]>
Commit | Line | Data |
---|---|---|
c1d7c514 | 1 | // SPDX-License-Identifier: GPL-2.0 |
0f9dd46c JB |
2 | /* |
3 | * Copyright (C) 2008 Red Hat. All rights reserved. | |
0f9dd46c JB |
4 | */ |
5 | ||
96303081 | 6 | #include <linux/pagemap.h> |
0f9dd46c | 7 | #include <linux/sched.h> |
f361bf4a | 8 | #include <linux/sched/signal.h> |
5a0e3ad6 | 9 | #include <linux/slab.h> |
96303081 | 10 | #include <linux/math64.h> |
6ab60601 | 11 | #include <linux/ratelimit.h> |
540adea3 | 12 | #include <linux/error-injection.h> |
84de76a2 | 13 | #include <linux/sched/mm.h> |
0f9dd46c | 14 | #include "ctree.h" |
fa9c0d79 CM |
15 | #include "free-space-cache.h" |
16 | #include "transaction.h" | |
0af3d00b | 17 | #include "disk-io.h" |
43be2146 | 18 | #include "extent_io.h" |
581bb050 | 19 | #include "inode-map.h" |
04216820 | 20 | #include "volumes.h" |
8719aaae | 21 | #include "space-info.h" |
86736342 | 22 | #include "delalloc-space.h" |
aac0023c | 23 | #include "block-group.h" |
b0643e59 | 24 | #include "discard.h" |
fa9c0d79 | 25 | |
0ef6447a | 26 | #define BITS_PER_BITMAP (PAGE_SIZE * 8UL) |
5d90c5c7 DZ |
27 | #define MAX_CACHE_BYTES_PER_GIG SZ_64K |
28 | #define FORCE_EXTENT_THRESHOLD SZ_1M | |
0f9dd46c | 29 | |
55507ce3 FM |
30 | struct btrfs_trim_range { |
31 | u64 start; | |
32 | u64 bytes; | |
33 | struct list_head list; | |
34 | }; | |
35 | ||
dfb79ddb DZ |
36 | static int count_bitmap_extents(struct btrfs_free_space_ctl *ctl, |
37 | struct btrfs_free_space *bitmap_info); | |
34d52cb6 | 38 | static int link_free_space(struct btrfs_free_space_ctl *ctl, |
0cb59c99 | 39 | struct btrfs_free_space *info); |
cd023e7b JB |
40 | static void unlink_free_space(struct btrfs_free_space_ctl *ctl, |
41 | struct btrfs_free_space *info); | |
afdb5718 JM |
42 | static int btrfs_wait_cache_io_root(struct btrfs_root *root, |
43 | struct btrfs_trans_handle *trans, | |
44 | struct btrfs_io_ctl *io_ctl, | |
45 | struct btrfs_path *path); | |
0cb59c99 | 46 | |
0414efae LZ |
47 | static struct inode *__lookup_free_space_inode(struct btrfs_root *root, |
48 | struct btrfs_path *path, | |
49 | u64 offset) | |
0af3d00b | 50 | { |
0b246afa | 51 | struct btrfs_fs_info *fs_info = root->fs_info; |
0af3d00b JB |
52 | struct btrfs_key key; |
53 | struct btrfs_key location; | |
54 | struct btrfs_disk_key disk_key; | |
55 | struct btrfs_free_space_header *header; | |
56 | struct extent_buffer *leaf; | |
57 | struct inode *inode = NULL; | |
84de76a2 | 58 | unsigned nofs_flag; |
0af3d00b JB |
59 | int ret; |
60 | ||
0af3d00b | 61 | key.objectid = BTRFS_FREE_SPACE_OBJECTID; |
0414efae | 62 | key.offset = offset; |
0af3d00b JB |
63 | key.type = 0; |
64 | ||
65 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
66 | if (ret < 0) | |
67 | return ERR_PTR(ret); | |
68 | if (ret > 0) { | |
b3b4aa74 | 69 | btrfs_release_path(path); |
0af3d00b JB |
70 | return ERR_PTR(-ENOENT); |
71 | } | |
72 | ||
73 | leaf = path->nodes[0]; | |
74 | header = btrfs_item_ptr(leaf, path->slots[0], | |
75 | struct btrfs_free_space_header); | |
76 | btrfs_free_space_key(leaf, header, &disk_key); | |
77 | btrfs_disk_key_to_cpu(&location, &disk_key); | |
b3b4aa74 | 78 | btrfs_release_path(path); |
0af3d00b | 79 | |
84de76a2 JB |
80 | /* |
81 | * We are often under a trans handle at this point, so we need to make | |
82 | * sure NOFS is set to keep us from deadlocking. | |
83 | */ | |
84 | nofs_flag = memalloc_nofs_save(); | |
0202e83f | 85 | inode = btrfs_iget_path(fs_info->sb, location.objectid, root, path); |
4222ea71 | 86 | btrfs_release_path(path); |
84de76a2 | 87 | memalloc_nofs_restore(nofs_flag); |
0af3d00b JB |
88 | if (IS_ERR(inode)) |
89 | return inode; | |
0af3d00b | 90 | |
528c0327 | 91 | mapping_set_gfp_mask(inode->i_mapping, |
c62d2555 MH |
92 | mapping_gfp_constraint(inode->i_mapping, |
93 | ~(__GFP_FS | __GFP_HIGHMEM))); | |
adae52b9 | 94 | |
0414efae LZ |
95 | return inode; |
96 | } | |
97 | ||
32da5386 | 98 | struct inode *lookup_free_space_inode(struct btrfs_block_group *block_group, |
7949f339 | 99 | struct btrfs_path *path) |
0414efae | 100 | { |
7949f339 | 101 | struct btrfs_fs_info *fs_info = block_group->fs_info; |
0414efae | 102 | struct inode *inode = NULL; |
5b0e95bf | 103 | u32 flags = BTRFS_INODE_NODATASUM | BTRFS_INODE_NODATACOW; |
0414efae LZ |
104 | |
105 | spin_lock(&block_group->lock); | |
106 | if (block_group->inode) | |
107 | inode = igrab(block_group->inode); | |
108 | spin_unlock(&block_group->lock); | |
109 | if (inode) | |
110 | return inode; | |
111 | ||
77ab86bf | 112 | inode = __lookup_free_space_inode(fs_info->tree_root, path, |
b3470b5d | 113 | block_group->start); |
0414efae LZ |
114 | if (IS_ERR(inode)) |
115 | return inode; | |
116 | ||
0af3d00b | 117 | spin_lock(&block_group->lock); |
5b0e95bf | 118 | if (!((BTRFS_I(inode)->flags & flags) == flags)) { |
0b246afa | 119 | btrfs_info(fs_info, "Old style space inode found, converting."); |
5b0e95bf JB |
120 | BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM | |
121 | BTRFS_INODE_NODATACOW; | |
2f356126 JB |
122 | block_group->disk_cache_state = BTRFS_DC_CLEAR; |
123 | } | |
124 | ||
300e4f8a | 125 | if (!block_group->iref) { |
0af3d00b JB |
126 | block_group->inode = igrab(inode); |
127 | block_group->iref = 1; | |
128 | } | |
129 | spin_unlock(&block_group->lock); | |
130 | ||
131 | return inode; | |
132 | } | |
133 | ||
48a3b636 ES |
134 | static int __create_free_space_inode(struct btrfs_root *root, |
135 | struct btrfs_trans_handle *trans, | |
136 | struct btrfs_path *path, | |
137 | u64 ino, u64 offset) | |
0af3d00b JB |
138 | { |
139 | struct btrfs_key key; | |
140 | struct btrfs_disk_key disk_key; | |
141 | struct btrfs_free_space_header *header; | |
142 | struct btrfs_inode_item *inode_item; | |
143 | struct extent_buffer *leaf; | |
5b0e95bf | 144 | u64 flags = BTRFS_INODE_NOCOMPRESS | BTRFS_INODE_PREALLOC; |
0af3d00b JB |
145 | int ret; |
146 | ||
0414efae | 147 | ret = btrfs_insert_empty_inode(trans, root, path, ino); |
0af3d00b JB |
148 | if (ret) |
149 | return ret; | |
150 | ||
5b0e95bf JB |
151 | /* We inline crc's for the free disk space cache */ |
152 | if (ino != BTRFS_FREE_INO_OBJECTID) | |
153 | flags |= BTRFS_INODE_NODATASUM | BTRFS_INODE_NODATACOW; | |
154 | ||
0af3d00b JB |
155 | leaf = path->nodes[0]; |
156 | inode_item = btrfs_item_ptr(leaf, path->slots[0], | |
157 | struct btrfs_inode_item); | |
158 | btrfs_item_key(leaf, &disk_key, path->slots[0]); | |
b159fa28 | 159 | memzero_extent_buffer(leaf, (unsigned long)inode_item, |
0af3d00b JB |
160 | sizeof(*inode_item)); |
161 | btrfs_set_inode_generation(leaf, inode_item, trans->transid); | |
162 | btrfs_set_inode_size(leaf, inode_item, 0); | |
163 | btrfs_set_inode_nbytes(leaf, inode_item, 0); | |
164 | btrfs_set_inode_uid(leaf, inode_item, 0); | |
165 | btrfs_set_inode_gid(leaf, inode_item, 0); | |
166 | btrfs_set_inode_mode(leaf, inode_item, S_IFREG | 0600); | |
5b0e95bf | 167 | btrfs_set_inode_flags(leaf, inode_item, flags); |
0af3d00b JB |
168 | btrfs_set_inode_nlink(leaf, inode_item, 1); |
169 | btrfs_set_inode_transid(leaf, inode_item, trans->transid); | |
0414efae | 170 | btrfs_set_inode_block_group(leaf, inode_item, offset); |
0af3d00b | 171 | btrfs_mark_buffer_dirty(leaf); |
b3b4aa74 | 172 | btrfs_release_path(path); |
0af3d00b JB |
173 | |
174 | key.objectid = BTRFS_FREE_SPACE_OBJECTID; | |
0414efae | 175 | key.offset = offset; |
0af3d00b | 176 | key.type = 0; |
0af3d00b JB |
177 | ret = btrfs_insert_empty_item(trans, root, path, &key, |
178 | sizeof(struct btrfs_free_space_header)); | |
179 | if (ret < 0) { | |
b3b4aa74 | 180 | btrfs_release_path(path); |
0af3d00b JB |
181 | return ret; |
182 | } | |
c9dc4c65 | 183 | |
0af3d00b JB |
184 | leaf = path->nodes[0]; |
185 | header = btrfs_item_ptr(leaf, path->slots[0], | |
186 | struct btrfs_free_space_header); | |
b159fa28 | 187 | memzero_extent_buffer(leaf, (unsigned long)header, sizeof(*header)); |
0af3d00b JB |
188 | btrfs_set_free_space_key(leaf, header, &disk_key); |
189 | btrfs_mark_buffer_dirty(leaf); | |
b3b4aa74 | 190 | btrfs_release_path(path); |
0af3d00b JB |
191 | |
192 | return 0; | |
193 | } | |
194 | ||
4ca75f1b | 195 | int create_free_space_inode(struct btrfs_trans_handle *trans, |
32da5386 | 196 | struct btrfs_block_group *block_group, |
0414efae LZ |
197 | struct btrfs_path *path) |
198 | { | |
199 | int ret; | |
200 | u64 ino; | |
201 | ||
4ca75f1b | 202 | ret = btrfs_find_free_objectid(trans->fs_info->tree_root, &ino); |
0414efae LZ |
203 | if (ret < 0) |
204 | return ret; | |
205 | ||
4ca75f1b | 206 | return __create_free_space_inode(trans->fs_info->tree_root, trans, path, |
b3470b5d | 207 | ino, block_group->start); |
0414efae LZ |
208 | } |
209 | ||
2ff7e61e | 210 | int btrfs_check_trunc_cache_free_space(struct btrfs_fs_info *fs_info, |
7b61cd92 | 211 | struct btrfs_block_rsv *rsv) |
0af3d00b | 212 | { |
c8174313 | 213 | u64 needed_bytes; |
7b61cd92 | 214 | int ret; |
c8174313 JB |
215 | |
216 | /* 1 for slack space, 1 for updating the inode */ | |
2bd36e7b JB |
217 | needed_bytes = btrfs_calc_insert_metadata_size(fs_info, 1) + |
218 | btrfs_calc_metadata_size(fs_info, 1); | |
c8174313 | 219 | |
7b61cd92 MX |
220 | spin_lock(&rsv->lock); |
221 | if (rsv->reserved < needed_bytes) | |
222 | ret = -ENOSPC; | |
223 | else | |
224 | ret = 0; | |
225 | spin_unlock(&rsv->lock); | |
4b286cd1 | 226 | return ret; |
7b61cd92 MX |
227 | } |
228 | ||
77ab86bf | 229 | int btrfs_truncate_free_space_cache(struct btrfs_trans_handle *trans, |
32da5386 | 230 | struct btrfs_block_group *block_group, |
7b61cd92 MX |
231 | struct inode *inode) |
232 | { | |
77ab86bf | 233 | struct btrfs_root *root = BTRFS_I(inode)->root; |
7b61cd92 | 234 | int ret = 0; |
35c76642 | 235 | bool locked = false; |
1bbc621e | 236 | |
1bbc621e | 237 | if (block_group) { |
21e75ffe JM |
238 | struct btrfs_path *path = btrfs_alloc_path(); |
239 | ||
240 | if (!path) { | |
241 | ret = -ENOMEM; | |
242 | goto fail; | |
243 | } | |
35c76642 | 244 | locked = true; |
1bbc621e CM |
245 | mutex_lock(&trans->transaction->cache_write_mutex); |
246 | if (!list_empty(&block_group->io_list)) { | |
247 | list_del_init(&block_group->io_list); | |
248 | ||
afdb5718 | 249 | btrfs_wait_cache_io(trans, block_group, path); |
1bbc621e CM |
250 | btrfs_put_block_group(block_group); |
251 | } | |
252 | ||
253 | /* | |
254 | * now that we've truncated the cache away, its no longer | |
255 | * setup or written | |
256 | */ | |
257 | spin_lock(&block_group->lock); | |
258 | block_group->disk_cache_state = BTRFS_DC_CLEAR; | |
259 | spin_unlock(&block_group->lock); | |
21e75ffe | 260 | btrfs_free_path(path); |
1bbc621e | 261 | } |
0af3d00b | 262 | |
6ef06d27 | 263 | btrfs_i_size_write(BTRFS_I(inode), 0); |
7caef267 | 264 | truncate_pagecache(inode, 0); |
0af3d00b JB |
265 | |
266 | /* | |
f7e9e8fc OS |
267 | * We skip the throttling logic for free space cache inodes, so we don't |
268 | * need to check for -EAGAIN. | |
0af3d00b JB |
269 | */ |
270 | ret = btrfs_truncate_inode_items(trans, root, inode, | |
271 | 0, BTRFS_EXTENT_DATA_KEY); | |
35c76642 FM |
272 | if (ret) |
273 | goto fail; | |
0af3d00b | 274 | |
82d5902d | 275 | ret = btrfs_update_inode(trans, root, inode); |
1bbc621e | 276 | |
1bbc621e | 277 | fail: |
35c76642 FM |
278 | if (locked) |
279 | mutex_unlock(&trans->transaction->cache_write_mutex); | |
79787eaa | 280 | if (ret) |
66642832 | 281 | btrfs_abort_transaction(trans, ret); |
c8174313 | 282 | |
82d5902d | 283 | return ret; |
0af3d00b JB |
284 | } |
285 | ||
1d480538 | 286 | static void readahead_cache(struct inode *inode) |
9d66e233 JB |
287 | { |
288 | struct file_ra_state *ra; | |
289 | unsigned long last_index; | |
290 | ||
291 | ra = kzalloc(sizeof(*ra), GFP_NOFS); | |
292 | if (!ra) | |
1d480538 | 293 | return; |
9d66e233 JB |
294 | |
295 | file_ra_state_init(ra, inode->i_mapping); | |
09cbfeaf | 296 | last_index = (i_size_read(inode) - 1) >> PAGE_SHIFT; |
9d66e233 JB |
297 | |
298 | page_cache_sync_readahead(inode->i_mapping, ra, NULL, 0, last_index); | |
299 | ||
300 | kfree(ra); | |
9d66e233 JB |
301 | } |
302 | ||
4c6d1d85 | 303 | static int io_ctl_init(struct btrfs_io_ctl *io_ctl, struct inode *inode, |
f15376df | 304 | int write) |
a67509c3 | 305 | { |
5349d6c3 MX |
306 | int num_pages; |
307 | int check_crcs = 0; | |
308 | ||
09cbfeaf | 309 | num_pages = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); |
5349d6c3 | 310 | |
4a0cc7ca | 311 | if (btrfs_ino(BTRFS_I(inode)) != BTRFS_FREE_INO_OBJECTID) |
5349d6c3 MX |
312 | check_crcs = 1; |
313 | ||
8f6c72a9 | 314 | /* Make sure we can fit our crcs and generation into the first page */ |
5349d6c3 | 315 | if (write && check_crcs && |
8f6c72a9 | 316 | (num_pages * sizeof(u32) + sizeof(u64)) > PAGE_SIZE) |
5349d6c3 MX |
317 | return -ENOSPC; |
318 | ||
4c6d1d85 | 319 | memset(io_ctl, 0, sizeof(struct btrfs_io_ctl)); |
5349d6c3 | 320 | |
31e818fe | 321 | io_ctl->pages = kcalloc(num_pages, sizeof(struct page *), GFP_NOFS); |
a67509c3 JB |
322 | if (!io_ctl->pages) |
323 | return -ENOMEM; | |
5349d6c3 MX |
324 | |
325 | io_ctl->num_pages = num_pages; | |
f15376df | 326 | io_ctl->fs_info = btrfs_sb(inode->i_sb); |
5349d6c3 | 327 | io_ctl->check_crcs = check_crcs; |
c9dc4c65 | 328 | io_ctl->inode = inode; |
5349d6c3 | 329 | |
a67509c3 JB |
330 | return 0; |
331 | } | |
663faf9f | 332 | ALLOW_ERROR_INJECTION(io_ctl_init, ERRNO); |
a67509c3 | 333 | |
4c6d1d85 | 334 | static void io_ctl_free(struct btrfs_io_ctl *io_ctl) |
a67509c3 JB |
335 | { |
336 | kfree(io_ctl->pages); | |
c9dc4c65 | 337 | io_ctl->pages = NULL; |
a67509c3 JB |
338 | } |
339 | ||
4c6d1d85 | 340 | static void io_ctl_unmap_page(struct btrfs_io_ctl *io_ctl) |
a67509c3 JB |
341 | { |
342 | if (io_ctl->cur) { | |
a67509c3 JB |
343 | io_ctl->cur = NULL; |
344 | io_ctl->orig = NULL; | |
345 | } | |
346 | } | |
347 | ||
4c6d1d85 | 348 | static void io_ctl_map_page(struct btrfs_io_ctl *io_ctl, int clear) |
a67509c3 | 349 | { |
b12d6869 | 350 | ASSERT(io_ctl->index < io_ctl->num_pages); |
a67509c3 | 351 | io_ctl->page = io_ctl->pages[io_ctl->index++]; |
2b108268 | 352 | io_ctl->cur = page_address(io_ctl->page); |
a67509c3 | 353 | io_ctl->orig = io_ctl->cur; |
09cbfeaf | 354 | io_ctl->size = PAGE_SIZE; |
a67509c3 | 355 | if (clear) |
619a9742 | 356 | clear_page(io_ctl->cur); |
a67509c3 JB |
357 | } |
358 | ||
4c6d1d85 | 359 | static void io_ctl_drop_pages(struct btrfs_io_ctl *io_ctl) |
a67509c3 JB |
360 | { |
361 | int i; | |
362 | ||
363 | io_ctl_unmap_page(io_ctl); | |
364 | ||
365 | for (i = 0; i < io_ctl->num_pages; i++) { | |
a1ee5a45 LZ |
366 | if (io_ctl->pages[i]) { |
367 | ClearPageChecked(io_ctl->pages[i]); | |
368 | unlock_page(io_ctl->pages[i]); | |
09cbfeaf | 369 | put_page(io_ctl->pages[i]); |
a1ee5a45 | 370 | } |
a67509c3 JB |
371 | } |
372 | } | |
373 | ||
7a195f6d | 374 | static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate) |
a67509c3 JB |
375 | { |
376 | struct page *page; | |
831fa14f | 377 | struct inode *inode = io_ctl->inode; |
a67509c3 JB |
378 | gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); |
379 | int i; | |
380 | ||
381 | for (i = 0; i < io_ctl->num_pages; i++) { | |
382 | page = find_or_create_page(inode->i_mapping, i, mask); | |
383 | if (!page) { | |
384 | io_ctl_drop_pages(io_ctl); | |
385 | return -ENOMEM; | |
386 | } | |
387 | io_ctl->pages[i] = page; | |
388 | if (uptodate && !PageUptodate(page)) { | |
389 | btrfs_readpage(NULL, page); | |
390 | lock_page(page); | |
3797136b JB |
391 | if (page->mapping != inode->i_mapping) { |
392 | btrfs_err(BTRFS_I(inode)->root->fs_info, | |
393 | "free space cache page truncated"); | |
394 | io_ctl_drop_pages(io_ctl); | |
395 | return -EIO; | |
396 | } | |
a67509c3 | 397 | if (!PageUptodate(page)) { |
efe120a0 FH |
398 | btrfs_err(BTRFS_I(inode)->root->fs_info, |
399 | "error reading free space cache"); | |
a67509c3 JB |
400 | io_ctl_drop_pages(io_ctl); |
401 | return -EIO; | |
402 | } | |
403 | } | |
404 | } | |
405 | ||
f7d61dcd JB |
406 | for (i = 0; i < io_ctl->num_pages; i++) { |
407 | clear_page_dirty_for_io(io_ctl->pages[i]); | |
408 | set_page_extent_mapped(io_ctl->pages[i]); | |
409 | } | |
410 | ||
a67509c3 JB |
411 | return 0; |
412 | } | |
413 | ||
4c6d1d85 | 414 | static void io_ctl_set_generation(struct btrfs_io_ctl *io_ctl, u64 generation) |
a67509c3 | 415 | { |
528c0327 | 416 | __le64 *val; |
a67509c3 JB |
417 | |
418 | io_ctl_map_page(io_ctl, 1); | |
419 | ||
420 | /* | |
5b0e95bf JB |
421 | * Skip the csum areas. If we don't check crcs then we just have a |
422 | * 64bit chunk at the front of the first page. | |
a67509c3 | 423 | */ |
5b0e95bf JB |
424 | if (io_ctl->check_crcs) { |
425 | io_ctl->cur += (sizeof(u32) * io_ctl->num_pages); | |
426 | io_ctl->size -= sizeof(u64) + (sizeof(u32) * io_ctl->num_pages); | |
427 | } else { | |
428 | io_ctl->cur += sizeof(u64); | |
429 | io_ctl->size -= sizeof(u64) * 2; | |
430 | } | |
a67509c3 JB |
431 | |
432 | val = io_ctl->cur; | |
433 | *val = cpu_to_le64(generation); | |
434 | io_ctl->cur += sizeof(u64); | |
a67509c3 JB |
435 | } |
436 | ||
4c6d1d85 | 437 | static int io_ctl_check_generation(struct btrfs_io_ctl *io_ctl, u64 generation) |
a67509c3 | 438 | { |
528c0327 | 439 | __le64 *gen; |
a67509c3 | 440 | |
5b0e95bf JB |
441 | /* |
442 | * Skip the crc area. If we don't check crcs then we just have a 64bit | |
443 | * chunk at the front of the first page. | |
444 | */ | |
445 | if (io_ctl->check_crcs) { | |
446 | io_ctl->cur += sizeof(u32) * io_ctl->num_pages; | |
447 | io_ctl->size -= sizeof(u64) + | |
448 | (sizeof(u32) * io_ctl->num_pages); | |
449 | } else { | |
450 | io_ctl->cur += sizeof(u64); | |
451 | io_ctl->size -= sizeof(u64) * 2; | |
452 | } | |
a67509c3 | 453 | |
a67509c3 JB |
454 | gen = io_ctl->cur; |
455 | if (le64_to_cpu(*gen) != generation) { | |
f15376df | 456 | btrfs_err_rl(io_ctl->fs_info, |
94647322 DS |
457 | "space cache generation (%llu) does not match inode (%llu)", |
458 | *gen, generation); | |
a67509c3 JB |
459 | io_ctl_unmap_page(io_ctl); |
460 | return -EIO; | |
461 | } | |
462 | io_ctl->cur += sizeof(u64); | |
5b0e95bf JB |
463 | return 0; |
464 | } | |
465 | ||
4c6d1d85 | 466 | static void io_ctl_set_crc(struct btrfs_io_ctl *io_ctl, int index) |
5b0e95bf JB |
467 | { |
468 | u32 *tmp; | |
469 | u32 crc = ~(u32)0; | |
470 | unsigned offset = 0; | |
471 | ||
472 | if (!io_ctl->check_crcs) { | |
473 | io_ctl_unmap_page(io_ctl); | |
474 | return; | |
475 | } | |
476 | ||
477 | if (index == 0) | |
cb54f257 | 478 | offset = sizeof(u32) * io_ctl->num_pages; |
5b0e95bf | 479 | |
4bb3c2e2 JT |
480 | crc = btrfs_crc32c(crc, io_ctl->orig + offset, PAGE_SIZE - offset); |
481 | btrfs_crc32c_final(crc, (u8 *)&crc); | |
5b0e95bf | 482 | io_ctl_unmap_page(io_ctl); |
2b108268 | 483 | tmp = page_address(io_ctl->pages[0]); |
5b0e95bf JB |
484 | tmp += index; |
485 | *tmp = crc; | |
5b0e95bf JB |
486 | } |
487 | ||
4c6d1d85 | 488 | static int io_ctl_check_crc(struct btrfs_io_ctl *io_ctl, int index) |
5b0e95bf JB |
489 | { |
490 | u32 *tmp, val; | |
491 | u32 crc = ~(u32)0; | |
492 | unsigned offset = 0; | |
493 | ||
494 | if (!io_ctl->check_crcs) { | |
495 | io_ctl_map_page(io_ctl, 0); | |
496 | return 0; | |
497 | } | |
498 | ||
499 | if (index == 0) | |
500 | offset = sizeof(u32) * io_ctl->num_pages; | |
501 | ||
2b108268 | 502 | tmp = page_address(io_ctl->pages[0]); |
5b0e95bf JB |
503 | tmp += index; |
504 | val = *tmp; | |
5b0e95bf JB |
505 | |
506 | io_ctl_map_page(io_ctl, 0); | |
4bb3c2e2 JT |
507 | crc = btrfs_crc32c(crc, io_ctl->orig + offset, PAGE_SIZE - offset); |
508 | btrfs_crc32c_final(crc, (u8 *)&crc); | |
5b0e95bf | 509 | if (val != crc) { |
f15376df | 510 | btrfs_err_rl(io_ctl->fs_info, |
94647322 | 511 | "csum mismatch on free space cache"); |
5b0e95bf JB |
512 | io_ctl_unmap_page(io_ctl); |
513 | return -EIO; | |
514 | } | |
515 | ||
a67509c3 JB |
516 | return 0; |
517 | } | |
518 | ||
4c6d1d85 | 519 | static int io_ctl_add_entry(struct btrfs_io_ctl *io_ctl, u64 offset, u64 bytes, |
a67509c3 JB |
520 | void *bitmap) |
521 | { | |
522 | struct btrfs_free_space_entry *entry; | |
523 | ||
524 | if (!io_ctl->cur) | |
525 | return -ENOSPC; | |
526 | ||
527 | entry = io_ctl->cur; | |
528 | entry->offset = cpu_to_le64(offset); | |
529 | entry->bytes = cpu_to_le64(bytes); | |
530 | entry->type = (bitmap) ? BTRFS_FREE_SPACE_BITMAP : | |
531 | BTRFS_FREE_SPACE_EXTENT; | |
532 | io_ctl->cur += sizeof(struct btrfs_free_space_entry); | |
533 | io_ctl->size -= sizeof(struct btrfs_free_space_entry); | |
534 | ||
