Loading...
Note: File does not exist in v6.8.
1/*
2 * fs/cifs/file.c
3 *
4 * vfs operations that deal with files
5 *
6 * Copyright (C) International Business Machines Corp., 2002,2010
7 * Author(s): Steve French (sfrench@us.ibm.com)
8 * Jeremy Allison (jra@samba.org)
9 *
10 * This library is free software; you can redistribute it and/or modify
11 * it under the terms of the GNU Lesser General Public License as published
12 * by the Free Software Foundation; either version 2.1 of the License, or
13 * (at your option) any later version.
14 *
15 * This library is distributed in the hope that it will be useful,
16 * but WITHOUT ANY WARRANTY; without even the implied warranty of
17 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
18 * the GNU Lesser General Public License for more details.
19 *
20 * You should have received a copy of the GNU Lesser General Public License
21 * along with this library; if not, write to the Free Software
22 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
23 */
24#include <linux/fs.h>
25#include <linux/backing-dev.h>
26#include <linux/stat.h>
27#include <linux/fcntl.h>
28#include <linux/pagemap.h>
29#include <linux/pagevec.h>
30#include <linux/writeback.h>
31#include <linux/task_io_accounting_ops.h>
32#include <linux/delay.h>
33#include <linux/mount.h>
34#include <linux/slab.h>
35#include <linux/swap.h>
36#include <asm/div64.h>
37#include "cifsfs.h"
38#include "cifspdu.h"
39#include "cifsglob.h"
40#include "cifsproto.h"
41#include "cifs_unicode.h"
42#include "cifs_debug.h"
43#include "cifs_fs_sb.h"
44#include "fscache.h"
45#include "smbdirect.h"
46
47static inline int cifs_convert_flags(unsigned int flags)
48{
49 if ((flags & O_ACCMODE) == O_RDONLY)
50 return GENERIC_READ;
51 else if ((flags & O_ACCMODE) == O_WRONLY)
52 return GENERIC_WRITE;
53 else if ((flags & O_ACCMODE) == O_RDWR) {
54 /* GENERIC_ALL is too much permission to request
55 can cause unnecessary access denied on create */
56 /* return GENERIC_ALL; */
57 return (GENERIC_READ | GENERIC_WRITE);
58 }
59
60 return (READ_CONTROL | FILE_WRITE_ATTRIBUTES | FILE_READ_ATTRIBUTES |
61 FILE_WRITE_EA | FILE_APPEND_DATA | FILE_WRITE_DATA |
62 FILE_READ_DATA);
63}
64
65static u32 cifs_posix_convert_flags(unsigned int flags)
66{
67 u32 posix_flags = 0;
68
69 if ((flags & O_ACCMODE) == O_RDONLY)
70 posix_flags = SMB_O_RDONLY;
71 else if ((flags & O_ACCMODE) == O_WRONLY)
72 posix_flags = SMB_O_WRONLY;
73 else if ((flags & O_ACCMODE) == O_RDWR)
74 posix_flags = SMB_O_RDWR;
75
76 if (flags & O_CREAT) {
77 posix_flags |= SMB_O_CREAT;
78 if (flags & O_EXCL)
79 posix_flags |= SMB_O_EXCL;
80 } else if (flags & O_EXCL)
81 cifs_dbg(FYI, "Application %s pid %d has incorrectly set O_EXCL flag but not O_CREAT on file open. Ignoring O_EXCL\n",
82 current->comm, current->tgid);
83
84 if (flags & O_TRUNC)
85 posix_flags |= SMB_O_TRUNC;
86 /* be safe and imply O_SYNC for O_DSYNC */
87 if (flags & O_DSYNC)
88 posix_flags |= SMB_O_SYNC;
89 if (flags & O_DIRECTORY)
90 posix_flags |= SMB_O_DIRECTORY;
91 if (flags & O_NOFOLLOW)
92 posix_flags |= SMB_O_NOFOLLOW;
93 if (flags & O_DIRECT)
94 posix_flags |= SMB_O_DIRECT;
95
96 return posix_flags;
97}
98
99static inline int cifs_get_disposition(unsigned int flags)
100{
101 if ((flags & (O_CREAT | O_EXCL)) == (O_CREAT | O_EXCL))
102 return FILE_CREATE;
103 else if ((flags & (O_CREAT | O_TRUNC)) == (O_CREAT | O_TRUNC))
104 return FILE_OVERWRITE_IF;
105 else if ((flags & O_CREAT) == O_CREAT)
106 return FILE_OPEN_IF;
107 else if ((flags & O_TRUNC) == O_TRUNC)
108 return FILE_OVERWRITE;
109 else
110 return FILE_OPEN;
111}
112
113int cifs_posix_open(char *full_path, struct inode **pinode,
114 struct super_block *sb, int mode, unsigned int f_flags,
115 __u32 *poplock, __u16 *pnetfid, unsigned int xid)
116{
117 int rc;
118 FILE_UNIX_BASIC_INFO *presp_data;
119 __u32 posix_flags = 0;
120 struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
121 struct cifs_fattr fattr;
122 struct tcon_link *tlink;
123 struct cifs_tcon *tcon;
124
125 cifs_dbg(FYI, "posix open %s\n", full_path);
126
127 presp_data = kzalloc(sizeof(FILE_UNIX_BASIC_INFO), GFP_KERNEL);
128 if (presp_data == NULL)
129 return -ENOMEM;
130
131 tlink = cifs_sb_tlink(cifs_sb);
132 if (IS_ERR(tlink)) {
133 rc = PTR_ERR(tlink);
134 goto posix_open_ret;
135 }
136
137 tcon = tlink_tcon(tlink);
138 mode &= ~current_umask();
139
140 posix_flags = cifs_posix_convert_flags(f_flags);
141 rc = CIFSPOSIXCreate(xid, tcon, posix_flags, mode, pnetfid, presp_data,
142 poplock, full_path, cifs_sb->local_nls,
143 cifs_remap(cifs_sb));
144 cifs_put_tlink(tlink);
145
146 if (rc)
147 goto posix_open_ret;
148
149 if (presp_data->Type == cpu_to_le32(-1))
150 goto posix_open_ret; /* open ok, caller does qpathinfo */
151
152 if (!pinode)
153 goto posix_open_ret; /* caller does not need info */
154
155 cifs_unix_basic_to_fattr(&fattr, presp_data, cifs_sb);
156
157 /* get new inode and set it up */
158 if (*pinode == NULL) {
159 cifs_fill_uniqueid(sb, &fattr);
160 *pinode = cifs_iget(sb, &fattr);
161 if (!*pinode) {
162 rc = -ENOMEM;
163 goto posix_open_ret;
164 }
165 } else {
166 cifs_fattr_to_inode(*pinode, &fattr);
167 }
168
169posix_open_ret:
170 kfree(presp_data);
171 return rc;
172}
173
174static int
175cifs_nt_open(char *full_path, struct inode *inode, struct cifs_sb_info *cifs_sb,
176 struct cifs_tcon *tcon, unsigned int f_flags, __u32 *oplock,
177 struct cifs_fid *fid, unsigned int xid)
178{
179 int rc;
180 int desired_access;
181 int disposition;
182 int create_options = CREATE_NOT_DIR;
183 FILE_ALL_INFO *buf;
184 struct TCP_Server_Info *server = tcon->ses->server;
185 struct cifs_open_parms oparms;
186
187 if (!server->ops->open)
188 return -ENOSYS;
189
190 desired_access = cifs_convert_flags(f_flags);
191
192/*********************************************************************
193 * open flag mapping table:
194 *
195 * POSIX Flag CIFS Disposition
196 * ---------- ----------------
197 * O_CREAT FILE_OPEN_IF
198 * O_CREAT | O_EXCL FILE_CREATE
199 * O_CREAT | O_TRUNC FILE_OVERWRITE_IF
200 * O_TRUNC FILE_OVERWRITE
201 * none of the above FILE_OPEN
202 *
203 * Note that there is not a direct match between disposition
204 * FILE_SUPERSEDE (ie create whether or not file exists although
205 * O_CREAT | O_TRUNC is similar but truncates the existing
206 * file rather than creating a new file as FILE_SUPERSEDE does
207 * (which uses the attributes / metadata passed in on open call)
208 *?
209 *? O_SYNC is a reasonable match to CIFS writethrough flag
210 *? and the read write flags match reasonably. O_LARGEFILE
211 *? is irrelevant because largefile support is always used
212 *? by this client. Flags O_APPEND, O_DIRECT, O_DIRECTORY,
213 * O_FASYNC, O_NOFOLLOW, O_NONBLOCK need further investigation
214 *********************************************************************/
215
216 disposition = cifs_get_disposition(f_flags);
217
218 /* BB pass O_SYNC flag through on file attributes .. BB */
219
220 buf = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
221 if (!buf)
222 return -ENOMEM;
223
224 if (backup_cred(cifs_sb))
225 create_options |= CREATE_OPEN_BACKUP_INTENT;
226
227 /* O_SYNC also has bit for O_DSYNC so following check picks up either */
228 if (f_flags & O_SYNC)
229 create_options |= CREATE_WRITE_THROUGH;
230
231 if (f_flags & O_DIRECT)
232 create_options |= CREATE_NO_BUFFER;
233
234 oparms.tcon = tcon;
235 oparms.cifs_sb = cifs_sb;
236 oparms.desired_access = desired_access;
237 oparms.create_options = create_options;
238 oparms.disposition = disposition;
239 oparms.path = full_path;
240 oparms.fid = fid;
241 oparms.reconnect = false;
242
243 rc = server->ops->open(xid, &oparms, oplock, buf);
244
245 if (rc)
246 goto out;
247
248 if (tcon->unix_ext)
249 rc = cifs_get_inode_info_unix(&inode, full_path, inode->i_sb,
250 xid);
251 else
252 rc = cifs_get_inode_info(&inode, full_path, buf, inode->i_sb,
253 xid, fid);
254
255out:
256 kfree(buf);
257 return rc;
258}
259
260static bool
261cifs_has_mand_locks(struct cifsInodeInfo *cinode)
262{
263 struct cifs_fid_locks *cur;
264 bool has_locks = false;
265
266 down_read(&cinode->lock_sem);
267 list_for_each_entry(cur, &cinode->llist, llist) {
268 if (!list_empty(&cur->locks)) {
269 has_locks = true;
270 break;
271 }
272 }
273 up_read(&cinode->lock_sem);
274 return has_locks;
275}
276
277struct cifsFileInfo *
278cifs_new_fileinfo(struct cifs_fid *fid, struct file *file,
279 struct tcon_link *tlink, __u32 oplock)
280{
281 struct dentry *dentry = file_dentry(file);
282 struct inode *inode = d_inode(dentry);
283 struct cifsInodeInfo *cinode = CIFS_I(inode);
284 struct cifsFileInfo *cfile;
285 struct cifs_fid_locks *fdlocks;
286 struct cifs_tcon *tcon = tlink_tcon(tlink);
287 struct TCP_Server_Info *server = tcon->ses->server;
288
289 cfile = kzalloc(sizeof(struct cifsFileInfo), GFP_KERNEL);
290 if (cfile == NULL)
291 return cfile;
292
293 fdlocks = kzalloc(sizeof(struct cifs_fid_locks), GFP_KERNEL);
294 if (!fdlocks) {
295 kfree(cfile);
296 return NULL;
297 }
298
299 INIT_LIST_HEAD(&fdlocks->locks);
300 fdlocks->cfile = cfile;
301 cfile->llist = fdlocks;
302 down_write(&cinode->lock_sem);
303 list_add(&fdlocks->llist, &cinode->llist);
304 up_write(&cinode->lock_sem);
305
306 cfile->count = 1;
307 cfile->pid = current->tgid;
308 cfile->uid = current_fsuid();
309 cfile->dentry = dget(dentry);
310 cfile->f_flags = file->f_flags;
311 cfile->invalidHandle = false;
312 cfile->tlink = cifs_get_tlink(tlink);
313 INIT_WORK(&cfile->oplock_break, cifs_oplock_break);
314 mutex_init(&cfile->fh_mutex);
315 spin_lock_init(&cfile->file_info_lock);
316
317 cifs_sb_active(inode->i_sb);
318
319 /*
320 * If the server returned a read oplock and we have mandatory brlocks,
321 * set oplock level to None.
322 */
323 if (server->ops->is_read_op(oplock) && cifs_has_mand_locks(cinode)) {
324 cifs_dbg(FYI, "Reset oplock val from read to None due to mand locks\n");
325 oplock = 0;
326 }
327
328 spin_lock(&tcon->open_file_lock);
329 if (fid->pending_open->oplock != CIFS_OPLOCK_NO_CHANGE && oplock)
330 oplock = fid->pending_open->oplock;
331 list_del(&fid->pending_open->olist);
332
333 fid->purge_cache = false;
334 server->ops->set_fid(cfile, fid, oplock);
335
336 list_add(&cfile->tlist, &tcon->openFileList);
337
338 /* if readable file instance put first in list*/
339 if (file->f_mode & FMODE_READ)
340 list_add(&cfile->flist, &cinode->openFileList);
341 else
342 list_add_tail(&cfile->flist, &cinode->openFileList);
343 spin_unlock(&tcon->open_file_lock);
344
345 if (fid->purge_cache)
346 cifs_zap_mapping(inode);
347
348 file->private_data = cfile;
349 return cfile;
350}
351
352struct cifsFileInfo *
353cifsFileInfo_get(struct cifsFileInfo *cifs_file)
354{
355 spin_lock(&cifs_file->file_info_lock);
356 cifsFileInfo_get_locked(cifs_file);
357 spin_unlock(&cifs_file->file_info_lock);
358 return cifs_file;
359}
360
361/*
362 * Release a reference on the file private data. This may involve closing
363 * the filehandle out on the server. Must be called without holding
364 * tcon->open_file_lock and cifs_file->file_info_lock.
365 */
366void cifsFileInfo_put(struct cifsFileInfo *cifs_file)
367{
368 struct inode *inode = d_inode(cifs_file->dentry);
369 struct cifs_tcon *tcon = tlink_tcon(cifs_file->tlink);
370 struct TCP_Server_Info *server = tcon->ses->server;
371 struct cifsInodeInfo *cifsi = CIFS_I(inode);
372 struct super_block *sb = inode->i_sb;
373 struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
374 struct cifsLockInfo *li, *tmp;
375 struct cifs_fid fid;
376 struct cifs_pending_open open;
377 bool oplock_break_cancelled;
378
379 spin_lock(&tcon->open_file_lock);
380
381 spin_lock(&cifs_file->file_info_lock);
382 if (--cifs_file->count > 0) {
383 spin_unlock(&cifs_file->file_info_lock);
384 spin_unlock(&tcon->open_file_lock);
385 return;
386 }
387 spin_unlock(&cifs_file->file_info_lock);
388
389 if (server->ops->get_lease_key)
390 server->ops->get_lease_key(inode, &fid);
391
392 /* store open in pending opens to make sure we don't miss lease break */
393 cifs_add_pending_open_locked(&fid, cifs_file->tlink, &open);
394
395 /* remove it from the lists */
396 list_del(&cifs_file->flist);
397 list_del(&cifs_file->tlist);
398
399 if (list_empty(&cifsi->openFileList)) {
400 cifs_dbg(FYI, "closing last open instance for inode %p\n",
401 d_inode(cifs_file->dentry));
402 /*
403 * In strict cache mode we need invalidate mapping on the last
404 * close because it may cause a error when we open this file
405 * again and get at least level II oplock.
406 */
407 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO)
408 set_bit(CIFS_INO_INVALID_MAPPING, &cifsi->flags);
409 cifs_set_oplock_level(cifsi, 0);
410 }
411
412 spin_unlock(&tcon->open_file_lock);
413
414 oplock_break_cancelled = cancel_work_sync(&cifs_file->oplock_break);
415
416 if (!tcon->need_reconnect && !cifs_file->invalidHandle) {
417 struct TCP_Server_Info *server = tcon->ses->server;
418 unsigned int xid;
419
420 xid = get_xid();
421 if (server->ops->close)
422 server->ops->close(xid, tcon, &cifs_file->fid);
423 _free_xid(xid);
424 }
425
426 if (oplock_break_cancelled)
427 cifs_done_oplock_break(cifsi);
428
429 cifs_del_pending_open(&open);
430
431 /*
432 * Delete any outstanding lock records. We'll lose them when the file
433 * is closed anyway.
434 */
435 down_write(&cifsi->lock_sem);
436 list_for_each_entry_safe(li, tmp, &cifs_file->llist->locks, llist) {
437 list_del(&li->llist);
438 cifs_del_lock_waiters(li);
439 kfree(li);
440 }
441 list_del(&cifs_file->llist->llist);
442 kfree(cifs_file->llist);
443 up_write(&cifsi->lock_sem);
444
445 cifs_put_tlink(cifs_file->tlink);
446 dput(cifs_file->dentry);
447 cifs_sb_deactive(sb);
448 kfree(cifs_file);
449}
450
451int cifs_open(struct inode *inode, struct file *file)
452
453{
454 int rc = -EACCES;
455 unsigned int xid;
456 __u32 oplock;
457 struct cifs_sb_info *cifs_sb;
458 struct TCP_Server_Info *server;
459 struct cifs_tcon *tcon;
460 struct tcon_link *tlink;
461 struct cifsFileInfo *cfile = NULL;
462 char *full_path = NULL;
463 bool posix_open_ok = false;
464 struct cifs_fid fid;
465 struct cifs_pending_open open;
466
467 xid = get_xid();
468
469 cifs_sb = CIFS_SB(inode->i_sb);
470 tlink = cifs_sb_tlink(cifs_sb);
471 if (IS_ERR(tlink)) {
472 free_xid(xid);
473 return PTR_ERR(tlink);
474 }
475 tcon = tlink_tcon(tlink);
476 server = tcon->ses->server;
477
478 full_path = build_path_from_dentry(file_dentry(file));
479 if (full_path == NULL) {
480 rc = -ENOMEM;
481 goto out;
482 }
483
484 cifs_dbg(FYI, "inode = 0x%p file flags are 0x%x for %s\n",
485 inode, file->f_flags, full_path);
486
487 if (file->f_flags & O_DIRECT &&
488 cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) {
489 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
490 file->f_op = &cifs_file_direct_nobrl_ops;
491 else
492 file->f_op = &cifs_file_direct_ops;
493 }
494
495 if (server->oplocks)
496 oplock = REQ_OPLOCK;
497 else
498 oplock = 0;
499
500 if (!tcon->broken_posix_open && tcon->unix_ext &&
501 cap_unix(tcon->ses) && (CIFS_UNIX_POSIX_PATH_OPS_CAP &
502 le64_to_cpu(tcon->fsUnixInfo.Capability))) {
503 /* can not refresh inode info since size could be stale */
504 rc = cifs_posix_open(full_path, &inode, inode->i_sb,
505 cifs_sb->mnt_file_mode /* ignored */,
506 file->f_flags, &oplock, &fid.netfid, xid);
507 if (rc == 0) {
508 cifs_dbg(FYI, "posix open succeeded\n");
509 posix_open_ok = true;
510 } else if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) {
511 if (tcon->ses->serverNOS)
512 cifs_dbg(VFS, "server %s of type %s returned unexpected error on SMB posix open, disabling posix open support. Check if server update available.\n",
513 tcon->ses->serverName,
514 tcon->ses->serverNOS);
515 tcon->broken_posix_open = true;
516 } else if ((rc != -EIO) && (rc != -EREMOTE) &&
517 (rc != -EOPNOTSUPP)) /* path not found or net err */
518 goto out;
519 /*
520 * Else fallthrough to retry open the old way on network i/o
521 * or DFS errors.
