]> git.ipfire.org Git - thirdparty/kernel/stable.git/blame - kernel/locking/lockdep.c
mm: introduce memalloc_nofs_{save,restore} API
[thirdparty/kernel/stable.git] / kernel / locking / lockdep.c
CommitLineData
fbb9ce95
IM
1/*
2 * kernel/lockdep.c
3 *
4 * Runtime locking correctness validator
5 *
6 * Started by Ingo Molnar:
7 *
4b32d0a4 8 * Copyright (C) 2006,2007 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
90eec103 9 * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra
fbb9ce95
IM
10 *
11 * this code maps all the lock dependencies as they occur in a live kernel
12 * and will warn about the following classes of locking bugs:
13 *
14 * - lock inversion scenarios
15 * - circular lock dependencies
16 * - hardirq/softirq safe/unsafe locking bugs
17 *
18 * Bugs are reported even if the current locking scenario does not cause
19 * any deadlock at this point.
20 *
21 * I.e. if anytime in the past two locks were taken in a different order,
22 * even if it happened for another task, even if those were different
23 * locks (but of the same class as this lock), this code will detect it.
24 *
25 * Thanks to Arjan van de Ven for coming up with the initial idea of
26 * mapping lock dependencies runtime.
27 */
a5e25883 28#define DISABLE_BRANCH_PROFILING
fbb9ce95
IM
29#include <linux/mutex.h>
30#include <linux/sched.h>
e6017571 31#include <linux/sched/clock.h>
29930025 32#include <linux/sched/task.h>
6d7225f0 33#include <linux/sched/mm.h>
fbb9ce95
IM
34#include <linux/delay.h>
35#include <linux/module.h>
36#include <linux/proc_fs.h>
37#include <linux/seq_file.h>
38#include <linux/spinlock.h>
39#include <linux/kallsyms.h>
40#include <linux/interrupt.h>
41#include <linux/stacktrace.h>
42#include <linux/debug_locks.h>
43#include <linux/irqflags.h>
99de055a 44#include <linux/utsname.h>
4b32d0a4 45#include <linux/hash.h>
81d68a96 46#include <linux/ftrace.h>
b4b136f4 47#include <linux/stringify.h>
d588e461 48#include <linux/bitops.h>
5a0e3ad6 49#include <linux/gfp.h>
d3d03d4f 50#include <linux/kmemcheck.h>
e7904a28 51#include <linux/random.h>
dfaaf3fa 52#include <linux/jhash.h>
af012961 53
fbb9ce95
IM
54#include <asm/sections.h>
55
56#include "lockdep_internals.h"
57
a8d154b0 58#define CREATE_TRACE_POINTS
67178767 59#include <trace/events/lock.h>
a8d154b0 60
f20786ff
PZ
61#ifdef CONFIG_PROVE_LOCKING
62int prove_locking = 1;
63module_param(prove_locking, int, 0644);
64#else
65#define prove_locking 0
66#endif
67
68#ifdef CONFIG_LOCK_STAT
69int lock_stat = 1;
70module_param(lock_stat, int, 0644);
71#else
72#define lock_stat 0
73#endif
74
fbb9ce95 75/*
74c383f1
IM
76 * lockdep_lock: protects the lockdep graph, the hashes and the
77 * class/list/hash allocators.
fbb9ce95
IM
78 *
79 * This is one of the rare exceptions where it's justified
80 * to use a raw spinlock - we really dont want the spinlock
74c383f1 81 * code to recurse back into the lockdep code...
fbb9ce95 82 */
edc35bd7 83static arch_spinlock_t lockdep_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;
74c383f1
IM
84
85static int graph_lock(void)
86{
0199c4e6 87 arch_spin_lock(&lockdep_lock);
74c383f1
IM
88 /*
89 * Make sure that if another CPU detected a bug while
90 * walking the graph we dont change it (while the other
91 * CPU is busy printing out stuff with the graph lock
92 * dropped already)
93 */
94 if (!debug_locks) {
0199c4e6 95 arch_spin_unlock(&lockdep_lock);
74c383f1
IM
96 return 0;
97 }
bb065afb
SR
98 /* prevent any recursions within lockdep from causing deadlocks */
99 current->lockdep_recursion++;
74c383f1
IM
100 return 1;
101}
102
103static inline int graph_unlock(void)
104{
0119fee4
PZ
105 if (debug_locks && !arch_spin_is_locked(&lockdep_lock)) {
106 /*
107 * The lockdep graph lock isn't locked while we expect it to
108 * be, we're confused now, bye!
109 */
381a2292 110 return DEBUG_LOCKS_WARN_ON(1);
0119fee4 111 }
381a2292 112
bb065afb 113 current->lockdep_recursion--;
0199c4e6 114 arch_spin_unlock(&lockdep_lock);
74c383f1
IM
115 return 0;
116}
117
118/*
119 * Turn lock debugging off and return with 0 if it was off already,
120 * and also release the graph lock:
121 */
122static inline int debug_locks_off_graph_unlock(void)
123{
124 int ret = debug_locks_off();
125
0199c4e6 126 arch_spin_unlock(&lockdep_lock);
74c383f1
IM
127
128 return ret;
129}
fbb9ce95 130
fbb9ce95 131unsigned long nr_list_entries;
af012961 132static struct lock_list list_entries[MAX_LOCKDEP_ENTRIES];
fbb9ce95 133
fbb9ce95
IM
134/*
135 * All data structures here are protected by the global debug_lock.
136 *
137 * Mutex key structs only get allocated, once during bootup, and never
138 * get freed - this significantly simplifies the debugging code.
139 */
140unsigned long nr_lock_classes;
141static struct lock_class lock_classes[MAX_LOCKDEP_KEYS];
142
f82b217e
DJ
143static inline struct lock_class *hlock_class(struct held_lock *hlock)
144{
145 if (!hlock->class_idx) {
0119fee4
PZ
146 /*
147 * Someone passed in garbage, we give up.
148 */
f82b217e
DJ
149 DEBUG_LOCKS_WARN_ON(1);
150 return NULL;
151 }
152 return lock_classes + hlock->class_idx - 1;
153}
154
f20786ff 155#ifdef CONFIG_LOCK_STAT
25528213 156static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], cpu_lock_stats);
f20786ff 157
3365e779
PZ
158static inline u64 lockstat_clock(void)
159{
c676329a 160 return local_clock();
3365e779
PZ
161}
162
c7e78cff 163static int lock_point(unsigned long points[], unsigned long ip)
f20786ff
PZ
164{
165 int i;
166
c7e78cff
PZ
167 for (i = 0; i < LOCKSTAT_POINTS; i++) {
168 if (points[i] == 0) {
169 points[i] = ip;
f20786ff
PZ
170 break;
171 }
c7e78cff 172 if (points[i] == ip)
f20786ff
PZ
173 break;
174 }
175
176 return i;
177}
178
3365e779 179static void lock_time_inc(struct lock_time *lt, u64 time)
f20786ff
PZ
180{
181 if (time > lt->max)
182 lt->max = time;
183
109d71c6 184 if (time < lt->min || !lt->nr)
f20786ff
PZ
185 lt->min = time;
186
187 lt->total += time;
188 lt->nr++;
189}
190
c46261de
PZ
191static inline void lock_time_add(struct lock_time *src, struct lock_time *dst)
192{
109d71c6
FR
193 if (!src->nr)
194 return;
195
196 if (src->max > dst->max)
197 dst->max = src->max;
198
199 if (src->min < dst->min || !dst->nr)
200 dst->min = src->min;
201
c46261de
PZ
202 dst->total += src->total;
203 dst->nr += src->nr;
204}
205
206struct lock_class_stats lock_stats(struct lock_class *class)
207{
208 struct lock_class_stats stats;
209 int cpu, i;
210
211 memset(&stats, 0, sizeof(struct lock_class_stats));
212 for_each_possible_cpu(cpu) {
213 struct lock_class_stats *pcs =
1871e52c 214 &per_cpu(cpu_lock_stats, cpu)[class - lock_classes];
c46261de
PZ
215
216 for (i = 0; i < ARRAY_SIZE(stats.contention_point); i++)
217 stats.contention_point[i] += pcs->contention_point[i];
218
c7e78cff
PZ
219 for (i = 0; i < ARRAY_SIZE(stats.contending_point); i++)
220 stats.contending_point[i] += pcs->contending_point[i];
221
c46261de
PZ
222 lock_time_add(&pcs->read_waittime, &stats.read_waittime);
223 lock_time_add(&pcs->write_waittime, &stats.write_waittime);
224
225 lock_time_add(&pcs->read_holdtime, &stats.read_holdtime);
226 lock_time_add(&pcs->write_holdtime, &stats.write_holdtime);
96645678
PZ
227
228 for (i = 0; i < ARRAY_SIZE(stats.bounces); i++)
229 stats.bounces[i] += pcs->bounces[i];
c46261de
PZ
230 }
231
232 return stats;
233}
234
235void clear_lock_stats(struct lock_class *class)
236{
237 int cpu;
238
239 for_each_possible_cpu(cpu) {
240 struct lock_class_stats *cpu_stats =
1871e52c 241 &per_cpu(cpu_lock_stats, cpu)[class - lock_classes];
c46261de
PZ
242
243 memset(cpu_stats, 0, sizeof(struct lock_class_stats));
244 }
245 memset(class->contention_point, 0, sizeof(class->contention_point));
c7e78cff 246 memset(class->contending_point, 0, sizeof(class->contending_point));
c46261de
PZ
247}
248
f20786ff
PZ
249static struct lock_class_stats *get_lock_stats(struct lock_class *class)
250{
1871e52c 251 return &get_cpu_var(cpu_lock_stats)[class - lock_classes];
f20786ff
PZ
252}
253
254static void put_lock_stats(struct lock_class_stats *stats)
255{
1871e52c 256 put_cpu_var(cpu_lock_stats);
f20786ff
PZ
257}
258
259static void lock_release_holdtime(struct held_lock *hlock)
260{
261 struct lock_class_stats *stats;
3365e779 262 u64 holdtime;
f20786ff
PZ
263
264 if (!lock_stat)
265 return;
266
3365e779 267 holdtime = lockstat_clock() - hlock->holdtime_stamp;
f20786ff 268
f82b217e 269 stats = get_lock_stats(hlock_class(hlock));
f20786ff
PZ
270 if (hlock->read)
271 lock_time_inc(&stats->read_holdtime, holdtime);
272 else
273 lock_time_inc(&stats->write_holdtime, holdtime);
274 put_lock_stats(stats);
275}
276#else
277static inline void lock_release_holdtime(struct held_lock *hlock)
278{
279}
280#endif
281
fbb9ce95
IM
282/*
283 * We keep a global list of all lock classes. The list only grows,
284 * never shrinks. The list is only accessed with the lockdep
285 * spinlock lock held.
286 */
287LIST_HEAD(all_lock_classes);
288
289/*
290 * The lockdep classes are in a hash-table as well, for fast lookup:
291 */
292#define CLASSHASH_BITS (MAX_LOCKDEP_KEYS_BITS - 1)
293#define CLASSHASH_SIZE (1UL << CLASSHASH_BITS)
4b32d0a4 294#define __classhashfn(key) hash_long((unsigned long)key, CLASSHASH_BITS)
fbb9ce95
IM
295#define classhashentry(key) (classhash_table + __classhashfn((key)))
296
a63f38cc 297static struct hlist_head classhash_table[CLASSHASH_SIZE];
fbb9ce95 298
fbb9ce95
IM
299/*
300 * We put the lock dependency chains into a hash-table as well, to cache
301 * their existence:
302 */
303#define CHAINHASH_BITS (MAX_LOCKDEP_CHAINS_BITS-1)
304#define CHAINHASH_SIZE (1UL << CHAINHASH_BITS)
4b32d0a4 305#define __chainhashfn(chain) hash_long(chain, CHAINHASH_BITS)
fbb9ce95
IM
306#define chainhashentry(chain) (chainhash_table + __chainhashfn((chain)))
307
a63f38cc 308static struct hlist_head chainhash_table[CHAINHASH_SIZE];
fbb9ce95
IM
309
310/*
311 * The hash key of the lock dependency chains is a hash itself too:
312 * it's a hash of all locks taken up to that lock, including that lock.
313 * It's a 64-bit hash, because it's important for the keys to be
314 * unique.
315 */
dfaaf3fa
PZ
316static inline u64 iterate_chain_key(u64 key, u32 idx)
317{
318 u32 k0 = key, k1 = key >> 32;
319
320 __jhash_mix(idx, k0, k1); /* Macro that modifies arguments! */
321
322 return k0 | (u64)k1 << 32;
323}
fbb9ce95 324
1d09daa5 325void lockdep_off(void)
fbb9ce95
IM
326{
327 current->lockdep_recursion++;
328}
fbb9ce95
IM
329EXPORT_SYMBOL(lockdep_off);
330
1d09daa5 331void lockdep_on(void)
fbb9ce95
IM
332{
333 current->lockdep_recursion--;
334}
fbb9ce95
IM
335EXPORT_SYMBOL(lockdep_on);
336
fbb9ce95
IM
337/*
338 * Debugging switches:
339 */
340
341#define VERBOSE 0
33e94e96 342#define VERY_VERBOSE 0
fbb9ce95
IM
343
344#if VERBOSE
345# define HARDIRQ_VERBOSE 1
346# define SOFTIRQ_VERBOSE 1
cf40bd16 347# define RECLAIM_VERBOSE 1
fbb9ce95
IM
348#else
349# define HARDIRQ_VERBOSE 0
350# define SOFTIRQ_VERBOSE 0
cf40bd16 351# define RECLAIM_VERBOSE 0
fbb9ce95
IM
352#endif
353
cf40bd16 354#if VERBOSE || HARDIRQ_VERBOSE || SOFTIRQ_VERBOSE || RECLAIM_VERBOSE
fbb9ce95
IM
355/*
356 * Quick filtering for interesting events:
357 */
358static int class_filter(struct lock_class *class)
359{
f9829cce
AK
360#if 0
361 /* Example */
fbb9ce95 362 if (class->name_version == 1 &&
f9829cce 363 !strcmp(class->name, "lockname"))
fbb9ce95
IM
364 return 1;
365 if (class->name_version == 1 &&
f9829cce 366 !strcmp(class->name, "&struct->lockfield"))
fbb9ce95 367 return 1;
f9829cce 368#endif
a6640897
IM
369 /* Filter everything else. 1 would be to allow everything else */
370 return 0;
fbb9ce95
IM
371}
372#endif
373
374static int verbose(struct lock_class *class)
375{
376#if VERBOSE
377 return class_filter(class);
378#endif
379 return 0;
380}
381
fbb9ce95
IM
382/*
383 * Stack-trace: tightly packed array of stack backtrace
74c383f1 384 * addresses. Protected by the graph_lock.
fbb9ce95
IM
385 */
386unsigned long nr_stack_trace_entries;
387static unsigned long stack_trace[MAX_STACK_TRACE_ENTRIES];
388
2c522836
DJ
389static void print_lockdep_off(const char *bug_msg)
390{
391 printk(KERN_DEBUG "%s\n", bug_msg);
392 printk(KERN_DEBUG "turning off the locking correctness validator.\n");
acf59377 393#ifdef CONFIG_LOCK_STAT
2c522836 394 printk(KERN_DEBUG "Please attach the output of /proc/lock_stat to the bug report\n");
acf59377 395#endif
2c522836
DJ
396}
397
fbb9ce95
IM
398static int save_trace(struct stack_trace *trace)
399{
400 trace->nr_entries = 0;
401 trace->max_entries = MAX_STACK_TRACE_ENTRIES - nr_stack_trace_entries;
402 trace->entries = stack_trace + nr_stack_trace_entries;
403
5a1b3999 404 trace->skip = 3;
5a1b3999 405
ab1b6f03 406 save_stack_trace(trace);
fbb9ce95 407
4f84f433
PZ
408 /*
409 * Some daft arches put -1 at the end to indicate its a full trace.
410 *
411 * <rant> this is buggy anyway, since it takes a whole extra entry so a
412 * complete trace that maxes out the entries provided will be reported
413 * as incomplete, friggin useless </rant>
414 */
ea5b41f9
TL
415 if (trace->nr_entries != 0 &&
416 trace->entries[trace->nr_entries-1] == ULONG_MAX)
4f84f433
PZ
417 trace->nr_entries--;
418
fbb9ce95
IM
419 trace->max_entries = trace->nr_entries;
420
421 nr_stack_trace_entries += trace->nr_entries;
fbb9ce95 422
4f84f433 423 if (nr_stack_trace_entries >= MAX_STACK_TRACE_ENTRIES-1) {
74c383f1
IM
424 if (!debug_locks_off_graph_unlock())
425 return 0;
426
2c522836 427 print_lockdep_off("BUG: MAX_STACK_TRACE_ENTRIES too low!");
74c383f1
IM
428 dump_stack();
429
fbb9ce95
IM
430 return 0;
431 }
432
433 return 1;
434}
435
436unsigned int nr_hardirq_chains;
437unsigned int nr_softirq_chains;
438unsigned int nr_process_chains;
439unsigned int max_lockdep_depth;
fbb9ce95
IM
440
441#ifdef CONFIG_DEBUG_LOCKDEP
fbb9ce95
IM
442/*
443 * Various lockdep statistics:
444 */
bd6d29c2 445DEFINE_PER_CPU(struct lockdep_stats, lockdep_stats);
fbb9ce95
IM
446#endif
447
448/*
449 * Locking printouts:
450 */
451
fabe9c42 452#define __USAGE(__STATE) \
b4b136f4
PZ
453 [LOCK_USED_IN_##__STATE] = "IN-"__stringify(__STATE)"-W", \
454 [LOCK_ENABLED_##__STATE] = __stringify(__STATE)"-ON-W", \
455 [LOCK_USED_IN_##__STATE##_READ] = "IN-"__stringify(__STATE)"-R",\
456 [LOCK_ENABLED_##__STATE##_READ] = __stringify(__STATE)"-ON-R",
fabe9c42 457
fbb9ce95
IM
458static const char *usage_str[] =
459{
fabe9c42
PZ
460#define LOCKDEP_STATE(__STATE) __USAGE(__STATE)
461#include "lockdep_states.h"
462#undef LOCKDEP_STATE
463 [LOCK_USED] = "INITIAL USE",
fbb9ce95
IM
464};
465
466const char * __get_key_name(struct lockdep_subclass_key *key, char *str)
467{
ffb45122 468 return kallsyms_lookup((unsigned long)key, NULL, NULL, NULL, str);
fbb9ce95
IM
469}
470
3ff176ca 471static inline unsigned long lock_flag(enum lock_usage_bit bit)
fbb9ce95 472{
3ff176ca
PZ
473 return 1UL << bit;
474}
fbb9ce95 475
3ff176ca
PZ
476static char get_usage_char(struct lock_class *class, enum lock_usage_bit bit)
477{
478 char c = '.';
479
480 if (class->usage_mask & lock_flag(bit + 2))
481 c = '+';
482 if (class->usage_mask & lock_flag(bit)) {
483 c = '-';
484 if (class->usage_mask & lock_flag(bit + 2))
485 c = '?';
fbb9ce95
IM
486 }
487
3ff176ca
PZ
488 return c;
489}
cf40bd16 490
f510b233 491void get_usage_chars(struct lock_class *class, char usage[LOCK_USAGE_CHARS])
3ff176ca 492{
f510b233 493 int i = 0;
cf40bd16 494
f510b233
PZ
495#define LOCKDEP_STATE(__STATE) \
496 usage[i++] = get_usage_char(class, LOCK_USED_IN_##__STATE); \
497 usage[i++] = get_usage_char(class, LOCK_USED_IN_##__STATE##_READ);
498#include "lockdep_states.h"
499#undef LOCKDEP_STATE
500
501 usage[i] = '\0';
fbb9ce95
IM
502}
503
e5e78d08 504static void __print_lock_name(struct lock_class *class)
3003eba3
SR
505{
506 char str[KSYM_NAME_LEN];
507 const char *name;
508
fbb9ce95
IM
509 name = class->name;
510 if (!name) {
511 name = __get_key_name(class->key, str);
f943fe0f 512 printk(KERN_CONT "%s", name);
fbb9ce95 513 } else {
f943fe0f 514 printk(KERN_CONT "%s", name);
fbb9ce95 515 if (class->name_version > 1)
f943fe0f 516 printk(KERN_CONT "#%d", class->name_version);
fbb9ce95 517 if (class->subclass)
f943fe0f 518 printk(KERN_CONT "/%d", class->subclass);
fbb9ce95 519 }
e5e78d08
SR
520}
521
522static void print_lock_name(struct lock_class *class)
523{
524 char usage[LOCK_USAGE_CHARS];
525
526 get_usage_chars(class, usage);
527
f943fe0f 528 printk(KERN_CONT " (");
e5e78d08 529 __print_lock_name(class);
f943fe0f 530 printk(KERN_CONT "){%s}", usage);
fbb9ce95
IM
531}
532
533static void print_lockdep_cache(struct lockdep_map *lock)
534{
535 const char *name;
9281acea 536 char str[KSYM_NAME_LEN];
fbb9ce95
IM
537
538 name = lock->name;
539 if (!name)
540 name = __get_key_name(lock->key->subkeys, str);
541
f943fe0f 542 printk(KERN_CONT "%s", name);
fbb9ce95
IM
543}
544
545static void print_lock(struct held_lock *hlock)
546{
d7bc3197
PZ
547 /*
548 * We can be called locklessly through debug_show_all_locks() so be
549 * extra careful, the hlock might have been released and cleared.
