aboutsummaryrefslogtreecommitdiffstats
path: root/kernel/irq/manage.c
blob: 46d6611a33bbe4a7997637168feb543acd0f0b8e (plain) (blame)
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
356
357
358
359
360
361
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
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
/*
 * linux/kernel/irq/manage.c
 *
 * Copyright (C) 1992, 1998-2006 Linus Torvalds, Ingo Molnar
 * Copyright (C) 2005-2006 Thomas Gleixner
 *
 * This file contains driver APIs to the irq subsystem.
 */

#include <linux/irq.h>
#include <linux/module.h>
#include <linux/random.h>
#include <linux/interrupt.h>
#include <linux/slab.h>

#include "internals.h"

#ifdef CONFIG_SMP

/**
 *	synchronize_irq - wait for pending IRQ handlers (on other CPUs)
 *	@irq: interrupt number to wait for
 *
 *	This function waits for any pending IRQ handlers for this interrupt
 *	to complete before returning. If you use this function while
 *	holding a resource the IRQ handler may need you will deadlock.
 *
 *	This function may be called - with care - from IRQ context.
 */
void synchronize_irq(unsigned int irq)
{
	struct irq_desc *desc = irq_desc + irq;
	unsigned int status;

	if (irq >= NR_IRQS)
		return;

	do {
		unsigned long flags;

		/*
		 * Wait until we're out of the critical section.  This might
		 * give the wrong answer due to the lack of memory barriers.
		 */
		while (desc->status & IRQ_INPROGRESS)
			cpu_relax();

		/* Ok, that indicated we're done: double-check carefully. */
		spin_lock_irqsave(&desc->lock, flags);
		status = desc->status;
		spin_unlock_irqrestore(&desc->lock, flags);

		/* Oops, that failed? */
	} while (status & IRQ_INPROGRESS);
}
EXPORT_SYMBOL(synchronize_irq);

/**
 *	irq_can_set_affinity - Check if the affinity of a given irq can be set
 *	@irq:		Interrupt to check
 *
 */
int irq_can_set_affinity(unsigned int irq)
{
	struct irq_desc *desc = irq_desc + irq;

	if (CHECK_IRQ_PER_CPU(desc->status) || !desc->chip ||
	    !desc->chip->set_affinity)
		return 0;

	return 1;
}

/**
 *	irq_set_affinity - Set the irq affinity of a given irq
 *	@irq:		Interrupt to set affinity
 *	@cpumask:	cpumask
 *
 */
int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
{
	struct irq_desc *desc = irq_desc + irq;

	if (!desc->chip->set_affinity)
		return -EINVAL;

	set_balance_irq_affinity(irq, cpumask);

#ifdef CONFIG_GENERIC_PENDING_IRQ
	set_pending_irq(irq, cpumask);
#else
	desc->affinity = cpumask;
	desc->chip->set_affinity(irq, cpumask);
#endif
	return 0;
}

#endif

/**
 *	disable_irq_nosync - disable an irq without waiting
 *	@irq: Interrupt to disable
 *
 *	Disable the selected interrupt line.  Disables and Enables are
 *	nested.
 *	Unlike disable_irq(), this function does not ensure existing
 *	instances of the IRQ handler have completed before returning.
 *
 *	This function may be called from IRQ context.
 */
void disable_irq_nosync(unsigned int irq)
{
	struct irq_desc *desc = irq_desc + irq;
	unsigned long flags;

	if (irq >= NR_IRQS)
		return;

	spin_lock_irqsave(&desc->lock, flags);
	if (!desc->depth++) {
		desc->status |= IRQ_DISABLED;
		desc->chip->disable(irq);
	}
	spin_unlock_irqrestore(&desc->lock, flags);
}
EXPORT_SYMBOL(disable_irq_nosync);

/**
 *	disable_irq - disable an irq and wait for completion
 *	@irq: Interrupt to disable
 *
 *	Disable the selected interrupt line.  Enables and Disables are
 *	nested.
 *	This function waits for any pending IRQ handlers for this interrupt
 *	to complete before returning. If you use this function while
 *	holding a resource the IRQ handler may need you will deadlock.
 *
 *	This function may be called - with care - from IRQ context.
 */
void disable_irq(unsigned int irq)
{
	struct irq_desc *desc = irq_desc + irq;

	if (irq >= NR_IRQS)
		return;

	disable_irq_nosync(irq);
	if (desc->action)
		synchronize_irq(irq);
}
EXPORT_SYMBOL(disable_irq);

static void __enable_irq(struct irq_desc *desc, unsigned int irq)
{
	switch (desc->depth) {
	case 0:
		printk(KERN_WARNING "Unbalanced enable for IRQ %d\n", irq);
		WARN_ON(1);
		break;
	case 1: {
		unsigned int status = desc->status & ~IRQ_DISABLED;

		/* Prevent probing on this irq: */
		desc->status = status | IRQ_NOPROBE;
		check_irq_resend(desc, irq);
		/* fall-through */
	}
	default:
		desc->depth--;
	}
}

/**
 *	enable_irq - enable handling of an irq
 *	@irq: Interrupt to enable
 *
 *	Undoes the effect of one call to disable_irq().  If this
 *	matches the last disable, processing of interrupts on this
 *	IRQ line is re-enabled.
 *
 *	This function may be called from IRQ context.
 */
void enable_irq(unsigned int irq)
{
	struct irq_desc *desc = irq_desc + irq;
	unsigned long flags;

	if (irq >= NR_IRQS)
		return;

	spin_lock_irqsave(&desc->lock, flags);
	__enable_irq(desc, irq);
	spin_unlock_irqrestore(&desc->lock, flags);
}
EXPORT_SYMBOL(enable_irq);

/**
 *	set_irq_wake - control irq power management wakeup
 *	@irq:	interrupt to control
 *	@on:	enable/disable power management wakeup
 *
 *	Enable/disable power management wakeup mode, which is
 *	disabled by default.  Enables and disables must match,
 *	just as they match for non-wakeup mode support.
 *
 *	Wakeup mode lets this IRQ wake the system from sleep
 *	states like "suspend to RAM".
 */
int set_irq_wake(unsigned int irq, unsigned int on)
{
	struct irq_desc *desc = irq_desc + irq;
	unsigned long flags;
	int ret = -ENXIO;
	int (*set_wake)(unsigned, unsigned) = desc->chip->set_wake;

	/* wakeup-capable irqs can be shared between drivers that
	 * don't need to have the same sleep mode behaviors.
	 */
	spin_lock_irqsave(&desc->lock, flags);
	if (on) {
		if (desc->wake_depth++ == 0)
			desc->status |= IRQ_WAKEUP;
		else
			set_wake = NULL;
	} else {
		if (desc->wake_depth == 0) {
			printk(KERN_WARNING "Unbalanced IRQ %d "
					"wake disable\n", irq);
			WARN_ON(1);
		} else if (--desc->wake_depth == 0)
			desc->status &= ~IRQ_WAKEUP;
		else
			set_wake = NULL;
	}
	if (set_wake)
		ret = desc->chip->set_wake(irq, on);
	spin_unlock_irqrestore(&desc->lock, flags);
	return ret;
}
EXPORT_SYMBOL(set_irq_wake);

/*
 * Internal function that tells the architecture code whether a
 * particular irq has been exclusively allocated or is available
 * for driver use.
 */
int can_request_irq(unsigned int irq, unsigned long irqflags)
{
	struct irqaction *action;

	if (irq >= NR_IRQS || irq_desc[irq].status & IRQ_NOREQUEST)
		return 0;

	action = irq_desc[irq].action;
	if (action)
		if (irqflags & action->flags & IRQF_SHARED)
			action = NULL;

	return !action;
}

void compat_irq_chip_set_default_handler(struct irq_desc *desc)
{
	/*
	 * If the architecture still has not overriden
	 * the flow handler then zap the default. This
	 * should catch incorrect flow-type setting.
	 */
	if (desc->handle_irq == &handle_bad_irq)
		desc->handle_irq = NULL;
}

/*
 * Internal function to register an irqaction - typically used to
 * allocate special interrupts that are part of the architecture.
 */
int setup_irq(unsigned int irq, struct irqaction *new)
{
	struct irq_desc *desc = irq_desc + irq;
	struct irqaction *old, **p;
	const char *old_name = NULL;
	unsigned long flags;
	int shared = 0;

	if (irq >= NR_IRQS)
		return -EINVAL;

	if (desc->chip == &no_irq_chip)
		return -ENOSYS;
	/*
	 * Some drivers like serial.c use request_irq() heavily,
	 * so we have to be careful not to interfere with a
	 * running system.
	 */
	if (new->flags & IRQF_SAMPLE_RANDOM) {
		/*
		 * This function might sleep, we want to call it first,
		 * outside of the atomic block.
		 * Yes, this might clear the entropy pool if the wrong
		 * driver is attempted to be loaded, without actually
		 * installing a new handler, but is this really a problem,
		 * only the sysadmin is able to do this.
		 */
		rand_initialize_irq(irq);
	}

	/*
	 * The following block of code has to be executed atomically
	 */
	spin_lock_irqsave(&desc->lock, flags);
	p = &desc->action;
	old = *p;
	if (old) {
		/*
		 * Can't share interrupts unless both agree to and are
		 * the same type (level, edge, polarity). So both flag
		 * fields must have IRQF_SHARED set and the bits which
		 * set the trigger type must match.
		 */
		if (!((old->flags & new->flags) & IRQF_SHARED) ||
		    ((old->flags ^ new->flags) & IRQF_TRIGGER_MASK)) {
			old_name = old->name;
			goto mismatch;
		}

#if defined(CONFIG_IRQ_PER_CPU)
		/* All handlers must agree on per-cpuness */
		if ((old->flags & IRQF_PERCPU) !=
		    (new->flags & IRQF_PERCPU))
			goto mismatch;
#endif

		/* add new interrupt at end of irq queue */
		do {
			p = &old->next;
			old = *p;
		} while (old);
		shared = 1;
	}

	*p = new;

	/* Exclude IRQ from balancing */
	if (new->flags & IRQF_NOBALANCING)
		desc->status |= IRQ_NO_BALANCING;

	if (!shared) {
		irq_chip_set_defaults(desc->chip);

#if defined(CONFIG_IRQ_PER_CPU)
		if (new->flags & IRQF_PERCPU)
			desc->status |= IRQ_PER_CPU;
#endif

		/* Setup the type (level, edge polarity) if configured: */
		if (new->flags & IRQF_TRIGGER_MASK) {
			if (desc->chip && desc->chip->set_type)
				desc->chip->set_type(irq,
						new->flags & IRQF_TRIGGER_MASK);
			else
				/*
				 * IRQF_TRIGGER_* but the PIC does not support
				 * multiple flow-types?
				 */
				printk(KERN_WARNING "No IRQF_TRIGGER set_type "
				       "function for IRQ %d (%s)\n", irq,
				       desc->chip ? desc->chip->name :
				       "unknown");
		} else
			compat_irq_chip_set_default_handler(desc);

		desc->status &= ~(IRQ_AUTODETECT | IRQ_WAITING |
				  IRQ_INPROGRESS | IRQ_SPURIOUS_DISABLED);

		if (!(desc->status & IRQ_NOAUTOEN)) {
			desc->depth = 0;
			desc->status &= ~IRQ_DISABLED;
			if (desc->chip->startup)
				desc->chip->startup(irq);
			else
				desc->chip->enable(irq);
		} else
			/* Undo nested disables: */
			desc->depth = 1;
	}
	/* Reset broken irq detection when installing new handler */
	desc->irq_count = 0;
	desc->irqs_unhandled = 0;

	/*
	 * Check whether we disabled the irq via the spurious handler
	 * before. Reenable it and give it another chance.
	 */
	if (shared && (desc->status & IRQ_SPURIOUS_DISABLED)) {
		desc->status &= ~IRQ_SPURIOUS_DISABLED;
		__enable_irq(desc, irq);
	}

	spin_unlock_irqrestore(&desc->lock, flags);

	new->irq = irq;
	register_irq_proc(irq);
	new->dir = NULL;
	register_handler_proc(irq, new);

	return 0;

mismatch:
#ifdef CONFIG_DEBUG_SHIRQ
	if (!(new->flags & IRQF_PROBE_SHARED)) {
		printk(KERN_ERR "IRQ handler type mismatch for IRQ %d\n", irq);
		if (old_name)
			printk(KERN_ERR "current handler: %s\n", old_name);
		dump_stack();
	}
#endif
	spin_unlock_irqrestore(&desc->lock, flags);
	return -EBUSY;
}

