4 * Runtime locking correctness validator
6 * Started by Ingo Molnar:
8 * Copyright (C) 2006,2007 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
9 * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra
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:
14 * - lock inversion scenarios
15 * - circular lock dependencies
16 * - hardirq/softirq safe/unsafe locking bugs
18 * Bugs are reported even if the current locking scenario does not cause
19 * any deadlock at this point.
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.
25 * Thanks to Arjan van de Ven for coming up with the initial idea of
26 * mapping lock dependencies runtime.
28 #define DISABLE_BRANCH_PROFILING
29 #include <linux/mutex.h>
30 #include <linux/sched.h>
31 #include <linux/sched/clock.h>
32 #include <linux/sched/task.h>
33 #include <linux/sched/mm.h>
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>
44 #include <linux/utsname.h>
45 #include <linux/hash.h>
46 #include <linux/ftrace.h>
47 #include <linux/stringify.h>
48 #include <linux/bitops.h>
49 #include <linux/gfp.h>
50 #include <linux/kmemcheck.h>
51 #include <linux/random.h>
52 #include <linux/jhash.h>
54 #include <asm/sections.h>
56 #include "lockdep_internals.h"
58 #define CREATE_TRACE_POINTS
59 #include <trace/events/lock.h>
61 #ifdef CONFIG_PROVE_LOCKING
62 int prove_locking
= 1;
63 module_param(prove_locking
, int, 0644);
65 #define prove_locking 0
68 #ifdef CONFIG_LOCK_STAT
70 module_param(lock_stat
, int, 0644);
76 * lockdep_lock: protects the lockdep graph, the hashes and the
77 * class/list/hash allocators.
79 * This is one of the rare exceptions where it's justified
80 * to use a raw spinlock - we really dont want the spinlock
81 * code to recurse back into the lockdep code...
83 static arch_spinlock_t lockdep_lock
= (arch_spinlock_t
)__ARCH_SPIN_LOCK_UNLOCKED
;
85 static int graph_lock(void)
87 arch_spin_lock(&lockdep_lock
);
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
95 arch_spin_unlock(&lockdep_lock
);
98 /* prevent any recursions within lockdep from causing deadlocks */
99 current
->lockdep_recursion
++;
103 static inline int graph_unlock(void)
105 if (debug_locks
&& !arch_spin_is_locked(&lockdep_lock
)) {
107 * The lockdep graph lock isn't locked while we expect it to
108 * be, we're confused now, bye!
110 return DEBUG_LOCKS_WARN_ON(1);
113 current
->lockdep_recursion
--;
114 arch_spin_unlock(&lockdep_lock
);
119 * Turn lock debugging off and return with 0 if it was off already,
120 * and also release the graph lock:
122 static inline int debug_locks_off_graph_unlock(void)
124 int ret
= debug_locks_off();
126 arch_spin_unlock(&lockdep_lock
);
131 unsigned long nr_list_entries
;
132 static struct lock_list list_entries
[MAX_LOCKDEP_ENTRIES
];
135 * All data structures here are protected by the global debug_lock.
137 * Mutex key structs only get allocated, once during bootup, and never
138 * get freed - this significantly simplifies the debugging code.
140 unsigned long nr_lock_classes
;
141 static struct lock_class lock_classes
[MAX_LOCKDEP_KEYS
];
143 static inline struct lock_class
*hlock_class(struct held_lock
*hlock
)
145 if (!hlock
->class_idx
) {
147 * Someone passed in garbage, we give up.
149 DEBUG_LOCKS_WARN_ON(1);
152 return lock_classes
+ hlock
->class_idx
- 1;
155 #ifdef CONFIG_LOCK_STAT
156 static DEFINE_PER_CPU(struct lock_class_stats
[MAX_LOCKDEP_KEYS
], cpu_lock_stats
);
158 static inline u64
lockstat_clock(void)
160 return local_clock();
163 static int lock_point(unsigned long points
[], unsigned long ip
)
167 for (i
= 0; i
< LOCKSTAT_POINTS
; i
++) {
168 if (points
[i
] == 0) {
179 static void lock_time_inc(struct lock_time
*lt
, u64 time
)
184 if (time
< lt
->min
|| !lt
->nr
)
191 static inline void lock_time_add(struct lock_time
*src
, struct lock_time
*dst
)
196 if (src
->max
> dst
->max
)
199 if (src
->min
< dst
->min
|| !dst
->nr
)
202 dst
->total
+= src
->total
;
206 struct lock_class_stats
lock_stats(struct lock_class
*class)
208 struct lock_class_stats stats
;
211 memset(&stats
, 0, sizeof(struct lock_class_stats
));
212 for_each_possible_cpu(cpu
) {
213 struct lock_class_stats
*pcs
=
214 &per_cpu(cpu_lock_stats
, cpu
)[class - lock_classes
];
216 for (i
= 0; i
< ARRAY_SIZE(stats
.contention_point
); i
++)
217 stats
.contention_point
[i
] += pcs
->contention_point
[i
];
219 for (i
= 0; i
< ARRAY_SIZE(stats
.contending_point
); i
++)
220 stats
.contending_point
[i
] += pcs
->contending_point
[i
];
222 lock_time_add(&pcs
->read_waittime
, &stats
.read_waittime
);
223 lock_time_add(&pcs
->write_waittime
, &stats
.write_waittime
);
225 lock_time_add(&pcs
->read_holdtime
, &stats
.read_holdtime
);
226 lock_time_add(&pcs
->write_holdtime
, &stats
.write_holdtime
);
228 for (i
= 0; i
< ARRAY_SIZE(stats
.bounces
); i
++)
229 stats
.bounces
[i
] += pcs
->bounces
[i
];
235 void clear_lock_stats(struct lock_class
*class)
239 for_each_possible_cpu(cpu
) {
240 struct lock_class_stats
*cpu_stats
=
241 &per_cpu(cpu_lock_stats
, cpu
)[class - lock_classes
];
243 memset(cpu_stats
, 0, sizeof(struct lock_class_stats
));
245 memset(class->contention_point
, 0, sizeof(class->contention_point
));
246 memset(class->contending_point
, 0, sizeof(class->contending_point
));
249 static struct lock_class_stats
*get_lock_stats(struct lock_class
*class)
251 return &get_cpu_var(cpu_lock_stats
)[class - lock_classes
];
254 static void put_lock_stats(struct lock_class_stats
*stats
)
256 put_cpu_var(cpu_lock_stats
);
259 static void lock_release_holdtime(struct held_lock
*hlock
)
261 struct lock_class_stats
*stats
;
267 holdtime
= lockstat_clock() - hlock
->holdtime_stamp
;
269 stats
= get_lock_stats(hlock_class(hlock
));
271 lock_time_inc(&stats
->read_holdtime
, holdtime
);
273 lock_time_inc(&stats
->write_holdtime
, holdtime
);
274 put_lock_stats(stats
);
277 static inline void lock_release_holdtime(struct held_lock
*hlock
)
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.
287 LIST_HEAD(all_lock_classes
);
290 * The lockdep classes are in a hash-table as well, for fast lookup:
292 #define CLASSHASH_BITS (MAX_LOCKDEP_KEYS_BITS - 1)
293 #define CLASSHASH_SIZE (1UL << CLASSHASH_BITS)
294 #define __classhashfn(key) hash_long((unsigned long)key, CLASSHASH_BITS)
295 #define classhashentry(key) (classhash_table + __classhashfn((key)))
297 static struct hlist_head classhash_table
[CLASSHASH_SIZE
];
300 * We put the lock dependency chains into a hash-table as well, to cache
303 #define CHAINHASH_BITS (MAX_LOCKDEP_CHAINS_BITS-1)
304 #define CHAINHASH_SIZE (1UL << CHAINHASH_BITS)
305 #define __chainhashfn(chain) hash_long(chain, CHAINHASH_BITS)
306 #define chainhashentry(chain) (chainhash_table + __chainhashfn((chain)))
308 static struct hlist_head chainhash_table
[CHAINHASH_SIZE
];
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
316 static inline u64
iterate_chain_key(u64 key
, u32 idx
)
318 u32 k0
= key
, k1
= key
>> 32;
320 __jhash_mix(idx
, k0
, k1
); /* Macro that modifies arguments! */
322 return k0
| (u64
)k1
<< 32;
325 void lockdep_off(void)
327 current
->lockdep_recursion
++;
329 EXPORT_SYMBOL(lockdep_off
);
331 void lockdep_on(void)
333 current
->lockdep_recursion
--;
335 EXPORT_SYMBOL(lockdep_on
);
338 * Debugging switches:
342 #define VERY_VERBOSE 0
345 # define HARDIRQ_VERBOSE 1
346 # define SOFTIRQ_VERBOSE 1
347 # define RECLAIM_VERBOSE 1
349 # define HARDIRQ_VERBOSE 0
350 # define SOFTIRQ_VERBOSE 0
351 # define RECLAIM_VERBOSE 0
354 #if VERBOSE || HARDIRQ_VERBOSE || SOFTIRQ_VERBOSE || RECLAIM_VERBOSE
356 * Quick filtering for interesting events:
358 static int class_filter(struct lock_class
*class)
362 if (class->name_version
== 1 &&
363 !strcmp(class->name
, "lockname"))
365 if (class->name_version
== 1 &&
366 !strcmp(class->name
, "&struct->lockfield"))
369 /* Filter everything else. 1 would be to allow everything else */
374 static int verbose(struct lock_class
*class)
377 return class_filter(class);
383 * Stack-trace: tightly packed array of stack backtrace
384 * addresses. Protected by the graph_lock.
386 unsigned long nr_stack_trace_entries
;
387 static unsigned long stack_trace
[MAX_STACK_TRACE_ENTRIES
];
389 static void print_lockdep_off(const char *bug_msg
)
391 printk(KERN_DEBUG
"%s\n", bug_msg
);
392 printk(KERN_DEBUG
"turning off the locking correctness validator.\n");
393 #ifdef CONFIG_LOCK_STAT
394 printk(KERN_DEBUG
"Please attach the output of /proc/lock_stat to the bug report\n");
398 static int save_trace(struct stack_trace
*trace
)
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
;
406 save_stack_trace(trace
);
409 * Some daft arches put -1 at the end to indicate its a full trace.
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>
415 if (trace
->nr_entries
!= 0 &&
416 trace
->entries
[trace
->nr_entries
-1] == ULONG_MAX
)
419 trace
->max_entries
= trace
->nr_entries
;
421 nr_stack_trace_entries
+= trace
->nr_entries
;
423 if (nr_stack_trace_entries
>= MAX_STACK_TRACE_ENTRIES
-1) {
424 if (!debug_locks_off_graph_unlock())
427 print_lockdep_off("BUG: MAX_STACK_TRACE_ENTRIES too low!");
436 unsigned int nr_hardirq_chains
;
437 unsigned int nr_softirq_chains
;
438 unsigned int nr_process_chains
;
439 unsigned int max_lockdep_depth
;
441 #ifdef CONFIG_DEBUG_LOCKDEP
443 * Various lockdep statistics:
445 DEFINE_PER_CPU(struct lockdep_stats
, lockdep_stats
);
452 #define __USAGE(__STATE) \
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",
458 static const char *usage_str
[] =
460 #define LOCKDEP_STATE(__STATE) __USAGE(__STATE)
461 #include "lockdep_states.h"
463 [LOCK_USED
] = "INITIAL USE",
466 const char * __get_key_name(struct lockdep_subclass_key
*key
, char *str
)
468 return kallsyms_lookup((unsigned long)key
, NULL
, NULL
, NULL
, str
);
471 static inline unsigned long lock_flag(enum lock_usage_bit bit
)
476 static char get_usage_char(struct lock_class
*class, enum lock_usage_bit bit
)
480 if (class->usage_mask
& lock_flag(bit
+ 2))
482 if (class->usage_mask
& lock_flag(bit
)) {
484 if (class->usage_mask
& lock_flag(bit
+ 2))
491 void get_usage_chars(struct lock_class
*class, char usage
[LOCK_USAGE_CHARS
])
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"
504 static void __print_lock_name(struct lock_class
*class)
506 char str
[KSYM_NAME_LEN
];
511 name
= __get_key_name(class->key
, str
);
512 printk(KERN_CONT
"%s", name
);
514 printk(KERN_CONT
"%s", name
);
515 if (class->name_version
> 1)
516 printk(KERN_CONT
"#%d", class->name_version
);
518 printk(KERN_CONT
"/%d", class->subclass
);
522 static void print_lock_name(struct lock_class
*class)
524 char usage
[LOCK_USAGE_CHARS
];
526 get_usage_chars(class, usage
);
528 printk(KERN_CONT
" (");
529 __print_lock_name(class);
530 printk(KERN_CONT
"){%s}", usage
);
533 static void print_lockdep_cache(struct lockdep_map
*lock
)
536 char str
[KSYM_NAME_LEN
];
540 name
= __get_key_name(lock
->key
->subkeys
, str
);
542 printk(KERN_CONT
"%s", name
);
545 static void print_lock(struct held_lock
*hlock
)
548 * We can be called locklessly through debug_show_all_locks() so be
549 * extra careful, the hlock might have been released and cleared.
551 unsigned int class_idx
= hlock
->class_idx
;
553 /* Don't re-read hlock->class_idx, can't use READ_ONCE() on bitfields: */
556 if (!class_idx
|| (class_idx
- 1) >= MAX_LOCKDEP_KEYS
) {
557 printk(KERN_CONT
"<RELEASED>\n");
561 print_lock_name(lock_classes
+ class_idx
- 1);
562 printk(KERN_CONT
", at: [<%p>] %pS\n",
563 (void *)hlock
->acquire_ip
, (void *)hlock
->acquire_ip
);
566 static void lockdep_print_held_locks(struct task_struct
*curr
)
568 int i
, depth
= curr
->lockdep_depth
;
571 printk("no locks held by %s/%d.\n", curr
->comm
, task_pid_nr(curr
));
574 printk("%d lock%s held by %s/%d:\n",
575 depth
, depth
> 1 ? "s" : "", curr
->comm
, task_pid_nr(curr
));
577 for (i
= 0; i
< depth
; i
++) {
579 print_lock(curr
->held_locks
+ i
);
583 static void print_kernel_ident(void)
585 printk("%s %.*s %s\n", init_utsname()->release
,
586 (int)strcspn(init_utsname()->version
, " "),
587 init_utsname()->version
,
591 static int very_verbose(struct lock_class
*class)
594 return class_filter(class);
600 * Is this the address of a static object:
603 static int static_obj(void *obj
)
605 unsigned long start
= (unsigned long) &_stext
,
606 end
= (unsigned long) &_end
,
607 addr
= (unsigned long) obj
;
612 if ((addr
>= start
) && (addr
< end
))
615 if (arch_is_kernel_data(addr
))
619 * in-kernel percpu var?
