Blame view
block/bio.c
44 KB
8c16567d8 block: switch all... |
1 |
// SPDX-License-Identifier: GPL-2.0 |
1da177e4c Linux-2.6.12-rc2 |
2 |
/* |
0fe234795 [PATCH] Update ax... |
3 |
* Copyright (C) 2001 Jens Axboe <axboe@kernel.dk> |
1da177e4c Linux-2.6.12-rc2 |
4 5 6 7 8 |
*/ #include <linux/mm.h> #include <linux/swap.h> #include <linux/bio.h> #include <linux/blkdev.h> |
a27bb332c aio: don't includ... |
9 |
#include <linux/uio.h> |
852c788f8 block: implement ... |
10 |
#include <linux/iocontext.h> |
1da177e4c Linux-2.6.12-rc2 |
11 12 13 |
#include <linux/slab.h> #include <linux/init.h> #include <linux/kernel.h> |
630d9c472 fs: reduce the us... |
14 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
15 16 |
#include <linux/mempool.h> #include <linux/workqueue.h> |
852c788f8 block: implement ... |
17 |
#include <linux/cgroup.h> |
08e18eab0 block: add bi_blk... |
18 |
#include <linux/blk-cgroup.h> |
b4c5875d3 block: Allow mapp... |
19 |
#include <linux/highmem.h> |
de6a78b60 block: Prevent hu... |
20 |
#include <linux/sched/sysctl.h> |
a892c8d52 block: Inline enc... |
21 |
#include <linux/blk-crypto.h> |
1da177e4c Linux-2.6.12-rc2 |
22 |
|
55782138e tracing/events: c... |
23 |
#include <trace/events/block.h> |
9e234eeaf blk-throttle: add... |
24 |
#include "blk.h" |
67b42d0bf rq-qos: introduce... |
25 |
#include "blk-rq-qos.h" |
0bfc24559 blktrace: port to... |
26 |
|
392ddc329 bio: add support ... |
27 28 29 30 31 |
/* * Test patch to inline a certain number of bi_io_vec's inside the bio * itself, to shrink a bio data allocation from two mempool calls to one */ #define BIO_INLINE_VECS 4 |
1da177e4c Linux-2.6.12-rc2 |
32 33 34 35 36 |
/* * if you change this list, also change bvec_alloc or things will * break badly! cannot be bigger than what you can fit into an * unsigned short */ |
bd5c4facf Fix slab name "bi... |
37 |
#define BV(x, n) { .nr_vecs = x, .name = "biovec-"#n } |
ed996a52c block: simplify a... |
38 |
static struct biovec_slab bvec_slabs[BVEC_POOL_NR] __read_mostly = { |
bd5c4facf Fix slab name "bi... |
39 |
BV(1, 1), BV(4, 4), BV(16, 16), BV(64, 64), BV(128, 128), BV(BIO_MAX_PAGES, max), |
1da177e4c Linux-2.6.12-rc2 |
40 41 42 43 |
}; #undef BV /* |
1da177e4c Linux-2.6.12-rc2 |
44 45 46 |
* fs_bio_set is the bio_set containing bio and iovec memory pools used by * IO code that does not need private memory pools. */ |
f4f8154a0 block: Use bioset... |
47 |
struct bio_set fs_bio_set; |
3f86a82ae block: Consolidat... |
48 |
EXPORT_SYMBOL(fs_bio_set); |
1da177e4c Linux-2.6.12-rc2 |
49 |
|
bb799ca02 bio: allow indivi... |
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 |
/* * Our slab pool management */ struct bio_slab { struct kmem_cache *slab; unsigned int slab_ref; unsigned int slab_size; char name[8]; }; static DEFINE_MUTEX(bio_slab_lock); static struct bio_slab *bio_slabs; static unsigned int bio_slab_nr, bio_slab_max; static struct kmem_cache *bio_find_or_create_slab(unsigned int extra_size) { unsigned int sz = sizeof(struct bio) + extra_size; struct kmem_cache *slab = NULL; |
389d7b26d bio: Fix potentia... |
67 |
struct bio_slab *bslab, *new_bio_slabs; |
386bc35a2 vfs: fix: don't i... |
68 |
unsigned int new_bio_slab_max; |
bb799ca02 bio: allow indivi... |
69 70 71 72 73 74 |
unsigned int i, entry = -1; mutex_lock(&bio_slab_lock); i = 0; while (i < bio_slab_nr) { |
f06f135d8 fs/bio.c: fix sha... |
75 |
bslab = &bio_slabs[i]; |
bb799ca02 bio: allow indivi... |
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 |
if (!bslab->slab && entry == -1) entry = i; else if (bslab->slab_size == sz) { slab = bslab->slab; bslab->slab_ref++; break; } i++; } if (slab) goto out_unlock; if (bio_slab_nr == bio_slab_max && entry == -1) { |
386bc35a2 vfs: fix: don't i... |
91 |
new_bio_slab_max = bio_slab_max << 1; |
389d7b26d bio: Fix potentia... |
92 |
new_bio_slabs = krealloc(bio_slabs, |
386bc35a2 vfs: fix: don't i... |
93 |
new_bio_slab_max * sizeof(struct bio_slab), |
389d7b26d bio: Fix potentia... |
94 95 |
GFP_KERNEL); if (!new_bio_slabs) |
bb799ca02 bio: allow indivi... |
96 |
goto out_unlock; |
386bc35a2 vfs: fix: don't i... |
97 |
bio_slab_max = new_bio_slab_max; |
389d7b26d bio: Fix potentia... |
98 |
bio_slabs = new_bio_slabs; |
bb799ca02 bio: allow indivi... |
99 100 101 102 103 104 105 |
} if (entry == -1) entry = bio_slab_nr++; bslab = &bio_slabs[entry]; snprintf(bslab->name, sizeof(bslab->name), "bio-%d", entry); |
6a2414836 block: use kmallo... |
106 107 |
slab = kmem_cache_create(bslab->name, sz, ARCH_KMALLOC_MINALIGN, SLAB_HWCACHE_ALIGN, NULL); |
bb799ca02 bio: allow indivi... |
108 109 |
if (!slab) goto out_unlock; |
bb799ca02 bio: allow indivi... |
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 |
bslab->slab = slab; bslab->slab_ref = 1; bslab->slab_size = sz; out_unlock: mutex_unlock(&bio_slab_lock); return slab; } static void bio_put_slab(struct bio_set *bs) { struct bio_slab *bslab = NULL; unsigned int i; mutex_lock(&bio_slab_lock); for (i = 0; i < bio_slab_nr; i++) { if (bs->bio_slab == bio_slabs[i].slab) { bslab = &bio_slabs[i]; break; } } if (WARN(!bslab, KERN_ERR "bio: unable to find slab! ")) goto out; WARN_ON(!bslab->slab_ref); if (--bslab->slab_ref) goto out; kmem_cache_destroy(bslab->slab); bslab->slab = NULL; out: mutex_unlock(&bio_slab_lock); } |
7ba1ba12e block: Block laye... |
147 148 |
unsigned int bvec_nr_vecs(unsigned short idx) { |
d6c02a9be block: bvec_nr_ve... |
149 |
return bvec_slabs[--idx].nr_vecs; |
7ba1ba12e block: Block laye... |
150 |
} |
9f060e223 block: Convert in... |
151 |
void bvec_free(mempool_t *pool, struct bio_vec *bv, unsigned int idx) |
bb799ca02 bio: allow indivi... |
152 |
{ |
ed996a52c block: simplify a... |
153 154 155 156 157 |
if (!idx) return; idx--; BIO_BUG_ON(idx >= BVEC_POOL_NR); |
bb799ca02 bio: allow indivi... |
158 |
|
ed996a52c block: simplify a... |
159 |
if (idx == BVEC_POOL_MAX) { |
9f060e223 block: Convert in... |
160 |
mempool_free(bv, pool); |
ed996a52c block: simplify a... |
161 |
} else { |
bb799ca02 bio: allow indivi... |
162 163 164 165 166 |
struct biovec_slab *bvs = bvec_slabs + idx; kmem_cache_free(bvs->slab, bv); } } |
9f060e223 block: Convert in... |
167 168 |
struct bio_vec *bvec_alloc(gfp_t gfp_mask, int nr, unsigned long *idx, mempool_t *pool) |
1da177e4c Linux-2.6.12-rc2 |
169 170 |
{ struct bio_vec *bvl; |
1da177e4c Linux-2.6.12-rc2 |
171 172 |
/* |
7ff9345ff bio: only mempool... |
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 |
* see comment near bvec_array define! */ switch (nr) { case 1: *idx = 0; break; case 2 ... 4: *idx = 1; break; case 5 ... 16: *idx = 2; break; case 17 ... 64: *idx = 3; break; case 65 ... 128: *idx = 4; break; case 129 ... BIO_MAX_PAGES: *idx = 5; break; default: return NULL; } /* * idx now points to the pool we want to allocate from. only the * 1-vec entry pool is mempool backed. */ |
ed996a52c block: simplify a... |
202 |
if (*idx == BVEC_POOL_MAX) { |
7ff9345ff bio: only mempool... |
203 |
fallback: |
9f060e223 block: Convert in... |
204 |
bvl = mempool_alloc(pool, gfp_mask); |
7ff9345ff bio: only mempool... |
205 206 |
} else { struct biovec_slab *bvs = bvec_slabs + *idx; |
d0164adc8 mm, page_alloc: d... |
207 |
gfp_t __gfp_mask = gfp_mask & ~(__GFP_DIRECT_RECLAIM | __GFP_IO); |
7ff9345ff bio: only mempool... |
208 |
|
0a0d96b03 block: add bio_km... |
209 |
/* |
7ff9345ff bio: only mempool... |
210 211 212 |
* Make this allocation restricted and don't dump info on * allocation failures, since we'll fallback to the mempool * in case of failure. |
0a0d96b03 block: add bio_km... |
213 |
*/ |
7ff9345ff bio: only mempool... |
214 |
__gfp_mask |= __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN; |
1da177e4c Linux-2.6.12-rc2 |
215 |
|
0a0d96b03 block: add bio_km... |
216 |
/* |
d0164adc8 mm, page_alloc: d... |
217 |
* Try a slab allocation. If this fails and __GFP_DIRECT_RECLAIM |
7ff9345ff bio: only mempool... |
218 |
* is set, retry with the 1-entry mempool |
0a0d96b03 block: add bio_km... |
219 |
*/ |
7ff9345ff bio: only mempool... |
220 |
bvl = kmem_cache_alloc(bvs->slab, __gfp_mask); |
d0164adc8 mm, page_alloc: d... |
221 |
if (unlikely(!bvl && (gfp_mask & __GFP_DIRECT_RECLAIM))) { |
ed996a52c block: simplify a... |
222 |
*idx = BVEC_POOL_MAX; |
7ff9345ff bio: only mempool... |
223 224 225 |
goto fallback; } } |
ed996a52c block: simplify a... |
226 |
(*idx)++; |
1da177e4c Linux-2.6.12-rc2 |
227 228 |
return bvl; } |
9ae3b3f52 block: provide bi... |
229 |
void bio_uninit(struct bio *bio) |
1da177e4c Linux-2.6.12-rc2 |
230 |
{ |
db9819c76 block: remove bio... |
231 232 233 234 235 236 |
#ifdef CONFIG_BLK_CGROUP if (bio->bi_blkg) { blkg_put(bio->bi_blkg); bio->bi_blkg = NULL; } #endif |
ece841abb block: fix memlea... |
237 238 |
if (bio_integrity(bio)) bio_integrity_free(bio); |
a892c8d52 block: Inline enc... |
239 240 |
bio_crypt_free_ctx(bio); |
4254bba17 block: Kill bi_de... |
241 |
} |
9ae3b3f52 block: provide bi... |
242 |
EXPORT_SYMBOL(bio_uninit); |
7ba1ba12e block: Block laye... |
243 |
|
4254bba17 block: Kill bi_de... |