/**
 *	free_irq - free an interrupt
 *	@irq: Interrupt line to free
 *	@dev_id: Device identity to free
 *
 *	Remove an interrupt handler. The handler is removed and if the
 *	interrupt line is no longer in use by any driver it is disabled.
 *	On a shared IRQ the caller must ensure the interrupt is disabled
 *	on the card it drives before calling this function. The function
 *	does not return until any executing interrupts for this IRQ
 *	have completed.
 *
 *	This function must not be called from interrupt context.
 */
void free_irq(unsigned int irq, void *dev_id)
{
	struct irq_desc *desc;
	struct irqaction **p;
	unsigned long flags;

	WARN_ON(in_interrupt());
	if (irq >= NR_IRQS)
		return;

	desc = irq_desc + irq;
	spin_lock_irqsave(&desc->lock, flags);
	p = &desc->action;
	for (;;) {
		struct irqaction *action = *p;

		if (action) {
			struct irqaction **pp = p;

			p = &action->next;
			if (action->dev_id != dev_id)
				continue;

			/* Found it - now remove it from the list of entries */
			*pp = action->next;

			/* Currently used only by UML, might disappear one day.*/
#ifdef CONFIG_IRQ_RELEASE_METHOD
			if (desc->chip->release)
				desc->chip->release(irq, dev_id);
#endif

			if (!desc->action) {
				desc->status |= IRQ_DISABLED;
				if (desc->chip->shutdown)
					desc->chip->shutdown(irq);
				else
					desc->chip->disable(irq);
			}
			spin_unlock_irqrestore(&desc->lock, flags);
			unregister_handler_proc(irq, action);

			/* Make sure it's not being used on another CPU */
			synchronize_irq(irq);
#ifdef CONFIG_DEBUG_SHIRQ
			/*
			 * It's a shared IRQ -- the driver ought to be
			 * prepared for it to happen even now it's
			 * being freed, so let's make sure....  We do
			 * this after actually deregistering it, to
			 * make sure that a 'real' IRQ doesn't run in
			 * parallel with our fake
			 */
			if (action->flags & IRQF_SHARED) {
				local_irq_save(flags);
				action->handler(irq, dev_id);
				local_irq_restore(flags);
			}
#endif
			kfree(action);
			return;
		}
		printk(KERN_ERR "Trying to free already-free IRQ %d\n", irq);
#ifdef CONFIG_DEBUG_SHIRQ
		dump_stack();
#endif
		spin_unlock_irqrestore(&desc->lock, flags);
		return;
	}
}
EXPORT_SYMBOL(free_irq);

/**
 *	request_irq - allocate an interrupt line
 *	@irq: Interrupt line to allocate
 *	@handler: Function to be called when the IRQ occurs
 *	@irqflags: Interrupt type flags
 *	@devname: An ascii name for the claiming device
 *	@dev_id: A cookie passed back to the handler function
 *
 *	This call allocates interrupt resources and enables the
 *	interrupt line and IRQ handling. From the point this
 *	call is made your handler function may be invoked. Since
 *	your handler function must clear any interrupt the board
 *	raises, you must take care both to initialise your hardware
 *	and to set up the interrupt handler in the right order.
 *
 *	Dev_id must be globally unique. Normally the address of the
 *	device data structure is used as the cookie. Since the handler
 *	receives this value it makes sense to use it.
 *
 *	If your interrupt is shared you must pass a non NULL dev_id
 *	as this is required when freeing the interrupt.
 *
 *	Flags:
 *
 *	IRQF_SHARED		Interrupt is shared
 *	IRQF_DISABLED	Disable local interrupts while processing
 *	IRQF_SAMPLE_RANDOM	The interrupt can be used for entropy
 *
 */
int request_irq(unsigned int irq, irq_handler_t handler,
		unsigned long irqflags, const char *devname, void *dev_id)
{
	struct irqaction *action;
	int retval;

#ifdef CONFIG_LOCKDEP
	/*
	 * Lockdep wants atomic interrupt handlers:
	 */
	irqflags |= IRQF_DISABLED;
#endif
	/*
	 * Sanity-check: shared interrupts must pass in a real dev-ID,
	 * otherwise we'll have trouble later trying to figure out
	 * which interrupt is which (messes up the interrupt freeing
	 * logic etc).
	 */
	if ((irqflags & IRQF_SHARED) && !dev_id)
		return -EINVAL;
	if (irq >= NR_IRQS)
		return -EINVAL;
	if (irq_desc[irq].status & IRQ_NOREQUEST)
		return -EINVAL;
	if (!handler)
		return -EINVAL;

	action = kmalloc(sizeof(struct irqaction), GFP_ATOMIC);
	if (!action)
		return -ENOMEM;

	action->handler = handler;
	action->flags = irqflags;
	cpus_clear(action->mask);
	action->name = devname;
	action->next = NULL;
	action->dev_id = dev_id;

	select_smp_affinity(irq);

#ifdef CONFIG_DEBUG_SHIRQ
	if (irqflags & IRQF_SHARED) {
		/*
		 * It's a shared IRQ -- the driver ought to be prepared for it
		 * to happen immediately, so let's make sure....
		 * We do this before actually registering it, to make sure that
		 * a 'real' IRQ doesn't run in parallel with our fake
		 */
		unsigned long flags;

		local_irq_save(flags);
		handler(irq, dev_id);
		local_irq_restore(flags);
	}
#endif

	retval = setup_irq(irq, action);
	if (retval)
		kfree(action);

	return retval;
}
EXPORT_SYMBOL(request_irq);
id='n2028' href='#n2028'>2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900
/*
 * Generic ring buffer
 *
 * Copyright (C) 2008 Steven Rostedt <srostedt@redhat.com>
 */
#include <linux/ring_buffer.h>
#include <linux/trace_clock.h>
#include <linux/ftrace_irq.h>
#include <linux/spinlock.h>
#include <linux/debugfs.h>
#include <linux/uaccess.h>
#include <linux/hardirq.h>
#include <linux/kmemcheck.h>
#include <linux/module.h>
#include <linux/percpu.h>
#include <linux/mutex.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/hash.h>
#include <linux/list.h>
#include <linux/cpu.h>
#include <linux/fs.h>

#include <asm/local.h>
#include "trace.h"

/*
 * The ring buffer header is special. We must manually up keep it.
 */
int ring_buffer_print_entry_header(struct trace_seq *s)
{
	int ret;

	ret = trace_seq_printf(s, "# compressed entry header\n");
	ret = trace_seq_printf(s, "\ttype_len    :    5 bits\n");
	ret = trace_seq_printf(s, "\ttime_delta  :   27 bits\n");
	ret = trace_seq_printf(s, "\tarray       :   32 bits\n");
	ret = trace_seq_printf(s, "\n");
	ret = trace_seq_printf(s, "\tpadding     : type == %d\n",
			       RINGBUF_TYPE_PADDING);
	ret = trace_seq_printf(s, "\ttime_extend : type == %d\n",
			       RINGBUF_TYPE_TIME_EXTEND);
	ret = trace_seq_printf(s, "\tdata max type_len  == %d\n",
			       RINGBUF_TYPE_DATA_TYPE_LEN_MAX);

	return ret;
}

/*
 * The ring buffer is made up of a list of pages. A separate list of pages is
 * allocated for each CPU. A writer may only write to a buffer that is
 * associated with the CPU it is currently executing on.  A reader may read
 * from any per cpu buffer.
 *
 * The reader is special. For each per cpu buffer, the reader has its own
 * reader page. When a reader has read the entire reader page, this reader
 * page is swapped with another page in the ring buffer.
 *
 * Now, as long as the writer is off the reader page, the reader can do what
 * ever it wants with that page. The writer will never write to that page
 * again (as long as it is out of the ring buffer).
 *
 * Here's some silly ASCII art.
 *
 *   +------+
 *   |reader|          RING BUFFER
 *   |page  |
 *   +------+        +---+   +---+   +---+
 *                   |   |-->|   |-->|   |
 *                   +---+   +---+   +---+
 *                     ^               |
 *                     |               |
 *                     +---------------+
 *
 *
 *   +------+
 *   |reader|          RING BUFFER
 *   |page  |------------------v
 *   +------+        +---+   +---+   +---+
 *                   |   |-->|   |-->|   |
 *                   +---+   +---+   +---+
 *                     ^               |
 *                     |               |
 *                     +---------------+
 *
 *
 *   +------+
 *   |reader|          RING BUFFER
 *   |page  |------------------v
 *   +------+        +---+   +---+   +---+
 *      ^            |   |-->|   |-->|   |
 *      |            +---+   +---+   +---+
 *      |                              |
 *      |                              |
 *      +------------------------------+
 *
 *
 *   +------+
 *   |buffer|          RING BUFFER
 *   |page  |------------------v
 *   +------+        +---+   +---+   +---+
 *      ^            |   |   |   |-->|   |
 *      |   New      +---+   +---+   +---+
 *      |  Reader------^               |
 *      |   page                       |
 *      +------------------------------+
 *
 *
 * After we make this swap, the reader can hand this page off to the splice
 * code and be done with it. It can even allocate a new page if it needs to
 * and swap that into the ring buffer.
 *
 * We will be using cmpxchg soon to make all this lockless.
 *
 */

/*
 * A fast way to enable or disable all ring buffers is to
 * call tracing_on or tracing_off. Turning off the ring buffers
 * prevents all ring buffers from being recorded to.
 * Turning this switch on, makes it OK to write to the
 * ring buffer, if the ring buffer is enabled itself.
 *
 * There's three layers that must be on in order to write
 * to the ring buffer.
 *
 * 1) This global flag must be set.
 * 2) The ring buffer must be enabled for recording.
 * 3) The per cpu buffer must be enabled for recording.
 *
 * In case of an anomaly, this global flag has a bit set that
 * will permantly disable all ring buffers.
 */

/*
 * Global flag to disable all recording to ring buffers
 *  This has two bits: ON, DISABLED
 *
 *  ON   DISABLED
 * ---- ----------
 *   0      0        : ring buffers are off
 *   1      0        : ring buffers are on
 *   X      1        : ring buffers are permanently disabled
 */

enum {
	RB_BUFFERS_ON_BIT	= 0,
	RB_BUFFERS_DISABLED_BIT	= 1,
};

enum {
	RB_BUFFERS_ON		= 1 << RB_BUFFERS_ON_BIT,
	RB_BUFFERS_DISABLED	= 1 << RB_BUFFERS_DISABLED_BIT,
};

static unsigned long ring_buffer_flags __read_mostly = RB_BUFFERS_ON;

#define BUF_PAGE_HDR_SIZE offsetof(struct buffer_data_page, data)

/**
 * tracing_on - enable all tracing buffers
 *
 * This function enables all tracing buffers that may have been
 * disabled with tracing_off.
 */
void tracing_on(void)
{
	set_bit(RB_BUFFERS_ON_BIT, &ring_buffer_flags);
}
EXPORT_SYMBOL_GPL(tracing_on);

/**
 * tracing_off - turn off all tracing buffers
 *
 * This function stops all tracing buffers from recording data.
 * It does not disable any overhead the tracers themselves may
 * be causing. This function simply causes all recording to
 * the ring buffers to fail.
 */
void tracing_off(void)
{
	clear_bit(RB_BUFFERS_ON_BIT, &ring_buffer_flags);
}
EXPORT_SYMBOL_GPL(tracing_off);

/**
 * tracing_off_permanent - permanently disable ring buffers
 *
 * This function, once called, will disable all ring buffers
 * permanently.
 */
void tracing_off_permanent(void)
{
	set_bit(RB_BUFFERS_DISABLED_BIT, &ring_buffer_flags);
}

/**
 * tracing_is_on - show state of ring buffers enabled
 */
int tracing_is_on(void)
{
	return ring_buffer_flags == RB_BUFFERS_ON;
}
EXPORT_SYMBOL_GPL(tracing_is_on);

#define RB_EVNT_HDR_SIZE (offsetof(struct ring_buffer_event, array))
#define RB_ALIGNMENT		4U
#define RB_MAX_SMALL_DATA	(RB_ALIGNMENT * RINGBUF_TYPE_DATA_TYPE_LEN_MAX)
#define RB_EVNT_MIN_SIZE	8U	/* two 32bit words */

#if !defined(CONFIG_64BIT) || defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
# define RB_FORCE_8BYTE_ALIGNMENT	0
# define RB_ARCH_ALIGNMENT		RB_ALIGNMENT
#else
# define RB_FORCE_8BYTE_ALIGNMENT	1
# define RB_ARCH_ALIGNMENT		8U
#endif

/* define RINGBUF_TYPE_DATA for 'case RINGBUF_TYPE_DATA:' */
#define RINGBUF_TYPE_DATA 0 ... RINGBUF_TYPE_DATA_TYPE_LEN_MAX

enum {
	RB_LEN_TIME_EXTEND = 8,
	RB_LEN_TIME_STAMP = 16,
};

static inline int rb_null_event(struct ring_buffer_event *event)
{
	return event->type_len == RINGBUF_TYPE_PADDING && !event->time_delta;
}

static void rb_event_set_padding(struct ring_buffer_event *event)
{
	/* padding has a NULL time_delta */
	event->type_len = RINGBUF_TYPE_PADDING;
	event->time_delta = 0;
}

static unsigned
rb_event_data_length(struct ring_buffer_event *event)
{
	unsigned length;

	if (event->type_len)
		length = event->type_len * RB_ALIGNMENT;
	else
		length = event->array[0];
	return length + RB_EVNT_HDR_SIZE;
}

