Blame view
fs/sync.c
9.89 KB
f79e2abb9 [PATCH] sys_sync_... |
1 2 3 4 5 6 7 |
/* * High-level sync()-related operations */ #include <linux/kernel.h> #include <linux/file.h> #include <linux/fs.h> |
5a0e3ad6a include cleanup: ... |
8 |
#include <linux/slab.h> |
f79e2abb9 [PATCH] sys_sync_... |
9 |
#include <linux/module.h> |
914e26379 [PATCH] severing ... |
10 |
#include <linux/sched.h> |
f79e2abb9 [PATCH] sys_sync_... |
11 12 13 14 |
#include <linux/writeback.h> #include <linux/syscalls.h> #include <linux/linkage.h> #include <linux/pagemap.h> |
cf9a2ae8d [PATCH] BLOCK: Mo... |
15 16 |
#include <linux/quotaops.h> #include <linux/buffer_head.h> |
5129a469a Catch filesystems... |
17 |
#include <linux/backing-dev.h> |
5a3e5cb8e vfs: Fix sys_sync... |
18 |
#include "internal.h" |
f79e2abb9 [PATCH] sys_sync_... |
19 20 21 |
#define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \ SYNC_FILE_RANGE_WAIT_AFTER) |
c15c54f5f vfs: Move syncing... |
22 |
/* |
d8a8559cd writeback: get ri... |
23 24 25 26 27 |
* Do the filesystem syncing work. For simple filesystems * writeback_inodes_sb(sb) just dirties buffers with inodes so we have to * submit IO for these buffers via __sync_blockdev(). This also speeds up the * wait == 1 case since in that case write_inode() functions do * sync_dirty_buffer() and thus effectively write one block at a time. |
c15c54f5f vfs: Move syncing... |
28 |
*/ |
60b0680fa vfs: Rename fsync... |
29 |
static int __sync_filesystem(struct super_block *sb, int wait) |
c15c54f5f vfs: Move syncing... |
30 |
{ |
32a88aa1b fs: Assign bdi in... |
31 32 33 34 |
/* * This should be safe, as we require bdi backing to actually * write out data in the first place */ |
5129a469a Catch filesystems... |
35 |
if (!sb->s_bdi || sb->s_bdi == &noop_backing_dev_info) |
32a88aa1b fs: Assign bdi in... |
36 |
return 0; |
5fb324ad2 quota: move code ... |
37 38 39 40 |
if (sb->s_qcop && sb->s_qcop->quota_sync) sb->s_qcop->quota_sync(sb, -1, wait); if (wait) |
d8a8559cd writeback: get ri... |
41 |
sync_inodes_sb(sb); |
5fb324ad2 quota: move code ... |
42 |
else |
0e3c9a228 Revert "writeback... |
43 |
writeback_inodes_sb(sb); |
5fb324ad2 quota: move code ... |
44 |
|
c15c54f5f vfs: Move syncing... |
45 46 47 48 49 50 51 52 53 54 |
if (sb->s_op->sync_fs) sb->s_op->sync_fs(sb, wait); return __sync_blockdev(sb->s_bdev, wait); } /* * Write out and wait upon all dirty data associated with this * superblock. Filesystem data as well as the underlying block * device. Takes the superblock lock. */ |
60b0680fa vfs: Rename fsync... |
55 |
int sync_filesystem(struct super_block *sb) |
c15c54f5f vfs: Move syncing... |
56 57 |
{ int ret; |
5af7926ff enforce ->sync_fs... |
58 59 60 61 62 63 64 65 66 67 68 |
/* * We need to be protected against the filesystem going from * r/o to r/w or vice versa. */ WARN_ON(!rwsem_is_locked(&sb->s_umount)); /* * No point in syncing out anything if the filesystem is read-only. */ if (sb->s_flags & MS_RDONLY) return 0; |
60b0680fa vfs: Rename fsync... |
69 |
ret = __sync_filesystem(sb, 0); |
c15c54f5f vfs: Move syncing... |
70 71 |
if (ret < 0) return ret; |
60b0680fa vfs: Rename fsync... |
72 |
return __sync_filesystem(sb, 1); |
c15c54f5f vfs: Move syncing... |
73 |
} |
60b0680fa vfs: Rename fsync... |
74 |
EXPORT_SYMBOL_GPL(sync_filesystem); |
c15c54f5f vfs: Move syncing... |
75 |
|
01a05b337 new helper: itera... |
76 77 78 79 80 |
static void sync_one_sb(struct super_block *sb, void *arg) { if (!(sb->s_flags & MS_RDONLY) && sb->s_bdi) __sync_filesystem(sb, *(int *)arg); } |
c15c54f5f vfs: Move syncing... |
81 82 83 |
/* * Sync all the data for all the filesystems (called by sys_sync() and * emergency sync) |
c15c54f5f vfs: Move syncing... |
84 85 86 |
*/ static void sync_filesystems(int wait) { |
01a05b337 new helper: itera... |
87 |
