4 * Copyright (C) 2008 Steven Rostedt <srostedt@redhat.com>
6 #include <linux/ring_buffer.h>
7 #include <linux/spinlock.h>
8 #include <linux/debugfs.h>
9 #include <linux/uaccess.h>
10 #include <linux/module.h>
11 #include <linux/percpu.h>
12 #include <linux/mutex.h>
13 #include <linux/sched.h> /* used for sched_clock() (for now) */
14 #include <linux/init.h>
15 #include <linux/hash.h>
16 #include <linux/list.h>
19 /* Up this if you want to test the TIME_EXTENTS and normalization */
23 u64
ring_buffer_time_stamp(int cpu
)
25 /* shift to debug/test normalization and TIME_EXTENTS */
26 return sched_clock() << DEBUG_SHIFT
;
29 void ring_buffer_normalize_time_stamp(int cpu
, u64
*ts
)
31 /* Just stupid testing the normalize function and deltas */
35 #define RB_EVNT_HDR_SIZE (sizeof(struct ring_buffer_event))
36 #define RB_ALIGNMENT_SHIFT 2
37 #define RB_ALIGNMENT (1 << RB_ALIGNMENT_SHIFT)
38 #define RB_MAX_SMALL_DATA 28
41 RB_LEN_TIME_EXTEND
= 8,
42 RB_LEN_TIME_STAMP
= 16,
45 /* inline for ring buffer fast paths */
46 static inline unsigned
47 rb_event_length(struct ring_buffer_event
*event
)
51 switch (event
->type
) {
52 case RINGBUF_TYPE_PADDING
:
56 case RINGBUF_TYPE_TIME_EXTEND
:
57 return RB_LEN_TIME_EXTEND
;
59 case RINGBUF_TYPE_TIME_STAMP
:
60 return RB_LEN_TIME_STAMP
;
62 case RINGBUF_TYPE_DATA
:
64 length
= event
->len
<< RB_ALIGNMENT_SHIFT
;
66 length
= event
->array
[0];
67 return length
+ RB_EVNT_HDR_SIZE
;
76 * ring_buffer_event_length - return the length of the event
77 * @event: the event to get the length of
79 unsigned ring_buffer_event_length(struct ring_buffer_event
*event
)
81 return rb_event_length(event
);
84 /* inline for ring buffer fast paths */
86 rb_event_data(struct ring_buffer_event
*event
)
88 BUG_ON(event
->type
!= RINGBUF_TYPE_DATA
);
89 /* If length is in len field, then array[0] has the data */
91 return (void *)&event
->array
[0];
92 /* Otherwise length is in array[0] and array[1] has the data */
93 return (void *)&event
->array
[1];
97 * ring_buffer_event_data - return the data of the event
98 * @event: the event to get the data from
100 void *ring_buffer_event_data(struct ring_buffer_event
*event
)
102 return rb_event_data(event
);
105 #define for_each_buffer_cpu(buffer, cpu) \
106 for_each_cpu_mask(cpu, buffer->cpumask)
109 #define TS_MASK ((1ULL << TS_SHIFT) - 1)
110 #define TS_DELTA_TEST (~TS_MASK)
113 * This hack stolen from mm/slob.c.
114 * We can store per page timing information in the page frame of the page.
115 * Thanks to Peter Zijlstra for suggesting this idea.
118 u64 time_stamp
; /* page time stamp */
119 local_t write
; /* index for next write */
120 local_t commit
; /* write commited index */
121 unsigned read
; /* index for next read */
122 struct list_head list
; /* list of free pages */
123 void *page
; /* Actual data page */
127 * Also stolen from mm/slob.c. Thanks to Mathieu Desnoyers for pointing
130 static inline void free_buffer_page(struct buffer_page
*bpage
)
133 __free_page(bpage
->page
);
138 * We need to fit the time_stamp delta into 27 bits.
140 static inline int test_time_stamp(u64 delta
)
142 if (delta
& TS_DELTA_TEST
)
147 #define BUF_PAGE_SIZE PAGE_SIZE
150 * head_page == tail_page && head == tail then buffer is empty.
152 struct ring_buffer_per_cpu
{
154 struct ring_buffer
*buffer
;
156 struct lock_class_key lock_key
;
157 struct list_head pages
;
158 struct buffer_page
*head_page
; /* read from head */
159 struct buffer_page
*tail_page
; /* write to tail */
160 struct buffer_page
*commit_page
; /* commited pages */
161 struct buffer_page
*reader_page
;
162 unsigned long overrun
;
163 unsigned long entries
;
166 atomic_t record_disabled
;
175 atomic_t record_disabled
;
179 struct ring_buffer_per_cpu
**buffers
;
182 struct ring_buffer_iter
{
183 struct ring_buffer_per_cpu
*cpu_buffer
;
185 struct buffer_page
*head_page
;
189 #define RB_WARN_ON(buffer, cond) \
191 if (unlikely(cond)) { \
192 atomic_inc(&buffer->record_disabled); \
197 #define RB_WARN_ON_RET(buffer, cond) \
199 if (unlikely(cond)) { \
200 atomic_inc(&buffer->record_disabled); \
206 #define RB_WARN_ON_ONCE(buffer, cond) \
209 if (unlikely(cond) && !once) { \
211 atomic_inc(&buffer->record_disabled); \
217 * check_pages - integrity check of buffer pages
218 * @cpu_buffer: CPU buffer with pages to test
220 * As a safty measure we check to make sure the data pages have not
223 static int rb_check_pages(struct ring_buffer_per_cpu
*cpu_buffer
)
225 struct list_head
*head
= &cpu_buffer
->pages
;
226 struct buffer_page
*page
, *tmp
;
228 RB_WARN_ON_RET(cpu_buffer
, head
->next
->prev
!= head
);
229 RB_WARN_ON_RET(cpu_buffer
, head
->prev
->next
!= head
);
231 list_for_each_entry_safe(page
, tmp
, head
, list
) {
232 RB_WARN_ON_RET(cpu_buffer
,
233 page
->list
.next
->prev
!= &page
->list
);
234 RB_WARN_ON_RET(cpu_buffer
,
235 page
->list
.prev
->next
!= &page
->list
);
241 static int rb_allocate_pages(struct ring_buffer_per_cpu
*cpu_buffer
,
244 struct list_head
*head
= &cpu_buffer
->pages
;
245 struct buffer_page
*page
, *tmp
;
250 for (i
= 0; i
< nr_pages
; i
++) {
251 page
= kzalloc_node(ALIGN(sizeof(*page
), cache_line_size()),
252 GFP_KERNEL
, cpu_to_node(cpu_buffer
->cpu
));
255 list_add(&page
->list
, &pages
);
257 addr
= __get_free_page(GFP_KERNEL
);
260 page
->page
= (void *)addr
;
263 list_splice(&pages
, head
);
265 rb_check_pages(cpu_buffer
);
270 list_for_each_entry_safe(page
, tmp
, &pages
, list
) {
271 list_del_init(&page
->list
);
272 free_buffer_page(page
);
277 static struct ring_buffer_per_cpu
*
278 rb_allocate_cpu_buffer(struct ring_buffer
*buffer
, int cpu
)
280 struct ring_buffer_per_cpu
*cpu_buffer
;
281 struct buffer_page
*page
;
285 cpu_buffer
= kzalloc_node(ALIGN(sizeof(*cpu_buffer
), cache_line_size()),
286 GFP_KERNEL
, cpu_to_node(cpu
));
290 cpu_buffer
->cpu
= cpu
;
291 cpu_buffer
->buffer
= buffer
;
292 spin_lock_init(&cpu_buffer
->lock
);
293 INIT_LIST_HEAD(&cpu_buffer
->pages
);
295 page
= kzalloc_node(ALIGN(sizeof(*page
), cache_line_size()),
296 GFP_KERNEL
, cpu_to_node(cpu
));
298 goto fail_free_buffer
;
300 cpu_buffer
->reader_page
= page
;
301 addr
= __get_free_page(GFP_KERNEL
);
303 goto fail_free_reader
;
304 page
->page
= (void *)addr
;
306 INIT_LIST_HEAD(&cpu_buffer
->reader_page
->list
);
308 ret
= rb_allocate_pages(cpu_buffer
, buffer
->pages
);
310 goto fail_free_reader
;
312 cpu_buffer
->head_page
313 = list_entry(cpu_buffer
->pages
.next
, struct buffer_page
, list
);
314 cpu_buffer
->tail_page
= cpu_buffer
->commit_page
= cpu_buffer
->head_page
;
319 free_buffer_page(cpu_buffer
->reader_page
);
326 static void rb_free_cpu_buffer(struct ring_buffer_per_cpu
*cpu_buffer
)
328 struct list_head
*head
= &cpu_buffer
->pages
;
329 struct buffer_page
*page
, *tmp
;
331 list_del_init(&cpu_buffer
->reader_page
->list
);
332 free_buffer_page(cpu_buffer
->reader_page
);
334 list_for_each_entry_safe(page
, tmp
, head
, list
) {
335 list_del_init(&page
->list
);
336 free_buffer_page(page
);
342 * Causes compile errors if the struct buffer_page gets bigger
343 * than the struct page.