/* inline for ring buffer fast paths */
static unsigned
rb_event_length(struct ring_buffer_event *event)
{
	switch (event->type_len) {
	case RINGBUF_TYPE_PADDING:
		if (rb_null_event(event))
			/* undefined */
			return -1;
		return  event->array[0] + RB_EVNT_HDR_SIZE;

	case RINGBUF_TYPE_TIME_EXTEND:
		return RB_LEN_TIME_EXTEND;

	case RINGBUF_TYPE_TIME_STAMP:
		return RB_LEN_TIME_STAMP;

	case RINGBUF_TYPE_DATA:
		return rb_event_data_length(event);
	default:
		BUG();
	}
	/* not hit */
	return 0;
}

/**
 * ring_buffer_event_length - return the length of the event
 * @event: the event to get the length of
 */
unsigned ring_buffer_event_length(struct ring_buffer_event *event)
{
	unsigned length = rb_event_length(event);
	if (event->type_len > RINGBUF_TYPE_DATA_TYPE_LEN_MAX)
		return length;
	length -= RB_EVNT_HDR_SIZE;
	if (length > RB_MAX_SMALL_DATA + sizeof(event->array[0]))
                length -= sizeof(event->array[0]);
	return length;
}
EXPORT_SYMBOL_GPL(ring_buffer_event_length);

/* inline for ring buffer fast paths */
static void *
rb_event_data(struct ring_buffer_event *event)
{
	BUG_ON(event->type_len > RINGBUF_TYPE_DATA_TYPE_LEN_MAX);
	/* If length is in len field, then array[0] has the data */
	if (event->type_len)
		return (void *)&event->array[0];
	/* Otherwise length is in array[0] and array[1] has the data */
	return (void *)&event->array[1];
}

/**
 * ring_buffer_event_data - return the data of the event
 * @event: the event to get the data from
 */
void *ring_buffer_event_data(struct ring_buffer_event *event)
{
	return rb_event_data(event);
}
EXPORT_SYMBOL_GPL(ring_buffer_event_data);

#define for_each_buffer_cpu(buffer, cpu)		\
	for_each_cpu(cpu, buffer->cpumask)

#define TS_SHIFT	27
#define TS_MASK		((1ULL << TS_SHIFT) - 1)
#define TS_DELTA_TEST	(~TS_MASK)

struct buffer_data_page {
	u64		 time_stamp;	/* page time stamp */
	local_t		 commit;	/* write committed index */
	unsigned char	 data[];	/* data of buffer page */
};

/*
 * Note, the buffer_page list must be first. The buffer pages
 * are allocated in cache lines, which means that each buffer
 * page will be at the beginning of a cache line, and thus
 * the least significant bits will be zero. We use this to
 * add flags in the list struct pointers, to make the ring buffer
 * lockless.
 */
struct buffer_page {
	struct list_head list;		/* list of buffer pages */
	local_t		 write;		/* index for next write */
	unsigned	 read;		/* index for next read */
	local_t		 entries;	/* entries on this page */
	struct buffer_data_page *page;	/* Actual data page */
};

/*
 * The buffer page counters, write and entries, must be reset
 * atomically when crossing page boundaries. To synchronize this
 * update, two counters are inserted into the number. One is
 * the actual counter for the write position or count on the page.
 *
 * The other is a counter of updaters. Before an update happens
 * the update partition of the counter is incremented. This will
 * allow the updater to update the counter atomically.
 *
 * The counter is 20 bits, and the state data is 12.
 */
#define RB_WRITE_MASK		0xfffff
#define RB_WRITE_INTCNT		(1 << 20)

static void rb_init_page(struct buffer_data_page *bpage)
{
	local_set(&bpage->commit, 0);
}

/**
 * ring_buffer_page_len - the size of data on the page.
 * @page: The page to read
 *
 * Returns the amount of data on the page, including buffer page header.
 */
size_t ring_buffer_page_len(void *page)
{
	return local_read(&((struct buffer_data_page *)page)->commit)
		+ BUF_PAGE_HDR_SIZE;
}

/*
 * Also stolen from mm/slob.c. Thanks to Mathieu Desnoyers for pointing
 * this issue out.
 */
static void free_buffer_page(struct buffer_page *bpage)
{
	free_page((unsigned long)bpage->page);
	kfree(bpage);
}

/*
 * We need to fit the time_stamp delta into 27 bits.
 */
static inline int test_time_stamp(u64 delta)
{
	if (delta & TS_DELTA_TEST)
		return 1;
	return 0;
}

#define BUF_PAGE_SIZE (PAGE_SIZE - BUF_PAGE_HDR_SIZE)

/* Max payload is BUF_PAGE_SIZE - header (8bytes) */
#define BUF_MAX_DATA_SIZE (BUF_PAGE_SIZE - (sizeof(u32) * 2))

/* Max number of timestamps that can fit on a page */
#define RB_TIMESTAMPS_PER_PAGE	(BUF_PAGE_SIZE / RB_LEN_TIME_STAMP)

int ring_buffer_print_page_header(struct trace_seq *s)
{
	struct buffer_data_page field;
	int ret;

	ret = trace_seq_printf(s, "\tfield: u64 timestamp;\t"
			       "offset:0;\tsize:%u;\tsigned:%u;\n",
			       (unsigned int)sizeof(field.time_stamp),
			       (unsigned int)is_signed_type(u64));

	ret = trace_seq_printf(s, "\tfield: local_t commit;\t"
			       "offset:%u;\tsize:%u;\tsigned:%u;\n",
			       (unsigned int)offsetof(typeof(field), commit),
			       (unsigned int)sizeof(field.commit),
			       (unsigned int)is_signed_type(long));

	ret = trace_seq_printf(s, "\tfield: char data;\t"
			       "offset:%u;\tsize:%u;\tsigned:%u;\n",
			       (unsigned int)offsetof(typeof(field), data),
			       (unsigned int)BUF_PAGE_SIZE,
			       (unsigned int)is_signed_type(char));

	return ret;
}

/*
 * head_page == tail_page && head == tail then buffer is empty.
 */
struct ring_buffer_per_cpu {
	int				cpu;
	struct ring_buffer		*buffer;
	spinlock_t			reader_lock;	/* serialize readers */
	arch_spinlock_t			lock;
	struct lock_class_key		lock_key;
	struct list_head		*pages;
	struct buffer_page		*head_page;	/* read from head */
	struct buffer_page		*tail_page;	/* write to tail */
	struct buffer_page		*commit_page;	/* committed pages */
	struct buffer_page		*reader_page;
	local_t				commit_overrun;
	local_t				overrun;
	local_t				entries;
	local_t				committing;
	local_t				commits;
	unsigned long			read;
	u64				write_stamp;
	u64				read_stamp;
	atomic_t			record_disabled;
};

struct ring_buffer {
	unsigned			pages;
	unsigned			flags;
	int				cpus;
	atomic_t			record_disabled;
	cpumask_var_t			cpumask;

	struct lock_class_key		*reader_lock_key;

	struct mutex			mutex;

	struct ring_buffer_per_cpu	**buffers;

#ifdef CONFIG_HOTPLUG_CPU
	struct notifier_block		cpu_notify;
#endif
	u64				(*clock)(void);
};

struct ring_buffer_iter {
	struct ring_buffer_per_cpu	*cpu_buffer;
	unsigned long			head;
	struct buffer_page		*head_page;
	struct buffer_page		*cache_reader_page;
	unsigned long			cache_read;
	u64				read_stamp;
};

/* buffer may be either ring_buffer or ring_buffer_per_cpu */
#define RB_WARN_ON(b, cond)						\
	({								\
		int _____ret = unlikely(cond);				\
		if (_____ret) {						\
			if (__same_type(*(b), struct ring_buffer_per_cpu)) { \
				struct ring_buffer_per_cpu *__b =	\
					(void *)b;			\
				atomic_inc(&__b->buffer->record_disabled); \
			} else						\
				atomic_inc(&b->record_disabled);	\
			WARN_ON(1);					\
		}							\
		_____ret;						\
	})

/* Up this if you want to test the TIME_EXTENTS and normalization */
#define DEBUG_SHIFT 0

static inline u64 rb_time_stamp(struct ring_buffer *buffer)
{
	/* shift to debug/test normalization and TIME_EXTENTS */
	return buffer->clock() << DEBUG_SHIFT;
}

u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu)
{
	u64 time;

	preempt_disable_notrace();
	time = rb_time_stamp(buffer);
	preempt_enable_no_resched_notrace();

	return time;
}
EXPORT_SYMBOL_GPL(ring_buffer_time_stamp);

void ring_buffer_normalize_time_stamp(struct ring_buffer *buffer,
				      int cpu, u64 *ts)
{
	/* Just stupid testing the normalize function and deltas */
	*ts >>= DEBUG_SHIFT;
}
EXPORT_SYMBOL_GPL(ring_buffer_normalize_time_stamp);

/*
 * Making the ring buffer lockless makes things tricky.
 * Although writes only happen on the CPU that they are on,
 * and they only need to worry about interrupts. Reads can
 * happen on any CPU.
 *
 * The reader page is always off the ring buffer, but when the
 * reader finishes with a page, it needs to swap its page with
 * a new one from the buffer. The reader needs to take from
 * the head (writes go to the tail). But if a writer is in overwrite
 * mode and wraps, it must push the head page forward.
 *
 * Here lies the problem.
 *
 * The reader must be careful to replace only the head page, and
 * not another one. As described at the top of the file in the
 * ASCII art, the reader sets its old page to point to the next
 * page after head. It then sets the page after head to point to
 * the old reader page. But if the writer moves the head page
 * during this operation, the reader could end up with the tail.
 *
 * We use cmpxchg to help prevent this race. We also do something
 * special with the page before head. We set the LSB to 1.
 *
 * When the writer must push the page forward, it will clear the
 * bit that points to the head page, move the head, and then set
 * the bit that points to the new head page.
 *
 * We also don't want an interrupt coming in and moving the head
 * page on another writer. Thus we use the second LSB to catch
 * that too. Thus:
 *
 * head->list->prev->next        bit 1          bit 0
 *                              -------        -------
 * Normal page                     0              0
 * Points to head page             0              1
 * New head page                   1              0
 *
 * Note we can not trust the prev pointer of the head page, because:
 *
 * +----+       +-----+        +-----+
 * |    |------>|  T  |---X--->|  N  |
 * |    |<------|     |        |     |
 * +----+       +-----+        +-----+
 *   ^                           ^ |
 *   |          +-----+          | |
 *   +----------|  R  |----------+ |
 *              |     |<-----------+
 *              +-----+
 *
 * Key:  ---X-->  HEAD flag set in pointer
 *         T      Tail page
 *         R      Reader page
 *         N      Next page
 *
 * (see __rb_reserve_next() to see where this happens)
 *
 *  What the above shows is that the reader just swapped out
 *  the reader page with a page in the buffer, but before it
 *  could make the new header point back to the new page added
 *  it was preempted by a writer. The writer moved forward onto
 *  the new page added by the reader and is about to move forward
 *  again.
 *
 *  You can see, it is legitimate for the previous pointer of
 *  the head (or any page) not to point back to itself. But only
 *  temporarially.
 */

#define RB_PAGE_NORMAL		0UL
#define RB_PAGE_HEAD		1UL
#define RB_PAGE_UPDATE		2UL


#define RB_FLAG_MASK		3UL

/* PAGE_MOVED is not part of the mask */
#define RB_PAGE_MOVED		4UL

/*
 * rb_list_head - remove any bit
 */
static struct list_head *rb_list_head(struct list_head *list)
{
	unsigned long val = (unsigned long)list;

	return (struct list_head *)(val & ~RB_FLAG_MASK);
}

/*
 * rb_is_head_page - test if the given page is the head page
 *
 * Because the reader may move the head_page pointer, we can
 * not trust what the head page is (it may be pointing to
 * the reader page). But if the next page is a header page,
 * its flags will be non zero.
 */
static int inline
rb_is_head_page(struct ring_buffer_per_cpu *cpu_buffer,
		struct buffer_page *page, struct list_head *list)
{
	unsigned long val;

	val = (unsigned long)list->next;

	if ((val & ~RB_FLAG_MASK) != (unsigned long)&page->list)
		return RB_PAGE_MOVED;

	return val & RB_FLAG_MASK;
}

/*
 * rb_is_reader_page
 *
 * The unique thing about the reader page, is that, if the
 * writer is ever on it, the previous pointer never points
 * back to the reader page.
 */
static int rb_is_reader_page(struct buffer_page *page)
{
	struct list_head *list = page->list.prev;

	return rb_list_head(list->next) != &page->list;
}

/*
 * rb_set_list_to_head - set a list_head to be pointing to head.
 */
static void rb_set_list_to_head(struct ring_buffer_per_cpu *cpu_buffer,
				struct list_head *list)
{
	unsigned long *ptr;

	ptr = (unsigned long *)&list->next;
	*ptr |= RB_PAGE_HEAD;
	*ptr &= ~RB_PAGE_UPDATE;
}