244 245 246 247 |
static void bio_free(struct bio *bio) { struct bio_set *bs = bio->bi_pool; void *p; |
9ae3b3f52 block: provide bi... |
248 |
bio_uninit(bio); |
4254bba17 block: Kill bi_de... |
249 250 |
if (bs) { |
8aa6ba2f6 block: Convert bi... |
251 |
bvec_free(&bs->bvec_pool, bio->bi_io_vec, BVEC_POOL_IDX(bio)); |
4254bba17 block: Kill bi_de... |
252 253 254 255 256 |
/* * If we have front padding, adjust the bio pointer before freeing */ p = bio; |
bb799ca02 bio: allow indivi... |
257 |
p -= bs->front_pad; |
8aa6ba2f6 block: Convert bi... |
258 |
mempool_free(p, &bs->bio_pool); |
4254bba17 block: Kill bi_de... |
259 260 261 262 |
} else { /* Bio was allocated by bio_kmalloc() */ kfree(bio); } |
3676347a5 [PATCH] kill bio-... |
263 |
} |
9ae3b3f52 block: provide bi... |
264 265 266 267 268 |
/* * Users of this function have their own bio allocation. Subsequently, * they must remember to pair any call to bio_init() with bio_uninit() * when IO has completed, or when the bio is released. */ |
3a83f4677 block: bio: pass ... |
269 270 |
void bio_init(struct bio *bio, struct bio_vec *table, unsigned short max_vecs) |
1da177e4c Linux-2.6.12-rc2 |
271 |
{ |
2b94de552 bio: use memset()... |
272 |
memset(bio, 0, sizeof(*bio)); |
c4cf5261f bio: skip atomic ... |
273 |
atomic_set(&bio->__bi_remaining, 1); |
dac56212e bio: skip atomic ... |
274 |
atomic_set(&bio->__bi_cnt, 1); |
3a83f4677 block: bio: pass ... |
275 276 277 |
bio->bi_io_vec = table; bio->bi_max_vecs = max_vecs; |
1da177e4c Linux-2.6.12-rc2 |
278 |
} |
a112a71d4 fs/bio.c: move EX... |
279 |
EXPORT_SYMBOL(bio_init); |
1da177e4c Linux-2.6.12-rc2 |
280 281 |
/** |
f44b48c76 block: Add bio_re... |
282 283 284 285 286 287 288 289 290 291 292 293 |
* bio_reset - reinitialize a bio * @bio: bio to reset * * Description: * After calling bio_reset(), @bio will be in the same state as a freshly * allocated bio returned bio bio_alloc_bioset() - the only fields that are * preserved are the ones that are initialized by bio_alloc_bioset(). See * comment in struct bio. */ void bio_reset(struct bio *bio) { unsigned long flags = bio->bi_flags & (~0UL << BIO_RESET_BITS); |
9ae3b3f52 block: provide bi... |
294 |
bio_uninit(bio); |
f44b48c76 block: Add bio_re... |
295 296 |
memset(bio, 0, BIO_RESET_BYTES); |
4246a0b63 block: add a bi_e... |
297 |
bio->bi_flags = flags; |
c4cf5261f bio: skip atomic ... |
298 |
atomic_set(&bio->__bi_remaining, 1); |
f44b48c76 block: Add bio_re... |
299 300 |
} EXPORT_SYMBOL(bio_reset); |
38f8baae8 block: factor out... |
301 |
static struct bio *__bio_chain_endio(struct bio *bio) |
196d38bcc block: Generic bi... |
302 |
{ |
4246a0b63 block: add a bi_e... |
303 |
struct bio *parent = bio->bi_private; |
4e4cbee93 block: switch bio... |
304 305 |
if (!parent->bi_status) parent->bi_status = bio->bi_status; |
196d38bcc block: Generic bi... |
306 |
bio_put(bio); |
38f8baae8 block: factor out... |
307 308 309 310 311 312 |
return parent; } static void bio_chain_endio(struct bio *bio) { bio_endio(__bio_chain_endio(bio)); |
196d38bcc block: Generic bi... |
313 314 315 316 |
} /** * bio_chain - chain bio completions |
1051a902f fs: fix new kerne... |
317 |
* @bio: the target bio |
5b874af62 block: bio: fix a... |
318 |
* @parent: the parent bio of @bio |
196d38bcc block: Generic bi... |
319 320 321 322 323 324 325 326 327 328 329 330 331 |
* * The caller won't have a bi_end_io called when @bio completes - instead, * @parent's bi_end_io won't be called until both @parent and @bio have * completed; the chained bio will also be freed when it completes. * * The caller must not set bi_private or bi_end_io in @bio. */ void bio_chain(struct bio *bio, struct bio *parent) { BUG_ON(bio->bi_private || bio->bi_end_io); bio->bi_private = parent; bio->bi_end_io = bio_chain_endio; |
c4cf5261f bio: skip atomic ... |
332 |
bio_inc_remaining(parent); |
196d38bcc block: Generic bi... |
333 334 |
} EXPORT_SYMBOL(bio_chain); |
df2cb6daa block: Avoid dead... |
335 336 337 338 339 340 341 342 343 344 345 346 |
static void bio_alloc_rescue(struct work_struct *work) { struct bio_set *bs = container_of(work, struct bio_set, rescue_work); struct bio *bio; while (1) { spin_lock(&bs->rescue_lock); bio = bio_list_pop(&bs->rescue_list); spin_unlock(&bs->rescue_lock); if (!bio) break; |
ed00aabd5 block: rename gen... |
347 |
submit_bio_noacct(bio); |
df2cb6daa block: Avoid dead... |
348 349 350 351 352 353 354 |
} } static void punt_bios_to_rescuer(struct bio_set *bs) { struct bio_list punt, nopunt; struct bio *bio; |
47e0fb461 blk: make the bio... |
355 356 |
if (WARN_ON_ONCE(!bs->rescue_workqueue)) return; |
df2cb6daa block: Avoid dead... |
357 358 359 360 361 362 363 364 365 366 367 368 369 |
/* * In order to guarantee forward progress we must punt only bios that * were allocated from this bio_set; otherwise, if there was a bio on * there for a stacking driver higher up in the stack, processing it * could require allocating bios from this bio_set, and doing that from * our own rescuer would be bad. * * Since bio lists are singly linked, pop them all instead of trying to * remove from the middle of the list: */ bio_list_init(&punt); bio_list_init(&nopunt); |
f5fe1b519 blk: Ensure users... |
370 |
while ((bio = bio_list_pop(¤t->bio_list[0]))) |
df2cb6daa block: Avoid dead... |
371 |
bio_list_add(bio->bi_pool == bs ? &punt : &nopunt, bio); |
f5fe1b519 blk: Ensure users... |
372 |
current->bio_list[0] = nopunt; |
df2cb6daa block: Avoid dead... |
373 |
|
f5fe1b519 blk: Ensure users... |
374 375 376 377 |
bio_list_init(&nopunt); while ((bio = bio_list_pop(¤t->bio_list[1]))) bio_list_add(bio->bi_pool == bs ? &punt : &nopunt, bio); current->bio_list[1] = nopunt; |
df2cb6daa block: Avoid dead... |
378 379 380 381 382 383 384 |
spin_lock(&bs->rescue_lock); bio_list_merge(&bs->rescue_list, &punt); spin_unlock(&bs->rescue_lock); queue_work(bs->rescue_workqueue, &bs->rescue_work); } |
f44b48c76 block: Add bio_re... |
385 |
/** |
1da177e4c Linux-2.6.12-rc2 |
386 |
* bio_alloc_bioset - allocate a bio for I/O |
519c8e9ff block: fix Sphinx... |
387 |
* @gfp_mask: the GFP_* mask given to the slab allocator |
1da177e4c Linux-2.6.12-rc2 |
388 |
* @nr_iovecs: number of iovecs to pre-allocate |
db18efac0 bio: Fix outdated... |
389 |
* @bs: the bio_set to allocate from. |
1da177e4c Linux-2.6.12-rc2 |
390 391 |
* * Description: |
3f86a82ae block: Consolidat... |
392 393 394 |
* If @bs is NULL, uses kmalloc() to allocate the bio; else the allocation is * backed by the @bs's mempool. * |
d0164adc8 mm, page_alloc: d... |
395 396 397 398 399 400 |
* When @bs is not NULL, if %__GFP_DIRECT_RECLAIM is set then bio_alloc will * always be able to allocate a bio. This is due to the mempool guarantees. * To make this work, callers must never allocate more than 1 bio at a time * from this pool. Callers that need to allocate more than 1 bio must always * submit the previously allocated bio for IO before attempting to allocate * a new one. Failure to do so can cause deadlocks under memory pressure. |
3f86a82ae block: Consolidat... |
401 |
* |
ed00aabd5 block: rename gen... |
402 |
* Note that when running under submit_bio_noacct() (i.e. any block |
df2cb6daa block: Avoid dead... |
403 |
* driver), bios are not submitted until after you return - see the code in |
ed00aabd5 block: rename gen... |
404 |
* submit_bio_noacct() that converts recursion into iteration, to prevent |
df2cb6daa block: Avoid dead... |
405 406 407 |
* stack overflows. * * This would normally mean allocating multiple bios under |
ed00aabd5 block: rename gen... |
408 |
* submit_bio_noacct() would be susceptible to deadlocks, but we have |
df2cb6daa block: Avoid dead... |
409 410 411 412 413 |
* deadlock avoidance code that resubmits any blocked bios from a rescuer * thread. * * However, we do not guarantee forward progress for allocations from other * mempools. Doing multiple allocations from the same mempool under |
ed00aabd5 block: rename gen... |
414 |
* submit_bio_noacct() should be avoided - instead, use bio_set's front_pad |
df2cb6daa block: Avoid dead... |
415 416 |
* for per bio allocations. * |
3f86a82ae block: Consolidat... |
417 418 419 |
* RETURNS: * Pointer to new bio on success, NULL on failure. */ |
7a88fa191 block: make nr_io... |
420 421 |
struct bio *bio_alloc_bioset(gfp_t gfp_mask, unsigned int nr_iovecs, struct bio_set *bs) |
1da177e4c Linux-2.6.12-rc2 |
422 |
{ |
df2cb6daa block: Avoid dead... |
423 |
gfp_t saved_gfp = gfp_mask; |
3f86a82ae block: Consolidat... |
424 425 |
unsigned front_pad; unsigned inline_vecs; |
34053979f block: cleanup bi... |
426 |
struct bio_vec *bvl = NULL; |
451a9ebf6 bio: fix bio_kmal... |
427 428 |
struct bio *bio; void *p; |
3f86a82ae block: Consolidat... |
429 430 431 |
if (!bs) { if (nr_iovecs > UIO_MAXIOV) return NULL; |
1f4fe21cf block: bio: Use s... |
432 |
p = kmalloc(struct_size(bio, bi_inline_vecs, nr_iovecs), gfp_mask); |
3f86a82ae block: Consolidat... |
433 434 435 |
front_pad = 0; inline_vecs = nr_iovecs; } else { |
d8f429e16 block: add bioset... |
