]> bbs.cooldavid.org Git - net-next-2.6.git/blame - kernel/relay.c
[PATCH] Use activate_mm() in fs/aio.c:use_mm()
[net-next-2.6.git] / kernel / relay.c
CommitLineData
b86ff981
JA
1/*
2 * Public API and common code for kernel->userspace relay file support.
3 *
4 * See Documentation/filesystems/relayfs.txt for an overview of relayfs.
5 *
6 * Copyright (C) 2002-2005 - Tom Zanussi (zanussi@us.ibm.com), IBM Corp
7 * Copyright (C) 1999-2005 - Karim Yaghmour (karim@opersys.com)
8 *
9 * Moved to kernel/relay.c by Paul Mundt, 2006.
10 *
11 * This file is released under the GPL.
12 */
13#include <linux/errno.h>
14#include <linux/stddef.h>
15#include <linux/slab.h>
16#include <linux/module.h>
17#include <linux/string.h>
18#include <linux/relay.h>
19#include <linux/vmalloc.h>
20#include <linux/mm.h>
21
22/*
23 * close() vm_op implementation for relay file mapping.
24 */
25static void relay_file_mmap_close(struct vm_area_struct *vma)
26{
27 struct rchan_buf *buf = vma->vm_private_data;
28 buf->chan->cb->buf_unmapped(buf, vma->vm_file);
29}
30
31/*
32 * nopage() vm_op implementation for relay file mapping.
33 */
34static struct page *relay_buf_nopage(struct vm_area_struct *vma,
35 unsigned long address,
36 int *type)
37{
38 struct page *page;
39 struct rchan_buf *buf = vma->vm_private_data;
40 unsigned long offset = address - vma->vm_start;
41
42 if (address > vma->vm_end)
43 return NOPAGE_SIGBUS; /* Disallow mremap */
44 if (!buf)
45 return NOPAGE_OOM;
46
47 page = vmalloc_to_page(buf->start + offset);
48 if (!page)
49 return NOPAGE_OOM;
50 get_page(page);
51
52 if (type)
53 *type = VM_FAULT_MINOR;
54
55 return page;
56}
57
58/*
59 * vm_ops for relay file mappings.
60 */
61static struct vm_operations_struct relay_file_mmap_ops = {
62 .nopage = relay_buf_nopage,
63 .close = relay_file_mmap_close,
64};
65
66/**
67 * relay_mmap_buf: - mmap channel buffer to process address space
68 * @buf: relay channel buffer
69 * @vma: vm_area_struct describing memory to be mapped
70 *
71 * Returns 0 if ok, negative on error
72 *
73 * Caller should already have grabbed mmap_sem.
74 */
75int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma)
76{
77 unsigned long length = vma->vm_end - vma->vm_start;
78 struct file *filp = vma->vm_file;
79
80 if (!buf)
81 return -EBADF;
82
83 if (length != (unsigned long)buf->chan->alloc_size)
84 return -EINVAL;
85
86 vma->vm_ops = &relay_file_mmap_ops;
87 vma->vm_private_data = buf;
88 buf->chan->cb->buf_mapped(buf, filp);
89
90 return 0;
91}
92
93/**
94 * relay_alloc_buf - allocate a channel buffer
95 * @buf: the buffer struct
96 * @size: total size of the buffer
97 *
4c78a663 98 * Returns a pointer to the resulting buffer, %NULL if unsuccessful. The
221415d7 99 * passed in size will get page aligned, if it isn't already.
b86ff981 100 */
221415d7 101static void *relay_alloc_buf(struct rchan_buf *buf, size_t *size)
b86ff981
JA
102{
103 void *mem;
104 unsigned int i, j, n_pages;
105
221415d7
JA
106 *size = PAGE_ALIGN(*size);
107 n_pages = *size >> PAGE_SHIFT;
b86ff981
JA
108
109 buf->page_array = kcalloc(n_pages, sizeof(struct page *), GFP_KERNEL);
110 if (!buf->page_array)
111 return NULL;
112
113 for (i = 0; i < n_pages; i++) {
114 buf->page_array[i] = alloc_page(GFP_KERNEL);
115 if (unlikely(!buf->page_array[i]))
116 goto depopulate;
117 }
118 mem = vmap(buf->page_array, n_pages, VM_MAP, PAGE_KERNEL);
119 if (!mem)
120 goto depopulate;
121
221415d7 122 memset(mem, 0, *size);
b86ff981
JA
123 buf->page_count = n_pages;
124 return mem;
125
126depopulate:
127 for (j = 0; j < i; j++)
128 __free_page(buf->page_array[j]);
129 kfree(buf->page_array);
130 return NULL;
131}
132
133/**
134 * relay_create_buf - allocate and initialize a channel buffer
4c78a663 135 * @chan: the relay channel
b86ff981 136 *
4c78a663 137 * Returns channel buffer if successful, %NULL otherwise.
