Blame view

fs/read_write.c 17.9 KB
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
1
2
3
4
5
6
7
8
9
10
11
12
  /*
   *  linux/fs/read_write.c
   *
   *  Copyright (C) 1991, 1992  Linus Torvalds
   */
  
  #include <linux/slab.h> 
  #include <linux/stat.h>
  #include <linux/fcntl.h>
  #include <linux/file.h>
  #include <linux/uio.h>
  #include <linux/smp_lock.h>
0eeca2830   Robert Love   [PATCH] inotify
13
  #include <linux/fsnotify.h>
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
14
15
16
  #include <linux/security.h>
  #include <linux/module.h>
  #include <linux/syscalls.h>
e28cc7157   Linus Torvalds   Relax the rw_veri...
17
  #include <linux/pagemap.h>
d6b29d7ce   Jens Axboe   splice: divorce t...
18
  #include <linux/splice.h>
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
19
  #include "read_write.h"
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
20
21
22
  
  #include <asm/uaccess.h>
  #include <asm/unistd.h>
4b6f5d20b   Arjan van de Ven   [PATCH] Make most...
23
  const struct file_operations generic_ro_fops = {
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
24
  	.llseek		= generic_file_llseek,
543ade1fc   Badari Pulavarty   [PATCH] Streamlin...
25
26
  	.read		= do_sync_read,
  	.aio_read	= generic_file_aio_read,
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
27
  	.mmap		= generic_file_readonly_mmap,
534f2aaa6   Jens Axboe   sys_sendfile: swi...
28
  	.splice_read	= generic_file_splice_read,
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
29
30
31
32
33
34
35
36
  };
  
  EXPORT_SYMBOL(generic_ro_fops);
  
  loff_t generic_file_llseek(struct file *file, loff_t offset, int origin)
  {
  	long long retval;
  	struct inode *inode = file->f_mapping->host;
1b1dcc1b5   Jes Sorensen   [PATCH] mutex sub...
37
  	mutex_lock(&inode->i_mutex);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
38
  	switch (origin) {
7b8e89249   Chris Snook   use symbolic cons...
39
  		case SEEK_END:
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
40
41
  			offset += inode->i_size;
  			break;
7b8e89249   Chris Snook   use symbolic cons...
42
  		case SEEK_CUR:
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
43
44
45
46
47
48
49
50
51
52
  			offset += file->f_pos;
  	}
  	retval = -EINVAL;
  	if (offset>=0 && offset<=inode->i_sb->s_maxbytes) {
  		if (offset != file->f_pos) {
  			file->f_pos = offset;
  			file->f_version = 0;
  		}
  		retval = offset;
  	}
1b1dcc1b5   Jes Sorensen   [PATCH] mutex sub...
53
  	mutex_unlock(&inode->i_mutex);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
54
55
56
57
58
59
60
61
62
63
64
  	return retval;
  }
  
  EXPORT_SYMBOL(generic_file_llseek);
  
  loff_t remote_llseek(struct file *file, loff_t offset, int origin)
  {
  	long long retval;
  
  	lock_kernel();
  	switch (origin) {
7b8e89249   Chris Snook   use symbolic cons...
65
  		case SEEK_END:
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
66
  			offset += i_size_read(file->f_path.dentry->d_inode);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
67
  			break;
7b8e89249   Chris Snook   use symbolic cons...
68
  		case SEEK_CUR:
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
69
70
71
  			offset += file->f_pos;
  	}
  	retval = -EINVAL;
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
72
  	if (offset>=0 && offset<=file->f_path.dentry->d_inode->i_sb->s_maxbytes) {
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
  		if (offset != file->f_pos) {
  			file->f_pos = offset;
  			file->f_version = 0;
  		}
  		retval = offset;
  	}
  	unlock_kernel();
  	return retval;
  }
  EXPORT_SYMBOL(remote_llseek);
  