522 */
523 }
524
525 if (server->ops->get_lease_key)
526 server->ops->get_lease_key(inode, &fid);
527
528 cifs_add_pending_open(&fid, tlink, &open);
529
530 if (!posix_open_ok) {
531 if (server->ops->get_lease_key)
532 server->ops->get_lease_key(inode, &fid);
533
534 rc = cifs_nt_open(full_path, inode, cifs_sb, tcon,
535 file->f_flags, &oplock, &fid, xid);
536 if (rc) {
537 cifs_del_pending_open(&open);
538 goto out;
539 }
540 }
541
542 cfile = cifs_new_fileinfo(&fid, file, tlink, oplock);
543 if (cfile == NULL) {
544 if (server->ops->close)
545 server->ops->close(xid, tcon, &fid);
546 cifs_del_pending_open(&open);
547 rc = -ENOMEM;
548 goto out;
549 }
550
551 cifs_fscache_set_inode_cookie(inode, file);
552
553 if ((oplock & CIFS_CREATE_ACTION) && !posix_open_ok && tcon->unix_ext) {
554 /*
555 * Time to set mode which we can not set earlier due to
556 * problems creating new read-only files.
557 */
558 struct cifs_unix_set_info_args args = {
559 .mode = inode->i_mode,
560 .uid = INVALID_UID, /* no change */
561 .gid = INVALID_GID, /* no change */
562 .ctime = NO_CHANGE_64,
563 .atime = NO_CHANGE_64,
564 .mtime = NO_CHANGE_64,
565 .device = 0,
566 };
567 CIFSSMBUnixSetFileInfo(xid, tcon, &args, fid.netfid,
568 cfile->pid);
569 }
570
571out:
572 kfree(full_path);
573 free_xid(xid);
574 cifs_put_tlink(tlink);
575 return rc;
576}
577
578static int cifs_push_posix_locks(struct cifsFileInfo *cfile);
579
580/*
581 * Try to reacquire byte range locks that were released when session
582 * to server was lost.
583 */
584static int
585cifs_relock_file(struct cifsFileInfo *cfile)
586{
587 struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb);
588 struct cifsInodeInfo *cinode = CIFS_I(d_inode(cfile->dentry));
589 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
590 int rc = 0;
591
592 down_read_nested(&cinode->lock_sem, SINGLE_DEPTH_NESTING);
593 if (cinode->can_cache_brlcks) {
594 /* can cache locks - no need to relock */
595 up_read(&cinode->lock_sem);
596 return rc;
597 }
598
599 if (cap_unix(tcon->ses) &&
600 (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) &&
601 ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0))
602 rc = cifs_push_posix_locks(cfile);
603 else
604 rc = tcon->ses->server->ops->push_mand_locks(cfile);
605
606 up_read(&cinode->lock_sem);
607 return rc;
608}
609
610static int
611cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush)
612{
613 int rc = -EACCES;
614 unsigned int xid;
615 __u32 oplock;
616 struct cifs_sb_info *cifs_sb;
617 struct cifs_tcon *tcon;
618 struct TCP_Server_Info *server;
619 struct cifsInodeInfo *cinode;
620 struct inode *inode;
621 char *full_path = NULL;
622 int desired_access;
623 int disposition = FILE_OPEN;
624 int create_options = CREATE_NOT_DIR;
625 struct cifs_open_parms oparms;
626
627 xid = get_xid();
628 mutex_lock(&cfile->fh_mutex);
629 if (!cfile->invalidHandle) {
630 mutex_unlock(&cfile->fh_mutex);
631 rc = 0;
632 free_xid(xid);
633 return rc;
634 }
635
636 inode = d_inode(cfile->dentry);
637 cifs_sb = CIFS_SB(inode->i_sb);
638 tcon = tlink_tcon(cfile->tlink);
639 server = tcon->ses->server;
640
641 /*
642 * Can not grab rename sem here because various ops, including those
643 * that already have the rename sem can end up causing writepage to get
644 * called and if the server was down that means we end up here, and we
645 * can never tell if the caller already has the rename_sem.
646 */
647 full_path = build_path_from_dentry(cfile->dentry);
648 if (full_path == NULL) {
649 rc = -ENOMEM;
650 mutex_unlock(&cfile->fh_mutex);
651 free_xid(xid);
652 return rc;
653 }
654
655 cifs_dbg(FYI, "inode = 0x%p file flags 0x%x for %s\n",
656 inode, cfile->f_flags, full_path);
657
658 if (tcon->ses->server->oplocks)
659 oplock = REQ_OPLOCK;
660 else
661 oplock = 0;
662
663 if (tcon->unix_ext && cap_unix(tcon->ses) &&
664 (CIFS_UNIX_POSIX_PATH_OPS_CAP &
665 le64_to_cpu(tcon->fsUnixInfo.Capability))) {
666 /*
667 * O_CREAT, O_EXCL and O_TRUNC already had their effect on the
668 * original open. Must mask them off for a reopen.
669 */
670 unsigned int oflags = cfile->f_flags &
671 ~(O_CREAT | O_EXCL | O_TRUNC);
672
673 rc = cifs_posix_open(full_path, NULL, inode->i_sb,
674 cifs_sb->mnt_file_mode /* ignored */,
675 oflags, &oplock, &cfile->fid.netfid, xid);
676 if (rc == 0) {
677 cifs_dbg(FYI, "posix reopen succeeded\n");
678 oparms.reconnect = true;
679 goto reopen_success;
680 }
681 /*
682 * fallthrough to retry open the old way on errors, especially
683 * in the reconnect path it is important to retry hard
684 */
685 }
686
687 desired_access = cifs_convert_flags(cfile->f_flags);
688
689 if (backup_cred(cifs_sb))
690 create_options |= CREATE_OPEN_BACKUP_INTENT;
691
692 if (server->ops->get_lease_key)
693 server->ops->get_lease_key(inode, &cfile->fid);
694
695 oparms.tcon = tcon;
696 oparms.cifs_sb = cifs_sb;
697 oparms.desired_access = desired_access;
698 oparms.create_options = create_options;
699 oparms.disposition = disposition;
700 oparms.path = full_path;
701 oparms.fid = &cfile->fid;
702 oparms.reconnect = true;
703
704 /*
705 * Can not refresh inode by passing in file_info buf to be returned by
706 * ops->open and then calling get_inode_info with returned buf since
707 * file might have write behind data that needs to be flushed and server
708 * version of file size can be stale. If we knew for sure that inode was
709 * not dirty locally we could do this.
710 */
711 rc = server->ops->open(xid, &oparms, &oplock, NULL);
712 if (rc == -ENOENT && oparms.reconnect == false) {
713 /* durable handle timeout is expired - open the file again */
714 rc = server->ops->open(xid, &oparms, &oplock, NULL);
715 /* indicate that we need to relock the file */
716 oparms.reconnect = true;
717 }
718
719 if (rc) {
720 mutex_unlock(&cfile->fh_mutex);
721 cifs_dbg(FYI, "cifs_reopen returned 0x%x\n", rc);
722 cifs_dbg(FYI, "oplock: %d\n", oplock);
723 goto reopen_error_exit;
724 }
725
726reopen_success:
727 cfile->invalidHandle = false;
728 mutex_unlock(&cfile->fh_mutex);
729 cinode = CIFS_I(inode);
730
731 if (can_flush) {
732 rc = filemap_write_and_wait(inode->i_mapping);
733 mapping_set_error(inode->i_mapping, rc);
734
735 if (tcon->unix_ext)
736 rc = cifs_get_inode_info_unix(&inode, full_path,
737 inode->i_sb, xid);
738 else
739 rc = cifs_get_inode_info(&inode, full_path, NULL,
740 inode->i_sb, xid, NULL);
741 }
742 /*
743 * Else we are writing out data to server already and could deadlock if
744 * we tried to flush data, and since we do not know if we have data that
745 * would invalidate the current end of file on the server we can not go
746 * to the server to get the new inode info.
747 */
748
749 /*
750 * If the server returned a read oplock and we have mandatory brlocks,
751 * set oplock level to None.
752 */
753 if (server->ops->is_read_op(oplock) && cifs_has_mand_locks(cinode)) {
754 cifs_dbg(FYI, "Reset oplock val from read to None due to mand locks\n");
755 oplock = 0;
756 }
757
758 server->ops->set_fid(cfile, &cfile->fid, oplock);
759 if (oparms.reconnect)
760 cifs_relock_file(cfile);
761
762reopen_error_exit:
763 kfree(full_path);
764 free_xid(xid);
765 return rc;
766}
767
768int cifs_close(struct inode *inode, struct file *file)
769{
770 if (file->private_data != NULL) {
771 cifsFileInfo_put(file->private_data);
772 file->private_data = NULL;
773 }
774
775 /* return code from the ->release op is always ignored */
776 return 0;
777}
778
779void
780cifs_reopen_persistent_handles(struct cifs_tcon *tcon)
781{
782 struct cifsFileInfo *open_file;
783 struct list_head *tmp;
784 struct list_head *tmp1;
785 struct list_head tmp_list;
786
787 if (!tcon->use_persistent || !tcon->need_reopen_files)
788 return;
789
790 tcon->need_reopen_files = false;
791
792 cifs_dbg(FYI, "Reopen persistent handles");
793 INIT_LIST_HEAD(&tmp_list);
794
795 /* list all files open on tree connection, reopen resilient handles */
796 spin_lock(&tcon->open_file_lock);
797 list_for_each(tmp, &tcon->openFileList) {
798 open_file = list_entry(tmp, struct cifsFileInfo, tlist);
799 if (!open_file->invalidHandle)
800 continue;
801 cifsFileInfo_get(open_file);
802 list_add_tail(&open_file->rlist, &tmp_list);
803 }
804 spin_unlock(&tcon->open_file_lock);
805
806 list_for_each_safe(tmp, tmp1, &tmp_list) {
807 open_file = list_entry(tmp, struct cifsFileInfo, rlist);
808 if (cifs_reopen_file(open_file, false /* do not flush */))
809 tcon->need_reopen_files = true;
810 list_del_init(&open_file->rlist);
811 cifsFileInfo_put(open_file);
812 }
813}
814
815int cifs_closedir(struct inode *inode, struct file *file)
816{
817 int rc = 0;
818 unsigned int xid;
819 struct cifsFileInfo *cfile = file->private_data;
820 struct cifs_tcon *tcon;
821 struct TCP_Server_Info *server;
822 char *buf;
823
824 cifs_dbg(FYI, "Closedir inode = 0x%p\n", inode);
825
826 if (cfile == NULL)
827 return rc;
828
829 xid = get_xid();
830 tcon = tlink_tcon(cfile->tlink);
831 server = tcon->ses->server;
832
833 cifs_dbg(FYI, "Freeing private data in close dir\n");
834 spin_lock(&cfile->file_info_lock);
835 if (server->ops->dir_needs_close(cfile)) {
836 cfile->invalidHandle = true;
837 spin_unlock(&cfile->file_info_lock);
838 if (server->ops->close_dir)
839 rc = server->ops->close_dir(xid, tcon, &cfile->fid);
840 else
841 rc = -ENOSYS;
842 cifs_dbg(FYI, "Closing uncompleted readdir with rc %d\n", rc);
843 /* not much we can do if it fails anyway, ignore rc */
844 rc = 0;
845 } else
846 spin_unlock(&cfile->file_info_lock);
847
848 buf = cfile->srch_inf.ntwrk_buf_start;
849 if (buf) {
850 cifs_dbg(FYI, "closedir free smb buf in srch struct\n");
851 cfile->srch_inf.ntwrk_buf_start = NULL;
852 if (cfile->srch_inf.smallBuf)
853 cifs_small_buf_release(buf);
854 else
855 cifs_buf_release(buf);
856 }
857
858 cifs_put_tlink(cfile->tlink);
859 kfree(file->private_data);
860 file->private_data = NULL;
861 /* BB can we lock the filestruct while this is going on? */
862 free_xid(xid);
863 return rc;
864}
865
866static struct cifsLockInfo *
867cifs_lock_init(__u64 offset, __u64 length, __u8 type)
868{
869 struct cifsLockInfo *lock =
870 kmalloc(sizeof(struct cifsLockInfo), GFP_KERNEL);
871 if (!lock)
872 return lock;
873 lock->offset = offset;
874 lock->length = length;
875 lock->type = type;
876 lock->pid = current->tgid;
877 INIT_LIST_HEAD(&lock->blist);
878 init_waitqueue_head(&lock->block_q);
879 return lock;
880}
881
882void
883cifs_del_lock_waiters(struct cifsLockInfo *lock)
884{
885 struct cifsLockInfo *li, *tmp;
886 list_for_each_entry_safe(li, tmp, &lock->blist, blist) {
887 list_del_init(&li->blist);
888 wake_up(&li->block_q);
889 }
890}
891
892#define CIFS_LOCK_OP 0
893#define CIFS_READ_OP 1
894#define CIFS_WRITE_OP 2
895
896/* @rw_check : 0 - no op, 1 - read, 2 - write */
897static bool
898cifs_find_fid_lock_conflict(struct cifs_fid_locks *fdlocks, __u64 offset,
899 __u64 length, __u8 type, struct cifsFileInfo *cfile,
900 struct cifsLockInfo **conf_lock, int rw_check)
901{
902 struct cifsLockInfo *li;
903 struct cifsFileInfo *cur_cfile = fdlocks->cfile;
904 struct TCP_Server_Info *server = tlink_tcon(cfile->tlink)->ses->server;
905
906 list_for_each_entry(li, &fdlocks->locks, llist) {
907 if (offset + length <= li->offset ||
908 offset >= li->offset + li->length)
909 continue;
910 if (rw_check != CIFS_LOCK_OP && current->tgid == li->pid &&
911 server->ops->compare_fids(cfile, cur_cfile)) {
912 /* shared lock prevents write op through the same fid */
913 if (!(li->type & server->vals->shared_lock_type) ||
914 rw_check != CIFS_WRITE_OP)
915 continue;
916 }
917 if ((type & server->vals->shared_lock_type) &&
918 ((server->ops->compare_fids(cfile, cur_cfile) &&
919 current->tgid == li->pid) || type == li->type))
920 continue;
921 if (conf_lock)
922 *conf_lock = li;
923 return true;
924 }
925 return false;
926}
927
928bool
929cifs_find_lock_conflict(struct cifsFileInfo *cfile, __u64 offset, __u64 length,
930 __u8 type, struct cifsLockInfo **conf_lock,
931 int rw_check)
932{
933 bool rc = false;
934 struct cifs_fid_locks *cur;
935 struct cifsInodeInfo *cinode = CIFS_I(d_inode(cfile->dentry));
936
937 list_for_each_entry(cur, &cinode->llist, llist) {
938 rc = cifs_find_fid_lock_conflict(cur, offset, length, type,
939 cfile, conf_lock, rw_check);
940 if (rc)
941 break;
942 }
943
944 return rc;
945}
946
947/*
948 * Check if there is another lock that prevents us to set the lock (mandatory
949 * style). If such a lock exists, update the flock structure with its
950 * properties. Otherwise, set the flock type to F_UNLCK if we can cache brlocks
951 * or leave it the same if we can't. Returns 0 if we don't need to request to
952 * the server or 1 otherwise.
953 */
954static int
955cifs_lock_test(struct cifsFileInfo *cfile, __u64 offset, __u64 length,
956 __u8 type, struct file_lock *flock)
957{
958 int rc = 0;
959 struct cifsLockInfo *conf_lock;
960 struct cifsInodeInfo *cinode = CIFS_I(d_inode(cfile->dentry));
961 struct TCP_Server_Info *server = tlink_tcon(cfile->tlink)->ses->server;
962 bool exist;
963
964 down_read(&cinode->lock_sem);
965
966 exist = cifs_find_lock_conflict(cfile, offset, length, type,
967 &conf_lock, CIFS_LOCK_OP);
968 if (exist) {
969 flock->fl_start = conf_lock->offset;
970 flock->fl_end = conf_lock->offset + conf_lock->length - 1;
971 flock->fl_pid = conf_lock->pid;
972 if (conf_lock->type & server->vals->shared_lock_type)
973 flock->fl_type = F_RDLCK;
974 else
975 flock->fl_type = F_WRLCK;
976 } else if (!cinode->can_cache_brlcks)
977 rc = 1;
978 else
979 flock->fl_type = F_UNLCK;
980
981 up_read(&cinode->lock_sem);
982 return rc;
983}
984
985static void
986cifs_lock_add(struct cifsFileInfo *cfile, struct cifsLockInfo *lock)
987{
988 struct cifsInodeInfo *cinode = CIFS_I(d_inode(cfile->dentry));
989 down_write(&cinode->lock_sem);
990 list_add_tail(&lock->llist, &cfile->llist->locks);
991 up_write(&cinode->lock_sem);
992}
993
994/*
995 * Set the byte-range lock (mandatory style). Returns:
996 * 1) 0, if we set the lock and don't need to request to the server;
997 * 2) 1, if no locks prevent us but we need to request to the server;
998 * 3) -EACCESS, if there is a lock that prevents us and wait is false.