b86ff981
JA
138 */
139struct rchan_buf *relay_create_buf(struct rchan *chan)
140{
141 struct rchan_buf *buf = kcalloc(1, sizeof(struct rchan_buf), GFP_KERNEL);
142 if (!buf)
143 return NULL;
144
145 buf->padding = kmalloc(chan->n_subbufs * sizeof(size_t *), GFP_KERNEL);
146 if (!buf->padding)
147 goto free_buf;
148
221415d7 149 buf->start = relay_alloc_buf(buf, &chan->alloc_size);
b86ff981
JA
150 if (!buf->start)
151 goto free_buf;
152
153 buf->chan = chan;
154 kref_get(&buf->chan->kref);
155 return buf;
156
157free_buf:
158 kfree(buf->padding);
159 kfree(buf);
160 return NULL;
161}
162
163/**
164 * relay_destroy_channel - free the channel struct
4c78a663 165 * @kref: target kernel reference that contains the relay channel
b86ff981
JA
166 *
167 * Should only be called from kref_put().
168 */
169void relay_destroy_channel(struct kref *kref)
170{
171 struct rchan *chan = container_of(kref, struct rchan, kref);
172 kfree(chan);
173}
174
175/**
176 * relay_destroy_buf - destroy an rchan_buf struct and associated buffer
177 * @buf: the buffer struct
178 */
179void relay_destroy_buf(struct rchan_buf *buf)
180{
181 struct rchan *chan = buf->chan;
182 unsigned int i;
183
184 if (likely(buf->start)) {
185 vunmap(buf->start);
186 for (i = 0; i < buf->page_count; i++)
187 __free_page(buf->page_array[i]);
188 kfree(buf->page_array);
189 }
190 kfree(buf->padding);
191 kfree(buf);
192 kref_put(&chan->kref, relay_destroy_channel);
193}
194
195/**
196 * relay_remove_buf - remove a channel buffer
4c78a663 197 * @kref: target kernel reference that contains the relay buffer
b86ff981
JA
198 *
199 * Removes the file from the fileystem, which also frees the
200 * rchan_buf_struct and the channel buffer. Should only be called from
201 * kref_put().
202 */
203void relay_remove_buf(struct kref *kref)
204{
205 struct rchan_buf *buf = container_of(kref, struct rchan_buf, kref);
206 buf->chan->cb->remove_buf_file(buf->dentry);
207 relay_destroy_buf(buf);
208}
209
210/**
211 * relay_buf_empty - boolean, is the channel buffer empty?
212 * @buf: channel buffer
213 *
214 * Returns 1 if the buffer is empty, 0 otherwise.
215 */
216int relay_buf_empty(struct rchan_buf *buf)
217{
218 return (buf->subbufs_produced - buf->subbufs_consumed) ? 0 : 1;
219}
220EXPORT_SYMBOL_GPL(relay_buf_empty);
221
222/**
223 * relay_buf_full - boolean, is the channel buffer full?
224 * @buf: channel buffer
225 *
226 * Returns 1 if the buffer is full, 0 otherwise.
227 */
228int relay_buf_full(struct rchan_buf *buf)
229{
230 size_t ready = buf->subbufs_produced - buf->subbufs_consumed;
231 return (ready >= buf->chan->n_subbufs) ? 1 : 0;
232}
233EXPORT_SYMBOL_GPL(relay_buf_full);
234
235/*
236 * High-level relay kernel API and associated functions.
237 */
238
239/*
240 * rchan_callback implementations defining default channel behavior. Used
241 * in place of corresponding NULL values in client callback struct.
242 */
243
244/*
245 * subbuf_start() default callback. Does nothing.
246 */
247static int subbuf_start_default_callback (struct rchan_buf *buf,
248 void *subbuf,
249 void *prev_subbuf,
250 size_t prev_padding)
251{
252 if (relay_buf_full(buf))
253 return 0;
254
255 return 1;
256}
257
258/*
259 * buf_mapped() default callback. Does nothing.