  loff_t no_llseek(struct file *file, loff_t offset, int origin)
  {
  	return -ESPIPE;
  }
  EXPORT_SYMBOL(no_llseek);
  
  loff_t default_llseek(struct file *file, loff_t offset, int origin)
  {
  	long long retval;
  
  	lock_kernel();
  	switch (origin) {
7b8e89249   Chris Snook   use symbolic cons...
96
  		case SEEK_END:
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
97
  			offset += i_size_read(file->f_path.dentry->d_inode);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
98
  			break;
7b8e89249   Chris Snook   use symbolic cons...
99
  		case SEEK_CUR:
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
  			offset += file->f_pos;
  	}
  	retval = -EINVAL;
  	if (offset >= 0) {
  		if (offset != file->f_pos) {
  			file->f_pos = offset;
  			file->f_version = 0;
  		}
  		retval = offset;
  	}
  	unlock_kernel();
  	return retval;
  }
  EXPORT_SYMBOL(default_llseek);
  
  loff_t vfs_llseek(struct file *file, loff_t offset, int origin)
  {
  	loff_t (*fn)(struct file *, loff_t, int);
  
  	fn = no_llseek;
  	if (file->f_mode & FMODE_LSEEK) {
  		fn = default_llseek;
  		if (file->f_op && file->f_op->llseek)
  			fn = file->f_op->llseek;
  	}
  	return fn(file, offset, origin);
  }
  EXPORT_SYMBOL(vfs_llseek);
  
  asmlinkage off_t sys_lseek(unsigned int fd, off_t offset, unsigned int origin)
  {
  	off_t retval;
  	struct file * file;
  	int fput_needed;
  
  	retval = -EBADF;
  	file = fget_light(fd, &fput_needed);
  	if (!file)
  		goto bad;
  
  	retval = -EINVAL;
1ae7075bc   Chris Snook   use use SEEK_MAX ...
141
  	if (origin <= SEEK_MAX) {
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
  		loff_t res = vfs_llseek(file, offset, origin);
  		retval = res;
  		if (res != (loff_t)retval)
  			retval = -EOVERFLOW;	/* LFS: should only happen on 32 bit platforms */
  	}
  	fput_light(file, fput_needed);
  bad:
  	return retval;
  }
  
  #ifdef __ARCH_WANT_SYS_LLSEEK
  asmlinkage long sys_llseek(unsigned int fd, unsigned long offset_high,
  			   unsigned long offset_low, loff_t __user * result,
  			   unsigned int origin)
  {
  	int retval;
  	struct file * file;
  	loff_t offset;
  	int fput_needed;
  
  	retval = -EBADF;
  	file = fget_light(fd, &fput_needed);
  	if (!file)
  		goto bad;
  
  	retval = -EINVAL;
1ae7075bc   Chris Snook   use use SEEK_MAX ...
168
  	if (origin > SEEK_MAX)
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
  		goto out_putf;
  
  	offset = vfs_llseek(file, ((loff_t) offset_high << 32) | offset_low,
  			origin);
  
  	retval = (int)offset;
  	if (offset >= 0) {
  		retval = -EFAULT;
  		if (!copy_to_user(result, &offset, sizeof(offset)))
  			retval = 0;
  	}
  out_putf:
  	fput_light(file, fput_needed);
  bad:
  	return retval;
  }
  #endif
e28cc7157   Linus Torvalds   Relax the rw_veri...
186
187
188
189
190
191
  /*
   * rw_verify_area doesn't like huge counts. We limit
   * them to something that fits in "int" so that others
   * won't have to do range checks all the time.
   */
  #define MAX_RW_COUNT (INT_MAX & PAGE_CACHE_MASK)
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
192
193
194
195
196
  
