aboutsummaryrefslogtreecommitdiff
path: root/arch/x86/kernel/cpu/mcheck/therm_throt.c
blob: 81c499eceb21d88f62499a86cb9b0034d8519858 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
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
150
151
152
153
154
155
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
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
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
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
350
351
352
353
354
355
/*
 * Thermal throttle event support code (such as syslog messaging and rate
 * limiting) that was factored out from x86_64 (mce_intel.c) and i386 (p4.c).
 *
 * This allows consistent reporting of CPU thermal throttle events.
 *
 * Maintains a counter in /sys that keeps track of the number of thermal
 * events, such that the user knows how bad the thermal problem might be
 * (since the logging to syslog and mcelog is rate limited).
 *
 * Author: Dmitriy Zavin (dmitriyz@google.com)
 *
 * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c.
 *          Inspired by Ross Biro's and Al Borchers' counter code.
 */
#include <linux/interrupt.h>
#include <linux/notifier.h>
#include <linux/jiffies.h>
#include <linux/kernel.h>
#include <linux/percpu.h>
#include <linux/sysdev.h>
#include <linux/types.h>
#include <linux/init.h>
#include <linux/smp.h>
#include <linux/cpu.h>

#include <asm/processor.h>
#include <asm/system.h>
#include <asm/apic.h>
#include <asm/idle.h>
#include <asm/mce.h>
#include <asm/msr.h>

/* How long to wait between reporting thermal events */
#define CHECK_INTERVAL		(300 * HZ)

/*
 * Current thermal throttling state:
 */
struct thermal_state {
	bool			is_throttled;

	u64			next_check;
	unsigned long		throttle_count;
	unsigned long		last_throttle_count;
};

static DEFINE_PER_CPU(struct thermal_state, thermal_state);

static atomic_t therm_throt_en	= ATOMIC_INIT(0);

static u32 lvtthmr_init __read_mostly;

#ifdef CONFIG_SYSFS
#define define_therm_throt_sysdev_one_ro(_name)				\
	static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL)

#define define_therm_throt_sysdev_show_func(name)			\
									\
static ssize_t therm_throt_sysdev_show_##name(				\
			struct sys_device *dev,				\
			struct sysdev_attribute *attr,			\
			char *buf)					\
{									\
	unsigned int cpu = dev->id;					\
	ssize_t ret;							\
									\
	preempt_disable();	/* CPU hotplug */			\
	if (cpu_online(cpu))						\
		ret = sprintf(buf, "%lu\n",				\
			      per_cpu(thermal_state, cpu).name);	\
	else								\
		ret = 0;						\
	preempt_enable();						\
									\
	return ret;							\
}

define_therm_throt_sysdev_show_func(throttle_count);
define_therm_throt_sysdev_one_ro(throttle_count);

static struct attribute *thermal_throttle_attrs[] = {
	&attr_throttle_count.attr,
	NULL
};

static struct attribute_group thermal_throttle_attr_group = {
	.attrs	= thermal_throttle_attrs,
	.name	= "thermal_throttle"
};
#endif /* CONFIG_SYSFS */

/***
 * therm_throt_process - Process thermal throttling event from interrupt
 * @curr: Whether the condition is current or not (boolean), since the
 *        thermal interrupt normally gets called both when the thermal
 *        event begins and once the event has ended.
 *
 * This function is called by the thermal interrupt after the
 * IRQ has been acknowledged.
 *
 * It will take care of rate limiting and printing messages to the syslog.
 *
 * Returns: 0 : Event should NOT be further logged, i.e. still in
 *              "timeout" from previous log message.
 *          1 : Event should be logged further, and a message has been
 *              printed to the syslog.
 */
static int therm_throt_process(bool is_throttled)
{
	struct thermal_state *state;
	unsigned int this_cpu;
	bool was_throttled;
	u64 now;

	this_cpu = smp_processor_id();
	now = get_jiffies_64();
	state = &per_cpu(thermal_state, this_cpu);

	was_throttled = state->is_throttled;
	state->is_throttled = is_throttled;

	if (is_throttled)
		state->throttle_count++;

	if (time_before64(now, state->next_check) &&
			state->throttle_count != state->last_throttle_count)
		return 0;

	state->next_check = now + CHECK_INTERVAL;
	state->last_throttle_count = state->throttle_count;

	/* if we just entered the thermal event */
	if (is_throttled) {
		printk(KERN_CRIT "CPU%d: Temperature above threshold, cpu clock throttled (total events = %lu)\n", this_cpu, state->throttle_count);

		add_taint(TAINT_MACHINE_CHECK);
		return 1;
	}
	if (was_throttled) {
		printk(KERN_INFO "CPU%d: Temperature/speed normal\n", this_cpu);
		return 1;
	}

	return 0;
}

#ifdef CONFIG_SYSFS
/* Add/Remove thermal_throttle interface for CPU device: */
static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev)
{
	return sysfs_create_group(&sys_dev->kobj,
				  &thermal_throttle_attr_group);
}

static __cpuinit void thermal_throttle_remove_dev(struct sys_device *sys_dev)
{
	sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group);
}

/* Mutex protecting device creation against CPU hotplug: */
static DEFINE_MUTEX(therm_cpu_lock);