/*
 * rb_head_page_activate - sets up head page
 */
static void rb_head_page_activate(struct ring_buffer_per_cpu *cpu_buffer)
{
	struct buffer_page *head;

	head = cpu_buffer->head_page;
	if (!head)
		return;

	/*
	 * Set the previous list pointer to have the HEAD flag.
	 */
	rb_set_list_to_head(cpu_buffer, head->list.prev);
}

static void rb_list_head_clear(struct list_head *list)
{
	unsigned long *ptr = (unsigned long *)&list->next;

	*ptr &= ~RB_FLAG_MASK;
}

/*
 * rb_head_page_dactivate - clears head page ptr (for free list)
 */
static void
rb_head_page_deactivate(struct ring_buffer_per_cpu *cpu_buffer)
{
	struct list_head *hd;

	/* Go through the whole list and clear any pointers found. */
	rb_list_head_clear(cpu_buffer->pages);

	list_for_each(hd, cpu_buffer->pages)
		rb_list_head_clear(hd);
}

static int rb_head_page_set(struct ring_buffer_per_cpu *cpu_buffer,
			    struct buffer_page *head,
			    struct buffer_page *prev,
			    int old_flag, int new_flag)
{
	struct list_head *list;
	unsigned long val = (unsigned long)&head->list;
	unsigned long ret;

	list = &prev->list;

	val &= ~RB_FLAG_MASK;

	ret = cmpxchg((unsigned long *)&list->next,
		      val | old_flag, val | new_flag);

	/* check if the reader took the page */
	if ((ret & ~RB_FLAG_MASK) != val)
		return RB_PAGE_MOVED;

	return ret & RB_FLAG_MASK;
}

static int rb_head_page_set_update(struct ring_buffer_per_cpu *cpu_buffer,
				   struct buffer_page *head,
				   struct buffer_page *prev,
				   int old_flag)
{
	return rb_head_page_set(cpu_buffer, head, prev,
				old_flag, RB_PAGE_UPDATE);
}

static int rb_head_page_set_head(struct ring_buffer_per_cpu *cpu_buffer,
				 struct buffer_page *head,
				 struct buffer_page *prev,
				 int old_flag)
{
	return rb_head_page_set(cpu_buffer, head, prev,
				old_flag, RB_PAGE_HEAD);
}

static int rb_head_page_set_normal(struct ring_buffer_per_cpu *cpu_buffer,
				   struct buffer_page *head,
				   struct buffer_page *prev,
				   int old_flag)
{
	return rb_head_page_set(cpu_buffer, head, prev,
				old_flag, RB_PAGE_NORMAL);
}

static inline void rb_inc_page(struct ring_buffer_per_cpu *cpu_buffer,
			       struct buffer_page **bpage)
{
	struct list_head *p = rb_list_head((*bpage)->list.next);

	*bpage = list_entry(p, struct buffer_page, list);
}

static struct buffer_page *
rb_set_head_page(struct ring_buffer_per_cpu *cpu_buffer)
{
	struct buffer_page *head;
	struct buffer_page *page;
	struct list_head *list;
	int i;

	if (RB_WARN_ON(cpu_buffer, !cpu_buffer->head_page))
		return NULL;

	/* sanity check */
	list = cpu_buffer->pages;
	if (RB_WARN_ON(cpu_buffer, rb_list_head(list->prev->next) != list))
		return NULL;

	page = head = cpu_buffer->head_page;
	/*
	 * It is possible that the writer moves the header behind
	 * where we started, and we miss in one loop.
	 * A second loop should grab the header, but we'll do
	 * three loops just because I'm paranoid.
	 */
	for (i = 0; i < 3; i++) {
		do {
			if (rb_is_head_page(cpu_buffer, page, page->list.prev)) {
				cpu_buffer->head_page = page;
				return page;
			}
			rb_inc_page(cpu_buffer, &page);
		} while (page != head);
	}

	RB_WARN_ON(cpu_buffer, 1);

	return NULL;
}

static int rb_head_page_replace(struct buffer_page *old,
				struct buffer_page *new)
{
	unsigned long *ptr = (unsigned long *)&old->list.prev->next;
	unsigned long val;
	unsigned long ret;

	val = *ptr & ~RB_FLAG_MASK;
	val |= RB_PAGE_HEAD;

	ret = cmpxchg(ptr, val, (unsigned long)&new->list);

	return ret == val;
}

/*
 * rb_tail_page_update - move the tail page forward
 *
 * Returns 1 if moved tail page, 0 if someone else did.
 */
static int rb_tail_page_update(struct ring_buffer_per_cpu *cpu_buffer,
			       struct buffer_page *tail_page,
			       struct buffer_page *next_page)
{
	struct buffer_page *old_tail;
	unsigned long old_entries;
	unsigned long old_write;
	int ret = 0;

	/*
	 * The tail page now needs to be moved forward.
	 *
	 * We need to reset the tail page, but without messing
	 * with possible erasing of data brought in by interrupts
	 * that have moved the tail page and are currently on it.
	 *
	 * We add a counter to the write field to denote this.
	 */
	old_write = local_add_return(RB_WRITE_INTCNT, &next_page->write);
	old_entries = local_add_return(RB_WRITE_INTCNT, &next_page->entries);

	/*
	 * Just make sure we have seen our old_write and synchronize
	 * with any interrupts that come in.
	 */
	barrier();

	/*
	 * If the tail page is still the same as what we think
	 * it is, then it is up to us to update the tail
	 * pointer.
	 */
	if (tail_page == cpu_buffer->tail_page) {
		/* Zero the write counter */
		unsigned long val = old_write & ~RB_WRITE_MASK;
		unsigned long eval = old_entries & ~RB_WRITE_MASK;

		/*
		 * This will only succeed if an interrupt did
		 * not come in and change it. In which case, we
		 * do not want to modify it.
		 *
		 * We add (void) to let the compiler know that we do not care
		 * about the return value of these functions. We use the
		 * cmpxchg to only update if an interrupt did not already
		 * do it for us. If the cmpxchg fails, we don't care.
		 */
		(void)local_cmpxchg(&next_page->write, old_write, val);
		(void)local_cmpxchg(&next_page->entries, old_entries, eval);

		/*
		 * No need to worry about races with clearing out the commit.
		 * it only can increment when a commit takes place. But that
		 * only happens in the outer most nested commit.
		 */
		local_set(&next_page->page->commit, 0);

		old_tail = cmpxchg(&cpu_buffer->tail_page,
				   tail_page, next_page);

		if (old_tail == tail_page)
			ret = 1;
	}

	return ret;
}

static int rb_check_bpage(struct ring_buffer_per_cpu *cpu_buffer,
			  struct buffer_page *bpage)
{
	unsigned long val = (unsigned long)bpage;

	if (RB_WARN_ON(cpu_buffer, val & RB_FLAG_MASK))
		return 1;

	return 0;
}

/**
 * rb_check_list - make sure a pointer to a list has the last bits zero
 */
static int rb_check_list(struct ring_buffer_per_cpu *cpu_buffer,
			 struct list_head *list)
{
	if (RB_WARN_ON(cpu_buffer, rb_list_head(list->prev) != list->prev))
		return 1;
	if (RB_WARN_ON(cpu_buffer, rb_list_head(list->next) != list->next))
		return 1;
	return 0;
}

/**
 * check_pages - integrity check of buffer pages
 * @cpu_buffer: CPU buffer with pages to test
 *
 * As a safety measure we check to make sure the data pages have not
 * been corrupted.
 */
static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer)
{
	struct list_head *head = cpu_buffer->pages;
	struct buffer_page *bpage, *tmp;

	rb_head_page_deactivate(cpu_buffer);

	if (RB_WARN_ON(cpu_buffer, head->next->prev != head))
		return -1;
	if (RB_WARN_ON(cpu_buffer, head->prev->next != head))
		return -1;

	if (rb_check_list(cpu_buffer, head))
		return -1;

	list_for_each_entry_safe(bpage, tmp, head, list) {
		if (RB_WARN_ON(cpu_buffer,
			       bpage->list.next->prev != &bpage->list))
			return -1;
		if (RB_WARN_ON(cpu_buffer,
			       bpage->list.prev->next != &bpage->list))
			return -1;
		if (rb_check_list(cpu_buffer, &bpage->list))
			return -1;
	}

	rb_head_page_activate(cpu_buffer);

	return 0;
}

static int rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer,
			     unsigned nr_pages)
{
	struct buffer_page *bpage, *tmp;
	unsigned long addr;
	LIST_HEAD(pages);
	unsigned i;

	WARN_ON(!nr_pages);

	for (i = 0; i < nr_pages; i++) {
		bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()),
				    GFP_KERNEL, cpu_to_node(cpu_buffer->cpu));
		if (!bpage)
			goto free_pages;

		rb_check_bpage(cpu_buffer, bpage);

		list_add(&bpage->list, &pages);

		addr = __get_free_page(GFP_KERNEL);
		if (!addr)
			goto free_pages;
		bpage->page = (void *)addr;
		rb_init_page(bpage->page);
	}

	/*
	 * The ring buffer page list is a circular list that does not
	 * start and end with a list head. All page list items point to
	 * other pages.
	 */
	cpu_buffer->pages = pages.next;
	list_del(&pages);

	rb_check_pages(cpu_buffer);

	return 0;

 free_pages:
	list_for_each_entry_safe(bpage, tmp, &pages, list) {
		list_del_init(&bpage->list);
		free_buffer_page(bpage);
	}
	return -ENOMEM;
}

static struct ring_buffer_per_cpu *
rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	struct buffer_page *bpage;
	unsigned long addr;
	int ret;

	cpu_buffer = kzalloc_node(ALIGN(sizeof(*cpu_buffer), cache_line_size()),
				  GFP_KERNEL, cpu_to_node(cpu));
	if (!cpu_buffer)
		return NULL;

	cpu_buffer->cpu = cpu;
	cpu_buffer->buffer = buffer;
	spin_lock_init(&cpu_buffer->reader_lock);
	lockdep_set_class(&cpu_buffer->reader_lock, buffer->reader_lock_key);
	cpu_buffer->lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;

	bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()),
			    GFP_KERNEL, cpu_to_node(cpu));
	if (!bpage)
		goto fail_free_buffer;

	rb_check_bpage(cpu_buffer, bpage);

	cpu_buffer->reader_page = bpage;
	addr = __get_free_page(GFP_KERNEL);
	if (!addr)
		goto fail_free_reader;
	bpage->page = (void *)addr;
	rb_init_page(bpage->page);

	INIT_LIST_HEAD(&cpu_buffer->reader_page->list);

	ret = rb_allocate_pages(cpu_buffer, buffer->pages);
	if (ret < 0)
		goto fail_free_reader;

	cpu_buffer->head_page
		= list_entry(cpu_buffer->pages, struct buffer_page, list);
	cpu_buffer->tail_page = cpu_buffer->commit_page = cpu_buffer->head_page;

	rb_head_page_activate(cpu_buffer);

	return cpu_buffer;

 fail_free_reader:
	free_buffer_page(cpu_buffer->reader_page);

 fail_free_buffer:
	kfree(cpu_buffer);
	return NULL;
}

static void rb_free_cpu_buffer(struct ring_buffer_per_cpu *cpu_buffer)
{
	struct list_head *head = cpu_buffer->pages;
	struct buffer_page *bpage, *tmp;

	free_buffer_page(cpu_buffer->reader_page);

	rb_head_page_deactivate(cpu_buffer);

	if (head) {
		list_for_each_entry_safe(bpage, tmp, head, list) {
			list_del_init(&bpage->list);
			free_buffer_page(bpage);
		}
		bpage = list_entry(head, struct buffer_page, list);
		free_buffer_page(bpage);
	}

	kfree(cpu_buffer);
}

#ifdef CONFIG_HOTPLUG_CPU
static int rb_cpu_notify(struct notifier_block *self,
			 unsigned long action, void *hcpu);
#endif