999 */
1000static int
1001cifs_lock_add_if(struct cifsFileInfo *cfile, struct cifsLockInfo *lock,
1002 bool wait)
1003{
1004 struct cifsLockInfo *conf_lock;
1005 struct cifsInodeInfo *cinode = CIFS_I(d_inode(cfile->dentry));
1006 bool exist;
1007 int rc = 0;
1008
1009try_again:
1010 exist = false;
1011 down_write(&cinode->lock_sem);
1012
1013 exist = cifs_find_lock_conflict(cfile, lock->offset, lock->length,
1014 lock->type, &conf_lock, CIFS_LOCK_OP);
1015 if (!exist && cinode->can_cache_brlcks) {
1016 list_add_tail(&lock->llist, &cfile->llist->locks);
1017 up_write(&cinode->lock_sem);
1018 return rc;
1019 }
1020
1021 if (!exist)
1022 rc = 1;
1023 else if (!wait)
1024 rc = -EACCES;
1025 else {
1026 list_add_tail(&lock->blist, &conf_lock->blist);
1027 up_write(&cinode->lock_sem);
1028 rc = wait_event_interruptible(lock->block_q,
1029 (lock->blist.prev == &lock->blist) &&
1030 (lock->blist.next == &lock->blist));
1031 if (!rc)
1032 goto try_again;
1033 down_write(&cinode->lock_sem);
1034 list_del_init(&lock->blist);
1035 }
1036
1037 up_write(&cinode->lock_sem);
1038 return rc;
1039}
1040
1041/*
1042 * Check if there is another lock that prevents us to set the lock (posix
1043 * style). If such a lock exists, update the flock structure with its
1044 * properties. Otherwise, set the flock type to F_UNLCK if we can cache brlocks
1045 * or leave it the same if we can't. Returns 0 if we don't need to request to
1046 * the server or 1 otherwise.
1047 */
1048static int
1049cifs_posix_lock_test(struct file *file, struct file_lock *flock)
1050{
1051 int rc = 0;
1052 struct cifsInodeInfo *cinode = CIFS_I(file_inode(file));
1053 unsigned char saved_type = flock->fl_type;
1054
1055 if ((flock->fl_flags & FL_POSIX) == 0)
1056 return 1;
1057
1058 down_read(&cinode->lock_sem);
1059 posix_test_lock(file, flock);
1060
1061 if (flock->fl_type == F_UNLCK && !cinode->can_cache_brlcks) {
1062 flock->fl_type = saved_type;
1063 rc = 1;
1064 }
1065
1066 up_read(&cinode->lock_sem);
1067 return rc;
1068}
1069
1070/*
1071 * Set the byte-range lock (posix style). Returns:
1072 * 1) 0, if we set the lock and don't need to request to the server;
1073 * 2) 1, if we need to request to the server;
1074 * 3) <0, if the error occurs while setting the lock.
1075 */
1076static int
1077cifs_posix_lock_set(struct file *file, struct file_lock *flock)
1078{
1079 struct cifsInodeInfo *cinode = CIFS_I(file_inode(file));
1080 int rc = 1;
1081
1082 if ((flock->fl_flags & FL_POSIX) == 0)
1083 return rc;
1084
1085try_again:
1086 down_write(&cinode->lock_sem);
1087 if (!cinode->can_cache_brlcks) {
1088 up_write(&cinode->lock_sem);
1089 return rc;
1090 }
1091
1092 rc = posix_lock_file(file, flock, NULL);
1093 up_write(&cinode->lock_sem);
1094 if (rc == FILE_LOCK_DEFERRED) {
1095 rc = wait_event_interruptible(flock->fl_wait, !flock->fl_next);
1096 if (!rc)
1097 goto try_again;
1098 posix_unblock_lock(flock);
1099 }
1100 return rc;
1101}
1102
1103int
1104cifs_push_mandatory_locks(struct cifsFileInfo *cfile)
1105{
1106 unsigned int xid;
1107 int rc = 0, stored_rc;
1108 struct cifsLockInfo *li, *tmp;
1109 struct cifs_tcon *tcon;
1110 unsigned int num, max_num, max_buf;
1111 LOCKING_ANDX_RANGE *buf, *cur;
1112 static const int types[] = {
1113 LOCKING_ANDX_LARGE_FILES,
1114 LOCKING_ANDX_SHARED_LOCK | LOCKING_ANDX_LARGE_FILES
1115 };
1116 int i;
1117
1118 xid = get_xid();
1119 tcon = tlink_tcon(cfile->tlink);
1120
1121 /*
1122 * Accessing maxBuf is racy with cifs_reconnect - need to store value
1123 * and check it for zero before using.
1124 */
1125 max_buf = tcon->ses->server->maxBuf;
1126 if (!max_buf) {
1127 free_xid(xid);
1128 return -EINVAL;
1129 }
1130
1131 max_num = (max_buf - sizeof(struct smb_hdr)) /
1132 sizeof(LOCKING_ANDX_RANGE);
1133 buf = kcalloc(max_num, sizeof(LOCKING_ANDX_RANGE), GFP_KERNEL);
1134 if (!buf) {
1135 free_xid(xid);
1136 return -ENOMEM;
1137 }
1138
1139 for (i = 0; i < 2; i++) {
1140 cur = buf;
1141 num = 0;
1142 list_for_each_entry_safe(li, tmp, &cfile->llist->locks, llist) {
1143 if (li->type != types[i])
1144 continue;
1145 cur->Pid = cpu_to_le16(li->pid);
1146 cur->LengthLow = cpu_to_le32((u32)li->length);
1147 cur->LengthHigh = cpu_to_le32((u32)(li->length>>32));
1148 cur->OffsetLow = cpu_to_le32((u32)li->offset);
1149 cur->OffsetHigh = cpu_to_le32((u32)(li->offset>>32));
1150 if (++num == max_num) {
1151 stored_rc = cifs_lockv(xid, tcon,
1152 cfile->fid.netfid,
1153 (__u8)li->type, 0, num,
1154 buf);
1155 if (stored_rc)
1156 rc = stored_rc;
1157 cur = buf;
1158 num = 0;
1159 } else
1160 cur++;
1161 }
1162
1163 if (num) {
1164 stored_rc = cifs_lockv(xid, tcon, cfile->fid.netfid,
1165 (__u8)types[i], 0, num, buf);
1166 if (stored_rc)
1167 rc = stored_rc;
1168 }
1169 }
1170
1171 kfree(buf);
1172 free_xid(xid);
1173 return rc;
1174}
1175
1176static __u32
1177hash_lockowner(fl_owner_t owner)
1178{
1179 return cifs_lock_secret ^ hash32_ptr((const void *)owner);
1180}
1181
1182struct lock_to_push {
1183 struct list_head llist;
1184 __u64 offset;
1185 __u64 length;
1186 __u32 pid;
1187 __u16 netfid;
1188 __u8 type;
1189};
1190
1191static int
1192cifs_push_posix_locks(struct cifsFileInfo *cfile)
1193{
1194 struct inode *inode = d_inode(cfile->dentry);
1195 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
1196 struct file_lock *flock;
1197 struct file_lock_context *flctx = inode->i_flctx;
1198 unsigned int count = 0, i;
1199 int rc = 0, xid, type;
1200 struct list_head locks_to_send, *el;
1201 struct lock_to_push *lck, *tmp;
1202 __u64 length;
1203
1204 xid = get_xid();
1205
1206 if (!flctx)
1207 goto out;
1208
1209 spin_lock(&flctx->flc_lock);
1210 list_for_each(el, &flctx->flc_posix) {
1211 count++;
1212 }
1213 spin_unlock(&flctx->flc_lock);
1214
1215 INIT_LIST_HEAD(&locks_to_send);
1216
1217 /*
1218 * Allocating count locks is enough because no FL_POSIX locks can be
1219 * added to the list while we are holding cinode->lock_sem that
1220 * protects locking operations of this inode.
1221 */
1222 for (i = 0; i < count; i++) {
1223 lck = kmalloc(sizeof(struct lock_to_push), GFP_KERNEL);
1224 if (!lck) {
1225 rc = -ENOMEM;
1226 goto err_out;
1227 }
1228 list_add_tail(&lck->llist, &locks_to_send);
1229 }
1230
1231 el = locks_to_send.next;
1232 spin_lock(&flctx->flc_lock);
1233 list_for_each_entry(flock, &flctx->flc_posix, fl_list) {
1234 if (el == &locks_to_send) {
1235 /*
1236 * The list ended. We don't have enough allocated
1237 * structures - something is really wrong.
1238 */
1239 cifs_dbg(VFS, "Can't push all brlocks!\n");
1240 break;
1241 }
1242 length = 1 + flock->fl_end - flock->fl_start;
1243 if (flock->fl_type == F_RDLCK || flock->fl_type == F_SHLCK)
1244 type = CIFS_RDLCK;
1245 else
1246 type = CIFS_WRLCK;
1247 lck = list_entry(el, struct lock_to_push, llist);
1248 lck->pid = hash_lockowner(flock->fl_owner);
1249 lck->netfid = cfile->fid.netfid;
1250 lck->length = length;
1251 lck->type = type;
1252 lck->offset = flock->fl_start;
1253 }
1254 spin_unlock(&flctx->flc_lock);
1255
1256 list_for_each_entry_safe(lck, tmp, &locks_to_send, llist) {
1257 int stored_rc;
1258
1259 stored_rc = CIFSSMBPosixLock(xid, tcon, lck->netfid, lck->pid,
1260 lck->offset, lck->length, NULL,
1261 lck->type, 0);
1262 if (stored_rc)
1263 rc = stored_rc;
1264 list_del(&lck->llist);
1265 kfree(lck);
1266 }
1267
1268out:
1269 free_xid(xid);
1270 return rc;
1271err_out:
1272 list_for_each_entry_safe(lck, tmp, &locks_to_send, llist) {
1273 list_del(&lck->llist);
1274 kfree(lck);
1275 }
1276 goto out;
1277}
1278
1279static int
1280cifs_push_locks(struct cifsFileInfo *cfile)
1281{
1282 struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb);
1283 struct cifsInodeInfo *cinode = CIFS_I(d_inode(cfile->dentry));
1284 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
1285 int rc = 0;
1286
1287 /* we are going to update can_cache_brlcks here - need a write access */
1288 down_write(&cinode->lock_sem);
1289 if (!cinode->can_cache_brlcks) {
1290 up_write(&cinode->lock_sem);
1291 return rc;
1292 }
1293
1294 if (cap_unix(tcon->ses) &&
1295 (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) &&
1296 ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0))
1297 rc = cifs_push_posix_locks(cfile);
1298 else
1299 rc = tcon->ses->server->ops->push_mand_locks(cfile);
1300
1301 cinode->can_cache_brlcks = false;
1302 up_write(&cinode->lock_sem);
1303 return rc;
1304}
1305
1306static void
1307cifs_read_flock(struct file_lock *flock, __u32 *type, int *lock, int *unlock,
1308 bool *wait_flag, struct TCP_Server_Info *server)
1309{
1310 if (flock->fl_flags & FL_POSIX)
1311 cifs_dbg(FYI, "Posix\n");
1312 if (flock->fl_flags & FL_FLOCK)
1313 cifs_dbg(FYI, "Flock\n");
1314 if (flock->fl_flags & FL_SLEEP) {
1315 cifs_dbg(FYI, "Blocking lock\n");
1316 *wait_flag = true;
1317 }
1318 if (flock->fl_flags & FL_ACCESS)
1319 cifs_dbg(FYI, "Process suspended by mandatory locking - not implemented yet\n");
1320 if (flock->fl_flags & FL_LEASE)
1321 cifs_dbg(FYI, "Lease on file - not implemented yet\n");
1322 if (flock->fl_flags &
1323 (~(FL_POSIX | FL_FLOCK | FL_SLEEP |
1324 FL_ACCESS | FL_LEASE | FL_CLOSE)))
1325 cifs_dbg(FYI, "Unknown lock flags 0x%x\n", flock->fl_flags);
1326
1327 *type = server->vals->large_lock_type;
1328 if (flock->fl_type == F_WRLCK) {
1329 cifs_dbg(FYI, "F_WRLCK\n");
1330 *type |= server->vals->exclusive_lock_type;
1331 *lock = 1;
1332 } else if (flock->fl_type == F_UNLCK) {
1333 cifs_dbg(FYI, "F_UNLCK\n");
1334 *type |= server->vals->unlock_lock_type;
1335 *unlock = 1;
1336 /* Check if unlock includes more than one lock range */
1337 } else if (flock->fl_type == F_RDLCK) {
1338 cifs_dbg(FYI, "F_RDLCK\n");
1339 *type |= server->vals->shared_lock_type;
1340 *lock = 1;
1341 } else if (flock->fl_type == F_EXLCK) {
1342 cifs_dbg(FYI, "F_EXLCK\n");
1343 *type |= server->vals->exclusive_lock_type;
1344 *lock = 1;
1345 } else if (flock->fl_type == F_SHLCK) {
1346 cifs_dbg(FYI, "F_SHLCK\n");
1347 *type |= server->vals->shared_lock_type;
1348 *lock = 1;
1349 } else
1350 cifs_dbg(FYI, "Unknown type of lock\n");
1351}
1352
1353static int
1354cifs_getlk(struct file *file, struct file_lock *flock, __u32 type,
1355 bool wait_flag, bool posix_lck, unsigned int xid)
1356{
1357 int rc = 0;
1358 __u64 length = 1 + flock->fl_end - flock->fl_start;
1359 struct cifsFileInfo *cfile = (struct cifsFileInfo *)file->private_data;
1360 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
1361 struct TCP_Server_Info *server = tcon->ses->server;
1362 __u16 netfid = cfile->fid.netfid;
1363
1364 if (posix_lck) {
1365 int posix_lock_type;
1366
1367 rc = cifs_posix_lock_test(file, flock);
1368 if (!rc)
1369 return rc;
1370
1371 if (type & server->vals->shared_lock_type)
1372 posix_lock_type = CIFS_RDLCK;
1373 else
1374 posix_lock_type = CIFS_WRLCK;
1375 rc = CIFSSMBPosixLock(xid, tcon, netfid,
1376 hash_lockowner(flock->fl_owner),
1377 flock->fl_start, length, flock,
1378 posix_lock_type, wait_flag);
1379 return rc;
1380 }
1381
1382 rc = cifs_lock_test(cfile, flock->fl_start, length, type, flock);
1383 if (!rc)
1384 return rc;
1385
1386 /* BB we could chain these into one lock request BB */
1387 rc = server->ops->mand_lock(xid, cfile, flock->fl_start, length, type,
1388 1, 0, false);
1389 if (rc == 0) {
1390 rc = server->ops->mand_lock(xid, cfile, flock->fl_start, length,
1391 type, 0, 1, false);
1392 flock->fl_type = F_UNLCK;
1393 if (rc != 0)
1394 cifs_dbg(VFS, "Error unlocking previously locked range %d during test of lock\n",
1395 rc);
1396 return 0;
1397 }
1398
1399 if (type & server->vals->shared_lock_type) {
1400 flock->fl_type = F_WRLCK;
1401 return 0;
1402 }
1403
1404 type &= ~server->vals->exclusive_lock_type;
1405
1406 rc = server->ops->mand_lock(xid, cfile, flock->fl_start, length,
1407 type | server->vals->shared_lock_type,
1408 1, 0, false);
1409 if (rc == 0) {
1410 rc = server->ops->mand_lock(xid, cfile, flock->fl_start, length,
1411 type | server->vals->shared_lock_type, 0, 1, false);
1412 flock->fl_type = F_RDLCK;
1413 if (rc != 0)
1414 cifs_dbg(VFS, "Error unlocking previously locked range %d during test of lock\n",
1415 rc);
1416 } else
1417 flock->fl_type = F_WRLCK;
1418
1419 return 0;
1420}
1421
1422void
1423cifs_move_llist(struct list_head *source, struct list_head *dest)
1424{
1425 struct list_head *li, *tmp;
1426 list_for_each_safe(li, tmp, source)
1427 list_move(li, dest);
1428}
1429
1430void
1431cifs_free_llist(struct list_head *llist)
1432{
1433 struct cifsLockInfo *li, *tmp;
1434 list_for_each_entry_safe(li, tmp, llist, llist) {
1435 cifs_del_lock_waiters(li);
1436 list_del(&li->llist);
1437 kfree(li);
1438 }
1439}
1440
1441int
1442cifs_unlock_range(struct cifsFileInfo *cfile, struct file_lock *flock,
1443 unsigned int xid)
1444{
1445 int rc = 0, stored_rc;
1446 static const int types[] = {
1447 LOCKING_ANDX_LARGE_FILES,
1448 LOCKING_ANDX_SHARED_LOCK | LOCKING_ANDX_LARGE_FILES
1449 };
1450 unsigned int i;
1451 unsigned int max_num, num, max_buf;
1452 LOCKING_ANDX_RANGE *buf, *cur;
1453 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
1454 struct cifsInodeInfo *cinode = CIFS_I(d_inode(cfile->dentry));
1455 struct cifsLockInfo *li, *tmp;
1456 __u64 length = 1 + flock->fl_end - flock->fl_start;
1457 struct list_head tmp_llist;
1458
1459 INIT_LIST_HEAD(&tmp_llist);
1460
1461 /*
1462 * Accessing maxBuf is racy with cifs_reconnect - need to store value
1463 * and check it for zero before using.
1464 */
1465 max_buf = tcon->ses->server->maxBuf;
1466 if (!max_buf)
1467 return -EINVAL;
1468
1469 max_num = (max_buf - sizeof(struct smb_hdr)) /
1470 sizeof(LOCKING_ANDX_RANGE);
1471 buf = kcalloc(max_num, sizeof(LOCKING_ANDX_RANGE), GFP_KERNEL);
1472 if (!buf)
1473 return -ENOMEM;
1474
1475 down_write(&cinode->lock_sem);
1476 for (i = 0; i < 2; i++) {
1477 cur = buf;
1478 num = 0;
1479 list_for_each_entry_safe(li, tmp, &cfile->llist->locks, llist) {
1480 if (flock->fl_start > li->offset ||
1481 (flock->fl_start + length) <
1482 (li->offset + li->length))
1483 continue;
1484 if (current->tgid != li->pid)
1485 continue;
1486 if (types[i] != li->type)
1487 continue;
1488 if (cinode->can_cache_brlcks) {
1489 /*
1490 * We can cache brlock requests - simply remove
1491 * a lock from the file's list.
