Merge branch 'pm-cpufreq'

* pm-cpufreq: (63 commits)
  intel_pstate: Clean up get_target_pstate_use_performance()
  intel_pstate: Use sample.core_avg_perf in get_avg_pstate()
  intel_pstate: Clarify average performance computation
  intel_pstate: Avoid unnecessary synchronize_sched() during initialization
  cpufreq: schedutil: Make default depend on CONFIG_SMP
  cpufreq: powernv: del_timer_sync when global and local pstate are equal
  cpufreq: powernv: Move smp_call_function_any() out of irq safe block
  intel_pstate: Clean up intel_pstate_get()
  cpufreq: schedutil: Make it depend on CONFIG_SMP
  cpufreq: governor: Fix handling of special cases in dbs_update()
  cpufreq: intel_pstate: Ignore _PPC processing under HWP
  cpufreq: arm_big_little: use generic OPP functions for {init, free}_opp_table
  cpufreq: tango: Use generic platdev driver
  cpufreq: Fix GOV_LIMITS handling for the userspace governor
  cpufreq: mvebu: Move cpufreq code into drivers/cpufreq/
  cpufreq: dt: Kill platform-data
  mvebu: Use dev_pm_opp_set_sharing_cpus() to mark OPP tables as shared
  cpufreq: dt: Identify cpu-sharing for platforms without operating-points-v2
  cpufreq: governor: Change confusing struct field and variable names
  cpufreq: intel_pstate: Enable PPC enforcement for servers
  ...

Documentation/kernel-parameters.txt

@@ -1661,6 +1661,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 		hwp_only
 			Only load intel_pstate on systems which support
 			hardware P state control (HWP) if available.
+		support_acpi_ppc
+			Enforce ACPI _PPC performance limits. If the Fixed ACPI
+			Description Table, specifies preferred power management
+			profile as "Enterprise Server" or "Performance Server",
+			then this feature is turned on by default.
 
 	intremap=	[X86-64, Intel-IOMMU]
 			on	enable Interrupt Remapping (default)

MAINTAINERS

@@ -1322,6 +1322,7 @@ F:	drivers/rtc/rtc-armada38x.c
 F:	arch/arm/boot/dts/armada*
 F:	arch/arm/boot/dts/kirkwood*
 F:	arch/arm64/boot/dts/marvell/armada*
+F:	drivers/cpufreq/mvebu-cpufreq.c
 
 
 ARM/Marvell Berlin SoC support

arch/arm/mach-berlin/berlin.c

@@ -18,11 +18,6 @@
 #include <asm/hardware/cache-l2x0.h>
 #include <asm/mach/arch.h>
 
-static void __init berlin_init_late(void)
-{
-	platform_device_register_simple("cpufreq-dt", -1, NULL, 0);
-}
-
 static const char * const berlin_dt_compat[] = {
 	"marvell,berlin",
 	NULL,
@@ -30,7 +25,6 @@ static const char * const berlin_dt_compat[] = {
 
 DT_MACHINE_START(BERLIN_DT, "Marvell Berlin")
 	.dt_compat	= berlin_dt_compat,
-	.init_late	= berlin_init_late,
 	/*
 	 * with DT probing for L2CCs, berlin_init_machine can be removed.
 	 * Note: 88DE3005 (Armada 1500-mini) uses pl310 l2cc

arch/arm/mach-exynos/exynos.c

@@ -213,33 +213,6 @@ static void __init exynos_init_irq(void)
 	exynos_map_pmu();
 }
 
-static const struct of_device_id exynos_cpufreq_matches[] = {
-	{ .compatible = "samsung,exynos3250", .data = "cpufreq-dt" },
-	{ .compatible = "samsung,exynos4210", .data = "cpufreq-dt" },
-	{ .compatible = "samsung,exynos4212", .data = "cpufreq-dt" },
-	{ .compatible = "samsung,exynos4412", .data = "cpufreq-dt" },
-	{ .compatible = "samsung,exynos5250", .data = "cpufreq-dt" },
-#ifndef CONFIG_BL_SWITCHER
-	{ .compatible = "samsung,exynos5420", .data = "cpufreq-dt" },
-	{ .compatible = "samsung,exynos5800", .data = "cpufreq-dt" },
-#endif
-	{ /* sentinel */ }
-};
-
-static void __init exynos_cpufreq_init(void)
-{
-	struct device_node *root = of_find_node_by_path("/");
-	const struct of_device_id *match;
-
-	match = of_match_node(exynos_cpufreq_matches, root);
-	if (!match) {
-		platform_device_register_simple("exynos-cpufreq", -1, NULL, 0);
-		return;
-	}
-
-	platform_device_register_simple(match->data, -1, NULL, 0);
-}
-
 static void __init exynos_dt_machine_init(void)
 {
 	/*
@@ -262,8 +235,6 @@ static void __init exynos_dt_machine_init(void)
 	    of_machine_is_compatible("samsung,exynos5250"))
 		platform_device_register(&exynos_cpuidle);
 
-	exynos_cpufreq_init();
-
 	of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL);
 }
 

arch/arm/mach-imx/imx27-dt.c

@@ -18,15 +18,6 @@
 #include "common.h"
 #include "mx27.h"
 
-static void __init imx27_dt_init(void)
-{
-	struct platform_device_info devinfo = { .name = "cpufreq-dt", };
-
-	of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL);
-
-	platform_device_register_full(&devinfo);
-}
-
 static const char * const imx27_dt_board_compat[] __initconst = {
 	"fsl,imx27",
 	NULL
@@ -36,6 +27,5 @@ DT_MACHINE_START(IMX27_DT, "Freescale i.MX27 (Device Tree Support)")
 	.map_io		= mx27_map_io,
 	.init_early	= imx27_init_early,
 	.init_irq	= mx27_init_irq,
-	.init_machine	= imx27_dt_init,
 	.dt_compat	= imx27_dt_board_compat,
 MACHINE_END

arch/arm/mach-imx/mach-imx51.c

@@ -50,13 +50,10 @@ static void __init imx51_ipu_mipi_setup(void)
 
 static void __init imx51_dt_init(void)
 {
-	struct platform_device_info devinfo = { .name = "cpufreq-dt", };
-
 	imx51_ipu_mipi_setup();
 	imx_src_init();
 
 	of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL);
-	platform_device_register_full(&devinfo);
 }
 
 static void __init imx51_init_late(void)

arch/arm/mach-imx/mach-imx53.c

@@ -40,8 +40,6 @@ static void __init imx53_dt_init(void)
 static void __init imx53_init_late(void)
 {
 	imx53_pm_init();
-
-	platform_device_register_simple("cpufreq-dt", -1, NULL, 0);
 }
 
 static const char * const imx53_dt_board_compat[] __initconst = {

arch/arm/mach-imx/mach-imx7d.c

@@ -105,11 +105,6 @@ static void __init imx7d_init_irq(void)
 	irqchip_init();
 }
 
-static void __init imx7d_init_late(void)
-{
-	platform_device_register_simple("cpufreq-dt", -1, NULL, 0);
-}
-
 static const char *const imx7d_dt_compat[] __initconst = {
 	"fsl,imx7d",
 	NULL,
@@ -117,7 +112,6 @@ static const char *const imx7d_dt_compat[] __initconst = {
 
 DT_MACHINE_START(IMX7D, "Freescale i.MX7 Dual (Device Tree)")
 	.init_irq	= imx7d_init_irq,
-	.init_late	= imx7d_init_late,
 	.init_machine	= imx7d_init_machine,
 	.dt_compat	= imx7d_dt_compat,
 MACHINE_END

arch/arm/mach-mvebu/pmsu.c

@@ -20,7 +20,6 @@
 
 #include <linux/clk.h>
 #include <linux/cpu_pm.h>
-#include <linux/cpufreq-dt.h>
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/io.h>
@@ -29,7 +28,6 @@
 #include <linux/of_address.h>
 #include <linux/of_device.h>
 #include <linux/platform_device.h>
-#include <linux/pm_opp.h>
 #include <linux/resource.h>
 #include <linux/slab.h>
 #include <linux/smp.h>
@@ -608,86 +606,3 @@ int mvebu_pmsu_dfs_request(int cpu)
 
 	return 0;
 }
-
-struct cpufreq_dt_platform_data cpufreq_dt_pd = {
-	.independent_clocks = true,
-};
-
-static int __init armada_xp_pmsu_cpufreq_init(void)
-{
-	struct device_node *np;
-	struct resource res;
-	int ret, cpu;
-
-	if (!of_machine_is_compatible("marvell,armadaxp"))
-		return 0;
-
-	/*
-	 * In order to have proper cpufreq handling, we need to ensure
-	 * that the Device Tree description of the CPU clock includes
-	 * the definition of the PMU DFS registers. If not, we do not
-	 * register the clock notifier and the cpufreq driver. This
-	 * piece of code is only for compatibility with old Device
-	 * Trees.
-	 */
-	np = of_find_compatible_node(NULL, NULL, "marvell,armada-xp-cpu-clock");
-	if (!np)
-		return 0;
-
-	ret = of_address_to_resource(np, 1, &res);
-	if (ret) {
-		pr_warn(FW_WARN "not enabling cpufreq, deprecated armada-xp-cpu-clock binding\n");
-		of_node_put(np);
-		return 0;
-	}
-
-	of_node_put(np);
-
-	/*
-	 * For each CPU, this loop registers the operating points
-	 * supported (which are the nominal CPU frequency and half of
-	 * it), and registers the clock notifier that will take care
-	 * of doing the PMSU part of a frequency transition.
-	 */
-	for_each_possible_cpu(cpu) {
-		struct device *cpu_dev;
-		struct clk *clk;
-		int ret;
-
-		cpu_dev = get_cpu_device(cpu);
-		if (!cpu_dev) {
-			pr_err("Cannot get CPU %d\n", cpu);
-			continue;
-		}
-
-		clk = clk_get(cpu_dev, 0);
-		if (IS_ERR(clk)) {
-			pr_err("Cannot get clock for CPU %d\n", cpu);
-			return PTR_ERR(clk);
-		}
-
-		/*
-		 * In case of a failure of dev_pm_opp_add(), we don't
-		 * bother with cleaning up the registered OPP (there's
-		 * no function to do so), and simply cancel the
-		 * registration of the cpufreq device.
-		 */
-		ret = dev_pm_opp_add(cpu_dev, clk_get_rate(clk), 0);
-		if (ret) {
-			clk_put(clk);
-			return ret;
-		}
-
-		ret = dev_pm_opp_add(cpu_dev, clk_get_rate(clk) / 2, 0);
-		if (ret) {
-			clk_put(clk);
-			return ret;
-		}
-	}
-
-	platform_device_register_data(NULL, "cpufreq-dt", -1,
-				      &cpufreq_dt_pd, sizeof(cpufreq_dt_pd));
-	return 0;
-}
-
-device_initcall(armada_xp_pmsu_cpufreq_init);

arch/arm/mach-omap2/pm.c

@@ -277,13 +277,10 @@ static void __init omap4_init_voltages(void)
 
 static inline void omap_init_cpufreq(void)
 {
-	struct platform_device_info devinfo = { };
+	struct platform_device_info devinfo = { .name = "omap-cpufreq" };
 
 	if (!of_have_populated_dt())
-		devinfo.name = "omap-cpufreq";
-	else
-		devinfo.name = "cpufreq-dt";
-	platform_device_register_full(&devinfo);
+		platform_device_register_full(&devinfo);
 }
 
 static int __init omap2_common_pm_init(void)

arch/arm/mach-rockchip/rockchip.c

@@ -74,7 +74,6 @@ static void __init rockchip_dt_init(void)
 {
 	rockchip_suspend_init();
 	of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL);
-	platform_device_register_simple("cpufreq-dt", 0, NULL, 0);
 }
 
 static const char * const rockchip_board_dt_compat[] = {

arch/arm/mach-shmobile/Makefile

@@ -38,7 +38,6 @@ smp-$(CONFIG_ARCH_EMEV2)	+= smp-emev2.o headsmp-scu.o platsmp-scu.o
 
 # PM objects
 obj-$(CONFIG_SUSPEND)		+= suspend.o
-obj-$(CONFIG_CPU_FREQ)		+= cpufreq.o
 obj-$(CONFIG_PM_RCAR)		+= pm-rcar.o
 obj-$(CONFIG_PM_RMOBILE)	+= pm-rmobile.o
 obj-$(CONFIG_ARCH_RCAR_GEN2)	+= pm-rcar-gen2.o

arch/arm/mach-shmobile/common.h

@@ -25,16 +25,9 @@ static inline int shmobile_suspend_init(void) { return 0; }
 static inline void shmobile_smp_apmu_suspend_init(void) { }
 #endif
 
-#ifdef CONFIG_CPU_FREQ
-int shmobile_cpufreq_init(void);
-#else
-static inline int shmobile_cpufreq_init(void) { return 0; }
-#endif
-
 static inline void __init shmobile_init_late(void)
 {
 	shmobile_suspend_init();
-	shmobile_cpufreq_init();
 }
 
 #endif /* __ARCH_MACH_COMMON_H */

arch/arm/mach-shmobile/cpufreq.c

@@ -1,19 +0,0 @@
-/*
- * CPUFreq support code for SH-Mobile ARM
- *
- *  Copyright (C) 2014 Gaku Inami
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-
-#include <linux/platform_device.h>
-
-#include "common.h"
-
-int __init shmobile_cpufreq_init(void)
-{
-	platform_device_register_simple("cpufreq-dt", -1, NULL, 0);
-	return 0;
-}

arch/arm/mach-sunxi/sunxi.c

@@ -17,11 +17,6 @@
 
 #include <asm/mach/arch.h>
 
-static void __init sunxi_dt_cpufreq_init(void)
-{
-	platform_device_register_simple("cpufreq-dt", -1, NULL, 0);
-}
-
 static const char * const sunxi_board_dt_compat[] = {
 	"allwinner,sun4i-a10",
 	"allwinner,sun5i-a10s",
@@ -32,7 +27,6 @@ static const char * const sunxi_board_dt_compat[] = {
 
 DT_MACHINE_START(SUNXI_DT, "Allwinner sun4i/sun5i Families")
 	.dt_compat	= sunxi_board_dt_compat,
-	.init_late	= sunxi_dt_cpufreq_init,
 MACHINE_END
 
 static const char * const sun6i_board_dt_compat[] = {
@@ -53,7 +47,6 @@ static void __init sun6i_timer_init(void)
 DT_MACHINE_START(SUN6I_DT, "Allwinner sun6i (A31) Family")
 	.init_time	= sun6i_timer_init,
 	.dt_compat	= sun6i_board_dt_compat,
-	.init_late	= sunxi_dt_cpufreq_init,
 MACHINE_END
 
 static const char * const sun7i_board_dt_compat[] = {
@@ -63,7 +56,6 @@ static const char * const sun7i_board_dt_compat[] = {
 
 DT_MACHINE_START(SUN7I_DT, "Allwinner sun7i (A20) Family")
 	.dt_compat	= sun7i_board_dt_compat,
-	.init_late	= sunxi_dt_cpufreq_init,
 MACHINE_END
 
 static const char * const sun8i_board_dt_compat[] = {
@@ -77,7 +69,6 @@ static const char * const sun8i_board_dt_compat[] = {
 DT_MACHINE_START(SUN8I_DT, "Allwinner sun8i Family")
 	.init_time	= sun6i_timer_init,
 	.dt_compat	= sun8i_board_dt_compat,
-	.init_late	= sunxi_dt_cpufreq_init,
 MACHINE_END
 
 static const char * const sun9i_board_dt_compat[] = {

arch/arm/mach-zynq/common.c

@@ -110,7 +110,6 @@ static void __init zynq_init_late(void)
  */
 static void __init zynq_init_machine(void)
 {
-	struct platform_device_info devinfo = { .name = "cpufreq-dt", };
 	struct soc_device_attribute *soc_dev_attr;
 	struct soc_device *soc_dev;
 	struct device *parent = NULL;
@@ -145,7 +144,6 @@ out:
 	of_platform_populate(NULL, of_default_bus_match_table, NULL, parent);
 
 	platform_device_register(&zynq_cpuidle_device);
-	platform_device_register_full(&devinfo);
 }
 
 static void __init zynq_timer_init(void)

drivers/cpufreq/Kconfig

@@ -18,7 +18,11 @@ config CPU_FREQ
 
 if CPU_FREQ
 
+config CPU_FREQ_GOV_ATTR_SET
+	bool
+
 config CPU_FREQ_GOV_COMMON
+	select CPU_FREQ_GOV_ATTR_SET
 	select IRQ_WORK
 	bool
 
@@ -103,6 +107,17 @@ config CPU_FREQ_DEFAULT_GOV_CONSERVATIVE
 	  Be aware that not all cpufreq drivers support the conservative
 	  governor. If unsure have a look at the help section of the
 	  driver. Fallback governor will be the performance governor.
+
+config CPU_FREQ_DEFAULT_GOV_SCHEDUTIL
+	bool "schedutil"
+	depends on SMP
+	select CPU_FREQ_GOV_SCHEDUTIL
+	select CPU_FREQ_GOV_PERFORMANCE
+	help
+	  Use the 'schedutil' CPUFreq governor by default. If unsure,
+	  have a look at the help section of that governor. The fallback
+	  governor will be 'performance'.
+
 endchoice
 
 config CPU_FREQ_GOV_PERFORMANCE
@@ -184,6 +199,26 @@ config CPU_FREQ_GOV_CONSERVATIVE
 
 	  If in doubt, say N.
 
+config CPU_FREQ_GOV_SCHEDUTIL
+	tristate "'schedutil' cpufreq policy governor"
+	depends on CPU_FREQ && SMP
+	select CPU_FREQ_GOV_ATTR_SET
+	select IRQ_WORK
+	help
+	  This governor makes decisions based on the utilization data provided
+	  by the scheduler.  It sets the CPU frequency to be proportional to
+	  the utilization/capacity ratio coming from the scheduler.  If the
+	  utilization is frequency-invariant, the new frequency is also
+	  proportional to the maximum available frequency.  If that is not the
+	  case, it is proportional to the current frequency of the CPU.  The
+	  frequency tipping point is at utilization/capacity equal to 80% in
+	  both cases.
+
+	  To compile this driver as a module, choose M here: the module will
+	  be called cpufreq_schedutil.
+
+	  If in doubt, say N.
+
 comment "CPU frequency scaling drivers"
 
 config CPUFREQ_DT
@@ -191,6 +226,7 @@ config CPUFREQ_DT
 	depends on HAVE_CLK && OF
 	# if CPU_THERMAL is on and THERMAL=m, CPUFREQ_DT cannot be =y:
 	depends on !CPU_THERMAL || THERMAL
+	select CPUFREQ_DT_PLATDEV
 	select PM_OPP
 	help
 	  This adds a generic DT based cpufreq driver for frequency management.
@@ -199,6 +235,15 @@ config CPUFREQ_DT
 
 	  If in doubt, say N.
 
+config CPUFREQ_DT_PLATDEV
+	bool
+	help
+	  This adds a generic DT based cpufreq platdev driver for frequency
+	  management.  This creates a 'cpufreq-dt' platform device, on the
+	  supported platforms.
+
+	  If in doubt, say N.
+
 if X86
 source "drivers/cpufreq/Kconfig.x86"
 endif

drivers/cpufreq/Kconfig.arm

@@ -50,15 +50,6 @@ config ARM_HIGHBANK_CPUFREQ
 
 	  If in doubt, say N.
 
-config ARM_HISI_ACPU_CPUFREQ
-	tristate "Hisilicon ACPU CPUfreq driver"
-	depends on ARCH_HISI && CPUFREQ_DT
-	select PM_OPP
-	help
-	  This enables the hisilicon ACPU CPUfreq driver.
-
-	  If in doubt, say N.
-
 config ARM_IMX6Q_CPUFREQ
 	tristate "Freescale i.MX6 cpufreq support"
 	depends on ARCH_MXC

drivers/cpufreq/Kconfig.x86

@@ -5,6 +5,7 @@
 config X86_INTEL_PSTATE
        bool "Intel P state control"
        depends on X86
+       select ACPI_PROCESSOR if ACPI
        help
           This driver provides a P state for Intel core processors.
 	  The driver implements an internal governor and will become

drivers/cpufreq/Makefile

@@ -11,8 +11,10 @@ obj-$(CONFIG_CPU_FREQ_GOV_USERSPACE)	+= cpufreq_userspace.o
 obj-$(CONFIG_CPU_FREQ_GOV_ONDEMAND)	+= cpufreq_ondemand.o
 obj-$(CONFIG_CPU_FREQ_GOV_CONSERVATIVE)	+= cpufreq_conservative.o
 obj-$(CONFIG_CPU_FREQ_GOV_COMMON)		+= cpufreq_governor.o
+obj-$(CONFIG_CPU_FREQ_GOV_ATTR_SET)	+= cpufreq_governor_attr_set.o
 
 obj-$(CONFIG_CPUFREQ_DT)		+= cpufreq-dt.o
+obj-$(CONFIG_CPUFREQ_DT_PLATDEV)	+= cpufreq-dt-platdev.o
 
 ##################################################################################
 # x86 drivers.
@@ -53,7 +55,6 @@ obj-$(CONFIG_ARCH_DAVINCI)		+= davinci-cpufreq.o
 obj-$(CONFIG_UX500_SOC_DB8500)		+= dbx500-cpufreq.o
 obj-$(CONFIG_ARM_EXYNOS5440_CPUFREQ)	+= exynos5440-cpufreq.o
 obj-$(CONFIG_ARM_HIGHBANK_CPUFREQ)	+= highbank-cpufreq.o
-obj-$(CONFIG_ARM_HISI_ACPU_CPUFREQ)	+= hisi-acpu-cpufreq.o
 obj-$(CONFIG_ARM_IMX6Q_CPUFREQ)		+= imx6q-cpufreq.o
 obj-$(CONFIG_ARM_INTEGRATOR)		+= integrator-cpufreq.o
 obj-$(CONFIG_ARM_KIRKWOOD_CPUFREQ)	+= kirkwood-cpufreq.o
@@ -78,6 +79,7 @@ obj-$(CONFIG_ARM_TEGRA20_CPUFREQ)	+= tegra20-cpufreq.o
 obj-$(CONFIG_ARM_TEGRA124_CPUFREQ)	+= tegra124-cpufreq.o
 obj-$(CONFIG_ARM_VEXPRESS_SPC_CPUFREQ)	+= vexpress-spc-cpufreq.o
 obj-$(CONFIG_ACPI_CPPC_CPUFREQ) += cppc_cpufreq.o
+obj-$(CONFIG_MACH_MVEBU_V7)		+= mvebu-cpufreq.o
 
 
 ##################################################################################

drivers/cpufreq/acpi-cpufreq.c

@@ -25,6 +25,8 @@
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -50,8 +52,6 @@ MODULE_AUTHOR("Paul Diefenbaugh, Dominik Brodowski");
 MODULE_DESCRIPTION("ACPI Processor P-States Driver");
 MODULE_LICENSE("GPL");
 
-#define PFX "acpi-cpufreq: "
-
 enum {
 	UNDEFINED_CAPABLE = 0,
 	SYSTEM_INTEL_MSR_CAPABLE,
@@ -65,7 +65,6 @@ enum {
 #define MSR_K7_HWCR_CPB_DIS	(1ULL << 25)
 
 struct acpi_cpufreq_data {
-	struct cpufreq_frequency_table *freq_table;
 	unsigned int resume;
 	unsigned int cpu_feature;
 	unsigned int acpi_perf_cpu;
@@ -200,8 +199,9 @@ static int check_amd_hwpstate_cpu(unsigned int cpuid)
 	return cpu_has(cpu, X86_FEATURE_HW_PSTATE);
 }
 
-static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)
+static unsigned extract_io(struct cpufreq_policy *policy, u32 value)
 {
+	struct acpi_cpufreq_data *data = policy->driver_data;
 	struct acpi_processor_performance *perf;
 	int i;
 
@@ -209,13 +209,14 @@ static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)
 
 	for (i = 0; i < perf->state_count; i++) {
 		if (value == perf->states[i].status)
-			return data->freq_table[i].frequency;
+			return policy->freq_table[i].frequency;
 	}
 	return 0;
 }
 
-static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
+static unsigned extract_msr(struct cpufreq_policy *policy, u32 msr)
 {
+	struct acpi_cpufreq_data *data = policy->driver_data;
 	struct cpufreq_frequency_table *pos;
 	struct acpi_processor_performance *perf;
 
@@ -226,20 +227,22 @@ static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
 
 	perf = to_perf_data(data);
 
-	cpufreq_for_each_entry(pos, data->freq_table)
+	cpufreq_for_each_entry(pos, policy->freq_table)
 		if (msr == perf->states[pos->driver_data].status)
 			return pos->frequency;
-	return data->freq_table[0].frequency;
+	return policy->freq_table[0].frequency;
 }
 
-static unsigned extract_freq(u32 val, struct acpi_cpufreq_data *data)
+static unsigned extract_freq(struct cpufreq_policy *policy, u32 val)
 {
+	struct acpi_cpufreq_data *data = policy->driver_data;
+
 	switch (data->cpu_feature) {
 	case SYSTEM_INTEL_MSR_CAPABLE:
 	case SYSTEM_AMD_MSR_CAPABLE:
-		return extract_msr(val, data);
+		return extract_msr(policy, val);
 	case SYSTEM_IO_CAPABLE:
-		return extract_io(val, data);
+		return extract_io(policy, val);
 	default:
 		return 0;
 	}
@@ -374,11 +377,11 @@ static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 		return 0;
 
 	data = policy->driver_data;
-	if (unlikely(!data || !data->freq_table))
+	if (unlikely(!data || !policy->freq_table))
 		return 0;
 
-	cached_freq = data->freq_table[to_perf_data(data)->state].frequency;
-	freq = extract_freq(get_cur_val(cpumask_of(cpu), data), data);
+	cached_freq = policy->freq_table[to_perf_data(data)->state].frequency;
+	freq = extract_freq(policy, get_cur_val(cpumask_of(cpu), data));
 	if (freq != cached_freq) {
 		/*
 		 * The dreaded BIOS frequency change behind our back.
@@ -392,14 +395,15 @@ static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 	return freq;
 }
 
-static unsigned int check_freqs(const struct cpumask *mask, unsigned int freq,
-				struct acpi_cpufreq_data *data)
+static unsigned int check_freqs(struct cpufreq_policy *policy,
+				const struct cpumask *mask, unsigned int freq)
 {
+	struct acpi_cpufreq_data *data = policy->driver_data;
 	unsigned int cur_freq;
 	unsigned int i;
 
 	for (i = 0; i < 100; i++) {
-		cur_freq = extract_freq(get_cur_val(mask, data), data);
+		cur_freq = extract_freq(policy, get_cur_val(mask, data));
 		if (cur_freq == freq)
 			return 1;
 		udelay(10);
@@ -416,12 +420,12 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 	unsigned int next_perf_state = 0; /* Index into perf table */
 	int result = 0;
 
-	if (unlikely(data == NULL || data->freq_table == NULL)) {
+	if (unlikely(!data)) {
 		return -ENODEV;
 	}
 
 	perf = to_perf_data(data);
-	next_perf_state = data->freq_table[index].driver_data;
+	next_perf_state = policy->freq_table[index].driver_data;
 	if (perf->state == next_perf_state) {
 		if (unlikely(data->resume)) {
 			pr_debug("Called after resume, resetting to P%d\n",
@@ -444,8 +448,8 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 	drv_write(data, mask, perf->states[next_perf_state].control);
 
 	if (acpi_pstate_strict) {
-		if (!check_freqs(mask, data->freq_table[index].frequency,
-					data)) {
+		if (!check_freqs(policy, mask,
+				 policy->freq_table[index].frequency)) {
 			pr_debug("acpi_cpufreq_target failed (%d)\n",
 				policy->cpu);
 			result = -EAGAIN;
@@ -458,6 +462,43 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 	return result;
 }
 
+unsigned int acpi_cpufreq_fast_switch(struct cpufreq_policy *policy,
+				      unsigned int target_freq)
+{
+	struct acpi_cpufreq_data *data = policy->driver_data;
+	struct acpi_processor_performance *perf;
+	struct cpufreq_frequency_table *entry;
+	unsigned int next_perf_state, next_freq, freq;
+
+	/*
+	 * Find the closest frequency above target_freq.
+	 *
+	 * The table is sorted in the reverse order with respect to the
+	 * frequency and all of the entries are valid (see the initialization).
+	 */
+	entry = policy->freq_table;
+	do {
+		entry++;
+		freq = entry->frequency;
+	} while (freq >= target_freq && freq != CPUFREQ_TABLE_END);
+	entry--;
+	next_freq = entry->frequency;
+	next_perf_state = entry->driver_data;
+
+	perf = to_perf_data(data);
+	if (perf->state == next_perf_state) {
+		if (unlikely(data->resume))
+			data->resume = 0;
+		else
+			return next_freq;
+	}
+
+	data->cpu_freq_write(&perf->control_register,
+			     perf->states[next_perf_state].control);
+	perf->state = next_perf_state;
+	return next_freq;
+}
+
 static unsigned long
 acpi_cpufreq_guess_freq(struct acpi_cpufreq_data *data, unsigned int cpu)
 {
@@ -611,10 +652,7 @@ static int acpi_cpufreq_blacklist(struct cpuinfo_x86 *c)
 		if ((c->x86 == 15) &&
 		    (c->x86_model == 6) &&
 		    (c->x86_mask == 8)) {
-			printk(KERN_INFO "acpi-cpufreq: Intel(R) "
-			    "Xeon(R) 7100 Errata AL30, processors may "
-			    "lock up on frequency changes: disabling "
-			    "acpi-cpufreq.\n");
+			pr_info("Intel(R) Xeon(R) 7100 Errata AL30, processors may lock up on frequency changes: disabling acpi-cpufreq\n");
 			return -ENODEV;
 		    }
 		}
@@ -631,6 +669,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	unsigned int result = 0;
 	struct cpuinfo_x86 *c = &cpu_data(policy->cpu);
 	struct acpi_processor_performance *perf;
+	struct cpufreq_frequency_table *freq_table;
 #ifdef CONFIG_SMP
 	static int blacklisted;
 #endif
@@ -690,7 +729,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 		cpumask_copy(data->freqdomain_cpus,
 			     topology_sibling_cpumask(cpu));
 		policy->shared_type = CPUFREQ_SHARED_TYPE_HW;
-		pr_info_once(PFX "overriding BIOS provided _PSD data\n");
+		pr_info_once("overriding BIOS provided _PSD data\n");
 	}
 #endif
 
@@ -742,9 +781,9 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 		goto err_unreg;
 	}
 
