1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
|
/*
* Copyright (C) 2003 Sistina Software
* Copyright (C) 2006 Red Hat GmbH
*
* This file is released under the GPL.
*/
#include "dm-io.h"
#include <linux/bio.h>
#include <linux/mempool.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/slab.h>
struct dm_io_client {
mempool_t *pool;
struct bio_set *bios;
};
/* FIXME: can we shrink this ? */
struct io {
unsigned long error;
atomic_t count;
struct task_struct *sleeper;
struct dm_io_client *client;
io_notify_fn callback;
void *context;
};
/*
* io contexts are only dynamically allocated for asynchronous
* io. Since async io is likely to be the majority of io we'll
* have the same number of io contexts as bios! (FIXME: must reduce this).
*/
static unsigned int pages_to_ios(unsigned int pages)
{
return 4 * pages; /* too many ? */
}
/*
* Create a client with mempool and bioset.
*/
struct dm_io_client *dm_io_client_create(unsigned num_pages)
{
unsigned ios = pages_to_ios(num_pages);
struct dm_io_client *client;
client = kmalloc(sizeof(*client), GFP_KERNEL);
if (!client)
return ERR_PTR(-ENOMEM);
client->pool = mempool_create_kmalloc_pool(ios, sizeof(struct io));
if (!client->pool)
goto bad;
client->bios = bioset_create(16, 16);
if (!client->bios)
goto bad;
return client;
bad:
if (client->pool)
mempool_destroy(client->pool);
kfree(client);
return ERR_PTR(-ENOMEM);
}
EXPORT_SYMBOL(dm_io_client_create);
int dm_io_client_resize(unsigned num_pages, struct dm_io_client *client)
{
return mempool_resize(client->pool, pages_to_ios(num_pages),
GFP_KERNEL);
}
EXPORT_SYMBOL(dm_io_client_resize);
void dm_io_client_destroy(struct dm_io_client *client)
{
mempool_destroy(client->pool);
bioset_free(client->bios);
kfree(client);
}
EXPORT_SYMBOL(dm_io_client_destroy);
/*-----------------------------------------------------------------
* We need to keep track of which region a bio is doing io for.
* In order to save a memory allocation we store this the last
* bvec which we know is unused (blech).
* XXX This is ugly and can OOPS with some configs... find another way.
*---------------------------------------------------------------*/
static inline void bio_set_region(struct bio *bio, unsigned region)
{
bio->bi_io_vec[bio->bi_max_vecs].bv_len = region;
}
static inline unsigned bio_get_region(struct bio *bio)
{
return bio->bi_io_vec[bio->bi_max_vecs].bv_len;
}
/*-----------------------------------------------------------------
* We need an io object to keep track of the number of bios that
* have been dispatched for a particular io.
*---------------------------------------------------------------*/
static void dec_count(struct io *io, unsigned int region, int error)
{
if (error)
set_bit(region, &io->error);
if (atomic_dec_and_test(&io->count)) {
if (io->sleeper)
wake_up_process(io->sleeper);
else {
unsigned long r = io->error;
io_notify_fn fn = io->callback;
void *context = io->context;
mempool_free(io, io->client->pool);
fn(r, context);
}
}
}
static void endio(struct bio *bio, int error)
{
struct io *io;
unsigned region;
if (error && bio_data_dir(bio) == READ)
zero_fill_bio(bio);
/*
* The bio destructor in bio_put() may use the io object.
*/
io = bio->bi_private;
region = bio_get_region(bio);
bio->bi_max_vecs++;
bio_put(bio);
dec_count(io, region, error);
}
/*-----------------------------------------------------------------
* These little objects provide an abstraction for getting a new
* destination page for io.
*---------------------------------------------------------------*/
struct dpages {
void (*get_page)(struct dpages *dp,
struct page **p, unsigned long *len, unsigned *offset);
void (*next_page)(struct dpages *dp);
unsigned context_u;
void *context_ptr;
};
/*
* Functions for getting the pages from a list.
*/
static void list_get_page(struct dpages *dp,
struct page **p, unsigned long *len, unsigned *offset)
{
unsigned o = dp->context_u;
struct page_list *pl = (struct page_list *) dp->context_ptr;
*p = pl->page;
*len = PAGE_SIZE - o;
*offset = o;
}
static void list_next_page(struct dpages *dp)
{
struct page_list *pl = (struct page_list *) dp->context_ptr;
dp->context_ptr = pl->next;
dp->context_u = 0;
}
static void list_dp_init(struct dpages *dp, struct page_list *pl, unsigned offset)
{
dp->get_page = list_get_page;
dp->next_page = list_next_page;
dp->context_u = offset;
dp->context_ptr = pl;
}
/*
* Functions for getting the pages from a bvec.
*/
static void bvec_get_page(struct dpages *dp,
struct page **p, unsigned long *len, unsigned *offset)
{
struct bio_vec *bvec = (struct bio_vec *) dp->context_ptr;
*p = bvec->bv_page;
*len = bvec->bv_len;
*offset = bvec->bv_offset;
}
static void bvec_next_page(struct dpages *dp)
{
struct bio_vec *bvec = (struct bio_vec *) dp->context_ptr;
dp->context_ptr = bvec + 1;
}
static void bvec_dp_init(struct dpages *dp, struct bio_vec *bvec)
{
dp->get_page = bvec_get_page;
dp->next_page = bvec_next_page;
dp->context_ptr = bvec;
}
/*
* Functions for getting the pages from a VMA.
*/
static void vm_get_page(struct dpages *dp,
struct page **p, unsigned long *len, unsigned *offset)
{
*p = vmalloc_to_page(dp->context_ptr);
*offset = dp->context_u;
*len = PAGE_SIZE - dp->context_u;
}
static void vm_next_page(struct dpages *dp)
{
dp->context_ptr += PAGE_SIZE - dp->context_u;
dp->context_u = 0;
}
static void vm_dp_init(struct dpages *dp, void *data)
{
dp->get_page = vm_get_page;
dp->next_page = vm_next_page;
dp->context_u = ((unsigned long) data) & (PAGE_SIZE - 1);
dp->context_ptr = data;
}
static void dm_bio_destructor(struct bio *bio)
{
struct io *io = bio->bi_private;
bio_free(bio, io->client->bios);
}
/*
* Functions for getting the pages from kernel memory.
*/
static void km_get_page(struct dpages *dp, struct page **p, unsigned long *len,
unsigned *offset)
{
*p = virt_to_page(dp->context_ptr);
*offset = dp->context_u;
*len = PAGE_SIZE - dp->context_u;
}
static void km_next_page(struct dpages *dp)
{
dp->context_ptr += PAGE_SIZE - dp->context_u;
dp->context_u = 0;
}
static void km_dp_init(struct dpages *dp,
|