  int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count)
  {
  	struct inode *inode;
  	loff_t pos;
163da958b   Eric Dumazet   [PATCH] FS: speed...
197
  	inode = file->f_path.dentry->d_inode;
e28cc7157   Linus Torvalds   Relax the rw_veri...
198
  	if (unlikely((ssize_t) count < 0))
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
199
200
201
202
  		goto Einval;
  	pos = *ppos;
  	if (unlikely((pos < 0) || (loff_t) (pos + count) < 0))
  		goto Einval;
a16877ca9   Pavel Emelyanov   Cleanup macros fo...
203
  	if (unlikely(inode->i_flock && mandatory_lock(inode))) {
e28cc7157   Linus Torvalds   Relax the rw_veri...
204
205
206
207
208
209
210
  		int retval = locks_mandatory_area(
  			read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE,
  			inode, file, pos, count);
  		if (retval < 0)
  			return retval;
  	}
  	return count > MAX_RW_COUNT ? MAX_RW_COUNT : count;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
211
212
213
214
  
  Einval:
  	return -EINVAL;
  }
63e688091   Benjamin LaHaise   [PATCH] aio: fix ...
215
216
217
218
219
220
221
222
223
  static void wait_on_retry_sync_kiocb(struct kiocb *iocb)
  {
  	set_current_state(TASK_UNINTERRUPTIBLE);
  	if (!kiocbIsKicked(iocb))
  		schedule();
  	else
  		kiocbClearKicked(iocb);
  	__set_current_state(TASK_RUNNING);
  }
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
224
225
  ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
  {
027445c37   Badari Pulavarty   [PATCH] Vectorize...
226
  	struct iovec iov = { .iov_base = buf, .iov_len = len };
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
227
228
229
230
231
  	struct kiocb kiocb;
  	ssize_t ret;
  
  	init_sync_kiocb(&kiocb, filp);
  	kiocb.ki_pos = *ppos;
027445c37   Badari Pulavarty   [PATCH] Vectorize...
232
233
234
235
236
237
  	kiocb.ki_left = len;
  
  	for (;;) {
  		ret = filp->f_op->aio_read(&kiocb, &iov, 1, kiocb.ki_pos);
  		if (ret != -EIOCBRETRY)
  			break;
63e688091   Benjamin LaHaise   [PATCH] aio: fix ...
238
  		wait_on_retry_sync_kiocb(&kiocb);
027445c37   Badari Pulavarty   [PATCH] Vectorize...
239
  	}
63e688091   Benjamin LaHaise   [PATCH] aio: fix ...
240

1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
  	if (-EIOCBQUEUED == ret)
  		ret = wait_on_sync_kiocb(&kiocb);
  	*ppos = kiocb.ki_pos;
  	return ret;
  }
  
  EXPORT_SYMBOL(do_sync_read);
  
  ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
  {
  	ssize_t ret;
  
  	if (!(file->f_mode & FMODE_READ))
  		return -EBADF;
  	if (!file->f_op || (!file->f_op->read && !file->f_op->aio_read))
  		return -EINVAL;
  	if (unlikely(!access_ok(VERIFY_WRITE, buf, count)))
  		return -EFAULT;
  
  	ret = rw_verify_area(READ, file, pos, count);
e28cc7157   Linus Torvalds   Relax the rw_veri...
261
262
  	if (ret >= 0) {
  		count = ret;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
263
264
265
266
267
268
269
  		ret = security_file_permission (file, MAY_READ);
  		if (!ret) {
  			if (file->f_op->read)
  				ret = file->f_op->read(file, buf, count, pos);
  			else
  				ret = do_sync_read(file, buf, count, pos);
  			if (ret > 0) {
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
270
  				fsnotify_access(file->f_path.dentry);
4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
271
  				add_rchar(current, ret);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
272
  			}
4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
273
  			inc_syscr(current);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
274
275
276
277
278
279
280
281
282
283
  		}
  	}
  
  	return ret;
  }
  
  EXPORT_SYMBOL(vfs_read);
  
  ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
  {
027445c37   Badari Pulavarty   [PATCH] Vectorize...
284
  	struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = len };
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
285
286
287
288
289
  	struct kiocb kiocb;
  	ssize_t ret;
  
  	init_sync_kiocb(&kiocb, filp);
  	kiocb.ki_pos = *ppos;
027445c37   Badari Pulavarty   [PATCH] Vectorize...
290
291
292
293
294
295
  	kiocb.ki_left = len;
  
  	for (;;) {
  		ret = filp->f_op->aio_write(&kiocb, &iov, 1, kiocb.ki_pos);
  		if (ret != -EIOCBRETRY)
  			break;
63e688091   Benjamin LaHaise   [PATCH] aio: fix ...
296
  		wait_on_retry_sync_kiocb(&kiocb);
027445c37   Badari Pulavarty   [PATCH] Vectorize...
297
  	}
63e688091   Benjamin LaHaise   [PATCH] aio: fix ...
298

1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
  	if (-EIOCBQUEUED == ret)
  		ret = wait_on_sync_kiocb(&kiocb);
  	*ppos = kiocb.ki_pos;
  	return ret;
  }
  
  EXPORT_SYMBOL(do_sync_write);
  
  ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_t *pos)
  {
  	ssize_t ret;
  
  	if (!(file->f_mode & FMODE_WRITE))
  		return -EBADF;
  	if (!file->f_op || (!file->f_op->write && !file->f_op->aio_write))
  		return -EINVAL;
  	if (unlikely(!access_ok(VERIFY_READ, buf, count)))
  		return -EFAULT;
  
  	ret = rw_verify_area(WRITE, file, pos, count);
e28cc7157   Linus Torvalds   Relax the rw_veri...
319
320
  	if (ret >= 0) {
  		count = ret;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
321
322
323
324
325
326
327
  		ret = security_file_permission (file, MAY_WRITE);
  		if (!ret) {
  			if (file->f_op->write)
  				ret = file->f_op->write(file, buf, count, pos);
  			else
  				ret = do_sync_write(file, buf, count, pos);
  			if (ret > 0) {
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
328
  				fsnotify_modify(file->f_path.dentry);
4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
329
  				add_wchar(current, ret);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
330
  			}
4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
331
  			inc_syscw(current);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
  		}
  	}
  
  	return ret;
  }
  
  EXPORT_SYMBOL(vfs_write);
  
  static inline loff_t file_pos_read(struct file *file)
  {
  	return file->f_pos;
  }
  
  static inline void file_pos_write(struct file *file, loff_t pos)
  {
  	file->f_pos = pos;
  }
  
  asmlinkage ssize_t sys_read(unsigned int fd, char __user * buf, size_t count)
  {
  	struct file *file;
  	ssize_t ret = -EBADF;
  	int fput_needed;
  
  	file = fget_light(fd, &fput_needed);
  	if (file) {
  		loff_t pos = file_pos_read(file);
  		ret = vfs_read(file, buf, count, &pos);
  		file_pos_write(file, pos);
  		fput_light(file, fput_needed);
  	}
  
  	return ret;
  }
  EXPORT_SYMBOL_GPL(sys_read);
  
  asmlinkage ssize_t sys_write(unsigned int fd, const char __user * buf, size_t count)
  {
  	struct file *file;
  	ssize_t ret = -EBADF;
  	int fput_needed;
  
  	file = fget_light(fd, &fput_needed);
  	if (file) {
  		loff_t pos = file_pos_read(file);
  		ret = vfs_write(file, buf, count, &pos);
  		file_pos_write(file, pos);
  		fput_light(file, fput_needed);
  	}
  
  	return ret;
  }
  
  asmlinkage ssize_t sys_pread64(unsigned int fd, char __user *buf,
  			     size_t count, loff_t pos)
  {
  	struct file *file;
  	ssize_t ret = -EBADF;
  	int fput_needed;
  
  	if (pos < 0)
  		return -EINVAL;
  
  	file = fget_light(fd, &fput_needed);
  	if (file) {
  		ret = -ESPIPE;
  		if (file->f_mode & FMODE_PREAD)
  			ret = vfs_read(file, buf, count, &pos);
  		fput_light(file, fput_needed);
  	}
  