260 */
261static void buf_mapped_default_callback(struct rchan_buf *buf,
262 struct file *filp)
263{
264}
265
266/*
267 * buf_unmapped() default callback. Does nothing.
268 */
269static void buf_unmapped_default_callback(struct rchan_buf *buf,
270 struct file *filp)
271{
272}
273
274/*
275 * create_buf_file_create() default callback. Does nothing.
276 */
277static struct dentry *create_buf_file_default_callback(const char *filename,
278 struct dentry *parent,
279 int mode,
280 struct rchan_buf *buf,
281 int *is_global)
282{
283 return NULL;
284}
285
286/*
287 * remove_buf_file() default callback. Does nothing.
288 */
289static int remove_buf_file_default_callback(struct dentry *dentry)
290{
291 return -EINVAL;
292}
293
294/* relay channel default callbacks */
295static struct rchan_callbacks default_channel_callbacks = {
296 .subbuf_start = subbuf_start_default_callback,
297 .buf_mapped = buf_mapped_default_callback,
298 .buf_unmapped = buf_unmapped_default_callback,
299 .create_buf_file = create_buf_file_default_callback,
300 .remove_buf_file = remove_buf_file_default_callback,
301};
302
303/**
304 * wakeup_readers - wake up readers waiting on a channel
305 * @private: the channel buffer
306 *
307 * This is the work function used to defer reader waking. The
308 * reason waking is deferred is that calling directly from write
309 * causes problems if you're writing from say the scheduler.
310 */
c4028958 311static void wakeup_readers(struct work_struct *work)
b86ff981 312{
c4028958
DH
313 struct rchan_buf *buf =
314 container_of(work, struct rchan_buf, wake_readers.work);
b86ff981
JA
315 wake_up_interruptible(&buf->read_wait);
316}
317
318/**
319 * __relay_reset - reset a channel buffer
320 * @buf: the channel buffer
321 * @init: 1 if this is a first-time initialization
322 *
323 * See relay_reset for description of effect.
324 */
325static inline void __relay_reset(struct rchan_buf *buf, unsigned int init)
326{
327 size_t i;
328
329 if (init) {
330 init_waitqueue_head(&buf->read_wait);
331 kref_init(&buf->kref);
c4028958 332 INIT_DELAYED_WORK(&buf->wake_readers, NULL);
b86ff981
JA
333 } else {
334 cancel_delayed_work(&buf->wake_readers);
335 flush_scheduled_work();
336 }
337
338 buf->subbufs_produced = 0;
339 buf->subbufs_consumed = 0;
340 buf->bytes_consumed = 0;
341 buf->finalized = 0;
342 buf->data = buf->start;
343 buf->offset = 0;
344
345 for (i = 0; i < buf->chan->n_subbufs; i++)
346 buf->padding[i] = 0;
347
348 buf->chan->cb->subbuf_start(buf, buf->data, NULL, 0);
349}
350
351/**
352 * relay_reset - reset the channel
353 * @chan: the channel
354 *
355 * This has the effect of erasing all data from all channel buffers
356 * and restarting the channel in its initial state. The buffers
357 * are not freed, so any mappings are still in effect.
358 *
359 * NOTE: Care should be taken that the channel isn't actually
360 * being used by anything when this call is made.
361 */
362void relay_reset(struct rchan *chan)
363{
364 unsigned int i;
365 struct rchan_buf *prev = NULL;
366
367 if (!chan)
368 return;
369
370 for (i = 0; i < NR_CPUS; i++) {
371 if (!chan->buf[i] || chan->buf[i] == prev)
372 break;
373 __relay_reset(chan->buf[i], 0);
374 prev = chan->buf[i];
375 }
376}
377EXPORT_SYMBOL_GPL(relay_reset);
378
4c78a663 379/*
b86ff981
JA
380 * relay_open_buf - create a new relay channel buffer
381 *
382 * Internal - used by relay_open().
383 */
384static struct rchan_buf *relay_open_buf(struct rchan *chan,
385 const char *filename,
386 struct dentry *parent,
387 int *is_global)
388{
389 struct rchan_buf *buf;
390 struct dentry *dentry;
391
392 if (*is_global)
393 return chan->buf[0];
394
395 buf = relay_create_buf(chan);
396 if (!buf)
397 return NULL;
398
399 /* Create file in fs */
400 dentry = chan->cb->create_buf_file(filename, parent, S_IRUSR,
401 buf, is_global);
402 if (!dentry) {
403 relay_destroy_buf(buf);
404 return NULL;
405 }
406
407 buf->dentry = dentry;
408 __relay_reset(buf, 1);
409
410 return buf;
411}
412
413/**
414 * relay_close_buf - close a channel buffer
415 * @buf: channel buffer
416 *
417 * Marks the buffer finalized and restores the default callbacks.