436 |
/* should not use nobvec bioset for nr_iovecs > 0 */ |
8aa6ba2f6 block: Convert bi... |
437 438 |
if (WARN_ON_ONCE(!mempool_initialized(&bs->bvec_pool) && nr_iovecs > 0)) |
d8f429e16 block: add bioset... |
439 |
return NULL; |
df2cb6daa block: Avoid dead... |
440 |
/* |
ed00aabd5 block: rename gen... |
441 |
* submit_bio_noacct() converts recursion to iteration; this |
df2cb6daa block: Avoid dead... |
442 443 444 445 446 447 |
* means if we're running beneath it, any bios we allocate and * submit will not be submitted (and thus freed) until after we * return. * * This exposes us to a potential deadlock if we allocate * multiple bios from the same bio_set() while running |
ed00aabd5 block: rename gen... |
448 |
* underneath submit_bio_noacct(). If we were to allocate |
df2cb6daa block: Avoid dead... |
449 450 451 452 453 454 455 |
* multiple bios (say a stacking block driver that was splitting * bios), we would deadlock if we exhausted the mempool's * reserve. * * We solve this, and guarantee forward progress, with a rescuer * workqueue per bio_set. If we go to allocate and there are * bios on current->bio_list, we first try the allocation |
d0164adc8 mm, page_alloc: d... |
456 457 458 |
* without __GFP_DIRECT_RECLAIM; if that fails, we punt those * bios we would be blocking to the rescuer workqueue before * we retry with the original gfp_flags. |
df2cb6daa block: Avoid dead... |
459 |
*/ |
f5fe1b519 blk: Ensure users... |
460 461 |
if (current->bio_list && (!bio_list_empty(¤t->bio_list[0]) || |
47e0fb461 blk: make the bio... |
462 463 |
!bio_list_empty(¤t->bio_list[1])) && bs->rescue_workqueue) |
d0164adc8 mm, page_alloc: d... |
464 |
gfp_mask &= ~__GFP_DIRECT_RECLAIM; |
df2cb6daa block: Avoid dead... |
465 |
|
8aa6ba2f6 block: Convert bi... |
466 |
p = mempool_alloc(&bs->bio_pool, gfp_mask); |
df2cb6daa block: Avoid dead... |
467 468 469 |
if (!p && gfp_mask != saved_gfp) { punt_bios_to_rescuer(bs); gfp_mask = saved_gfp; |
8aa6ba2f6 block: Convert bi... |
470 |
p = mempool_alloc(&bs->bio_pool, gfp_mask); |
df2cb6daa block: Avoid dead... |
471 |
} |
3f86a82ae block: Consolidat... |
472 473 474 |
front_pad = bs->front_pad; inline_vecs = BIO_INLINE_VECS; } |
451a9ebf6 bio: fix bio_kmal... |
475 476 |
if (unlikely(!p)) return NULL; |
1da177e4c Linux-2.6.12-rc2 |
477 |
|
3f86a82ae block: Consolidat... |
478 |
bio = p + front_pad; |
3a83f4677 block: bio: pass ... |
479 |
bio_init(bio, NULL, 0); |
34053979f block: cleanup bi... |
480 |
|
3f86a82ae block: Consolidat... |
481 |
if (nr_iovecs > inline_vecs) { |
ed996a52c block: simplify a... |
482 |
unsigned long idx = 0; |
8aa6ba2f6 block: Convert bi... |
483 |
bvl = bvec_alloc(gfp_mask, nr_iovecs, &idx, &bs->bvec_pool); |
df2cb6daa block: Avoid dead... |
484 485 486 |
if (!bvl && gfp_mask != saved_gfp) { punt_bios_to_rescuer(bs); gfp_mask = saved_gfp; |
8aa6ba2f6 block: Convert bi... |
487 |
bvl = bvec_alloc(gfp_mask, nr_iovecs, &idx, &bs->bvec_pool); |
df2cb6daa block: Avoid dead... |
488 |
} |
34053979f block: cleanup bi... |
489 490 |
if (unlikely(!bvl)) goto err_free; |
a38352e0a block: Add an exp... |
491 |
|
ed996a52c block: simplify a... |
492 |
bio->bi_flags |= idx << BVEC_POOL_OFFSET; |
3f86a82ae block: Consolidat... |
493 494 |
} else if (nr_iovecs) { bvl = bio->bi_inline_vecs; |
1da177e4c Linux-2.6.12-rc2 |
495 |
} |
3f86a82ae block: Consolidat... |
496 497 |
bio->bi_pool = bs; |
34053979f block: cleanup bi... |
498 |
bio->bi_max_vecs = nr_iovecs; |
34053979f block: cleanup bi... |
499 |
bio->bi_io_vec = bvl; |
1da177e4c Linux-2.6.12-rc2 |
500 |
return bio; |
34053979f block: cleanup bi... |
501 502 |
err_free: |
8aa6ba2f6 block: Convert bi... |
503 |
mempool_free(p, &bs->bio_pool); |
34053979f block: cleanup bi... |
504 |
return NULL; |
1da177e4c Linux-2.6.12-rc2 |
505 |
} |
a112a71d4 fs/bio.c: move EX... |
506 |
EXPORT_SYMBOL(bio_alloc_bioset); |
1da177e4c Linux-2.6.12-rc2 |
507 |
|
38a72dac4 block: Add bio_co... |
508 |
void zero_fill_bio_iter(struct bio *bio, struct bvec_iter start) |
1da177e4c Linux-2.6.12-rc2 |
509 510 |
{ unsigned long flags; |
7988613b0 block: Convert bi... |
511 512 |
struct bio_vec bv; struct bvec_iter iter; |
1da177e4c Linux-2.6.12-rc2 |
513 |
|
38a72dac4 block: Add bio_co... |
514 |
__bio_for_each_segment(bv, bio, iter, start) { |
7988613b0 block: Convert bi... |
515 516 517 |
char *data = bvec_kmap_irq(&bv, &flags); memset(data, 0, bv.bv_len); flush_dcache_page(bv.bv_page); |
1da177e4c Linux-2.6.12-rc2 |
518 519 520 |
bvec_kunmap_irq(data, &flags); } } |
38a72dac4 block: Add bio_co... |
521 |
EXPORT_SYMBOL(zero_fill_bio_iter); |
1da177e4c Linux-2.6.12-rc2 |
522 |
|
83c9c5471 fs: move guard_bi... |
523 524 525 526 527 528 529 530 531 532 |
/** * bio_truncate - truncate the bio to small size of @new_size * @bio: the bio to be truncated * @new_size: new size for truncating the bio * * Description: * Truncate the bio to new size of @new_size. If bio_op(bio) is * REQ_OP_READ, zero the truncated part. This function should only * be used for handling corner cases, such as bio eod. */ |
85a8ce62c block: add bio_tr... |
533 534 535 536 537 538 539 540 541 |
void bio_truncate(struct bio *bio, unsigned new_size) { struct bio_vec bv; struct bvec_iter iter; unsigned int done = 0; bool truncated = false; if (new_size >= bio->bi_iter.bi_size) return; |
83c9c5471 fs: move guard_bi... |
542 |
if (bio_op(bio) != REQ_OP_READ) |
85a8ce62c block: add bio_tr... |
543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 |
goto exit; bio_for_each_segment(bv, bio, iter) { if (done + bv.bv_len > new_size) { unsigned offset; if (!truncated) offset = new_size - done; else offset = 0; zero_user(bv.bv_page, offset, bv.bv_len - offset); truncated = true; } done += bv.bv_len; } exit: /* * Don't touch bvec table here and make it really immutable, since * fs bio user has to retrieve all pages via bio_for_each_segment_all * in its .end_bio() callback. * * It is enough to truncate bio by updating .bi_size since we can make * correct bvec with the updated .bi_size for drivers. */ bio->bi_iter.bi_size = new_size; } |
1da177e4c Linux-2.6.12-rc2 |
570 |
/** |
29125ed62 block: move guard... |
571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 |
* guard_bio_eod - truncate a BIO to fit the block device * @bio: bio to truncate * * This allows us to do IO even on the odd last sectors of a device, even if the * block size is some multiple of the physical sector size. * * We'll just truncate the bio to the size of the device, and clear the end of * the buffer head manually. Truly out-of-range accesses will turn into actual * I/O errors, this only handles the "we need to be able to do I/O at the final * sector" case. */ void guard_bio_eod(struct bio *bio) { sector_t maxsector; struct hd_struct *part; rcu_read_lock(); part = __disk_get_part(bio->bi_disk, bio->bi_partno); if (part) maxsector = part_nr_sects_read(part); else maxsector = get_capacity(bio->bi_disk); rcu_read_unlock(); if (!maxsector) return; /* * If the *whole* IO is past the end of the device, * let it through, and the IO layer will turn it into * an EIO. */ if (unlikely(bio->bi_iter.bi_sector >= maxsector)) return; maxsector -= bio->bi_iter.bi_sector; if (likely((bio->bi_iter.bi_size >> 9) <= maxsector)) return; bio_truncate(bio, maxsector << 9); } /** |
1da177e4c Linux-2.6.12-rc2 |
614 615 616 617 618 |
* bio_put - release a reference to a bio * @bio: bio to release reference to * * Description: * Put a reference to a &struct bio, either one you have gotten with |
9b10f6a9c block: remove bio... |
619 |
* bio_alloc, bio_get or bio_clone_*. The last put of a bio will free it. |
1da177e4c Linux-2.6.12-rc2 |
620 621 622 |
**/ void bio_put(struct bio *bio) { |
dac56212e bio: skip atomic ... |
623 |
if (!bio_flagged(bio, BIO_REFFED)) |
4254bba17 block: Kill bi_de... |
624 |
bio_free(bio); |
dac56212e bio: skip atomic ... |
625 626 627 628 629 630 631 632 633 |
else { BIO_BUG_ON(!atomic_read(&bio->__bi_cnt)); /* * last put frees it */ if (atomic_dec_and_test(&bio->__bi_cnt)) bio_free(bio); } |
1da177e4c Linux-2.6.12-rc2 |
634 |
} |
a112a71d4 fs/bio.c: move EX... |
635 |
EXPORT_SYMBOL(bio_put); |
1da177e4c Linux-2.6.12-rc2 |
636 |
|
1da177e4c Linux-2.6.12-rc2 |
637 |
/** |
59d276fe0 block: Add bio_cl... |
638 639 640 641 642 643 644 645 646 647 648 649 |
* __bio_clone_fast - clone a bio that shares the original bio's biovec * @bio: destination bio * @bio_src: bio to clone * * Clone a &bio. Caller will own the returned bio, but not * the actual data it points to. Reference count of returned * bio will be one. * * Caller must ensure that @bio_src is not freed before @bio. */ void __bio_clone_fast(struct bio *bio, struct bio *bio_src) { |
ed996a52c block: simplify a... |
650 |
BUG_ON(bio->bi_pool && BVEC_POOL_IDX(bio)); |
59d276fe0 block: Add bio_cl... |
651 652 |
/* |
74d46992e block: replace bi... |
653 |
* most users will be overriding ->bi_disk with a new target, |
59d276fe0 block: Add bio_cl... |
654 655 |
* so we don't set nor calculate new physical/hw segment counts here */ |