-	data->freq_table = kzalloc(sizeof(*data->freq_table) *
+	freq_table = kzalloc(sizeof(*freq_table) *
 		    (perf->state_count+1), GFP_KERNEL);
-	if (!data->freq_table) {
+	if (!freq_table) {
 		result = -ENOMEM;
 		goto err_unreg;
 	}
@@ -762,30 +801,29 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	if (perf->control_register.space_id == ACPI_ADR_SPACE_FIXED_HARDWARE &&
 	    policy->cpuinfo.transition_latency > 20 * 1000) {
 		policy->cpuinfo.transition_latency = 20 * 1000;
-		printk_once(KERN_INFO
-			    "P-state transition latency capped at 20 uS\n");
+		pr_info_once("P-state transition latency capped at 20 uS\n");
 	}
 
 	/* table init */
 	for (i = 0; i < perf->state_count; i++) {
 		if (i > 0 && perf->states[i].core_frequency >=
-		    data->freq_table[valid_states-1].frequency / 1000)
+		    freq_table[valid_states-1].frequency / 1000)
 			continue;
 
-		data->freq_table[valid_states].driver_data = i;
-		data->freq_table[valid_states].frequency =
+		freq_table[valid_states].driver_data = i;
+		freq_table[valid_states].frequency =
 		    perf->states[i].core_frequency * 1000;
 		valid_states++;
 	}
-	data->freq_table[valid_states].frequency = CPUFREQ_TABLE_END;
+	freq_table[valid_states].frequency = CPUFREQ_TABLE_END;
 	perf->state = 0;
 
-	result = cpufreq_table_validate_and_show(policy, data->freq_table);
+	result = cpufreq_table_validate_and_show(policy, freq_table);
 	if (result)
 		goto err_freqfree;
 
 	if (perf->states[0].core_frequency * 1000 != policy->cpuinfo.max_freq)
-		printk(KERN_WARNING FW_WARN "P-state 0 is not max freq\n");
+		pr_warn(FW_WARN "P-state 0 is not max freq\n");
 
 	switch (perf->control_register.space_id) {
 	case ACPI_ADR_SPACE_SYSTEM_IO:
@@ -821,10 +859,13 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	 */
 	data->resume = 1;
 
+	policy->fast_switch_possible = !acpi_pstate_strict &&
+		!(policy_is_shared(policy) && policy->shared_type != CPUFREQ_SHARED_TYPE_ANY);
+
 	return result;
 
 err_freqfree:
-	kfree(data->freq_table);
+	kfree(freq_table);
 err_unreg:
 	acpi_processor_unregister_performance(cpu);
 err_free_mask:
@@ -842,13 +883,12 @@ static int acpi_cpufreq_cpu_exit(struct cpufreq_policy *policy)
 
 	pr_debug("acpi_cpufreq_cpu_exit\n");
 
-	if (data) {
-		policy->driver_data = NULL;
-		acpi_processor_unregister_performance(data->acpi_perf_cpu);
-		free_cpumask_var(data->freqdomain_cpus);
-		kfree(data->freq_table);
-		kfree(data);
-	}
+	policy->fast_switch_possible = false;
+	policy->driver_data = NULL;
+	acpi_processor_unregister_performance(data->acpi_perf_cpu);
+	free_cpumask_var(data->freqdomain_cpus);
+	kfree(policy->freq_table);
+	kfree(data);
 
 	return 0;
 }
@@ -876,6 +916,7 @@ static struct freq_attr *acpi_cpufreq_attr[] = {
 static struct cpufreq_driver acpi_cpufreq_driver = {
 	.verify		= cpufreq_generic_frequency_table_verify,
 	.target_index	= acpi_cpufreq_target,
+	.fast_switch	= acpi_cpufreq_fast_switch,
 	.bios_limit	= acpi_processor_get_bios_limit,
 	.init		= acpi_cpufreq_cpu_init,
 	.exit		= acpi_cpufreq_cpu_exit,

drivers/cpufreq/arm_big_little.c

@@ -298,7 +298,8 @@ static int merge_cluster_tables(void)
 	return 0;
 }
 
-static void _put_cluster_clk_and_freq_table(struct device *cpu_dev)
+static void _put_cluster_clk_and_freq_table(struct device *cpu_dev,
+					    const struct cpumask *cpumask)
 {
 	u32 cluster = raw_cpu_to_cluster(cpu_dev->id);
 
@@ -308,11 +309,12 @@ static void _put_cluster_clk_and_freq_table(struct device *cpu_dev)
 	clk_put(clk[cluster]);
 	dev_pm_opp_free_cpufreq_table(cpu_dev, &freq_table[cluster]);
 	if (arm_bL_ops->free_opp_table)
-		arm_bL_ops->free_opp_table(cpu_dev);
+		arm_bL_ops->free_opp_table(cpumask);
 	dev_dbg(cpu_dev, "%s: cluster: %d\n", __func__, cluster);
 }
 
-static void put_cluster_clk_and_freq_table(struct device *cpu_dev)
+static void put_cluster_clk_and_freq_table(struct device *cpu_dev,
+					   const struct cpumask *cpumask)
 {
 	u32 cluster = cpu_to_cluster(cpu_dev->id);
 	int i;
@@ -321,7 +323,7 @@ static void put_cluster_clk_and_freq_table(struct device *cpu_dev)
 		return;
 
 	if (cluster < MAX_CLUSTERS)
-		return _put_cluster_clk_and_freq_table(cpu_dev);
+		return _put_cluster_clk_and_freq_table(cpu_dev, cpumask);
 
 	for_each_present_cpu(i) {
 		struct device *cdev = get_cpu_device(i);
@@ -330,14 +332,15 @@ static void put_cluster_clk_and_freq_table(struct device *cpu_dev)
 			return;
 		}
 
-		_put_cluster_clk_and_freq_table(cdev);
+		_put_cluster_clk_and_freq_table(cdev, cpumask);
 	}
 
 	/* free virtual table */
 	kfree(freq_table[cluster]);
 }
 
-static int _get_cluster_clk_and_freq_table(struct device *cpu_dev)
+static int _get_cluster_clk_and_freq_table(struct device *cpu_dev,
+					   const struct cpumask *cpumask)
 {
 	u32 cluster = raw_cpu_to_cluster(cpu_dev->id);
 	int ret;
@@ -345,7 +348,7 @@ static int _get_cluster_clk_and_freq_table(struct device *cpu_dev)
 	if (freq_table[cluster])
 		return 0;
 
-	ret = arm_bL_ops->init_opp_table(cpu_dev);
+	ret = arm_bL_ops->init_opp_table(cpumask);
 	if (ret) {
 		dev_err(cpu_dev, "%s: init_opp_table failed, cpu: %d, err: %d\n",
 				__func__, cpu_dev->id, ret);
@@ -374,14 +377,15 @@ static int _get_cluster_clk_and_freq_table(struct device *cpu_dev)
 
 free_opp_table:
 	if (arm_bL_ops->free_opp_table)
-		arm_bL_ops->free_opp_table(cpu_dev);
+		arm_bL_ops->free_opp_table(cpumask);
 out:
 	dev_err(cpu_dev, "%s: Failed to get data for cluster: %d\n", __func__,
 			cluster);
 	return ret;
 }
 
-static int get_cluster_clk_and_freq_table(struct device *cpu_dev)
+static int get_cluster_clk_and_freq_table(struct device *cpu_dev,
+					  const struct cpumask *cpumask)
 {
 	u32 cluster = cpu_to_cluster(cpu_dev->id);
 	int i, ret;
@@ -390,7 +394,7 @@ static int get_cluster_clk_and_freq_table(struct device *cpu_dev)
 		return 0;
 
 	if (cluster < MAX_CLUSTERS) {
-		ret = _get_cluster_clk_and_freq_table(cpu_dev);
+		ret = _get_cluster_clk_and_freq_table(cpu_dev, cpumask);
 		if (ret)
 			atomic_dec(&cluster_usage[cluster]);
 		return ret;
@@ -407,7 +411,7 @@ static int get_cluster_clk_and_freq_table(struct device *cpu_dev)
 			return -ENODEV;
 		}
 
-		ret = _get_cluster_clk_and_freq_table(cdev);
+		ret = _get_cluster_clk_and_freq_table(cdev, cpumask);
 		if (ret)
 			goto put_clusters;
 	}
@@ -433,7 +437,7 @@ put_clusters:
 			return -ENODEV;
 		}
 
-		_put_cluster_clk_and_freq_table(cdev);
+		_put_cluster_clk_and_freq_table(cdev, cpumask);
 	}
 
 	atomic_dec(&cluster_usage[cluster]);
@@ -455,18 +459,6 @@ static int bL_cpufreq_init(struct cpufreq_policy *policy)
 		return -ENODEV;
 	}
 
-	ret = get_cluster_clk_and_freq_table(cpu_dev);
-	if (ret)
-		return ret;
-
-	ret = cpufreq_table_validate_and_show(policy, freq_table[cur_cluster]);
-	if (ret) {
-		dev_err(cpu_dev, "CPU %d, cluster: %d invalid freq table\n",
-				policy->cpu, cur_cluster);
-		put_cluster_clk_and_freq_table(cpu_dev);
-		return ret;
-	}
-
 	if (cur_cluster < MAX_CLUSTERS) {
 		int cpu;
 
@@ -479,6 +471,18 @@ static int bL_cpufreq_init(struct cpufreq_policy *policy)
 		per_cpu(physical_cluster, policy->cpu) = A15_CLUSTER;
 	}
 
+	ret = get_cluster_clk_and_freq_table(cpu_dev, policy->cpus);
+	if (ret)
+		return ret;
+
+	ret = cpufreq_table_validate_and_show(policy, freq_table[cur_cluster]);
+	if (ret) {
+		dev_err(cpu_dev, "CPU %d, cluster: %d invalid freq table\n",
+			policy->cpu, cur_cluster);
+		put_cluster_clk_and_freq_table(cpu_dev, policy->cpus);
+		return ret;
+	}
+
 	if (arm_bL_ops->get_transition_latency)
 		policy->cpuinfo.transition_latency =
 			arm_bL_ops->get_transition_latency(cpu_dev);
@@ -509,7 +513,7 @@ static int bL_cpufreq_exit(struct cpufreq_policy *policy)
 		return -ENODEV;
 	}
 
-	put_cluster_clk_and_freq_table(cpu_dev);
+	put_cluster_clk_and_freq_table(cpu_dev, policy->related_cpus);
 	dev_dbg(cpu_dev, "%s: Exited, cpu: %d\n", __func__, policy->cpu);
 
 	return 0;

drivers/cpufreq/arm_big_little.h

@@ -30,11 +30,11 @@ struct cpufreq_arm_bL_ops {
 	 * This must set opp table for cpu_dev in a similar way as done by
 	 * dev_pm_opp_of_add_table().
 	 */
-	int (*init_opp_table)(struct device *cpu_dev);
+	int (*init_opp_table)(const struct cpumask *cpumask);
 
 	/* Optional */
 	int (*get_transition_latency)(struct device *cpu_dev);
-	void (*free_opp_table)(struct device *cpu_dev);
+	void (*free_opp_table)(const struct cpumask *cpumask);
 };
 
 int bL_cpufreq_register(struct cpufreq_arm_bL_ops *ops);

drivers/cpufreq/arm_big_little_dt.c

@@ -43,23 +43,6 @@ static struct device_node *get_cpu_node_with_valid_op(int cpu)
 	return np;
 }
 
-static int dt_init_opp_table(struct device *cpu_dev)
-{
-	struct device_node *np;
-	int ret;
-
-	np = of_node_get(cpu_dev->of_node);
-	if (!np) {
-		pr_err("failed to find cpu%d node\n", cpu_dev->id);
-		return -ENOENT;
-	}
-
-	ret = dev_pm_opp_of_add_table(cpu_dev);
-	of_node_put(np);
-
-	return ret;
-}
-
 static int dt_get_transition_latency(struct device *cpu_dev)
 {
 	struct device_node *np;
@@ -81,8 +64,8 @@ static int dt_get_transition_latency(struct device *cpu_dev)
 static struct cpufreq_arm_bL_ops dt_bL_ops = {
 	.name	= "dt-bl",
 	.get_transition_latency = dt_get_transition_latency,
-	.init_opp_table = dt_init_opp_table,
-	.free_opp_table = dev_pm_opp_of_remove_table,
+	.init_opp_table = dev_pm_opp_of_cpumask_add_table,
+	.free_opp_table = dev_pm_opp_of_cpumask_remove_table,
 };
 
 static int generic_bL_probe(struct platform_device *pdev)

drivers/cpufreq/cppc_cpufreq.c

@@ -173,4 +173,25 @@ out:
 	return -ENODEV;
 }
 
+static void __exit cppc_cpufreq_exit(void)
+{
+	struct cpudata *cpu;
+	int i;
+
+	cpufreq_unregister_driver(&cppc_cpufreq_driver);
+
+	for_each_possible_cpu(i) {
+		cpu = all_cpu_data[i];
+		free_cpumask_var(cpu->shared_cpu_map);
+		kfree(cpu);
+	}
+
+	kfree(all_cpu_data);
+}
+
+module_exit(cppc_cpufreq_exit);
+MODULE_AUTHOR("Ashwin Chaugule");
+MODULE_DESCRIPTION("CPUFreq driver based on the ACPI CPPC v5.0+ spec");
+MODULE_LICENSE("GPL");
+
 late_initcall(cppc_cpufreq_init);

drivers/cpufreq/cpufreq-dt-platdev.c

@@ -0,0 +1,94 @@
+/*
+ * Copyright (C) 2016 Linaro.
+ * Viresh Kumar <viresh.kumar@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/err.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+static const struct of_device_id machines[] __initconst = {
+	{ .compatible = "allwinner,sun4i-a10", },
+	{ .compatible = "allwinner,sun5i-a10s", },
+	{ .compatible = "allwinner,sun5i-a13", },
+	{ .compatible = "allwinner,sun5i-r8", },
+	{ .compatible = "allwinner,sun6i-a31", },
+	{ .compatible = "allwinner,sun6i-a31s", },
+	{ .compatible = "allwinner,sun7i-a20", },
+	{ .compatible = "allwinner,sun8i-a23", },
+	{ .compatible = "allwinner,sun8i-a33", },
+	{ .compatible = "allwinner,sun8i-a83t", },
+	{ .compatible = "allwinner,sun8i-h3", },
+
+	{ .compatible = "hisilicon,hi6220", },
+
+	{ .compatible = "fsl,imx27", },
+	{ .compatible = "fsl,imx51", },
+	{ .compatible = "fsl,imx53", },
+	{ .compatible = "fsl,imx7d", },
+
+	{ .compatible = "marvell,berlin", },
+
+	{ .compatible = "samsung,exynos3250", },
+	{ .compatible = "samsung,exynos4210", },
+	{ .compatible = "samsung,exynos4212", },
+	{ .compatible = "samsung,exynos4412", },
+	{ .compatible = "samsung,exynos5250", },
+#ifndef CONFIG_BL_SWITCHER
+	{ .compatible = "samsung,exynos5420", },
+	{ .compatible = "samsung,exynos5800", },
+#endif
+
+	{ .compatible = "renesas,emev2", },
+	{ .compatible = "renesas,r7s72100", },
+	{ .compatible = "renesas,r8a73a4", },
+	{ .compatible = "renesas,r8a7740", },
+	{ .compatible = "renesas,r8a7778", },
+	{ .compatible = "renesas,r8a7779", },
+	{ .compatible = "renesas,r8a7790", },
+	{ .compatible = "renesas,r8a7791", },
+	{ .compatible = "renesas,r8a7793", },
+	{ .compatible = "renesas,r8a7794", },
+	{ .compatible = "renesas,sh73a0", },
+
+	{ .compatible = "rockchip,rk2928", },
+	{ .compatible = "rockchip,rk3036", },
+	{ .compatible = "rockchip,rk3066a", },
+	{ .compatible = "rockchip,rk3066b", },
+	{ .compatible = "rockchip,rk3188", },
+	{ .compatible = "rockchip,rk3228", },
+	{ .compatible = "rockchip,rk3288", },
+	{ .compatible = "rockchip,rk3366", },
+	{ .compatible = "rockchip,rk3368", },
+	{ .compatible = "rockchip,rk3399", },
+
+	{ .compatible = "sigma,tango4" },
+
+	{ .compatible = "ti,omap2", },
+	{ .compatible = "ti,omap3", },
+	{ .compatible = "ti,omap4", },
+	{ .compatible = "ti,omap5", },
+
+	{ .compatible = "xlnx,zynq-7000", },
+};
+
+static int __init cpufreq_dt_platdev_init(void)
+{
+	struct device_node *np = of_find_node_by_path("/");
+
+	if (!np)
+		return -ENODEV;
+
+	if (!of_match_node(machines, np))
+		return -ENODEV;
+
+	of_node_put(of_root);
+
+	return PTR_ERR_OR_ZERO(platform_device_register_simple("cpufreq-dt", -1,
+							       NULL, 0));
+}
+device_initcall(cpufreq_dt_platdev_init);

drivers/cpufreq/cpufreq-dt.c

@@ -15,7 +15,6 @@
 #include <linux/cpu.h>
 #include <linux/cpu_cooling.h>
 #include <linux/cpufreq.h>
-#include <linux/cpufreq-dt.h>
 #include <linux/cpumask.h>
 #include <linux/err.h>
 #include <linux/module.h>
@@ -147,7 +146,7 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 	struct clk *cpu_clk;
 	struct dev_pm_opp *suspend_opp;
 	unsigned int transition_latency;
-	bool opp_v1 = false;
+	bool fallback = false;
 	const char *name;
 	int ret;
 
@@ -167,14 +166,16 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 	/* Get OPP-sharing information from "operating-points-v2" bindings */
 	ret = dev_pm_opp_of_get_sharing_cpus(cpu_dev, policy->cpus);
 	if (ret) {
+		if (ret != -ENOENT)
+			goto out_put_clk;
+
 		/*
 		 * operating-points-v2 not supported, fallback to old method of
-		 * finding shared-OPPs for backward compatibility.
+		 * finding shared-OPPs for backward compatibility if the
+		 * platform hasn't set sharing CPUs.
 		 */
-		if (ret == -ENOENT)
-			opp_v1 = true;
-		else
-			goto out_put_clk;
+		if (dev_pm_opp_get_sharing_cpus(cpu_dev, policy->cpus))
+			fallback = true;
 	}
 
 	/*
@@ -214,11 +215,8 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 		goto out_free_opp;
 	}
 
-	if (opp_v1) {
-		struct cpufreq_dt_platform_data *pd = cpufreq_get_driver_data();
-
-		if (!pd || !pd->independent_clocks)
-			cpumask_setall(policy->cpus);
+	if (fallback) {
+		cpumask_setall(policy->cpus);
 
 		/*
 		 * OPP tables are initialized only for policy->cpu, do it for

drivers/cpufreq/cpufreq-nforce2.c

@@ -7,6 +7,8 @@
  *  BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous*
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
@@ -56,8 +58,6 @@ MODULE_PARM_DESC(fid, "CPU multiplier to use (11.5 = 115)");
 MODULE_PARM_DESC(min_fsb,
 		"Minimum FSB to use, if not defined: current FSB - 50");
 
-#define PFX "cpufreq-nforce2: "
-
 /**
  * nforce2_calc_fsb - calculate FSB
  * @pll: PLL value
@@ -174,13 +174,13 @@ static int nforce2_set_fsb(unsigned int fsb)
 	int pll = 0;
 
 	if ((fsb > max_fsb) || (fsb < NFORCE2_MIN_FSB)) {
-		printk(KERN_ERR PFX "FSB %d is out of range!\n", fsb);
+		pr_err("FSB %d is out of range!\n", fsb);
 		return -EINVAL;
 	}
 
 	tfsb = nforce2_fsb_read(0);
 	if (!tfsb) {
-		printk(KERN_ERR PFX "Error while reading the FSB\n");
+		pr_err("Error while reading the FSB\n");
 		return -EINVAL;
 	}
 
@@ -276,8 +276,7 @@ static int nforce2_target(struct cpufreq_policy *policy,
 	/* local_irq_save(flags); */
 
 	if (nforce2_set_fsb(target_fsb) < 0)
-		printk(KERN_ERR PFX "Changing FSB to %d failed\n",
-			target_fsb);
+		pr_err("Changing FSB to %d failed\n", target_fsb);
 	else
 		pr_debug("Changed FSB successfully to %d\n",
 			target_fsb);
@@ -325,8 +324,7 @@ static int nforce2_cpu_init(struct cpufreq_policy *policy)
 	/* FIX: Get FID from CPU */
 	if (!fid) {
 		if (!cpu_khz) {
-			printk(KERN_WARNING PFX
-			"cpu_khz not set, can't calculate multiplier!\n");
+			pr_warn("cpu_khz not set, can't calculate multiplier!\n");
 			return -ENODEV;
 		}
 
@@ -341,8 +339,8 @@ static int nforce2_cpu_init(struct cpufreq_policy *policy)
 		}
 	}
 
-	printk(KERN_INFO PFX "FSB currently at %i MHz, FID %d.%d\n", fsb,
-	       fid / 10, fid % 10);
+	pr_info("FSB currently at %i MHz, FID %d.%d\n",
+		fsb, fid / 10, fid % 10);
 
 	/* Set maximum FSB to FSB at boot time */
 	max_fsb = nforce2_fsb_read(1);
@@ -401,11 +399,9 @@ static int nforce2_detect_chipset(void)
 	if (nforce2_dev == NULL)
 		return -ENODEV;
 
-	printk(KERN_INFO PFX "Detected nForce2 chipset revision %X\n",
-	       nforce2_dev->revision);
-	printk(KERN_INFO PFX
-	       "FSB changing is maybe unstable and can lead to "
-	       "crashes and data loss.\n");
+	pr_info("Detected nForce2 chipset revision %X\n",
+		nforce2_dev->revision);
+	pr_info("FSB changing is maybe unstable and can lead to crashes and data loss\n");
 
 	return 0;
 }
@@ -423,7 +419,7 @@ static int __init nforce2_init(void)
 
 	/* detect chipset */
 	if (nforce2_detect_chipset()) {
-		printk(KERN_INFO PFX "No nForce2 chipset.\n");
+		pr_info("No nForce2 chipset\n");
 		return -ENODEV;
 	}
 

drivers/cpufreq/cpufreq.c

@@ -78,6 +78,11 @@ static int cpufreq_governor(struct cpufreq_policy *policy, unsigned int event);
 static unsigned int __cpufreq_get(struct cpufreq_policy *policy);
 static int cpufreq_start_governor(struct cpufreq_policy *policy);
 
+static inline int cpufreq_exit_governor(struct cpufreq_policy *policy)
+{
+	return cpufreq_governor(policy, CPUFREQ_GOV_POLICY_EXIT);
+}
+
 /**
  * Two notifier lists: the "policy" list is involved in the
  * validation process for a new CPU frequency policy; the
@@ -429,6 +434,73 @@ void cpufreq_freq_transition_end(struct cpufreq_policy *policy,
 }
 EXPORT_SYMBOL_GPL(cpufreq_freq_transition_end);
 
+/*
+ * Fast frequency switching status count.  Positive means "enabled", negative
+ * means "disabled" and 0 means "not decided yet".
+ */
+static int cpufreq_fast_switch_count;
+static DEFINE_MUTEX(cpufreq_fast_switch_lock);
+
+static void cpufreq_list_transition_notifiers(void)
+{
+	struct notifier_block *nb;
+
+	pr_info("Registered transition notifiers:\n");
+
+	mutex_lock(&cpufreq_transition_notifier_list.mutex);
+
+	for (nb = cpufreq_transition_notifier_list.head; nb; nb = nb->next)
+		pr_info("%pF\n", nb->notifier_call);
+
+	mutex_unlock(&cpufreq_transition_notifier_list.mutex);
+}
+
+/**
+ * cpufreq_enable_fast_switch - Enable fast frequency switching for policy.
+ * @policy: cpufreq policy to enable fast frequency switching for.
+ *
+ * Try to enable fast frequency switching for @policy.
+ *
+ * The attempt will fail if there is at least one transition notifier registered
+ * at this point, as fast frequency switching is quite fundamentally at odds
+ * with transition notifiers.  Thus if successful, it will make registration of
+ * transition notifiers fail going forward.
+ */
+void cpufreq_enable_fast_switch(struct cpufreq_policy *policy)
+{
+	lockdep_assert_held(&policy->rwsem);
+
+	if (!policy->fast_switch_possible)
+		return;
+
+	mutex_lock(&cpufreq_fast_switch_lock);
+	if (cpufreq_fast_switch_count >= 0) {
+		cpufreq_fast_switch_count++;
+		policy->fast_switch_enabled = true;
+	} else {
+		pr_warn("CPU%u: Fast frequency switching not enabled\n",
+			policy->cpu);
+		cpufreq_list_transition_notifiers();
+	}
+	mutex_unlock(&cpufreq_fast_switch_lock);
+}
+EXPORT_SYMBOL_GPL(cpufreq_enable_fast_switch);
+
+/**
+ * cpufreq_disable_fast_switch - Disable fast frequency switching for policy.
+ * @policy: cpufreq policy to disable fast frequency switching for.
+ */
+void cpufreq_disable_fast_switch(struct cpufreq_policy *policy)
+{
+	mutex_lock(&cpufreq_fast_switch_lock);
+	if (policy->fast_switch_enabled) {
+		policy->fast_switch_enabled = false;
+		if (!WARN_ON(cpufreq_fast_switch_count <= 0))
+			cpufreq_fast_switch_count--;
+	}
+	mutex_unlock(&cpufreq_fast_switch_lock);
+}
+EXPORT_SYMBOL_GPL(cpufreq_disable_fast_switch);
 
 /*********************************************************************
  *                          SYSFS INTERFACE                          *
@@ -1248,26 +1320,24 @@ out_free_policy:
  */
 static int cpufreq_add_dev(struct device *dev, struct subsys_interface *sif)
 {
+	struct cpufreq_policy *policy;
 	unsigned cpu = dev->id;
-	int ret;
 
 	dev_dbg(dev, "%s: adding CPU%u\n", __func__, cpu);
 
-	if (cpu_online(cpu)) {
-		ret = cpufreq_online(cpu);
-	} else {
-		/*
-		 * A hotplug notifier will follow and we will handle it as CPU
-		 * online then.  For now, just create the sysfs link, unless
-		 * there is no policy or the link is already present.
-		 */
-		struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu);
+	if (cpu_online(cpu))
+		return cpufreq_online(cpu);
 
-		ret = policy && !cpumask_test_and_set_cpu(cpu, policy->real_cpus)
-			? add_cpu_dev_symlink(policy, cpu) : 0;
-	}
+	/*
+	 * A hotplug notifier will follow and we will handle it as CPU online
+	 * then.  For now, just create the sysfs link, unless there is no policy
+	 * or the link is already present.
+	 */
+	policy = per_cpu(cpufreq_cpu_data, cpu);
+	if (!policy || cpumask_test_and_set_cpu(cpu, policy->real_cpus))
+		return 0;
 
-	return ret;
+	return add_cpu_dev_symlink(policy, cpu);
 }
 
 static void cpufreq_offline(unsigned int cpu)
@@ -1319,7 +1389,7 @@ static void cpufreq_offline(unsigned int cpu)
 
 	/* If cpu is last user of policy, free policy */
 	if (has_target()) {
-		ret = cpufreq_governor(policy, CPUFREQ_GOV_POLICY_EXIT);
+		ret = cpufreq_exit_governor(policy);
 		if (ret)
 			pr_err("%s: Failed to exit governor\n", __func__);
 	}
@@ -1447,8 +1517,12 @@ static unsigned int __cpufreq_get(struct cpufreq_policy *policy)
 
 	ret_freq = cpufreq_driver->get(policy->cpu);
 
-	/* Updating inactive policies is invalid, so avoid doing that. */
-	if (unlikely(policy_is_inactive(policy)))
+	/*
+	 * Updating inactive policies is invalid, so avoid doing that.  Also
+	 * if fast frequency switching is used with the given policy, the check
+	 * against policy->cur is pointless, so skip it in that case too.
+	 */
+	if (unlikely(policy_is_inactive(policy)) || policy->fast_switch_enabled)
 		return ret_freq;
 
 	if (ret_freq && policy->cur &&
@@ -1679,8 +1753,18 @@ int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list)
 
 	switch (list) {
 	case CPUFREQ_TRANSITION_NOTIFIER:
+		mutex_lock(&cpufreq_fast_switch_lock);
+
+		if (cpufreq_fast_switch_count > 0) {
+			mutex_unlock(&cpufreq_fast_switch_lock);
+			return -EBUSY;
+		}
 		ret = srcu_notifier_chain_register(
 				&cpufreq_transition_notifier_list, nb);
+		if (!ret)
+			cpufreq_fast_switch_count--;
+
+		mutex_unlock(&cpufreq_fast_switch_lock);
 		break;
 	case CPUFREQ_POLICY_NOTIFIER:
 		ret = blocking_notifier_chain_register(
@@ -1713,8 +1797,14 @@ int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list)
 
 	switch (list) {
 	case CPUFREQ_TRANSITION_NOTIFIER:
+		mutex_lock(&cpufreq_fast_switch_lock);
+
 		ret = srcu_notifier_chain_unregister(
 				&cpufreq_transition_notifier_list, nb);
+		if (!ret && !WARN_ON(cpufreq_fast_switch_count >= 0))
+			cpufreq_fast_switch_count++;
+
+		mutex_unlock(&cpufreq_fast_switch_lock);
 		break;
 	case CPUFREQ_POLICY_NOTIFIER:
 		ret = blocking_notifier_chain_unregister(
@@ -1733,6 +1823,37 @@ EXPORT_SYMBOL(cpufreq_unregister_notifier);
  *                              GOVERNORS                            *
  *********************************************************************/
 
+/**
+ * cpufreq_driver_fast_switch - Carry out a fast CPU frequency switch.
+ * @policy: cpufreq policy to switch the frequency for.
+ * @target_freq: New frequency to set (may be approximate).
+ *
+ * Carry out a fast frequency switch without sleeping.
+ *
+ * The driver's ->fast_switch() callback invoked by this function must be
+ * suitable for being called from within RCU-sched read-side critical sections
+ * and it is expected to select the minimum available frequency greater than or
+ * equal to @target_freq (CPUFREQ_RELATION_L).
+ *
+ * This function must not be called if policy->fast_switch_enabled is unset.
+ *
+ * Governors calling this function must guarantee that it will never be invoked
+ * twice in parallel for the same policy and that it will never be called in
+ * parallel with either ->target() or ->target_index() for the same policy.
+ *
+ * If CPUFREQ_ENTRY_INVALID is returned by the driver's ->fast_switch()
+ * callback to indicate an error condition, the hardware configuration must be
+ * preserved.
+ */
+unsigned int cpufreq_driver_fast_switch(struct cpufreq_policy *policy,
+					unsigned int target_freq)
+{
+	clamp_val(target_freq, policy->min, policy->max);
+
+	return cpufreq_driver->fast_switch(policy, target_freq);
+}
+EXPORT_SYMBOL_GPL(cpufreq_driver_fast_switch);
+
 /* Must set freqs->new to intermediate frequency */
 static int __target_intermediate(struct cpufreq_policy *policy,
 				 struct cpufreq_freqs *freqs, int index)
@@ -2108,7 +2229,7 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy,
 			return ret;
 		}
 
