Commit | Line | Data |
---|---|---|
e1152c37 PMF |
1 | /* |
2 | * ltt/ltt-relay.c | |
3 | * | |
4 | * (C) Copyright 2005-2008 - Mathieu Desnoyers (mathieu.desnoyers@polymtl.ca) | |
5 | * | |
6 | * LTTng lockless buffer space management (reader/writer). | |
7 | * | |
8 | * Author: | |
9 | * Mathieu Desnoyers (mathieu.desnoyers@polymtl.ca) | |
10 | * | |
11 | * Inspired from LTT : | |
12 | * Karim Yaghmour (karim@opersys.com) | |
13 | * Tom Zanussi (zanussi@us.ibm.com) | |
14 | * Bob Wisniewski (bob@watson.ibm.com) | |
15 | * And from K42 : | |
16 | * Bob Wisniewski (bob@watson.ibm.com) | |
17 | * | |
18 | * Changelog: | |
19 | * 08/10/08, Cleanup. | |
20 | * 19/10/05, Complete lockless mechanism. | |
21 | * 27/05/05, Modular redesign and rewrite. | |
22 | * | |
23 | * Userspace reader semantic : | |
24 | * while (poll fd != POLLHUP) { | |
25 | * - ioctl RELAY_GET_SUBBUF_SIZE | |
26 | * while (1) { | |
27 | * - ioctl GET_SUBBUF | |
28 | * - splice 1 subbuffer worth of data to a pipe | |
29 | * - splice the data from pipe to disk/network | |
30 | * - ioctl PUT_SUBBUF, check error value | |
31 | * if err val < 0, previous subbuffer was corrupted. | |
32 | * } | |
33 | * } | |
34 | */ | |
35 | ||
36 | #include <linux/time.h> | |
37 | #include <linux/ltt-tracer.h> | |
38 | #include <linux/ltt-relay.h> | |
39 | #include <linux/module.h> | |
40 | #include <linux/string.h> | |
41 | #include <linux/slab.h> | |
42 | #include <linux/init.h> | |
43 | #include <linux/rcupdate.h> | |
44 | #include <linux/sched.h> | |
45 | #include <linux/bitops.h> | |
46 | #include <linux/fs.h> | |
47 | #include <linux/smp_lock.h> | |
48 | #include <linux/debugfs.h> | |
49 | #include <linux/stat.h> | |
50 | #include <linux/cpu.h> | |
51 | #include <linux/pipe_fs_i.h> | |
52 | #include <linux/splice.h> | |
53 | #include <asm/atomic.h> | |
54 | #include <asm/local.h> | |
55 | ||
56 | #if 0 | |
57 | #define printk_dbg(fmt, args...) printk(fmt, args) | |
58 | #else | |
59 | #define printk_dbg(fmt, args...) | |
60 | #endif | |
61 | ||
62 | /* LTTng lockless logging buffer info */ | |
63 | struct ltt_channel_buf_struct { | |
64 | /* First 32 bytes cache-hot cacheline */ | |
65 | local_t offset; /* Current offset in the buffer */ | |
66 | local_t *commit_count; /* Commit count per sub-buffer */ | |
67 | atomic_long_t consumed; /* | |
68 | * Current offset in the buffer | |
69 | * standard atomic access (shared) | |
70 | */ | |
71 | unsigned long last_tsc; /* | |
72 | * Last timestamp written in the buffer. | |
73 | */ | |
74 | /* End of first 32 bytes cacheline */ | |
75 | atomic_long_t active_readers; /* | |
76 | * Active readers count | |
77 | * standard atomic access (shared) | |
78 | */ | |
79 | local_t events_lost; | |
80 | local_t corrupted_subbuffers; | |
81 | spinlock_t full_lock; /* | |
82 | * buffer full condition spinlock, only | |
83 | * for userspace tracing blocking mode | |
84 | * synchronization with reader. | |
85 | */ | |
86 | wait_queue_head_t write_wait; /* | |
87 | * Wait queue for blocking user space | |
88 | * writers | |
89 | */ | |
90 | atomic_t wakeup_readers; /* Boolean : wakeup readers waiting ? */ | |
91 | } ____cacheline_aligned; | |
92 | ||
93 | /* | |
94 | * Last TSC comparison functions. Check if the current TSC overflows | |
95 | * LTT_TSC_BITS bits from the last TSC read. Reads and writes last_tsc | |
96 | * atomically. | |
97 | */ | |
98 | ||
99 | #if (BITS_PER_LONG == 32) | |
100 | static inline void save_last_tsc(struct ltt_channel_buf_struct *ltt_buf, | |
101 | u64 tsc) | |
102 | { | |
103 | ltt_buf->last_tsc = (unsigned long)(tsc >> LTT_TSC_BITS); | |
104 | } | |
105 | ||
106 | static inline int last_tsc_overflow(struct ltt_channel_buf_struct *ltt_buf, | |
107 | u64 tsc) | |
108 | { | |
109 | unsigned long tsc_shifted = (unsigned long)(tsc >> LTT_TSC_BITS); | |
110 | ||
111 | if (unlikely((tsc_shifted - ltt_buf->last_tsc))) | |
112 | return 1; | |
113 | else | |
114 | return 0; | |
115 | } | |
116 | #else | |
117 | static inline void save_last_tsc(struct ltt_channel_buf_struct *ltt_buf, | |
118 | u64 tsc) | |
119 | { | |
120 | ltt_buf->last_tsc = (unsigned long)tsc; | |
121 | } | |
122 | ||
123 | static inline int last_tsc_overflow(struct ltt_channel_buf_struct *ltt_buf, | |
124 | u64 tsc) | |
125 | { | |
126 | if (unlikely((tsc - ltt_buf->last_tsc) >> LTT_TSC_BITS)) | |
127 | return 1; | |
128 | else | |
129 | return 0; | |
130 | } | |
131 | #endif | |
132 | ||
133 | static struct file_operations ltt_file_operations; | |
134 | ||
135 | /* | |
136 | * A switch is done during tracing or as a final flush after tracing (so it | |
137 | * won't write in the new sub-buffer). | |
138 | */ | |
139 | enum force_switch_mode { FORCE_ACTIVE, FORCE_FLUSH }; | |
140 | ||
141 | static int ltt_relay_create_buffer(struct ltt_trace_struct *trace, | |
142 | struct ltt_channel_struct *ltt_chan, | |
143 | struct rchan_buf *buf, | |
144 | unsigned int cpu, | |
145 | unsigned int n_subbufs); | |
146 | ||
147 | static void ltt_relay_destroy_buffer(struct ltt_channel_struct *ltt_chan, | |
148 | unsigned int cpu); | |
149 | ||
150 | static void ltt_force_switch(struct rchan_buf *buf, | |
151 | enum force_switch_mode mode); | |
152 | ||
153 | /* | |
154 | * Trace callbacks | |
155 | */ | |
156 | static void ltt_buffer_begin_callback(struct rchan_buf *buf, | |
157 | u64 tsc, unsigned int subbuf_idx) | |
158 | { | |
159 | struct ltt_channel_struct *channel = | |
160 | (struct ltt_channel_struct *)buf->chan->private_data; | |
161 | struct ltt_subbuffer_header *header = | |
162 | (struct ltt_subbuffer_header *) | |
163 | ltt_relay_offset_address(buf, | |
164 | subbuf_idx * buf->chan->subbuf_size); | |
165 | ||
166 | header->cycle_count_begin = tsc; | |
167 | header->lost_size = 0xFFFFFFFF; /* for debugging */ | |
168 | header->buf_size = buf->chan->subbuf_size; | |
169 | ltt_write_trace_header(channel->trace, header); | |
170 | } | |
171 | ||
172 | /* | |
173 | * offset is assumed to never be 0 here : never deliver a completely empty | |
174 | * subbuffer. The lost size is between 0 and subbuf_size-1. | |
175 | */ | |
176 | static notrace void ltt_buffer_end_callback(struct rchan_buf *buf, | |
177 | u64 tsc, unsigned int offset, unsigned int subbuf_idx) | |
178 | { | |
179 | struct ltt_channel_struct *channel = | |
180 | (struct ltt_channel_struct *)buf->chan->private_data; | |
181 | struct ltt_channel_buf_struct *ltt_buf = | |
182 | percpu_ptr(channel->buf, buf->cpu); | |
183 | struct ltt_subbuffer_header *header = | |
184 | (struct ltt_subbuffer_header *) | |
185 | ltt_relay_offset_address(buf, | |
186 | subbuf_idx * buf->chan->subbuf_size); | |
187 | ||
188 | header->lost_size = SUBBUF_OFFSET((buf->chan->subbuf_size - offset), | |
189 | buf->chan); | |
190 | header->cycle_count_end = tsc; | |
191 | header->events_lost = local_read(<t_buf->events_lost); | |
192 | header->subbuf_corrupt = local_read(<t_buf->corrupted_subbuffers); | |
193 | } | |
194 | ||
195 | static notrace void ltt_deliver(struct rchan_buf *buf, unsigned int subbuf_idx, | |
196 | void *subbuf) | |
197 | { | |
198 | struct ltt_channel_struct *channel = | |
199 | (struct ltt_channel_struct *)buf->chan->private_data; | |
200 | struct ltt_channel_buf_struct *ltt_buf = | |
201 | percpu_ptr(channel->buf, buf->cpu); | |
202 | ||
203 | atomic_set(<t_buf->wakeup_readers, 1); | |
204 | } | |
205 | ||
206 | static struct dentry *ltt_create_buf_file_callback(const char *filename, | |
207 | struct dentry *parent, int mode, | |
208 | struct rchan_buf *buf) | |
209 | { | |
210 | struct ltt_channel_struct *ltt_chan; | |
211 | int err; | |
212 | struct dentry *dentry; | |
213 | ||
214 | ltt_chan = buf->chan->private_data; | |
215 | err = ltt_relay_create_buffer(ltt_chan->trace, ltt_chan, | |
216 | buf, buf->cpu, | |
217 | buf->chan->n_subbufs); | |
218 | if (err) | |
219 | return ERR_PTR(err); | |
220 | ||
221 | dentry = debugfs_create_file(filename, mode, parent, buf, | |
