sync.c 9.91 KB
Newer Older
1 2 3 4 5 6 7
/*
 * High-level sync()-related operations
 */

#include <linux/kernel.h>
#include <linux/file.h>
#include <linux/fs.h>
8
#include <linux/slab.h>
9
#include <linux/export.h>
10
#include <linux/namei.h>
11
#include <linux/sched.h>
12 13 14 15
#include <linux/writeback.h>
#include <linux/syscalls.h>
#include <linux/linkage.h>
#include <linux/pagemap.h>
16
#include <linux/quotaops.h>
17
#include <linux/backing-dev.h>
18
#include "internal.h"
19 20 21 22

#define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \
			SYNC_FILE_RANGE_WAIT_AFTER)

23
/*
24 25 26 27 28
 * 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.
29
 */
30
static int __sync_filesystem(struct super_block *sb, int wait)
31
{
32
	if (wait)
33
		sync_inodes_sb(sb);
34
	else
35
		writeback_inodes_sb(sb, WB_REASON_SYNC);
36

37 38 39 40 41 42 43 44 45 46
	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.
 */
47
int sync_filesystem(struct super_block *sb)
48 49 50
{
	int ret;

51 52 53 54 55 56 57 58 59 60 61 62
	/*
	 * 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;

63
	ret = __sync_filesystem(sb, 0);
64 65
	if (ret < 0)
		return ret;
66
	return __sync_filesystem(sb, 1);
67
}
68
EXPORT_SYMBOL(sync_filesystem);
69

70
static void sync_inodes_one_sb(struct super_block *sb, void *arg)
Al Viro's avatar
Al Viro committed
71
{
72
	if (!(sb->s_flags & MS_RDONLY))
73
		sync_inodes_sb(sb);
Al Viro's avatar
Al Viro committed
74
}
75 76 77 78 79 80 81

static void sync_fs_one_sb(struct super_block *sb, void *arg)
{
	if (!(sb->s_flags & MS_RDONLY) && sb->s_op->sync_fs)
		sb->s_op->sync_fs(sb, *(int *)arg);
}

82
static void fdatawrite_one_bdev(struct block_device *bdev, void *arg)
83
{
84
	filemap_fdatawrite(bdev->bd_inode->i_mapping);
85 86
}

87
static void fdatawait_one_bdev(struct block_device *bdev, void *arg)
88
{
89 90 91 92 93 94
	/*
	 * We keep the error status of individual mapping so that
	 * applications can catch the writeback error using fsync(2).
	 * See filemap_fdatawait_keep_errors() for details.
	 */
	filemap_fdatawait_keep_errors(bdev->bd_inode->i_mapping);
95 96
}

97
/*
98 99 100 101 102 103 104 105
 * Sync everything. We start by waking flusher threads so that most of
 * writeback runs on all devices in parallel. Then we sync all inodes reliably
 * which effectively also waits for all flusher threads to finish doing
 * writeback. At this point all data is on disk so metadata should be stable
 * and we tell filesystems to sync their metadata via ->sync_fs() calls.
 * Finally, we writeout all block devices because some filesystems (e.g. ext2)
 * just write metadata (such as inodes or bitmaps) to block device page cache
 * and do not sync it on their own in ->sync_fs().
106
 */
107
SYSCALL_DEFINE0(sync)
108
{
109 110
	int nowait = 0, wait = 1;

111
	wakeup_flusher_threads(0, WB_REASON_SYNC);
112
	iterate_supers(sync_inodes_one_sb, NULL);
113
	iterate_supers(sync_fs_one_sb, &nowait);
114
	iterate_supers(sync_fs_one_sb, &wait);
115 116
	iterate_bdevs(fdatawrite_one_bdev, NULL);
	iterate_bdevs(fdatawait_one_bdev, NULL);
117 118 119 120 121
	if (unlikely(laptop_mode))
		laptop_sync_completion();
	return 0;
}

122 123
static void do_sync_work(struct work_struct *work)
{
124 125
	int nowait = 0;

126 127 128 129
	/*
	 * Sync twice to reduce the possibility we skipped some inodes / pages
	 * because they were temporarily locked
	 */
130 131
	iterate_supers(sync_inodes_one_sb, &nowait);
	iterate_supers(sync_fs_one_sb, &nowait);
132
	iterate_bdevs(fdatawrite_one_bdev, NULL);
133 134
	iterate_supers(sync_inodes_one_sb, &nowait);
	iterate_supers(sync_fs_one_sb, &nowait);
135
	iterate_bdevs(fdatawrite_one_bdev, NULL);
136
	printk("Emergency Sync complete\n");
137 138 139
	kfree(work);
}

140 141
void emergency_sync(void)
{
142 143 144 145 146 147 148
	struct work_struct *work;

	work = kmalloc(sizeof(*work), GFP_ATOMIC);
	if (work) {
		INIT_WORK(work, do_sync_work);
		schedule_work(work);
	}
149 150
}