535 | if (io_ctl->size >= sizeof(struct btrfs_free_space_entry)) | |
536 | return 0; | |
537 | ||
5b0e95bf | 538 | io_ctl_set_crc(io_ctl, io_ctl->index - 1); |
a67509c3 JB |
539 | |
540 | /* No more pages to map */ | |
541 | if (io_ctl->index >= io_ctl->num_pages) | |
542 | return 0; | |
543 | ||
544 | /* map the next page */ | |
545 | io_ctl_map_page(io_ctl, 1); | |
546 | return 0; | |
547 | } | |
548 | ||
4c6d1d85 | 549 | static int io_ctl_add_bitmap(struct btrfs_io_ctl *io_ctl, void *bitmap) |
a67509c3 JB |
550 | { |
551 | if (!io_ctl->cur) | |
552 | return -ENOSPC; | |
553 | ||
554 | /* | |
555 | * If we aren't at the start of the current page, unmap this one and | |
556 | * map the next one if there is any left. | |
557 | */ | |
558 | if (io_ctl->cur != io_ctl->orig) { | |
5b0e95bf | 559 | io_ctl_set_crc(io_ctl, io_ctl->index - 1); |
a67509c3 JB |
560 | if (io_ctl->index >= io_ctl->num_pages) |
561 | return -ENOSPC; | |
562 | io_ctl_map_page(io_ctl, 0); | |
563 | } | |
564 | ||
69d24804 | 565 | copy_page(io_ctl->cur, bitmap); |
5b0e95bf | 566 | io_ctl_set_crc(io_ctl, io_ctl->index - 1); |
a67509c3 JB |
567 | if (io_ctl->index < io_ctl->num_pages) |
568 | io_ctl_map_page(io_ctl, 0); | |
569 | return 0; | |
570 | } | |
571 | ||
4c6d1d85 | 572 | static void io_ctl_zero_remaining_pages(struct btrfs_io_ctl *io_ctl) |
a67509c3 | 573 | { |
5b0e95bf JB |
574 | /* |
575 | * If we're not on the boundary we know we've modified the page and we | |
576 | * need to crc the page. | |
577 | */ | |
578 | if (io_ctl->cur != io_ctl->orig) | |
579 | io_ctl_set_crc(io_ctl, io_ctl->index - 1); | |
580 | else | |
581 | io_ctl_unmap_page(io_ctl); | |
a67509c3 JB |
582 | |
583 | while (io_ctl->index < io_ctl->num_pages) { | |
584 | io_ctl_map_page(io_ctl, 1); | |
5b0e95bf | 585 | io_ctl_set_crc(io_ctl, io_ctl->index - 1); |
a67509c3 JB |
586 | } |
587 | } | |
588 | ||
4c6d1d85 | 589 | static int io_ctl_read_entry(struct btrfs_io_ctl *io_ctl, |
5b0e95bf | 590 | struct btrfs_free_space *entry, u8 *type) |
a67509c3 JB |
591 | { |
592 | struct btrfs_free_space_entry *e; | |
2f120c05 JB |
593 | int ret; |
594 | ||
595 | if (!io_ctl->cur) { | |
596 | ret = io_ctl_check_crc(io_ctl, io_ctl->index); | |
597 | if (ret) | |
598 | return ret; | |
599 | } | |
a67509c3 JB |
600 | |
601 | e = io_ctl->cur; | |
602 | entry->offset = le64_to_cpu(e->offset); | |
603 | entry->bytes = le64_to_cpu(e->bytes); | |
5b0e95bf | 604 | *type = e->type; |
a67509c3 JB |
605 | io_ctl->cur += sizeof(struct btrfs_free_space_entry); |
606 | io_ctl->size -= sizeof(struct btrfs_free_space_entry); | |
607 | ||
608 | if (io_ctl->size >= sizeof(struct btrfs_free_space_entry)) | |
5b0e95bf | 609 | return 0; |
a67509c3 JB |
610 | |
611 | io_ctl_unmap_page(io_ctl); | |
612 | ||
2f120c05 | 613 | return 0; |
a67509c3 JB |
614 | } |
615 | ||
4c6d1d85 | 616 | static int io_ctl_read_bitmap(struct btrfs_io_ctl *io_ctl, |
5b0e95bf | 617 | struct btrfs_free_space *entry) |
a67509c3 | 618 | { |
5b0e95bf JB |
619 | int ret; |
620 | ||
5b0e95bf JB |
621 | ret = io_ctl_check_crc(io_ctl, io_ctl->index); |
622 | if (ret) | |
623 | return ret; | |
624 | ||
69d24804 | 625 | copy_page(entry->bitmap, io_ctl->cur); |
a67509c3 | 626 | io_ctl_unmap_page(io_ctl); |
5b0e95bf JB |
627 | |
628 | return 0; | |
a67509c3 JB |
629 | } |
630 | ||
cd023e7b JB |
631 | /* |
632 | * Since we attach pinned extents after the fact we can have contiguous sections | |
633 | * of free space that are split up in entries. This poses a problem with the | |
634 | * tree logging stuff since it could have allocated across what appears to be 2 | |
635 | * entries since we would have merged the entries when adding the pinned extents | |
636 | * back to the free space cache. So run through the space cache that we just | |
637 | * loaded and merge contiguous entries. This will make the log replay stuff not | |
638 | * blow up and it will make for nicer allocator behavior. | |
639 | */ | |
640 | static void merge_space_tree(struct btrfs_free_space_ctl *ctl) | |
641 | { | |
642 | struct btrfs_free_space *e, *prev = NULL; | |
643 | struct rb_node *n; | |
644 | ||
645 | again: | |
646 | spin_lock(&ctl->tree_lock); | |
647 | for (n = rb_first(&ctl->free_space_offset); n; n = rb_next(n)) { | |
648 | e = rb_entry(n, struct btrfs_free_space, offset_index); | |
649 | if (!prev) | |
650 | goto next; | |
651 | if (e->bitmap || prev->bitmap) | |
652 | goto next; | |
653 | if (prev->offset + prev->bytes == e->offset) { | |
654 | unlink_free_space(ctl, prev); | |
655 | unlink_free_space(ctl, e); | |
656 | prev->bytes += e->bytes; | |
657 | kmem_cache_free(btrfs_free_space_cachep, e); | |
658 | link_free_space(ctl, prev); | |
659 | prev = NULL; | |
660 | spin_unlock(&ctl->tree_lock); | |
661 | goto again; | |
662 | } | |
663 | next: | |
664 | prev = e; | |
665 | } | |
666 | spin_unlock(&ctl->tree_lock); | |
667 | } | |
668 | ||
48a3b636 ES |
669 | static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, |
670 | struct btrfs_free_space_ctl *ctl, | |
671 | struct btrfs_path *path, u64 offset) | |
9d66e233 | 672 | { |
3ffbd68c | 673 | struct btrfs_fs_info *fs_info = root->fs_info; |
9d66e233 JB |
674 | struct btrfs_free_space_header *header; |
675 | struct extent_buffer *leaf; | |
4c6d1d85 | 676 | struct btrfs_io_ctl io_ctl; |
9d66e233 | 677 | struct btrfs_key key; |
a67509c3 | 678 | struct btrfs_free_space *e, *n; |
b76808fc | 679 | LIST_HEAD(bitmaps); |
9d66e233 JB |
680 | u64 num_entries; |
681 | u64 num_bitmaps; | |
682 | u64 generation; | |
a67509c3 | 683 | u8 type; |
f6a39829 | 684 | int ret = 0; |
9d66e233 | 685 | |
9d66e233 | 686 | /* Nothing in the space cache, goodbye */ |
0414efae | 687 | if (!i_size_read(inode)) |
a67509c3 | 688 | return 0; |
9d66e233 JB |
689 | |
690 | key.objectid = BTRFS_FREE_SPACE_OBJECTID; | |
0414efae | 691 | key.offset = offset; |
9d66e233 JB |
692 | key.type = 0; |
693 | ||
694 | ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); | |
0414efae | 695 | if (ret < 0) |
a67509c3 | 696 | return 0; |
0414efae | 697 | else if (ret > 0) { |
945d8962 | 698 | btrfs_release_path(path); |
a67509c3 | 699 | return 0; |
9d66e233 JB |
700 | } |
701 | ||
0414efae LZ |
702 | ret = -1; |
703 | ||
9d66e233 JB |
704 | leaf = path->nodes[0]; |
705 | header = btrfs_item_ptr(leaf, path->slots[0], | |
706 | struct btrfs_free_space_header); | |
707 | num_entries = btrfs_free_space_entries(leaf, header); | |
708 | num_bitmaps = btrfs_free_space_bitmaps(leaf, header); | |
709 | generation = btrfs_free_space_generation(leaf, header); | |
945d8962 | 710 | btrfs_release_path(path); |
9d66e233 | 711 | |
e570fd27 | 712 | if (!BTRFS_I(inode)->generation) { |
0b246afa | 713 | btrfs_info(fs_info, |
913e1535 | 714 | "the free space cache file (%llu) is invalid, skip it", |
e570fd27 MX |
715 | offset); |
716 | return 0; | |
717 | } | |
718 | ||
9d66e233 | 719 | if (BTRFS_I(inode)->generation != generation) { |
0b246afa JM |
720 | btrfs_err(fs_info, |
721 | "free space inode generation (%llu) did not match free space cache generation (%llu)", | |
722 | BTRFS_I(inode)->generation, generation); | |
a67509c3 | 723 | return 0; |
9d66e233 JB |
724 | } |
725 | ||
726 | if (!num_entries) | |
a67509c3 | 727 | return 0; |
9d66e233 | 728 | |
f15376df | 729 | ret = io_ctl_init(&io_ctl, inode, 0); |
706efc66 LZ |
730 | if (ret) |
731 | return ret; | |
732 | ||
1d480538 | 733 | readahead_cache(inode); |
9d66e233 | 734 | |
7a195f6d | 735 | ret = io_ctl_prepare_pages(&io_ctl, true); |
a67509c3 JB |
736 | if (ret) |
737 | goto out; | |
9d66e233 | 738 | |
5b0e95bf JB |
739 | ret = io_ctl_check_crc(&io_ctl, 0); |
740 | if (ret) | |
741 | goto free_cache; | |
742 | ||
a67509c3 JB |
743 | ret = io_ctl_check_generation(&io_ctl, generation); |
744 | if (ret) | |
745 | goto free_cache; | |
9d66e233 | 746 | |
a67509c3 JB |
747 | while (num_entries) { |
748 | e = kmem_cache_zalloc(btrfs_free_space_cachep, | |
749 | GFP_NOFS); | |
750 | if (!e) | |
9d66e233 | 751 | goto free_cache; |
9d66e233 | 752 | |
5b0e95bf JB |
753 | ret = io_ctl_read_entry(&io_ctl, e, &type); |
754 | if (ret) { | |
755 | kmem_cache_free(btrfs_free_space_cachep, e); | |
756 | goto free_cache; | |
757 | } | |
758 | ||
a7ccb255 DZ |
759 | /* |
760 | * Sync discard ensures that the free space cache is always | |
761 | * trimmed. So when reading this in, the state should reflect | |
b0643e59 DZ |
762 | * that. We also do this for async as a stop gap for lack of |
763 | * persistence. | |
a7ccb255 | 764 | */ |
b0643e59 DZ |
765 | if (btrfs_test_opt(fs_info, DISCARD_SYNC) || |
766 | btrfs_test_opt(fs_info, DISCARD_ASYNC)) | |
a7ccb255 DZ |
767 | e->trim_state = BTRFS_TRIM_STATE_TRIMMED; |
768 | ||
a67509c3 JB |
769 | if (!e->bytes) { |
770 | kmem_cache_free(btrfs_free_space_cachep, e); | |
771 | goto free_cache; | |
9d66e233 | 772 | } |
a67509c3 JB |
773 | |
774 | if (type == BTRFS_FREE_SPACE_EXTENT) { | |
775 | spin_lock(&ctl->tree_lock); | |
776 | ret = link_free_space(ctl, e); | |
777 | spin_unlock(&ctl->tree_lock); | |
778 | if (ret) { | |
0b246afa | 779 | btrfs_err(fs_info, |
c2cf52eb | 780 | "Duplicate entries in free space cache, dumping"); |
a67509c3 | 781 | kmem_cache_free(btrfs_free_space_cachep, e); |
9d66e233 JB |
782 | goto free_cache; |
783 | } | |
a67509c3 | 784 | } else { |
b12d6869 | 785 | ASSERT(num_bitmaps); |
a67509c3 | 786 | num_bitmaps--; |
3acd4850 CL |
787 | e->bitmap = kmem_cache_zalloc( |
788 | btrfs_free_space_bitmap_cachep, GFP_NOFS); | |
a67509c3 JB |
789 | if (!e->bitmap) { |
790 | kmem_cache_free( | |
791 | btrfs_free_space_cachep, e); | |
9d66e233 JB |
792 | goto free_cache; |
793 | } | |
a67509c3 JB |
794 | spin_lock(&ctl->tree_lock); |
795 | ret = link_free_space(ctl, e); | |
796 | ctl->total_bitmaps++; | |
797 | ctl->op->recalc_thresholds(ctl); | |
798 | spin_unlock(&ctl->tree_lock); | |
799 | if (ret) { | |
0b246afa | 800 | btrfs_err(fs_info, |
c2cf52eb | 801 | "Duplicate entries in free space cache, dumping"); |
dc89e982 | 802 | kmem_cache_free(btrfs_free_space_cachep, e); |
9d66e233 JB |
803 | goto free_cache; |
804 | } | |
a67509c3 | 805 | list_add_tail(&e->list, &bitmaps); |
9d66e233 JB |
806 | } |
807 | ||
a67509c3 JB |
808 | num_entries--; |
809 | } | |
9d66e233 | 810 | |
2f120c05 JB |
811 | io_ctl_unmap_page(&io_ctl); |
812 | ||
a67509c3 JB |
813 | /* |
814 | * We add the bitmaps at the end of the entries in order that | |
815 | * the bitmap entries are added to the cache. | |
816 | */ | |
817 | list_for_each_entry_safe(e, n, &bitmaps, list) { | |
9d66e233 | 818 | list_del_init(&e->list); |
5b0e95bf JB |
819 | ret = io_ctl_read_bitmap(&io_ctl, e); |
820 | if (ret) | |
821 | goto free_cache; | |
dfb79ddb | 822 | e->bitmap_extents = count_bitmap_extents(ctl, e); |
5dc7c10b | 823 | if (!btrfs_free_space_trimmed(e)) { |
dfb79ddb DZ |
824 | ctl->discardable_extents[BTRFS_STAT_CURR] += |
825 | e->bitmap_extents; | |
5dc7c10b DZ |
826 | ctl->discardable_bytes[BTRFS_STAT_CURR] += e->bytes; |
827 | } | |
9d66e233 JB |
828 | } |
829 | ||
a67509c3 | 830 | io_ctl_drop_pages(&io_ctl); |
cd023e7b | 831 | merge_space_tree(ctl); |
9d66e233 JB |
832 | ret = 1; |
833 | out: | |
dfb79ddb | 834 | btrfs_discard_update_discardable(ctl->private, ctl); |
a67509c3 | 835 | io_ctl_free(&io_ctl); |
9d66e233 | 836 | return ret; |
9d66e233 | 837 | free_cache: |
a67509c3 | 838 | io_ctl_drop_pages(&io_ctl); |
0414efae | 839 | __btrfs_remove_free_space_cache(ctl); |
9d66e233 JB |
840 | goto out; |
841 | } | |
842 | ||
32da5386 | 843 | int load_free_space_cache(struct btrfs_block_group *block_group) |
0cb59c99 | 844 | { |
bb6cb1c5 | 845 | struct btrfs_fs_info *fs_info = block_group->fs_info; |
34d52cb6 | 846 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
0414efae LZ |
847 | struct inode *inode; |
848 | struct btrfs_path *path; | |
5b0e95bf | 849 | int ret = 0; |
0414efae | 850 | bool matched; |
bf38be65 | 851 | u64 used = block_group->used; |
0414efae | 852 | |
0414efae LZ |
853 | /* |
854 | * If this block group has been marked to be cleared for one reason or | |
855 | * another then we can't trust the on disk cache, so just return. | |
856 | */ | |
9d66e233 | 857 | spin_lock(&block_group->lock); |
0414efae LZ |
858 | if (block_group->disk_cache_state != BTRFS_DC_WRITTEN) { |
859 | spin_unlock(&block_group->lock); | |
860 | return 0; | |
861 | } | |
9d66e233 | 862 | spin_unlock(&block_group->lock); |
0414efae LZ |
863 | |
864 | path = btrfs_alloc_path(); | |
865 | if (!path) | |
866 | return 0; | |
d53ba474 JB |
867 | path->search_commit_root = 1; |
868 | path->skip_locking = 1; | |
0414efae | 869 | |
4222ea71 FM |
870 | /* |
871 | * We must pass a path with search_commit_root set to btrfs_iget in | |
872 | * order to avoid a deadlock when allocating extents for the tree root. | |
873 | * | |
874 | * When we are COWing an extent buffer from the tree root, when looking | |
875 | * for a free extent, at extent-tree.c:find_free_extent(), we can find | |
876 | * block group without its free space cache loaded. When we find one | |
877 | * we must load its space cache which requires reading its free space | |
878 | * cache's inode item from the root tree. If this inode item is located | |
879 | * in the same leaf that we started COWing before, then we end up in | |
880 | * deadlock on the extent buffer (trying to read lock it when we | |
881 | * previously write locked it). | |
882 | * | |
883 | * It's safe to read the inode item using the commit root because | |
884 | * block groups, once loaded, stay in memory forever (until they are | |
885 | * removed) as well as their space caches once loaded. New block groups | |
886 | * once created get their ->cached field set to BTRFS_CACHE_FINISHED so | |
887 | * we will never try to read their inode item while the fs is mounted. | |
888 | */ | |
7949f339 | 889 | inode = lookup_free_space_inode(block_group, path); |
0414efae LZ |
890 | if (IS_ERR(inode)) { |
891 | btrfs_free_path(path); | |
892 | return 0; | |
893 | } | |
894 | ||
5b0e95bf JB |
895 | /* We may have converted the inode and made the cache invalid. */ |
896 | spin_lock(&block_group->lock); | |
897 | if (block_group->disk_cache_state != BTRFS_DC_WRITTEN) { | |
898 | spin_unlock(&block_group->lock); | |
a7e221e9 | 899 | btrfs_free_path(path); |
5b0e95bf JB |
900 | goto out; |
901 | } | |
902 | spin_unlock(&block_group->lock); | |
903 | ||
0414efae | 904 | ret = __load_free_space_cache(fs_info->tree_root, inode, ctl, |
b3470b5d | 905 | path, block_group->start); |
0414efae LZ |
906 | btrfs_free_path(path); |
907 | if (ret <= 0) | |
908 | goto out; | |
909 | ||
910 | spin_lock(&ctl->tree_lock); | |
b3470b5d | 911 | matched = (ctl->free_space == (block_group->length - used - |
0414efae LZ |
912 | block_group->bytes_super)); |
913 | spin_unlock(&ctl->tree_lock); | |
914 | ||
915 | if (!matched) { | |
916 | __btrfs_remove_free_space_cache(ctl); | |
5d163e0e JM |
917 | btrfs_warn(fs_info, |
918 | "block group %llu has wrong amount of free space", | |
b3470b5d | 919 | block_group->start); |
0414efae LZ |
920 | ret = -1; |
921 | } | |
922 | out: | |
923 | if (ret < 0) { | |
924 | /* This cache is bogus, make sure it gets cleared */ | |
925 | spin_lock(&block_group->lock); | |
926 | block_group->disk_cache_state = BTRFS_DC_CLEAR; | |
927 | spin_unlock(&block_group->lock); | |
82d5902d | 928 | ret = 0; |
0414efae | 929 | |
5d163e0e JM |
930 | btrfs_warn(fs_info, |
931 | "failed to load free space cache for block group %llu, rebuilding it now", | |
b3470b5d | 932 | block_group->start); |
0414efae LZ |
933 | } |
934 | ||
935 | iput(inode); | |
936 | return ret; | |
9d66e233 JB |
937 | } |
938 | ||
d4452bc5 | 939 | static noinline_for_stack |
4c6d1d85 | 940 | int write_cache_extent_entries(struct btrfs_io_ctl *io_ctl, |
d4452bc5 | 941 | struct btrfs_free_space_ctl *ctl, |
32da5386 | 942 | struct btrfs_block_group *block_group, |
d4452bc5 CM |
943 | int *entries, int *bitmaps, |
944 | struct list_head *bitmap_list) | |
0cb59c99 | 945 | { |
c09544e0 | 946 | int ret; |
d4452bc5 | 947 | struct btrfs_free_cluster *cluster = NULL; |
1bbc621e | 948 | struct btrfs_free_cluster *cluster_locked = NULL; |
d4452bc5 | 949 | struct rb_node *node = rb_first(&ctl->free_space_offset); |
55507ce3 | 950 | struct btrfs_trim_range *trim_entry; |
be1a12a0 | 951 | |
43be2146 | 952 | /* Get the cluster for this block_group if it exists */ |
d4452bc5 | 953 | if (block_group && !list_empty(&block_group->cluster_list)) { |
43be2146 JB |
954 | cluster = list_entry(block_group->cluster_list.next, |
955 | struct btrfs_free_cluster, | |
956 | block_group_list); | |
d4452bc5 | 957 | } |
43be2146 | 958 | |
f75b130e | 959 | if (!node && cluster) { |
1bbc621e CM |
960 | cluster_locked = cluster; |
961 | spin_lock(&cluster_locked->lock); | |
f75b130e JB |
962 | node = rb_first(&cluster->root); |
963 | cluster = NULL; | |
964 | } | |
965 | ||
a67509c3 JB |
966 | /* Write out the extent entries */ |
967 | while (node) { | |
968 | struct btrfs_free_space *e; | |
0cb59c99 | 969 | |
a67509c3 | 970 | e = rb_entry(node, struct btrfs_free_space, offset_index); |
d4452bc5 | 971 | *entries += 1; |
0cb59c99 | 972 | |
d4452bc5 | 973 | ret = io_ctl_add_entry(io_ctl, e->offset, e->bytes, |
a67509c3 JB |
974 | e->bitmap); |
975 | if (ret) | |
d4452bc5 | 976 | goto fail; |
2f356126 | 977 | |
a67509c3 | 978 | if (e->bitmap) { |
d4452bc5 CM |
979 | list_add_tail(&e->list, bitmap_list); |
980 | *bitmaps += 1; | |
2f356126 | 981 | } |
a67509c3 JB |
982 | node = rb_next(node); |
983 | if (!node && cluster) { | |
984 | node = rb_first(&cluster->root); | |
1bbc621e CM |
985 | cluster_locked = cluster; |
986 | spin_lock(&cluster_locked->lock); | |
a67509c3 | 987 | cluster = NULL; |
43be2146 | 988 | } |
a67509c3 | 989 | } |
1bbc621e CM |
990 | if (cluster_locked) { |
991 | spin_unlock(&cluster_locked->lock); | |
992 | cluster_locked = NULL; | |
993 | } | |
55507ce3 FM |
994 | |
995 | /* | |
996 | * Make sure we don't miss any range that was removed from our rbtree | |
997 | * because trimming is running. Otherwise after a umount+mount (or crash | |
998 | * after committing the transaction) we would leak free space and get | |
999 | * an inconsistent free space cache report from fsck. | |
1000 | */ | |
1001 | list_for_each_entry(trim_entry, &ctl->trimming_ranges, list) { | |
1002 | ret = io_ctl_add_entry(io_ctl, trim_entry->start, | |
1003 | trim_entry->bytes, NULL); | |
1004 | if (ret) | |
1005 | goto fail; | |
1006 | *entries += 1; | |
1007 | } | |
1008 | ||
d4452bc5 CM |
1009 | return 0; |
1010 | fail: | |
1bbc621e CM |
1011 | if (cluster_locked) |
1012 | spin_unlock(&cluster_locked->lock); | |
d4452bc5 CM |
1013 | return -ENOSPC; |
1014 | } | |
1015 | ||
1016 | static noinline_for_stack int | |
1017 | update_cache_item(struct btrfs_trans_handle *trans, | |
1018 | struct btrfs_root *root, | |
1019 | struct inode *inode, | |
1020 | struct btrfs_path *path, u64 offset, | |
1021 | int entries, int bitmaps) | |
1022 | { | |
1023 | struct btrfs_key key; | |
1024 | struct btrfs_free_space_header *header; | |
1025 | struct extent_buffer *leaf; | |
1026 | int ret; | |
1027 | ||
1028 | key.objectid = BTRFS_FREE_SPACE_OBJECTID; | |
1029 | key.offset = offset; | |
1030 | key.type = 0; | |
1031 | ||
1032 | ret = btrfs_search_slot(trans, root, &key, path, 0, 1); | |
1033 | if (ret < 0) { | |
1034 | clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1, | |
e182163d | 1035 | EXTENT_DELALLOC, 0, 0, NULL); |
d4452bc5 CM |
1036 | goto fail; |
1037 | } | |
1038 | leaf = path->nodes[0]; | |
1039 | if (ret > 0) { | |
1040 | struct btrfs_key found_key; | |
1041 | ASSERT(path->slots[0]); | |
1042 | path->slots[0]--; | |
1043 | btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); | |
1044 | if (found_key.objectid != BTRFS_FREE_SPACE_OBJECTID || | |
1045 | found_key.offset != offset) { | |
1046 | clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, | |
e182163d OS |
1047 | inode->i_size - 1, EXTENT_DELALLOC, 0, |
1048 | 0, NULL); | |
d4452bc5 CM |
1049 | btrfs_release_path(path); |
1050 | goto fail; | |
1051 | } | |
1052 | } | |
1053 | ||
1054 | BTRFS_I(inode)->generation = trans->transid; | |
1055 | header = btrfs_item_ptr(leaf, path->slots[0], | |
1056 | struct btrfs_free_space_header); | |
1057 | btrfs_set_free_space_entries(leaf, header, entries); | |
1058 | btrfs_set_free_space_bitmaps(leaf, header, bitmaps); | |
1059 | btrfs_set_free_space_generation(leaf, header, trans->transid); | |
1060 | btrfs_mark_buffer_dirty(leaf); | |
1061 | btrfs_release_path(path); | |
1062 | ||
1063 | return 0; | |
1064 | ||
1065 | fail: | |
