sync.c 9.9 KB
Newer Older
A
Andrew Morton 已提交
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>
A
Andrew Morton 已提交
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>
J
Jörn Engel 已提交
17
#include <linux/backing-dev.h>
18
#include "internal.h"
A
Andrew Morton 已提交
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
{
J
Jens Axboe 已提交
32 33 34 35
	/*
	 * This should be safe, as we require bdi backing to actually
	 * write out data in the first place
	 */
36
	if (sb->s_bdi == &noop_backing_dev_info)
J
Jens Axboe 已提交
37 38
		return 0;

39 40 41 42
	if (sb->s_qcop && sb->s_qcop->quota_sync)
		sb->s_qcop->quota_sync(sb, -1, wait);

	if (wait)
43
		sync_inodes_sb(sb);
44
	else
45
		writeback_inodes_sb(sb, WB_REASON_SYNC);
46

47 48 49 50 51 52 53 54 55 56
	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.
 */
57
int sync_filesystem(struct super_block *sb)
58 59 60
{
	int ret;

61 62 63 64 65 66 67 68 69 70 71 72
	/*
	 * 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;

73
	ret = __sync_filesystem(sb, 0);
74 75
	if (ret < 0)
		return ret;
76
	return __sync_filesystem(sb, 1);
77
}
78
EXPORT_SYMBOL_GPL(sync_filesystem);
79

A
Al Viro 已提交
80 81
static void sync_one_sb(struct super_block *sb, void *arg)
{
82
	if (!(sb->s_flags & MS_RDONLY))
A
Al Viro 已提交
83 84
		__sync_filesystem(sb, *(int *)arg);
}
85 86 87 88 89 90
/*
 * Sync all the data for all the filesystems (called by sys_sync() and
 * emergency sync)
 */
static void sync_filesystems(int wait)
{
A
Al Viro 已提交
91
	iterate_supers(sync_one_sb, &wait);
92 93
}

94 95 96 97
/*
 * sync everything.  Start out by waking pdflush, because that writes back
 * all queues in parallel.
 */
98
SYSCALL_DEFINE0(sync)
99
{
100
	wakeup_flusher_threads(0, WB_REASON_SYNC);
101 102
	sync_filesystems(0);
	sync_filesystems(1);
103 104 105 106 107
	if (unlikely(laptop_mode))
		laptop_sync_completion();
	return 0;
}

108 109
static void do_sync_work(struct work_struct *work)
{
110 111 112 113 114 115 116
	/*
	 * 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\n");
117 118 119
	kfree(work);
}

120 121
void emergency_sync(void)
{
122 123 124 125 126 127 128
	struct work_struct *work;

	work = kmalloc(sizeof(*work), GFP_ATOMIC);
	if (work) {
		INIT_WORK(work, do_sync_work);
		schedule_work(work);
	}
129 130
}

131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153
/*
 * sync a single super
 */
SYSCALL_DEFINE1(syncfs, int, fd)
{
	struct file *file;
	struct super_block *sb;
	int ret;
	int fput_needed;

	file = fget_light(fd, &fput_needed);
	if (!file)
		return -EBADF;
	sb = file->f_dentry->d_sb;

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

	fput_light(file, fput_needed);
	return ret;
}

C
Christoph Hellwig 已提交
154
/**
155
 * vfs_fsync_range - helper to sync a range of data & metadata to disk
C
Christoph Hellwig 已提交
156
 * @file:		file to sync
157 158 159
 * @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
C
Christoph Hellwig 已提交
160
 *
161 162 163
 * 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.
C
Christoph Hellwig 已提交
164
 */
165
int vfs_fsync_range(struct file *file, loff_t start, loff_t end, int datasync)
166
{
167 168 169
	if (!file->f_op || !file->f_op->fsync)
		return -EINVAL;
	return file->f_op->fsync(file, start, end, datasync);
170
}
171 172 173 174 175 176 177 178 179 180
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.
 */
181
int vfs_fsync(struct file *file, int datasync)
182
{
183
	return vfs_fsync_range(file, 0, LLONG_MAX, datasync);
184
}
C
Christoph Hellwig 已提交
185
EXPORT_SYMBOL(vfs_fsync);
186

C
Christoph Hellwig 已提交
187
static int do_fsync(unsigned int fd, int datasync)
188 189 190 191 192 193
{
	struct file *file;
	int ret = -EBADF;

	file = fget(fd);
	if (file) {
194
		ret = vfs_fsync(file, datasync);
195 196 197 198 199
		fput(file);
	}
	return ret;
}

200
SYSCALL_DEFINE1(fsync, unsigned int, fd)
201
{
C
Christoph Hellwig 已提交
202
	return do_fsync(fd, 0);
203 204
}

205
SYSCALL_DEFINE1(fdatasync, unsigned int, fd)
206
{
C
Christoph Hellwig 已提交
207
	return do_fsync(fd, 1);
208 209
}

210 211 212 213 214 215 216 217 218 219
/**
 * 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)
{
220
	if (!(file->f_flags & O_DSYNC) && !IS_SYNC(file->f_mapping->host))
221
		return 0;
222
	return vfs_fsync_range(file, pos, pos + count - 1,
223
			       (file->f_flags & __O_SYNC) ? 0 : 1);
224 225 226
}
EXPORT_SYMBOL(generic_write_sync);

A
Andrew Morton 已提交
227 228 229 230 231 232 233 234 235 236 237
/*
 * 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
238 239
 * range which are not presently under writeback. Note that this may block for
 * significant periods due to exhaustion of disk request structures.
A
Andrew Morton 已提交
240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273
 *
 * 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.
 */
274 275
SYSCALL_DEFINE(sync_file_range)(int fd, loff_t offset, loff_t nbytes,
				unsigned int flags)
A
Andrew Morton 已提交
276 277 278
{
	int ret;
	struct file *file;
279
	struct address_space *mapping;
A
Andrew Morton 已提交
280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
	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)
315
		endbyte = LLONG_MAX;
A
Andrew Morton 已提交
316 317 318 319 320 321 322 323
	else
		endbyte--;		/* inclusive */

	ret = -EBADF;
	file = fget_light(fd, &fput_needed);
	if (!file)
		goto out;

324
	i_mode = file->f_path.dentry->d_inode->i_mode;
A
Andrew Morton 已提交
325 326 327 328 329
	ret = -ESPIPE;
	if (!S_ISREG(i_mode) && !S_ISBLK(i_mode) && !S_ISDIR(i_mode) &&
			!S_ISLNK(i_mode))
		goto out_put;

330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351
	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);

A
Andrew Morton 已提交
352 353 354 355 356
out_put:
	fput_light(file, fput_needed);
out:
	return ret;
}
357 358 359 360 361 362 363 364 365
#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
A
Andrew Morton 已提交
366

367 368
/* It would be nice if people remember that not all the world's an i386
   when they introduce new system calls */
369 370
SYSCALL_DEFINE(sync_file_range2)(int fd, unsigned int flags,
				 loff_t offset, loff_t nbytes)
371 372 373
{
	return sys_sync_file_range(fd, offset, nbytes, flags);
}
374 375 376 377 378 379 380 381 382
#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