621 if (is_kernel_percpu_address(addr
))
625 * module static or percpu var?
627 return is_module_address(addr
) || is_module_percpu_address(addr
);
632 * To make lock name printouts unique, we calculate a unique
633 * class->name_version generation counter:
635 static int count_matching_names(struct lock_class
*new_class
)
637 struct lock_class
*class;
640 if (!new_class
->name
)
643 list_for_each_entry_rcu(class, &all_lock_classes
, lock_entry
) {
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
);
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.
658 static inline struct lock_class
*
659 look_up_lock_class(struct lockdep_map
*lock
, unsigned int subclass
)
661 struct lockdep_subclass_key
*key
;
662 struct hlist_head
*hash_head
;
663 struct lock_class
*class;
664 bool is_static
= false;
666 if (unlikely(subclass
>= MAX_LOCKDEP_SUBCLASSES
)) {
669 "BUG: looking up invalid subclass: %u\n", subclass
);
671 "turning off the locking correctness validator.\n");
677 * Static locks do not have their class-keys yet - for them the key
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
682 if (unlikely(!lock
->key
)) {
683 unsigned long can_addr
, addr
= (unsigned long)lock
;
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
;
692 return ERR_PTR(-EINVAL
);
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.
702 BUILD_BUG_ON(sizeof(struct lock_class_key
) >
703 sizeof(struct lockdep_map
));
705 key
= lock
->key
->subkeys
+ subclass
;
707 hash_head
= classhashentry(key
);
710 * We do an RCU walk of the hash, see lockdep_free_key_range().
712 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
715 hlist_for_each_entry_rcu(class, hash_head
, hash_entry
) {
716 if (class->key
== key
) {
718 * Huh! same key, different name? Did someone trample
719 * on some memory? We're most confused.
721 WARN_ON_ONCE(class->name
!= lock
->name
);
726 return is_static
|| static_obj(lock
->key
) ? NULL
: ERR_PTR(-EINVAL
);
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.
734 static struct lock_class
*
735 register_lock_class(struct lockdep_map
*lock
, unsigned int subclass
, int force
)
737 struct lockdep_subclass_key
*key
;
738 struct hlist_head
*hash_head
;
739 struct lock_class
*class;
741 DEBUG_LOCKS_WARN_ON(!irqs_disabled());
743 class = look_up_lock_class(lock
, subclass
);
744 if (likely(!IS_ERR_OR_NULL(class)))
745 goto out_set_class_cache
;
748 * Debug-check: all keys must be persistent!
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");
759 key
= lock
->key
->subkeys
+ subclass
;
760 hash_head
= classhashentry(key
);
766 * We have to do the hash-walk again, to avoid races
769 hlist_for_each_entry_rcu(class, hash_head
, hash_entry
) {
770 if (class->key
== key
)
775 * Allocate a new key from the static array, and add it to
778 if (nr_lock_classes
>= MAX_LOCKDEP_KEYS
) {
779 if (!debug_locks_off_graph_unlock()) {
783 print_lockdep_off("BUG: MAX_LOCKDEP_KEYS too low!");
787 class = lock_classes
+ nr_lock_classes
++;
788 debug_atomic_inc(nr_unused_locks
);
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);
797 * We use RCU's safe list-add method to make
798 * parallel walking of the hash-list safe:
800 hlist_add_head_rcu(&class->hash_entry
, hash_head
);
802 * Add it to the global list of classes:
804 list_add_tail_rcu(&class->lock_entry
, &all_lock_classes
);
806 if (verbose(class)) {
809 printk("\nnew class %p: %s", class->key
, class->name
);
810 if (class->name_version
> 1)
811 printk(KERN_CONT
"#%d", class->name_version
);
812 printk(KERN_CONT
"\n");
823 if (!subclass
|| force
)
824 lock
->class_cache
[0] = class;
825 else if (subclass
< NR_LOCKDEP_CACHING_CLASSES
)
826 lock
->class_cache
[subclass
] = class;
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.
832 if (DEBUG_LOCKS_WARN_ON(class->subclass
!= subclass
))
838 #ifdef CONFIG_PROVE_LOCKING
840 * Allocate a lockdep entry. (assumes the graph_lock held, returns
841 * with NULL on failure)
843 static struct lock_list
*alloc_list_entry(void)
845 if (nr_list_entries
>= MAX_LOCKDEP_ENTRIES
) {
846 if (!debug_locks_off_graph_unlock())
849 print_lockdep_off("BUG: MAX_LOCKDEP_ENTRIES too low!");
853 return list_entries
+ nr_list_entries
++;
857 * Add a new dependency to the head of the list:
859 static int add_lock_to_list(struct lock_class
*this, struct list_head
*head
,
860 unsigned long ip
, int distance
,
861 struct stack_trace
*trace
)
863 struct lock_list
*entry
;
865 * Lock not present yet - get a new dependency struct and
866 * add it to the list:
868 entry
= alloc_list_entry();
873 entry
->distance
= distance
;
874 entry
->trace
= *trace
;
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().
880 list_add_tail_rcu(&entry
->entry
, head
);
886 * For good efficiency of modular, we use power of 2
888 #define MAX_CIRCULAR_QUEUE_SIZE 4096UL
889 #define CQ_MASK (MAX_CIRCULAR_QUEUE_SIZE-1)
892 * The circular_queue and helpers is used to implement the
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.
897 struct circular_queue
{
898 unsigned long element
[MAX_CIRCULAR_QUEUE_SIZE
];
899 unsigned int front
, rear
;
902 static struct circular_queue lock_cq
;
904 unsigned int max_bfs_queue_depth
;
906 static unsigned int lockdep_dependency_gen_id
;
908 static inline void __cq_init(struct circular_queue
*cq
)
910 cq
->front
= cq
->rear
= 0;
911 lockdep_dependency_gen_id
++;
914 static inline int __cq_empty(struct circular_queue
*cq
)
916 return (cq
->front
== cq
->rear
);
919 static inline int __cq_full(struct circular_queue
*cq
)
921 return ((cq
->rear
+ 1) & CQ_MASK
) == cq
->front
;
924 static inline int __cq_enqueue(struct circular_queue
*cq
, unsigned long elem
)
929 cq
->element
[cq
->rear
] = elem
;
930 cq
->rear
= (cq
->rear
+ 1) & CQ_MASK
;
934 static inline int __cq_dequeue(struct circular_queue
*cq
, unsigned long *elem
)
939 *elem
= cq
->element
[cq
->front
];
940 cq
->front
= (cq
->front
+ 1) & CQ_MASK
;
944 static inline unsigned int __cq_get_elem_count(struct circular_queue
*cq
)
946 return (cq
->rear
- cq
->front
) & CQ_MASK
;
949 static inline void mark_lock_accessed(struct lock_list
*lock
,
950 struct lock_list
*parent
)
954 nr
= lock
- list_entries
;
955 WARN_ON(nr
>= nr_list_entries
); /* Out-of-bounds, input fail */
956 lock
->parent
= parent
;
957 lock
->class->dep_gen_id
= lockdep_dependency_gen_id
;
960 static inline unsigned long lock_accessed(struct lock_list
*lock
)
964 nr
= lock
- list_entries
;
965 WARN_ON(nr
>= nr_list_entries
); /* Out-of-bounds, input fail */
966 return lock
->class->dep_gen_id
== lockdep_dependency_gen_id
;
969 static inline struct lock_list
*get_lock_parent(struct lock_list
*child
)
971 return child
->parent
;
974 static inline int get_lock_depth(struct lock_list
*child
)
977 struct lock_list
*parent
;
979 while ((parent
= get_lock_parent(child
))) {
986 static int __bfs(struct lock_list
*source_entry
,
988 int (*match
)(struct lock_list
*entry
, void *data
),
989 struct lock_list
**target_entry
,
992 struct lock_list
*entry
;
993 struct list_head
*head
;
994 struct circular_queue
*cq
= &lock_cq
;
997 if (match(source_entry
, data
)) {
998 *target_entry
= source_entry
;
1004 head
= &source_entry
->class->locks_after
;
1006 head
= &source_entry
->class->locks_before
;
1008 if (list_empty(head
))
1012 __cq_enqueue(cq
, (unsigned long)source_entry
);
1014 while (!__cq_empty(cq
)) {
1015 struct lock_list
*lock
;
1017 __cq_dequeue(cq
, (unsigned long *)&lock
);
1025 head
= &lock
->class->locks_after
;
1027 head
= &lock
->class->locks_before
;
1029 DEBUG_LOCKS_WARN_ON(!irqs_disabled());
1031 list_for_each_entry_rcu(entry
, head
, entry
) {
1032 if (!lock_accessed(entry
)) {
1033 unsigned int cq_depth
;
1034 mark_lock_accessed(entry
, lock
);
1035 if (match(entry
, data
)) {
1036 *target_entry
= entry
;
1041 if (__cq_enqueue(cq
, (unsigned long)entry
)) {
1045 cq_depth
= __cq_get_elem_count(cq
);
1046 if (max_bfs_queue_depth
< cq_depth
)
1047 max_bfs_queue_depth
= cq_depth
;
1055 static inline int __bfs_forwards(struct lock_list
*src_entry
,
1057 int (*match
)(struct lock_list
*entry
, void *data
),
1058 struct lock_list
**target_entry
)
1060 return __bfs(src_entry
, data
, match
, target_entry
, 1);
1064 static inline int __bfs_backwards(struct lock_list
*src_entry
,
1066 int (*match
)(struct lock_list
*entry
, void *data
),
1067 struct lock_list
**target_entry
)
1069 return __bfs(src_entry
, data
, match
, target_entry
, 0);
1074 * Recursive, forwards-direction lock-dependency checking, used for
1075 * both noncyclic checking and for hardirq-unsafe/softirq-unsafe
1080 * Print a dependency chain entry (this is only done when a deadlock
1081 * has been detected):
1084 print_circular_bug_entry(struct lock_list
*target
, int depth
)
1086 if (debug_locks_silent
)
1088 printk("\n-> #%u", depth
);
1089 print_lock_name(target
->class);
1090 printk(KERN_CONT
":\n");
1091 print_stack_trace(&target
->trace
, 6);
1097 print_circular_lock_scenario(struct held_lock
*src
,
1098 struct held_lock
*tgt
,
1099 struct lock_list
*prt
)
1101 struct lock_class
*source
= hlock_class(src
);
1102 struct lock_class
*target
= hlock_class(tgt
);
1103 struct lock_class
*parent
= prt
->class;
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.
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.
1118 if (parent
!= source
) {
1119 printk("Chain exists of:\n ");
1120 __print_lock_name(source
);
1121 printk(KERN_CONT
" --> ");
1122 __print_lock_name(parent
);
1123 printk(KERN_CONT
" --> ");
1124 __print_lock_name(target
);
1125 printk(KERN_CONT
"\n\n");
1128 printk(" Possible unsafe locking scenario:\n\n");
1129 printk(" CPU0 CPU1\n");
1130 printk(" ---- ----\n");
1132 __print_lock_name(target
);
1133 printk(KERN_CONT
");\n");
1135 __print_lock_name(parent
);
1136 printk(KERN_CONT
");\n");
1138 __print_lock_name(target
);
1139 printk(KERN_CONT
");\n");
1141 __print_lock_name(source
);
1142 printk(KERN_CONT
");\n");
1143 printk("\n *** DEADLOCK ***\n\n");
1147 * When a circular dependency is detected, print the
1151 print_circular_bug_header(struct lock_list
*entry
, unsigned int depth
,
1152 struct held_lock
*check_src
,
1153 struct held_lock
*check_tgt
)
1155 struct task_struct
*curr
= current
;
1157 if (debug_locks_silent
)
1161 pr_warn("======================================================\n");
1162 pr_warn("WARNING: possible circular locking dependency detected\n");
1163 print_kernel_ident();
1164 pr_warn("------------------------------------------------------\n");
1165 printk("%s/%d is trying to acquire lock:\n",
1166 curr
->comm
, task_pid_nr(curr
));
1167 print_lock(check_src
);
1168 printk("\nbut task is already holding lock:\n");
1169 print_lock(check_tgt
);
1170 printk("\nwhich lock already depends on the new lock.\n\n");
1171 printk("\nthe existing dependency chain (in reverse order) is:\n");
1173 print_circular_bug_entry(entry
, depth
);
1178 static inline int class_equal(struct lock_list
*entry
, void *data
)
1180 return entry
->class == data
;
1183 static 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
)
1188 struct task_struct
*curr
= current
;
1189 struct lock_list
*parent
;
1190 struct lock_list
*first_parent
;
1193 if (!debug_locks_off_graph_unlock() || debug_locks_silent
)
1196 if (!save_trace(&this->trace
))
1199 depth
= get_lock_depth(target
);
1201 print_circular_bug_header(target
, depth
, check_src
, check_tgt
);
1203 parent
= get_lock_parent(target
);
1204 first_parent
= parent
;
1207 print_circular_bug_entry(parent
, --depth
);
1208 parent
= get_lock_parent(parent
);
1211 printk("\nother info that might help us debug this:\n\n");
1212 print_circular_lock_scenario(check_src
, check_tgt
,
1215 lockdep_print_held_locks(curr
);
1217 printk("\nstack backtrace:\n");
1223 static noinline
int print_bfs_bug(int ret
)
1225 if (!debug_locks_off_graph_unlock())
1229 * Breadth-first-search failed, graph got corrupted?