1066 | return -1; | |
1067 | } | |
1068 | ||
6701bdb3 | 1069 | static noinline_for_stack int write_pinned_extent_entries( |
6b45f641 | 1070 | struct btrfs_trans_handle *trans, |
32da5386 | 1071 | struct btrfs_block_group *block_group, |
4c6d1d85 | 1072 | struct btrfs_io_ctl *io_ctl, |
5349d6c3 | 1073 | int *entries) |
d4452bc5 CM |
1074 | { |
1075 | u64 start, extent_start, extent_end, len; | |
d4452bc5 CM |
1076 | struct extent_io_tree *unpin = NULL; |
1077 | int ret; | |
43be2146 | 1078 | |
5349d6c3 MX |
1079 | if (!block_group) |
1080 | return 0; | |
1081 | ||
a67509c3 JB |
1082 | /* |
1083 | * We want to add any pinned extents to our free space cache | |
1084 | * so we don't leak the space | |
d4452bc5 | 1085 | * |
db804f23 LZ |
1086 | * We shouldn't have switched the pinned extents yet so this is the |
1087 | * right one | |
1088 | */ | |
fe119a6e | 1089 | unpin = &trans->transaction->pinned_extents; |
db804f23 | 1090 | |
b3470b5d | 1091 | start = block_group->start; |
db804f23 | 1092 | |
b3470b5d | 1093 | while (start < block_group->start + block_group->length) { |
db804f23 LZ |
1094 | ret = find_first_extent_bit(unpin, start, |
1095 | &extent_start, &extent_end, | |
e6138876 | 1096 | EXTENT_DIRTY, NULL); |
5349d6c3 MX |
1097 | if (ret) |
1098 | return 0; | |
0cb59c99 | 1099 | |
a67509c3 | 1100 | /* This pinned extent is out of our range */ |
b3470b5d | 1101 | if (extent_start >= block_group->start + block_group->length) |
5349d6c3 | 1102 | return 0; |
2f356126 | 1103 | |
db804f23 | 1104 | extent_start = max(extent_start, start); |
b3470b5d DS |
1105 | extent_end = min(block_group->start + block_group->length, |
1106 | extent_end + 1); | |
db804f23 | 1107 | len = extent_end - extent_start; |
0cb59c99 | 1108 | |
d4452bc5 CM |
1109 | *entries += 1; |
1110 | ret = io_ctl_add_entry(io_ctl, extent_start, len, NULL); | |
a67509c3 | 1111 | if (ret) |
5349d6c3 | 1112 | return -ENOSPC; |
0cb59c99 | 1113 | |
db804f23 | 1114 | start = extent_end; |
a67509c3 | 1115 | } |
0cb59c99 | 1116 | |
5349d6c3 MX |
1117 | return 0; |
1118 | } | |
1119 | ||
1120 | static noinline_for_stack int | |
4c6d1d85 | 1121 | write_bitmap_entries(struct btrfs_io_ctl *io_ctl, struct list_head *bitmap_list) |
5349d6c3 | 1122 | { |
7ae1681e | 1123 | struct btrfs_free_space *entry, *next; |
5349d6c3 MX |
1124 | int ret; |
1125 | ||
0cb59c99 | 1126 | /* Write out the bitmaps */ |
7ae1681e | 1127 | list_for_each_entry_safe(entry, next, bitmap_list, list) { |
d4452bc5 | 1128 | ret = io_ctl_add_bitmap(io_ctl, entry->bitmap); |
a67509c3 | 1129 | if (ret) |
5349d6c3 | 1130 | return -ENOSPC; |
0cb59c99 | 1131 | list_del_init(&entry->list); |
be1a12a0 JB |
1132 | } |
1133 | ||
5349d6c3 MX |
1134 | return 0; |
1135 | } | |
0cb59c99 | 1136 | |
5349d6c3 MX |
1137 | static int flush_dirty_cache(struct inode *inode) |
1138 | { | |
1139 | int ret; | |
be1a12a0 | 1140 | |
0ef8b726 | 1141 | ret = btrfs_wait_ordered_range(inode, 0, (u64)-1); |
5349d6c3 | 1142 | if (ret) |
0ef8b726 | 1143 | clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, inode->i_size - 1, |
e182163d | 1144 | EXTENT_DELALLOC, 0, 0, NULL); |
0cb59c99 | 1145 | |
5349d6c3 | 1146 | return ret; |
d4452bc5 CM |
1147 | } |
1148 | ||
1149 | static void noinline_for_stack | |
a3bdccc4 | 1150 | cleanup_bitmap_list(struct list_head *bitmap_list) |
d4452bc5 | 1151 | { |
7ae1681e | 1152 | struct btrfs_free_space *entry, *next; |
5349d6c3 | 1153 | |
7ae1681e | 1154 | list_for_each_entry_safe(entry, next, bitmap_list, list) |
d4452bc5 | 1155 | list_del_init(&entry->list); |
a3bdccc4 CM |
1156 | } |
1157 | ||
1158 | static void noinline_for_stack | |
1159 | cleanup_write_cache_enospc(struct inode *inode, | |
1160 | struct btrfs_io_ctl *io_ctl, | |
7bf1a159 | 1161 | struct extent_state **cached_state) |
a3bdccc4 | 1162 | { |
d4452bc5 CM |
1163 | io_ctl_drop_pages(io_ctl); |
1164 | unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0, | |
e43bbe5e | 1165 | i_size_read(inode) - 1, cached_state); |
d4452bc5 | 1166 | } |
549b4fdb | 1167 | |
afdb5718 JM |
1168 | static int __btrfs_wait_cache_io(struct btrfs_root *root, |
1169 | struct btrfs_trans_handle *trans, | |
32da5386 | 1170 | struct btrfs_block_group *block_group, |
afdb5718 JM |
1171 | struct btrfs_io_ctl *io_ctl, |
1172 | struct btrfs_path *path, u64 offset) | |
c9dc4c65 CM |
1173 | { |
1174 | int ret; | |
1175 | struct inode *inode = io_ctl->inode; | |
1176 | ||
1bbc621e CM |
1177 | if (!inode) |
1178 | return 0; | |
1179 | ||
c9dc4c65 CM |
1180 | /* Flush the dirty pages in the cache file. */ |
1181 | ret = flush_dirty_cache(inode); | |
1182 | if (ret) | |
1183 | goto out; | |
1184 | ||
1185 | /* Update the cache item to tell everyone this cache file is valid. */ | |
1186 | ret = update_cache_item(trans, root, inode, path, offset, | |
1187 | io_ctl->entries, io_ctl->bitmaps); | |
1188 | out: | |
1189 | io_ctl_free(io_ctl); | |
1190 | if (ret) { | |
1191 | invalidate_inode_pages2(inode->i_mapping); | |
1192 | BTRFS_I(inode)->generation = 0; | |
1193 | if (block_group) { | |
1afb648e | 1194 | #ifdef CONFIG_BTRFS_DEBUG |
3ffbd68c | 1195 | btrfs_err(root->fs_info, |
0b246afa | 1196 | "failed to write free space cache for block group %llu", |
b3470b5d | 1197 | block_group->start); |
c9dc4c65 CM |
1198 | #endif |
1199 | } | |
1200 | } | |
1201 | btrfs_update_inode(trans, root, inode); | |
1202 | ||
1203 | if (block_group) { | |
1bbc621e CM |
1204 | /* the dirty list is protected by the dirty_bgs_lock */ |
1205 | spin_lock(&trans->transaction->dirty_bgs_lock); | |
1206 | ||
1207 | /* the disk_cache_state is protected by the block group lock */ | |
c9dc4c65 CM |
1208 | spin_lock(&block_group->lock); |
1209 | ||
1210 | /* | |
1211 | * only mark this as written if we didn't get put back on | |
1bbc621e CM |
1212 | * the dirty list while waiting for IO. Otherwise our |
1213 | * cache state won't be right, and we won't get written again | |
c9dc4c65 CM |
1214 | */ |
1215 | if (!ret && list_empty(&block_group->dirty_list)) | |
1216 | block_group->disk_cache_state = BTRFS_DC_WRITTEN; | |
1217 | else if (ret) | |
1218 | block_group->disk_cache_state = BTRFS_DC_ERROR; | |
1219 | ||
1220 | spin_unlock(&block_group->lock); | |
1bbc621e | 1221 | spin_unlock(&trans->transaction->dirty_bgs_lock); |
c9dc4c65 CM |
1222 | io_ctl->inode = NULL; |
1223 | iput(inode); | |
1224 | } | |
1225 | ||
1226 | return ret; | |
1227 | ||
1228 | } | |
1229 | ||
afdb5718 JM |
1230 | static int btrfs_wait_cache_io_root(struct btrfs_root *root, |
1231 | struct btrfs_trans_handle *trans, | |
1232 | struct btrfs_io_ctl *io_ctl, | |
1233 | struct btrfs_path *path) | |
1234 | { | |
1235 | return __btrfs_wait_cache_io(root, trans, NULL, io_ctl, path, 0); | |
1236 | } | |
1237 | ||
1238 | int btrfs_wait_cache_io(struct btrfs_trans_handle *trans, | |
32da5386 | 1239 | struct btrfs_block_group *block_group, |
afdb5718 JM |
1240 | struct btrfs_path *path) |
1241 | { | |
1242 | return __btrfs_wait_cache_io(block_group->fs_info->tree_root, trans, | |
1243 | block_group, &block_group->io_ctl, | |
b3470b5d | 1244 | path, block_group->start); |
afdb5718 JM |
1245 | } |
1246 | ||
d4452bc5 CM |
1247 | /** |
1248 | * __btrfs_write_out_cache - write out cached info to an inode | |
1249 | * @root - the root the inode belongs to | |
1250 | * @ctl - the free space cache we are going to write out | |
1251 | * @block_group - the block_group for this cache if it belongs to a block_group | |
1252 | * @trans - the trans handle | |
d4452bc5 CM |
1253 | * |
1254 | * This function writes out a free space cache struct to disk for quick recovery | |
8cd1e731 | 1255 | * on mount. This will return 0 if it was successful in writing the cache out, |
b8605454 | 1256 | * or an errno if it was not. |
d4452bc5 CM |
1257 | */ |
1258 | static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, | |
1259 | struct btrfs_free_space_ctl *ctl, | |
32da5386 | 1260 | struct btrfs_block_group *block_group, |
c9dc4c65 | 1261 | struct btrfs_io_ctl *io_ctl, |
0e8d931a | 1262 | struct btrfs_trans_handle *trans) |
d4452bc5 CM |
1263 | { |
1264 | struct extent_state *cached_state = NULL; | |
5349d6c3 | 1265 | LIST_HEAD(bitmap_list); |
d4452bc5 CM |
1266 | int entries = 0; |
1267 | int bitmaps = 0; | |
1268 | int ret; | |
c9dc4c65 | 1269 | int must_iput = 0; |
d4452bc5 CM |
1270 | |
1271 | if (!i_size_read(inode)) | |
b8605454 | 1272 | return -EIO; |
d4452bc5 | 1273 | |
c9dc4c65 | 1274 | WARN_ON(io_ctl->pages); |
f15376df | 1275 | ret = io_ctl_init(io_ctl, inode, 1); |
d4452bc5 | 1276 | if (ret) |
b8605454 | 1277 | return ret; |
d4452bc5 | 1278 | |
e570fd27 MX |
1279 | if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA)) { |
1280 | down_write(&block_group->data_rwsem); | |
1281 | spin_lock(&block_group->lock); | |
1282 | if (block_group->delalloc_bytes) { | |
1283 | block_group->disk_cache_state = BTRFS_DC_WRITTEN; | |
1284 | spin_unlock(&block_group->lock); | |
1285 | up_write(&block_group->data_rwsem); | |
1286 | BTRFS_I(inode)->generation = 0; | |
1287 | ret = 0; | |
c9dc4c65 | 1288 | must_iput = 1; |
e570fd27 MX |
1289 | goto out; |
1290 | } | |
1291 | spin_unlock(&block_group->lock); | |
1292 | } | |
1293 | ||
d4452bc5 | 1294 | /* Lock all pages first so we can lock the extent safely. */ |
7a195f6d | 1295 | ret = io_ctl_prepare_pages(io_ctl, false); |
b8605454 | 1296 | if (ret) |
b77000ed | 1297 | goto out_unlock; |
d4452bc5 CM |
1298 | |
1299 | lock_extent_bits(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1, | |
ff13db41 | 1300 | &cached_state); |
d4452bc5 | 1301 | |
c9dc4c65 | 1302 | io_ctl_set_generation(io_ctl, trans->transid); |
d4452bc5 | 1303 | |
55507ce3 | 1304 | mutex_lock(&ctl->cache_writeout_mutex); |
5349d6c3 | 1305 | /* Write out the extent entries in the free space cache */ |
1bbc621e | 1306 | spin_lock(&ctl->tree_lock); |
c9dc4c65 | 1307 | ret = write_cache_extent_entries(io_ctl, ctl, |
d4452bc5 CM |
1308 | block_group, &entries, &bitmaps, |
1309 | &bitmap_list); | |
a3bdccc4 CM |
1310 | if (ret) |
1311 | goto out_nospc_locked; | |
d4452bc5 | 1312 | |
5349d6c3 MX |
1313 | /* |
1314 | * Some spaces that are freed in the current transaction are pinned, | |
1315 | * they will be added into free space cache after the transaction is | |
1316 | * committed, we shouldn't lose them. | |
1bbc621e CM |
1317 | * |
1318 | * If this changes while we are working we'll get added back to | |
1319 | * the dirty list and redo it. No locking needed | |
5349d6c3 | 1320 | */ |
6b45f641 | 1321 | ret = write_pinned_extent_entries(trans, block_group, io_ctl, &entries); |
a3bdccc4 CM |
1322 | if (ret) |
1323 | goto out_nospc_locked; | |
5349d6c3 | 1324 | |
55507ce3 FM |
1325 | /* |
1326 | * At last, we write out all the bitmaps and keep cache_writeout_mutex | |
1327 | * locked while doing it because a concurrent trim can be manipulating | |
1328 | * or freeing the bitmap. | |
1329 | */ | |
c9dc4c65 | 1330 | ret = write_bitmap_entries(io_ctl, &bitmap_list); |
1bbc621e | 1331 | spin_unlock(&ctl->tree_lock); |
55507ce3 | 1332 | mutex_unlock(&ctl->cache_writeout_mutex); |
5349d6c3 MX |
1333 | if (ret) |
1334 | goto out_nospc; | |
1335 | ||
1336 | /* Zero out the rest of the pages just to make sure */ | |
c9dc4c65 | 1337 | io_ctl_zero_remaining_pages(io_ctl); |
d4452bc5 | 1338 | |
5349d6c3 | 1339 | /* Everything is written out, now we dirty the pages in the file. */ |
2ff7e61e JM |
1340 | ret = btrfs_dirty_pages(inode, io_ctl->pages, io_ctl->num_pages, 0, |
1341 | i_size_read(inode), &cached_state); | |
5349d6c3 | 1342 | if (ret) |
d4452bc5 | 1343 | goto out_nospc; |
5349d6c3 | 1344 | |
e570fd27 MX |
1345 | if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA)) |
1346 | up_write(&block_group->data_rwsem); | |
5349d6c3 MX |
1347 | /* |
1348 | * Release the pages and unlock the extent, we will flush | |
1349 | * them out later | |
1350 | */ | |
c9dc4c65 | 1351 | io_ctl_drop_pages(io_ctl); |
5349d6c3 MX |
1352 | |
1353 | unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0, | |
e43bbe5e | 1354 | i_size_read(inode) - 1, &cached_state); |
5349d6c3 | 1355 | |
c9dc4c65 CM |
1356 | /* |
1357 | * at this point the pages are under IO and we're happy, | |
1358 | * The caller is responsible for waiting on them and updating the | |
1359 | * the cache and the inode | |
1360 | */ | |
1361 | io_ctl->entries = entries; | |
1362 | io_ctl->bitmaps = bitmaps; | |
1363 | ||
1364 | ret = btrfs_fdatawrite_range(inode, 0, (u64)-1); | |
5349d6c3 | 1365 | if (ret) |
d4452bc5 CM |
1366 | goto out; |
1367 | ||
c9dc4c65 CM |
1368 | return 0; |
1369 | ||
a3bdccc4 CM |
1370 | out_nospc_locked: |
1371 | cleanup_bitmap_list(&bitmap_list); | |
1372 | spin_unlock(&ctl->tree_lock); | |
1373 | mutex_unlock(&ctl->cache_writeout_mutex); | |
1374 | ||
a67509c3 | 1375 | out_nospc: |
7bf1a159 | 1376 | cleanup_write_cache_enospc(inode, io_ctl, &cached_state); |
e570fd27 | 1377 | |
b77000ed | 1378 | out_unlock: |
e570fd27 MX |
1379 | if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA)) |
1380 | up_write(&block_group->data_rwsem); | |
1381 | ||
fd8efa81 JT |
1382 | out: |
1383 | io_ctl->inode = NULL; | |
1384 | io_ctl_free(io_ctl); | |
1385 | if (ret) { | |
1386 | invalidate_inode_pages2(inode->i_mapping); | |
1387 | BTRFS_I(inode)->generation = 0; | |
1388 | } | |
1389 | btrfs_update_inode(trans, root, inode); | |
1390 | if (must_iput) | |
1391 | iput(inode); | |
1392 | return ret; | |
0414efae LZ |
1393 | } |
1394 | ||
fe041534 | 1395 | int btrfs_write_out_cache(struct btrfs_trans_handle *trans, |
32da5386 | 1396 | struct btrfs_block_group *block_group, |
0414efae LZ |
1397 | struct btrfs_path *path) |
1398 | { | |
fe041534 | 1399 | struct btrfs_fs_info *fs_info = trans->fs_info; |
0414efae LZ |
1400 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
1401 | struct inode *inode; | |
1402 | int ret = 0; | |
1403 | ||
0414efae LZ |
1404 | spin_lock(&block_group->lock); |
1405 | if (block_group->disk_cache_state < BTRFS_DC_SETUP) { | |
1406 | spin_unlock(&block_group->lock); | |
e570fd27 MX |
1407 | return 0; |
1408 | } | |
0414efae LZ |
1409 | spin_unlock(&block_group->lock); |
1410 | ||
7949f339 | 1411 | inode = lookup_free_space_inode(block_group, path); |
0414efae LZ |
1412 | if (IS_ERR(inode)) |
1413 | return 0; | |
1414 | ||
77ab86bf JM |
1415 | ret = __btrfs_write_out_cache(fs_info->tree_root, inode, ctl, |
1416 | block_group, &block_group->io_ctl, trans); | |
c09544e0 | 1417 | if (ret) { |
1afb648e | 1418 | #ifdef CONFIG_BTRFS_DEBUG |
0b246afa JM |
1419 | btrfs_err(fs_info, |
1420 | "failed to write free space cache for block group %llu", | |
b3470b5d | 1421 | block_group->start); |
c09544e0 | 1422 | #endif |
c9dc4c65 CM |
1423 | spin_lock(&block_group->lock); |
1424 | block_group->disk_cache_state = BTRFS_DC_ERROR; | |
1425 | spin_unlock(&block_group->lock); | |
1426 | ||
1427 | block_group->io_ctl.inode = NULL; | |
1428 | iput(inode); | |
0414efae LZ |
1429 | } |
1430 | ||
c9dc4c65 CM |
1431 | /* |
1432 | * if ret == 0 the caller is expected to call btrfs_wait_cache_io | |
1433 | * to wait for IO and put the inode | |
1434 | */ | |
1435 | ||
0cb59c99 JB |
1436 | return ret; |
1437 | } | |
1438 | ||
34d52cb6 | 1439 | static inline unsigned long offset_to_bit(u64 bitmap_start, u32 unit, |
96303081 | 1440 | u64 offset) |
0f9dd46c | 1441 | { |
b12d6869 | 1442 | ASSERT(offset >= bitmap_start); |
96303081 | 1443 | offset -= bitmap_start; |
34d52cb6 | 1444 | return (unsigned long)(div_u64(offset, unit)); |
96303081 | 1445 | } |
0f9dd46c | 1446 | |
34d52cb6 | 1447 | static inline unsigned long bytes_to_bits(u64 bytes, u32 unit) |
96303081 | 1448 | { |
34d52cb6 | 1449 | return (unsigned long)(div_u64(bytes, unit)); |
96303081 | 1450 | } |
0f9dd46c | 1451 | |
34d52cb6 | 1452 | static inline u64 offset_to_bitmap(struct btrfs_free_space_ctl *ctl, |
96303081 JB |
1453 | u64 offset) |
1454 | { | |
1455 | u64 bitmap_start; | |
0ef6447a | 1456 | u64 bytes_per_bitmap; |
0f9dd46c | 1457 | |
34d52cb6 LZ |
1458 | bytes_per_bitmap = BITS_PER_BITMAP * ctl->unit; |
1459 | bitmap_start = offset - ctl->start; | |
0ef6447a | 1460 | bitmap_start = div64_u64(bitmap_start, bytes_per_bitmap); |
96303081 | 1461 | bitmap_start *= bytes_per_bitmap; |
34d52cb6 | 1462 | bitmap_start += ctl->start; |
0f9dd46c | 1463 | |
96303081 | 1464 | return bitmap_start; |
0f9dd46c JB |
1465 | } |
1466 | ||
96303081 JB |
1467 | static int tree_insert_offset(struct rb_root *root, u64 offset, |
1468 | struct rb_node *node, int bitmap) | |
0f9dd46c JB |
1469 | { |
1470 | struct rb_node **p = &root->rb_node; | |
1471 | struct rb_node *parent = NULL; | |
1472 | struct btrfs_free_space *info; | |
1473 | ||
1474 | while (*p) { | |
1475 | parent = *p; | |
96303081 | 1476 | info = rb_entry(parent, struct btrfs_free_space, offset_index); |
0f9dd46c | 1477 | |
96303081 | 1478 | if (offset < info->offset) { |
0f9dd46c | 1479 | p = &(*p)->rb_left; |
96303081 | 1480 | } else if (offset > info->offset) { |
0f9dd46c | 1481 | p = &(*p)->rb_right; |
96303081 JB |
1482 | } else { |
1483 | /* | |
1484 | * we could have a bitmap entry and an extent entry | |
1485 | * share the same offset. If this is the case, we want | |
1486 | * the extent entry to always be found first if we do a | |
1487 | * linear search through the tree, since we want to have | |
1488 | * the quickest allocation time, and allocating from an | |
1489 | * extent is faster than allocating from a bitmap. So | |
1490 | * if we're inserting a bitmap and we find an entry at | |
1491 | * this offset, we want to go right, or after this entry | |
1492 | * logically. If we are inserting an extent and we've | |
1493 | * found a bitmap, we want to go left, or before | |
1494 | * logically. | |
1495 | */ | |
1496 | if (bitmap) { | |
207dde82 JB |
1497 | if (info->bitmap) { |
1498 | WARN_ON_ONCE(1); | |
1499 | return -EEXIST; | |
1500 | } | |
96303081 JB |
1501 | p = &(*p)->rb_right; |
1502 | } else { | |
207dde82 JB |
1503 | if (!info->bitmap) { |
1504 | WARN_ON_ONCE(1); | |
1505 | return -EEXIST; | |
1506 | } | |
96303081 JB |
1507 | p = &(*p)->rb_left; |
1508 | } | |
1509 | } | |
0f9dd46c JB |
1510 | } |
1511 | ||
1512 | rb_link_node(node, parent, p); | |
1513 | rb_insert_color(node, root); | |
1514 | ||
1515 | return 0; | |
1516 | } | |
1517 | ||
1518 | /* | |
70cb0743 JB |
1519 | * searches the tree for the given offset. |
1520 | * | |
96303081 JB |
1521 | * fuzzy - If this is set, then we are trying to make an allocation, and we just |
1522 | * want a section that has at least bytes size and comes at or after the given | |
1523 | * offset. | |
0f9dd46c | 1524 | */ |
96303081 | 1525 | static struct btrfs_free_space * |
34d52cb6 | 1526 | tree_search_offset(struct btrfs_free_space_ctl *ctl, |
96303081 | 1527 | u64 offset, int bitmap_only, int fuzzy) |
0f9dd46c | 1528 | { |
34d52cb6 | 1529 | struct rb_node *n = ctl->free_space_offset.rb_node; |
96303081 JB |
1530 | struct btrfs_free_space *entry, *prev = NULL; |
1531 | ||
1532 | /* find entry that is closest to the 'offset' */ | |
1533 | while (1) { | |
1534 | if (!n) { | |
1535 | entry = NULL; | |
1536 | break; | |
1537 | } | |
0f9dd46c | 1538 | |
0f9dd46c | 1539 | entry = rb_entry(n, struct btrfs_free_space, offset_index); |
96303081 | 1540 | prev = entry; |
0f9dd46c | 1541 | |
96303081 | 1542 | if (offset < entry->offset) |
0f9dd46c | 1543 | n = n->rb_left; |
96303081 | 1544 | else if (offset > entry->offset) |
0f9dd46c | 1545 | n = n->rb_right; |
96303081 | 1546 | else |
0f9dd46c | 1547 | break; |
0f9dd46c JB |
1548 | } |
1549 | ||
96303081 JB |
1550 | if (bitmap_only) { |
1551 | if (!entry) | |
1552 | return NULL; | |
1553 | if (entry->bitmap) | |
1554 | return entry; | |
0f9dd46c | 1555 | |
96303081 JB |
1556 | /* |
1557 | * bitmap entry and extent entry may share same offset, | |
1558 | * in that case, bitmap entry comes after extent entry. | |
1559 | */ | |
1560 | n = rb_next(n); | |
1561 | if (!n) | |
1562 | return NULL; | |
1563 | entry = rb_entry(n, struct btrfs_free_space, offset_index); | |
1564 | if (entry->offset != offset) | |
1565 | return NULL; | |
0f9dd46c | 1566 | |
96303081 JB |
1567 | WARN_ON(!entry->bitmap); |
1568 | return entry; | |
1569 | } else if (entry) { | |
1570 | if (entry->bitmap) { | |
0f9dd46c | 1571 | /* |
96303081 JB |
1572 | * if previous extent entry covers the offset, |
1573 | * we should return it instead of the bitmap entry | |
0f9dd46c | 1574 | */ |
de6c4115 MX |
1575 | n = rb_prev(&entry->offset_index); |
1576 | if (n) { | |
96303081 JB |
1577 | prev = rb_entry(n, struct btrfs_free_space, |
1578 | offset_index); | |