418 * The channel buffer and channel buffer data structure are then freed
419 * automatically when the last reference is given up.
420 */
421static inline void relay_close_buf(struct rchan_buf *buf)
422{
423 buf->finalized = 1;
424 cancel_delayed_work(&buf->wake_readers);
425 flush_scheduled_work();
426 kref_put(&buf->kref, relay_remove_buf);
427}
428
429static inline void setup_callbacks(struct rchan *chan,
430 struct rchan_callbacks *cb)
431{
432 if (!cb) {
433 chan->cb = &default_channel_callbacks;
434 return;
435 }
436
437 if (!cb->subbuf_start)
438 cb->subbuf_start = subbuf_start_default_callback;
439 if (!cb->buf_mapped)
440 cb->buf_mapped = buf_mapped_default_callback;
441 if (!cb->buf_unmapped)
442 cb->buf_unmapped = buf_unmapped_default_callback;
443 if (!cb->create_buf_file)
444 cb->create_buf_file = create_buf_file_default_callback;
445 if (!cb->remove_buf_file)
446 cb->remove_buf_file = remove_buf_file_default_callback;
447 chan->cb = cb;
448}
449
450/**
451 * relay_open - create a new relay channel
452 * @base_filename: base name of files to create
4c78a663 453 * @parent: dentry of parent directory, %NULL for root directory
b86ff981
JA
454 * @subbuf_size: size of sub-buffers
455 * @n_subbufs: number of sub-buffers
456 * @cb: client callback functions
457 *
4c78a663 458 * Returns channel pointer if successful, %NULL otherwise.
b86ff981
JA
459 *
460 * Creates a channel buffer for each cpu using the sizes and
461 * attributes specified. The created channel buffer files
462 * will be named base_filename0...base_filenameN-1. File
463 * permissions will be S_IRUSR.
464 */
465struct rchan *relay_open(const char *base_filename,
466 struct dentry *parent,
467 size_t subbuf_size,
468 size_t n_subbufs,
469 struct rchan_callbacks *cb)
470{
471 unsigned int i;
472 struct rchan *chan;
473 char *tmpname;
474 int is_global = 0;
475
476 if (!base_filename)
477 return NULL;
478
479 if (!(subbuf_size && n_subbufs))
480 return NULL;
481
482 chan = kcalloc(1, sizeof(struct rchan), GFP_KERNEL);
483 if (!chan)
484 return NULL;
485
486 chan->version = RELAYFS_CHANNEL_VERSION;
487 chan->n_subbufs = n_subbufs;
488 chan->subbuf_size = subbuf_size;
489 chan->alloc_size = FIX_SIZE(subbuf_size * n_subbufs);
490 setup_callbacks(chan, cb);
491 kref_init(&chan->kref);
492
493 tmpname = kmalloc(NAME_MAX + 1, GFP_KERNEL);
494 if (!tmpname)
495 goto free_chan;
496
497 for_each_online_cpu(i) {
498 sprintf(tmpname, "%s%d", base_filename, i);
499 chan->buf[i] = relay_open_buf(chan, tmpname, parent,
500 &is_global);
501 if (!chan->buf[i])
502 goto free_bufs;
503
504 chan->buf[i]->cpu = i;
505 }
506
507 kfree(tmpname);
508 return chan;
509
510free_bufs:
511 for (i = 0; i < NR_CPUS; i++) {
512 if (!chan->buf[i])
513 break;
514 relay_close_buf(chan->buf[i]);
515 if (is_global)
516 break;
517 }
518 kfree(tmpname);
519
520free_chan:
521 kref_put(&chan->kref, relay_destroy_channel);
522 return NULL;
523}
524EXPORT_SYMBOL_GPL(relay_open);
525
526/**
527 * relay_switch_subbuf - switch to a new sub-buffer
528 * @buf: channel buffer
529 * @length: size of current event
530 *
531 * Returns either the length passed in or 0 if full.
532 *
533 * Performs sub-buffer-switch tasks such as invoking callbacks,
534 * updating padding counts, waking up readers, etc.