  	return ret;
  }
  
  asmlinkage ssize_t sys_pwrite64(unsigned int fd, const char __user *buf,
  			      size_t count, loff_t pos)
  {
  	struct file *file;
  	ssize_t ret = -EBADF;
  	int fput_needed;
  
  	if (pos < 0)
  		return -EINVAL;
  
  	file = fget_light(fd, &fput_needed);
  	if (file) {
  		ret = -ESPIPE;
  		if (file->f_mode & FMODE_PWRITE)  
  			ret = vfs_write(file, buf, count, &pos);
  		fput_light(file, fput_needed);
  	}
  
  	return ret;
  }
  
  /*
   * Reduce an iovec's length in-place.  Return the resulting number of segments
   */
  unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to)
  {
  	unsigned long seg = 0;
  	size_t len = 0;
  
  	while (seg < nr_segs) {
  		seg++;
  		if (len + iov->iov_len >= to) {
  			iov->iov_len = to - len;
  			break;
  		}
  		len += iov->iov_len;
  		iov++;
  	}
  	return seg;
  }
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
  ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov,
  		unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn)
  {
  	struct kiocb kiocb;
  	ssize_t ret;
  
  	init_sync_kiocb(&kiocb, filp);
  	kiocb.ki_pos = *ppos;
  	kiocb.ki_left = len;
  	kiocb.ki_nbytes = len;
  
  	for (;;) {
  		ret = fn(&kiocb, iov, nr_segs, kiocb.ki_pos);
  		if (ret != -EIOCBRETRY)
  			break;
  		wait_on_retry_sync_kiocb(&kiocb);
  	}
  
  	if (ret == -EIOCBQUEUED)
  		ret = wait_on_sync_kiocb(&kiocb);
  	*ppos = kiocb.ki_pos;
  	return ret;
  }
  
  /* Do it by hand, with file-ops */
  ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov,
  		unsigned long nr_segs, loff_t *ppos, io_fn_t fn)
  {
  	struct iovec *vector = iov;
  	ssize_t ret = 0;
  
  	while (nr_segs > 0) {
  		void __user *base;
  		size_t len;
  		ssize_t nr;
  
  		base = vector->iov_base;
  		len = vector->iov_len;
  		vector++;
  		nr_segs--;
  
  		nr = fn(filp, base, len, ppos);
  
  		if (nr < 0) {
  			if (!ret)
  				ret = nr;
  			break;
  		}
  		ret += nr;
  		if (nr != len)
  			break;
  	}
  
  	return ret;
  }
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
501
502
  /* A write operation does a read from user space and vice versa */
  #define vrfy_dir(type) ((type) == READ ? VERIFY_WRITE : VERIFY_READ)
eed4e51fb   Badari Pulavarty   [PATCH] Add vecto...
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
  ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
  			      unsigned long nr_segs, unsigned long fast_segs,
  			      struct iovec *fast_pointer,
  			      struct iovec **ret_pointer)
    {
  	unsigned long seg;
    	ssize_t ret;
  	struct iovec *iov = fast_pointer;
  
    	/*
    	 * SuS says "The readv() function *may* fail if the iovcnt argument
    	 * was less than or equal to 0, or greater than {IOV_MAX}.  Linux has
    	 * traditionally returned zero for zero segments, so...
    	 */
  	if (nr_segs == 0) {
  		ret = 0;
    		goto out;
  	}
  
    	/*
    	 * First get the "struct iovec" from user memory and
    	 * verify all the pointers
    	 */
  	if (nr_segs > UIO_MAXIOV) {
  		ret = -EINVAL;
    		goto out;
  	}
  	if (nr_segs > fast_segs) {
    		iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL);
  		if (iov == NULL) {
  			ret = -ENOMEM;
    			goto out;
  		}
    	}
  	if (copy_from_user(iov, uvector, nr_segs*sizeof(*uvector))) {
  		ret = -EFAULT;
    		goto out;
  	}
  