de6c4115 MX |
1579 | if (!prev->bitmap && |
1580 | prev->offset + prev->bytes > offset) | |
1581 | entry = prev; | |
0f9dd46c | 1582 | } |
96303081 JB |
1583 | } |
1584 | return entry; | |
1585 | } | |
1586 | ||
1587 | if (!prev) | |
1588 | return NULL; | |
1589 | ||
1590 | /* find last entry before the 'offset' */ | |
1591 | entry = prev; | |
1592 | if (entry->offset > offset) { | |
1593 | n = rb_prev(&entry->offset_index); | |
1594 | if (n) { | |
1595 | entry = rb_entry(n, struct btrfs_free_space, | |
1596 | offset_index); | |
b12d6869 | 1597 | ASSERT(entry->offset <= offset); |
0f9dd46c | 1598 | } else { |
96303081 JB |
1599 | if (fuzzy) |
1600 | return entry; | |
1601 | else | |
1602 | return NULL; | |
0f9dd46c JB |
1603 | } |
1604 | } | |
1605 | ||
96303081 | 1606 | if (entry->bitmap) { |
de6c4115 MX |
1607 | n = rb_prev(&entry->offset_index); |
1608 | if (n) { | |
96303081 JB |
1609 | prev = rb_entry(n, struct btrfs_free_space, |
1610 | offset_index); | |
de6c4115 MX |
1611 | if (!prev->bitmap && |
1612 | prev->offset + prev->bytes > offset) | |
1613 | return prev; | |
96303081 | 1614 | } |
34d52cb6 | 1615 | if (entry->offset + BITS_PER_BITMAP * ctl->unit > offset) |
96303081 JB |
1616 | return entry; |
1617 | } else if (entry->offset + entry->bytes > offset) | |
1618 | return entry; | |
1619 | ||
1620 | if (!fuzzy) | |
1621 | return NULL; | |
1622 | ||
1623 | while (1) { | |
1624 | if (entry->bitmap) { | |
1625 | if (entry->offset + BITS_PER_BITMAP * | |
34d52cb6 | 1626 | ctl->unit > offset) |
96303081 JB |
1627 | break; |
1628 | } else { | |
1629 | if (entry->offset + entry->bytes > offset) | |
1630 | break; | |
1631 | } | |
1632 | ||
1633 | n = rb_next(&entry->offset_index); | |
1634 | if (!n) | |
1635 | return NULL; | |
1636 | entry = rb_entry(n, struct btrfs_free_space, offset_index); | |
1637 | } | |
1638 | return entry; | |
0f9dd46c JB |
1639 | } |
1640 | ||
f333adb5 | 1641 | static inline void |
34d52cb6 | 1642 | __unlink_free_space(struct btrfs_free_space_ctl *ctl, |
f333adb5 | 1643 | struct btrfs_free_space *info) |
0f9dd46c | 1644 | { |
34d52cb6 LZ |
1645 | rb_erase(&info->offset_index, &ctl->free_space_offset); |
1646 | ctl->free_extents--; | |
dfb79ddb | 1647 | |
5dc7c10b | 1648 | if (!info->bitmap && !btrfs_free_space_trimmed(info)) { |
dfb79ddb | 1649 | ctl->discardable_extents[BTRFS_STAT_CURR]--; |
5dc7c10b DZ |
1650 | ctl->discardable_bytes[BTRFS_STAT_CURR] -= info->bytes; |
1651 | } | |
f333adb5 LZ |
1652 | } |
1653 | ||
34d52cb6 | 1654 | static void unlink_free_space(struct btrfs_free_space_ctl *ctl, |
f333adb5 LZ |
1655 | struct btrfs_free_space *info) |
1656 | { | |
34d52cb6 LZ |
1657 | __unlink_free_space(ctl, info); |
1658 | ctl->free_space -= info->bytes; | |
0f9dd46c JB |
1659 | } |
1660 | ||
34d52cb6 | 1661 | static int link_free_space(struct btrfs_free_space_ctl *ctl, |
0f9dd46c JB |
1662 | struct btrfs_free_space *info) |
1663 | { | |
1664 | int ret = 0; | |
1665 | ||
b12d6869 | 1666 | ASSERT(info->bytes || info->bitmap); |
34d52cb6 | 1667 | ret = tree_insert_offset(&ctl->free_space_offset, info->offset, |
96303081 | 1668 | &info->offset_index, (info->bitmap != NULL)); |
0f9dd46c JB |
1669 | if (ret) |
1670 | return ret; | |
1671 | ||
5dc7c10b | 1672 | if (!info->bitmap && !btrfs_free_space_trimmed(info)) { |
dfb79ddb | 1673 | ctl->discardable_extents[BTRFS_STAT_CURR]++; |
5dc7c10b DZ |
1674 | ctl->discardable_bytes[BTRFS_STAT_CURR] += info->bytes; |
1675 | } | |
dfb79ddb | 1676 | |
34d52cb6 LZ |
1677 | ctl->free_space += info->bytes; |
1678 | ctl->free_extents++; | |
96303081 JB |
1679 | return ret; |
1680 | } | |
1681 | ||
34d52cb6 | 1682 | static void recalculate_thresholds(struct btrfs_free_space_ctl *ctl) |
96303081 | 1683 | { |
32da5386 | 1684 | struct btrfs_block_group *block_group = ctl->private; |
25891f79 JB |
1685 | u64 max_bytes; |
1686 | u64 bitmap_bytes; | |
1687 | u64 extent_bytes; | |
b3470b5d | 1688 | u64 size = block_group->length; |
0ef6447a FX |
1689 | u64 bytes_per_bg = BITS_PER_BITMAP * ctl->unit; |
1690 | u64 max_bitmaps = div64_u64(size + bytes_per_bg - 1, bytes_per_bg); | |
34d52cb6 | 1691 | |
0ef6447a | 1692 | max_bitmaps = max_t(u64, max_bitmaps, 1); |
dde5740f | 1693 | |
b12d6869 | 1694 | ASSERT(ctl->total_bitmaps <= max_bitmaps); |
96303081 JB |
1695 | |
1696 | /* | |
5d90c5c7 DZ |
1697 | * We are trying to keep the total amount of memory used per 1GiB of |
1698 | * space to be MAX_CACHE_BYTES_PER_GIG. However, with a reclamation | |
1699 | * mechanism of pulling extents >= FORCE_EXTENT_THRESHOLD out of | |
1700 | * bitmaps, we may end up using more memory than this. | |
96303081 | 1701 | */ |
ee22184b | 1702 | if (size < SZ_1G) |
8eb2d829 LZ |
1703 | max_bytes = MAX_CACHE_BYTES_PER_GIG; |
1704 | else | |
ee22184b | 1705 | max_bytes = MAX_CACHE_BYTES_PER_GIG * div_u64(size, SZ_1G); |
96303081 | 1706 | |
5d90c5c7 | 1707 | bitmap_bytes = ctl->total_bitmaps * ctl->unit; |
96303081 | 1708 | |
25891f79 | 1709 | /* |
f8c269d7 | 1710 | * we want the extent entry threshold to always be at most 1/2 the max |
25891f79 JB |
1711 | * bytes we can have, or whatever is less than that. |
1712 | */ | |
1713 | extent_bytes = max_bytes - bitmap_bytes; | |
f8c269d7 | 1714 | extent_bytes = min_t(u64, extent_bytes, max_bytes >> 1); |
96303081 | 1715 | |
34d52cb6 | 1716 | ctl->extents_thresh = |
f8c269d7 | 1717 | div_u64(extent_bytes, sizeof(struct btrfs_free_space)); |
96303081 JB |
1718 | } |
1719 | ||
bb3ac5a4 MX |
1720 | static inline void __bitmap_clear_bits(struct btrfs_free_space_ctl *ctl, |
1721 | struct btrfs_free_space *info, | |
1722 | u64 offset, u64 bytes) | |
96303081 | 1723 | { |
dfb79ddb DZ |
1724 | unsigned long start, count, end; |
1725 | int extent_delta = -1; | |
96303081 | 1726 | |
34d52cb6 LZ |
1727 | start = offset_to_bit(info->offset, ctl->unit, offset); |
1728 | count = bytes_to_bits(bytes, ctl->unit); | |
dfb79ddb DZ |
1729 | end = start + count; |
1730 | ASSERT(end <= BITS_PER_BITMAP); | |
96303081 | 1731 | |
f38b6e75 | 1732 | bitmap_clear(info->bitmap, start, count); |
96303081 JB |
1733 | |
1734 | info->bytes -= bytes; | |
553cceb4 JB |
1735 | if (info->max_extent_size > ctl->unit) |
1736 | info->max_extent_size = 0; | |
dfb79ddb DZ |
1737 | |
1738 | if (start && test_bit(start - 1, info->bitmap)) | |
1739 | extent_delta++; | |
1740 | ||
1741 | if (end < BITS_PER_BITMAP && test_bit(end, info->bitmap)) | |
1742 | extent_delta++; | |
1743 | ||
1744 | info->bitmap_extents += extent_delta; | |
5dc7c10b | 1745 | if (!btrfs_free_space_trimmed(info)) { |
dfb79ddb | 1746 | ctl->discardable_extents[BTRFS_STAT_CURR] += extent_delta; |
5dc7c10b DZ |
1747 | ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes; |
1748 | } | |
bb3ac5a4 MX |
1749 | } |
1750 | ||
1751 | static void bitmap_clear_bits(struct btrfs_free_space_ctl *ctl, | |
1752 | struct btrfs_free_space *info, u64 offset, | |
1753 | u64 bytes) | |
1754 | { | |
1755 | __bitmap_clear_bits(ctl, info, offset, bytes); | |
34d52cb6 | 1756 | ctl->free_space -= bytes; |
96303081 JB |
1757 | } |
1758 | ||
34d52cb6 | 1759 | static void bitmap_set_bits(struct btrfs_free_space_ctl *ctl, |
817d52f8 JB |
1760 | struct btrfs_free_space *info, u64 offset, |
1761 | u64 bytes) | |
96303081 | 1762 | { |
dfb79ddb DZ |
1763 | unsigned long start, count, end; |
1764 | int extent_delta = 1; | |
96303081 | 1765 | |
34d52cb6 LZ |
1766 | start = offset_to_bit(info->offset, ctl->unit, offset); |
1767 | count = bytes_to_bits(bytes, ctl->unit); | |
dfb79ddb DZ |
1768 | end = start + count; |
1769 | ASSERT(end <= BITS_PER_BITMAP); | |
96303081 | 1770 | |
f38b6e75 | 1771 | bitmap_set(info->bitmap, start, count); |
96303081 JB |
1772 | |
1773 | info->bytes += bytes; | |
34d52cb6 | 1774 | ctl->free_space += bytes; |
dfb79ddb DZ |
1775 | |
1776 | if (start && test_bit(start - 1, info->bitmap)) | |
1777 | extent_delta--; | |
1778 | ||
1779 | if (end < BITS_PER_BITMAP && test_bit(end, info->bitmap)) | |
1780 | extent_delta--; | |
1781 | ||
1782 | info->bitmap_extents += extent_delta; | |
5dc7c10b | 1783 | if (!btrfs_free_space_trimmed(info)) { |
dfb79ddb | 1784 | ctl->discardable_extents[BTRFS_STAT_CURR] += extent_delta; |
5dc7c10b DZ |
1785 | ctl->discardable_bytes[BTRFS_STAT_CURR] += bytes; |
1786 | } | |
96303081 JB |
1787 | } |
1788 | ||
a4820398 MX |
1789 | /* |
1790 | * If we can not find suitable extent, we will use bytes to record | |
1791 | * the size of the max extent. | |
1792 | */ | |
34d52cb6 | 1793 | static int search_bitmap(struct btrfs_free_space_ctl *ctl, |
96303081 | 1794 | struct btrfs_free_space *bitmap_info, u64 *offset, |
0584f718 | 1795 | u64 *bytes, bool for_alloc) |
96303081 JB |
1796 | { |
1797 | unsigned long found_bits = 0; | |
a4820398 | 1798 | unsigned long max_bits = 0; |
96303081 JB |
1799 | unsigned long bits, i; |
1800 | unsigned long next_zero; | |
a4820398 | 1801 | unsigned long extent_bits; |
96303081 | 1802 | |
cef40483 JB |
1803 | /* |
1804 | * Skip searching the bitmap if we don't have a contiguous section that | |
1805 | * is large enough for this allocation. | |
1806 | */ | |
0584f718 JB |
1807 | if (for_alloc && |
1808 | bitmap_info->max_extent_size && | |
cef40483 JB |
1809 | bitmap_info->max_extent_size < *bytes) { |
1810 | *bytes = bitmap_info->max_extent_size; | |
1811 | return -1; | |
1812 | } | |
1813 | ||
34d52cb6 | 1814 | i = offset_to_bit(bitmap_info->offset, ctl->unit, |
96303081 | 1815 | max_t(u64, *offset, bitmap_info->offset)); |
34d52cb6 | 1816 | bits = bytes_to_bits(*bytes, ctl->unit); |
96303081 | 1817 | |
ebb3dad4 | 1818 | for_each_set_bit_from(i, bitmap_info->bitmap, BITS_PER_BITMAP) { |
0584f718 JB |
1819 | if (for_alloc && bits == 1) { |
1820 | found_bits = 1; | |
1821 | break; | |
1822 | } | |
96303081 JB |
1823 | next_zero = find_next_zero_bit(bitmap_info->bitmap, |
1824 | BITS_PER_BITMAP, i); | |
a4820398 MX |
1825 | extent_bits = next_zero - i; |
1826 | if (extent_bits >= bits) { | |
1827 | found_bits = extent_bits; | |
96303081 | 1828 | break; |
a4820398 MX |
1829 | } else if (extent_bits > max_bits) { |
1830 | max_bits = extent_bits; | |
96303081 JB |
1831 | } |
1832 | i = next_zero; | |
1833 | } | |
1834 | ||
1835 | if (found_bits) { | |
34d52cb6 LZ |
1836 | *offset = (u64)(i * ctl->unit) + bitmap_info->offset; |
1837 | *bytes = (u64)(found_bits) * ctl->unit; | |
96303081 JB |
1838 | return 0; |
1839 | } | |
1840 | ||
a4820398 | 1841 | *bytes = (u64)(max_bits) * ctl->unit; |
cef40483 | 1842 | bitmap_info->max_extent_size = *bytes; |
96303081 JB |
1843 | return -1; |
1844 | } | |
1845 | ||
ad22cf6e JB |
1846 | static inline u64 get_max_extent_size(struct btrfs_free_space *entry) |
1847 | { | |
1848 | if (entry->bitmap) | |
1849 | return entry->max_extent_size; | |
1850 | return entry->bytes; | |
1851 | } | |
1852 | ||
a4820398 | 1853 | /* Cache the size of the max extent in bytes */ |
34d52cb6 | 1854 | static struct btrfs_free_space * |
53b381b3 | 1855 | find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes, |
a4820398 | 1856 | unsigned long align, u64 *max_extent_size) |
96303081 JB |
1857 | { |
1858 | struct btrfs_free_space *entry; | |
1859 | struct rb_node *node; | |
53b381b3 DW |
1860 | u64 tmp; |
1861 | u64 align_off; | |
96303081 JB |
1862 | int ret; |
1863 | ||
34d52cb6 | 1864 | if (!ctl->free_space_offset.rb_node) |
a4820398 | 1865 | goto out; |
96303081 | 1866 | |
34d52cb6 | 1867 | entry = tree_search_offset(ctl, offset_to_bitmap(ctl, *offset), 0, 1); |
96303081 | 1868 | if (!entry) |
a4820398 | 1869 | goto out; |
96303081 JB |
1870 | |
1871 | for (node = &entry->offset_index; node; node = rb_next(node)) { | |
1872 | entry = rb_entry(node, struct btrfs_free_space, offset_index); | |
a4820398 | 1873 | if (entry->bytes < *bytes) { |
ad22cf6e JB |
1874 | *max_extent_size = max(get_max_extent_size(entry), |
1875 | *max_extent_size); | |
96303081 | 1876 | continue; |
a4820398 | 1877 | } |
96303081 | 1878 | |
53b381b3 DW |
1879 | /* make sure the space returned is big enough |
1880 | * to match our requested alignment | |
1881 | */ | |
1882 | if (*bytes >= align) { | |
a4820398 | 1883 | tmp = entry->offset - ctl->start + align - 1; |
47c5713f | 1884 | tmp = div64_u64(tmp, align); |
53b381b3 DW |
1885 | tmp = tmp * align + ctl->start; |
1886 | align_off = tmp - entry->offset; | |
1887 | } else { | |
1888 | align_off = 0; | |
1889 | tmp = entry->offset; | |
1890 | } | |
1891 | ||
a4820398 | 1892 | if (entry->bytes < *bytes + align_off) { |
ad22cf6e JB |
1893 | *max_extent_size = max(get_max_extent_size(entry), |
1894 | *max_extent_size); | |
53b381b3 | 1895 | continue; |
a4820398 | 1896 | } |
53b381b3 | 1897 | |
96303081 | 1898 | if (entry->bitmap) { |
a4820398 MX |
1899 | u64 size = *bytes; |
1900 | ||
0584f718 | 1901 | ret = search_bitmap(ctl, entry, &tmp, &size, true); |
53b381b3 DW |
1902 | if (!ret) { |
1903 | *offset = tmp; | |
a4820398 | 1904 | *bytes = size; |
96303081 | 1905 | return entry; |
ad22cf6e JB |
1906 | } else { |
1907 | *max_extent_size = | |
1908 | max(get_max_extent_size(entry), | |
1909 | *max_extent_size); | |
53b381b3 | 1910 | } |
96303081 JB |
1911 | continue; |
1912 | } | |
1913 | ||
53b381b3 DW |
1914 | *offset = tmp; |
1915 | *bytes = entry->bytes - align_off; | |
96303081 JB |
1916 | return entry; |
1917 | } | |
a4820398 | 1918 | out: |
96303081 JB |
1919 | return NULL; |
1920 | } | |
1921 | ||
dfb79ddb DZ |
1922 | static int count_bitmap_extents(struct btrfs_free_space_ctl *ctl, |
1923 | struct btrfs_free_space *bitmap_info) | |
1924 | { | |
1925 | struct btrfs_block_group *block_group = ctl->private; | |
1926 | u64 bytes = bitmap_info->bytes; | |
1927 | unsigned int rs, re; | |
1928 | int count = 0; | |
1929 | ||
1930 | if (!block_group || !bytes) | |
1931 | return count; | |
1932 | ||
1933 | bitmap_for_each_set_region(bitmap_info->bitmap, rs, re, 0, | |
1934 | BITS_PER_BITMAP) { | |
1935 | bytes -= (rs - re) * ctl->unit; | |
1936 | count++; | |
1937 | ||
1938 | if (!bytes) | |
1939 | break; | |
1940 | } | |
1941 | ||
1942 | return count; | |
1943 | } | |
1944 | ||
34d52cb6 | 1945 | static void add_new_bitmap(struct btrfs_free_space_ctl *ctl, |
96303081 JB |
1946 | struct btrfs_free_space *info, u64 offset) |
1947 | { | |
34d52cb6 | 1948 | info->offset = offset_to_bitmap(ctl, offset); |
f019f426 | 1949 | info->bytes = 0; |
dfb79ddb | 1950 | info->bitmap_extents = 0; |
f2d0f676 | 1951 | INIT_LIST_HEAD(&info->list); |
34d52cb6 LZ |
1952 | link_free_space(ctl, info); |
1953 | ctl->total_bitmaps++; | |
96303081 | 1954 | |
34d52cb6 | 1955 | ctl->op->recalc_thresholds(ctl); |
96303081 JB |
1956 | } |
1957 | ||
34d52cb6 | 1958 | static void free_bitmap(struct btrfs_free_space_ctl *ctl, |
edf6e2d1 LZ |
1959 | struct btrfs_free_space *bitmap_info) |
1960 | { | |
27f0afc7 DZ |
1961 | /* |
1962 | * Normally when this is called, the bitmap is completely empty. However, | |
1963 | * if we are blowing up the free space cache for one reason or another | |
1964 | * via __btrfs_remove_free_space_cache(), then it may not be freed and | |
1965 | * we may leave stats on the table. | |
1966 | */ | |
1967 | if (bitmap_info->bytes && !btrfs_free_space_trimmed(bitmap_info)) { | |
1968 | ctl->discardable_extents[BTRFS_STAT_CURR] -= | |
1969 | bitmap_info->bitmap_extents; | |
1970 | ctl->discardable_bytes[BTRFS_STAT_CURR] -= bitmap_info->bytes; | |
1971 | ||
1972 | } | |
34d52cb6 | 1973 | unlink_free_space(ctl, bitmap_info); |
3acd4850 | 1974 | kmem_cache_free(btrfs_free_space_bitmap_cachep, bitmap_info->bitmap); |
dc89e982 | 1975 | kmem_cache_free(btrfs_free_space_cachep, bitmap_info); |
34d52cb6 LZ |
1976 | ctl->total_bitmaps--; |
1977 | ctl->op->recalc_thresholds(ctl); | |
edf6e2d1 LZ |
1978 | } |
1979 | ||
34d52cb6 | 1980 | static noinline int remove_from_bitmap(struct btrfs_free_space_ctl *ctl, |
96303081 JB |
1981 | struct btrfs_free_space *bitmap_info, |
1982 | u64 *offset, u64 *bytes) | |
1983 | { | |
1984 | u64 end; | |
6606bb97 JB |
1985 | u64 search_start, search_bytes; |
1986 | int ret; | |
96303081 JB |
1987 | |
1988 | again: | |
34d52cb6 | 1989 | end = bitmap_info->offset + (u64)(BITS_PER_BITMAP * ctl->unit) - 1; |
96303081 | 1990 | |
6606bb97 | 1991 | /* |
bdb7d303 JB |
1992 | * We need to search for bits in this bitmap. We could only cover some |
1993 | * of the extent in this bitmap thanks to how we add space, so we need | |
1994 | * to search for as much as it as we can and clear that amount, and then | |
1995 | * go searching for the next bit. | |
6606bb97 JB |
1996 | */ |
1997 | search_start = *offset; | |
bdb7d303 | 1998 | search_bytes = ctl->unit; |
13dbc089 | 1999 | search_bytes = min(search_bytes, end - search_start + 1); |
0584f718 JB |
2000 | ret = search_bitmap(ctl, bitmap_info, &search_start, &search_bytes, |
2001 | false); | |
b50c6e25 JB |
2002 | if (ret < 0 || search_start != *offset) |
2003 | return -EINVAL; | |
6606bb97 | 2004 | |
bdb7d303 JB |
2005 | /* We may have found more bits than what we need */ |
2006 | search_bytes = min(search_bytes, *bytes); | |
2007 | ||
2008 | /* Cannot clear past the end of the bitmap */ | |
2009 | search_bytes = min(search_bytes, end - search_start + 1); | |
2010 | ||
2011 | bitmap_clear_bits(ctl, bitmap_info, search_start, search_bytes); | |
2012 | *offset += search_bytes; | |
2013 | *bytes -= search_bytes; | |
96303081 JB |
2014 | |
2015 | if (*bytes) { | |
6606bb97 | 2016 | struct rb_node *next = rb_next(&bitmap_info->offset_index); |
edf6e2d1 | 2017 | if (!bitmap_info->bytes) |
34d52cb6 | 2018 | free_bitmap(ctl, bitmap_info); |
96303081 | 2019 | |
6606bb97 JB |
2020 | /* |
2021 | * no entry after this bitmap, but we still have bytes to | |
2022 | * remove, so something has gone wrong. | |
2023 | */ | |
2024 | if (!next) | |
96303081 JB |
2025 | return -EINVAL; |
2026 | ||
6606bb97 JB |
2027 | bitmap_info = rb_entry(next, struct btrfs_free_space, |
2028 | offset_index); | |
2029 | ||
2030 | /* | |
2031 | * if the next entry isn't a bitmap we need to return to let the | |
2032 | * extent stuff do its work. | |
2033 | */ | |
96303081 JB |
2034 | if (!bitmap_info->bitmap) |
2035 | return -EAGAIN; | |
2036 | ||
6606bb97 JB |
2037 | /* |
2038 | * Ok the next item is a bitmap, but it may not actually hold | |
2039 | * the information for the rest of this free space stuff, so | |
2040 | * look for it, and if we don't find it return so we can try | |
2041 | * everything over again. | |
2042 | */ | |
2043 | search_start = *offset; | |
bdb7d303 | 2044 | search_bytes = ctl->unit; |
34d52cb6 | 2045 | ret = search_bitmap(ctl, bitmap_info, &search_start, |
0584f718 | 2046 | &search_bytes, false); |
6606bb97 JB |
2047 | if (ret < 0 || search_start != *offset) |
2048 | return -EAGAIN; | |
2049 | ||
96303081 | 2050 | goto again; |
edf6e2d1 | 2051 | } else if (!bitmap_info->bytes) |
34d52cb6 | 2052 | free_bitmap(ctl, bitmap_info); |
96303081 JB |
2053 | |
2054 | return 0; | |
2055 | } | |
2056 | ||
2cdc342c JB |
2057 | static u64 add_bytes_to_bitmap(struct btrfs_free_space_ctl *ctl, |
2058 | struct btrfs_free_space *info, u64 offset, | |
da080fe1 | 2059 | u64 bytes, enum btrfs_trim_state trim_state) |
2cdc342c JB |
2060 | { |
2061 | u64 bytes_to_set = 0; | |
2062 | u64 end; | |
2063 | ||
da080fe1 DZ |
2064 | /* |
2065 | * This is a tradeoff to make bitmap trim state minimal. We mark the | |
2066 | * whole bitmap untrimmed if at any point we add untrimmed regions. | |
2067 | */ | |
dfb79ddb | 2068 | if (trim_state == BTRFS_TRIM_STATE_UNTRIMMED) { |
5dc7c10b | 2069 | if (btrfs_free_space_trimmed(info)) { |
dfb79ddb DZ |
2070 | ctl->discardable_extents[BTRFS_STAT_CURR] += |
2071 | info->bitmap_extents; | |
5dc7c10b DZ |
2072 | ctl->discardable_bytes[BTRFS_STAT_CURR] += info->bytes; |
2073 | } | |
da080fe1 | 2074 | info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; |
dfb79ddb | 2075 | } |
da080fe1 | 2076 | |
2cdc342c JB |
2077 | end = info->offset + (u64)(BITS_PER_BITMAP * ctl->unit); |
2078 | ||
2079 | bytes_to_set = min(end - offset, bytes); | |
2080 | ||
2081 | bitmap_set_bits(ctl, info, offset, bytes_to_set); | |
2082 | ||
cef40483 JB |
2083 | /* |
2084 | * We set some bytes, we have no idea what the max extent size is | |
2085 | * anymore. | |
2086 | */ | |
2087 | info->max_extent_size = 0; | |
2088 | ||
2cdc342c JB |
2089 | return bytes_to_set; |
2090 | ||
2091 | } | |
2092 | ||
34d52cb6 LZ |
2093 | static bool use_bitmap(struct btrfs_free_space_ctl *ctl, |
2094 | struct btrfs_free_space *info) | |
96303081 | 2095 | { |
32da5386 | 2096 | struct btrfs_block_group *block_group = ctl->private; |