1231 WARN(1, "lockdep bfs error:%d\n", ret
);
1236 static int noop_count(struct lock_list
*entry
, void *data
)
1238 (*(unsigned long *)data
)++;
1242 static unsigned long __lockdep_count_forward_deps(struct lock_list
*this)
1244 unsigned long count
= 0;
1245 struct lock_list
*uninitialized_var(target_entry
);
1247 __bfs_forwards(this, (void *)&count
, noop_count
, &target_entry
);
1251 unsigned long lockdep_count_forward_deps(struct lock_class
*class)
1253 unsigned long ret
, flags
;
1254 struct lock_list
this;
1259 local_irq_save(flags
);
1260 arch_spin_lock(&lockdep_lock
);
1261 ret
= __lockdep_count_forward_deps(&this);
1262 arch_spin_unlock(&lockdep_lock
);
1263 local_irq_restore(flags
);
1268 static unsigned long __lockdep_count_backward_deps(struct lock_list
*this)
1270 unsigned long count
= 0;
1271 struct lock_list
*uninitialized_var(target_entry
);
1273 __bfs_backwards(this, (void *)&count
, noop_count
, &target_entry
);
1278 unsigned long lockdep_count_backward_deps(struct lock_class
*class)
1280 unsigned long ret
, flags
;
1281 struct lock_list
this;
1286 local_irq_save(flags
);
1287 arch_spin_lock(&lockdep_lock
);
1288 ret
= __lockdep_count_backward_deps(&this);
1289 arch_spin_unlock(&lockdep_lock
);
1290 local_irq_restore(flags
);
1296 * Prove that the dependency graph starting at <entry> can not
1297 * lead to <target>. Print an error and return 0 if it does.
1300 check_noncircular(struct lock_list
*root
, struct lock_class
*target
,
1301 struct lock_list
**target_entry
)
1305 debug_atomic_inc(nr_cyclic_checks
);
1307 result
= __bfs_forwards(root
, target
, class_equal
, target_entry
);
1312 #if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING)
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.
1319 static inline int usage_match(struct lock_list
*entry
, void *bit
)
1321 return entry
->class->usage_mask
& (1 << (enum lock_usage_bit
)bit
);
1327 * Find a node in the forwards-direction dependency sub-graph starting
1328 * at @root->class that matches @bit.
1330 * Return 0 if such a node exists in the subgraph, and put that node
1331 * into *@target_entry.
1333 * Return 1 otherwise and keep *@target_entry unchanged.
1334 * Return <0 on error.
1337 find_usage_forwards(struct lock_list
*root
, enum lock_usage_bit bit
,
1338 struct lock_list
**target_entry
)
1342 debug_atomic_inc(nr_find_usage_forwards_checks
);
1344 result
= __bfs_forwards(root
, (void *)bit
, usage_match
, target_entry
);
1350 * Find a node in the backwards-direction dependency sub-graph starting
1351 * at @root->class that matches @bit.
1353 * Return 0 if such a node exists in the subgraph, and put that node
1354 * into *@target_entry.
1356 * Return 1 otherwise and keep *@target_entry unchanged.
1357 * Return <0 on error.
1360 find_usage_backwards(struct lock_list
*root
, enum lock_usage_bit bit
,
1361 struct lock_list
**target_entry
)
1365 debug_atomic_inc(nr_find_usage_backwards_checks
);
1367 result
= __bfs_backwards(root
, (void *)bit
, usage_match
, target_entry
);
1372 static void print_lock_class_header(struct lock_class
*class, int depth
)
1376 printk("%*s->", depth
, "");
1377 print_lock_name(class);
1378 printk(KERN_CONT
" ops: %lu", class->ops
);
1379 printk(KERN_CONT
" {\n");
1381 for (bit
= 0; bit
< LOCK_USAGE_STATES
; bit
++) {
1382 if (class->usage_mask
& (1 << bit
)) {
1385 len
+= printk("%*s %s", depth
, "", usage_str
[bit
]);
1386 len
+= printk(KERN_CONT
" at:\n");
1387 print_stack_trace(class->usage_traces
+ bit
, len
);
1390 printk("%*s }\n", depth
, "");
1392 printk("%*s ... key at: [<%p>] %pS\n",
1393 depth
, "", class->key
, class->key
);
1397 * printk the shortest lock dependencies from @start to @end in reverse order:
1400 print_shortest_lock_dependencies(struct lock_list
*leaf
,
1401 struct lock_list
*root
)
1403 struct lock_list
*entry
= leaf
;
1406 /*compute depth from generated tree by BFS*/
1407 depth
= get_lock_depth(leaf
);
1410 print_lock_class_header(entry
->class, depth
);
1411 printk("%*s ... acquired at:\n", depth
, "");
1412 print_stack_trace(&entry
->trace
, 2);
1415 if (depth
== 0 && (entry
!= root
)) {
1416 printk("lockdep:%s bad path found in chain graph\n", __func__
);
1420 entry
= get_lock_parent(entry
);
1422 } while (entry
&& (depth
>= 0));
1428 print_irq_lock_scenario(struct lock_list
*safe_entry
,
1429 struct lock_list
*unsafe_entry
,
1430 struct lock_class
*prev_class
,
1431 struct lock_class
*next_class
)
1433 struct lock_class
*safe_class
= safe_entry
->class;
1434 struct lock_class
*unsafe_class
= unsafe_entry
->class;
1435 struct lock_class
*middle_class
= prev_class
;
1437 if (middle_class
== safe_class
)
1438 middle_class
= next_class
;
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.
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.
1453 if (middle_class
!= unsafe_class
) {
1454 printk("Chain exists of:\n ");
1455 __print_lock_name(safe_class
);
1456 printk(KERN_CONT
" --> ");
1457 __print_lock_name(middle_class
);
1458 printk(KERN_CONT
" --> ");
1459 __print_lock_name(unsafe_class
);
1460 printk(KERN_CONT
"\n\n");
1463 printk(" Possible interrupt unsafe locking scenario:\n\n");
1464 printk(" CPU0 CPU1\n");
1465 printk(" ---- ----\n");
1467 __print_lock_name(unsafe_class
);
1468 printk(KERN_CONT
");\n");
1469 printk(" local_irq_disable();\n");
1471 __print_lock_name(safe_class
);
1472 printk(KERN_CONT
");\n");
1474 __print_lock_name(middle_class
);
1475 printk(KERN_CONT
");\n");
1476 printk(" <Interrupt>\n");
1478 __print_lock_name(safe_class
);
1479 printk(KERN_CONT
");\n");
1480 printk("\n *** DEADLOCK ***\n\n");
1484 print_bad_irq_dependency(struct task_struct
*curr
,
1485 struct lock_list
*prev_root
,
1486 struct lock_list
*next_root
,
1487 struct lock_list
*backwards_entry
,
1488 struct lock_list
*forwards_entry
,
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
)
1495 if (!debug_locks_off_graph_unlock() || debug_locks_silent
)
1499 pr_warn("=====================================================\n");
1500 pr_warn("WARNING: %s-safe -> %s-unsafe lock order detected\n",
1501 irqclass
, irqclass
);
1502 print_kernel_ident();
1503 pr_warn("-----------------------------------------------------\n");
1504 printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] is trying to acquire:\n",
1505 curr
->comm
, task_pid_nr(curr
),
1506 curr
->hardirq_context
, hardirq_count() >> HARDIRQ_SHIFT
,
1507 curr
->softirq_context
, softirq_count() >> SOFTIRQ_SHIFT
,
1508 curr
->hardirqs_enabled
,
1509 curr
->softirqs_enabled
);
1512 printk("\nand this task is already holding:\n");
1514 printk("which would create a new lock dependency:\n");
1515 print_lock_name(hlock_class(prev
));
1516 printk(KERN_CONT
" ->");
1517 print_lock_name(hlock_class(next
));
1518 printk(KERN_CONT
"\n");
1520 printk("\nbut this new dependency connects a %s-irq-safe lock:\n",
1522 print_lock_name(backwards_entry
->class);
1523 printk("\n... which became %s-irq-safe at:\n", irqclass
);
1525 print_stack_trace(backwards_entry
->class->usage_traces
+ bit1
, 1);
1527 printk("\nto a %s-irq-unsafe lock:\n", irqclass
);
1528 print_lock_name(forwards_entry
->class);
1529 printk("\n... which became %s-irq-unsafe at:\n", irqclass
);
1532 print_stack_trace(forwards_entry
->class->usage_traces
+ bit2
, 1);
1534 printk("\nother info that might help us debug this:\n\n");
1535 print_irq_lock_scenario(backwards_entry
, forwards_entry
,
1536 hlock_class(prev
), hlock_class(next
));
1538 lockdep_print_held_locks(curr
);
1540 printk("\nthe dependencies between %s-irq-safe lock and the holding lock:\n", irqclass
);
1541 if (!save_trace(&prev_root
->trace
))
1543 print_shortest_lock_dependencies(backwards_entry
, prev_root
);
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
))
1549 print_shortest_lock_dependencies(forwards_entry
, next_root
);
1551 printk("\nstack backtrace:\n");
1558 check_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
)
1563 struct lock_list
this, that
;
1564 struct lock_list
*uninitialized_var(target_entry
);
1565 struct lock_list
*uninitialized_var(target_entry1
);
1569 this.class = hlock_class(prev
);
1570 ret
= find_usage_backwards(&this, bit_backwards
, &target_entry
);
1572 return print_bfs_bug(ret
);
1577 that
.class = hlock_class(next
);
1578 ret
= find_usage_forwards(&that
, bit_forwards
, &target_entry1
);
1580 return print_bfs_bug(ret
);
1584 return print_bad_irq_dependency(curr
, &this, &that
,
1585 target_entry
, target_entry1
,
1587 bit_backwards
, bit_forwards
, irqclass
);
1590 static const char *state_names
[] = {
1591 #define LOCKDEP_STATE(__STATE) \
1592 __stringify(__STATE),
1593 #include "lockdep_states.h"
1594 #undef LOCKDEP_STATE
1597 static const char *state_rnames
[] = {
1598 #define LOCKDEP_STATE(__STATE) \
1599 __stringify(__STATE)"-READ",
1600 #include "lockdep_states.h"
1601 #undef LOCKDEP_STATE
1604 static inline const char *state_name(enum lock_usage_bit bit
)
1606 return (bit
& 1) ? state_rnames
[bit
>> 2] : state_names
[bit
>> 2];
1609 static int exclusive_bit(int new_bit
)
1617 * bit 0 - write/read
1618 * bit 1 - used_in/enabled
1622 int state
= new_bit
& ~3;
1623 int dir
= new_bit
& 2;
1626 * keep state, bit flip the direction and strip read.
1628 return state
| (dir
^ 2);
1631 static int check_irq_usage(struct task_struct
*curr
, struct held_lock
*prev
,
1632 struct held_lock
*next
, enum lock_usage_bit bit
)
1635 * Prove that the new dependency does not connect a hardirq-safe
1636 * lock with a hardirq-unsafe lock - to achieve this we search
1637 * the backwards-subgraph starting at <prev>, and the
1638 * forwards-subgraph starting at <next>:
1640 if (!check_usage(curr
, prev
, next
, bit
,
1641 exclusive_bit(bit
), state_name(bit
)))
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
1649 * the backwards-subgraph starting at <prev>, and the
1650 * forwards-subgraph starting at <next>:
1652 if (!check_usage(curr
, prev
, next
, bit
,
1653 exclusive_bit(bit
), state_name(bit
)))
1660 check_prev_add_irq(struct task_struct
*curr
, struct held_lock
*prev
,
1661 struct held_lock
*next
)
1663 #define LOCKDEP_STATE(__STATE) \
1664 if (!check_irq_usage(curr, prev, next, LOCK_USED_IN_##__STATE)) \
1666 #include "lockdep_states.h"
1667 #undef LOCKDEP_STATE
1672 static void inc_chains(void)
1674 if (current
->hardirq_context
)
1675 nr_hardirq_chains
++;
1677 if (current
->softirq_context
)
1678 nr_softirq_chains
++;
1680 nr_process_chains
++;
1687 check_prev_add_irq(struct task_struct
*curr
, struct held_lock
*prev
,
1688 struct held_lock
*next
)
1693 static inline void inc_chains(void)
1695 nr_process_chains
++;
1701 print_deadlock_scenario(struct held_lock
*nxt
,
1702 struct held_lock
*prv
)
1704 struct lock_class
*next
= hlock_class(nxt
);
1705 struct lock_class
*prev
= hlock_class(prv
);
1707 printk(" Possible unsafe locking scenario:\n\n");
1711 __print_lock_name(prev
);
1712 printk(KERN_CONT
");\n");
1714 __print_lock_name(next
);
1715 printk(KERN_CONT
");\n");
1716 printk("\n *** DEADLOCK ***\n\n");
1717 printk(" May be due to missing lock nesting notation\n\n");
1721 print_deadlock_bug(struct task_struct
*curr
, struct held_lock
*prev
,
1722 struct held_lock
*next
)
1724 if (!debug_locks_off_graph_unlock() || debug_locks_silent
)
1728 pr_warn("============================================\n");
1729 pr_warn("WARNING: possible recursive locking detected\n");
1730 print_kernel_ident();
1731 pr_warn("--------------------------------------------\n");
1732 printk("%s/%d is trying to acquire lock:\n",
1733 curr
->comm
, task_pid_nr(curr
));
1735 printk("\nbut task is already holding lock:\n");
1738 printk("\nother info that might help us debug this:\n");
1739 print_deadlock_scenario(next
, prev
);
1740 lockdep_print_held_locks(curr
);
1742 printk("\nstack backtrace:\n");
1749 * Check whether we are holding such a class already.
1751 * (Note that this has to be done separately, because the graph cannot
1752 * detect such classes of deadlocks.)
1754 * Returns: 0 on deadlock detected, 1 on OK, 2 on recursive read
1757 check_deadlock(struct task_struct
*curr
, struct held_lock
*next
,
1758 struct lockdep_map
*next_instance
, int read
)
1760 struct held_lock
*prev
;
1761 struct held_lock
*nest
= NULL
;
1764 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
1765 prev
= curr
->held_locks
+ i
;
1767 if (prev
->instance
== next
->nest_lock
)
1770 if (hlock_class(prev
) != hlock_class(next
))
1774 * Allow read-after-read recursion of the same
1775 * lock class (i.e. read_lock(lock)+read_lock(lock)):
1777 if ((read
== 2) && prev
->read
)
1781 * We're holding the nest_lock, which serializes this lock's
1782 * nesting behaviour.