550 */
551 unsigned int class_idx = hlock->class_idx;
552
553 /* Don't re-read hlock->class_idx, can't use READ_ONCE() on bitfields: */
554 barrier();
555
556 if (!class_idx || (class_idx - 1) >= MAX_LOCKDEP_KEYS) {
f943fe0f 557 printk(KERN_CONT "<RELEASED>\n");
d7bc3197
PZ
558 return;
559 }
560
561 print_lock_name(lock_classes + class_idx - 1);
f943fe0f
DV
562 printk(KERN_CONT ", at: [<%p>] %pS\n",
563 (void *)hlock->acquire_ip, (void *)hlock->acquire_ip);
fbb9ce95
IM
564}
565
566static void lockdep_print_held_locks(struct task_struct *curr)
567{
568 int i, depth = curr->lockdep_depth;
569
570 if (!depth) {
ba25f9dc 571 printk("no locks held by %s/%d.\n", curr->comm, task_pid_nr(curr));
fbb9ce95
IM
572 return;
573 }
574 printk("%d lock%s held by %s/%d:\n",
ba25f9dc 575 depth, depth > 1 ? "s" : "", curr->comm, task_pid_nr(curr));
fbb9ce95
IM
576
577 for (i = 0; i < depth; i++) {
578 printk(" #%d: ", i);
579 print_lock(curr->held_locks + i);
580 }
581}
fbb9ce95 582
fbdc4b9a 583static void print_kernel_ident(void)
8e18257d 584{
fbdc4b9a 585 printk("%s %.*s %s\n", init_utsname()->release,
8e18257d 586 (int)strcspn(init_utsname()->version, " "),
fbdc4b9a
BH
587 init_utsname()->version,
588 print_tainted());
8e18257d
PZ
589}
590
591static int very_verbose(struct lock_class *class)
592{
593#if VERY_VERBOSE
594 return class_filter(class);
595#endif
596 return 0;
597}
598
fbb9ce95 599/*
8e18257d 600 * Is this the address of a static object:
fbb9ce95 601 */
8dce7a9a 602#ifdef __KERNEL__
8e18257d 603static int static_obj(void *obj)
fbb9ce95 604{
8e18257d
PZ
605 unsigned long start = (unsigned long) &_stext,
606 end = (unsigned long) &_end,
607 addr = (unsigned long) obj;
8e18257d 608
fbb9ce95 609 /*
8e18257d 610 * static variable?
fbb9ce95 611 */
8e18257d
PZ
612 if ((addr >= start) && (addr < end))
613 return 1;
fbb9ce95 614
2a9ad18d
MF
615 if (arch_is_kernel_data(addr))
616 return 1;
617
fbb9ce95 618 /*
10fad5e4 619 * in-kernel percpu var?
fbb9ce95 620 */
10fad5e4
TH
621 if (is_kernel_percpu_address(addr))
622 return 1;
fbb9ce95 623
8e18257d 624 /*
10fad5e4 625 * module static or percpu var?
8e18257d 626 */
10fad5e4 627 return is_module_address(addr) || is_module_percpu_address(addr);
99de055a 628}
8dce7a9a 629#endif
99de055a 630
fbb9ce95 631/*
8e18257d
PZ
632 * To make lock name printouts unique, we calculate a unique
633 * class->name_version generation counter:
fbb9ce95 634 */
8e18257d 635static int count_matching_names(struct lock_class *new_class)
fbb9ce95 636{
8e18257d
PZ
637 struct lock_class *class;
638 int count = 0;
fbb9ce95 639
8e18257d 640 if (!new_class->name)
fbb9ce95
IM
641 return 0;
642
35a9393c 643 list_for_each_entry_rcu(class, &all_lock_classes, lock_entry) {
8e18257d
PZ
644 if (new_class->key - new_class->subclass == class->key)
645 return class->name_version;
646 if (class->name && !strcmp(class->name, new_class->name))
647 count = max(count, class->name_version);
648 }
fbb9ce95 649
8e18257d 650 return count + 1;
fbb9ce95
IM
651}
652
8e18257d
PZ
653/*
654 * Register a lock's class in the hash-table, if the class is not present
655 * yet. Otherwise we look it up. We cache the result in the lock object
656 * itself, so actual lookup of the hash should be once per lock object.
657 */
658static inline struct lock_class *
659look_up_lock_class(struct lockdep_map *lock, unsigned int subclass)
fbb9ce95 660{
8e18257d 661 struct lockdep_subclass_key *key;
a63f38cc 662 struct hlist_head *hash_head;
8e18257d 663 struct lock_class *class;
383776fa 664 bool is_static = false;
fbb9ce95 665
4ba053c0
HM
666 if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) {
667 debug_locks_off();
668 printk(KERN_ERR
669 "BUG: looking up invalid subclass: %u\n", subclass);
670 printk(KERN_ERR
671 "turning off the locking correctness validator.\n");
672 dump_stack();
673 return NULL;
674 }
675
8e18257d
PZ
676 /*
677 * Static locks do not have their class-keys yet - for them the key
383776fa
TG
678 * is the lock object itself. If the lock is in the per cpu area,
679 * the canonical address of the lock (per cpu offset removed) is
680 * used.
8e18257d 681 */
383776fa
TG
682 if (unlikely(!lock->key)) {
683 unsigned long can_addr, addr = (unsigned long)lock;
684
685 if (__is_kernel_percpu_address(addr, &can_addr))
686 lock->key = (void *)can_addr;
687 else if (__is_module_percpu_address(addr, &can_addr))
688 lock->key = (void *)can_addr;
689 else if (static_obj(lock))
690 lock->key = (void *)lock;
691 else
692 return ERR_PTR(-EINVAL);
693 is_static = true;
694 }
fbb9ce95 695
8e18257d
PZ
696 /*
697 * NOTE: the class-key must be unique. For dynamic locks, a static
698 * lock_class_key variable is passed in through the mutex_init()
699 * (or spin_lock_init()) call - which acts as the key. For static
700 * locks we use the lock object itself as the key.
701 */
4b32d0a4
PZ
702 BUILD_BUG_ON(sizeof(struct lock_class_key) >
703 sizeof(struct lockdep_map));
fbb9ce95 704
8e18257d 705 key = lock->key->subkeys + subclass;
ca268c69 706
8e18257d 707 hash_head = classhashentry(key);
74c383f1 708
8e18257d 709 /*
35a9393c 710 * We do an RCU walk of the hash, see lockdep_free_key_range().
8e18257d 711 */
35a9393c
PZ
712 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
713 return NULL;
714
a63f38cc 715 hlist_for_each_entry_rcu(class, hash_head, hash_entry) {
4b32d0a4 716 if (class->key == key) {
0119fee4
PZ
717 /*
718 * Huh! same key, different name? Did someone trample
719 * on some memory? We're most confused.
720 */
4b32d0a4 721 WARN_ON_ONCE(class->name != lock->name);
8e18257d 722 return class;
4b32d0a4
PZ
723 }
724 }
fbb9ce95 725
383776fa 726 return is_static || static_obj(lock->key) ? NULL : ERR_PTR(-EINVAL);
fbb9ce95
IM
727}
728
729/*
8e18257d
PZ
730 * Register a lock's class in the hash-table, if the class is not present
731 * yet. Otherwise we look it up. We cache the result in the lock object
732 * itself, so actual lookup of the hash should be once per lock object.
fbb9ce95 733 */
c003ed92 734static struct lock_class *
8e18257d 735register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
fbb9ce95 736{
8e18257d 737 struct lockdep_subclass_key *key;
a63f38cc 738 struct hlist_head *hash_head;
8e18257d 739 struct lock_class *class;
35a9393c
PZ
740
741 DEBUG_LOCKS_WARN_ON(!irqs_disabled());
8e18257d
PZ
742
743 class = look_up_lock_class(lock, subclass);
383776fa 744 if (likely(!IS_ERR_OR_NULL(class)))
87cdee71 745 goto out_set_class_cache;
8e18257d
PZ
746
747 /*
748 * Debug-check: all keys must be persistent!
383776fa
TG
749 */
750 if (IS_ERR(class)) {
8e18257d
PZ
751 debug_locks_off();
752 printk("INFO: trying to register non-static key.\n");
753 printk("the code is fine but needs lockdep annotation.\n");
754 printk("turning off the locking correctness validator.\n");
755 dump_stack();
8e18257d
PZ
756 return NULL;
757 }
758
759 key = lock->key->subkeys + subclass;
760 hash_head = classhashentry(key);
761
8e18257d 762 if (!graph_lock()) {
8e18257d
PZ
763 return NULL;
764 }
765 /*
766 * We have to do the hash-walk again, to avoid races
767 * with another CPU:
768 */
a63f38cc 769 hlist_for_each_entry_rcu(class, hash_head, hash_entry) {
8e18257d
PZ
770 if (class->key == key)
771 goto out_unlock_set;
35a9393c
PZ
772 }
773
8e18257d
PZ
774 /*
775 * Allocate a new key from the static array, and add it to
776 * the hash:
777 */
778 if (nr_lock_classes >= MAX_LOCKDEP_KEYS) {
779 if (!debug_locks_off_graph_unlock()) {
8e18257d
PZ
780 return NULL;
781 }
8e18257d 782
2c522836 783 print_lockdep_off("BUG: MAX_LOCKDEP_KEYS too low!");
eedeeabd 784 dump_stack();
8e18257d
PZ
785 return NULL;
786 }
787 class = lock_classes + nr_lock_classes++;
bd6d29c2 788 debug_atomic_inc(nr_unused_locks);
8e18257d
PZ
789 class->key = key;
790 class->name = lock->name;
791 class->subclass = subclass;
792 INIT_LIST_HEAD(&class->lock_entry);
793 INIT_LIST_HEAD(&class->locks_before);
794 INIT_LIST_HEAD(&class->locks_after);
795 class->name_version = count_matching_names(class);
796 /*
797 * We use RCU's safe list-add method to make
798 * parallel walking of the hash-list safe:
799 */
a63f38cc 800 hlist_add_head_rcu(&class->hash_entry, hash_head);
1481197b
DF
801 /*
802 * Add it to the global list of classes:
803 */
804 list_add_tail_rcu(&class->lock_entry, &all_lock_classes);
8e18257d
PZ
805
806 if (verbose(class)) {
807 graph_unlock();
8e18257d
PZ
808
809 printk("\nnew class %p: %s", class->key, class->name);
810 if (class->name_version > 1)
f943fe0f
DV
811 printk(KERN_CONT "#%d", class->name_version);
812 printk(KERN_CONT "\n");
8e18257d
PZ
813 dump_stack();
814
8e18257d 815 if (!graph_lock()) {
8e18257d
PZ
816 return NULL;
817 }
818 }
819out_unlock_set:
820 graph_unlock();
8e18257d 821
87cdee71 822out_set_class_cache:
8e18257d 823 if (!subclass || force)
62016250
HM
824 lock->class_cache[0] = class;
825 else if (subclass < NR_LOCKDEP_CACHING_CLASSES)
826 lock->class_cache[subclass] = class;
8e18257d 827
0119fee4
PZ
828 /*
829 * Hash collision, did we smoke some? We found a class with a matching
830 * hash but the subclass -- which is hashed in -- didn't match.
831 */
8e18257d
PZ
832 if (DEBUG_LOCKS_WARN_ON(class->subclass != subclass))
833 return NULL;
834
835 return class;
836}
837
838#ifdef CONFIG_PROVE_LOCKING
839/*
840 * Allocate a lockdep entry. (assumes the graph_lock held, returns
841 * with NULL on failure)
842 */
843static struct lock_list *alloc_list_entry(void)
844{
845 if (nr_list_entries >= MAX_LOCKDEP_ENTRIES) {
846 if (!debug_locks_off_graph_unlock())
847 return NULL;
848
2c522836 849 print_lockdep_off("BUG: MAX_LOCKDEP_ENTRIES too low!");
eedeeabd 850 dump_stack();
8e18257d
PZ
851 return NULL;
852 }
853 return list_entries + nr_list_entries++;
854}
855
856/*
857 * Add a new dependency to the head of the list:
858 */
83f06168
TE
859static int add_lock_to_list(struct lock_class *this, struct list_head *head,
860 unsigned long ip, int distance,
861 struct stack_trace *trace)
8e18257d
PZ
862{
863 struct lock_list *entry;
864 /*
865 * Lock not present yet - get a new dependency struct and
866 * add it to the list:
867 */
868 entry = alloc_list_entry();
869 if (!entry)
870 return 0;
871
74870172
ZY
872 entry->class = this;
873 entry->distance = distance;
4726f2a6 874 entry->trace = *trace;
8e18257d 875 /*
35a9393c
PZ
876 * Both allocation and removal are done under the graph lock; but
877 * iteration is under RCU-sched; see look_up_lock_class() and
878 * lockdep_free_key_range().
8e18257d
PZ
879 */
880 list_add_tail_rcu(&entry->entry, head);
881
882 return 1;
883}
884
98c33edd
PZ
885/*
886 * For good efficiency of modular, we use power of 2
887 */
af012961
PZ
888#define MAX_CIRCULAR_QUEUE_SIZE 4096UL
889#define CQ_MASK (MAX_CIRCULAR_QUEUE_SIZE-1)
890
98c33edd
PZ
891/*
892 * The circular_queue and helpers is used to implement the
af012961
PZ
893 * breadth-first search(BFS)algorithem, by which we can build
894 * the shortest path from the next lock to be acquired to the
895 * previous held lock if there is a circular between them.
98c33edd 896 */
af012961
PZ
897struct circular_queue {
898 unsigned long element[MAX_CIRCULAR_QUEUE_SIZE];
899 unsigned int front, rear;
900};
901
902static struct circular_queue lock_cq;
af012961 903
12f3dfd0 904unsigned int max_bfs_queue_depth;
af012961 905
e351b660
ML
906static unsigned int lockdep_dependency_gen_id;
907
af012961
PZ
908static inline void __cq_init(struct circular_queue *cq)
909{
910 cq->front = cq->rear = 0;
e351b660 911 lockdep_dependency_gen_id++;
af012961
PZ
912}
913
914static inline int __cq_empty(struct circular_queue *cq)
915{
916 return (cq->front == cq->rear);
917}
918
919static inline int __cq_full(struct circular_queue *cq)
920{
921 return ((cq->rear + 1) & CQ_MASK) == cq->front;
922}
923
924static inline int __cq_enqueue(struct circular_queue *cq, unsigned long elem)
925{
926 if (__cq_full(cq))
927 return -1;
928
929 cq->element[cq->rear] = elem;
930 cq->rear = (cq->rear + 1) & CQ_MASK;
931 return 0;
932}
933
934static inline int __cq_dequeue(struct circular_queue *cq, unsigned long *elem)
935{
936 if (__cq_empty(cq))
937 return -1;
938
939 *elem = cq->element[cq->front];
940 cq->front = (cq->front + 1) & CQ_MASK;
941 return 0;
942}
943
944static inline unsigned int __cq_get_elem_count(struct circular_queue *cq)
945{
946 return (cq->rear - cq->front) & CQ_MASK;
947}
948
949static inline void mark_lock_accessed(struct lock_list *lock,
950 struct lock_list *parent)
951{
952 unsigned long nr;
98c33edd 953
af012961 954 nr = lock - list_entries;
0119fee4 955 WARN_ON(nr >= nr_list_entries); /* Out-of-bounds, input fail */
af012961 956 lock->parent = parent;
e351b660 957 lock->class->dep_gen_id = lockdep_dependency_gen_id;
af012961
PZ
958}
959
960static inline unsigned long lock_accessed(struct lock_list *lock)
961{
962 unsigned long nr;
98c33edd 963
af012961 964 nr = lock - list_entries;
0119fee4 965 WARN_ON(nr >= nr_list_entries); /* Out-of-bounds, input fail */
e351b660 966 return lock->class->dep_gen_id == lockdep_dependency_gen_id;
af012961
PZ
967}
968
969static inline struct lock_list *get_lock_parent(struct lock_list *child)
970{
971 return child->parent;
972}
973
974static inline int get_lock_depth(struct lock_list *child)
975{
976 int depth = 0;
977 struct lock_list *parent;
978
979 while ((parent = get_lock_parent(child))) {
980 child = parent;
981 depth++;
982 }
983 return depth;
984}
985
9e2d551e 986static int __bfs(struct lock_list *source_entry,
af012961
PZ
987 void *data,
988 int (*match)(struct lock_list *entry, void *data),
989 struct lock_list **target_entry,
990 int forward)
c94aa5ca
ML
991{
992 struct lock_list *entry;
d588e461 993 struct list_head *head;
c94aa5ca
ML
994 struct circular_queue *cq = &lock_cq;
995 int ret = 1;
996
9e2d551e 997 if (match(source_entry, data)) {
c94aa5ca
ML
998 *target_entry = source_entry;
999 ret = 0;
1000 goto exit;
1001 }
1002
d588e461
ML
1003 if (forward)
1004 head = &source_entry->class->locks_after;
1005 else
1006 head = &source_entry->class->locks_before;
1007
1008 if (list_empty(head))
1009 goto exit;
1010
1011 __cq_init(cq);
c94aa5ca
ML
1012 __cq_enqueue(cq, (unsigned long)source_entry);
1013
1014 while (!__cq_empty(cq)) {
1015 struct lock_list *lock;
c94aa5ca
ML
1016
1017 __cq_dequeue(cq, (unsigned long *)&lock);
1018
1019 if (!lock->class) {
1020 ret = -2;
1021 goto exit;
1022 }
1023
1024 if (forward)
1025 head = &lock->class->locks_after;
1026 else
1027 head = &lock->class->locks_before;
1028
35a9393c
PZ
1029 DEBUG_LOCKS_WARN_ON(!irqs_disabled());
1030
1031 list_for_each_entry_rcu(entry, head, entry) {
c94aa5ca 1032 if (!lock_accessed(entry)) {
12f3dfd0 1033 unsigned int cq_depth;
c94aa5ca 1034 mark_lock_accessed(entry, lock);
9e2d551e 1035 if (match(entry, data)) {
c94aa5ca
ML
1036 *target_entry = entry;
1037 ret = 0;
1038 goto exit;
1039 }
1040
1041 if (__cq_enqueue(cq, (unsigned long)entry)) {
1042 ret = -1;
1043 goto exit;
1044 }
12f3dfd0
ML
1045 cq_depth = __cq_get_elem_count(cq);
1046 if (max_bfs_queue_depth < cq_depth)
1047 max_bfs_queue_depth = cq_depth;
c94aa5ca
ML
1048 }
1049 }
1050 }
1051exit:
1052 return ret;
1053}
1054
d7aaba14 1055static inline int __bfs_forwards(struct lock_list *src_entry,
9e2d551e
ML
1056 void *data,
1057 int (*match)(struct lock_list *entry, void *data),
1058 struct lock_list **target_entry)
c94aa5ca 1059{
9e2d551e 1060 return __bfs(src_entry, data, match, target_entry, 1);
c94aa5ca
ML
1061
1062}
1063
d7aaba14 1064static inline int __bfs_backwards(struct lock_list *src_entry,
9e2d551e
ML
1065 void *data,
1066 int (*match)(struct lock_list *entry, void *data),
1067 struct lock_list **target_entry)
c94aa5ca 1068{
9e2d551e 1069 return __bfs(src_entry, data, match, target_entry, 0);
c94aa5ca
ML
1070
1071}
1072
8e18257d
PZ
1073/*
1074 * Recursive, forwards-direction lock-dependency checking, used for
1075 * both noncyclic checking and for hardirq-unsafe/softirq-unsafe
1076 * checking.
8e18257d 1077 */
8e18257d
PZ
1078
1079/*
1080 * Print a dependency chain entry (this is only done when a deadlock
1081 * has been detected):
1082 */
1083static noinline int
24208ca7 1084print_circular_bug_entry(struct lock_list *target, int depth)
8e18257d
PZ
1085{
1086 if (debug_locks_silent)
1087 return 0;
1088 printk("\n-> #%u", depth);
1089 print_lock_name(target->class);
f943fe0f 1090 printk(KERN_CONT ":\n");
8e18257d
PZ
1091 print_stack_trace(&target->trace, 6);
1092
1093 return 0;
1094}
1095
f4185812
SR
1096static void
1097print_circular_lock_scenario(struct held_lock *src,
1098 struct held_lock *tgt,
1099 struct lock_list *prt)
1100{
1101 struct lock_class *source = hlock_class(src);
1102 struct lock_class *target = hlock_class(tgt);
1103 struct lock_class *parent = prt->class;
1104
1105 /*
1106 * A direct locking problem where unsafe_class lock is taken
1107 * directly by safe_class lock, then all we need to show
1108 * is the deadlock scenario, as it is obvious that the
1109 * unsafe lock is taken under the safe lock.
1110 *
1111 * But if there is a chain instead, where the safe lock takes
1112 * an intermediate lock (middle_class) where this lock is
1113 * not the same as the safe lock, then the lock chain is
1114 * used to describe the problem. Otherwise we would need
1115 * to show a different CPU case for each link in the chain
1116 * from the safe_class lock to the unsafe_class lock.
1117 */
1118 if (parent != source) {
1119 printk("Chain exists of:\n ");
1120 __print_lock_name(source);
f943fe0f 1121 printk(KERN_CONT " --> ");
f4185812 1122 __print_lock_name(parent);
f943fe0f 1123 printk(KERN_CONT " --> ");
f4185812 1124 __print_lock_name(target);
f943fe0f 1125 printk(KERN_CONT "\n\n");
f4185812
SR
1126 }
1127
1128 printk(" Possible unsafe locking scenario:\n\n");
1129 printk(" CPU0 CPU1\n");
1130 printk(" ---- ----\n");
1131 printk(" lock(");
1132 __print_lock_name(target);
f943fe0f 1133 printk(KERN_CONT ");\n");
f4185812
SR
1134 printk(" lock(");
1135 __print_lock_name(parent);
f943fe0f 1136 printk(KERN_CONT ");\n");
f4185812
SR
1137 printk(" lock(");
1138 __print_lock_name(target);
f943fe0f 1139 printk(KERN_CONT ");\n");
f4185812
SR
1140 printk(" lock(");
1141 __print_lock_name(source);
f943fe0f 1142 printk(KERN_CONT ");\n");
f4185812
SR
1143 printk("\n *** DEADLOCK ***\n\n");
1144}
1145
8e18257d
PZ
1146/*
1147 * When a circular dependency is detected, print the
1148 * header first:
1149 */
1150static noinline int
db0002a3
ML
1151print_circular_bug_header(struct lock_list *entry, unsigned int depth,
1152 struct held_lock *check_src,
1153 struct held_lock *check_tgt)
8e18257d
PZ
1154{
1155 struct task_struct *curr = current;
1156
c94aa5ca 1157 if (debug_locks_silent)
8e18257d
PZ
1158 return 0;
1159
b3fbab05
PM
1160 printk("\n");
1161 printk("======================================================\n");
1162 printk("[ INFO: possible circular locking dependency detected ]\n");
fbdc4b9a 1163 print_kernel_ident();
b3fbab05 1164 printk("-------------------------------------------------------\n");
8e18257d 1165 printk("%s/%d is trying to acquire lock:\n",
ba25f9dc 1166 curr->comm, task_pid_nr(curr));
db0002a3 1167 print_lock(check_src);
8e18257d 1168 printk("\nbut task is already holding lock:\n");
db0002a3 1169 print_lock(check_tgt);
8e18257d
PZ
1170 printk("\nwhich lock already depends on the new lock.\n\n");
1171 printk("\nthe existing dependency chain (in reverse order) is:\n");
1172
1173 print_circular_bug_entry(entry, depth);
1174
1175 return 0;
1176}
1177
9e2d551e
ML
1178static inline int class_equal(struct lock_list *entry, void *data)
1179{
1180 return entry->class == data;
1181}
1182
db0002a3
ML
1183static noinline int print_circular_bug(struct lock_list *this,
1184 struct lock_list *target,
1185 struct held_lock *check_src,
1186 struct held_lock *check_tgt)
8e18257d
PZ
1187{
1188 struct task_struct *curr = current;
c94aa5ca 1189 struct lock_list *parent;
f4185812 1190 struct lock_list *first_parent;
24208ca7 1191 int depth;
8e18257d 1192
c94aa5ca 1193 if (!debug_locks_off_graph_unlock() || debug_locks_silent)
8e18257d
PZ
1194 return 0;
1195
db0002a3 1196 if (!save_trace(&this->trace))
8e18257d
PZ
1197 return 0;
1198
c94aa5ca
ML
1199 depth = get_lock_depth(target);
1200
db0002a3 1201 print_circular_bug_header(target, depth, check_src, check_tgt);
c94aa5ca
ML
1202
1203 parent = get_lock_parent(target);
f4185812 1204 first_parent = parent;
c94aa5ca
ML
1205
1206 while (parent) {
1207 print_circular_bug_entry(parent, --depth);
1208 parent = get_lock_parent(parent);
1209 }
8e18257d
PZ
1210
1211 printk("\nother info that might help us debug this:\n\n");
f4185812
SR
1212 print_circular_lock_scenario(check_src, check_tgt,
1213 first_parent);
1214
8e18257d
PZ
1215 lockdep_print_held_locks(curr);
1216
1217 printk("\nstack backtrace:\n");
1218 dump_stack();
1219
1220 return 0;
1221}
1222
db0002a3
ML
1223static noinline int print_bfs_bug(int ret)
1224{
1225 if (!debug_locks_off_graph_unlock())
1226 return 0;
1227
0119fee4
PZ
1228 /*
1229 * Breadth-first-search failed, graph got corrupted?
