]> git.proxmox.com Git - qemu.git/blob - block/qcow2-cache.c
qcow2: Add QcowCache
[qemu.git] / block / qcow2-cache.c
1 /*
2 * L2/refcount table cache for the QCOW2 format
3 *
4 * Copyright (c) 2010 Kevin Wolf <kwolf@redhat.com>
5 *
6 * Permission is hereby granted, free of charge, to any person obtaining a copy
7 * of this software and associated documentation files (the "Software"), to deal
8 * in the Software without restriction, including without limitation the rights
9 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
10 * copies of the Software, and to permit persons to whom the Software is
11 * furnished to do so, subject to the following conditions:
12 *
13 * The above copyright notice and this permission notice shall be included in
14 * all copies or substantial portions of the Software.
15 *
16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
19 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
22 * THE SOFTWARE.
23 */
24
25 #include "block_int.h"
26 #include "qemu-common.h"
27 #include "qcow2.h"
28
29 typedef struct Qcow2CachedTable {
30 void* table;
31 int64_t offset;
32 bool dirty;
33 int cache_hits;
34 int ref;
35 } Qcow2CachedTable;
36
37 struct Qcow2Cache {
38 int size;
39 Qcow2CachedTable* entries;
40 struct Qcow2Cache* depends;
41 bool writethrough;
42 };
43
44 Qcow2Cache *qcow2_cache_create(BlockDriverState *bs, int num_tables,
45 bool writethrough)
46 {
47 BDRVQcowState *s = bs->opaque;
48 Qcow2Cache *c;
49 int i;
50
51 c = qemu_mallocz(sizeof(*c));
52 c->size = num_tables;
53 c->entries = qemu_mallocz(sizeof(*c->entries) * num_tables);
54 c->writethrough = writethrough;
55
56 for (i = 0; i < c->size; i++) {
57 c->entries[i].table = qemu_blockalign(bs, s->cluster_size);
58 }
59
60 return c;
61 }
62
63 int qcow2_cache_destroy(BlockDriverState* bs, Qcow2Cache *c)
64 {
65 int i;
66
67 for (i = 0; i < c->size; i++) {
68 assert(c->entries[i].ref == 0);
69 qemu_vfree(c->entries[i].table);
70 }
71
72 qemu_free(c->entries);
73 qemu_free(c);
74
75 return 0;
76 }
77
78 static int qcow2_cache_flush_dependency(BlockDriverState *bs, Qcow2Cache *c)
79 {
80 int ret;
81
82 ret = qcow2_cache_flush(bs, c->depends);
83 if (ret < 0) {
84 return ret;
85 }
86
87 c->depends = NULL;
88 return 0;
89 }
90
91 static int qcow2_cache_entry_flush(BlockDriverState *bs, Qcow2Cache *c, int i)
92 {
93 BDRVQcowState *s = bs->opaque;
94 int ret;
95
96 if (!c->entries[i].dirty || !c->entries[i].offset) {
97 return 0;
98 }
99
100 if (c->depends) {
101 ret = qcow2_cache_flush_dependency(bs, c);
102 if (ret < 0) {
103 return ret;
104 }
105 }
106
107 ret = bdrv_pwrite(bs->file, c->entries[i].offset, c->entries[i].table,
108 s->cluster_size);
109 if (ret < 0) {
110 return ret;
111 }
112
113 c->entries[i].dirty = false;
114
115 return 0;
116 }
117
118 int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c)
119 {
120 int result = 0;
121 int ret;
122 int i;
123
124 for (i = 0; i < c->size; i++) {
125 ret = qcow2_cache_entry_flush(bs, c, i);
126 if (ret < 0 && result != -ENOSPC) {
127 result = ret;
128 }
129 }
130
131 if (result == 0) {
132 ret = bdrv_flush(bs->file);
133 if (ret < 0) {
134 result = ret;
135 }
136 }
137
138 return result;
139 }
140
141 int qcow2_cache_set_dependency(BlockDriverState *bs, Qcow2Cache *c,
142 Qcow2Cache *dependency)