222 | <t_file_operations); | |
223 | if (!dentry) | |
224 | goto error; | |
225 | return dentry; | |
226 | error: | |
227 | ltt_relay_destroy_buffer(ltt_chan, buf->cpu); | |
228 | return NULL; | |
229 | } | |
230 | ||
231 | static int ltt_remove_buf_file_callback(struct dentry *dentry) | |
232 | { | |
233 | struct rchan_buf *buf = dentry->d_inode->i_private; | |
234 | struct ltt_channel_struct *ltt_chan = buf->chan->private_data; | |
235 | ||
236 | debugfs_remove(dentry); | |
237 | ltt_relay_destroy_buffer(ltt_chan, buf->cpu); | |
238 | ||
239 | return 0; | |
240 | } | |
241 | ||
242 | /* | |
243 | * Wake writers : | |
244 | * | |
245 | * This must be done after the trace is removed from the RCU list so that there | |
246 | * are no stalled writers. | |
247 | */ | |
248 | static void ltt_relay_wake_writers(struct ltt_channel_buf_struct *ltt_buf) | |
249 | { | |
250 | ||
251 | if (waitqueue_active(<t_buf->write_wait)) | |
252 | wake_up_interruptible(<t_buf->write_wait); | |
253 | } | |
254 | ||
255 | /* | |
256 | * This function should not be called from NMI interrupt context | |
257 | */ | |
258 | static notrace void ltt_buf_unfull(struct rchan_buf *buf, | |
259 | unsigned int subbuf_idx, | |
260 | long offset) | |
261 | { | |
262 | struct ltt_channel_struct *ltt_channel = | |
263 | (struct ltt_channel_struct *)buf->chan->private_data; | |
264 | struct ltt_channel_buf_struct *ltt_buf = | |
265 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
266 | ||
267 | ltt_relay_wake_writers(ltt_buf); | |
268 | } | |
269 | ||
270 | /** | |
271 | * ltt_open - open file op for ltt files | |
272 | * @inode: opened inode | |
273 | * @file: opened file | |
274 | * | |
275 | * Open implementation. Makes sure only one open instance of a buffer is | |
276 | * done at a given moment. | |
277 | */ | |
278 | static int ltt_open(struct inode *inode, struct file *file) | |
279 | { | |
280 | struct rchan_buf *buf = inode->i_private; | |
281 | struct ltt_channel_struct *ltt_channel = | |
282 | (struct ltt_channel_struct *)buf->chan->private_data; | |
283 | struct ltt_channel_buf_struct *ltt_buf = | |
284 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
285 | ||
286 | if (!atomic_long_add_unless(<t_buf->active_readers, 1, 1)) | |
287 | return -EBUSY; | |
288 | return ltt_relay_file_operations.open(inode, file); | |
289 | } | |
290 | ||
291 | /** | |
292 | * ltt_release - release file op for ltt files | |
293 | * @inode: opened inode | |
294 | * @file: opened file | |
295 | * | |
296 | * Release implementation. | |
297 | */ | |
298 | static int ltt_release(struct inode *inode, struct file *file) | |
299 | { | |
300 | struct rchan_buf *buf = inode->i_private; | |
301 | struct ltt_channel_struct *ltt_channel = | |
302 | (struct ltt_channel_struct *)buf->chan->private_data; | |
303 | struct ltt_channel_buf_struct *ltt_buf = | |
304 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
305 | int ret; | |
306 | ||
307 | WARN_ON(atomic_long_read(<t_buf->active_readers) != 1); | |
308 | atomic_long_dec(<t_buf->active_readers); | |
309 | ret = ltt_relay_file_operations.release(inode, file); | |
310 | WARN_ON(ret); | |
311 | return ret; | |
312 | } | |
313 | ||
314 | /** | |
315 | * ltt_poll - file op for ltt files | |
316 | * @filp: the file | |
317 | * @wait: poll table | |
318 | * | |
319 | * Poll implementation. | |
320 | */ | |
321 | static unsigned int ltt_poll(struct file *filp, poll_table *wait) | |
322 | { | |
323 | unsigned int mask = 0; | |
324 | struct inode *inode = filp->f_dentry->d_inode; | |
325 | struct rchan_buf *buf = inode->i_private; | |
326 | struct ltt_channel_struct *ltt_channel = | |
327 | (struct ltt_channel_struct *)buf->chan->private_data; | |
328 | struct ltt_channel_buf_struct *ltt_buf = | |
329 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
330 | ||
331 | if (filp->f_mode & FMODE_READ) { | |
332 | poll_wait_set_exclusive(wait); | |
333 | poll_wait(filp, &buf->read_wait, wait); | |
334 | ||
335 | WARN_ON(atomic_long_read(<t_buf->active_readers) != 1); | |
336 | if (SUBBUF_TRUNC(local_read(<t_buf->offset), | |
337 | buf->chan) | |
338 | - SUBBUF_TRUNC(atomic_long_read(<t_buf->consumed), | |
339 | buf->chan) | |
340 | == 0) { | |
341 | if (buf->finalized) | |
342 | return POLLHUP; | |
343 | else | |
344 | return 0; | |
345 | } else { | |
346 | struct rchan *rchan = | |
347 | ltt_channel->trans_channel_data; | |
348 | if (SUBBUF_TRUNC(local_read(<t_buf->offset), | |
349 | buf->chan) | |
350 | - SUBBUF_TRUNC(atomic_long_read( | |
351 | <t_buf->consumed), | |
352 | buf->chan) | |
353 | >= rchan->alloc_size) | |
354 | return POLLPRI | POLLRDBAND; | |
355 | else | |
356 | return POLLIN | POLLRDNORM; | |
357 | } | |
358 | } | |
359 | return mask; | |
360 | } | |
361 | ||
362 | static int ltt_do_get_subbuf(struct rchan_buf *buf, struct ltt_channel_buf_struct *ltt_buf, long *pconsumed_old) | |
363 | { | |
364 | long consumed_old, consumed_idx, commit_count, write_offset; | |
365 | consumed_old = atomic_long_read(<t_buf->consumed); | |
366 | consumed_idx = SUBBUF_INDEX(consumed_old, buf->chan); | |
367 | commit_count = local_read(<t_buf->commit_count[consumed_idx]); | |
368 | /* | |
369 | * Make sure we read the commit count before reading the buffer | |
370 | * data and the write offset. Correct consumed offset ordering | |
371 | * wrt commit count is insured by the use of cmpxchg to update | |
372 | * the consumed offset. | |
373 | */ | |
374 | smp_rmb(); | |
375 | write_offset = local_read(<t_buf->offset); | |
376 | /* | |
377 | * Check that the subbuffer we are trying to consume has been | |
378 | * already fully committed. | |
379 | */ | |
380 | if (((commit_count - buf->chan->subbuf_size) | |
381 | & ltt_channel->commit_count_mask) | |
382 | - (BUFFER_TRUNC(consumed_old, buf->chan) | |
383 | >> ltt_channel->n_subbufs_order) | |
384 | != 0) { | |
385 | return -EAGAIN; | |
386 | } | |
387 | /* | |
388 | * Check that we are not about to read the same subbuffer in | |
389 | * which the writer head is. | |
390 | */ | |
391 | if ((SUBBUF_TRUNC(write_offset, buf->chan) | |
392 | - SUBBUF_TRUNC(consumed_old, buf->chan)) | |
393 | == 0) { | |
394 | return -EAGAIN; | |
395 | } | |
396 | ||
397 | *pconsumed_old = consumed_old; | |
398 | return 0; | |
399 | } | |
400 | ||
401 | static int ltt_do_put_subbuf(struct rchan_buf *buf, struct ltt_channel_buf_struct *ltt_buf, u32 uconsumed_old) | |
402 | { | |
403 | long consumed_new, consumed_old; | |
404 | ||
405 | consumed_old = atomic_long_read(<t_buf->consumed); | |
406 | consumed_old = consumed_old & (~0xFFFFFFFFL); | |
407 | consumed_old = consumed_old | uconsumed_old; | |
408 | consumed_new = SUBBUF_ALIGN(consumed_old, buf->chan); | |
409 | ||
410 | spin_lock(<t_buf->full_lock); | |
411 | if (atomic_long_cmpxchg(<t_buf->consumed, consumed_old, | |
412 | consumed_new) | |
413 | != consumed_old) { | |
414 | /* We have been pushed by the writer : the last | |
415 | * buffer read _is_ corrupted! It can also | |
416 | * happen if this is a buffer we never got. */ | |
417 | spin_unlock(<t_buf->full_lock); | |
418 | return -EIO; | |
419 | } else { | |
420 | /* tell the client that buffer is now unfull */ | |
421 | int index; | |
422 | long data; | |
423 | index = SUBBUF_INDEX(consumed_old, buf->chan); | |
424 | data = BUFFER_OFFSET(consumed_old, buf->chan); | |
425 | ltt_buf_unfull(buf, index, data); | |
426 | spin_unlock(<t_buf->full_lock); | |
427 | } | |
428 | return 0; | |
429 | } | |
430 | ||
431 | /** | |
432 | * ltt_ioctl - control on the debugfs file | |
433 | * | |
434 | * @inode: the inode | |
435 | * @filp: the file | |
436 | * @cmd: the command | |
437 | * @arg: command arg | |
438 | * | |
439 | * This ioctl implements three commands necessary for a minimal | |
440 | * producer/consumer implementation : | |
441 | * RELAY_GET_SUBBUF | |
442 | * Get the next sub buffer that can be read. It never blocks. | |
443 | * RELAY_PUT_SUBBUF | |
444 | * Release the currently read sub-buffer. Parameter is the last | |