345 extern int ring_buffer_page_too_big(void);
348 * ring_buffer_alloc - allocate a new ring_buffer
349 * @size: the size in bytes that is needed.
350 * @flags: attributes to set for the ring buffer.
352 * Currently the only flag that is available is the RB_FL_OVERWRITE
353 * flag. This flag means that the buffer will overwrite old data
354 * when the buffer wraps. If this flag is not set, the buffer will
355 * drop data when the tail hits the head.
357 struct ring_buffer
*ring_buffer_alloc(unsigned long size
, unsigned flags
)
359 struct ring_buffer
*buffer
;
363 /* Paranoid! Optimizes out when all is well */
364 if (sizeof(struct buffer_page
) > sizeof(struct page
))
365 ring_buffer_page_too_big();
368 /* keep it in its own cache line */
369 buffer
= kzalloc(ALIGN(sizeof(*buffer
), cache_line_size()),
374 buffer
->pages
= DIV_ROUND_UP(size
, BUF_PAGE_SIZE
);
375 buffer
->flags
= flags
;
377 /* need at least two pages */
378 if (buffer
->pages
== 1)
381 buffer
->cpumask
= cpu_possible_map
;
382 buffer
->cpus
= nr_cpu_ids
;
384 bsize
= sizeof(void *) * nr_cpu_ids
;
385 buffer
->buffers
= kzalloc(ALIGN(bsize
, cache_line_size()),
387 if (!buffer
->buffers
)
388 goto fail_free_buffer
;
390 for_each_buffer_cpu(buffer
, cpu
) {
391 buffer
->buffers
[cpu
] =
392 rb_allocate_cpu_buffer(buffer
, cpu
);
393 if (!buffer
->buffers
[cpu
])
394 goto fail_free_buffers
;
397 mutex_init(&buffer
->mutex
);
402 for_each_buffer_cpu(buffer
, cpu
) {
403 if (buffer
->buffers
[cpu
])
404 rb_free_cpu_buffer(buffer
->buffers
[cpu
]);
406 kfree(buffer
->buffers
);
414 * ring_buffer_free - free a ring buffer.
415 * @buffer: the buffer to free.
418 ring_buffer_free(struct ring_buffer
*buffer
)
422 for_each_buffer_cpu(buffer
, cpu
)
423 rb_free_cpu_buffer(buffer
->buffers
[cpu
]);
428 static void rb_reset_cpu(struct ring_buffer_per_cpu
*cpu_buffer
);
431 rb_remove_pages(struct ring_buffer_per_cpu
*cpu_buffer
, unsigned nr_pages
)
433 struct buffer_page
*page
;
437 atomic_inc(&cpu_buffer
->record_disabled
);
440 for (i
= 0; i
< nr_pages
; i
++) {
441 BUG_ON(list_empty(&cpu_buffer
->pages
));
442 p
= cpu_buffer
->pages
.next
;
443 page
= list_entry(p
, struct buffer_page
, list
);
444 list_del_init(&page
->list
);
445 free_buffer_page(page
);
447 BUG_ON(list_empty(&cpu_buffer
->pages
));
449 rb_reset_cpu(cpu_buffer
);
451 rb_check_pages(cpu_buffer
);
453 atomic_dec(&cpu_buffer
->record_disabled
);
458 rb_insert_pages(struct ring_buffer_per_cpu
*cpu_buffer
,
459 struct list_head
*pages
, unsigned nr_pages
)
461 struct buffer_page
*page
;
465 atomic_inc(&cpu_buffer
->record_disabled
);
468 for (i
= 0; i
< nr_pages
; i
++) {
469 BUG_ON(list_empty(pages
));
471 page
= list_entry(p
, struct buffer_page
, list
);
472 list_del_init(&page
->list
);
473 list_add_tail(&page
->list
, &cpu_buffer
->pages
);
475 rb_reset_cpu(cpu_buffer
);
477 rb_check_pages(cpu_buffer
);
479 atomic_dec(&cpu_buffer
->record_disabled
);
483 * ring_buffer_resize - resize the ring buffer
484 * @buffer: the buffer to resize.
485 * @size: the new size.
487 * The tracer is responsible for making sure that the buffer is
488 * not being used while changing the size.
489 * Note: We may be able to change the above requirement by using
490 * RCU synchronizations.
492 * Minimum size is 2 * BUF_PAGE_SIZE.
494 * Returns -1 on failure.
