diff options
Diffstat (limited to 'drivers/cpuidle')
| -rw-r--r-- | drivers/cpuidle/Kconfig | 5 | ||||
| -rw-r--r-- | drivers/cpuidle/Kconfig.arm | 19 | ||||
| -rw-r--r-- | drivers/cpuidle/Kconfig.mips | 17 | ||||
| -rw-r--r-- | drivers/cpuidle/Makefile | 7 | ||||
| -rw-r--r-- | drivers/cpuidle/coupled.c | 4 | ||||
| -rw-r--r-- | drivers/cpuidle/cpuidle-armada-370-xp.c | 93 | ||||
| -rw-r--r-- | drivers/cpuidle/cpuidle-clps711x.c | 64 | ||||
| -rw-r--r-- | drivers/cpuidle/cpuidle-cps.c | 186 | ||||
| -rw-r--r-- | drivers/cpuidle/cpuidle-exynos.c | 99 | ||||
| -rw-r--r-- | drivers/cpuidle/cpuidle-powernv.c | 105 | ||||
| -rw-r--r-- | drivers/cpuidle/cpuidle-pseries.c | 6 | ||||
| -rw-r--r-- | drivers/cpuidle/cpuidle.c | 127 | ||||
| -rw-r--r-- | drivers/cpuidle/driver.c | 9 | ||||
| -rw-r--r-- | drivers/cpuidle/governors/menu.c | 92 | ||||
| -rw-r--r-- | drivers/cpuidle/sysfs.c | 3 |
15 files changed, 736 insertions, 100 deletions
diff --git a/drivers/cpuidle/Kconfig b/drivers/cpuidle/Kconfig index f04e25f6c98..1b96fb91d32 100644 --- a/drivers/cpuidle/Kconfig +++ b/drivers/cpuidle/Kconfig @@ -35,6 +35,11 @@ depends on ARM source "drivers/cpuidle/Kconfig.arm" endmenu +menu "MIPS CPU Idle Drivers" +depends on MIPS +source "drivers/cpuidle/Kconfig.mips" +endmenu + menu "POWERPC CPU Idle Drivers" depends on PPC source "drivers/cpuidle/Kconfig.powerpc" diff --git a/drivers/cpuidle/Kconfig.arm b/drivers/cpuidle/Kconfig.arm index d988948a89a..b6d69e899f5 100644 --- a/drivers/cpuidle/Kconfig.arm +++ b/drivers/cpuidle/Kconfig.arm @@ -1,6 +1,11 @@ # # ARM CPU Idle drivers # +config ARM_ARMADA_370_XP_CPUIDLE + bool "CPU Idle Driver for Armada 370/XP family processors" + depends on ARCH_MVEBU + help + Select this to enable cpuidle on Armada 370/XP processors. config ARM_BIG_LITTLE_CPUIDLE bool "Support for ARM big.LITTLE processors" @@ -13,6 +18,12 @@ config ARM_BIG_LITTLE_CPUIDLE define different C-states for little and big cores through the multiple CPU idle drivers infrastructure. +config ARM_CLPS711X_CPUIDLE + bool "CPU Idle Driver for CLPS711X processors" + depends on ARCH_CLPS711X || COMPILE_TEST + help + Select this to enable cpuidle on Cirrus Logic CLPS711X SOCs. + config ARM_HIGHBANK_CPUIDLE bool "CPU Idle Driver for Calxeda processors" depends on ARM_PSCI @@ -22,7 +33,7 @@ config ARM_HIGHBANK_CPUIDLE config ARM_KIRKWOOD_CPUIDLE bool "CPU Idle Driver for Marvell Kirkwood SoCs" - depends on ARCH_KIRKWOOD + depends on ARCH_KIRKWOOD || MACH_KIRKWOOD help This adds the CPU Idle driver for Marvell Kirkwood SoCs. @@ -44,3 +55,9 @@ config ARM_AT91_CPUIDLE depends on ARCH_AT91 help Select this to enable cpuidle for AT91 processors + +config ARM_EXYNOS_CPUIDLE + bool "Cpu Idle Driver for the Exynos processors" + depends on ARCH_EXYNOS + help + Select this to enable cpuidle for Exynos processors diff --git a/drivers/cpuidle/Kconfig.mips b/drivers/cpuidle/Kconfig.mips new file mode 100644 index 00000000000..0e70ee28a5c --- /dev/null +++ b/drivers/cpuidle/Kconfig.mips @@ -0,0 +1,17 @@ +# +# MIPS CPU Idle Drivers +# +config MIPS_CPS_CPUIDLE + bool "CPU Idle driver for MIPS CPS platforms" + depends on CPU_IDLE + depends on SYS_SUPPORTS_MIPS_CPS + select ARCH_NEEDS_CPU_IDLE_COUPLED if MIPS_MT + select GENERIC_CLOCKEVENTS_BROADCAST if SMP + select MIPS_CPS_PM + default y + help + Select this option to enable processor idle state management + through cpuidle for systems built around the MIPS Coherent + Processing System (CPS) architecture. In order to make use of + the deepest idle states you will need to ensure that you are + also using the CONFIG_MIPS_CPS SMP implementation. diff --git a/drivers/cpuidle/Makefile b/drivers/cpuidle/Makefile index f71ae1b373c..d8bb1ff7256 100644 --- a/drivers/cpuidle/Makefile +++ b/drivers/cpuidle/Makefile @@ -7,12 +7,19 @@ obj-$(CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED) += coupled.o ################################################################################## # ARM SoC drivers +obj-$(CONFIG_ARM_ARMADA_370_XP_CPUIDLE) += cpuidle-armada-370-xp.o obj-$(CONFIG_ARM_BIG_LITTLE_CPUIDLE) += cpuidle-big_little.o +obj-$(CONFIG_ARM_CLPS711X_CPUIDLE) += cpuidle-clps711x.o obj-$(CONFIG_ARM_HIGHBANK_CPUIDLE) += cpuidle-calxeda.o obj-$(CONFIG_ARM_KIRKWOOD_CPUIDLE) += cpuidle-kirkwood.o obj-$(CONFIG_ARM_ZYNQ_CPUIDLE) += cpuidle-zynq.o obj-$(CONFIG_ARM_U8500_CPUIDLE) += cpuidle-ux500.o