445 | * put subbuffer (returned by GET_SUBBUF). | |
446 | * RELAY_GET_N_BUBBUFS | |
447 | * returns the number of sub buffers in the per cpu channel. | |
448 | * RELAY_GET_SUBBUF_SIZE | |
449 | * returns the size of the sub buffers. | |
450 | */ | |
451 | static int ltt_ioctl(struct inode *inode, struct file *filp, | |
452 | unsigned int cmd, unsigned long arg) | |
453 | { | |
454 | struct rchan_buf *buf = inode->i_private; | |
455 | struct ltt_channel_struct *ltt_channel = | |
456 | (struct ltt_channel_struct *)buf->chan->private_data; | |
457 | struct ltt_channel_buf_struct *ltt_buf = | |
458 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
459 | u32 __user *argp = (u32 __user *)arg; | |
460 | ||
461 | WARN_ON(atomic_long_read(<t_buf->active_readers) != 1); | |
462 | switch (cmd) { | |
463 | case RELAY_GET_SUBBUF: | |
464 | { | |
465 | int ret; | |
466 | ret = ltt_do_get_subbuf(buf, ltt_buf, &consumed_old); | |
467 | if(ret < 0) | |
468 | return ret; | |
469 | return put_user((u32)consumed_old, argp); | |
470 | } | |
471 | case RELAY_PUT_SUBBUF: | |
472 | { | |
473 | int ret; | |
474 | u32 uconsumed_old; | |
475 | ret = get_user(uconsumed_old, argp); | |
476 | if (ret) | |
477 | return ret; /* will return -EFAULT */ | |
478 | return ltt_do_put_subbuf(buf, ltt_buf, uconsumed_old); | |
479 | } | |
480 | case RELAY_GET_N_SUBBUFS: | |
481 | return put_user((u32)buf->chan->n_subbufs, argp); | |
482 | break; | |
483 | case RELAY_GET_SUBBUF_SIZE: | |
484 | return put_user((u32)buf->chan->subbuf_size, argp); | |
485 | break; | |
486 | default: | |
487 | return -ENOIOCTLCMD; | |
488 | } | |
489 | return 0; | |
490 | } | |
491 | ||
492 | #ifdef CONFIG_COMPAT | |
493 | static long ltt_compat_ioctl(struct file *file, unsigned int cmd, | |
494 | unsigned long arg) | |
495 | { | |
496 | long ret = -ENOIOCTLCMD; | |
497 | ||
498 | lock_kernel(); | |
499 | ret = ltt_ioctl(file->f_dentry->d_inode, file, cmd, arg); | |
500 | unlock_kernel(); | |
501 | ||
502 | return ret; | |
503 | } | |
504 | #endif | |
505 | ||
506 | static void ltt_relay_pipe_buf_release(struct pipe_inode_info *pipe, | |
507 | struct pipe_buffer *pbuf) | |
508 | { | |
509 | } | |
510 | ||
511 | static struct pipe_buf_operations ltt_relay_pipe_buf_ops = { | |
512 | .can_merge = 0, | |
513 | .map = generic_pipe_buf_map, | |
514 | .unmap = generic_pipe_buf_unmap, | |
515 | .confirm = generic_pipe_buf_confirm, | |
516 | .release = ltt_relay_pipe_buf_release, | |
517 | .steal = generic_pipe_buf_steal, | |
518 | .get = generic_pipe_buf_get, | |
519 | }; | |
520 | ||
521 | static void ltt_relay_page_release(struct splice_pipe_desc *spd, unsigned int i) | |
522 | { | |
523 | } | |
524 | ||
525 | /* | |
526 | * subbuf_splice_actor - splice up to one subbuf's worth of data | |
527 | */ | |
528 | static int subbuf_splice_actor(struct file *in, | |
529 | loff_t *ppos, | |
530 | struct pipe_inode_info *pipe, | |
531 | size_t len, | |
532 | unsigned int flags) | |
533 | { | |
534 | struct rchan_buf *buf = in->private_data; | |
535 | struct ltt_channel_struct *ltt_channel = | |
536 | (struct ltt_channel_struct *)buf->chan->private_data; | |
537 | struct ltt_channel_buf_struct *ltt_buf = | |
538 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
539 | unsigned int poff, subbuf_pages, nr_pages; | |
540 | struct page *pages[PIPE_BUFFERS]; | |
541 | struct partial_page partial[PIPE_BUFFERS]; | |
542 | struct splice_pipe_desc spd = { | |
543 | .pages = pages, | |
544 | .nr_pages = 0, | |
545 | .partial = partial, | |
546 | .flags = flags, | |
547 | .ops = <t_relay_pipe_buf_ops, | |
548 | .spd_release = ltt_relay_page_release, | |
549 | }; | |
550 | long consumed_old, consumed_idx, roffset; | |
551 | unsigned long bytes_avail; | |
552 | ||
553 | /* | |
554 | * Check that a GET_SUBBUF ioctl has been done before. | |
555 | */ | |
556 | WARN_ON(atomic_long_read(<t_buf->active_readers) != 1); | |
557 | consumed_old = atomic_long_read(<t_buf->consumed); | |
558 | consumed_old += *ppos; | |
559 | consumed_idx = SUBBUF_INDEX(consumed_old, buf->chan); | |
560 | ||
561 | /* | |
562 | * Adjust read len, if longer than what is available | |
563 | */ | |
564 | bytes_avail = SUBBUF_TRUNC(local_read(<t_buf->offset), buf->chan) | |
565 | - consumed_old; | |
566 | WARN_ON(bytes_avail > buf->chan->alloc_size); | |
567 | len = min_t(size_t, len, bytes_avail); | |
568 | subbuf_pages = bytes_avail >> PAGE_SHIFT; | |
569 | nr_pages = min_t(unsigned int, subbuf_pages, PIPE_BUFFERS); | |
570 | roffset = consumed_old & PAGE_MASK; | |
571 | poff = consumed_old & ~PAGE_MASK; | |
572 | printk_dbg(KERN_DEBUG "SPLICE actor len %zu pos %zd write_pos %ld\n", | |
573 | len, (ssize_t)*ppos, local_read(<t_buf->offset)); | |
574 | ||
575 | for (; spd.nr_pages < nr_pages; spd.nr_pages++) { | |
576 | unsigned int this_len; | |
577 | struct buf_page *page; | |
578 | ||
579 | if (!len) | |
580 | break; | |
581 | printk_dbg(KERN_DEBUG "SPLICE actor loop len %zu roffset %ld\n", | |
582 | len, roffset); | |
583 | ||
584 | this_len = PAGE_SIZE - poff; | |
585 | page = ltt_relay_read_get_page(buf, roffset); | |
586 | spd.pages[spd.nr_pages] = page->page; | |
587 | spd.partial[spd.nr_pages].offset = poff; | |
588 | spd.partial[spd.nr_pages].len = this_len; | |
589 | ||
590 | poff = 0; | |
591 | roffset += PAGE_SIZE; | |
592 | len -= this_len; | |
593 | } | |
594 | ||
595 | if (!spd.nr_pages) | |
596 | return 0; | |
597 | ||
598 | return splice_to_pipe(pipe, &spd); | |
599 | } | |
600 | ||
601 | static ssize_t ltt_relay_file_splice_read(struct file *in, | |
602 | loff_t *ppos, | |
603 | struct pipe_inode_info *pipe, | |
604 | size_t len, | |
605 | unsigned int flags) | |
606 | { | |
607 | ssize_t spliced; | |
608 | int ret; | |
609 | ||
610 | ret = 0; | |
611 | spliced = 0; | |
612 | ||
613 | printk_dbg(KERN_DEBUG "SPLICE read len %zu pos %zd\n", | |
614 | len, (ssize_t)*ppos); | |
615 | while (len && !spliced) { | |
616 | ret = subbuf_splice_actor(in, ppos, pipe, len, flags); | |
617 | printk_dbg(KERN_DEBUG "SPLICE read loop ret %d\n", ret); | |
618 | if (ret < 0) | |
619 | break; | |
620 | else if (!ret) { | |
621 | if (flags & SPLICE_F_NONBLOCK) | |
622 | ret = -EAGAIN; | |
623 | break; | |
624 | } | |
625 | ||
626 | *ppos += ret; | |
627 | if (ret > len) | |
628 | len = 0; | |
629 | else | |
630 | len -= ret; | |
631 | spliced += ret; | |
632 | } | |
633 | ||
634 | if (spliced) | |
635 | return spliced; | |
636 | ||
637 | return ret; | |
638 | } | |
639 | ||
640 | static void ltt_relay_print_subbuffer_errors( | |
641 | struct ltt_channel_struct *ltt_chan, | |
642 | long cons_off, unsigned int cpu) | |
643 | { | |
644 | struct rchan *rchan = ltt_chan->trans_channel_data; | |
645 | struct ltt_channel_buf_struct *ltt_buf = | |
646 | percpu_ptr(ltt_chan->buf, cpu); | |
647 | long cons_idx, commit_count, write_offset; | |
648 | ||
649 | cons_idx = SUBBUF_INDEX(cons_off, rchan); | |
650 | commit_count = local_read(<t_buf->commit_count[cons_idx]); | |
651 | /* | |
652 | * No need to order commit_count and write_offset reads because we | |
653 | * execute after trace is stopped when there are no readers left. | |
654 | */ | |
655 | write_offset = local_read(<t_buf->offset); | |
656 | printk(KERN_WARNING | |
657 | "LTT : unread channel %s offset is %ld " | |
658 | "and cons_off : %ld (cpu %u)\n", | |
659 | ltt_chan->channel_name, write_offset, cons_off, cpu); | |
660 | /* Check each sub-buffer for non filled commit count */ | |
661 | if (((commit_count - rchan->subbuf_size) & ltt_chan->commit_count_mask) | |
662 | - (BUFFER_TRUNC(cons_off, rchan) >> ltt_chan->n_subbufs_order) | |
663 | != 0) | |
664 | printk(KERN_ALERT | |
665 | "LTT : %s : subbuffer %lu has non filled " | |
666 | "commit count %lu.\n", | |
667 | ltt_chan->channel_name, cons_idx, commit_count); | |
668 | printk(KERN_ALERT "LTT : %s : commit count : %lu, subbuf size %zd\n", | |
669 | ltt_chan->channel_name, commit_count, | |