535 */
536size_t relay_switch_subbuf(struct rchan_buf *buf, size_t length)
537{
538 void *old, *new;
539 size_t old_subbuf, new_subbuf;
540
541 if (unlikely(length > buf->chan->subbuf_size))
542 goto toobig;
543
544 if (buf->offset != buf->chan->subbuf_size + 1) {
545 buf->prev_padding = buf->chan->subbuf_size - buf->offset;
546 old_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
547 buf->padding[old_subbuf] = buf->prev_padding;
548 buf->subbufs_produced++;
221415d7
JA
549 buf->dentry->d_inode->i_size += buf->chan->subbuf_size -
550 buf->padding[old_subbuf];
551 smp_mb();
b86ff981 552 if (waitqueue_active(&buf->read_wait)) {
c4028958
DH
553 PREPARE_DELAYED_WORK(&buf->wake_readers,
554 wakeup_readers);
b86ff981
JA
555 schedule_delayed_work(&buf->wake_readers, 1);
556 }
557 }
558
559 old = buf->data;
560 new_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
561 new = buf->start + new_subbuf * buf->chan->subbuf_size;
562 buf->offset = 0;
563 if (!buf->chan->cb->subbuf_start(buf, new, old, buf->prev_padding)) {
564 buf->offset = buf->chan->subbuf_size + 1;
565 return 0;
566 }
567 buf->data = new;
568 buf->padding[new_subbuf] = 0;
569
570 if (unlikely(length + buf->offset > buf->chan->subbuf_size))
571 goto toobig;
572
573 return length;
574
575toobig:
576 buf->chan->last_toobig = length;
577 return 0;
578}
579EXPORT_SYMBOL_GPL(relay_switch_subbuf);
580
581/**
582 * relay_subbufs_consumed - update the buffer's sub-buffers-consumed count
583 * @chan: the channel
584 * @cpu: the cpu associated with the channel buffer to update
585 * @subbufs_consumed: number of sub-buffers to add to current buf's count
586 *
587 * Adds to the channel buffer's consumed sub-buffer count.
588 * subbufs_consumed should be the number of sub-buffers newly consumed,
589 * not the total consumed.
590 *
4c78a663 591 * NOTE: Kernel clients don't need to call this function if the channel
b86ff981
JA
592 * mode is 'overwrite'.
593 */
594void relay_subbufs_consumed(struct rchan *chan,
595 unsigned int cpu,
596 size_t subbufs_consumed)
597{
598 struct rchan_buf *buf;
599
600 if (!chan)
601 return;
602
603 if (cpu >= NR_CPUS || !chan->buf[cpu])
604 return;
605
606 buf = chan->buf[cpu];
607 buf->subbufs_consumed += subbufs_consumed;
608 if (buf->subbufs_consumed > buf->subbufs_produced)
609 buf->subbufs_consumed = buf->subbufs_produced;
610}
611EXPORT_SYMBOL_GPL(relay_subbufs_consumed);
612
613/**
614 * relay_close - close the channel
615 * @chan: the channel
616 *
617 * Closes all channel buffers and frees the channel.
618 */
619void relay_close(struct rchan *chan)
620{
621 unsigned int i;
622 struct rchan_buf *prev = NULL;
623
624 if (!chan)
625 return;
626
627 for (i = 0; i < NR_CPUS; i++) {
628 if (!chan->buf[i] || chan->buf[i] == prev)
629 break;
630 relay_close_buf(chan->buf[i]);
631 prev = chan->buf[i];
632 }
633
634 if (chan->last_toobig)
635 printk(KERN_WARNING "relay: one or more items not logged "
636 "[item size (%Zd) > sub-buffer size (%Zd)]\n",
637 chan->last_toobig, chan->subbuf_size);
638
639 kref_put(&chan->kref, relay_destroy_channel);
640}
641EXPORT_SYMBOL_GPL(relay_close);
642
643/**
644 * relay_flush - close the channel
645 * @chan: the channel
646 *
4c78a663 647 * Flushes all channel buffers, i.e. forces buffer switch.
b86ff981
JA
648 */
649void relay_flush(struct rchan *chan)
650{
651 unsigned int i;
652 struct rchan_buf *prev = NULL;
653
654 if (!chan)
655 return;
656
657 for (i = 0; i < NR_CPUS; i++) {
658 if (!chan->buf[i] || chan->buf[i] == prev)
659 break;
660 relay_switch_subbuf(chan->buf[i], 0);
661 prev = chan->buf[i];
662 }
663}
664EXPORT_SYMBOL_GPL(relay_flush);
665
666/**
667 * relay_file_open - open file op for relay files
668 * @inode: the inode
669 * @filp: the file
670 *
671 * Increments the channel buffer refcount.