1787 return print_deadlock_bug(curr
, prev
, next
);
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:
1796 * - would the adding of the <prev> -> <next> dependency create a
1797 * circular dependency in the graph? [== circular deadlock]
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]
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]
1809 * any of these scenarios could lead to a deadlock.
1811 * Then if all the validations pass, we add the forwards and backwards
1815 check_prev_add(struct task_struct
*curr
, struct held_lock
*prev
,
1816 struct held_lock
*next
, int distance
, int *stack_saved
)
1818 struct lock_list
*entry
;
1820 struct lock_list
this;
1821 struct lock_list
*uninitialized_var(target_entry
);
1823 * Static variable, serialized by the graph_lock().
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.
1829 static struct stack_trace trace
;
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>.)
1837 * We are using global variables to control the recursion, to
1838 * keep the stackframe size of the recursive functions low:
1840 this.class = hlock_class(next
);
1842 ret
= check_noncircular(&this, hlock_class(prev
), &target_entry
);
1844 return print_circular_bug(&this, target_entry
, next
, prev
);
1845 else if (unlikely(ret
< 0))
1846 return print_bfs_bug(ret
);
1848 if (!check_prev_add_irq(curr
, prev
, next
))
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.
1859 if (next
->read
== 2 || prev
->read
== 2)
1862 * Is the <prev> -> <next> dependency already present?
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.)
1869 list_for_each_entry(entry
, &hlock_class(prev
)->locks_after
, entry
) {
1870 if (entry
->class == hlock_class(next
)) {
1872 entry
->distance
= 1;
1877 if (!*stack_saved
) {
1878 if (!save_trace(&trace
))
1884 * Ok, all validations passed, add the new lock
1885 * to the previous lock's dependency list:
1887 ret
= add_lock_to_list(hlock_class(next
),
1888 &hlock_class(prev
)->locks_after
,
1889 next
->acquire_ip
, distance
, &trace
);
1894 ret
= add_lock_to_list(hlock_class(prev
),
1895 &hlock_class(next
)->locks_before
,
1896 next
->acquire_ip
, distance
, &trace
);
1901 * Debugging printouts:
1903 if (verbose(hlock_class(prev
)) || verbose(hlock_class(next
))) {
1904 /* We drop graph lock, so another thread can overwrite trace. */
1907 printk("\n new dependency: ");
1908 print_lock_name(hlock_class(prev
));
1909 printk(KERN_CONT
" => ");
1910 print_lock_name(hlock_class(next
));
1911 printk(KERN_CONT
"\n");
1913 return graph_lock();
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.
1925 check_prevs_add(struct task_struct
*curr
, struct held_lock
*next
)
1927 int depth
= curr
->lockdep_depth
;
1928 int stack_saved
= 0;
1929 struct held_lock
*hlock
;
1934 * Depth must not be zero for a non-head lock:
1939 * At least two relevant locks must exist for this
1942 if (curr
->held_locks
[depth
].irq_context
!=
1943 curr
->held_locks
[depth
-1].irq_context
)
1947 int distance
= curr
->lockdep_depth
- depth
+ 1;
1948 hlock
= curr
->held_locks
+ depth
- 1;
1950 * Only non-recursive-read entries get new dependencies
1953 if (hlock
->read
!= 2 && hlock
->check
) {
1954 if (!check_prev_add(curr
, hlock
, next
,
1955 distance
, &stack_saved
))
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:
1963 if (!hlock
->trylock
)
1968 * End of lock-stack?
1973 * Stop the search if we cross into another context:
1975 if (curr
->held_locks
[depth
].irq_context
!=
1976 curr
->held_locks
[depth
-1].irq_context
)
1981 if (!debug_locks_off_graph_unlock())
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.
1994 unsigned long nr_lock_chains
;
1995 struct lock_chain lock_chains
[MAX_LOCKDEP_CHAINS
];
1996 int nr_chain_hlocks
;
1997 static u16 chain_hlocks
[MAX_LOCKDEP_CHAIN_HLOCKS
];
1999 struct lock_class
*lock_chain_get_class(struct lock_chain
*chain
, int i
)
2001 return lock_classes
+ chain_hlocks
[chain
->base
+ i
];
2005 * Returns the index of the first held_lock of the current chain
2007 static inline int get_first_held_lock(struct task_struct
*curr
,
2008 struct held_lock
*hlock
)
2011 struct held_lock
*hlock_curr
;
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
)
2023 #ifdef CONFIG_DEBUG_LOCKDEP
2025 * Returns the next chain_key iteration
2027 static u64
print_chain_key_iteration(int class_idx
, u64 chain_key
)
2029 u64 new_chain_key
= iterate_chain_key(chain_key
, class_idx
);
2031 printk(" class_idx:%d -> chain_key:%016Lx",
2033 (unsigned long long)new_chain_key
);
2034 return new_chain_key
;
2038 print_chain_keys_held_locks(struct task_struct
*curr
, struct held_lock
*hlock_next
)
2040 struct held_lock
*hlock
;
2042 int depth
= curr
->lockdep_depth
;
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
);
2053 print_chain_key_iteration(hlock_next
->class_idx
, chain_key
);
2054 print_lock(hlock_next
);
2057 static void print_chain_keys_chain(struct lock_chain
*chain
)
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
);
2068 print_lock_name(lock_classes
+ class_id
);
2073 static void print_collision(struct task_struct
*curr
,
2074 struct held_lock
*hlock_next
,
2075 struct lock_chain
*chain
)
2078 pr_warn("============================\n");
2079 pr_warn("WARNING: chain_key collision\n");
2080 print_kernel_ident();
2081 pr_warn("----------------------------\n");
2082 printk("%s/%d: ", current
->comm
, task_pid_nr(current
));
2083 printk("Hash chain already cached but the contents don't match!\n");
2085 printk("Held locks:");
2086 print_chain_keys_held_locks(curr
, hlock_next
);
2088 printk("Locks in cached chain:");
2089 print_chain_keys_chain(chain
);
2091 printk("\nstack backtrace:\n");
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
2102 static int check_no_collision(struct task_struct
*curr
,
2103 struct held_lock
*hlock
,
2104 struct lock_chain
*chain
)
2106 #ifdef CONFIG_DEBUG_LOCKDEP
2109 i
= get_first_held_lock(curr
, hlock
);
2111 if (DEBUG_LOCKS_WARN_ON(chain
->depth
!= curr
->lockdep_depth
- (i
- 1))) {
2112 print_collision(curr
, hlock
, chain
);
2116 for (j
= 0; j
< chain
->depth
- 1; j
++, i
++) {
2117 id
= curr
->held_locks
[i
].class_idx
- 1;
2119 if (DEBUG_LOCKS_WARN_ON(chain_hlocks
[chain
->base
+ j
] != id
)) {
2120 print_collision(curr
, hlock
, chain
);
2129 * Look up a dependency chain. If the key is not present yet then
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.)
2134 static inline int lookup_chain_cache(struct task_struct
*curr
,
2135 struct held_lock
*hlock
,
2138 struct lock_class
*class = hlock_class(hlock
);
2139 struct hlist_head
*hash_head
= chainhashentry(chain_key
);
2140 struct lock_chain
*chain
;
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.
2148 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2151 * We can walk it lock-free, because entries only get added
2154 hlist_for_each_entry_rcu(chain
, hash_head
, entry
) {
2155 if (chain
->chain_key
== chain_key
) {
2157 debug_atomic_inc(chain_lookup_hits
);
2158 if (!check_no_collision(curr
, hlock
, chain
))
2161 if (very_verbose(class))
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
);
2169 if (very_verbose(class))
2170 printk("\nnew hash chain, key: %016Lx tail class: [%p] %s\n",
2171 (unsigned long long)chain_key
, class->key
, class->name
);
2173 * Allocate a new chain entry from the static array, and add
2179 * We have to walk the chain again locked - to avoid duplicates:
2181 hlist_for_each_entry(chain
, hash_head
, entry
) {
2182 if (chain
->chain_key
== chain_key
) {
2187 if (unlikely(nr_lock_chains
>= MAX_LOCKDEP_CHAINS
)) {
2188 if (!debug_locks_off_graph_unlock())
2191 print_lockdep_off("BUG: MAX_LOCKDEP_CHAINS too low!");
2195 chain
= lock_chains
+ nr_lock_chains
++;
2196 chain
->chain_key
= chain_key
;
2197 chain
->irq_context
= hlock
->irq_context
;
2198 i
= get_first_held_lock(curr
, hlock
);
2199 chain
->depth
= curr
->lockdep_depth
+ 1 - i
;
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
));
2205 if (likely(nr_chain_hlocks
+ chain
->depth
<= MAX_LOCKDEP_CHAIN_HLOCKS
)) {
2206 chain
->base
= nr_chain_hlocks
;
2207 for (j
= 0; j
< chain
->depth
- 1; j
++, i
++) {
2208 int lock_id
= curr
->held_locks
[i
].class_idx
- 1;
2209 chain_hlocks
[chain
->base
+ j
] = lock_id
;
2211 chain_hlocks
[chain
->base
+ j
] = class - lock_classes
;
2214 if (nr_chain_hlocks
< MAX_LOCKDEP_CHAIN_HLOCKS
)
2215 nr_chain_hlocks
+= chain
->depth
;
2217 #ifdef CONFIG_DEBUG_LOCKDEP
2219 * Important for check_no_collision().
2221 if (unlikely(nr_chain_hlocks
> MAX_LOCKDEP_CHAIN_HLOCKS
)) {
2222 if (!debug_locks_off_graph_unlock())
2225 print_lockdep_off("BUG: MAX_LOCKDEP_CHAIN_HLOCKS too low!");
2231 hlist_add_head_rcu(&chain
->entry
, hash_head
);
2232 debug_atomic_inc(chain_lookup_misses
);
2238 static int validate_chain(struct task_struct
*curr
, struct lockdep_map
*lock
,
2239 struct held_lock
*hlock
, int chain_head
, u64 chain_key
)
2242 * Trylock needs to maintain the stack of held locks, but it
2243 * does not add new dependencies, because trylock can be done
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)
2251 if (!hlock
->trylock
&& hlock
->check
&&
2252 lookup_chain_cache(curr
, hlock
, chain_key
)) {
2254 * Check whether last held lock:
2256 * - is irq-safe, if this lock is irq-unsafe
2257 * - is softirq-safe, if this lock is hardirq-unsafe
2259 * And check whether the new lock's dependency graph
2260 * could lead back to the previous lock.
2262 * any of these scenarios could lead to a deadlock. If
2265 int ret
= check_deadlock(curr
, hlock
, lock
, hlock
->read
);
2270 * Mark recursive read, as we jump over it when
2271 * building dependencies (just like we jump over
2277 * Add dependency only if this lock is not the head
2278 * of the chain, and if it's not a secondary read-lock:
2280 if (!chain_head
&& ret
!= 2)
2281 if (!check_prevs_add(curr
, hlock
))
2285 /* after lookup_chain_cache(): */
2286 if (unlikely(!debug_locks
))
2292 static inline int validate_chain(struct task_struct
*curr
,
2293 struct lockdep_map
*lock
, struct held_lock
*hlock
,
2294 int chain_head
, u64 chain_key
)
2301 * We are building curr_chain_key incrementally, so double-check
2302 * it from scratch, to make sure that it's done correctly:
2304 static void check_chain_key(struct task_struct
*curr
)
2306 #ifdef CONFIG_DEBUG_LOCKDEP
2307 struct held_lock
*hlock
, *prev_hlock
= NULL
;
2311 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
2312 hlock
= curr
->held_locks
+ i
;
2313 if (chain_key
!= hlock
->prev_chain_key
) {
2316 * We got mighty confused, our chain keys don't match
2317 * with what we expect, someone trample on our task state?
2319 WARN(1, "hm#1, depth: %u [%u], %016Lx != %016Lx\n",
2320 curr
->lockdep_depth
, i
,
2321 (unsigned long long)chain_key
,
2322 (unsigned long long)hlock
->prev_chain_key
);
2326 * Whoops ran out of static storage again?
2328 if (DEBUG_LOCKS_WARN_ON(hlock
->class_idx
> MAX_LOCKDEP_KEYS
))
2331 if (prev_hlock
&& (prev_hlock
->irq_context
!=
2332 hlock
->irq_context
))
2334 chain_key
= iterate_chain_key(chain_key
, hlock
->class_idx
);
2337 if (chain_key
!= curr
->curr_chain_key
) {
2340 * More smoking hash instead of calculating it, damn see these
2341 * numbers float.. I bet that a pink elephant stepped on my memory.