1230 */
db0002a3
ML
1231 WARN(1, "lockdep bfs error:%d\n", ret);
1232
1233 return 0;
1234}
1235
ef681026 1236static int noop_count(struct lock_list *entry, void *data)
419ca3f1 1237{
ef681026
ML
1238 (*(unsigned long *)data)++;
1239 return 0;
1240}
419ca3f1 1241
5216d530 1242static unsigned long __lockdep_count_forward_deps(struct lock_list *this)
ef681026
ML
1243{
1244 unsigned long count = 0;
1245 struct lock_list *uninitialized_var(target_entry);
419ca3f1 1246
ef681026 1247 __bfs_forwards(this, (void *)&count, noop_count, &target_entry);
419ca3f1 1248
ef681026 1249 return count;
419ca3f1 1250}
419ca3f1
DM
1251unsigned long lockdep_count_forward_deps(struct lock_class *class)
1252{
1253 unsigned long ret, flags;
ef681026
ML
1254 struct lock_list this;
1255
1256 this.parent = NULL;
1257 this.class = class;
419ca3f1
DM
1258
1259 local_irq_save(flags);
0199c4e6 1260 arch_spin_lock(&lockdep_lock);
ef681026 1261 ret = __lockdep_count_forward_deps(&this);
0199c4e6 1262 arch_spin_unlock(&lockdep_lock);
419ca3f1
DM
1263 local_irq_restore(flags);
1264
1265 return ret;
1266}
1267
5216d530 1268static unsigned long __lockdep_count_backward_deps(struct lock_list *this)
419ca3f1 1269{
ef681026
ML
1270 unsigned long count = 0;
1271 struct lock_list *uninitialized_var(target_entry);
419ca3f1 1272
ef681026 1273 __bfs_backwards(this, (void *)&count, noop_count, &target_entry);
419ca3f1 1274
ef681026 1275 return count;
419ca3f1
DM
1276}
1277
1278unsigned long lockdep_count_backward_deps(struct lock_class *class)
1279{
1280 unsigned long ret, flags;
ef681026
ML
1281 struct lock_list this;
1282
1283 this.parent = NULL;
1284 this.class = class;
419ca3f1
DM
1285
1286 local_irq_save(flags);
0199c4e6 1287 arch_spin_lock(&lockdep_lock);
ef681026 1288 ret = __lockdep_count_backward_deps(&this);
0199c4e6 1289 arch_spin_unlock(&lockdep_lock);
419ca3f1
DM
1290 local_irq_restore(flags);
1291
1292 return ret;
1293}
1294
8e18257d
PZ
1295/*
1296 * Prove that the dependency graph starting at <entry> can not
1297 * lead to <target>. Print an error and return 0 if it does.
1298 */
1299static noinline int
db0002a3
ML
1300check_noncircular(struct lock_list *root, struct lock_class *target,
1301 struct lock_list **target_entry)
8e18257d 1302{
db0002a3 1303 int result;
8e18257d 1304
bd6d29c2 1305 debug_atomic_inc(nr_cyclic_checks);
419ca3f1 1306
d7aaba14 1307 result = __bfs_forwards(root, target, class_equal, target_entry);
fbb9ce95 1308
db0002a3
ML
1309 return result;
1310}
c94aa5ca 1311
81d68a96 1312#if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING)
fbb9ce95
IM
1313/*
1314 * Forwards and backwards subgraph searching, for the purposes of
1315 * proving that two subgraphs can be connected by a new dependency
1316 * without creating any illegal irq-safe -> irq-unsafe lock dependency.
1317 */
fbb9ce95 1318
d7aaba14
ML
1319static inline int usage_match(struct lock_list *entry, void *bit)
1320{
1321 return entry->class->usage_mask & (1 << (enum lock_usage_bit)bit);
1322}
1323
1324
1325
fbb9ce95
IM
1326/*
1327 * Find a node in the forwards-direction dependency sub-graph starting
d7aaba14 1328 * at @root->class that matches @bit.
fbb9ce95 1329 *
d7aaba14
ML
1330 * Return 0 if such a node exists in the subgraph, and put that node
1331 * into *@target_entry.
fbb9ce95 1332 *
d7aaba14
ML
1333 * Return 1 otherwise and keep *@target_entry unchanged.
1334 * Return <0 on error.
fbb9ce95 1335 */
d7aaba14
ML
1336static int
1337find_usage_forwards(struct lock_list *root, enum lock_usage_bit bit,
1338 struct lock_list **target_entry)
fbb9ce95 1339{
d7aaba14 1340 int result;
fbb9ce95 1341
bd6d29c2 1342 debug_atomic_inc(nr_find_usage_forwards_checks);
fbb9ce95 1343
d7aaba14
ML
1344 result = __bfs_forwards(root, (void *)bit, usage_match, target_entry);
1345
1346 return result;
fbb9ce95
IM
1347}
1348
1349/*
1350 * Find a node in the backwards-direction dependency sub-graph starting
d7aaba14 1351 * at @root->class that matches @bit.
fbb9ce95 1352 *
d7aaba14
ML
1353 * Return 0 if such a node exists in the subgraph, and put that node
1354 * into *@target_entry.
fbb9ce95 1355 *
d7aaba14
ML
1356 * Return 1 otherwise and keep *@target_entry unchanged.
1357 * Return <0 on error.
fbb9ce95 1358 */
d7aaba14
ML
1359static int
1360find_usage_backwards(struct lock_list *root, enum lock_usage_bit bit,
1361 struct lock_list **target_entry)
fbb9ce95 1362{
d7aaba14 1363 int result;
fbb9ce95 1364
bd6d29c2 1365 debug_atomic_inc(nr_find_usage_backwards_checks);
fbb9ce95 1366
d7aaba14 1367 result = __bfs_backwards(root, (void *)bit, usage_match, target_entry);
f82b217e 1368
d7aaba14 1369 return result;
fbb9ce95
IM
1370}
1371
af012961
PZ
1372static void print_lock_class_header(struct lock_class *class, int depth)
1373{
1374 int bit;
1375
1376 printk("%*s->", depth, "");
1377 print_lock_name(class);
f943fe0f
DV
1378 printk(KERN_CONT " ops: %lu", class->ops);
1379 printk(KERN_CONT " {\n");
af012961
PZ
1380
1381 for (bit = 0; bit < LOCK_USAGE_STATES; bit++) {
1382 if (class->usage_mask & (1 << bit)) {
1383 int len = depth;
1384
1385 len += printk("%*s %s", depth, "", usage_str[bit]);
f943fe0f 1386 len += printk(KERN_CONT " at:\n");
af012961
PZ
1387 print_stack_trace(class->usage_traces + bit, len);
1388 }
1389 }
1390 printk("%*s }\n", depth, "");
1391
f943fe0f
DV
1392 printk("%*s ... key at: [<%p>] %pS\n",
1393 depth, "", class->key, class->key);
af012961
PZ
1394}
1395
1396/*
1397 * printk the shortest lock dependencies from @start to @end in reverse order:
1398 */
1399static void __used
1400print_shortest_lock_dependencies(struct lock_list *leaf,
1401 struct lock_list *root)
1402{
1403 struct lock_list *entry = leaf;
1404 int depth;
1405
1406 /*compute depth from generated tree by BFS*/
1407 depth = get_lock_depth(leaf);
1408
1409 do {
1410 print_lock_class_header(entry->class, depth);
1411 printk("%*s ... acquired at:\n", depth, "");
1412 print_stack_trace(&entry->trace, 2);
1413 printk("\n");
1414
1415 if (depth == 0 && (entry != root)) {
6be8c393 1416 printk("lockdep:%s bad path found in chain graph\n", __func__);
af012961
PZ
1417 break;
1418 }
1419
1420 entry = get_lock_parent(entry);
1421 depth--;
1422 } while (entry && (depth >= 0));
1423
1424 return;
1425}
d7aaba14 1426
3003eba3
SR
1427static void
1428print_irq_lock_scenario(struct lock_list *safe_entry,
1429 struct lock_list *unsafe_entry,
dad3d743
SR
1430 struct lock_class *prev_class,
1431 struct lock_class *next_class)
3003eba3
SR
1432{
1433 struct lock_class *safe_class = safe_entry->class;
1434 struct lock_class *unsafe_class = unsafe_entry->class;
dad3d743 1435 struct lock_class *middle_class = prev_class;
3003eba3
SR
1436
1437 if (middle_class == safe_class)
dad3d743 1438 middle_class = next_class;
3003eba3
SR
1439
1440 /*
1441 * A direct locking problem where unsafe_class lock is taken
1442 * directly by safe_class lock, then all we need to show
1443 * is the deadlock scenario, as it is obvious that the
1444 * unsafe lock is taken under the safe lock.
1445 *
1446 * But if there is a chain instead, where the safe lock takes
1447 * an intermediate lock (middle_class) where this lock is
1448 * not the same as the safe lock, then the lock chain is
1449 * used to describe the problem. Otherwise we would need
1450 * to show a different CPU case for each link in the chain
1451 * from the safe_class lock to the unsafe_class lock.
1452 */
1453 if (middle_class != unsafe_class) {
1454 printk("Chain exists of:\n ");
1455 __print_lock_name(safe_class);
f943fe0f 1456 printk(KERN_CONT " --> ");
3003eba3 1457 __print_lock_name(middle_class);
f943fe0f 1458 printk(KERN_CONT " --> ");
3003eba3 1459 __print_lock_name(unsafe_class);
f943fe0f 1460 printk(KERN_CONT "\n\n");
3003eba3
SR
1461 }
1462
1463 printk(" Possible interrupt unsafe locking scenario:\n\n");
1464 printk(" CPU0 CPU1\n");
1465 printk(" ---- ----\n");
1466 printk(" lock(");
1467 __print_lock_name(unsafe_class);
f943fe0f 1468 printk(KERN_CONT ");\n");
3003eba3
SR
1469 printk(" local_irq_disable();\n");
1470 printk(" lock(");
1471 __print_lock_name(safe_class);
f943fe0f 1472 printk(KERN_CONT ");\n");
3003eba3
SR
1473 printk(" lock(");
1474 __print_lock_name(middle_class);
f943fe0f 1475 printk(KERN_CONT ");\n");
3003eba3
SR
1476 printk(" <Interrupt>\n");
1477 printk(" lock(");
1478 __print_lock_name(safe_class);
f943fe0f 1479 printk(KERN_CONT ");\n");
3003eba3
SR
1480 printk("\n *** DEADLOCK ***\n\n");
1481}
1482
fbb9ce95
IM
1483static int
1484print_bad_irq_dependency(struct task_struct *curr,
24208ca7
ML
1485 struct lock_list *prev_root,
1486 struct lock_list *next_root,
1487 struct lock_list *backwards_entry,
1488 struct lock_list *forwards_entry,
fbb9ce95
IM
1489 struct held_lock *prev,
1490 struct held_lock *next,
1491 enum lock_usage_bit bit1,
1492 enum lock_usage_bit bit2,
1493 const char *irqclass)
1494{
74c383f1 1495 if (!debug_locks_off_graph_unlock() || debug_locks_silent)
fbb9ce95
IM
1496 return 0;
1497
b3fbab05
PM
1498 printk("\n");
1499 printk("======================================================\n");
1500 printk("[ INFO: %s-safe -> %s-unsafe lock order detected ]\n",
fbb9ce95 1501 irqclass, irqclass);
fbdc4b9a 1502 print_kernel_ident();
b3fbab05 1503 printk("------------------------------------------------------\n");
fbb9ce95 1504 printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] is trying to acquire:\n",
ba25f9dc 1505 curr->comm, task_pid_nr(curr),
fbb9ce95
IM
1506 curr->hardirq_context, hardirq_count() >> HARDIRQ_SHIFT,
1507 curr->softirq_context, softirq_count() >> SOFTIRQ_SHIFT,
1508 curr->hardirqs_enabled,
1509 curr->softirqs_enabled);
1510 print_lock(next);
1511
1512 printk("\nand this task is already holding:\n");
1513 print_lock(prev);
1514 printk("which would create a new lock dependency:\n");
f82b217e 1515 print_lock_name(hlock_class(prev));
f943fe0f 1516 printk(KERN_CONT " ->");
f82b217e 1517 print_lock_name(hlock_class(next));
f943fe0f 1518 printk(KERN_CONT "\n");
fbb9ce95
IM
1519
1520 printk("\nbut this new dependency connects a %s-irq-safe lock:\n",
1521 irqclass);
24208ca7 1522 print_lock_name(backwards_entry->class);
fbb9ce95
IM
1523 printk("\n... which became %s-irq-safe at:\n", irqclass);
1524
24208ca7 1525 print_stack_trace(backwards_entry->class->usage_traces + bit1, 1);
fbb9ce95
IM
1526
1527 printk("\nto a %s-irq-unsafe lock:\n", irqclass);
24208ca7 1528 print_lock_name(forwards_entry->class);
fbb9ce95
IM
1529 printk("\n... which became %s-irq-unsafe at:\n", irqclass);
1530 printk("...");
1531
24208ca7 1532 print_stack_trace(forwards_entry->class->usage_traces + bit2, 1);
fbb9ce95
IM
1533
1534 printk("\nother info that might help us debug this:\n\n");
dad3d743
SR
1535 print_irq_lock_scenario(backwards_entry, forwards_entry,
1536 hlock_class(prev), hlock_class(next));
3003eba3 1537
fbb9ce95
IM
1538 lockdep_print_held_locks(curr);
1539
f943fe0f 1540 printk("\nthe dependencies between %s-irq-safe lock and the holding lock:\n", irqclass);
24208ca7
ML
1541 if (!save_trace(&prev_root->trace))
1542 return 0;
1543 print_shortest_lock_dependencies(backwards_entry, prev_root);
fbb9ce95 1544
24208ca7
ML
1545 printk("\nthe dependencies between the lock to be acquired");
1546 printk(" and %s-irq-unsafe lock:\n", irqclass);
1547 if (!save_trace(&next_root->trace))
1548 return 0;
1549 print_shortest_lock_dependencies(forwards_entry, next_root);
fbb9ce95
IM
1550
1551 printk("\nstack backtrace:\n");
1552 dump_stack();
1553
1554 return 0;
1555}
1556
1557static int
1558check_usage(struct task_struct *curr, struct held_lock *prev,
1559 struct held_lock *next, enum lock_usage_bit bit_backwards,
1560 enum lock_usage_bit bit_forwards, const char *irqclass)
1561{
1562 int ret;
24208ca7 1563 struct lock_list this, that;
d7aaba14 1564 struct lock_list *uninitialized_var(target_entry);
24208ca7 1565 struct lock_list *uninitialized_var(target_entry1);
d7aaba14
ML
1566
1567 this.parent = NULL;
1568
1569 this.class = hlock_class(prev);
1570 ret = find_usage_backwards(&this, bit_backwards, &target_entry);
af012961
PZ
1571 if (ret < 0)
1572 return print_bfs_bug(ret);
1573 if (ret == 1)
1574 return ret;
d7aaba14 1575
24208ca7
ML
1576 that.parent = NULL;
1577 that.class = hlock_class(next);
1578 ret = find_usage_forwards(&that, bit_forwards, &target_entry1);
af012961
PZ
1579 if (ret < 0)
1580 return print_bfs_bug(ret);
1581 if (ret == 1)
1582 return ret;
fbb9ce95 1583
24208ca7
ML
1584 return print_bad_irq_dependency(curr, &this, &that,
1585 target_entry, target_entry1,
1586 prev, next,
fbb9ce95
IM
1587 bit_backwards, bit_forwards, irqclass);
1588}
1589
4f367d8a
PZ
1590static const char *state_names[] = {
1591#define LOCKDEP_STATE(__STATE) \
b4b136f4 1592 __stringify(__STATE),
4f367d8a
PZ
1593#include "lockdep_states.h"
1594#undef LOCKDEP_STATE
1595};
1596
1597static const char *state_rnames[] = {
1598#define LOCKDEP_STATE(__STATE) \
b4b136f4 1599 __stringify(__STATE)"-READ",
4f367d8a
PZ
1600#include "lockdep_states.h"
1601#undef LOCKDEP_STATE
1602};
1603
1604static inline const char *state_name(enum lock_usage_bit bit)
8e18257d 1605{
4f367d8a
PZ
1606 return (bit & 1) ? state_rnames[bit >> 2] : state_names[bit >> 2];
1607}
8e18257d 1608
4f367d8a
PZ
1609static int exclusive_bit(int new_bit)
1610{
8e18257d 1611 /*
4f367d8a
PZ
1612 * USED_IN
1613 * USED_IN_READ
1614 * ENABLED
1615 * ENABLED_READ
1616 *
1617 * bit 0 - write/read
1618 * bit 1 - used_in/enabled
1619 * bit 2+ state
8e18257d 1620 */
4f367d8a
PZ
1621
1622 int state = new_bit & ~3;
1623 int dir = new_bit & 2;
8e18257d
PZ
1624
1625 /*
4f367d8a 1626 * keep state, bit flip the direction and strip read.
8e18257d 1627 */
4f367d8a
PZ
1628 return state | (dir ^ 2);
1629}
1630
1631static int check_irq_usage(struct task_struct *curr, struct held_lock *prev,
1632 struct held_lock *next, enum lock_usage_bit bit)
1633{
8e18257d 1634 /*
4f367d8a
PZ
1635 * Prove that the new dependency does not connect a hardirq-safe
1636 * lock with a hardirq-unsafe lock - to achieve this we search
8e18257d
PZ
1637 * the backwards-subgraph starting at <prev>, and the
1638 * forwards-subgraph starting at <next>:
1639 */
4f367d8a
PZ
1640 if (!check_usage(curr, prev, next, bit,
1641 exclusive_bit(bit), state_name(bit)))
8e18257d
PZ
1642 return 0;
1643
4f367d8a
PZ
1644 bit++; /* _READ */
1645
cf40bd16 1646 /*
4f367d8a
PZ
1647 * Prove that the new dependency does not connect a hardirq-safe-read
1648 * lock with a hardirq-unsafe lock - to achieve this we search
cf40bd16
NP
1649 * the backwards-subgraph starting at <prev>, and the
1650 * forwards-subgraph starting at <next>:
1651 */
4f367d8a
PZ
1652 if (!check_usage(curr, prev, next, bit,
1653 exclusive_bit(bit), state_name(bit)))
cf40bd16
NP
1654 return 0;
1655
4f367d8a
PZ
1656 return 1;
1657}
1658
1659static int
1660check_prev_add_irq(struct task_struct *curr, struct held_lock *prev,
1661 struct held_lock *next)
1662{
1663#define LOCKDEP_STATE(__STATE) \
1664 if (!check_irq_usage(curr, prev, next, LOCK_USED_IN_##__STATE)) \
cf40bd16 1665 return 0;
4f367d8a
PZ
1666#include "lockdep_states.h"
1667#undef LOCKDEP_STATE
cf40bd16 1668
8e18257d
PZ
1669 return 1;
1670}
1671
1672static void inc_chains(void)
1673{
1674 if (current->hardirq_context)
1675 nr_hardirq_chains++;
1676 else {
1677 if (current->softirq_context)
1678 nr_softirq_chains++;
1679 else
1680 nr_process_chains++;
1681 }
1682}
1683
1684#else
1685
1686static inline int
1687check_prev_add_irq(struct task_struct *curr, struct held_lock *prev,
1688 struct held_lock *next)
1689{
1690 return 1;
1691}
1692
1693static inline void inc_chains(void)
1694{
1695 nr_process_chains++;
1696}
1697
fbb9ce95
IM
1698#endif
1699
48702ecf
SR
1700static void
1701print_deadlock_scenario(struct held_lock *nxt,
1702 struct held_lock *prv)
1703{
1704 struct lock_class *next = hlock_class(nxt);
1705 struct lock_class *prev = hlock_class(prv);
1706
1707 printk(" Possible unsafe locking scenario:\n\n");
1708 printk(" CPU0\n");
1709 printk(" ----\n");
1710 printk(" lock(");
1711 __print_lock_name(prev);
f943fe0f 1712 printk(KERN_CONT ");\n");
48702ecf
SR
1713 printk(" lock(");
1714 __print_lock_name(next);
f943fe0f 1715 printk(KERN_CONT ");\n");
48702ecf
SR
1716 printk("\n *** DEADLOCK ***\n\n");
1717 printk(" May be due to missing lock nesting notation\n\n");
1718}
1719
fbb9ce95
IM
1720static int
1721print_deadlock_bug(struct task_struct *curr, struct held_lock *prev,
1722 struct held_lock *next)
1723{
74c383f1 1724 if (!debug_locks_off_graph_unlock() || debug_locks_silent)
fbb9ce95
IM
1725 return 0;
1726
b3fbab05
PM
1727 printk("\n");
1728 printk("=============================================\n");
1729 printk("[ INFO: possible recursive locking detected ]\n");
fbdc4b9a 1730 print_kernel_ident();
b3fbab05 1731 printk("---------------------------------------------\n");
fbb9ce95 1732 printk("%s/%d is trying to acquire lock:\n",
ba25f9dc 1733 curr->comm, task_pid_nr(curr));
fbb9ce95
IM
1734 print_lock(next);
1735 printk("\nbut task is already holding lock:\n");
1736 print_lock(prev);
1737
1738 printk("\nother info that might help us debug this:\n");
48702ecf 1739 print_deadlock_scenario(next, prev);
fbb9ce95
IM
1740 lockdep_print_held_locks(curr);
1741
1742 printk("\nstack backtrace:\n");
1743 dump_stack();
1744
1745 return 0;
1746}
1747
1748/*
1749 * Check whether we are holding such a class already.