670 | rchan->subbuf_size); | |
671 | } | |
672 | ||
673 | static void ltt_relay_print_errors(struct ltt_trace_struct *trace, | |
674 | struct ltt_channel_struct *ltt_chan, int cpu) | |
675 | { | |
676 | struct rchan *rchan = ltt_chan->trans_channel_data; | |
677 | struct ltt_channel_buf_struct *ltt_buf = | |
678 | percpu_ptr(ltt_chan->buf, cpu); | |
679 | long cons_off; | |
680 | ||
681 | for (cons_off = atomic_long_read(<t_buf->consumed); | |
682 | (SUBBUF_TRUNC(local_read(<t_buf->offset), | |
683 | rchan) | |
684 | - cons_off) > 0; | |
685 | cons_off = SUBBUF_ALIGN(cons_off, rchan)) | |
686 | ltt_relay_print_subbuffer_errors(ltt_chan, cons_off, cpu); | |
687 | } | |
688 | ||
689 | static void ltt_relay_print_buffer_errors(struct ltt_channel_struct *ltt_chan, | |
690 | unsigned int cpu) | |
691 | { | |
692 | struct ltt_trace_struct *trace = ltt_chan->trace; | |
693 | struct ltt_channel_buf_struct *ltt_buf = | |
694 | percpu_ptr(ltt_chan->buf, cpu); | |
695 | ||
696 | if (local_read(<t_buf->events_lost)) | |
697 | printk(KERN_ALERT | |
698 | "LTT : %s : %ld events lost " | |
699 | "in %s channel (cpu %u).\n", | |
700 | ltt_chan->channel_name, | |
701 | local_read(<t_buf->events_lost), | |
702 | ltt_chan->channel_name, cpu); | |
703 | if (local_read(<t_buf->corrupted_subbuffers)) | |
704 | printk(KERN_ALERT | |
705 | "LTT : %s : %ld corrupted subbuffers " | |
706 | "in %s channel (cpu %u).\n", | |
707 | ltt_chan->channel_name, | |
708 | local_read(<t_buf->corrupted_subbuffers), | |
709 | ltt_chan->channel_name, cpu); | |
710 | ||
711 | ltt_relay_print_errors(trace, ltt_chan, cpu); | |
712 | } | |
713 | ||
714 | static void ltt_relay_remove_dirs(struct ltt_trace_struct *trace) | |
715 | { | |
716 | debugfs_remove(trace->dentry.trace_root); | |
717 | } | |
718 | ||
719 | static void ltt_relay_release_channel(struct kref *kref) | |
720 | { | |
721 | struct ltt_channel_struct *ltt_chan = container_of(kref, | |
722 | struct ltt_channel_struct, kref); | |
723 | percpu_free(ltt_chan->buf); | |
724 | } | |
725 | ||
726 | /* | |
727 | * Create ltt buffer. | |
728 | */ | |
729 | static int ltt_relay_create_buffer(struct ltt_trace_struct *trace, | |
730 | struct ltt_channel_struct *ltt_chan, struct rchan_buf *buf, | |
731 | unsigned int cpu, unsigned int n_subbufs) | |
732 | { | |
733 | struct ltt_channel_buf_struct *ltt_buf = | |
734 | percpu_ptr(ltt_chan->buf, cpu); | |
735 | unsigned int j; | |
736 | ||
737 | ltt_buf->commit_count = | |
738 | kzalloc_node(sizeof(ltt_buf->commit_count) * n_subbufs, | |
739 | GFP_KERNEL, cpu_to_node(cpu)); | |
740 | if (!ltt_buf->commit_count) | |
741 | return -ENOMEM; | |
742 | kref_get(&trace->kref); | |
743 | kref_get(&trace->ltt_transport_kref); | |
744 | kref_get(<t_chan->kref); | |
745 | local_set(<t_buf->offset, ltt_subbuffer_header_size()); | |
746 | atomic_long_set(<t_buf->consumed, 0); | |
747 | atomic_long_set(<t_buf->active_readers, 0); | |
748 | for (j = 0; j < n_subbufs; j++) | |
749 | local_set(<t_buf->commit_count[j], 0); | |
750 | init_waitqueue_head(<t_buf->write_wait); | |
751 | atomic_set(<t_buf->wakeup_readers, 0); | |
752 | spin_lock_init(<t_buf->full_lock); | |
753 | ||
754 | ltt_buffer_begin_callback(buf, trace->start_tsc, 0); | |
755 | /* atomic_add made on local variable on data that belongs to | |
756 | * various CPUs : ok because tracing not started (for this cpu). */ | |
757 | local_add(ltt_subbuffer_header_size(), <t_buf->commit_count[0]); | |
758 | ||
759 | local_set(<t_buf->events_lost, 0); | |
760 | local_set(<t_buf->corrupted_subbuffers, 0); | |
761 | ||
762 | return 0; | |
763 | } | |
764 | ||
765 | static void ltt_relay_destroy_buffer(struct ltt_channel_struct *ltt_chan, | |
766 | unsigned int cpu) | |
767 | { | |
768 | struct ltt_trace_struct *trace = ltt_chan->trace; | |
769 | struct ltt_channel_buf_struct *ltt_buf = | |
770 | percpu_ptr(ltt_chan->buf, cpu); | |
771 | ||
772 | kref_put(<t_chan->trace->ltt_transport_kref, | |
773 | ltt_release_transport); | |
774 | ltt_relay_print_buffer_errors(ltt_chan, cpu); | |
775 | kfree(ltt_buf->commit_count); | |
776 | ltt_buf->commit_count = NULL; | |
777 | kref_put(<t_chan->kref, ltt_relay_release_channel); | |
778 | kref_put(&trace->kref, ltt_release_trace); | |
779 | wake_up_interruptible(&trace->kref_wq); | |
780 | } | |
781 | ||
782 | /* | |
783 | * Create channel. | |
784 | */ | |
785 | static int ltt_relay_create_channel(const char *trace_name, | |
786 | struct ltt_trace_struct *trace, struct dentry *dir, | |
787 | const char *channel_name, struct ltt_channel_struct *ltt_chan, | |
788 | unsigned int subbuf_size, unsigned int n_subbufs, | |
789 | int overwrite) | |
790 | { | |
791 | char *tmpname; | |
792 | unsigned int tmpname_len; | |
793 | int err = 0; | |
794 | ||
795 | tmpname = kmalloc(PATH_MAX, GFP_KERNEL); | |
796 | if (!tmpname) | |
797 | return EPERM; | |
798 | if (overwrite) { | |
799 | strncpy(tmpname, LTT_FLIGHT_PREFIX, PATH_MAX-1); | |
800 | strncat(tmpname, channel_name, | |
801 | PATH_MAX-1-sizeof(LTT_FLIGHT_PREFIX)); | |
802 | } else { | |
803 | strncpy(tmpname, channel_name, PATH_MAX-1); | |
804 | } | |
805 | strncat(tmpname, "_", PATH_MAX-1-strlen(tmpname)); | |
806 | ||
807 | kref_init(<t_chan->kref); | |
808 | ||
809 | ltt_chan->trace = trace; | |
810 | ltt_chan->buffer_begin = ltt_buffer_begin_callback; | |
811 | ltt_chan->buffer_end = ltt_buffer_end_callback; | |
812 | ltt_chan->overwrite = overwrite; | |
813 | ltt_chan->n_subbufs_order = get_count_order(n_subbufs); | |
814 | ltt_chan->commit_count_mask = (~0UL >> ltt_chan->n_subbufs_order); | |
815 | ltt_chan->buf = percpu_alloc_mask(sizeof(struct ltt_channel_buf_struct), | |
816 | GFP_KERNEL, cpu_possible_map); | |
817 | if (!ltt_chan->buf) | |
818 | goto ltt_percpu_alloc_error; | |
819 | ltt_chan->trans_channel_data = ltt_relay_open(tmpname, | |
820 | dir, | |
821 | subbuf_size, | |
822 | n_subbufs, | |
823 | &trace->callbacks, | |
824 | ltt_chan); | |
825 | tmpname_len = strlen(tmpname); | |
826 | if (tmpname_len > 0) { | |
827 | /* Remove final _ for pretty printing */ | |
828 | tmpname[tmpname_len-1] = '\0'; | |
829 | } | |
830 | if (ltt_chan->trans_channel_data == NULL) { | |
831 | printk(KERN_ERR "LTT : Can't open %s channel for trace %s\n", | |
832 | tmpname, trace_name); | |
833 | goto relay_open_error; | |
834 | } | |
835 | ||
836 | err = 0; | |
837 | goto end; | |
838 | ||
839 | relay_open_error: | |
840 | percpu_free(ltt_chan->buf); | |
841 | ltt_percpu_alloc_error: | |
842 | err = EPERM; | |
843 | end: | |
844 | kfree(tmpname); | |
845 | return err; | |
846 | } | |
847 | ||
848 | static int ltt_relay_create_dirs(struct ltt_trace_struct *new_trace) | |
849 | { | |
850 | new_trace->dentry.trace_root = debugfs_create_dir(new_trace->trace_name, | |
851 | get_ltt_root()); | |
852 | if (new_trace->dentry.trace_root == NULL) { | |
853 | printk(KERN_ERR "LTT : Trace directory name %s already taken\n", | |
854 | new_trace->trace_name); | |
855 | return EEXIST; | |
856 | } | |
857 | ||
858 | new_trace->callbacks.create_buf_file = ltt_create_buf_file_callback; | |
859 | new_trace->callbacks.remove_buf_file = ltt_remove_buf_file_callback; | |
860 | ||
861 | return 0; | |
862 | } | |
863 | ||
864 | /* | |
865 | * LTTng channel flush function. | |
866 | * | |
867 | * Must be called when no tracing is active in the channel, because of | |
868 | * accesses across CPUs. | |
869 | */ | |
870 | static notrace void ltt_relay_buffer_flush(struct rchan_buf *buf) | |
871 | { | |
872 | buf->finalized = 1; | |
873 | ltt_force_switch(buf, FORCE_FLUSH); | |
874 | } | |
875 | ||
876 | static void ltt_relay_async_wakeup_chan(struct ltt_channel_struct *ltt_channel) | |
877 | { | |
878 | unsigned int i; | |
879 | struct rchan *rchan = ltt_channel->trans_channel_data; | |
880 | ||
881 | for_each_possible_cpu(i) { | |
882 | struct ltt_channel_buf_struct *ltt_buf = | |
883 | percpu_ptr(ltt_channel->buf, i); | |
884 | ||