1492 */
1493 list_del(&li->llist);
1494 cifs_del_lock_waiters(li);
1495 kfree(li);
1496 continue;
1497 }
1498 cur->Pid = cpu_to_le16(li->pid);
1499 cur->LengthLow = cpu_to_le32((u32)li->length);
1500 cur->LengthHigh = cpu_to_le32((u32)(li->length>>32));
1501 cur->OffsetLow = cpu_to_le32((u32)li->offset);
1502 cur->OffsetHigh = cpu_to_le32((u32)(li->offset>>32));
1503 /*
1504 * We need to save a lock here to let us add it again to
1505 * the file's list if the unlock range request fails on
1506 * the server.
1507 */
1508 list_move(&li->llist, &tmp_llist);
1509 if (++num == max_num) {
1510 stored_rc = cifs_lockv(xid, tcon,
1511 cfile->fid.netfid,
1512 li->type, num, 0, buf);
1513 if (stored_rc) {
1514 /*
1515 * We failed on the unlock range
1516 * request - add all locks from the tmp
1517 * list to the head of the file's list.
1518 */
1519 cifs_move_llist(&tmp_llist,
1520 &cfile->llist->locks);
1521 rc = stored_rc;
1522 } else
1523 /*
1524 * The unlock range request succeed -
1525 * free the tmp list.
1526 */
1527 cifs_free_llist(&tmp_llist);
1528 cur = buf;
1529 num = 0;
1530 } else
1531 cur++;
1532 }
1533 if (num) {
1534 stored_rc = cifs_lockv(xid, tcon, cfile->fid.netfid,
1535 types[i], num, 0, buf);
1536 if (stored_rc) {
1537 cifs_move_llist(&tmp_llist,
1538 &cfile->llist->locks);
1539 rc = stored_rc;
1540 } else
1541 cifs_free_llist(&tmp_llist);
1542 }
1543 }
1544
1545 up_write(&cinode->lock_sem);
1546 kfree(buf);
1547 return rc;
1548}
1549
1550static int
1551cifs_setlk(struct file *file, struct file_lock *flock, __u32 type,
1552 bool wait_flag, bool posix_lck, int lock, int unlock,
1553 unsigned int xid)
1554{
1555 int rc = 0;
1556 __u64 length = 1 + flock->fl_end - flock->fl_start;
1557 struct cifsFileInfo *cfile = (struct cifsFileInfo *)file->private_data;
1558 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
1559 struct TCP_Server_Info *server = tcon->ses->server;
1560 struct inode *inode = d_inode(cfile->dentry);
1561
1562 if (posix_lck) {
1563 int posix_lock_type;
1564
1565 rc = cifs_posix_lock_set(file, flock);
1566 if (!rc || rc < 0)
1567 return rc;
1568
1569 if (type & server->vals->shared_lock_type)
1570 posix_lock_type = CIFS_RDLCK;
1571 else
1572 posix_lock_type = CIFS_WRLCK;
1573
1574 if (unlock == 1)
1575 posix_lock_type = CIFS_UNLCK;
1576
1577 rc = CIFSSMBPosixLock(xid, tcon, cfile->fid.netfid,
1578 hash_lockowner(flock->fl_owner),
1579 flock->fl_start, length,
1580 NULL, posix_lock_type, wait_flag);
1581 goto out;
1582 }
1583
1584 if (lock) {
1585 struct cifsLockInfo *lock;
1586
1587 lock = cifs_lock_init(flock->fl_start, length, type);
1588 if (!lock)
1589 return -ENOMEM;
1590
1591 rc = cifs_lock_add_if(cfile, lock, wait_flag);
1592 if (rc < 0) {
1593 kfree(lock);
1594 return rc;
1595 }
1596 if (!rc)
1597 goto out;
1598
1599 /*
1600 * Windows 7 server can delay breaking lease from read to None
1601 * if we set a byte-range lock on a file - break it explicitly
1602 * before sending the lock to the server to be sure the next
1603 * read won't conflict with non-overlapted locks due to
1604 * pagereading.
1605 */
1606 if (!CIFS_CACHE_WRITE(CIFS_I(inode)) &&
1607 CIFS_CACHE_READ(CIFS_I(inode))) {
1608 cifs_zap_mapping(inode);
1609 cifs_dbg(FYI, "Set no oplock for inode=%p due to mand locks\n",
1610 inode);
1611 CIFS_I(inode)->oplock = 0;
1612 }
1613
1614 rc = server->ops->mand_lock(xid, cfile, flock->fl_start, length,
1615 type, 1, 0, wait_flag);
1616 if (rc) {
1617 kfree(lock);
1618 return rc;
1619 }
1620
1621 cifs_lock_add(cfile, lock);
1622 } else if (unlock)
1623 rc = server->ops->mand_unlock_range(cfile, flock, xid);
1624
1625out:
1626 if (flock->fl_flags & FL_POSIX && !rc)
1627 rc = locks_lock_file_wait(file, flock);
1628 return rc;
1629}
1630
1631int cifs_lock(struct file *file, int cmd, struct file_lock *flock)
1632{
1633 int rc, xid;
1634 int lock = 0, unlock = 0;
1635 bool wait_flag = false;
1636 bool posix_lck = false;
1637 struct cifs_sb_info *cifs_sb;
1638 struct cifs_tcon *tcon;
1639 struct cifsInodeInfo *cinode;
1640 struct cifsFileInfo *cfile;
1641 __u16 netfid;
1642 __u32 type;
1643
1644 rc = -EACCES;
1645 xid = get_xid();
1646
1647 cifs_dbg(FYI, "Lock parm: 0x%x flockflags: 0x%x flocktype: 0x%x start: %lld end: %lld\n",
1648 cmd, flock->fl_flags, flock->fl_type,
1649 flock->fl_start, flock->fl_end);
1650
1651 cfile = (struct cifsFileInfo *)file->private_data;
1652 tcon = tlink_tcon(cfile->tlink);
1653
1654 cifs_read_flock(flock, &type, &lock, &unlock, &wait_flag,
1655 tcon->ses->server);
1656
1657 cifs_sb = CIFS_FILE_SB(file);
1658 netfid = cfile->fid.netfid;
1659 cinode = CIFS_I(file_inode(file));
1660
1661 if (cap_unix(tcon->ses) &&
1662 (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) &&
1663 ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0))
1664 posix_lck = true;
1665 /*
1666 * BB add code here to normalize offset and length to account for
1667 * negative length which we can not accept over the wire.
1668 */
1669 if (IS_GETLK(cmd)) {
1670 rc = cifs_getlk(file, flock, type, wait_flag, posix_lck, xid);
1671 free_xid(xid);
1672 return rc;
1673 }
1674
1675 if (!lock && !unlock) {
1676 /*
1677 * if no lock or unlock then nothing to do since we do not
1678 * know what it is
1679 */
1680 free_xid(xid);
1681 return -EOPNOTSUPP;
1682 }
1683
1684 rc = cifs_setlk(file, flock, type, wait_flag, posix_lck, lock, unlock,
1685 xid);
1686 free_xid(xid);
1687 return rc;
1688}
1689
1690/*
1691 * update the file size (if needed) after a write. Should be called with
1692 * the inode->i_lock held
1693 */
1694void
1695cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset,
1696 unsigned int bytes_written)
1697{
1698 loff_t end_of_write = offset + bytes_written;
1699
1700 if (end_of_write > cifsi->server_eof)
1701 cifsi->server_eof = end_of_write;
1702}
1703
1704static ssize_t
1705cifs_write(struct cifsFileInfo *open_file, __u32 pid, const char *write_data,
1706 size_t write_size, loff_t *offset)
1707{
1708 int rc = 0;
1709 unsigned int bytes_written = 0;
1710 unsigned int total_written;
1711 struct cifs_sb_info *cifs_sb;
1712 struct cifs_tcon *tcon;
1713 struct TCP_Server_Info *server;
1714 unsigned int xid;
1715 struct dentry *dentry = open_file->dentry;
1716 struct cifsInodeInfo *cifsi = CIFS_I(d_inode(dentry));
1717 struct cifs_io_parms io_parms;
1718
1719 cifs_sb = CIFS_SB(dentry->d_sb);
1720
1721 cifs_dbg(FYI, "write %zd bytes to offset %lld of %pd\n",
1722 write_size, *offset, dentry);
1723
1724 tcon = tlink_tcon(open_file->tlink);
1725 server = tcon->ses->server;
1726
1727 if (!server->ops->sync_write)
1728 return -ENOSYS;
1729
1730 xid = get_xid();
1731
1732 for (total_written = 0; write_size > total_written;
1733 total_written += bytes_written) {
1734 rc = -EAGAIN;
1735 while (rc == -EAGAIN) {
1736 struct kvec iov[2];
1737 unsigned int len;
1738
1739 if (open_file->invalidHandle) {
1740 /* we could deadlock if we called
1741 filemap_fdatawait from here so tell
1742 reopen_file not to flush data to
1743 server now */
1744 rc = cifs_reopen_file(open_file, false);
1745 if (rc != 0)
1746 break;
1747 }
1748
1749 len = min(server->ops->wp_retry_size(d_inode(dentry)),
1750 (unsigned int)write_size - total_written);
1751 /* iov[0] is reserved for smb header */
1752 iov[1].iov_base = (char *)write_data + total_written;
1753 iov[1].iov_len = len;
1754 io_parms.pid = pid;
1755 io_parms.tcon = tcon;
1756 io_parms.offset = *offset;
1757 io_parms.length = len;
1758 rc = server->ops->sync_write(xid, &open_file->fid,
1759 &io_parms, &bytes_written, iov, 1);
1760 }
1761 if (rc || (bytes_written == 0)) {
1762 if (total_written)
1763 break;
1764 else {
1765 free_xid(xid);
1766 return rc;
1767 }
1768 } else {
1769 spin_lock(&d_inode(dentry)->i_lock);
1770 cifs_update_eof(cifsi, *offset, bytes_written);
1771 spin_unlock(&d_inode(dentry)->i_lock);
1772 *offset += bytes_written;
1773 }
1774 }
1775
1776 cifs_stats_bytes_written(tcon, total_written);
1777
1778 if (total_written > 0) {
1779 spin_lock(&d_inode(dentry)->i_lock);
1780 if (*offset > d_inode(dentry)->i_size)
1781 i_size_write(d_inode(dentry), *offset);
1782 spin_unlock(&d_inode(dentry)->i_lock);
1783 }
1784 mark_inode_dirty_sync(d_inode(dentry));
1785 free_xid(xid);
1786 return total_written;
1787}
1788
1789struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode,
1790 bool fsuid_only)
1791{
1792 struct cifsFileInfo *open_file = NULL;
1793 struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb);
1794 struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb);
1795
1796 /* only filter by fsuid on multiuser mounts */
1797 if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER))
1798 fsuid_only = false;
1799
1800 spin_lock(&tcon->open_file_lock);
1801 /* we could simply get the first_list_entry since write-only entries
1802 are always at the end of the list but since the first entry might
1803 have a close pending, we go through the whole list */
1804 list_for_each_entry(open_file, &cifs_inode->openFileList, flist) {
1805 if (fsuid_only && !uid_eq(open_file->uid, current_fsuid()))
1806 continue;
1807 if (OPEN_FMODE(open_file->f_flags) & FMODE_READ) {
1808 if (!open_file->invalidHandle) {
1809 /* found a good file */
1810 /* lock it so it will not be closed on us */
1811 cifsFileInfo_get(open_file);
1812 spin_unlock(&tcon->open_file_lock);
1813 return open_file;
1814 } /* else might as well continue, and look for
1815 another, or simply have the caller reopen it
1816 again rather than trying to fix this handle */
1817 } else /* write only file */
1818 break; /* write only files are last so must be done */
1819 }
1820 spin_unlock(&tcon->open_file_lock);
1821 return NULL;
1822}
1823
1824struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode,
1825 bool fsuid_only)
1826{
1827 struct cifsFileInfo *open_file, *inv_file = NULL;
1828 struct cifs_sb_info *cifs_sb;
1829 struct cifs_tcon *tcon;
1830 bool any_available = false;
1831 int rc;
1832 unsigned int refind = 0;
1833
1834 /* Having a null inode here (because mapping->host was set to zero by
1835 the VFS or MM) should not happen but we had reports of on oops (due to
1836 it being zero) during stress testcases so we need to check for it */
1837
1838 if (cifs_inode == NULL) {
1839 cifs_dbg(VFS, "Null inode passed to cifs_writeable_file\n");
1840 dump_stack();
1841 return NULL;
1842 }
1843
1844 cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb);
1845 tcon = cifs_sb_master_tcon(cifs_sb);
1846
1847 /* only filter by fsuid on multiuser mounts */
1848 if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER))
1849 fsuid_only = false;
1850
1851 spin_lock(&tcon->open_file_lock);
1852refind_writable:
1853 if (refind > MAX_REOPEN_ATT) {
1854 spin_unlock(&tcon->open_file_lock);
1855 return NULL;
1856 }
1857 list_for_each_entry(open_file, &cifs_inode->openFileList, flist) {
1858 if (!any_available && open_file->pid != current->tgid)
1859 continue;
1860 if (fsuid_only && !uid_eq(open_file->uid, current_fsuid()))
1861 continue;
1862 if (OPEN_FMODE(open_file->f_flags) & FMODE_WRITE) {
1863 if (!open_file->invalidHandle) {
1864 /* found a good writable file */
1865 cifsFileInfo_get(open_file);
1866 spin_unlock(&tcon->open_file_lock);
1867 return open_file;
1868 } else {
1869 if (!inv_file)
1870 inv_file = open_file;
1871 }
1872 }
1873 }
1874 /* couldn't find useable FH with same pid, try any available */
1875 if (!any_available) {
1876 any_available = true;
1877 goto refind_writable;
1878 }
1879
1880 if (inv_file) {
1881 any_available = false;
1882 cifsFileInfo_get(inv_file);
1883 }
1884
1885 spin_unlock(&tcon->open_file_lock);
1886
1887 if (inv_file) {
1888 rc = cifs_reopen_file(inv_file, false);
1889 if (!rc)
1890 return inv_file;
1891 else {
1892 spin_lock(&tcon->open_file_lock);
1893 list_move_tail(&inv_file->flist,
1894 &cifs_inode->openFileList);
1895 spin_unlock(&tcon->open_file_lock);
1896 cifsFileInfo_put(inv_file);
1897 ++refind;
1898 inv_file = NULL;
1899 spin_lock(&tcon->open_file_lock);
1900 goto refind_writable;
1901 }
1902 }
1903
1904 return NULL;
1905}
1906
1907static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to)
1908{
1909 struct address_space *mapping = page->mapping;
1910 loff_t offset = (loff_t)page->index << PAGE_SHIFT;
1911 char *write_data;
1912 int rc = -EFAULT;
1913 int bytes_written = 0;
1914 struct inode *inode;
1915 struct cifsFileInfo *open_file;
1916
1917 if (!mapping || !mapping->host)
1918 return -EFAULT;
1919
1920 inode = page->mapping->host;
1921
1922 offset += (loff_t)from;
1923 write_data = kmap(page);
1924 write_data += from;
1925
1926 if ((to > PAGE_SIZE) || (from > to)) {
1927 kunmap(page);
1928 return -EIO;
1929 }
1930
1931 /* racing with truncate? */
1932 if (offset > mapping->host->i_size) {
1933 kunmap(page);
1934 return 0; /* don't care */
1935 }
1936
1937 /* check to make sure that we are not extending the file */
1938 if (mapping->host->i_size - offset < (loff_t)to)
1939 to = (unsigned)(mapping->host->i_size - offset);
1940
1941 open_file = find_writable_file(CIFS_I(mapping->host), false);
1942 if (open_file) {
1943 bytes_written = cifs_write(open_file, open_file->pid,
1944 write_data, to - from, &offset);
1945 cifsFileInfo_put(open_file);
1946 /* Does mm or vfs already set times? */
1947 inode->i_atime = inode->i_mtime = current_time(inode);
1948 if ((bytes_written > 0) && (offset))
1949 rc = 0;
1950 else if (bytes_written < 0)
1951 rc = bytes_written;
1952 } else {
1953 cifs_dbg(FYI, "No writeable filehandles for inode\n");
1954 rc = -EIO;
1955 }
1956
1957 kunmap(page);
1958 return rc;
1959}
1960
1961static struct cifs_writedata *
1962wdata_alloc_and_fillpages(pgoff_t tofind, struct address_space *mapping,
1963 pgoff_t end, pgoff_t *index,
1964 unsigned int *found_pages)
1965{
1966 struct cifs_writedata *wdata;
1967
1968 wdata = cifs_writedata_alloc((unsigned int)tofind,
1969 cifs_writev_complete);
1970 if (!wdata)
1971 return NULL;
1972
1973 *found_pages = find_get_pages_range_tag(mapping, index, end,
1974 PAGECACHE_TAG_DIRTY, tofind, wdata->pages);
1975 return wdata;
1976}
1977
1978static unsigned int
1979wdata_prepare_pages(struct cifs_writedata *wdata, unsigned int found_pages,
1980 struct address_space *mapping,
1981 struct writeback_control *wbc,
1982 pgoff_t end, pgoff_t *index, pgoff_t *next, bool *done)
1983{
1984 unsigned int nr_pages = 0, i;
1985 struct page *page;
1986
1987 for (i = 0; i < found_pages; i++) {
1988 page = wdata->pages[i];
1989 /*
1990 * At this point we hold neither the i_pages lock nor the
1991 * page lock: the page may be truncated or invalidated
1992 * (changing page->mapping to NULL), or even swizzled
1993 * back from swapper_space to tmpfs file mapping
1994 */
1995
1996 if (nr_pages == 0)
1997 lock_page(page);
1998 else if (!trylock_page(page))
1999 break;
2000
2001 if (unlikely(page->mapping != mapping)) {
2002 unlock_page(page);
2003 break;
2004 }
2005
2006 if (!wbc->range_cyclic && page->index > end) {
2007 *done = true;
2008 unlock_page(page);
2009 break;
2010 }
2011
2012 if (*next && (page->index != *next)) {
2013 /* Not next consecutive page */
2014 unlock_page(page);
2015 break;
2016 }
2017
2018 if (wbc->sync_mode != WB_SYNC_NONE)
2019 wait_on_page_writeback(page);
2020
2021 if (PageWriteback(page) ||
2022 !clear_page_dirty_for_io(page)) {
2023 unlock_page(page);
2024 break;
2025 }
2026
2027 /*
2028 * This actually clears the dirty bit in the radix tree.