74d46992e block: replace bi... |
656 |
bio->bi_disk = bio_src->bi_disk; |
62530ed8b bio: ensure __bio... |
657 |
bio->bi_partno = bio_src->bi_partno; |
b7c44ed9d block: manipulate... |
658 |
bio_set_flag(bio, BIO_CLONED); |
111be8839 block-throttle: a... |
659 660 |
if (bio_flagged(bio_src, BIO_THROTTLED)) bio_set_flag(bio, BIO_THROTTLED); |
1eff9d322 block: rename bio... |
661 |
bio->bi_opf = bio_src->bi_opf; |
ca474b738 block: copy iopri... |
662 |
bio->bi_ioprio = bio_src->bi_ioprio; |
cb6934f8e block: add suppor... |
663 |
bio->bi_write_hint = bio_src->bi_write_hint; |
59d276fe0 block: Add bio_cl... |
664 665 |
bio->bi_iter = bio_src->bi_iter; bio->bi_io_vec = bio_src->bi_io_vec; |
20bd723ec block: add missin... |
666 |
|
db6638d7d blkcg: remove bio... |
667 |
bio_clone_blkg_association(bio, bio_src); |
e439bedf6 blkcg: consolidat... |
668 |
blkcg_bio_issue_init(bio); |
59d276fe0 block: Add bio_cl... |
669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 |
} EXPORT_SYMBOL(__bio_clone_fast); /** * bio_clone_fast - clone a bio that shares the original bio's biovec * @bio: bio to clone * @gfp_mask: allocation priority * @bs: bio_set to allocate from * * Like __bio_clone_fast, only also allocates the returned bio */ struct bio *bio_clone_fast(struct bio *bio, gfp_t gfp_mask, struct bio_set *bs) { struct bio *b; b = bio_alloc_bioset(gfp_mask, 0, bs); if (!b) return NULL; __bio_clone_fast(b, bio); |
07560151d block: make bio_c... |
689 690 |
if (bio_crypt_clone(b, bio, gfp_mask) < 0) goto err_put; |
a892c8d52 block: Inline enc... |
691 |
|
07560151d block: make bio_c... |
692 693 694 |
if (bio_integrity(bio) && bio_integrity_clone(b, bio, gfp_mask) < 0) goto err_put; |
59d276fe0 block: Add bio_cl... |
695 696 |
return b; |
07560151d block: make bio_c... |
697 698 699 700 |
err_put: bio_put(b); return NULL; |
59d276fe0 block: Add bio_cl... |
701 702 |
} EXPORT_SYMBOL(bio_clone_fast); |
5cbd28e3c block: move disk_... |
703 704 705 706 707 |
const char *bio_devname(struct bio *bio, char *buf) { return disk_name(bio->bi_disk, bio->bi_partno, buf); } EXPORT_SYMBOL(bio_devname); |
5919482e2 block: check if p... |
708 709 |
static inline bool page_is_mergeable(const struct bio_vec *bv, struct page *page, unsigned int len, unsigned int off, |
ff896738b block: return fro... |
710 |
bool *same_page) |
5919482e2 block: check if p... |
711 |
{ |
d81665198 block: Fix page_i... |
712 713 |
size_t bv_end = bv->bv_offset + bv->bv_len; phys_addr_t vec_end_addr = page_to_phys(bv->bv_page) + bv_end - 1; |
5919482e2 block: check if p... |
714 715 716 717 718 719 |
phys_addr_t page_addr = page_to_phys(page); if (vec_end_addr + 1 != page_addr + off) return false; if (xen_domain() && !xen_biovec_phys_mergeable(bv, page)) return false; |
52d52d1c9 block: only allow... |
720 |
|
ff896738b block: return fro... |
721 |
*same_page = ((vec_end_addr & PAGE_MASK) == page_addr); |
d81665198 block: Fix page_i... |
722 723 724 |
if (*same_page) return true; return (bv->bv_page + bv_end / PAGE_SIZE) == (page + off / PAGE_SIZE); |
5919482e2 block: check if p... |
725 |
} |
e45811057 block: rename __b... |
726 727 728 729 730 731 732 733 |
/* * Try to merge a page into a segment, while obeying the hardware segment * size limit. This is not for normal read/write bios, but for passthrough * or Zone Append operations that we can't split. */ static bool bio_try_merge_hw_seg(struct request_queue *q, struct bio *bio, struct page *page, unsigned len, unsigned offset, bool *same_page) |
489fbbcb5 block: enable mul... |
734 |
{ |
384209cd5 block: create a b... |
735 |
struct bio_vec *bv = &bio->bi_io_vec[bio->bi_vcnt - 1]; |
489fbbcb5 block: enable mul... |
736 737 738 739 740 741 |
unsigned long mask = queue_segment_boundary(q); phys_addr_t addr1 = page_to_phys(bv->bv_page) + bv->bv_offset; phys_addr_t addr2 = page_to_phys(page) + offset + len - 1; if ((addr1 | mask) != (addr2 | mask)) return false; |
489fbbcb5 block: enable mul... |
742 743 |
if (bv->bv_len + len > queue_max_segment_size(q)) return false; |
384209cd5 block: create a b... |
744 |
return __bio_try_merge_page(bio, page, len, offset, same_page); |
489fbbcb5 block: enable mul... |
745 |
} |
f45958756 block: remove bio... |
746 |
/** |
e45811057 block: rename __b... |
747 748 749 750 751 752 753 754 |
* bio_add_hw_page - attempt to add a page to a bio with hw constraints * @q: the target queue * @bio: destination bio * @page: page to add * @len: vec entry length * @offset: vec entry offset * @max_sectors: maximum number of sectors that can be added * @same_page: return if the segment has been merged inside the same page |
c66a14d07 block: simplify b... |
755 |
* |
e45811057 block: rename __b... |
756 757 |
* Add a page to a bio while respecting the hardware max_sectors, max_segment * and gap limitations. |
1da177e4c Linux-2.6.12-rc2 |
758 |
*/ |
e45811057 block: rename __b... |
759 |
int bio_add_hw_page(struct request_queue *q, struct bio *bio, |
190470871 block: put the sa... |
760 |
struct page *page, unsigned int len, unsigned int offset, |
e45811057 block: rename __b... |
761 |
unsigned int max_sectors, bool *same_page) |
1da177e4c Linux-2.6.12-rc2 |
762 |
{ |
1da177e4c Linux-2.6.12-rc2 |
763 |
struct bio_vec *bvec; |
e45811057 block: rename __b... |
764 |
if (WARN_ON_ONCE(bio_flagged(bio, BIO_CLONED))) |
1da177e4c Linux-2.6.12-rc2 |
765 |
return 0; |
e45811057 block: rename __b... |
766 |
if (((bio->bi_iter.bi_size + len) >> 9) > max_sectors) |
1da177e4c Linux-2.6.12-rc2 |
767 |
return 0; |
80cfd548e [BLOCK] bio: chec... |
768 |
if (bio->bi_vcnt > 0) { |
e45811057 block: rename __b... |
769 |
if (bio_try_merge_hw_seg(q, bio, page, len, offset, same_page)) |
384209cd5 block: create a b... |
770 |
return len; |
320ea869a block: improve th... |
771 772 773 774 775 |
/* * If the queue doesn't support SG gaps and adding this segment * would create a gap, disallow it. */ |
384209cd5 block: create a b... |
776 |
bvec = &bio->bi_io_vec[bio->bi_vcnt - 1]; |
320ea869a block: improve th... |
777 778 |
if (bvec_gap_to_prev(q, bvec, offset)) return 0; |
80cfd548e [BLOCK] bio: chec... |
779 |
} |
79d08f89b block: fix .bi_si... |
780 |
if (bio_full(bio, len)) |
1da177e4c Linux-2.6.12-rc2 |
781 |
return 0; |
14ccb66b3 block: remove the... |
782 |
if (bio->bi_vcnt >= queue_max_segments(q)) |
489fbbcb5 block: enable mul... |
783 |
return 0; |
fcbf6a087 bio: modify __bio... |
784 785 786 787 788 |
bvec = &bio->bi_io_vec[bio->bi_vcnt]; bvec->bv_page = page; bvec->bv_len = len; bvec->bv_offset = offset; bio->bi_vcnt++; |
dcdca753c block: clean up _... |
789 |
bio->bi_iter.bi_size += len; |
1da177e4c Linux-2.6.12-rc2 |
790 791 |
return len; } |
190470871 block: put the sa... |
792 |
|
e45811057 block: rename __b... |
793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 |
/** * bio_add_pc_page - attempt to add page to passthrough bio * @q: the target queue * @bio: destination bio * @page: page to add * @len: vec entry length * @offset: vec entry offset * * Attempt to add a page to the bio_vec maplist. This can fail for a * number of reasons, such as the bio being full or target block device * limitations. The target block device must allow bio's up to PAGE_SIZE, * so it is always possible to add a single page to an empty bio. * * This should only be used by passthrough bios. */ |
190470871 block: put the sa... |
808 809 810 |
int bio_add_pc_page(struct request_queue *q, struct bio *bio, struct page *page, unsigned int len, unsigned int offset) { |
d1916c86c block: move same ... |
811 |
bool same_page = false; |
e45811057 block: rename __b... |
812 813 |
return bio_add_hw_page(q, bio, page, len, offset, queue_max_hw_sectors(q), &same_page); |
190470871 block: put the sa... |
814 |
} |
a112a71d4 fs/bio.c: move EX... |
815 |
EXPORT_SYMBOL(bio_add_pc_page); |
6e68af666 [SCSI] Convert SC... |
816 817 |
/** |
0aa69fd32 block: add a lowe... |
818 819 |
* __bio_try_merge_page - try appending data to an existing bvec. * @bio: destination bio |
551879a48 block: clarify th... |
820 |
* @page: start page to add |
0aa69fd32 block: add a lowe... |
821 |
* @len: length of the data to add |
551879a48 block: clarify th... |
822 |
* @off: offset of the data relative to @page |
ff896738b block: return fro... |
823 |
* @same_page: return if the segment has been merged inside the same page |
1da177e4c Linux-2.6.12-rc2 |
824 |
* |
0aa69fd32 block: add a lowe... |
825 |
* Try to add the data at @page + @off to the last bvec of @bio. This is a |
3cf148891 block: bio: delet... |
826 |
* useful optimisation for file systems with a block size smaller than the |
0aa69fd32 block: add a lowe... |
827 828 |
* page size. * |
551879a48 block: clarify th... |
829 830 |
* Warn if (@len, @off) crosses pages in case that @same_page is true. * |
0aa69fd32 block: add a lowe... |
831 |
* Return %true on success or %false on failure. |
1da177e4c Linux-2.6.12-rc2 |
832 |
*/ |
0aa69fd32 block: add a lowe... |
833 |
bool __bio_try_merge_page(struct bio *bio, struct page *page, |