672 */
673static int relay_file_open(struct inode *inode, struct file *filp)
674{
8e18e294 675 struct rchan_buf *buf = inode->i_private;
b86ff981
JA
676 kref_get(&buf->kref);
677 filp->private_data = buf;
678
679 return 0;
680}
681
682/**
683 * relay_file_mmap - mmap file op for relay files
684 * @filp: the file
685 * @vma: the vma describing what to map
686 *
687 * Calls upon relay_mmap_buf to map the file into user space.
688 */
689static int relay_file_mmap(struct file *filp, struct vm_area_struct *vma)
690{
691 struct rchan_buf *buf = filp->private_data;
692 return relay_mmap_buf(buf, vma);
693}
694
695/**
696 * relay_file_poll - poll file op for relay files
697 * @filp: the file
698 * @wait: poll table
699 *
700 * Poll implemention.
701 */
702static unsigned int relay_file_poll(struct file *filp, poll_table *wait)
703{
704 unsigned int mask = 0;
705 struct rchan_buf *buf = filp->private_data;
706
707 if (buf->finalized)
708 return POLLERR;
709
710 if (filp->f_mode & FMODE_READ) {
711 poll_wait(filp, &buf->read_wait, wait);
712 if (!relay_buf_empty(buf))
713 mask |= POLLIN | POLLRDNORM;
714 }
715
716 return mask;
717}
718
719/**
720 * relay_file_release - release file op for relay files
721 * @inode: the inode
722 * @filp: the file
723 *
724 * Decrements the channel refcount, as the filesystem is
725 * no longer using it.
726 */
727static int relay_file_release(struct inode *inode, struct file *filp)
728{
729 struct rchan_buf *buf = filp->private_data;
730 kref_put(&buf->kref, relay_remove_buf);
731
732 return 0;
733}
734
4c78a663 735/*
b86ff981
JA
736 * relay_file_read_consume - update the consumed count for the buffer
737 */
738static void relay_file_read_consume(struct rchan_buf *buf,
739 size_t read_pos,
740 size_t bytes_consumed)
741{
742 size_t subbuf_size = buf->chan->subbuf_size;
743 size_t n_subbufs = buf->chan->n_subbufs;
744 size_t read_subbuf;
745
746 if (buf->bytes_consumed + bytes_consumed > subbuf_size) {
747 relay_subbufs_consumed(buf->chan, buf->cpu, 1);
748 buf->bytes_consumed = 0;
749 }
750
751 buf->bytes_consumed += bytes_consumed;
752 read_subbuf = read_pos / buf->chan->subbuf_size;
753 if (buf->bytes_consumed + buf->padding[read_subbuf] == subbuf_size) {
754 if ((read_subbuf == buf->subbufs_produced % n_subbufs) &&
755 (buf->offset == subbuf_size))
756 return;
757 relay_subbufs_consumed(buf->chan, buf->cpu, 1);
758 buf->bytes_consumed = 0;
759 }
760}
761
4c78a663 762/*
b86ff981
JA
763 * relay_file_read_avail - boolean, are there unconsumed bytes available?