obj-$(CONFIG_ARM_AT91_CPUIDLE) += cpuidle-at91.o +obj-$(CONFIG_ARM_EXYNOS_CPUIDLE) += cpuidle-exynos.o + +############################################################################### +# MIPS drivers +obj-$(CONFIG_MIPS_CPS_CPUIDLE) += cpuidle-cps.o ############################################################################### # POWERPC drivers diff --git a/drivers/cpuidle/coupled.c b/drivers/cpuidle/coupled.c index e952936418d..73fe2f8d7f9 100644 --- a/drivers/cpuidle/coupled.c +++ b/drivers/cpuidle/coupled.c @@ -159,7 +159,7 @@ void cpuidle_coupled_parallel_barrier(struct cpuidle_device *dev, atomic_t *a) { int n = dev->coupled->online_count; - smp_mb__before_atomic_inc(); + smp_mb__before_atomic(); atomic_inc(a); while (atomic_read(a) < n) @@ -323,7 +323,7 @@ static void cpuidle_coupled_poke(int cpu) struct call_single_data *csd = &per_cpu(cpuidle_coupled_poke_cb, cpu); if (!cpumask_test_and_set_cpu(cpu, &cpuidle_coupled_poke_pending)) - __smp_call_function_single(cpu, csd, 0); + smp_call_function_single_async(cpu, csd); } /** diff --git a/drivers/cpuidle/cpuidle-armada-370-xp.c b/drivers/cpuidle/cpuidle-armada-370-xp.c new file mode 100644 index 00000000000..a5fba0287bf --- /dev/null +++ b/drivers/cpuidle/cpuidle-armada-370-xp.c @@ -0,0 +1,93 @@ +/* + * Marvell Armada 370 and Armada XP SoC cpuidle driver + * + * Copyright (C) 2014 Marvell + * + * Nadav Haklai <nadavh@marvell.com> + * Gregory CLEMENT <gregory.clement@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. + * + * Maintainer: Gregory CLEMENT <gregory.clement@free-electrons.com> + */ + +#include <linux/cpu_pm.h> +#include <linux/cpuidle.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/suspend.h> +#include <linux/platform_device.h> +#include <asm/cpuidle.h> + +#define ARMADA_370_XP_MAX_STATES 3 +#define ARMADA_370_XP_FLAG_DEEP_IDLE 0x10000 + +static int (*armada_370_xp_cpu_suspend)(int); + +static int armada_370_xp_enter_idle(struct cpuidle_device *dev, + struct cpuidle_driver *drv, + int index) +{ + int ret; + bool deepidle = false; + cpu_pm_enter(); + + if (drv->states[index].flags & ARMADA_370_XP_FLAG_DEEP_IDLE) + deepidle = true; + + ret = armada_370_xp_cpu_suspend(deepidle); + if (ret) + return ret; + + cpu_pm_exit(); + + return index; +} + +static struct cpuidle_driver armada_370_xp_idle_driver = { + .name = "armada_370_xp_idle", + .states[0] = ARM_CPUIDLE_WFI_STATE, + .states[1] = { + .enter = armada_370_xp_enter_idle, + .exit_latency = 10, + .power_usage = 50, + .target_residency = 100, + .flags = CPUIDLE_FLAG_TIME_VALID, + .name = "Idle", + .desc = "CPU power down", + }, + .states[2] = { + .enter = armada_370_xp_enter_idle, + .exit_latency = 100, + .power_usage = 5, + .target_residency = 1000, + .flags = CPUIDLE_FLAG_TIME_VALID | + ARMADA_370_XP_FLAG_DEEP_IDLE, + .name = "Deep idle", + .desc = "CPU and L2 Fabric power down", + }, + .state_count = ARMADA_370_XP_MAX_STATES, +}; + +static int armada_370_xp_cpuidle_probe(struct platform_device *pdev) +{ + + armada_370_xp_cpu_suspend = (void *)(pdev->dev.platform_data); + return cpuidle_register(&armada_370_xp_idle_driver, NULL); +} + +static struct platform_driver armada_370_xp_cpuidle_plat_driver = { + .driver = { + .name = "cpuidle-armada-370-xp", + .owner = THIS_MODULE, + }, + .probe = armada_370_xp_cpuidle_probe, +}; + +module_platform_driver(armada_370_xp_cpuidle_plat_driver); + +MODULE_AUTHOR("Gregory CLEMENT <gregory.clement@free-electrons.com>"); +MODULE_DESCRIPTION("Armada 370/XP cpu idle driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/cpuidle/cpuidle-clps711x.c b/drivers/cpuidle/cpuidle-clps711x.c new file mode 100644 index 00000000000..5243811daa6 --- /dev/null +++ b/drivers/cpuidle/cpuidle-clps711x.c @@ -0,0 +1,64 @@ +/* + * CLPS711X CPU idle driver + * + * Copyright (C) 2014 Alexander Shiyan <shc_work@mail.ru> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/cpuidle.