496 int ring_buffer_resize(struct ring_buffer
*buffer
, unsigned long size
)
498 struct ring_buffer_per_cpu
*cpu_buffer
;
499 unsigned nr_pages
, rm_pages
, new_pages
;
500 struct buffer_page
*page
, *tmp
;
501 unsigned long buffer_size
;
506 size
= DIV_ROUND_UP(size
, BUF_PAGE_SIZE
);
507 size
*= BUF_PAGE_SIZE
;
508 buffer_size
= buffer
->pages
* BUF_PAGE_SIZE
;
510 /* we need a minimum of two pages */
511 if (size
< BUF_PAGE_SIZE
* 2)
512 size
= BUF_PAGE_SIZE
* 2;
514 if (size
== buffer_size
)
517 mutex_lock(&buffer
->mutex
);
519 nr_pages
= DIV_ROUND_UP(size
, BUF_PAGE_SIZE
);
521 if (size
< buffer_size
) {
523 /* easy case, just free pages */
524 BUG_ON(nr_pages
>= buffer
->pages
);
526 rm_pages
= buffer
->pages
- nr_pages
;
528 for_each_buffer_cpu(buffer
, cpu
) {
529 cpu_buffer
= buffer
->buffers
[cpu
];
530 rb_remove_pages(cpu_buffer
, rm_pages
);
536 * This is a bit more difficult. We only want to add pages
537 * when we can allocate enough for all CPUs. We do this
538 * by allocating all the pages and storing them on a local
539 * link list. If we succeed in our allocation, then we
540 * add these pages to the cpu_buffers. Otherwise we just free
541 * them all and return -ENOMEM;
543 BUG_ON(nr_pages
<= buffer
->pages
);
544 new_pages
= nr_pages
- buffer
->pages
;
546 for_each_buffer_cpu(buffer
, cpu
) {
547 for (i
= 0; i
< new_pages
; i
++) {
548 page
= kzalloc_node(ALIGN(sizeof(*page
),
550 GFP_KERNEL
, cpu_to_node(cpu
));
553 list_add(&page
->list
, &pages
);
554 addr
= __get_free_page(GFP_KERNEL
);
557 page
->page
= (void *)addr
;
561 for_each_buffer_cpu(buffer
, cpu
) {
562 cpu_buffer
= buffer
->buffers
[cpu
];
563 rb_insert_pages(cpu_buffer
, &pages
, new_pages
);
566 BUG_ON(!list_empty(&pages
));
569 buffer
->pages
= nr_pages
;
570 mutex_unlock(&buffer
->mutex
);
575 list_for_each_entry_safe(page
, tmp
, &pages
, list
) {
576 list_del_init(&page
->list
);
577 free_buffer_page(page
);
582 static inline int rb_null_event(struct ring_buffer_event
*event
)
584 return event
->type
== RINGBUF_TYPE_PADDING
;
587 static inline void *__rb_page_index(struct buffer_page
*page
, unsigned index
)
589 return page
->page
+ index
;
592 static inline struct ring_buffer_event
*
593 rb_reader_event(struct ring_buffer_per_cpu
*cpu_buffer
)
595 return __rb_page_index(cpu_buffer
->reader_page
,
596 cpu_buffer
->reader_page
->read
);
599 static inline struct ring_buffer_event
*
600 rb_head_event(struct ring_buffer_per_cpu
*cpu_buffer
)
602 return __rb_page_index(cpu_buffer
->head_page
,
603 cpu_buffer
->head_page
->read
);
606 static inline struct ring_buffer_event
*
607 rb_iter_head_event(struct ring_buffer_iter
*iter
)
609 return __rb_page_index(iter
->head_page
, iter
->head
);
612 static inline unsigned rb_page_write(struct buffer_page
*bpage
)
614 return local_read(&bpage
->write
);
617 static inline unsigned rb_page_commit(struct buffer_page
*bpage
)
619 return local_read(&bpage
->commit
);
622 /* Size is determined by what has been commited */
623 static inline unsigned rb_page_size(struct buffer_page
*bpage
)
625 return rb_page_commit(bpage
);
628 static inline unsigned
629 rb_commit_index(struct ring_buffer_per_cpu
*cpu_buffer
)
631 return rb_page_commit(cpu_buffer
->commit_page
);
634 static inline unsigned rb_head_size(struct ring_buffer_per_cpu
*cpu_buffer
)
636 return rb_page_commit(cpu_buffer
->head_page
);
640 * When the tail hits the head and the buffer is in overwrite mode,
641 * the head jumps to the next page and all content on the previous
642 * page is discarded. But before doing so, we update the overrun
643 * variable of the buffer.
645 static void rb_update_overflow(struct ring_buffer_per_cpu
*cpu_buffer
)
647 struct ring_buffer_event
*event
;
650 for (head
= 0; head
< rb_head_size(cpu_buffer
);
651 head
+= rb_event_length(event
)) {
653 event
= __rb_page_index(cpu_buffer
->head_page
, head
);
654 BUG_ON(rb_null_event(event
));
655 /* Only count data entries */
656 if (event
->type
!= RINGBUF_TYPE_DATA
)
658 cpu_buffer
->overrun
++;
659 cpu_buffer
->entries
--;
663 static inline void rb_inc_page(struct ring_buffer_per_cpu
*cpu_buffer
,
664 struct buffer_page
**page
)
666 struct list_head
*p
= (*page
)->list
.next
;
668 if (p
== &cpu_buffer
->pages
)
671 *page
= list_entry(p
, struct buffer_page
, list
);
674 static inline unsigned
675 rb_event_index(struct ring_buffer_event
*event
)
677 unsigned long addr
= (unsigned long)event
;
679 return (addr
& ~PAGE_MASK
) - (PAGE_SIZE
- BUF_PAGE_SIZE
);
683 rb_is_commit(struct ring_buffer_per_cpu
*cpu_buffer
,
684 struct ring_buffer_event
*event
)
686 unsigned long addr
= (unsigned long)event
;
689 index
= rb_event_index(event
);
692 return cpu_buffer
->commit_page
->page
== (void *)addr
&&
693 rb_commit_index(cpu_buffer
) == index
;
697 rb_set_commit_event(struct ring_buffer_per_cpu
*cpu_buffer
,
698 struct ring_buffer_event
*event
)
700 unsigned long addr
= (unsigned long)event
;
703 index
= rb_event_index(event
);
706 while (cpu_buffer
->commit_page
->page
!= (void *)addr
) {
707 RB_WARN_ON(cpu_buffer
,
708 cpu_buffer
->commit_page
== cpu_buffer
->tail_page
);
709 cpu_buffer
->commit_page
->commit
=
710 cpu_buffer
->commit_page
->write
;
711 rb_inc_page(cpu_buffer
, &cpu_buffer
->commit_page
);
712 cpu_buffer
->write_stamp
= cpu_buffer
->commit_page
->time_stamp
;
715 /* Now set the commit to the event's index */
716 local_set(&cpu_buffer
->commit_page
->commit
, index
);
720 rb_set_commit_to_write(struct ring_buffer_per_cpu
*cpu_buffer
)
723 * We only race with interrupts and NMIs on this CPU.
724 * If we own the commit event, then we can commit
725 * all others that interrupted us, since the interruptions
726 * are in stack format (they finish before they come
727 * back to us). This allows us to do a simple loop to
728 * assign the commit to the tail.
730 while (cpu_buffer
->commit_page
!= cpu_buffer
->tail_page
) {
731 cpu_buffer
->commit_page
->commit
=
732 cpu_buffer
->commit_page
->write
;
733 rb_inc_page(cpu_buffer
, &cpu_buffer
->commit_page
);
734 cpu_buffer
->write_stamp
= cpu_buffer
->commit_page
->time_stamp
;
735 /* add barrier to keep gcc from optimizing too much */
738 while (rb_commit_index(cpu_buffer
) !=
739 rb_page_write(cpu_buffer
->commit_page
)) {
740 cpu_buffer
->commit_page
->commit
=
741 cpu_buffer
->commit_page
->write
;
746 static void rb_reset_reader_page(struct ring_buffer_per_cpu
*cpu_buffer
)
748 cpu_buffer
->read_stamp
= cpu_buffer
->reader_page
->time_stamp
;
749 cpu_buffer
->reader_page
->read
= 0;
752 static inline void rb_inc_iter(struct ring_buffer_iter
*iter
)
754 struct ring_buffer_per_cpu
*cpu_buffer
= iter
->cpu_buffer
;
757 * The iterator could be on the reader page (it starts there).
758 * But the head could have moved, since the reader was
759 * found. Check for this case and assign the iterator
760 * to the head page instead of next.