2029 * See cifs_writepage() for more commentary.
2030 */
2031 set_page_writeback(page);
2032 if (page_offset(page) >= i_size_read(mapping->host)) {
2033 *done = true;
2034 unlock_page(page);
2035 end_page_writeback(page);
2036 break;
2037 }
2038
2039 wdata->pages[i] = page;
2040 *next = page->index + 1;
2041 ++nr_pages;
2042 }
2043
2044 /* reset index to refind any pages skipped */
2045 if (nr_pages == 0)
2046 *index = wdata->pages[0]->index + 1;
2047
2048 /* put any pages we aren't going to use */
2049 for (i = nr_pages; i < found_pages; i++) {
2050 put_page(wdata->pages[i]);
2051 wdata->pages[i] = NULL;
2052 }
2053
2054 return nr_pages;
2055}
2056
2057static int
2058wdata_send_pages(struct cifs_writedata *wdata, unsigned int nr_pages,
2059 struct address_space *mapping, struct writeback_control *wbc)
2060{
2061 int rc = 0;
2062 struct TCP_Server_Info *server;
2063 unsigned int i;
2064
2065 wdata->sync_mode = wbc->sync_mode;
2066 wdata->nr_pages = nr_pages;
2067 wdata->offset = page_offset(wdata->pages[0]);
2068 wdata->pagesz = PAGE_SIZE;
2069 wdata->tailsz = min(i_size_read(mapping->host) -
2070 page_offset(wdata->pages[nr_pages - 1]),
2071 (loff_t)PAGE_SIZE);
2072 wdata->bytes = ((nr_pages - 1) * PAGE_SIZE) + wdata->tailsz;
2073
2074 if (wdata->cfile != NULL)
2075 cifsFileInfo_put(wdata->cfile);
2076 wdata->cfile = find_writable_file(CIFS_I(mapping->host), false);
2077 if (!wdata->cfile) {
2078 cifs_dbg(VFS, "No writable handles for inode\n");
2079 rc = -EBADF;
2080 } else {
2081 wdata->pid = wdata->cfile->pid;
2082 server = tlink_tcon(wdata->cfile->tlink)->ses->server;
2083 rc = server->ops->async_writev(wdata, cifs_writedata_release);
2084 }
2085
2086 for (i = 0; i < nr_pages; ++i)
2087 unlock_page(wdata->pages[i]);
2088
2089 return rc;
2090}
2091
2092static int cifs_writepages(struct address_space *mapping,
2093 struct writeback_control *wbc)
2094{
2095 struct cifs_sb_info *cifs_sb = CIFS_SB(mapping->host->i_sb);
2096 struct TCP_Server_Info *server;
2097 bool done = false, scanned = false, range_whole = false;
2098 pgoff_t end, index;
2099 struct cifs_writedata *wdata;
2100 int rc = 0;
2101
2102 /*
2103 * If wsize is smaller than the page cache size, default to writing
2104 * one page at a time via cifs_writepage
2105 */
2106 if (cifs_sb->wsize < PAGE_SIZE)
2107 return generic_writepages(mapping, wbc);
2108
2109 if (wbc->range_cyclic) {
2110 index = mapping->writeback_index; /* Start from prev offset */
2111 end = -1;
2112 } else {
2113 index = wbc->range_start >> PAGE_SHIFT;
2114 end = wbc->range_end >> PAGE_SHIFT;
2115 if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
2116 range_whole = true;
2117 scanned = true;
2118 }
2119 server = cifs_sb_master_tcon(cifs_sb)->ses->server;
2120retry:
2121 while (!done && index <= end) {
2122 unsigned int i, nr_pages, found_pages, wsize, credits;
2123 pgoff_t next = 0, tofind, saved_index = index;
2124
2125 rc = server->ops->wait_mtu_credits(server, cifs_sb->wsize,
2126 &wsize, &credits);
2127 if (rc)
2128 break;
2129
2130 tofind = min((wsize / PAGE_SIZE) - 1, end - index) + 1;
2131
2132 wdata = wdata_alloc_and_fillpages(tofind, mapping, end, &index,
2133 &found_pages);
2134 if (!wdata) {
2135 rc = -ENOMEM;
2136 add_credits_and_wake_if(server, credits, 0);
2137 break;
2138 }
2139
2140 if (found_pages == 0) {
2141 kref_put(&wdata->refcount, cifs_writedata_release);
2142 add_credits_and_wake_if(server, credits, 0);
2143 break;
2144 }
2145
2146 nr_pages = wdata_prepare_pages(wdata, found_pages, mapping, wbc,
2147 end, &index, &next, &done);
2148
2149 /* nothing to write? */
2150 if (nr_pages == 0) {
2151 kref_put(&wdata->refcount, cifs_writedata_release);
2152 add_credits_and_wake_if(server, credits, 0);
2153 continue;
2154 }
2155
2156 wdata->credits = credits;
2157
2158 rc = wdata_send_pages(wdata, nr_pages, mapping, wbc);
2159
2160 /* send failure -- clean up the mess */
2161 if (rc != 0) {
2162 add_credits_and_wake_if(server, wdata->credits, 0);
2163 for (i = 0; i < nr_pages; ++i) {
2164 if (rc == -EAGAIN)
2165 redirty_page_for_writepage(wbc,
2166 wdata->pages[i]);
2167 else
2168 SetPageError(wdata->pages[i]);
2169 end_page_writeback(wdata->pages[i]);
2170 put_page(wdata->pages[i]);
2171 }
2172 if (rc != -EAGAIN)
2173 mapping_set_error(mapping, rc);
2174 }
2175 kref_put(&wdata->refcount, cifs_writedata_release);
2176
2177 if (wbc->sync_mode == WB_SYNC_ALL && rc == -EAGAIN) {
2178 index = saved_index;
2179 continue;
2180 }
2181
2182 wbc->nr_to_write -= nr_pages;
2183 if (wbc->nr_to_write <= 0)
2184 done = true;
2185
2186 index = next;
2187 }
2188
2189 if (!scanned && !done) {
2190 /*
2191 * We hit the last page and there is more work to be done: wrap
2192 * back to the start of the file
2193 */
2194 scanned = true;
2195 index = 0;
2196 goto retry;
2197 }
2198
2199 if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
2200 mapping->writeback_index = index;
2201
2202 return rc;
2203}
2204
2205static int
2206cifs_writepage_locked(struct page *page, struct writeback_control *wbc)
2207{
2208 int rc;
2209 unsigned int xid;
2210
2211 xid = get_xid();
2212/* BB add check for wbc flags */
2213 get_page(page);
2214 if (!PageUptodate(page))
2215 cifs_dbg(FYI, "ppw - page not up to date\n");
2216
2217 /*
2218 * Set the "writeback" flag, and clear "dirty" in the radix tree.
2219 *
2220 * A writepage() implementation always needs to do either this,
2221 * or re-dirty the page with "redirty_page_for_writepage()" in
2222 * the case of a failure.
2223 *
2224 * Just unlocking the page will cause the radix tree tag-bits
2225 * to fail to update with the state of the page correctly.
2226 */
2227 set_page_writeback(page);
2228retry_write:
2229 rc = cifs_partialpagewrite(page, 0, PAGE_SIZE);
2230 if (rc == -EAGAIN) {
2231 if (wbc->sync_mode == WB_SYNC_ALL)
2232 goto retry_write;
2233 redirty_page_for_writepage(wbc, page);
2234 } else if (rc != 0) {
2235 SetPageError(page);
2236 mapping_set_error(page->mapping, rc);
2237 } else {
2238 SetPageUptodate(page);
2239 }
2240 end_page_writeback(page);
2241 put_page(page);
2242 free_xid(xid);
2243 return rc;
2244}
2245
2246static int cifs_writepage(struct page *page, struct writeback_control *wbc)
2247{
2248 int rc = cifs_writepage_locked(page, wbc);
2249 unlock_page(page);
2250 return rc;
2251}
2252
2253static int cifs_write_end(struct file *file, struct address_space *mapping,
2254 loff_t pos, unsigned len, unsigned copied,
2255 struct page *page, void *fsdata)
2256{
2257 int rc;
2258 struct inode *inode = mapping->host;
2259 struct cifsFileInfo *cfile = file->private_data;
2260 struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb);
2261 __u32 pid;
2262
2263 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
2264 pid = cfile->pid;
2265 else
2266 pid = current->tgid;
2267
2268 cifs_dbg(FYI, "write_end for page %p from pos %lld with %d bytes\n",
2269 page, pos, copied);
2270
2271 if (PageChecked(page)) {
2272 if (copied == len)
2273 SetPageUptodate(page);
2274 ClearPageChecked(page);
2275 } else if (!PageUptodate(page) && copied == PAGE_SIZE)
2276 SetPageUptodate(page);
2277
2278 if (!PageUptodate(page)) {
2279 char *page_data;
2280 unsigned offset = pos & (PAGE_SIZE - 1);
2281 unsigned int xid;
2282
2283 xid = get_xid();
2284 /* this is probably better than directly calling
2285 partialpage_write since in this function the file handle is
2286 known which we might as well leverage */
2287 /* BB check if anything else missing out of ppw
2288 such as updating last write time */
2289 page_data = kmap(page);
2290 rc = cifs_write(cfile, pid, page_data + offset, copied, &pos);
2291 /* if (rc < 0) should we set writebehind rc? */
2292 kunmap(page);
2293
2294 free_xid(xid);
2295 } else {
2296 rc = copied;
2297 pos += copied;
2298 set_page_dirty(page);
2299 }
2300
2301 if (rc > 0) {
2302 spin_lock(&inode->i_lock);
2303 if (pos > inode->i_size)
2304 i_size_write(inode, pos);
2305 spin_unlock(&inode->i_lock);
2306 }
2307
2308 unlock_page(page);
2309 put_page(page);
2310
2311 return rc;
2312}
2313
2314int cifs_strict_fsync(struct file *file, loff_t start, loff_t end,
2315 int datasync)
2316{
2317 unsigned int xid;
2318 int rc = 0;
2319 struct cifs_tcon *tcon;
2320 struct TCP_Server_Info *server;
2321 struct cifsFileInfo *smbfile = file->private_data;
2322 struct inode *inode = file_inode(file);
2323 struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
2324
2325 rc = file_write_and_wait_range(file, start, end);
2326 if (rc)
2327 return rc;
2328 inode_lock(inode);
2329
2330 xid = get_xid();
2331
2332 cifs_dbg(FYI, "Sync file - name: %pD datasync: 0x%x\n",
2333 file, datasync);
2334
2335 if (!CIFS_CACHE_READ(CIFS_I(inode))) {
2336 rc = cifs_zap_mapping(inode);
2337 if (rc) {
2338 cifs_dbg(FYI, "rc: %d during invalidate phase\n", rc);
2339 rc = 0; /* don't care about it in fsync */
2340 }
2341 }
2342
2343 tcon = tlink_tcon(smbfile->tlink);
2344 if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) {
2345 server = tcon->ses->server;
2346 if (server->ops->flush)
2347 rc = server->ops->flush(xid, tcon, &smbfile->fid);
2348 else
2349 rc = -ENOSYS;
2350 }
2351
2352 free_xid(xid);
2353 inode_unlock(inode);
2354 return rc;
2355}
2356
2357int cifs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
2358{
2359 unsigned int xid;
2360 int rc = 0;
2361 struct cifs_tcon *tcon;
2362 struct TCP_Server_Info *server;
2363 struct cifsFileInfo *smbfile = file->private_data;
2364 struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file);
2365 struct inode *inode = file->f_mapping->host;
2366
2367 rc = file_write_and_wait_range(file, start, end);
2368 if (rc)
2369 return rc;
2370 inode_lock(inode);
2371
2372 xid = get_xid();
2373
2374 cifs_dbg(FYI, "Sync file - name: %pD datasync: 0x%x\n",
2375 file, datasync);
2376
2377 tcon = tlink_tcon(smbfile->tlink);
2378 if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) {
2379 server = tcon->ses->server;
2380 if (server->ops->flush)
2381 rc = server->ops->flush(xid, tcon, &smbfile->fid);
2382 else
2383 rc = -ENOSYS;
2384 }
2385
2386 free_xid(xid);
2387 inode_unlock(inode);
2388 return rc;
2389}
2390
2391/*
2392 * As file closes, flush all cached write data for this inode checking
2393 * for write behind errors.
2394 */
2395int cifs_flush(struct file *file, fl_owner_t id)
2396{
2397 struct inode *inode = file_inode(file);
2398 int rc = 0;
2399
2400 if (file->f_mode & FMODE_WRITE)
2401 rc = filemap_write_and_wait(inode->i_mapping);
2402
2403 cifs_dbg(FYI, "Flush inode %p file %p rc %d\n", inode, file, rc);
2404
2405 return rc;
2406}
2407
2408static int
2409cifs_write_allocate_pages(struct page **pages, unsigned long num_pages)
2410{
2411 int rc = 0;
2412 unsigned long i;
2413
2414 for (i = 0; i < num_pages; i++) {
2415 pages[i] = alloc_page(GFP_KERNEL|__GFP_HIGHMEM);
2416 if (!pages[i]) {
2417 /*
2418 * save number of pages we have already allocated and
2419 * return with ENOMEM error
2420 */
2421 num_pages = i;
2422 rc = -ENOMEM;
2423 break;
2424 }
2425 }
2426
2427 if (rc) {
2428 for (i = 0; i < num_pages; i++)
2429 put_page(pages[i]);
2430 }
2431 return rc;
2432}
2433
2434static inline
2435size_t get_numpages(const size_t wsize, const size_t len, size_t *cur_len)
2436{
2437 size_t num_pages;
2438 size_t clen;
2439
2440 clen = min_t(const size_t, len, wsize);
2441 num_pages = DIV_ROUND_UP(clen, PAGE_SIZE);
2442
2443 if (cur_len)
2444 *cur_len = clen;
2445
2446 return num_pages;
2447}
2448
2449static void
2450cifs_uncached_writedata_release(struct kref *refcount)
2451{
2452 int i;
2453 struct cifs_writedata *wdata = container_of(refcount,
2454 struct cifs_writedata, refcount);
2455
2456 kref_put(&wdata->ctx->refcount, cifs_aio_ctx_release);
2457 for (i = 0; i < wdata->nr_pages; i++)
2458 put_page(wdata->pages[i]);
2459 cifs_writedata_release(refcount);
2460}
2461
2462static void collect_uncached_write_data(struct cifs_aio_ctx *ctx);
2463
2464static void
2465cifs_uncached_writev_complete(struct work_struct *work)
2466{
2467 struct cifs_writedata *wdata = container_of(work,
2468 struct cifs_writedata, work);
2469 struct inode *inode = d_inode(wdata->cfile->dentry);
2470 struct cifsInodeInfo *cifsi = CIFS_I(inode);
2471
2472 spin_lock(&inode->i_lock);
2473 cifs_update_eof(cifsi, wdata->offset, wdata->bytes);
2474 if (cifsi->server_eof > inode->i_size)
2475 i_size_write(inode, cifsi->server_eof);
2476 spin_unlock(&inode->i_lock);
2477
2478 complete(&wdata->done);
2479 collect_uncached_write_data(wdata->ctx);
2480 /* the below call can possibly free the last ref to aio ctx */
2481 kref_put(&wdata->refcount, cifs_uncached_writedata_release);
2482}
2483
2484static int
2485wdata_fill_from_iovec(struct cifs_writedata *wdata, struct iov_iter *from,
2486 size_t *len, unsigned long *num_pages)
2487{
2488 size_t save_len, copied, bytes, cur_len = *len;
2489 unsigned long i, nr_pages = *num_pages;
2490
2491 save_len = cur_len;
2492 for (i = 0; i < nr_pages; i++) {
2493 bytes = min_t(const size_t, cur_len, PAGE_SIZE);
2494 copied = copy_page_from_iter(wdata->pages[i], 0, bytes, from);
2495 cur_len -= copied;
2496 /*
2497 * If we didn't copy as much as we expected, then that
2498 * may mean we trod into an unmapped area. Stop copying
2499 * at that point. On the next pass through the big
2500 * loop, we'll likely end up getting a zero-length
2501 * write and bailing out of it.
2502 */
2503 if (copied < bytes)
2504 break;
2505 }
2506 cur_len = save_len - cur_len;
2507 *len = cur_len;
2508
2509 /*
2510 * If we have no data to send, then that probably means that
2511 * the copy above failed altogether. That's most likely because
2512 * the address in the iovec was bogus. Return -EFAULT and let
2513 * the caller free anything we allocated and bail out.
2514 */
2515 if (!cur_len)
2516 return -EFAULT;
2517
2518 /*
2519 * i + 1 now represents the number of pages we actually used in
2520 * the copy phase above.
2521 */
2522 *num_pages = i + 1;
2523 return 0;
2524}
2525
2526static int
2527cifs_write_from_iter(loff_t offset, size_t len, struct iov_iter *from,
2528 struct cifsFileInfo *open_file,
2529 struct cifs_sb_info *cifs_sb, struct list_head *wdata_list,
2530 struct cifs_aio_ctx *ctx)
2531{
2532 int rc = 0;
2533 size_t cur_len;
2534 unsigned long nr_pages, num_pages, i;
2535 struct cifs_writedata *wdata;
2536 struct iov_iter saved_from = *from;
2537 loff_t saved_offset = offset;
2538 pid_t pid;
2539 struct TCP_Server_Info *server;
2540
2541 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
2542 pid = open_file->pid;
2543 else
2544 pid = current->tgid;
2545
2546 server = tlink_tcon(open_file->tlink)->ses->server;
2547
2548 do {
2549 unsigned int wsize, credits;
2550
2551 rc = server->ops->wait_mtu_credits(server, cifs_sb->wsize,
2552 &wsize, &credits);
2553 if (rc)
2554 break;
2555
2556 nr_pages = get_numpages(wsize, len, &cur_len);
2557 wdata = cifs_writedata_alloc(nr_pages,
2558 cifs_uncached_writev_complete);
2559 if (!wdata) {
2560 rc = -ENOMEM;
2561 add_credits_and_wake_if(server, credits, 0);
2562 break;
2563 }
2564
2565 rc = cifs_write_allocate_pages(wdata->pages, nr_pages);
2566 if (rc) {
2567 kfree(wdata);
2568 add_credits_and_wake_if(server, credits, 0);
2569 break;
2570 }
2571
2572 num_pages = nr_pages;
2573 rc = wdata_fill_from_iovec(wdata, from, &cur_len, &num_pages);
2574 if (rc) {
2575 for (i = 0; i < nr_pages; i++)
2576 put_page(wdata->pages[i]);
2577 kfree(wdata);
2578 add_credits_and_wake_if(server, credits, 0);
2579 break;
2580 }
2581
2582 /*
2583 * Bring nr_pages down to the number of pages we actually used,
2584 * and free any pages that we didn't use.