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/platform_device.h> + +#define CLPS711X_CPUIDLE_NAME "clps711x-cpuidle" + +static void __iomem *clps711x_halt; + +static int clps711x_cpuidle_halt(struct cpuidle_device *dev, + struct cpuidle_driver *drv, int index) +{ + writel(0xaa, clps711x_halt); + + return index; +} + +static struct cpuidle_driver clps711x_idle_driver = { + .name = CLPS711X_CPUIDLE_NAME, + .owner = THIS_MODULE, + .states[0] = { + .name = "HALT", + .desc = "CLPS711X HALT", + .enter = clps711x_cpuidle_halt, + .exit_latency = 1, + }, + .state_count = 1, +}; + +static int __init clps711x_cpuidle_probe(struct platform_device *pdev) +{ + struct resource *res; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + clps711x_halt = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(clps711x_halt)) + return PTR_ERR(clps711x_halt); + + return cpuidle_register(&clps711x_idle_driver, NULL); +} + +static struct platform_driver clps711x_cpuidle_driver = { + .driver = { + .name = CLPS711X_CPUIDLE_NAME, + .owner = THIS_MODULE, + }, +}; +module_platform_driver_probe(clps711x_cpuidle_driver, clps711x_cpuidle_probe); + +MODULE_AUTHOR("Alexander Shiyan <shc_work@mail.ru>"); +MODULE_DESCRIPTION("CLPS711X CPU idle driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/cpuidle/cpuidle-cps.c b/drivers/cpuidle/cpuidle-cps.c new file mode 100644 index 00000000000..fc7b62720de --- /dev/null +++ b/drivers/cpuidle/cpuidle-cps.c @@ -0,0 +1,186 @@ +/* + * Copyright (C) 2014 Imagination Technologies + * Author: Paul Burton <paul.burton@imgtec.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#include <linux/cpu_pm.h> +#include <linux/cpuidle.h> +#include <linux/init.h> + +#include <asm/idle.h> +#include <asm/pm-cps.h> + +/* Enumeration of the various idle states this driver may enter */ +enum cps_idle_state { + STATE_WAIT = 0, /* MIPS wait instruction, coherent */ + STATE_NC_WAIT, /* MIPS wait instruction, non-coherent */ + STATE_CLOCK_GATED, /* Core clock gated */ + STATE_POWER_GATED, /* Core power gated */ + STATE_COUNT +}; + +static int cps_nc_enter(struct cpuidle_device *dev, + struct cpuidle_driver *drv, int index) +{ + enum cps_pm_state pm_state; + int err; + + /* + * At least one core must remain powered up & clocked in order for the + * system to have any hope of functioning. + * + * TODO: don't treat core 0 specially, just prevent the final core + * TODO: remap interrupt affinity temporarily + */ + if (!cpu_data[dev->cpu].core && (index > STATE_NC_WAIT)) + index = STATE_NC_WAIT; + + /* Select the appropriate cps_pm_state */ + switch (index) { + case STATE_NC_WAIT: + pm_state = CPS_PM_NC_WAIT; + break; + case STATE_CLOCK_GATED: + pm_state = CPS_PM_CLOCK_GATED; + break; + case STATE_POWER_GATED: + pm_state = CPS_PM_POWER_GATED; + break; + default: + BUG(); + return -EINVAL; + } + + /* Notify listeners the CPU is about to power down */ + if ((pm_state == CPS_PM_POWER_GATED) && cpu_pm_enter()) + return -EINTR; + + /* Enter that state */ + err = cps_pm_enter_state(pm_state); + + /* Notify listeners the CPU is back up */ + if (pm_state == CPS_PM_POWER_GATED) + cpu_pm_exit(); + + return err ?: index; +} + +static struct cpuidle_driver cps_driver = { + .name = "cpc_cpuidle", + .owner = THIS_MODULE, + .states = { + [STATE_WAIT] = MIPS_CPUIDLE_WAIT_STATE, + [STATE_NC_WAIT] = { + .enter = cps_nc_enter, + .exit_latency = 200, + .target_residency = 450, + .flags = CPUIDLE_FLAG_TIME_VALID, + .name = "nc-wait", + .desc = "non-coherent MIPS wait", + }, + [STATE_CLOCK_GATED] = { + .enter = cps_nc_enter, + .exit_latency = 300, + .target_residency = 700, + .flags = CPUIDLE_FLAG_TIME_VALID | + CPUIDLE_FLAG_TIMER_STOP, + .name = "clock-gated", + .desc = "core clock gated", + }, + [STATE_POWER_GATED] = { + .enter = cps_nc_enter, + .exit_latency = 600, + .target_residency = 1000, + .flags = CPUIDLE_FLAG_TIME_VALID | + CPUIDLE_FLAG_TIMER_STOP, + .name = "power-gated", + .desc = "core power gated", + }, + }, + .state_count = STATE_COUNT, + .safe_state_index = 0, +}; + +static void __init cps_cpuidle_unregister(void) +{ + int cpu; + struct cpuidle_device *device; + + for_each_possible_cpu(cpu) { + device = &per_cpu(cpuidle_dev, cpu); + cpuidle_unregister_device(device); + } + + cpuidle_unregister_driver(&cps_driver); +} + +static int __init cps_cpuidle_init(void) +{ + int err, cpu, core, i; + struct cpuidle_device *device; + + /* Detect supported states */ + if (!cps_pm_support_state(CPS_PM_POWER_GATED)) + cps_driver.state_count = STATE_CLOCK_GATED + 1; + if (!cps_pm_support_state(CPS_PM_CLOCK_GATED)) + cps_driver.state_count = STATE_NC_WAIT + 1; + if (!cps_pm_support_state(CPS_PM_NC_WAIT)) + cps_driver.state_count = STATE_WAIT + 1; + + /* Inform the user if some states are unavailable */ + if (cps_driver.state_count < STATE_COUNT) { + pr_info("cpuidle-cps: limited to "); + switch (cps_driver.state_count - 