-		ret = cpufreq_governor(policy, CPUFREQ_GOV_POLICY_EXIT);
+		ret = cpufreq_exit_governor(policy);
 		if (ret) {
 			pr_err("%s: Failed to Exit Governor: %s (%d)\n",
 			       __func__, old_gov->name, ret);
@@ -2125,7 +2246,7 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy,
 			pr_debug("cpufreq: governor change\n");
 			return 0;
 		}
-		cpufreq_governor(policy, CPUFREQ_GOV_POLICY_EXIT);
+		cpufreq_exit_governor(policy);
 	}
 
 	/* new governor failed, so re-start old one */
@@ -2193,16 +2314,13 @@ static int cpufreq_cpu_callback(struct notifier_block *nfb,
 
 	switch (action & ~CPU_TASKS_FROZEN) {
 	case CPU_ONLINE:
+	case CPU_DOWN_FAILED:
 		cpufreq_online(cpu);
 		break;
 
 	case CPU_DOWN_PREPARE:
 		cpufreq_offline(cpu);
 		break;
-
-	case CPU_DOWN_FAILED:
-		cpufreq_online(cpu);
-		break;
 	}
 	return NOTIFY_OK;
 }

drivers/cpufreq/cpufreq_conservative.c

@@ -129,9 +129,10 @@ static struct notifier_block cs_cpufreq_notifier_block = {
 /************************** sysfs interface ************************/
 static struct dbs_governor cs_dbs_gov;
 
-static ssize_t store_sampling_down_factor(struct dbs_data *dbs_data,
-		const char *buf, size_t count)
+static ssize_t store_sampling_down_factor(struct gov_attr_set *attr_set,
+					  const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	unsigned int input;
 	int ret;
 	ret = sscanf(buf, "%u", &input);
@@ -143,9 +144,10 @@ static ssize_t store_sampling_down_factor(struct dbs_data *dbs_data,
 	return count;
 }
 
-static ssize_t store_up_threshold(struct dbs_data *dbs_data, const char *buf,
-		size_t count)
+static ssize_t store_up_threshold(struct gov_attr_set *attr_set,
+				  const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	struct cs_dbs_tuners *cs_tuners = dbs_data->tuners;
 	unsigned int input;
 	int ret;
@@ -158,9 +160,10 @@ static ssize_t store_up_threshold(struct dbs_data *dbs_data, const char *buf,
 	return count;
 }
 
-static ssize_t store_down_threshold(struct dbs_data *dbs_data, const char *buf,
-		size_t count)
+static ssize_t store_down_threshold(struct gov_attr_set *attr_set,
+				    const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	struct cs_dbs_tuners *cs_tuners = dbs_data->tuners;
 	unsigned int input;
 	int ret;
@@ -175,9 +178,10 @@ static ssize_t store_down_threshold(struct dbs_data *dbs_data, const char *buf,
 	return count;
 }
 
-static ssize_t store_ignore_nice_load(struct dbs_data *dbs_data,
-		const char *buf, size_t count)
+static ssize_t store_ignore_nice_load(struct gov_attr_set *attr_set,
+				      const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	unsigned int input;
 	int ret;
 
@@ -199,9 +203,10 @@ static ssize_t store_ignore_nice_load(struct dbs_data *dbs_data,
 	return count;
 }
 
-static ssize_t store_freq_step(struct dbs_data *dbs_data, const char *buf,
-		size_t count)
+static ssize_t store_freq_step(struct gov_attr_set *attr_set, const char *buf,
+			       size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	struct cs_dbs_tuners *cs_tuners = dbs_data->tuners;
 	unsigned int input;
 	int ret;

drivers/cpufreq/cpufreq_governor.c

@@ -43,9 +43,10 @@ static DEFINE_MUTEX(gov_dbs_data_mutex);
  * This must be called with dbs_data->mutex held, otherwise traversing
  * policy_dbs_list isn't safe.
  */
-ssize_t store_sampling_rate(struct dbs_data *dbs_data, const char *buf,
+ssize_t store_sampling_rate(struct gov_attr_set *attr_set, const char *buf,
 			    size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	struct policy_dbs_info *policy_dbs;
 	unsigned int rate;
 	int ret;
@@ -59,7 +60,7 @@ ssize_t store_sampling_rate(struct dbs_data *dbs_data, const char *buf,
 	 * We are operating under dbs_data->mutex and so the list and its
 	 * entries can't be freed concurrently.
 	 */
-	list_for_each_entry(policy_dbs, &dbs_data->policy_dbs_list, list) {
+	list_for_each_entry(policy_dbs, &attr_set->policy_list, list) {
 		mutex_lock(&policy_dbs->timer_mutex);
 		/*
 		 * On 32-bit architectures this may race with the
@@ -96,13 +97,13 @@ void gov_update_cpu_data(struct dbs_data *dbs_data)
 {
 	struct policy_dbs_info *policy_dbs;
 
-	list_for_each_entry(policy_dbs, &dbs_data->policy_dbs_list, list) {
+	list_for_each_entry(policy_dbs, &dbs_data->attr_set.policy_list, list) {
 		unsigned int j;
 
 		for_each_cpu(j, policy_dbs->policy->cpus) {
 			struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
 
-			j_cdbs->prev_cpu_idle = get_cpu_idle_time(j, &j_cdbs->prev_cpu_wall,
+			j_cdbs->prev_cpu_idle = get_cpu_idle_time(j, &j_cdbs->prev_update_time,
 								  dbs_data->io_is_busy);
 			if (dbs_data->ignore_nice_load)
 				j_cdbs->prev_cpu_nice = kcpustat_cpu(j).cpustat[CPUTIME_NICE];
@@ -111,54 +112,6 @@ void gov_update_cpu_data(struct dbs_data *dbs_data)
 }
 EXPORT_SYMBOL_GPL(gov_update_cpu_data);
 
-static inline struct dbs_data *to_dbs_data(struct kobject *kobj)
-{
-	return container_of(kobj, struct dbs_data, kobj);
-}
-
-static inline struct governor_attr *to_gov_attr(struct attribute *attr)
-{
-	return container_of(attr, struct governor_attr, attr);
-}
-
-static ssize_t governor_show(struct kobject *kobj, struct attribute *attr,
-			     char *buf)
-{
-	struct dbs_data *dbs_data = to_dbs_data(kobj);
-	struct governor_attr *gattr = to_gov_attr(attr);
-
-	return gattr->show(dbs_data, buf);
-}
-
-static ssize_t governor_store(struct kobject *kobj, struct attribute *attr,
-			      const char *buf, size_t count)
-{
-	struct dbs_data *dbs_data = to_dbs_data(kobj);
-	struct governor_attr *gattr = to_gov_attr(attr);
-	int ret = -EBUSY;
-
-	mutex_lock(&dbs_data->mutex);
-
-	if (dbs_data->usage_count)
-		ret = gattr->store(dbs_data, buf, count);
-
-	mutex_unlock(&dbs_data->mutex);
-
-	return ret;
-}
-
-/*
- * Sysfs Ops for accessing governor attributes.
- *
- * All show/store invocations for governor specific sysfs attributes, will first
- * call the below show/store callbacks and the attribute specific callback will
- * be called from within it.
- */
-static const struct sysfs_ops governor_sysfs_ops = {
-	.show	= governor_show,
-	.store	= governor_store,
-};
-
 unsigned int dbs_update(struct cpufreq_policy *policy)
 {
 	struct policy_dbs_info *policy_dbs = policy->governor_data;
@@ -184,14 +137,14 @@ unsigned int dbs_update(struct cpufreq_policy *policy)
 	/* Get Absolute Load */
 	for_each_cpu(j, policy->cpus) {
 		struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
-		u64 cur_wall_time, cur_idle_time;
-		unsigned int idle_time, wall_time;
+		u64 update_time, cur_idle_time;
+		unsigned int idle_time, time_elapsed;
 		unsigned int load;
 
-		cur_idle_time = get_cpu_idle_time(j, &cur_wall_time, io_busy);
+		cur_idle_time = get_cpu_idle_time(j, &update_time, io_busy);
 
-		wall_time = cur_wall_time - j_cdbs->prev_cpu_wall;
-		j_cdbs->prev_cpu_wall = cur_wall_time;
+		time_elapsed = update_time - j_cdbs->prev_update_time;
+		j_cdbs->prev_update_time = update_time;
 
 		idle_time = cur_idle_time - j_cdbs->prev_cpu_idle;
 		j_cdbs->prev_cpu_idle = cur_idle_time;
@@ -203,47 +156,62 @@ unsigned int dbs_update(struct cpufreq_policy *policy)
 			j_cdbs->prev_cpu_nice = cur_nice;
 		}
 
-		if (unlikely(!wall_time || wall_time < idle_time))
-			continue;
-
-		/*
-		 * If the CPU had gone completely idle, and a task just woke up
-		 * on this CPU now, it would be unfair to calculate 'load' the
-		 * usual way for this elapsed time-window, because it will show
-		 * near-zero load, irrespective of how CPU intensive that task
-		 * actually is. This is undesirable for latency-sensitive bursty
-		 * workloads.
-		 *
-		 * To avoid this, we reuse the 'load' from the previous
-		 * time-window and give this task a chance to start with a
-		 * reasonably high CPU frequency. (However, we shouldn't over-do
-		 * this copy, lest we get stuck at a high load (high frequency)
-		 * for too long, even when the current system load has actually
-		 * dropped down. So we perform the copy only once, upon the
-		 * first wake-up from idle.)
-		 *
-		 * Detecting this situation is easy: the governor's utilization
-		 * update handler would not have run during CPU-idle periods.
-		 * Hence, an unusually large 'wall_time' (as compared to the
-		 * sampling rate) indicates this scenario.
-		 *
-		 * prev_load can be zero in two cases and we must recalculate it
-		 * for both cases:
-		 * - during long idle intervals
-		 * - explicitly set to zero
-		 */
-		if (unlikely(wall_time > (2 * sampling_rate) &&
-			     j_cdbs->prev_load)) {
+		if (unlikely(!time_elapsed)) {
+			/*
+			 * That can only happen when this function is called
+			 * twice in a row with a very short interval between the
+			 * calls, so the previous load value can be used then.
+			 */
 			load = j_cdbs->prev_load;
-
+		} else if (unlikely(time_elapsed > 2 * sampling_rate &&
+				    j_cdbs->prev_load)) {
 			/*
-			 * Perform a destructive copy, to ensure that we copy
-			 * the previous load only once, upon the first wake-up
-			 * from idle.
+			 * If the CPU had gone completely idle and a task has
+			 * just woken up on this CPU now, it would be unfair to
+			 * calculate 'load' the usual way for this elapsed
+			 * time-window, because it would show near-zero load,
+			 * irrespective of how CPU intensive that task actually
+			 * was. This is undesirable for latency-sensitive bursty
+			 * workloads.
+			 *
+			 * To avoid this, reuse the 'load' from the previous
+			 * time-window and give this task a chance to start with
+			 * a reasonably high CPU frequency. However, that
+			 * shouldn't be over-done, lest we get stuck at a high
+			 * load (high frequency) for too long, even when the
+			 * current system load has actually dropped down, so
+			 * clear prev_load to guarantee that the load will be
+			 * computed again next time.
+			 *
+			 * Detecting this situation is easy: the governor's
+			 * utilization update handler would not have run during
+			 * CPU-idle periods.  Hence, an unusually large
+			 * 'time_elapsed' (as compared to the sampling rate)
+			 * indicates this scenario.
 			 */
+			load = j_cdbs->prev_load;
 			j_cdbs->prev_load = 0;
 		} else {
-			load = 100 * (wall_time - idle_time) / wall_time;
+			if (time_elapsed >= idle_time) {
+				load = 100 * (time_elapsed - idle_time) / time_elapsed;
+			} else {
+				/*
+				 * That can happen if idle_time is returned by
+				 * get_cpu_idle_time_jiffy().  In that case
+				 * idle_time is roughly equal to the difference
+				 * between time_elapsed and "busy time" obtained
+				 * from CPU statistics.  Then, the "busy time"
+				 * can end up being greater than time_elapsed
+				 * (for example, if jiffies_64 and the CPU
+				 * statistics are updated by different CPUs),
+				 * so idle_time may in fact be negative.  That
+				 * means, though, that the CPU was busy all
+				 * the time (on the rough average) during the
+				 * last sampling interval and 100 can be
+				 * returned as the load.
+				 */
+				load = (int)idle_time < 0 ? 100 : 0;
+			}
 			j_cdbs->prev_load = load;
 		}
 
@@ -254,43 +222,6 @@ unsigned int dbs_update(struct cpufreq_policy *policy)
 }
 EXPORT_SYMBOL_GPL(dbs_update);
 
-static void gov_set_update_util(struct policy_dbs_info *policy_dbs,
-				unsigned int delay_us)
-{
-	struct cpufreq_policy *policy = policy_dbs->policy;
-	int cpu;
-
-	gov_update_sample_delay(policy_dbs, delay_us);
-	policy_dbs->last_sample_time = 0;
-
-	for_each_cpu(cpu, policy->cpus) {
-		struct cpu_dbs_info *cdbs = &per_cpu(cpu_dbs, cpu);
-
-		cpufreq_set_update_util_data(cpu, &cdbs->update_util);
-	}
-}
-
-static inline void gov_clear_update_util(struct cpufreq_policy *policy)
-{
-	int i;
-
-	for_each_cpu(i, policy->cpus)
-		cpufreq_set_update_util_data(i, NULL);
-
-	synchronize_sched();
-}
-
-static void gov_cancel_work(struct cpufreq_policy *policy)
-{
-	struct policy_dbs_info *policy_dbs = policy->governor_data;
-
-	gov_clear_update_util(policy_dbs->policy);
-	irq_work_sync(&policy_dbs->irq_work);
-	cancel_work_sync(&policy_dbs->work);
-	atomic_set(&policy_dbs->work_count, 0);
-	policy_dbs->work_in_progress = false;
-}
-
 static void dbs_work_handler(struct work_struct *work)
 {
 	struct policy_dbs_info *policy_dbs;
@@ -378,6 +309,44 @@ static void dbs_update_util_handler(struct update_util_data *data, u64 time,
 	irq_work_queue(&policy_dbs->irq_work);
 }
 
+static void gov_set_update_util(struct policy_dbs_info *policy_dbs,
+				unsigned int delay_us)
+{
+	struct cpufreq_policy *policy = policy_dbs->policy;
+	int cpu;
+
+	gov_update_sample_delay(policy_dbs, delay_us);
+	policy_dbs->last_sample_time = 0;
+
+	for_each_cpu(cpu, policy->cpus) {
+		struct cpu_dbs_info *cdbs = &per_cpu(cpu_dbs, cpu);
+
+		cpufreq_add_update_util_hook(cpu, &cdbs->update_util,
+					     dbs_update_util_handler);
+	}
+}
+
+static inline void gov_clear_update_util(struct cpufreq_policy *policy)
+{
+	int i;
+
+	for_each_cpu(i, policy->cpus)
+		cpufreq_remove_update_util_hook(i);
+
+	synchronize_sched();
+}
+
+static void gov_cancel_work(struct cpufreq_policy *policy)
+{
+	struct policy_dbs_info *policy_dbs = policy->governor_data;
+
+	gov_clear_update_util(policy_dbs->policy);
+	irq_work_sync(&policy_dbs->irq_work);
+	cancel_work_sync(&policy_dbs->work);
+	atomic_set(&policy_dbs->work_count, 0);
+	policy_dbs->work_in_progress = false;
+}
+
 static struct policy_dbs_info *alloc_policy_dbs_info(struct cpufreq_policy *policy,
 						     struct dbs_governor *gov)
 {
@@ -400,7 +369,6 @@ static struct policy_dbs_info *alloc_policy_dbs_info(struct cpufreq_policy *poli
 		struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
 
 		j_cdbs->policy_dbs = policy_dbs;
-		j_cdbs->update_util.func = dbs_update_util_handler;
 	}
 	return policy_dbs;
 }
@@ -449,10 +417,7 @@ static int cpufreq_governor_init(struct cpufreq_policy *policy)
 		policy_dbs->dbs_data = dbs_data;
 		policy->governor_data = policy_dbs;
 
-		mutex_lock(&dbs_data->mutex);
-		dbs_data->usage_count++;
-		list_add(&policy_dbs->list, &dbs_data->policy_dbs_list);
-		mutex_unlock(&dbs_data->mutex);
+		gov_attr_set_get(&dbs_data->attr_set, &policy_dbs->list);
 		goto out;
 	}
 
@@ -462,8 +427,7 @@ static int cpufreq_governor_init(struct cpufreq_policy *policy)
 		goto free_policy_dbs_info;
 	}
 
-	INIT_LIST_HEAD(&dbs_data->policy_dbs_list);
-	mutex_init(&dbs_data->mutex);
+	gov_attr_set_init(&dbs_data->attr_set, &policy_dbs->list);
 
 	ret = gov->init(dbs_data, !policy->governor->initialized);
 	if (ret)
@@ -483,14 +447,11 @@ static int cpufreq_governor_init(struct cpufreq_policy *policy)
 	if (!have_governor_per_policy())
 		gov->gdbs_data = dbs_data;
 
-	policy->governor_data = policy_dbs;
-
 	policy_dbs->dbs_data = dbs_data;
-	dbs_data->usage_count = 1;
-	list_add(&policy_dbs->list, &dbs_data->policy_dbs_list);
+	policy->governor_data = policy_dbs;
 
 	gov->kobj_type.sysfs_ops = &governor_sysfs_ops;
-	ret = kobject_init_and_add(&dbs_data->kobj, &gov->kobj_type,
+	ret = kobject_init_and_add(&dbs_data->attr_set.kobj, &gov->kobj_type,
 				   get_governor_parent_kobj(policy),
 				   "%s", gov->gov.name);
 	if (!ret)
@@ -519,29 +480,21 @@ static int cpufreq_governor_exit(struct cpufreq_policy *policy)
 	struct dbs_governor *gov = dbs_governor_of(policy);
 	struct policy_dbs_info *policy_dbs = policy->governor_data;
 	struct dbs_data *dbs_data = policy_dbs->dbs_data;
-	int count;
+	unsigned int count;
 
 	/* Protect gov->gdbs_data against concurrent updates. */
 	mutex_lock(&gov_dbs_data_mutex);
 
-	mutex_lock(&dbs_data->mutex);
-	list_del(&policy_dbs->list);
-	count = --dbs_data->usage_count;
-	mutex_unlock(&dbs_data->mutex);
+	count = gov_attr_set_put(&dbs_data->attr_set, &policy_dbs->list);
 
-	if (!count) {
-		kobject_put(&dbs_data->kobj);
-
-		policy->governor_data = NULL;
+	policy->governor_data = NULL;
 
+	if (!count) {
 		if (!have_governor_per_policy())
 			gov->gdbs_data = NULL;
 
 		gov->exit(dbs_data, policy->governor->initialized == 1);
-		mutex_destroy(&dbs_data->mutex);
 		kfree(dbs_data);
-	} else {
-		policy->governor_data = NULL;
 	}
 
 	free_policy_dbs_info(policy_dbs, gov);
@@ -570,12 +523,12 @@ static int cpufreq_governor_start(struct cpufreq_policy *policy)
 
 	for_each_cpu(j, policy->cpus) {
 		struct cpu_dbs_info *j_cdbs = &per_cpu(cpu_dbs, j);
-		unsigned int prev_load;
 
-		j_cdbs->prev_cpu_idle = get_cpu_idle_time(j, &j_cdbs->prev_cpu_wall, io_busy);
-
-		prev_load = j_cdbs->prev_cpu_wall - j_cdbs->prev_cpu_idle;
-		j_cdbs->prev_load = 100 * prev_load / (unsigned int)j_cdbs->prev_cpu_wall;
+		j_cdbs->prev_cpu_idle = get_cpu_idle_time(j, &j_cdbs->prev_update_time, io_busy);
+		/*
+		 * Make the first invocation of dbs_update() compute the load.
+		 */
+		j_cdbs->prev_load = 0;
 
 		if (ignore_nice)
 			j_cdbs->prev_cpu_nice = kcpustat_cpu(j).cpustat[CPUTIME_NICE];

drivers/cpufreq/cpufreq_governor.h

@@ -24,20 +24,6 @@
 #include <linux/module.h>
 #include <linux/mutex.h>
 
-/*
- * The polling frequency depends on the capability of the processor. Default
- * polling frequency is 1000 times the transition latency of the processor. The
- * governor will work on any processor with transition latency <= 10ms, using
- * appropriate sampling rate.
- *
- * For CPUs with transition latency > 10ms (mostly drivers with CPUFREQ_ETERNAL)
- * this governor will not work. All times here are in us (micro seconds).
- */
-#define MIN_SAMPLING_RATE_RATIO			(2)
-#define LATENCY_MULTIPLIER			(1000)
-#define MIN_LATENCY_MULTIPLIER			(20)
-#define TRANSITION_LATENCY_LIMIT		(10 * 1000 * 1000)
-
 /* Ondemand Sampling types */
 enum {OD_NORMAL_SAMPLE, OD_SUB_SAMPLE};
 
@@ -52,7 +38,7 @@ enum {OD_NORMAL_SAMPLE, OD_SUB_SAMPLE};
 
 /* Governor demand based switching data (per-policy or global). */
 struct dbs_data {
-	int usage_count;
+	struct gov_attr_set attr_set;
 	void *tuners;
 	unsigned int min_sampling_rate;
 	unsigned int ignore_nice_load;
@@ -60,37 +46,27 @@ struct dbs_data {
 	unsigned int sampling_down_factor;
 	unsigned int up_threshold;
 	unsigned int io_is_busy;
-
-	struct kobject kobj;
-	struct list_head policy_dbs_list;
-	/*
-	 * Protect concurrent updates to governor tunables from sysfs,
-	 * policy_dbs_list and usage_count.
-	 */
-	struct mutex mutex;
 };
 
-/* Governor's specific attributes */
-struct dbs_data;
-struct governor_attr {
-	struct attribute attr;
-	ssize_t (*show)(struct dbs_data *dbs_data, char *buf);
-	ssize_t (*store)(struct dbs_data *dbs_data, const char *buf,
-			 size_t count);
-};
+static inline struct dbs_data *to_dbs_data(struct gov_attr_set *attr_set)
+{
+	return container_of(attr_set, struct dbs_data, attr_set);
+}
 
 #define gov_show_one(_gov, file_name)					\
 static ssize_t show_##file_name						\
-(struct dbs_data *dbs_data, char *buf)					\
+(struct gov_attr_set *attr_set, char *buf)				\
 {									\
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);		\
 	struct _gov##_dbs_tuners *tuners = dbs_data->tuners;		\
 	return sprintf(buf, "%u\n", tuners->file_name);			\
 }
 
 #define gov_show_one_common(file_name)					\
 static ssize_t show_##file_name						\
-(struct dbs_data *dbs_data, char *buf)					\
+(struct gov_attr_set *attr_set, char *buf)				\
 {									\
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);		\
 	return sprintf(buf, "%u\n", dbs_data->file_name);		\
 }
 
@@ -135,7 +111,7 @@ static inline void gov_update_sample_delay(struct policy_dbs_info *policy_dbs,
 /* Per cpu structures */
 struct cpu_dbs_info {
 	u64 prev_cpu_idle;
-	u64 prev_cpu_wall;
+	u64 prev_update_time;
 	u64 prev_cpu_nice;
 	/*
 	 * Used to keep track of load in the previous interval. However, when
@@ -184,7 +160,7 @@ void od_register_powersave_bias_handler(unsigned int (*f)
 		(struct cpufreq_policy *, unsigned int, unsigned int),
 		unsigned int powersave_bias);
 void od_unregister_powersave_bias_handler(void);
-ssize_t store_sampling_rate(struct dbs_data *dbs_data, const char *buf,
+ssize_t store_sampling_rate(struct gov_attr_set *attr_set, const char *buf,
 			    size_t count);
 void gov_update_cpu_data(struct dbs_data *dbs_data);
 #endif /* _CPUFREQ_GOVERNOR_H */

drivers/cpufreq/cpufreq_governor_attr_set.c

@@ -0,0 +1,84 @@
+/*
+ * Abstract code for CPUFreq governor tunable sysfs attributes.
+ *
+ * Copyright (C) 2016, Intel Corporation
+ * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include "cpufreq_governor.h"
+
+static inline struct gov_attr_set *to_gov_attr_set(struct kobject *kobj)
+{
+	return container_of(kobj, struct gov_attr_set, kobj);
+}
+
+static inline struct governor_attr *to_gov_attr(struct attribute *attr)
+{
+	return container_of(attr, struct governor_attr, attr);
+}
+
+static ssize_t governor_show(struct kobject *kobj, struct attribute *attr,
+			     char *buf)
+{
+	struct governor_attr *gattr = to_gov_attr(attr);
+
+	return gattr->show(to_gov_attr_set(kobj), buf);
+}
+
+static ssize_t governor_store(struct kobject *kobj, struct attribute *attr,
+			      const char *buf, size_t count)
+{
+	struct gov_attr_set *attr_set = to_gov_attr_set(kobj);
+	struct governor_attr *gattr = to_gov_attr(attr);
+	int ret;
+
+	mutex_lock(&attr_set->update_lock);
+	ret = attr_set->usage_count ? gattr->store(attr_set, buf, count) : -EBUSY;
+	mutex_unlock(&attr_set->update_lock);
+	return ret;
+}
+
+const struct sysfs_ops governor_sysfs_ops = {
+	.show	= governor_show,
+	.store	= governor_store,
+};
+EXPORT_SYMBOL_GPL(governor_sysfs_ops);
+
+void gov_attr_set_init(struct gov_attr_set *attr_set, struct list_head *list_node)
+{
+	INIT_LIST_HEAD(&attr_set->policy_list);
+	mutex_init(&attr_set->update_lock);
+	attr_set->usage_count = 1;
+	list_add(list_node, &attr_set->policy_list);
+}
+EXPORT_SYMBOL_GPL(gov_attr_set_init);
+
+void gov_attr_set_get(struct gov_attr_set *attr_set, struct list_head *list_node)
+{
+	mutex_lock(&attr_set->update_lock);
+	attr_set->usage_count++;
+	list_add(list_node, &attr_set->policy_list);
+	mutex_unlock(&attr_set->update_lock);
+}
+EXPORT_SYMBOL_GPL(gov_attr_set_get);
+
+unsigned int gov_attr_set_put(struct gov_attr_set *attr_set, struct list_head *list_node)
+{
+	unsigned int count;
+
+	mutex_lock(&attr_set->update_lock);
+	list_del(list_node);
+	count = --attr_set->usage_count;
+	mutex_unlock(&attr_set->update_lock);
+	if (count)
+		return count;
+
+	kobject_put(&attr_set->kobj);
+	mutex_destroy(&attr_set->update_lock);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gov_attr_set_put);

drivers/cpufreq/cpufreq_ondemand.c

@@ -207,9 +207,10 @@ static unsigned int od_dbs_timer(struct cpufreq_policy *policy)
 /************************** sysfs interface ************************/
 static struct dbs_governor od_dbs_gov;
 
-static ssize_t store_io_is_busy(struct dbs_data *dbs_data, const char *buf,
-		size_t count)
+static ssize_t store_io_is_busy(struct gov_attr_set *attr_set, const char *buf,
+				size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	unsigned int input;
 	int ret;
 
@@ -224,9 +225,10 @@ static ssize_t store_io_is_busy(struct dbs_data *dbs_data, const char *buf,
 	return count;
 }
 
-static ssize_t store_up_threshold(struct dbs_data *dbs_data, const char *buf,
-		size_t count)
+static ssize_t store_up_threshold(struct gov_attr_set *attr_set,
+				  const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	unsigned int input;
 	int ret;
 	ret = sscanf(buf, "%u", &input);
@@ -240,9 +242,10 @@ static ssize_t store_up_threshold(struct dbs_data *dbs_data, const char *buf,
 	return count;
 }
 
-static ssize_t store_sampling_down_factor(struct dbs_data *dbs_data,
-		const char *buf, size_t count)
+static ssize_t store_sampling_down_factor(struct gov_attr_set *attr_set,
+					  const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	struct policy_dbs_info *policy_dbs;
 	unsigned int input;
 	int ret;
@@ -254,7 +257,7 @@ static ssize_t store_sampling_down_factor(struct dbs_data *dbs_data,
 	dbs_data->sampling_down_factor = input;
 
 	/* Reset down sampling multiplier in case it was active */
-	list_for_each_entry(policy_dbs, &dbs_data->policy_dbs_list, list) {
+	list_for_each_entry(policy_dbs, &attr_set->policy_list, list) {
 		/*
 		 * Doing this without locking might lead to using different
 		 * rate_mult values in od_update() and od_dbs_timer().
@@ -267,9 +270,10 @@ static ssize_t store_sampling_down_factor(struct dbs_data *dbs_data,
 	return count;
 }
 
-static ssize_t store_ignore_nice_load(struct dbs_data *dbs_data,
-		const char *buf, size_t count)
+static ssize_t store_ignore_nice_load(struct gov_attr_set *attr_set,
+				      const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	unsigned int input;
 	int ret;
 
@@ -291,9 +295,10 @@ static ssize_t store_ignore_nice_load(struct dbs_data *dbs_data,
 	return count;
 }
 
-static ssize_t store_powersave_bias(struct dbs_data *dbs_data, const char *buf,
-		size_t count)
+static ssize_t store_powersave_bias(struct gov_attr_set *attr_set,
+				    const char *buf, size_t count)
 {
+	struct dbs_data *dbs_data = to_dbs_data(attr_set);
 	struct od_dbs_tuners *od_tuners = dbs_data->tuners;
 	struct policy_dbs_info *policy_dbs;
 	unsigned int input;
@@ -308,7 +313,7 @@ static ssize_t store_powersave_bias(struct dbs_data *dbs_data, const char *buf,
 
 	od_tuners->powersave_bias = input;
 