764 */
765static int relay_file_read_avail(struct rchan_buf *buf, size_t read_pos)
766{
b86ff981
JA
767 size_t subbuf_size = buf->chan->subbuf_size;
768 size_t n_subbufs = buf->chan->n_subbufs;
221415d7
JA
769 size_t produced = buf->subbufs_produced;
770 size_t consumed = buf->subbufs_consumed;
b86ff981 771
221415d7 772 relay_file_read_consume(buf, read_pos, 0);
b86ff981 773
221415d7
JA
774 if (unlikely(buf->offset > subbuf_size)) {
775 if (produced == consumed)
776 return 0;
777 return 1;
b86ff981
JA
778 }
779
221415d7
JA
780 if (unlikely(produced - consumed >= n_subbufs)) {
781 consumed = (produced / n_subbufs) * n_subbufs;
782 buf->subbufs_consumed = consumed;
783 }
784
785 produced = (produced % n_subbufs) * subbuf_size + buf->offset;
786 consumed = (consumed % n_subbufs) * subbuf_size + buf->bytes_consumed;
787
788 if (consumed > produced)
789 produced += n_subbufs * subbuf_size;
790
791 if (consumed == produced)
b86ff981
JA
792 return 0;
793
b86ff981
JA
794 return 1;
795}
796
797/**
798 * relay_file_read_subbuf_avail - return bytes available in sub-buffer
4c78a663
RD
799 * @read_pos: file read position
800 * @buf: relay channel buffer
b86ff981
JA
801 */
802static size_t relay_file_read_subbuf_avail(size_t read_pos,
803 struct rchan_buf *buf)
804{
805 size_t padding, avail = 0;
806 size_t read_subbuf, read_offset, write_subbuf, write_offset;
807 size_t subbuf_size = buf->chan->subbuf_size;
808
809 write_subbuf = (buf->data - buf->start) / subbuf_size;
810 write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
811 read_subbuf = read_pos / subbuf_size;
812 read_offset = read_pos % subbuf_size;
813 padding = buf->padding[read_subbuf];
814
815 if (read_subbuf == write_subbuf) {
816 if (read_offset + padding < write_offset)
817 avail = write_offset - (read_offset + padding);
818 } else
819 avail = (subbuf_size - padding) - read_offset;
820
821 return avail;
822}
823
824/**
825 * relay_file_read_start_pos - find the first available byte to read
4c78a663
RD
826 * @read_pos: file read position
827 * @buf: relay channel buffer
b86ff981
JA
828 *
829 * If the read_pos is in the middle of padding, return the
830 * position of the first actually available byte, otherwise
831 * return the original value.
832 */
833static size_t relay_file_read_start_pos(size_t read_pos,
834 struct rchan_buf *buf)
835{
836 size_t read_subbuf, padding, padding_start, padding_end;
837 size_t subbuf_size = buf->chan->subbuf_size;
838 size_t n_subbufs = buf->chan->n_subbufs;
839
840 read_subbuf = read_pos / subbuf_size;
841 padding = buf->padding[read_subbuf];
842 padding_start = (read_subbuf + 1) * subbuf_size - padding;
843 padding_end = (read_subbuf + 1) * subbuf_size;
844 if (read_pos >= padding_start && read_pos < padding_end) {
845 read_subbuf = (read_subbuf + 1) % n_subbufs;
846 read_pos = read_subbuf * subbuf_size;
847 }
848
849 return read_pos;
850}
851
852/**
853 * relay_file_read_end_pos - return the new read position
4c78a663
RD
854 * @read_pos: file read position
855 * @buf: relay channel buffer
856 * @count: number of bytes to be read
b86ff981
JA
857 */
858static size_t relay_file_read_end_pos(struct rchan_buf *buf,
859 size_t read_pos,
860 size_t count)
861{
862 size_t read_subbuf, padding, end_pos;
863 size_t subbuf_size = buf->chan->subbuf_size;
864 size_t n_subbufs = buf->chan->n_subbufs;
865
866 read_subbuf = read_pos / subbuf_size;
867 padding = buf->padding[read_subbuf];
868 if (read_pos % subbuf_size + count + padding == subbuf_size)
869 end_pos = (read_subbuf + 1) * subbuf_size;
870 else
871 end_pos = read_pos + count;
872 if (end_pos >= subbuf_size * n_subbufs)
873 end_pos = 0;
874
875 return end_pos;
876}
877
4c78a663 878/*
6dac40a7 879 * subbuf_read_actor - read up to one subbuf's worth of data
b86ff981 880 */
6dac40a7
TZ
881static int subbuf_read_actor(size_t read_start,
882 struct rchan_buf *buf,
883 size_t avail,
884 read_descriptor_t *desc,
885 read_actor_t actor)
b86ff981 886{