2585 */
2586 for ( ; nr_pages > num_pages; nr_pages--)
2587 put_page(wdata->pages[nr_pages - 1]);
2588
2589 wdata->sync_mode = WB_SYNC_ALL;
2590 wdata->nr_pages = nr_pages;
2591 wdata->offset = (__u64)offset;
2592 wdata->cfile = cifsFileInfo_get(open_file);
2593 wdata->pid = pid;
2594 wdata->bytes = cur_len;
2595 wdata->pagesz = PAGE_SIZE;
2596 wdata->tailsz = cur_len - ((nr_pages - 1) * PAGE_SIZE);
2597 wdata->credits = credits;
2598 wdata->ctx = ctx;
2599 kref_get(&ctx->refcount);
2600
2601 if (!wdata->cfile->invalidHandle ||
2602 !(rc = cifs_reopen_file(wdata->cfile, false)))
2603 rc = server->ops->async_writev(wdata,
2604 cifs_uncached_writedata_release);
2605 if (rc) {
2606 add_credits_and_wake_if(server, wdata->credits, 0);
2607 kref_put(&wdata->refcount,
2608 cifs_uncached_writedata_release);
2609 if (rc == -EAGAIN) {
2610 *from = saved_from;
2611 iov_iter_advance(from, offset - saved_offset);
2612 continue;
2613 }
2614 break;
2615 }
2616
2617 list_add_tail(&wdata->list, wdata_list);
2618 offset += cur_len;
2619 len -= cur_len;
2620 } while (len > 0);
2621
2622 return rc;
2623}
2624
2625static void collect_uncached_write_data(struct cifs_aio_ctx *ctx)
2626{
2627 struct cifs_writedata *wdata, *tmp;
2628 struct cifs_tcon *tcon;
2629 struct cifs_sb_info *cifs_sb;
2630 struct dentry *dentry = ctx->cfile->dentry;
2631 unsigned int i;
2632 int rc;
2633
2634 tcon = tlink_tcon(ctx->cfile->tlink);
2635 cifs_sb = CIFS_SB(dentry->d_sb);
2636
2637 mutex_lock(&ctx->aio_mutex);
2638
2639 if (list_empty(&ctx->list)) {
2640 mutex_unlock(&ctx->aio_mutex);
2641 return;
2642 }
2643
2644 rc = ctx->rc;
2645 /*
2646 * Wait for and collect replies for any successful sends in order of
2647 * increasing offset. Once an error is hit, then return without waiting
2648 * for any more replies.
2649 */
2650restart_loop:
2651 list_for_each_entry_safe(wdata, tmp, &ctx->list, list) {
2652 if (!rc) {
2653 if (!try_wait_for_completion(&wdata->done)) {
2654 mutex_unlock(&ctx->aio_mutex);
2655 return;
2656 }
2657
2658 if (wdata->result)
2659 rc = wdata->result;
2660 else
2661 ctx->total_len += wdata->bytes;
2662
2663 /* resend call if it's a retryable error */
2664 if (rc == -EAGAIN) {
2665 struct list_head tmp_list;
2666 struct iov_iter tmp_from = ctx->iter;
2667
2668 INIT_LIST_HEAD(&tmp_list);
2669 list_del_init(&wdata->list);
2670
2671 iov_iter_advance(&tmp_from,
2672 wdata->offset - ctx->pos);
2673
2674 rc = cifs_write_from_iter(wdata->offset,
2675 wdata->bytes, &tmp_from,
2676 ctx->cfile, cifs_sb, &tmp_list,
2677 ctx);
2678
2679 list_splice(&tmp_list, &ctx->list);
2680
2681 kref_put(&wdata->refcount,
2682 cifs_uncached_writedata_release);
2683 goto restart_loop;
2684 }
2685 }
2686 list_del_init(&wdata->list);
2687 kref_put(&wdata->refcount, cifs_uncached_writedata_release);
2688 }
2689
2690 for (i = 0; i < ctx->npages; i++)
2691 put_page(ctx->bv[i].bv_page);
2692
2693 cifs_stats_bytes_written(tcon, ctx->total_len);
2694 set_bit(CIFS_INO_INVALID_MAPPING, &CIFS_I(dentry->d_inode)->flags);
2695
2696 ctx->rc = (rc == 0) ? ctx->total_len : rc;
2697
2698 mutex_unlock(&ctx->aio_mutex);
2699
2700 if (ctx->iocb && ctx->iocb->ki_complete)
2701 ctx->iocb->ki_complete(ctx->iocb, ctx->rc, 0);
2702 else
2703 complete(&ctx->done);
2704}
2705
2706ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from)
2707{
2708 struct file *file = iocb->ki_filp;
2709 ssize_t total_written = 0;
2710 struct cifsFileInfo *cfile;
2711 struct cifs_tcon *tcon;
2712 struct cifs_sb_info *cifs_sb;
2713 struct cifs_aio_ctx *ctx;
2714 struct iov_iter saved_from = *from;
2715 int rc;
2716
2717 /*
2718 * BB - optimize the way when signing is disabled. We can drop this
2719 * extra memory-to-memory copying and use iovec buffers for constructing
2720 * write request.
2721 */
2722
2723 rc = generic_write_checks(iocb, from);
2724 if (rc <= 0)
2725 return rc;
2726
2727 cifs_sb = CIFS_FILE_SB(file);
2728 cfile = file->private_data;
2729 tcon = tlink_tcon(cfile->tlink);
2730
2731 if (!tcon->ses->server->ops->async_writev)
2732 return -ENOSYS;
2733
2734 ctx = cifs_aio_ctx_alloc();
2735 if (!ctx)
2736 return -ENOMEM;
2737
2738 ctx->cfile = cifsFileInfo_get(cfile);
2739
2740 if (!is_sync_kiocb(iocb))
2741 ctx->iocb = iocb;
2742
2743 ctx->pos = iocb->ki_pos;
2744
2745 rc = setup_aio_ctx_iter(ctx, from, WRITE);
2746 if (rc) {
2747 kref_put(&ctx->refcount, cifs_aio_ctx_release);
2748 return rc;
2749 }
2750
2751 /* grab a lock here due to read response handlers can access ctx */
2752 mutex_lock(&ctx->aio_mutex);
2753
2754 rc = cifs_write_from_iter(iocb->ki_pos, ctx->len, &saved_from,
2755 cfile, cifs_sb, &ctx->list, ctx);
2756
2757 /*
2758 * If at least one write was successfully sent, then discard any rc
2759 * value from the later writes. If the other write succeeds, then
2760 * we'll end up returning whatever was written. If it fails, then
2761 * we'll get a new rc value from that.
2762 */
2763 if (!list_empty(&ctx->list))
2764 rc = 0;
2765
2766 mutex_unlock(&ctx->aio_mutex);
2767
2768 if (rc) {
2769 kref_put(&ctx->refcount, cifs_aio_ctx_release);
2770 return rc;
2771 }
2772
2773 if (!is_sync_kiocb(iocb)) {
2774 kref_put(&ctx->refcount, cifs_aio_ctx_release);
2775 return -EIOCBQUEUED;
2776 }
2777
2778 rc = wait_for_completion_killable(&ctx->done);
2779 if (rc) {
2780 mutex_lock(&ctx->aio_mutex);
2781 ctx->rc = rc = -EINTR;
2782 total_written = ctx->total_len;
2783 mutex_unlock(&ctx->aio_mutex);
2784 } else {
2785 rc = ctx->rc;
2786 total_written = ctx->total_len;
2787 }
2788
2789 kref_put(&ctx->refcount, cifs_aio_ctx_release);
2790
2791 if (unlikely(!total_written))
2792 return rc;
2793
2794 iocb->ki_pos += total_written;
2795 return total_written;
2796}
2797
2798static ssize_t
2799cifs_writev(struct kiocb *iocb, struct iov_iter *from)
2800{
2801 struct file *file = iocb->ki_filp;
2802 struct cifsFileInfo *cfile = (struct cifsFileInfo *)file->private_data;
2803 struct inode *inode = file->f_mapping->host;
2804 struct cifsInodeInfo *cinode = CIFS_I(inode);
2805 struct TCP_Server_Info *server = tlink_tcon(cfile->tlink)->ses->server;
2806 ssize_t rc;
2807
2808 inode_lock(inode);
2809 /*
2810 * We need to hold the sem to be sure nobody modifies lock list
2811 * with a brlock that prevents writing.
2812 */
2813 down_read(&cinode->lock_sem);
2814
2815 rc = generic_write_checks(iocb, from);
2816 if (rc <= 0)
2817 goto out;
2818
2819 if (!cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(from),
2820 server->vals->exclusive_lock_type, NULL,
2821 CIFS_WRITE_OP))
2822 rc = __generic_file_write_iter(iocb, from);
2823 else
2824 rc = -EACCES;
2825out:
2826 up_read(&cinode->lock_sem);
2827 inode_unlock(inode);
2828
2829 if (rc > 0)
2830 rc = generic_write_sync(iocb, rc);
2831 return rc;
2832}
2833
2834ssize_t
2835cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from)
2836{
2837 struct inode *inode = file_inode(iocb->ki_filp);
2838 struct cifsInodeInfo *cinode = CIFS_I(inode);
2839 struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
2840 struct cifsFileInfo *cfile = (struct cifsFileInfo *)
2841 iocb->ki_filp->private_data;
2842 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
2843 ssize_t written;
2844
2845 written = cifs_get_writer(cinode);
2846 if (written)
2847 return written;
2848
2849 if (CIFS_CACHE_WRITE(cinode)) {
2850 if (cap_unix(tcon->ses) &&
2851 (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability))
2852 && ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) {
2853 written = generic_file_write_iter(iocb, from);
2854 goto out;
2855 }
2856 written = cifs_writev(iocb, from);
2857 goto out;
2858 }
2859 /*
2860 * For non-oplocked files in strict cache mode we need to write the data
2861 * to the server exactly from the pos to pos+len-1 rather than flush all
2862 * affected pages because it may cause a error with mandatory locks on
2863 * these pages but not on the region from pos to ppos+len-1.
2864 */
2865 written = cifs_user_writev(iocb, from);
2866 if (written > 0 && CIFS_CACHE_READ(cinode)) {
2867 /*
2868 * Windows 7 server can delay breaking level2 oplock if a write
2869 * request comes - break it on the client to prevent reading
2870 * an old data.
2871 */
2872 cifs_zap_mapping(inode);
2873 cifs_dbg(FYI, "Set no oplock for inode=%p after a write operation\n",
2874 inode);
2875 cinode->oplock = 0;
2876 }
2877out:
2878 cifs_put_writer(cinode);
2879 return written;
2880}
2881
2882static struct cifs_readdata *
2883cifs_readdata_alloc(unsigned int nr_pages, work_func_t complete)
2884{
2885 struct cifs_readdata *rdata;
2886
2887 rdata = kzalloc(sizeof(*rdata) + (sizeof(struct page *) * nr_pages),
2888 GFP_KERNEL);
2889 if (rdata != NULL) {
2890 kref_init(&rdata->refcount);
2891 INIT_LIST_HEAD(&rdata->list);
2892 init_completion(&rdata->done);
2893 INIT_WORK(&rdata->work, complete);
2894 }
2895
2896 return rdata;
2897}
2898
2899void
2900cifs_readdata_release(struct kref *refcount)
2901{
2902 struct cifs_readdata *rdata = container_of(refcount,
2903 struct cifs_readdata, refcount);
2904#ifdef CONFIG_CIFS_SMB_DIRECT
2905 if (rdata->mr) {
2906 smbd_deregister_mr(rdata->mr);
2907 rdata->mr = NULL;
2908 }
2909#endif
2910 if (rdata->cfile)
2911 cifsFileInfo_put(rdata->cfile);
2912
2913 kfree(rdata);
2914}
2915
2916static int
2917cifs_read_allocate_pages(struct cifs_readdata *rdata, unsigned int nr_pages)
2918{
2919 int rc = 0;
2920 struct page *page;
2921 unsigned int i;
2922
2923 for (i = 0; i < nr_pages; i++) {
2924 page = alloc_page(GFP_KERNEL|__GFP_HIGHMEM);
2925 if (!page) {
2926 rc = -ENOMEM;
2927 break;
2928 }
2929 rdata->pages[i] = page;
2930 }
2931
2932 if (rc) {
2933 for (i = 0; i < nr_pages; i++) {
2934 put_page(rdata->pages[i]);
2935 rdata->pages[i] = NULL;
2936 }
2937 }
2938 return rc;
2939}
2940
2941static void
2942cifs_uncached_readdata_release(struct kref *refcount)
2943{
2944 struct cifs_readdata *rdata = container_of(refcount,
2945 struct cifs_readdata, refcount);
2946 unsigned int i;
2947
2948 kref_put(&rdata->ctx->refcount, cifs_aio_ctx_release);
2949 for (i = 0; i < rdata->nr_pages; i++) {
2950 put_page(rdata->pages[i]);
2951 rdata->pages[i] = NULL;
2952 }
2953 cifs_readdata_release(refcount);
2954}
2955
2956/**
2957 * cifs_readdata_to_iov - copy data from pages in response to an iovec
2958 * @rdata: the readdata response with list of pages holding data
2959 * @iter: destination for our data
2960 *
2961 * This function copies data from a list of pages in a readdata response into
2962 * an array of iovecs. It will first calculate where the data should go
2963 * based on the info in the readdata and then copy the data into that spot.
2964 */
2965static int
2966cifs_readdata_to_iov(struct cifs_readdata *rdata, struct iov_iter *iter)
2967{
2968 size_t remaining = rdata->got_bytes;
2969 unsigned int i;
2970
2971 for (i = 0; i < rdata->nr_pages; i++) {
2972 struct page *page = rdata->pages[i];
2973 size_t copy = min_t(size_t, remaining, PAGE_SIZE);
2974 size_t written;
2975
2976 if (unlikely(iter->type & ITER_PIPE)) {
2977 void *addr = kmap_atomic(page);
2978
2979 written = copy_to_iter(addr, copy, iter);
2980 kunmap_atomic(addr);
2981 } else
2982 written = copy_page_to_iter(page, 0, copy, iter);
2983 remaining -= written;
2984 if (written < copy && iov_iter_count(iter) > 0)
2985 break;
2986 }
2987 return remaining ? -EFAULT : 0;
2988}
2989
2990static void collect_uncached_read_data(struct cifs_aio_ctx *ctx);
2991
2992static void
2993cifs_uncached_readv_complete(struct work_struct *work)
2994{
2995 struct cifs_readdata *rdata = container_of(work,
2996 struct cifs_readdata, work);
2997
2998 complete(&rdata->done);
2999 collect_uncached_read_data(rdata->ctx);
3000 /* the below call can possibly free the last ref to aio ctx */
3001 kref_put(&rdata->refcount, cifs_uncached_readdata_release);
3002}
3003
3004static int
3005uncached_fill_pages(struct TCP_Server_Info *server,
3006 struct cifs_readdata *rdata, struct iov_iter *iter,
3007 unsigned int len)
3008{
3009 int result = 0;
3010 unsigned int i;
3011 unsigned int nr_pages = rdata->nr_pages;
3012
3013 rdata->got_bytes = 0;
3014 rdata->tailsz = PAGE_SIZE;
3015 for (i = 0; i < nr_pages; i++) {
3016 struct page *page = rdata->pages[i];
3017 size_t n;
3018
3019 if (len <= 0) {
3020 /* no need to hold page hostage */
3021 rdata->pages[i] = NULL;
3022 rdata->nr_pages--;
3023 put_page(page);
3024 continue;
3025 }
3026 n = len;
3027 if (len >= PAGE_SIZE) {
3028 /* enough data to fill the page */
3029 n = PAGE_SIZE;
3030 len -= n;
3031 } else {
3032 zero_user(page, len, PAGE_SIZE - len);
3033 rdata->tailsz = len;
3034 len = 0;
3035 }
3036 if (iter)
3037 result = copy_page_from_iter(page, 0, n, iter);
3038#ifdef CONFIG_CIFS_SMB_DIRECT
3039 else if (rdata->mr)
3040 result = n;
3041#endif
3042 else
3043 result = cifs_read_page_from_socket(server, page, n);
3044 if (result < 0)
3045 break;
3046
3047 rdata->got_bytes += result;
3048 }
3049
3050 return rdata->got_bytes > 0 && result != -ECONNABORTED ?