/**
 * ring_buffer_alloc - allocate a new ring_buffer
 * @size: the size in bytes per cpu that is needed.
 * @flags: attributes to set for the ring buffer.
 *
 * Currently the only flag that is available is the RB_FL_OVERWRITE
 * flag. This flag means that the buffer will overwrite old data
 * when the buffer wraps. If this flag is not set, the buffer will
 * drop data when the tail hits the head.
 */
struct ring_buffer *__ring_buffer_alloc(unsigned long size, unsigned flags,
					struct lock_class_key *key)
{
	struct ring_buffer *buffer;
	int bsize;
	int cpu;

	/* keep it in its own cache line */
	buffer = kzalloc(ALIGN(sizeof(*buffer), cache_line_size()),
			 GFP_KERNEL);
	if (!buffer)
		return NULL;

	if (!alloc_cpumask_var(&buffer->cpumask, GFP_KERNEL))
		goto fail_free_buffer;

	buffer->pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE);
	buffer->flags = flags;
	buffer->clock = trace_clock_local;
	buffer->reader_lock_key = key;

	/* need at least two pages */
	if (buffer->pages < 2)
		buffer->pages = 2;

	/*
	 * In case of non-hotplug cpu, if the ring-buffer is allocated
	 * in early initcall, it will not be notified of secondary cpus.
	 * In that off case, we need to allocate for all possible cpus.
	 */
#ifdef CONFIG_HOTPLUG_CPU
	get_online_cpus();
	cpumask_copy(buffer->cpumask, cpu_online_mask);
#else
	cpumask_copy(buffer->cpumask, cpu_possible_mask);
#endif
	buffer->cpus = nr_cpu_ids;

	bsize = sizeof(void *) * nr_cpu_ids;
	buffer->buffers = kzalloc(ALIGN(bsize, cache_line_size()),
				  GFP_KERNEL);
	if (!buffer->buffers)
		goto fail_free_cpumask;

	for_each_buffer_cpu(buffer, cpu) {
		buffer->buffers[cpu] =
			rb_allocate_cpu_buffer(buffer, cpu);
		if (!buffer->buffers[cpu])
			goto fail_free_buffers;
	}

#ifdef CONFIG_HOTPLUG_CPU
	buffer->cpu_notify.notifier_call = rb_cpu_notify;
	buffer->cpu_notify.priority = 0;
	register_cpu_notifier(&buffer->cpu_notify);
#endif

	put_online_cpus();
	mutex_init(&buffer->mutex);

	return buffer;

 fail_free_buffers:
	for_each_buffer_cpu(buffer, cpu) {
		if (buffer->buffers[cpu])
			rb_free_cpu_buffer(buffer->buffers[cpu]);
	}
	kfree(buffer->buffers);

 fail_free_cpumask:
	free_cpumask_var(buffer->cpumask);
	put_online_cpus();

 fail_free_buffer:
	kfree(buffer);
	return NULL;
}
EXPORT_SYMBOL_GPL(__ring_buffer_alloc);

/**
 * ring_buffer_free - free a ring buffer.
 * @buffer: the buffer to free.
 */
void
ring_buffer_free(struct ring_buffer *buffer)
{
	int cpu;

	get_online_cpus();

#ifdef CONFIG_HOTPLUG_CPU
	unregister_cpu_notifier(&buffer->cpu_notify);
#endif

	for_each_buffer_cpu(buffer, cpu)
		rb_free_cpu_buffer(buffer->buffers[cpu]);

	put_online_cpus();

	kfree(buffer->buffers);
	free_cpumask_var(buffer->cpumask);

	kfree(buffer);
}
EXPORT_SYMBOL_GPL(ring_buffer_free);

void ring_buffer_set_clock(struct ring_buffer *buffer,
			   u64 (*clock)(void))
{
	buffer->clock = clock;
}

static void rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer);

static void
rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages)
{
	struct buffer_page *bpage;
	struct list_head *p;
	unsigned i;

	spin_lock_irq(&cpu_buffer->reader_lock);
	rb_head_page_deactivate(cpu_buffer);

	for (i = 0; i < nr_pages; i++) {
		if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages)))
			goto out;
		p = cpu_buffer->pages->next;
		bpage = list_entry(p, struct buffer_page, list);
		list_del_init(&bpage->list);
		free_buffer_page(bpage);
	}
	if (RB_WARN_ON(cpu_buffer, list_empty(cpu_buffer->pages)))
		goto out;

	rb_reset_cpu(cpu_buffer);
	rb_check_pages(cpu_buffer);

out:
	spin_unlock_irq(&cpu_buffer->reader_lock);
}

static void
rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer,
		struct list_head *pages, unsigned nr_pages)
{
	struct buffer_page *bpage;
	struct list_head *p;
	unsigned i;

	spin_lock_irq(&cpu_buffer->reader_lock);
	rb_head_page_deactivate(cpu_buffer);

	for (i = 0; i < nr_pages; i++) {
		if (RB_WARN_ON(cpu_buffer, list_empty(pages)))
			goto out;
		p = pages->next;
		bpage = list_entry(p, struct buffer_page, list);
		list_del_init(&bpage->list);
		list_add_tail(&bpage->list, cpu_buffer->pages);
	}
	rb_reset_cpu(cpu_buffer);
	rb_check_pages(cpu_buffer);

out:
	spin_unlock_irq(&cpu_buffer->reader_lock);
}

/**
 * ring_buffer_resize - resize the ring buffer
 * @buffer: the buffer to resize.
 * @size: the new size.
 *
 * Minimum size is 2 * BUF_PAGE_SIZE.
 *
 * Returns -1 on failure.
 */
int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	unsigned nr_pages, rm_pages, new_pages;
	struct buffer_page *bpage, *tmp;
	unsigned long buffer_size;
	unsigned long addr;
	LIST_HEAD(pages);
	int i, cpu;

	/*
	 * Always succeed at resizing a non-existent buffer:
	 */
	if (!buffer)
		return size;

	size = DIV_ROUND_UP(size, BUF_PAGE_SIZE);
	size *= BUF_PAGE_SIZE;
	buffer_size = buffer->pages * BUF_PAGE_SIZE;

	/* we need a minimum of two pages */
	if (size < BUF_PAGE_SIZE * 2)
		size = BUF_PAGE_SIZE * 2;

	if (size == buffer_size)
		return size;

	atomic_inc(&buffer->record_disabled);

	/* Make sure all writers are done with this buffer. */
	synchronize_sched();

	mutex_lock(&buffer->mutex);
	get_online_cpus();

	nr_pages = DIV_ROUND_UP(size, BUF_PAGE_SIZE);

	if (size < buffer_size) {

		/* easy case, just free pages */
		if (RB_WARN_ON(buffer, nr_pages >= buffer->pages))
			goto out_fail;

		rm_pages = buffer->pages - nr_pages;

		for_each_buffer_cpu(buffer, cpu) {
			cpu_buffer = buffer->buffers[cpu];
			rb_remove_pages(cpu_buffer, rm_pages);
		}
		goto out;
	}

	/*
	 * This is a bit more difficult. We only want to add pages
	 * when we can allocate enough for all CPUs. We do this
	 * by allocating all the pages and storing them on a local
	 * link list. If we succeed in our allocation, then we
	 * add these pages to the cpu_buffers. Otherwise we just free
	 * them all and return -ENOMEM;
	 */
	if (RB_WARN_ON(buffer, nr_pages <= buffer->pages))
		goto out_fail;

	new_pages = nr_pages - buffer->pages;

	for_each_buffer_cpu(buffer, cpu) {
		for (i = 0; i < new_pages; i++) {
			bpage = kzalloc_node(ALIGN(sizeof(*bpage),
						  cache_line_size()),
					    GFP_KERNEL, cpu_to_node(cpu));
			if (!bpage)
				goto free_pages;
			list_add(&bpage->list, &pages);
			addr = __get_free_page(GFP_KERNEL);
			if (!addr)
				goto free_pages;
			bpage->page = (void *)addr;
			rb_init_page(bpage->page);
		}
	}

	for_each_buffer_cpu(buffer, cpu) {
		cpu_buffer = buffer->buffers[cpu];
		rb_insert_pages(cpu_buffer, &pages, new_pages);
	}

	if (RB_WARN_ON(buffer, !list_empty(&pages)))
		goto out_fail;

 out:
	buffer->pages = nr_pages;
	put_online_cpus();
	mutex_unlock(&buffer->mutex);

	atomic_dec(&buffer->record_disabled);

	return size;

 free_pages:
	list_for_each_entry_safe(bpage, tmp, &pages, list) {
		list_del_init(&bpage->list);
		free_buffer_page(bpage);
	}
	put_online_cpus();
	mutex_unlock(&buffer->mutex);
	atomic_dec(&buffer->record_disabled);
	return -ENOMEM;

	/*
	 * Something went totally wrong, and we are too paranoid
	 * to even clean up the mess.
	 */
 out_fail:
	put_online_cpus();
	mutex_unlock(&buffer->mutex);
	atomic_dec(&buffer->record_disabled);
	return -1;
}
EXPORT_SYMBOL_GPL(ring_buffer_resize);

static inline void *
__rb_data_page_index(struct buffer_data_page *bpage, unsigned index)
{
	return bpage->data + index;
}

static inline void *__rb_page_index(struct buffer_page *bpage, unsigned index)
{
	return bpage->page->data + index;
}

static inline struct ring_buffer_event *
rb_reader_event(struct ring_buffer_per_cpu *cpu_buffer)
{
	return __rb_page_index(cpu_buffer->reader_page,
			       cpu_buffer->reader_page->read);
}

static inline struct ring_buffer_event *
rb_iter_head_event(struct ring_buffer_iter *iter)
{
	return __rb_page_index(iter->head_page, iter->head);
}

static inline unsigned long rb_page_write(struct buffer_page *bpage)
{
	return local_read(&bpage->write) & RB_WRITE_MASK;
}

static inline unsigned rb_page_commit(struct buffer_page *bpage)
{
	return local_read(&bpage->page->commit);
}

static inline unsigned long rb_page_entries(struct buffer_page *bpage)
{
	return local_read(&bpage->entries) & RB_WRITE_MASK;
}

/* Size is determined by what has been commited */
static inline unsigned rb_page_size(struct buffer_page *bpage)
{
	return rb_page_commit(bpage);
}

static inline unsigned
rb_commit_index(struct ring_buffer_per_cpu *cpu_buffer)
{
	return rb_page_commit(cpu_buffer->commit_page);
}

static inline unsigned
rb_event_index(struct ring_buffer_event *event)
{
	unsigned long addr = (unsigned long)event;

	return (addr & ~PAGE_MASK) - BUF_PAGE_HDR_SIZE;
}

static inline int
rb_event_is_commit(struct ring_buffer_per_cpu *cpu_buffer,
		   struct ring_buffer_event *event)
{
	unsigned long addr = (unsigned long)event;
	unsigned long index;

	index = rb_event_index(event);
	addr &= PAGE_MASK;

	return cpu_buffer->commit_page->page == (void *)addr &&
		rb_commit_index(cpu_buffer) == index;
}

static void
rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer)
{
	unsigned long max_count;

	/*
	 * We only race with interrupts and NMIs on this CPU.
	 * If we own the commit event, then we can commit
	 * all others that interrupted us, since the interruptions
	 * are in stack format (they finish before they come
	 * back to us). This allows us to do a simple loop to
	 * assign the commit to the tail.
	 */
 again:
	max_count = cpu_buffer->buffer->pages * 100;

	while (cpu_buffer->commit_page != cpu_buffer->tail_page) {
		if (RB_WARN_ON(cpu_buffer, !(--max_count)))
			return;
		if (RB_WARN_ON(cpu_buffer,
			       rb_is_reader_page(cpu_buffer->tail_page)))
			return;
		local_set(&cpu_buffer->commit_page->page->commit,
			  rb_page_write(cpu_buffer->commit_page));
		rb_inc_page(cpu_buffer, &cpu_buffer->commit_page);
		cpu_buffer->write_stamp =
			cpu_buffer->commit_page->page->time_stamp;
		/* add barrier to keep gcc from optimizing too much */
		barrier();
	}
	while (rb_commit_index(cpu_buffer) !=
	       rb_page_write(cpu_buffer->commit_page)) {

		local_set(&cpu_buffer->commit_page->page->commit,
			  rb_page_write(cpu_buffer->commit_page));
		RB_WARN_ON(cpu_buffer,
			   local_read(&cpu_buffer->commit_page->page->commit) &
			   ~RB_WRITE_MASK);
		barrier();
	}

	/* again, keep gcc from optimizing */
	barrier();

	/*
	 * If an interrupt came in just after the first while loop
	 * and pushed the tail page forward, we will be left with
	 * a dangling commit that will never go forward.
	 */
	if (unlikely(cpu_buffer->commit_page != cpu_buffer->tail_page))
		goto again;
}

static void rb_reset_reader_page(struct ring_buffer_per_cpu *cpu_buffer)
{
	cpu_buffer->read_stamp = cpu_buffer->reader_page->page->time_stamp;
	cpu_buffer->reader_page->read = 0;
}

static void rb_inc_iter(struct ring_buffer_iter *iter)
{
	struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer;