0b246afa | 2097 | struct btrfs_fs_info *fs_info = block_group->fs_info; |
d0bd4560 JB |
2098 | bool forced = false; |
2099 | ||
2100 | #ifdef CONFIG_BTRFS_DEBUG | |
2ff7e61e | 2101 | if (btrfs_should_fragment_free_space(block_group)) |
d0bd4560 JB |
2102 | forced = true; |
2103 | #endif | |
96303081 | 2104 | |
5d90c5c7 DZ |
2105 | /* This is a way to reclaim large regions from the bitmaps. */ |
2106 | if (!forced && info->bytes >= FORCE_EXTENT_THRESHOLD) | |
2107 | return false; | |
2108 | ||
96303081 JB |
2109 | /* |
2110 | * If we are below the extents threshold then we can add this as an | |
2111 | * extent, and don't have to deal with the bitmap | |
2112 | */ | |
d0bd4560 | 2113 | if (!forced && ctl->free_extents < ctl->extents_thresh) { |
32cb0840 JB |
2114 | /* |
2115 | * If this block group has some small extents we don't want to | |
2116 | * use up all of our free slots in the cache with them, we want | |
01327610 | 2117 | * to reserve them to larger extents, however if we have plenty |
32cb0840 JB |
2118 | * of cache left then go ahead an dadd them, no sense in adding |
2119 | * the overhead of a bitmap if we don't have to. | |
2120 | */ | |
f9bb615a DZ |
2121 | if (info->bytes <= fs_info->sectorsize * 8) { |
2122 | if (ctl->free_extents * 3 <= ctl->extents_thresh) | |
34d52cb6 | 2123 | return false; |
32cb0840 | 2124 | } else { |
34d52cb6 | 2125 | return false; |
32cb0840 JB |
2126 | } |
2127 | } | |
96303081 JB |
2128 | |
2129 | /* | |
dde5740f JB |
2130 | * The original block groups from mkfs can be really small, like 8 |
2131 | * megabytes, so don't bother with a bitmap for those entries. However | |
2132 | * some block groups can be smaller than what a bitmap would cover but | |
2133 | * are still large enough that they could overflow the 32k memory limit, | |
2134 | * so allow those block groups to still be allowed to have a bitmap | |
2135 | * entry. | |
96303081 | 2136 | */ |
b3470b5d | 2137 | if (((BITS_PER_BITMAP * ctl->unit) >> 1) > block_group->length) |
34d52cb6 LZ |
2138 | return false; |
2139 | ||
2140 | return true; | |
2141 | } | |
2142 | ||
20e5506b | 2143 | static const struct btrfs_free_space_op free_space_op = { |
2cdc342c JB |
2144 | .recalc_thresholds = recalculate_thresholds, |
2145 | .use_bitmap = use_bitmap, | |
2146 | }; | |
2147 | ||
34d52cb6 LZ |
2148 | static int insert_into_bitmap(struct btrfs_free_space_ctl *ctl, |
2149 | struct btrfs_free_space *info) | |
2150 | { | |
2151 | struct btrfs_free_space *bitmap_info; | |
32da5386 | 2152 | struct btrfs_block_group *block_group = NULL; |
34d52cb6 | 2153 | int added = 0; |
2cdc342c | 2154 | u64 bytes, offset, bytes_added; |
da080fe1 | 2155 | enum btrfs_trim_state trim_state; |
34d52cb6 | 2156 | int ret; |
96303081 JB |
2157 | |
2158 | bytes = info->bytes; | |
2159 | offset = info->offset; | |
da080fe1 | 2160 | trim_state = info->trim_state; |
96303081 | 2161 | |
34d52cb6 LZ |
2162 | if (!ctl->op->use_bitmap(ctl, info)) |
2163 | return 0; | |
2164 | ||
2cdc342c JB |
2165 | if (ctl->op == &free_space_op) |
2166 | block_group = ctl->private; | |
38e87880 | 2167 | again: |
2cdc342c JB |
2168 | /* |
2169 | * Since we link bitmaps right into the cluster we need to see if we | |
2170 | * have a cluster here, and if so and it has our bitmap we need to add | |
2171 | * the free space to that bitmap. | |
2172 | */ | |
2173 | if (block_group && !list_empty(&block_group->cluster_list)) { | |
2174 | struct btrfs_free_cluster *cluster; | |
2175 | struct rb_node *node; | |
2176 | struct btrfs_free_space *entry; | |
2177 | ||
2178 | cluster = list_entry(block_group->cluster_list.next, | |
2179 | struct btrfs_free_cluster, | |
2180 | block_group_list); | |
2181 | spin_lock(&cluster->lock); | |
2182 | node = rb_first(&cluster->root); | |
2183 | if (!node) { | |
2184 | spin_unlock(&cluster->lock); | |
38e87880 | 2185 | goto no_cluster_bitmap; |
2cdc342c JB |
2186 | } |
2187 | ||
2188 | entry = rb_entry(node, struct btrfs_free_space, offset_index); | |
2189 | if (!entry->bitmap) { | |
2190 | spin_unlock(&cluster->lock); | |
38e87880 | 2191 | goto no_cluster_bitmap; |
2cdc342c JB |
2192 | } |
2193 | ||
2194 | if (entry->offset == offset_to_bitmap(ctl, offset)) { | |
da080fe1 DZ |
2195 | bytes_added = add_bytes_to_bitmap(ctl, entry, offset, |
2196 | bytes, trim_state); | |
2cdc342c JB |
2197 | bytes -= bytes_added; |
2198 | offset += bytes_added; | |
2199 | } | |
2200 | spin_unlock(&cluster->lock); | |
2201 | if (!bytes) { | |
2202 | ret = 1; | |
2203 | goto out; | |
2204 | } | |
2205 | } | |
38e87880 CM |
2206 | |
2207 | no_cluster_bitmap: | |
34d52cb6 | 2208 | bitmap_info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), |
96303081 JB |
2209 | 1, 0); |
2210 | if (!bitmap_info) { | |
b12d6869 | 2211 | ASSERT(added == 0); |
96303081 JB |
2212 | goto new_bitmap; |
2213 | } | |
2214 | ||
da080fe1 DZ |
2215 | bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes, |
2216 | trim_state); | |
2cdc342c JB |
2217 | bytes -= bytes_added; |
2218 | offset += bytes_added; | |
2219 | added = 0; | |
96303081 JB |
2220 | |
2221 | if (!bytes) { | |
2222 | ret = 1; | |
2223 | goto out; | |
2224 | } else | |
2225 | goto again; | |
2226 | ||
2227 | new_bitmap: | |
2228 | if (info && info->bitmap) { | |
34d52cb6 | 2229 | add_new_bitmap(ctl, info, offset); |
96303081 JB |
2230 | added = 1; |
2231 | info = NULL; | |
2232 | goto again; | |
2233 | } else { | |
34d52cb6 | 2234 | spin_unlock(&ctl->tree_lock); |
96303081 JB |
2235 | |
2236 | /* no pre-allocated info, allocate a new one */ | |
2237 | if (!info) { | |
dc89e982 JB |
2238 | info = kmem_cache_zalloc(btrfs_free_space_cachep, |
2239 | GFP_NOFS); | |
96303081 | 2240 | if (!info) { |
34d52cb6 | 2241 | spin_lock(&ctl->tree_lock); |
96303081 JB |
2242 | ret = -ENOMEM; |
2243 | goto out; | |
2244 | } | |
2245 | } | |
2246 | ||
2247 | /* allocate the bitmap */ | |
3acd4850 CL |
2248 | info->bitmap = kmem_cache_zalloc(btrfs_free_space_bitmap_cachep, |
2249 | GFP_NOFS); | |
da080fe1 | 2250 | info->trim_state = BTRFS_TRIM_STATE_TRIMMED; |
34d52cb6 | 2251 | spin_lock(&ctl->tree_lock); |
96303081 JB |
2252 | if (!info->bitmap) { |
2253 | ret = -ENOMEM; | |
2254 | goto out; | |
2255 | } | |
2256 | goto again; | |
2257 | } | |
2258 | ||
2259 | out: | |
2260 | if (info) { | |
3acd4850 CL |
2261 | if (info->bitmap) |
2262 | kmem_cache_free(btrfs_free_space_bitmap_cachep, | |
2263 | info->bitmap); | |
dc89e982 | 2264 | kmem_cache_free(btrfs_free_space_cachep, info); |
96303081 | 2265 | } |
0f9dd46c JB |
2266 | |
2267 | return ret; | |
2268 | } | |
2269 | ||
a7ccb255 DZ |
2270 | /* |
2271 | * Free space merging rules: | |
2272 | * 1) Merge trimmed areas together | |
2273 | * 2) Let untrimmed areas coalesce with trimmed areas | |
2274 | * 3) Always pull neighboring regions from bitmaps | |
2275 | * | |
2276 | * The above rules are for when we merge free space based on btrfs_trim_state. | |
2277 | * Rules 2 and 3 are subtle because they are suboptimal, but are done for the | |
2278 | * same reason: to promote larger extent regions which makes life easier for | |
2279 | * find_free_extent(). Rule 2 enables coalescing based on the common path | |
2280 | * being returning free space from btrfs_finish_extent_commit(). So when free | |
2281 | * space is trimmed, it will prevent aggregating trimmed new region and | |
2282 | * untrimmed regions in the rb_tree. Rule 3 is purely to obtain larger extents | |
2283 | * and provide find_free_extent() with the largest extents possible hoping for | |
2284 | * the reuse path. | |
2285 | */ | |
945d8962 | 2286 | static bool try_merge_free_space(struct btrfs_free_space_ctl *ctl, |
f333adb5 | 2287 | struct btrfs_free_space *info, bool update_stat) |
0f9dd46c | 2288 | { |
120d66ee LZ |
2289 | struct btrfs_free_space *left_info; |
2290 | struct btrfs_free_space *right_info; | |
2291 | bool merged = false; | |
2292 | u64 offset = info->offset; | |
2293 | u64 bytes = info->bytes; | |
a7ccb255 | 2294 | const bool is_trimmed = btrfs_free_space_trimmed(info); |
6226cb0a | 2295 | |
0f9dd46c JB |
2296 | /* |
2297 | * first we want to see if there is free space adjacent to the range we | |
2298 | * are adding, if there is remove that struct and add a new one to | |
2299 | * cover the entire range | |
2300 | */ | |
34d52cb6 | 2301 | right_info = tree_search_offset(ctl, offset + bytes, 0, 0); |
96303081 JB |
2302 | if (right_info && rb_prev(&right_info->offset_index)) |
2303 | left_info = rb_entry(rb_prev(&right_info->offset_index), | |
2304 | struct btrfs_free_space, offset_index); | |
2305 | else | |
34d52cb6 | 2306 | left_info = tree_search_offset(ctl, offset - 1, 0, 0); |
0f9dd46c | 2307 | |
a7ccb255 DZ |
2308 | /* See try_merge_free_space() comment. */ |
2309 | if (right_info && !right_info->bitmap && | |
2310 | (!is_trimmed || btrfs_free_space_trimmed(right_info))) { | |
f333adb5 | 2311 | if (update_stat) |
34d52cb6 | 2312 | unlink_free_space(ctl, right_info); |
f333adb5 | 2313 | else |
34d52cb6 | 2314 | __unlink_free_space(ctl, right_info); |
6226cb0a | 2315 | info->bytes += right_info->bytes; |
dc89e982 | 2316 | kmem_cache_free(btrfs_free_space_cachep, right_info); |
120d66ee | 2317 | merged = true; |
0f9dd46c JB |
2318 | } |
2319 | ||
a7ccb255 | 2320 | /* See try_merge_free_space() comment. */ |
96303081 | 2321 | if (left_info && !left_info->bitmap && |
a7ccb255 DZ |
2322 | left_info->offset + left_info->bytes == offset && |
2323 | (!is_trimmed || btrfs_free_space_trimmed(left_info))) { | |
f333adb5 | 2324 | if (update_stat) |
34d52cb6 | 2325 | unlink_free_space(ctl, left_info); |
f333adb5 | 2326 | else |
34d52cb6 | 2327 | __unlink_free_space(ctl, left_info); |
6226cb0a JB |
2328 | info->offset = left_info->offset; |
2329 | info->bytes += left_info->bytes; | |
dc89e982 | 2330 | kmem_cache_free(btrfs_free_space_cachep, left_info); |
120d66ee | 2331 | merged = true; |
0f9dd46c JB |
2332 | } |
2333 | ||
120d66ee LZ |
2334 | return merged; |
2335 | } | |
2336 | ||
20005523 FM |
2337 | static bool steal_from_bitmap_to_end(struct btrfs_free_space_ctl *ctl, |
2338 | struct btrfs_free_space *info, | |
2339 | bool update_stat) | |
2340 | { | |
2341 | struct btrfs_free_space *bitmap; | |
2342 | unsigned long i; | |
2343 | unsigned long j; | |
2344 | const u64 end = info->offset + info->bytes; | |
2345 | const u64 bitmap_offset = offset_to_bitmap(ctl, end); | |
2346 | u64 bytes; | |
2347 | ||
2348 | bitmap = tree_search_offset(ctl, bitmap_offset, 1, 0); | |
2349 | if (!bitmap) | |
2350 | return false; | |
2351 | ||
2352 | i = offset_to_bit(bitmap->offset, ctl->unit, end); | |
2353 | j = find_next_zero_bit(bitmap->bitmap, BITS_PER_BITMAP, i); | |
2354 | if (j == i) | |
2355 | return false; | |
2356 | bytes = (j - i) * ctl->unit; | |
2357 | info->bytes += bytes; | |
2358 | ||
a7ccb255 DZ |
2359 | /* See try_merge_free_space() comment. */ |
2360 | if (!btrfs_free_space_trimmed(bitmap)) | |
2361 | info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; | |
2362 | ||
20005523 FM |
2363 | if (update_stat) |
2364 | bitmap_clear_bits(ctl, bitmap, end, bytes); | |
2365 | else | |
2366 | __bitmap_clear_bits(ctl, bitmap, end, bytes); | |
2367 | ||
2368 | if (!bitmap->bytes) | |
2369 | free_bitmap(ctl, bitmap); | |
2370 | ||
2371 | return true; | |
2372 | } | |
2373 | ||
2374 | static bool steal_from_bitmap_to_front(struct btrfs_free_space_ctl *ctl, | |
2375 | struct btrfs_free_space *info, | |
2376 | bool update_stat) | |
2377 | { | |
2378 | struct btrfs_free_space *bitmap; | |
2379 | u64 bitmap_offset; | |
2380 | unsigned long i; | |
2381 | unsigned long j; | |
2382 | unsigned long prev_j; | |
2383 | u64 bytes; | |
2384 | ||
2385 | bitmap_offset = offset_to_bitmap(ctl, info->offset); | |
2386 | /* If we're on a boundary, try the previous logical bitmap. */ | |
2387 | if (bitmap_offset == info->offset) { | |
2388 | if (info->offset == 0) | |
2389 | return false; | |
2390 | bitmap_offset = offset_to_bitmap(ctl, info->offset - 1); | |
2391 | } | |
2392 | ||
2393 | bitmap = tree_search_offset(ctl, bitmap_offset, 1, 0); | |
2394 | if (!bitmap) | |
2395 | return false; | |
2396 | ||
2397 | i = offset_to_bit(bitmap->offset, ctl->unit, info->offset) - 1; | |
2398 | j = 0; | |
2399 | prev_j = (unsigned long)-1; | |
2400 | for_each_clear_bit_from(j, bitmap->bitmap, BITS_PER_BITMAP) { | |
2401 | if (j > i) | |
2402 | break; | |
2403 | prev_j = j; | |
2404 | } | |
2405 | if (prev_j == i) | |
2406 | return false; | |
2407 | ||
2408 | if (prev_j == (unsigned long)-1) | |
2409 | bytes = (i + 1) * ctl->unit; | |
2410 | else | |
2411 | bytes = (i - prev_j) * ctl->unit; | |
2412 | ||
2413 | info->offset -= bytes; | |
2414 | info->bytes += bytes; | |
2415 | ||
a7ccb255 DZ |
2416 | /* See try_merge_free_space() comment. */ |
2417 | if (!btrfs_free_space_trimmed(bitmap)) | |
2418 | info->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; | |
2419 | ||
20005523 FM |
2420 | if (update_stat) |
2421 | bitmap_clear_bits(ctl, bitmap, info->offset, bytes); | |
2422 | else | |
2423 | __bitmap_clear_bits(ctl, bitmap, info->offset, bytes); | |
2424 | ||
2425 | if (!bitmap->bytes) | |
2426 | free_bitmap(ctl, bitmap); | |
2427 | ||
2428 | return true; | |
2429 | } | |
2430 | ||
2431 | /* | |
2432 | * We prefer always to allocate from extent entries, both for clustered and | |
2433 | * non-clustered allocation requests. So when attempting to add a new extent | |
2434 | * entry, try to see if there's adjacent free space in bitmap entries, and if | |
2435 | * there is, migrate that space from the bitmaps to the extent. | |
2436 | * Like this we get better chances of satisfying space allocation requests | |
2437 | * because we attempt to satisfy them based on a single cache entry, and never | |
2438 | * on 2 or more entries - even if the entries represent a contiguous free space | |
2439 | * region (e.g. 1 extent entry + 1 bitmap entry starting where the extent entry | |
2440 | * ends). | |
2441 | */ | |
2442 | static void steal_from_bitmap(struct btrfs_free_space_ctl *ctl, | |
2443 | struct btrfs_free_space *info, | |
2444 | bool update_stat) | |
2445 | { | |
2446 | /* | |
2447 | * Only work with disconnected entries, as we can change their offset, | |
2448 | * and must be extent entries. | |
2449 | */ | |
2450 | ASSERT(!info->bitmap); | |
2451 | ASSERT(RB_EMPTY_NODE(&info->offset_index)); | |
2452 | ||
2453 | if (ctl->total_bitmaps > 0) { | |
2454 | bool stole_end; | |
2455 | bool stole_front = false; | |
2456 | ||
2457 | stole_end = steal_from_bitmap_to_end(ctl, info, update_stat); | |
2458 | if (ctl->total_bitmaps > 0) | |
2459 | stole_front = steal_from_bitmap_to_front(ctl, info, | |
2460 | update_stat); | |
2461 | ||
2462 | if (stole_end || stole_front) | |
2463 | try_merge_free_space(ctl, info, update_stat); | |
2464 | } | |
2465 | } | |
2466 | ||
ab8d0fc4 JM |
2467 | int __btrfs_add_free_space(struct btrfs_fs_info *fs_info, |
2468 | struct btrfs_free_space_ctl *ctl, | |
a7ccb255 DZ |
2469 | u64 offset, u64 bytes, |
2470 | enum btrfs_trim_state trim_state) | |
120d66ee | 2471 | { |
b0643e59 | 2472 | struct btrfs_block_group *block_group = ctl->private; |
120d66ee LZ |
2473 | struct btrfs_free_space *info; |
2474 | int ret = 0; | |
7fe6d45e | 2475 | u64 filter_bytes = bytes; |
120d66ee | 2476 | |
dc89e982 | 2477 | info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); |
120d66ee LZ |
2478 | if (!info) |
2479 | return -ENOMEM; | |
2480 | ||
2481 | info->offset = offset; | |
2482 | info->bytes = bytes; | |
a7ccb255 | 2483 | info->trim_state = trim_state; |
20005523 | 2484 | RB_CLEAR_NODE(&info->offset_index); |
120d66ee | 2485 | |
34d52cb6 | 2486 | spin_lock(&ctl->tree_lock); |
120d66ee | 2487 | |
34d52cb6 | 2488 | if (try_merge_free_space(ctl, info, true)) |
120d66ee LZ |
2489 | goto link; |
2490 | ||
2491 | /* | |
2492 | * There was no extent directly to the left or right of this new | |
2493 | * extent then we know we're going to have to allocate a new extent, so | |
2494 | * before we do that see if we need to drop this into a bitmap | |
2495 | */ | |
34d52cb6 | 2496 | ret = insert_into_bitmap(ctl, info); |
120d66ee LZ |
2497 | if (ret < 0) { |
2498 | goto out; | |
2499 | } else if (ret) { | |
2500 | ret = 0; | |
2501 | goto out; | |
2502 | } | |
2503 | link: | |
20005523 FM |
2504 | /* |
2505 | * Only steal free space from adjacent bitmaps if we're sure we're not | |
2506 | * going to add the new free space to existing bitmap entries - because | |
2507 | * that would mean unnecessary work that would be reverted. Therefore | |
2508 | * attempt to steal space from bitmaps if we're adding an extent entry. | |
2509 | */ | |
2510 | steal_from_bitmap(ctl, info, true); | |
2511 | ||
7fe6d45e DZ |
2512 | filter_bytes = max(filter_bytes, info->bytes); |
2513 | ||
34d52cb6 | 2514 | ret = link_free_space(ctl, info); |
0f9dd46c | 2515 | if (ret) |
dc89e982 | 2516 | kmem_cache_free(btrfs_free_space_cachep, info); |
96303081 | 2517 | out: |
dfb79ddb | 2518 | btrfs_discard_update_discardable(block_group, ctl); |
34d52cb6 | 2519 | spin_unlock(&ctl->tree_lock); |
6226cb0a | 2520 | |
0f9dd46c | 2521 | if (ret) { |
ab8d0fc4 | 2522 | btrfs_crit(fs_info, "unable to add free space :%d", ret); |
b12d6869 | 2523 | ASSERT(ret != -EEXIST); |
0f9dd46c JB |
2524 | } |
2525 | ||
7fe6d45e DZ |
2526 | if (trim_state != BTRFS_TRIM_STATE_TRIMMED) { |
2527 | btrfs_discard_check_filter(block_group, filter_bytes); | |
b0643e59 | 2528 | btrfs_discard_queue_work(&fs_info->discard_ctl, block_group); |
7fe6d45e | 2529 | } |
b0643e59 | 2530 | |
0f9dd46c JB |
2531 | return ret; |
2532 | } | |
2533 | ||
32da5386 | 2534 | int btrfs_add_free_space(struct btrfs_block_group *block_group, |
478b4d9f JB |
2535 | u64 bytenr, u64 size) |
2536 | { | |
a7ccb255 DZ |
2537 | enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; |
2538 | ||
2539 | if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC)) | |
2540 | trim_state = BTRFS_TRIM_STATE_TRIMMED; | |
2541 | ||
478b4d9f JB |
2542 | return __btrfs_add_free_space(block_group->fs_info, |
2543 | block_group->free_space_ctl, | |
a7ccb255 | 2544 | bytenr, size, trim_state); |
478b4d9f JB |
2545 | } |
2546 | ||
b0643e59 DZ |
2547 | /* |
2548 | * This is a subtle distinction because when adding free space back in general, | |
2549 | * we want it to be added as untrimmed for async. But in the case where we add | |
2550 | * it on loading of a block group, we want to consider it trimmed. | |
2551 | */ | |
2552 | int btrfs_add_free_space_async_trimmed(struct btrfs_block_group *block_group, | |
2553 | u64 bytenr, u64 size) | |
2554 | { | |
2555 | enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; | |
2556 | ||
2557 | if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC) || | |
2558 | btrfs_test_opt(block_group->fs_info, DISCARD_ASYNC)) | |
2559 | trim_state = BTRFS_TRIM_STATE_TRIMMED; | |
2560 | ||
2561 | return __btrfs_add_free_space(block_group->fs_info, | |
2562 | block_group->free_space_ctl, | |
2563 | bytenr, size, trim_state); | |
2564 | } | |
2565 | ||
32da5386 | 2566 | int btrfs_remove_free_space(struct btrfs_block_group *block_group, |
6226cb0a | 2567 | u64 offset, u64 bytes) |
0f9dd46c | 2568 | { |
34d52cb6 | 2569 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
0f9dd46c | 2570 | struct btrfs_free_space *info; |
b0175117 JB |
2571 | int ret; |
2572 | bool re_search = false; | |
0f9dd46c | 2573 | |
34d52cb6 | 2574 | spin_lock(&ctl->tree_lock); |
6226cb0a | 2575 | |
96303081 | 2576 | again: |
b0175117 | 2577 | ret = 0; |
bdb7d303 JB |
2578 | if (!bytes) |
2579 | goto out_lock; | |
2580 | ||
34d52cb6 | 2581 | info = tree_search_offset(ctl, offset, 0, 0); |
96303081 | 2582 | if (!info) { |
6606bb97 JB |
2583 | /* |
2584 | * oops didn't find an extent that matched the space we wanted | |
2585 | * to remove, look for a bitmap instead | |
2586 | */ | |
34d52cb6 | 2587 | info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), |
6606bb97 JB |
2588 | 1, 0); |
2589 | if (!info) { | |
b0175117 JB |
2590 | /* |
2591 | * If we found a partial bit of our free space in a | |
2592 | * bitmap but then couldn't find the other part this may | |
2593 | * be a problem, so WARN about it. | |
24a70313 | 2594 | */ |
b0175117 | 2595 | WARN_ON(re_search); |
6606bb97 JB |
2596 | goto out_lock; |
2597 | } | |
96303081 JB |
2598 | } |
2599 | ||
b0175117 | 2600 | re_search = false; |
bdb7d303 | 2601 | if (!info->bitmap) { |
34d52cb6 | 2602 | unlink_free_space(ctl, info); |
bdb7d303 JB |
2603 | if (offset == info->offset) { |
2604 | u64 to_free = min(bytes, info->bytes); | |
2605 | ||
2606 | info->bytes -= to_free; | |