1750 *
1751 * (Note that this has to be done separately, because the graph cannot
1752 * detect such classes of deadlocks.)
1753 *
1754 * Returns: 0 on deadlock detected, 1 on OK, 2 on recursive read
1755 */
1756static int
1757check_deadlock(struct task_struct *curr, struct held_lock *next,
1758 struct lockdep_map *next_instance, int read)
1759{
1760 struct held_lock *prev;
7531e2f3 1761 struct held_lock *nest = NULL;
fbb9ce95
IM
1762 int i;
1763
1764 for (i = 0; i < curr->lockdep_depth; i++) {
1765 prev = curr->held_locks + i;
7531e2f3
PZ
1766
1767 if (prev->instance == next->nest_lock)
1768 nest = prev;
1769
f82b217e 1770 if (hlock_class(prev) != hlock_class(next))
fbb9ce95 1771 continue;
7531e2f3 1772
fbb9ce95
IM
1773 /*
1774 * Allow read-after-read recursion of the same
6c9076ec 1775 * lock class (i.e. read_lock(lock)+read_lock(lock)):
fbb9ce95 1776 */
6c9076ec 1777 if ((read == 2) && prev->read)
fbb9ce95 1778 return 2;
7531e2f3
PZ
1779
1780 /*
1781 * We're holding the nest_lock, which serializes this lock's
1782 * nesting behaviour.
1783 */
1784 if (nest)
1785 return 2;
1786
fbb9ce95
IM
1787 return print_deadlock_bug(curr, prev, next);
1788 }
1789 return 1;
1790}
1791
1792/*
1793 * There was a chain-cache miss, and we are about to add a new dependency
1794 * to a previous lock. We recursively validate the following rules:
1795 *
1796 * - would the adding of the <prev> -> <next> dependency create a
1797 * circular dependency in the graph? [== circular deadlock]
1798 *
1799 * - does the new prev->next dependency connect any hardirq-safe lock
1800 * (in the full backwards-subgraph starting at <prev>) with any
1801 * hardirq-unsafe lock (in the full forwards-subgraph starting at
1802 * <next>)? [== illegal lock inversion with hardirq contexts]
1803 *
1804 * - does the new prev->next dependency connect any softirq-safe lock
1805 * (in the full backwards-subgraph starting at <prev>) with any
1806 * softirq-unsafe lock (in the full forwards-subgraph starting at
1807 * <next>)? [== illegal lock inversion with softirq contexts]
1808 *
1809 * any of these scenarios could lead to a deadlock.
1810 *
1811 * Then if all the validations pass, we add the forwards and backwards
1812 * dependency.
1813 */
1814static int
1815check_prev_add(struct task_struct *curr, struct held_lock *prev,
8a5fd564 1816 struct held_lock *next, int distance, int *stack_saved)
fbb9ce95
IM
1817{
1818 struct lock_list *entry;
1819 int ret;
db0002a3
ML
1820 struct lock_list this;
1821 struct lock_list *uninitialized_var(target_entry);
4726f2a6
YZ
1822 /*
1823 * Static variable, serialized by the graph_lock().
1824 *
1825 * We use this static variable to save the stack trace in case
1826 * we call into this function multiple times due to encountering
1827 * trylocks in the held lock stack.
1828 */
1829 static struct stack_trace trace;
fbb9ce95
IM
1830
1831 /*
1832 * Prove that the new <prev> -> <next> dependency would not
1833 * create a circular dependency in the graph. (We do this by
1834 * forward-recursing into the graph starting at <next>, and
1835 * checking whether we can reach <prev>.)
1836 *
1837 * We are using global variables to control the recursion, to
1838 * keep the stackframe size of the recursive functions low:
1839 */
db0002a3
ML
1840 this.class = hlock_class(next);
1841 this.parent = NULL;
1842 ret = check_noncircular(&this, hlock_class(prev), &target_entry);
1843 if (unlikely(!ret))
1844 return print_circular_bug(&this, target_entry, next, prev);
1845 else if (unlikely(ret < 0))
1846 return print_bfs_bug(ret);
c94aa5ca 1847
8e18257d 1848 if (!check_prev_add_irq(curr, prev, next))
fbb9ce95
IM
1849 return 0;
1850
fbb9ce95
IM
1851 /*
1852 * For recursive read-locks we do all the dependency checks,
1853 * but we dont store read-triggered dependencies (only
1854 * write-triggered dependencies). This ensures that only the
1855 * write-side dependencies matter, and that if for example a
1856 * write-lock never takes any other locks, then the reads are
1857 * equivalent to a NOP.
1858 */
1859 if (next->read == 2 || prev->read == 2)
1860 return 1;
1861 /*
1862 * Is the <prev> -> <next> dependency already present?
1863 *
1864 * (this may occur even though this is a new chain: consider
1865 * e.g. the L1 -> L2 -> L3 -> L4 and the L5 -> L1 -> L2 -> L3
1866 * chains - the second one will be new, but L1 already has
1867 * L2 added to its dependency list, due to the first chain.)
1868 */
f82b217e
DJ
1869 list_for_each_entry(entry, &hlock_class(prev)->locks_after, entry) {
1870 if (entry->class == hlock_class(next)) {
068135e6
JB
1871 if (distance == 1)
1872 entry->distance = 1;
fbb9ce95 1873 return 2;
068135e6 1874 }
fbb9ce95
IM
1875 }
1876
8a5fd564
DV
1877 if (!*stack_saved) {
1878 if (!save_trace(&trace))
1879 return 0;
1880 *stack_saved = 1;
1881 }
4726f2a6 1882
fbb9ce95
IM
1883 /*
1884 * Ok, all validations passed, add the new lock
1885 * to the previous lock's dependency list:
1886 */
83f06168 1887 ret = add_lock_to_list(hlock_class(next),
f82b217e 1888 &hlock_class(prev)->locks_after,
4726f2a6 1889 next->acquire_ip, distance, &trace);
068135e6 1890
fbb9ce95
IM
1891 if (!ret)
1892 return 0;
910b1b2e 1893
83f06168 1894 ret = add_lock_to_list(hlock_class(prev),
f82b217e 1895 &hlock_class(next)->locks_before,
4726f2a6 1896 next->acquire_ip, distance, &trace);
910b1b2e
JP
1897 if (!ret)
1898 return 0;
fbb9ce95
IM
1899
1900 /*
8e18257d
PZ
1901 * Debugging printouts:
1902 */
f82b217e 1903 if (verbose(hlock_class(prev)) || verbose(hlock_class(next))) {
8a5fd564
DV
1904 /* We drop graph lock, so another thread can overwrite trace. */
1905 *stack_saved = 0;
8e18257d
PZ
1906 graph_unlock();
1907 printk("\n new dependency: ");
f82b217e 1908 print_lock_name(hlock_class(prev));
f943fe0f 1909 printk(KERN_CONT " => ");
f82b217e 1910 print_lock_name(hlock_class(next));
f943fe0f 1911 printk(KERN_CONT "\n");
fbb9ce95 1912 dump_stack();
8e18257d 1913 return graph_lock();
fbb9ce95 1914 }
8e18257d
PZ
1915 return 1;
1916}
fbb9ce95 1917
8e18257d
PZ
1918/*
1919 * Add the dependency to all directly-previous locks that are 'relevant'.
1920 * The ones that are relevant are (in increasing distance from curr):
1921 * all consecutive trylock entries and the final non-trylock entry - or
1922 * the end of this context's lock-chain - whichever comes first.
1923 */
1924static int
1925check_prevs_add(struct task_struct *curr, struct held_lock *next)
1926{
1927 int depth = curr->lockdep_depth;
8a5fd564 1928 int stack_saved = 0;
8e18257d 1929 struct held_lock *hlock;
d6d897ce 1930
fbb9ce95 1931 /*
8e18257d
PZ
1932 * Debugging checks.
1933 *
1934 * Depth must not be zero for a non-head lock:
fbb9ce95 1935 */
8e18257d
PZ
1936 if (!depth)
1937 goto out_bug;
fbb9ce95 1938 /*
8e18257d
PZ
1939 * At least two relevant locks must exist for this
1940 * to be a head:
fbb9ce95 1941 */
8e18257d
PZ
1942 if (curr->held_locks[depth].irq_context !=
1943 curr->held_locks[depth-1].irq_context)
1944 goto out_bug;
74c383f1 1945
8e18257d
PZ
1946 for (;;) {
1947 int distance = curr->lockdep_depth - depth + 1;
1b5ff816 1948 hlock = curr->held_locks + depth - 1;
8e18257d
PZ
1949 /*
1950 * Only non-recursive-read entries get new dependencies
1951 * added:
1952 */
1b5ff816 1953 if (hlock->read != 2 && hlock->check) {
4726f2a6 1954 if (!check_prev_add(curr, hlock, next,
8a5fd564 1955 distance, &stack_saved))
8e18257d
PZ
1956 return 0;
1957 /*
1958 * Stop after the first non-trylock entry,
1959 * as non-trylock entries have added their
1960 * own direct dependencies already, so this
1961 * lock is connected to them indirectly:
1962 */
1963 if (!hlock->trylock)
1964 break;
74c383f1 1965 }
8e18257d
PZ
1966 depth--;
1967 /*
1968 * End of lock-stack?
1969 */
1970 if (!depth)
1971 break;
1972 /*
1973 * Stop the search if we cross into another context:
1974 */
1975 if (curr->held_locks[depth].irq_context !=
1976 curr->held_locks[depth-1].irq_context)
1977 break;
fbb9ce95 1978 }
8e18257d
PZ
1979 return 1;
1980out_bug:
1981 if (!debug_locks_off_graph_unlock())
1982 return 0;
fbb9ce95 1983
0119fee4
PZ
1984 /*
1985 * Clearly we all shouldn't be here, but since we made it we
1986 * can reliable say we messed up our state. See the above two
1987 * gotos for reasons why we could possibly end up here.
1988 */
8e18257d 1989 WARN_ON(1);
fbb9ce95 1990
8e18257d 1991 return 0;
fbb9ce95
IM
1992}
1993
8e18257d 1994unsigned long nr_lock_chains;
443cd507 1995struct lock_chain lock_chains[MAX_LOCKDEP_CHAINS];
cd1a28e8 1996int nr_chain_hlocks;
443cd507
HY
1997static u16 chain_hlocks[MAX_LOCKDEP_CHAIN_HLOCKS];
1998
1999struct lock_class *lock_chain_get_class(struct lock_chain *chain, int i)
2000{
2001 return lock_classes + chain_hlocks[chain->base + i];
2002}
8e18257d 2003
9e4e7554
IM
2004/*
2005 * Returns the index of the first held_lock of the current chain
2006 */
2007static inline int get_first_held_lock(struct task_struct *curr,
2008 struct held_lock *hlock)
2009{
2010 int i;
2011 struct held_lock *hlock_curr;
2012
2013 for (i = curr->lockdep_depth - 1; i >= 0; i--) {
2014 hlock_curr = curr->held_locks + i;
2015 if (hlock_curr->irq_context != hlock->irq_context)
2016 break;
2017
2018 }
2019
2020 return ++i;
2021}
2022
5c8a010c 2023#ifdef CONFIG_DEBUG_LOCKDEP
39e2e173
AAF
2024/*
2025 * Returns the next chain_key iteration
2026 */
2027static u64 print_chain_key_iteration(int class_idx, u64 chain_key)
2028{
2029 u64 new_chain_key = iterate_chain_key(chain_key, class_idx);
2030
2031 printk(" class_idx:%d -> chain_key:%016Lx",
2032 class_idx,
2033 (unsigned long long)new_chain_key);
2034 return new_chain_key;
2035}
2036
2037static void
2038print_chain_keys_held_locks(struct task_struct *curr, struct held_lock *hlock_next)
2039{
2040 struct held_lock *hlock;
2041 u64 chain_key = 0;
2042 int depth = curr->lockdep_depth;
2043 int i;
2044
2045 printk("depth: %u\n", depth + 1);
2046 for (i = get_first_held_lock(curr, hlock_next); i < depth; i++) {
2047 hlock = curr->held_locks + i;
2048 chain_key = print_chain_key_iteration(hlock->class_idx, chain_key);
2049
2050 print_lock(hlock);
2051 }
2052
2053 print_chain_key_iteration(hlock_next->class_idx, chain_key);
2054 print_lock(hlock_next);
2055}
2056
2057static void print_chain_keys_chain(struct lock_chain *chain)
2058{
2059 int i;
2060 u64 chain_key = 0;
2061 int class_id;
2062
2063 printk("depth: %u\n", chain->depth);
2064 for (i = 0; i < chain->depth; i++) {
2065 class_id = chain_hlocks[chain->base + i];
2066 chain_key = print_chain_key_iteration(class_id + 1, chain_key);
2067
2068 print_lock_name(lock_classes + class_id);
2069 printk("\n");
2070 }
2071}
2072
2073static void print_collision(struct task_struct *curr,
2074 struct held_lock *hlock_next,
2075 struct lock_chain *chain)
2076{
2077 printk("\n");
2078 printk("======================\n");
2079 printk("[chain_key collision ]\n");
2080 print_kernel_ident();
2081 printk("----------------------\n");
2082 printk("%s/%d: ", current->comm, task_pid_nr(current));
2083 printk("Hash chain already cached but the contents don't match!\n");
2084
2085 printk("Held locks:");
2086 print_chain_keys_held_locks(curr, hlock_next);
2087
2088 printk("Locks in cached chain:");
2089 print_chain_keys_chain(chain);
2090
2091 printk("\nstack backtrace:\n");
2092 dump_stack();
2093}
5c8a010c 2094#endif
39e2e173 2095
9e4e7554
IM
2096/*
2097 * Checks whether the chain and the current held locks are consistent
2098 * in depth and also in content. If they are not it most likely means
2099 * that there was a collision during the calculation of the chain_key.
2100 * Returns: 0 not passed, 1 passed
2101 */
2102static int check_no_collision(struct task_struct *curr,
2103 struct held_lock *hlock,
2104 struct lock_chain *chain)
2105{
2106#ifdef CONFIG_DEBUG_LOCKDEP
2107 int i, j, id;
2108
2109 i = get_first_held_lock(curr, hlock);
2110
39e2e173
AAF
2111 if (DEBUG_LOCKS_WARN_ON(chain->depth != curr->lockdep_depth - (i - 1))) {
2112 print_collision(curr, hlock, chain);
9e4e7554 2113 return 0;
39e2e173 2114 }
9e4e7554
IM
2115
2116 for (j = 0; j < chain->depth - 1; j++, i++) {
2117 id = curr->held_locks[i].class_idx - 1;
2118
39e2e173
AAF
2119 if (DEBUG_LOCKS_WARN_ON(chain_hlocks[chain->base + j] != id)) {
2120 print_collision(curr, hlock, chain);
9e4e7554 2121 return 0;
39e2e173 2122 }
9e4e7554
IM
2123 }
2124#endif
2125 return 1;
2126}
2127
fbb9ce95
IM
2128/*
2129 * Look up a dependency chain. If the key is not present yet then
9e860d00
JP
2130 * add it and return 1 - in this case the new dependency chain is
2131 * validated. If the key is already hashed, return 0.
2132 * (On return with 1 graph_lock is held.)
fbb9ce95 2133 */
443cd507
HY
2134static inline int lookup_chain_cache(struct task_struct *curr,
2135 struct held_lock *hlock,
2136 u64 chain_key)
fbb9ce95 2137{
f82b217e 2138 struct lock_class *class = hlock_class(hlock);
a63f38cc 2139 struct hlist_head *hash_head = chainhashentry(chain_key);
fbb9ce95 2140 struct lock_chain *chain;
e0944ee6 2141 int i, j;
fbb9ce95 2142
0119fee4
PZ
2143 /*
2144 * We might need to take the graph lock, ensure we've got IRQs
2145 * disabled to make this an IRQ-safe lock.. for recursion reasons
2146 * lockdep won't complain about its own locking errors.
2147 */
381a2292
JP
2148 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2149 return 0;
fbb9ce95
IM
2150 /*
2151 * We can walk it lock-free, because entries only get added
2152 * to the hash:
2153 */
a63f38cc 2154 hlist_for_each_entry_rcu(chain, hash_head, entry) {
fbb9ce95
IM
2155 if (chain->chain_key == chain_key) {
2156cache_hit:
bd6d29c2 2157 debug_atomic_inc(chain_lookup_hits);
9e4e7554
IM
2158 if (!check_no_collision(curr, hlock, chain))
2159 return 0;
2160
81fc685a 2161 if (very_verbose(class))
755cd900
AM
2162 printk("\nhash chain already cached, key: "
2163 "%016Lx tail class: [%p] %s\n",
2164 (unsigned long long)chain_key,
2165 class->key, class->name);
fbb9ce95
IM
2166 return 0;
2167 }
2168 }
81fc685a 2169 if (very_verbose(class))
755cd900
AM
2170 printk("\nnew hash chain, key: %016Lx tail class: [%p] %s\n",
2171 (unsigned long long)chain_key, class->key, class->name);
fbb9ce95
IM
2172 /*
2173 * Allocate a new chain entry from the static array, and add
2174 * it to the hash:
2175 */
74c383f1
IM
2176 if (!graph_lock())
2177 return 0;
fbb9ce95
IM
2178 /*
2179 * We have to walk the chain again locked - to avoid duplicates:
2180 */
a63f38cc 2181 hlist_for_each_entry(chain, hash_head, entry) {
fbb9ce95 2182 if (chain->chain_key == chain_key) {
74c383f1 2183 graph_unlock();
fbb9ce95
IM
2184 goto cache_hit;
2185 }
2186 }
2187 if (unlikely(nr_lock_chains >= MAX_LOCKDEP_CHAINS)) {
74c383f1
IM
2188 if (!debug_locks_off_graph_unlock())
2189 return 0;
2190
2c522836 2191 print_lockdep_off("BUG: MAX_LOCKDEP_CHAINS too low!");
eedeeabd 2192 dump_stack();
fbb9ce95
IM
2193 return 0;
2194 }
2195 chain = lock_chains + nr_lock_chains++;
2196 chain->chain_key = chain_key;
443cd507 2197 chain->irq_context = hlock->irq_context;
9e4e7554 2198 i = get_first_held_lock(curr, hlock);
443cd507 2199 chain->depth = curr->lockdep_depth + 1 - i;
75dd602a
PZ
2200
2201 BUILD_BUG_ON((1UL << 24) <= ARRAY_SIZE(chain_hlocks));
2202 BUILD_BUG_ON((1UL << 6) <= ARRAY_SIZE(curr->held_locks));
2203 BUILD_BUG_ON((1UL << 8*sizeof(chain_hlocks[0])) <= ARRAY_SIZE(lock_classes));
2204
e0944ee6
SR
2205 if (likely(nr_chain_hlocks + chain->depth <= MAX_LOCKDEP_CHAIN_HLOCKS)) {
2206 chain->base = nr_chain_hlocks;
443cd507 2207 for (j = 0; j < chain->depth - 1; j++, i++) {
f82b217e 2208 int lock_id = curr->held_locks[i].class_idx - 1;
443cd507
HY
2209 chain_hlocks[chain->base + j] = lock_id;
2210 }
2211 chain_hlocks[chain->base + j] = class - lock_classes;
2212 }
75dd602a
PZ
2213
2214 if (nr_chain_hlocks < MAX_LOCKDEP_CHAIN_HLOCKS)
2215 nr_chain_hlocks += chain->depth;
2216
2217#ifdef CONFIG_DEBUG_LOCKDEP
2218 /*
2219 * Important for check_no_collision().
2220 */
2221 if (unlikely(nr_chain_hlocks > MAX_LOCKDEP_CHAIN_HLOCKS)) {
f9af456a 2222 if (!debug_locks_off_graph_unlock())
75dd602a
PZ
2223 return 0;
2224
2225 print_lockdep_off("BUG: MAX_LOCKDEP_CHAIN_HLOCKS too low!");
2226 dump_stack();
2227 return 0;
2228 }
2229#endif
2230
a63f38cc 2231 hlist_add_head_rcu(&chain->entry, hash_head);
bd6d29c2 2232 debug_atomic_inc(chain_lookup_misses);
8e18257d
PZ
2233 inc_chains();
2234
2235 return 1;
2236}
2237
2238static int validate_chain(struct task_struct *curr, struct lockdep_map *lock,
4e6045f1 2239 struct held_lock *hlock, int chain_head, u64 chain_key)
8e18257d
PZ
2240{
2241 /*
2242 * Trylock needs to maintain the stack of held locks, but it
2243 * does not add new dependencies, because trylock can be done
2244 * in any order.
2245 *
2246 * We look up the chain_key and do the O(N^2) check and update of
2247 * the dependencies only if this is a new dependency chain.
2248 * (If lookup_chain_cache() returns with 1 it acquires
2249 * graph_lock for us)
2250 */
fb9edbe9 2251 if (!hlock->trylock && hlock->check &&
443cd507 2252 lookup_chain_cache(curr, hlock, chain_key)) {
8e18257d
PZ
2253 /*
2254 * Check whether last held lock:
2255 *
2256 * - is irq-safe, if this lock is irq-unsafe
2257 * - is softirq-safe, if this lock is hardirq-unsafe
2258 *
2259 * And check whether the new lock's dependency graph
2260 * could lead back to the previous lock.
2261 *
2262 * any of these scenarios could lead to a deadlock. If
2263 * All validations
2264 */
2265 int ret = check_deadlock(curr, hlock, lock, hlock->read);
2266
2267 if (!ret)
2268 return 0;
2269 /*
2270 * Mark recursive read, as we jump over it when
2271 * building dependencies (just like we jump over
2272 * trylock entries):
2273 */
2274 if (ret == 2)
2275 hlock->read = 2;
2276 /*
2277 * Add dependency only if this lock is not the head
2278 * of the chain, and if it's not a secondary read-lock:
2279 */
2280 if (!chain_head && ret != 2)
2281 if (!check_prevs_add(curr, hlock))
2282 return 0;
2283 graph_unlock();
2284 } else
2285 /* after lookup_chain_cache(): */
2286 if (unlikely(!debug_locks))
2287 return 0;
fbb9ce95
IM
2288
2289 return 1;
2290}
8e18257d
PZ
2291#else
2292static inline int validate_chain(struct task_struct *curr,
2293 struct lockdep_map *lock, struct held_lock *hlock,
3aa416b0 2294 int chain_head, u64 chain_key)
8e18257d
PZ
2295{
2296 return 1;
2297}
ca58abcb 2298#endif
fbb9ce95
IM
2299
2300/*
2301 * We are building curr_chain_key incrementally, so double-check
2302 * it from scratch, to make sure that it's done correctly:
2303 */
1d09daa5 2304static void check_chain_key(struct task_struct *curr)
fbb9ce95
IM
2305{
2306#ifdef CONFIG_DEBUG_LOCKDEP
2307 struct held_lock *hlock, *prev_hlock = NULL;
5f18ab5c 2308 unsigned int i;
fbb9ce95
IM
2309 u64 chain_key = 0;
2310
2311 for (i = 0; i < curr->lockdep_depth; i++) {
2312 hlock = curr->held_locks + i;
2313 if (chain_key != hlock->prev_chain_key) {
2314 debug_locks_off();
0119fee4
PZ
2315 /*
2316 * We got mighty confused, our chain keys don't match
2317 * with what we expect, someone trample on our task state?
