cpuidle.c 8.2 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14
/*
 * cpuidle.c - core cpuidle infrastructure
 *
 * (C) 2006-2007 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
 *               Shaohua Li <shaohua.li@intel.com>
 *               Adam Belay <abelay@novell.com>
 *
 * This code is licenced under the GPL.
 */

#include <linux/kernel.h>
#include <linux/mutex.h>
#include <linux/sched.h>
#include <linux/notifier.h>
M
Mark Gross 已提交
15
#include <linux/pm_qos_params.h>
16 17
#include <linux/cpu.h>
#include <linux/cpuidle.h>
18
#include <linux/ktime.h>
19 20 21 22 23 24 25 26 27 28 29

#include "cpuidle.h"

DEFINE_PER_CPU(struct cpuidle_device *, cpuidle_devices);

DEFINE_MUTEX(cpuidle_lock);
LIST_HEAD(cpuidle_detected_devices);
static void (*pm_idle_old)(void);

static int enabled_devices;

V
Venki Pallipadi 已提交
30 31 32 33 34 35 36 37 38 39 40
#if defined(CONFIG_ARCH_HAS_CPU_IDLE_WAIT)
static void cpuidle_kick_cpus(void)
{
	cpu_idle_wait();
}
#elif defined(CONFIG_SMP)
# error "Arch needs cpu_idle_wait() equivalent here"
#else /* !CONFIG_ARCH_HAS_CPU_IDLE_WAIT && !CONFIG_SMP */
static void cpuidle_kick_cpus(void) {}
#endif

41 42
static int __cpuidle_register_device(struct cpuidle_device *dev);

43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
/**
 * cpuidle_idle_call - the main idle loop
 *
 * NOTE: no locks or semaphores should be used here
 */
static void cpuidle_idle_call(void)
{
	struct cpuidle_device *dev = __get_cpu_var(cpuidle_devices);
	struct cpuidle_state *target_state;
	int next_state;

	/* check if the device is ready */
	if (!dev || !dev->enabled) {
		if (pm_idle_old)
			pm_idle_old();
		else
59 60 61
#if defined(CONFIG_ARCH_HAS_DEFAULT_IDLE)
			default_idle();
#else
62
			local_irq_enable();
63
#endif
64 65 66 67 68 69 70 71 72 73 74
		return;
	}

	/* ask the governor for the next state */
	next_state = cpuidle_curr_governor->select(dev);
	if (need_resched())
		return;
	target_state = &dev->states[next_state];

	/* enter the state and update stats */
	dev->last_state = target_state;
75 76 77 78
	dev->last_residency = target_state->enter(dev, target_state);
	if (dev->last_state)
		target_state = dev->last_state;

79
	target_state->time += (unsigned long long)dev->last_residency;
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
	target_state->usage++;

	/* give the governor an opportunity to reflect on the outcome */
	if (cpuidle_curr_governor->reflect)
		cpuidle_curr_governor->reflect(dev);
}

/**
 * cpuidle_install_idle_handler - installs the cpuidle idle loop handler
 */
void cpuidle_install_idle_handler(void)
{
	if (enabled_devices && (pm_idle != cpuidle_idle_call)) {
		/* Make sure all changes finished before we switch to new idle */
		smp_wmb();
		pm_idle = cpuidle_idle_call;
	}
}

/**
 * cpuidle_uninstall_idle_handler - uninstalls the cpuidle idle loop handler
 */
void cpuidle_uninstall_idle_handler(void)
{
104
	if (enabled_devices && pm_idle_old && (pm_idle != pm_idle_old)) {
105
		pm_idle = pm_idle_old;
V
Venki Pallipadi 已提交
106
		cpuidle_kick_cpus();
107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149
	}
}

/**
 * cpuidle_pause_and_lock - temporarily disables CPUIDLE
 */
void cpuidle_pause_and_lock(void)
{
	mutex_lock(&cpuidle_lock);
	cpuidle_uninstall_idle_handler();
}