2607 | info->offset += to_free; | |
2608 | if (info->bytes) { | |
2609 | ret = link_free_space(ctl, info); | |
2610 | WARN_ON(ret); | |
2611 | } else { | |
2612 | kmem_cache_free(btrfs_free_space_cachep, info); | |
2613 | } | |
0f9dd46c | 2614 | |
bdb7d303 JB |
2615 | offset += to_free; |
2616 | bytes -= to_free; | |
2617 | goto again; | |
2618 | } else { | |
2619 | u64 old_end = info->bytes + info->offset; | |
9b49c9b9 | 2620 | |
bdb7d303 | 2621 | info->bytes = offset - info->offset; |
34d52cb6 | 2622 | ret = link_free_space(ctl, info); |
96303081 JB |
2623 | WARN_ON(ret); |
2624 | if (ret) | |
2625 | goto out_lock; | |
96303081 | 2626 | |
bdb7d303 JB |
2627 | /* Not enough bytes in this entry to satisfy us */ |
2628 | if (old_end < offset + bytes) { | |
2629 | bytes -= old_end - offset; | |
2630 | offset = old_end; | |
2631 | goto again; | |
2632 | } else if (old_end == offset + bytes) { | |
2633 | /* all done */ | |
2634 | goto out_lock; | |
2635 | } | |
2636 | spin_unlock(&ctl->tree_lock); | |
2637 | ||
a7ccb255 DZ |
2638 | ret = __btrfs_add_free_space(block_group->fs_info, ctl, |
2639 | offset + bytes, | |
2640 | old_end - (offset + bytes), | |
2641 | info->trim_state); | |
bdb7d303 JB |
2642 | WARN_ON(ret); |
2643 | goto out; | |
2644 | } | |
0f9dd46c | 2645 | } |
96303081 | 2646 | |
34d52cb6 | 2647 | ret = remove_from_bitmap(ctl, info, &offset, &bytes); |
b0175117 JB |
2648 | if (ret == -EAGAIN) { |
2649 | re_search = true; | |
96303081 | 2650 | goto again; |
b0175117 | 2651 | } |
96303081 | 2652 | out_lock: |
dfb79ddb | 2653 | btrfs_discard_update_discardable(block_group, ctl); |
34d52cb6 | 2654 | spin_unlock(&ctl->tree_lock); |
0f9dd46c | 2655 | out: |
25179201 JB |
2656 | return ret; |
2657 | } | |
2658 | ||
32da5386 | 2659 | void btrfs_dump_free_space(struct btrfs_block_group *block_group, |
0f9dd46c JB |
2660 | u64 bytes) |
2661 | { | |
0b246afa | 2662 | struct btrfs_fs_info *fs_info = block_group->fs_info; |
34d52cb6 | 2663 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
0f9dd46c JB |
2664 | struct btrfs_free_space *info; |
2665 | struct rb_node *n; | |
2666 | int count = 0; | |
2667 | ||
9084cb6a | 2668 | spin_lock(&ctl->tree_lock); |
34d52cb6 | 2669 | for (n = rb_first(&ctl->free_space_offset); n; n = rb_next(n)) { |
0f9dd46c | 2670 | info = rb_entry(n, struct btrfs_free_space, offset_index); |
f6175efa | 2671 | if (info->bytes >= bytes && !block_group->ro) |
0f9dd46c | 2672 | count++; |
0b246afa | 2673 | btrfs_crit(fs_info, "entry offset %llu, bytes %llu, bitmap %s", |
efe120a0 | 2674 | info->offset, info->bytes, |
96303081 | 2675 | (info->bitmap) ? "yes" : "no"); |
0f9dd46c | 2676 | } |
9084cb6a | 2677 | spin_unlock(&ctl->tree_lock); |
0b246afa | 2678 | btrfs_info(fs_info, "block group has cluster?: %s", |
96303081 | 2679 | list_empty(&block_group->cluster_list) ? "no" : "yes"); |
0b246afa | 2680 | btrfs_info(fs_info, |
efe120a0 | 2681 | "%d blocks of free space at or bigger than bytes is", count); |
0f9dd46c JB |
2682 | } |
2683 | ||
32da5386 | 2684 | void btrfs_init_free_space_ctl(struct btrfs_block_group *block_group) |
0f9dd46c | 2685 | { |
0b246afa | 2686 | struct btrfs_fs_info *fs_info = block_group->fs_info; |
34d52cb6 | 2687 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
0f9dd46c | 2688 | |
34d52cb6 | 2689 | spin_lock_init(&ctl->tree_lock); |
0b246afa | 2690 | ctl->unit = fs_info->sectorsize; |
b3470b5d | 2691 | ctl->start = block_group->start; |
34d52cb6 LZ |
2692 | ctl->private = block_group; |
2693 | ctl->op = &free_space_op; | |
55507ce3 FM |
2694 | INIT_LIST_HEAD(&ctl->trimming_ranges); |
2695 | mutex_init(&ctl->cache_writeout_mutex); | |
0f9dd46c | 2696 | |
34d52cb6 LZ |
2697 | /* |
2698 | * we only want to have 32k of ram per block group for keeping | |
2699 | * track of free space, and if we pass 1/2 of that we want to | |
2700 | * start converting things over to using bitmaps | |
2701 | */ | |
ee22184b | 2702 | ctl->extents_thresh = (SZ_32K / 2) / sizeof(struct btrfs_free_space); |
0f9dd46c JB |
2703 | } |
2704 | ||
fa9c0d79 CM |
2705 | /* |
2706 | * for a given cluster, put all of its extents back into the free | |
2707 | * space cache. If the block group passed doesn't match the block group | |
2708 | * pointed to by the cluster, someone else raced in and freed the | |
2709 | * cluster already. In that case, we just return without changing anything | |
2710 | */ | |
2711 | static int | |
2712 | __btrfs_return_cluster_to_free_space( | |
32da5386 | 2713 | struct btrfs_block_group *block_group, |
fa9c0d79 CM |
2714 | struct btrfs_free_cluster *cluster) |
2715 | { | |
34d52cb6 | 2716 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
fa9c0d79 CM |
2717 | struct btrfs_free_space *entry; |
2718 | struct rb_node *node; | |
2719 | ||
2720 | spin_lock(&cluster->lock); | |
2721 | if (cluster->block_group != block_group) | |
2722 | goto out; | |
2723 | ||
96303081 | 2724 | cluster->block_group = NULL; |
fa9c0d79 | 2725 | cluster->window_start = 0; |
96303081 | 2726 | list_del_init(&cluster->block_group_list); |
96303081 | 2727 | |
fa9c0d79 | 2728 | node = rb_first(&cluster->root); |
96303081 | 2729 | while (node) { |
4e69b598 JB |
2730 | bool bitmap; |
2731 | ||
fa9c0d79 CM |
2732 | entry = rb_entry(node, struct btrfs_free_space, offset_index); |
2733 | node = rb_next(&entry->offset_index); | |
2734 | rb_erase(&entry->offset_index, &cluster->root); | |
20005523 | 2735 | RB_CLEAR_NODE(&entry->offset_index); |
4e69b598 JB |
2736 | |
2737 | bitmap = (entry->bitmap != NULL); | |
20005523 | 2738 | if (!bitmap) { |
dfb79ddb | 2739 | /* Merging treats extents as if they were new */ |
5dc7c10b | 2740 | if (!btrfs_free_space_trimmed(entry)) { |
dfb79ddb | 2741 | ctl->discardable_extents[BTRFS_STAT_CURR]--; |
5dc7c10b DZ |
2742 | ctl->discardable_bytes[BTRFS_STAT_CURR] -= |
2743 | entry->bytes; | |
2744 | } | |
dfb79ddb | 2745 | |
34d52cb6 | 2746 | try_merge_free_space(ctl, entry, false); |
20005523 | 2747 | steal_from_bitmap(ctl, entry, false); |
dfb79ddb DZ |
2748 | |
2749 | /* As we insert directly, update these statistics */ | |
5dc7c10b | 2750 | if (!btrfs_free_space_trimmed(entry)) { |
dfb79ddb | 2751 | ctl->discardable_extents[BTRFS_STAT_CURR]++; |
5dc7c10b DZ |
2752 | ctl->discardable_bytes[BTRFS_STAT_CURR] += |
2753 | entry->bytes; | |
2754 | } | |
20005523 | 2755 | } |
34d52cb6 | 2756 | tree_insert_offset(&ctl->free_space_offset, |
4e69b598 | 2757 | entry->offset, &entry->offset_index, bitmap); |
fa9c0d79 | 2758 | } |
6bef4d31 | 2759 | cluster->root = RB_ROOT; |
96303081 | 2760 | |
fa9c0d79 CM |
2761 | out: |
2762 | spin_unlock(&cluster->lock); | |
96303081 | 2763 | btrfs_put_block_group(block_group); |
fa9c0d79 CM |
2764 | return 0; |
2765 | } | |
2766 | ||
48a3b636 ES |
2767 | static void __btrfs_remove_free_space_cache_locked( |
2768 | struct btrfs_free_space_ctl *ctl) | |
0f9dd46c JB |
2769 | { |
2770 | struct btrfs_free_space *info; | |
2771 | struct rb_node *node; | |
581bb050 | 2772 | |
581bb050 LZ |
2773 | while ((node = rb_last(&ctl->free_space_offset)) != NULL) { |
2774 | info = rb_entry(node, struct btrfs_free_space, offset_index); | |
9b90f513 JB |
2775 | if (!info->bitmap) { |
2776 | unlink_free_space(ctl, info); | |
2777 | kmem_cache_free(btrfs_free_space_cachep, info); | |
2778 | } else { | |
2779 | free_bitmap(ctl, info); | |
2780 | } | |
351810c1 DS |
2781 | |
2782 | cond_resched_lock(&ctl->tree_lock); | |
581bb050 | 2783 | } |
09655373 CM |
2784 | } |
2785 | ||
2786 | void __btrfs_remove_free_space_cache(struct btrfs_free_space_ctl *ctl) | |
2787 | { | |
2788 | spin_lock(&ctl->tree_lock); | |
2789 | __btrfs_remove_free_space_cache_locked(ctl); | |
27f0afc7 DZ |
2790 | if (ctl->private) |
2791 | btrfs_discard_update_discardable(ctl->private, ctl); | |
581bb050 LZ |
2792 | spin_unlock(&ctl->tree_lock); |
2793 | } | |
2794 | ||
32da5386 | 2795 | void btrfs_remove_free_space_cache(struct btrfs_block_group *block_group) |
581bb050 LZ |
2796 | { |
2797 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; | |
fa9c0d79 | 2798 | struct btrfs_free_cluster *cluster; |
96303081 | 2799 | struct list_head *head; |
0f9dd46c | 2800 | |
34d52cb6 | 2801 | spin_lock(&ctl->tree_lock); |
96303081 JB |
2802 | while ((head = block_group->cluster_list.next) != |
2803 | &block_group->cluster_list) { | |
2804 | cluster = list_entry(head, struct btrfs_free_cluster, | |
2805 | block_group_list); | |
fa9c0d79 CM |
2806 | |
2807 | WARN_ON(cluster->block_group != block_group); | |
2808 | __btrfs_return_cluster_to_free_space(block_group, cluster); | |
351810c1 DS |
2809 | |
2810 | cond_resched_lock(&ctl->tree_lock); | |
fa9c0d79 | 2811 | } |
09655373 | 2812 | __btrfs_remove_free_space_cache_locked(ctl); |
dfb79ddb | 2813 | btrfs_discard_update_discardable(block_group, ctl); |
34d52cb6 | 2814 | spin_unlock(&ctl->tree_lock); |
fa9c0d79 | 2815 | |
0f9dd46c JB |
2816 | } |
2817 | ||
6e80d4f8 DZ |
2818 | /** |
2819 | * btrfs_is_free_space_trimmed - see if everything is trimmed | |
2820 | * @block_group: block_group of interest | |
2821 | * | |
2822 | * Walk @block_group's free space rb_tree to determine if everything is trimmed. | |
2823 | */ | |
2824 | bool btrfs_is_free_space_trimmed(struct btrfs_block_group *block_group) | |
2825 | { | |
2826 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; | |
2827 | struct btrfs_free_space *info; | |
2828 | struct rb_node *node; | |
2829 | bool ret = true; | |
2830 | ||
2831 | spin_lock(&ctl->tree_lock); | |
2832 | node = rb_first(&ctl->free_space_offset); | |
2833 | ||
2834 | while (node) { | |
2835 | info = rb_entry(node, struct btrfs_free_space, offset_index); | |
2836 | ||
2837 | if (!btrfs_free_space_trimmed(info)) { | |
2838 | ret = false; | |
2839 | break; | |
2840 | } | |
2841 | ||
2842 | node = rb_next(node); | |
2843 | } | |
2844 | ||
2845 | spin_unlock(&ctl->tree_lock); | |
2846 | return ret; | |
2847 | } | |
2848 | ||
32da5386 | 2849 | u64 btrfs_find_space_for_alloc(struct btrfs_block_group *block_group, |
a4820398 MX |
2850 | u64 offset, u64 bytes, u64 empty_size, |
2851 | u64 *max_extent_size) | |
0f9dd46c | 2852 | { |
34d52cb6 | 2853 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
9ddf648f DZ |
2854 | struct btrfs_discard_ctl *discard_ctl = |
2855 | &block_group->fs_info->discard_ctl; | |
6226cb0a | 2856 | struct btrfs_free_space *entry = NULL; |
96303081 | 2857 | u64 bytes_search = bytes + empty_size; |
6226cb0a | 2858 | u64 ret = 0; |
53b381b3 DW |
2859 | u64 align_gap = 0; |
2860 | u64 align_gap_len = 0; | |
a7ccb255 | 2861 | enum btrfs_trim_state align_gap_trim_state = BTRFS_TRIM_STATE_UNTRIMMED; |
0f9dd46c | 2862 | |
34d52cb6 | 2863 | spin_lock(&ctl->tree_lock); |
53b381b3 | 2864 | entry = find_free_space(ctl, &offset, &bytes_search, |
a4820398 | 2865 | block_group->full_stripe_len, max_extent_size); |
6226cb0a | 2866 | if (!entry) |
96303081 JB |
2867 | goto out; |
2868 | ||
2869 | ret = offset; | |
2870 | if (entry->bitmap) { | |
34d52cb6 | 2871 | bitmap_clear_bits(ctl, entry, offset, bytes); |
9ddf648f DZ |
2872 | |
2873 | if (!btrfs_free_space_trimmed(entry)) | |
2874 | atomic64_add(bytes, &discard_ctl->discard_bytes_saved); | |
2875 | ||
edf6e2d1 | 2876 | if (!entry->bytes) |
34d52cb6 | 2877 | free_bitmap(ctl, entry); |
96303081 | 2878 | } else { |
34d52cb6 | 2879 | unlink_free_space(ctl, entry); |
53b381b3 DW |
2880 | align_gap_len = offset - entry->offset; |
2881 | align_gap = entry->offset; | |
a7ccb255 | 2882 | align_gap_trim_state = entry->trim_state; |
53b381b3 | 2883 | |
9ddf648f DZ |
2884 | if (!btrfs_free_space_trimmed(entry)) |
2885 | atomic64_add(bytes, &discard_ctl->discard_bytes_saved); | |
2886 | ||
53b381b3 DW |
2887 | entry->offset = offset + bytes; |
2888 | WARN_ON(entry->bytes < bytes + align_gap_len); | |
2889 | ||
2890 | entry->bytes -= bytes + align_gap_len; | |
6226cb0a | 2891 | if (!entry->bytes) |
dc89e982 | 2892 | kmem_cache_free(btrfs_free_space_cachep, entry); |
6226cb0a | 2893 | else |
34d52cb6 | 2894 | link_free_space(ctl, entry); |
6226cb0a | 2895 | } |
96303081 | 2896 | out: |
dfb79ddb | 2897 | btrfs_discard_update_discardable(block_group, ctl); |
34d52cb6 | 2898 | spin_unlock(&ctl->tree_lock); |
817d52f8 | 2899 | |
53b381b3 | 2900 | if (align_gap_len) |
ab8d0fc4 | 2901 | __btrfs_add_free_space(block_group->fs_info, ctl, |
a7ccb255 DZ |
2902 | align_gap, align_gap_len, |
2903 | align_gap_trim_state); | |
0f9dd46c JB |
2904 | return ret; |
2905 | } | |
fa9c0d79 CM |
2906 | |
2907 | /* | |
2908 | * given a cluster, put all of its extents back into the free space | |
2909 | * cache. If a block group is passed, this function will only free | |
2910 | * a cluster that belongs to the passed block group. | |
2911 | * | |
2912 | * Otherwise, it'll get a reference on the block group pointed to by the | |
2913 | * cluster and remove the cluster from it. | |
2914 | */ | |
2915 | int btrfs_return_cluster_to_free_space( | |
32da5386 | 2916 | struct btrfs_block_group *block_group, |
fa9c0d79 CM |
2917 | struct btrfs_free_cluster *cluster) |
2918 | { | |
34d52cb6 | 2919 | struct btrfs_free_space_ctl *ctl; |
fa9c0d79 CM |
2920 | int ret; |
2921 | ||
2922 | /* first, get a safe pointer to the block group */ | |
2923 | spin_lock(&cluster->lock); | |
2924 | if (!block_group) { | |
2925 | block_group = cluster->block_group; | |
2926 | if (!block_group) { | |
2927 | spin_unlock(&cluster->lock); | |
2928 | return 0; | |
2929 | } | |
2930 | } else if (cluster->block_group != block_group) { | |
2931 | /* someone else has already freed it don't redo their work */ | |
2932 | spin_unlock(&cluster->lock); | |
2933 | return 0; | |
2934 | } | |
2935 | atomic_inc(&block_group->count); | |
2936 | spin_unlock(&cluster->lock); | |
2937 | ||
34d52cb6 LZ |
2938 | ctl = block_group->free_space_ctl; |
2939 | ||
fa9c0d79 | 2940 | /* now return any extents the cluster had on it */ |
34d52cb6 | 2941 | spin_lock(&ctl->tree_lock); |
fa9c0d79 | 2942 | ret = __btrfs_return_cluster_to_free_space(block_group, cluster); |
34d52cb6 | 2943 | spin_unlock(&ctl->tree_lock); |
fa9c0d79 | 2944 | |
6e80d4f8 DZ |
2945 | btrfs_discard_queue_work(&block_group->fs_info->discard_ctl, block_group); |
2946 | ||
fa9c0d79 CM |
2947 | /* finally drop our ref */ |
2948 | btrfs_put_block_group(block_group); | |
2949 | return ret; | |
2950 | } | |
2951 | ||
32da5386 | 2952 | static u64 btrfs_alloc_from_bitmap(struct btrfs_block_group *block_group, |
96303081 | 2953 | struct btrfs_free_cluster *cluster, |
4e69b598 | 2954 | struct btrfs_free_space *entry, |
a4820398 MX |
2955 | u64 bytes, u64 min_start, |
2956 | u64 *max_extent_size) | |
96303081 | 2957 | { |
34d52cb6 | 2958 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
96303081 JB |
2959 | int err; |
2960 | u64 search_start = cluster->window_start; | |
2961 | u64 search_bytes = bytes; | |
2962 | u64 ret = 0; | |
2963 | ||
96303081 JB |
2964 | search_start = min_start; |
2965 | search_bytes = bytes; | |
2966 | ||
0584f718 | 2967 | err = search_bitmap(ctl, entry, &search_start, &search_bytes, true); |
a4820398 | 2968 | if (err) { |
ad22cf6e JB |
2969 | *max_extent_size = max(get_max_extent_size(entry), |
2970 | *max_extent_size); | |
4e69b598 | 2971 | return 0; |
a4820398 | 2972 | } |
96303081 JB |
2973 | |
2974 | ret = search_start; | |
bb3ac5a4 | 2975 | __bitmap_clear_bits(ctl, entry, ret, bytes); |
96303081 JB |
2976 | |
2977 | return ret; | |
2978 | } | |
2979 | ||
fa9c0d79 CM |
2980 | /* |
2981 | * given a cluster, try to allocate 'bytes' from it, returns 0 | |
2982 | * if it couldn't find anything suitably large, or a logical disk offset | |
2983 | * if things worked out | |
2984 | */ | |
32da5386 | 2985 | u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group, |
fa9c0d79 | 2986 | struct btrfs_free_cluster *cluster, u64 bytes, |
a4820398 | 2987 | u64 min_start, u64 *max_extent_size) |
fa9c0d79 | 2988 | { |
34d52cb6 | 2989 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
9ddf648f DZ |
2990 | struct btrfs_discard_ctl *discard_ctl = |
2991 | &block_group->fs_info->discard_ctl; | |
fa9c0d79 CM |
2992 | struct btrfs_free_space *entry = NULL; |
2993 | struct rb_node *node; | |
2994 | u64 ret = 0; | |
2995 | ||
2996 | spin_lock(&cluster->lock); | |
2997 | if (bytes > cluster->max_size) | |
2998 | goto out; | |
2999 | ||
3000 | if (cluster->block_group != block_group) | |
3001 | goto out; | |
3002 | ||
3003 | node = rb_first(&cluster->root); | |
3004 | if (!node) | |
3005 | goto out; | |
3006 | ||
3007 | entry = rb_entry(node, struct btrfs_free_space, offset_index); | |
67871254 | 3008 | while (1) { |
ad22cf6e JB |
3009 | if (entry->bytes < bytes) |
3010 | *max_extent_size = max(get_max_extent_size(entry), | |
3011 | *max_extent_size); | |
a4820398 | 3012 | |
4e69b598 JB |
3013 | if (entry->bytes < bytes || |
3014 | (!entry->bitmap && entry->offset < min_start)) { | |
fa9c0d79 CM |
3015 | node = rb_next(&entry->offset_index); |
3016 | if (!node) | |
3017 | break; | |
3018 | entry = rb_entry(node, struct btrfs_free_space, | |
3019 | offset_index); | |
3020 | continue; | |
3021 | } | |
fa9c0d79 | 3022 | |
4e69b598 JB |
3023 | if (entry->bitmap) { |
3024 | ret = btrfs_alloc_from_bitmap(block_group, | |
3025 | cluster, entry, bytes, | |
a4820398 MX |
3026 | cluster->window_start, |
3027 | max_extent_size); | |
4e69b598 | 3028 | if (ret == 0) { |
4e69b598 JB |
3029 | node = rb_next(&entry->offset_index); |
3030 | if (!node) | |
3031 | break; | |
3032 | entry = rb_entry(node, struct btrfs_free_space, | |
3033 | offset_index); | |
3034 | continue; | |
3035 | } | |
9b230628 | 3036 | cluster->window_start += bytes; |
4e69b598 | 3037 | } else { |
4e69b598 JB |
3038 | ret = entry->offset; |
3039 | ||
3040 | entry->offset += bytes; | |
3041 | entry->bytes -= bytes; | |
3042 | } | |
fa9c0d79 | 3043 | |
5e71b5d5 | 3044 | if (entry->bytes == 0) |
fa9c0d79 | 3045 | rb_erase(&entry->offset_index, &cluster->root); |
fa9c0d79 CM |
3046 | break; |
3047 | } | |
3048 | out: | |
3049 | spin_unlock(&cluster->lock); | |
96303081 | 3050 | |
5e71b5d5 LZ |
3051 | if (!ret) |
3052 | return 0; | |
3053 | ||
34d52cb6 | 3054 | spin_lock(&ctl->tree_lock); |
5e71b5d5 | 3055 | |
9ddf648f DZ |
3056 | if (!btrfs_free_space_trimmed(entry)) |
3057 | atomic64_add(bytes, &discard_ctl->discard_bytes_saved); | |
3058 | ||
34d52cb6 | 3059 | ctl->free_space -= bytes; |
5dc7c10b DZ |
3060 | if (!entry->bitmap && !btrfs_free_space_trimmed(entry)) |
3061 | ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes; | |
5e71b5d5 | 3062 | if (entry->bytes == 0) { |
34d52cb6 | 3063 | ctl->free_extents--; |
4e69b598 | 3064 | if (entry->bitmap) { |
3acd4850 CL |
3065 | kmem_cache_free(btrfs_free_space_bitmap_cachep, |
3066 | entry->bitmap); | |
34d52cb6 LZ |
3067 | ctl->total_bitmaps--; |
3068 | ctl->op->recalc_thresholds(ctl); | |
dfb79ddb DZ |
3069 | } else if (!btrfs_free_space_trimmed(entry)) { |
3070 | ctl->discardable_extents[BTRFS_STAT_CURR]--; | |
4e69b598 | 3071 | } |
dc89e982 | 3072 | kmem_cache_free(btrfs_free_space_cachep, entry); |
5e71b5d5 LZ |
3073 | } |
3074 | ||
34d52cb6 | 3075 | spin_unlock(&ctl->tree_lock); |
5e71b5d5 | 3076 | |
fa9c0d79 CM |
3077 | return ret; |
3078 | } | |
3079 | ||
32da5386 | 3080 | static int btrfs_bitmap_cluster(struct btrfs_block_group *block_group, |
96303081 JB |
3081 | struct btrfs_free_space *entry, |
3082 | struct btrfs_free_cluster *cluster, | |
1bb91902 AO |
3083 | u64 offset, u64 bytes, |
3084 | u64 cont1_bytes, u64 min_bytes) | |
96303081 | 3085 | { |
34d52cb6 | 3086 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
96303081 JB |
3087 | unsigned long next_zero; |
3088 | unsigned long i; | |
1bb91902 AO |
3089 | unsigned long want_bits; |
3090 | unsigned long min_bits; | |
96303081 | 3091 | unsigned long found_bits; |
cef40483 | 3092 | unsigned long max_bits = 0; |
96303081 JB |
3093 | unsigned long start = 0; |
3094 | unsigned long total_found = 0; | |
4e69b598 | 3095 | int ret; |
96303081 | 3096 | |
96009762 | 3097 | i = offset_to_bit(entry->offset, ctl->unit, |
96303081 | 3098 | max_t(u64, offset, entry->offset)); |
96009762 WSH |
3099 | want_bits = bytes_to_bits(bytes, ctl->unit); |
3100 | min_bits = bytes_to_bits(min_bytes, ctl->unit); | |
96303081 | 3101 | |
cef40483 JB |
3102 | /* |
3103 | * Don't bother looking for a cluster in this bitmap if it's heavily | |
3104 | * fragmented. | |
3105 | */ | |
3106 | if (entry->max_extent_size && | |
3107 | entry->max_extent_size < cont1_bytes) | |
3108 | return -ENOSPC; | |
96303081 JB |
3109 | again: |
3110 | found_bits = 0; | |
ebb3dad4 | 3111 | for_each_set_bit_from(i, entry->bitmap, BITS_PER_BITMAP) { |