	/*
	 * The iterator could be on the reader page (it starts there).
	 * But the head could have moved, since the reader was
	 * found. Check for this case and assign the iterator
	 * to the head page instead of next.
	 */
	if (iter->head_page == cpu_buffer->reader_page)
		iter->head_page = rb_set_head_page(cpu_buffer);
	else
		rb_inc_page(cpu_buffer, &iter->head_page);

	iter->read_stamp = iter->head_page->page->time_stamp;
	iter->head = 0;
}

/**
 * ring_buffer_update_event - update event type and data
 * @event: the even to update
 * @type: the type of event
 * @length: the size of the event field in the ring buffer
 *
 * Update the type and data fields of the event. The length
 * is the actual size that is written to the ring buffer,
 * and with this, we can determine what to place into the
 * data field.
 */
static void
rb_update_event(struct ring_buffer_event *event,
			 unsigned type, unsigned length)
{
	event->type_len = type;

	switch (type) {

	case RINGBUF_TYPE_PADDING:
	case RINGBUF_TYPE_TIME_EXTEND:
	case RINGBUF_TYPE_TIME_STAMP:
		break;

	case 0:
		length -= RB_EVNT_HDR_SIZE;
		if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT)
			event->array[0] = length;
		else
			event->type_len = DIV_ROUND_UP(length, RB_ALIGNMENT);
		break;
	default:
		BUG();
	}
}

/*
 * rb_handle_head_page - writer hit the head page
 *
 * Returns: +1 to retry page
 *           0 to continue
 *          -1 on error
 */
static int
rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer,
		    struct buffer_page *tail_page,
		    struct buffer_page *next_page)
{
	struct buffer_page *new_head;
	int entries;
	int type;
	int ret;

	entries = rb_page_entries(next_page);

	/*
	 * The hard part is here. We need to move the head
	 * forward, and protect against both readers on
	 * other CPUs and writers coming in via interrupts.
	 */
	type = rb_head_page_set_update(cpu_buffer, next_page, tail_page,
				       RB_PAGE_HEAD);

	/*
	 * type can be one of four:
	 *  NORMAL - an interrupt already moved it for us
	 *  HEAD   - we are the first to get here.
	 *  UPDATE - we are the interrupt interrupting
	 *           a current move.
	 *  MOVED  - a reader on another CPU moved the next
	 *           pointer to its reader page. Give up
	 *           and try again.
	 */

	switch (type) {
	case RB_PAGE_HEAD:
		/*
		 * We changed the head to UPDATE, thus
		 * it is our responsibility to update
		 * the counters.
		 */
		local_add(entries, &cpu_buffer->overrun);

		/*
		 * The entries will be zeroed out when we move the
		 * tail page.
		 */

		/* still more to do */
		break;

	case RB_PAGE_UPDATE:
		/*
		 * This is an interrupt that interrupt the
		 * previous update. Still more to do.
		 */
		break;
	case RB_PAGE_NORMAL:
		/*
		 * An interrupt came in before the update
		 * and processed this for us.
		 * Nothing left to do.
		 */
		return 1;
	case RB_PAGE_MOVED:
		/*
		 * The reader is on another CPU and just did
		 * a swap with our next_page.
		 * Try again.
		 */
		return 1;
	default:
		RB_WARN_ON(cpu_buffer, 1); /* WTF??? */
		return -1;
	}

	/*
	 * Now that we are here, the old head pointer is
	 * set to UPDATE. This will keep the reader from
	 * swapping the head page with the reader page.
	 * The reader (on another CPU) will spin till
	 * we are finished.
	 *
	 * We just need to protect against interrupts
	 * doing the job. We will set the next pointer
	 * to HEAD. After that, we set the old pointer
	 * to NORMAL, but only if it was HEAD before.
	 * otherwise we are an interrupt, and only
	 * want the outer most commit to reset it.
	 */
	new_head = next_page;
	rb_inc_page(cpu_buffer, &new_head);

	ret = rb_head_page_set_head(cpu_buffer, new_head, next_page,
				    RB_PAGE_NORMAL);

	/*
	 * Valid returns are:
	 *  HEAD   - an interrupt came in and already set it.
	 *  NORMAL - One of two things:
	 *            1) We really set it.
	 *            2) A bunch of interrupts came in and moved
	 *               the page forward again.
	 */
	switch (ret) {
	case RB_PAGE_HEAD:
	case RB_PAGE_NORMAL:
		/* OK */
		break;
	default:
		RB_WARN_ON(cpu_buffer, 1);
		return -1;
	}

	/*
	 * It is possible that an interrupt came in,
	 * set the head up, then more interrupts came in
	 * and moved it again. When we get back here,
	 * the page would have been set to NORMAL but we
	 * just set it back to HEAD.
	 *
	 * How do you detect this? Well, if that happened
	 * the tail page would have moved.
	 */
	if (ret == RB_PAGE_NORMAL) {
		/*
		 * If the tail had moved passed next, then we need
		 * to reset the pointer.
		 */
		if (cpu_buffer->tail_page != tail_page &&
		    cpu_buffer->tail_page != next_page)
			rb_head_page_set_normal(cpu_buffer, new_head,
						next_page,
						RB_PAGE_HEAD);
	}

	/*
	 * If this was the outer most commit (the one that
	 * changed the original pointer from HEAD to UPDATE),
	 * then it is up to us to reset it to NORMAL.
	 */
	if (type == RB_PAGE_HEAD) {
		ret = rb_head_page_set_normal(cpu_buffer, next_page,
					      tail_page,
					      RB_PAGE_UPDATE);
		if (RB_WARN_ON(cpu_buffer,
			       ret != RB_PAGE_UPDATE))
			return -1;
	}

	return 0;
}

static unsigned rb_calculate_event_length(unsigned length)
{
	struct ring_buffer_event event; /* Used only for sizeof array */

	/* zero length can cause confusions */
	if (!length)
		length = 1;

	if (length > RB_MAX_SMALL_DATA || RB_FORCE_8BYTE_ALIGNMENT)
		length += sizeof(event.array[0]);

	length += RB_EVNT_HDR_SIZE;
	length = ALIGN(length, RB_ARCH_ALIGNMENT);

	return length;
}

static inline void
rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer,
	      struct buffer_page *tail_page,
	      unsigned long tail, unsigned long length)
{
	struct ring_buffer_event *event;

	/*
	 * Only the event that crossed the page boundary
	 * must fill the old tail_page with padding.
	 */
	if (tail >= BUF_PAGE_SIZE) {
		local_sub(length, &tail_page->write);
		return;
	}

	event = __rb_page_index(tail_page, tail);
	kmemcheck_annotate_bitfield(event, bitfield);

	/*
	 * If this event is bigger than the minimum size, then
	 * we need to be careful that we don't subtract the
	 * write counter enough to allow another writer to slip
	 * in on this page.
	 * We put in a discarded commit instead, to make sure
	 * that this space is not used again.
	 *
	 * If we are less than the minimum size, we don't need to
	 * worry about it.
	 */
	if (tail > (BUF_PAGE_SIZE - RB_EVNT_MIN_SIZE)) {
		/* No room for any events */

		/* Mark the rest of the page with padding */
		rb_event_set_padding(event);

		/* Set the write back to the previous setting */
		local_sub(length, &tail_page->write);
		return;
	}

	/* Put in a discarded event */
	event->array[0] = (BUF_PAGE_SIZE - tail) - RB_EVNT_HDR_SIZE;
	event->type_len = RINGBUF_TYPE_PADDING;
	/* time delta must be non zero */
	event->time_delta = 1;

	/* Set write to end of buffer */
	length = (tail + length) - BUF_PAGE_SIZE;
	local_sub(length, &tail_page->write);
}

static struct ring_buffer_event *
rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer,
	     unsigned long length, unsigned long tail,
	     struct buffer_page *tail_page, u64 *ts)
{
	struct buffer_page *commit_page = cpu_buffer->commit_page;
	struct ring_buffer *buffer = cpu_buffer->buffer;
	struct buffer_page *next_page;
	int ret;

	next_page = tail_page;

	rb_inc_page(cpu_buffer, &next_page);

	/*
	 * If for some reason, we had an interrupt storm that made
	 * it all the way around the buffer, bail, and warn
	 * about it.
	 */
	if (unlikely(next_page == commit_page)) {
		local_inc(&cpu_buffer->commit_overrun);
		goto out_reset;
	}

	/*
	 * This is where the fun begins!
	 *
	 * We are fighting against races between a reader that
	 * could be on another CPU trying to swap its reader
	 * page with the buffer head.
	 *
	 * We are also fighting against interrupts coming in and
	 * moving the head or tail on us as well.
	 *
	 * If the next page is the head page then we have filled
	 * the buffer, unless the commit page is still on the
	 * reader page.
	 */
	if (rb_is_head_page(cpu_buffer, next_page, &tail_page->list)) {

		/*
		 * If the commit is not on the reader page, then
		 * move the header page.
		 */
		if (!rb_is_reader_page(cpu_buffer->commit_page)) {
			/*
			 * If we are not in overwrite mode,
			 * this is easy, just stop here.
			 */
			if (!(buffer->flags & RB_FL_OVERWRITE))
				goto out_reset;

			ret = rb_handle_head_page(cpu_buffer,
						  tail_page,
						  next_page);
			if (ret < 0)
				goto out_reset;
			if (ret)
				goto out_again;
		} else {
			/*
			 * We need to be careful here too. The
			 * commit page could still be on the reader
			 * page. We could have a small buffer, and
			 * have filled up the buffer with events
			 * from interrupts and such, and wrapped.
			 *
			 * Note, if the tail page is also the on the
			 * reader_page, we let it move out.
			 */
			if (unlikely((cpu_buffer->commit_page !=
				      cpu_buffer->tail_page) &&
				     (cpu_buffer->commit_page ==
				      cpu_buffer->reader_page))) {
				local_inc(&cpu_buffer->commit_overrun);
				goto out_reset;
			}
		}
	}

	ret = rb_tail_page_update(cpu_buffer, tail_page, next_page);
	if (ret) {
		/*
		 * Nested commits always have zero deltas, so
		 * just reread the time stamp
		 */
		*ts = rb_time_stamp(buffer);
		next_page->page->time_stamp = *ts;
	}

 out_again:

	rb_reset_tail(cpu_buffer, tail_page, tail, length);

	/* fail and let the caller try again */
	return ERR_PTR(-EAGAIN);

 out_reset:
	/* reset write */
	rb_reset_tail(cpu_buffer, tail_page, tail, length);

	return NULL;
}

static struct ring_buffer_event *
__rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer,
		  unsigned type, unsigned long length, u64 *ts)
{
	struct buffer_page *tail_page;
	struct ring_buffer_event *event;
	unsigned long tail, write;

	tail_page = cpu_buffer->tail_page;
	write = local_add_return(length, &tail_page->write);

	/* set write to only the index of the write */
	write &= RB_WRITE_MASK;
	tail = write - length;

	/* See if we shot pass the end of this buffer page */
	if (write > BUF_PAGE_SIZE)
		return rb_move_tail(cpu_buffer, length, tail,
				    tail_page, ts);

	/* We reserved something on the buffer */

	event = __rb_page_index(tail_page, tail);
	kmemcheck_annotate_bitfield(event, bitfield);
	rb_update_event(event, type, length);

	/* The passed in type is zero for DATA */
	if (likely(!type))
		local_inc(&tail_page->entries);

	/*
	 * If this is the first commit on the page, then update
	 * its timestamp.
	 */
	if (!tail)
		tail_page->page->time_stamp = *ts;

	return event;
}

static inline int
rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer,
		  struct ring_buffer_event *event)
{
	unsigned long new_index, old_index;
	struct buffer_page *bpage;
	unsigned long index;
	unsigned long addr;

	new_index = rb_event_index(event);
	old_index = new_index + rb_event_length(event);
	addr = (unsigned long)event;
	addr &= PAGE_MASK;

	bpage = cpu_buffer->tail_page;

	if (bpage->page == (void *)addr && rb_page_write(bpage) == old_index) {
		unsigned long write_mask =
			local_read(&bpage->write) & ~RB_WRITE_MASK;
		/*
		 * This is on the tail page. It is possible that
		 * a write could come in and move the tail page
		 * and write to the next page. That is fine
		 * because we just shorten what is on this page.
		 */
		old_index += write_mask;
		new_index += write_mask;
		index = local_cmpxchg(&bpage->write, old_index, new_index);
		if (index == old_index)
			return 1;
	}