143 {
144 int ret;
145
146 if (dependency->depends) {
147 ret = qcow2_cache_flush_dependency(bs, dependency);
148 if (ret < 0) {
149 return ret;
150 }
151 }
152
153 if (c->depends && (c->depends != dependency)) {
154 ret = qcow2_cache_flush_dependency(bs, c);
155 if (ret < 0) {
156 return ret;
157 }
158 }
159
160 c->depends = dependency;
161 return 0;
162 }
163
164 static int qcow2_cache_find_entry_to_replace(Qcow2Cache *c)
165 {
166 int i;
167 int min_count = INT_MAX;
168 int min_index = -1;
169
170
171 for (i = 0; i < c->size; i++) {
172 if (c->entries[i].ref) {
173 continue;
174 }
175
176 if (c->entries[i].cache_hits < min_count) {
177 min_index = i;
178 min_count = c->entries[i].cache_hits;
179 }
180
181 /* Give newer hits priority */
182 /* TODO Check how to optimize the replacement strategy */
183 c->entries[i].cache_hits /= 2;
184 }
185
186 if (min_index == -1) {
187 /* This can't happen in current synchronous code, but leave the check
188 * here as a reminder for whoever starts using AIO with the cache */
189 abort();
190 }
191 return min_index;
192 }
193
194 static int qcow2_cache_do_get(BlockDriverState *bs, Qcow2Cache *c,
195 uint64_t offset, void **table, bool read_from_disk)
196 {
197 BDRVQcowState *s = bs->opaque;
198 int i;
199 int ret;
200
201 /* Check if the table is already cached */
202 for (i = 0; i < c->size; i++) {
203 if (c->entries[i].offset == offset) {
204 goto found;
205 }
206 }
207
208 /* If not, write a table back and replace it */
209 i = qcow2_cache_find_entry_to_replace(c);
210 if (i < 0) {
211 return i;
212 }
213
214 ret = qcow2_cache_entry_flush(bs, c, i);
215 if (ret < 0) {
216 return ret;
217 }
218
219 c->entries[i].offset = 0;
220 if (read_from_disk) {
221 ret = bdrv_pread(bs->file, offset, c->entries[i].table, s->cluster_size);
222 if (ret < 0) {
223 return ret;
224 }
225 }
226
227 /* Give the table some hits for the start so that it won't be replaced
228 * immediately. The number 32 is completely arbitrary. */
229 c->entries[i].cache_hits = 32;
230 c->entries[i].offset = offset;
231
232 /* And return the right table */
233 found:
234 c->entries[i].cache_hits++;
235 c->entries[i].ref++;
236 *table = c->entries[i].table;
237 return 0;
238 }
239
240 int qcow2_cache_get(BlockDriverState *bs, Qcow2Cache *c, uint64_t offset,
241 void **table)
242 {
243 return qcow2_cache_do_get(bs, c, offset, table, true);
244 }
245
246 int qcow2_cache_get_empty(BlockDriverState *bs, Qcow2Cache *c, uint64_t offset,
247 void **table)
248 {
249 return qcow2_cache_do_get(bs, c, offset, table, false);
250 }
251
252 int qcow2_cache_put(BlockDriverState *bs, Qcow2Cache *c, void **table)
253 {
254 int i;
255
256 for (i = 0; i < c->size; i++) {
257 if (c->entries[i].table == *table) {
258 goto found;
259 }
260 }
261 return -ENOENT;
262
263 found:
264 c->entries[i].ref--;
265 *table = NULL;
266
267 assert(c->entries[i].ref >= 0);
268
269 if (c->writethrough) {
270 return qcow2_cache_entry_flush(bs, c, i);
271 } else {
272 return 0;
273 }
274 }
275
276 void qcow2_cache_entry_mark_dirty(Qcow2Cache *c, void *table)
277 {
278 int i;
279
280 for (i = 0; i < c->size; i++) {
281 if (c->entries[i].table == table) {
282 goto found;
283 }
284 }
285 abort();
286
287 found:
288 c->entries[i].dirty = true;
289 }
290