762 if (iter
->head_page
== cpu_buffer
->reader_page
)
763 iter
->head_page
= cpu_buffer
->head_page
;
765 rb_inc_page(cpu_buffer
, &iter
->head_page
);
767 iter
->read_stamp
= iter
->head_page
->time_stamp
;
772 * ring_buffer_update_event - update event type and data
773 * @event: the even to update
774 * @type: the type of event
775 * @length: the size of the event field in the ring buffer
777 * Update the type and data fields of the event. The length
778 * is the actual size that is written to the ring buffer,
779 * and with this, we can determine what to place into the
783 rb_update_event(struct ring_buffer_event
*event
,
784 unsigned type
, unsigned length
)
790 case RINGBUF_TYPE_PADDING
:
793 case RINGBUF_TYPE_TIME_EXTEND
:
795 (RB_LEN_TIME_EXTEND
+ (RB_ALIGNMENT
-1))
796 >> RB_ALIGNMENT_SHIFT
;
799 case RINGBUF_TYPE_TIME_STAMP
:
801 (RB_LEN_TIME_STAMP
+ (RB_ALIGNMENT
-1))
802 >> RB_ALIGNMENT_SHIFT
;
805 case RINGBUF_TYPE_DATA
:
806 length
-= RB_EVNT_HDR_SIZE
;
807 if (length
> RB_MAX_SMALL_DATA
) {
809 event
->array
[0] = length
;
812 (length
+ (RB_ALIGNMENT
-1))
813 >> RB_ALIGNMENT_SHIFT
;
820 static inline unsigned rb_calculate_event_length(unsigned length
)
822 struct ring_buffer_event event
; /* Used only for sizeof array */
824 /* zero length can cause confusions */
828 if (length
> RB_MAX_SMALL_DATA
)
829 length
+= sizeof(event
.array
[0]);
831 length
+= RB_EVNT_HDR_SIZE
;
832 length
= ALIGN(length
, RB_ALIGNMENT
);
837 static struct ring_buffer_event
*
838 __rb_reserve_next(struct ring_buffer_per_cpu
*cpu_buffer
,
839 unsigned type
, unsigned long length
, u64
*ts
)
841 struct buffer_page
*tail_page
, *head_page
, *reader_page
;
842 unsigned long tail
, write
;
843 struct ring_buffer
*buffer
= cpu_buffer
->buffer
;
844 struct ring_buffer_event
*event
;
847 tail_page
= cpu_buffer
->tail_page
;
848 write
= local_add_return(length
, &tail_page
->write
);
849 tail
= write
- length
;
851 /* See if we shot pass the end of this buffer page */
852 if (write
> BUF_PAGE_SIZE
) {
853 struct buffer_page
*next_page
= tail_page
;
855 spin_lock_irqsave(&cpu_buffer
->lock
, flags
);
857 rb_inc_page(cpu_buffer
, &next_page
);
859 head_page
= cpu_buffer
->head_page
;
860 reader_page
= cpu_buffer
->reader_page
;
862 /* we grabbed the lock before incrementing */
863 RB_WARN_ON(cpu_buffer
, next_page
== reader_page
);
866 * If for some reason, we had an interrupt storm that made
867 * it all the way around the buffer, bail, and warn
870 if (unlikely(next_page
== cpu_buffer
->commit_page
)) {
875 if (next_page
== head_page
) {
876 if (!(buffer
->flags
& RB_FL_OVERWRITE
)) {
878 if (tail
<= BUF_PAGE_SIZE
)
879 local_set(&tail_page
->write
, tail
);
883 /* tail_page has not moved yet? */
884 if (tail_page
== cpu_buffer
->tail_page
) {
885 /* count overflows */
886 rb_update_overflow(cpu_buffer
);
888 rb_inc_page(cpu_buffer
, &head_page
);
889 cpu_buffer
->head_page
= head_page
;
890 cpu_buffer
->head_page
->read
= 0;
895 * If the tail page is still the same as what we think
896 * it is, then it is up to us to update the tail
899 if (tail_page
== cpu_buffer
->tail_page
) {
900 local_set(&next_page
->write
, 0);
901 local_set(&next_page
->commit
, 0);
902 cpu_buffer
->tail_page
= next_page
;
904 /* reread the time stamp */
905 *ts
= ring_buffer_time_stamp(cpu_buffer
->cpu
);
906 cpu_buffer
->tail_page
->time_stamp
= *ts
;
910 * The actual tail page has moved forward.
912 if (tail
< BUF_PAGE_SIZE
) {
913 /* Mark the rest of the page with padding */
914 event
= __rb_page_index(tail_page
, tail
);
915 event
->type
= RINGBUF_TYPE_PADDING
;
918 if (tail
<= BUF_PAGE_SIZE
)
919 /* Set the write back to the previous setting */
920 local_set(&tail_page
->write
, tail
);
923 * If this was a commit entry that failed,
926 if (tail_page
== cpu_buffer
->commit_page
&&
927 tail
== rb_commit_index(cpu_buffer
)) {
928 rb_set_commit_to_write(cpu_buffer
);
931 spin_unlock_irqrestore(&cpu_buffer
->lock
, flags
);
933 /* fail and let the caller try again */
934 return ERR_PTR(-EAGAIN
);
937 /* We reserved something on the buffer */
939 BUG_ON(write
> BUF_PAGE_SIZE
);
941 event
= __rb_page_index(tail_page
, tail
);
942 rb_update_event(event
, type
, length
);
945 * If this is a commit and the tail is zero, then update
946 * this page's time stamp.
948 if (!tail
&& rb_is_commit(cpu_buffer
, event
))
949 cpu_buffer
->commit_page
->time_stamp
= *ts
;
954 spin_unlock_irqrestore(&cpu_buffer
->lock
, flags
);
959 rb_add_time_stamp(struct ring_buffer_per_cpu
*cpu_buffer
,
962 struct ring_buffer_event
*event
;
966 if (unlikely(*delta
> (1ULL << 59) && !once
++)) {
967 printk(KERN_WARNING
"Delta way too big! %llu"
968 " ts=%llu write stamp = %llu\n",
969 *delta
, *ts
, cpu_buffer
->write_stamp
);
974 * The delta is too big, we to add a
977 event
= __rb_reserve_next(cpu_buffer
,
978 RINGBUF_TYPE_TIME_EXTEND
,
984 if (PTR_ERR(event
) == -EAGAIN
)
987 /* Only a commited time event can update the write stamp */
988 if (rb_is_commit(cpu_buffer
, event
)) {
990 * If this is the first on the page, then we need to
991 * update the page itself, and just put in a zero.
993 if (rb_event_index(event
)) {
994 event
->time_delta
= *delta
& TS_MASK
;
995 event
->array
[0] = *delta
>> TS_SHIFT
;
997 cpu_buffer
->commit_page
->time_stamp
= *ts
;
998 event
->time_delta
= 0;
1001 cpu_buffer
->write_stamp
= *ts
;
1002 /* let the caller know this was the commit */
1005 /* Darn, this is just wasted space */
1006 event
->time_delta
= 0;
1007 event
->array
[0] = 0;
1016 static struct ring_buffer_event
*
1017 rb_reserve_next_event(struct ring_buffer_per_cpu
*cpu_buffer
,
1018 unsigned type
, unsigned long length
)
1020 struct ring_buffer_event
*event
;
1025 ts
= ring_buffer_time_stamp(cpu_buffer
->cpu
);
1028 * Only the first commit can update the timestamp.
1029 * Yes there is a race here. If an interrupt comes in
1030 * just after the conditional and it traces too, then it
1031 * will also check the deltas. More than one timestamp may
1032 * also be made. But only the entry that did the actual
1033 * commit will be something other than zero.