EXPORT_SYMBOL_GPL(cpuidle_pause_and_lock);

/**
 * cpuidle_resume_and_unlock - resumes CPUIDLE operation
 */
void cpuidle_resume_and_unlock(void)
{
	cpuidle_install_idle_handler();
	mutex_unlock(&cpuidle_lock);
}

EXPORT_SYMBOL_GPL(cpuidle_resume_and_unlock);

/**
 * cpuidle_enable_device - enables idle PM for a CPU
 * @dev: the CPU
 *
 * This function must be called between cpuidle_pause_and_lock and
 * cpuidle_resume_and_unlock when used externally.
 */
int cpuidle_enable_device(struct cpuidle_device *dev)
{
	int ret, i;

	if (dev->enabled)
		return 0;
	if (!cpuidle_curr_driver || !cpuidle_curr_governor)
		return -EIO;
	if (!dev->state_count)
		return -EINVAL;

150 151 152 153 154 155
	if (dev->registered == 0) {
		ret = __cpuidle_register_device(dev);
		if (ret)
			return ret;
	}

156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209
	if ((ret = cpuidle_add_state_sysfs(dev)))
		return ret;

	if (cpuidle_curr_governor->enable &&
	    (ret = cpuidle_curr_governor->enable(dev)))
		goto fail_sysfs;

	for (i = 0; i < dev->state_count; i++) {
		dev->states[i].usage = 0;
		dev->states[i].time = 0;
	}
	dev->last_residency = 0;
	dev->last_state = NULL;

	smp_wmb();

	dev->enabled = 1;

	enabled_devices++;
	return 0;

fail_sysfs:
	cpuidle_remove_state_sysfs(dev);

	return ret;
}

EXPORT_SYMBOL_GPL(cpuidle_enable_device);

/**
 * cpuidle_disable_device - disables idle PM for a CPU
 * @dev: the CPU
 *
 * This function must be called between cpuidle_pause_and_lock and
 * cpuidle_resume_and_unlock when used externally.
 */
void cpuidle_disable_device(struct cpuidle_device *dev)
{
	if (!dev->enabled)
		return;
	if (!cpuidle_curr_driver || !cpuidle_curr_governor)
		return;

	dev->enabled = 0;

	if (cpuidle_curr_governor->disable)
		cpuidle_curr_governor->disable(dev);

	cpuidle_remove_state_sysfs(dev);
	enabled_devices--;
}

EXPORT_SYMBOL_GPL(cpuidle_disable_device);

210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236
#ifdef CONFIG_ARCH_HAS_CPU_RELAX
static int poll_idle(struct cpuidle_device *dev, struct cpuidle_state *st)
{
	ktime_t	t1, t2;
	s64 diff;
	int ret;

	t1 = ktime_get();
	local_irq_enable();
	while (!need_resched())
		cpu_relax();

	t2 = ktime_get();
	diff = ktime_to_us(ktime_sub(t2, t1));
	if (diff > INT_MAX)
		diff = INT_MAX;

	ret = (int) diff;
	return ret;
}

static void poll_idle_init(struct cpuidle_device *dev)
{
	struct cpuidle_state *state = &dev->states[0];

	cpuidle_set_statedata(state, NULL);

237 238
	snprintf(state->name, CPUIDLE_NAME_LEN, "C0");
	snprintf(state->desc, CPUIDLE_DESC_LEN, "CPUIDLE CORE POLL IDLE");
239 240 241
	state->exit_latency = 0;
	state->target_residency = 0;
	state->power_usage = -1;
242
	state->flags = CPUIDLE_FLAG_POLL;
243 244 245 246 247 248
	state->enter = poll_idle;
}
#else
static void poll_idle_init(struct cpuidle_device *dev) {}
#endif /* CONFIG_ARCH_HAS_CPU_RELAX */