2343 WARN(1, "hm#2, depth: %u [%u], %016Lx != %016Lx\n",
2344 curr
->lockdep_depth
, i
,
2345 (unsigned long long)chain_key
,
2346 (unsigned long long)curr
->curr_chain_key
);
2352 print_usage_bug_scenario(struct held_lock
*lock
)
2354 struct lock_class
*class = hlock_class(lock
);
2356 printk(" Possible unsafe locking scenario:\n\n");
2360 __print_lock_name(class);
2361 printk(KERN_CONT
");\n");
2362 printk(" <Interrupt>\n");
2364 __print_lock_name(class);
2365 printk(KERN_CONT
");\n");
2366 printk("\n *** DEADLOCK ***\n\n");
2370 print_usage_bug(struct task_struct
*curr
, struct held_lock
*this,
2371 enum lock_usage_bit prev_bit
, enum lock_usage_bit new_bit
)
2373 if (!debug_locks_off_graph_unlock() || debug_locks_silent
)
2377 pr_warn("================================\n");
2378 pr_warn("WARNING: inconsistent lock state\n");
2379 print_kernel_ident();
2380 pr_warn("--------------------------------\n");
2382 printk("inconsistent {%s} -> {%s} usage.\n",
2383 usage_str
[prev_bit
], usage_str
[new_bit
]);
2385 printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] takes:\n",
2386 curr
->comm
, task_pid_nr(curr
),
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
));
2393 printk("{%s} state was registered at:\n", usage_str
[prev_bit
]);
2394 print_stack_trace(hlock_class(this)->usage_traces
+ prev_bit
, 1);
2396 print_irqtrace_events(curr
);
2397 printk("\nother info that might help us debug this:\n");
2398 print_usage_bug_scenario(this);
2400 lockdep_print_held_locks(curr
);
2402 printk("\nstack backtrace:\n");
2409 * Print out an error if an invalid bit is set:
2412 valid_state(struct task_struct
*curr
, struct held_lock
*this,
2413 enum lock_usage_bit new_bit
, enum lock_usage_bit bad_bit
)
2415 if (unlikely(hlock_class(this)->usage_mask
& (1 << bad_bit
)))
2416 return print_usage_bug(curr
, this, bad_bit
, new_bit
);
2420 static int mark_lock(struct task_struct
*curr
, struct held_lock
*this,
2421 enum lock_usage_bit new_bit
);
2423 #if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING)
2426 * print irq inversion bug:
2429 print_irq_inversion_bug(struct task_struct
*curr
,
2430 struct lock_list
*root
, struct lock_list
*other
,
2431 struct held_lock
*this, int forwards
,
2432 const char *irqclass
)
2434 struct lock_list
*entry
= other
;
2435 struct lock_list
*middle
= NULL
;
2438 if (!debug_locks_off_graph_unlock() || debug_locks_silent
)
2442 pr_warn("========================================================\n");
2443 pr_warn("WARNING: possible irq lock inversion dependency detected\n");
2444 print_kernel_ident();
2445 pr_warn("--------------------------------------------------------\n");
2446 printk("%s/%d just changed the state of lock:\n",
2447 curr
->comm
, task_pid_nr(curr
));
2450 printk("but this lock took another, %s-unsafe lock in the past:\n", irqclass
);
2452 printk("but this lock was taken by another, %s-safe lock in the past:\n", irqclass
);
2453 print_lock_name(other
->class);
2454 printk("\n\nand interrupts could create inverse lock ordering between them.\n\n");
2456 printk("\nother info that might help us debug this:\n");
2458 /* Find a middle lock (if one exists) */
2459 depth
= get_lock_depth(other
);
2461 if (depth
== 0 && (entry
!= root
)) {
2462 printk("lockdep:%s bad path found in chain graph\n", __func__
);
2466 entry
= get_lock_parent(entry
);
2468 } while (entry
&& entry
!= root
&& (depth
>= 0));
2470 print_irq_lock_scenario(root
, other
,
2471 middle
? middle
->class : root
->class, other
->class);
2473 print_irq_lock_scenario(other
, root
,
2474 middle
? middle
->class : other
->class, root
->class);
2476 lockdep_print_held_locks(curr
);
2478 printk("\nthe shortest dependencies between 2nd lock and 1st lock:\n");
2479 if (!save_trace(&root
->trace
))
2481 print_shortest_lock_dependencies(other
, root
);
2483 printk("\nstack backtrace:\n");
2490 * Prove that in the forwards-direction subgraph starting at <this>
2491 * there is no lock matching <mask>:
2494 check_usage_forwards(struct task_struct
*curr
, struct held_lock
*this,
2495 enum lock_usage_bit bit
, const char *irqclass
)
2498 struct lock_list root
;
2499 struct lock_list
*uninitialized_var(target_entry
);
2502 root
.class = hlock_class(this);
2503 ret
= find_usage_forwards(&root
, bit
, &target_entry
);
2505 return print_bfs_bug(ret
);
2509 return print_irq_inversion_bug(curr
, &root
, target_entry
,
2514 * Prove that in the backwards-direction subgraph starting at <this>
2515 * there is no lock matching <mask>:
2518 check_usage_backwards(struct task_struct
*curr
, struct held_lock
*this,
2519 enum lock_usage_bit bit
, const char *irqclass
)
2522 struct lock_list root
;
2523 struct lock_list
*uninitialized_var(target_entry
);
2526 root
.class = hlock_class(this);
2527 ret
= find_usage_backwards(&root
, bit
, &target_entry
);
2529 return print_bfs_bug(ret
);
2533 return print_irq_inversion_bug(curr
, &root
, target_entry
,
2537 void print_irqtrace_events(struct task_struct
*curr
)
2539 printk("irq event stamp: %u\n", curr
->irq_events
);
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
);
2554 static int HARDIRQ_verbose(struct lock_class
*class)
2557 return class_filter(class);
2562 static int SOFTIRQ_verbose(struct lock_class
*class)
2565 return class_filter(class);
2570 static int RECLAIM_FS_verbose(struct lock_class
*class)
2573 return class_filter(class);
2578 #define STRICT_READ_CHECKS 1
2580 static int (*state_verbose_f
[])(struct lock_class
*class) = {
2581 #define LOCKDEP_STATE(__STATE) \
2583 #include "lockdep_states.h"
2584 #undef LOCKDEP_STATE
2587 static inline int state_verbose(enum lock_usage_bit bit
,
2588 struct lock_class
*class)
2590 return state_verbose_f
[bit
>> 2](class);
2593 typedef int (*check_usage_f
)(struct task_struct
*, struct held_lock
*,
2594 enum lock_usage_bit bit
, const char *name
);
2597 mark_lock_irq(struct task_struct
*curr
, struct held_lock
*this,
2598 enum lock_usage_bit new_bit
)
2600 int excl_bit
= exclusive_bit(new_bit
);
2601 int read
= new_bit
& 1;
2602 int dir
= new_bit
& 2;
2605 * mark USED_IN has to look forwards -- to ensure no dependency
2606 * has ENABLED state, which would allow recursion deadlocks.
2608 * mark ENABLED has to look backwards -- to ensure no dependee
2609 * has USED_IN state, which, again, would allow recursion deadlocks.
2611 check_usage_f usage
= dir
?
2612 check_usage_backwards
: check_usage_forwards
;
2615 * Validate that this particular lock does not have conflicting
2618 if (!valid_state(curr
, this, new_bit
, excl_bit
))
2622 * Validate that the lock dependencies don't have conflicting usage
2625 if ((!read
|| !dir
|| STRICT_READ_CHECKS
) &&
2626 !usage(curr
, this, excl_bit
, state_name(new_bit
& ~1)))
2630 * Check for read in write conflicts
2633 if (!valid_state(curr
, this, new_bit
, excl_bit
+ 1))
2636 if (STRICT_READ_CHECKS
&&
2637 !usage(curr
, this, excl_bit
+ 1,
2638 state_name(new_bit
+ 1)))
2642 if (state_verbose(new_bit
, hlock_class(this)))
2649 #define LOCKDEP_STATE(__STATE) __STATE,
2650 #include "lockdep_states.h"
2651 #undef LOCKDEP_STATE
2655 * Mark all held locks with a usage bit:
2658 mark_held_locks(struct task_struct
*curr
, enum mark_type mark
)
2660 enum lock_usage_bit usage_bit
;
2661 struct held_lock
*hlock
;
2664 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
2665 hlock
= curr
->held_locks
+ i
;
2667 usage_bit
= 2 + (mark
<< 2); /* ENABLED */
2669 usage_bit
+= 1; /* READ */
2671 BUG_ON(usage_bit
>= LOCK_USAGE_STATES
);
2676 if (!mark_lock(curr
, hlock
, usage_bit
))
2684 * Hardirqs will be enabled:
2686 static void __trace_hardirqs_on_caller(unsigned long ip
)
2688 struct task_struct
*curr
= current
;
2690 /* we'll do an OFF -> ON transition: */
2691 curr
->hardirqs_enabled
= 1;
2694 * We are going to turn hardirqs on, so set the
2695 * usage bit for all held locks:
2697 if (!mark_held_locks(curr
, HARDIRQ
))
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)
2704 if (curr
->softirqs_enabled
)
2705 if (!mark_held_locks(curr
, SOFTIRQ
))
2708 curr
->hardirq_enable_ip
= ip
;
2709 curr
->hardirq_enable_event
= ++curr
->irq_events
;
2710 debug_atomic_inc(hardirqs_on_events
);
2713 __visible
void trace_hardirqs_on_caller(unsigned long ip
)
2715 time_hardirqs_on(CALLER_ADDR0
, ip
);
2717 if (unlikely(!debug_locks
|| current
->lockdep_recursion
))
2720 if (unlikely(current
->hardirqs_enabled
)) {
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.
2726 __debug_atomic_inc(redundant_hardirqs_on
);
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.
2735 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2739 * See the fine text that goes along with this variable definition.
2741 if (DEBUG_LOCKS_WARN_ON(unlikely(early_boot_irqs_disabled
)))
2745 * Can't allow enabling interrupts while in an interrupt handler,
2746 * that's general bad form and such. Recursion, limited stack etc..
2748 if (DEBUG_LOCKS_WARN_ON(current
->hardirq_context
))
2751 current
->lockdep_recursion
= 1;
2752 __trace_hardirqs_on_caller(ip
);
2753 current
->lockdep_recursion
= 0;
2755 EXPORT_SYMBOL(trace_hardirqs_on_caller
);
2757 void trace_hardirqs_on(void)
2759 trace_hardirqs_on_caller(CALLER_ADDR0
);
2761 EXPORT_SYMBOL(trace_hardirqs_on
);
2764 * Hardirqs were disabled:
2766 __visible
void trace_hardirqs_off_caller(unsigned long ip
)
2768 struct task_struct
*curr
= current
;
2770 time_hardirqs_off(CALLER_ADDR0
, ip
);
2772 if (unlikely(!debug_locks
|| current
->lockdep_recursion
))
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.
2779 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2782 if (curr
->hardirqs_enabled
) {
2784 * We have done an ON -> OFF transition:
2786 curr
->hardirqs_enabled
= 0;
2787 curr
->hardirq_disable_ip
= ip
;
2788 curr
->hardirq_disable_event
= ++curr
->irq_events
;
2789 debug_atomic_inc(hardirqs_off_events
);
2791 debug_atomic_inc(redundant_hardirqs_off
);
2793 EXPORT_SYMBOL(trace_hardirqs_off_caller
);
2795 void trace_hardirqs_off(void)
2797 trace_hardirqs_off_caller(CALLER_ADDR0
);
2799 EXPORT_SYMBOL(trace_hardirqs_off
);
2802 * Softirqs will be enabled:
2804 void trace_softirqs_on(unsigned long ip
)
2806 struct task_struct
*curr
= current
;
2808 if (unlikely(!debug_locks
|| current
->lockdep_recursion
))
2812 * We fancy IRQs being disabled here, see softirq.c, avoids
2813 * funny state and nesting things.
2815 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2818 if (curr
->softirqs_enabled
) {
2819 debug_atomic_inc(redundant_softirqs_on
);
2823 current
->lockdep_recursion
= 1;
2825 * We'll do an OFF -> ON transition:
2827 curr
->softirqs_enabled
= 1;
2828 curr
->softirq_enable_ip
= ip
;
2829 curr
->softirq_enable_event
= ++curr
->irq_events
;
2830 debug_atomic_inc(softirqs_on_events
);
2832 * We are going to turn softirqs on, so set the
2833 * usage bit for all held locks, if hardirqs are
2836 if (curr
->hardirqs_enabled
)
2837 mark_held_locks(curr
, SOFTIRQ
);
2838 current
->lockdep_recursion
= 0;
2842 * Softirqs were disabled:
2844 void trace_softirqs_off(unsigned long ip
)
2846 struct task_struct
*curr
= current
;
2848 if (unlikely(!debug_locks
|| current
->lockdep_recursion
))
2852 * We fancy IRQs being disabled here, see softirq.c
2854 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
2857 if (curr
->softirqs_enabled
) {
2859 * We have done an ON -> OFF transition:
2861 curr
->softirqs_enabled
= 0;
2862 curr
->softirq_disable_ip
= ip
;
2863 curr
->softirq_disable_event
= ++curr
->irq_events
;
2864 debug_atomic_inc(softirqs_off_events
);
2866 * Whoops, we wanted softirqs off, so why aren't they?
2868 DEBUG_LOCKS_WARN_ON(!softirq_count());
2870 debug_atomic_inc(redundant_softirqs_off
);
2873 static void __lockdep_trace_alloc(gfp_t gfp_mask
, unsigned long flags
)
2875 struct task_struct
*curr
= current
;
2877 if (unlikely(!debug_locks
))
2880 gfp_mask
= current_gfp_context(gfp_mask
);
2882 /* no reclaim without waiting on it */
2883 if (!(gfp_mask
& __GFP_DIRECT_RECLAIM
))
2886 /* this guy won't enter reclaim */
2887 if ((curr
->flags
& PF_MEMALLOC
) && !(gfp_mask
& __GFP_NOMEMALLOC
))
2890 /* We're only interested __GFP_FS allocations for now */
2891 if (!(gfp_mask
& __GFP_FS
) || (curr
->flags
& PF_MEMALLOC_NOFS
))
2895 * Oi! Can't be having __GFP_FS allocations with IRQs disabled.