1035 if (cpu_buffer
->tail_page
== cpu_buffer
->commit_page
&&
1036 rb_page_write(cpu_buffer
->tail_page
) ==
1037 rb_commit_index(cpu_buffer
)) {
1039 delta
= ts
- cpu_buffer
->write_stamp
;
1041 /* make sure this delta is calculated here */
1044 /* Did the write stamp get updated already? */
1045 if (unlikely(ts
< cpu_buffer
->write_stamp
))
1048 if (test_time_stamp(delta
)) {
1050 commit
= rb_add_time_stamp(cpu_buffer
, &ts
, &delta
);
1052 if (commit
== -EBUSY
)
1055 if (commit
== -EAGAIN
)
1058 RB_WARN_ON(cpu_buffer
, commit
< 0);
1061 /* Non commits have zero deltas */
1064 event
= __rb_reserve_next(cpu_buffer
, type
, length
, &ts
);
1065 if (PTR_ERR(event
) == -EAGAIN
)
1069 if (unlikely(commit
))
1071 * Ouch! We needed a timestamp and it was commited. But
1072 * we didn't get our event reserved.
1074 rb_set_commit_to_write(cpu_buffer
);
1079 * If the timestamp was commited, make the commit our entry
1080 * now so that we will update it when needed.
1083 rb_set_commit_event(cpu_buffer
, event
);
1084 else if (!rb_is_commit(cpu_buffer
, event
))
1087 event
->time_delta
= delta
;
1092 static DEFINE_PER_CPU(int, rb_need_resched
);
1095 * ring_buffer_lock_reserve - reserve a part of the buffer
1096 * @buffer: the ring buffer to reserve from
1097 * @length: the length of the data to reserve (excluding event header)
1098 * @flags: a pointer to save the interrupt flags
1100 * Returns a reseverd event on the ring buffer to copy directly to.
1101 * The user of this interface will need to get the body to write into
1102 * and can use the ring_buffer_event_data() interface.
1104 * The length is the length of the data needed, not the event length
1105 * which also includes the event header.
1107 * Must be paired with ring_buffer_unlock_commit, unless NULL is returned.
1108 * If NULL is returned, then nothing has been allocated or locked.
1110 struct ring_buffer_event
*
1111 ring_buffer_lock_reserve(struct ring_buffer
*buffer
,
1112 unsigned long length
,
1113 unsigned long *flags
)
1115 struct ring_buffer_per_cpu
*cpu_buffer
;
1116 struct ring_buffer_event
*event
;
1119 if (atomic_read(&buffer
->record_disabled
))
1122 /* If we are tracing schedule, we don't want to recurse */
1123 resched
= need_resched();
1124 preempt_disable_notrace();
1126 cpu
= raw_smp_processor_id();
1128 if (!cpu_isset(cpu
, buffer
->cpumask
))
1131 cpu_buffer
= buffer
->buffers
[cpu
];
1133 if (atomic_read(&cpu_buffer
->record_disabled
))
1136 length
= rb_calculate_event_length(length
);
1137 if (length
> BUF_PAGE_SIZE
)
1140 event
= rb_reserve_next_event(cpu_buffer
, RINGBUF_TYPE_DATA
, length
);
1145 * Need to store resched state on this cpu.
1146 * Only the first needs to.
1149 if (preempt_count() == 1)
1150 per_cpu(rb_need_resched
, cpu
) = resched
;
1156 preempt_enable_notrace();
1158 preempt_enable_notrace();
1162 static void rb_commit(struct ring_buffer_per_cpu
*cpu_buffer
,
1163 struct ring_buffer_event
*event
)
1165 cpu_buffer
->entries
++;
1167 /* Only process further if we own the commit */
1168 if (!rb_is_commit(cpu_buffer
, event
))
1171 cpu_buffer
->write_stamp
+= event
->time_delta
;
1173 rb_set_commit_to_write(cpu_buffer
);
1177 * ring_buffer_unlock_commit - commit a reserved
1178 * @buffer: The buffer to commit to
1179 * @event: The event pointer to commit.
1180 * @flags: the interrupt flags received from ring_buffer_lock_reserve.
1182 * This commits the data to the ring buffer, and releases any locks held.
1184 * Must be paired with ring_buffer_lock_reserve.
1186 int ring_buffer_unlock_commit(struct ring_buffer
*buffer
,
1187 struct ring_buffer_event
*event
,
1188 unsigned long flags
)
1190 struct ring_buffer_per_cpu
*cpu_buffer
;
1191 int cpu
= raw_smp_processor_id();
1193 cpu_buffer
= buffer
->buffers
[cpu
];
1195 rb_commit(cpu_buffer
, event
);
1198 * Only the last preempt count needs to restore preemption.
1200 if (preempt_count() == 1) {
1201 if (per_cpu(rb_need_resched
, cpu
))
1202 preempt_enable_no_resched_notrace();
1204 preempt_enable_notrace();
1206 preempt_enable_no_resched_notrace();
1212 * ring_buffer_write - write data to the buffer without reserving
1213 * @buffer: The ring buffer to write to.
1214 * @length: The length of the data being written (excluding the event header)
1215 * @data: The data to write to the buffer.
1217 * This is like ring_buffer_lock_reserve and ring_buffer_unlock_commit as
1218 * one function. If you already have the data to write to the buffer, it
1219 * may be easier to simply call this function.
1221 * Note, like ring_buffer_lock_reserve, the length is the length of the data
1222 * and not the length of the event which would hold the header.
1224 int ring_buffer_write(struct ring_buffer
*buffer
,
1225 unsigned long length
,
1228 struct ring_buffer_per_cpu
*cpu_buffer
;
1229 struct ring_buffer_event
*event
;
1230 unsigned long event_length
;
1235 if (atomic_read(&buffer
->record_disabled
))
1238 resched
= need_resched();
1239 preempt_disable_notrace();
1241 cpu
= raw_smp_processor_id();
1243 if (!cpu_isset(cpu
, buffer
->cpumask
))
1246 cpu_buffer
= buffer
->buffers
[cpu
];
1248 if (atomic_read(&cpu_buffer
->record_disabled
))
1251 event_length
= rb_calculate_event_length(length
);
1252 event
= rb_reserve_next_event(cpu_buffer
,
1253 RINGBUF_TYPE_DATA
, event_length
);
1257 body
= rb_event_data(event
);
1259 memcpy(body
, data
, length
);
1261 rb_commit(cpu_buffer
, event
);
1266 preempt_enable_no_resched_notrace();
1268 preempt_enable_notrace();
1273 static inline int rb_per_cpu_empty(struct ring_buffer_per_cpu
*cpu_buffer
)
1275 struct buffer_page
*reader
= cpu_buffer
->reader_page
;
1276 struct buffer_page
*head
= cpu_buffer
->head_page
;
1277 struct buffer_page
*commit
= cpu_buffer
->commit_page
;
1279 return reader
->read
== rb_page_commit(reader
) &&
1280 (commit
== reader
||
1282 head
->read
== rb_page_commit(commit
)));
1286 * ring_buffer_record_disable - stop all writes into the buffer
1287 * @buffer: The ring buffer to stop writes to.
1289 * This prevents all writes to the buffer. Any attempt to write
1290 * to the buffer after this will fail and return NULL.
1292 * The caller should call synchronize_sched() after this.
1294 void ring_buffer_record_disable(struct ring_buffer
*buffer
)
1296 atomic_inc(&buffer
->record_disabled
);
1300 * ring_buffer_record_enable - enable writes to the buffer
1301 * @buffer: The ring buffer to enable writes
1303 * Note, multiple disables will need the same number of enables
1304 * to truely enable the writing (much like preempt_disable).