iterate_supers(sync_one_sb, &wait); |
c15c54f5f vfs: Move syncing... |
88 |
} |
3beab0b42 sys_sync(): fix 1... |
89 90 91 92 |
/* * sync everything. Start out by waking pdflush, because that writes back * all queues in parallel. */ |
5cee5815d vfs: Make sys_syn... |
93 |
SYSCALL_DEFINE0(sync) |
cf9a2ae8d [PATCH] BLOCK: Mo... |
94 |
{ |
03ba3782e writeback: switch... |
95 |
wakeup_flusher_threads(0); |
5cee5815d vfs: Make sys_syn... |
96 97 |
sync_filesystems(0); sync_filesystems(1); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
98 99 |
if (unlikely(laptop_mode)) laptop_sync_completion(); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
100 101 |
return 0; } |
a2a9537ac Get rid of pdflus... |
102 103 |
static void do_sync_work(struct work_struct *work) { |
5cee5815d vfs: Make sys_syn... |
104 105 106 107 108 109 110 111 |
/* * Sync twice to reduce the possibility we skipped some inodes / pages * because they were temporarily locked */ sync_filesystems(0); sync_filesystems(0); printk("Emergency Sync complete "); |
a2a9537ac Get rid of pdflus... |
112 113 |
kfree(work); } |
cf9a2ae8d [PATCH] BLOCK: Mo... |
114 115 |
void emergency_sync(void) { |
a2a9537ac Get rid of pdflus... |
116 117 118 119 120 121 122 |
struct work_struct *work; work = kmalloc(sizeof(*work), GFP_ATOMIC); if (work) { INIT_WORK(work, do_sync_work); schedule_work(work); } |
cf9a2ae8d [PATCH] BLOCK: Mo... |
123 |
} |
4c728ef58 add a vfs_fsync h... |
124 |
/** |
148f948ba vfs: Introduce ne... |
125 |
* vfs_fsync_range - helper to sync a range of data & metadata to disk |
4c728ef58 add a vfs_fsync h... |
126 |
* @file: file to sync |
148f948ba vfs: Introduce ne... |
127 128 129 |
* @start: offset in bytes of the beginning of data range to sync * @end: offset in bytes of the end of data range (inclusive) * @datasync: perform only datasync |
4c728ef58 add a vfs_fsync h... |
130 |
* |
148f948ba vfs: Introduce ne... |
131 132 133 |
* Write back data in range @start..@end and metadata for @file to disk. If * @datasync is set only metadata needed to access modified file data is * written. |
4c728ef58 add a vfs_fsync h... |
134 |
*/ |
8018ab057 sanitize vfs_fsyn... |
135 |
int vfs_fsync_range(struct file *file, loff_t start, loff_t end, int datasync) |
cf9a2ae8d [PATCH] BLOCK: Mo... |
136 |
{ |
8018ab057 sanitize vfs_fsyn... |
137 |
struct address_space *mapping = file->f_mapping; |
4c728ef58 add a vfs_fsync h... |
138 |
int err, ret; |
8018ab057 sanitize vfs_fsyn... |
139 |
if (!file->f_op || !file->f_op->fsync) { |
cf9a2ae8d [PATCH] BLOCK: Mo... |
140 141 142 |
ret = -EINVAL; goto out; } |
2daea67e9 fsync: wait for d... |
143 |
ret = filemap_write_and_wait_range(mapping, start, end); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
144 145 146 147 148 149 |
/* * We need to protect against concurrent writers, which could cause * livelocks in fsync_buffers_list(). */ mutex_lock(&mapping->host->i_mutex); |
7ea808591 drop unused dentr... |
150 |
err = file->f_op->fsync(file, datasync); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
151 152 153 |
if (!ret) ret = err; mutex_unlock(&mapping->host->i_mutex); |
148f948ba vfs: Introduce ne... |
154 |
|
cf9a2ae8d [PATCH] BLOCK: Mo... |
155 156 157 |
out: return ret; } |
148f948ba vfs: Introduce ne... |
158 159 160 161 162 |
EXPORT_SYMBOL(vfs_fsync_range); /** * vfs_fsync - perform a fsync or fdatasync on a file * @file: file to sync |
148f948ba vfs: Introduce ne... |
163 164 165 166 |
* @datasync: only perform a fdatasync operation * * Write back data and metadata for @file to disk. If @datasync is * set only metadata needed to access modified file data is written. |
148f948ba vfs: Introduce ne... |
167 |
*/ |
8018ab057 sanitize vfs_fsyn... |
168 |
int vfs_fsync(struct file *file, int datasync) |
148f948ba vfs: Introduce ne... |
169 |
{ |
8018ab057 sanitize vfs_fsyn... |