2318 */
2df8b1d6 2319 WARN(1, "hm#1, depth: %u [%u], %016Lx != %016Lx\n",
fbb9ce95
IM
2320 curr->lockdep_depth, i,
2321 (unsigned long long)chain_key,
2322 (unsigned long long)hlock->prev_chain_key);
fbb9ce95
IM
2323 return;
2324 }
0119fee4
PZ
2325 /*
2326 * Whoops ran out of static storage again?
2327 */
5f18ab5c 2328 if (DEBUG_LOCKS_WARN_ON(hlock->class_idx > MAX_LOCKDEP_KEYS))
381a2292
JP
2329 return;
2330
fbb9ce95
IM
2331 if (prev_hlock && (prev_hlock->irq_context !=
2332 hlock->irq_context))
2333 chain_key = 0;
5f18ab5c 2334 chain_key = iterate_chain_key(chain_key, hlock->class_idx);
fbb9ce95
IM
2335 prev_hlock = hlock;
2336 }
2337 if (chain_key != curr->curr_chain_key) {
2338 debug_locks_off();
0119fee4
PZ
2339 /*
2340 * More smoking hash instead of calculating it, damn see these
2341 * numbers float.. I bet that a pink elephant stepped on my memory.
2342 */
2df8b1d6 2343 WARN(1, "hm#2, depth: %u [%u], %016Lx != %016Lx\n",
fbb9ce95
IM
2344 curr->lockdep_depth, i,
2345 (unsigned long long)chain_key,
2346 (unsigned long long)curr->curr_chain_key);
fbb9ce95
IM
2347 }
2348#endif
2349}
2350
282b5c2f
SR
2351static void
2352print_usage_bug_scenario(struct held_lock *lock)
2353{
2354 struct lock_class *class = hlock_class(lock);
2355
2356 printk(" Possible unsafe locking scenario:\n\n");
2357 printk(" CPU0\n");
2358 printk(" ----\n");
2359 printk(" lock(");
2360 __print_lock_name(class);
f943fe0f 2361 printk(KERN_CONT ");\n");
282b5c2f
SR
2362 printk(" <Interrupt>\n");
2363 printk(" lock(");
2364 __print_lock_name(class);
f943fe0f 2365 printk(KERN_CONT ");\n");
282b5c2f
SR
2366 printk("\n *** DEADLOCK ***\n\n");
2367}
2368
8e18257d
PZ
2369static int
2370print_usage_bug(struct task_struct *curr, struct held_lock *this,
2371 enum lock_usage_bit prev_bit, enum lock_usage_bit new_bit)
2372{
2373 if (!debug_locks_off_graph_unlock() || debug_locks_silent)
2374 return 0;
2375
b3fbab05
PM
2376 printk("\n");
2377 printk("=================================\n");
2378 printk("[ INFO: inconsistent lock state ]\n");
fbdc4b9a 2379 print_kernel_ident();
b3fbab05 2380 printk("---------------------------------\n");
8e18257d
PZ
2381
2382 printk("inconsistent {%s} -> {%s} usage.\n",
2383 usage_str[prev_bit], usage_str[new_bit]);
2384
2385 printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] takes:\n",
ba25f9dc 2386 curr->comm, task_pid_nr(curr),
8e18257d
PZ
2387 trace_hardirq_context(curr), hardirq_count() >> HARDIRQ_SHIFT,
2388 trace_softirq_context(curr), softirq_count() >> SOFTIRQ_SHIFT,
2389 trace_hardirqs_enabled(curr),
2390 trace_softirqs_enabled(curr));
2391 print_lock(this);
2392
2393 printk("{%s} state was registered at:\n", usage_str[prev_bit]);
f82b217e 2394 print_stack_trace(hlock_class(this)->usage_traces + prev_bit, 1);
8e18257d
PZ
2395
2396 print_irqtrace_events(curr);
2397 printk("\nother info that might help us debug this:\n");
282b5c2f
SR
2398 print_usage_bug_scenario(this);
2399
8e18257d
PZ
2400 lockdep_print_held_locks(curr);
2401
2402 printk("\nstack backtrace:\n");
2403 dump_stack();
2404
2405 return 0;
2406}
2407
2408/*
2409 * Print out an error if an invalid bit is set:
2410 */
2411static inline int
2412valid_state(struct task_struct *curr, struct held_lock *this,
2413 enum lock_usage_bit new_bit, enum lock_usage_bit bad_bit)
2414{
f82b217e 2415 if (unlikely(hlock_class(this)->usage_mask & (1 << bad_bit)))
8e18257d
PZ
2416 return print_usage_bug(curr, this, bad_bit, new_bit);
2417 return 1;
2418}
2419
2420static int mark_lock(struct task_struct *curr, struct held_lock *this,
2421 enum lock_usage_bit new_bit);
2422
81d68a96 2423#if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING)
fbb9ce95
IM
2424
2425/*
2426 * print irq inversion bug:
2427 */
2428static int
24208ca7
ML
2429print_irq_inversion_bug(struct task_struct *curr,
2430 struct lock_list *root, struct lock_list *other,
fbb9ce95
IM
2431 struct held_lock *this, int forwards,
2432 const char *irqclass)
2433{
dad3d743
SR
2434 struct lock_list *entry = other;
2435 struct lock_list *middle = NULL;
2436 int depth;
2437
74c383f1 2438 if (!debug_locks_off_graph_unlock() || debug_locks_silent)
fbb9ce95
IM
2439 return 0;
2440
b3fbab05
PM
2441 printk("\n");
2442 printk("=========================================================\n");
2443 printk("[ INFO: possible irq lock inversion dependency detected ]\n");
fbdc4b9a 2444 print_kernel_ident();
b3fbab05 2445 printk("---------------------------------------------------------\n");
fbb9ce95 2446 printk("%s/%d just changed the state of lock:\n",
ba25f9dc 2447 curr->comm, task_pid_nr(curr));
fbb9ce95
IM
2448 print_lock(this);
2449 if (forwards)
26575e28 2450 printk("but this lock took another, %s-unsafe lock in the past:\n", irqclass);
fbb9ce95 2451 else
26575e28 2452 printk("but this lock was taken by another, %s-safe lock in the past:\n", irqclass);
24208ca7 2453 print_lock_name(other->class);
fbb9ce95
IM
2454 printk("\n\nand interrupts could create inverse lock ordering between them.\n\n");
2455
2456 printk("\nother info that might help us debug this:\n");
dad3d743
SR
2457
2458 /* Find a middle lock (if one exists) */
2459 depth = get_lock_depth(other);
2460 do {
2461 if (depth == 0 && (entry != root)) {
2462 printk("lockdep:%s bad path found in chain graph\n", __func__);
2463 break;
2464 }
2465 middle = entry;
2466 entry = get_lock_parent(entry);
2467 depth--;
2468 } while (entry && entry != root && (depth >= 0));
2469 if (forwards)
2470 print_irq_lock_scenario(root, other,
2471 middle ? middle->class : root->class, other->class);
2472 else
2473 print_irq_lock_scenario(other, root,
2474 middle ? middle->class : other->class, root->class);
2475
fbb9ce95
IM
2476 lockdep_print_held_locks(curr);
2477
24208ca7
ML
2478 printk("\nthe shortest dependencies between 2nd lock and 1st lock:\n");
2479 if (!save_trace(&root->trace))
2480 return 0;
2481 print_shortest_lock_dependencies(other, root);
fbb9ce95
IM
2482
2483 printk("\nstack backtrace:\n");
2484 dump_stack();
2485
2486 return 0;
2487}
2488
2489/*
2490 * Prove that in the forwards-direction subgraph starting at <this>
2491 * there is no lock matching <mask>:
2492 */
2493static int
2494check_usage_forwards(struct task_struct *curr, struct held_lock *this,
2495 enum lock_usage_bit bit, const char *irqclass)
2496{
2497 int ret;
d7aaba14
ML
2498 struct lock_list root;
2499 struct lock_list *uninitialized_var(target_entry);
fbb9ce95 2500
d7aaba14
ML
2501 root.parent = NULL;
2502 root.class = hlock_class(this);
2503 ret = find_usage_forwards(&root, bit, &target_entry);
af012961
PZ
2504 if (ret < 0)
2505 return print_bfs_bug(ret);
2506 if (ret == 1)
2507 return ret;
fbb9ce95 2508
24208ca7 2509 return print_irq_inversion_bug(curr, &root, target_entry,
d7aaba14 2510 this, 1, irqclass);
fbb9ce95
IM
2511}
2512
2513/*
2514 * Prove that in the backwards-direction subgraph starting at <this>
2515 * there is no lock matching <mask>:
2516 */
2517static int
2518check_usage_backwards(struct task_struct *curr, struct held_lock *this,
2519 enum lock_usage_bit bit, const char *irqclass)
2520{
2521 int ret;
d7aaba14
ML
2522 struct lock_list root;
2523 struct lock_list *uninitialized_var(target_entry);
fbb9ce95 2524
d7aaba14
ML
2525 root.parent = NULL;
2526 root.class = hlock_class(this);
2527 ret = find_usage_backwards(&root, bit, &target_entry);
af012961
PZ
2528 if (ret < 0)
2529 return print_bfs_bug(ret);
2530 if (ret == 1)
2531 return ret;
fbb9ce95 2532
24208ca7 2533 return print_irq_inversion_bug(curr, &root, target_entry,
48d50674 2534 this, 0, irqclass);
fbb9ce95
IM
2535}
2536
3117df04 2537void print_irqtrace_events(struct task_struct *curr)
fbb9ce95
IM
2538{
2539 printk("irq event stamp: %u\n", curr->irq_events);
f943fe0f
DV
2540 printk("hardirqs last enabled at (%u): [<%p>] %pS\n",
2541 curr->hardirq_enable_event, (void *)curr->hardirq_enable_ip,
2542 (void *)curr->hardirq_enable_ip);
2543 printk("hardirqs last disabled at (%u): [<%p>] %pS\n",
2544 curr->hardirq_disable_event, (void *)curr->hardirq_disable_ip,
2545 (void *)curr->hardirq_disable_ip);
2546 printk("softirqs last enabled at (%u): [<%p>] %pS\n",
2547 curr->softirq_enable_event, (void *)curr->softirq_enable_ip,
2548 (void *)curr->softirq_enable_ip);
2549 printk("softirqs last disabled at (%u): [<%p>] %pS\n",
2550 curr->softirq_disable_event, (void *)curr->softirq_disable_ip,
2551 (void *)curr->softirq_disable_ip);
fbb9ce95
IM
2552}
2553
cd95302d 2554static int HARDIRQ_verbose(struct lock_class *class)
fbb9ce95 2555{
8e18257d
PZ
2556#if HARDIRQ_VERBOSE
2557 return class_filter(class);
2558#endif
fbb9ce95
IM
2559 return 0;
2560}
2561
cd95302d 2562static int SOFTIRQ_verbose(struct lock_class *class)
fbb9ce95 2563{
8e18257d
PZ
2564#if SOFTIRQ_VERBOSE
2565 return class_filter(class);
2566#endif
2567 return 0;
fbb9ce95
IM
2568}
2569
cd95302d 2570static int RECLAIM_FS_verbose(struct lock_class *class)
cf40bd16
NP
2571{
2572#if RECLAIM_VERBOSE
2573 return class_filter(class);
2574#endif
2575 return 0;
2576}
2577
fbb9ce95
IM
2578#define STRICT_READ_CHECKS 1
2579
cd95302d
PZ
2580static int (*state_verbose_f[])(struct lock_class *class) = {
2581#define LOCKDEP_STATE(__STATE) \
2582 __STATE##_verbose,
2583#include "lockdep_states.h"
2584#undef LOCKDEP_STATE
2585};
2586
2587static inline int state_verbose(enum lock_usage_bit bit,
2588 struct lock_class *class)
2589{
2590 return state_verbose_f[bit >> 2](class);
2591}
2592
42c50d54
PZ
2593typedef int (*check_usage_f)(struct task_struct *, struct held_lock *,
2594 enum lock_usage_bit bit, const char *name);
2595
6a6904d3 2596static int
1c21f14e
PZ
2597mark_lock_irq(struct task_struct *curr, struct held_lock *this,
2598 enum lock_usage_bit new_bit)
6a6904d3 2599{
f989209e 2600 int excl_bit = exclusive_bit(new_bit);
9d3651a2 2601 int read = new_bit & 1;
42c50d54
PZ
2602 int dir = new_bit & 2;
2603
38aa2714
PZ
2604 /*
2605 * mark USED_IN has to look forwards -- to ensure no dependency
2606 * has ENABLED state, which would allow recursion deadlocks.
2607 *
2608 * mark ENABLED has to look backwards -- to ensure no dependee
2609 * has USED_IN state, which, again, would allow recursion deadlocks.
2610 */
42c50d54
PZ
2611 check_usage_f usage = dir ?
2612 check_usage_backwards : check_usage_forwards;
f989209e 2613
38aa2714
PZ
2614 /*
2615 * Validate that this particular lock does not have conflicting
2616 * usage states.
2617 */
6a6904d3
PZ
2618 if (!valid_state(curr, this, new_bit, excl_bit))
2619 return 0;
42c50d54 2620
38aa2714
PZ
2621 /*
2622 * Validate that the lock dependencies don't have conflicting usage
2623 * states.
2624 */
2625 if ((!read || !dir || STRICT_READ_CHECKS) &&
1c21f14e 2626 !usage(curr, this, excl_bit, state_name(new_bit & ~1)))
6a6904d3 2627 return 0;
780e820b 2628
38aa2714
PZ
2629 /*
2630 * Check for read in write conflicts
2631 */
2632 if (!read) {
2633 if (!valid_state(curr, this, new_bit, excl_bit + 1))
2634 return 0;
2635
2636 if (STRICT_READ_CHECKS &&
4f367d8a
PZ
2637 !usage(curr, this, excl_bit + 1,
2638 state_name(new_bit + 1)))
38aa2714
PZ
2639 return 0;
2640 }
780e820b 2641
cd95302d 2642 if (state_verbose(new_bit, hlock_class(this)))
6a6904d3
PZ
2643 return 2;
2644
2645 return 1;
2646}
2647
cf40bd16 2648enum mark_type {
36bfb9bb
PZ
2649#define LOCKDEP_STATE(__STATE) __STATE,
2650#include "lockdep_states.h"
2651#undef LOCKDEP_STATE
cf40bd16
NP
2652};
2653
fbb9ce95
IM
2654/*
2655 * Mark all held locks with a usage bit:
2656 */
1d09daa5 2657static int
cf40bd16 2658mark_held_locks(struct task_struct *curr, enum mark_type mark)
fbb9ce95
IM
2659{
2660 enum lock_usage_bit usage_bit;
2661 struct held_lock *hlock;
2662 int i;
2663
2664 for (i = 0; i < curr->lockdep_depth; i++) {
2665 hlock = curr->held_locks + i;
2666
cf2ad4d1
PZ
2667 usage_bit = 2 + (mark << 2); /* ENABLED */
2668 if (hlock->read)
2669 usage_bit += 1; /* READ */
2670
2671 BUG_ON(usage_bit >= LOCK_USAGE_STATES);
cf40bd16 2672
34d0ed5e 2673 if (!hlock->check)
efbe2eee
PZ
2674 continue;
2675
4ff773bb 2676 if (!mark_lock(curr, hlock, usage_bit))
fbb9ce95
IM
2677 return 0;
2678 }
2679
2680 return 1;
2681}
2682
fbb9ce95
IM
2683/*
2684 * Hardirqs will be enabled:
2685 */
dd4e5d3a 2686static void __trace_hardirqs_on_caller(unsigned long ip)
fbb9ce95
IM
2687{
2688 struct task_struct *curr = current;
fbb9ce95 2689
fbb9ce95
IM
2690 /* we'll do an OFF -> ON transition: */
2691 curr->hardirqs_enabled = 1;
fbb9ce95 2692
fbb9ce95
IM
2693 /*
2694 * We are going to turn hardirqs on, so set the
2695 * usage bit for all held locks:
2696 */
cf40bd16 2697 if (!mark_held_locks(curr, HARDIRQ))
fbb9ce95
IM
2698 return;
2699 /*
2700 * If we have softirqs enabled, then set the usage
2701 * bit for all held locks. (disabled hardirqs prevented
2702 * this bit from being set before)
2703 */
2704 if (curr->softirqs_enabled)
cf40bd16 2705 if (!mark_held_locks(curr, SOFTIRQ))
fbb9ce95
IM
2706 return;
2707
8e18257d
PZ
2708 curr->hardirq_enable_ip = ip;
2709 curr->hardirq_enable_event = ++curr->irq_events;
bd6d29c2 2710 debug_atomic_inc(hardirqs_on_events);
8e18257d 2711}
dd4e5d3a 2712
b35f8305 2713__visible void trace_hardirqs_on_caller(unsigned long ip)
dd4e5d3a
PZ
2714{
2715 time_hardirqs_on(CALLER_ADDR0, ip);
2716
2717 if (unlikely(!debug_locks || current->lockdep_recursion))
2718 return;
2719
7d36b26b
PZ
2720 if (unlikely(current->hardirqs_enabled)) {
2721 /*
2722 * Neither irq nor preemption are disabled here
2723 * so this is racy by nature but losing one hit
2724 * in a stat is not a big deal.
2725 */
2726 __debug_atomic_inc(redundant_hardirqs_on);
2727 return;
2728 }
2729
0119fee4
PZ
2730 /*
2731 * We're enabling irqs and according to our state above irqs weren't
2732 * already enabled, yet we find the hardware thinks they are in fact
2733 * enabled.. someone messed up their IRQ state tracing.
2734 */
dd4e5d3a
PZ
2735 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2736 return;
2737
0119fee4
PZ
2738 /*
2739 * See the fine text that goes along with this variable definition.
2740 */
7d36b26b
PZ
2741 if (DEBUG_LOCKS_WARN_ON(unlikely(early_boot_irqs_disabled)))
2742 return;
2743
0119fee4
PZ
2744 /*
2745 * Can't allow enabling interrupts while in an interrupt handler,
2746 * that's general bad form and such. Recursion, limited stack etc..
2747 */
7d36b26b
PZ
2748 if (DEBUG_LOCKS_WARN_ON(current->hardirq_context))
2749 return;
2750
dd4e5d3a
PZ
2751 current->lockdep_recursion = 1;
2752 __trace_hardirqs_on_caller(ip);
2753 current->lockdep_recursion = 0;
2754}
81d68a96 2755EXPORT_SYMBOL(trace_hardirqs_on_caller);
8e18257d 2756
1d09daa5 2757void trace_hardirqs_on(void)
81d68a96
SR
2758{
2759 trace_hardirqs_on_caller(CALLER_ADDR0);
2760}
8e18257d
PZ
2761EXPORT_SYMBOL(trace_hardirqs_on);
2762
2763/*
2764 * Hardirqs were disabled:
2765 */
b35f8305 2766__visible void trace_hardirqs_off_caller(unsigned long ip)
8e18257d
PZ
2767{
2768 struct task_struct *curr = current;
2769
6afe40b4 2770 time_hardirqs_off(CALLER_ADDR0, ip);
81d68a96 2771
8e18257d
PZ
2772 if (unlikely(!debug_locks || current->lockdep_recursion))
2773 return;
2774
0119fee4
PZ
2775 /*
2776 * So we're supposed to get called after you mask local IRQs, but for
2777 * some reason the hardware doesn't quite think you did a proper job.
2778 */
8e18257d
PZ
2779 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2780 return;
2781
2782 if (curr->hardirqs_enabled) {
2783 /*
2784 * We have done an ON -> OFF transition:
2785 */
2786 curr->hardirqs_enabled = 0;
6afe40b4 2787 curr->hardirq_disable_ip = ip;
8e18257d 2788 curr->hardirq_disable_event = ++curr->irq_events;
bd6d29c2 2789 debug_atomic_inc(hardirqs_off_events);
8e18257d 2790 } else
bd6d29c2 2791 debug_atomic_inc(redundant_hardirqs_off);
8e18257d 2792}
81d68a96 2793EXPORT_SYMBOL(trace_hardirqs_off_caller);
8e18257d 2794
1d09daa5 2795void trace_hardirqs_off(void)
81d68a96
SR
2796{
2797 trace_hardirqs_off_caller(CALLER_ADDR0);
2798}
8e18257d
PZ
2799EXPORT_SYMBOL(trace_hardirqs_off);
2800
2801/*
2802 * Softirqs will be enabled:
2803 */
2804void trace_softirqs_on(unsigned long ip)
2805{
2806 struct task_struct *curr = current;
2807
dd4e5d3a 2808 if (unlikely(!debug_locks || current->lockdep_recursion))
8e18257d
PZ
2809 return;
2810
0119fee4
PZ
2811 /*
2812 * We fancy IRQs being disabled here, see softirq.c, avoids
2813 * funny state and nesting things.
2814 */
8e18257d
PZ
2815 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2816 return;
2817
2818 if (curr->softirqs_enabled) {
bd6d29c2 2819 debug_atomic_inc(redundant_softirqs_on);
8e18257d
PZ
2820 return;
2821 }
2822
dd4e5d3a 2823 current->lockdep_recursion = 1;
8e18257d
PZ
2824 /*
2825 * We'll do an OFF -> ON transition:
2826 */
2827 curr->softirqs_enabled = 1;
2828 curr->softirq_enable_ip = ip;
2829 curr->softirq_enable_event = ++curr->irq_events;
bd6d29c2 2830 debug_atomic_inc(softirqs_on_events);
8e18257d
PZ
2831 /*
2832 * We are going to turn softirqs on, so set the
2833 * usage bit for all held locks, if hardirqs are
2834 * enabled too:
2835 */
2836 if (curr->hardirqs_enabled)
cf40bd16 2837 mark_held_locks(curr, SOFTIRQ);
dd4e5d3a 2838 current->lockdep_recursion = 0;
8e18257d
PZ
2839}
2840
2841/*
2842 * Softirqs were disabled:
2843 */
2844void trace_softirqs_off(unsigned long ip)
2845{
2846 struct task_struct *curr = current;
2847
dd4e5d3a 2848 if (unlikely(!debug_locks || current->lockdep_recursion))
8e18257d
PZ
2849 return;
2850
0119fee4
PZ
2851 /*
2852 * We fancy IRQs being disabled here, see softirq.c
2853 */
8e18257d
PZ
2854 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2855 return;
2856
2857 if (curr->softirqs_enabled) {
2858 /*
2859 * We have done an ON -> OFF transition:
2860 */
2861 curr->softirqs_enabled = 0;
2862 curr->softirq_disable_ip = ip;
2863 curr->softirq_disable_event = ++curr->irq_events;
bd6d29c2 2864 debug_atomic_inc(softirqs_off_events);
0119fee4
PZ
2865 /*
2866 * Whoops, we wanted softirqs off, so why aren't they?