1) { + case STATE_WAIT: + pr_cont("coherent wait\n"); + break; + case STATE_NC_WAIT: + pr_cont("non-coherent wait\n"); + break; + case STATE_CLOCK_GATED: + pr_cont("clock gating\n"); + break; + } + } + + /* + * Set the coupled flag on the appropriate states if this system + * requires it. + */ + if (coupled_coherence) + for (i = STATE_NC_WAIT; i < cps_driver.state_count; i++) + cps_driver.states[i].flags |= CPUIDLE_FLAG_COUPLED; + + err = cpuidle_register_driver(&cps_driver); + if (err) { + pr_err("Failed to register CPS cpuidle driver\n"); + return err; + } + + for_each_possible_cpu(cpu) { + core = cpu_data[cpu].core; + device = &per_cpu(cpuidle_dev, cpu); + device->cpu = cpu; +#ifdef CONFIG_MIPS_MT + cpumask_copy(&device->coupled_cpus, &cpu_sibling_map[cpu]); +#endif + + err = cpuidle_register_device(device); + if (err) { + pr_err("Failed to register CPU%d cpuidle device\n", + cpu); + goto err_out; + } + } + + return 0; +err_out: + cps_cpuidle_unregister(); + return err; +} +device_initcall(cps_cpuidle_init); diff --git a/drivers/cpuidle/cpuidle-exynos.c b/drivers/cpuidle/cpuidle-exynos.c new file mode 100644 index 00000000000..7c015126382 --- /dev/null +++ b/drivers/cpuidle/cpuidle-exynos.c @@ -0,0 +1,99 @@ +/* linux/arch/arm/mach-exynos/cpuidle.c + * + * Copyright (c) 2011 Samsung Electronics Co., Ltd. + * http://www.samsung.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/cpuidle.h> +#include <linux/cpu_pm.h> +#include <linux/export.h> +#include <linux/module.h> +#include <linux/platform_device.h> + +#include <asm/proc-fns.h> +#include <asm/suspend.h> +#include <asm/cpuidle.h> + +static void (*exynos_enter_aftr)(void); + +static int idle_finisher(unsigned long flags) +{ + exynos_enter_aftr(); + cpu_do_idle(); + + return 1; +} + +static int exynos_enter_core0_aftr(struct cpuidle_device *dev, + struct cpuidle_driver *drv, + int index) +{ + cpu_pm_enter(); + cpu_suspend(0, idle_finisher); + cpu_pm_exit(); + + return index; +} + +static int exynos_enter_lowpower(struct cpuidle_device *dev, + struct cpuidle_driver *drv, + int index) +{ + int new_index = index; + + /* AFTR can only be entered when cores other than CPU0 are offline */ + if (num_online_cpus() > 1 || dev->cpu != 0) + new_index = drv->safe_state_index; + + if (new_index == 0) + return arm_cpuidle_simple_enter(dev, drv, new_index); + else + return exynos_enter_core0_aftr(dev, drv, new_index); +} + +static struct cpuidle_driver exynos_idle_driver = { + .name = "exynos_idle", + .owner = THIS_MODULE, + .states = { + [0] = ARM_CPUIDLE_WFI_STATE, + [1] = { + .enter = exynos_enter_lowpower, + .exit_latency = 300, + .target_residency = 100000, + .flags = CPUIDLE_FLAG_TIME_VALID, + .name = "C1", + .desc = "ARM power down", + }, + }, + .state_count = 2, + .safe_state_index = 0, +}; + +static int exynos_cpuidle_probe(struct platform_device *pdev) +{ + int ret; + + exynos_enter_aftr = (void *)(pdev->dev.platform_data); + + ret = cpuidle_register(&exynos_idle_driver, NULL); + if (ret) { + dev_err(&pdev->dev, "failed to register cpuidle driver\n"); + return ret; + } + + return 0; +} + +static struct platform_driver exynos_cpuidle_driver = { + .probe = exynos_cpuidle_probe, + .driver = { + .name = "exynos_cpuidle", + .owner = THIS_MODULE, + }, +}; + +module_platform_driver(exynos_cpuidle_driver); diff --git a/drivers/cpuidle/cpuidle-powernv.c b/drivers/cpuidle/cpuidle-powernv.c index 78fd174c57e..74f5788d50b 100644 --- a/drivers/cpuidle/cpuidle-powernv.c +++ b/drivers/cpuidle/cpuidle-powernv.c @@ -11,9 +11,18 @@ #include <linux/cpuidle.h> #include <linux/cpu.h> #include <linux/notifier.h> +#include <linux/clockchips.h> +#include <linux/of.h> #include <asm/machdep.h> #include <asm/firmware.h> +#include <asm/runlatch.h> + +/* Flags and constants used in PowerNV platform */ + +#define MAX_POWERNV_IDLE_STATES 8 +#define IDLE_USE_INST_NAP 0x00010000 /* Use nap instruction */ +#define IDLE_USE_INST_SLEEP 0x00020000 /* Use sleep instruction */ struct cpuidle_driver powernv_idle_driver = { .name = "powernv_idle", @@ -30,12 +39,14 @@ static int snooze_loop(struct cpuidle_device *dev, local_irq_enable(); set_thread_flag(TIF_POLLING_NRFLAG); + ppc64_runlatch_off(); while (!need_resched()) { HMT_low(); HMT_very_low(); } HMT_medium(); + ppc64_runlatch_on(); clear_thread_flag(TIF_POLLING_NRFLAG); smp_mb(); return index; @@ -45,14 +56,40 @@ static int nap_loop(struct