-	list_for_each_entry(policy_dbs, &dbs_data->policy_dbs_list, list)
+	list_for_each_entry(policy_dbs, &attr_set->policy_list, list)
 		ondemand_powersave_bias_init(policy_dbs->policy);
 
 	return count;

drivers/cpufreq/cpufreq_userspace.c

@@ -17,6 +17,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
+#include <linux/slab.h>
 
 static DEFINE_PER_CPU(unsigned int, cpu_is_managed);
 static DEFINE_MUTEX(userspace_mutex);
@@ -31,6 +32,7 @@ static DEFINE_MUTEX(userspace_mutex);
 static int cpufreq_set(struct cpufreq_policy *policy, unsigned int freq)
 {
 	int ret = -EINVAL;
+	unsigned int *setspeed = policy->governor_data;
 
 	pr_debug("cpufreq_set for cpu %u, freq %u kHz\n", policy->cpu, freq);
 
@@ -38,6 +40,8 @@ static int cpufreq_set(struct cpufreq_policy *policy, unsigned int freq)
 	if (!per_cpu(cpu_is_managed, policy->cpu))
 		goto err;
 
+	*setspeed = freq;
+
 	ret = __cpufreq_driver_target(policy, freq, CPUFREQ_RELATION_L);
  err:
 	mutex_unlock(&userspace_mutex);
@@ -49,19 +53,45 @@ static ssize_t show_speed(struct cpufreq_policy *policy, char *buf)
 	return sprintf(buf, "%u\n", policy->cur);
 }
 
+static int cpufreq_userspace_policy_init(struct cpufreq_policy *policy)
+{
+	unsigned int *setspeed;
+
+	setspeed = kzalloc(sizeof(*setspeed), GFP_KERNEL);
+	if (!setspeed)
+		return -ENOMEM;
+
+	policy->governor_data = setspeed;
+	return 0;
+}
+
 static int cpufreq_governor_userspace(struct cpufreq_policy *policy,
 				   unsigned int event)
 {
+	unsigned int *setspeed = policy->governor_data;
 	unsigned int cpu = policy->cpu;
 	int rc = 0;
 
+	if (event == CPUFREQ_GOV_POLICY_INIT)
+		return cpufreq_userspace_policy_init(policy);
+
+	if (!setspeed)
+		return -EINVAL;
+
 	switch (event) {
+	case CPUFREQ_GOV_POLICY_EXIT:
+		mutex_lock(&userspace_mutex);
+		policy->governor_data = NULL;
+		kfree(setspeed);
+		mutex_unlock(&userspace_mutex);
+		break;
 	case CPUFREQ_GOV_START:
 		BUG_ON(!policy->cur);
 		pr_debug("started managing cpu %u\n", cpu);
 
 		mutex_lock(&userspace_mutex);
 		per_cpu(cpu_is_managed, cpu) = 1;
+		*setspeed = policy->cur;
 		mutex_unlock(&userspace_mutex);
 		break;
 	case CPUFREQ_GOV_STOP:
@@ -69,20 +99,23 @@ static int cpufreq_governor_userspace(struct cpufreq_policy *policy,
 
 		mutex_lock(&userspace_mutex);
 		per_cpu(cpu_is_managed, cpu) = 0;
+		*setspeed = 0;
 		mutex_unlock(&userspace_mutex);
 		break;
 	case CPUFREQ_GOV_LIMITS:
 		mutex_lock(&userspace_mutex);
-		pr_debug("limit event for cpu %u: %u - %u kHz, currently %u kHz\n",
-			cpu, policy->min, policy->max,
-			policy->cur);
+		pr_debug("limit event for cpu %u: %u - %u kHz, currently %u kHz, last set to %u kHz\n",
+			cpu, policy->min, policy->max, policy->cur, *setspeed);
 
-		if (policy->max < policy->cur)
+		if (policy->max < *setspeed)
 			__cpufreq_driver_target(policy, policy->max,
 						CPUFREQ_RELATION_H);
-		else if (policy->min > policy->cur)
+		else if (policy->min > *setspeed)
 			__cpufreq_driver_target(policy, policy->min,
 						CPUFREQ_RELATION_L);
+		else
+			__cpufreq_driver_target(policy, *setspeed,
+						CPUFREQ_RELATION_L);
 		mutex_unlock(&userspace_mutex);
 		break;
 	}

drivers/cpufreq/e_powersaver.c

@@ -6,6 +6,8 @@
  *  BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous*
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -20,7 +22,7 @@
 #include <asm/msr.h>
 #include <asm/tsc.h>
 
-#if defined CONFIG_ACPI_PROCESSOR || defined CONFIG_ACPI_PROCESSOR_MODULE
+#if IS_ENABLED(CONFIG_ACPI_PROCESSOR)
 #include <linux/acpi.h>
 #include <acpi/processor.h>
 #endif
@@ -33,7 +35,7 @@
 
 struct eps_cpu_data {
 	u32 fsb;
-#if defined CONFIG_ACPI_PROCESSOR || defined CONFIG_ACPI_PROCESSOR_MODULE
+#if IS_ENABLED(CONFIG_ACPI_PROCESSOR)
 	u32 bios_limit;
 #endif
 	struct cpufreq_frequency_table freq_table[];
@@ -46,7 +48,7 @@ static int freq_failsafe_off;
 static int voltage_failsafe_off;
 static int set_max_voltage;
 
-#if defined CONFIG_ACPI_PROCESSOR || defined CONFIG_ACPI_PROCESSOR_MODULE
+#if IS_ENABLED(CONFIG_ACPI_PROCESSOR)
 static int ignore_acpi_limit;
 
 static struct acpi_processor_performance *eps_acpi_cpu_perf;
@@ -141,11 +143,9 @@ static int eps_set_state(struct eps_cpu_data *centaur,
 	/* Print voltage and multiplier */
 	rdmsr(MSR_IA32_PERF_STATUS, lo, hi);
 	current_voltage = lo & 0xff;
-	printk(KERN_INFO "eps: Current voltage = %dmV\n",
-		current_voltage * 16 + 700);
+	pr_info("Current voltage = %dmV\n", current_voltage * 16 + 700);
 	current_multiplier = (lo >> 8) & 0xff;
-	printk(KERN_INFO "eps: Current multiplier = %d\n",
-		current_multiplier);
+	pr_info("Current multiplier = %d\n", current_multiplier);
 	}
 #endif
 	return 0;
@@ -166,7 +166,7 @@ static int eps_target(struct cpufreq_policy *policy, unsigned int index)
 	dest_state = centaur->freq_table[index].driver_data & 0xffff;
 	ret = eps_set_state(centaur, policy, dest_state);
 	if (ret)
-		printk(KERN_ERR "eps: Timeout!\n");
+		pr_err("Timeout!\n");
 	return ret;
 }
 
@@ -186,7 +186,7 @@ static int eps_cpu_init(struct cpufreq_policy *policy)
 	int k, step, voltage;
 	int ret;
 	int states;
-#if defined CONFIG_ACPI_PROCESSOR || defined CONFIG_ACPI_PROCESSOR_MODULE
+#if IS_ENABLED(CONFIG_ACPI_PROCESSOR)
 	unsigned int limit;
 #endif
 
@@ -194,36 +194,36 @@ static int eps_cpu_init(struct cpufreq_policy *policy)
 		return -ENODEV;
 
 	/* Check brand */
-	printk(KERN_INFO "eps: Detected VIA ");
+	pr_info("Detected VIA ");
 
 	switch (c->x86_model) {
 	case 10:
 		rdmsr(0x1153, lo, hi);
 		brand = (((lo >> 2) ^ lo) >> 18) & 3;
-		printk(KERN_CONT "Model A ");
+		pr_cont("Model A ");
 		break;
 	case 13:
 		rdmsr(0x1154, lo, hi);
 		brand = (((lo >> 4) ^ (lo >> 2))) & 0x000000ff;
-		printk(KERN_CONT "Model D ");
+		pr_cont("Model D ");
 		break;
 	}
 
 	switch (brand) {
 	case EPS_BRAND_C7M:
-		printk(KERN_CONT "C7-M\n");
+		pr_cont("C7-M\n");
 		break;
 	case EPS_BRAND_C7:
-		printk(KERN_CONT "C7\n");
+		pr_cont("C7\n");
 		break;
 	case EPS_BRAND_EDEN:
-		printk(KERN_CONT "Eden\n");
+		pr_cont("Eden\n");
 		break;
 	case EPS_BRAND_C7D:
-		printk(KERN_CONT "C7-D\n");
+		pr_cont("C7-D\n");
 		break;
 	case EPS_BRAND_C3:
-		printk(KERN_CONT "C3\n");
+		pr_cont("C3\n");
 		return -ENODEV;
 		break;
 	}
@@ -235,7 +235,7 @@ static int eps_cpu_init(struct cpufreq_policy *policy)
 		/* Can be locked at 0 */
 		rdmsrl(MSR_IA32_MISC_ENABLE, val);
 		if (!(val & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) {
-			printk(KERN_INFO "eps: Can't enable Enhanced PowerSaver\n");
+			pr_info("Can't enable Enhanced PowerSaver\n");
 			return -ENODEV;
 		}
 	}
@@ -243,22 +243,19 @@ static int eps_cpu_init(struct cpufreq_policy *policy)
 	/* Print voltage and multiplier */
 	rdmsr(MSR_IA32_PERF_STATUS, lo, hi);
 	current_voltage = lo & 0xff;
-	printk(KERN_INFO "eps: Current voltage = %dmV\n",
-			current_voltage * 16 + 700);
+	pr_info("Current voltage = %dmV\n", current_voltage * 16 + 700);
 	current_multiplier = (lo >> 8) & 0xff;
-	printk(KERN_INFO "eps: Current multiplier = %d\n", current_multiplier);
+	pr_info("Current multiplier = %d\n", current_multiplier);
 
 	/* Print limits */
 	max_voltage = hi & 0xff;
-	printk(KERN_INFO "eps: Highest voltage = %dmV\n",
-			max_voltage * 16 + 700);
+	pr_info("Highest voltage = %dmV\n", max_voltage * 16 + 700);
 	max_multiplier = (hi >> 8) & 0xff;
-	printk(KERN_INFO "eps: Highest multiplier = %d\n", max_multiplier);
+	pr_info("Highest multiplier = %d\n", max_multiplier);
 	min_voltage = (hi >> 16) & 0xff;
-	printk(KERN_INFO "eps: Lowest voltage = %dmV\n",
-			min_voltage * 16 + 700);
+	pr_info("Lowest voltage = %dmV\n", min_voltage * 16 + 700);
 	min_multiplier = (hi >> 24) & 0xff;
-	printk(KERN_INFO "eps: Lowest multiplier = %d\n", min_multiplier);
+	pr_info("Lowest multiplier = %d\n", min_multiplier);
 
 	/* Sanity checks */
 	if (current_multiplier == 0 || max_multiplier == 0
@@ -276,34 +273,30 @@ static int eps_cpu_init(struct cpufreq_policy *policy)
 
 	/* Check for systems using underclocked CPU */
 	if (!freq_failsafe_off && max_multiplier != current_multiplier) {
-		printk(KERN_INFO "eps: Your processor is running at different "
-			"frequency then its maximum. Aborting.\n");
-		printk(KERN_INFO "eps: You can use freq_failsafe_off option "
-			"to disable this check.\n");
+		pr_info("Your processor is running at different frequency then its maximum. Aborting.\n");
+		pr_info("You can use freq_failsafe_off option to disable this check.\n");
 		return -EINVAL;
 	}
 	if (!voltage_failsafe_off && max_voltage != current_voltage) {
-		printk(KERN_INFO "eps: Your processor is running at different "
-			"voltage then its maximum. Aborting.\n");
-		printk(KERN_INFO "eps: You can use voltage_failsafe_off "
-			"option to disable this check.\n");
+		pr_info("Your processor is running at different voltage then its maximum. Aborting.\n");
+		pr_info("You can use voltage_failsafe_off option to disable this check.\n");
 		return -EINVAL;
 	}
 
 	/* Calc FSB speed */
 	fsb = cpu_khz / current_multiplier;
 
-#if defined CONFIG_ACPI_PROCESSOR || defined CONFIG_ACPI_PROCESSOR_MODULE
+#if IS_ENABLED(CONFIG_ACPI_PROCESSOR)
 	/* Check for ACPI processor speed limit */
 	if (!ignore_acpi_limit && !eps_acpi_init()) {
 		if (!acpi_processor_get_bios_limit(policy->cpu, &limit)) {
-			printk(KERN_INFO "eps: ACPI limit %u.%uGHz\n",
+			pr_info("ACPI limit %u.%uGHz\n",
 				limit/1000000,
 				(limit%1000000)/10000);
 			eps_acpi_exit(policy);
 			/* Check if max_multiplier is in BIOS limits */
 			if (limit && max_multiplier * fsb > limit) {
-				printk(KERN_INFO "eps: Aborting.\n");
+				pr_info("Aborting\n");
 				return -EINVAL;
 			}
 		}
@@ -319,8 +312,7 @@ static int eps_cpu_init(struct cpufreq_policy *policy)
 		v = (set_max_voltage - 700) / 16;
 		/* Check if voltage is within limits */
 		if (v >= min_voltage && v <= max_voltage) {
-			printk(KERN_INFO "eps: Setting %dmV as maximum.\n",
-				v * 16 + 700);
+			pr_info("Setting %dmV as maximum\n", v * 16 + 700);
 			max_voltage = v;
 		}
 	}
@@ -341,7 +333,7 @@ static int eps_cpu_init(struct cpufreq_policy *policy)
 
 	/* Copy basic values */
 	centaur->fsb = fsb;
-#if defined CONFIG_ACPI_PROCESSOR || defined CONFIG_ACPI_PROCESSOR_MODULE
+#if IS_ENABLED(CONFIG_ACPI_PROCESSOR)
 	centaur->bios_limit = limit;
 #endif
 
@@ -426,7 +418,7 @@ module_param(freq_failsafe_off, int, 0644);
 MODULE_PARM_DESC(freq_failsafe_off, "Disable current vs max frequency check");
 module_param(voltage_failsafe_off, int, 0644);
 MODULE_PARM_DESC(voltage_failsafe_off, "Disable current vs max voltage check");
-#if defined CONFIG_ACPI_PROCESSOR || defined CONFIG_ACPI_PROCESSOR_MODULE
+#if IS_ENABLED(CONFIG_ACPI_PROCESSOR)
 module_param(ignore_acpi_limit, int, 0644);
 MODULE_PARM_DESC(ignore_acpi_limit, "Don't check ACPI's processor speed limit");
 #endif

drivers/cpufreq/elanfreq.c

@@ -16,6 +16,8 @@
  *
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -185,7 +187,7 @@ static int elanfreq_cpu_init(struct cpufreq_policy *policy)
 static int __init elanfreq_setup(char *str)
 {
 	max_freq = simple_strtoul(str, &str, 0);
-	printk(KERN_WARNING "You're using the deprecated elanfreq command line option. Use elanfreq.max_freq instead, please!\n");
+	pr_warn("You're using the deprecated elanfreq command line option. Use elanfreq.max_freq instead, please!\n");
 	return 1;
 }
 __setup("elanfreq=", elanfreq_setup);

drivers/cpufreq/hisi-acpu-cpufreq.c

@@ -1,42 +0,0 @@
-/*
- * Hisilicon Platforms Using ACPU CPUFreq Support
- *
- * Copyright (c) 2015 Hisilicon Limited.
- * Copyright (c) 2015 Linaro Limited.
- *
- * Leo Yan <leo.yan@linaro.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed "as is" WITHOUT ANY WARRANTY of any
- * kind, whether express or implied; without even the implied warranty
- * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/err.h>
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/of.h>
-#include <linux/platform_device.h>
-
-static int __init hisi_acpu_cpufreq_driver_init(void)
-{
-	struct platform_device *pdev;
-
-	if (!of_machine_is_compatible("hisilicon,hi6220"))
-		return -ENODEV;
-
-	pdev = platform_device_register_simple("cpufreq-dt", -1, NULL, 0);
-	return PTR_ERR_OR_ZERO(pdev);
-}
-module_init(hisi_acpu_cpufreq_driver_init);
-
-MODULE_AUTHOR("Leo Yan <leo.yan@linaro.org>");
-MODULE_DESCRIPTION("Hisilicon acpu cpufreq driver");
-MODULE_LICENSE("GPL v2");

drivers/cpufreq/ia64-acpi-cpufreq.c

@@ -8,6 +8,8 @@
  *      Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/module.h>
@@ -118,8 +120,7 @@ processor_get_freq (
 
 	if (ret) {
 		set_cpus_allowed_ptr(current, &saved_mask);
-		printk(KERN_WARNING "get performance failed with error %d\n",
-		       ret);
+		pr_warn("get performance failed with error %d\n", ret);
 		ret = 0;
 		goto migrate_end;
 	}
@@ -177,7 +178,7 @@ processor_set_freq (
 
 	ret = processor_set_pstate(value);
 	if (ret) {
-		printk(KERN_WARNING "Transition failed with error %d\n", ret);
+		pr_warn("Transition failed with error %d\n", ret);
 		retval = -ENODEV;
 		goto migrate_end;
 	}
@@ -291,8 +292,7 @@ acpi_cpufreq_cpu_init (
 	/* notify BIOS that we exist */
 	acpi_processor_notify_smm(THIS_MODULE);
 
-	printk(KERN_INFO "acpi-cpufreq: CPU%u - ACPI performance management "
-	       "activated.\n", cpu);
+	pr_info("CPU%u - ACPI performance management activated\n", cpu);
 
 	for (i = 0; i < data->acpi_data.state_count; i++)
 		pr_debug("     %cP%d: %d MHz, %d mW, %d uS, %d uS, 0x%x 0x%x\n",

drivers/cpufreq/intel_pstate.c

@@ -10,6 +10,8 @@
  * of the License.
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/kernel_stat.h>
 #include <linux/module.h>
@@ -39,10 +41,17 @@
 #define ATOM_TURBO_RATIOS	0x66c
 #define ATOM_TURBO_VIDS		0x66d
 
+#ifdef CONFIG_ACPI
+#include <acpi/processor.h>
+#endif
+
 #define FRAC_BITS 8
 #define int_tofp(X) ((int64_t)(X) << FRAC_BITS)
 #define fp_toint(X) ((X) >> FRAC_BITS)
 
+#define EXT_BITS 6
+#define EXT_FRAC_BITS (EXT_BITS + FRAC_BITS)
+
 static inline int32_t mul_fp(int32_t x, int32_t y)
 {
 	return ((int64_t)x * (int64_t)y) >> FRAC_BITS;
@@ -64,12 +73,22 @@ static inline int ceiling_fp(int32_t x)
 	return ret;
 }
 
+static inline u64 mul_ext_fp(u64 x, u64 y)
+{
+	return (x * y) >> EXT_FRAC_BITS;
+}
+
+static inline u64 div_ext_fp(u64 x, u64 y)
+{
+	return div64_u64(x << EXT_FRAC_BITS, y);
+}
+
 /**
  * struct sample -	Store performance sample
- * @core_pct_busy:	Ratio of APERF/MPERF in percent, which is actual
+ * @core_avg_perf:	Ratio of APERF/MPERF which is the actual average
  *			performance during last sample period
  * @busy_scaled:	Scaled busy value which is used to calculate next
- *			P state. This can be different than core_pct_busy
+ *			P state. This can be different than core_avg_perf
  *			to account for cpu idle period
  * @aperf:		Difference of actual performance frequency clock count
  *			read from APERF MSR between last and current sample
@@ -84,7 +103,7 @@ static inline int ceiling_fp(int32_t x)
  * data for choosing next P State.
  */
 struct sample {
-	int32_t core_pct_busy;
+	int32_t core_avg_perf;
 	int32_t busy_scaled;
 	u64 aperf;
 	u64 mperf;
@@ -162,6 +181,7 @@ struct _pid {
  * struct cpudata -	Per CPU instance data storage
  * @cpu:		CPU number for this instance data
  * @update_util:	CPUFreq utility callback information
+ * @update_util_set:	CPUFreq utility callback is set
  * @pstate:		Stores P state limits for this CPU
  * @vid:		Stores VID limits for this CPU
  * @pid:		Stores PID parameters for this CPU
@@ -172,6 +192,8 @@ struct _pid {
  * @prev_cummulative_iowait: IO Wait time difference from last and
  *			current sample
  * @sample:		Storage for storing last Sample data
+ * @acpi_perf_data:	Stores ACPI perf information read from _PSS
+ * @valid_pss_table:	Set to true for valid ACPI _PSS entries found
  *
  * This structure stores per CPU instance data for all CPUs.
  */
@@ -179,6 +201,7 @@ struct cpudata {
 	int cpu;
 
 	struct update_util_data update_util;
+	bool   update_util_set;
 
 	struct pstate_data pstate;
 	struct vid_data vid;
@@ -190,6 +213,10 @@ struct cpudata {
 	u64	prev_tsc;
 	u64	prev_cummulative_iowait;
 	struct sample sample;
+#ifdef CONFIG_ACPI
+	struct acpi_processor_performance acpi_perf_data;
+	bool valid_pss_table;
+#endif
 };
 
 static struct cpudata **all_cpu_data;
@@ -258,6 +285,9 @@ static struct pstate_adjust_policy pid_params;
 static struct pstate_funcs pstate_funcs;
 static int hwp_active;
 
+#ifdef CONFIG_ACPI
+static bool acpi_ppc;
+#endif
 
 /**
  * struct perf_limits - Store user and policy limits
@@ -331,6 +361,124 @@ static struct perf_limits *limits = &performance_limits;
 static struct perf_limits *limits = &powersave_limits;
 #endif
 
+#ifdef CONFIG_ACPI
+
+static bool intel_pstate_get_ppc_enable_status(void)
+{
+	if (acpi_gbl_FADT.preferred_profile == PM_ENTERPRISE_SERVER ||
+	    acpi_gbl_FADT.preferred_profile == PM_PERFORMANCE_SERVER)
+		return true;
+
+	return acpi_ppc;
+}
+
+/*
+ * The max target pstate ratio is a 8 bit value in both PLATFORM_INFO MSR and
+ * in TURBO_RATIO_LIMIT MSR, which pstate driver stores in max_pstate and
+ * max_turbo_pstate fields. The PERF_CTL MSR contains 16 bit value for P state
+ * ratio, out of it only high 8 bits are used. For example 0x1700 is setting
+ * target ratio 0x17. The _PSS control value stores in a format which can be
+ * directly written to PERF_CTL MSR. But in intel_pstate driver this shift
+ * occurs during write to PERF_CTL (E.g. for cores core_set_pstate()).
+ * This function converts the _PSS control value to intel pstate driver format
+ * for comparison and assignment.
+ */
+static int convert_to_native_pstate_format(struct cpudata *cpu, int index)
+{
+	return cpu->acpi_perf_data.states[index].control >> 8;
+}
+
+static void intel_pstate_init_acpi_perf_limits(struct cpufreq_policy *policy)
+{
+	struct cpudata *cpu;
+	int turbo_pss_ctl;
+	int ret;
+	int i;
+
+	if (hwp_active)
+		return;
+
+	if (!intel_pstate_get_ppc_enable_status())
+		return;
+
+	cpu = all_cpu_data[policy->cpu];
+
+	ret = acpi_processor_register_performance(&cpu->acpi_perf_data,
+						  policy->cpu);
+	if (ret)
+		return;
+
+	/*
+	 * Check if the control value in _PSS is for PERF_CTL MSR, which should
+	 * guarantee that the states returned by it map to the states in our
+	 * list directly.
+	 */
+	if (cpu->acpi_perf_data.control_register.space_id !=
+						ACPI_ADR_SPACE_FIXED_HARDWARE)
+		goto err;
+
+	/*
+	 * If there is only one entry _PSS, simply ignore _PSS and continue as
+	 * usual without taking _PSS into account
+	 */
+	if (cpu->acpi_perf_data.state_count < 2)
+		goto err;
+
+	pr_debug("CPU%u - ACPI _PSS perf data\n", policy->cpu);
+	for (i = 0; i < cpu->acpi_perf_data.state_count; i++) {
+		pr_debug("     %cP%d: %u MHz, %u mW, 0x%x\n",
+			 (i == cpu->acpi_perf_data.state ? '*' : ' '), i,
+			 (u32) cpu->acpi_perf_data.states[i].core_frequency,
+			 (u32) cpu->acpi_perf_data.states[i].power,
+			 (u32) cpu->acpi_perf_data.states[i].control);
+	}
+
+	/*
+	 * The _PSS table doesn't contain whole turbo frequency range.
+	 * This just contains +1 MHZ above the max non turbo frequency,
+	 * with control value corresponding to max turbo ratio. But
+	 * when cpufreq set policy is called, it will call with this
+	 * max frequency, which will cause a reduced performance as
+	 * this driver uses real max turbo frequency as the max
+	 * frequency. So correct this frequency in _PSS table to
+	 * correct max turbo frequency based on the turbo ratio.
+	 * Also need to convert to MHz as _PSS freq is in MHz.
+	 */
+	turbo_pss_ctl = convert_to_native_pstate_format(cpu, 0);
+	if (turbo_pss_ctl > cpu->pstate.max_pstate)
+		cpu->acpi_perf_data.states[0].core_frequency =
+					policy->cpuinfo.max_freq / 1000;
+	cpu->valid_pss_table = true;
+	pr_info("_PPC limits will be enforced\n");
+
+	return;
+
+ err:
+	cpu->valid_pss_table = false;
+	acpi_processor_unregister_performance(policy->cpu);
+}
+
+static void intel_pstate_exit_perf_limits(struct cpufreq_policy *policy)
+{
+	struct cpudata *cpu;
+
+	cpu = all_cpu_data[policy->cpu];
+	if (!cpu->valid_pss_table)
+		return;
+
+	acpi_processor_unregister_performance(policy->cpu);
+}
+
+#else
+static void intel_pstate_init_acpi_perf_limits(struct cpufreq_policy *policy)
+{
+}
+
+static void intel_pstate_exit_perf_limits(struct cpufreq_policy *policy)
+{
+}
+#endif
+
 static inline void pid_reset(struct _pid *pid, int setpoint, int busy,
 			     int deadband, int integral) {
 	pid->setpoint = int_tofp(setpoint);
@@ -341,17 +489,17 @@ static inline void pid_reset(struct _pid *pid, int setpoint, int busy,
 
 static inline void pid_p_gain_set(struct _pid *pid, int percent)
 {
-	pid->p_gain = div_fp(int_tofp(percent), int_tofp(100));
+	pid->p_gain = div_fp(percent, 100);
 }
 
 static inline void pid_i_gain_set(struct _pid *pid, int percent)
 {
-	pid->i_gain = div_fp(int_tofp(percent), int_tofp(100));
+	pid->i_gain = div_fp(percent, 100);
 }
 
 static inline void pid_d_gain_set(struct _pid *pid, int percent)
 {
-	pid->d_gain = div_fp(int_tofp(percent), int_tofp(100));
+	pid->d_gain = div_fp(percent, 100);
 }
 
 static signed int pid_calc(struct _pid *pid, int32_t busy)
@@ -537,7 +685,7 @@ static ssize_t show_turbo_pct(struct kobject *kobj,
 
 	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_fp = div_fp(no_turbo, total);
 	turbo_pct = 100 - fp_toint(mul_fp(turbo_fp, int_tofp(100)));
 	return sprintf(buf, "%u\n", turbo_pct);
 }
@@ -579,7 +727,7 @@ static ssize_t store_no_turbo(struct kobject *a, struct attribute *b,
 
 	update_turbo_state();
 	if (limits->turbo_disabled) {
-		pr_warn("intel_pstate: Turbo disabled by BIOS or unavailable on processor\n");
+		pr_warn("Turbo disabled by BIOS or unavailable on processor\n");
 		return -EPERM;
 	}
 
@@ -608,8 +756,7 @@ static ssize_t store_max_perf_pct(struct kobject *a, struct attribute *b,
 				   limits->max_perf_pct);
 	limits->max_perf_pct = max(limits->min_perf_pct,
 				   limits->max_perf_pct);
-	limits->max_perf = div_fp(int_tofp(limits->max_perf_pct),
-				  int_tofp(100));
+	limits->max_perf = div_fp(limits->max_perf_pct, 100);
 
 	if (hwp_active)
 		intel_pstate_hwp_set_online_cpus();
@@ -633,8 +780,7 @@ static ssize_t store_min_perf_pct(struct kobject *a, struct attribute *b,
 				   limits->min_perf_pct);
 	limits->min_perf_pct = min(limits->max_perf_pct,
 				   limits->min_perf_pct);
-	limits->min_perf = div_fp(int_tofp(limits->min_perf_pct),
-				  int_tofp(100));
+	limits->min_perf = div_fp(limits->min_perf_pct, 100);
 
 	if (hwp_active)
 		intel_pstate_hwp_set_online_cpus();
@@ -1019,15 +1165,11 @@ static void intel_pstate_get_cpu_pstates(struct cpudata *cpu)
 	intel_pstate_set_min_pstate(cpu);
 }
 
-static inline void intel_pstate_calc_busy(struct cpudata *cpu)
+static inline void intel_pstate_calc_avg_perf(struct cpudata *cpu)
 {
 	struct sample *sample = &cpu->sample;
-	int64_t core_pct;
 
-	core_pct = int_tofp(sample->aperf) * int_tofp(100);
-	core_pct = div64_u64(core_pct, int_tofp(sample->mperf));
-
-	sample->core_pct_busy = (int32_t)core_pct;
+	sample->core_avg_perf = div_ext_fp(sample->aperf, sample->mperf);
 }
 
 static inline bool intel_pstate_sample(struct cpudata *cpu, u64 time)
@@ -1070,9 +1212,14 @@ static inline bool intel_pstate_sample(struct cpudata *cpu, u64 time)
 
 static inline int32_t get_avg_frequency(struct cpudata *cpu)
 {
-	return fp_toint(mul_fp(cpu->sample.core_pct_busy,
-			       int_tofp(cpu->pstate.max_pstate_physical *
-						cpu->pstate.scaling / 100)));
+	return mul_ext_fp(cpu->sample.core_avg_perf,
+			  cpu->pstate.max_pstate_physical * cpu->pstate.scaling);
+}
+
+static inline int32_t get_avg_pstate(struct cpudata *cpu)
+{
+	return mul_ext_fp(cpu->pstate.max_pstate_physical,
+			  cpu->sample.core_avg_perf);
 }
 
 static inline int32_t get_target_pstate_use_cpu_load(struct cpudata *cpu)
@@ -1107,49 +1254,43 @@ static inline int32_t get_target_pstate_use_cpu_load(struct cpudata *cpu)
 	cpu_load = div64_u64(int_tofp(100) * mperf, sample->tsc);
 	cpu->sample.busy_scaled = cpu_load;
 