ff896738b block: return fro... |
834 |
unsigned int len, unsigned int off, bool *same_page) |
1da177e4c Linux-2.6.12-rc2 |
835 |
{ |
c66a14d07 block: simplify b... |
836 |
if (WARN_ON_ONCE(bio_flagged(bio, BIO_CLONED))) |
0aa69fd32 block: add a lowe... |
837 |
return false; |
762380ad9 block: add notion... |
838 |
|
cc90bc684 block: fix "check... |
839 |
if (bio->bi_vcnt > 0) { |
0aa69fd32 block: add a lowe... |
840 |
struct bio_vec *bv = &bio->bi_io_vec[bio->bi_vcnt - 1]; |
5919482e2 block: check if p... |
841 842 |
if (page_is_mergeable(bv, page, len, off, same_page)) { |
2cd896a5e block: Set same_p... |
843 844 |
if (bio->bi_iter.bi_size > UINT_MAX - len) { *same_page = false; |
cc90bc684 block: fix "check... |
845 |
return false; |
2cd896a5e block: Set same_p... |
846 |
} |
5919482e2 block: check if p... |
847 848 849 850 |
bv->bv_len += len; bio->bi_iter.bi_size += len; return true; } |
c66a14d07 block: simplify b... |
851 |
} |
0aa69fd32 block: add a lowe... |
852 853 854 |
return false; } EXPORT_SYMBOL_GPL(__bio_try_merge_page); |
c66a14d07 block: simplify b... |
855 |
|
0aa69fd32 block: add a lowe... |
856 |
/** |
551879a48 block: clarify th... |
857 |
* __bio_add_page - add page(s) to a bio in a new segment |
0aa69fd32 block: add a lowe... |
858 |
* @bio: destination bio |
551879a48 block: clarify th... |
859 860 861 |
* @page: start page to add * @len: length of the data to add, may cross pages * @off: offset of the data relative to @page, may cross pages |
0aa69fd32 block: add a lowe... |
862 863 864 865 866 867 868 869 |
* * Add the data at @page + @off to @bio as a new bvec. The caller must ensure * that @bio has space for another bvec. */ void __bio_add_page(struct bio *bio, struct page *page, unsigned int len, unsigned int off) { struct bio_vec *bv = &bio->bi_io_vec[bio->bi_vcnt]; |
c66a14d07 block: simplify b... |
870 |
|
0aa69fd32 block: add a lowe... |
871 |
WARN_ON_ONCE(bio_flagged(bio, BIO_CLONED)); |
79d08f89b block: fix .bi_si... |
872 |
WARN_ON_ONCE(bio_full(bio, len)); |
0aa69fd32 block: add a lowe... |
873 874 875 876 |
bv->bv_page = page; bv->bv_offset = off; bv->bv_len = len; |
c66a14d07 block: simplify b... |
877 |
|
c66a14d07 block: simplify b... |
878 |
bio->bi_iter.bi_size += len; |
0aa69fd32 block: add a lowe... |
879 |
bio->bi_vcnt++; |
b8e24a930 block: annotate r... |
880 881 882 |
if (!bio_flagged(bio, BIO_WORKINGSET) && unlikely(PageWorkingset(page))) bio_set_flag(bio, BIO_WORKINGSET); |
0aa69fd32 block: add a lowe... |
883 884 885 886 |
} EXPORT_SYMBOL_GPL(__bio_add_page); /** |
551879a48 block: clarify th... |
887 |
* bio_add_page - attempt to add page(s) to bio |
0aa69fd32 block: add a lowe... |
888 |
* @bio: destination bio |
551879a48 block: clarify th... |
889 890 891 |
* @page: start page to add * @len: vec entry length, may cross pages * @offset: vec entry offset relative to @page, may cross pages |
0aa69fd32 block: add a lowe... |
892 |
* |
551879a48 block: clarify th... |
893 |
* Attempt to add page(s) to the bio_vec maplist. This will only fail |
0aa69fd32 block: add a lowe... |
894 895 896 897 898 |
* if either bio->bi_vcnt == bio->bi_max_vecs or it's a cloned bio. */ int bio_add_page(struct bio *bio, struct page *page, unsigned int len, unsigned int offset) { |
ff896738b block: return fro... |
899 900 901 |
bool same_page = false; if (!__bio_try_merge_page(bio, page, len, offset, &same_page)) { |
79d08f89b block: fix .bi_si... |
902 |
if (bio_full(bio, len)) |
0aa69fd32 block: add a lowe... |
903 904 905 |
return 0; __bio_add_page(bio, page, len, offset); } |
c66a14d07 block: simplify b... |
906 |
return len; |
1da177e4c Linux-2.6.12-rc2 |
907 |
} |
a112a71d4 fs/bio.c: move EX... |
908 |
EXPORT_SYMBOL(bio_add_page); |
1da177e4c Linux-2.6.12-rc2 |
909 |
|
d241a95f3 block: optionally... |
910 |
void bio_release_pages(struct bio *bio, bool mark_dirty) |
7321ecbfc block: change how... |
911 912 913 |
{ struct bvec_iter_all iter_all; struct bio_vec *bvec; |
7321ecbfc block: change how... |
914 |
|
b2d0d9913 block: move the B... |
915 916 |
if (bio_flagged(bio, BIO_NO_PAGE_REF)) return; |
d241a95f3 block: optionally... |
917 918 919 |
bio_for_each_segment_all(bvec, bio, iter_all) { if (mark_dirty && !PageCompound(bvec->bv_page)) set_page_dirty_lock(bvec->bv_page); |
7321ecbfc block: change how... |
920 |
put_page(bvec->bv_page); |
d241a95f3 block: optionally... |
921 |
} |
7321ecbfc block: change how... |
922 |
} |
29b2a3aa2 block: export bio... |
923 |
EXPORT_SYMBOL_GPL(bio_release_pages); |
7321ecbfc block: change how... |
924 |
|
6d0c48aed block: implement ... |
925 926 927 928 929 930 931 932 933 934 935 936 |
static int __bio_iov_bvec_add_pages(struct bio *bio, struct iov_iter *iter) { const struct bio_vec *bv = iter->bvec; unsigned int len; size_t size; if (WARN_ON_ONCE(iter->iov_offset > bv->bv_len)) return -EINVAL; len = min_t(size_t, bv->bv_len - iter->iov_offset, iter->count); size = bio_add_page(bio, bv->bv_page, len, bv->bv_offset + iter->iov_offset); |
a10584c3c block: refactor _... |
937 938 |
if (unlikely(size != len)) return -EINVAL; |
a10584c3c block: refactor _... |
939 940 |
iov_iter_advance(iter, size); return 0; |
6d0c48aed block: implement ... |
941 |
} |
576ed9135 block: use bio_ad... |
942 |
#define PAGE_PTRS_PER_BVEC (sizeof(struct bio_vec) / sizeof(struct page *)) |
2cefe4dba block: add bio_io... |
943 |
/** |
17d51b10d block: bio_iov_it... |
944 |
* __bio_iov_iter_get_pages - pin user or kernel pages and add them to a bio |
2cefe4dba block: add bio_io... |
945 946 947 |
* @bio: bio to add pages to * @iter: iov iterator describing the region to be mapped * |
17d51b10d block: bio_iov_it... |
948 |
* Pins pages from *iter and appends them to @bio's bvec array. The |
2cefe4dba block: add bio_io... |
949 |
* pages will have to be released using put_page() when done. |
17d51b10d block: bio_iov_it... |
950 |
* For multi-segment *iter, this function only adds pages from the |
3cf148891 block: bio: delet... |
951 |
* next non-empty segment of the iov iterator. |
2cefe4dba block: add bio_io... |
952 |
*/ |
17d51b10d block: bio_iov_it... |
953 |
static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) |
2cefe4dba block: add bio_io... |
954 |
{ |
576ed9135 block: use bio_ad... |
955 956 |
unsigned short nr_pages = bio->bi_max_vecs - bio->bi_vcnt; unsigned short entries_left = bio->bi_max_vecs - bio->bi_vcnt; |
2cefe4dba block: add bio_io... |
957 958 |
struct bio_vec *bv = bio->bi_io_vec + bio->bi_vcnt; struct page **pages = (struct page **)bv; |
456918049 block: fix page l... |
959 |
bool same_page = false; |
576ed9135 block: use bio_ad... |
960 961 |
ssize_t size, left; unsigned len, i; |
b403ea240 block: bio_iov_it... |
962 |
size_t offset; |
576ed9135 block: use bio_ad... |
963 964 965 966 967 968 969 970 |
/* * Move page array up in the allocated memory for the bio vecs as far as * possible so that we can start filling biovecs from the beginning * without overwriting the temporary page array. */ BUILD_BUG_ON(PAGE_PTRS_PER_BVEC < 2); pages += entries_left * (PAGE_PTRS_PER_BVEC - 1); |
2cefe4dba block: add bio_io... |
971 972 973 974 |
size = iov_iter_get_pages(iter, pages, LONG_MAX, nr_pages, &offset); if (unlikely(size <= 0)) return size ? size : -EFAULT; |
2cefe4dba block: add bio_io... |
975 |
|
576ed9135 block: use bio_ad... |
976 977 |
for (left = size, i = 0; left > 0; left -= len, i++) { struct page *page = pages[i]; |
2cefe4dba block: add bio_io... |
978 |
|
576ed9135 block: use bio_ad... |
979 |
len = min_t(size_t, PAGE_SIZE - offset, left); |
456918049 block: fix page l... |
980 981 982 983 984 |
if (__bio_try_merge_page(bio, page, len, offset, &same_page)) { if (same_page) put_page(page); } else { |
79d08f89b block: fix .bi_si... |
985 |
if (WARN_ON_ONCE(bio_full(bio, len))) |
456918049 block: fix page l... |
986 987 988 |
return -EINVAL; __bio_add_page(bio, page, len, offset); } |
576ed9135 block: use bio_ad... |
989 |
offset = 0; |
2cefe4dba block: add bio_io... |
990 |
} |
2cefe4dba block: add bio_io... |
991 992 993 |
iov_iter_advance(iter, size); return 0; } |
17d51b10d block: bio_iov_it... |
994 |
|
0512a75b9 block: Introduce ... |
995 996 997 998 999 1000 1001 1002 1003 1004 1005 |
static int __bio_iov_append_get_pages(struct bio *bio, struct iov_iter *iter) { unsigned short nr_pages = bio->bi_max_vecs - bio->bi_vcnt; unsigned short entries_left = bio->bi_max_vecs - bio->bi_vcnt; struct request_queue *q = bio->bi_disk->queue; unsigned int max_append_sectors = queue_max_zone_append_sectors(q); struct bio_vec *bv = bio->bi_io_vec + bio->bi_vcnt; struct page **pages = (struct page **)bv; ssize_t size, left; unsigned len, i; size_t offset; |
4977d121b block: advance io... |
1006 |
int ret = 0; |
0512a75b9 block: Introduce ... |
1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 |