2897 if (DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags
)))
2900 /* Disable lockdep if explicitly requested */
2901 if (gfp_mask
& __GFP_NOLOCKDEP
)
2904 mark_held_locks(curr
, RECLAIM_FS
);
2907 static void check_flags(unsigned long flags
);
2909 void lockdep_trace_alloc(gfp_t gfp_mask
)
2911 unsigned long flags
;
2913 if (unlikely(current
->lockdep_recursion
))
2916 raw_local_irq_save(flags
);
2918 current
->lockdep_recursion
= 1;
2919 __lockdep_trace_alloc(gfp_mask
, flags
);
2920 current
->lockdep_recursion
= 0;
2921 raw_local_irq_restore(flags
);
2924 static int mark_irqflags(struct task_struct
*curr
, struct held_lock
*hlock
)
2927 * If non-trylock use in a hardirq or softirq context, then
2928 * mark the lock as used in these contexts:
2930 if (!hlock
->trylock
) {
2932 if (curr
->hardirq_context
)
2933 if (!mark_lock(curr
, hlock
,
2934 LOCK_USED_IN_HARDIRQ_READ
))
2936 if (curr
->softirq_context
)
2937 if (!mark_lock(curr
, hlock
,
2938 LOCK_USED_IN_SOFTIRQ_READ
))
2941 if (curr
->hardirq_context
)
2942 if (!mark_lock(curr
, hlock
, LOCK_USED_IN_HARDIRQ
))
2944 if (curr
->softirq_context
)
2945 if (!mark_lock(curr
, hlock
, LOCK_USED_IN_SOFTIRQ
))
2949 if (!hlock
->hardirqs_off
) {
2951 if (!mark_lock(curr
, hlock
,
2952 LOCK_ENABLED_HARDIRQ_READ
))
2954 if (curr
->softirqs_enabled
)
2955 if (!mark_lock(curr
, hlock
,
2956 LOCK_ENABLED_SOFTIRQ_READ
))
2959 if (!mark_lock(curr
, hlock
,
2960 LOCK_ENABLED_HARDIRQ
))
2962 if (curr
->softirqs_enabled
)
2963 if (!mark_lock(curr
, hlock
,
2964 LOCK_ENABLED_SOFTIRQ
))
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
2975 if (!hlock
->trylock
&& (curr
->lockdep_reclaim_gfp
& __GFP_FS
)) {
2977 if (!mark_lock(curr
, hlock
, LOCK_USED_IN_RECLAIM_FS_READ
))
2980 if (!mark_lock(curr
, hlock
, LOCK_USED_IN_RECLAIM_FS
))
2988 static inline unsigned int task_irq_context(struct task_struct
*task
)
2990 return 2 * !!task
->hardirq_context
+ !!task
->softirq_context
;
2993 static int separate_irq_context(struct task_struct
*curr
,
2994 struct held_lock
*hlock
)
2996 unsigned int depth
= curr
->lockdep_depth
;
2999 * Keep track of points where we cross into an interrupt context:
3002 struct held_lock
*prev_hlock
;
3004 prev_hlock
= curr
->held_locks
+ depth
-1;
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'):
3010 if (prev_hlock
->irq_context
!= hlock
->irq_context
)
3016 #else /* defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) */
3019 int mark_lock_irq(struct task_struct
*curr
, struct held_lock
*this,
3020 enum lock_usage_bit new_bit
)
3022 WARN_ON(1); /* Impossible innit? when we don't have TRACE_IRQFLAG */
3026 static inline int mark_irqflags(struct task_struct
*curr
,
3027 struct held_lock
*hlock
)
3032 static inline unsigned int task_irq_context(struct task_struct
*task
)
3037 static inline int separate_irq_context(struct task_struct
*curr
,
3038 struct held_lock
*hlock
)
3043 void lockdep_trace_alloc(gfp_t gfp_mask
)
3047 #endif /* defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) */
3050 * Mark a lock with a usage bit, and validate the state transition:
3052 static int mark_lock(struct task_struct
*curr
, struct held_lock
*this,
3053 enum lock_usage_bit new_bit
)
3055 unsigned int new_mask
= 1 << new_bit
, ret
= 1;
3058 * If already set then do not dirty the cacheline,
3059 * nor do any checks:
3061 if (likely(hlock_class(this)->usage_mask
& new_mask
))
3067 * Make sure we didn't race:
3069 if (unlikely(hlock_class(this)->usage_mask
& new_mask
)) {
3074 hlock_class(this)->usage_mask
|= new_mask
;
3076 if (!save_trace(hlock_class(this)->usage_traces
+ new_bit
))
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
3087 ret
= mark_lock_irq(curr
, this, new_bit
);
3092 debug_atomic_dec(nr_unused_locks
);
3095 if (!debug_locks_off_graph_unlock())
3104 * We must printk outside of the graph_lock:
3107 printk("\nmarked lock as {%s}:\n", usage_str
[new_bit
]);
3109 print_irqtrace_events(curr
);
3117 * Initialize a lock instance's lock-class mapping info:
3119 void lockdep_init_map(struct lockdep_map
*lock
, const char *name
,
3120 struct lock_class_key
*key
, int subclass
)
3124 kmemcheck_mark_initialized(lock
, sizeof(*lock
));
3126 for (i
= 0; i
< NR_LOCKDEP_CACHING_CLASSES
; i
++)
3127 lock
->class_cache
[i
] = NULL
;
3129 #ifdef CONFIG_LOCK_STAT
3130 lock
->cpu
= raw_smp_processor_id();
3134 * Can't be having no nameless bastards around this place!
3136 if (DEBUG_LOCKS_WARN_ON(!name
)) {
3137 lock
->name
= "NULL";
3144 * No key, no joy, we need to hash something.
3146 if (DEBUG_LOCKS_WARN_ON(!key
))
3149 * Sanity check, the lock-class key must be persistent:
3151 if (!static_obj(key
)) {
3152 printk("BUG: key %p not in .data!\n", key
);
3154 * What it says above ^^^^^, I suggest you read it.
3156 DEBUG_LOCKS_WARN_ON(1);
3161 if (unlikely(!debug_locks
))
3165 unsigned long flags
;
3167 if (DEBUG_LOCKS_WARN_ON(current
->lockdep_recursion
))
3170 raw_local_irq_save(flags
);
3171 current
->lockdep_recursion
= 1;
3172 register_lock_class(lock
, subclass
, 1);
3173 current
->lockdep_recursion
= 0;
3174 raw_local_irq_restore(flags
);
3177 EXPORT_SYMBOL_GPL(lockdep_init_map
);
3179 struct lock_class_key __lockdep_no_validate__
;
3180 EXPORT_SYMBOL_GPL(__lockdep_no_validate__
);
3183 print_lock_nested_lock_not_held(struct task_struct
*curr
,
3184 struct held_lock
*hlock
,
3187 if (!debug_locks_off())
3189 if (debug_locks_silent
)
3193 pr_warn("==================================\n");
3194 pr_warn("WARNING: Nested lock was not taken\n");
3195 print_kernel_ident();
3196 pr_warn("----------------------------------\n");
3198 printk("%s/%d is trying to lock:\n", curr
->comm
, task_pid_nr(curr
));
3201 printk("\nbut this task is not holding:\n");
3202 printk("%s\n", hlock
->nest_lock
->name
);
3204 printk("\nstack backtrace:\n");
3207 printk("\nother info that might help us debug this:\n");
3208 lockdep_print_held_locks(curr
);
3210 printk("\nstack backtrace:\n");
3216 static int __lock_is_held(struct lockdep_map
*lock
, int read
);
3219 * This gets called for every mutex_lock*()/spin_lock*() operation.
3220 * We maintain the dependency maps and validate the locking attempt:
3222 static int __lock_acquire(struct lockdep_map
*lock
, unsigned int subclass
,
3223 int trylock
, int read
, int check
, int hardirqs_off
,
3224 struct lockdep_map
*nest_lock
, unsigned long ip
,
3225 int references
, int pin_count
)
3227 struct task_struct
*curr
= current
;
3228 struct lock_class
*class = NULL
;
3229 struct held_lock
*hlock
;
3235 if (unlikely(!debug_locks
))
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?
3243 if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
3246 if (!prove_locking
|| lock
->key
== &__lockdep_no_validate__
)
3249 if (subclass
< NR_LOCKDEP_CACHING_CLASSES
)
3250 class = lock
->class_cache
[subclass
];
3254 if (unlikely(!class)) {
3255 class = register_lock_class(lock
, subclass
, 0);
3259 atomic_inc((atomic_t
*)&class->ops
);
3260 if (very_verbose(class)) {
3261 printk("\nacquire class [%p] %s", class->key
, class->name
);
3262 if (class->name_version
> 1)
3263 printk(KERN_CONT
"#%d", class->name_version
);
3264 printk(KERN_CONT
"\n");
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)
3273 depth
= curr
->lockdep_depth
;
3275 * Ran out of static storage for our per-task lock stack again have we?
3277 if (DEBUG_LOCKS_WARN_ON(depth
>= MAX_LOCK_DEPTH
))
3280 class_idx
= class - lock_classes
+ 1;
3283 hlock
= curr
->held_locks
+ depth
- 1;
3284 if (hlock
->class_idx
== class_idx
&& nest_lock
) {
3285 if (hlock
->references
) {
3287 * Check: unsigned int references:12, overflow.
3289 if (DEBUG_LOCKS_WARN_ON(hlock
->references
== (1 << 12)-1))
3292 hlock
->references
++;
3294 hlock
->references
= 2;
3301 hlock
= curr
->held_locks
+ depth
;
3303 * Plain impossible, we just registered it and checked it weren't no
3304 * NULL like.. I bet this mushroom I ate was good!
3306 if (DEBUG_LOCKS_WARN_ON(!class))
3308 hlock
->class_idx
= class_idx
;
3309 hlock
->acquire_ip
= ip
;
3310 hlock
->instance
= lock
;
3311 hlock
->nest_lock
= nest_lock
;
3312 hlock
->irq_context
= task_irq_context(curr
);
3313 hlock
->trylock
= trylock
;
3315 hlock
->check
= check
;
3316 hlock
->hardirqs_off
= !!hardirqs_off
;
3317 hlock
->references
= references
;
3318 #ifdef CONFIG_LOCK_STAT
3319 hlock
->waittime_stamp
= 0;
3320 hlock
->holdtime_stamp
= lockstat_clock();
3322 hlock
->pin_count
= pin_count
;
3324 if (check
&& !mark_irqflags(curr
, hlock
))
3327 /* mark it as used: */
3328 if (!mark_lock(curr
, hlock
, LOCK_USED
))
3332 * Calculate the chain hash: it's the combined hash of all the
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
3338 * The 'key ID' is what is the most compact key value to drive
3339 * the hash, not class->key.
3342 * Whoops, we did it again.. ran straight out of our static allocation.
3344 if (DEBUG_LOCKS_WARN_ON(class_idx
> MAX_LOCKDEP_KEYS
))
3347 chain_key
= curr
->curr_chain_key
;
3350 * How can we have a chain hash when we ain't got no keys?!
3352 if (DEBUG_LOCKS_WARN_ON(chain_key
!= 0))
3357 hlock
->prev_chain_key
= chain_key
;
3358 if (separate_irq_context(curr
, hlock
)) {
3362 chain_key
= iterate_chain_key(chain_key
, class_idx
);
3364 if (nest_lock
&& !__lock_is_held(nest_lock
, -1))
3365 return print_lock_nested_lock_not_held(curr
, hlock
, ip
);
3367 if (!validate_chain(curr
, lock
, hlock
, chain_head
, chain_key
))
3370 curr
->curr_chain_key
= chain_key
;
3371 curr
->lockdep_depth
++;
3372 check_chain_key(curr
);
3373 #ifdef CONFIG_DEBUG_LOCKDEP
3374 if (unlikely(!debug_locks
))
3377 if (unlikely(curr
->lockdep_depth
>= MAX_LOCK_DEPTH
)) {
3379 print_lockdep_off("BUG: MAX_LOCK_DEPTH too low!");
3380 printk(KERN_DEBUG
"depth: %i max: %lu!\n",
3381 curr
->lockdep_depth
, MAX_LOCK_DEPTH
);
3383 lockdep_print_held_locks(current
);
3384 debug_show_all_locks();
3390 if (unlikely(curr
->lockdep_depth
> max_lockdep_depth
))
3391 max_lockdep_depth
= curr
->lockdep_depth
;
3397 print_unlock_imbalance_bug(struct task_struct
*curr
, struct lockdep_map
*lock
,
3400 if (!debug_locks_off())
3402 if (debug_locks_silent
)
3406 pr_warn("=====================================\n");
3407 pr_warn("WARNING: bad unlock balance detected!\n");
3408 print_kernel_ident();
3409 pr_warn("-------------------------------------\n");
3410 printk("%s/%d is trying to release lock (",
3411 curr
->comm
, task_pid_nr(curr
));
3412 print_lockdep_cache(lock
);
3413 printk(KERN_CONT
") at:\n");
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
);
3419 printk("\nstack backtrace:\n");
3425 static int match_held_lock(struct held_lock
*hlock
, struct lockdep_map
*lock
)
3427 if (hlock
->instance
== lock
)
3430 if (hlock
->references
) {
3431 struct lock_class
*class = lock
->class_cache
[0];
3434 class = look_up_lock_class(lock
, 0);
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.
3442 if (IS_ERR_OR_NULL(class))
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.
3450 if (DEBUG_LOCKS_WARN_ON(!hlock
->nest_lock
))
3453 if (hlock
->class_idx
== class - lock_classes
+ 1)
3460 /* @depth must not be zero */
3461 static struct held_lock
*find_held_lock(struct task_struct
*curr
,
3462 struct lockdep_map
*lock
,
3463 unsigned int depth
, int *idx
)
3465 struct held_lock
*ret
, *hlock
, *prev_hlock
;
3469 hlock
= curr
->held_locks
+ i
;
3471 if (match_held_lock(hlock
, lock
))
3475 for (i
--, prev_hlock
= hlock
--;
3477 i
--, prev_hlock
= hlock
--) {
3479 * We must not cross into another context:
3481 if (prev_hlock
->irq_context
!= hlock
->irq_context
) {
3485 if (match_held_lock(hlock
, lock
)) {
3496 static int reacquire_held_locks(struct task_struct
*curr
, unsigned int depth
,
3499 struct held_lock
*hlock
;
3501 for (hlock
= curr
->held_locks
+ idx
; idx
< depth
; idx
++, hlock
++) {
3502 if (!__lock_acquire(hlock
->instance
,
3503 hlock_class(hlock
)->subclass
,
3505 hlock
->read
, hlock
->check
,
3506 hlock
->hardirqs_off
,
3507 hlock
->nest_lock
, hlock
->acquire_ip
,
3508 hlock
->references
, hlock
->pin_count
))
3515 __lock_set_class(struct lockdep_map
*lock
, const char *name
,
3516 struct lock_class_key
*key
, unsigned int subclass
,
3519 struct task_struct
*curr
= current
;
3520 struct held_lock
*hlock
;
3521 struct lock_class
*class;
3525 depth
= curr
->lockdep_depth
;
3527 * This function is about (re)setting the class of a held lock,
3528 * yet we're not actually holding any locks. Naughty user!
3530 if (DEBUG_LOCKS_WARN_ON(!depth
))
3533 hlock
= find_held_lock(curr
, lock
, depth
, &i
);
3535 return print_unlock_imbalance_bug(curr
, lock
, ip
);
3537 lockdep_init_map(lock
, name
, key
, 0);
3538 class = register_lock_class(lock
, subclass
, 0);
3539 hlock
->class_idx
= class - lock_classes
+ 1;
3541 curr
->lockdep_depth
= i
;
3542 curr
->curr_chain_key
= hlock
->prev_chain_key
;
3544 if (reacquire_held_locks(curr
, depth
, i
))
3548 * I took it apart and put it back together again, except now I have
3549 * these 'spare' parts.. where shall I put them.