96303081 JB |
3112 | next_zero = find_next_zero_bit(entry->bitmap, |
3113 | BITS_PER_BITMAP, i); | |
1bb91902 | 3114 | if (next_zero - i >= min_bits) { |
96303081 | 3115 | found_bits = next_zero - i; |
cef40483 JB |
3116 | if (found_bits > max_bits) |
3117 | max_bits = found_bits; | |
96303081 JB |
3118 | break; |
3119 | } | |
cef40483 JB |
3120 | if (next_zero - i > max_bits) |
3121 | max_bits = next_zero - i; | |
96303081 JB |
3122 | i = next_zero; |
3123 | } | |
3124 | ||
cef40483 JB |
3125 | if (!found_bits) { |
3126 | entry->max_extent_size = (u64)max_bits * ctl->unit; | |
4e69b598 | 3127 | return -ENOSPC; |
cef40483 | 3128 | } |
96303081 | 3129 | |
1bb91902 | 3130 | if (!total_found) { |
96303081 | 3131 | start = i; |
b78d09bc | 3132 | cluster->max_size = 0; |
96303081 JB |
3133 | } |
3134 | ||
3135 | total_found += found_bits; | |
3136 | ||
96009762 WSH |
3137 | if (cluster->max_size < found_bits * ctl->unit) |
3138 | cluster->max_size = found_bits * ctl->unit; | |
96303081 | 3139 | |
1bb91902 AO |
3140 | if (total_found < want_bits || cluster->max_size < cont1_bytes) { |
3141 | i = next_zero + 1; | |
96303081 JB |
3142 | goto again; |
3143 | } | |
3144 | ||
96009762 | 3145 | cluster->window_start = start * ctl->unit + entry->offset; |
34d52cb6 | 3146 | rb_erase(&entry->offset_index, &ctl->free_space_offset); |
4e69b598 JB |
3147 | ret = tree_insert_offset(&cluster->root, entry->offset, |
3148 | &entry->offset_index, 1); | |
b12d6869 | 3149 | ASSERT(!ret); /* -EEXIST; Logic error */ |
96303081 | 3150 | |
3f7de037 | 3151 | trace_btrfs_setup_cluster(block_group, cluster, |
96009762 | 3152 | total_found * ctl->unit, 1); |
96303081 JB |
3153 | return 0; |
3154 | } | |
3155 | ||
4e69b598 JB |
3156 | /* |
3157 | * This searches the block group for just extents to fill the cluster with. | |
1bb91902 AO |
3158 | * Try to find a cluster with at least bytes total bytes, at least one |
3159 | * extent of cont1_bytes, and other clusters of at least min_bytes. | |
4e69b598 | 3160 | */ |
3de85bb9 | 3161 | static noinline int |
32da5386 | 3162 | setup_cluster_no_bitmap(struct btrfs_block_group *block_group, |
3de85bb9 JB |
3163 | struct btrfs_free_cluster *cluster, |
3164 | struct list_head *bitmaps, u64 offset, u64 bytes, | |
1bb91902 | 3165 | u64 cont1_bytes, u64 min_bytes) |
4e69b598 | 3166 | { |
34d52cb6 | 3167 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
4e69b598 JB |
3168 | struct btrfs_free_space *first = NULL; |
3169 | struct btrfs_free_space *entry = NULL; | |
4e69b598 JB |
3170 | struct btrfs_free_space *last; |
3171 | struct rb_node *node; | |
4e69b598 JB |
3172 | u64 window_free; |
3173 | u64 max_extent; | |
3f7de037 | 3174 | u64 total_size = 0; |
4e69b598 | 3175 | |
34d52cb6 | 3176 | entry = tree_search_offset(ctl, offset, 0, 1); |
4e69b598 JB |
3177 | if (!entry) |
3178 | return -ENOSPC; | |
3179 | ||
3180 | /* | |
3181 | * We don't want bitmaps, so just move along until we find a normal | |
3182 | * extent entry. | |
3183 | */ | |
1bb91902 AO |
3184 | while (entry->bitmap || entry->bytes < min_bytes) { |
3185 | if (entry->bitmap && list_empty(&entry->list)) | |
86d4a77b | 3186 | list_add_tail(&entry->list, bitmaps); |
4e69b598 JB |
3187 | node = rb_next(&entry->offset_index); |
3188 | if (!node) | |
3189 | return -ENOSPC; | |
3190 | entry = rb_entry(node, struct btrfs_free_space, offset_index); | |
3191 | } | |
3192 | ||
4e69b598 JB |
3193 | window_free = entry->bytes; |
3194 | max_extent = entry->bytes; | |
3195 | first = entry; | |
3196 | last = entry; | |
4e69b598 | 3197 | |
1bb91902 AO |
3198 | for (node = rb_next(&entry->offset_index); node; |
3199 | node = rb_next(&entry->offset_index)) { | |
4e69b598 JB |
3200 | entry = rb_entry(node, struct btrfs_free_space, offset_index); |
3201 | ||
86d4a77b JB |
3202 | if (entry->bitmap) { |
3203 | if (list_empty(&entry->list)) | |
3204 | list_add_tail(&entry->list, bitmaps); | |
4e69b598 | 3205 | continue; |
86d4a77b JB |
3206 | } |
3207 | ||
1bb91902 AO |
3208 | if (entry->bytes < min_bytes) |
3209 | continue; | |
3210 | ||
3211 | last = entry; | |
3212 | window_free += entry->bytes; | |
3213 | if (entry->bytes > max_extent) | |
4e69b598 | 3214 | max_extent = entry->bytes; |
4e69b598 JB |
3215 | } |
3216 | ||
1bb91902 AO |
3217 | if (window_free < bytes || max_extent < cont1_bytes) |
3218 | return -ENOSPC; | |
3219 | ||
4e69b598 JB |
3220 | cluster->window_start = first->offset; |
3221 | ||
3222 | node = &first->offset_index; | |
3223 | ||
3224 | /* | |
3225 | * now we've found our entries, pull them out of the free space | |
3226 | * cache and put them into the cluster rbtree | |
3227 | */ | |
3228 | do { | |
3229 | int ret; | |
3230 | ||
3231 | entry = rb_entry(node, struct btrfs_free_space, offset_index); | |
3232 | node = rb_next(&entry->offset_index); | |
1bb91902 | 3233 | if (entry->bitmap || entry->bytes < min_bytes) |
4e69b598 JB |
3234 | continue; |
3235 | ||
34d52cb6 | 3236 | rb_erase(&entry->offset_index, &ctl->free_space_offset); |
4e69b598 JB |
3237 | ret = tree_insert_offset(&cluster->root, entry->offset, |
3238 | &entry->offset_index, 0); | |
3f7de037 | 3239 | total_size += entry->bytes; |
b12d6869 | 3240 | ASSERT(!ret); /* -EEXIST; Logic error */ |
4e69b598 JB |
3241 | } while (node && entry != last); |
3242 | ||
3243 | cluster->max_size = max_extent; | |
3f7de037 | 3244 | trace_btrfs_setup_cluster(block_group, cluster, total_size, 0); |
4e69b598 JB |
3245 | return 0; |
3246 | } | |
3247 | ||
3248 | /* | |
3249 | * This specifically looks for bitmaps that may work in the cluster, we assume | |
3250 | * that we have already failed to find extents that will work. | |
3251 | */ | |
3de85bb9 | 3252 | static noinline int |
32da5386 | 3253 | setup_cluster_bitmap(struct btrfs_block_group *block_group, |
3de85bb9 JB |
3254 | struct btrfs_free_cluster *cluster, |
3255 | struct list_head *bitmaps, u64 offset, u64 bytes, | |
1bb91902 | 3256 | u64 cont1_bytes, u64 min_bytes) |
4e69b598 | 3257 | { |
34d52cb6 | 3258 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
1b9b922a | 3259 | struct btrfs_free_space *entry = NULL; |
4e69b598 | 3260 | int ret = -ENOSPC; |
0f0fbf1d | 3261 | u64 bitmap_offset = offset_to_bitmap(ctl, offset); |
4e69b598 | 3262 | |
34d52cb6 | 3263 | if (ctl->total_bitmaps == 0) |
4e69b598 JB |
3264 | return -ENOSPC; |
3265 | ||
0f0fbf1d LZ |
3266 | /* |
3267 | * The bitmap that covers offset won't be in the list unless offset | |
3268 | * is just its start offset. | |
3269 | */ | |
1b9b922a CM |
3270 | if (!list_empty(bitmaps)) |
3271 | entry = list_first_entry(bitmaps, struct btrfs_free_space, list); | |
3272 | ||
3273 | if (!entry || entry->offset != bitmap_offset) { | |
0f0fbf1d LZ |
3274 | entry = tree_search_offset(ctl, bitmap_offset, 1, 0); |
3275 | if (entry && list_empty(&entry->list)) | |
3276 | list_add(&entry->list, bitmaps); | |
3277 | } | |
3278 | ||
86d4a77b | 3279 | list_for_each_entry(entry, bitmaps, list) { |
357b9784 | 3280 | if (entry->bytes < bytes) |
86d4a77b JB |
3281 | continue; |
3282 | ret = btrfs_bitmap_cluster(block_group, entry, cluster, offset, | |
1bb91902 | 3283 | bytes, cont1_bytes, min_bytes); |
86d4a77b JB |
3284 | if (!ret) |
3285 | return 0; | |
3286 | } | |
3287 | ||
3288 | /* | |
52621cb6 LZ |
3289 | * The bitmaps list has all the bitmaps that record free space |
3290 | * starting after offset, so no more search is required. | |
86d4a77b | 3291 | */ |
52621cb6 | 3292 | return -ENOSPC; |
4e69b598 JB |
3293 | } |
3294 | ||
fa9c0d79 CM |
3295 | /* |
3296 | * here we try to find a cluster of blocks in a block group. The goal | |
1bb91902 | 3297 | * is to find at least bytes+empty_size. |
fa9c0d79 CM |
3298 | * We might not find them all in one contiguous area. |
3299 | * | |
3300 | * returns zero and sets up cluster if things worked out, otherwise | |
3301 | * it returns -enospc | |
3302 | */ | |
32da5386 | 3303 | int btrfs_find_space_cluster(struct btrfs_block_group *block_group, |
fa9c0d79 CM |
3304 | struct btrfs_free_cluster *cluster, |
3305 | u64 offset, u64 bytes, u64 empty_size) | |
3306 | { | |
2ceeae2e | 3307 | struct btrfs_fs_info *fs_info = block_group->fs_info; |
34d52cb6 | 3308 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
86d4a77b | 3309 | struct btrfs_free_space *entry, *tmp; |
52621cb6 | 3310 | LIST_HEAD(bitmaps); |
fa9c0d79 | 3311 | u64 min_bytes; |
1bb91902 | 3312 | u64 cont1_bytes; |
fa9c0d79 CM |
3313 | int ret; |
3314 | ||
1bb91902 AO |
3315 | /* |
3316 | * Choose the minimum extent size we'll require for this | |
3317 | * cluster. For SSD_SPREAD, don't allow any fragmentation. | |
3318 | * For metadata, allow allocates with smaller extents. For | |
3319 | * data, keep it dense. | |
3320 | */ | |
0b246afa | 3321 | if (btrfs_test_opt(fs_info, SSD_SPREAD)) { |
1bb91902 | 3322 | cont1_bytes = min_bytes = bytes + empty_size; |
451d7585 | 3323 | } else if (block_group->flags & BTRFS_BLOCK_GROUP_METADATA) { |
1bb91902 | 3324 | cont1_bytes = bytes; |
0b246afa | 3325 | min_bytes = fs_info->sectorsize; |
1bb91902 AO |
3326 | } else { |
3327 | cont1_bytes = max(bytes, (bytes + empty_size) >> 2); | |
0b246afa | 3328 | min_bytes = fs_info->sectorsize; |
1bb91902 | 3329 | } |
fa9c0d79 | 3330 | |
34d52cb6 | 3331 | spin_lock(&ctl->tree_lock); |
7d0d2e8e JB |
3332 | |
3333 | /* | |
3334 | * If we know we don't have enough space to make a cluster don't even | |
3335 | * bother doing all the work to try and find one. | |
3336 | */ | |
1bb91902 | 3337 | if (ctl->free_space < bytes) { |
34d52cb6 | 3338 | spin_unlock(&ctl->tree_lock); |
7d0d2e8e JB |
3339 | return -ENOSPC; |
3340 | } | |
3341 | ||
fa9c0d79 CM |
3342 | spin_lock(&cluster->lock); |
3343 | ||
3344 | /* someone already found a cluster, hooray */ | |
3345 | if (cluster->block_group) { | |
3346 | ret = 0; | |
3347 | goto out; | |
3348 | } | |
fa9c0d79 | 3349 | |
3f7de037 JB |
3350 | trace_btrfs_find_cluster(block_group, offset, bytes, empty_size, |
3351 | min_bytes); | |
3352 | ||
86d4a77b | 3353 | ret = setup_cluster_no_bitmap(block_group, cluster, &bitmaps, offset, |
1bb91902 AO |
3354 | bytes + empty_size, |
3355 | cont1_bytes, min_bytes); | |
4e69b598 | 3356 | if (ret) |
86d4a77b | 3357 | ret = setup_cluster_bitmap(block_group, cluster, &bitmaps, |
1bb91902 AO |
3358 | offset, bytes + empty_size, |
3359 | cont1_bytes, min_bytes); | |
86d4a77b JB |
3360 | |
3361 | /* Clear our temporary list */ | |
3362 | list_for_each_entry_safe(entry, tmp, &bitmaps, list) | |
3363 | list_del_init(&entry->list); | |
fa9c0d79 | 3364 | |
4e69b598 JB |
3365 | if (!ret) { |
3366 | atomic_inc(&block_group->count); | |
3367 | list_add_tail(&cluster->block_group_list, | |
3368 | &block_group->cluster_list); | |
3369 | cluster->block_group = block_group; | |
3f7de037 JB |
3370 | } else { |
3371 | trace_btrfs_failed_cluster_setup(block_group); | |
fa9c0d79 | 3372 | } |
fa9c0d79 CM |
3373 | out: |
3374 | spin_unlock(&cluster->lock); | |
34d52cb6 | 3375 | spin_unlock(&ctl->tree_lock); |
fa9c0d79 CM |
3376 | |
3377 | return ret; | |
3378 | } | |
3379 | ||
3380 | /* | |
3381 | * simple code to zero out a cluster | |
3382 | */ | |
3383 | void btrfs_init_free_cluster(struct btrfs_free_cluster *cluster) | |
3384 | { | |
3385 | spin_lock_init(&cluster->lock); | |
3386 | spin_lock_init(&cluster->refill_lock); | |
6bef4d31 | 3387 | cluster->root = RB_ROOT; |
fa9c0d79 | 3388 | cluster->max_size = 0; |
c759c4e1 | 3389 | cluster->fragmented = false; |
fa9c0d79 CM |
3390 | INIT_LIST_HEAD(&cluster->block_group_list); |
3391 | cluster->block_group = NULL; | |
3392 | } | |
3393 | ||
32da5386 | 3394 | static int do_trimming(struct btrfs_block_group *block_group, |
7fe1e641 | 3395 | u64 *total_trimmed, u64 start, u64 bytes, |
55507ce3 | 3396 | u64 reserved_start, u64 reserved_bytes, |
b0643e59 | 3397 | enum btrfs_trim_state reserved_trim_state, |
55507ce3 | 3398 | struct btrfs_trim_range *trim_entry) |
f7039b1d | 3399 | { |
7fe1e641 | 3400 | struct btrfs_space_info *space_info = block_group->space_info; |
f7039b1d | 3401 | struct btrfs_fs_info *fs_info = block_group->fs_info; |
55507ce3 | 3402 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
7fe1e641 LZ |
3403 | int ret; |
3404 | int update = 0; | |
b0643e59 DZ |
3405 | const u64 end = start + bytes; |
3406 | const u64 reserved_end = reserved_start + reserved_bytes; | |
3407 | enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; | |
7fe1e641 | 3408 | u64 trimmed = 0; |
f7039b1d | 3409 | |
7fe1e641 LZ |
3410 | spin_lock(&space_info->lock); |
3411 | spin_lock(&block_group->lock); | |
3412 | if (!block_group->ro) { | |
3413 | block_group->reserved += reserved_bytes; | |
3414 | space_info->bytes_reserved += reserved_bytes; | |
3415 | update = 1; | |
3416 | } | |
3417 | spin_unlock(&block_group->lock); | |
3418 | spin_unlock(&space_info->lock); | |
3419 | ||
2ff7e61e | 3420 | ret = btrfs_discard_extent(fs_info, start, bytes, &trimmed); |
b0643e59 | 3421 | if (!ret) { |
7fe1e641 | 3422 | *total_trimmed += trimmed; |
b0643e59 DZ |
3423 | trim_state = BTRFS_TRIM_STATE_TRIMMED; |
3424 | } | |
7fe1e641 | 3425 | |
55507ce3 | 3426 | mutex_lock(&ctl->cache_writeout_mutex); |
b0643e59 DZ |
3427 | if (reserved_start < start) |
3428 | __btrfs_add_free_space(fs_info, ctl, reserved_start, | |
3429 | start - reserved_start, | |
3430 | reserved_trim_state); | |
3431 | if (start + bytes < reserved_start + reserved_bytes) | |
3432 | __btrfs_add_free_space(fs_info, ctl, end, reserved_end - end, | |
3433 | reserved_trim_state); | |
3434 | __btrfs_add_free_space(fs_info, ctl, start, bytes, trim_state); | |
55507ce3 FM |
3435 | list_del(&trim_entry->list); |
3436 | mutex_unlock(&ctl->cache_writeout_mutex); | |
7fe1e641 LZ |
3437 | |
3438 | if (update) { | |
3439 | spin_lock(&space_info->lock); | |
3440 | spin_lock(&block_group->lock); | |
3441 | if (block_group->ro) | |
3442 | space_info->bytes_readonly += reserved_bytes; | |
3443 | block_group->reserved -= reserved_bytes; | |
3444 | space_info->bytes_reserved -= reserved_bytes; | |
7fe1e641 | 3445 | spin_unlock(&block_group->lock); |
8f63a840 | 3446 | spin_unlock(&space_info->lock); |
7fe1e641 LZ |
3447 | } |
3448 | ||
3449 | return ret; | |
3450 | } | |
3451 | ||
2bee7eb8 DZ |
3452 | /* |
3453 | * If @async is set, then we will trim 1 region and return. | |
3454 | */ | |
32da5386 | 3455 | static int trim_no_bitmap(struct btrfs_block_group *block_group, |
2bee7eb8 DZ |
3456 | u64 *total_trimmed, u64 start, u64 end, u64 minlen, |
3457 | bool async) | |
7fe1e641 | 3458 | { |
19b2a2c7 DZ |
3459 | struct btrfs_discard_ctl *discard_ctl = |
3460 | &block_group->fs_info->discard_ctl; | |
7fe1e641 LZ |
3461 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
3462 | struct btrfs_free_space *entry; | |
3463 | struct rb_node *node; | |
3464 | int ret = 0; | |
3465 | u64 extent_start; | |
3466 | u64 extent_bytes; | |
b0643e59 | 3467 | enum btrfs_trim_state extent_trim_state; |
7fe1e641 | 3468 | u64 bytes; |
19b2a2c7 | 3469 | const u64 max_discard_size = READ_ONCE(discard_ctl->max_discard_size); |
f7039b1d LD |
3470 | |
3471 | while (start < end) { | |
55507ce3 FM |
3472 | struct btrfs_trim_range trim_entry; |
3473 | ||
3474 | mutex_lock(&ctl->cache_writeout_mutex); | |
34d52cb6 | 3475 | spin_lock(&ctl->tree_lock); |
f7039b1d | 3476 | |
2bee7eb8 DZ |
3477 | if (ctl->free_space < minlen) |
3478 | goto out_unlock; | |
f7039b1d | 3479 | |
34d52cb6 | 3480 | entry = tree_search_offset(ctl, start, 0, 1); |
2bee7eb8 DZ |
3481 | if (!entry) |
3482 | goto out_unlock; | |
f7039b1d | 3483 | |
2bee7eb8 DZ |
3484 | /* Skip bitmaps and if async, already trimmed entries */ |
3485 | while (entry->bitmap || | |
3486 | (async && btrfs_free_space_trimmed(entry))) { | |
7fe1e641 | 3487 | node = rb_next(&entry->offset_index); |
2bee7eb8 DZ |
3488 | if (!node) |
3489 | goto out_unlock; | |
7fe1e641 LZ |
3490 | entry = rb_entry(node, struct btrfs_free_space, |
3491 | offset_index); | |
f7039b1d LD |
3492 | } |
3493 | ||
2bee7eb8 DZ |
3494 | if (entry->offset >= end) |
3495 | goto out_unlock; | |
f7039b1d | 3496 | |
7fe1e641 LZ |
3497 | extent_start = entry->offset; |
3498 | extent_bytes = entry->bytes; | |
b0643e59 | 3499 | extent_trim_state = entry->trim_state; |
4aa9ad52 DZ |
3500 | if (async) { |
3501 | start = entry->offset; | |
3502 | bytes = entry->bytes; | |
3503 | if (bytes < minlen) { | |
3504 | spin_unlock(&ctl->tree_lock); | |
3505 | mutex_unlock(&ctl->cache_writeout_mutex); | |
3506 | goto next; | |
3507 | } | |
3508 | unlink_free_space(ctl, entry); | |
7fe6d45e DZ |
3509 | /* |
3510 | * Let bytes = BTRFS_MAX_DISCARD_SIZE + X. | |
3511 | * If X < BTRFS_ASYNC_DISCARD_MIN_FILTER, we won't trim | |
3512 | * X when we come back around. So trim it now. | |
3513 | */ | |
3514 | if (max_discard_size && | |
3515 | bytes >= (max_discard_size + | |
3516 | BTRFS_ASYNC_DISCARD_MIN_FILTER)) { | |
19b2a2c7 DZ |
3517 | bytes = max_discard_size; |
3518 | extent_bytes = max_discard_size; | |
3519 | entry->offset += max_discard_size; | |
3520 | entry->bytes -= max_discard_size; | |
4aa9ad52 DZ |
3521 | link_free_space(ctl, entry); |
3522 | } else { | |
3523 | kmem_cache_free(btrfs_free_space_cachep, entry); | |
3524 | } | |
3525 | } else { | |
3526 | start = max(start, extent_start); | |
3527 | bytes = min(extent_start + extent_bytes, end) - start; | |
3528 | if (bytes < minlen) { | |
3529 | spin_unlock(&ctl->tree_lock); | |
3530 | mutex_unlock(&ctl->cache_writeout_mutex); | |
3531 | goto next; | |
3532 | } | |
f7039b1d | 3533 | |
4aa9ad52 DZ |
3534 | unlink_free_space(ctl, entry); |
3535 | kmem_cache_free(btrfs_free_space_cachep, entry); | |
3536 | } | |
7fe1e641 | 3537 | |
34d52cb6 | 3538 | spin_unlock(&ctl->tree_lock); |
55507ce3 FM |
3539 | trim_entry.start = extent_start; |
3540 | trim_entry.bytes = extent_bytes; | |
3541 | list_add_tail(&trim_entry.list, &ctl->trimming_ranges); | |
3542 | mutex_unlock(&ctl->cache_writeout_mutex); | |
f7039b1d | 3543 | |
7fe1e641 | 3544 | ret = do_trimming(block_group, total_trimmed, start, bytes, |
b0643e59 DZ |
3545 | extent_start, extent_bytes, extent_trim_state, |
3546 | &trim_entry); | |
2bee7eb8 DZ |
3547 | if (ret) { |
3548 | block_group->discard_cursor = start + bytes; | |
7fe1e641 | 3549 | break; |
2bee7eb8 | 3550 | } |
7fe1e641 LZ |
3551 | next: |
3552 | start += bytes; | |
2bee7eb8 DZ |
3553 | block_group->discard_cursor = start; |
3554 | if (async && *total_trimmed) | |
3555 | break; | |
f7039b1d | 3556 | |
7fe1e641 LZ |
3557 | if (fatal_signal_pending(current)) { |
3558 | ret = -ERESTARTSYS; | |
3559 | break; | |
3560 | } | |
3561 | ||
3562 | cond_resched(); | |
3563 | } | |
2bee7eb8 DZ |
3564 | |
3565 | return ret; | |
3566 | ||
3567 | out_unlock: | |
3568 | block_group->discard_cursor = btrfs_block_group_end(block_group); | |
3569 | spin_unlock(&ctl->tree_lock); | |
3570 | mutex_unlock(&ctl->cache_writeout_mutex); | |
3571 | ||
7fe1e641 LZ |
3572 | return ret; |
3573 | } | |
3574 | ||
da080fe1 DZ |
3575 | /* |
3576 | * If we break out of trimming a bitmap prematurely, we should reset the | |
3577 | * trimming bit. In a rather contrieved case, it's possible to race here so | |
3578 | * reset the state to BTRFS_TRIM_STATE_UNTRIMMED. | |
3579 | * | |
3580 | * start = start of bitmap | |
3581 | * end = near end of bitmap | |
3582 | * | |
3583 | * Thread 1: Thread 2: | |
3584 | * trim_bitmaps(start) | |
3585 | * trim_bitmaps(end) | |
3586 | * end_trimming_bitmap() | |
3587 | * reset_trimming_bitmap() | |
3588 | */ | |
3589 | static void reset_trimming_bitmap(struct btrfs_free_space_ctl *ctl, u64 offset) | |
3590 | { | |
3591 | struct btrfs_free_space *entry; | |
3592 | ||
3593 | spin_lock(&ctl->tree_lock); | |
3594 | entry = tree_search_offset(ctl, offset, 1, 0); | |
dfb79ddb | 3595 | if (entry) { |
5dc7c10b | 3596 | if (btrfs_free_space_trimmed(entry)) { |
dfb79ddb DZ |
3597 | ctl->discardable_extents[BTRFS_STAT_CURR] += |
3598 | entry->bitmap_extents; | |
5dc7c10b DZ |
3599 | ctl->discardable_bytes[BTRFS_STAT_CURR] += entry->bytes; |
3600 | } | |
da080fe1 | 3601 | entry->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; |
dfb79ddb DZ |
3602 | } |
3603 | ||
da080fe1 DZ |
3604 | spin_unlock(&ctl->tree_lock); |
3605 | } | |
3606 | ||
dfb79ddb DZ |
3607 | static void end_trimming_bitmap(struct btrfs_free_space_ctl *ctl, |
3608 | struct btrfs_free_space *entry) | |
da080fe1 | 3609 | { |
dfb79ddb | 3610 | if (btrfs_free_space_trimming_bitmap(entry)) { |
da080fe1 | 3611 | entry->trim_state = BTRFS_TRIM_STATE_TRIMMED; |
dfb79ddb DZ |
3612 | ctl->discardable_extents[BTRFS_STAT_CURR] -= |