885 | if (atomic_read(<t_buf->wakeup_readers) == 1) { | |
886 | atomic_set(<t_buf->wakeup_readers, 0); | |
887 | wake_up_interruptible(&rchan->buf[i]->read_wait); | |
888 | } | |
889 | } | |
890 | } | |
891 | ||
892 | static void ltt_relay_finish_buffer(struct ltt_channel_struct *ltt_channel, | |
893 | unsigned int cpu) | |
894 | { | |
895 | struct rchan *rchan = ltt_channel->trans_channel_data; | |
896 | ||
897 | if (rchan->buf[cpu]) { | |
898 | struct ltt_channel_buf_struct *ltt_buf = | |
899 | percpu_ptr(ltt_channel->buf, cpu); | |
900 | ltt_relay_buffer_flush(rchan->buf[cpu]); | |
901 | ltt_relay_wake_writers(ltt_buf); | |
902 | } | |
903 | } | |
904 | ||
905 | ||
906 | static void ltt_relay_finish_channel(struct ltt_channel_struct *ltt_channel) | |
907 | { | |
908 | unsigned int i; | |
909 | ||
910 | for_each_possible_cpu(i) | |
911 | ltt_relay_finish_buffer(ltt_channel, i); | |
912 | } | |
913 | ||
914 | static void ltt_relay_remove_channel(struct ltt_channel_struct *channel) | |
915 | { | |
916 | struct rchan *rchan = channel->trans_channel_data; | |
917 | ||
918 | ltt_relay_close(rchan); | |
919 | kref_put(&channel->kref, ltt_relay_release_channel); | |
920 | } | |
921 | ||
922 | struct ltt_reserve_switch_offsets { | |
923 | long begin, end, old; | |
924 | long begin_switch, end_switch_current, end_switch_old; | |
925 | long commit_count, reserve_commit_diff; | |
926 | size_t before_hdr_pad, size; | |
927 | }; | |
928 | ||
929 | /* | |
930 | * Returns : | |
931 | * 0 if ok | |
932 | * !0 if execution must be aborted. | |
933 | */ | |
934 | static inline int ltt_relay_try_reserve( | |
935 | struct ltt_channel_struct *ltt_channel, | |
936 | struct ltt_channel_buf_struct *ltt_buf, struct rchan *rchan, | |
937 | struct rchan_buf *buf, | |
938 | struct ltt_reserve_switch_offsets *offsets, size_t data_size, | |
939 | u64 *tsc, unsigned int *rflags, int largest_align) | |
940 | { | |
941 | offsets->begin = local_read(<t_buf->offset); | |
942 | offsets->old = offsets->begin; | |
943 | offsets->begin_switch = 0; | |
944 | offsets->end_switch_current = 0; | |
945 | offsets->end_switch_old = 0; | |
946 | ||
947 | *tsc = trace_clock_read64(); | |
948 | if (last_tsc_overflow(ltt_buf, *tsc)) | |
949 | *rflags = LTT_RFLAG_ID_SIZE_TSC; | |
950 | ||
951 | if (SUBBUF_OFFSET(offsets->begin, buf->chan) == 0) { | |
952 | offsets->begin_switch = 1; /* For offsets->begin */ | |
953 | } else { | |
954 | offsets->size = ltt_get_header_size(ltt_channel, | |
955 | offsets->begin, data_size, | |
956 | &offsets->before_hdr_pad, *rflags); | |
957 | offsets->size += ltt_align(offsets->begin + offsets->size, | |
958 | largest_align) | |
959 | + data_size; | |
960 | if ((SUBBUF_OFFSET(offsets->begin, buf->chan) + offsets->size) | |
961 | > buf->chan->subbuf_size) { | |
962 | offsets->end_switch_old = 1; /* For offsets->old */ | |
963 | offsets->begin_switch = 1; /* For offsets->begin */ | |
964 | } | |
965 | } | |
966 | if (offsets->begin_switch) { | |
967 | long subbuf_index; | |
968 | ||
969 | if (offsets->end_switch_old) | |
970 | offsets->begin = SUBBUF_ALIGN(offsets->begin, | |
971 | buf->chan); | |
972 | offsets->begin = offsets->begin + ltt_subbuffer_header_size(); | |
973 | /* Test new buffer integrity */ | |
974 | subbuf_index = SUBBUF_INDEX(offsets->begin, buf->chan); | |
975 | offsets->reserve_commit_diff = | |
976 | (BUFFER_TRUNC(offsets->begin, buf->chan) | |
977 | >> ltt_channel->n_subbufs_order) | |
978 | - (local_read(<t_buf->commit_count[subbuf_index]) | |
979 | & ltt_channel->commit_count_mask); | |
980 | if (offsets->reserve_commit_diff == 0) { | |
981 | /* Next buffer not corrupted. */ | |
982 | if (!ltt_channel->overwrite && | |
983 | (SUBBUF_TRUNC(offsets->begin, buf->chan) | |
984 | - SUBBUF_TRUNC(atomic_long_read( | |
985 | <t_buf->consumed), | |
986 | buf->chan)) | |
987 | >= rchan->alloc_size) { | |
988 | /* | |
989 | * We do not overwrite non consumed buffers | |
990 | * and we are full : event is lost. | |
991 | */ | |
992 | local_inc(<t_buf->events_lost); | |
993 | return -1; | |
994 | } else { | |
995 | /* | |
996 | * next buffer not corrupted, we are either in | |
997 | * overwrite mode or the buffer is not full. | |
998 | * It's safe to write in this new subbuffer. | |
999 | */ | |
1000 | } | |
1001 | } else { | |
1002 | /* | |
1003 | * Next subbuffer corrupted. Force pushing reader even | |
1004 | * in normal mode. It's safe to write in this new | |
1005 | * subbuffer. | |
1006 | */ | |
1007 | } | |
1008 | offsets->size = ltt_get_header_size(ltt_channel, | |
1009 | offsets->begin, data_size, | |
1010 | &offsets->before_hdr_pad, *rflags); | |
1011 | offsets->size += ltt_align(offsets->begin + offsets->size, | |
1012 | largest_align) | |
1013 | + data_size; | |
1014 | if ((SUBBUF_OFFSET(offsets->begin, buf->chan) + offsets->size) | |
1015 | > buf->chan->subbuf_size) { | |
1016 | /* | |
1017 | * Event too big for subbuffers, report error, don't | |
1018 | * complete the sub-buffer switch. | |
1019 | */ | |
1020 | local_inc(<t_buf->events_lost); | |
1021 | return -1; | |
1022 | } else { | |
1023 | /* | |
1024 | * We just made a successful buffer switch and the event | |
1025 | * fits in the new subbuffer. Let's write. | |
1026 | */ | |
1027 | } | |
1028 | } else { | |
1029 | /* | |
1030 | * Event fits in the current buffer and we are not on a switch | |
1031 | * boundary. It's safe to write. | |
1032 | */ | |
1033 | } | |
1034 | offsets->end = offsets->begin + offsets->size; | |
1035 | ||
1036 | if ((SUBBUF_OFFSET(offsets->end, buf->chan)) == 0) { | |
1037 | /* | |
1038 | * The offset_end will fall at the very beginning of the next | |
1039 | * subbuffer. | |
1040 | */ | |
1041 | offsets->end_switch_current = 1; /* For offsets->begin */ | |
1042 | } | |
1043 | return 0; | |
1044 | } | |
1045 | ||
1046 | /* | |
1047 | * Returns : | |
1048 | * 0 if ok | |
1049 | * !0 if execution must be aborted. | |
1050 | */ | |
1051 | static inline int ltt_relay_try_switch( | |
1052 | enum force_switch_mode mode, | |
1053 | struct ltt_channel_struct *ltt_channel, | |
1054 | struct ltt_channel_buf_struct *ltt_buf, struct rchan *rchan, | |
1055 | struct rchan_buf *buf, | |
1056 | struct ltt_reserve_switch_offsets *offsets, | |
1057 | u64 *tsc) | |
1058 | { | |
1059 | long subbuf_index; | |
1060 | ||
1061 | offsets->begin = local_read(<t_buf->offset); | |
1062 | offsets->old = offsets->begin; | |
1063 | offsets->begin_switch = 0; | |
1064 | offsets->end_switch_old = 0; | |
1065 | ||
1066 | *tsc = trace_clock_read64(); | |
1067 | ||
1068 | if (SUBBUF_OFFSET(offsets->begin, buf->chan) != 0) { | |
1069 | offsets->begin = SUBBUF_ALIGN(offsets->begin, buf->chan); | |
1070 | offsets->end_switch_old = 1; | |
1071 | } else { | |
1072 | /* we do not have to switch : buffer is empty */ | |
1073 | return -1; | |
1074 | } | |
1075 | if (mode == FORCE_ACTIVE) | |
1076 | offsets->begin += ltt_subbuffer_header_size(); | |
1077 | /* | |
1078 | * Always begin_switch in FORCE_ACTIVE mode. | |
1079 | * Test new buffer integrity | |
1080 | */ | |
1081 | subbuf_index = SUBBUF_INDEX(offsets->begin, buf->chan); | |
1082 | offsets->reserve_commit_diff = | |
1083 | (BUFFER_TRUNC(offsets->begin, buf->chan) | |
1084 | >> ltt_channel->n_subbufs_order) | |
1085 | - (local_read(<t_buf->commit_count[subbuf_index]) | |
1086 | & ltt_channel->commit_count_mask); | |
1087 | if (offsets->reserve_commit_diff == 0) { | |
1088 | /* Next buffer not corrupted. */ | |
1089 | if (mode == FORCE_ACTIVE | |
1090 | && !ltt_channel->overwrite | |
1091 | && offsets->begin - atomic_long_read(<t_buf->consumed) | |
1092 | >= rchan->alloc_size) { | |
1093 | /* | |
1094 | * We do not overwrite non consumed buffers and we are | |
1095 | * full : ignore switch while tracing is active. | |
1096 | */ | |
1097 | return -1; | |
1098 | } | |
1099 | } else { | |
1100 | /* | |
1101 | * Next subbuffer corrupted. Force pushing reader even in normal | |
1102 | * mode | |
1103 | */ | |
1104 | } | |