3551 if (DEBUG_LOCKS_WARN_ON(curr
->lockdep_depth
!= depth
))
3556 static int __lock_downgrade(struct lockdep_map
*lock
, unsigned long ip
)
3558 struct task_struct
*curr
= current
;
3559 struct held_lock
*hlock
;
3563 depth
= curr
->lockdep_depth
;
3565 * This function is about (re)setting the class of a held lock,
3566 * yet we're not actually holding any locks. Naughty user!
3568 if (DEBUG_LOCKS_WARN_ON(!depth
))
3571 hlock
= find_held_lock(curr
, lock
, depth
, &i
);
3573 return print_unlock_imbalance_bug(curr
, lock
, ip
);
3575 curr
->lockdep_depth
= i
;
3576 curr
->curr_chain_key
= hlock
->prev_chain_key
;
3578 WARN(hlock
->read
, "downgrading a read lock");
3580 hlock
->acquire_ip
= ip
;
3582 if (reacquire_held_locks(curr
, depth
, i
))
3586 * I took it apart and put it back together again, except now I have
3587 * these 'spare' parts.. where shall I put them.
3589 if (DEBUG_LOCKS_WARN_ON(curr
->lockdep_depth
!= depth
))
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()).
3599 * @nested is an hysterical artifact, needs a tree wide cleanup.
3602 __lock_release(struct lockdep_map
*lock
, int nested
, unsigned long ip
)
3604 struct task_struct
*curr
= current
;
3605 struct held_lock
*hlock
;
3609 if (unlikely(!debug_locks
))
3612 depth
= curr
->lockdep_depth
;
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?
3617 if (DEBUG_LOCKS_WARN_ON(depth
<= 0))
3618 return print_unlock_imbalance_bug(curr
, lock
, ip
);
3621 * Check whether the lock exists in the current stack
3624 hlock
= find_held_lock(curr
, lock
, depth
, &i
);
3626 return print_unlock_imbalance_bug(curr
, lock
, ip
);
3628 if (hlock
->instance
== lock
)
3629 lock_release_holdtime(hlock
);
3631 WARN(hlock
->pin_count
, "releasing a pinned lock\n");
3633 if (hlock
->references
) {
3634 hlock
->references
--;
3635 if (hlock
->references
) {
3637 * We had, and after removing one, still have
3638 * references, the current lock stack is still
3639 * valid. We're done!
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:
3651 curr
->lockdep_depth
= i
;
3652 curr
->curr_chain_key
= hlock
->prev_chain_key
;
3654 if (reacquire_held_locks(curr
, depth
, i
+ 1))
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...
3661 if (DEBUG_LOCKS_WARN_ON(curr
->lockdep_depth
!= depth
- 1))
3667 static int __lock_is_held(struct lockdep_map
*lock
, int read
)
3669 struct task_struct
*curr
= current
;
3672 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
3673 struct held_lock
*hlock
= curr
->held_locks
+ i
;
3675 if (match_held_lock(hlock
, lock
)) {
3676 if (read
== -1 || hlock
->read
== read
)
3686 static struct pin_cookie
__lock_pin_lock(struct lockdep_map
*lock
)
3688 struct pin_cookie cookie
= NIL_COOKIE
;
3689 struct task_struct
*curr
= current
;
3692 if (unlikely(!debug_locks
))
3695 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
3696 struct held_lock
*hlock
= curr
->held_locks
+ i
;
3698 if (match_held_lock(hlock
, lock
)) {
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.
3704 cookie
.val
= 1 + (prandom_u32() >> 16);
3705 hlock
->pin_count
+= cookie
.val
;
3710 WARN(1, "pinning an unheld lock\n");
3714 static void __lock_repin_lock(struct lockdep_map
*lock
, struct pin_cookie cookie
)
3716 struct task_struct
*curr
= current
;
3719 if (unlikely(!debug_locks
))
3722 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
3723 struct held_lock
*hlock
= curr
->held_locks
+ i
;
3725 if (match_held_lock(hlock
, lock
)) {
3726 hlock
->pin_count
+= cookie
.val
;
3731 WARN(1, "pinning an unheld lock\n");
3734 static void __lock_unpin_lock(struct lockdep_map
*lock
, struct pin_cookie cookie
)
3736 struct task_struct
*curr
= current
;
3739 if (unlikely(!debug_locks
))
3742 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
3743 struct held_lock
*hlock
= curr
->held_locks
+ i
;
3745 if (match_held_lock(hlock
, lock
)) {
3746 if (WARN(!hlock
->pin_count
, "unpinning an unpinned lock\n"))
3749 hlock
->pin_count
-= cookie
.val
;
3751 if (WARN((int)hlock
->pin_count
< 0, "pin count corrupted\n"))
3752 hlock
->pin_count
= 0;
3758 WARN(1, "unpinning an unheld lock\n");
3762 * Check whether we follow the irq-flags state precisely:
3764 static void check_flags(unsigned long flags
)
3766 #if defined(CONFIG_PROVE_LOCKING) && defined(CONFIG_DEBUG_LOCKDEP) && \
3767 defined(CONFIG_TRACE_IRQFLAGS)
3771 if (irqs_disabled_flags(flags
)) {
3772 if (DEBUG_LOCKS_WARN_ON(current
->hardirqs_enabled
)) {
3773 printk("possible reason: unannotated irqs-off.\n");
3776 if (DEBUG_LOCKS_WARN_ON(!current
->hardirqs_enabled
)) {
3777 printk("possible reason: unannotated irqs-on.\n");
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:
3786 if (!hardirq_count()) {
3787 if (softirq_count()) {
3788 /* like the above, but with softirqs */
3789 DEBUG_LOCKS_WARN_ON(current
->softirqs_enabled
);
3791 /* lick the above, does it taste good? */
3792 DEBUG_LOCKS_WARN_ON(!current
->softirqs_enabled
);
3797 print_irqtrace_events(current
);
3801 void lock_set_class(struct lockdep_map
*lock
, const char *name
,
3802 struct lock_class_key
*key
, unsigned int subclass
,
3805 unsigned long flags
;
3807 if (unlikely(current
->lockdep_recursion
))
3810 raw_local_irq_save(flags
);
3811 current
->lockdep_recursion
= 1;
3813 if (__lock_set_class(lock
, name
, key
, subclass
, ip
))
3814 check_chain_key(current
);
3815 current
->lockdep_recursion
= 0;
3816 raw_local_irq_restore(flags
);
3818 EXPORT_SYMBOL_GPL(lock_set_class
);
3820 void lock_downgrade(struct lockdep_map
*lock
, unsigned long ip
)
3822 unsigned long flags
;
3824 if (unlikely(current
->lockdep_recursion
))
3827 raw_local_irq_save(flags
);
3828 current
->lockdep_recursion
= 1;
3830 if (__lock_downgrade(lock
, ip
))
3831 check_chain_key(current
);
3832 current
->lockdep_recursion
= 0;
3833 raw_local_irq_restore(flags
);
3835 EXPORT_SYMBOL_GPL(lock_downgrade
);
3838 * We are not always called with irqs disabled - do that here,
3839 * and also avoid lockdep recursion:
3841 void lock_acquire(struct lockdep_map
*lock
, unsigned int subclass
,
3842 int trylock
, int read
, int check
,
3843 struct lockdep_map
*nest_lock
, unsigned long ip
)
3845 unsigned long flags
;
3847 if (unlikely(current
->lockdep_recursion
))
3850 raw_local_irq_save(flags
);
3853 current
->lockdep_recursion
= 1;
3854 trace_lock_acquire(lock
, subclass
, trylock
, read
, check
, nest_lock
, ip
);
3855 __lock_acquire(lock
, subclass
, trylock
, read
, check
,
3856 irqs_disabled_flags(flags
), nest_lock
, ip
, 0, 0);
3857 current
->lockdep_recursion
= 0;
3858 raw_local_irq_restore(flags
);
3860 EXPORT_SYMBOL_GPL(lock_acquire
);
3862 void lock_release(struct lockdep_map
*lock
, int nested
,
3865 unsigned long flags
;
3867 if (unlikely(current
->lockdep_recursion
))
3870 raw_local_irq_save(flags
);
3872 current
->lockdep_recursion
= 1;
3873 trace_lock_release(lock
, ip
);
3874 if (__lock_release(lock
, nested
, ip
))
3875 check_chain_key(current
);
3876 current
->lockdep_recursion
= 0;
3877 raw_local_irq_restore(flags
);
3879 EXPORT_SYMBOL_GPL(lock_release
);
3881 int lock_is_held_type(struct lockdep_map
*lock
, int read
)
3883 unsigned long flags
;
3886 if (unlikely(current
->lockdep_recursion
))
3887 return 1; /* avoid false negative lockdep_assert_held() */
3889 raw_local_irq_save(flags
);
3892 current
->lockdep_recursion
= 1;
3893 ret
= __lock_is_held(lock
, read
);
3894 current
->lockdep_recursion
= 0;
3895 raw_local_irq_restore(flags
);
3899 EXPORT_SYMBOL_GPL(lock_is_held_type
);
3901 struct pin_cookie
lock_pin_lock(struct lockdep_map
*lock
)
3903 struct pin_cookie cookie
= NIL_COOKIE
;
3904 unsigned long flags
;
3906 if (unlikely(current
->lockdep_recursion
))
3909 raw_local_irq_save(flags
);
3912 current
->lockdep_recursion
= 1;
3913 cookie
= __lock_pin_lock(lock
);
3914 current
->lockdep_recursion
= 0;
3915 raw_local_irq_restore(flags
);
3919 EXPORT_SYMBOL_GPL(lock_pin_lock
);
3921 void lock_repin_lock(struct lockdep_map
*lock
, struct pin_cookie cookie
)
3923 unsigned long flags
;
3925 if (unlikely(current
->lockdep_recursion
))
3928 raw_local_irq_save(flags
);
3931 current
->lockdep_recursion
= 1;
3932 __lock_repin_lock(lock
, cookie
);
3933 current
->lockdep_recursion
= 0;
3934 raw_local_irq_restore(flags
);
3936 EXPORT_SYMBOL_GPL(lock_repin_lock
);
3938 void lock_unpin_lock(struct lockdep_map
*lock
, struct pin_cookie cookie
)
3940 unsigned long flags
;
3942 if (unlikely(current
->lockdep_recursion
))
3945 raw_local_irq_save(flags
);
3948 current
->lockdep_recursion
= 1;
3949 __lock_unpin_lock(lock
, cookie
);
3950 current
->lockdep_recursion
= 0;
3951 raw_local_irq_restore(flags
);
3953 EXPORT_SYMBOL_GPL(lock_unpin_lock
);
3955 void lockdep_set_current_reclaim_state(gfp_t gfp_mask
)
3957 current
->lockdep_reclaim_gfp
= current_gfp_context(gfp_mask
);
3959 EXPORT_SYMBOL_GPL(lockdep_set_current_reclaim_state
);
3961 void lockdep_clear_current_reclaim_state(void)
3963 current
->lockdep_reclaim_gfp
= 0;
3965 EXPORT_SYMBOL_GPL(lockdep_clear_current_reclaim_state
);
3967 #ifdef CONFIG_LOCK_STAT
3969 print_lock_contention_bug(struct task_struct
*curr
, struct lockdep_map
*lock
,
3972 if (!debug_locks_off())
3974 if (debug_locks_silent
)
3978 pr_warn("=================================\n");
3979 pr_warn("WARNING: bad contention detected!\n");
3980 print_kernel_ident();
3981 pr_warn("---------------------------------\n");
3982 printk("%s/%d is trying to contend lock (",
3983 curr
->comm
, task_pid_nr(curr
));
3984 print_lockdep_cache(lock
);
3985 printk(KERN_CONT
") at:\n");
3987 printk("but there are no locks held!\n");
3988 printk("\nother info that might help us debug this:\n");
3989 lockdep_print_held_locks(curr
);
3991 printk("\nstack backtrace:\n");
3998 __lock_contended(struct lockdep_map
*lock
, unsigned long ip
)
4000 struct task_struct
*curr
= current
;
4001 struct held_lock
*hlock
;
4002 struct lock_class_stats
*stats
;
4004 int i
, contention_point
, contending_point
;
4006 depth
= curr
->lockdep_depth
;
4008 * Whee, we contended on this lock, except it seems we're not
4009 * actually trying to acquire anything much at all..
4011 if (DEBUG_LOCKS_WARN_ON(!depth
))
4014 hlock
= find_held_lock(curr
, lock
, depth
, &i
);
4016 print_lock_contention_bug(curr
, lock
, ip
);
4020 if (hlock
->instance
!= lock
)
4023 hlock
->waittime_stamp
= lockstat_clock();
4025 contention_point
= lock_point(hlock_class(hlock
)->contention_point
, ip
);
4026 contending_point
= lock_point(hlock_class(hlock
)->contending_point
,
4029 stats
= get_lock_stats(hlock_class(hlock
));
4030 if (contention_point
< LOCKSTAT_POINTS
)
4031 stats
->contention_point
[contention_point
]++;
4032 if (contending_point
< LOCKSTAT_POINTS
)
4033 stats
->contending_point
[contending_point
]++;
4034 if (lock
->cpu
!= smp_processor_id())
4035 stats
->bounces
[bounce_contended
+ !!hlock
->read
]++;
4036 put_lock_stats(stats
);
4040 __lock_acquired(struct lockdep_map
*lock
, unsigned long ip
)
4042 struct task_struct
*curr
= current
;
4043 struct held_lock
*hlock
;
4044 struct lock_class_stats
*stats
;
4046 u64 now
, waittime
= 0;
4049 depth
= curr
->lockdep_depth
;
4051 * Yay, we acquired ownership of this lock we didn't try to
4052 * acquire, how the heck did that happen?