2867 */
8e18257d
PZ
2868 DEBUG_LOCKS_WARN_ON(!softirq_count());
2869 } else
bd6d29c2 2870 debug_atomic_inc(redundant_softirqs_off);
8e18257d
PZ
2871}
2872
2f850181 2873static void __lockdep_trace_alloc(gfp_t gfp_mask, unsigned long flags)
cf40bd16
NP
2874{
2875 struct task_struct *curr = current;
2876
2877 if (unlikely(!debug_locks))
2878 return;
2879
7dea19f9 2880 gfp_mask = current_gfp_context(gfp_mask);
6d7225f0 2881
cf40bd16 2882 /* no reclaim without waiting on it */
d0164adc 2883 if (!(gfp_mask & __GFP_DIRECT_RECLAIM))
cf40bd16
NP
2884 return;
2885
2886 /* this guy won't enter reclaim */
2887 if ((curr->flags & PF_MEMALLOC) && !(gfp_mask & __GFP_NOMEMALLOC))
2888 return;
2889
2890 /* We're only interested __GFP_FS allocations for now */
7dea19f9 2891 if (!(gfp_mask & __GFP_FS) || (curr->flags & PF_MEMALLOC_NOFS))
cf40bd16
NP
2892 return;
2893
0119fee4
PZ
2894 /*
2895 * Oi! Can't be having __GFP_FS allocations with IRQs disabled.
2896 */
2f850181 2897 if (DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags)))
cf40bd16
NP
2898 return;
2899
7e784422
MH
2900 /* Disable lockdep if explicitly requested */
2901 if (gfp_mask & __GFP_NOLOCKDEP)
2902 return;
2903
cf40bd16
NP
2904 mark_held_locks(curr, RECLAIM_FS);
2905}
2906
2f850181
PZ
2907static void check_flags(unsigned long flags);
2908
2909void lockdep_trace_alloc(gfp_t gfp_mask)
2910{
2911 unsigned long flags;
2912
2913 if (unlikely(current->lockdep_recursion))
2914 return;
2915
2916 raw_local_irq_save(flags);
2917 check_flags(flags);
2918 current->lockdep_recursion = 1;
2919 __lockdep_trace_alloc(gfp_mask, flags);
2920 current->lockdep_recursion = 0;
2921 raw_local_irq_restore(flags);
2922}
2923
8e18257d
PZ
2924static int mark_irqflags(struct task_struct *curr, struct held_lock *hlock)
2925{
2926 /*
2927 * If non-trylock use in a hardirq or softirq context, then
2928 * mark the lock as used in these contexts:
2929 */
2930 if (!hlock->trylock) {
2931 if (hlock->read) {
2932 if (curr->hardirq_context)
2933 if (!mark_lock(curr, hlock,
2934 LOCK_USED_IN_HARDIRQ_READ))
2935 return 0;
2936 if (curr->softirq_context)
2937 if (!mark_lock(curr, hlock,
2938 LOCK_USED_IN_SOFTIRQ_READ))
2939 return 0;
2940 } else {
2941 if (curr->hardirq_context)
2942 if (!mark_lock(curr, hlock, LOCK_USED_IN_HARDIRQ))
2943 return 0;
2944 if (curr->softirq_context)
2945 if (!mark_lock(curr, hlock, LOCK_USED_IN_SOFTIRQ))
2946 return 0;
2947 }
2948 }
2949 if (!hlock->hardirqs_off) {
2950 if (hlock->read) {
2951 if (!mark_lock(curr, hlock,
4fc95e86 2952 LOCK_ENABLED_HARDIRQ_READ))
8e18257d
PZ
2953 return 0;
2954 if (curr->softirqs_enabled)
2955 if (!mark_lock(curr, hlock,
4fc95e86 2956 LOCK_ENABLED_SOFTIRQ_READ))
8e18257d
PZ
2957 return 0;
2958 } else {
2959 if (!mark_lock(curr, hlock,
4fc95e86 2960 LOCK_ENABLED_HARDIRQ))
8e18257d
PZ
2961 return 0;
2962 if (curr->softirqs_enabled)
2963 if (!mark_lock(curr, hlock,
4fc95e86 2964 LOCK_ENABLED_SOFTIRQ))
8e18257d
PZ
2965 return 0;
2966 }
2967 }
2968
cf40bd16
NP
2969 /*
2970 * We reuse the irq context infrastructure more broadly as a general
2971 * context checking code. This tests GFP_FS recursion (a lock taken
2972 * during reclaim for a GFP_FS allocation is held over a GFP_FS
2973 * allocation).
2974 */
2975 if (!hlock->trylock && (curr->lockdep_reclaim_gfp & __GFP_FS)) {
2976 if (hlock->read) {
2977 if (!mark_lock(curr, hlock, LOCK_USED_IN_RECLAIM_FS_READ))
2978 return 0;
2979 } else {
2980 if (!mark_lock(curr, hlock, LOCK_USED_IN_RECLAIM_FS))
2981 return 0;
2982 }
2983 }
2984
8e18257d
PZ
2985 return 1;
2986}
2987
c2469756
BF
2988static inline unsigned int task_irq_context(struct task_struct *task)
2989{
2990 return 2 * !!task->hardirq_context + !!task->softirq_context;
2991}
2992
8e18257d
PZ
2993static int separate_irq_context(struct task_struct *curr,
2994 struct held_lock *hlock)
2995{
2996 unsigned int depth = curr->lockdep_depth;
2997
2998 /*
2999 * Keep track of points where we cross into an interrupt context:
3000 */
8e18257d
PZ
3001 if (depth) {
3002 struct held_lock *prev_hlock;
3003
3004 prev_hlock = curr->held_locks + depth-1;
3005 /*
3006 * If we cross into another context, reset the
3007 * hash key (this also prevents the checking and the
3008 * adding of the dependency to 'prev'):
3009 */
3010 if (prev_hlock->irq_context != hlock->irq_context)
3011 return 1;
3012 }
3013 return 0;
fbb9ce95
IM
3014}
3015
0119fee4 3016#else /* defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) */
fbb9ce95 3017
8e18257d
PZ
3018static inline
3019int mark_lock_irq(struct task_struct *curr, struct held_lock *this,
3020 enum lock_usage_bit new_bit)
fbb9ce95 3021{
0119fee4 3022 WARN_ON(1); /* Impossible innit? when we don't have TRACE_IRQFLAG */
8e18257d
PZ
3023 return 1;
3024}
fbb9ce95 3025
8e18257d
PZ
3026static inline int mark_irqflags(struct task_struct *curr,
3027 struct held_lock *hlock)
3028{
3029 return 1;
3030}
fbb9ce95 3031
c2469756
BF
3032static inline unsigned int task_irq_context(struct task_struct *task)
3033{
3034 return 0;
3035}
3036
8e18257d
PZ
3037static inline int separate_irq_context(struct task_struct *curr,
3038 struct held_lock *hlock)
3039{
3040 return 0;
fbb9ce95
IM
3041}
3042
868a23a8
PZ
3043void lockdep_trace_alloc(gfp_t gfp_mask)
3044{
3045}
3046
0119fee4 3047#endif /* defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) */
fbb9ce95
IM
3048
3049/*
8e18257d 3050 * Mark a lock with a usage bit, and validate the state transition:
fbb9ce95 3051 */
1d09daa5 3052static int mark_lock(struct task_struct *curr, struct held_lock *this,
0764d23c 3053 enum lock_usage_bit new_bit)
fbb9ce95 3054{
8e18257d 3055 unsigned int new_mask = 1 << new_bit, ret = 1;
fbb9ce95
IM
3056
3057 /*
8e18257d
PZ
3058 * If already set then do not dirty the cacheline,
3059 * nor do any checks:
fbb9ce95 3060 */
f82b217e 3061 if (likely(hlock_class(this)->usage_mask & new_mask))
8e18257d
PZ
3062 return 1;
3063
3064 if (!graph_lock())
3065 return 0;
fbb9ce95 3066 /*
25985edc 3067 * Make sure we didn't race:
fbb9ce95 3068 */
f82b217e 3069 if (unlikely(hlock_class(this)->usage_mask & new_mask)) {
8e18257d
PZ
3070 graph_unlock();
3071 return 1;
3072 }
fbb9ce95 3073
f82b217e 3074 hlock_class(this)->usage_mask |= new_mask;
fbb9ce95 3075
f82b217e 3076 if (!save_trace(hlock_class(this)->usage_traces + new_bit))
8e18257d 3077 return 0;
fbb9ce95 3078
8e18257d 3079 switch (new_bit) {
5346417e
PZ
3080#define LOCKDEP_STATE(__STATE) \
3081 case LOCK_USED_IN_##__STATE: \
3082 case LOCK_USED_IN_##__STATE##_READ: \
3083 case LOCK_ENABLED_##__STATE: \
3084 case LOCK_ENABLED_##__STATE##_READ:
3085#include "lockdep_states.h"
3086#undef LOCKDEP_STATE
8e18257d
PZ
3087 ret = mark_lock_irq(curr, this, new_bit);
3088 if (!ret)
3089 return 0;
3090 break;
3091 case LOCK_USED:
bd6d29c2 3092 debug_atomic_dec(nr_unused_locks);
8e18257d
PZ
3093 break;
3094 default:
3095 if (!debug_locks_off_graph_unlock())
3096 return 0;
3097 WARN_ON(1);
3098 return 0;
3099 }
fbb9ce95 3100
8e18257d
PZ
3101 graph_unlock();
3102
3103 /*
3104 * We must printk outside of the graph_lock:
3105 */
3106 if (ret == 2) {
3107 printk("\nmarked lock as {%s}:\n", usage_str[new_bit]);
3108 print_lock(this);
3109 print_irqtrace_events(curr);
3110 dump_stack();
3111 }
3112
3113 return ret;
3114}
fbb9ce95
IM
3115
3116/*
3117 * Initialize a lock instance's lock-class mapping info:
3118 */
3119void lockdep_init_map(struct lockdep_map *lock, const char *name,
4dfbb9d8 3120 struct lock_class_key *key, int subclass)
fbb9ce95 3121{
d3d03d4f
YZ
3122 int i;
3123
3124 kmemcheck_mark_initialized(lock, sizeof(*lock));
3125
3126 for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++)
3127 lock->class_cache[i] = NULL;
62016250 3128
c8a25005
PZ
3129#ifdef CONFIG_LOCK_STAT
3130 lock->cpu = raw_smp_processor_id();
3131#endif
3132
0119fee4
PZ
3133 /*
3134 * Can't be having no nameless bastards around this place!
3135 */
c8a25005
PZ
3136 if (DEBUG_LOCKS_WARN_ON(!name)) {
3137 lock->name = "NULL";
fbb9ce95 3138 return;
c8a25005
PZ
3139 }
3140
3141 lock->name = name;
fbb9ce95 3142
0119fee4
PZ
3143 /*
3144 * No key, no joy, we need to hash something.
3145 */
fbb9ce95
IM
3146 if (DEBUG_LOCKS_WARN_ON(!key))
3147 return;
fbb9ce95
IM
3148 /*
3149 * Sanity check, the lock-class key must be persistent:
3150 */
3151 if (!static_obj(key)) {
3152 printk("BUG: key %p not in .data!\n", key);
0119fee4
PZ
3153 /*
3154 * What it says above ^^^^^, I suggest you read it.
3155 */
fbb9ce95
IM
3156 DEBUG_LOCKS_WARN_ON(1);
3157 return;
3158 }
fbb9ce95 3159 lock->key = key;
c8a25005
PZ
3160
3161 if (unlikely(!debug_locks))
3162 return;
3163
35a9393c
PZ
3164 if (subclass) {
3165 unsigned long flags;
3166
3167 if (DEBUG_LOCKS_WARN_ON(current->lockdep_recursion))
3168 return;
3169
3170 raw_local_irq_save(flags);
3171 current->lockdep_recursion = 1;
4dfbb9d8 3172 register_lock_class(lock, subclass, 1);
35a9393c
PZ
3173 current->lockdep_recursion = 0;
3174 raw_local_irq_restore(flags);
3175 }
fbb9ce95 3176}
fbb9ce95
IM
3177EXPORT_SYMBOL_GPL(lockdep_init_map);
3178
1704f47b 3179struct lock_class_key __lockdep_no_validate__;
ea6749c7 3180EXPORT_SYMBOL_GPL(__lockdep_no_validate__);
1704f47b 3181
d0945950
ML
3182static int
3183print_lock_nested_lock_not_held(struct task_struct *curr,
3184 struct held_lock *hlock,
3185 unsigned long ip)
3186{
3187 if (!debug_locks_off())
3188 return 0;
3189 if (debug_locks_silent)
3190 return 0;
3191
3192 printk("\n");
3193 printk("==================================\n");
3194 printk("[ BUG: Nested lock was not taken ]\n");
3195 print_kernel_ident();
3196 printk("----------------------------------\n");
3197
3198 printk("%s/%d is trying to lock:\n", curr->comm, task_pid_nr(curr));
3199 print_lock(hlock);
3200
3201 printk("\nbut this task is not holding:\n");
3202 printk("%s\n", hlock->nest_lock->name);
3203
3204 printk("\nstack backtrace:\n");
3205 dump_stack();
3206
3207 printk("\nother info that might help us debug this:\n");
3208 lockdep_print_held_locks(curr);
3209
3210 printk("\nstack backtrace:\n");
3211 dump_stack();
3212
3213 return 0;
3214}
3215
f8319483 3216static int __lock_is_held(struct lockdep_map *lock, int read);
d0945950 3217
fbb9ce95
IM
3218/*
3219 * This gets called for every mutex_lock*()/spin_lock*() operation.
3220 * We maintain the dependency maps and validate the locking attempt:
3221 */
3222static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass,
3223 int trylock, int read, int check, int hardirqs_off,
bb97a91e 3224 struct lockdep_map *nest_lock, unsigned long ip,
21199f27 3225 int references, int pin_count)
fbb9ce95
IM
3226{
3227 struct task_struct *curr = current;
d6d897ce 3228 struct lock_class *class = NULL;
fbb9ce95 3229 struct held_lock *hlock;
5f18ab5c 3230 unsigned int depth;
fbb9ce95 3231 int chain_head = 0;
bb97a91e 3232 int class_idx;
fbb9ce95
IM
3233 u64 chain_key;
3234
3235 if (unlikely(!debug_locks))
3236 return 0;
3237
0119fee4
PZ
3238 /*
3239 * Lockdep should run with IRQs disabled, otherwise we could
3240 * get an interrupt which would want to take locks, which would
3241 * end up in lockdep and have you got a head-ache already?
3242 */
fbb9ce95
IM
3243 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
3244 return 0;
3245
fb9edbe9
ON
3246 if (!prove_locking || lock->key == &__lockdep_no_validate__)
3247 check = 0;
1704f47b 3248
62016250
HM
3249 if (subclass < NR_LOCKDEP_CACHING_CLASSES)
3250 class = lock->class_cache[subclass];
d6d897ce 3251 /*
62016250 3252 * Not cached?
d6d897ce 3253 */
fbb9ce95 3254 if (unlikely(!class)) {
4dfbb9d8 3255 class = register_lock_class(lock, subclass, 0);
fbb9ce95
IM
3256 if (!class)
3257 return 0;
3258 }
bd6d29c2 3259 atomic_inc((atomic_t *)&class->ops);
fbb9ce95
IM
3260 if (very_verbose(class)) {
3261 printk("\nacquire class [%p] %s", class->key, class->name);
3262 if (class->name_version > 1)
f943fe0f
DV
3263 printk(KERN_CONT "#%d", class->name_version);
3264 printk(KERN_CONT "\n");
fbb9ce95
IM
3265 dump_stack();
3266 }
3267
3268 /*
3269 * Add the lock to the list of currently held locks.
3270 * (we dont increase the depth just yet, up until the
3271 * dependency checks are done)
3272 */
3273 depth = curr->lockdep_depth;
0119fee4
PZ
3274 /*
3275 * Ran out of static storage for our per-task lock stack again have we?
3276 */
fbb9ce95
IM
3277 if (DEBUG_LOCKS_WARN_ON(depth >= MAX_LOCK_DEPTH))
3278 return 0;
3279
bb97a91e
PZ
3280 class_idx = class - lock_classes + 1;
3281
3282 if (depth) {
3283 hlock = curr->held_locks + depth - 1;
3284 if (hlock->class_idx == class_idx && nest_lock) {
7fb4a2ce
PZ
3285 if (hlock->references) {
3286 /*
3287 * Check: unsigned int references:12, overflow.
3288 */
3289 if (DEBUG_LOCKS_WARN_ON(hlock->references == (1 << 12)-1))
3290 return 0;
3291
bb97a91e 3292 hlock->references++;
7fb4a2ce 3293 } else {
bb97a91e 3294 hlock->references = 2;
7fb4a2ce 3295 }
bb97a91e
PZ
3296
3297 return 1;
3298 }
3299 }
3300
fbb9ce95 3301 hlock = curr->held_locks + depth;
0119fee4
PZ
3302 /*
3303 * Plain impossible, we just registered it and checked it weren't no
3304 * NULL like.. I bet this mushroom I ate was good!
3305 */
f82b217e
DJ
3306 if (DEBUG_LOCKS_WARN_ON(!class))
3307 return 0;
bb97a91e 3308 hlock->class_idx = class_idx;
fbb9ce95
IM
3309 hlock->acquire_ip = ip;
3310 hlock->instance = lock;
7531e2f3 3311 hlock->nest_lock = nest_lock;
c2469756 3312 hlock->irq_context = task_irq_context(curr);
fbb9ce95
IM
3313 hlock->trylock = trylock;
3314 hlock->read = read;
3315 hlock->check = check;
6951b12a 3316 hlock->hardirqs_off = !!hardirqs_off;
bb97a91e 3317 hlock->references = references;
f20786ff
PZ
3318#ifdef CONFIG_LOCK_STAT
3319 hlock->waittime_stamp = 0;
3365e779 3320 hlock->holdtime_stamp = lockstat_clock();
f20786ff 3321#endif
21199f27 3322 hlock->pin_count = pin_count;
fbb9ce95 3323
fb9edbe9 3324 if (check && !mark_irqflags(curr, hlock))
8e18257d
PZ
3325 return 0;
3326
fbb9ce95 3327 /* mark it as used: */
4ff773bb 3328 if (!mark_lock(curr, hlock, LOCK_USED))
fbb9ce95 3329 return 0;
8e18257d 3330
fbb9ce95 3331 /*
17aacfb9 3332 * Calculate the chain hash: it's the combined hash of all the
fbb9ce95
IM
3333 * lock keys along the dependency chain. We save the hash value
3334 * at every step so that we can get the current hash easily
3335 * after unlock. The chain hash is then used to cache dependency
3336 * results.
3337 *
3338 * The 'key ID' is what is the most compact key value to drive
3339 * the hash, not class->key.
3340 */
0119fee4
PZ
3341 /*
3342 * Whoops, we did it again.. ran straight out of our static allocation.
3343 */
5f18ab5c 3344 if (DEBUG_LOCKS_WARN_ON(class_idx > MAX_LOCKDEP_KEYS))
fbb9ce95
IM
3345 return 0;
3346
3347 chain_key = curr->curr_chain_key;
3348 if (!depth) {
0119fee4
PZ
3349 /*
3350 * How can we have a chain hash when we ain't got no keys?!
3351 */
fbb9ce95
IM
3352 if (DEBUG_LOCKS_WARN_ON(chain_key != 0))
3353 return 0;
3354 chain_head = 1;
3355 }
3356
3357 hlock->prev_chain_key = chain_key;
8e18257d
PZ
3358 if (separate_irq_context(curr, hlock)) {
3359 chain_key = 0;
3360 chain_head = 1;
fbb9ce95 3361 }
5f18ab5c 3362 chain_key = iterate_chain_key(chain_key, class_idx);
fbb9ce95 3363
f8319483 3364 if (nest_lock && !__lock_is_held(nest_lock, -1))
d0945950
ML
3365 return print_lock_nested_lock_not_held(curr, hlock, ip);
3366
3aa416b0 3367 if (!validate_chain(curr, lock, hlock, chain_head, chain_key))
8e18257d 3368 return 0;
381a2292 3369
3aa416b0 3370 curr->curr_chain_key = chain_key;
fbb9ce95
IM
3371 curr->lockdep_depth++;
3372 check_chain_key(curr);
60e114d1
JP
3373#ifdef CONFIG_DEBUG_LOCKDEP
3374 if (unlikely(!debug_locks))
3375 return 0;
3376#endif
fbb9ce95
IM
3377 if (unlikely(curr->lockdep_depth >= MAX_LOCK_DEPTH)) {
3378 debug_locks_off();
2c522836
DJ
3379 print_lockdep_off("BUG: MAX_LOCK_DEPTH too low!");
3380 printk(KERN_DEBUG "depth: %i max: %lu!\n",
c0540606 3381 curr->lockdep_depth, MAX_LOCK_DEPTH);
c0540606
BG
3382
3383 lockdep_print_held_locks(current);
3384 debug_show_all_locks();
eedeeabd 3385 dump_stack();
c0540606 3386
fbb9ce95
IM
3387 return 0;
3388 }
381a2292 3389
fbb9ce95
IM
3390 if (unlikely(curr->lockdep_depth > max_lockdep_depth))
3391 max_lockdep_depth = curr->lockdep_depth;
3392
3393 return 1;
3394}
3395
3396static int
f86f7554 3397print_unlock_imbalance_bug(struct task_struct *curr, struct lockdep_map *lock,
fbb9ce95
IM
3398 unsigned long ip)
3399{
3400 if (!debug_locks_off())
3401 return 0;
3402 if (debug_locks_silent)
3403 return 0;
3404
b3fbab05
PM
3405 printk("\n");
3406 printk("=====================================\n");
3407 printk("[ BUG: bad unlock balance detected! ]\n");
fbdc4b9a 3408 print_kernel_ident();
b3fbab05 3409 printk("-------------------------------------\n");
fbb9ce95 3410 printk("%s/%d is trying to release lock (",
ba25f9dc 3411 curr->comm, task_pid_nr(curr));
fbb9ce95 3412 print_lockdep_cache(lock);
f943fe0f 3413 printk(KERN_CONT ") at:\n");
fbb9ce95
IM
3414 print_ip_sym(ip);
3415 printk("but there are no more locks to release!\n");
3416 printk("\nother info that might help us debug this:\n");
3417 lockdep_print_held_locks(curr);
3418
3419 printk("\nstack backtrace:\n");
3420 dump_stack();
3421
3422 return 0;
3423}
3424
bb97a91e
PZ
3425static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock)
3426{
3427 if (hlock->instance == lock)
3428 return 1;
3429
3430 if (hlock->references) {
62016250 3431 struct lock_class *class = lock->class_cache[0];
bb97a91e
PZ
3432
3433 if (!class)
3434 class = look_up_lock_class(lock, 0);
3435
80e0401e
PZ
3436 /*
3437 * If look_up_lock_class() failed to find a class, we're trying
3438 * to test if we hold a lock that has never yet been acquired.