1105 | offsets->end = offsets->begin; | |
1106 | return 0; | |
1107 | } | |
1108 | ||
1109 | static inline void ltt_reserve_push_reader( | |
1110 | struct ltt_channel_struct *ltt_channel, | |
1111 | struct ltt_channel_buf_struct *ltt_buf, | |
1112 | struct rchan *rchan, | |
1113 | struct rchan_buf *buf, | |
1114 | struct ltt_reserve_switch_offsets *offsets) | |
1115 | { | |
1116 | long consumed_old, consumed_new; | |
1117 | ||
1118 | do { | |
1119 | consumed_old = atomic_long_read(<t_buf->consumed); | |
1120 | /* | |
1121 | * If buffer is in overwrite mode, push the reader consumed | |
1122 | * count if the write position has reached it and we are not | |
1123 | * at the first iteration (don't push the reader farther than | |
1124 | * the writer). This operation can be done concurrently by many | |
1125 | * writers in the same buffer, the writer being at the farthest | |
1126 | * write position sub-buffer index in the buffer being the one | |
1127 | * which will win this loop. | |
1128 | * If the buffer is not in overwrite mode, pushing the reader | |
1129 | * only happens if a sub-buffer is corrupted. | |
1130 | */ | |
1131 | if ((SUBBUF_TRUNC(offsets->end-1, buf->chan) | |
1132 | - SUBBUF_TRUNC(consumed_old, buf->chan)) | |
1133 | >= rchan->alloc_size) | |
1134 | consumed_new = SUBBUF_ALIGN(consumed_old, buf->chan); | |
1135 | else { | |
1136 | consumed_new = consumed_old; | |
1137 | break; | |
1138 | } | |
1139 | } while (atomic_long_cmpxchg(<t_buf->consumed, consumed_old, | |
1140 | consumed_new) != consumed_old); | |
1141 | ||
1142 | if (consumed_old != consumed_new) { | |
1143 | /* | |
1144 | * Reader pushed : we are the winner of the push, we can | |
1145 | * therefore reequilibrate reserve and commit. Atomic increment | |
1146 | * of the commit count permits other writers to play around | |
1147 | * with this variable before us. We keep track of | |
1148 | * corrupted_subbuffers even in overwrite mode : | |
1149 | * we never want to write over a non completely committed | |
1150 | * sub-buffer : possible causes : the buffer size is too low | |
1151 | * compared to the unordered data input, or there is a writer | |
1152 | * that died between the reserve and the commit. | |
1153 | */ | |
1154 | if (offsets->reserve_commit_diff) { | |
1155 | /* | |
1156 | * We have to alter the sub-buffer commit count. | |
1157 | * We do not deliver the previous subbuffer, given it | |
1158 | * was either corrupted or not consumed (overwrite | |
1159 | * mode). | |
1160 | */ | |
1161 | local_add(offsets->reserve_commit_diff, | |
1162 | <t_buf->commit_count[ | |
1163 | SUBBUF_INDEX(offsets->begin, | |
1164 | buf->chan)]); | |
1165 | if (!ltt_channel->overwrite | |
1166 | || offsets->reserve_commit_diff | |
1167 | != rchan->subbuf_size) { | |
1168 | /* | |
1169 | * The reserve commit diff was not subbuf_size : | |
1170 | * it means the subbuffer was partly written to | |
1171 | * and is therefore corrupted. If it is multiple | |
1172 | * of subbuffer size and we are in flight | |
1173 | * recorder mode, we are skipping over a whole | |
1174 | * subbuffer. | |
1175 | */ | |
1176 | local_inc(<t_buf->corrupted_subbuffers); | |
1177 | } | |
1178 | } | |
1179 | } | |
1180 | } | |
1181 | ||
1182 | ||
1183 | /* | |
1184 | * ltt_reserve_switch_old_subbuf: switch old subbuffer | |
1185 | * | |
1186 | * Concurrency safe because we are the last and only thread to alter this | |
1187 | * sub-buffer. As long as it is not delivered and read, no other thread can | |
1188 | * alter the offset, alter the reserve_count or call the | |
1189 | * client_buffer_end_callback on this sub-buffer. | |
1190 | * | |
1191 | * The only remaining threads could be the ones with pending commits. They will | |
1192 | * have to do the deliver themselves. Not concurrency safe in overwrite mode. | |
1193 | * We detect corrupted subbuffers with commit and reserve counts. We keep a | |
1194 | * corrupted sub-buffers count and push the readers across these sub-buffers. | |
1195 | * | |
1196 | * Not concurrency safe if a writer is stalled in a subbuffer and another writer | |
1197 | * switches in, finding out it's corrupted. The result will be than the old | |
1198 | * (uncommited) subbuffer will be declared corrupted, and that the new subbuffer | |
1199 | * will be declared corrupted too because of the commit count adjustment. | |
1200 | * | |
1201 | * Note : offset_old should never be 0 here. | |
1202 | */ | |
1203 | static inline void ltt_reserve_switch_old_subbuf( | |
1204 | struct ltt_channel_struct *ltt_channel, | |
1205 | struct ltt_channel_buf_struct *ltt_buf, struct rchan *rchan, | |
1206 | struct rchan_buf *buf, | |
1207 | struct ltt_reserve_switch_offsets *offsets, u64 *tsc) | |
1208 | { | |
1209 | long oldidx = SUBBUF_INDEX(offsets->old - 1, rchan); | |
1210 | ||
1211 | ltt_channel->buffer_end(buf, *tsc, offsets->old, oldidx); | |
1212 | /* Must write buffer end before incrementing commit count */ | |
1213 | smp_wmb(); | |
1214 | offsets->commit_count = | |
1215 | local_add_return(rchan->subbuf_size | |
1216 | - (SUBBUF_OFFSET(offsets->old - 1, rchan) | |
1217 | + 1), | |
1218 | <t_buf->commit_count[oldidx]); | |
1219 | if ((BUFFER_TRUNC(offsets->old - 1, rchan) | |
1220 | >> ltt_channel->n_subbufs_order) | |
1221 | - ((offsets->commit_count - rchan->subbuf_size) | |
1222 | & ltt_channel->commit_count_mask) == 0) | |
1223 | ltt_deliver(buf, oldidx, NULL); | |
1224 | } | |
1225 | ||
1226 | /* | |
1227 | * ltt_reserve_switch_new_subbuf: Populate new subbuffer. | |
1228 | * | |
1229 | * This code can be executed unordered : writers may already have written to the | |
1230 | * sub-buffer before this code gets executed, caution. The commit makes sure | |
1231 | * that this code is executed before the deliver of this sub-buffer. | |
1232 | */ | |
1233 | static inline void ltt_reserve_switch_new_subbuf( | |
1234 | struct ltt_channel_struct *ltt_channel, | |
1235 | struct ltt_channel_buf_struct *ltt_buf, struct rchan *rchan, | |
1236 | struct rchan_buf *buf, | |
1237 | struct ltt_reserve_switch_offsets *offsets, u64 *tsc) | |
1238 | { | |
1239 | long beginidx = SUBBUF_INDEX(offsets->begin, rchan); | |
1240 | ||
1241 | ltt_channel->buffer_begin(buf, *tsc, beginidx); | |
1242 | /* Must write buffer end before incrementing commit count */ | |
1243 | smp_wmb(); | |
1244 | offsets->commit_count = local_add_return(ltt_subbuffer_header_size(), | |
1245 | <t_buf->commit_count[beginidx]); | |
1246 | /* Check if the written buffer has to be delivered */ | |
1247 | if ((BUFFER_TRUNC(offsets->begin, rchan) | |
1248 | >> ltt_channel->n_subbufs_order) | |
1249 | - ((offsets->commit_count - rchan->subbuf_size) | |
1250 | & ltt_channel->commit_count_mask) == 0) | |
1251 | ltt_deliver(buf, beginidx, NULL); | |
1252 | } | |
1253 | ||
1254 | ||
1255 | /* | |
1256 | * ltt_reserve_end_switch_current: finish switching current subbuffer | |
1257 | * | |
1258 | * Concurrency safe because we are the last and only thread to alter this | |
1259 | * sub-buffer. As long as it is not delivered and read, no other thread can | |
1260 | * alter the offset, alter the reserve_count or call the | |
1261 | * client_buffer_end_callback on this sub-buffer. | |
1262 | * | |
1263 | * The only remaining threads could be the ones with pending commits. They will | |
1264 | * have to do the deliver themselves. Not concurrency safe in overwrite mode. | |
1265 | * We detect corrupted subbuffers with commit and reserve counts. We keep a | |
1266 | * corrupted sub-buffers count and push the readers across these sub-buffers. | |
1267 | * | |
1268 | * Not concurrency safe if a writer is stalled in a subbuffer and another writer | |
1269 | * switches in, finding out it's corrupted. The result will be than the old | |
1270 | * (uncommited) subbuffer will be declared corrupted, and that the new subbuffer | |
1271 | * will be declared corrupted too because of the commit count adjustment. | |
1272 | */ | |