if (WARN_ON_ONCE(!max_append_sectors)) return 0; /* * Move page array up in the allocated memory for the bio vecs as far as * possible so that we can start filling biovecs from the beginning * without overwriting the temporary page array. */ BUILD_BUG_ON(PAGE_PTRS_PER_BVEC < 2); pages += entries_left * (PAGE_PTRS_PER_BVEC - 1); size = iov_iter_get_pages(iter, pages, LONG_MAX, nr_pages, &offset); if (unlikely(size <= 0)) return size ? size : -EFAULT; for (left = size, i = 0; left > 0; left -= len, i++) { struct page *page = pages[i]; bool same_page = false; len = min_t(size_t, PAGE_SIZE - offset, left); if (bio_add_hw_page(q, bio, page, len, offset, |
4977d121b block: advance io... |
1029 1030 1031 1032 |
max_append_sectors, &same_page) != len) { ret = -EINVAL; break; } |
0512a75b9 block: Introduce ... |
1033 1034 1035 1036 |
if (same_page) put_page(page); offset = 0; } |
4977d121b block: advance io... |
1037 1038 |
iov_iter_advance(iter, size - left); return ret; |
0512a75b9 block: Introduce ... |
1039 |
} |
17d51b10d block: bio_iov_it... |
1040 |
/** |
6d0c48aed block: implement ... |
1041 |
* bio_iov_iter_get_pages - add user or kernel pages to a bio |
17d51b10d block: bio_iov_it... |
1042 |
* @bio: bio to add pages to |
6d0c48aed block: implement ... |
1043 1044 1045 1046 1047 |
* @iter: iov iterator describing the region to be added * * This takes either an iterator pointing to user memory, or one pointing to * kernel pages (BVEC iterator). If we're adding user pages, we pin them and * map them into the kernel. On IO completion, the caller should put those |
399254aaf block: add BIO_NO... |
1048 1049 1050 1051 1052 1053 |
* pages. If we're adding kernel pages, and the caller told us it's safe to * do so, we just have to add the pages to the bio directly. We don't grab an * extra reference to those pages (the user should already have that), and we * don't put the page on IO completion. The caller needs to check if the bio is * flagged BIO_NO_PAGE_REF on IO completion. If it isn't, then pages should be * released. |
17d51b10d block: bio_iov_it... |
1054 |
* |
17d51b10d block: bio_iov_it... |
1055 |
* The function tries, but does not guarantee, to pin as many pages as |
5cd3ddc18 docs: bio: fix a ... |
1056 |
* fit into the bio, or are requested in @iter, whatever is smaller. If |
6d0c48aed block: implement ... |
1057 1058 |
* MM encounters an error pinning the requested pages, it stops. Error * is returned only if 0 pages could be pinned. |
17d51b10d block: bio_iov_it... |
1059 1060 1061 |
*/ int bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) { |
6d0c48aed block: implement ... |
1062 |
const bool is_bvec = iov_iter_is_bvec(iter); |
14eacf12d block: don't allo... |
1063 1064 1065 1066 |
int ret; if (WARN_ON_ONCE(bio->bi_vcnt)) return -EINVAL; |
17d51b10d block: bio_iov_it... |
1067 1068 |
do { |
0512a75b9 block: Introduce ... |
1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 |
if (bio_op(bio) == REQ_OP_ZONE_APPEND) { if (WARN_ON_ONCE(is_bvec)) return -EINVAL; ret = __bio_iov_append_get_pages(bio, iter); } else { if (is_bvec) ret = __bio_iov_bvec_add_pages(bio, iter); else ret = __bio_iov_iter_get_pages(bio, iter); } |
79d08f89b block: fix .bi_si... |
1079 |
} while (!ret && iov_iter_count(iter) && !bio_full(bio, 0)); |
17d51b10d block: bio_iov_it... |
1080 |
|
b62074307 block: never take... |
1081 |
if (is_bvec) |
7321ecbfc block: change how... |
1082 |
bio_set_flag(bio, BIO_NO_PAGE_REF); |
14eacf12d block: don't allo... |
1083 |
return bio->bi_vcnt ? 0 : ret; |
17d51b10d block: bio_iov_it... |
1084 |
} |
29b2a3aa2 block: export bio... |
1085 |
EXPORT_SYMBOL_GPL(bio_iov_iter_get_pages); |
2cefe4dba block: add bio_io... |
1086 |
|
4246a0b63 block: add a bi_e... |
1087 |
static void submit_bio_wait_endio(struct bio *bio) |
9e882242c block: Add submit... |
1088 |
{ |
65e53aab6 block: Use DECLAR... |
1089 |
complete(bio->bi_private); |
9e882242c block: Add submit... |
1090 1091 1092 1093 |
} /** * submit_bio_wait - submit a bio, and wait until it completes |
9e882242c block: Add submit... |
1094 1095 1096 1097 |
* @bio: The &struct bio which describes the I/O * * Simple wrapper around submit_bio(). Returns 0 on success, or the error from * bio_endio() on failure. |
3d289d688 block: Add commen... |
1098 1099 1100 1101 |
* * WARNING: Unlike to how submit_bio() is usually used, this function does not * result in bio reference to be consumed. The caller must drop the reference * on his own. |
9e882242c block: Add submit... |
1102 |
*/ |
4e49ea4a3 block/fs/drivers:... |
1103 |
int submit_bio_wait(struct bio *bio) |
9e882242c block: Add submit... |
1104 |
{ |
e319e1fbd block, locking/lo... |
1105 |
DECLARE_COMPLETION_ONSTACK_MAP(done, bio->bi_disk->lockdep_map); |
de6a78b60 block: Prevent hu... |
1106 |
unsigned long hang_check; |
9e882242c block: Add submit... |
1107 |
|
65e53aab6 block: Use DECLAR... |
1108 |
bio->bi_private = &done; |
9e882242c block: Add submit... |
1109 |
bio->bi_end_io = submit_bio_wait_endio; |
1eff9d322 block: rename bio... |
1110 |
bio->bi_opf |= REQ_SYNC; |
4e49ea4a3 block/fs/drivers:... |
1111 |
submit_bio(bio); |
de6a78b60 block: Prevent hu... |
1112 1113 1114 1115 1116 1117 1118 1119 1120 |
/* Prevent hang_check timer from firing at us during very long I/O */ hang_check = sysctl_hung_task_timeout_secs; if (hang_check) while (!wait_for_completion_io_timeout(&done, hang_check * (HZ/2))) ; else wait_for_completion_io(&done); |
9e882242c block: Add submit... |
1121 |
|
65e53aab6 block: Use DECLAR... |
1122 |
return blk_status_to_errno(bio->bi_status); |
9e882242c block: Add submit... |
1123 1124 |
} EXPORT_SYMBOL(submit_bio_wait); |
054bdf646 block: Add bio_ad... |
1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 |
/** * bio_advance - increment/complete a bio by some number of bytes * @bio: bio to advance * @bytes: number of bytes to complete * * This updates bi_sector, bi_size and bi_idx; if the number of bytes to * complete doesn't align with a bvec boundary, then bv_len and bv_offset will * be updated on the last bvec as well. * * @bio will then represent the remaining, uncompleted portion of the io. */ void bio_advance(struct bio *bio, unsigned bytes) { if (bio_integrity(bio)) bio_integrity_advance(bio, bytes); |
a892c8d52 block: Inline enc... |
1140 |
bio_crypt_advance(bio, bytes); |
4550dd6c6 block: Immutable ... |
1141 |
bio_advance_iter(bio, &bio->bi_iter, bytes); |
054bdf646 block: Add bio_ad... |
1142 1143 |
} EXPORT_SYMBOL(bio_advance); |
45db54d58 block: Split out ... |
1144 1145 |
void bio_copy_data_iter(struct bio *dst, struct bvec_iter *dst_iter, struct bio *src, struct bvec_iter *src_iter) |
16ac3d63e block: Add bio_co... |
1146 |
{ |
1cb9dda4f block: Convert bi... |
1147 |
struct bio_vec src_bv, dst_bv; |
16ac3d63e block: Add bio_co... |
1148 |
void *src_p, *dst_p; |
1cb9dda4f block: Convert bi... |
1149 |
unsigned bytes; |
16ac3d63e block: Add bio_co... |
1150 |
|
45db54d58 block: Split out ... |
1151 1152 1153 |
while (src_iter->bi_size && dst_iter->bi_size) { src_bv = bio_iter_iovec(src, *src_iter); dst_bv = bio_iter_iovec(dst, *dst_iter); |
1cb9dda4f block: Convert bi... |
1154 1155 |
bytes = min(src_bv.bv_len, dst_bv.bv_len); |
16ac3d63e block: Add bio_co... |
1156 |
|
1cb9dda4f block: Convert bi... |
1157 1158 |
src_p = kmap_atomic(src_bv.bv_page); dst_p = kmap_atomic(dst_bv.bv_page); |
16ac3d63e block: Add bio_co... |
1159 |
|
1cb9dda4f block: Convert bi... |
1160 1161 |
memcpy(dst_p + dst_bv.bv_offset, src_p + src_bv.bv_offset, |
16ac3d63e block: Add bio_co... |
1162 1163 1164 1165 |
bytes); kunmap_atomic(dst_p); kunmap_atomic(src_p); |
6e6e811d7 block: Add missin... |
1166 |
flush_dcache_page(dst_bv.bv_page); |
45db54d58 block: Split out ... |
1167 1168 |
bio_advance_iter(src, src_iter, bytes); bio_advance_iter(dst, dst_iter, bytes); |
16ac3d63e block: Add bio_co... |
1169 1170 |
} } |
38a72dac4 block: Add bio_co... |
1171 1172 1173 |
EXPORT_SYMBOL(bio_copy_data_iter); /** |
45db54d58 block: Split out ... |
1174 1175 1176 |
* bio_copy_data - copy contents of data buffers from one bio to another * @src: source bio * @dst: destination bio |
38a72dac4 block: Add bio_co... |
1177 1178 1179 1180 1181 1182 |
* * Stops when it reaches the end of either @src or @dst - that is, copies * min(src->bi_size, dst->bi_size) bytes (or the equivalent for lists of bios). */ void bio_copy_data(struct bio *dst, struct bio *src) { |
45db54d58 block: Split out ... |
1183 1184 1185 1186 |
struct bvec_iter src_iter = src->bi_iter; struct bvec_iter dst_iter = dst->bi_iter; bio_copy_data_iter(dst, &dst_iter, src, &src_iter); |
38a72dac4 block: Add bio_co... |
1187 |
} |
16ac3d63e block: Add bio_co... |
1188 |
EXPORT_SYMBOL(bio_copy_data); |
45db54d58 block: Split out ... |
1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 |
/** * bio_list_copy_data - copy contents of data buffers from one chain of bios to * another * @src: source bio list * @dst: destination bio list * * Stops when it reaches the end of either the @src list or @dst list - that is, * copies min(src->bi_size, dst->bi_size) bytes (or the equivalent for lists of * bios). */ void bio_list_copy_data(struct bio *dst, struct bio *src) { struct bvec_iter src_iter = src->bi_iter; struct bvec_iter dst_iter = dst->bi_iter; while (1) { if (!src_iter.bi_size) { src = src->bi_next; if (!src) break; src_iter = src->bi_iter; } if (!dst_iter.bi_size) { dst = dst->bi_next; if (!dst) break; dst_iter = dst->bi_iter; } bio_copy_data_iter(dst, &dst_iter, src, &src_iter); } } EXPORT_SYMBOL(bio_list_copy_data); |