	/* could not discard */
	return 0;
}

static int
rb_add_time_stamp(struct ring_buffer_per_cpu *cpu_buffer,
		  u64 *ts, u64 *delta)
{
	struct ring_buffer_event *event;
	static int once;
	int ret;

	if (unlikely(*delta > (1ULL << 59) && !once++)) {
		printk(KERN_WARNING "Delta way too big! %llu"
		       " ts=%llu write stamp = %llu\n",
		       (unsigned long long)*delta,
		       (unsigned long long)*ts,
		       (unsigned long long)cpu_buffer->write_stamp);
		WARN_ON(1);
	}

	/*
	 * The delta is too big, we to add a
	 * new timestamp.
	 */
	event = __rb_reserve_next(cpu_buffer,
				  RINGBUF_TYPE_TIME_EXTEND,
				  RB_LEN_TIME_EXTEND,
				  ts);
	if (!event)
		return -EBUSY;

	if (PTR_ERR(event) == -EAGAIN)
		return -EAGAIN;

	/* Only a commited time event can update the write stamp */
	if (rb_event_is_commit(cpu_buffer, event)) {
		/*
		 * If this is the first on the page, then it was
		 * updated with the page itself. Try to discard it
		 * and if we can't just make it zero.
		 */
		if (rb_event_index(event)) {
			event->time_delta = *delta & TS_MASK;
			event->array[0] = *delta >> TS_SHIFT;
		} else {
			/* try to discard, since we do not need this */
			if (!rb_try_to_discard(cpu_buffer, event)) {
				/* nope, just zero it */
				event->time_delta = 0;
				event->array[0] = 0;
			}
		}
		cpu_buffer->write_stamp = *ts;
		/* let the caller know this was the commit */
		ret = 1;
	} else {
		/* Try to discard the event */
		if (!rb_try_to_discard(cpu_buffer, event)) {
			/* Darn, this is just wasted space */
			event->time_delta = 0;
			event->array[0] = 0;
		}
		ret = 0;
	}

	*delta = 0;

	return ret;
}

static void rb_start_commit(struct ring_buffer_per_cpu *cpu_buffer)
{
	local_inc(&cpu_buffer->committing);
	local_inc(&cpu_buffer->commits);
}

static void rb_end_commit(struct ring_buffer_per_cpu *cpu_buffer)
{
	unsigned long commits;

	if (RB_WARN_ON(cpu_buffer,
		       !local_read(&cpu_buffer->committing)))
		return;

 again:
	commits = local_read(&cpu_buffer->commits);
	/* synchronize with interrupts */
	barrier();
	if (local_read(&cpu_buffer->committing) == 1)
		rb_set_commit_to_write(cpu_buffer);

	local_dec(&cpu_buffer->committing);

	/* synchronize with interrupts */
	barrier();

	/*
	 * Need to account for interrupts coming in between the
	 * updating of the commit page and the clearing of the
	 * committing counter.
	 */
	if (unlikely(local_read(&cpu_buffer->commits) != commits) &&
	    !local_read(&cpu_buffer->committing)) {
		local_inc(&cpu_buffer->committing);
		goto again;
	}
}

static struct ring_buffer_event *
rb_reserve_next_event(struct ring_buffer *buffer,
		      struct ring_buffer_per_cpu *cpu_buffer,
		      unsigned long length)
{
	struct ring_buffer_event *event;
	u64 ts, delta = 0;
	int commit = 0;
	int nr_loops = 0;

	rb_start_commit(cpu_buffer);

#ifdef CONFIG_RING_BUFFER_ALLOW_SWAP
	/*
	 * Due to the ability to swap a cpu buffer from a buffer
	 * it is possible it was swapped before we committed.
	 * (committing stops a swap). We check for it here and
	 * if it happened, we have to fail the write.
	 */
	barrier();
	if (unlikely(ACCESS_ONCE(cpu_buffer->buffer) != buffer)) {
		local_dec(&cpu_buffer->committing);
		local_dec(&cpu_buffer->commits);
		return NULL;
	}
#endif

	length = rb_calculate_event_length(length);
 again:
	/*
	 * We allow for interrupts to reenter here and do a trace.
	 * If one does, it will cause this original code to loop
	 * back here. Even with heavy interrupts happening, this
	 * should only happen a few times in a row. If this happens
	 * 1000 times in a row, there must be either an interrupt
	 * storm or we have something buggy.
	 * Bail!
	 */
	if (RB_WARN_ON(cpu_buffer, ++nr_loops > 1000))
		goto out_fail;

	ts = rb_time_stamp(cpu_buffer->buffer);

	/*
	 * Only the first commit can update the timestamp.
	 * Yes there is a race here. If an interrupt comes in
	 * just after the conditional and it traces too, then it
	 * will also check the deltas. More than one timestamp may
	 * also be made. But only the entry that did the actual
	 * commit will be something other than zero.
	 */
	if (likely(cpu_buffer->tail_page == cpu_buffer->commit_page &&
		   rb_page_write(cpu_buffer->tail_page) ==
		   rb_commit_index(cpu_buffer))) {
		u64 diff;

		diff = ts - cpu_buffer->write_stamp;

		/* make sure this diff is calculated here */
		barrier();

		/* Did the write stamp get updated already? */
		if (unlikely(ts < cpu_buffer->write_stamp))
			goto get_event;

		delta = diff;
		if (unlikely(test_time_stamp(delta))) {

			commit = rb_add_time_stamp(cpu_buffer, &ts, &delta);
			if (commit == -EBUSY)
				goto out_fail;

			if (commit == -EAGAIN)
				goto again;

			RB_WARN_ON(cpu_buffer, commit < 0);
		}
	}

 get_event:
	event = __rb_reserve_next(cpu_buffer, 0, length, &ts);
	if (unlikely(PTR_ERR(event) == -EAGAIN))
		goto again;

	if (!event)
		goto out_fail;

	if (!rb_event_is_commit(cpu_buffer, event))
		delta = 0;

	event->time_delta = delta;

	return event;

 out_fail:
	rb_end_commit(cpu_buffer);
	return NULL;
}

#ifdef CONFIG_TRACING

#define TRACE_RECURSIVE_DEPTH 16

static int trace_recursive_lock(void)
{
	current->trace_recursion++;

	if (likely(current->trace_recursion < TRACE_RECURSIVE_DEPTH))
		return 0;

	/* Disable all tracing before we do anything else */
	tracing_off_permanent();

	printk_once(KERN_WARNING "Tracing recursion: depth[%ld]:"
		    "HC[%lu]:SC[%lu]:NMI[%lu]\n",
		    current->trace_recursion,
		    hardirq_count() >> HARDIRQ_SHIFT,
		    softirq_count() >> SOFTIRQ_SHIFT,
		    in_nmi());

	WARN_ON_ONCE(1);
	return -1;
}

static void trace_recursive_unlock(void)
{
	WARN_ON_ONCE(!current->trace_recursion);

	current->trace_recursion--;
}

#else

#define trace_recursive_lock()		(0)
#define trace_recursive_unlock()	do { } while (0)

#endif

static DEFINE_PER_CPU(int, rb_need_resched);

/**
 * ring_buffer_lock_reserve - reserve a part of the buffer
 * @buffer: the ring buffer to reserve from
 * @length: the length of the data to reserve (excluding event header)
 *
 * Returns a reseverd event on the ring buffer to copy directly to.
 * The user of this interface will need to get the body to write into
 * and can use the ring_buffer_event_data() interface.
 *
 * The length is the length of the data needed, not the event length
 * which also includes the event header.
 *
 * Must be paired with ring_buffer_unlock_commit, unless NULL is returned.
 * If NULL is returned, then nothing has been allocated or locked.
 */
struct ring_buffer_event *
ring_buffer_lock_reserve(struct ring_buffer *buffer, unsigned long length)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	struct ring_buffer_event *event;
	int cpu, resched;

	if (ring_buffer_flags != RB_BUFFERS_ON)
		return NULL;

	/* If we are tracing schedule, we don't want to recurse */
	resched = ftrace_preempt_disable();

	if (atomic_read(&buffer->record_disabled))
		goto out_nocheck;

	if (trace_recursive_lock())
		goto out_nocheck;

	cpu = raw_smp_processor_id();

	if (!cpumask_test_cpu(cpu, buffer->cpumask))
		goto out;

	cpu_buffer = buffer->buffers[cpu];

	if (atomic_read(&cpu_buffer->record_disabled))
		goto out;

	if (length > BUF_MAX_DATA_SIZE)
		goto out;

	event = rb_reserve_next_event(buffer, cpu_buffer, length);
	if (!event)
		goto out;

	/*
	 * Need to store resched state on this cpu.
	 * Only the first needs to.
	 */

	if (preempt_count() == 1)
		per_cpu(rb_need_resched, cpu) = resched;

	return event;

 out:
	trace_recursive_unlock();

 out_nocheck:
	ftrace_preempt_enable(resched);
	return NULL;
}
EXPORT_SYMBOL_GPL(ring_buffer_lock_reserve);

static void
rb_update_write_stamp(struct ring_buffer_per_cpu *cpu_buffer,
		      struct ring_buffer_event *event)
{
	/*
	 * The event first in the commit queue updates the
	 * time stamp.
	 */
	if (rb_event_is_commit(cpu_buffer, event))
		cpu_buffer->write_stamp += event->time_delta;
}

static void rb_commit(struct ring_buffer_per_cpu *cpu_buffer,
		      struct ring_buffer_event *event)
{
	local_inc(&cpu_buffer->entries);
	rb_update_write_stamp(cpu_buffer, event);
	rb_end_commit(cpu_buffer);
}

/**
 * ring_buffer_unlock_commit - commit a reserved
 * @buffer: The buffer to commit to
 * @event: The event pointer to commit.
 *
 * This commits the data to the ring buffer, and releases any locks held.
 *
 * Must be paired with ring_buffer_lock_reserve.
 */
int ring_buffer_unlock_commit(struct ring_buffer *buffer,
			      struct ring_buffer_event *event)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	int cpu = raw_smp_processor_id();

	cpu_buffer = buffer->buffers[cpu];

	rb_commit(cpu_buffer, event);

	trace_recursive_unlock();

	/*
	 * Only the last preempt count needs to restore preemption.
	 */
	if (preempt_count() == 1)
		ftrace_preempt_enable(per_cpu(rb_need_resched, cpu));
	else
		preempt_enable_no_resched_notrace();

	return 0;
}
EXPORT_SYMBOL_GPL(ring_buffer_unlock_commit);

static inline void rb_event_discard(struct ring_buffer_event *event)
{
	/* array[0] holds the actual length for the discarded event */
	event->array[0] = rb_event_data_length(event) - RB_EVNT_HDR_SIZE;
	event->type_len = RINGBUF_TYPE_PADDING;
	/* time delta must be non zero */
	if (!event->time_delta)
		event->time_delta = 1;
}

/*
 * Decrement the entries to the page that an event is on.
 * The event does not even need to exist, only the pointer
 * to the page it is on. This may only be called before the commit
 * takes place.
 */
static inline void
rb_decrement_entry(struct ring_buffer_per_cpu *cpu_buffer,
		   struct ring_buffer_event *event)
{
	unsigned long addr = (unsigned long)event;
	struct buffer_page *bpage = cpu_buffer->commit_page;
	struct buffer_page *start;

	addr &= PAGE_MASK;

	/* Do the likely case first */
	if (likely(bpage->page == (void *)addr)) {
		local_dec(&bpage->entries);
		return;
	}

	/*
	 * Because the commit page may be on the reader page we
	 * start with the next page and check the end loop there.
	 */
	rb_inc_page(cpu_buffer, &bpage);
	start = bpage;
	do {
		if (bpage->page == (void *)addr) {
			local_dec(&bpage->entries);
			return;
		}
		rb_inc_page(cpu_buffer, &bpage);
	} while (bpage != start);

	/* commit not part of this buffer?? */
	RB_WARN_ON(cpu_buffer, 1);
}

/**
 * ring_buffer_commit_discard - discard an event that has not been committed
 * @buffer: the ring buffer
 * @event: non committed event to discard
 *
 * Sometimes an event that is in the ring buffer needs to be ignored.
 * This function lets the user discard an event in the ring buffer
 * and then that event will not be read later.
 *
 * This function only works if it is called before the the item has been
 * committed. It will try to free the event from the ring buffer
 * if another event has not been added behind it.
 *
 * If another event has been added behind it, it will set the event
 * up as discarded, and perform the commit.
 *
 * If this function is called, do not call ring_buffer_unlock_commit on
 * the event.
 */
void ring_buffer_discard_commit(struct ring_buffer *buffer,
				struct ring_buffer_event *event)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	int cpu;

	/* The event is discarded regardless */
	rb_event_discard(event);

	cpu = smp_processor_id();
	cpu_buffer = buffer->buffers[cpu];

	/*
	 * This must only be called if the event has not been
	 * committed yet. Thus we can assume that preemption
	 * is still disabled.
	 */
	RB_WARN_ON(buffer, !local_read(&cpu_buffer->committing));

	rb_decrement_entry(cpu_buffer, event);
	if (rb_try_to_discard(cpu_buffer, event))
		goto out;