3613 | entry->bitmap_extents; | |
5dc7c10b | 3614 | ctl->discardable_bytes[BTRFS_STAT_CURR] -= entry->bytes; |
dfb79ddb | 3615 | } |
da080fe1 DZ |
3616 | } |
3617 | ||
2bee7eb8 DZ |
3618 | /* |
3619 | * If @async is set, then we will trim 1 region and return. | |
3620 | */ | |
32da5386 | 3621 | static int trim_bitmaps(struct btrfs_block_group *block_group, |
2bee7eb8 | 3622 | u64 *total_trimmed, u64 start, u64 end, u64 minlen, |
7fe6d45e | 3623 | u64 maxlen, bool async) |
7fe1e641 | 3624 | { |
19b2a2c7 DZ |
3625 | struct btrfs_discard_ctl *discard_ctl = |
3626 | &block_group->fs_info->discard_ctl; | |
7fe1e641 LZ |
3627 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
3628 | struct btrfs_free_space *entry; | |
3629 | int ret = 0; | |
3630 | int ret2; | |
3631 | u64 bytes; | |
3632 | u64 offset = offset_to_bitmap(ctl, start); | |
19b2a2c7 | 3633 | const u64 max_discard_size = READ_ONCE(discard_ctl->max_discard_size); |
7fe1e641 LZ |
3634 | |
3635 | while (offset < end) { | |
3636 | bool next_bitmap = false; | |
55507ce3 | 3637 | struct btrfs_trim_range trim_entry; |
7fe1e641 | 3638 | |
55507ce3 | 3639 | mutex_lock(&ctl->cache_writeout_mutex); |
7fe1e641 LZ |
3640 | spin_lock(&ctl->tree_lock); |
3641 | ||
3642 | if (ctl->free_space < minlen) { | |
2bee7eb8 DZ |
3643 | block_group->discard_cursor = |
3644 | btrfs_block_group_end(block_group); | |
7fe1e641 | 3645 | spin_unlock(&ctl->tree_lock); |
55507ce3 | 3646 | mutex_unlock(&ctl->cache_writeout_mutex); |
7fe1e641 LZ |
3647 | break; |
3648 | } | |
3649 | ||
3650 | entry = tree_search_offset(ctl, offset, 1, 0); | |
7fe6d45e DZ |
3651 | /* |
3652 | * Bitmaps are marked trimmed lossily now to prevent constant | |
3653 | * discarding of the same bitmap (the reason why we are bound | |
3654 | * by the filters). So, retrim the block group bitmaps when we | |
3655 | * are preparing to punt to the unused_bgs list. This uses | |
3656 | * @minlen to determine if we are in BTRFS_DISCARD_INDEX_UNUSED | |
3657 | * which is the only discard index which sets minlen to 0. | |
3658 | */ | |
3659 | if (!entry || (async && minlen && start == offset && | |
2bee7eb8 | 3660 | btrfs_free_space_trimmed(entry))) { |
7fe1e641 | 3661 | spin_unlock(&ctl->tree_lock); |
55507ce3 | 3662 | mutex_unlock(&ctl->cache_writeout_mutex); |
7fe1e641 LZ |
3663 | next_bitmap = true; |
3664 | goto next; | |
3665 | } | |
3666 | ||
da080fe1 DZ |
3667 | /* |
3668 | * Async discard bitmap trimming begins at by setting the start | |
3669 | * to be key.objectid and the offset_to_bitmap() aligns to the | |
3670 | * start of the bitmap. This lets us know we are fully | |
3671 | * scanning the bitmap rather than only some portion of it. | |
3672 | */ | |
3673 | if (start == offset) | |
3674 | entry->trim_state = BTRFS_TRIM_STATE_TRIMMING; | |
3675 | ||
7fe1e641 | 3676 | bytes = minlen; |
0584f718 | 3677 | ret2 = search_bitmap(ctl, entry, &start, &bytes, false); |
7fe1e641 | 3678 | if (ret2 || start >= end) { |
da080fe1 | 3679 | /* |
7fe6d45e DZ |
3680 | * We lossily consider a bitmap trimmed if we only skip |
3681 | * over regions <= BTRFS_ASYNC_DISCARD_MIN_FILTER. | |
da080fe1 | 3682 | */ |
7fe6d45e | 3683 | if (ret2 && minlen <= BTRFS_ASYNC_DISCARD_MIN_FILTER) |
dfb79ddb | 3684 | end_trimming_bitmap(ctl, entry); |
da080fe1 DZ |
3685 | else |
3686 | entry->trim_state = BTRFS_TRIM_STATE_UNTRIMMED; | |
7fe1e641 | 3687 | spin_unlock(&ctl->tree_lock); |
55507ce3 | 3688 | mutex_unlock(&ctl->cache_writeout_mutex); |
7fe1e641 LZ |
3689 | next_bitmap = true; |
3690 | goto next; | |
3691 | } | |
3692 | ||
2bee7eb8 DZ |
3693 | /* |
3694 | * We already trimmed a region, but are using the locking above | |
3695 | * to reset the trim_state. | |
3696 | */ | |
3697 | if (async && *total_trimmed) { | |
3698 | spin_unlock(&ctl->tree_lock); | |
3699 | mutex_unlock(&ctl->cache_writeout_mutex); | |
3700 | goto out; | |
3701 | } | |
3702 | ||
7fe1e641 | 3703 | bytes = min(bytes, end - start); |
7fe6d45e | 3704 | if (bytes < minlen || (async && maxlen && bytes > maxlen)) { |
7fe1e641 | 3705 | spin_unlock(&ctl->tree_lock); |
55507ce3 | 3706 | mutex_unlock(&ctl->cache_writeout_mutex); |
7fe1e641 LZ |
3707 | goto next; |
3708 | } | |
3709 | ||
7fe6d45e DZ |
3710 | /* |
3711 | * Let bytes = BTRFS_MAX_DISCARD_SIZE + X. | |
3712 | * If X < @minlen, we won't trim X when we come back around. | |
3713 | * So trim it now. We differ here from trimming extents as we | |
3714 | * don't keep individual state per bit. | |
3715 | */ | |
3716 | if (async && | |
3717 | max_discard_size && | |
3718 | bytes > (max_discard_size + minlen)) | |
19b2a2c7 | 3719 | bytes = max_discard_size; |
4aa9ad52 | 3720 | |
7fe1e641 LZ |
3721 | bitmap_clear_bits(ctl, entry, start, bytes); |
3722 | if (entry->bytes == 0) | |
3723 | free_bitmap(ctl, entry); | |
3724 | ||
3725 | spin_unlock(&ctl->tree_lock); | |
55507ce3 FM |
3726 | trim_entry.start = start; |
3727 | trim_entry.bytes = bytes; | |
3728 | list_add_tail(&trim_entry.list, &ctl->trimming_ranges); | |
3729 | mutex_unlock(&ctl->cache_writeout_mutex); | |
7fe1e641 LZ |
3730 | |
3731 | ret = do_trimming(block_group, total_trimmed, start, bytes, | |
b0643e59 | 3732 | start, bytes, 0, &trim_entry); |
da080fe1 DZ |
3733 | if (ret) { |
3734 | reset_trimming_bitmap(ctl, offset); | |
2bee7eb8 DZ |
3735 | block_group->discard_cursor = |
3736 | btrfs_block_group_end(block_group); | |
7fe1e641 | 3737 | break; |
da080fe1 | 3738 | } |
7fe1e641 LZ |
3739 | next: |
3740 | if (next_bitmap) { | |
3741 | offset += BITS_PER_BITMAP * ctl->unit; | |
da080fe1 | 3742 | start = offset; |
7fe1e641 LZ |
3743 | } else { |
3744 | start += bytes; | |
f7039b1d | 3745 | } |
2bee7eb8 | 3746 | block_group->discard_cursor = start; |
f7039b1d LD |
3747 | |
3748 | if (fatal_signal_pending(current)) { | |
da080fe1 DZ |
3749 | if (start != offset) |
3750 | reset_trimming_bitmap(ctl, offset); | |
f7039b1d LD |
3751 | ret = -ERESTARTSYS; |
3752 | break; | |
3753 | } | |
3754 | ||
3755 | cond_resched(); | |
3756 | } | |
3757 | ||
2bee7eb8 DZ |
3758 | if (offset >= end) |
3759 | block_group->discard_cursor = end; | |
3760 | ||
3761 | out: | |
f7039b1d LD |
3762 | return ret; |
3763 | } | |
581bb050 | 3764 | |
32da5386 | 3765 | int btrfs_trim_block_group(struct btrfs_block_group *block_group, |
e33e17ee JM |
3766 | u64 *trimmed, u64 start, u64 end, u64 minlen) |
3767 | { | |
da080fe1 | 3768 | struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; |
e33e17ee | 3769 | int ret; |
da080fe1 | 3770 | u64 rem = 0; |
e33e17ee JM |
3771 | |
3772 | *trimmed = 0; | |
3773 | ||
3774 | spin_lock(&block_group->lock); | |
3775 | if (block_group->removed) { | |
04216820 | 3776 | spin_unlock(&block_group->lock); |
e33e17ee | 3777 | return 0; |
04216820 | 3778 | } |
6b7304af | 3779 | btrfs_freeze_block_group(block_group); |
e33e17ee JM |
3780 | spin_unlock(&block_group->lock); |
3781 | ||
2bee7eb8 | 3782 | ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, false); |
e33e17ee JM |
3783 | if (ret) |
3784 | goto out; | |
7fe1e641 | 3785 | |
7fe6d45e | 3786 | ret = trim_bitmaps(block_group, trimmed, start, end, minlen, 0, false); |
da080fe1 DZ |
3787 | div64_u64_rem(end, BITS_PER_BITMAP * ctl->unit, &rem); |
3788 | /* If we ended in the middle of a bitmap, reset the trimming flag */ | |
3789 | if (rem) | |
3790 | reset_trimming_bitmap(ctl, offset_to_bitmap(ctl, end)); | |
e33e17ee | 3791 | out: |
6b7304af | 3792 | btrfs_unfreeze_block_group(block_group); |
7fe1e641 LZ |
3793 | return ret; |
3794 | } | |
3795 | ||
2bee7eb8 DZ |
3796 | int btrfs_trim_block_group_extents(struct btrfs_block_group *block_group, |
3797 | u64 *trimmed, u64 start, u64 end, u64 minlen, | |
3798 | bool async) | |
3799 | { | |
3800 | int ret; | |
3801 | ||
3802 | *trimmed = 0; | |
3803 | ||
3804 | spin_lock(&block_group->lock); | |
3805 | if (block_group->removed) { | |
3806 | spin_unlock(&block_group->lock); | |
3807 | return 0; | |
3808 | } | |
6b7304af | 3809 | btrfs_freeze_block_group(block_group); |
2bee7eb8 DZ |
3810 | spin_unlock(&block_group->lock); |
3811 | ||
3812 | ret = trim_no_bitmap(block_group, trimmed, start, end, minlen, async); | |
6b7304af | 3813 | btrfs_unfreeze_block_group(block_group); |
2bee7eb8 DZ |
3814 | |
3815 | return ret; | |
3816 | } | |
3817 | ||
3818 | int btrfs_trim_block_group_bitmaps(struct btrfs_block_group *block_group, | |
3819 | u64 *trimmed, u64 start, u64 end, u64 minlen, | |
7fe6d45e | 3820 | u64 maxlen, bool async) |
2bee7eb8 DZ |
3821 | { |
3822 | int ret; | |
3823 | ||
3824 | *trimmed = 0; | |
3825 | ||
3826 | spin_lock(&block_group->lock); | |
3827 | if (block_group->removed) { | |
3828 | spin_unlock(&block_group->lock); | |
3829 | return 0; | |
3830 | } | |
6b7304af | 3831 | btrfs_freeze_block_group(block_group); |
2bee7eb8 DZ |
3832 | spin_unlock(&block_group->lock); |
3833 | ||
7fe6d45e DZ |
3834 | ret = trim_bitmaps(block_group, trimmed, start, end, minlen, maxlen, |
3835 | async); | |
3836 | ||
6b7304af | 3837 | btrfs_unfreeze_block_group(block_group); |
2bee7eb8 DZ |
3838 | |
3839 | return ret; | |
3840 | } | |
3841 | ||
581bb050 LZ |
3842 | /* |
3843 | * Find the left-most item in the cache tree, and then return the | |
3844 | * smallest inode number in the item. | |
3845 | * | |
3846 | * Note: the returned inode number may not be the smallest one in | |
3847 | * the tree, if the left-most item is a bitmap. | |
3848 | */ | |
3849 | u64 btrfs_find_ino_for_alloc(struct btrfs_root *fs_root) | |
3850 | { | |
3851 | struct btrfs_free_space_ctl *ctl = fs_root->free_ino_ctl; | |
3852 | struct btrfs_free_space *entry = NULL; | |
3853 | u64 ino = 0; | |
3854 | ||
3855 | spin_lock(&ctl->tree_lock); | |
3856 | ||
3857 | if (RB_EMPTY_ROOT(&ctl->free_space_offset)) | |
3858 | goto out; | |
3859 | ||
3860 | entry = rb_entry(rb_first(&ctl->free_space_offset), | |
3861 | struct btrfs_free_space, offset_index); | |
3862 | ||
3863 | if (!entry->bitmap) { | |
3864 | ino = entry->offset; | |
3865 | ||
3866 | unlink_free_space(ctl, entry); | |
3867 | entry->offset++; | |
3868 | entry->bytes--; | |
3869 | if (!entry->bytes) | |
3870 | kmem_cache_free(btrfs_free_space_cachep, entry); | |
3871 | else | |
3872 | link_free_space(ctl, entry); | |
3873 | } else { | |
3874 | u64 offset = 0; | |
3875 | u64 count = 1; | |
3876 | int ret; | |
3877 | ||
0584f718 | 3878 | ret = search_bitmap(ctl, entry, &offset, &count, true); |
79787eaa | 3879 | /* Logic error; Should be empty if it can't find anything */ |
b12d6869 | 3880 | ASSERT(!ret); |
581bb050 LZ |
3881 | |
3882 | ino = offset; | |
3883 | bitmap_clear_bits(ctl, entry, offset, 1); | |
3884 | if (entry->bytes == 0) | |
3885 | free_bitmap(ctl, entry); | |
3886 | } | |
3887 | out: | |
3888 | spin_unlock(&ctl->tree_lock); | |
3889 | ||
3890 | return ino; | |
3891 | } | |
82d5902d LZ |
3892 | |
3893 | struct inode *lookup_free_ino_inode(struct btrfs_root *root, | |
3894 | struct btrfs_path *path) | |
3895 | { | |
3896 | struct inode *inode = NULL; | |
3897 | ||
57cdc8db DS |
3898 | spin_lock(&root->ino_cache_lock); |
3899 | if (root->ino_cache_inode) | |
3900 | inode = igrab(root->ino_cache_inode); | |
3901 | spin_unlock(&root->ino_cache_lock); | |
82d5902d LZ |
3902 | if (inode) |
3903 | return inode; | |
3904 | ||
3905 | inode = __lookup_free_space_inode(root, path, 0); | |
3906 | if (IS_ERR(inode)) | |
3907 | return inode; | |
3908 | ||
57cdc8db | 3909 | spin_lock(&root->ino_cache_lock); |
7841cb28 | 3910 | if (!btrfs_fs_closing(root->fs_info)) |
57cdc8db DS |
3911 | root->ino_cache_inode = igrab(inode); |
3912 | spin_unlock(&root->ino_cache_lock); | |
82d5902d LZ |
3913 | |
3914 | return inode; | |
3915 | } | |
3916 | ||
3917 | int create_free_ino_inode(struct btrfs_root *root, | |
3918 | struct btrfs_trans_handle *trans, | |
3919 | struct btrfs_path *path) | |
3920 | { | |
3921 | return __create_free_space_inode(root, trans, path, | |
3922 | BTRFS_FREE_INO_OBJECTID, 0); | |
3923 | } | |
3924 | ||
3925 | int load_free_ino_cache(struct btrfs_fs_info *fs_info, struct btrfs_root *root) | |
3926 | { | |
3927 | struct btrfs_free_space_ctl *ctl = root->free_ino_ctl; | |
3928 | struct btrfs_path *path; | |
3929 | struct inode *inode; | |
3930 | int ret = 0; | |
3931 | u64 root_gen = btrfs_root_generation(&root->root_item); | |
3932 | ||
0b246afa | 3933 | if (!btrfs_test_opt(fs_info, INODE_MAP_CACHE)) |
4b9465cb CM |
3934 | return 0; |
3935 | ||
82d5902d LZ |
3936 | /* |
3937 | * If we're unmounting then just return, since this does a search on the | |
3938 | * normal root and not the commit root and we could deadlock. | |
3939 | */ | |
7841cb28 | 3940 | if (btrfs_fs_closing(fs_info)) |
82d5902d LZ |
3941 | return 0; |
3942 | ||
3943 | path = btrfs_alloc_path(); | |
3944 | if (!path) | |
3945 | return 0; | |
3946 | ||
3947 | inode = lookup_free_ino_inode(root, path); | |
3948 | if (IS_ERR(inode)) | |
3949 | goto out; | |
3950 | ||
3951 | if (root_gen != BTRFS_I(inode)->generation) | |
3952 | goto out_put; | |
3953 | ||
3954 | ret = __load_free_space_cache(root, inode, ctl, path, 0); | |
3955 | ||
3956 | if (ret < 0) | |
c2cf52eb SK |
3957 | btrfs_err(fs_info, |
3958 | "failed to load free ino cache for root %llu", | |
3959 | root->root_key.objectid); | |
82d5902d LZ |
3960 | out_put: |
3961 | iput(inode); | |
3962 | out: | |
3963 | btrfs_free_path(path); | |
3964 | return ret; | |
3965 | } | |
3966 | ||
3967 | int btrfs_write_out_ino_cache(struct btrfs_root *root, | |
3968 | struct btrfs_trans_handle *trans, | |
53645a91 FDBM |
3969 | struct btrfs_path *path, |
3970 | struct inode *inode) | |
82d5902d | 3971 | { |
0b246afa | 3972 | struct btrfs_fs_info *fs_info = root->fs_info; |
82d5902d | 3973 | struct btrfs_free_space_ctl *ctl = root->free_ino_ctl; |
82d5902d | 3974 | int ret; |
c9dc4c65 | 3975 | struct btrfs_io_ctl io_ctl; |
e43699d4 | 3976 | bool release_metadata = true; |
82d5902d | 3977 | |
0b246afa | 3978 | if (!btrfs_test_opt(fs_info, INODE_MAP_CACHE)) |
4b9465cb CM |
3979 | return 0; |
3980 | ||
85db36cf | 3981 | memset(&io_ctl, 0, sizeof(io_ctl)); |
0e8d931a | 3982 | ret = __btrfs_write_out_cache(root, inode, ctl, NULL, &io_ctl, trans); |
e43699d4 FM |
3983 | if (!ret) { |
3984 | /* | |
3985 | * At this point writepages() didn't error out, so our metadata | |
3986 | * reservation is released when the writeback finishes, at | |
3987 | * inode.c:btrfs_finish_ordered_io(), regardless of it finishing | |
3988 | * with or without an error. | |
3989 | */ | |
3990 | release_metadata = false; | |
afdb5718 | 3991 | ret = btrfs_wait_cache_io_root(root, trans, &io_ctl, path); |
e43699d4 | 3992 | } |
85db36cf | 3993 | |
c09544e0 | 3994 | if (ret) { |
e43699d4 | 3995 | if (release_metadata) |
691fa059 | 3996 | btrfs_delalloc_release_metadata(BTRFS_I(inode), |
43b18595 | 3997 | inode->i_size, true); |
1afb648e | 3998 | #ifdef CONFIG_BTRFS_DEBUG |
0b246afa JM |
3999 | btrfs_err(fs_info, |
4000 | "failed to write free ino cache for root %llu", | |
4001 | root->root_key.objectid); | |
c09544e0 JB |
4002 | #endif |
4003 | } | |
82d5902d | 4004 | |
82d5902d LZ |
4005 | return ret; |
4006 | } | |
74255aa0 JB |
4007 | |
4008 | #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS | |
dc11dd5d JB |
4009 | /* |
4010 | * Use this if you need to make a bitmap or extent entry specifically, it | |
4011 | * doesn't do any of the merging that add_free_space does, this acts a lot like | |
4012 | * how the free space cache loading stuff works, so you can get really weird | |
4013 | * configurations. | |
4014 | */ | |
32da5386 | 4015 | int test_add_free_space_entry(struct btrfs_block_group *cache, |
dc11dd5d | 4016 | u64 offset, u64 bytes, bool bitmap) |
74255aa0 | 4017 | { |
dc11dd5d JB |
4018 | struct btrfs_free_space_ctl *ctl = cache->free_space_ctl; |
4019 | struct btrfs_free_space *info = NULL, *bitmap_info; | |
4020 | void *map = NULL; | |
da080fe1 | 4021 | enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_TRIMMED; |
dc11dd5d JB |
4022 | u64 bytes_added; |
4023 | int ret; | |
74255aa0 | 4024 | |
dc11dd5d JB |
4025 | again: |
4026 | if (!info) { | |
4027 | info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); | |
4028 | if (!info) | |
4029 | return -ENOMEM; | |
74255aa0 JB |
4030 | } |
4031 | ||
dc11dd5d JB |
4032 | if (!bitmap) { |
4033 | spin_lock(&ctl->tree_lock); | |
4034 | info->offset = offset; | |
4035 | info->bytes = bytes; | |
cef40483 | 4036 | info->max_extent_size = 0; |
dc11dd5d JB |
4037 | ret = link_free_space(ctl, info); |
4038 | spin_unlock(&ctl->tree_lock); | |
4039 | if (ret) | |
4040 | kmem_cache_free(btrfs_free_space_cachep, info); | |
4041 | return ret; | |
4042 | } | |
4043 | ||
4044 | if (!map) { | |
3acd4850 | 4045 | map = kmem_cache_zalloc(btrfs_free_space_bitmap_cachep, GFP_NOFS); |
dc11dd5d JB |
4046 | if (!map) { |
4047 | kmem_cache_free(btrfs_free_space_cachep, info); | |
4048 | return -ENOMEM; | |
4049 | } | |
4050 | } | |
4051 | ||
4052 | spin_lock(&ctl->tree_lock); | |
4053 | bitmap_info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), | |
4054 | 1, 0); | |
4055 | if (!bitmap_info) { | |
4056 | info->bitmap = map; | |
4057 | map = NULL; | |
4058 | add_new_bitmap(ctl, info, offset); | |
4059 | bitmap_info = info; | |
20005523 | 4060 | info = NULL; |
dc11dd5d | 4061 | } |
74255aa0 | 4062 | |
da080fe1 DZ |
4063 | bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes, |
4064 | trim_state); | |
cef40483 | 4065 | |
dc11dd5d JB |
4066 | bytes -= bytes_added; |
4067 | offset += bytes_added; | |
4068 | spin_unlock(&ctl->tree_lock); | |
74255aa0 | 4069 | |
dc11dd5d JB |
4070 | if (bytes) |
4071 | goto again; | |
74255aa0 | 4072 | |
20005523 FM |
4073 | if (info) |
4074 | kmem_cache_free(btrfs_free_space_cachep, info); | |
3acd4850 CL |
4075 | if (map) |
4076 | kmem_cache_free(btrfs_free_space_bitmap_cachep, map); | |
dc11dd5d | 4077 | return 0; |
74255aa0 JB |
4078 | } |
4079 | ||
4080 | /* | |
4081 | * Checks to see if the given range is in the free space cache. This is really | |
4082 | * just used to check the absence of space, so if there is free space in the | |
4083 | * range at all we will return 1. | |
4084 | */ | |
32da5386 | 4085 | int test_check_exists(struct btrfs_block_group *cache, |
dc11dd5d | 4086 | u64 offset, u64 bytes) |
74255aa0 JB |
4087 | { |
4088 | struct btrfs_free_space_ctl *ctl = cache->free_space_ctl; | |
4089 | struct btrfs_free_space *info; | |
4090 | int ret = 0; | |
4091 | ||
4092 | spin_lock(&ctl->tree_lock); | |
4093 | info = tree_search_offset(ctl, offset, 0, 0); | |
4094 | if (!info) { | |
4095 | info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), | |
4096 | 1, 0); | |
4097 | if (!info) | |
4098 | goto out; | |
4099 | } | |
4100 | ||
4101 | have_info: | |
4102 | if (info->bitmap) { | |
4103 | u64 bit_off, bit_bytes; | |
4104 | struct rb_node *n; | |
4105 | struct btrfs_free_space *tmp; | |
4106 | ||
4107 | bit_off = offset; | |
4108 | bit_bytes = ctl->unit; | |
0584f718 | 4109 | ret = search_bitmap(ctl, info, &bit_off, &bit_bytes, false); |
74255aa0 JB |
4110 | if (!ret) { |
4111 | if (bit_off == offset) { | |
4112 | ret = 1; | |
4113 | goto out; | |
4114 | } else if (bit_off > offset && | |
4115 | offset + bytes > bit_off) { | |
4116 | ret = 1; | |
4117 | goto out; | |
4118 | } | |
4119 | } | |
4120 | ||
4121 | n = rb_prev(&info->offset_index); | |
4122 | while (n) { | |
4123 | tmp = rb_entry(n, struct btrfs_free_space, | |
4124 | offset_index); | |
4125 | if (tmp->offset + tmp->bytes < offset) | |
4126 | break; | |
4127 | if (offset + bytes < tmp->offset) { | |
5473e0c4 | 4128 | n = rb_prev(&tmp->offset_index); |
74255aa0 JB |
4129 | continue; |
4130 | } | |
4131 | info = tmp; | |
4132 | goto have_info; | |
4133 | } | |
4134 | ||
4135 | n = rb_next(&info->offset_index); | |
4136 | while (n) { | |
4137 | tmp = rb_entry(n, struct btrfs_free_space, | |
4138 | offset_index); | |
4139 | if (offset + bytes < tmp->offset) | |
4140 | break; | |
4141 | if (tmp->offset + tmp->bytes < offset) { | |
5473e0c4 | 4142 | n = rb_next(&tmp->offset_index); |
74255aa0 JB |
4143 | continue; |
4144 | } | |
4145 | info = tmp; | |
4146 | goto have_info; | |
4147 | } | |
4148 | ||
20005523 | 4149 | ret = 0; |
74255aa0 JB |
4150 | goto out; |
4151 | } | |
4152 | ||
4153 | if (info->offset == offset) { | |
4154 | ret = 1; | |
4155 | goto out; | |
4156 | } | |
4157 | ||
4158 | if (offset > info->offset && offset < info->offset + info->bytes) | |
4159 | ret = 1; | |
4160 | out: | |
4161 | spin_unlock(&ctl->tree_lock); | |
4162 | return ret; | |
4163 | } | |
dc11dd5d | 4164 | #endif /* CONFIG_BTRFS_FS_RUN_SANITY_TESTS */ |