3051 rdata->got_bytes : result;
3052}
3053
3054static int
3055cifs_uncached_read_into_pages(struct TCP_Server_Info *server,
3056 struct cifs_readdata *rdata, unsigned int len)
3057{
3058 return uncached_fill_pages(server, rdata, NULL, len);
3059}
3060
3061static int
3062cifs_uncached_copy_into_pages(struct TCP_Server_Info *server,
3063 struct cifs_readdata *rdata,
3064 struct iov_iter *iter)
3065{
3066 return uncached_fill_pages(server, rdata, iter, iter->count);
3067}
3068
3069static int
3070cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file,
3071 struct cifs_sb_info *cifs_sb, struct list_head *rdata_list,
3072 struct cifs_aio_ctx *ctx)
3073{
3074 struct cifs_readdata *rdata;
3075 unsigned int npages, rsize, credits;
3076 size_t cur_len;
3077 int rc;
3078 pid_t pid;
3079 struct TCP_Server_Info *server;
3080
3081 server = tlink_tcon(open_file->tlink)->ses->server;
3082
3083 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
3084 pid = open_file->pid;
3085 else
3086 pid = current->tgid;
3087
3088 do {
3089 rc = server->ops->wait_mtu_credits(server, cifs_sb->rsize,
3090 &rsize, &credits);
3091 if (rc)
3092 break;
3093
3094 cur_len = min_t(const size_t, len, rsize);
3095 npages = DIV_ROUND_UP(cur_len, PAGE_SIZE);
3096
3097 /* allocate a readdata struct */
3098 rdata = cifs_readdata_alloc(npages,
3099 cifs_uncached_readv_complete);
3100 if (!rdata) {
3101 add_credits_and_wake_if(server, credits, 0);
3102 rc = -ENOMEM;
3103 break;
3104 }
3105
3106 rc = cifs_read_allocate_pages(rdata, npages);
3107 if (rc)
3108 goto error;
3109
3110 rdata->cfile = cifsFileInfo_get(open_file);
3111 rdata->nr_pages = npages;
3112 rdata->offset = offset;
3113 rdata->bytes = cur_len;
3114 rdata->pid = pid;
3115 rdata->pagesz = PAGE_SIZE;
3116 rdata->read_into_pages = cifs_uncached_read_into_pages;
3117 rdata->copy_into_pages = cifs_uncached_copy_into_pages;
3118 rdata->credits = credits;
3119 rdata->ctx = ctx;
3120 kref_get(&ctx->refcount);
3121
3122 if (!rdata->cfile->invalidHandle ||
3123 !(rc = cifs_reopen_file(rdata->cfile, true)))
3124 rc = server->ops->async_readv(rdata);
3125error:
3126 if (rc) {
3127 add_credits_and_wake_if(server, rdata->credits, 0);
3128 kref_put(&rdata->refcount,
3129 cifs_uncached_readdata_release);
3130 if (rc == -EAGAIN)
3131 continue;
3132 break;
3133 }
3134
3135 list_add_tail(&rdata->list, rdata_list);
3136 offset += cur_len;
3137 len -= cur_len;
3138 } while (len > 0);
3139
3140 return rc;
3141}
3142
3143static void
3144collect_uncached_read_data(struct cifs_aio_ctx *ctx)
3145{
3146 struct cifs_readdata *rdata, *tmp;
3147 struct iov_iter *to = &ctx->iter;
3148 struct cifs_sb_info *cifs_sb;
3149 struct cifs_tcon *tcon;
3150 unsigned int i;
3151 int rc;
3152
3153 tcon = tlink_tcon(ctx->cfile->tlink);
3154 cifs_sb = CIFS_SB(ctx->cfile->dentry->d_sb);
3155
3156 mutex_lock(&ctx->aio_mutex);
3157
3158 if (list_empty(&ctx->list)) {
3159 mutex_unlock(&ctx->aio_mutex);
3160 return;
3161 }
3162
3163 rc = ctx->rc;
3164 /* the loop below should proceed in the order of increasing offsets */
3165again:
3166 list_for_each_entry_safe(rdata, tmp, &ctx->list, list) {
3167 if (!rc) {
3168 if (!try_wait_for_completion(&rdata->done)) {
3169 mutex_unlock(&ctx->aio_mutex);
3170 return;
3171 }
3172
3173 if (rdata->result == -EAGAIN) {
3174 /* resend call if it's a retryable error */
3175 struct list_head tmp_list;
3176 unsigned int got_bytes = rdata->got_bytes;
3177
3178 list_del_init(&rdata->list);
3179 INIT_LIST_HEAD(&tmp_list);
3180
3181 /*
3182 * Got a part of data and then reconnect has
3183 * happened -- fill the buffer and continue
3184 * reading.
3185 */
3186 if (got_bytes && got_bytes < rdata->bytes) {
3187 rc = cifs_readdata_to_iov(rdata, to);
3188 if (rc) {
3189 kref_put(&rdata->refcount,
3190 cifs_uncached_readdata_release);
3191 continue;
3192 }
3193 }
3194
3195 rc = cifs_send_async_read(
3196 rdata->offset + got_bytes,
3197 rdata->bytes - got_bytes,
3198 rdata->cfile, cifs_sb,
3199 &tmp_list, ctx);
3200
3201 list_splice(&tmp_list, &ctx->list);
3202
3203 kref_put(&rdata->refcount,
3204 cifs_uncached_readdata_release);
3205 goto again;
3206 } else if (rdata->result)
3207 rc = rdata->result;
3208 else
3209 rc = cifs_readdata_to_iov(rdata, to);
3210
3211 /* if there was a short read -- discard anything left */
3212 if (rdata->got_bytes && rdata->got_bytes < rdata->bytes)
3213 rc = -ENODATA;
3214 }
3215 list_del_init(&rdata->list);
3216 kref_put(&rdata->refcount, cifs_uncached_readdata_release);
3217 }
3218
3219 for (i = 0; i < ctx->npages; i++) {
3220 if (ctx->should_dirty)
3221 set_page_dirty(ctx->bv[i].bv_page);
3222 put_page(ctx->bv[i].bv_page);
3223 }
3224
3225 ctx->total_len = ctx->len - iov_iter_count(to);
3226
3227 cifs_stats_bytes_read(tcon, ctx->total_len);
3228
3229 /* mask nodata case */
3230 if (rc == -ENODATA)
3231 rc = 0;
3232
3233 ctx->rc = (rc == 0) ? ctx->total_len : rc;
3234
3235 mutex_unlock(&ctx->aio_mutex);
3236
3237 if (ctx->iocb && ctx->iocb->ki_complete)
3238 ctx->iocb->ki_complete(ctx->iocb, ctx->rc, 0);
3239 else
3240 complete(&ctx->done);
3241}
3242
3243ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to)
3244{
3245 struct file *file = iocb->ki_filp;
3246 ssize_t rc;
3247 size_t len;
3248 ssize_t total_read = 0;
3249 loff_t offset = iocb->ki_pos;
3250 struct cifs_sb_info *cifs_sb;
3251 struct cifs_tcon *tcon;
3252 struct cifsFileInfo *cfile;
3253 struct cifs_aio_ctx *ctx;
3254
3255 len = iov_iter_count(to);
3256 if (!len)
3257 return 0;
3258
3259 cifs_sb = CIFS_FILE_SB(file);
3260 cfile = file->private_data;
3261 tcon = tlink_tcon(cfile->tlink);
3262
3263 if (!tcon->ses->server->ops->async_readv)
3264 return -ENOSYS;
3265
3266 if ((file->f_flags & O_ACCMODE) == O_WRONLY)
3267 cifs_dbg(FYI, "attempting read on write only file instance\n");
3268
3269 ctx = cifs_aio_ctx_alloc();
3270 if (!ctx)
3271 return -ENOMEM;
3272
3273 ctx->cfile = cifsFileInfo_get(cfile);
3274
3275 if (!is_sync_kiocb(iocb))
3276 ctx->iocb = iocb;
3277
3278 if (to->type == ITER_IOVEC)
3279 ctx->should_dirty = true;
3280
3281 rc = setup_aio_ctx_iter(ctx, to, READ);
3282 if (rc) {
3283 kref_put(&ctx->refcount, cifs_aio_ctx_release);
3284 return rc;
3285 }
3286
3287 len = ctx->len;
3288
3289 /* grab a lock here due to read response handlers can access ctx */
3290 mutex_lock(&ctx->aio_mutex);
3291
3292 rc = cifs_send_async_read(offset, len, cfile, cifs_sb, &ctx->list, ctx);
3293
3294 /* if at least one read request send succeeded, then reset rc */
3295 if (!list_empty(&ctx->list))
3296 rc = 0;
3297
3298 mutex_unlock(&ctx->aio_mutex);
3299
3300 if (rc) {
3301 kref_put(&ctx->refcount, cifs_aio_ctx_release);
3302 return rc;
3303 }
3304
3305 if (!is_sync_kiocb(iocb)) {
3306 kref_put(&ctx->refcount, cifs_aio_ctx_release);
3307 return -EIOCBQUEUED;
3308 }
3309
3310 rc = wait_for_completion_killable(&ctx->done);
3311 if (rc) {
3312 mutex_lock(&ctx->aio_mutex);
3313 ctx->rc = rc = -EINTR;
3314 total_read = ctx->total_len;
3315 mutex_unlock(&ctx->aio_mutex);
3316 } else {
3317 rc = ctx->rc;
3318 total_read = ctx->total_len;
3319 }
3320
3321 kref_put(&ctx->refcount, cifs_aio_ctx_release);
3322
3323 if (total_read) {
3324 iocb->ki_pos += total_read;
3325 return total_read;
3326 }
3327 return rc;
3328}
3329
3330ssize_t
3331cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to)
3332{
3333 struct inode *inode = file_inode(iocb->ki_filp);
3334 struct cifsInodeInfo *cinode = CIFS_I(inode);
3335 struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
3336 struct cifsFileInfo *cfile = (struct cifsFileInfo *)
3337 iocb->ki_filp->private_data;
3338 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
3339 int rc = -EACCES;
3340
3341 /*
3342 * In strict cache mode we need to read from the server all the time
3343 * if we don't have level II oplock because the server can delay mtime
3344 * change - so we can't make a decision about inode invalidating.
3345 * And we can also fail with pagereading if there are mandatory locks
3346 * on pages affected by this read but not on the region from pos to
3347 * pos+len-1.
3348 */
3349 if (!CIFS_CACHE_READ(cinode))
3350 return cifs_user_readv(iocb, to);
3351
3352 if (cap_unix(tcon->ses) &&
3353 (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) &&
3354 ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0))
3355 return generic_file_read_iter(iocb, to);
3356
3357 /*
3358 * We need to hold the sem to be sure nobody modifies lock list
3359 * with a brlock that prevents reading.
3360 */
3361 down_read(&cinode->lock_sem);
3362 if (!cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(to),
3363 tcon->ses->server->vals->shared_lock_type,
3364 NULL, CIFS_READ_OP))
3365 rc = generic_file_read_iter(iocb, to);
3366 up_read(&cinode->lock_sem);
3367 return rc;
3368}
3369
3370static ssize_t
3371cifs_read(struct file *file, char *read_data, size_t read_size, loff_t *offset)
3372{
3373 int rc = -EACCES;
3374 unsigned int bytes_read = 0;
3375 unsigned int total_read;
3376 unsigned int current_read_size;
3377 unsigned int rsize;
3378 struct cifs_sb_info *cifs_sb;
3379 struct cifs_tcon *tcon;
3380 struct TCP_Server_Info *server;
3381 unsigned int xid;
3382 char *cur_offset;
3383 struct cifsFileInfo *open_file;
3384 struct cifs_io_parms io_parms;
3385 int buf_type = CIFS_NO_BUFFER;
3386 __u32 pid;
3387
3388 xid = get_xid();
3389 cifs_sb = CIFS_FILE_SB(file);
3390
3391 /* FIXME: set up handlers for larger reads and/or convert to async */
3392 rsize = min_t(unsigned int, cifs_sb->rsize, CIFSMaxBufSize);
3393
3394 if (file->private_data == NULL) {
3395 rc = -EBADF;
3396 free_xid(xid);
3397 return rc;
3398 }
3399 open_file = file->private_data;
3400 tcon = tlink_tcon(open_file->tlink);
3401 server = tcon->ses->server;
3402
3403 if (!server->ops->sync_read) {
3404 free_xid(xid);
3405 return -ENOSYS;
3406 }
3407
3408 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
3409 pid = open_file->pid;
3410 else
3411 pid = current->tgid;
3412
3413 if ((file->f_flags & O_ACCMODE) == O_WRONLY)
3414 cifs_dbg(FYI, "attempting read on write only file instance\n");
3415
3416 for (total_read = 0, cur_offset = read_data; read_size > total_read;
3417 total_read += bytes_read, cur_offset += bytes_read) {
3418 do {
3419 current_read_size = min_t(uint, read_size - total_read,
3420 rsize);
3421 /*
3422 * For windows me and 9x we do not want to request more
3423 * than it negotiated since it will refuse the read
3424 * then.
3425 */
3426 if ((tcon->ses) && !(tcon->ses->capabilities &
3427 tcon->ses->server->vals->cap_large_files)) {
3428 current_read_size = min_t(uint,
3429 current_read_size, CIFSMaxBufSize);
3430 }
3431 if (open_file->invalidHandle) {
3432 rc = cifs_reopen_file(open_file, true);
3433 if (rc != 0)
3434 break;
3435 }
3436 io_parms.pid = pid;
3437 io_parms.tcon = tcon;
3438 io_parms.offset = *offset;
3439 io_parms.length = current_read_size;
3440 rc = server->ops->sync_read(xid, &open_file->fid, &io_parms,
3441 &bytes_read, &cur_offset,
3442 &buf_type);
3443 } while (rc == -EAGAIN);
3444
3445 if (rc || (bytes_read == 0)) {
3446 if (total_read) {
3447 break;
3448 } else {
3449 free_xid(xid);
3450 return rc;
3451 }
3452 } else {
3453 cifs_stats_bytes_read(tcon, total_read);
3454 *offset += bytes_read;
3455 }
3456 }
3457 free_xid(xid);
3458 return total_read;
3459}
3460
3461/*
3462 * If the page is mmap'ed into a process' page tables, then we need to make
3463 * sure that it doesn't change while being written back.
3464 */
3465static vm_fault_t
3466cifs_page_mkwrite(struct vm_fault *vmf)
3467{
3468 struct page *page = vmf->page;
3469
3470 lock_page(page);
3471 return VM_FAULT_LOCKED;
3472}
3473
3474static const struct vm_operations_struct cifs_file_vm_ops = {
3475 .fault = filemap_fault,
3476 .map_pages = filemap_map_pages,
3477 .page_mkwrite = cifs_page_mkwrite,
3478};
3479
3480int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma)
3481{
3482 int xid, rc = 0;
3483 struct inode *inode = file_inode(file);
3484
3485 xid = get_xid();
3486
3487 if (!CIFS_CACHE_READ(CIFS_I(inode)))
3488 rc = cifs_zap_mapping(inode);
3489 if (!rc)
3490 rc = generic_file_mmap(file, vma);
3491 if (!rc)
3492 vma->vm_ops = &cifs_file_vm_ops;
3493
3494 free_xid(xid);
3495 return rc;
3496}
3497
3498int cifs_file_mmap(struct file *file, struct vm_area_struct *vma)
3499{
3500 int rc, xid;
3501
3502 xid = get_xid();
3503
3504 rc = cifs_revalidate_file(file);
3505 if (rc)
3506 cifs_dbg(FYI, "Validation prior to mmap failed, error=%d\n",
3507 rc);
3508 if (!rc)
3509 rc = generic_file_mmap(file, vma);
3510 if (!rc)
3511 vma->vm_ops = &cifs_file_vm_ops;
3512
3513 free_xid(xid);
3514 return rc;
3515}
3516
3517static void
3518cifs_readv_complete(struct work_struct *work)
3519{
3520 unsigned int i, got_bytes;
3521 struct cifs_readdata *rdata = container_of(work,
3522 struct cifs_readdata, work);
3523
3524 got_bytes = rdata->got_bytes;
3525 for (i = 0; i < rdata->nr_pages; i++) {
3526 struct page *page = rdata->pages[i];
3527
3528 lru_cache_add_file(page);
3529
3530 if (rdata->result == 0 ||
3531 (rdata->result == -EAGAIN && got_bytes)) {
3532 flush_dcache_page(page);
3533 SetPageUptodate(page);
3534 }
3535
3536 unlock_page(page);
3537
3538 if (rdata->result == 0 ||
3539 (rdata->result == -EAGAIN && got_bytes))
3540 cifs_readpage_to_fscache(rdata->mapping->host, page);
3541
3542 got_bytes -= min_t(unsigned int, PAGE_SIZE, got_bytes);
3543
3544 put_page(page);
3545 rdata->pages[i] = NULL;
3546 }
3547 kref_put(&rdata->refcount, cifs_readdata_release);
3548}
3549
3550static int
3551readpages_fill_pages(struct TCP_Server_Info *server,
3552 struct cifs_readdata *rdata, struct iov_iter *iter,
3553 unsigned int len)
3554{
3555 int result = 0;
3556 unsigned int i;
3557 u64 eof;
3558 pgoff_t eof_index;
3559 unsigned int nr_pages = rdata->nr_pages;
3560
3561 /* determine the eof that the server (probably) has */
3562 eof = CIFS_I(rdata->mapping->host)->server_eof;
3563 eof_index = eof ? (eof - 1) >> PAGE_SHIFT : 0;
3564 cifs_dbg(FYI, "eof=%llu eof_index=%lu\n", eof, eof_index);
3565
3566 rdata->got_bytes = 0;
3567 rdata->tailsz = PAGE_SIZE;
3568 for (i = 0; i < nr_pages; i++) {
3569 struct page *page = rdata->pages[i];
3570 size_t n = PAGE_SIZE;
3571
3572 if (len >= PAGE_SIZE) {
3573 len -= PAGE_SIZE;
3574 } else if (len > 0) {
3575 /* enough for partial page, fill and zero the rest */
3576 zero_user(page, len, PAGE_SIZE - len);
3577 n = rdata->tailsz = len;
3578 len = 0;
3579 } else if (page->index > eof_index) {
3580 /*
3581 * The VFS will not try to do readahead past the
3582 * i_size, but it's possible that we have outstanding
3583 * writes with gaps in the middle and the i_size hasn't
3584 * caught up yet. Populate those with zeroed out pages
3585 * to prevent the VFS from repeatedly attempting to
3586 * fill them until the writes are flushed.
3587 */
3588 zero_user(page, 0, PAGE_SIZE);
3589 lru_cache_add_file(page);
3590 flush_dcache_page(page);
3591 SetPageUptodate(page);
3592 unlock_page(page);
3593 put_page(page);
3594 rdata->pages[i] = NULL;
3595 rdata->nr_pages--;
3596 continue;
3597 } else {
3598 /* no need to hold page hostage */
3599 lru_cache_add_file(page);
3600 unlock_page(page);
3601 put_page(page);
3602 rdata->pages[i] = NULL;
3603 rdata->nr_pages--;
3604 continue;
3605 }
3606
3607 if (iter)
3608 result = copy_page_from_iter(page, 0, n, iter);
3609#ifdef CONFIG_CIFS_SMB_DIRECT
3610 else if (rdata->mr)
3611 result = n;
3612#endif
3613 else
3614 result = cifs_read_page_from_socket(server, page, n);
3615 if (result < 0)
3616 break;
3617
3618 rdata->got_bytes += result;
3619 }
3620
3621 return rdata->got_bytes > 0 && result != -ECONNABORTED ?