	/*
	 * The commit is still visible by the reader, so we
	 * must still update the timestamp.
	 */
	rb_update_write_stamp(cpu_buffer, event);
 out:
	rb_end_commit(cpu_buffer);

	trace_recursive_unlock();

	/*
	 * Only the last preempt count needs to restore preemption.
	 */
	if (preempt_count() == 1)
		ftrace_preempt_enable(per_cpu(rb_need_resched, cpu));
	else
		preempt_enable_no_resched_notrace();

}
EXPORT_SYMBOL_GPL(ring_buffer_discard_commit);

/**
 * ring_buffer_write - write data to the buffer without reserving
 * @buffer: The ring buffer to write to.
 * @length: The length of the data being written (excluding the event header)
 * @data: The data to write to the buffer.
 *
 * This is like ring_buffer_lock_reserve and ring_buffer_unlock_commit as
 * one function. If you already have the data to write to the buffer, it
 * may be easier to simply call this function.
 *
 * Note, like ring_buffer_lock_reserve, the length is the length of the data
 * and not the length of the event which would hold the header.
 */
int ring_buffer_write(struct ring_buffer *buffer,
			unsigned long length,
			void *data)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	struct ring_buffer_event *event;
	void *body;
	int ret = -EBUSY;
	int cpu, resched;

	if (ring_buffer_flags != RB_BUFFERS_ON)
		return -EBUSY;

	resched = ftrace_preempt_disable();

	if (atomic_read(&buffer->record_disabled))
		goto out;

	cpu = raw_smp_processor_id();

	if (!cpumask_test_cpu(cpu, buffer->cpumask))
		goto out;

	cpu_buffer = buffer->buffers[cpu];

	if (atomic_read(&cpu_buffer->record_disabled))
		goto out;

	if (length > BUF_MAX_DATA_SIZE)
		goto out;

	event = rb_reserve_next_event(buffer, cpu_buffer, length);
	if (!event)
		goto out;

	body = rb_event_data(event);

	memcpy(body, data, length);

	rb_commit(cpu_buffer, event);

	ret = 0;
 out:
	ftrace_preempt_enable(resched);

	return ret;
}
EXPORT_SYMBOL_GPL(ring_buffer_write);

static int rb_per_cpu_empty(struct ring_buffer_per_cpu *cpu_buffer)
{
	struct buffer_page *reader = cpu_buffer->reader_page;
	struct buffer_page *head = rb_set_head_page(cpu_buffer);
	struct buffer_page *commit = cpu_buffer->commit_page;

	/* In case of error, head will be NULL */
	if (unlikely(!head))
		return 1;

	return reader->read == rb_page_commit(reader) &&
		(commit == reader ||
		 (commit == head &&
		  head->read == rb_page_commit(commit)));
}

/**
 * ring_buffer_record_disable - stop all writes into the buffer
 * @buffer: The ring buffer to stop writes to.
 *
 * This prevents all writes to the buffer. Any attempt to write
 * to the buffer after this will fail and return NULL.
 *
 * The caller should call synchronize_sched() after this.
 */
void ring_buffer_record_disable(struct ring_buffer *buffer)
{
	atomic_inc(&buffer->record_disabled);
}
EXPORT_SYMBOL_GPL(ring_buffer_record_disable);

/**
 * ring_buffer_record_enable - enable writes to the buffer
 * @buffer: The ring buffer to enable writes
 *
 * Note, multiple disables will need the same number of enables
 * to truly enable the writing (much like preempt_disable).
 */
void ring_buffer_record_enable(struct ring_buffer *buffer)
{
	atomic_dec(&buffer->record_disabled);
}
EXPORT_SYMBOL_GPL(ring_buffer_record_enable);

/**
 * ring_buffer_record_disable_cpu - stop all writes into the cpu_buffer
 * @buffer: The ring buffer to stop writes to.
 * @cpu: The CPU buffer to stop
 *
 * This prevents all writes to the buffer. Any attempt to write
 * to the buffer after this will fail and return NULL.
 *
 * The caller should call synchronize_sched() after this.
 */
void ring_buffer_record_disable_cpu(struct ring_buffer *buffer, int cpu)
{
	struct ring_buffer_per_cpu *cpu_buffer;

	if (!cpumask_test_cpu(cpu, buffer->cpumask))
		return;

	cpu_buffer = buffer->buffers[cpu];
	atomic_inc(&cpu_buffer->record_disabled);
}
EXPORT_SYMBOL_GPL(ring_buffer_record_disable_cpu);

/**
 * ring_buffer_record_enable_cpu - enable writes to the buffer
 * @buffer: The ring buffer to enable writes
 * @cpu: The CPU to enable.
 *
 * Note, multiple disables will need the same number of enables
 * to truly enable the writing (much like preempt_disable).
 */
void ring_buffer_record_enable_cpu(struct ring_buffer *buffer, int cpu)
{
	struct ring_buffer_per_cpu *cpu_buffer;

	if (!cpumask_test_cpu(cpu, buffer->cpumask))
		return;

	cpu_buffer = buffer->buffers[cpu];
	atomic_dec(&cpu_buffer->record_disabled);
}
EXPORT_SYMBOL_GPL(ring_buffer_record_enable_cpu);

/**
 * ring_buffer_entries_cpu - get the number of entries in a cpu buffer
 * @buffer: The ring buffer
 * @cpu: The per CPU buffer to get the entries from.
 */
unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	unsigned long ret;

	if (!cpumask_test_cpu(cpu, buffer->cpumask))
		return 0;

	cpu_buffer = buffer->buffers[cpu];
	ret = (local_read(&cpu_buffer->entries) - local_read(&cpu_buffer->overrun))
		- cpu_buffer->read;

	return ret;
}
EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu);

/**
 * ring_buffer_overrun_cpu - get the number of overruns in a cpu_buffer
 * @buffer: The ring buffer
 * @cpu: The per CPU buffer to get the number of overruns from
 */
unsigned long ring_buffer_overrun_cpu(struct ring_buffer *buffer, int cpu)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	unsigned long ret;

	if (!cpumask_test_cpu(cpu, buffer->cpumask))
		return 0;

	cpu_buffer = buffer->buffers[cpu];
	ret = local_read(&cpu_buffer->overrun);

	return ret;
}
EXPORT_SYMBOL_GPL(ring_buffer_overrun_cpu);

/**
 * ring_buffer_commit_overrun_cpu - get the number of overruns caused by commits
 * @buffer: The ring buffer
 * @cpu: The per CPU buffer to get the number of overruns from
 */
unsigned long
ring_buffer_commit_overrun_cpu(struct ring_buffer *buffer, int cpu)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	unsigned long ret;

	if (!cpumask_test_cpu(cpu, buffer->cpumask))
		return 0;

	cpu_buffer = buffer->buffers[cpu];
	ret = local_read(&cpu_buffer->commit_overrun);

	return ret;
}
EXPORT_SYMBOL_GPL(ring_buffer_commit_overrun_cpu);

/**
 * ring_buffer_entries - get the number of entries in a buffer
 * @buffer: The ring buffer
 *
 * Returns the total number of entries in the ring buffer
 * (all CPU entries)
 */
unsigned long ring_buffer_entries(struct ring_buffer *buffer)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	unsigned long entries = 0;
	int cpu;

	/* if you care about this being correct, lock the buffer */
	for_each_buffer_cpu(buffer, cpu) {
		cpu_buffer = buffer->buffers[cpu];
		entries += (local_read(&cpu_buffer->entries) -
			    local_read(&cpu_buffer->overrun)) - cpu_buffer->read;
	}

	return entries;
}
EXPORT_SYMBOL_GPL(ring_buffer_entries);

/**
 * ring_buffer_overruns - get the number of overruns in buffer
 * @buffer: The ring buffer
 *
 * Returns the total number of overruns in the ring buffer
 * (all CPU entries)
 */
unsigned long ring_buffer_overruns(struct ring_buffer *buffer)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	unsigned long overruns = 0;
	int cpu;

	/* if you care about this being correct, lock the buffer */
	for_each_buffer_cpu(buffer, cpu) {
		cpu_buffer = buffer->buffers[cpu];
		overruns += local_read(&cpu_buffer->overrun);
	}

	return overruns;
}
EXPORT_SYMBOL_GPL(ring_buffer_overruns);

static void rb_iter_reset(struct ring_buffer_iter *iter)
{
	struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer;

	/* Iterator usage is expected to have record disabled */
	if (list_empty(&cpu_buffer->reader_page->list)) {
		iter->head_page = rb_set_head_page(cpu_buffer);
		if (unlikely(!iter->head_page))
			return;
		iter->head = iter->head_page->read;
	} else {
		iter->head_page = cpu_buffer->reader_page;
		iter->head = cpu_buffer->reader_page->read;
	}
	if (iter->head)
		iter->read_stamp = cpu_buffer->read_stamp;
	else
		iter->read_stamp = iter->head_page->page->time_stamp;
	iter->cache_reader_page = cpu_buffer->reader_page;
	iter->cache_read = cpu_buffer->read;
}

/**
 * ring_buffer_iter_reset - reset an iterator
 * @iter: The iterator to reset
 *
 * Resets the iterator, so that it will start from the beginning
 * again.
 */
void ring_buffer_iter_reset(struct ring_buffer_iter *iter)
{
	struct ring_buffer_per_cpu *cpu_buffer;
	unsigned long flags;

	if (!iter)
		return;

	cpu_buffer = iter->cpu_buffer;

	spin_lock_irqsave(&cpu_buffer->reader_lock, flags);
	rb_iter_reset(iter);
	spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags);
}
EXPORT_SYMBOL_GPL(ring_buffer_iter_reset);

/**
 * ring_buffer_iter_empty - check if an iterator has no more to read
 * @iter: The iterator to check
 */
int ring_buffer_iter_empty(struct ring_buffer_iter *iter)
{
	struct ring_buffer_per_cpu *cpu_buffer;

	cpu_buffer = iter->cpu_buffer;

	return iter->head_page == cpu_buffer->commit_page &&
		iter->head == rb_commit_index(cpu_buffer);
}
EXPORT_SYMBOL_GPL(ring_buffer_iter_empty);

static void
rb_update_read_stamp(struct ring_buffer_per_cpu *cpu_buffer,
		     struct ring_buffer_event *event)
{
	u64 delta;

	switch (event->type_len) {
	case RINGBUF_TYPE_PADDING:
		return;

	case RINGBUF_TYPE_TIME_EXTEND:
		delta = event->array[0];
		delta <<= TS_SHIFT;
		delta += event->time_delta;
		cpu_buffer->read_stamp += delta;
		return;

	case RINGBUF_TYPE_TIME_STAMP:
		/* FIXME: not implemented */
		return;

	case RINGBUF_TYPE_DATA:
		cpu_buffer->read_stamp += event->time_delta;
		return;

	default:
		BUG();
	}
	return;
}

static void
rb_update_iter_read_stamp(struct ring_buffer_iter *iter,
			  struct ring_buffer_event *event)
{
	u64 delta;

	switch (event->type_len) {
	case RINGBUF_TYPE_PADDING:
		return;

	case RINGBUF_TYPE_TIME_EXTEND:
		delta = event->array[0];
		delta <<= TS_SHIFT;
		delta += event->time_delta;
		iter->read_stamp += delta;
		return;

	case RINGBUF_TYPE_TIME_STAMP:
		/* FIXME: not implemented */
		return;

	case RINGBUF_TYPE_DATA:
		iter->read_stamp += event->time_delta;
		return;

	default:
		BUG();
	}
	return;
}

static struct buffer_page *
rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer)
{
	struct buffer_page *reader = NULL;
	unsigned long flags;
	int nr_loops = 0;
	int ret;

	local_irq_save(flags);
	arch_spin_lock(&cpu_buffer->lock);

 again:
	/*
	 * This should normally only loop twice. But because the
	 * start of the reader inserts an empty page, it causes
	 * a case where we will loop three times. There should be no
	 * reason to loop four times (that I know of).
	 */
	if (RB_WARN_ON(cpu_buffer, ++nr_loops > 3)) {
		reader = NULL;
		goto out;
	}

	reader = cpu_buffer->reader_page;

	/* If there's more to read, return this page */
	if (cpu_buffer->reader_page->read < rb_page_size(reader))
		goto out;

	/* Never should we have an index greater than the size */
	if (RB_WARN_ON(cpu_buffer,
		       cpu_buffer->reader_page->read > rb_page_size(reader)))
		goto out;

	/* check if we caught up to the tail */
	reader = NULL;
	if (cpu_buffer->commit_page == cpu_buffer->reader_page)
		goto out;

	/*
	 * Reset the reader page to size zero.
	 */
	local_set(&cpu_buffer->reader_page->write, 0);