cpuidle_device *dev, struct cpuidle_driver *drv, int index) { + ppc64_runlatch_off(); power7_idle(); + ppc64_runlatch_on(); + return index; +} + +static int fastsleep_loop(struct cpuidle_device *dev, + struct cpuidle_driver *drv, + int index) +{ + unsigned long old_lpcr = mfspr(SPRN_LPCR); + unsigned long new_lpcr; + + if (unlikely(system_state < SYSTEM_RUNNING)) + return index; + + new_lpcr = old_lpcr; + /* Do not exit powersave upon decrementer as we've setup the timer + * offload. + */ + new_lpcr &= ~LPCR_PECE1; + + mtspr(SPRN_LPCR, new_lpcr); + power7_sleep(); + + mtspr(SPRN_LPCR, old_lpcr); + return index; } /* * States for dedicated partition case. */ -static struct cpuidle_state powernv_states[] = { +static struct cpuidle_state powernv_states[MAX_POWERNV_IDLE_STATES] = { { /* Snooze */ .name = "snooze", .desc = "snooze", @@ -60,13 +97,6 @@ static struct cpuidle_state powernv_states[] = { .exit_latency = 0, .target_residency = 0, .enter = &snooze_loop }, - { /* NAP */ - .name = "NAP", - .desc = "NAP", - .flags = CPUIDLE_FLAG_TIME_VALID, - .exit_latency = 10, - .target_residency = 100, - .enter = &nap_loop }, }; static int powernv_cpuidle_add_cpu_notifier(struct notifier_block *n, @@ -127,19 +157,74 @@ static int powernv_cpuidle_driver_init(void) return 0; } +static int powernv_add_idle_states(void) +{ + struct device_node *power_mgt; + struct property *prop; + int nr_idle_states = 1; /* Snooze */ + int dt_idle_states; + u32 *flags; + int i; + + /* Currently we have snooze statically defined */ + + power_mgt = of_find_node_by_path("/ibm,opal/power-mgt"); + if (!power_mgt) { + pr_warn("opal: PowerMgmt Node not found\n"); + return nr_idle_states; + } + + prop = of_find_property(power_mgt, "ibm,cpu-idle-state-flags", NULL); + if (!prop) { + pr_warn("DT-PowerMgmt: missing ibm,cpu-idle-state-flags\n"); + return nr_idle_states; + } + + dt_idle_states = prop->length / sizeof(u32); + flags = (u32 *) prop->value; + + for (i = 0; i < dt_idle_states; i++) { + + if (flags[i] & IDLE_USE_INST_NAP) { + /* Add NAP state */ + strcpy(powernv_states[nr_idle_states].name, "Nap"); + strcpy(powernv_states[nr_idle_states].desc, "Nap"); + powernv_states[nr_idle_states].flags = CPUIDLE_FLAG_TIME_VALID; + powernv_states[nr_idle_states].exit_latency = 10; + powernv_states[nr_idle_states].target_residency = 100; + powernv_states[nr_idle_states].enter = &nap_loop; + nr_idle_states++; + } + + if (flags[i] & IDLE_USE_INST_SLEEP) { + /* Add FASTSLEEP state */ + strcpy(powernv_states[nr_idle_states].name, "FastSleep"); + strcpy(powernv_states[nr_idle_states].desc, "FastSleep"); + powernv_states[nr_idle_states].flags = + CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TIMER_STOP; + powernv_states[nr_idle_states].exit_latency = 300; + powernv_states[nr_idle_states].target_residency = 1000000; + powernv_states[nr_idle_states].enter = &fastsleep_loop; + nr_idle_states++; + } + } + + return nr_idle_states; +} + /* * powernv_idle_probe() * Choose state table for shared versus dedicated partition */ static int powernv_idle_probe(void) { - if (cpuidle_disable != IDLE_NO_OVERRIDE) return -ENODEV; if (firmware_has_feature(FW_FEATURE_OPALv3)) { cpuidle_state_table = powernv_states; - max_idle_state = ARRAY_SIZE(powernv_states); + /* Device tree can indicate more idle states */ + max_idle_state = powernv_add_idle_states(); } else return -ENODEV; diff --git a/drivers/cpuidle/cpuidle-pseries.c b/drivers/cpuidle/cpuidle-pseries.c index 7ab564aa0b1..6f7b0195688 100644 --- a/drivers/cpuidle/cpuidle-pseries.c +++ b/drivers/cpuidle/cpuidle-pseries.c @@ -17,6 +17,7 @@ #include <asm/reg.h> #include <asm/machdep.h> #include <asm/firmware.h> +#include <asm/runlatch.h> #include <asm/plpar_wrappers.h> struct cpuidle_driver pseries_idle_driver = { @@ -29,6 +30,7 @@ static struct cpuidle_state *cpuidle_state_table; static inline void idle_loop_prolog(unsigned long *in_purr) { + ppc64_runlatch_off(); *in_purr = mfspr(SPRN_PURR); /* * Indicate to the HV that we are idle. Now would be @@ -45,6 +47,10 @@ static inline void idle_loop_epilog(unsigned long in_purr) wait_cycles += mfspr(SPRN_PURR) - in_purr; get_lppaca()->wait_state_cycles = cpu_to_be64(wait_cycles); get_lppaca()->idle = 0; + + if (irqs_disabled()) + local_irq_enable(); + ppc64_runlatch_on(); } static int snooze_loop(struct cpuidle_device *dev, diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index