1273 | static inline void ltt_reserve_end_switch_current( | |
1274 | struct ltt_channel_struct *ltt_channel, | |
1275 | struct ltt_channel_buf_struct *ltt_buf, struct rchan *rchan, | |
1276 | struct rchan_buf *buf, | |
1277 | struct ltt_reserve_switch_offsets *offsets, u64 *tsc) | |
1278 | { | |
1279 | long endidx = SUBBUF_INDEX(offsets->end - 1, rchan); | |
1280 | ||
1281 | ltt_channel->buffer_end(buf, *tsc, offsets->end, endidx); | |
1282 | /* Must write buffer begin before incrementing commit count */ | |
1283 | smp_wmb(); | |
1284 | offsets->commit_count = | |
1285 | local_add_return(rchan->subbuf_size | |
1286 | - (SUBBUF_OFFSET(offsets->end - 1, rchan) | |
1287 | + 1), | |
1288 | <t_buf->commit_count[endidx]); | |
1289 | if ((BUFFER_TRUNC(offsets->end - 1, rchan) | |
1290 | >> ltt_channel->n_subbufs_order) | |
1291 | - ((offsets->commit_count - rchan->subbuf_size) | |
1292 | & ltt_channel->commit_count_mask) == 0) | |
1293 | ltt_deliver(buf, endidx, NULL); | |
1294 | } | |
1295 | ||
1296 | /** | |
1297 | * ltt_relay_reserve_slot - Atomic slot reservation in a LTTng buffer. | |
1298 | * @trace: the trace structure to log to. | |
1299 | * @ltt_channel: channel structure | |
1300 | * @transport_data: data structure specific to ltt relay | |
1301 | * @data_size: size of the variable length data to log. | |
1302 | * @slot_size: pointer to total size of the slot (out) | |
1303 | * @buf_offset : pointer to reserved buffer offset (out) | |
1304 | * @tsc: pointer to the tsc at the slot reservation (out) | |
1305 | * @cpu: cpuid | |
1306 | * | |
1307 | * Return : -ENOSPC if not enough space, else returns 0. | |
1308 | * It will take care of sub-buffer switching. | |
1309 | */ | |
1310 | static notrace int ltt_relay_reserve_slot(struct ltt_trace_struct *trace, | |
1311 | struct ltt_channel_struct *ltt_channel, void **transport_data, | |
1312 | size_t data_size, size_t *slot_size, long *buf_offset, u64 *tsc, | |
1313 | unsigned int *rflags, int largest_align, int cpu) | |
1314 | { | |
1315 | struct rchan *rchan = ltt_channel->trans_channel_data; | |
1316 | struct rchan_buf *buf = *transport_data = | |
1317 | rchan->buf[cpu]; | |
1318 | struct ltt_channel_buf_struct *ltt_buf = | |
1319 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
1320 | struct ltt_reserve_switch_offsets offsets; | |
1321 | ||
1322 | offsets.reserve_commit_diff = 0; | |
1323 | offsets.size = 0; | |
1324 | ||
1325 | /* | |
1326 | * Perform retryable operations. | |
1327 | */ | |
1328 | if (__get_cpu_var(ltt_nesting) > 4) { | |
1329 | local_inc(<t_buf->events_lost); | |
1330 | return -EPERM; | |
1331 | } | |
1332 | do { | |
1333 | if (ltt_relay_try_reserve(ltt_channel, ltt_buf, | |
1334 | rchan, buf, &offsets, data_size, tsc, rflags, | |
1335 | largest_align)) | |
1336 | return -ENOSPC; | |
1337 | } while (local_cmpxchg(<t_buf->offset, offsets.old, | |
1338 | offsets.end) != offsets.old); | |
1339 | ||
1340 | /* | |
1341 | * Atomically update last_tsc. This update races against concurrent | |
1342 | * atomic updates, but the race will always cause supplementary full TSC | |
1343 | * events, never the opposite (missing a full TSC event when it would be | |
1344 | * needed). | |
1345 | */ | |
1346 | save_last_tsc(ltt_buf, *tsc); | |
1347 | ||
1348 | /* | |
1349 | * Push the reader if necessary | |
1350 | */ | |
1351 | ltt_reserve_push_reader(ltt_channel, ltt_buf, rchan, buf, &offsets); | |
1352 | ||
1353 | /* | |
1354 | * Switch old subbuffer if needed. | |
1355 | */ | |
1356 | if (offsets.end_switch_old) | |
1357 | ltt_reserve_switch_old_subbuf(ltt_channel, ltt_buf, rchan, buf, | |
1358 | &offsets, tsc); | |
1359 | ||
1360 | /* | |
1361 | * Populate new subbuffer. | |
1362 | */ | |
1363 | if (offsets.begin_switch) | |
1364 | ltt_reserve_switch_new_subbuf(ltt_channel, ltt_buf, rchan, | |
1365 | buf, &offsets, tsc); | |
1366 | ||
1367 | if (offsets.end_switch_current) | |
1368 | ltt_reserve_end_switch_current(ltt_channel, ltt_buf, rchan, | |
1369 | buf, &offsets, tsc); | |
1370 | ||
1371 | *slot_size = offsets.size; | |
1372 | *buf_offset = offsets.begin + offsets.before_hdr_pad; | |
1373 | return 0; | |
1374 | } | |
1375 | ||
1376 | /* | |
1377 | * Force a sub-buffer switch for a per-cpu buffer. This operation is | |
1378 | * completely reentrant : can be called while tracing is active with | |
1379 | * absolutely no lock held. | |
1380 | * | |
1381 | * Note, however, that as a local_cmpxchg is used for some atomic | |
1382 | * operations, this function must be called from the CPU which owns the buffer | |
1383 | * for a ACTIVE flush. | |
1384 | */ | |
1385 | static notrace void ltt_force_switch(struct rchan_buf *buf, | |
1386 | enum force_switch_mode mode) | |
1387 | { | |
1388 | struct ltt_channel_struct *ltt_channel = | |
1389 | (struct ltt_channel_struct *)buf->chan->private_data; | |
1390 | struct ltt_channel_buf_struct *ltt_buf = | |
1391 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
1392 | struct rchan *rchan = ltt_channel->trans_channel_data; | |
1393 | struct ltt_reserve_switch_offsets offsets; | |
1394 | u64 tsc; | |
1395 | ||
1396 | offsets.reserve_commit_diff = 0; | |
1397 | offsets.size = 0; | |
1398 | ||
1399 | /* | |
1400 | * Perform retryable operations. | |
1401 | */ | |
1402 | do { | |
1403 | if (ltt_relay_try_switch(mode, ltt_channel, ltt_buf, | |
1404 | rchan, buf, &offsets, &tsc)) | |
1405 | return; | |
1406 | } while (local_cmpxchg(<t_buf->offset, offsets.old, | |
1407 | offsets.end) != offsets.old); | |
1408 | ||
1409 | /* | |
1410 | * Atomically update last_tsc. This update races against concurrent | |
1411 | * atomic updates, but the race will always cause supplementary full TSC | |
1412 | * events, never the opposite (missing a full TSC event when it would be | |
1413 | * needed). | |
1414 | */ | |
1415 | save_last_tsc(ltt_buf, tsc); | |
1416 | ||
1417 | /* | |
1418 | * Push the reader if necessary | |
1419 | */ | |
1420 | if (mode == FORCE_ACTIVE) | |
1421 | ltt_reserve_push_reader(ltt_channel, ltt_buf, rchan, | |
1422 | buf, &offsets); | |
1423 | ||
1424 | /* | |
1425 | * Switch old subbuffer if needed. | |
1426 | */ | |
1427 | if (offsets.end_switch_old) | |
1428 | ltt_reserve_switch_old_subbuf(ltt_channel, ltt_buf, rchan, buf, | |
1429 | &offsets, &tsc); | |
1430 | ||
1431 | /* | |
1432 | * Populate new subbuffer. | |
1433 | */ | |
1434 | if (mode == FORCE_ACTIVE) | |
1435 | ltt_reserve_switch_new_subbuf(ltt_channel, | |
1436 | ltt_buf, rchan, buf, &offsets, &tsc); | |
1437 | } | |
1438 | ||
1439 | /* | |
1440 | * for flight recording. must be called after relay_commit. | |
1441 | * This function decrements de subbuffer's lost_size each time the commit count | |
1442 | * reaches back the reserve offset (module subbuffer size). It is useful for | |
1443 | * crash dump. | |
1444 | * We use slot_size - 1 to make sure we deal correctly with the case where we | |
1445 | * fill the subbuffer completely (so the subbuf index stays in the previous | |
1446 | * subbuffer). | |
1447 | */ | |
1448 | #ifdef CONFIG_LTT_VMCORE | |
1449 | static inline void ltt_write_commit_counter(struct rchan_buf *buf, | |
1450 | long buf_offset, size_t slot_size) | |
1451 | { | |
1452 | struct ltt_channel_struct *ltt_channel = | |
1453 | (struct ltt_channel_struct *)buf->chan->private_data; | |
1454 | struct ltt_channel_buf_struct *ltt_buf = | |
1455 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
1456 | struct ltt_subbuffer_header *header; | |
1457 | long offset, subbuf_idx, commit_count; | |
1458 | uint32_t lost_old, lost_new; | |
1459 | ||
1460 | subbuf_idx = SUBBUF_INDEX(buf_offset - 1, buf->chan); | |
1461 | offset = buf_offset + slot_size; | |
1462 | header = (struct ltt_subbuffer_header *) | |
1463 | ltt_relay_offset_address(buf, | |
1464 | subbuf_idx * buf->chan->subbuf_size); | |
1465 | for (;;) { | |
1466 | lost_old = header->lost_size; | |
1467 | commit_count = | |
1468 | local_read(<t_buf->commit_count[subbuf_idx]); | |