170 |
return vfs_fsync_range(file, 0, LLONG_MAX, datasync); |
148f948ba vfs: Introduce ne... |
171 |
} |
4c728ef58 add a vfs_fsync h... |
172 |
EXPORT_SYMBOL(vfs_fsync); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
173 |
|
4c728ef58 add a vfs_fsync h... |
174 |
static int do_fsync(unsigned int fd, int datasync) |
cf9a2ae8d [PATCH] BLOCK: Mo... |
175 176 177 178 179 180 |
{ struct file *file; int ret = -EBADF; file = fget(fd); if (file) { |
8018ab057 sanitize vfs_fsyn... |
181 |
ret = vfs_fsync(file, datasync); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
182 183 184 185 |
fput(file); } return ret; } |
a5f8fa9e9 [CVE-2009-0029] S... |
186 |
SYSCALL_DEFINE1(fsync, unsigned int, fd) |
cf9a2ae8d [PATCH] BLOCK: Mo... |
187 |
{ |
4c728ef58 add a vfs_fsync h... |
188 |
return do_fsync(fd, 0); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
189 |
} |
a5f8fa9e9 [CVE-2009-0029] S... |
190 |
SYSCALL_DEFINE1(fdatasync, unsigned int, fd) |
cf9a2ae8d [PATCH] BLOCK: Mo... |
191 |
{ |
4c728ef58 add a vfs_fsync h... |
192 |
return do_fsync(fd, 1); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
193 |
} |
148f948ba vfs: Introduce ne... |
194 195 196 197 198 199 200 201 202 203 |
/** * generic_write_sync - perform syncing after a write if file / inode is sync * @file: file to which the write happened * @pos: offset where the write started * @count: length of the write * * This is just a simple wrapper about our general syncing function. */ int generic_write_sync(struct file *file, loff_t pos, loff_t count) { |
6b2f3d1f7 vfs: Implement pr... |
204 |
if (!(file->f_flags & O_DSYNC) && !IS_SYNC(file->f_mapping->host)) |
148f948ba vfs: Introduce ne... |
205 |
return 0; |
8018ab057 sanitize vfs_fsyn... |
206 |
return vfs_fsync_range(file, pos, pos + count - 1, |
6b2f3d1f7 vfs: Implement pr... |
207 |
(file->f_flags & __O_SYNC) ? 0 : 1); |
148f948ba vfs: Introduce ne... |
208 209 |
} EXPORT_SYMBOL(generic_write_sync); |
cf9a2ae8d [PATCH] BLOCK: Mo... |
210 |
/* |
f79e2abb9 [PATCH] sys_sync_... |
211 212 213 214 215 216 217 218 219 220 |
* sys_sync_file_range() permits finely controlled syncing over a segment of * a file in the range offset .. (offset+nbytes-1) inclusive. If nbytes is * zero then sys_sync_file_range() will operate from offset out to EOF. * * The flag bits are: * * SYNC_FILE_RANGE_WAIT_BEFORE: wait upon writeout of all pages in the range * before performing the write. * * SYNC_FILE_RANGE_WRITE: initiate writeout of all those dirty pages in the |
cce770815 SYNC_FILE_RANGE_W... |
221 222 |
* range which are not presently under writeback. Note that this may block for * significant periods due to exhaustion of disk request structures. |
f79e2abb9 [PATCH] sys_sync_... |
223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 |
* * SYNC_FILE_RANGE_WAIT_AFTER: wait upon writeout of all pages in the range * after performing the write. * * Useful combinations of the flag bits are: * * SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE: ensures that all pages * in the range which were dirty on entry to sys_sync_file_range() are placed * under writeout. This is a start-write-for-data-integrity operation. * * SYNC_FILE_RANGE_WRITE: start writeout of all dirty pages in the range which * are not presently under writeout. This is an asynchronous flush-to-disk * operation. Not suitable for data integrity operations. * * SYNC_FILE_RANGE_WAIT_BEFORE (or SYNC_FILE_RANGE_WAIT_AFTER): wait for * completion of writeout of all pages in the range. This will be used after an * earlier SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE operation to wait * for that operation to complete and to return the result. * * SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE|SYNC_FILE_RANGE_WAIT_AFTER: * a traditional sync() operation. This is