/* Get notified when a cpu comes on/off. Be hotplug friendly. */
static __cpuinit int
thermal_throttle_cpu_callback(struct notifier_block *nfb,
			      unsigned long action,
			      void *hcpu)
{
	unsigned int cpu = (unsigned long)hcpu;
	struct sys_device *sys_dev;
	int err = 0;

	sys_dev = get_cpu_sysdev(cpu);

	switch (action) {
	case CPU_UP_PREPARE:
	case CPU_UP_PREPARE_FROZEN:
		mutex_lock(&therm_cpu_lock);
		err = thermal_throttle_add_dev(sys_dev);
		mutex_unlock(&therm_cpu_lock);
		WARN_ON(err);
		break;
	case CPU_UP_CANCELED:
	case CPU_UP_CANCELED_FROZEN:
	case CPU_DEAD:
	case CPU_DEAD_FROZEN:
		mutex_lock(&therm_cpu_lock);
		thermal_throttle_remove_dev(sys_dev);
		mutex_unlock(&therm_cpu_lock);
		break;
	}
	return err ? NOTIFY_BAD : NOTIFY_OK;
}

static struct notifier_block thermal_throttle_cpu_notifier __cpuinitdata =
{
	.notifier_call = thermal_throttle_cpu_callback,
};

static __init int thermal_throttle_init_device(void)
{
	unsigned int cpu = 0;
	int err;

	if (!atomic_read(&therm_throt_en))
		return 0;

	register_hotcpu_notifier(&thermal_throttle_cpu_notifier);

#ifdef CONFIG_HOTPLUG_CPU
	mutex_lock(&therm_cpu_lock);
#endif
	/* connect live CPUs to sysfs */
	for_each_online_cpu(cpu) {
		err = thermal_throttle_add_dev(get_cpu_sysdev(cpu));
		WARN_ON(err);
	}
#ifdef CONFIG_HOTPLUG_CPU
	mutex_unlock(&therm_cpu_lock);
#endif

	return 0;
}
device_initcall(thermal_throttle_init_device);

#endif /* CONFIG_SYSFS */

/* Thermal transition interrupt handler */
static void intel_thermal_interrupt(void)
{
	__u64 msr_val;

	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
	if (therm_throt_process((msr_val & THERM_STATUS_PROCHOT) != 0))
		mce_log_therm_throt_event(msr_val);
}

static void unexpected_thermal_interrupt(void)
{
	printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n",
			smp_processor_id());
	add_taint(TAINT_MACHINE_CHECK);
}

static void (*smp_thermal_vector)(void) = unexpected_thermal_interrupt;

asmlinkage void smp_thermal_interrupt(struct pt_regs *regs)
{
	exit_idle();
	irq_enter();
	inc_irq_stat(irq_thermal_count);
	smp_thermal_vector();
	irq_exit();
	/* Ack only at the end to avoid potential reentry */
	ack_APIC_irq();
}

/* Thermal monitoring depends on APIC, ACPI and clock modulation */
static int intel_thermal_supported(struct cpuinfo_x86 *c)
{
	if (!cpu_has_apic)
		return 0;
	if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC))
		return 0;
	return 1;
}

void __init mcheck_intel_therm_init(void)
{
	/*
	 * This function is only called on boot CPU. Save the init thermal
	 * LVT value on BSP and use that value to restore APs' thermal LVT
	 * entry BIOS programmed later
	 */
	if (intel_thermal_supported(&boot_cpu_data))
		lvtthmr_init = apic_read(APIC_LVTTHMR);
}

void intel_init_thermal(struct cpuinfo_x86 *c)
{
	unsigned int cpu = smp_processor_id();
	int tm2 = 0;
	u32 l, h;

	if (!intel_thermal_supported(c))
		return;

	/*
	 * First check if its enabled already, in which case there might
	 * be some SMM goo which handles it, so we can't even put a handler
	 * since it might be delivered via SMI already:
	 */
	rdmsr(MSR_IA32_MISC_ENABLE, l, h);

	/*
	 * The initial value of thermal LVT entries on all APs always reads
	 * 0x10000 because APs are woken up by BSP issuing INIT-SIPI-SIPI
	 * sequence to them and LVT registers are reset to 0s except for
	 * the mask bits which are set to 1s when APs receive INIT IPI.
	 * Always restore the value that BIOS has programmed on AP based on
	 * BSP's info we saved since BIOS is always setting the same value
	 * for all threads/cores
	 */
	apic_write(APIC_LVTTHMR, lvtthmr_init);

	h = lvtthmr_init;

	if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) {
		printk(KERN_DEBUG
		       "CPU%d: Thermal monitoring handled by SMI\n", cpu);
		return;
	}

	/* Check whether a vector already exists */
	if (h & APIC_VECTOR_MASK) {
		printk(KERN_DEBUG
		       "CPU%d: Thermal LVT vector (%#x) already installed\n",
		       cpu, (h & APIC_VECTOR_MASK));
		return;
	}

	/* early Pentium M models use different method for enabling TM2 */
	if (cpu_has(c, X86_FEATURE_TM2)) {
		if (c->x86 == 6 && (c->x86_model == 9 || c->x86_model == 13)) {
			rdmsr(MSR_THERM2_CTL, l, h);
			if (l & MSR_THERM2_CTL_TM_SELECT)
				tm2 = 1;
		} else if (l & MSR_IA32_MISC_ENABLE_TM2)
			tm2 = 1;
	}

	/* We'll mask the thermal vector in the lapic till we're ready: */
	h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED;
	apic_write(APIC_LVTTHMR, h);

	rdmsr(MSR_IA32_THERM_INTERRUPT, l, h);
	wrmsr(MSR_IA32_THERM_INTERRUPT,
		l | (THERM_INT_LOW_ENABLE | THERM_INT_HIGH_ENABLE), h);

	smp_thermal_vector = intel_thermal_interrupt;

	rdmsr(MSR_IA32_MISC_ENABLE, l, h);
	wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h);

	/* Unmask the thermal vector: */
	l = apic_read(APIC_LVTTHMR);
	apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED);

	printk_once(KERN_INFO "CPU0: Thermal monitoring enabled (%s)\n",
		       tm2 ? "TM2" : "TM1");

	/* enable thermal throttle processing */
	atomic_set(&therm_throt_en, 1);
}