-	return cpu->pstate.current_pstate - pid_calc(&cpu->pid, cpu_load);
+	return get_avg_pstate(cpu) - pid_calc(&cpu->pid, cpu_load);
 }
 
 static inline int32_t get_target_pstate_use_performance(struct cpudata *cpu)
 {
-	int32_t core_busy, max_pstate, current_pstate, sample_ratio;
+	int32_t perf_scaled, max_pstate, current_pstate, sample_ratio;
 	u64 duration_ns;
 
 	/*
-	 * core_busy is the ratio of actual performance to max
-	 * max_pstate is the max non turbo pstate available
-	 * current_pstate was the pstate that was requested during
-	 * 	the last sample period.
-	 *
-	 * We normalize core_busy, which was our actual percent
-	 * performance to what we requested during the last sample
-	 * period. The result will be a percentage of busy at a
-	 * specified pstate.
+	 * perf_scaled is the average performance during the last sampling
+	 * period scaled by the ratio of the maximum P-state to the P-state
+	 * requested last time (in percent).  That measures the system's
+	 * response to the previous P-state selection.
 	 */
-	core_busy = cpu->sample.core_pct_busy;
-	max_pstate = int_tofp(cpu->pstate.max_pstate_physical);
-	current_pstate = int_tofp(cpu->pstate.current_pstate);
-	core_busy = mul_fp(core_busy, div_fp(max_pstate, current_pstate));
+	max_pstate = cpu->pstate.max_pstate_physical;
+	current_pstate = cpu->pstate.current_pstate;
+	perf_scaled = mul_ext_fp(cpu->sample.core_avg_perf,
+			       div_fp(100 * max_pstate, current_pstate));
 
 	/*
 	 * Since our utilization update callback will not run unless we are
 	 * in C0, check if the actual elapsed time is significantly greater (3x)
 	 * than our sample interval.  If it is, then we were idle for a long
-	 * enough period of time to adjust our busyness.
+	 * enough period of time to adjust our performance metric.
 	 */
 	duration_ns = cpu->sample.time - cpu->last_sample_time;
 	if ((s64)duration_ns > pid_params.sample_rate_ns * 3) {
-		sample_ratio = div_fp(int_tofp(pid_params.sample_rate_ns),
-				      int_tofp(duration_ns));
-		core_busy = mul_fp(core_busy, sample_ratio);
+		sample_ratio = div_fp(pid_params.sample_rate_ns, duration_ns);
+		perf_scaled = mul_fp(perf_scaled, sample_ratio);
 	} else {
 		sample_ratio = div_fp(100 * cpu->sample.mperf, cpu->sample.tsc);
 		if (sample_ratio < int_tofp(1))
-			core_busy = 0;
+			perf_scaled = 0;
 	}
 
-	cpu->sample.busy_scaled = core_busy;
-	return cpu->pstate.current_pstate - pid_calc(&cpu->pid, core_busy);
+	cpu->sample.busy_scaled = perf_scaled;
+	return cpu->pstate.current_pstate - pid_calc(&cpu->pid, perf_scaled);
 }
 
 static inline void intel_pstate_update_pstate(struct cpudata *cpu, int pstate)
@@ -1179,7 +1320,7 @@ static inline void intel_pstate_adjust_busy_pstate(struct cpudata *cpu)
 	intel_pstate_update_pstate(cpu, target_pstate);
 
 	sample = &cpu->sample;
-	trace_pstate_sample(fp_toint(sample->core_pct_busy),
+	trace_pstate_sample(mul_ext_fp(100, sample->core_avg_perf),
 		fp_toint(sample->busy_scaled),
 		from,
 		cpu->pstate.current_pstate,
@@ -1199,7 +1340,7 @@ static void intel_pstate_update_util(struct update_util_data *data, u64 time,
 		bool sample_taken = intel_pstate_sample(cpu, time);
 
 		if (sample_taken) {
-			intel_pstate_calc_busy(cpu);
+			intel_pstate_calc_avg_perf(cpu);
 			if (!hwp_active)
 				intel_pstate_adjust_busy_pstate(cpu);
 		}
@@ -1261,23 +1402,16 @@ static int intel_pstate_init_cpu(unsigned int cpunum)
 
 	intel_pstate_busy_pid_reset(cpu);
 
-	cpu->update_util.func = intel_pstate_update_util;
-
-	pr_debug("intel_pstate: controlling: cpu %d\n", cpunum);
+	pr_debug("controlling: cpu %d\n", cpunum);
 
 	return 0;
 }
 
 static unsigned int intel_pstate_get(unsigned int cpu_num)
 {
-	struct sample *sample;
-	struct cpudata *cpu;
+	struct cpudata *cpu = all_cpu_data[cpu_num];
 
-	cpu = all_cpu_data[cpu_num];
-	if (!cpu)
-		return 0;
-	sample = &cpu->sample;
-	return get_avg_frequency(cpu);
+	return cpu ? get_avg_frequency(cpu) : 0;
 }
 
 static void intel_pstate_set_update_util_hook(unsigned int cpu_num)
@@ -1286,12 +1420,20 @@ static void intel_pstate_set_update_util_hook(unsigned int cpu_num)
 
 	/* Prevent intel_pstate_update_util() from using stale data. */
 	cpu->sample.time = 0;
-	cpufreq_set_update_util_data(cpu_num, &cpu->update_util);
+	cpufreq_add_update_util_hook(cpu_num, &cpu->update_util,
+				     intel_pstate_update_util);
+	cpu->update_util_set = true;
 }
 
 static void intel_pstate_clear_update_util_hook(unsigned int cpu)
 {
-	cpufreq_set_update_util_data(cpu, NULL);
+	struct cpudata *cpu_data = all_cpu_data[cpu];
+
+	if (!cpu_data->update_util_set)
+		return;
+
+	cpufreq_remove_update_util_hook(cpu);
+	cpu_data->update_util_set = false;
 	synchronize_sched();
 }
 
@@ -1311,20 +1453,31 @@ static void intel_pstate_set_performance_limits(struct perf_limits *limits)
 
 static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 {
+	struct cpudata *cpu;
+
 	if (!policy->cpuinfo.max_freq)
 		return -ENODEV;
 
 	intel_pstate_clear_update_util_hook(policy->cpu);
 
+	cpu = all_cpu_data[0];
+	if (cpu->pstate.max_pstate_physical > cpu->pstate.max_pstate) {
+		if (policy->max < policy->cpuinfo.max_freq &&
+		    policy->max > cpu->pstate.max_pstate * cpu->pstate.scaling) {
+			pr_debug("policy->max > max non turbo frequency\n");
+			policy->max = policy->cpuinfo.max_freq;
+		}
+	}
+
 	if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) {
 		limits = &performance_limits;
 		if (policy->max >= policy->cpuinfo.max_freq) {
-			pr_debug("intel_pstate: set performance\n");
+			pr_debug("set performance\n");
 			intel_pstate_set_performance_limits(limits);
 			goto out;
 		}
 	} else {
-		pr_debug("intel_pstate: set powersave\n");
+		pr_debug("set powersave\n");
 		limits = &powersave_limits;
 	}
 
@@ -1348,10 +1501,8 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 	/* Make sure min_perf_pct <= max_perf_pct */
 	limits->min_perf_pct = min(limits->max_perf_pct, limits->min_perf_pct);
 
-	limits->min_perf = div_fp(int_tofp(limits->min_perf_pct),
-				  int_tofp(100));
-	limits->max_perf = div_fp(int_tofp(limits->max_perf_pct),
-				  int_tofp(100));
+	limits->min_perf = div_fp(limits->min_perf_pct, 100);
+	limits->max_perf = div_fp(limits->max_perf_pct, 100);
 
  out:
 	intel_pstate_set_update_util_hook(policy->cpu);
@@ -1377,7 +1528,7 @@ static void intel_pstate_stop_cpu(struct cpufreq_policy *policy)
 	int cpu_num = policy->cpu;
 	struct cpudata *cpu = all_cpu_data[cpu_num];
 
-	pr_debug("intel_pstate: CPU %d exiting\n", cpu_num);
+	pr_debug("CPU %d exiting\n", cpu_num);
 
 	intel_pstate_clear_update_util_hook(cpu_num);
 
@@ -1410,12 +1561,20 @@ static int intel_pstate_cpu_init(struct cpufreq_policy *policy)
 	policy->cpuinfo.min_freq = cpu->pstate.min_pstate * cpu->pstate.scaling;
 	policy->cpuinfo.max_freq =
 		cpu->pstate.turbo_pstate * cpu->pstate.scaling;
+	intel_pstate_init_acpi_perf_limits(policy);
 	policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL;
 	cpumask_set_cpu(policy->cpu, policy->cpus);
 
 	return 0;
 }
 
+static int intel_pstate_cpu_exit(struct cpufreq_policy *policy)
+{
+	intel_pstate_exit_perf_limits(policy);
+
+	return 0;
+}
+
 static struct cpufreq_driver intel_pstate_driver = {
 	.flags		= CPUFREQ_CONST_LOOPS,
 	.verify		= intel_pstate_verify_policy,
@@ -1423,6 +1582,7 @@ static struct cpufreq_driver intel_pstate_driver = {
 	.resume		= intel_pstate_hwp_set_policy,
 	.get		= intel_pstate_get,
 	.init		= intel_pstate_cpu_init,
+	.exit		= intel_pstate_cpu_exit,
 	.stop_cpu	= intel_pstate_stop_cpu,
 	.name		= "intel_pstate",
 };
@@ -1466,8 +1626,7 @@ static void copy_cpu_funcs(struct pstate_funcs *funcs)
 
 }
 
-#if IS_ENABLED(CONFIG_ACPI)
-#include <acpi/processor.h>
+#ifdef CONFIG_ACPI
 
 static bool intel_pstate_no_acpi_pss(void)
 {
@@ -1623,7 +1782,7 @@ hwp_cpu_matched:
 	if (intel_pstate_platform_pwr_mgmt_exists())
 		return -ENODEV;
 
-	pr_info("Intel P-state driver initializing.\n");
+	pr_info("Intel P-state driver initializing\n");
 
 	all_cpu_data = vzalloc(sizeof(void *) * num_possible_cpus());
 	if (!all_cpu_data)
@@ -1640,7 +1799,7 @@ hwp_cpu_matched:
 	intel_pstate_sysfs_expose_params();
 
 	if (hwp_active)
-		pr_info("intel_pstate: HWP enabled\n");
+		pr_info("HWP enabled\n");
 
 	return rc;
 out:
@@ -1666,13 +1825,19 @@ static int __init intel_pstate_setup(char *str)
 	if (!strcmp(str, "disable"))
 		no_load = 1;
 	if (!strcmp(str, "no_hwp")) {
-		pr_info("intel_pstate: HWP disabled\n");
+		pr_info("HWP disabled\n");
 		no_hwp = 1;
 	}
 	if (!strcmp(str, "force"))
 		force_load = 1;
 	if (!strcmp(str, "hwp_only"))
 		hwp_only = 1;
+
+#ifdef CONFIG_ACPI
+	if (!strcmp(str, "support_acpi_ppc"))
+		acpi_ppc = true;
+#endif
+
 	return 0;
 }
 early_param("intel_pstate", intel_pstate_setup);

drivers/cpufreq/longhaul.c

@@ -21,6 +21,8 @@
  *  BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous*
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
@@ -40,8 +42,6 @@
 
 #include "longhaul.h"
 
-#define PFX "longhaul: "
-
 #define TYPE_LONGHAUL_V1	1
 #define TYPE_LONGHAUL_V2	2
 #define TYPE_POWERSAVER		3
@@ -347,14 +347,13 @@ retry_loop:
 	freqs.new = calc_speed(longhaul_get_cpu_mult());
 	/* Check if requested frequency is set. */
 	if (unlikely(freqs.new != speed)) {
-		printk(KERN_INFO PFX "Failed to set requested frequency!\n");
+		pr_info("Failed to set requested frequency!\n");
 		/* Revision ID = 1 but processor is expecting revision key
 		 * equal to 0. Jumpers at the bottom of processor will change
 		 * multiplier and FSB, but will not change bits in Longhaul
 		 * MSR nor enable voltage scaling. */
 		if (!revid_errata) {
-			printk(KERN_INFO PFX "Enabling \"Ignore Revision ID\" "
-						"option.\n");
+			pr_info("Enabling \"Ignore Revision ID\" option\n");
 			revid_errata = 1;
 			msleep(200);
 			goto retry_loop;
@@ -364,11 +363,10 @@ retry_loop:
 		 * but it doesn't change frequency. I tried poking various
 		 * bits in northbridge registers, but without success. */
 		if (longhaul_flags & USE_ACPI_C3) {
-			printk(KERN_INFO PFX "Disabling ACPI C3 support.\n");
+			pr_info("Disabling ACPI C3 support\n");
 			longhaul_flags &= ~USE_ACPI_C3;
 			if (revid_errata) {
-				printk(KERN_INFO PFX "Disabling \"Ignore "
-						"Revision ID\" option.\n");
+				pr_info("Disabling \"Ignore Revision ID\" option\n");
 				revid_errata = 0;
 			}
 			msleep(200);
@@ -379,7 +377,7 @@ retry_loop:
 		 * RevID = 1. RevID errata will make things right. Just
 		 * to be 100% sure. */
 		if (longhaul_version == TYPE_LONGHAUL_V2) {
-			printk(KERN_INFO PFX "Switching to Longhaul ver. 1\n");
+			pr_info("Switching to Longhaul ver. 1\n");
 			longhaul_version = TYPE_LONGHAUL_V1;
 			msleep(200);
 			goto retry_loop;
@@ -387,8 +385,7 @@ retry_loop:
 	}
 
 	if (!bm_timeout) {
-		printk(KERN_INFO PFX "Warning: Timeout while waiting for "
-				"idle PCI bus.\n");
+		pr_info("Warning: Timeout while waiting for idle PCI bus\n");
 		return -EBUSY;
 	}
 
@@ -433,12 +430,12 @@ static int longhaul_get_ranges(void)
 	/* Get current frequency */
 	mult = longhaul_get_cpu_mult();
 	if (mult == -1) {
-		printk(KERN_INFO PFX "Invalid (reserved) multiplier!\n");
+		pr_info("Invalid (reserved) multiplier!\n");
 		return -EINVAL;
 	}
 	fsb = guess_fsb(mult);
 	if (fsb == 0) {
-		printk(KERN_INFO PFX "Invalid (reserved) FSB!\n");
+		pr_info("Invalid (reserved) FSB!\n");
 		return -EINVAL;
 	}
 	/* Get max multiplier - as we always did.
@@ -468,11 +465,11 @@ static int longhaul_get_ranges(void)
 		 print_speed(highest_speed/1000));
 
 	if (lowest_speed == highest_speed) {
-		printk(KERN_INFO PFX "highestspeed == lowest, aborting.\n");
+		pr_info("highestspeed == lowest, aborting\n");
 		return -EINVAL;
 	}
 	if (lowest_speed > highest_speed) {
-		printk(KERN_INFO PFX "nonsense! lowest (%d > %d) !\n",
+		pr_info("nonsense! lowest (%d > %d) !\n",
 			lowest_speed, highest_speed);
 		return -EINVAL;
 	}
@@ -538,16 +535,16 @@ static void longhaul_setup_voltagescaling(void)
 
 	rdmsrl(MSR_VIA_LONGHAUL, longhaul.val);
 	if (!(longhaul.bits.RevisionID & 1)) {
-		printk(KERN_INFO PFX "Voltage scaling not supported by CPU.\n");
+		pr_info("Voltage scaling not supported by CPU\n");
 		return;
 	}
 
 	if (!longhaul.bits.VRMRev) {
-		printk(KERN_INFO PFX "VRM 8.5\n");
+		pr_info("VRM 8.5\n");
 		vrm_mV_table = &vrm85_mV[0];
 		mV_vrm_table = &mV_vrm85[0];
 	} else {
-		printk(KERN_INFO PFX "Mobile VRM\n");
+		pr_info("Mobile VRM\n");
 		if (cpu_model < CPU_NEHEMIAH)
 			return;
 		vrm_mV_table = &mobilevrm_mV[0];
@@ -558,27 +555,21 @@ static void longhaul_setup_voltagescaling(void)
 	maxvid = vrm_mV_table[longhaul.bits.MaximumVID];
 
 	if (minvid.mV == 0 || maxvid.mV == 0 || minvid.mV > maxvid.mV) {
-		printk(KERN_INFO PFX "Bogus values Min:%d.%03d Max:%d.%03d. "
-					"Voltage scaling disabled.\n",
-					minvid.mV/1000, minvid.mV%1000,
-					maxvid.mV/1000, maxvid.mV%1000);
+		pr_info("Bogus values Min:%d.%03d Max:%d.%03d - Voltage scaling disabled\n",
+			minvid.mV/1000, minvid.mV%1000,
+			maxvid.mV/1000, maxvid.mV%1000);
 		return;
 	}
 
 	if (minvid.mV == maxvid.mV) {
-		printk(KERN_INFO PFX "Claims to support voltage scaling but "
-				"min & max are both %d.%03d. "
-				"Voltage scaling disabled\n",
-				maxvid.mV/1000, maxvid.mV%1000);
+		pr_info("Claims to support voltage scaling but min & max are both %d.%03d - Voltage scaling disabled\n",
+			maxvid.mV/1000, maxvid.mV%1000);
 		return;
 	}
 
 	/* How many voltage steps*/
 	numvscales = maxvid.pos - minvid.pos + 1;
-	printk(KERN_INFO PFX
-		"Max VID=%d.%03d  "
-		"Min VID=%d.%03d, "
-		"%d possible voltage scales\n",
+	pr_info("Max VID=%d.%03d  Min VID=%d.%03d, %d possible voltage scales\n",
 		maxvid.mV/1000, maxvid.mV%1000,
 		minvid.mV/1000, minvid.mV%1000,
 		numvscales);
@@ -617,12 +608,12 @@ static void longhaul_setup_voltagescaling(void)
 			pos = minvid.pos;
 		freq_pos->driver_data |= mV_vrm_table[pos] << 8;
 		vid = vrm_mV_table[mV_vrm_table[pos]];
-		printk(KERN_INFO PFX "f: %d kHz, index: %d, vid: %d mV\n",
+		pr_info("f: %d kHz, index: %d, vid: %d mV\n",
 			speed, (int)(freq_pos - longhaul_table), vid.mV);
 	}
 
 	can_scale_voltage = 1;
-	printk(KERN_INFO PFX "Voltage scaling enabled.\n");
+	pr_info("Voltage scaling enabled\n");
 }
 
 
@@ -720,8 +711,7 @@ static int enable_arbiter_disable(void)
 			pci_write_config_byte(dev, reg, pci_cmd);
 			pci_read_config_byte(dev, reg, &pci_cmd);
 			if (!(pci_cmd & 1<<7)) {
-				printk(KERN_ERR PFX
-					"Can't enable access to port 0x22.\n");
+				pr_err("Can't enable access to port 0x22\n");
 				status = 0;
 			}
 		}
@@ -758,8 +748,7 @@ static int longhaul_setup_southbridge(void)
 		if (pci_cmd & 1 << 7) {
 			pci_read_config_dword(dev, 0x88, &acpi_regs_addr);
 			acpi_regs_addr &= 0xff00;
-			printk(KERN_INFO PFX "ACPI I/O at 0x%x\n",
-					acpi_regs_addr);
+			pr_info("ACPI I/O at 0x%x\n", acpi_regs_addr);
 		}
 
 		pci_dev_put(dev);
@@ -853,14 +842,14 @@ static int longhaul_cpu_init(struct cpufreq_policy *policy)
 			longhaul_version = TYPE_LONGHAUL_V1;
 	}
 
-	printk(KERN_INFO PFX "VIA %s CPU detected.  ", cpuname);
+	pr_info("VIA %s CPU detected.  ", cpuname);
 	switch (longhaul_version) {
 	case TYPE_LONGHAUL_V1:
 	case TYPE_LONGHAUL_V2:
-		printk(KERN_CONT "Longhaul v%d supported.\n", longhaul_version);
+		pr_cont("Longhaul v%d supported\n", longhaul_version);
 		break;
 	case TYPE_POWERSAVER:
-		printk(KERN_CONT "Powersaver supported.\n");
+		pr_cont("Powersaver supported\n");
 		break;
 	};
 
@@ -889,15 +878,14 @@ static int longhaul_cpu_init(struct cpufreq_policy *policy)
 	if (!(longhaul_flags & USE_ACPI_C3
 	     || longhaul_flags & USE_NORTHBRIDGE)
 	    && ((pr == NULL) || !(pr->flags.bm_control))) {
-		printk(KERN_ERR PFX
-			"No ACPI support. Unsupported northbridge.\n");
+		pr_err("No ACPI support: Unsupported northbridge\n");
 		return -ENODEV;
 	}
 
 	if (longhaul_flags & USE_NORTHBRIDGE)
-		printk(KERN_INFO PFX "Using northbridge support.\n");
+		pr_info("Using northbridge support\n");
 	if (longhaul_flags & USE_ACPI_C3)
-		printk(KERN_INFO PFX "Using ACPI support.\n");
+		pr_info("Using ACPI support\n");
 
 	ret = longhaul_get_ranges();
 	if (ret != 0)
@@ -934,20 +922,18 @@ static int __init longhaul_init(void)
 		return -ENODEV;
 
 	if (!enable) {
-		printk(KERN_ERR PFX "Option \"enable\" not set. Aborting.\n");
+		pr_err("Option \"enable\" not set - Aborting\n");
 		return -ENODEV;
 	}
 #ifdef CONFIG_SMP
 	if (num_online_cpus() > 1) {
-		printk(KERN_ERR PFX "More than 1 CPU detected, "
-				"longhaul disabled.\n");
+		pr_err("More than 1 CPU detected, longhaul disabled\n");
 		return -ENODEV;
 	}
 #endif
 #ifdef CONFIG_X86_IO_APIC
 	if (cpu_has_apic) {
-		printk(KERN_ERR PFX "APIC detected. Longhaul is currently "
-				"broken in this configuration.\n");
+		pr_err("APIC detected. Longhaul is currently broken in this configuration.\n");
 		return -ENODEV;
 	}
 #endif
@@ -955,7 +941,7 @@ static int __init longhaul_init(void)
 	case 6 ... 9:
 		return cpufreq_register_driver(&longhaul_driver);
 	case 10:
-		printk(KERN_ERR PFX "Use acpi-cpufreq driver for VIA C7\n");
+		pr_err("Use acpi-cpufreq driver for VIA C7\n");
 	default:
 		;
 	}

drivers/cpufreq/loongson2_cpufreq.c

@@ -10,6 +10,9 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/cpufreq.h>
 #include <linux/module.h>
 #include <linux/err.h>
@@ -76,7 +79,7 @@ static int loongson2_cpufreq_cpu_init(struct cpufreq_policy *policy)
 
 	cpuclk = clk_get(NULL, "cpu_clk");
 	if (IS_ERR(cpuclk)) {
-		printk(KERN_ERR "cpufreq: couldn't get CPU clk\n");
+		pr_err("couldn't get CPU clk\n");
 		return PTR_ERR(cpuclk);
 	}
 
@@ -163,7 +166,7 @@ static int __init cpufreq_init(void)
 	if (ret)
 		return ret;
 
-	pr_info("cpufreq: Loongson-2F CPU frequency driver.\n");
+	pr_info("Loongson-2F CPU frequency driver\n");
 
 	cpufreq_register_notifier(&loongson2_cpufreq_notifier_block,
 				  CPUFREQ_TRANSITION_NOTIFIER);

drivers/cpufreq/maple-cpufreq.c

@@ -13,6 +13,8 @@
 
 #undef DEBUG
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/errno.h>
@@ -174,7 +176,7 @@ static int __init maple_cpufreq_init(void)
 	/* Get first CPU node */
 	cpunode = of_cpu_device_node_get(0);
 	if (cpunode == NULL) {
-		printk(KERN_ERR "cpufreq: Can't find any CPU 0 node\n");
+		pr_err("Can't find any CPU 0 node\n");
 		goto bail_noprops;
 	}
 
@@ -182,8 +184,7 @@ static int __init maple_cpufreq_init(void)
 	/* we actually don't care on which CPU to access PVR */
 	pvr_hi = PVR_VER(mfspr(SPRN_PVR));
 	if (pvr_hi != 0x3c && pvr_hi != 0x44) {
-		printk(KERN_ERR "cpufreq: Unsupported CPU version (%x)\n",
-				pvr_hi);
+		pr_err("Unsupported CPU version (%x)\n", pvr_hi);
 		goto bail_noprops;
 	}
 
@@ -222,8 +223,8 @@ static int __init maple_cpufreq_init(void)
 	maple_pmode_cur = -1;
 	maple_scom_switch_freq(maple_scom_query_freq());
 
-	printk(KERN_INFO "Registering Maple CPU frequency driver\n");
-	printk(KERN_INFO "Low: %d Mhz, High: %d Mhz, Cur: %d MHz\n",
+	pr_info("Registering Maple CPU frequency driver\n");
+	pr_info("Low: %d Mhz, High: %d Mhz, Cur: %d MHz\n",
 		maple_cpu_freqs[1].frequency/1000,
 		maple_cpu_freqs[0].frequency/1000,
 		maple_cpu_freqs[maple_pmode_cur].frequency/1000);

drivers/cpufreq/mt8173-cpufreq.c

@@ -59,11 +59,8 @@ static LIST_HEAD(dvfs_info_list);
 static struct mtk_cpu_dvfs_info *mtk_cpu_dvfs_info_lookup(int cpu)
 {
 	struct mtk_cpu_dvfs_info *info;
-	struct list_head *list;
-
-	list_for_each(list, &dvfs_info_list) {
-		info = list_entry(list, struct mtk_cpu_dvfs_info, list_head);
 
+	list_for_each_entry(info, &dvfs_info_list, list_head) {
 		if (cpumask_test_cpu(cpu, &info->cpus))
 			return info;
 	}
@@ -524,8 +521,7 @@ static struct cpufreq_driver mt8173_cpufreq_driver = {
 
 static int mt8173_cpufreq_probe(struct platform_device *pdev)
 {
-	struct mtk_cpu_dvfs_info *info;
-	struct list_head *list, *tmp;
+	struct mtk_cpu_dvfs_info *info, *tmp;
 	int cpu, ret;
 
 	for_each_possible_cpu(cpu) {
@@ -559,11 +555,9 @@ static int mt8173_cpufreq_probe(struct platform_device *pdev)
 	return 0;
 
 release_dvfs_info_list:
-	list_for_each_safe(list, tmp, &dvfs_info_list) {
-		info = list_entry(list, struct mtk_cpu_dvfs_info, list_head);
-
+	list_for_each_entry_safe(info, tmp, &dvfs_info_list, list_head) {
 		mtk_cpu_dvfs_info_release(info);
-		list_del(list);
+		list_del(&info->list_head);
 	}
 
 	return ret;

drivers/cpufreq/mvebu-cpufreq.c

@@ -0,0 +1,107 @@
+/*
+ * CPUFreq support for Armada 370/XP platforms.
+ *
+ * Copyright (C) 2012-2016 Marvell
+ *
+ * Yehuda Yitschak <yehuday@marvell.com>
+ * Gregory Clement <gregory.clement@free-electrons.com>
+ * Thomas Petazzoni <thomas.petazzoni@free-electrons.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.
+ */
+
+#define pr_fmt(fmt) "mvebu-pmsu: " fmt
+
+#include <linux/clk.h>
+#include <linux/cpu.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/of_address.h>
+#include <linux/platform_device.h>
+#include <linux/pm_opp.h>
+#include <linux/resource.h>
+
+static int __init armada_xp_pmsu_cpufreq_init(void)
+{
+	struct device_node *np;
+	struct resource res;
+	int ret, cpu;
+
+	if (!of_machine_is_compatible("marvell,armadaxp"))
+		return 0;
+
+	/*
+	 * In order to have proper cpufreq handling, we need to ensure
+	 * that the Device Tree description of the CPU clock includes
+	 * the definition of the PMU DFS registers. If not, we do not
+	 * register the clock notifier and the cpufreq driver. This
+	 * piece of code is only for compatibility with old Device
+	 * Trees.
+	 */
+	np = of_find_compatible_node(NULL, NULL, "marvell,armada-xp-cpu-clock");
+	if (!np)
+		return 0;
+
+	ret = of_address_to_resource(np, 1, &res);
+	if (ret) {
+		pr_warn(FW_WARN "not enabling cpufreq, deprecated armada-xp-cpu-clock binding\n");
+		of_node_put(np);
+		return 0;
+	}
+
+	of_node_put(np);
+
+	/*
+	 * For each CPU, this loop registers the operating points
+	 * supported (which are the nominal CPU frequency and half of
+	 * it), and registers the clock notifier that will take care
+	 * of doing the PMSU part of a frequency transition.
+	 */
+	for_each_possible_cpu(cpu) {
+		struct device *cpu_dev;
+		struct clk *clk;
+		int ret;
+
+		cpu_dev = get_cpu_device(cpu);
+		if (!cpu_dev) {
+			pr_err("Cannot get CPU %d\n", cpu);
+			continue;
+		}
+
+		clk = clk_get(cpu_dev, 0);
+		if (IS_ERR(clk)) {
+			pr_err("Cannot get clock for CPU %d\n", cpu);
+			return PTR_ERR(clk);
+		}
+
+		/*
+		 * In case of a failure of dev_pm_opp_add(), we don't
+		 * bother with cleaning up the registered OPP (there's
+		 * no function to do so), and simply cancel the
+		 * registration of the cpufreq device.
+		 */
+		ret = dev_pm_opp_add(cpu_dev, clk_get_rate(clk), 0);
+		if (ret) {
+			clk_put(clk);
+			return ret;
+		}
+
+		ret = dev_pm_opp_add(cpu_dev, clk_get_rate(clk) / 2, 0);
+		if (ret) {
+			clk_put(clk);
+			return ret;
+		}
+
+		ret = dev_pm_opp_set_sharing_cpus(cpu_dev,
+						  cpumask_of(cpu_dev->id));
+		if (ret)
+			dev_err(cpu_dev, "%s: failed to mark OPPs as shared: %d\n",
+				__func__, ret);
+	}
+
+	platform_device_register_simple("cpufreq-dt", -1, NULL, 0);
+	return 0;
+}
+device_initcall(armada_xp_pmsu_cpufreq_init);

drivers/cpufreq/omap-cpufreq.c

@@ -13,6 +13,9 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
@@ -163,13 +166,13 @@ static int omap_cpufreq_probe(struct platform_device *pdev)
 {
 	mpu_dev = get_cpu_device(0);
 	if (!mpu_dev) {
-		pr_warning("%s: unable to get the mpu device\n", __func__);
+		pr_warn("%s: unable to get the MPU device\n", __func__);
 		return -EINVAL;
 	}
 
 	mpu_reg = regulator_get(mpu_dev, "vcc");
 	if (IS_ERR(mpu_reg)) {
-		pr_warning("%s: unable to get MPU regulator\n", __func__);
+		pr_warn("%s: unable to get MPU regulator\n", __func__);
 		mpu_reg = NULL;
 	} else {
 		/* 

drivers/cpufreq/p4-clockmod.c

@@ -20,6 +20,8 @@
  *
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -35,8 +37,6 @@
 