491221f88 block: export bio... |
1225 |
void bio_free_pages(struct bio *bio) |
1dfa0f68c block: add a help... |
1226 1227 |
{ struct bio_vec *bvec; |
6dc4f100c block: allow bio_... |
1228 |
struct bvec_iter_all iter_all; |
1dfa0f68c block: add a help... |
1229 |
|
2b070cfe5 block: remove the... |
1230 |
bio_for_each_segment_all(bvec, bio, iter_all) |
1dfa0f68c block: add a help... |
1231 1232 |
__free_page(bvec->bv_page); } |
491221f88 block: export bio... |
1233 |
EXPORT_SYMBOL(bio_free_pages); |
1dfa0f68c block: add a help... |
1234 |
|
1da177e4c Linux-2.6.12-rc2 |
1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 |
/* * bio_set_pages_dirty() and bio_check_pages_dirty() are support functions * for performing direct-IO in BIOs. * * The problem is that we cannot run set_page_dirty() from interrupt context * because the required locks are not interrupt-safe. So what we can do is to * mark the pages dirty _before_ performing IO. And in interrupt context, * check that the pages are still dirty. If so, fine. If not, redirty them * in process context. * * We special-case compound pages here: normally this means reads into hugetlb * pages. The logic in here doesn't really work right for compound pages * because the VM does not uniformly chase down the head page in all cases. * But dirtiness of compound pages is pretty meaningless anyway: the VM doesn't * handle them at all. So we skip compound pages here at an early stage. * * Note that this code is very hard to test under normal circumstances because * direct-io pins the pages with get_user_pages(). This makes * is_page_cache_freeable return false, and the VM will not clean the pages. |
0d5c3eba2 vfs: nuke pdflush... |
1254 |
* But other code (eg, flusher threads) could clean the pages if they are mapped |
1da177e4c Linux-2.6.12-rc2 |
1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 |
* pagecache. * * Simply disabling the call to bio_set_pages_dirty() is a good way to test the * deferred bio dirtying paths. */ /* * bio_set_pages_dirty() will mark all the bio's pages as dirty. */ void bio_set_pages_dirty(struct bio *bio) { |
cb34e057a block: Convert so... |
1266 |
struct bio_vec *bvec; |
6dc4f100c block: allow bio_... |
1267 |
struct bvec_iter_all iter_all; |
1da177e4c Linux-2.6.12-rc2 |
1268 |
|
2b070cfe5 block: remove the... |
1269 |
bio_for_each_segment_all(bvec, bio, iter_all) { |
3bb509831 block: bio_set_pa... |
1270 1271 |
if (!PageCompound(bvec->bv_page)) set_page_dirty_lock(bvec->bv_page); |
1da177e4c Linux-2.6.12-rc2 |
1272 1273 |
} } |
1da177e4c Linux-2.6.12-rc2 |
1274 1275 1276 1277 |
/* * bio_check_pages_dirty() will check that all the BIO's pages are still dirty. * If they are, then fine. If, however, some pages are clean then they must * have been written out during the direct-IO read. So we take another ref on |
24d5493f2 block: simplify b... |
1278 |
* the BIO and re-dirty the pages in process context. |
1da177e4c Linux-2.6.12-rc2 |
1279 1280 |
* * It is expected that bio_check_pages_dirty() will wholly own the BIO from |
ea1754a08 mm, fs: remove re... |
1281 1282 |
* here on. It will run one put_page() against each page and will run one * bio_put() against the BIO. |
1da177e4c Linux-2.6.12-rc2 |
1283 |
*/ |
65f27f384 WorkStruct: Pass ... |
1284 |
static void bio_dirty_fn(struct work_struct *work); |
1da177e4c Linux-2.6.12-rc2 |
1285 |
|
65f27f384 WorkStruct: Pass ... |
1286 |
static DECLARE_WORK(bio_dirty_work, bio_dirty_fn); |
1da177e4c Linux-2.6.12-rc2 |
1287 1288 1289 1290 1291 1292 |
static DEFINE_SPINLOCK(bio_dirty_lock); static struct bio *bio_dirty_list; /* * This runs in process context */ |
65f27f384 WorkStruct: Pass ... |
1293 |
static void bio_dirty_fn(struct work_struct *work) |
1da177e4c Linux-2.6.12-rc2 |
1294 |
{ |
24d5493f2 block: simplify b... |
1295 |
struct bio *bio, *next; |
1da177e4c Linux-2.6.12-rc2 |
1296 |
|
24d5493f2 block: simplify b... |
1297 1298 |
spin_lock_irq(&bio_dirty_lock); next = bio_dirty_list; |
1da177e4c Linux-2.6.12-rc2 |
1299 |
bio_dirty_list = NULL; |
24d5493f2 block: simplify b... |
1300 |
spin_unlock_irq(&bio_dirty_lock); |
1da177e4c Linux-2.6.12-rc2 |
1301 |
|
24d5493f2 block: simplify b... |
1302 1303 |
while ((bio = next) != NULL) { next = bio->bi_private; |
1da177e4c Linux-2.6.12-rc2 |
1304 |
|
d241a95f3 block: optionally... |
1305 |
bio_release_pages(bio, true); |
1da177e4c Linux-2.6.12-rc2 |
1306 |
bio_put(bio); |
1da177e4c Linux-2.6.12-rc2 |
1307 1308 1309 1310 1311 |
} } void bio_check_pages_dirty(struct bio *bio) { |
cb34e057a block: Convert so... |
1312 |
struct bio_vec *bvec; |
24d5493f2 block: simplify b... |
1313 |
unsigned long flags; |
6dc4f100c block: allow bio_... |
1314 |
struct bvec_iter_all iter_all; |
1da177e4c Linux-2.6.12-rc2 |
1315 |
|
2b070cfe5 block: remove the... |
1316 |
bio_for_each_segment_all(bvec, bio, iter_all) { |
24d5493f2 block: simplify b... |
1317 1318 |
if (!PageDirty(bvec->bv_page) && !PageCompound(bvec->bv_page)) goto defer; |
1da177e4c Linux-2.6.12-rc2 |
1319 |
} |
d241a95f3 block: optionally... |
1320 |
bio_release_pages(bio, false); |
24d5493f2 block: simplify b... |
1321 1322 1323 1324 1325 1326 1327 1328 |
bio_put(bio); return; defer: spin_lock_irqsave(&bio_dirty_lock, flags); bio->bi_private = bio_dirty_list; bio_dirty_list = bio; spin_unlock_irqrestore(&bio_dirty_lock, flags); schedule_work(&bio_dirty_work); |
1da177e4c Linux-2.6.12-rc2 |
1329 |
} |
c4cf5261f bio: skip atomic ... |
1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 |
static inline bool bio_remaining_done(struct bio *bio) { /* * If we're not chaining, then ->__bi_remaining is always 1 and * we always end io on the first invocation. */ if (!bio_flagged(bio, BIO_CHAIN)) return true; BUG_ON(atomic_read(&bio->__bi_remaining) <= 0); |
326e1dbb5 block: remove man... |
1340 |
if (atomic_dec_and_test(&bio->__bi_remaining)) { |
b7c44ed9d block: manipulate... |
1341 |
bio_clear_flag(bio, BIO_CHAIN); |
c4cf5261f bio: skip atomic ... |
1342 |
return true; |
326e1dbb5 block: remove man... |
1343 |
} |
c4cf5261f bio: skip atomic ... |
1344 1345 1346 |
return false; } |
1da177e4c Linux-2.6.12-rc2 |
1347 1348 1349 |
/** * bio_endio - end I/O on a bio * @bio: bio |
1da177e4c Linux-2.6.12-rc2 |
1350 1351 |
* * Description: |
4246a0b63 block: add a bi_e... |
1352 1353 1354 |
* bio_endio() will end I/O on the whole bio. bio_endio() is the preferred * way to end I/O on a bio. No one should call bi_end_io() directly on a * bio unless they own it and thus know that it has an end_io function. |
fbbaf700e block: trace comp... |
1355 1356 1357 1358 1359 |
* * bio_endio() can be called several times on a bio that has been chained * using bio_chain(). The ->bi_end_io() function will only be called the * last time. At this point the BLK_TA_COMPLETE tracing event will be * generated if BIO_TRACE_COMPLETION is set. |
1da177e4c Linux-2.6.12-rc2 |
1360 |
**/ |
4246a0b63 block: add a bi_e... |
1361 |
void bio_endio(struct bio *bio) |
1da177e4c Linux-2.6.12-rc2 |
1362 |
{ |
ba8c6967b block: cleanup bi... |
1363 |
again: |
2b8855171 block: bio_remain... |
1364 |
if (!bio_remaining_done(bio)) |
ba8c6967b block: cleanup bi... |
1365 |
return; |
7c20f1168 bio-integrity: st... |
1366 1367 |
if (!bio_integrity_endio(bio)) return; |
1da177e4c Linux-2.6.12-rc2 |
1368 |
|
67b42d0bf rq-qos: introduce... |
1369 1370 |
if (bio->bi_disk) rq_qos_done_bio(bio->bi_disk->queue, bio); |
ba8c6967b block: cleanup bi... |
1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 |
/* * Need to have a real endio function for chained bios, otherwise * various corner cases will break (like stacking block devices that * save/restore bi_end_io) - however, we want to avoid unbounded * recursion and blowing the stack. Tail call optimization would * handle this, but compiling with frame pointers also disables * gcc's sibling call optimization. */ if (bio->bi_end_io == bio_chain_endio) { bio = __bio_chain_endio(bio); goto again; |
196d38bcc block: Generic bi... |
1382 |
} |
ba8c6967b block: cleanup bi... |
1383 |
|
74d46992e block: replace bi... |
1384 |
if (bio->bi_disk && bio_flagged(bio, BIO_TRACE_COMPLETION)) { |
d24de76af block: remove the... |
1385 |
trace_block_bio_complete(bio->bi_disk->queue, bio); |
fbbaf700e block: trace comp... |
1386 1387 |
bio_clear_flag(bio, BIO_TRACE_COMPLETION); } |
9e234eeaf blk-throttle: add... |
1388 |
blk_throtl_bio_endio(bio); |
b222dd2fd block: call bio_u... |
1389 1390 |
/* release cgroup info */ bio_uninit(bio); |
ba8c6967b block: cleanup bi... |
1391 1392 |
if (bio->bi_end_io) bio->bi_end_io(bio); |
1da177e4c Linux-2.6.12-rc2 |
1393 |
} |
a112a71d4 fs/bio.c: move EX... |
1394 |
EXPORT_SYMBOL(bio_endio); |