    	/*
  	 * According to the Single Unix Specification we should return EINVAL
  	 * if an element length is < 0 when cast to ssize_t or if the
  	 * total length would overflow the ssize_t return value of the
  	 * system call.
    	 */
  	ret = 0;
    	for (seg = 0; seg < nr_segs; seg++) {
    		void __user *buf = iov[seg].iov_base;
    		ssize_t len = (ssize_t)iov[seg].iov_len;
  
  		/* see if we we're about to use an invalid len or if
  		 * it's about to overflow ssize_t */
  		if (len < 0 || (ret + len < ret)) {
  			ret = -EINVAL;
    			goto out;
  		}
  		if (unlikely(!access_ok(vrfy_dir(type), buf, len))) {
  			ret = -EFAULT;
    			goto out;
  		}
  
  		ret += len;
    	}
  out:
  	*ret_pointer = iov;
  	return ret;
  }
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
570
571
572
573
  static ssize_t do_readv_writev(int type, struct file *file,
  			       const struct iovec __user * uvector,
  			       unsigned long nr_segs, loff_t *pos)
  {
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
574
575
  	size_t tot_len;
  	struct iovec iovstack[UIO_FASTIOV];
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
576
  	struct iovec *iov = iovstack;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
577
  	ssize_t ret;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
578
579
  	io_fn_t fn;
  	iov_fn_t fnv;
eed4e51fb   Badari Pulavarty   [PATCH] Add vecto...
580
581
  	if (!file->f_op) {
  		ret = -EINVAL;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
582
  		goto out;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
583
  	}
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
584

eed4e51fb   Badari Pulavarty   [PATCH] Add vecto...
585
586
587
  	ret = rw_copy_check_uvector(type, uvector, nr_segs,
  			ARRAY_SIZE(iovstack), iovstack, &iov);
  	if (ret <= 0)
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
588
  		goto out;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
589

eed4e51fb   Badari Pulavarty   [PATCH] Add vecto...
590
  	tot_len = ret;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
591
  	ret = rw_verify_area(type, file, pos, tot_len);
e28cc7157   Linus Torvalds   Relax the rw_veri...
592
  	if (ret < 0)
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
593
  		goto out;
411b67b4b   Kostik Belousov   [PATCH] readv/wri...
594
595
596
  	ret = security_file_permission(file, type == READ ? MAY_READ : MAY_WRITE);
  	if (ret)
  		goto out;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
597
598
599
600
  
  	fnv = NULL;
  	if (type == READ) {
  		fn = file->f_op->read;
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
601
  		fnv = file->f_op->aio_read;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
602
603
  	} else {
  		fn = (io_fn_t)file->f_op->write;
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
604
  		fnv = file->f_op->aio_write;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
605
  	}
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
606
607
608
609
610
  	if (fnv)
  		ret = do_sync_readv_writev(file, iov, nr_segs, tot_len,
  						pos, fnv);
  	else
  		ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
611

1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
612
613
614
  out:
  	if (iov != iovstack)
  		kfree(iov);
0eeca2830   Robert Love   [PATCH] inotify
615
616
  	if ((ret + (type == READ)) > 0) {
  		if (type == READ)
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
617
  			fsnotify_access(file->f_path.dentry);
0eeca2830   Robert Love   [PATCH] inotify
618
  		else
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
619
  			fsnotify_modify(file->f_path.dentry);
0eeca2830   Robert Love   [PATCH] inotify
620
  	}
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
621
  	return ret;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
622
623
624
625
626
627
628
  }
  
  ssize_t vfs_readv(struct file *file, const struct iovec __user *vec,
  		  unsigned long vlen, loff_t *pos)
  {
  	if (!(file->f_mode & FMODE_READ))
  		return -EBADF;
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
629
  	if (!file->f_op || (!file->f_op->aio_read && !file->f_op->read))
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
630
631
632
633
634
635
636
637
638
639
640
641
  		return -EINVAL;
  