3439 * Clearly if the lock hasn't been acquired _ever_, we're not
3440 * holding it either, so report failure.
3441 */
383776fa 3442 if (IS_ERR_OR_NULL(class))
bb97a91e
PZ
3443 return 0;
3444
0119fee4
PZ
3445 /*
3446 * References, but not a lock we're actually ref-counting?
3447 * State got messed up, follow the sites that change ->references
3448 * and try to make sense of it.
3449 */
bb97a91e
PZ
3450 if (DEBUG_LOCKS_WARN_ON(!hlock->nest_lock))
3451 return 0;
3452
3453 if (hlock->class_idx == class - lock_classes + 1)
3454 return 1;
3455 }
3456
3457 return 0;
3458}
3459
41c2c5b8
O
3460/* @depth must not be zero */
3461static struct held_lock *find_held_lock(struct task_struct *curr,
3462 struct lockdep_map *lock,
3463 unsigned int depth, int *idx)
3464{
3465 struct held_lock *ret, *hlock, *prev_hlock;
3466 int i;
3467
3468 i = depth - 1;
3469 hlock = curr->held_locks + i;
3470 ret = hlock;
3471 if (match_held_lock(hlock, lock))
3472 goto out;
3473
3474 ret = NULL;
3475 for (i--, prev_hlock = hlock--;
3476 i >= 0;
3477 i--, prev_hlock = hlock--) {
3478 /*
3479 * We must not cross into another context:
3480 */
3481 if (prev_hlock->irq_context != hlock->irq_context) {
3482 ret = NULL;
3483 break;
3484 }
3485 if (match_held_lock(hlock, lock)) {
3486 ret = hlock;
3487 break;
3488 }
3489 }
3490
3491out:
3492 *idx = i;
3493 return ret;
3494}
3495
e969970b
O
3496static int reacquire_held_locks(struct task_struct *curr, unsigned int depth,
3497 int idx)
3498{
3499 struct held_lock *hlock;
3500
3501 for (hlock = curr->held_locks + idx; idx < depth; idx++, hlock++) {
3502 if (!__lock_acquire(hlock->instance,
3503 hlock_class(hlock)->subclass,
3504 hlock->trylock,
3505 hlock->read, hlock->check,
3506 hlock->hardirqs_off,
3507 hlock->nest_lock, hlock->acquire_ip,
3508 hlock->references, hlock->pin_count))
3509 return 1;
3510 }
3511 return 0;
3512}
3513
64aa348e 3514static int
00ef9f73
PZ
3515__lock_set_class(struct lockdep_map *lock, const char *name,
3516 struct lock_class_key *key, unsigned int subclass,
3517 unsigned long ip)
64aa348e
PZ
3518{
3519 struct task_struct *curr = current;
41c2c5b8 3520 struct held_lock *hlock;
64aa348e
PZ
3521 struct lock_class *class;
3522 unsigned int depth;
3523 int i;
3524
3525 depth = curr->lockdep_depth;
0119fee4
PZ
3526 /*
3527 * This function is about (re)setting the class of a held lock,
3528 * yet we're not actually holding any locks. Naughty user!
3529 */
64aa348e
PZ
3530 if (DEBUG_LOCKS_WARN_ON(!depth))
3531 return 0;
3532
41c2c5b8
O
3533 hlock = find_held_lock(curr, lock, depth, &i);
3534 if (!hlock)
3535 return print_unlock_imbalance_bug(curr, lock, ip);
64aa348e 3536
00ef9f73 3537 lockdep_init_map(lock, name, key, 0);
64aa348e 3538 class = register_lock_class(lock, subclass, 0);
f82b217e 3539 hlock->class_idx = class - lock_classes + 1;
64aa348e
PZ
3540
3541 curr->lockdep_depth = i;
3542 curr->curr_chain_key = hlock->prev_chain_key;
3543
e969970b
O
3544 if (reacquire_held_locks(curr, depth, i))
3545 return 0;
64aa348e 3546
0119fee4
PZ
3547 /*
3548 * I took it apart and put it back together again, except now I have
3549 * these 'spare' parts.. where shall I put them.
3550 */
64aa348e
PZ
3551 if (DEBUG_LOCKS_WARN_ON(curr->lockdep_depth != depth))
3552 return 0;
3553 return 1;
3554}
3555
6419c4af
O
3556static int __lock_downgrade(struct lockdep_map *lock, unsigned long ip)
3557{
3558 struct task_struct *curr = current;
3559 struct held_lock *hlock;
3560 unsigned int depth;
3561 int i;
3562
3563 depth = curr->lockdep_depth;
3564 /*
3565 * This function is about (re)setting the class of a held lock,
3566 * yet we're not actually holding any locks. Naughty user!
3567 */
3568 if (DEBUG_LOCKS_WARN_ON(!depth))
3569 return 0;
3570
3571 hlock = find_held_lock(curr, lock, depth, &i);
3572 if (!hlock)
3573 return print_unlock_imbalance_bug(curr, lock, ip);
3574
3575 curr->lockdep_depth = i;
3576 curr->curr_chain_key = hlock->prev_chain_key;
3577
3578 WARN(hlock->read, "downgrading a read lock");
3579 hlock->read = 1;
3580 hlock->acquire_ip = ip;
3581
3582 if (reacquire_held_locks(curr, depth, i))
3583 return 0;
3584
3585 /*
3586 * I took it apart and put it back together again, except now I have
3587 * these 'spare' parts.. where shall I put them.
3588 */
3589 if (DEBUG_LOCKS_WARN_ON(curr->lockdep_depth != depth))
3590 return 0;
3591 return 1;
3592}
3593
fbb9ce95 3594/*
e0f56fd7
PZ
3595 * Remove the lock to the list of currently held locks - this gets
3596 * called on mutex_unlock()/spin_unlock*() (or on a failed
3597 * mutex_lock_interruptible()).
3598 *
3599 * @nested is an hysterical artifact, needs a tree wide cleanup.
fbb9ce95
IM
3600 */
3601static int
e0f56fd7 3602__lock_release(struct lockdep_map *lock, int nested, unsigned long ip)
fbb9ce95 3603{
e0f56fd7 3604 struct task_struct *curr = current;
41c2c5b8 3605 struct held_lock *hlock;
fbb9ce95
IM
3606 unsigned int depth;
3607 int i;
3608
e0f56fd7
PZ
3609 if (unlikely(!debug_locks))
3610 return 0;
3611
fbb9ce95 3612 depth = curr->lockdep_depth;
0119fee4
PZ
3613 /*
3614 * So we're all set to release this lock.. wait what lock? We don't
3615 * own any locks, you've been drinking again?
3616 */
e0f56fd7
PZ
3617 if (DEBUG_LOCKS_WARN_ON(depth <= 0))
3618 return print_unlock_imbalance_bug(curr, lock, ip);
fbb9ce95 3619
e0f56fd7
PZ
3620 /*
3621 * Check whether the lock exists in the current stack
3622 * of held locks:
3623 */
41c2c5b8
O
3624 hlock = find_held_lock(curr, lock, depth, &i);
3625 if (!hlock)
3626 return print_unlock_imbalance_bug(curr, lock, ip);
fbb9ce95 3627
bb97a91e
PZ
3628 if (hlock->instance == lock)
3629 lock_release_holdtime(hlock);
3630
a24fc60d
PZ
3631 WARN(hlock->pin_count, "releasing a pinned lock\n");
3632
bb97a91e
PZ
3633 if (hlock->references) {
3634 hlock->references--;
3635 if (hlock->references) {
3636 /*
3637 * We had, and after removing one, still have
3638 * references, the current lock stack is still
3639 * valid. We're done!
3640 */
3641 return 1;
3642 }
3643 }
f20786ff 3644
fbb9ce95
IM
3645 /*
3646 * We have the right lock to unlock, 'hlock' points to it.
3647 * Now we remove it from the stack, and add back the other
3648 * entries (if any), recalculating the hash along the way:
3649 */
bb97a91e 3650
fbb9ce95
IM
3651 curr->lockdep_depth = i;
3652 curr->curr_chain_key = hlock->prev_chain_key;
3653
e969970b
O
3654 if (reacquire_held_locks(curr, depth, i + 1))
3655 return 0;
fbb9ce95 3656
0119fee4
PZ
3657 /*
3658 * We had N bottles of beer on the wall, we drank one, but now
3659 * there's not N-1 bottles of beer left on the wall...
3660 */
fbb9ce95
IM
3661 if (DEBUG_LOCKS_WARN_ON(curr->lockdep_depth != depth - 1))
3662 return 0;
f20786ff 3663
fbb9ce95
IM
3664 return 1;
3665}
3666
f8319483 3667static int __lock_is_held(struct lockdep_map *lock, int read)
fbb9ce95 3668{
f607c668
PZ
3669 struct task_struct *curr = current;
3670 int i;
fbb9ce95 3671
f607c668 3672 for (i = 0; i < curr->lockdep_depth; i++) {
bb97a91e 3673 struct held_lock *hlock = curr->held_locks + i;
fbb9ce95 3674
f8319483
PZ
3675 if (match_held_lock(hlock, lock)) {
3676 if (read == -1 || hlock->read == read)
3677 return 1;
3678
3679 return 0;
3680 }
f607c668 3681 }
f20786ff 3682
f607c668 3683 return 0;
fbb9ce95
IM
3684}
3685
e7904a28
PZ
3686static struct pin_cookie __lock_pin_lock(struct lockdep_map *lock)
3687{
3688 struct pin_cookie cookie = NIL_COOKIE;
3689 struct task_struct *curr = current;
3690 int i;
3691
3692 if (unlikely(!debug_locks))
3693 return cookie;
3694
3695 for (i = 0; i < curr->lockdep_depth; i++) {
3696 struct held_lock *hlock = curr->held_locks + i;
3697
3698 if (match_held_lock(hlock, lock)) {
3699 /*
3700 * Grab 16bits of randomness; this is sufficient to not
3701 * be guessable and still allows some pin nesting in
3702 * our u32 pin_count.
3703 */
3704 cookie.val = 1 + (prandom_u32() >> 16);
3705 hlock->pin_count += cookie.val;
3706 return cookie;
3707 }
3708 }
3709
3710 WARN(1, "pinning an unheld lock\n");
3711 return cookie;
3712}
3713
3714static void __lock_repin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
fbb9ce95
IM
3715{
3716 struct task_struct *curr = current;
a24fc60d 3717 int i;
fbb9ce95 3718
a24fc60d 3719 if (unlikely(!debug_locks))
fbb9ce95
IM
3720 return;
3721
a24fc60d
PZ
3722 for (i = 0; i < curr->lockdep_depth; i++) {
3723 struct held_lock *hlock = curr->held_locks + i;
3724
3725 if (match_held_lock(hlock, lock)) {
e7904a28 3726 hlock->pin_count += cookie.val;
fbb9ce95 3727 return;
a24fc60d 3728 }
fbb9ce95
IM
3729 }
3730
a24fc60d 3731 WARN(1, "pinning an unheld lock\n");
fbb9ce95
IM
3732}
3733
e7904a28 3734static void __lock_unpin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
f607c668
PZ
3735{
3736 struct task_struct *curr = current;
3737 int i;
3738
a24fc60d
PZ
3739 if (unlikely(!debug_locks))
3740 return;
3741
f607c668 3742 for (i = 0; i < curr->lockdep_depth; i++) {
bb97a91e
PZ
3743 struct held_lock *hlock = curr->held_locks + i;
3744
a24fc60d
PZ
3745 if (match_held_lock(hlock, lock)) {
3746 if (WARN(!hlock->pin_count, "unpinning an unpinned lock\n"))
3747 return;
3748
e7904a28
PZ
3749 hlock->pin_count -= cookie.val;
3750
3751 if (WARN((int)hlock->pin_count < 0, "pin count corrupted\n"))
3752 hlock->pin_count = 0;
3753
a24fc60d
PZ
3754 return;
3755 }
f607c668
PZ
3756 }
3757
a24fc60d 3758 WARN(1, "unpinning an unheld lock\n");
f607c668
PZ
3759}
3760
fbb9ce95
IM
3761/*
3762 * Check whether we follow the irq-flags state precisely:
3763 */
1d09daa5 3764static void check_flags(unsigned long flags)
fbb9ce95 3765{
992860e9
IM
3766#if defined(CONFIG_PROVE_LOCKING) && defined(CONFIG_DEBUG_LOCKDEP) && \
3767 defined(CONFIG_TRACE_IRQFLAGS)
fbb9ce95
IM
3768 if (!debug_locks)
3769 return;
3770
5f9fa8a6
IM
3771 if (irqs_disabled_flags(flags)) {
3772 if (DEBUG_LOCKS_WARN_ON(current->hardirqs_enabled)) {
3773 printk("possible reason: unannotated irqs-off.\n");
3774 }
3775 } else {
3776 if (DEBUG_LOCKS_WARN_ON(!current->hardirqs_enabled)) {
3777 printk("possible reason: unannotated irqs-on.\n");
3778 }
3779 }
fbb9ce95
IM
3780
3781 /*
3782 * We dont accurately track softirq state in e.g.
3783 * hardirq contexts (such as on 4KSTACKS), so only
3784 * check if not in hardirq contexts:
3785 */
3786 if (!hardirq_count()) {
0119fee4
PZ
3787 if (softirq_count()) {
3788 /* like the above, but with softirqs */
fbb9ce95 3789 DEBUG_LOCKS_WARN_ON(current->softirqs_enabled);
0119fee4
PZ
3790 } else {
3791 /* lick the above, does it taste good? */
fbb9ce95 3792 DEBUG_LOCKS_WARN_ON(!current->softirqs_enabled);
0119fee4 3793 }
fbb9ce95
IM
3794 }
3795
3796 if (!debug_locks)
3797 print_irqtrace_events(current);
3798#endif
3799}
3800
00ef9f73
PZ
3801void lock_set_class(struct lockdep_map *lock, const char *name,
3802 struct lock_class_key *key, unsigned int subclass,
3803 unsigned long ip)
64aa348e
PZ
3804{
3805 unsigned long flags;
3806
3807 if (unlikely(current->lockdep_recursion))
3808 return;
3809
3810 raw_local_irq_save(flags);
3811 current->lockdep_recursion = 1;
3812 check_flags(flags);
00ef9f73 3813 if (__lock_set_class(lock, name, key, subclass, ip))
64aa348e
PZ
3814 check_chain_key(current);
3815 current->lockdep_recursion = 0;
3816 raw_local_irq_restore(flags);
3817}
00ef9f73 3818EXPORT_SYMBOL_GPL(lock_set_class);
64aa348e 3819
6419c4af
O
3820void lock_downgrade(struct lockdep_map *lock, unsigned long ip)
3821{
3822 unsigned long flags;
3823
3824 if (unlikely(current->lockdep_recursion))
3825 return;
3826
3827 raw_local_irq_save(flags);
3828 current->lockdep_recursion = 1;
3829 check_flags(flags);
3830 if (__lock_downgrade(lock, ip))
3831 check_chain_key(current);
3832 current->lockdep_recursion = 0;
3833 raw_local_irq_restore(flags);
3834}
3835EXPORT_SYMBOL_GPL(lock_downgrade);
3836
fbb9ce95
IM
3837/*
3838 * We are not always called with irqs disabled - do that here,
3839 * and also avoid lockdep recursion:
3840 */
1d09daa5 3841void lock_acquire(struct lockdep_map *lock, unsigned int subclass,
7531e2f3
PZ
3842 int trylock, int read, int check,
3843 struct lockdep_map *nest_lock, unsigned long ip)
fbb9ce95
IM
3844{
3845 unsigned long flags;
3846
3847 if (unlikely(current->lockdep_recursion))
3848 return;
3849
3850 raw_local_irq_save(flags);
3851 check_flags(flags);
3852
3853 current->lockdep_recursion = 1;
db2c4c77 3854 trace_lock_acquire(lock, subclass, trylock, read, check, nest_lock, ip);
fbb9ce95 3855 __lock_acquire(lock, subclass, trylock, read, check,
21199f27 3856 irqs_disabled_flags(flags), nest_lock, ip, 0, 0);
fbb9ce95
IM
3857 current->lockdep_recursion = 0;
3858 raw_local_irq_restore(flags);
3859}
fbb9ce95
IM
3860EXPORT_SYMBOL_GPL(lock_acquire);
3861
1d09daa5 3862void lock_release(struct lockdep_map *lock, int nested,
0764d23c 3863 unsigned long ip)
fbb9ce95
IM
3864{
3865 unsigned long flags;
3866
3867 if (unlikely(current->lockdep_recursion))
3868 return;
3869
3870 raw_local_irq_save(flags);
3871 check_flags(flags);
3872 current->lockdep_recursion = 1;
93135439 3873 trace_lock_release(lock, ip);
e0f56fd7
PZ
3874 if (__lock_release(lock, nested, ip))
3875 check_chain_key(current);
fbb9ce95
IM
3876 current->lockdep_recursion = 0;
3877 raw_local_irq_restore(flags);
3878}
fbb9ce95
IM
3879EXPORT_SYMBOL_GPL(lock_release);
3880
f8319483 3881int lock_is_held_type(struct lockdep_map *lock, int read)
f607c668
PZ
3882{
3883 unsigned long flags;
3884 int ret = 0;
3885
3886 if (unlikely(current->lockdep_recursion))
f2513cde 3887 return 1; /* avoid false negative lockdep_assert_held() */
f607c668
PZ
3888
3889 raw_local_irq_save(flags);
3890 check_flags(flags);
3891
3892 current->lockdep_recursion = 1;
f8319483 3893 ret = __lock_is_held(lock, read);
f607c668
PZ
3894 current->lockdep_recursion = 0;
3895 raw_local_irq_restore(flags);
3896
3897 return ret;
3898}
f8319483 3899EXPORT_SYMBOL_GPL(lock_is_held_type);
f607c668 3900
e7904a28 3901struct pin_cookie lock_pin_lock(struct lockdep_map *lock)
a24fc60d 3902{
e7904a28 3903 struct pin_cookie cookie = NIL_COOKIE;
a24fc60d
PZ
3904 unsigned long flags;
3905
3906 if (unlikely(current->lockdep_recursion))
e7904a28 3907 return cookie;
a24fc60d
PZ
3908
3909 raw_local_irq_save(flags);
3910 check_flags(flags);
3911
3912 current->lockdep_recursion = 1;
e7904a28 3913 cookie = __lock_pin_lock(lock);
a24fc60d
PZ
3914 current->lockdep_recursion = 0;
3915 raw_local_irq_restore(flags);
e7904a28
PZ
3916
3917 return cookie;
a24fc60d
PZ
3918}
3919EXPORT_SYMBOL_GPL(lock_pin_lock);
3920
e7904a28
PZ
3921void lock_repin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
3922{
3923 unsigned long flags;
3924
3925 if (unlikely(current->lockdep_recursion))
3926 return;
3927
3928 raw_local_irq_save(flags);
3929 check_flags(flags);
3930
3931 current->lockdep_recursion = 1;
3932 __lock_repin_lock(lock, cookie);
3933 current->lockdep_recursion = 0;
3934 raw_local_irq_restore(flags);
3935}
3936EXPORT_SYMBOL_GPL(lock_repin_lock);
3937
3938void lock_unpin_lock(struct lockdep_map *lock, struct pin_cookie cookie)
a24fc60d
PZ
3939{
3940 unsigned long flags;
3941
3942 if (unlikely(current->lockdep_recursion))
3943 return;
3944
3945 raw_local_irq_save(flags);
3946 check_flags(flags);
3947
3948 current->lockdep_recursion = 1;
e7904a28 3949 __lock_unpin_lock(lock, cookie);
a24fc60d
PZ
3950 current->lockdep_recursion = 0;
3951 raw_local_irq_restore(flags);
3952}
3953EXPORT_SYMBOL_GPL(lock_unpin_lock);
3954
cf40bd16
NP
3955void lockdep_set_current_reclaim_state(gfp_t gfp_mask)
3956{
7dea19f9 3957 current->lockdep_reclaim_gfp = current_gfp_context(gfp_mask);
cf40bd16
NP
3958}
3959
3960void lockdep_clear_current_reclaim_state(void)
3961{
3962 current->lockdep_reclaim_gfp = 0;
3963}
3964
f20786ff
PZ
3965#ifdef CONFIG_LOCK_STAT
3966static int
3967print_lock_contention_bug(struct task_struct *curr, struct lockdep_map *lock,
3968 unsigned long ip)
3969{
3970 if (!debug_locks_off())
3971 return 0;
3972 if (debug_locks_silent)
3973 return 0;
3974
b3fbab05
PM
3975 printk("\n");
3976 printk("=================================\n");
3977 printk("[ BUG: bad contention detected! ]\n");
fbdc4b9a 3978 print_kernel_ident();
b3fbab05 3979 printk("---------------------------------\n");
f20786ff 3980 printk("%s/%d is trying to contend lock (",
ba25f9dc 3981 curr->comm, task_pid_nr(curr));
f20786ff 3982 print_lockdep_cache(lock);
f943fe0f 3983 printk(KERN_CONT ") at:\n");
f20786ff
PZ
3984 print_ip_sym(ip);
3985 printk("but there are no locks held!\n");
3986 printk("\nother info that might help us debug this:\n");
3987 lockdep_print_held_locks(curr);
3988
3989 printk("\nstack backtrace:\n");
3990 dump_stack();
3991
3992 return 0;
3993}
3994
3995static void
3996__lock_contended(struct lockdep_map *lock, unsigned long ip)
3997{
3998 struct task_struct *curr = current;
41c2c5b8 3999 struct held_lock *hlock;
f20786ff
PZ
4000 struct lock_class_stats *stats;
4001 unsigned int depth;
c7e78cff 4002 int i, contention_point, contending_point;
f20786ff
PZ
4003
4004 depth = curr->lockdep_depth;
0119fee4
PZ
4005 /*
4006 * Whee, we contended on this lock, except it seems we're not
4007 * actually trying to acquire anything much at all..