1306 void ring_buffer_record_enable(struct ring_buffer
*buffer
)
1308 atomic_dec(&buffer
->record_disabled
);
1312 * ring_buffer_record_disable_cpu - stop all writes into the cpu_buffer
1313 * @buffer: The ring buffer to stop writes to.
1314 * @cpu: The CPU buffer to stop
1316 * This prevents all writes to the buffer. Any attempt to write
1317 * to the buffer after this will fail and return NULL.
1319 * The caller should call synchronize_sched() after this.
1321 void ring_buffer_record_disable_cpu(struct ring_buffer
*buffer
, int cpu
)
1323 struct ring_buffer_per_cpu
*cpu_buffer
;
1325 if (!cpu_isset(cpu
, buffer
->cpumask
))
1328 cpu_buffer
= buffer
->buffers
[cpu
];
1329 atomic_inc(&cpu_buffer
->record_disabled
);
1333 * ring_buffer_record_enable_cpu - enable writes to the buffer
1334 * @buffer: The ring buffer to enable writes
1335 * @cpu: The CPU to enable.
1337 * Note, multiple disables will need the same number of enables
1338 * to truely enable the writing (much like preempt_disable).
1340 void ring_buffer_record_enable_cpu(struct ring_buffer
*buffer
, int cpu
)
1342 struct ring_buffer_per_cpu
*cpu_buffer
;
1344 if (!cpu_isset(cpu
, buffer
->cpumask
))
1347 cpu_buffer
= buffer
->buffers
[cpu
];
1348 atomic_dec(&cpu_buffer
->record_disabled
);
1352 * ring_buffer_entries_cpu - get the number of entries in a cpu buffer
1353 * @buffer: The ring buffer
1354 * @cpu: The per CPU buffer to get the entries from.
1356 unsigned long ring_buffer_entries_cpu(struct ring_buffer
*buffer
, int cpu
)
1358 struct ring_buffer_per_cpu
*cpu_buffer
;
1360 if (!cpu_isset(cpu
, buffer
->cpumask
))
1363 cpu_buffer
= buffer
->buffers
[cpu
];
1364 return cpu_buffer
->entries
;
1368 * ring_buffer_overrun_cpu - get the number of overruns in a cpu_buffer
1369 * @buffer: The ring buffer
1370 * @cpu: The per CPU buffer to get the number of overruns from
1372 unsigned long ring_buffer_overrun_cpu(struct ring_buffer
*buffer
, int cpu
)
1374 struct ring_buffer_per_cpu
*cpu_buffer
;
1376 if (!cpu_isset(cpu
, buffer
->cpumask
))
1379 cpu_buffer
= buffer
->buffers
[cpu
];
1380 return cpu_buffer
->overrun
;
1384 * ring_buffer_entries - get the number of entries in a buffer
1385 * @buffer: The ring buffer
1387 * Returns the total number of entries in the ring buffer
1390 unsigned long ring_buffer_entries(struct ring_buffer
*buffer
)
1392 struct ring_buffer_per_cpu
*cpu_buffer
;
1393 unsigned long entries
= 0;
1396 /* if you care about this being correct, lock the buffer */
1397 for_each_buffer_cpu(buffer
, cpu
) {
1398 cpu_buffer
= buffer
->buffers
[cpu
];
1399 entries
+= cpu_buffer
->entries
;
1406 * ring_buffer_overrun_cpu - get the number of overruns in buffer
1407 * @buffer: The ring buffer
1409 * Returns the total number of overruns in the ring buffer
1412 unsigned long ring_buffer_overruns(struct ring_buffer
*buffer
)
1414 struct ring_buffer_per_cpu
*cpu_buffer
;
1415 unsigned long overruns
= 0;
1418 /* if you care about this being correct, lock the buffer */
1419 for_each_buffer_cpu(buffer
, cpu
) {
1420 cpu_buffer
= buffer
->buffers
[cpu
];
1421 overruns
+= cpu_buffer
->overrun
;
1428 * ring_buffer_iter_reset - reset an iterator
1429 * @iter: The iterator to reset
1431 * Resets the iterator, so that it will start from the beginning
1434 void ring_buffer_iter_reset(struct ring_buffer_iter
*iter
)
1436 struct ring_buffer_per_cpu
*cpu_buffer
= iter
->cpu_buffer
;
1438 /* Iterator usage is expected to have record disabled */
1439 if (list_empty(&cpu_buffer
->reader_page
->list
)) {
1440 iter
->head_page
= cpu_buffer
->head_page
;
1441 iter
->head
= cpu_buffer
->head_page
->read
;
1443 iter
->head_page
= cpu_buffer
->reader_page
;
1444 iter
->head
= cpu_buffer
->reader_page
->read
;
1447 iter
->read_stamp
= cpu_buffer
->read_stamp
;
1449 iter
->read_stamp
= iter
->head_page
->time_stamp
;
1453 * ring_buffer_iter_empty - check if an iterator has no more to read
1454 * @iter: The iterator to check
1456 int ring_buffer_iter_empty(struct ring_buffer_iter
*iter
)
1458 struct ring_buffer_per_cpu
*cpu_buffer
;
1460 cpu_buffer
= iter
->cpu_buffer
;
1462 return iter
->head_page
== cpu_buffer
->commit_page
&&
1463 iter
->head
== rb_commit_index(cpu_buffer
);
1467 rb_update_read_stamp(struct ring_buffer_per_cpu
*cpu_buffer
,
1468 struct ring_buffer_event
*event
)
1472 switch (event
->type
) {
1473 case RINGBUF_TYPE_PADDING
:
1476 case RINGBUF_TYPE_TIME_EXTEND
:
1477 delta
= event
->array
[0];
1479 delta
+= event
->time_delta
;
1480 cpu_buffer
->read_stamp
+= delta
;
1483 case RINGBUF_TYPE_TIME_STAMP
:
1484 /* FIXME: not implemented */
1487 case RINGBUF_TYPE_DATA
:
1488 cpu_buffer
->read_stamp
+= event
->time_delta
;
1498 rb_update_iter_read_stamp(struct ring_buffer_iter
*iter
,
1499 struct ring_buffer_event
*event
)
1503 switch (event
->type
) {
1504 case RINGBUF_TYPE_PADDING
:
1507 case RINGBUF_TYPE_TIME_EXTEND
:
1508 delta
= event
->array
[0];
1510 delta
+= event
->time_delta
;
1511 iter
->read_stamp
+= delta
;
1514 case RINGBUF_TYPE_TIME_STAMP
:
1515 /* FIXME: not implemented */
1518 case RINGBUF_TYPE_DATA
:
1519 iter
->read_stamp
+= event
->time_delta
;
1528 static struct buffer_page
*
1529 rb_get_reader_page(struct ring_buffer_per_cpu
*cpu_buffer
)
1531 struct buffer_page
*reader
= NULL
;
1532 unsigned long flags
;
1534 spin_lock_irqsave(&cpu_buffer
->lock
, flags
);
1537 reader
= cpu_buffer
->reader_page
;
1539 /* If there's more to read, return this page */
1540 if (cpu_buffer
->reader_page
->read
< rb_page_size(reader
))
1543 /* Never should we have an index greater than the size */
1544 RB_WARN_ON(cpu_buffer
,
1545 cpu_buffer
->reader_page
->read
> rb_page_size(reader
));
1547 /* check if we caught up to the tail */
1549 if (cpu_buffer
->commit_page
== cpu_buffer
->reader_page
)
1553 * Splice the empty reader page into the list around the head.