a55e68f2cfc..cb7019977c5 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -32,6 +32,7 @@ LIST_HEAD(cpuidle_detected_devices); static int enabled_devices; static int off __read_mostly; static int initialized __read_mostly; +static bool use_deepest_state __read_mostly; int cpuidle_disabled(void) { @@ -65,6 +66,45 @@ int cpuidle_play_dead(void) } /** + * cpuidle_use_deepest_state - Enable/disable the "deepest idle" mode. + * @enable: Whether enable or disable the feature. + * + * If the "deepest idle" mode is enabled, cpuidle will ignore the governor and + * always use the state with the greatest exit latency (out of the states that + * are not disabled). + * + * This function can only be called after cpuidle_pause() to avoid races. + */ +void cpuidle_use_deepest_state(bool enable) +{ + use_deepest_state = enable; +} + +/** + * cpuidle_find_deepest_state - Find the state of the greatest exit latency. + * @drv: cpuidle driver for a given CPU. + * @dev: cpuidle device for a given CPU. + */ +static int cpuidle_find_deepest_state(struct cpuidle_driver *drv, + struct cpuidle_device *dev) +{ + unsigned int latency_req = 0; + int i, ret = CPUIDLE_DRIVER_STATE_START - 1; + + for (i = CPUIDLE_DRIVER_STATE_START; i < drv->state_count; i++) { + struct cpuidle_state *s = &drv->states[i]; + struct cpuidle_state_usage *su = &dev->states_usage[i]; + + if (s->disabled || su->disable || s->exit_latency <= latency_req) + continue; + + latency_req = s->exit_latency; + ret = i; + } + return ret; +} + +/** * cpuidle_enter_state - enter the state and update stats * @dev: cpuidle device for this cpu * @drv: cpuidle driver for this cpu @@ -85,7 +125,8 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv, time_end = ktime_get(); - local_irq_enable(); + if (!cpuidle_state_is_coupled(dev, drv, entered_state)) + local_irq_enable(); diff = ktime_to_us(ktime_sub(time_end, time_start)); if (diff > INT_MAX) @@ -108,61 +149,57 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv, } /** - * cpuidle_idle_call - the main idle loop + * cpuidle_select - ask the cpuidle framework to choose an idle state * - * NOTE: no locks or semaphores should be used here - * return non-zero on failure + * @drv: the cpuidle driver + * @dev: the cpuidle device + * + * Returns the index of the idle state. */ -int cpuidle_idle_call(void) +int cpuidle_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) { - struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices); - struct cpuidle_driver *drv; - int next_state, entered_state; - bool broadcast; - if (off || !initialized) return -ENODEV; - /* check if the device is ready */ - if (!dev || !dev->enabled) + if (!drv || !dev || !dev->enabled) return -EBUSY; - drv = cpuidle_get_cpu_driver(dev); + if (unlikely(use_deepest_state)) + return cpuidle_find_deepest_state(drv, dev); - /* ask the governor for the next state */ - next_state = cpuidle_curr_governor->select(drv, dev); - if (need_resched()) { - dev->last_residency = 0; - /* give the governor an opportunity to reflect on the outcome */ - if (cpuidle_curr_governor->reflect) - cpuidle_curr_governor->reflect(dev, next_state); - local_irq_enable(); - return 0; - } - - trace_cpu_idle_rcuidle(next_state, dev->cpu); - - broadcast = !!(drv->states[next_state].flags & CPUIDLE_FLAG_TIMER_STOP); - - if (broadcast) - clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &dev->cpu); - - if (cpuidle_state_is_coupled(dev, drv, next_state)) - entered_state = cpuidle_enter_state_coupled(dev, drv, - next_state); - else - entered_state = cpuidle_enter_state(dev, drv, next_state); - - if (broadcast) - clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, &dev->cpu); - - trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, dev->cpu); + return cpuidle_curr_governor->select(drv, dev); +} - /* give the governor an opportunity to reflect on the outcome */ - if (cpuidle_curr_governor->reflect) - cpuidle_curr_governor->reflect(dev, entered_state); +/** + * cpuidle_enter - enter into the specified idle state + * + * @drv: the cpuidle driver tied with the cpu + * @dev: the cpuidle device + * @index: the index in the idle state table + * + * Returns the index in the idle state, < 0 in case of error. + * The error code depends on the backend driver + */ +int cpuidle_enter(struct cpuidle_driver *drv, struct cpuidle_device *dev, + int index) +{ + if (cpuidle_state_is_coupled(dev, drv, index)) + return cpuidle_enter_state_coupled(dev, drv, index); + return cpuidle_enter_state(dev, drv, index); +} - return 0; +/** + * cpuidle_reflect - tell the underlying governor what was the state + * we were in + * + * @dev : the cpuidle device + * @index: the index in the idle state table + * + */ +void cpuidle_reflect(struct cpuidle_device *dev, int index) +{ + if (cpuidle_curr_governor->reflect && !unlikely(use_deepest_state)) + cpuidle_curr_governor->reflect(dev, index); } /** diff --git a/drivers/cpuidle/driver.c b/drivers/cpuidle/driver.c index 06dbe7c8619..9634f20e392 100644 --- a/drivers/cpuidle/driver.c +++ b/drivers/cpuidle/driver.c @@ -187,8 +187,11 @@ static int poll_idle(struct cpuidle_device *dev, t1 = ktime_get(); local_irq_enable(); - while (!need_resched()) - cpu_relax(); + if (!current_set_polling_and_test()) { + while (!need_resched()) + cpu_relax(); + } + current_clr_polling(); t2 = ktime_get(); diff = ktime_to_us(ktime_sub(t2, t1)); @@ -209,7 +212,7 @@ static void poll_idle_init(struct cpuidle_driver *drv) state->exit_latency = 0; state->target_residency = 0; state->power_usage = -1; - state->flags = 0; + state->flags = CPUIDLE_FLAG_TIME_VALID; state->enter = poll_idle; state->disabled = false; } diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index cf7f2f0e4ef..c4f80c15a48 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -122,9 +122,8 @@ struct menu_device { int last_state_idx; int needs_update; - unsigned int expected_us; + unsigned int next_timer_us; unsigned int predicted_us; - unsigned int exit_us; unsigned int bucket; unsigned int correction_factor[BUCKETS]; unsigned int intervals[INTERVALS]; @@ -257,7 +256,7 @@ again: stddev = int_sqrt(stddev); if (((avg > stddev * 6) && (divisor * 4 >= INTERVALS * 3)) || stddev <= 20) { - if (data->expected_us > avg) + if (data->next_timer_us > avg) data->predicted_us = avg; return; } @@ -289,7 +288,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) struct menu_device *data = &__get_cpu_var(menu_devices); int latency_req = pm_qos_request(PM_QOS_CPU_DMA_LATENCY); int i; - int multiplier; + unsigned int interactivity_req; struct timespec t; if (data->needs_update) { @@ -297,8 +296,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) data->needs_update = 0; } - data->last_state_idx = 0; - data->exit_us = 0; + data->last_state_idx = CPUIDLE_DRIVER_STATE_START - 1; /* Special case when user has set very strict latency requirement */ if (unlikely(latency_req == 0)) @@ -306,37 +304,37 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) /* determine the expected residency time, round up */ t = ktime_to_timespec(tick_nohz_get_sleep_length()); - data->expected_us = + data->next_timer_us = t.tv_sec * USEC_PER_SEC + t.tv_nsec / NSEC_PER_USEC; - data->bucket = which_bucket(data->expected_us); - - multiplier = performance_multiplier(); - - /* - * if the correction factor is 0 (eg first time init or cpu hotplug - * etc), we actually want to start out with a unity factor. - */ - if (data->correction_factor[data->bucket] == 0) - data->correction_factor[data->bucket] = RESOLUTION * DECAY; + data->bucket = which_bucket(data->next_timer_us); /* * Force the result of multiplication to be 64 bits even if both * operands are 32 bits. * Make sure to round up for half microseconds. */ - data->predicted_us = div_round64((uint64_t)data->expected_us * + data->predicted_us = div_round64((uint64_t)data->next_timer_us * data->correction_factor[data->bucket], RESOLUTION * DECAY); get_typical_interval(data); /* + * Performance multiplier defines a minimum predicted idle + * duration / latency ratio. Adjust the latency limit if + * necessary. + */ + interactivity_req = data->predicted_us / performance_multiplier(); + if (latency_req > interactivity_req) + latency_req = interactivity_req; + + /* * We want to default to C1 (hlt), not to busy polling * unless the timer is happening really really soon. */ - if (data->expected_us > 5 && + if (data->next_timer_us > 5 && !drv->states[CPUIDLE_DRIVER_STATE_START].disabled && dev->states_usage[CPUIDLE_DRIVER_STATE_START].disable == 0) data->last_state_idx = CPUIDLE_DRIVER_STATE_START; @@ -355,11 +353,8 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev) continue; if (s->exit_latency > latency_req) continue; - if (s->exit_latency * multiplier > data->predicted_us) - continue; data->last_state_idx = i; - data->exit_us = s->exit_latency; } return data->last_state_idx; @@ -390,36 +385,47 @@ static void menu_update(struct cpuidle_driver *drv, struct cpuidle_device *dev) { struct menu_device *data = &__get_cpu_var(menu_devices); int last_idx = data->last_state_idx; - unsigned int last_idle_us = cpuidle_get_last_residency(dev); struct cpuidle_state *target = &drv->states[last_idx]; unsigned int measured_us; unsigned int new_factor; /* - * Ugh, this idle state doesn't support residency measurements, so we - * are basically lost in the dark. As a compromise, assume we slept - * for the whole expected time. + * Try to figure out how much time passed between entry to low + * power state and occurrence of the wakeup event. + * + * If the entered idle state didn't support residency measurements, + * we are basically lost in the dark how much time passed. + * As a compromise, assume we slept for the whole expected time. + * + * Any measured amount of time will include the exit latency. + * Since we are interested in when the wakeup begun, not when it + * was completed, we must substract the exit latency. However, if + * the measured amount of time is less than the exit latency, + * assume the state was never reached and the exit latency is 0. */ - if (unlikely(!(target->flags & CPUIDLE_FLAG_TIME_VALID))) - last_idle_us = data->expected_us; - + if (unlikely(!(target->flags & CPUIDLE_FLAG_TIME_VALID))) { + /* Use timer value as is */ + measured_us = data->next_timer_us; - measured_us = last_idle_us; + } else { + /* Use measured value */ + measured_us = cpuidle_get_last_residency(dev); - /* - * We correct for the exit latency; we are assuming here that the - * exit latency happens after the event that we're interested in. - */ - if (measured_us > data->exit_us) - measured_us -= data->exit_us; + /* Deduct exit latency */ + if (measured_us > target->exit_latency) + measured_us -= target->exit_latency; + /* Make sure our coefficients do not exceed unity */ + if (measured_us > data->next_timer_us) + measured_us = data->next_timer_us; + } /* Update our correction ratio */ new_factor = data->correction_factor[data->bucket]; new_factor -= new_factor / DECAY; - if (data->expected_us > 0 && measured_us < MAX_INTERESTING) - new_factor += RESOLUTION * measured_us / data->expected_us; + if (data->next_timer_us > 0 && measured_us < MAX_INTERESTING) + new_factor += RESOLUTION * measured_us / data->next_timer_us; else /* * we were idle so long that we count it as a perfect @@ -439,7 +445,7 @@ static void menu_update(struct cpuidle_driver *drv, struct cpuidle_device *dev) data->correction_factor[data->bucket] = new_factor; /* update the repeating-pattern data */ - data->intervals[data->interval_ptr++] = last_idle_us; + data->intervals[data->interval_ptr++] = measured_us; if (data->interval_ptr >= INTERVALS) data->interval_ptr = 0; } @@ -453,9 +459,17 @@ static int menu_enable_device(struct cpuidle_driver *drv, struct cpuidle_device *dev) { struct menu_device *data = &per_cpu(menu_devices, dev->cpu); + int i; memset(data, 0, sizeof(struct menu_device)); + /* + * if the correction factor is 0 (eg first time init or cpu hotplug + * etc), we actually want to start out with a unity factor. + */ + for(i = 0; i < BUCKETS; i++) + data->correction_factor[i] = RESOLUTION * DECAY; + return 0; } diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c index e918b6d0caf..efe2f175168 100644 --- a/drivers/cpuidle/sysfs.c +++ b/drivers/cpuidle/sysfs.c @@ -293,6 +293,7 @@ static ssize_t show_state_##_name(struct cpuidle_state *state, \ } define_show_state_function(exit_latency) +define_show_state_function(target_residency) define_show_state_function(power_usage) define_show_state_ull_function(usage) define_show_state_ull_function(time) @@ -304,6 +305,7 @@ define_store_state_ull_function(disable) define_one_state_ro(name, show_state_name); define_one_state_ro(desc, show_state_desc); define_one_state_ro(latency, show_state_exit_latency); +define_one_state_ro(residency, show_state_target_residency); define_one_state_ro(power, show_state_power_usage); define_one_state_ro(usage, show_state_usage); define_one_state_ro(time, show_state_time); @@ -313,6 +315,7 @@ static struct attribute *cpuidle_state_default_attrs[] = { &attr_name.attr, &attr_desc.attr, &attr_latency.attr, + &attr_residency.attr, &attr_power.attr, &attr_usage.attr, &attr_time.attr, |