4008 */
f20786ff
PZ
4009 if (DEBUG_LOCKS_WARN_ON(!depth))
4010 return;
4011
41c2c5b8
O
4012 hlock = find_held_lock(curr, lock, depth, &i);
4013 if (!hlock) {
4014 print_lock_contention_bug(curr, lock, ip);
4015 return;
f20786ff 4016 }
f20786ff 4017
bb97a91e
PZ
4018 if (hlock->instance != lock)
4019 return;
4020
3365e779 4021 hlock->waittime_stamp = lockstat_clock();
f20786ff 4022
c7e78cff
PZ
4023 contention_point = lock_point(hlock_class(hlock)->contention_point, ip);
4024 contending_point = lock_point(hlock_class(hlock)->contending_point,
4025 lock->ip);
f20786ff 4026
f82b217e 4027 stats = get_lock_stats(hlock_class(hlock));
c7e78cff
PZ
4028 if (contention_point < LOCKSTAT_POINTS)
4029 stats->contention_point[contention_point]++;
4030 if (contending_point < LOCKSTAT_POINTS)
4031 stats->contending_point[contending_point]++;
96645678
PZ
4032 if (lock->cpu != smp_processor_id())
4033 stats->bounces[bounce_contended + !!hlock->read]++;
f20786ff
PZ
4034 put_lock_stats(stats);
4035}
4036
4037static void
c7e78cff 4038__lock_acquired(struct lockdep_map *lock, unsigned long ip)
f20786ff
PZ
4039{
4040 struct task_struct *curr = current;
41c2c5b8 4041 struct held_lock *hlock;
f20786ff
PZ
4042 struct lock_class_stats *stats;
4043 unsigned int depth;
3365e779 4044 u64 now, waittime = 0;
96645678 4045 int i, cpu;
f20786ff
PZ
4046
4047 depth = curr->lockdep_depth;
0119fee4
PZ
4048 /*
4049 * Yay, we acquired ownership of this lock we didn't try to
4050 * acquire, how the heck did that happen?
4051 */
f20786ff
PZ
4052 if (DEBUG_LOCKS_WARN_ON(!depth))
4053 return;
4054
41c2c5b8
O
4055 hlock = find_held_lock(curr, lock, depth, &i);
4056 if (!hlock) {
4057 print_lock_contention_bug(curr, lock, _RET_IP_);
4058 return;
f20786ff 4059 }
f20786ff 4060
bb97a91e
PZ
4061 if (hlock->instance != lock)
4062 return;
4063
96645678
PZ
4064 cpu = smp_processor_id();
4065 if (hlock->waittime_stamp) {
3365e779 4066 now = lockstat_clock();
96645678
PZ
4067 waittime = now - hlock->waittime_stamp;
4068 hlock->holdtime_stamp = now;
4069 }
f20786ff 4070
883a2a31 4071 trace_lock_acquired(lock, ip);
2062501a 4072
f82b217e 4073 stats = get_lock_stats(hlock_class(hlock));
96645678
PZ
4074 if (waittime) {
4075 if (hlock->read)
4076 lock_time_inc(&stats->read_waittime, waittime);
4077 else
4078 lock_time_inc(&stats->write_waittime, waittime);
4079 }
4080 if (lock->cpu != cpu)
4081 stats->bounces[bounce_acquired + !!hlock->read]++;
f20786ff 4082 put_lock_stats(stats);
96645678
PZ
4083
4084 lock->cpu = cpu;
c7e78cff 4085 lock->ip = ip;
f20786ff
PZ
4086}
4087
4088void lock_contended(struct lockdep_map *lock, unsigned long ip)
4089{
4090 unsigned long flags;
4091
4092 if (unlikely(!lock_stat))
4093 return;
4094
4095 if (unlikely(current->lockdep_recursion))
4096 return;
4097
4098 raw_local_irq_save(flags);
4099 check_flags(flags);
4100 current->lockdep_recursion = 1;
db2c4c77 4101 trace_lock_contended(lock, ip);
f20786ff
PZ
4102 __lock_contended(lock, ip);
4103 current->lockdep_recursion = 0;
4104 raw_local_irq_restore(flags);
4105}
4106EXPORT_SYMBOL_GPL(lock_contended);
4107
c7e78cff 4108void lock_acquired(struct lockdep_map *lock, unsigned long ip)
f20786ff
PZ
4109{
4110 unsigned long flags;
4111
4112 if (unlikely(!lock_stat))
4113 return;
4114
4115 if (unlikely(current->lockdep_recursion))
4116 return;
4117
4118 raw_local_irq_save(flags);
4119 check_flags(flags);
4120 current->lockdep_recursion = 1;
c7e78cff 4121 __lock_acquired(lock, ip);
f20786ff
PZ
4122 current->lockdep_recursion = 0;
4123 raw_local_irq_restore(flags);
4124}
4125EXPORT_SYMBOL_GPL(lock_acquired);
4126#endif
4127
fbb9ce95
IM
4128/*
4129 * Used by the testsuite, sanitize the validator state
4130 * after a simulated failure:
4131 */
4132
4133void lockdep_reset(void)
4134{
4135 unsigned long flags;
23d95a03 4136 int i;
fbb9ce95
IM
4137
4138 raw_local_irq_save(flags);
4139 current->curr_chain_key = 0;
4140 current->lockdep_depth = 0;
4141 current->lockdep_recursion = 0;
4142 memset(current->held_locks, 0, MAX_LOCK_DEPTH*sizeof(struct held_lock));
4143 nr_hardirq_chains = 0;
4144 nr_softirq_chains = 0;
4145 nr_process_chains = 0;
4146 debug_locks = 1;
23d95a03 4147 for (i = 0; i < CHAINHASH_SIZE; i++)
a63f38cc 4148 INIT_HLIST_HEAD(chainhash_table + i);
fbb9ce95
IM
4149 raw_local_irq_restore(flags);
4150}
4151
4152static void zap_class(struct lock_class *class)
4153{
4154 int i;
4155
4156 /*
4157 * Remove all dependencies this lock is
4158 * involved in:
4159 */
4160 for (i = 0; i < nr_list_entries; i++) {
4161 if (list_entries[i].class == class)
4162 list_del_rcu(&list_entries[i].entry);
4163 }
4164 /*
4165 * Unhash the class and remove it from the all_lock_classes list:
4166 */
a63f38cc 4167 hlist_del_rcu(&class->hash_entry);
fbb9ce95
IM
4168 list_del_rcu(&class->lock_entry);
4169
cee34d88
PZ
4170 RCU_INIT_POINTER(class->key, NULL);
4171 RCU_INIT_POINTER(class->name, NULL);
fbb9ce95
IM
4172}
4173
fabe874a 4174static inline int within(const void *addr, void *start, unsigned long size)
fbb9ce95
IM
4175{
4176 return addr >= start && addr < start + size;
4177}
4178
35a9393c
PZ
4179/*
4180 * Used in module.c to remove lock classes from memory that is going to be
4181 * freed; and possibly re-used by other modules.
4182 *
4183 * We will have had one sync_sched() before getting here, so we're guaranteed
4184 * nobody will look up these exact classes -- they're properly dead but still
4185 * allocated.
4186 */
fbb9ce95
IM
4187void lockdep_free_key_range(void *start, unsigned long size)
4188{
35a9393c 4189 struct lock_class *class;
a63f38cc 4190 struct hlist_head *head;
fbb9ce95
IM
4191 unsigned long flags;
4192 int i;
5a26db5b 4193 int locked;
fbb9ce95
IM
4194
4195 raw_local_irq_save(flags);
5a26db5b 4196 locked = graph_lock();
fbb9ce95
IM
4197
4198 /*
4199 * Unhash all classes that were created by this module:
4200 */
4201 for (i = 0; i < CLASSHASH_SIZE; i++) {
4202 head = classhash_table + i;
a63f38cc 4203 hlist_for_each_entry_rcu(class, head, hash_entry) {
fbb9ce95
IM
4204 if (within(class->key, start, size))
4205 zap_class(class);
fabe874a
AV
4206 else if (within(class->name, start, size))
4207 zap_class(class);
4208 }
fbb9ce95
IM
4209 }
4210
5a26db5b
NP
4211 if (locked)
4212 graph_unlock();
fbb9ce95 4213 raw_local_irq_restore(flags);
35a9393c
PZ
4214
4215 /*
4216 * Wait for any possible iterators from look_up_lock_class() to pass
4217 * before continuing to free the memory they refer to.
4218 *
4219 * sync_sched() is sufficient because the read-side is IRQ disable.
4220 */
4221 synchronize_sched();
4222
4223 /*
4224 * XXX at this point we could return the resources to the pool;
4225 * instead we leak them. We would need to change to bitmap allocators
4226 * instead of the linear allocators we have now.
4227 */
fbb9ce95
IM
4228}
4229
4230void lockdep_reset_lock(struct lockdep_map *lock)
4231{
35a9393c 4232 struct lock_class *class;
a63f38cc 4233 struct hlist_head *head;
fbb9ce95
IM
4234 unsigned long flags;
4235 int i, j;
5a26db5b 4236 int locked;
fbb9ce95
IM
4237
4238 raw_local_irq_save(flags);
fbb9ce95
IM
4239
4240 /*
d6d897ce
IM
4241 * Remove all classes this lock might have:
4242 */
4243 for (j = 0; j < MAX_LOCKDEP_SUBCLASSES; j++) {
4244 /*
4245 * If the class exists we look it up and zap it:
4246 */
4247 class = look_up_lock_class(lock, j);
383776fa 4248 if (!IS_ERR_OR_NULL(class))
d6d897ce
IM
4249 zap_class(class);
4250 }
4251 /*
4252 * Debug check: in the end all mapped classes should
4253 * be gone.
fbb9ce95 4254 */
5a26db5b 4255 locked = graph_lock();
fbb9ce95
IM
4256 for (i = 0; i < CLASSHASH_SIZE; i++) {
4257 head = classhash_table + i;
a63f38cc 4258 hlist_for_each_entry_rcu(class, head, hash_entry) {
62016250
HM
4259 int match = 0;
4260
4261 for (j = 0; j < NR_LOCKDEP_CACHING_CLASSES; j++)
4262 match |= class == lock->class_cache[j];
4263
4264 if (unlikely(match)) {
0119fee4
PZ
4265 if (debug_locks_off_graph_unlock()) {
4266 /*
4267 * We all just reset everything, how did it match?
4268 */
74c383f1 4269 WARN_ON(1);
0119fee4 4270 }
d6d897ce 4271 goto out_restore;
fbb9ce95
IM
4272 }
4273 }
4274 }
5a26db5b
NP
4275 if (locked)
4276 graph_unlock();
d6d897ce
IM
4277
4278out_restore:
fbb9ce95
IM
4279 raw_local_irq_restore(flags);
4280}
4281
fbb9ce95
IM
4282void __init lockdep_info(void)
4283{
4284 printk("Lock dependency validator: Copyright (c) 2006 Red Hat, Inc., Ingo Molnar\n");
4285
b0788caf 4286 printk("... MAX_LOCKDEP_SUBCLASSES: %lu\n", MAX_LOCKDEP_SUBCLASSES);
fbb9ce95
IM
4287 printk("... MAX_LOCK_DEPTH: %lu\n", MAX_LOCK_DEPTH);
4288 printk("... MAX_LOCKDEP_KEYS: %lu\n", MAX_LOCKDEP_KEYS);
b0788caf 4289 printk("... CLASSHASH_SIZE: %lu\n", CLASSHASH_SIZE);
fbb9ce95
IM
4290 printk("... MAX_LOCKDEP_ENTRIES: %lu\n", MAX_LOCKDEP_ENTRIES);
4291 printk("... MAX_LOCKDEP_CHAINS: %lu\n", MAX_LOCKDEP_CHAINS);
4292 printk("... CHAINHASH_SIZE: %lu\n", CHAINHASH_SIZE);
4293
4294 printk(" memory used by lock dependency info: %lu kB\n",
4295 (sizeof(struct lock_class) * MAX_LOCKDEP_KEYS +
4296 sizeof(struct list_head) * CLASSHASH_SIZE +
4297 sizeof(struct lock_list) * MAX_LOCKDEP_ENTRIES +
4298 sizeof(struct lock_chain) * MAX_LOCKDEP_CHAINS +
90629209 4299 sizeof(struct list_head) * CHAINHASH_SIZE
4dd861d6 4300#ifdef CONFIG_PROVE_LOCKING
e351b660 4301 + sizeof(struct circular_queue)
4dd861d6 4302#endif
90629209 4303 ) / 1024
4dd861d6 4304 );
fbb9ce95
IM
4305
4306 printk(" per task-struct memory footprint: %lu bytes\n",
4307 sizeof(struct held_lock) * MAX_LOCK_DEPTH);
fbb9ce95
IM
4308}
4309
fbb9ce95
IM
4310static void
4311print_freed_lock_bug(struct task_struct *curr, const void *mem_from,
55794a41 4312 const void *mem_to, struct held_lock *hlock)
fbb9ce95
IM
4313{
4314 if (!debug_locks_off())
4315 return;
4316 if (debug_locks_silent)
4317 return;
4318
b3fbab05
PM
4319 printk("\n");
4320 printk("=========================\n");
4321 printk("[ BUG: held lock freed! ]\n");
fbdc4b9a 4322 print_kernel_ident();
b3fbab05 4323 printk("-------------------------\n");
fbb9ce95 4324 printk("%s/%d is freeing memory %p-%p, with a lock still held there!\n",
ba25f9dc 4325 curr->comm, task_pid_nr(curr), mem_from, mem_to-1);
55794a41 4326 print_lock(hlock);
fbb9ce95
IM
4327 lockdep_print_held_locks(curr);
4328
4329 printk("\nstack backtrace:\n");
4330 dump_stack();
4331}
4332
54561783
ON
4333static inline int not_in_range(const void* mem_from, unsigned long mem_len,
4334 const void* lock_from, unsigned long lock_len)
4335{
4336 return lock_from + lock_len <= mem_from ||
4337 mem_from + mem_len <= lock_from;
4338}
4339
fbb9ce95
IM
4340/*
4341 * Called when kernel memory is freed (or unmapped), or if a lock
4342 * is destroyed or reinitialized - this code checks whether there is
4343 * any held lock in the memory range of <from> to <to>:
4344 */
4345void debug_check_no_locks_freed(const void *mem_from, unsigned long mem_len)
4346{
fbb9ce95
IM
4347 struct task_struct *curr = current;
4348 struct held_lock *hlock;
4349 unsigned long flags;
4350 int i;
4351
4352 if (unlikely(!debug_locks))
4353 return;
4354
4355 local_irq_save(flags);
4356 for (i = 0; i < curr->lockdep_depth; i++) {
4357 hlock = curr->held_locks + i;
4358
54561783
ON
4359 if (not_in_range(mem_from, mem_len, hlock->instance,
4360 sizeof(*hlock->instance)))
fbb9ce95
IM
4361 continue;
4362
54561783 4363 print_freed_lock_bug(curr, mem_from, mem_from + mem_len, hlock);
fbb9ce95
IM
4364 break;
4365 }
4366 local_irq_restore(flags);
4367}
ed07536e 4368EXPORT_SYMBOL_GPL(debug_check_no_locks_freed);
fbb9ce95 4369
1b1d2fb4 4370static void print_held_locks_bug(void)
fbb9ce95
IM
4371{
4372 if (!debug_locks_off())
4373 return;
4374 if (debug_locks_silent)
4375 return;
4376
b3fbab05
PM
4377 printk("\n");
4378 printk("=====================================\n");
1b1d2fb4
CC
4379 printk("[ BUG: %s/%d still has locks held! ]\n",
4380 current->comm, task_pid_nr(current));
fbdc4b9a 4381 print_kernel_ident();
b3fbab05 4382 printk("-------------------------------------\n");
1b1d2fb4 4383 lockdep_print_held_locks(current);
fbb9ce95
IM
4384 printk("\nstack backtrace:\n");
4385 dump_stack();
4386}
4387
1b1d2fb4 4388void debug_check_no_locks_held(void)
fbb9ce95 4389{
1b1d2fb4
CC
4390 if (unlikely(current->lockdep_depth > 0))
4391 print_held_locks_bug();
fbb9ce95 4392}
1b1d2fb4 4393EXPORT_SYMBOL_GPL(debug_check_no_locks_held);
fbb9ce95 4394
8dce7a9a 4395#ifdef __KERNEL__
fbb9ce95
IM
4396void debug_show_all_locks(void)
4397{
4398 struct task_struct *g, *p;
4399 int count = 10;
4400 int unlock = 1;
4401
9c35dd7f
JP
4402 if (unlikely(!debug_locks)) {
4403 printk("INFO: lockdep is turned off.\n");
4404 return;
4405 }
fbb9ce95
IM
4406 printk("\nShowing all locks held in the system:\n");
4407
4408 /*
4409 * Here we try to get the tasklist_lock as hard as possible,
4410 * if not successful after 2 seconds we ignore it (but keep
4411 * trying). This is to enable a debug printout even if a
4412 * tasklist_lock-holding task deadlocks or crashes.
4413 */
4414retry:
4415 if (!read_trylock(&tasklist_lock)) {
4416 if (count == 10)
4417 printk("hm, tasklist_lock locked, retrying... ");
4418 if (count) {
4419 count--;
4420 printk(" #%d", 10-count);
4421 mdelay(200);
4422 goto retry;
4423 }
4424 printk(" ignoring it.\n");
4425 unlock = 0;
46fec7ac 4426 } else {
4427 if (count != 10)
4428 printk(KERN_CONT " locked it.\n");
fbb9ce95 4429 }
fbb9ce95
IM
4430
4431 do_each_thread(g, p) {
85684873
IM
4432 /*
4433 * It's not reliable to print a task's held locks
4434 * if it's not sleeping (or if it's not the current
4435 * task):
4436 */
4437 if (p->state == TASK_RUNNING && p != current)
4438 continue;
fbb9ce95
IM
4439 if (p->lockdep_depth)
4440 lockdep_print_held_locks(p);
4441 if (!unlock)
4442 if (read_trylock(&tasklist_lock))
4443 unlock = 1;
4444 } while_each_thread(g, p);
4445
4446 printk("\n");
4447 printk("=============================================\n\n");
4448
4449 if (unlock)
4450 read_unlock(&tasklist_lock);
4451}
fbb9ce95 4452EXPORT_SYMBOL_GPL(debug_show_all_locks);
8dce7a9a 4453#endif
fbb9ce95 4454
82a1fcb9
IM
4455/*
4456 * Careful: only use this function if you are sure that
4457 * the task cannot run in parallel!
4458 */
f1b499f0 4459void debug_show_held_locks(struct task_struct *task)
fbb9ce95 4460{
9c35dd7f
JP
4461 if (unlikely(!debug_locks)) {
4462 printk("INFO: lockdep is turned off.\n");
4463 return;
4464 }
fbb9ce95
IM
4465 lockdep_print_held_locks(task);
4466}
fbb9ce95 4467EXPORT_SYMBOL_GPL(debug_show_held_locks);
b351d164 4468
722a9f92 4469asmlinkage __visible void lockdep_sys_exit(void)
b351d164
PZ
4470{
4471 struct task_struct *curr = current;
4472
4473 if (unlikely(curr->lockdep_depth)) {
4474 if (!debug_locks_off())
4475 return;
b3fbab05
PM
4476 printk("\n");
4477 printk("================================================\n");
4478 printk("[ BUG: lock held when returning to user space! ]\n");
fbdc4b9a 4479 print_kernel_ident();
b3fbab05 4480 printk("------------------------------------------------\n");
b351d164
PZ
4481 printk("%s/%d is leaving the kernel with locks still held!\n",
4482 curr->comm, curr->pid);
4483 lockdep_print_held_locks(curr);
4484 }
4485}
0632eb3d 4486
b3fbab05 4487void lockdep_rcu_suspicious(const char *file, const int line, const char *s)
0632eb3d
PM
4488{
4489 struct task_struct *curr = current;
4490
2b3fc35f 4491#ifndef CONFIG_PROVE_RCU_REPEATEDLY
0632eb3d
PM
4492 if (!debug_locks_off())
4493 return;
2b3fc35f
LJ
4494#endif /* #ifdef CONFIG_PROVE_RCU_REPEATEDLY */
4495 /* Note: the following can be executed concurrently, so be careful. */
b3fbab05 4496 printk("\n");
4d4f88fa
PM
4497 pr_err("===============================\n");
4498 pr_err("[ ERR: suspicious RCU usage. ]\n");
fbdc4b9a 4499 print_kernel_ident();
4d4f88fa
PM
4500 pr_err("-------------------------------\n");
4501 pr_err("%s:%d %s!\n", file, line, s);
4502 pr_err("\nother info that might help us debug this:\n\n");
4503 pr_err("\n%srcu_scheduler_active = %d, debug_locks = %d\n",
c5fdcec9
PM
4504 !rcu_lockdep_current_cpu_online()
4505 ? "RCU used illegally from offline CPU!\n"
5c173eb8 4506 : !rcu_is_watching()
c5fdcec9
PM
4507 ? "RCU used illegally from idle CPU!\n"
4508 : "",
4509 rcu_scheduler_active, debug_locks);
0464e937
FW
4510
4511 /*
4512 * If a CPU is in the RCU-free window in idle (ie: in the section
4513 * between rcu_idle_enter() and rcu_idle_exit(), then RCU
4514 * considers that CPU to be in an "extended quiescent state",
4515 * which means that RCU will be completely ignoring that CPU.
4516 * Therefore, rcu_read_lock() and friends have absolutely no
4517 * effect on a CPU running in that state. In other words, even if
4518 * such an RCU-idle CPU has called rcu_read_lock(), RCU might well
4519 * delete data structures out from under it. RCU really has no
4520 * choice here: we need to keep an RCU-free window in idle where
4521 * the CPU may possibly enter into low power mode. This way we can
4522 * notice an extended quiescent state to other CPUs that started a grace
4523 * period. Otherwise we would delay any grace period as long as we run
4524 * in the idle task.
4525 *
4526 * So complain bitterly if someone does call rcu_read_lock(),
4527 * rcu_read_lock_bh() and so on from extended quiescent states.
4528 */
5c173eb8 4529 if (!rcu_is_watching())
0464e937
FW
4530 printk("RCU used illegally from extended quiescent state!\n");
4531
0632eb3d
PM
4532 lockdep_print_held_locks(curr);
4533 printk("\nstack backtrace:\n");
4534 dump_stack();
4535}
b3fbab05 4536EXPORT_SYMBOL_GPL(lockdep_rcu_suspicious);