1da177e4c Linux-2.6.12-rc2 |
1395 |
|
196d38bcc block: Generic bi... |
1396 |
/** |
20d0189b1 block: Introduce ... |
1397 1398 1399 1400 1401 1402 1403 1404 1405 |
* bio_split - split a bio * @bio: bio to split * @sectors: number of sectors to split from the front of @bio * @gfp: gfp mask * @bs: bio set to allocate from * * Allocates and returns a new bio which represents @sectors from the start of * @bio, and updates @bio to represent the remaining sectors. * |
f3f5da624 block: Do a full ... |
1406 |
* Unless this is a discard request the newly allocated bio will point |
dad775845 block: Document t... |
1407 1408 |
* to @bio's bi_io_vec. It is the caller's responsibility to ensure that * neither @bio nor @bs are freed before the split bio. |
20d0189b1 block: Introduce ... |
1409 1410 1411 1412 |
*/ struct bio *bio_split(struct bio *bio, int sectors, gfp_t gfp, struct bio_set *bs) { |
f341a4d38 block: remove use... |
1413 |
struct bio *split; |
20d0189b1 block: Introduce ... |
1414 1415 1416 |
BUG_ON(sectors <= 0); BUG_ON(sectors >= bio_sectors(bio)); |
0512a75b9 block: Introduce ... |
1417 1418 1419 |
/* Zone append commands cannot be split */ if (WARN_ON_ONCE(bio_op(bio) == REQ_OP_ZONE_APPEND)) return NULL; |
f9d03f96b block: improve ha... |
1420 |
split = bio_clone_fast(bio, gfp, bs); |
20d0189b1 block: Introduce ... |
1421 1422 1423 1424 1425 1426 |
if (!split) return NULL; split->bi_iter.bi_size = sectors << 9; if (bio_integrity(split)) |
fbd08e767 bio-integrity: fi... |
1427 |
bio_integrity_trim(split); |
20d0189b1 block: Introduce ... |
1428 1429 |
bio_advance(bio, split->bi_iter.bi_size); |
fbbaf700e block: trace comp... |
1430 |
if (bio_flagged(bio, BIO_TRACE_COMPLETION)) |
20d59023c block: Set BIO_TR... |
1431 |
bio_set_flag(split, BIO_TRACE_COMPLETION); |
fbbaf700e block: trace comp... |
1432 |
|
20d0189b1 block: Introduce ... |
1433 1434 1435 |
return split; } EXPORT_SYMBOL(bio_split); |
ad3316bf4 block: Find bio s... |
1436 |
/** |
6678d83f1 block: Consolidat... |
1437 1438 1439 1440 1441 1442 1443 1444 1445 |
* bio_trim - trim a bio * @bio: bio to trim * @offset: number of sectors to trim from the front of @bio * @size: size we want to trim @bio to, in sectors */ void bio_trim(struct bio *bio, int offset, int size) { /* 'bio' is a cloned bio which we need to trim to match * the given offset and size. |
6678d83f1 block: Consolidat... |
1446 |
*/ |
6678d83f1 block: Consolidat... |
1447 1448 |
size <<= 9; |
4f024f379 block: Abstract o... |
1449 |
if (offset == 0 && size == bio->bi_iter.bi_size) |
6678d83f1 block: Consolidat... |
1450 |
return; |
6678d83f1 block: Consolidat... |
1451 |
bio_advance(bio, offset << 9); |
4f024f379 block: Abstract o... |
1452 |
bio->bi_iter.bi_size = size; |
376a78abf bio-integrity: bi... |
1453 1454 |
if (bio_integrity(bio)) |
fbd08e767 bio-integrity: fi... |
1455 |
bio_integrity_trim(bio); |
376a78abf bio-integrity: bi... |
1456 |
|
6678d83f1 block: Consolidat... |
1457 1458 |
} EXPORT_SYMBOL_GPL(bio_trim); |
1da177e4c Linux-2.6.12-rc2 |
1459 1460 1461 1462 |
/* * create memory pools for biovec's in a bio_set. * use the global biovec slabs created for general use. */ |
8aa6ba2f6 block: Convert bi... |
1463 |
int biovec_init_pool(mempool_t *pool, int pool_entries) |
1da177e4c Linux-2.6.12-rc2 |
1464 |
{ |
ed996a52c block: simplify a... |
1465 |
struct biovec_slab *bp = bvec_slabs + BVEC_POOL_MAX; |
1da177e4c Linux-2.6.12-rc2 |
1466 |
|
8aa6ba2f6 block: Convert bi... |
1467 |
return mempool_init_slab_pool(pool, pool_entries, bp->slab); |
1da177e4c Linux-2.6.12-rc2 |
1468 |
} |
917a38c71 block: Add bioset... |
1469 1470 1471 1472 1473 1474 1475 |
/* * bioset_exit - exit a bioset initialized with bioset_init() * * May be called on a zeroed but uninitialized bioset (i.e. allocated with * kzalloc()). */ void bioset_exit(struct bio_set *bs) |
1da177e4c Linux-2.6.12-rc2 |
1476 |
{ |
df2cb6daa block: Avoid dead... |
1477 1478 |
if (bs->rescue_workqueue) destroy_workqueue(bs->rescue_workqueue); |
917a38c71 block: Add bioset... |
1479 |
bs->rescue_workqueue = NULL; |
df2cb6daa block: Avoid dead... |
1480 |
|
8aa6ba2f6 block: Convert bi... |
1481 1482 |
mempool_exit(&bs->bio_pool); mempool_exit(&bs->bvec_pool); |
9f060e223 block: Convert in... |
1483 |
|
7878cba9f block: Create bip... |
1484 |
bioset_integrity_free(bs); |
917a38c71 block: Add bioset... |
1485 1486 1487 1488 1489 |
if (bs->bio_slab) bio_put_slab(bs); bs->bio_slab = NULL; } EXPORT_SYMBOL(bioset_exit); |
1da177e4c Linux-2.6.12-rc2 |
1490 |
|
011067b05 blk: replace bios... |
1491 |
/** |
917a38c71 block: Add bioset... |
1492 |
* bioset_init - Initialize a bio_set |
dad085275 block: Drop biose... |
1493 |
* @bs: pool to initialize |
917a38c71 block: Add bioset... |
1494 1495 1496 1497 1498 |
* @pool_size: Number of bio and bio_vecs to cache in the mempool * @front_pad: Number of bytes to allocate in front of the returned bio * @flags: Flags to modify behavior, currently %BIOSET_NEED_BVECS * and %BIOSET_NEED_RESCUER * |
dad085275 block: Drop biose... |
1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 |
* Description: * Set up a bio_set to be used with @bio_alloc_bioset. Allows the caller * to ask for a number of bytes to be allocated in front of the bio. * Front pad allocation is useful for embedding the bio inside * another structure, to avoid allocating extra data to go with the bio. * Note that the bio must be embedded at the END of that structure always, * or things will break badly. * If %BIOSET_NEED_BVECS is set in @flags, a separate pool will be allocated * for allocating iovecs. This pool is not needed e.g. for bio_clone_fast(). * If %BIOSET_NEED_RESCUER is set, a workqueue is created which can be used to * dispatch queued requests when the mempool runs out of space. * |
917a38c71 block: Add bioset... |
1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 |
*/ int bioset_init(struct bio_set *bs, unsigned int pool_size, unsigned int front_pad, int flags) { unsigned int back_pad = BIO_INLINE_VECS * sizeof(struct bio_vec); bs->front_pad = front_pad; spin_lock_init(&bs->rescue_lock); bio_list_init(&bs->rescue_list); INIT_WORK(&bs->rescue_work, bio_alloc_rescue); bs->bio_slab = bio_find_or_create_slab(front_pad + back_pad); if (!bs->bio_slab) return -ENOMEM; if (mempool_init_slab_pool(&bs->bio_pool, pool_size, bs->bio_slab)) goto bad; if ((flags & BIOSET_NEED_BVECS) && biovec_init_pool(&bs->bvec_pool, pool_size)) goto bad; if (!(flags & BIOSET_NEED_RESCUER)) return 0; bs->rescue_workqueue = alloc_workqueue("bioset", WQ_MEM_RECLAIM, 0); if (!bs->rescue_workqueue) goto bad; return 0; bad: bioset_exit(bs); return -ENOMEM; } EXPORT_SYMBOL(bioset_init); |
28e89fd91 block: add bioset... |
1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 |
/* * Initialize and setup a new bio_set, based on the settings from * another bio_set. */ int bioset_init_from_src(struct bio_set *bs, struct bio_set *src) { int flags; flags = 0; if (src->bvec_pool.min_nr) flags |= BIOSET_NEED_BVECS; if (src->rescue_workqueue) flags |= BIOSET_NEED_RESCUER; return bioset_init(bs, src->bio_pool.min_nr, src->front_pad, flags); } EXPORT_SYMBOL(bioset_init_from_src); |
1da177e4c Linux-2.6.12-rc2 |
1566 1567 1568 |
static void __init biovec_init_slabs(void) { int i; |
ed996a52c block: simplify a... |
1569 |
for (i = 0; i < BVEC_POOL_NR; i++) { |
1da177e4c Linux-2.6.12-rc2 |
1570 1571 |
int size; struct biovec_slab *bvs = bvec_slabs + i; |
a7fcd37cd block: don't crea... |
1572 1573 1574 1575 |
if (bvs->nr_vecs <= BIO_INLINE_VECS) { bvs->slab = NULL; continue; } |
a7fcd37cd block: don't crea... |
1576 |
|
1da177e4c Linux-2.6.12-rc2 |
1577 1578 |
size = bvs->nr_vecs * sizeof(struct bio_vec); bvs->slab = kmem_cache_create(bvs->name, size, 0, |
20c2df83d mm: Remove slab d... |
1579 |
SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); |
1da177e4c Linux-2.6.12-rc2 |
1580 1581 1582 1583 1584 |
} } static int __init init_bio(void) { |
bb799ca02 bio: allow indivi... |
1585 1586 |
bio_slab_max = 2; bio_slab_nr = 0; |
6396bb221 treewide: kzalloc... |
1587 1588 |
bio_slabs = kcalloc(bio_slab_max, sizeof(struct bio_slab), GFP_KERNEL); |
2b24e6f63 block: bio: ensur... |
1589 1590 |
BUILD_BUG_ON(BIO_FLAG_LAST > BVEC_POOL_OFFSET); |
bb799ca02 bio: allow indivi... |
1591 1592 1593 |
if (!bio_slabs) panic("bio: can't allocate bios "); |
1da177e4c Linux-2.6.12-rc2 |
1594 |
|
7878cba9f block: Create bip... |
1595 |
bio_integrity_init(); |
1da177e4c Linux-2.6.12-rc2 |
1596 |
biovec_init_slabs(); |
f4f8154a0 block: Use bioset... |
1597 |
if (bioset_init(&fs_bio_set, BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS)) |
1da177e4c Linux-2.6.12-rc2 |
1598 1599 |
panic("bio: can't allocate bios "); |
f4f8154a0 block: Use bioset... |
1600 |
if (bioset_integrity_create(&fs_bio_set, BIO_POOL_SIZE)) |
a91a2785b block: Require su... |
1601 1602 |
panic("bio: can't create integrity pool "); |
1da177e4c Linux-2.6.12-rc2 |
1603 1604 |
return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1605 |
subsys_initcall(init_bio); |