1469 | /* SUBBUF_OFFSET includes commit_count_mask */ | |
1470 | if (!SUBBUF_OFFSET(offset - commit_count, buf->chan)) { | |
1471 | lost_new = (uint32_t)buf->chan->subbuf_size | |
1472 | - SUBBUF_OFFSET(commit_count, buf->chan); | |
1473 | lost_old = cmpxchg_local(&header->lost_size, lost_old, | |
1474 | lost_new); | |
1475 | if (lost_old <= lost_new) | |
1476 | break; | |
1477 | } else { | |
1478 | break; | |
1479 | } | |
1480 | } | |
1481 | } | |
1482 | #else | |
1483 | static inline void ltt_write_commit_counter(struct rchan_buf *buf, | |
1484 | long buf_offset, size_t slot_size) | |
1485 | { | |
1486 | } | |
1487 | #endif | |
1488 | ||
1489 | /* | |
1490 | * Atomic unordered slot commit. Increments the commit count in the | |
1491 | * specified sub-buffer, and delivers it if necessary. | |
1492 | * | |
1493 | * Parameters: | |
1494 | * | |
1495 | * @ltt_channel : channel structure | |
1496 | * @transport_data: transport-specific data | |
1497 | * @buf_offset : offset following the event header. | |
1498 | * @slot_size : size of the reserved slot. | |
1499 | */ | |
1500 | static notrace void ltt_relay_commit_slot( | |
1501 | struct ltt_channel_struct *ltt_channel, | |
1502 | void **transport_data, long buf_offset, size_t slot_size) | |
1503 | { | |
1504 | struct rchan_buf *buf = *transport_data; | |
1505 | struct ltt_channel_buf_struct *ltt_buf = | |
1506 | percpu_ptr(ltt_channel->buf, buf->cpu); | |
1507 | struct rchan *rchan = buf->chan; | |
1508 | long offset_end = buf_offset; | |
1509 | long endidx = SUBBUF_INDEX(offset_end - 1, rchan); | |
1510 | long commit_count; | |
1511 | ||
1512 | /* Must write slot data before incrementing commit count */ | |
1513 | smp_wmb(); | |
1514 | commit_count = local_add_return(slot_size, | |
1515 | <t_buf->commit_count[endidx]); | |
1516 | /* Check if all commits have been done */ | |
1517 | if ((BUFFER_TRUNC(offset_end - 1, rchan) | |
1518 | >> ltt_channel->n_subbufs_order) | |
1519 | - ((commit_count - rchan->subbuf_size) | |
1520 | & ltt_channel->commit_count_mask) == 0) | |
1521 | ltt_deliver(buf, endidx, NULL); | |
1522 | /* | |
1523 | * Update lost_size for each commit. It's needed only for extracting | |
1524 | * ltt buffers from vmcore, after crash. | |
1525 | */ | |
1526 | ltt_write_commit_counter(buf, buf_offset, slot_size); | |
1527 | } | |
1528 | ||
1529 | /* | |
1530 | * This is called with preemption disabled when user space has requested | |
1531 | * blocking mode. If one of the active traces has free space below a | |
1532 | * specific threshold value, we reenable preemption and block. | |
1533 | */ | |
1534 | static int ltt_relay_user_blocking(struct ltt_trace_struct *trace, | |
1535 | unsigned int chan_index, size_t data_size, | |
1536 | struct user_dbg_data *dbg) | |
1537 | { | |
1538 | struct rchan *rchan; | |
1539 | struct ltt_channel_buf_struct *ltt_buf; | |
1540 | struct ltt_channel_struct *channel; | |
1541 | struct rchan_buf *relay_buf; | |
1542 | int cpu; | |
1543 | DECLARE_WAITQUEUE(wait, current); | |
1544 | ||
1545 | channel = &trace->channels[chan_index]; | |
1546 | rchan = channel->trans_channel_data; | |
1547 | cpu = smp_processor_id(); | |
1548 | relay_buf = rchan->buf[cpu]; | |
1549 | ltt_buf = percpu_ptr(channel->buf, cpu); | |
1550 | ||
1551 | /* | |
1552 | * Check if data is too big for the channel : do not | |
1553 | * block for it. | |
1554 | */ | |
1555 | if (LTT_RESERVE_CRITICAL + data_size > relay_buf->chan->subbuf_size) | |
1556 | return 0; | |
1557 | ||
1558 | /* | |
1559 | * If free space too low, we block. We restart from the | |
1560 | * beginning after we resume (cpu id may have changed | |
1561 | * while preemption is active). | |
1562 | */ | |
1563 | spin_lock(<t_buf->full_lock); | |
1564 | if (!channel->overwrite) { | |
1565 | dbg->write = local_read(<t_buf->offset); | |
1566 | dbg->read = atomic_long_read(<t_buf->consumed); | |
1567 | dbg->avail_size = dbg->write + LTT_RESERVE_CRITICAL + data_size | |
1568 | - SUBBUF_TRUNC(dbg->read, | |
1569 | relay_buf->chan); | |
1570 | if (dbg->avail_size > rchan->alloc_size) { | |
1571 | __set_current_state(TASK_INTERRUPTIBLE); | |
1572 | add_wait_queue(<t_buf->write_wait, &wait); | |
1573 | spin_unlock(<t_buf->full_lock); | |
1574 | preempt_enable(); | |
1575 | schedule(); | |
1576 | __set_current_state(TASK_RUNNING); | |
1577 | remove_wait_queue(<t_buf->write_wait, &wait); | |
1578 | if (signal_pending(current)) | |
1579 | return -ERESTARTSYS; | |
1580 | preempt_disable(); | |
1581 | return 1; | |
1582 | } | |
1583 | } | |
1584 | spin_unlock(<t_buf->full_lock); | |
1585 | return 0; | |
1586 | } | |
1587 | ||
1588 | static void ltt_relay_print_user_errors(struct ltt_trace_struct *trace, | |
1589 | unsigned int chan_index, size_t data_size, | |
1590 | struct user_dbg_data *dbg, int cpu) | |
1591 | { | |
1592 | struct rchan *rchan; | |
1593 | struct ltt_channel_buf_struct *ltt_buf; | |
1594 | struct ltt_channel_struct *channel; | |
1595 | struct rchan_buf *relay_buf; | |
1596 | ||
1597 | channel = &trace->channels[chan_index]; | |
1598 | rchan = channel->trans_channel_data; | |
1599 | relay_buf = rchan->buf[cpu]; | |
1600 | ltt_buf = percpu_ptr(channel->buf, cpu); | |
1601 | ||
1602 | printk(KERN_ERR "Error in LTT usertrace : " | |
1603 | "buffer full : event lost in blocking " | |
1604 | "mode. Increase LTT_RESERVE_CRITICAL.\n"); | |
1605 | printk(KERN_ERR "LTT nesting level is %u.\n", | |
1606 | per_cpu(ltt_nesting, cpu)); | |
1607 | printk(KERN_ERR "LTT avail size %lu.\n", | |
1608 | dbg->avail_size); | |
1609 | printk(KERN_ERR "avai write : %lu, read : %lu\n", | |
1610 | dbg->write, dbg->read); | |
1611 | ||
1612 | dbg->write = local_read(<t_buf->offset); | |
1613 | dbg->read = atomic_long_read(<t_buf->consumed); | |
1614 | ||
1615 | printk(KERN_ERR "LTT cur size %lu.\n", | |
1616 | dbg->write + LTT_RESERVE_CRITICAL + data_size | |
1617 | - SUBBUF_TRUNC(dbg->read, relay_buf->chan)); | |
1618 | printk(KERN_ERR "cur write : %lu, read : %lu\n", | |
1619 | dbg->write, dbg->read); | |
1620 | } | |
1621 | ||
1622 | static struct ltt_transport ltt_relay_transport = { | |
1623 | .name = "relay", | |
1624 | .owner = THIS_MODULE, | |
1625 | .ops = { | |
1626 | .create_dirs = ltt_relay_create_dirs, | |
1627 | .remove_dirs = ltt_relay_remove_dirs, | |
1628 | .create_channel = ltt_relay_create_channel, | |
1629 | .finish_channel = ltt_relay_finish_channel, | |
1630 | .remove_channel = ltt_relay_remove_channel, | |
1631 | .wakeup_channel = ltt_relay_async_wakeup_chan, | |
1632 | .commit_slot = ltt_relay_commit_slot, | |
1633 | .reserve_slot = ltt_relay_reserve_slot, | |
1634 | .user_blocking = ltt_relay_user_blocking, | |
1635 | .user_errors = ltt_relay_print_user_errors, | |
1636 | }, | |
1637 | }; | |
1638 | ||
1639 | static int __init ltt_relay_init(void) | |
1640 | { | |
1641 | printk(KERN_INFO "LTT : ltt-relay init\n"); | |
1642 | ||
1643 | ltt_file_operations = ltt_relay_file_operations; | |
1644 | ltt_file_operations.owner = THIS_MODULE; | |
1645 | ltt_file_operations.open = ltt_open; | |
1646 | ltt_file_operations.release = ltt_release; | |
1647 | ltt_file_operations.poll = ltt_poll; | |
1648 | ltt_file_operations.splice_read = ltt_relay_file_splice_read, | |
1649 | ltt_file_operations.ioctl = ltt_ioctl; | |
1650 | #ifdef CONFIG_COMPAT | |
1651 | ltt_file_operations.compat_ioctl = ltt_compat_ioctl; | |
1652 | #endif | |
1653 | ||
1654 | ltt_transport_register(<t_relay_transport); | |
1655 | ||
1656 | return 0; | |
1657 | } | |
1658 | ||
1659 | static void __exit ltt_relay_exit(void) | |
1660 | { | |
1661 | printk(KERN_INFO "LTT : ltt-relay exit\n"); | |
1662 | ||
1663 | ltt_transport_unregister(<t_relay_transport); | |
1664 | } | |
1665 | ||
1666 | module_init(ltt_relay_init); | |
1667 | module_exit(ltt_relay_exit); | |
1668 | ||
1669 | MODULE_LICENSE("GPL"); | |
1670 | MODULE_AUTHOR("Mathieu Desnoyers"); | |
1671 | MODULE_DESCRIPTION("Linux Trace Toolkit Next Generation Lockless Relay"); |