 #include "speedstep-lib.h"
 
-#define PFX	"p4-clockmod: "
-
 /*
  * Duty Cycle (3bits), note DC_DISABLE is not specified in
  * intel docs i just use it to mean disable
@@ -124,11 +124,7 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c)
 {
 	if (c->x86 == 0x06) {
 		if (cpu_has(c, X86_FEATURE_EST))
-			printk_once(KERN_WARNING PFX "Warning: EST-capable "
-			       "CPU detected. The acpi-cpufreq module offers "
-			       "voltage scaling in addition to frequency "
-			       "scaling. You should use that instead of "
-			       "p4-clockmod, if possible.\n");
+			pr_warn_once("Warning: EST-capable CPU detected. The acpi-cpufreq module offers voltage scaling in addition to frequency scaling. You should use that instead of p4-clockmod, if possible.\n");
 		switch (c->x86_model) {
 		case 0x0E: /* Core */
 		case 0x0F: /* Core Duo */
@@ -152,11 +148,7 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c)
 	p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS;
 
 	if (speedstep_detect_processor() == SPEEDSTEP_CPU_P4M) {
-		printk(KERN_WARNING PFX "Warning: Pentium 4-M detected. "
-		       "The speedstep-ich or acpi cpufreq modules offer "
-		       "voltage scaling in addition of frequency scaling. "
-		       "You should use either one instead of p4-clockmod, "
-		       "if possible.\n");
+		pr_warn("Warning: Pentium 4-M detected. The speedstep-ich or acpi cpufreq modules offer voltage scaling in addition of frequency scaling. You should use either one instead of p4-clockmod, if possible.\n");
 		return speedstep_get_frequency(SPEEDSTEP_CPU_P4M);
 	}
 
@@ -265,8 +257,7 @@ static int __init cpufreq_p4_init(void)
 
 	ret = cpufreq_register_driver(&p4clockmod_driver);
 	if (!ret)
-		printk(KERN_INFO PFX "P4/Xeon(TM) CPU On-Demand Clock "
-				"Modulation available\n");
+		pr_info("P4/Xeon(TM) CPU On-Demand Clock Modulation available\n");
 
 	return ret;
 }

drivers/cpufreq/pmac32-cpufreq.c

@@ -13,6 +13,8 @@
  *
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/errno.h>
@@ -481,13 +483,13 @@ static int pmac_cpufreq_init_MacRISC3(struct device_node *cpunode)
 		freqs = of_get_property(cpunode, "bus-frequencies", &lenp);
 		lenp /= sizeof(u32);
 		if (freqs == NULL || lenp != 2) {
-			printk(KERN_ERR "cpufreq: bus-frequencies incorrect or missing\n");
+			pr_err("bus-frequencies incorrect or missing\n");
 			return 1;
 		}
 		ratio = of_get_property(cpunode, "processor-to-bus-ratio*2",
 						NULL);
 		if (ratio == NULL) {
-			printk(KERN_ERR "cpufreq: processor-to-bus-ratio*2 missing\n");
+			pr_err("processor-to-bus-ratio*2 missing\n");
 			return 1;
 		}
 
@@ -550,7 +552,7 @@ static int pmac_cpufreq_init_7447A(struct device_node *cpunode)
 	if (volt_gpio_np)
 		voltage_gpio = read_gpio(volt_gpio_np);
 	if (!voltage_gpio){
-		printk(KERN_ERR "cpufreq: missing cpu-vcore-select gpio\n");
+		pr_err("missing cpu-vcore-select gpio\n");
 		return 1;
 	}
 
@@ -675,9 +677,9 @@ out:
 	pmac_cpu_freqs[CPUFREQ_HIGH].frequency = hi_freq;
 	ppc_proc_freq = cur_freq * 1000ul;
 
-	printk(KERN_INFO "Registering PowerMac CPU frequency driver\n");
-	printk(KERN_INFO "Low: %d Mhz, High: %d Mhz, Boot: %d Mhz\n",
-	       low_freq/1000, hi_freq/1000, cur_freq/1000);
+	pr_info("Registering PowerMac CPU frequency driver\n");
+	pr_info("Low: %d Mhz, High: %d Mhz, Boot: %d Mhz\n",
+		low_freq/1000, hi_freq/1000, cur_freq/1000);
 
 	return cpufreq_register_driver(&pmac_cpufreq_driver);
 }

drivers/cpufreq/pmac64-cpufreq.c

@@ -12,6 +12,8 @@
 
 #undef DEBUG
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/errno.h>
@@ -138,7 +140,7 @@ static void g5_vdnap_switch_volt(int speed_mode)
 		usleep_range(1000, 1000);
 	}
 	if (done == 0)
-		printk(KERN_WARNING "cpufreq: Timeout in clock slewing !\n");
+		pr_warn("Timeout in clock slewing !\n");
 }
 
 
@@ -266,7 +268,7 @@ static int g5_pfunc_switch_freq(int speed_mode)
 		rc = pmf_call_one(pfunc_cpu_setfreq_low, NULL);
 
 	if (rc)
-		printk(KERN_WARNING "cpufreq: pfunc switch error %d\n", rc);
+		pr_warn("pfunc switch error %d\n", rc);
 
 	/* It's an irq GPIO so we should be able to just block here,
 	 * I'll do that later after I've properly tested the IRQ code for
@@ -282,7 +284,7 @@ static int g5_pfunc_switch_freq(int speed_mode)
 		usleep_range(500, 500);
 	}
 	if (done == 0)
-		printk(KERN_WARNING "cpufreq: Timeout in clock slewing !\n");
+		pr_warn("Timeout in clock slewing !\n");
 
 	/* If frequency is going down, last ramp the voltage */
 	if (speed_mode > g5_pmode_cur)
@@ -368,7 +370,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpunode)
 	}
 	pvr_hi = (*valp) >> 16;
 	if (pvr_hi != 0x3c && pvr_hi != 0x44) {
-		printk(KERN_ERR "cpufreq: Unsupported CPU version\n");
+		pr_err("Unsupported CPU version\n");
 		goto bail_noprops;
 	}
 
@@ -403,8 +405,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpunode)
 
 		root = of_find_node_by_path("/");
 		if (root == NULL) {
-			printk(KERN_ERR "cpufreq: Can't find root of "
-			       "device tree\n");
+			pr_err("Can't find root of device tree\n");
 			goto bail_noprops;
 		}
 		pfunc_set_vdnap0 = pmf_find_function(root, "set-vdnap0");
@@ -412,8 +413,7 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpunode)
 			pmf_find_function(root, "slewing-done");
 		if (pfunc_set_vdnap0 == NULL ||
 		    pfunc_vdnap0_complete == NULL) {
-			printk(KERN_ERR "cpufreq: Can't find required "
-			       "platform function\n");
+			pr_err("Can't find required platform function\n");
 			goto bail_noprops;
 		}
 
@@ -453,10 +453,10 @@ static int __init g5_neo2_cpufreq_init(struct device_node *cpunode)
 	g5_pmode_cur = -1;
 	g5_switch_freq(g5_query_freq());
 
-	printk(KERN_INFO "Registering G5 CPU frequency driver\n");
-	printk(KERN_INFO "Frequency method: %s, Voltage method: %s\n",
-	       freq_method, volt_method);
-	printk(KERN_INFO "Low: %d Mhz, High: %d Mhz, Cur: %d MHz\n",
+	pr_info("Registering G5 CPU frequency driver\n");
+	pr_info("Frequency method: %s, Voltage method: %s\n",
+		freq_method, volt_method);
+	pr_info("Low: %d Mhz, High: %d Mhz, Cur: %d MHz\n",
 		g5_cpu_freqs[1].frequency/1000,
 		g5_cpu_freqs[0].frequency/1000,
 		g5_cpu_freqs[g5_pmode_cur].frequency/1000);
@@ -493,7 +493,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode)
 	if (cpuid != NULL)
 		eeprom = of_get_property(cpuid, "cpuid", NULL);
 	if (eeprom == NULL) {
-		printk(KERN_ERR "cpufreq: Can't find cpuid EEPROM !\n");
+		pr_err("Can't find cpuid EEPROM !\n");
 		rc = -ENODEV;
 		goto bail;
 	}
@@ -511,7 +511,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode)
 		break;
 	}
 	if (hwclock == NULL) {
-		printk(KERN_ERR "cpufreq: Can't find i2c clock chip !\n");
+		pr_err("Can't find i2c clock chip !\n");
 		rc = -ENODEV;
 		goto bail;
 	}
@@ -539,7 +539,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode)
 	/* Check we have minimum requirements */
 	if (pfunc_cpu_getfreq == NULL || pfunc_cpu_setfreq_high == NULL ||
 	    pfunc_cpu_setfreq_low == NULL || pfunc_slewing_done == NULL) {
-		printk(KERN_ERR "cpufreq: Can't find platform functions !\n");
+		pr_err("Can't find platform functions !\n");
 		rc = -ENODEV;
 		goto bail;
 	}
@@ -567,7 +567,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode)
 	/* Get max frequency from device-tree */
 	valp = of_get_property(cpunode, "clock-frequency", NULL);
 	if (!valp) {
-		printk(KERN_ERR "cpufreq: Can't find CPU frequency !\n");
+		pr_err("Can't find CPU frequency !\n");
 		rc = -ENODEV;
 		goto bail;
 	}
@@ -583,8 +583,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode)
 
 	/* Check for machines with no useful settings */
 	if (il == ih) {
-		printk(KERN_WARNING "cpufreq: No low frequency mode available"
-		       " on this model !\n");
+		pr_warn("No low frequency mode available on this model !\n");
 		rc = -ENODEV;
 		goto bail;
 	}
@@ -595,7 +594,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode)
 
 	/* Sanity check */
 	if (min_freq >= max_freq || min_freq < 1000) {
-		printk(KERN_ERR "cpufreq: Can't calculate low frequency !\n");
+		pr_err("Can't calculate low frequency !\n");
 		rc = -ENXIO;
 		goto bail;
 	}
@@ -619,10 +618,10 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode)
 	g5_pmode_cur = -1;
 	g5_switch_freq(g5_query_freq());
 
-	printk(KERN_INFO "Registering G5 CPU frequency driver\n");
-	printk(KERN_INFO "Frequency method: i2c/pfunc, "
-	       "Voltage method: %s\n", has_volt ? "i2c/pfunc" : "none");
-	printk(KERN_INFO "Low: %d Mhz, High: %d Mhz, Cur: %d MHz\n",
+	pr_info("Registering G5 CPU frequency driver\n");
+	pr_info("Frequency method: i2c/pfunc, Voltage method: %s\n",
+		has_volt ? "i2c/pfunc" : "none");
+	pr_info("Low: %d Mhz, High: %d Mhz, Cur: %d MHz\n",
 		g5_cpu_freqs[1].frequency/1000,
 		g5_cpu_freqs[0].frequency/1000,
 		g5_cpu_freqs[g5_pmode_cur].frequency/1000);
@@ -654,7 +653,7 @@ static int __init g5_cpufreq_init(void)
 	/* Get first CPU node */
 	cpunode = of_cpu_device_node_get(0);
 	if (cpunode == NULL) {
-		pr_err("cpufreq: Can't find any CPU node\n");
+		pr_err("Can't find any CPU node\n");
 		return -ENODEV;
 	}
 

drivers/cpufreq/powernow-k6.c

@@ -8,6 +8,8 @@
  *  BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous*
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -22,7 +24,6 @@
 #define POWERNOW_IOPORT 0xfff0          /* it doesn't matter where, as long
 					   as it is unused */
 
-#define PFX "powernow-k6: "
 static unsigned int                     busfreq;   /* FSB, in 10 kHz */
 static unsigned int                     max_multiplier;
 
@@ -141,7 +142,7 @@ static int powernow_k6_target(struct cpufreq_policy *policy,
 {
 
 	if (clock_ratio[best_i].driver_data > max_multiplier) {
-		printk(KERN_ERR PFX "invalid target frequency\n");
+		pr_err("invalid target frequency\n");
 		return -EINVAL;
 	}
 
@@ -175,13 +176,14 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy)
 				max_multiplier = param_max_multiplier;
 				goto have_max_multiplier;
 			}
-		printk(KERN_ERR "powernow-k6: invalid max_multiplier parameter, valid parameters 20, 30, 35, 40, 45, 50, 55, 60\n");
+		pr_err("invalid max_multiplier parameter, valid parameters 20, 30, 35, 40, 45, 50, 55, 60\n");
 		return -EINVAL;
 	}
 
 	if (!max_multiplier) {
-		printk(KERN_WARNING "powernow-k6: unknown frequency %u, cannot determine current multiplier\n", khz);
-		printk(KERN_WARNING "powernow-k6: use module parameters max_multiplier and bus_frequency\n");
+		pr_warn("unknown frequency %u, cannot determine current multiplier\n",
+			khz);
+		pr_warn("use module parameters max_multiplier and bus_frequency\n");
 		return -EOPNOTSUPP;
 	}
 
@@ -193,7 +195,7 @@ have_max_multiplier:
 			busfreq = param_busfreq / 10;
 			goto have_busfreq;
 		}
-		printk(KERN_ERR "powernow-k6: invalid bus_frequency parameter, allowed range 50000 - 150000 kHz\n");
+		pr_err("invalid bus_frequency parameter, allowed range 50000 - 150000 kHz\n");
 		return -EINVAL;
 	}
 
@@ -275,7 +277,7 @@ static int __init powernow_k6_init(void)
 		return -ENODEV;
 
 	if (!request_region(POWERNOW_IOPORT, 16, "PowerNow!")) {
-		printk(KERN_INFO PFX "PowerNow IOPORT region already used.\n");
+		pr_info("PowerNow IOPORT region already used\n");
 		return -EIO;
 	}
 

drivers/cpufreq/powernow-k7.c

@@ -13,6 +13,8 @@
  *  - We disable half multipliers if ACPI is used on A0 stepping CPUs.
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
@@ -35,9 +37,6 @@
 
 #include "powernow-k7.h"
 
-#define PFX "powernow: "
-
-
 struct psb_s {
 	u8 signature[10];
 	u8 tableversion;
@@ -127,14 +126,13 @@ static int check_powernow(void)
 	maxei = cpuid_eax(0x80000000);
 	if (maxei < 0x80000007) {	/* Any powernow info ? */
 #ifdef MODULE
-		printk(KERN_INFO PFX "No powernow capabilities detected\n");
+		pr_info("No powernow capabilities detected\n");
 #endif
 		return 0;
 	}
 
 	if ((c->x86_model == 6) && (c->x86_mask == 0)) {
-		printk(KERN_INFO PFX "K7 660[A0] core detected, "
-				"enabling errata workarounds\n");
+		pr_info("K7 660[A0] core detected, enabling errata workarounds\n");
 		have_a0 = 1;
 	}
 
@@ -144,22 +142,22 @@ static int check_powernow(void)
 	if (!(edx & (1 << 1 | 1 << 2)))
 		return 0;
 
-	printk(KERN_INFO PFX "PowerNOW! Technology present. Can scale: ");
+	pr_info("PowerNOW! Technology present. Can scale: ");
 
 	if (edx & 1 << 1) {
-		printk("frequency");
+		pr_cont("frequency");
 		can_scale_bus = 1;
 	}
 
 	if ((edx & (1 << 1 | 1 << 2)) == 0x6)
-		printk(" and ");
+		pr_cont(" and ");
 
 	if (edx & 1 << 2) {
-		printk("voltage");
+		pr_cont("voltage");
 		can_scale_vid = 1;
 	}
 
-	printk(".\n");
+	pr_cont("\n");
 	return 1;
 }
 
@@ -427,16 +425,14 @@ err1:
 err05:
 	kfree(acpi_processor_perf);
 err0:
-	printk(KERN_WARNING PFX "ACPI perflib can not be used on "
-			"this platform\n");
+	pr_warn("ACPI perflib can not be used on this platform\n");
 	acpi_processor_perf = NULL;
 	return retval;
 }
 #else
 static int powernow_acpi_init(void)
 {
-	printk(KERN_INFO PFX "no support for ACPI processor found."
-	       "  Please recompile your kernel with ACPI processor\n");
+	pr_info("no support for ACPI processor found - please recompile your kernel with ACPI processor\n");
 	return -EINVAL;
 }
 #endif
@@ -468,8 +464,7 @@ static int powernow_decode_bios(int maxfid, int startvid)
 			psb = (struct psb_s *) p;
 			pr_debug("Table version: 0x%x\n", psb->tableversion);
 			if (psb->tableversion != 0x12) {
-				printk(KERN_INFO PFX "Sorry, only v1.2 tables"
-						" supported right now\n");
+				pr_info("Sorry, only v1.2 tables supported right now\n");
 				return -ENODEV;
 			}
 
@@ -481,10 +476,8 @@ static int powernow_decode_bios(int maxfid, int startvid)
 
 			latency = psb->settlingtime;
 			if (latency < 100) {
-				printk(KERN_INFO PFX "BIOS set settling time "
-						"to %d microseconds. "
-						"Should be at least 100. "
-						"Correcting.\n", latency);
+				pr_info("BIOS set settling time to %d microseconds. Should be at least 100. Correcting.\n",
+					latency);
 				latency = 100;
 			}
 			pr_debug("Settling Time: %d microseconds.\n",
@@ -516,10 +509,9 @@ static int powernow_decode_bios(int maxfid, int startvid)
 						p += 2;
 				}
 			}
-			printk(KERN_INFO PFX "No PST tables match this cpuid "
-					"(0x%x)\n", etuple);
-			printk(KERN_INFO PFX "This is indicative of a broken "
-					"BIOS.\n");
+			pr_info("No PST tables match this cpuid (0x%x)\n",
+				etuple);
+			pr_info("This is indicative of a broken BIOS\n");
 
 			return -EINVAL;
 		}
@@ -552,7 +544,7 @@ static int fixup_sgtc(void)
 	sgtc = 100 * m * latency;
 	sgtc = sgtc / 3;
 	if (sgtc > 0xfffff) {
-		printk(KERN_WARNING PFX "SGTC too large %d\n", sgtc);
+		pr_warn("SGTC too large %d\n", sgtc);
 		sgtc = 0xfffff;
 	}
 	return sgtc;
@@ -574,14 +566,10 @@ static unsigned int powernow_get(unsigned int cpu)
 
 static int acer_cpufreq_pst(const struct dmi_system_id *d)
 {
-	printk(KERN_WARNING PFX
-		"%s laptop with broken PST tables in BIOS detected.\n",
+	pr_warn("%s laptop with broken PST tables in BIOS detected\n",
 		d->ident);
-	printk(KERN_WARNING PFX
-		"You need to downgrade to 3A21 (09/09/2002), or try a newer "
-		"BIOS than 3A71 (01/20/2003)\n");
-	printk(KERN_WARNING PFX
-		"cpufreq scaling has been disabled as a result of this.\n");
+	pr_warn("You need to downgrade to 3A21 (09/09/2002), or try a newer BIOS than 3A71 (01/20/2003)\n");
+	pr_warn("cpufreq scaling has been disabled as a result of this\n");
 	return 0;
 }
 
@@ -616,40 +604,38 @@ static int powernow_cpu_init(struct cpufreq_policy *policy)
 
 	fsb = (10 * cpu_khz) / fid_codes[fidvidstatus.bits.CFID];
 	if (!fsb) {
-		printk(KERN_WARNING PFX "can not determine bus frequency\n");
+		pr_warn("can not determine bus frequency\n");
 		return -EINVAL;
 	}
 	pr_debug("FSB: %3dMHz\n", fsb/1000);
 
 	if (dmi_check_system(powernow_dmi_table) || acpi_force) {
-		printk(KERN_INFO PFX "PSB/PST known to be broken.  "
-				"Trying ACPI instead\n");
+		pr_info("PSB/PST known to be broken - trying ACPI instead\n");
 		result = powernow_acpi_init();
 	} else {
 		result = powernow_decode_bios(fidvidstatus.bits.MFID,
 				fidvidstatus.bits.SVID);
 		if (result) {
-			printk(KERN_INFO PFX "Trying ACPI perflib\n");
+			pr_info("Trying ACPI perflib\n");
 			maximum_speed = 0;
 			minimum_speed = -1;
 			latency = 0;
 			result = powernow_acpi_init();
 			if (result) {
-				printk(KERN_INFO PFX
-					"ACPI and legacy methods failed\n");
+				pr_info("ACPI and legacy methods failed\n");
 			}
 		} else {
 			/* SGTC use the bus clock as timer */
 			latency = fixup_sgtc();
-			printk(KERN_INFO PFX "SGTC: %d\n", latency);
+			pr_info("SGTC: %d\n", latency);
 		}
 	}
 
 	if (result)
 		return result;
 
-	printk(KERN_INFO PFX "Minimum speed %d MHz. Maximum speed %d MHz.\n",
-				minimum_speed/1000, maximum_speed/1000);
+	pr_info("Minimum speed %d MHz - Maximum speed %d MHz\n",
+		minimum_speed/1000, maximum_speed/1000);
 
 	policy->cpuinfo.transition_latency =
 		cpufreq_scale(2000000UL, fsb, latency);

drivers/cpufreq/powernv-cpufreq.c

@@ -36,12 +36,56 @@
 #include <asm/reg.h>
 #include <asm/smp.h> /* Required for cpu_sibling_mask() in UP configs */
 #include <asm/opal.h>
+#include <linux/timer.h>
 
 #define POWERNV_MAX_PSTATES	256
 #define PMSR_PSAFE_ENABLE	(1UL << 30)
 #define PMSR_SPR_EM_DISABLE	(1UL << 31)
 #define PMSR_MAX(x)		((x >> 32) & 0xFF)
 
+#define MAX_RAMP_DOWN_TIME				5120
+/*
+ * On an idle system we want the global pstate to ramp-down from max value to
+ * min over a span of ~5 secs. Also we want it to initially ramp-down slowly and
+ * then ramp-down rapidly later on.
+ *
+ * This gives a percentage rampdown for time elapsed in milliseconds.
+ * ramp_down_percentage = ((ms * ms) >> 18)
+ *			~= 3.8 * (sec * sec)
+ *
+ * At 0 ms	ramp_down_percent = 0
+ * At 5120 ms	ramp_down_percent = 100
+ */
+#define ramp_down_percent(time)		((time * time) >> 18)
+
+/* Interval after which the timer is queued to bring down global pstate */
+#define GPSTATE_TIMER_INTERVAL				2000
+
+/**
+ * struct global_pstate_info -	Per policy data structure to maintain history of
+ *				global pstates
+ * @highest_lpstate:		The local pstate from which we are ramping down
+ * @elapsed_time:		Time in ms spent in ramping down from
+ *				highest_lpstate
+ * @last_sampled_time:		Time from boot in ms when global pstates were
+ *				last set
+ * @last_lpstate,last_gpstate:	Last set values for local and global pstates
+ * @timer:			Is used for ramping down if cpu goes idle for
+ *				a long time with global pstate held high
+ * @gpstate_lock:		A spinlock to maintain synchronization between
+ *				routines called by the timer handler and
+ *				governer's target_index calls
+ */
+struct global_pstate_info {
+	int highest_lpstate;
+	unsigned int elapsed_time;
+	unsigned int last_sampled_time;
+	int last_lpstate;
+	int last_gpstate;
+	spinlock_t gpstate_lock;
+	struct timer_list timer;
+};
+
 static struct cpufreq_frequency_table powernv_freqs[POWERNV_MAX_PSTATES+1];
 static bool rebooting, throttled, occ_reset;
 
@@ -94,6 +138,17 @@ static struct powernv_pstate_info {
 	int nr_pstates;
 } powernv_pstate_info;
 
+static inline void reset_gpstates(struct cpufreq_policy *policy)
+{
+	struct global_pstate_info *gpstates = policy->driver_data;
+
+	gpstates->highest_lpstate = 0;
+	gpstates->elapsed_time = 0;
+	gpstates->last_sampled_time = 0;
+	gpstates->last_lpstate = 0;
+	gpstates->last_gpstate = 0;
+}
+
 /*
  * Initialize the freq table based on data obtained
  * from the firmware passed via device-tree
@@ -285,6 +340,7 @@ static inline void set_pmspr(unsigned long sprn, unsigned long val)
 struct powernv_smp_call_data {
 	unsigned int freq;
 	int pstate_id;
+	int gpstate_id;
 };
 
 /*
@@ -343,19 +399,21 @@ static unsigned int powernv_cpufreq_get(unsigned int cpu)
  * (struct powernv_smp_call_data *) and the pstate_id which needs to be set
  * on this CPU should be present in freq_data->pstate_id.
  */
-static void set_pstate(void *freq_data)
+static void set_pstate(void *data)
 {
 	unsigned long val;
-	unsigned long pstate_ul =
-		((struct powernv_smp_call_data *) freq_data)->pstate_id;
+	struct powernv_smp_call_data *freq_data = data;
+	unsigned long pstate_ul = freq_data->pstate_id;
+	unsigned long gpstate_ul = freq_data->gpstate_id;
 
 	val = get_pmspr(SPRN_PMCR);
 	val = val & 0x0000FFFFFFFFFFFFULL;
 
 	pstate_ul = pstate_ul & 0xFF;
+	gpstate_ul = gpstate_ul & 0xFF;
 
 	/* Set both global(bits 56..63) and local(bits 48..55) PStates */
-	val = val | (pstate_ul << 56) | (pstate_ul << 48);
+	val = val | (gpstate_ul << 56) | (pstate_ul << 48);
 
 	pr_debug("Setting cpu %d pmcr to %016lX\n",
 			raw_smp_processor_id(), val);
@@ -424,6 +482,111 @@ next:
 	}
 }
 
+/**
+ * calc_global_pstate - Calculate global pstate
+ * @elapsed_time:	Elapsed time in milliseconds
+ * @local_pstate:	New local pstate
+ * @highest_lpstate:	pstate from which its ramping down
+ *
+ * Finds the appropriate global pstate based on the pstate from which its
+ * ramping down and the time elapsed in ramping down. It follows a quadratic
+ * equation which ensures that it reaches ramping down to pmin in 5sec.
+ */
+static inline int calc_global_pstate(unsigned int elapsed_time,
+				     int highest_lpstate, int local_pstate)
+{
+	int pstate_diff;
+
+	/*
+	 * Using ramp_down_percent we get the percentage of rampdown
+	 * that we are expecting to be dropping. Difference between
+	 * highest_lpstate and powernv_pstate_info.min will give a absolute
+	 * number of how many pstates we will drop eventually by the end of
+	 * 5 seconds, then just scale it get the number pstates to be dropped.
+	 */
+	pstate_diff =  ((int)ramp_down_percent(elapsed_time) *
+			(highest_lpstate - powernv_pstate_info.min)) / 100;
+
+	/* Ensure that global pstate is >= to local pstate */
+	if (highest_lpstate - pstate_diff < local_pstate)
+		return local_pstate;
+	else
+		return highest_lpstate - pstate_diff;
+}
+
+static inline void  queue_gpstate_timer(struct global_pstate_info *gpstates)
+{
+	unsigned int timer_interval;
+
+	/*
+	 * Setting up timer to fire after GPSTATE_TIMER_INTERVAL ms, But
+	 * if it exceeds MAX_RAMP_DOWN_TIME ms for ramp down time.
+	 * Set timer such that it fires exactly at MAX_RAMP_DOWN_TIME
+	 * seconds of ramp down time.
+	 */
+	if ((gpstates->elapsed_time + GPSTATE_TIMER_INTERVAL)
+	     > MAX_RAMP_DOWN_TIME)
+		timer_interval = MAX_RAMP_DOWN_TIME - gpstates->elapsed_time;
+	else
+		timer_interval = GPSTATE_TIMER_INTERVAL;
+
+	mod_timer_pinned(&gpstates->timer, jiffies +
+			msecs_to_jiffies(timer_interval));
+}
+
+/**
+ * gpstate_timer_handler
+ *
+ * @data: pointer to cpufreq_policy on which timer was queued
+ *
+ * This handler brings down the global pstate closer to the local pstate
+ * according quadratic equation. Queues a new timer if it is still not equal
+ * to local pstate
+ */
+void gpstate_timer_handler(unsigned long data)
+{
+	struct cpufreq_policy *policy = (struct cpufreq_policy *)data;
+	struct global_pstate_info *gpstates = policy->driver_data;
+	int gpstate_id;
+	unsigned int time_diff = jiffies_to_msecs(jiffies)
+					- gpstates->last_sampled_time;
+	struct powernv_smp_call_data freq_data;
+
+	if (!spin_trylock(&gpstates->gpstate_lock))
+		return;
+
+	gpstates->last_sampled_time += time_diff;
+	gpstates->elapsed_time += time_diff;
+	freq_data.pstate_id = gpstates->last_lpstate;
+
+	if ((gpstates->last_gpstate == freq_data.pstate_id) ||
+	    (gpstates->elapsed_time > MAX_RAMP_DOWN_TIME)) {
+		gpstate_id = freq_data.pstate_id;
+		reset_gpstates(policy);
+		gpstates->highest_lpstate = freq_data.pstate_id;
+	} else {
+		gpstate_id = calc_global_pstate(gpstates->elapsed_time,
+						gpstates->highest_lpstate,
+						freq_data.pstate_id);
+	}
+
+	/*
+	 * If local pstate is equal to global pstate, rampdown is over
+	 * So timer is not required to be queued.
+	 */
+	if (gpstate_id != freq_data.pstate_id)
+		queue_gpstate_timer(gpstates);
+
+	freq_data.gpstate_id = gpstate_id;
+	gpstates->last_gpstate = freq_data.gpstate_id;
+	gpstates->last_lpstate = freq_data.pstate_id;
+
+	spin_unlock(&gpstates->gpstate_lock);
+
+	/* Timer may get migrated to a different cpu on cpu hot unplug */
+	smp_call_function_any(policy->cpus, set_pstate, &freq_data, 1);
+}
+
 /*
  * powernv_cpufreq_target_index: Sets the frequency corresponding to
  * the cpufreq table entry indexed by new_index on the cpus in the
@@ -433,6 +596,8 @@ static int powernv_cpufreq_target_index(struct cpufreq_policy *policy,
 					unsigned int new_index)
 {
 	struct powernv_smp_call_data freq_data;
+	unsigned int cur_msec, gpstate_id;
+	struct global_pstate_info *gpstates = policy->driver_data;
 
 	if (unlikely(rebooting) && new_index != get_nominal_index())
 		return 0;
@@ -440,28 +605,81 @@ static int powernv_cpufreq_target_index(struct cpufreq_policy *policy,
 	if (!throttled)
 		powernv_cpufreq_throttle_check(NULL);
 