1554 * Reset the reader page to size zero.
1557 reader
= cpu_buffer
->head_page
;
1558 cpu_buffer
->reader_page
->list
.next
= reader
->list
.next
;
1559 cpu_buffer
->reader_page
->list
.prev
= reader
->list
.prev
;
1561 local_set(&cpu_buffer
->reader_page
->write
, 0);
1562 local_set(&cpu_buffer
->reader_page
->commit
, 0);
1564 /* Make the reader page now replace the head */
1565 reader
->list
.prev
->next
= &cpu_buffer
->reader_page
->list
;
1566 reader
->list
.next
->prev
= &cpu_buffer
->reader_page
->list
;
1569 * If the tail is on the reader, then we must set the head
1570 * to the inserted page, otherwise we set it one before.
1572 cpu_buffer
->head_page
= cpu_buffer
->reader_page
;
1574 if (cpu_buffer
->commit_page
!= reader
)
1575 rb_inc_page(cpu_buffer
, &cpu_buffer
->head_page
);
1577 /* Finally update the reader page to the new head */
1578 cpu_buffer
->reader_page
= reader
;
1579 rb_reset_reader_page(cpu_buffer
);
1584 spin_unlock_irqrestore(&cpu_buffer
->lock
, flags
);
1589 static void rb_advance_reader(struct ring_buffer_per_cpu
*cpu_buffer
)
1591 struct ring_buffer_event
*event
;
1592 struct buffer_page
*reader
;
1595 reader
= rb_get_reader_page(cpu_buffer
);
1597 /* This function should not be called when buffer is empty */
1600 event
= rb_reader_event(cpu_buffer
);
1602 if (event
->type
== RINGBUF_TYPE_DATA
)
1603 cpu_buffer
->entries
--;
1605 rb_update_read_stamp(cpu_buffer
, event
);
1607 length
= rb_event_length(event
);
1608 cpu_buffer
->reader_page
->read
+= length
;
1611 static void rb_advance_iter(struct ring_buffer_iter
*iter
)
1613 struct ring_buffer
*buffer
;
1614 struct ring_buffer_per_cpu
*cpu_buffer
;
1615 struct ring_buffer_event
*event
;
1618 cpu_buffer
= iter
->cpu_buffer
;
1619 buffer
= cpu_buffer
->buffer
;
1622 * Check if we are at the end of the buffer.
1624 if (iter
->head
>= rb_page_size(iter
->head_page
)) {
1625 BUG_ON(iter
->head_page
== cpu_buffer
->commit_page
);
1630 event
= rb_iter_head_event(iter
);
1632 length
= rb_event_length(event
);
1635 * This should not be called to advance the header if we are
1636 * at the tail of the buffer.
1638 BUG_ON((iter
->head_page
== cpu_buffer
->commit_page
) &&
1639 (iter
->head
+ length
> rb_commit_index(cpu_buffer
)));
1641 rb_update_iter_read_stamp(iter
, event
);
1643 iter
->head
+= length
;
1645 /* check for end of page padding */
1646 if ((iter
->head
>= rb_page_size(iter
->head_page
)) &&
1647 (iter
->head_page
!= cpu_buffer
->commit_page
))
1648 rb_advance_iter(iter
);
1652 * ring_buffer_peek - peek at the next event to be read
1653 * @buffer: The ring buffer to read
1654 * @cpu: The cpu to peak at
1655 * @ts: The timestamp counter of this event.
1657 * This will return the event that will be read next, but does
1658 * not consume the data.
1660 struct ring_buffer_event
*
1661 ring_buffer_peek(struct ring_buffer
*buffer
, int cpu
, u64
*ts
)
1663 struct ring_buffer_per_cpu
*cpu_buffer
;
1664 struct ring_buffer_event
*event
;
1665 struct buffer_page
*reader
;
1667 if (!cpu_isset(cpu
, buffer
->cpumask
))
1670 cpu_buffer
= buffer
->buffers
[cpu
];
1673 reader
= rb_get_reader_page(cpu_buffer
);
1677 event
= rb_reader_event(cpu_buffer
);
1679 switch (event
->type
) {
1680 case RINGBUF_TYPE_PADDING
:
1681 RB_WARN_ON(cpu_buffer
, 1);
1682 rb_advance_reader(cpu_buffer
);
1685 case RINGBUF_TYPE_TIME_EXTEND
:
1686 /* Internal data, OK to advance */
1687 rb_advance_reader(cpu_buffer
);
1690 case RINGBUF_TYPE_TIME_STAMP
:
1691 /* FIXME: not implemented */
1692 rb_advance_reader(cpu_buffer
);
1695 case RINGBUF_TYPE_DATA
:
1697 *ts
= cpu_buffer
->read_stamp
+ event
->time_delta
;
1698 ring_buffer_normalize_time_stamp(cpu_buffer
->cpu
, ts
);
1710 * ring_buffer_iter_peek - peek at the next event to be read
1711 * @iter: The ring buffer iterator
1712 * @ts: The timestamp counter of this event.
1714 * This will return the event that will be read next, but does
1715 * not increment the iterator.
1717 struct ring_buffer_event
*
1718 ring_buffer_iter_peek(struct ring_buffer_iter
*iter
, u64
*ts
)
1720 struct ring_buffer
*buffer
;
1721 struct ring_buffer_per_cpu
*cpu_buffer
;
1722 struct ring_buffer_event
*event
;
1724 if (ring_buffer_iter_empty(iter
))
1727 cpu_buffer
= iter
->cpu_buffer
;
1728 buffer
= cpu_buffer
->buffer
;
1731 if (rb_per_cpu_empty(cpu_buffer
))
1734 event
= rb_iter_head_event(iter
);
1736 switch (event
->type
) {
1737 case RINGBUF_TYPE_PADDING
:
1741 case RINGBUF_TYPE_TIME_EXTEND
:
1742 /* Internal data, OK to advance */
1743 rb_advance_iter(iter
);
1746 case RINGBUF_TYPE_TIME_STAMP
:
1747 /* FIXME: not implemented */
1748 rb_advance_iter(iter
);
1751 case RINGBUF_TYPE_DATA
:
1753 *ts
= iter
->read_stamp
+ event
->time_delta
;
1754 ring_buffer_normalize_time_stamp(cpu_buffer
->cpu
, ts
);
1766 * ring_buffer_consume - return an event and consume it
1767 * @buffer: The ring buffer to get the next event from
1769 * Returns the next event in the ring buffer, and that event is consumed.
1770 * Meaning, that sequential reads will keep returning a different event,
1771 * and eventually empty the ring buffer if the producer is slower.
1773 struct ring_buffer_event
*
1774 ring_buffer_consume(struct ring_buffer
*buffer
, int cpu
, u64
*ts
)
1776 struct ring_buffer_per_cpu
*cpu_buffer
;
1777 struct ring_buffer_event
*event
;
1779 if (!cpu_isset(cpu
, buffer
->cpumask
))
1782 event
= ring_buffer_peek(buffer
, cpu
, ts
);
1786 cpu_buffer
= buffer
->buffers
[cpu
];
1787 rb_advance_reader(cpu_buffer
);
1793 * ring_buffer_read_start - start a non consuming read of the buffer
1794 * @buffer: The ring buffer to read from
1795 * @cpu: The cpu buffer to iterate over
1797 * This starts up an iteration through the buffer. It also disables
1798 * the recording to the buffer until the reading is finished.