249
/**
250 251
 * __cpuidle_register_device - internal register function called before register
 * and enable routines
252
 * @dev: the cpu
253 254
 *
 * cpuidle_lock mutex must be held before this is called
255
 */
256
static int __cpuidle_register_device(struct cpuidle_device *dev)
257 258 259 260 261 262 263 264 265 266 267
{
	int ret;
	struct sys_device *sys_dev = get_cpu_sysdev((unsigned long)dev->cpu);

	if (!sys_dev)
		return -EINVAL;
	if (!try_module_get(cpuidle_curr_driver->owner))
		return -EINVAL;

	init_completion(&dev->kobj_unregister);

268 269
	poll_idle_init(dev);

270 271 272 273 274 275 276
	per_cpu(cpuidle_devices, dev->cpu) = dev;
	list_add(&dev->device_list, &cpuidle_detected_devices);
	if ((ret = cpuidle_add_sysfs(sys_dev))) {
		module_put(cpuidle_curr_driver->owner);
		return ret;
	}

277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295
	dev->registered = 1;
	return 0;
}

/**
 * cpuidle_register_device - registers a CPU's idle PM feature
 * @dev: the cpu
 */
int cpuidle_register_device(struct cpuidle_device *dev)
{
	int ret;

	mutex_lock(&cpuidle_lock);

	if ((ret = __cpuidle_register_device(dev))) {
		mutex_unlock(&cpuidle_lock);
		return ret;
	}

296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
	cpuidle_enable_device(dev);
	cpuidle_install_idle_handler();

	mutex_unlock(&cpuidle_lock);

	return 0;

}

EXPORT_SYMBOL_GPL(cpuidle_register_device);

/**
 * cpuidle_unregister_device - unregisters a CPU's idle PM feature
 * @dev: the cpu
 */
void cpuidle_unregister_device(struct cpuidle_device *dev)
{
	struct sys_device *sys_dev = get_cpu_sysdev((unsigned long)dev->cpu);

315 316 317
	if (dev->registered == 0)
		return;

318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349
	cpuidle_pause_and_lock();

	cpuidle_disable_device(dev);

	cpuidle_remove_sysfs(sys_dev);
	list_del(&dev->device_list);
	wait_for_completion(&dev->kobj_unregister);
	per_cpu(cpuidle_devices, dev->cpu) = NULL;

	cpuidle_resume_and_unlock();

	module_put(cpuidle_curr_driver->owner);
}

EXPORT_SYMBOL_GPL(cpuidle_unregister_device);

#ifdef CONFIG_SMP

static void smp_callback(void *v)
{
	/* we already woke the CPU up, nothing more to do */
}

/*
 * This function gets called when a part of the kernel has a new latency
 * requirement.  This means we need to get all processors out of their C-state,
 * and then recalculate a new suitable C-state. Just do a cross-cpu IPI; that
 * wakes them all right up.
 */
static int cpuidle_latency_notify(struct notifier_block *b,
		unsigned long l, void *v)
{
350
	smp_call_function(smp_callback, NULL, 1);
351 352 353 354 355 356 357
	return NOTIFY_OK;
}

static struct notifier_block cpuidle_latency_notifier = {
	.notifier_call = cpuidle_latency_notify,
};

M
Mark Gross 已提交
358 359 360 361
static inline void latency_notifier_init(struct notifier_block *n)
{
	pm_qos_add_notifier(PM_QOS_CPU_DMA_LATENCY, n);
}
362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387

#else /* CONFIG_SMP */

#define latency_notifier_init(x) do { } while (0)

#endif /* CONFIG_SMP */

/**
 * cpuidle_init - core initializer
 */
static int __init cpuidle_init(void)
{
	int ret;

	pm_idle_old = pm_idle;

	ret = cpuidle_add_class_sysfs(&cpu_sysdev_class);
	if (ret)
		return ret;

	latency_notifier_init(&cpuidle_latency_notifier);

	return 0;
}

core_initcall(cpuidle_init);