+	cur_msec = jiffies_to_msecs(get_jiffies_64());
+
+	spin_lock(&gpstates->gpstate_lock);
 	freq_data.pstate_id = powernv_freqs[new_index].driver_data;
 
+	if (!gpstates->last_sampled_time) {
+		gpstate_id = freq_data.pstate_id;
+		gpstates->highest_lpstate = freq_data.pstate_id;
+		goto gpstates_done;
+	}
+
+	if (gpstates->last_gpstate > freq_data.pstate_id) {
+		gpstates->elapsed_time += cur_msec -
+						 gpstates->last_sampled_time;
+
+		/*
+		 * If its has been ramping down for more than MAX_RAMP_DOWN_TIME
+		 * we should be resetting all global pstate related data. Set it
+		 * equal to local pstate to start fresh.
+		 */
+		if (gpstates->elapsed_time > MAX_RAMP_DOWN_TIME) {
+			reset_gpstates(policy);
+			gpstates->highest_lpstate = freq_data.pstate_id;
+			gpstate_id = freq_data.pstate_id;
+		} else {
+		/* Elaspsed_time is less than 5 seconds, continue to rampdown */
+			gpstate_id = calc_global_pstate(gpstates->elapsed_time,
+							gpstates->highest_lpstate,
+							freq_data.pstate_id);
+		}
+	} else {
+		reset_gpstates(policy);
+		gpstates->highest_lpstate = freq_data.pstate_id;
+		gpstate_id = freq_data.pstate_id;
+	}
+
+	/*
+	 * If local pstate is equal to global pstate, rampdown is over
+	 * So timer is not required to be queued.
+	 */
+	if (gpstate_id != freq_data.pstate_id)
+		queue_gpstate_timer(gpstates);
+	else
+		del_timer_sync(&gpstates->timer);
+
+gpstates_done:
+	freq_data.gpstate_id = gpstate_id;
+	gpstates->last_sampled_time = cur_msec;
+	gpstates->last_gpstate = freq_data.gpstate_id;
+	gpstates->last_lpstate = freq_data.pstate_id;
+
+	spin_unlock(&gpstates->gpstate_lock);
+
 	/*
 	 * Use smp_call_function to send IPI and execute the
 	 * mtspr on target CPU.  We could do that without IPI
 	 * if current CPU is within policy->cpus (core)
 	 */
 	smp_call_function_any(policy->cpus, set_pstate, &freq_data, 1);
-
 	return 0;
 }
 
 static int powernv_cpufreq_cpu_init(struct cpufreq_policy *policy)
 {
-	int base, i;
+	int base, i, ret;
+	struct kernfs_node *kn;
+	struct global_pstate_info *gpstates;
 
 	base = cpu_first_thread_sibling(policy->cpu);
 
 	for (i = 0; i < threads_per_core; i++)
 		cpumask_set_cpu(base + i, policy->cpus);
 
-	if (!policy->driver_data) {
+	kn = kernfs_find_and_get(policy->kobj.sd, throttle_attr_grp.name);
+	if (!kn) {
 		int ret;
 
 		ret = sysfs_create_group(&policy->kobj, &throttle_attr_grp);
@@ -470,13 +688,37 @@ static int powernv_cpufreq_cpu_init(struct cpufreq_policy *policy)
 				policy->cpu);
 			return ret;
 		}
-		/*
-		 * policy->driver_data is used as a flag for one-time
-		 * creation of throttle sysfs files.
-		 */
-		policy->driver_data = policy;
+	} else {
+		kernfs_put(kn);
 	}
-	return cpufreq_table_validate_and_show(policy, powernv_freqs);
+
+	gpstates =  kzalloc(sizeof(*gpstates), GFP_KERNEL);
+	if (!gpstates)
+		return -ENOMEM;
+
+	policy->driver_data = gpstates;
+
+	/* initialize timer */
+	init_timer_deferrable(&gpstates->timer);
+	gpstates->timer.data = (unsigned long)policy;
+	gpstates->timer.function = gpstate_timer_handler;
+	gpstates->timer.expires = jiffies +
+				msecs_to_jiffies(GPSTATE_TIMER_INTERVAL);
+	spin_lock_init(&gpstates->gpstate_lock);
+	ret = cpufreq_table_validate_and_show(policy, powernv_freqs);
+
+	if (ret < 0)
+		kfree(policy->driver_data);
+
+	return ret;
+}
+
+static int powernv_cpufreq_cpu_exit(struct cpufreq_policy *policy)
+{
+	/* timer is deleted in cpufreq_cpu_stop() */
+	kfree(policy->driver_data);
+
+	return 0;
 }
 
 static int powernv_cpufreq_reboot_notifier(struct notifier_block *nb,
@@ -604,15 +846,19 @@ static struct notifier_block powernv_cpufreq_opal_nb = {
 static void powernv_cpufreq_stop_cpu(struct cpufreq_policy *policy)
 {
 	struct powernv_smp_call_data freq_data;
+	struct global_pstate_info *gpstates = policy->driver_data;
 
 	freq_data.pstate_id = powernv_pstate_info.min;
+	freq_data.gpstate_id = powernv_pstate_info.min;
 	smp_call_function_single(policy->cpu, set_pstate, &freq_data, 1);
+	del_timer_sync(&gpstates->timer);
 }
 
 static struct cpufreq_driver powernv_cpufreq_driver = {
 	.name		= "powernv-cpufreq",
 	.flags		= CPUFREQ_CONST_LOOPS,
 	.init		= powernv_cpufreq_cpu_init,
+	.exit		= powernv_cpufreq_cpu_exit,
 	.verify		= cpufreq_generic_frequency_table_verify,
 	.target_index	= powernv_cpufreq_target_index,
 	.get		= powernv_cpufreq_get,

drivers/cpufreq/ppc_cbe_cpufreq.h

@@ -17,7 +17,7 @@ int cbe_cpufreq_get_pmode(int cpu);
 
 int cbe_cpufreq_set_pmode_pmi(int cpu, unsigned int pmode);
 
-#if defined(CONFIG_CPU_FREQ_CBE_PMI) || defined(CONFIG_CPU_FREQ_CBE_PMI_MODULE)
+#if IS_ENABLED(CONFIG_CPU_FREQ_CBE_PMI)
 extern bool cbe_cpufreq_has_pmi;
 #else
 #define cbe_cpufreq_has_pmi (0)

drivers/cpufreq/ppc_cbe_cpufreq_pmi.c

@@ -23,7 +23,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/timer.h>
-#include <linux/module.h>
+#include <linux/init.h>
 #include <linux/of_platform.h>
 
 #include <asm/processor.h>
@@ -142,15 +142,4 @@ static int __init cbe_cpufreq_pmi_init(void)
 
 	return 0;
 }
-
-static void __exit cbe_cpufreq_pmi_exit(void)
-{
-	cpufreq_unregister_notifier(&pmi_notifier_block, CPUFREQ_POLICY_NOTIFIER);
-	pmi_unregister_handler(&cbe_pmi_handler);
-}
-
-module_init(cbe_cpufreq_pmi_init);
-module_exit(cbe_cpufreq_pmi_exit);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Christian Krafft <krafft@de.ibm.com>");
+device_initcall(cbe_cpufreq_pmi_init);

drivers/cpufreq/pxa2xx-cpufreq.c

@@ -29,6 +29,8 @@
  *
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/sched.h>
@@ -186,8 +188,7 @@ static int pxa_cpufreq_change_voltage(const struct pxa_freqs *pxa_freq)
 
 	ret = regulator_set_voltage(vcc_core, vmin, vmax);
 	if (ret)
-		pr_err("cpufreq: Failed to set vcc_core in [%dmV..%dmV]\n",
-		       vmin, vmax);
+		pr_err("Failed to set vcc_core in [%dmV..%dmV]\n", vmin, vmax);
 	return ret;
 }
 
@@ -195,10 +196,10 @@ static void __init pxa_cpufreq_init_voltages(void)
 {
 	vcc_core = regulator_get(NULL, "vcc_core");
 	if (IS_ERR(vcc_core)) {
-		pr_info("cpufreq: Didn't find vcc_core regulator\n");
+		pr_info("Didn't find vcc_core regulator\n");
 		vcc_core = NULL;
 	} else {
-		pr_info("cpufreq: Found vcc_core regulator\n");
+		pr_info("Found vcc_core regulator\n");
 	}
 }
 #else
@@ -233,9 +234,8 @@ static void pxa27x_guess_max_freq(void)
 {
 	if (!pxa27x_maxfreq) {
 		pxa27x_maxfreq = 416000;
-		printk(KERN_INFO "PXA CPU 27x max frequency not defined "
-		       "(pxa27x_maxfreq), assuming pxa271 with %dkHz maxfreq\n",
-		       pxa27x_maxfreq);
+		pr_info("PXA CPU 27x max frequency not defined (pxa27x_maxfreq), assuming pxa271 with %dkHz maxfreq\n",
+			pxa27x_maxfreq);
 	} else {
 		pxa27x_maxfreq *= 1000;
 	}
@@ -408,7 +408,7 @@ static int pxa_cpufreq_init(struct cpufreq_policy *policy)
 	 */
 	if (cpu_is_pxa25x()) {
 		find_freq_tables(&pxa255_freq_table, &pxa255_freqs);
-		pr_info("PXA255 cpufreq using %s frequency table\n",
+		pr_info("using %s frequency table\n",
 			pxa255_turbo_table ? "turbo" : "run");
 
 		cpufreq_table_validate_and_show(policy, pxa255_freq_table);
@@ -417,7 +417,7 @@ static int pxa_cpufreq_init(struct cpufreq_policy *policy)
 		cpufreq_table_validate_and_show(policy, pxa27x_freq_table);
 	}
 
-	printk(KERN_INFO "PXA CPU frequency change support initialized\n");
+	pr_info("frequency change support initialized\n");
 
 	return 0;
 }

drivers/cpufreq/qoriq-cpufreq.c

@@ -301,10 +301,11 @@ err_np:
 	return -ENODEV;
 }
 
-static int __exit qoriq_cpufreq_cpu_exit(struct cpufreq_policy *policy)
+static int qoriq_cpufreq_cpu_exit(struct cpufreq_policy *policy)
 {
 	struct cpu_data *data = policy->driver_data;
 
+	cpufreq_cooling_unregister(data->cdev);
 	kfree(data->pclk);
 	kfree(data->table);
 	kfree(data);
@@ -333,8 +334,8 @@ static void qoriq_cpufreq_ready(struct cpufreq_policy *policy)
 		cpud->cdev = of_cpufreq_cooling_register(np,
 							 policy->related_cpus);
 
-		if (IS_ERR(cpud->cdev)) {
-			pr_err("Failed to register cooling device cpu%d: %ld\n",
+		if (IS_ERR(cpud->cdev) && PTR_ERR(cpud->cdev) != -ENOSYS) {
+			pr_err("cpu%d is not running as cooling device: %ld\n",
 					policy->cpu, PTR_ERR(cpud->cdev));
 
 			cpud->cdev = NULL;
@@ -348,7 +349,7 @@ static struct cpufreq_driver qoriq_cpufreq_driver = {
 	.name		= "qoriq_cpufreq",
 	.flags		= CPUFREQ_CONST_LOOPS,
 	.init		= qoriq_cpufreq_cpu_init,
-	.exit		= __exit_p(qoriq_cpufreq_cpu_exit),
+	.exit		= qoriq_cpufreq_cpu_exit,
 	.verify		= cpufreq_generic_frequency_table_verify,
 	.target_index	= qoriq_cpufreq_target,
 	.get		= cpufreq_generic_get,

drivers/cpufreq/s3c2412-cpufreq.c

@@ -10,6 +10,8 @@
  * published by the Free Software Foundation.
 */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
@@ -197,21 +199,20 @@ static int s3c2412_cpufreq_add(struct device *dev,
 
 	hclk = clk_get(NULL, "hclk");
 	if (IS_ERR(hclk)) {
-		printk(KERN_ERR "%s: cannot find hclk clock\n", __func__);
+		pr_err("cannot find hclk clock\n");
 		return -ENOENT;
 	}
 
 	fclk = clk_get(NULL, "fclk");
 	if (IS_ERR(fclk)) {
-		printk(KERN_ERR "%s: cannot find fclk clock\n", __func__);
+		pr_err("cannot find fclk clock\n");
 		goto err_fclk;
 	}
 
 	fclk_rate = clk_get_rate(fclk);
 	if (fclk_rate > 200000000) {
-		printk(KERN_INFO
-		       "%s: fclk %ld MHz, assuming 266MHz capable part\n",
-		       __func__, fclk_rate / 1000000);
+		pr_info("fclk %ld MHz, assuming 266MHz capable part\n",
+			fclk_rate / 1000000);
 		s3c2412_cpufreq_info.max.fclk = 266000000;
 		s3c2412_cpufreq_info.max.hclk = 133000000;
 		s3c2412_cpufreq_info.max.pclk =  66000000;
@@ -219,13 +220,13 @@ static int s3c2412_cpufreq_add(struct device *dev,
 
 	armclk = clk_get(NULL, "armclk");
 	if (IS_ERR(armclk)) {
-		printk(KERN_ERR "%s: cannot find arm clock\n", __func__);
+		pr_err("cannot find arm clock\n");
 		goto err_armclk;
 	}
 
 	xtal = clk_get(NULL, "xtal");
 	if (IS_ERR(xtal)) {
-		printk(KERN_ERR "%s: cannot find xtal clock\n", __func__);
+		pr_err("cannot find xtal clock\n");
 		goto err_xtal;
 	}
 

drivers/cpufreq/s3c2440-cpufreq.c

@@ -11,6 +11,8 @@
  * published by the Free Software Foundation.
 */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
@@ -66,7 +68,7 @@ static int s3c2440_cpufreq_calcdivs(struct s3c_cpufreq_config *cfg)
 		     __func__, fclk, armclk, hclk_max);
 
 	if (armclk > fclk) {
-		printk(KERN_WARNING "%s: armclk > fclk\n", __func__);
+		pr_warn("%s: armclk > fclk\n", __func__);
 		armclk = fclk;
 	}
 
@@ -273,7 +275,7 @@ static int s3c2440_cpufreq_add(struct device *dev,
 	armclk = s3c_cpufreq_clk_get(NULL, "armclk");
 
 	if (IS_ERR(xtal) || IS_ERR(hclk) || IS_ERR(fclk) || IS_ERR(armclk)) {
-		printk(KERN_ERR "%s: failed to get clocks\n", __func__);
+		pr_err("%s: failed to get clocks\n", __func__);
 		return -ENOENT;
 	}
 

drivers/cpufreq/s3c24xx-cpufreq-debugfs.c

@@ -10,6 +10,8 @@
  * published by the Free Software Foundation.
 */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/init.h>
 #include <linux/export.h>
 #include <linux/interrupt.h>
@@ -178,7 +180,7 @@ static int __init s3c_freq_debugfs_init(void)
 {
 	dbgfs_root = debugfs_create_dir("s3c-cpufreq", NULL);
 	if (IS_ERR(dbgfs_root)) {
-		printk(KERN_ERR "%s: error creating debugfs root\n", __func__);
+		pr_err("%s: error creating debugfs root\n", __func__);
 		return PTR_ERR(dbgfs_root);
 	}
 

drivers/cpufreq/s3c24xx-cpufreq.c

@@ -10,6 +10,8 @@
  * published by the Free Software Foundation.
 */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
@@ -175,7 +177,7 @@ static int s3c_cpufreq_settarget(struct cpufreq_policy *policy,
 	cpu_new.freq.fclk = cpu_new.pll.frequency;
 
 	if (s3c_cpufreq_calcdivs(&cpu_new) < 0) {
-		printk(KERN_ERR "no divisors for %d\n", target_freq);
+		pr_err("no divisors for %d\n", target_freq);
 		goto err_notpossible;
 	}
 
@@ -187,7 +189,7 @@ static int s3c_cpufreq_settarget(struct cpufreq_policy *policy,
 
 	if (cpu_new.freq.hclk != cpu_cur.freq.hclk) {
 		if (s3c_cpufreq_calcio(&cpu_new) < 0) {
-			printk(KERN_ERR "%s: no IO timings\n", __func__);
+			pr_err("%s: no IO timings\n", __func__);
 			goto err_notpossible;
 		}
 	}
@@ -262,7 +264,7 @@ static int s3c_cpufreq_settarget(struct cpufreq_policy *policy,
 	return 0;
 
  err_notpossible:
-	printk(KERN_ERR "no compatible settings for %d\n", target_freq);
+	pr_err("no compatible settings for %d\n", target_freq);
 	return -EINVAL;
 }
 
@@ -331,7 +333,7 @@ static int s3c_cpufreq_target(struct cpufreq_policy *policy,
 						     &index);
 
 		if (ret < 0) {
-			printk(KERN_ERR "%s: no PLL available\n", __func__);
+			pr_err("%s: no PLL available\n", __func__);
 			goto err_notpossible;
 		}
 
@@ -346,7 +348,7 @@ static int s3c_cpufreq_target(struct cpufreq_policy *policy,
 	return s3c_cpufreq_settarget(policy, target_freq, pll);
 
  err_notpossible:
-	printk(KERN_ERR "no compatible settings for %d\n", target_freq);
+	pr_err("no compatible settings for %d\n", target_freq);
 	return -EINVAL;
 }
 
@@ -356,7 +358,7 @@ struct clk *s3c_cpufreq_clk_get(struct device *dev, const char *name)
 
 	clk = clk_get(dev, name);
 	if (IS_ERR(clk))
-		printk(KERN_ERR "cpufreq: failed to get clock '%s'\n", name);
+		pr_err("failed to get clock '%s'\n", name);
 
 	return clk;
 }
@@ -378,15 +380,16 @@ static int __init s3c_cpufreq_initclks(void)
 
 	if (IS_ERR(clk_fclk) || IS_ERR(clk_hclk) || IS_ERR(clk_pclk) ||
 	    IS_ERR(_clk_mpll) || IS_ERR(clk_arm) || IS_ERR(_clk_xtal)) {
-		printk(KERN_ERR "%s: could not get clock(s)\n", __func__);
+		pr_err("%s: could not get clock(s)\n", __func__);
 		return -ENOENT;
 	}
 
-	printk(KERN_INFO "%s: clocks f=%lu,h=%lu,p=%lu,a=%lu\n", __func__,
-	       clk_get_rate(clk_fclk) / 1000,
-	       clk_get_rate(clk_hclk) / 1000,
-	       clk_get_rate(clk_pclk) / 1000,
-	       clk_get_rate(clk_arm) / 1000);
+	pr_info("%s: clocks f=%lu,h=%lu,p=%lu,a=%lu\n",
+		__func__,
+		clk_get_rate(clk_fclk) / 1000,
+		clk_get_rate(clk_hclk) / 1000,
+		clk_get_rate(clk_pclk) / 1000,
+		clk_get_rate(clk_arm) / 1000);
 
 	return 0;
 }
@@ -424,7 +427,7 @@ static int s3c_cpufreq_resume(struct cpufreq_policy *policy)
 
 	ret = s3c_cpufreq_settarget(NULL, suspend_freq, &suspend_pll);
 	if (ret) {
-		printk(KERN_ERR "%s: failed to reset pll/freq\n", __func__);
+		pr_err("%s: failed to reset pll/freq\n", __func__);
 		return ret;
 	}
 
@@ -449,13 +452,12 @@ static struct cpufreq_driver s3c24xx_driver = {
 int s3c_cpufreq_register(struct s3c_cpufreq_info *info)
 {
 	if (!info || !info->name) {
-		printk(KERN_ERR "%s: failed to pass valid information\n",
-		       __func__);
+		pr_err("%s: failed to pass valid information\n", __func__);
 		return -EINVAL;
 	}
 
-	printk(KERN_INFO "S3C24XX CPU Frequency driver, %s cpu support\n",
-	       info->name);
+	pr_info("S3C24XX CPU Frequency driver, %s cpu support\n",
+		info->name);
 
 	/* check our driver info has valid data */
 
@@ -478,7 +480,7 @@ int __init s3c_cpufreq_setboard(struct s3c_cpufreq_board *board)
 	struct s3c_cpufreq_board *ours;
 
 	if (!board) {
-		printk(KERN_INFO "%s: no board data\n", __func__);
+		pr_info("%s: no board data\n", __func__);
 		return -EINVAL;
 	}
 
@@ -487,7 +489,7 @@ int __init s3c_cpufreq_setboard(struct s3c_cpufreq_board *board)
 
 	ours = kzalloc(sizeof(*ours), GFP_KERNEL);
 	if (ours == NULL) {
-		printk(KERN_ERR "%s: no memory\n", __func__);
+		pr_err("%s: no memory\n", __func__);
 		return -ENOMEM;
 	}
 
@@ -502,15 +504,15 @@ static int __init s3c_cpufreq_auto_io(void)
 	int ret;
 
 	if (!cpu_cur.info->get_iotiming) {
-		printk(KERN_ERR "%s: get_iotiming undefined\n", __func__);
+		pr_err("%s: get_iotiming undefined\n", __func__);
 		return -ENOENT;
 	}
 
-	printk(KERN_INFO "%s: working out IO settings\n", __func__);
+	pr_info("%s: working out IO settings\n", __func__);
 
 	ret = (cpu_cur.info->get_iotiming)(&cpu_cur, &s3c24xx_iotiming);
 	if (ret)
-		printk(KERN_ERR "%s: failed to get timings\n", __func__);
+		pr_err("%s: failed to get timings\n", __func__);
 
 	return ret;
 }
@@ -561,7 +563,7 @@ static void s3c_cpufreq_update_loctkime(void)
 	val = calc_locktime(rate, cpu_cur.info->locktime_u) << bits;
 	val |= calc_locktime(rate, cpu_cur.info->locktime_m);
 
-	printk(KERN_INFO "%s: new locktime is 0x%08x\n", __func__, val);
+	pr_info("%s: new locktime is 0x%08x\n", __func__, val);
 	__raw_writel(val, S3C2410_LOCKTIME);
 }
 
@@ -580,7 +582,7 @@ static int s3c_cpufreq_build_freq(void)
 
 	ftab = kzalloc(sizeof(*ftab) * size, GFP_KERNEL);
 	if (!ftab) {
-		printk(KERN_ERR "%s: no memory for tables\n", __func__);
+		pr_err("%s: no memory for tables\n", __func__);
 		return -ENOMEM;
 	}
 
@@ -608,15 +610,14 @@ static int __init s3c_cpufreq_initcall(void)
 		if (cpu_cur.board->auto_io) {
 			ret = s3c_cpufreq_auto_io();
 			if (ret) {
-				printk(KERN_ERR "%s: failed to get io timing\n",
+				pr_err("%s: failed to get io timing\n",
 				       __func__);
 				goto out;
 			}
 		}
 
 		if (cpu_cur.board->need_io && !cpu_cur.info->set_iotiming) {
-			printk(KERN_ERR "%s: no IO support registered\n",
-			       __func__);
+			pr_err("%s: no IO support registered\n", __func__);
 			ret = -EINVAL;
 			goto out;
 		}
@@ -666,9 +667,9 @@ int s3c_plltab_register(struct cpufreq_frequency_table *plls,
 		vals += plls_no;
 		vals->frequency = CPUFREQ_TABLE_END;
 
-		printk(KERN_INFO "cpufreq: %d PLL entries\n", plls_no);
+		pr_info("%d PLL entries\n", plls_no);
 	} else
-		printk(KERN_ERR "cpufreq: no memory for PLL tables\n");
+		pr_err("no memory for PLL tables\n");
 
 	return vals ? 0 : -ENOMEM;
 }

drivers/cpufreq/s5pv210-cpufreq.c

@@ -9,6 +9,8 @@
  * published by the Free Software Foundation.
 */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
@@ -205,7 +207,7 @@ static void s5pv210_set_refresh(enum s5pv210_dmc_port ch, unsigned long freq)
 	} else if (ch == DMC1) {
 		reg = (dmc_base[1] + 0x30);
 	} else {
-		printk(KERN_ERR "Cannot find DMC port\n");
+		pr_err("Cannot find DMC port\n");
 		return;
 	}
 
@@ -534,7 +536,7 @@ static int s5pv210_cpu_init(struct cpufreq_policy *policy)
 	mem_type = check_mem_type(dmc_base[0]);
 
 	if ((mem_type != LPDDR) && (mem_type != LPDDR2)) {
-		printk(KERN_ERR "CPUFreq doesn't support this memory type\n");
+		pr_err("CPUFreq doesn't support this memory type\n");
 		ret = -EINVAL;
 		goto out_dmc1;
 	}
@@ -635,13 +637,13 @@ static int s5pv210_cpufreq_probe(struct platform_device *pdev)
 
 	arm_regulator = regulator_get(NULL, "vddarm");
 	if (IS_ERR(arm_regulator)) {
-		pr_err("failed to get regulator vddarm");
+		pr_err("failed to get regulator vddarm\n");
 		return PTR_ERR(arm_regulator);
 	}
 
 	int_regulator = regulator_get(NULL, "vddint");
 	if (IS_ERR(int_regulator)) {
-		pr_err("failed to get regulator vddint");
+		pr_err("failed to get regulator vddint\n");
 		regulator_put(arm_regulator);
 		return PTR_ERR(int_regulator);
 	}

drivers/cpufreq/sc520_freq.c

@@ -13,6 +13,8 @@
  *	2005-03-30: - initial revision
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -30,8 +32,6 @@
 
 static __u8 __iomem *cpuctl;
 
-#define PFX "sc520_freq: "
-
 static struct cpufreq_frequency_table sc520_freq_table[] = {
 	{0, 0x01,	100000},
 	{0, 0x02,	133000},
@@ -44,8 +44,8 @@ static unsigned int sc520_freq_get_cpu_frequency(unsigned int cpu)
 
 	switch (clockspeed_reg & 0x03) {
 	default:
-		printk(KERN_ERR PFX "error: cpuctl register has unexpected "
-				"value %02x\n", clockspeed_reg);
+		pr_err("error: cpuctl register has unexpected value %02x\n",
+		       clockspeed_reg);
 	case 0x01:
 		return 100000;
 	case 0x02:
@@ -112,7 +112,7 @@ static int __init sc520_freq_init(void)
 
 	cpuctl = ioremap((unsigned long)(MMCR_BASE + OFFS_CPUCTL), 1);
 	if (!cpuctl) {
-		printk(KERN_ERR "sc520_freq: error: failed to remap memory\n");
+		pr_err("sc520_freq: error: failed to remap memory\n");
 		return -ENOMEM;
 	}
 

drivers/cpufreq/scpi-cpufreq.c

@@ -18,6 +18,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <linux/cpu.h>
 #include <linux/cpufreq.h>
 #include <linux/module.h>
 #include <linux/platform_device.h>
@@ -38,10 +39,20 @@ static struct scpi_dvfs_info *scpi_get_dvfs_info(struct device *cpu_dev)
 	return scpi_ops->dvfs_get_info(domain);
 }
 
-static int scpi_opp_table_ops(struct device *cpu_dev, bool remove)
+static int scpi_get_transition_latency(struct device *cpu_dev)
 {
-	int idx, ret = 0;
+	struct scpi_dvfs_info *info = scpi_get_dvfs_info(cpu_dev);
+
+	if (IS_ERR(info))
+		return PTR_ERR(info);
+	return info->latency;
+}
+
+static int scpi_init_opp_table(const struct cpumask *cpumask)
+{
+	int idx, ret;
 	struct scpi_opp *opp;
+	struct device *cpu_dev = get_cpu_device(cpumask_first(cpumask));
 	struct scpi_dvfs_info *info = scpi_get_dvfs_info(cpu_dev);
 
 	if (IS_ERR(info))
@@ -51,11 +62,7 @@ static int scpi_opp_table_ops(struct device *cpu_dev, bool remove)
 		return -EIO;
 
 	for (opp = info->opps, idx = 0; idx < info->count; idx++, opp++) {
-		if (remove)
-			dev_pm_opp_remove(cpu_dev, opp->freq);
-		else
-			ret = dev_pm_opp_add(cpu_dev, opp->freq,
-					     opp->m_volt * 1000);
+		ret = dev_pm_opp_add(cpu_dev, opp->freq, opp->m_volt * 1000);
 		if (ret) {
 			dev_warn(cpu_dev, "failed to add opp %uHz %umV\n",
 				 opp->freq, opp->m_volt);
@@ -64,33 +71,19 @@ static int scpi_opp_table_ops(struct device *cpu_dev, bool remove)
 			return ret;
 		}
 	}
-	return ret;
-}
 
-static int scpi_get_transition_latency(struct device *cpu_dev)
-{
-	struct scpi_dvfs_info *info = scpi_get_dvfs_info(cpu_dev);
-
-	if (IS_ERR(info))
-		return PTR_ERR(info);
-	return info->latency;
-}
-
-static int scpi_init_opp_table(struct device *cpu_dev)
-{
-	return scpi_opp_table_ops(cpu_dev, false);
-}
-
-static void scpi_free_opp_table(struct device *cpu_dev)
-{
-	scpi_opp_table_ops(cpu_dev, true);
+	ret = dev_pm_opp_set_sharing_cpus(cpu_dev, cpumask);
+	if (ret)
+		dev_err(cpu_dev, "%s: failed to mark OPPs as shared: %d\n",
+			__func__, ret);
+	return ret;
 }
 
 static struct cpufreq_arm_bL_ops scpi_cpufreq_ops = {
 	.name	= "scpi",
 	.get_transition_latency = scpi_get_transition_latency,
 	.init_opp_table = scpi_init_opp_table,
-	.free_opp_table = scpi_free_opp_table,
+	.free_opp_table = dev_pm_opp_cpumask_remove_table,
 };
 
 static int scpi_cpufreq_probe(struct platform_device *pdev)

drivers/cpufreq/speedstep-centrino.c

@@ -13,6 +13,8 @@
  * Copyright (C) 2003 Jeremy Fitzhardinge <jeremy@goop.org>
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -27,7 +29,6 @@
 #include <asm/cpufeature.h>
 #include <asm/cpu_device_id.h>
 
-#define PFX		"speedstep-centrino: "
 #define MAINTAINER	"linux-pm@vger.kernel.org"
 