4054 if (DEBUG_LOCKS_WARN_ON(!depth
))
4057 hlock
= find_held_lock(curr
, lock
, depth
, &i
);
4059 print_lock_contention_bug(curr
, lock
, _RET_IP_
);
4063 if (hlock
->instance
!= lock
)
4066 cpu
= smp_processor_id();
4067 if (hlock
->waittime_stamp
) {
4068 now
= lockstat_clock();
4069 waittime
= now
- hlock
->waittime_stamp
;
4070 hlock
->holdtime_stamp
= now
;
4073 trace_lock_acquired(lock
, ip
);
4075 stats
= get_lock_stats(hlock_class(hlock
));
4078 lock_time_inc(&stats
->read_waittime
, waittime
);
4080 lock_time_inc(&stats
->write_waittime
, waittime
);
4082 if (lock
->cpu
!= cpu
)
4083 stats
->bounces
[bounce_acquired
+ !!hlock
->read
]++;
4084 put_lock_stats(stats
);
4090 void lock_contended(struct lockdep_map
*lock
, unsigned long ip
)
4092 unsigned long flags
;
4094 if (unlikely(!lock_stat
))
4097 if (unlikely(current
->lockdep_recursion
))
4100 raw_local_irq_save(flags
);
4102 current
->lockdep_recursion
= 1;
4103 trace_lock_contended(lock
, ip
);
4104 __lock_contended(lock
, ip
);
4105 current
->lockdep_recursion
= 0;
4106 raw_local_irq_restore(flags
);
4108 EXPORT_SYMBOL_GPL(lock_contended
);
4110 void lock_acquired(struct lockdep_map
*lock
, unsigned long ip
)
4112 unsigned long flags
;
4114 if (unlikely(!lock_stat
))
4117 if (unlikely(current
->lockdep_recursion
))
4120 raw_local_irq_save(flags
);
4122 current
->lockdep_recursion
= 1;
4123 __lock_acquired(lock
, ip
);
4124 current
->lockdep_recursion
= 0;
4125 raw_local_irq_restore(flags
);
4127 EXPORT_SYMBOL_GPL(lock_acquired
);
4131 * Used by the testsuite, sanitize the validator state
4132 * after a simulated failure:
4135 void lockdep_reset(void)
4137 unsigned long flags
;
4140 raw_local_irq_save(flags
);
4141 current
->curr_chain_key
= 0;
4142 current
->lockdep_depth
= 0;
4143 current
->lockdep_recursion
= 0;
4144 memset(current
->held_locks
, 0, MAX_LOCK_DEPTH
*sizeof(struct held_lock
));
4145 nr_hardirq_chains
= 0;
4146 nr_softirq_chains
= 0;
4147 nr_process_chains
= 0;
4149 for (i
= 0; i
< CHAINHASH_SIZE
; i
++)
4150 INIT_HLIST_HEAD(chainhash_table
+ i
);
4151 raw_local_irq_restore(flags
);
4154 static void zap_class(struct lock_class
*class)
4159 * Remove all dependencies this lock is
4162 for (i
= 0; i
< nr_list_entries
; i
++) {
4163 if (list_entries
[i
].class == class)
4164 list_del_rcu(&list_entries
[i
].entry
);
4167 * Unhash the class and remove it from the all_lock_classes list:
4169 hlist_del_rcu(&class->hash_entry
);
4170 list_del_rcu(&class->lock_entry
);
4172 RCU_INIT_POINTER(class->key
, NULL
);
4173 RCU_INIT_POINTER(class->name
, NULL
);
4176 static inline int within(const void *addr
, void *start
, unsigned long size
)
4178 return addr
>= start
&& addr
< start
+ size
;
4182 * Used in module.c to remove lock classes from memory that is going to be
4183 * freed; and possibly re-used by other modules.
4185 * We will have had one sync_sched() before getting here, so we're guaranteed
4186 * nobody will look up these exact classes -- they're properly dead but still
4189 void lockdep_free_key_range(void *start
, unsigned long size
)
4191 struct lock_class
*class;
4192 struct hlist_head
*head
;
4193 unsigned long flags
;
4197 raw_local_irq_save(flags
);
4198 locked
= graph_lock();
4201 * Unhash all classes that were created by this module:
4203 for (i
= 0; i
< CLASSHASH_SIZE
; i
++) {
4204 head
= classhash_table
+ i
;
4205 hlist_for_each_entry_rcu(class, head
, hash_entry
) {
4206 if (within(class->key
, start
, size
))
4208 else if (within(class->name
, start
, size
))
4215 raw_local_irq_restore(flags
);
4218 * Wait for any possible iterators from look_up_lock_class() to pass
4219 * before continuing to free the memory they refer to.
4221 * sync_sched() is sufficient because the read-side is IRQ disable.
4223 synchronize_sched();
4226 * XXX at this point we could return the resources to the pool;
4227 * instead we leak them. We would need to change to bitmap allocators
4228 * instead of the linear allocators we have now.
4232 void lockdep_reset_lock(struct lockdep_map
*lock
)
4234 struct lock_class
*class;
4235 struct hlist_head
*head
;
4236 unsigned long flags
;
4240 raw_local_irq_save(flags
);
4243 * Remove all classes this lock might have:
4245 for (j
= 0; j
< MAX_LOCKDEP_SUBCLASSES
; j
++) {
4247 * If the class exists we look it up and zap it:
4249 class = look_up_lock_class(lock
, j
);
4250 if (!IS_ERR_OR_NULL(class))
4254 * Debug check: in the end all mapped classes should
4257 locked
= graph_lock();
4258 for (i
= 0; i
< CLASSHASH_SIZE
; i
++) {
4259 head
= classhash_table
+ i
;
4260 hlist_for_each_entry_rcu(class, head
, hash_entry
) {
4263 for (j
= 0; j
< NR_LOCKDEP_CACHING_CLASSES
; j
++)
4264 match
|= class == lock
->class_cache
[j
];
4266 if (unlikely(match
)) {
4267 if (debug_locks_off_graph_unlock()) {
4269 * We all just reset everything, how did it match?
4281 raw_local_irq_restore(flags
);
4284 void __init
lockdep_info(void)
4286 printk("Lock dependency validator: Copyright (c) 2006 Red Hat, Inc., Ingo Molnar\n");
4288 printk("... MAX_LOCKDEP_SUBCLASSES: %lu\n", MAX_LOCKDEP_SUBCLASSES
);
4289 printk("... MAX_LOCK_DEPTH: %lu\n", MAX_LOCK_DEPTH
);
4290 printk("... MAX_LOCKDEP_KEYS: %lu\n", MAX_LOCKDEP_KEYS
);
4291 printk("... CLASSHASH_SIZE: %lu\n", CLASSHASH_SIZE
);
4292 printk("... MAX_LOCKDEP_ENTRIES: %lu\n", MAX_LOCKDEP_ENTRIES
);
4293 printk("... MAX_LOCKDEP_CHAINS: %lu\n", MAX_LOCKDEP_CHAINS
);
4294 printk("... CHAINHASH_SIZE: %lu\n", CHAINHASH_SIZE
);
4296 printk(" memory used by lock dependency info: %lu kB\n",
4297 (sizeof(struct lock_class
) * MAX_LOCKDEP_KEYS
+
4298 sizeof(struct list_head
) * CLASSHASH_SIZE
+
4299 sizeof(struct lock_list
) * MAX_LOCKDEP_ENTRIES
+
4300 sizeof(struct lock_chain
) * MAX_LOCKDEP_CHAINS
+
4301 sizeof(struct list_head
) * CHAINHASH_SIZE
4302 #ifdef CONFIG_PROVE_LOCKING
4303 + sizeof(struct circular_queue
)
4308 printk(" per task-struct memory footprint: %lu bytes\n",
4309 sizeof(struct held_lock
) * MAX_LOCK_DEPTH
);
4313 print_freed_lock_bug(struct task_struct
*curr
, const void *mem_from
,
4314 const void *mem_to
, struct held_lock
*hlock
)
4316 if (!debug_locks_off())
4318 if (debug_locks_silent
)
4322 pr_warn("=========================\n");
4323 pr_warn("WARNING: held lock freed!\n");
4324 print_kernel_ident();
4325 pr_warn("-------------------------\n");
4326 printk("%s/%d is freeing memory %p-%p, with a lock still held there!\n",
4327 curr
->comm
, task_pid_nr(curr
), mem_from
, mem_to
-1);
4329 lockdep_print_held_locks(curr
);
4331 printk("\nstack backtrace:\n");
4335 static inline int not_in_range(const void* mem_from
, unsigned long mem_len
,
4336 const void* lock_from
, unsigned long lock_len
)
4338 return lock_from
+ lock_len
<= mem_from
||
4339 mem_from
+ mem_len
<= lock_from
;
4343 * Called when kernel memory is freed (or unmapped), or if a lock
4344 * is destroyed or reinitialized - this code checks whether there is
4345 * any held lock in the memory range of <from> to <to>:
4347 void debug_check_no_locks_freed(const void *mem_from
, unsigned long mem_len
)
4349 struct task_struct
*curr
= current
;
4350 struct held_lock
*hlock
;
4351 unsigned long flags
;
4354 if (unlikely(!debug_locks
))
4357 local_irq_save(flags
);
4358 for (i
= 0; i
< curr
->lockdep_depth
; i
++) {
4359 hlock
= curr
->held_locks
+ i
;
4361 if (not_in_range(mem_from
, mem_len
, hlock
->instance
,
4362 sizeof(*hlock
->instance
)))
4365 print_freed_lock_bug(curr
, mem_from
, mem_from
+ mem_len
, hlock
);
4368 local_irq_restore(flags
);
4370 EXPORT_SYMBOL_GPL(debug_check_no_locks_freed
);
4372 static void print_held_locks_bug(void)
4374 if (!debug_locks_off())
4376 if (debug_locks_silent
)
4380 pr_warn("====================================\n");
4381 pr_warn("WARNING: %s/%d still has locks held!\n",
4382 current
->comm
, task_pid_nr(current
));
4383 print_kernel_ident();
4384 pr_warn("------------------------------------\n");
4385 lockdep_print_held_locks(current
);
4386 printk("\nstack backtrace:\n");
4390 void debug_check_no_locks_held(void)
4392 if (unlikely(current
->lockdep_depth
> 0))
4393 print_held_locks_bug();
4395 EXPORT_SYMBOL_GPL(debug_check_no_locks_held
);
4398 void debug_show_all_locks(void)
4400 struct task_struct
*g
, *p
;
4404 if (unlikely(!debug_locks
)) {
4405 printk("INFO: lockdep is turned off.\n");
4408 printk("\nShowing all locks held in the system:\n");
4411 * Here we try to get the tasklist_lock as hard as possible,
4412 * if not successful after 2 seconds we ignore it (but keep
4413 * trying). This is to enable a debug printout even if a
4414 * tasklist_lock-holding task deadlocks or crashes.
4417 if (!read_trylock(&tasklist_lock
)) {
4419 printk("hm, tasklist_lock locked, retrying... ");
4422 printk(" #%d", 10-count
);
4426 printk(" ignoring it.\n");
4430 printk(KERN_CONT
" locked it.\n");
4433 do_each_thread(g
, p
) {
4435 * It's not reliable to print a task's held locks
4436 * if it's not sleeping (or if it's not the current
4439 if (p
->state
== TASK_RUNNING
&& p
!= current
)
4441 if (p
->lockdep_depth
)
4442 lockdep_print_held_locks(p
);
4444 if (read_trylock(&tasklist_lock
))
4446 } while_each_thread(g
, p
);
4449 pr_warn("=============================================\n\n");
4452 read_unlock(&tasklist_lock
);
4454 EXPORT_SYMBOL_GPL(debug_show_all_locks
);
4458 * Careful: only use this function if you are sure that
4459 * the task cannot run in parallel!
4461 void debug_show_held_locks(struct task_struct
*task
)
4463 if (unlikely(!debug_locks
)) {
4464 printk("INFO: lockdep is turned off.\n");
4467 lockdep_print_held_locks(task
);
4469 EXPORT_SYMBOL_GPL(debug_show_held_locks
);
4471 asmlinkage __visible
void lockdep_sys_exit(void)
4473 struct task_struct
*curr
= current
;
4475 if (unlikely(curr
->lockdep_depth
)) {
4476 if (!debug_locks_off())
4479 pr_warn("================================================\n");
4480 pr_warn("WARNING: lock held when returning to user space!\n");
4481 print_kernel_ident();
4482 pr_warn("------------------------------------------------\n");
4483 printk("%s/%d is leaving the kernel with locks still held!\n",
4484 curr
->comm
, curr
->pid
);
4485 lockdep_print_held_locks(curr
);
4489 void lockdep_rcu_suspicious(const char *file
, const int line
, const char *s
)
4491 struct task_struct
*curr
= current
;
4493 #ifndef CONFIG_PROVE_RCU_REPEATEDLY
4494 if (!debug_locks_off())
4496 #endif /* #ifdef CONFIG_PROVE_RCU_REPEATEDLY */
4497 /* Note: the following can be executed concurrently, so be careful. */
4499 pr_warn("=============================\n");
4500 pr_warn("WARNING: suspicious RCU usage\n");
4501 print_kernel_ident();
4502 pr_warn("-----------------------------\n");
4503 printk("%s:%d %s!\n", file
, line
, s
);
4504 printk("\nother info that might help us debug this:\n\n");
4505 printk("\n%srcu_scheduler_active = %d, debug_locks = %d\n",
4506 !rcu_lockdep_current_cpu_online()
4507 ? "RCU used illegally from offline CPU!\n"
4508 : !rcu_is_watching()
4509 ? "RCU used illegally from idle CPU!\n"
4511 rcu_scheduler_active
, debug_locks
);
4514 * If a CPU is in the RCU-free window in idle (ie: in the section
4515 * between rcu_idle_enter() and rcu_idle_exit(), then RCU
4516 * considers that CPU to be in an "extended quiescent state",
4517 * which means that RCU will be completely ignoring that CPU.
4518 * Therefore, rcu_read_lock() and friends have absolutely no
4519 * effect on a CPU running in that state. In other words, even if
4520 * such an RCU-idle CPU has called rcu_read_lock(), RCU might well
4521 * delete data structures out from under it. RCU really has no
4522 * choice here: we need to keep an RCU-free window in idle where
4523 * the CPU may possibly enter into low power mode. This way we can
4524 * notice an extended quiescent state to other CPUs that started a grace
4525 * period. Otherwise we would delay any grace period as long as we run
4528 * So complain bitterly if someone does call rcu_read_lock(),
4529 * rcu_read_lock_bh() and so on from extended quiescent states.
4531 if (!rcu_is_watching())
4532 printk("RCU used illegally from extended quiescent state!\n");
4534 lockdep_print_held_locks(curr
);
4535 printk("\nstack backtrace:\n");
4538 EXPORT_SYMBOL_GPL(lockdep_rcu_suspicious
);