tbl[i].frequency = CPUFREQ_ENTRY_INVALID;
+		else
+			freq_tbl[i].frequency = freq;
+		dev_dbg(ls1x_cpufreq.dev,
+			"cpufreq table: index %d: frequency %d\n", i,
+			freq_tbl[i].frequency);
+	}
+	freq_tbl[i].frequency = CPUFREQ_TABLE_END;
+
+	policy->clk = ls1x_cpufreq.clk;
+	ret = cpufreq_generic_init(policy, freq_tbl, 0);
+	if (ret)
+		kfree(freq_tbl);
+out:
+	return ret;
+}
+
+static int ls1x_cpufreq_exit(struct cpufreq_policy *policy)
+{
+	kfree(policy->freq_table);
+	return 0;
+}
+
+static struct cpufreq_driver ls1x_cpufreq_driver = {
+	.name		= "cpufreq-ls1x",
+	.flags		= CPUFREQ_STICKY | CPUFREQ_NEED_INITIAL_FREQ_CHECK,
+	.verify		= cpufreq_generic_frequency_table_verify,
+	.target_index	= ls1x_cpufreq_target,
+	.get		= cpufreq_generic_get,
+	.init		= ls1x_cpufreq_init,
+	.exit		= ls1x_cpufreq_exit,
+	.attr		= cpufreq_generic_attr,
+};
+
+static int ls1x_cpufreq_remove(struct platform_device *pdev)
+{
+	cpufreq_unregister_notifier(&ls1x_cpufreq_notifier_block,
+				    CPUFREQ_TRANSITION_NOTIFIER);
+	cpufreq_unregister_driver(&ls1x_cpufreq_driver);
+
+	return 0;
+}
+
+static int ls1x_cpufreq_probe(struct platform_device *pdev)
+{
+	struct plat_ls1x_cpufreq *pdata = pdev->dev.platform_data;
+	struct clk *clk;
+	int ret;
+
+	if (!pdata || !pdata->clk_name || !pdata->osc_clk_name)
+		return -EINVAL;
+
+	ls1x_cpufreq.dev = &pdev->dev;
+
+	clk = devm_clk_get(&pdev->dev, pdata->clk_name);
+	if (IS_ERR(clk)) {
+		dev_err(ls1x_cpufreq.dev, "unable to get %s clock\n",
+			pdata->clk_name);
+		ret = PTR_ERR(clk);
+		goto out;
+	}
+	ls1x_cpufreq.clk = clk;
+
+	clk = clk_get_parent(clk);
+	if (IS_ERR(clk)) {
+		dev_err(ls1x_cpufreq.dev, "unable to get parent of %s clock\n",
+			__clk_get_name(ls1x_cpufreq.clk));
+		ret = PTR_ERR(clk);
+		goto out;
+	}
+	ls1x_cpufreq.mux_clk = clk;
+
+	clk = clk_get_parent(clk);
+	if (IS_ERR(clk)) {
+		dev_err(ls1x_cpufreq.dev, "unable to get parent of %s clock\n",
+			__clk_get_name(ls1x_cpufreq.mux_clk));
+		ret = PTR_ERR(clk);
+		goto out;
+	}
+	ls1x_cpufreq.pll_clk = clk;
+
+	clk = devm_clk_get(&pdev->dev, pdata->osc_clk_name);
+	if (IS_ERR(clk)) {
+		dev_err(ls1x_cpufreq.dev, "unable to get %s clock\n",
+			pdata->osc_clk_name);
+		ret = PTR_ERR(clk);
+		goto out;
+	}
+	ls1x_cpufreq.osc_clk = clk;
+
+	ls1x_cpufreq.max_freq = pdata->max_freq;
+	ls1x_cpufreq.min_freq = pdata->min_freq;
+
+	ret = cpufreq_register_driver(&ls1x_cpufreq_driver);
+	if (ret) {
+		dev_err(ls1x_cpufreq.dev,
+			"failed to register cpufreq driver: %d\n", ret);
+		goto out;
+	}
+
+	ret = cpufreq_register_notifier(&ls1x_cpufreq_notifier_block,
+					CPUFREQ_TRANSITION_NOTIFIER);
+
+	if (!ret)
+		goto out;
+
+	dev_err(ls1x_cpufreq.dev, "failed to register cpufreq notifier: %d\n",
+		ret);
+
+	cpufreq_unregister_driver(&ls1x_cpufreq_driver);
+out:
+	return ret;
+}
+
+static struct platform_driver ls1x_cpufreq_platdrv = {
+	.driver = {
+		.name	= "ls1x-cpufreq",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= ls1x_cpufreq_probe,
+	.remove		= ls1x_cpufreq_remove,
+};
+
+module_platform_driver(ls1x_cpufreq_platdrv);
+
+MODULE_AUTHOR("Kelvin Cheung <keguang.zhang@gmail.com>");
+MODULE_DESCRIPTION("Loongson 1 CPUFreq driver");
+MODULE_LICENSE("GPL");

drivers/cpufreq/pcc-cpufreq.c

@@ -603,6 +603,13 @@ static void __exit pcc_cpufreq_exit(void)
 	free_percpu(pcc_cpu_info);
 }
 