 #define INTEL_MSR_RANGE	(0xffff)
@@ -386,8 +387,7 @@ static int centrino_cpu_init(struct cpufreq_policy *policy)
 		/* check to see if it stuck */
 		rdmsr(MSR_IA32_MISC_ENABLE, l, h);
 		if (!(l & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) {
-			printk(KERN_INFO PFX
-				"couldn't enable Enhanced SpeedStep\n");
+			pr_info("couldn't enable Enhanced SpeedStep\n");
 			return -ENODEV;
 		}
 	}

drivers/cpufreq/speedstep-ich.c

@@ -18,6 +18,8 @@
  *                        SPEEDSTEP - DEFINITIONS                    *
  *********************************************************************/
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -68,13 +70,13 @@ static int speedstep_find_register(void)
 	/* get PMBASE */
 	pci_read_config_dword(speedstep_chipset_dev, 0x40, &pmbase);
 	if (!(pmbase & 0x01)) {
-		printk(KERN_ERR "speedstep-ich: could not find speedstep register\n");
+		pr_err("could not find speedstep register\n");
 		return -ENODEV;
 	}
 
 	pmbase &= 0xFFFFFFFE;
 	if (!pmbase) {
-		printk(KERN_ERR "speedstep-ich: could not find speedstep register\n");
+		pr_err("could not find speedstep register\n");
 		return -ENODEV;
 	}
 
@@ -136,7 +138,7 @@ static void speedstep_set_state(unsigned int state)
 		pr_debug("change to %u MHz succeeded\n",
 			speedstep_get_frequency(speedstep_processor) / 1000);
 	else
-		printk(KERN_ERR "cpufreq: change failed - I/O error\n");
+		pr_err("change failed - I/O error\n");
 
 	return;
 }

drivers/cpufreq/speedstep-lib.c

@@ -8,6 +8,8 @@
  *  BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous*
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
@@ -153,7 +155,7 @@ static unsigned int pentium_core_get_frequency(void)
 		fsb = 333333;
 		break;
 	default:
-		printk(KERN_ERR "PCORE - MSR_FSB_FREQ undefined value");
+		pr_err("PCORE - MSR_FSB_FREQ undefined value\n");
 	}
 
 	rdmsr(MSR_IA32_EBL_CR_POWERON, msr_lo, msr_tmp);
@@ -453,11 +455,8 @@ unsigned int speedstep_get_freqs(enum speedstep_processor processor,
 		 */
 		if (*transition_latency > 10000000 ||
 		    *transition_latency < 50000) {
-			printk(KERN_WARNING PFX "frequency transition "
-					"measured seems out of range (%u "
-					"nSec), falling back to a safe one of"
-					"%u nSec.\n",
-					*transition_latency, 500000);
+			pr_warn("frequency transition measured seems out of range (%u nSec), falling back to a safe one of %u nSec\n",
+				*transition_latency, 500000);
 			*transition_latency = 500000;
 		}
 	}

drivers/cpufreq/speedstep-smi.c

@@ -12,6 +12,8 @@
  *                        SPEEDSTEP - DEFINITIONS                    *
  *********************************************************************/
 
+#define pr_fmt(fmt) "cpufreq: " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
@@ -204,9 +206,8 @@ static void speedstep_set_state(unsigned int state)
 			(speedstep_freqs[new_state].frequency / 1000),
 			retry, result);
 	else
-		printk(KERN_ERR "cpufreq: change to state %u "
-			"failed with new_state %u and result %u\n",
-			state, new_state, result);
+		pr_err("change to state %u failed with new_state %u and result %u\n",
+		       state, new_state, result);
 
 	return;
 }

drivers/cpufreq/tegra124-cpufreq.c

@@ -14,7 +14,6 @@
 #define pr_fmt(fmt)	KBUILD_MODNAME ": " fmt
 
 #include <linux/clk.h>
-#include <linux/cpufreq-dt.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
@@ -69,10 +68,6 @@ static void tegra124_cpu_switch_to_pllx(struct tegra124_cpufreq_priv *priv)
 	clk_set_parent(priv->cpu_clk, priv->pllx_clk);
 }
 
-static struct cpufreq_dt_platform_data cpufreq_dt_pd = {
-	.independent_clocks = false,
-};
-
 static int tegra124_cpufreq_probe(struct platform_device *pdev)
 {
 	struct tegra124_cpufreq_priv *priv;
@@ -129,8 +124,6 @@ static int tegra124_cpufreq_probe(struct platform_device *pdev)
 
 	cpufreq_dt_devinfo.name = "cpufreq-dt";
 	cpufreq_dt_devinfo.parent = &pdev->dev;
-	cpufreq_dt_devinfo.data = &cpufreq_dt_pd;
-	cpufreq_dt_devinfo.size_data = sizeof(cpufreq_dt_pd);
 
 	priv->cpufreq_dt_pdev =
 		platform_device_register_full(&cpufreq_dt_devinfo);

drivers/cpufreq/vexpress-spc-cpufreq.c

@@ -18,6 +18,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <linux/cpu.h>
 #include <linux/cpufreq.h>
 #include <linux/module.h>
 #include <linux/platform_device.h>
@@ -26,8 +27,9 @@
 
 #include "arm_big_little.h"
 
-static int ve_spc_init_opp_table(struct device *cpu_dev)
+static int ve_spc_init_opp_table(const struct cpumask *cpumask)
 {
+	struct device *cpu_dev = get_cpu_device(cpumask_first(cpumask));
 	/*
 	 * platform specific SPC code must initialise the opp table
 	 * so just check if the OPP count is non-zero

include/linux/cpufreq-dt.h

@@ -1,22 +0,0 @@
-/*
- * Copyright (C) 2014 Marvell
- * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#ifndef __CPUFREQ_DT_H__
-#define __CPUFREQ_DT_H__
-
-struct cpufreq_dt_platform_data {
-	/*
-	 * True when each CPU has its own clock to control its
-	 * frequency, false when all CPUs are controlled by a single
-	 * clock.
-	 */
-	bool independent_clocks;
-};
-
-#endif /* __CPUFREQ_DT_H__ */

include/linux/cpufreq.h

@@ -102,6 +102,17 @@ struct cpufreq_policy {
 	 */
 	struct rw_semaphore	rwsem;
 
+	/*
+	 * Fast switch flags:
+	 * - fast_switch_possible should be set by the driver if it can
+	 *   guarantee that frequency can be changed on any CPU sharing the
+	 *   policy and that the change will affect all of the policy CPUs then.
+	 * - fast_switch_enabled is to be set by governors that support fast
+	 *   freqnency switching with the help of cpufreq_enable_fast_switch().
+	 */
+	bool			fast_switch_possible;
+	bool			fast_switch_enabled;
+
 	/* Synchronization for frequency transitions */
 	bool			transition_ongoing; /* Tracks transition status */
 	spinlock_t		transition_lock;
@@ -156,6 +167,8 @@ int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu);
 int cpufreq_update_policy(unsigned int cpu);
 bool have_governor_per_policy(void);
 struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy);
+void cpufreq_enable_fast_switch(struct cpufreq_policy *policy);
+void cpufreq_disable_fast_switch(struct cpufreq_policy *policy);
 #else
 static inline unsigned int cpufreq_get(unsigned int cpu)
 {
@@ -236,6 +249,8 @@ struct cpufreq_driver {
 				  unsigned int relation);	/* Deprecated */
 	int		(*target_index)(struct cpufreq_policy *policy,
 					unsigned int index);
+	unsigned int	(*fast_switch)(struct cpufreq_policy *policy,
+				       unsigned int target_freq);
 	/*
 	 * Only for drivers with target_index() and CPUFREQ_ASYNC_NOTIFICATION
 	 * unset.
@@ -426,6 +441,20 @@ static inline unsigned long cpufreq_scale(unsigned long old, u_int div,
 #define CPUFREQ_POLICY_POWERSAVE	(1)
 #define CPUFREQ_POLICY_PERFORMANCE	(2)
 
+/*
+ * The polling frequency depends on the capability of the processor. Default
+ * polling frequency is 1000 times the transition latency of the processor. The
+ * ondemand governor will work on any processor with transition latency <= 10ms,
+ * using appropriate sampling rate.
+ *
+ * For CPUs with transition latency > 10ms (mostly drivers with CPUFREQ_ETERNAL)
+ * the ondemand governor will not work. All times here are in us (microseconds).
+ */
+#define MIN_SAMPLING_RATE_RATIO		(2)
+#define LATENCY_MULTIPLIER		(1000)
+#define MIN_LATENCY_MULTIPLIER		(20)
+#define TRANSITION_LATENCY_LIMIT	(10 * 1000 * 1000)
+
 /* Governor Events */
 #define CPUFREQ_GOV_START	1
 #define CPUFREQ_GOV_STOP	2
@@ -450,6 +479,8 @@ struct cpufreq_governor {
 };
 
 /* Pass a target to the cpufreq driver */
+unsigned int cpufreq_driver_fast_switch(struct cpufreq_policy *policy,
+					unsigned int target_freq);
 int cpufreq_driver_target(struct cpufreq_policy *policy,
 				 unsigned int target_freq,
 				 unsigned int relation);
@@ -462,6 +493,29 @@ void cpufreq_unregister_governor(struct cpufreq_governor *governor);
 struct cpufreq_governor *cpufreq_default_governor(void);
 struct cpufreq_governor *cpufreq_fallback_governor(void);
 
+/* Governor attribute set */
+struct gov_attr_set {
+	struct kobject kobj;
+	struct list_head policy_list;
+	struct mutex update_lock;
+	int usage_count;
+};
+
+/* sysfs ops for cpufreq governors */
+extern const struct sysfs_ops governor_sysfs_ops;
+
+void gov_attr_set_init(struct gov_attr_set *attr_set, struct list_head *list_node);
+void gov_attr_set_get(struct gov_attr_set *attr_set, struct list_head *list_node);
+unsigned int gov_attr_set_put(struct gov_attr_set *attr_set, struct list_head *list_node);
+
+/* Governor sysfs attribute */
+struct governor_attr {
+	struct attribute attr;
+	ssize_t (*show)(struct gov_attr_set *attr_set, char *buf);
+	ssize_t (*store)(struct gov_attr_set *attr_set, const char *buf,
+			 size_t count);
+};
+
 /*********************************************************************
  *                     FREQUENCY TABLE HELPERS                       *
  *********************************************************************/

include/linux/sched.h

@@ -3240,7 +3240,10 @@ struct update_util_data {
 		     u64 time, unsigned long util, unsigned long max);
 };
 
-void cpufreq_set_update_util_data(int cpu, struct update_util_data *data);
+void cpufreq_add_update_util_hook(int cpu, struct update_util_data *data,
+			void (*func)(struct update_util_data *data, u64 time,
+				     unsigned long util, unsigned long max));
+void cpufreq_remove_update_util_hook(int cpu);
 #endif /* CONFIG_CPU_FREQ */
 
 #endif

kernel/sched/Makefile

@@ -24,3 +24,4 @@ obj-$(CONFIG_SCHEDSTATS) += stats.o
 obj-$(CONFIG_SCHED_DEBUG) += debug.o
 obj-$(CONFIG_CGROUP_CPUACCT) += cpuacct.o
 obj-$(CONFIG_CPU_FREQ) += cpufreq.o
+obj-$(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) += cpufreq_schedutil.o

kernel/sched/cpufreq.c

@@ -14,24 +14,50 @@
 DEFINE_PER_CPU(struct update_util_data *, cpufreq_update_util_data);
 
 /**
- * cpufreq_set_update_util_data - Populate the CPU's update_util_data pointer.
+ * cpufreq_add_update_util_hook - Populate the CPU's update_util_data pointer.
  * @cpu: The CPU to set the pointer for.
  * @data: New pointer value.
+ * @func: Callback function to set for the CPU.
  *
- * Set and publish the update_util_data pointer for the given CPU.  That pointer
- * points to a struct update_util_data object containing a callback function
- * to call from cpufreq_update_util().  That function will be called from an RCU
- * read-side critical section, so it must not sleep.
+ * Set and publish the update_util_data pointer for the given CPU.
  *
- * Callers must use RCU-sched callbacks to free any memory that might be
- * accessed via the old update_util_data pointer or invoke synchronize_sched()
- * right after this function to avoid use-after-free.
+ * The update_util_data pointer of @cpu is set to @data and the callback
+ * function pointer in the target struct update_util_data is set to @func.
+ * That function will be called by cpufreq_update_util() from RCU-sched
+ * read-side critical sections, so it must not sleep.  @data will always be
+ * passed to it as the first argument which allows the function to get to the
+ * target update_util_data structure and its container.
+ *
+ * The update_util_data pointer of @cpu must be NULL when this function is
+ * called or it will WARN() and return with no effect.
  */
-void cpufreq_set_update_util_data(int cpu, struct update_util_data *data)
+void cpufreq_add_update_util_hook(int cpu, struct update_util_data *data,
+			void (*func)(struct update_util_data *data, u64 time,
+				     unsigned long util, unsigned long max))
 {
-	if (WARN_ON(data && !data->func))
+	if (WARN_ON(!data || !func))
 		return;
 
+	if (WARN_ON(per_cpu(cpufreq_update_util_data, cpu)))
+		return;
+
+	data->func = func;
 	rcu_assign_pointer(per_cpu(cpufreq_update_util_data, cpu), data);
 }
-EXPORT_SYMBOL_GPL(cpufreq_set_update_util_data);
+EXPORT_SYMBOL_GPL(cpufreq_add_update_util_hook);
+
+/**
+ * cpufreq_remove_update_util_hook - Clear the CPU's update_util_data pointer.
+ * @cpu: The CPU to clear the pointer for.
+ *
+ * Clear the update_util_data pointer for the given CPU.
+ *
+ * Callers must use RCU-sched callbacks to free any memory that might be
+ * accessed via the old update_util_data pointer or invoke synchronize_sched()
+ * right after this function to avoid use-after-free.
+ */
+void cpufreq_remove_update_util_hook(int cpu)
+{
+	rcu_assign_pointer(per_cpu(cpufreq_update_util_data, cpu), NULL);
+}
+EXPORT_SYMBOL_GPL(cpufreq_remove_update_util_hook);

kernel/sched/cpufreq_schedutil.c

@@ -0,0 +1,530 @@
+/*
+ * CPUFreq governor based on scheduler-provided CPU utilization data.
+ *
+ * Copyright (C) 2016, Intel Corporation
+ * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/cpufreq.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <trace/events/power.h>
+
+#include "sched.h"
+
+struct sugov_tunables {
+	struct gov_attr_set attr_set;
+	unsigned int rate_limit_us;
+};
+
+struct sugov_policy {
+	struct cpufreq_policy *policy;
+
+	struct sugov_tunables *tunables;
+	struct list_head tunables_hook;
+
+	raw_spinlock_t update_lock;  /* For shared policies */
+	u64 last_freq_update_time;
+	s64 freq_update_delay_ns;
+	unsigned int next_freq;
+
+	/* The next fields are only needed if fast switch cannot be used. */
+	struct irq_work irq_work;
+	struct work_struct work;
+	struct mutex work_lock;
+	bool work_in_progress;
+
+	bool need_freq_update;
+};
+
+struct sugov_cpu {
+	struct update_util_data update_util;
+	struct sugov_policy *sg_policy;
+
+	/* The fields below are only needed when sharing a policy. */
+	unsigned long util;
+	unsigned long max;
+	u64 last_update;
+};
+
+static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
+
+/************************ Governor internals ***********************/
+
+static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
+{
+	s64 delta_ns;
+
+	if (sg_policy->work_in_progress)
+		return false;
+
+	if (unlikely(sg_policy->need_freq_update)) {
+		sg_policy->need_freq_update = false;
+		/*
+		 * This happens when limits change, so forget the previous
+		 * next_freq value and force an update.
+		 */
+		sg_policy->next_freq = UINT_MAX;
+		return true;
+	}
+
+	delta_ns = time - sg_policy->last_freq_update_time;
+	return delta_ns >= sg_policy->freq_update_delay_ns;
+}
+
+static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,
+				unsigned int next_freq)
+{
+	struct cpufreq_policy *policy = sg_policy->policy;
+
+	sg_policy->last_freq_update_time = time;
+
+	if (policy->fast_switch_enabled) {
+		if (sg_policy->next_freq == next_freq) {
+			trace_cpu_frequency(policy->cur, smp_processor_id());
+			return;
+		}
+		sg_policy->next_freq = next_freq;
+		next_freq = cpufreq_driver_fast_switch(policy, next_freq);
+		if (next_freq == CPUFREQ_ENTRY_INVALID)
+			return;
+
+		policy->cur = next_freq;
+		trace_cpu_frequency(next_freq, smp_processor_id());
+	} else if (sg_policy->next_freq != next_freq) {
+		sg_policy->next_freq = next_freq;
+		sg_policy->work_in_progress = true;
+		irq_work_queue(&sg_policy->irq_work);
+	}
+}
+
+/**
+ * get_next_freq - Compute a new frequency for a given cpufreq policy.
+ * @policy: cpufreq policy object to compute the new frequency for.
+ * @util: Current CPU utilization.
+ * @max: CPU capacity.
+ *
+ * If the utilization is frequency-invariant, choose the new frequency to be
+ * proportional to it, that is
+ *
+ * next_freq = C * max_freq * util / max
+ *
+ * Otherwise, approximate the would-be frequency-invariant utilization by
+ * util_raw * (curr_freq / max_freq) which leads to
+ *
+ * next_freq = C * curr_freq * util_raw / max
+ *
+ * Take C = 1.25 for the frequency tipping point at (util / max) = 0.8.
+ */
+static unsigned int get_next_freq(struct cpufreq_policy *policy,
+				  unsigned long util, unsigned long max)
+{
+	unsigned int freq = arch_scale_freq_invariant() ?
+				policy->cpuinfo.max_freq : policy->cur;
+
+	return (freq + (freq >> 2)) * util / max;
+}
+
+static void sugov_update_single(struct update_util_data *hook, u64 time,
+				unsigned long util, unsigned long max)
+{
+	struct sugov_cpu *sg_cpu = container_of(hook, struct sugov_cpu, update_util);
+	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
+	struct cpufreq_policy *policy = sg_policy->policy;
+	unsigned int next_f;
+
+	if (!sugov_should_update_freq(sg_policy, time))
+		return;
+
+	next_f = util == ULONG_MAX ? policy->cpuinfo.max_freq :
+			get_next_freq(policy, util, max);
+	sugov_update_commit(sg_policy, time, next_f);
+}
+
+static unsigned int sugov_next_freq_shared(struct sugov_policy *sg_policy,
+					   unsigned long util, unsigned long max)
+{
+	struct cpufreq_policy *policy = sg_policy->policy;
+	unsigned int max_f = policy->cpuinfo.max_freq;
+	u64 last_freq_update_time = sg_policy->last_freq_update_time;
+	unsigned int j;
+
+	if (util == ULONG_MAX)
+		return max_f;
+
+	for_each_cpu(j, policy->cpus) {
+		struct sugov_cpu *j_sg_cpu;
+		unsigned long j_util, j_max;
+		s64 delta_ns;
+
+		if (j == smp_processor_id())
+			continue;
+
+		j_sg_cpu = &per_cpu(sugov_cpu, j);
+		/*
+		 * If the CPU utilization was last updated before the previous
+		 * frequency update and the time elapsed between the last update
+		 * of the CPU utilization and the last frequency update is long
+		 * enough, don't take the CPU into account as it probably is
+		 * idle now.
+		 */
+		delta_ns = last_freq_update_time - j_sg_cpu->last_update;
+		if (delta_ns > TICK_NSEC)
+			continue;
+
+		j_util = j_sg_cpu->util;
+		if (j_util == ULONG_MAX)
+			return max_f;
+
+		j_max = j_sg_cpu->max;
+		if (j_util * max > j_max * util) {
+			util = j_util;
+			max = j_max;
+		}
+	}
+
+	return get_next_freq(policy, util, max);
+}
+
+static void sugov_update_shared(struct update_util_data *hook, u64 time,
+				unsigned long util, unsigned long max)
+{
+	struct sugov_cpu *sg_cpu = container_of(hook, struct sugov_cpu, update_util);
+	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
+	unsigned int next_f;
+
+	raw_spin_lock(&sg_policy->update_lock);
+
+	sg_cpu->util = util;
+	sg_cpu->max = max;
+	sg_cpu->last_update = time;
+
+	if (sugov_should_update_freq(sg_policy, time)) {
+		next_f = sugov_next_freq_shared(sg_policy, util, max);
+		sugov_update_commit(sg_policy, time, next_f);
+	}
+
+	raw_spin_unlock(&sg_policy->update_lock);
+}
+
+static void sugov_work(struct work_struct *work)
+{
+	struct sugov_policy *sg_policy = container_of(work, struct sugov_policy, work);
+
+	mutex_lock(&sg_policy->work_lock);
+	__cpufreq_driver_target(sg_policy->policy, sg_policy->next_freq,
+				CPUFREQ_RELATION_L);
+	mutex_unlock(&sg_policy->work_lock);
+
+	sg_policy->work_in_progress = false;
+}
+
+static void sugov_irq_work(struct irq_work *irq_work)
+{
+	struct sugov_policy *sg_policy;
+
+	sg_policy = container_of(irq_work, struct sugov_policy, irq_work);
+	schedule_work_on(smp_processor_id(), &sg_policy->work);
+}
+
+/************************** sysfs interface ************************/
+
+static struct sugov_tunables *global_tunables;
+static DEFINE_MUTEX(global_tunables_lock);
+
+static inline struct sugov_tunables *to_sugov_tunables(struct gov_attr_set *attr_set)
+{
+	return container_of(attr_set, struct sugov_tunables, attr_set);
+}
+
+static ssize_t rate_limit_us_show(struct gov_attr_set *attr_set, char *buf)
+{
+	struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+
+	return sprintf(buf, "%u\n", tunables->rate_limit_us);
+}
+
+static ssize_t rate_limit_us_store(struct gov_attr_set *attr_set, const char *buf,
+				   size_t count)
+{
+	struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+	struct sugov_policy *sg_policy;
+	unsigned int rate_limit_us;
+
+	if (kstrtouint(buf, 10, &rate_limit_us))
+		return -EINVAL;
+
+	tunables->rate_limit_us = rate_limit_us;
+
+	list_for_each_entry(sg_policy, &attr_set->policy_list, tunables_hook)
+		sg_policy->freq_update_delay_ns = rate_limit_us * NSEC_PER_USEC;
+
+	return count;
+}
+
+static struct governor_attr rate_limit_us = __ATTR_RW(rate_limit_us);
+
+static struct attribute *sugov_attributes[] = {
+	&rate_limit_us.attr,
+	NULL
+};
+
+static struct kobj_type sugov_tunables_ktype = {
+	.default_attrs = sugov_attributes,
+	.sysfs_ops = &governor_sysfs_ops,
+};
+
+/********************** cpufreq governor interface *********************/
+
+static struct cpufreq_governor schedutil_gov;
+
+static struct sugov_policy *sugov_policy_alloc(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy;
+
+	sg_policy = kzalloc(sizeof(*sg_policy), GFP_KERNEL);
+	if (!sg_policy)
+		return NULL;
+
+	sg_policy->policy = policy;
+	init_irq_work(&sg_policy->irq_work, sugov_irq_work);
+	INIT_WORK(&sg_policy->work, sugov_work);
+	mutex_init(&sg_policy->work_lock);
+	raw_spin_lock_init(&sg_policy->update_lock);
+	return sg_policy;
+}
+
+static void sugov_policy_free(struct sugov_policy *sg_policy)
+{
+	mutex_destroy(&sg_policy->work_lock);
+	kfree(sg_policy);
+}
+
+static struct sugov_tunables *sugov_tunables_alloc(struct sugov_policy *sg_policy)
+{
+	struct sugov_tunables *tunables;
+
+	tunables = kzalloc(sizeof(*tunables), GFP_KERNEL);
+	if (tunables) {
+		gov_attr_set_init(&tunables->attr_set, &sg_policy->tunables_hook);
+		if (!have_governor_per_policy())
+			global_tunables = tunables;
+	}
+	return tunables;
+}
+
+static void sugov_tunables_free(struct sugov_tunables *tunables)
+{
+	if (!have_governor_per_policy())
+		global_tunables = NULL;
+
+	kfree(tunables);
+}
+
+static int sugov_init(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy;
+	struct sugov_tunables *tunables;
+	unsigned int lat;
+	int ret = 0;
+
+	/* State should be equivalent to EXIT */
+	if (policy->governor_data)
+		return -EBUSY;
+
+	sg_policy = sugov_policy_alloc(policy);
+	if (!sg_policy)
+		return -ENOMEM;
+
+	mutex_lock(&global_tunables_lock);
+
+	if (global_tunables) {
+		if (WARN_ON(have_governor_per_policy())) {
+			ret = -EINVAL;
+			goto free_sg_policy;
+		}
+		policy->governor_data = sg_policy;
+		sg_policy->tunables = global_tunables;
+
+		gov_attr_set_get(&global_tunables->attr_set, &sg_policy->tunables_hook);
+		goto out;
+	}
+
+	tunables = sugov_tunables_alloc(sg_policy);
+	if (!tunables) {
+		ret = -ENOMEM;
+		goto free_sg_policy;
+	}
+
+	tunables->rate_limit_us = LATENCY_MULTIPLIER;
+	lat = policy->cpuinfo.transition_latency / NSEC_PER_USEC;
+	if (lat)
+		tunables->rate_limit_us *= lat;
+
+	policy->governor_data = sg_policy;
+	sg_policy->tunables = tunables;
+
+	ret = kobject_init_and_add(&tunables->attr_set.kobj, &sugov_tunables_ktype,
+				   get_governor_parent_kobj(policy), "%s",
+				   schedutil_gov.name);
+	if (ret)
+		goto fail;
+
+ out:
+	mutex_unlock(&global_tunables_lock);
+
+	cpufreq_enable_fast_switch(policy);
+	return 0;
+
+ fail:
+	policy->governor_data = NULL;
+	sugov_tunables_free(tunables);
+
+ free_sg_policy:
+	mutex_unlock(&global_tunables_lock);
+
+	sugov_policy_free(sg_policy);
+	pr_err("cpufreq: schedutil governor initialization failed (error %d)\n", ret);
+	return ret;
+}
+
+static int sugov_exit(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+	struct sugov_tunables *tunables = sg_policy->tunables;
+	unsigned int count;
+
+	cpufreq_disable_fast_switch(policy);
+
+	mutex_lock(&global_tunables_lock);
+
+	count = gov_attr_set_put(&tunables->attr_set, &sg_policy->tunables_hook);
+	policy->governor_data = NULL;
+	if (!count)
+		sugov_tunables_free(tunables);
+
+	mutex_unlock(&global_tunables_lock);
+
+	sugov_policy_free(sg_policy);
+	return 0;
+}
+
+static int sugov_start(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+	unsigned int cpu;
+
+	sg_policy->freq_update_delay_ns = sg_policy->tunables->rate_limit_us * NSEC_PER_USEC;
+	sg_policy->last_freq_update_time = 0;
+	sg_policy->next_freq = UINT_MAX;
+	sg_policy->work_in_progress = false;
+	sg_policy->need_freq_update = false;
+
+	for_each_cpu(cpu, policy->cpus) {
+		struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu);
+
+		sg_cpu->sg_policy = sg_policy;
+		if (policy_is_shared(policy)) {
+			sg_cpu->util = ULONG_MAX;
+			sg_cpu->max = 0;
+			sg_cpu->last_update = 0;
+			cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util,
+						     sugov_update_shared);
+		} else {
+			cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util,
+						     sugov_update_single);
+		}
+	}
+	return 0;
+}
+
+static int sugov_stop(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+	unsigned int cpu;
+
+	for_each_cpu(cpu, policy->cpus)
+		cpufreq_remove_update_util_hook(cpu);
+
+	synchronize_sched();
+
+	irq_work_sync(&sg_policy->irq_work);
+	cancel_work_sync(&sg_policy->work);
+	return 0;
+}
+
+static int sugov_limits(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+
+	if (!policy->fast_switch_enabled) {
+		mutex_lock(&sg_policy->work_lock);
+
+		if (policy->max < policy->cur)
+			__cpufreq_driver_target(policy, policy->max,
+						CPUFREQ_RELATION_H);
+		else if (policy->min > policy->cur)
+			__cpufreq_driver_target(policy, policy->min,
+						CPUFREQ_RELATION_L);
+
+		mutex_unlock(&sg_policy->work_lock);
+	}
+
+	sg_policy->need_freq_update = true;
+	return 0;
+}
+
+int sugov_governor(struct cpufreq_policy *policy, unsigned int event)
+{
+	if (event == CPUFREQ_GOV_POLICY_INIT) {
+		return sugov_init(policy);
+	} else if (policy->governor_data) {
+		switch (event) {
+		case CPUFREQ_GOV_POLICY_EXIT:
+			return sugov_exit(policy);
+		case CPUFREQ_GOV_START:
+			return sugov_start(policy);
+		case CPUFREQ_GOV_STOP:
+			return sugov_stop(policy);
+		case CPUFREQ_GOV_LIMITS:
+			return sugov_limits(policy);
+		}
+	}
+	return -EINVAL;
+}
+
+static struct cpufreq_governor schedutil_gov = {
+	.name = "schedutil",
+	.governor = sugov_governor,
+	.owner = THIS_MODULE,
+};
+
+static int __init sugov_module_init(void)
+{
+	return cpufreq_register_governor(&schedutil_gov);
+}
+
+static void __exit sugov_module_exit(void)
+{
+	cpufreq_unregister_governor(&schedutil_gov);
+}
+
+MODULE_AUTHOR("Rafael J. Wysocki <rafael.j.wysocki@intel.com>");
+MODULE_DESCRIPTION("Utilization-based CPU frequency selection");
+MODULE_LICENSE("GPL");
+
+#ifdef CONFIG_CPU_FREQ_DEFAULT_GOV_SCHEDUTIL
+struct cpufreq_governor *cpufreq_default_governor(void)
+{
+	return &schedutil_gov;
+}
+
+fs_initcall(sugov_module_init);
+#else
+module_init(sugov_module_init);
+#endif
+module_exit(sugov_module_exit);

kernel/sched/sched.h

@@ -1842,6 +1842,14 @@ static inline void cpufreq_update_util(u64 time, unsigned long util, unsigned lo
 static inline void cpufreq_trigger_update(u64 time) {}
 #endif /* CONFIG_CPU_FREQ */
 
+#ifdef arch_scale_freq_capacity
+#ifndef arch_scale_freq_invariant
+#define arch_scale_freq_invariant()	(true)
+#endif
+#else /* arch_scale_freq_capacity */
+#define arch_scale_freq_invariant()	(false)
+#endif
+
 static inline void account_reset_rq(struct rq *rq)
 {
 #ifdef CONFIG_IRQ_TIME_ACCOUNTING

kernel/trace/power-traces.c

@@ -15,5 +15,6 @@
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(suspend_resume);
 EXPORT_TRACEPOINT_SYMBOL_GPL(cpu_idle);
+EXPORT_TRACEPOINT_SYMBOL_GPL(cpu_frequency);
 EXPORT_TRACEPOINT_SYMBOL_GPL(powernv_throttle);