  	return do_readv_writev(READ, file, vec, vlen, pos);
  }
  
  EXPORT_SYMBOL(vfs_readv);
  
  ssize_t vfs_writev(struct file *file, const struct iovec __user *vec,
  		   unsigned long vlen, loff_t *pos)
  {
  	if (!(file->f_mode & FMODE_WRITE))
  		return -EBADF;
ee0b3e671   Badari Pulavarty   [PATCH] Remove re...
642
  	if (!file->f_op || (!file->f_op->aio_write && !file->f_op->write))
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
  		return -EINVAL;
  
  	return do_readv_writev(WRITE, file, vec, vlen, pos);
  }
  
  EXPORT_SYMBOL(vfs_writev);
  
  asmlinkage ssize_t
  sys_readv(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
  {
  	struct file *file;
  	ssize_t ret = -EBADF;
  	int fput_needed;
  
  	file = fget_light(fd, &fput_needed);
  	if (file) {
  		loff_t pos = file_pos_read(file);
  		ret = vfs_readv(file, vec, vlen, &pos);
  		file_pos_write(file, pos);
  		fput_light(file, fput_needed);
  	}
  
  	if (ret > 0)
4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
666
667
  		add_rchar(current, ret);
  	inc_syscr(current);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
  	return ret;
  }
  
  asmlinkage ssize_t
  sys_writev(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
  {
  	struct file *file;
  	ssize_t ret = -EBADF;
  	int fput_needed;
  
  	file = fget_light(fd, &fput_needed);
  	if (file) {
  		loff_t pos = file_pos_read(file);
  		ret = vfs_writev(file, vec, vlen, &pos);
  		file_pos_write(file, pos);
  		fput_light(file, fput_needed);
  	}
  
  	if (ret > 0)
4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
687
688
  		add_wchar(current, ret);
  	inc_syscw(current);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
689
690
691
692
693
694
695
696
697
698
  	return ret;
  }
  
  static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
  			   size_t count, loff_t max)
  {
  	struct file * in_file, * out_file;
  	struct inode * in_inode, * out_inode;
  	loff_t pos;
  	ssize_t retval;
534f2aaa6   Jens Axboe   sys_sendfile: swi...
699
  	int fput_needed_in, fput_needed_out, fl;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
700
701
702
703
704
705
706
707
708
709
710
  
  	/*
  	 * Get input file, and verify that it is ok..
  	 */
  	retval = -EBADF;
  	in_file = fget_light(in_fd, &fput_needed_in);
  	if (!in_file)
  		goto out;
  	if (!(in_file->f_mode & FMODE_READ))
  		goto fput_in;
  	retval = -EINVAL;
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
711
  	in_inode = in_file->f_path.dentry->d_inode;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
712
713
  	if (!in_inode)
  		goto fput_in;
d96e6e716   Jens Axboe   Remove remnants o...
714
  	if (!in_file->f_op || !in_file->f_op->splice_read)
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
715
716
717
718
719
720
721
722
  		goto fput_in;
  	retval = -ESPIPE;
  	if (!ppos)
  		ppos = &in_file->f_pos;
  	else
  		if (!(in_file->f_mode & FMODE_PREAD))
  			goto fput_in;
  	retval = rw_verify_area(READ, in_file, ppos, count);
e28cc7157   Linus Torvalds   Relax the rw_veri...
723
  	if (retval < 0)
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
724
  		goto fput_in;
e28cc7157   Linus Torvalds   Relax the rw_veri...
725
  	count = retval;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
  
  	retval = security_file_permission (in_file, MAY_READ);
  	if (retval)
  		goto fput_in;
  