1799 * This prevents the reading from being corrupted. This is not
1800 * a consuming read, so a producer is not expected.
1802 * Must be paired with ring_buffer_finish.
1804 struct ring_buffer_iter
*
1805 ring_buffer_read_start(struct ring_buffer
*buffer
, int cpu
)
1807 struct ring_buffer_per_cpu
*cpu_buffer
;
1808 struct ring_buffer_iter
*iter
;
1809 unsigned long flags
;
1811 if (!cpu_isset(cpu
, buffer
->cpumask
))
1814 iter
= kmalloc(sizeof(*iter
), GFP_KERNEL
);
1818 cpu_buffer
= buffer
->buffers
[cpu
];
1820 iter
->cpu_buffer
= cpu_buffer
;
1822 atomic_inc(&cpu_buffer
->record_disabled
);
1823 synchronize_sched();
1825 spin_lock_irqsave(&cpu_buffer
->lock
, flags
);
1826 ring_buffer_iter_reset(iter
);
1827 spin_unlock_irqrestore(&cpu_buffer
->lock
, flags
);
1833 * ring_buffer_finish - finish reading the iterator of the buffer
1834 * @iter: The iterator retrieved by ring_buffer_start
1836 * This re-enables the recording to the buffer, and frees the
1840 ring_buffer_read_finish(struct ring_buffer_iter
*iter
)
1842 struct ring_buffer_per_cpu
*cpu_buffer
= iter
->cpu_buffer
;
1844 atomic_dec(&cpu_buffer
->record_disabled
);
1849 * ring_buffer_read - read the next item in the ring buffer by the iterator
1850 * @iter: The ring buffer iterator
1851 * @ts: The time stamp of the event read.
1853 * This reads the next event in the ring buffer and increments the iterator.
1855 struct ring_buffer_event
*
1856 ring_buffer_read(struct ring_buffer_iter
*iter
, u64
*ts
)
1858 struct ring_buffer_event
*event
;
1860 event
= ring_buffer_iter_peek(iter
, ts
);
1864 rb_advance_iter(iter
);
1870 * ring_buffer_size - return the size of the ring buffer (in bytes)
1871 * @buffer: The ring buffer.
1873 unsigned long ring_buffer_size(struct ring_buffer
*buffer
)
1875 return BUF_PAGE_SIZE
* buffer
->pages
;
1879 rb_reset_cpu(struct ring_buffer_per_cpu
*cpu_buffer
)
1881 cpu_buffer
->head_page
1882 = list_entry(cpu_buffer
->pages
.next
, struct buffer_page
, list
);
1883 local_set(&cpu_buffer
->head_page
->write
, 0);
1884 local_set(&cpu_buffer
->head_page
->commit
, 0);
1886 cpu_buffer
->head_page
->read
= 0;
1888 cpu_buffer
->tail_page
= cpu_buffer
->head_page
;
1889 cpu_buffer
->commit_page
= cpu_buffer
->head_page
;
1891 INIT_LIST_HEAD(&cpu_buffer
->reader_page
->list
);
1892 local_set(&cpu_buffer
->reader_page
->write
, 0);
1893 local_set(&cpu_buffer
->reader_page
->commit
, 0);
1894 cpu_buffer
->reader_page
->read
= 0;
1896 cpu_buffer
->overrun
= 0;
1897 cpu_buffer
->entries
= 0;
1901 * ring_buffer_reset_cpu - reset a ring buffer per CPU buffer
1902 * @buffer: The ring buffer to reset a per cpu buffer of
1903 * @cpu: The CPU buffer to be reset
1905 void ring_buffer_reset_cpu(struct ring_buffer
*buffer
, int cpu
)
1907 struct ring_buffer_per_cpu
*cpu_buffer
= buffer
->buffers
[cpu
];
1908 unsigned long flags
;
1910 if (!cpu_isset(cpu
, buffer
->cpumask
))
1913 spin_lock_irqsave(&cpu_buffer
->lock
, flags
);
1915 rb_reset_cpu(cpu_buffer
);
1917 spin_unlock_irqrestore(&cpu_buffer
->lock
, flags
);
1921 * ring_buffer_reset - reset a ring buffer
1922 * @buffer: The ring buffer to reset all cpu buffers
1924 void ring_buffer_reset(struct ring_buffer
*buffer
)
1928 for_each_buffer_cpu(buffer
, cpu
)
1929 ring_buffer_reset_cpu(buffer
, cpu
);
1933 * rind_buffer_empty - is the ring buffer empty?
1934 * @buffer: The ring buffer to test
1936 int ring_buffer_empty(struct ring_buffer
*buffer
)
1938 struct ring_buffer_per_cpu
*cpu_buffer
;
1941 /* yes this is racy, but if you don't like the race, lock the buffer */
1942 for_each_buffer_cpu(buffer
, cpu
) {
1943 cpu_buffer
= buffer
->buffers
[cpu
];
1944 if (!rb_per_cpu_empty(cpu_buffer
))
1951 * ring_buffer_empty_cpu - is a cpu buffer of a ring buffer empty?
1952 * @buffer: The ring buffer
1953 * @cpu: The CPU buffer to test
1955 int ring_buffer_empty_cpu(struct ring_buffer
*buffer
, int cpu
)
1957 struct ring_buffer_per_cpu
*cpu_buffer
;
1959 if (!cpu_isset(cpu
, buffer
->cpumask
))
1962 cpu_buffer
= buffer
->buffers
[cpu
];
1963 return rb_per_cpu_empty(cpu_buffer
);
1967 * ring_buffer_swap_cpu - swap a CPU buffer between two ring buffers
1968 * @buffer_a: One buffer to swap with
1969 * @buffer_b: The other buffer to swap with
1971 * This function is useful for tracers that want to take a "snapshot"
1972 * of a CPU buffer and has another back up buffer lying around.
1973 * it is expected that the tracer handles the cpu buffer not being
1974 * used at the moment.
1976 int ring_buffer_swap_cpu(struct ring_buffer
*buffer_a
,
1977 struct ring_buffer
*buffer_b
, int cpu
)
1979 struct ring_buffer_per_cpu
*cpu_buffer_a
;
1980 struct ring_buffer_per_cpu
*cpu_buffer_b
;
1982 if (!cpu_isset(cpu
, buffer_a
->cpumask
) ||
1983 !cpu_isset(cpu
, buffer_b
->cpumask
))
1986 /* At least make sure the two buffers are somewhat the same */
1987 if (buffer_a
->size
!= buffer_b
->size
||
1988 buffer_a
->pages
!= buffer_b
->pages
)
1991 cpu_buffer_a
= buffer_a
->buffers
[cpu
];
1992 cpu_buffer_b
= buffer_b
->buffers
[cpu
];
1995 * We can't do a synchronize_sched here because this
1996 * function can be called in atomic context.
1997 * Normally this will be called from the same CPU as cpu.
1998 * If not it's up to the caller to protect this.
2000 atomic_inc(&cpu_buffer_a
->record_disabled
);
2001 atomic_inc(&cpu_buffer_b
->record_disabled
);
2003 buffer_a
->buffers
[cpu
] = cpu_buffer_b
;
2004 buffer_b
->buffers
[cpu
] = cpu_buffer_a
;
2006 cpu_buffer_b
->buffer
= buffer_a
;
2007 cpu_buffer_a
->buffer
= buffer_b
;
2009 atomic_dec(&cpu_buffer_a
->record_disabled
);
2010 atomic_dec(&cpu_buffer_b
->record_disabled
);