+static const struct acpi_device_id processor_device_ids[] = {
+	{ACPI_PROCESSOR_OBJECT_HID, },
+	{ACPI_PROCESSOR_DEVICE_HID, },
+	{},
+};
+MODULE_DEVICE_TABLE(acpi, processor_device_ids);
+
 MODULE_AUTHOR("Matthew Garrett, Naga Chumbalkar");
 MODULE_VERSION(PCC_VERSION);
 MODULE_DESCRIPTION("Processor Clocking Control interface driver");

include/linux/cpufreq.h

@@ -217,26 +217,26 @@ __ATTR(_name, 0644, show_##_name, store_##_name)
 
 
 struct cpufreq_driver {
-	char			name[CPUFREQ_NAME_LEN];
-	u8			flags;
-	void			*driver_data;
+	char		name[CPUFREQ_NAME_LEN];
+	u8		flags;
+	void		*driver_data;
 
 	/* needed by all drivers */
-	int	(*init)		(struct cpufreq_policy *policy);
-	int	(*verify)	(struct cpufreq_policy *policy);
+	int		(*init)(struct cpufreq_policy *policy);
+	int		(*verify)(struct cpufreq_policy *policy);
 
 	/* define one out of two */
-	int	(*setpolicy)	(struct cpufreq_policy *policy);
+	int		(*setpolicy)(struct cpufreq_policy *policy);
 
 	/*
 	 * On failure, should always restore frequency to policy->restore_freq
 	 * (i.e. old freq).
 	 */
-	int	(*target)	(struct cpufreq_policy *policy,	/* Deprecated */
-				 unsigned int target_freq,
-				 unsigned int relation);
-	int	(*target_index)	(struct cpufreq_policy *policy,
-				 unsigned int index);
+	int		(*target)(struct cpufreq_policy *policy,
+				  unsigned int target_freq,
+				  unsigned int relation);	/* Deprecated */
+	int		(*target_index)(struct cpufreq_policy *policy,
+					unsigned int index);
 	/*
 	 * Only for drivers with target_index() and CPUFREQ_ASYNC_NOTIFICATION
 	 * unset.
@@ -252,27 +252,31 @@ struct cpufreq_driver {
 	 * wish to switch to intermediate frequency for some target frequency.
 	 * In that case core will directly call ->target_index().
 	 */
-	unsigned int (*get_intermediate)(struct cpufreq_policy *policy,
-					 unsigned int index);
-	int	(*target_intermediate)(struct cpufreq_policy *policy,
-				       unsigned int index);
+	unsigned int	(*get_intermediate)(struct cpufreq_policy *policy,
+					    unsigned int index);
+	int		(*target_intermediate)(struct cpufreq_policy *policy,
+					       unsigned int index);
 
 	/* should be defined, if possible */
-	unsigned int	(*get)	(unsigned int cpu);
+	unsigned int	(*get)(unsigned int cpu);
 
 	/* optional */
-	int	(*bios_limit)	(int cpu, unsigned int *limit);
+	int		(*bios_limit)(int cpu, unsigned int *limit);
+
+	int		(*exit)(struct cpufreq_policy *policy);
+	void		(*stop_cpu)(struct cpufreq_policy *policy);
+	int		(*suspend)(struct cpufreq_policy *policy);
+	int		(*resume)(struct cpufreq_policy *policy);
+
+	/* Will be called after the driver is fully initialized */
+	void		(*ready)(struct cpufreq_policy *policy);
 
-	int	(*exit)		(struct cpufreq_policy *policy);
-	void	(*stop_cpu)	(struct cpufreq_policy *policy);
-	int	(*suspend)	(struct cpufreq_policy *policy);
-	int	(*resume)	(struct cpufreq_policy *policy);
-	struct freq_attr	**attr;
+	struct freq_attr **attr;
 
 	/* platform specific boost support code */
-	bool                    boost_supported;
-	bool                    boost_enabled;
-	int     (*set_boost)    (int state);
+	bool		boost_supported;
+	bool		boost_enabled;
+	int		(*set_boost)(int state);
 };
 
 /* flags */