  	/*
  	 * Get output file, and verify that it is ok..
  	 */
  	retval = -EBADF;
  	out_file = fget_light(out_fd, &fput_needed_out);
  	if (!out_file)
  		goto fput_in;
  	if (!(out_file->f_mode & FMODE_WRITE))
  		goto fput_out;
  	retval = -EINVAL;
  	if (!out_file->f_op || !out_file->f_op->sendpage)
  		goto fput_out;
0f7fc9e4d   Josef "Jeff" Sipek   [PATCH] VFS: chan...
743
  	out_inode = out_file->f_path.dentry->d_inode;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
744
  	retval = rw_verify_area(WRITE, out_file, &out_file->f_pos, count);
e28cc7157   Linus Torvalds   Relax the rw_veri...
745
  	if (retval < 0)
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
746
  		goto fput_out;
e28cc7157   Linus Torvalds   Relax the rw_veri...
747
  	count = retval;
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
  
  	retval = security_file_permission (out_file, MAY_WRITE);
  	if (retval)
  		goto fput_out;
  
  	if (!max)
  		max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes);
  
  	pos = *ppos;
  	retval = -EINVAL;
  	if (unlikely(pos < 0))
  		goto fput_out;
  	if (unlikely(pos + count > max)) {
  		retval = -EOVERFLOW;
  		if (pos >= max)
  			goto fput_out;
  		count = max - pos;
  	}
d96e6e716   Jens Axboe   Remove remnants o...
766
  	fl = 0;
534f2aaa6   Jens Axboe   sys_sendfile: swi...
767
  #if 0
d96e6e716   Jens Axboe   Remove remnants o...
768
769
770
771
772
773
774
775
  	/*
  	 * We need to debate whether we can enable this or not. The
  	 * man page documents EAGAIN return for the output at least,
  	 * and the application is arguably buggy if it doesn't expect
  	 * EAGAIN on a non-blocking file descriptor.
  	 */
  	if (in_file->f_flags & O_NONBLOCK)
  		fl = SPLICE_F_NONBLOCK;
534f2aaa6   Jens Axboe   sys_sendfile: swi...
776
  #endif
d96e6e716   Jens Axboe   Remove remnants o...
777
  	retval = do_splice_direct(in_file, ppos, out_file, count, fl);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
778
779
  
  	if (retval > 0) {
4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
780
781
  		add_rchar(current, retval);
  		add_wchar(current, retval);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
782
  	}
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
783

4b98d11b4   Alexey Dobriyan   [PATCH] ifdef ->r...
784
785
  	inc_syscr(current);
  	inc_syscw(current);
1da177e4c   Linus Torvalds   Linux-2.6.12-rc2
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
  	if (*ppos > max)
  		retval = -EOVERFLOW;
  
  fput_out:
  	fput_light(out_file, fput_needed_out);
  fput_in:
  	fput_light(in_file, fput_needed_in);
  out:
  	return retval;
  }
  
  asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd, off_t __user *offset, size_t count)
  {
  	loff_t pos;
  	off_t off;
  	ssize_t ret;
  
  	if (offset) {
  		if (unlikely(get_user(off, offset)))
  			return -EFAULT;
  		pos = off;
  		ret = do_sendfile(out_fd, in_fd, &pos, count, MAX_NON_LFS);
  		if (unlikely(put_user(pos, offset)))
  			return -EFAULT;
  		return ret;
  	}
  
  	return do_sendfile(out_fd, in_fd, NULL, count, 0);
  }
  
  asmlinkage ssize_t sys_sendfile64(int out_fd, int in_fd, loff_t __user *offset, size_t count)
  {
  	loff_t pos;
  	ssize_t ret;
  
  	if (offset) {
  		if (unlikely(copy_from_user(&pos, offset, sizeof(loff_t))))
  			return -EFAULT;
  		ret = do_sendfile(out_fd, in_fd, &pos, count, 0);
  		if (unlikely(put_user(pos, offset)))
  			return -EFAULT;
  		return ret;
  	}
  
  	return do_sendfile(out_fd, in_fd, NULL, count, 0);
  }