151 152 153 154 155
/*
 * sync a single super
 */
SYSCALL_DEFINE1(syncfs, int, fd)
{
156
	struct fd f = fdget(fd);
157 158 159
	struct super_block *sb;
	int ret;

160
	if (!f.file)
161
		return -EBADF;
Al Viro's avatar
Al Viro committed
162
	sb = f.file->f_path.dentry->d_sb;
163 164 165 166 167

	down_read(&sb->s_umount);
	ret = sync_filesystem(sb);
	up_read(&sb->s_umount);

168
	fdput(f);
169 170 171
	return ret;
}

172
/**
173
 * vfs_fsync_range - helper to sync a range of data & metadata to disk
174
 * @file:		file to sync
175 176 177
 * @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
178
 *
179 180 181
 * 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.
182
 */
183
int vfs_fsync_range(struct file *file, loff_t start, loff_t end, int datasync)
184
{
185 186
	struct inode *inode = file->f_mapping->host;

Al Viro's avatar
Al Viro committed
187
	if (!file->f_op->fsync)
188
		return -EINVAL;
189 190 191 192 193 194
	if (!datasync && (inode->i_state & I_DIRTY_TIME)) {
		spin_lock(&inode->i_lock);
		inode->i_state &= ~I_DIRTY_TIME;
		spin_unlock(&inode->i_lock);
		mark_inode_dirty_sync(inode);
	}
195
	return call_fsync(file, start, end, datasync);
196
}
197 198 199 200 201 202 203 204 205 206
EXPORT_SYMBOL(vfs_fsync_range);

/**
 * vfs_fsync - perform a fsync or fdatasync on a file
 * @file:		file to sync
 * @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.
 */
207
int vfs_fsync(struct file *file, int datasync)
208
{
209
	return vfs_fsync_range(file, 0, LLONG_MAX, datasync);
210
}
211
EXPORT_SYMBOL(vfs_fsync);
212

213
static int do_fsync(unsigned int fd, int datasync)
214
{
215
	struct fd f = fdget(fd);
216 217
	int ret = -EBADF;

218 219 220
	if (f.file) {
		ret = vfs_fsync(f.file, datasync);
		fdput(f);
221 222 223 224
	}
	return ret;
}

225
SYSCALL_DEFINE1(fsync, unsigned int, fd)
226
{
227
	return do_fsync(fd, 0);
228 229
}

230
SYSCALL_DEFINE1(fdatasync, unsigned int, fd)
231
{
232
	return do_fsync(fd, 1);
233 234
}

235 236 237 238 239 240 241 242 243 244 245
/*
 * 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
246 247
 * range which are not presently under writeback. Note that this may block for
 * significant periods due to exhaustion of disk request structures.
248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
 *
 * 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.
 */
282 283
SYSCALL_DEFINE4(sync_file_range, int, fd, loff_t, offset, loff_t, nbytes,
				unsigned int, flags)
284 285
{
	int ret;
286
	struct fd f;
287
	struct address_space *mapping;
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304
	loff_t endbyte;			/* inclusive */
	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) {
305
		if (offset >= (0x100000000ULL << PAGE_SHIFT)) {
306 307 308 309 310 311 312
			/*
			 * The range starts outside a 32 bit machine's
			 * pagecache addressing capabilities.  Let it "succeed"
			 */
			ret = 0;
			goto out;
		}
313
		if (endbyte >= (0x100000000ULL << PAGE_SHIFT)) {
314 315 316 317 318 319 320 321
			/*
			 * Out to EOF
			 */
			nbytes = 0;
		}
	}

	if (nbytes == 0)
322
		endbyte = LLONG_MAX;
323 324 325 326
	else
		endbyte--;		/* inclusive */

	ret = -EBADF;
327 328
	f = fdget(fd);
	if (!f.file)
329 330
		goto out;

Al Viro's avatar
Al Viro committed
331
	i_mode = file_inode(f.file)->i_mode;
332 333 334 335 336
	ret = -ESPIPE;
	if (!S_ISREG(i_mode) && !S_ISBLK(i_mode) && !S_ISDIR(i_mode) &&
			!S_ISLNK(i_mode))
		goto out_put;

337
	mapping = f.file->f_mapping;
338 339 340 341 342 343 344 345 346 347 348 349 350
	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) {
351 352
		ret = __filemap_fdatawrite_range(mapping, offset, endbyte,
						 WB_SYNC_NONE);
353 354 355 356 357 358 359
		if (ret < 0)
			goto out_put;
	}

	if (flags & SYNC_FILE_RANGE_WAIT_AFTER)
		ret = filemap_fdatawait_range(mapping, offset, endbyte);

360
out_put:
361
	fdput(f);
362 363 364 365
out:
	return ret;
}

366 367
/* It would be nice if people remember that not all the world's an i386
   when they introduce new system calls */
368 369
SYSCALL_DEFINE4(sync_file_range2, int, fd, unsigned int, flags,
				 loff_t, offset, loff_t, nbytes)
370 371 372
{
	return sys_sync_file_range(fd, offset, nbytes, flags);
}