b86ff981 887 void *from;
6dac40a7 888 int ret = 0;
b86ff981
JA
889
890 from = buf->start + read_start;
6dac40a7 891 ret = avail;
ba2397ef 892 if (copy_to_user(desc->arg.buf, from, avail)) {
6dac40a7
TZ
893 desc->error = -EFAULT;
894 ret = 0;
b86ff981 895 }
6dac40a7
TZ
896 desc->arg.data += ret;
897 desc->written += ret;
898 desc->count -= ret;
899
b86ff981
JA
900 return ret;
901}
902
4c78a663 903/*
6dac40a7
TZ
904 * subbuf_send_actor - send up to one subbuf's worth of data
905 */
906static int subbuf_send_actor(size_t read_start,
907 struct rchan_buf *buf,
908 size_t avail,
909 read_descriptor_t *desc,
910 read_actor_t actor)
221415d7 911{
221415d7
JA
912 unsigned long pidx, poff;
913 unsigned int subbuf_pages;
6dac40a7 914 int ret = 0;
221415d7
JA
915
916 subbuf_pages = buf->chan->alloc_size >> PAGE_SHIFT;
917 pidx = (read_start / PAGE_SIZE) % subbuf_pages;
918 poff = read_start & ~PAGE_MASK;
6dac40a7 919 while (avail) {
221415d7
JA
920 struct page *p = buf->page_array[pidx];
921 unsigned int len;
922
923 len = PAGE_SIZE - poff;
6dac40a7
TZ
924 if (len > avail)
925 len = avail;
221415d7 926
6dac40a7
TZ
927 len = actor(desc, p, poff, len);
928 if (desc->error)
221415d7 929 break;
221415d7 930
6dac40a7 931 avail -= len;
221415d7
JA
932 ret += len;
933 poff = 0;
934 pidx = (pidx + 1) % subbuf_pages;
935 }
936
221415d7
JA
937 return ret;
938}
939
6dac40a7
TZ
940typedef int (*subbuf_actor_t) (size_t read_start,
941 struct rchan_buf *buf,
942 size_t avail,
943 read_descriptor_t *desc,
944 read_actor_t actor);
945
4c78a663 946/*
6dac40a7
TZ
947 * relay_file_read_subbufs - read count bytes, bridging subbuf boundaries
948 */
949static inline ssize_t relay_file_read_subbufs(struct file *filp,
950 loff_t *ppos,
6dac40a7
TZ
951 subbuf_actor_t subbuf_actor,
952 read_actor_t actor,
ba2397ef 953 read_descriptor_t *desc)
221415d7 954{
6dac40a7
TZ
955 struct rchan_buf *buf = filp->private_data;
956 size_t read_start, avail;
6dac40a7 957 int ret;
221415d7 958
ba2397ef 959 if (!desc->count)
221415d7
JA
960 return 0;
961
f3a43f3f 962 mutex_lock(&filp->f_path.dentry->d_inode->i_mutex);
221415d7 963 do {
6dac40a7
TZ
964 if (!relay_file_read_avail(buf, *ppos))
965 break;
966
967 read_start = relay_file_read_start_pos(*ppos, buf);
968 avail = relay_file_read_subbuf_avail(read_start, buf);
969 if (!avail)
221415d7 970 break;
221415d7 971
ba2397ef
AV
972 avail = min(desc->count, avail);
973 ret = subbuf_actor(read_start, buf, avail, desc, actor);
974 if (desc->error < 0)
6dac40a7
TZ
975 break;
976
977 if (ret) {
978 relay_file_read_consume(buf, read_start, ret);
979 *ppos = relay_file_read_end_pos(buf, read_start, ret);
980 }
ba2397ef 981 } while (desc->count && ret);
f3a43f3f 982 mutex_unlock(&filp->f_path.dentry->d_inode->i_mutex);
6dac40a7 983
ba2397ef 984 return desc->written;
6dac40a7
TZ
985}
986
987static ssize_t relay_file_read(struct file *filp,
988 char __user *buffer,
989 size_t count,
990 loff_t *ppos)
991{
ba2397ef
AV
992 read_descriptor_t desc;
993 desc.written = 0;
994 desc.count = count;
995 desc.arg.buf = buffer;
996 desc.error = 0;
997 return relay_file_read_subbufs(filp, ppos, subbuf_read_actor,
998 NULL, &desc);
6dac40a7
TZ
999}
1000
1001static ssize_t relay_file_sendfile(struct file *filp,
1002 loff_t *ppos,
1003 size_t count,
1004 read_actor_t actor,
1005 void *target)
1006{
ba2397ef
AV
1007 read_descriptor_t desc;
1008 desc.written = 0;
1009 desc.count = count;
1010 desc.arg.data = target;
1011 desc.error = 0;
1012 return relay_file_read_subbufs(filp, ppos, subbuf_send_actor,
1013 actor, &desc);
221415d7
JA
1014}
1015
15ad7cdc 1016const struct file_operations relay_file_operations = {
b86ff981
JA
1017 .open = relay_file_open,
1018 .poll = relay_file_poll,
1019 .mmap = relay_file_mmap,
1020 .read = relay_file_read,
1021 .llseek = no_llseek,
1022 .release = relay_file_release,
221415d7 1023 .sendfile = relay_file_sendfile,
b86ff981
JA
1024};
1025EXPORT_SYMBOL_GPL(relay_file_operations);