a write-for-data-integrity operation * which will ensure that all pages in the range which were dirty on entry to * sys_sync_file_range() are committed to disk. * * * SYNC_FILE_RANGE_WAIT_BEFORE and SYNC_FILE_RANGE_WAIT_AFTER will detect any * I/O errors or ENOSPC conditions and will return those to the caller, after * clearing the EIO and ENOSPC flags in the address_space. * * It should be noted that none of these operations write out the file's * metadata. So unless the application is strictly performing overwrites of * already-instantiated disk blocks, there are no guarantees here that the data * will be available after a crash. */ |
6673e0c3f [CVE-2009-0029] S... |
257 258 |
SYSCALL_DEFINE(sync_file_range)(int fd, loff_t offset, loff_t nbytes, unsigned int flags) |
f79e2abb9 [PATCH] sys_sync_... |
259 260 261 |
{ int ret; struct file *file; |
7a0ad10c3 fold do_sync_file... |
262 |
struct address_space *mapping; |
f79e2abb9 [PATCH] sys_sync_... |
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 |
loff_t endbyte; /* inclusive */ int fput_needed; umode_t i_mode; ret = -EINVAL; if (flags & ~VALID_FLAGS) goto out; endbyte = offset + nbytes; if ((s64)offset < 0) goto out; if ((s64)endbyte < 0) goto out; if (endbyte < offset) goto out; if (sizeof(pgoff_t) == 4) { if (offset >= (0x100000000ULL << PAGE_CACHE_SHIFT)) { /* * The range starts outside a 32 bit machine's * pagecache addressing capabilities. Let it "succeed" */ ret = 0; goto out; } if (endbyte >= (0x100000000ULL << PAGE_CACHE_SHIFT)) { /* * Out to EOF */ nbytes = 0; } } if (nbytes == 0) |
111ebb6e6 [PATCH] writeback... |
298 |
endbyte = LLONG_MAX; |
f79e2abb9 [PATCH] sys_sync_... |
299 300 301 302 303 304 305 |
else endbyte--; /* inclusive */ ret = -EBADF; file = fget_light(fd, &fput_needed); if (!file) goto out; |
0f7fc9e4d [PATCH] VFS: chan... |
306 |
i_mode = file->f_path.dentry->d_inode->i_mode; |
f79e2abb9 [PATCH] sys_sync_... |
307 308 309 310 |
ret = -ESPIPE; if (!S_ISREG(i_mode) && !S_ISBLK(i_mode) && !S_ISDIR(i_mode) && !S_ISLNK(i_mode)) goto out_put; |
7a0ad10c3 fold do_sync_file... |
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 |
mapping = file->f_mapping; if (!mapping) { ret = -EINVAL; goto out_put; } ret = 0; if (flags & SYNC_FILE_RANGE_WAIT_BEFORE) { ret = filemap_fdatawait_range(mapping, offset, endbyte); if (ret < 0) goto out_put; } if (flags & SYNC_FILE_RANGE_WRITE) { ret = filemap_fdatawrite_range(mapping, offset, endbyte); if (ret < 0) goto out_put; } if (flags & SYNC_FILE_RANGE_WAIT_AFTER) ret = filemap_fdatawait_range(mapping, offset, endbyte); |
f79e2abb9 [PATCH] sys_sync_... |
332 333 334 335 336 |
out_put: fput_light(file, fput_needed); out: return ret; } |
6673e0c3f [CVE-2009-0029] S... |
337 338 339 340 341 342 343 344 345 |
#ifdef CONFIG_HAVE_SYSCALL_WRAPPERS asmlinkage long SyS_sync_file_range(long fd, loff_t offset, loff_t nbytes, long flags) { return SYSC_sync_file_range((int) fd, offset, nbytes, (unsigned int) flags); } SYSCALL_ALIAS(sys_sync_file_range, SyS_sync_file_range); #endif |
f79e2abb9 [PATCH] sys_sync_... |
346 |
|
edd5cd4a9 Introduce fixed s... |
347 348 |
/* It would be nice if people remember that not all the world's an i386 when they introduce new system calls */ |
6673e0c3f [CVE-2009-0029] S... |
349 350 |
SYSCALL_DEFINE(sync_file_range2)(int fd, unsigned int flags, loff_t offset, loff_t nbytes) |
edd5cd4a9 Introduce fixed s... |
351 352 353 |
{ return sys_sync_file_range(fd, offset, nbytes, flags); } |
6673e0c3f [CVE-2009-0029] S... |
354 355 356 357 358 359 360 361 362 |
#ifdef CONFIG_HAVE_SYSCALL_WRAPPERS asmlinkage long SyS_sync_file_range2(long fd, long flags, loff_t offset, loff_t nbytes) { return SYSC_sync_file_range2((int) fd, (unsigned int) flags, offset, nbytes); } SYSCALL_ALIAS(sys_sync_file_range2, SyS_sync_file_range2); #endif |