3622 rdata->got_bytes : result;
3623}
3624
3625static int
3626cifs_readpages_read_into_pages(struct TCP_Server_Info *server,
3627 struct cifs_readdata *rdata, unsigned int len)
3628{
3629 return readpages_fill_pages(server, rdata, NULL, len);
3630}
3631
3632static int
3633cifs_readpages_copy_into_pages(struct TCP_Server_Info *server,
3634 struct cifs_readdata *rdata,
3635 struct iov_iter *iter)
3636{
3637 return readpages_fill_pages(server, rdata, iter, iter->count);
3638}
3639
3640static int
3641readpages_get_pages(struct address_space *mapping, struct list_head *page_list,
3642 unsigned int rsize, struct list_head *tmplist,
3643 unsigned int *nr_pages, loff_t *offset, unsigned int *bytes)
3644{
3645 struct page *page, *tpage;
3646 unsigned int expected_index;
3647 int rc;
3648 gfp_t gfp = readahead_gfp_mask(mapping);
3649
3650 INIT_LIST_HEAD(tmplist);
3651
3652 page = list_entry(page_list->prev, struct page, lru);
3653
3654 /*
3655 * Lock the page and put it in the cache. Since no one else
3656 * should have access to this page, we're safe to simply set
3657 * PG_locked without checking it first.
3658 */
3659 __SetPageLocked(page);
3660 rc = add_to_page_cache_locked(page, mapping,
3661 page->index, gfp);
3662
3663 /* give up if we can't stick it in the cache */
3664 if (rc) {
3665 __ClearPageLocked(page);
3666 return rc;
3667 }
3668
3669 /* move first page to the tmplist */
3670 *offset = (loff_t)page->index << PAGE_SHIFT;
3671 *bytes = PAGE_SIZE;
3672 *nr_pages = 1;
3673 list_move_tail(&page->lru, tmplist);
3674
3675 /* now try and add more pages onto the request */
3676 expected_index = page->index + 1;
3677 list_for_each_entry_safe_reverse(page, tpage, page_list, lru) {
3678 /* discontinuity ? */
3679 if (page->index != expected_index)
3680 break;
3681
3682 /* would this page push the read over the rsize? */
3683 if (*bytes + PAGE_SIZE > rsize)
3684 break;
3685
3686 __SetPageLocked(page);
3687 if (add_to_page_cache_locked(page, mapping, page->index, gfp)) {
3688 __ClearPageLocked(page);
3689 break;
3690 }
3691 list_move_tail(&page->lru, tmplist);
3692 (*bytes) += PAGE_SIZE;
3693 expected_index++;
3694 (*nr_pages)++;
3695 }
3696 return rc;
3697}
3698
3699static int cifs_readpages(struct file *file, struct address_space *mapping,
3700 struct list_head *page_list, unsigned num_pages)
3701{
3702 int rc;
3703 struct list_head tmplist;
3704 struct cifsFileInfo *open_file = file->private_data;
3705 struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file);
3706 struct TCP_Server_Info *server;
3707 pid_t pid;
3708
3709 /*
3710 * Reads as many pages as possible from fscache. Returns -ENOBUFS
3711 * immediately if the cookie is negative
3712 *
3713 * After this point, every page in the list might have PG_fscache set,
3714 * so we will need to clean that up off of every page we don't use.
3715 */
3716 rc = cifs_readpages_from_fscache(mapping->host, mapping, page_list,
3717 &num_pages);
3718 if (rc == 0)
3719 return rc;
3720
3721 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
3722 pid = open_file->pid;
3723 else
3724 pid = current->tgid;
3725
3726 rc = 0;
3727 server = tlink_tcon(open_file->tlink)->ses->server;
3728
3729 cifs_dbg(FYI, "%s: file=%p mapping=%p num_pages=%u\n",
3730 __func__, file, mapping, num_pages);
3731
3732 /*
3733 * Start with the page at end of list and move it to private
3734 * list. Do the same with any following pages until we hit
3735 * the rsize limit, hit an index discontinuity, or run out of
3736 * pages. Issue the async read and then start the loop again
3737 * until the list is empty.
3738 *
3739 * Note that list order is important. The page_list is in
3740 * the order of declining indexes. When we put the pages in
3741 * the rdata->pages, then we want them in increasing order.
3742 */
3743 while (!list_empty(page_list)) {
3744 unsigned int i, nr_pages, bytes, rsize;
3745 loff_t offset;
3746 struct page *page, *tpage;
3747 struct cifs_readdata *rdata;
3748 unsigned credits;
3749
3750 rc = server->ops->wait_mtu_credits(server, cifs_sb->rsize,
3751 &rsize, &credits);
3752 if (rc)
3753 break;
3754
3755 /*
3756 * Give up immediately if rsize is too small to read an entire
3757 * page. The VFS will fall back to readpage. We should never
3758 * reach this point however since we set ra_pages to 0 when the
3759 * rsize is smaller than a cache page.
3760 */
3761 if (unlikely(rsize < PAGE_SIZE)) {
3762 add_credits_and_wake_if(server, credits, 0);
3763 return 0;
3764 }
3765
3766 rc = readpages_get_pages(mapping, page_list, rsize, &tmplist,
3767 &nr_pages, &offset, &bytes);
3768 if (rc) {
3769 add_credits_and_wake_if(server, credits, 0);
3770 break;
3771 }
3772
3773 rdata = cifs_readdata_alloc(nr_pages, cifs_readv_complete);
3774 if (!rdata) {
3775 /* best to give up if we're out of mem */
3776 list_for_each_entry_safe(page, tpage, &tmplist, lru) {
3777 list_del(&page->lru);
3778 lru_cache_add_file(page);
3779 unlock_page(page);
3780 put_page(page);
3781 }
3782 rc = -ENOMEM;
3783 add_credits_and_wake_if(server, credits, 0);
3784 break;
3785 }
3786
3787 rdata->cfile = cifsFileInfo_get(open_file);
3788 rdata->mapping = mapping;
3789 rdata->offset = offset;
3790 rdata->bytes = bytes;
3791 rdata->pid = pid;
3792 rdata->pagesz = PAGE_SIZE;
3793 rdata->read_into_pages = cifs_readpages_read_into_pages;
3794 rdata->copy_into_pages = cifs_readpages_copy_into_pages;
3795 rdata->credits = credits;
3796
3797 list_for_each_entry_safe(page, tpage, &tmplist, lru) {
3798 list_del(&page->lru);
3799 rdata->pages[rdata->nr_pages++] = page;
3800 }
3801
3802 if (!rdata->cfile->invalidHandle ||
3803 !(rc = cifs_reopen_file(rdata->cfile, true)))
3804 rc = server->ops->async_readv(rdata);
3805 if (rc) {
3806 add_credits_and_wake_if(server, rdata->credits, 0);
3807 for (i = 0; i < rdata->nr_pages; i++) {
3808 page = rdata->pages[i];
3809 lru_cache_add_file(page);
3810 unlock_page(page);
3811 put_page(page);
3812 }
3813 /* Fallback to the readpage in error/reconnect cases */
3814 kref_put(&rdata->refcount, cifs_readdata_release);
3815 break;
3816 }
3817
3818 kref_put(&rdata->refcount, cifs_readdata_release);
3819 }
3820
3821 /* Any pages that have been shown to fscache but didn't get added to
3822 * the pagecache must be uncached before they get returned to the
3823 * allocator.
3824 */
3825 cifs_fscache_readpages_cancel(mapping->host, page_list);
3826 return rc;
3827}
3828
3829/*
3830 * cifs_readpage_worker must be called with the page pinned
3831 */
3832static int cifs_readpage_worker(struct file *file, struct page *page,
3833 loff_t *poffset)
3834{
3835 char *read_data;
3836 int rc;
3837
3838 /* Is the page cached? */
3839 rc = cifs_readpage_from_fscache(file_inode(file), page);
3840 if (rc == 0)
3841 goto read_complete;
3842
3843 read_data = kmap(page);
3844 /* for reads over a certain size could initiate async read ahead */
3845
3846 rc = cifs_read(file, read_data, PAGE_SIZE, poffset);
3847
3848 if (rc < 0)
3849 goto io_error;
3850 else
3851 cifs_dbg(FYI, "Bytes read %d\n", rc);
3852
3853 file_inode(file)->i_atime =
3854 current_time(file_inode(file));
3855
3856 if (PAGE_SIZE > rc)
3857 memset(read_data + rc, 0, PAGE_SIZE - rc);
3858
3859 flush_dcache_page(page);
3860 SetPageUptodate(page);
3861
3862 /* send this page to the cache */
3863 cifs_readpage_to_fscache(file_inode(file), page);
3864
3865 rc = 0;
3866
3867io_error:
3868 kunmap(page);
3869 unlock_page(page);
3870
3871read_complete:
3872 return rc;
3873}
3874
3875static int cifs_readpage(struct file *file, struct page *page)
3876{
3877 loff_t offset = (loff_t)page->index << PAGE_SHIFT;
3878 int rc = -EACCES;
3879 unsigned int xid;
3880
3881 xid = get_xid();
3882
3883 if (file->private_data == NULL) {
3884 rc = -EBADF;
3885 free_xid(xid);
3886 return rc;
3887 }
3888
3889 cifs_dbg(FYI, "readpage %p at offset %d 0x%x\n",
3890 page, (int)offset, (int)offset);
3891
3892 rc = cifs_readpage_worker(file, page, &offset);
3893
3894 free_xid(xid);
3895 return rc;
3896}
3897
3898static int is_inode_writable(struct cifsInodeInfo *cifs_inode)
3899{
3900 struct cifsFileInfo *open_file;
3901 struct cifs_tcon *tcon =
3902 cifs_sb_master_tcon(CIFS_SB(cifs_inode->vfs_inode.i_sb));
3903
3904 spin_lock(&tcon->open_file_lock);
3905 list_for_each_entry(open_file, &cifs_inode->openFileList, flist) {
3906 if (OPEN_FMODE(open_file->f_flags) & FMODE_WRITE) {
3907 spin_unlock(&tcon->open_file_lock);
3908 return 1;
3909 }
3910 }
3911 spin_unlock(&tcon->open_file_lock);
3912 return 0;
3913}
3914
3915/* We do not want to update the file size from server for inodes
3916 open for write - to avoid races with writepage extending
3917 the file - in the future we could consider allowing
3918 refreshing the inode only on increases in the file size
3919 but this is tricky to do without racing with writebehind
3920 page caching in the current Linux kernel design */
3921bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file)
3922{
3923 if (!cifsInode)
3924 return true;
3925
3926 if (is_inode_writable(cifsInode)) {
3927 /* This inode is open for write at least once */
3928 struct cifs_sb_info *cifs_sb;
3929
3930 cifs_sb = CIFS_SB(cifsInode->vfs_inode.i_sb);
3931 if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) {
3932 /* since no page cache to corrupt on directio
3933 we can change size safely */
3934 return true;
3935 }
3936
3937 if (i_size_read(&cifsInode->vfs_inode) < end_of_file)
3938 return true;
3939
3940 return false;
3941 } else
3942 return true;
3943}
3944
3945static int cifs_write_begin(struct file *file, struct address_space *mapping,
3946 loff_t pos, unsigned len, unsigned flags,
3947 struct page **pagep, void **fsdata)
3948{
3949 int oncethru = 0;
3950 pgoff_t index = pos >> PAGE_SHIFT;
3951 loff_t offset = pos & (PAGE_SIZE - 1);
3952 loff_t page_start = pos & PAGE_MASK;
3953 loff_t i_size;
3954 struct page *page;
3955 int rc = 0;
3956
3957 cifs_dbg(FYI, "write_begin from %lld len %d\n", (long long)pos, len);
3958
3959start:
3960 page = grab_cache_page_write_begin(mapping, index, flags);
3961 if (!page) {
3962 rc = -ENOMEM;
3963 goto out;
3964 }
3965
3966 if (PageUptodate(page))
3967 goto out;
3968
3969 /*
3970 * If we write a full page it will be up to date, no need to read from
3971 * the server. If the write is short, we'll end up doing a sync write
3972 * instead.
3973 */
3974 if (len == PAGE_SIZE)
3975 goto out;
3976
3977 /*
3978 * optimize away the read when we have an oplock, and we're not
3979 * expecting to use any of the data we'd be reading in. That
3980 * is, when the page lies beyond the EOF, or straddles the EOF
3981 * and the write will cover all of the existing data.
3982 */
3983 if (CIFS_CACHE_READ(CIFS_I(mapping->host))) {
3984 i_size = i_size_read(mapping->host);
3985 if (page_start >= i_size ||
3986 (offset == 0 && (pos + len) >= i_size)) {
3987 zero_user_segments(page, 0, offset,
3988 offset + len,
3989 PAGE_SIZE);
3990 /*
3991 * PageChecked means that the parts of the page
3992 * to which we're not writing are considered up
3993 * to date. Once the data is copied to the
3994 * page, it can be set uptodate.
3995 */
3996 SetPageChecked(page);
3997 goto out;
3998 }
3999 }
4000
4001 if ((file->f_flags & O_ACCMODE) != O_WRONLY && !oncethru) {
4002 /*
4003 * might as well read a page, it is fast enough. If we get
4004 * an error, we don't need to return it. cifs_write_end will
4005 * do a sync write instead since PG_uptodate isn't set.
4006 */
4007 cifs_readpage_worker(file, page, &page_start);
4008 put_page(page);
4009 oncethru = 1;
4010 goto start;
4011 } else {
4012 /* we could try using another file handle if there is one -
4013 but how would we lock it to prevent close of that handle
4014 racing with this read? In any case
4015 this will be written out by write_end so is fine */
4016 }
4017out:
4018 *pagep = page;
4019 return rc;
4020}
4021
4022static int cifs_release_page(struct page *page, gfp_t gfp)
4023{
4024 if (PagePrivate(page))
4025 return 0;
4026
4027 return cifs_fscache_release_page(page, gfp);
4028}
4029
4030static void cifs_invalidate_page(struct page *page, unsigned int offset,
4031 unsigned int length)
4032{
4033 struct cifsInodeInfo *cifsi = CIFS_I(page->mapping->host);
4034
4035 if (offset == 0 && length == PAGE_SIZE)
4036 cifs_fscache_invalidate_page(page, &cifsi->vfs_inode);
4037}
4038
4039static int cifs_launder_page(struct page *page)
4040{
4041 int rc = 0;
4042 loff_t range_start = page_offset(page);
4043 loff_t range_end = range_start + (loff_t)(PAGE_SIZE - 1);
4044 struct writeback_control wbc = {
4045 .sync_mode = WB_SYNC_ALL,
4046 .nr_to_write = 0,
4047 .range_start = range_start,
4048 .range_end = range_end,
4049 };
4050
4051 cifs_dbg(FYI, "Launder page: %p\n", page);
4052
4053 if (clear_page_dirty_for_io(page))
4054 rc = cifs_writepage_locked(page, &wbc);
4055
4056 cifs_fscache_invalidate_page(page, page->mapping->host);
4057 return rc;
4058}
4059
4060void cifs_oplock_break(struct work_struct *work)
4061{
4062 struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo,
4063 oplock_break);
4064 struct inode *inode = d_inode(cfile->dentry);
4065 struct cifsInodeInfo *cinode = CIFS_I(inode);
4066 struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
4067 struct TCP_Server_Info *server = tcon->ses->server;
4068 int rc = 0;
4069
4070 wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS,
4071 TASK_UNINTERRUPTIBLE);
4072
4073 server->ops->downgrade_oplock(server, cinode,
4074 test_bit(CIFS_INODE_DOWNGRADE_OPLOCK_TO_L2, &cinode->flags));
4075
4076 if (!CIFS_CACHE_WRITE(cinode) && CIFS_CACHE_READ(cinode) &&
4077 cifs_has_mand_locks(cinode)) {
4078 cifs_dbg(FYI, "Reset oplock to None for inode=%p due to mand locks\n",
4079 inode);
4080 cinode->oplock = 0;
4081 }
4082
4083 if (inode && S_ISREG(inode->i_mode)) {
4084 if (CIFS_CACHE_READ(cinode))
4085 break_lease(inode, O_RDONLY);
4086 else
4087 break_lease(inode, O_WRONLY);
4088 rc = filemap_fdatawrite(inode->i_mapping);
4089 if (!CIFS_CACHE_READ(cinode)) {
4090 rc = filemap_fdatawait(inode->i_mapping);
4091 mapping_set_error(inode->i_mapping, rc);
4092 cifs_zap_mapping(inode);
4093 }
4094 cifs_dbg(FYI, "Oplock flush inode %p rc %d\n", inode, rc);
4095 }
4096
4097 rc = cifs_push_locks(cfile);
4098 if (rc)
4099 cifs_dbg(VFS, "Push locks rc = %d\n", rc);
4100
4101 /*
4102 * releasing stale oplock after recent reconnect of smb session using
4103 * a now incorrect file handle is not a data integrity issue but do
4104 * not bother sending an oplock release if session to server still is
4105 * disconnected since oplock already released by the server
4106 */
4107 if (!cfile->oplock_break_cancelled) {
4108 rc = tcon->ses->server->ops->oplock_response(tcon, &cfile->fid,
4109 cinode);
4110 cifs_dbg(FYI, "Oplock release rc = %d\n", rc);
4111 }
4112 cifs_done_oplock_break(cinode);
4113}
4114
4115/*
4116 * The presence of cifs_direct_io() in the address space ops vector
4117 * allowes open() O_DIRECT flags which would have failed otherwise.
4118 *
4119 * In the non-cached mode (mount with cache=none), we shunt off direct read and write requests
4120 * so this method should never be called.
4121 *
4122 * Direct IO is not yet supported in the cached mode.
4123 */
4124static ssize_t
4125cifs_direct_io(struct kiocb *iocb, struct iov_iter *iter)
4126{
4127 /*
4128 * FIXME
4129 * Eventually need to support direct IO for non forcedirectio mounts
4130 */
4131 return -EINVAL;
4132}
4133
4134
4135const struct address_space_operations cifs_addr_ops = {
4136 .readpage = cifs_readpage,
4137 .readpages = cifs_readpages,
4138 .writepage = cifs_writepage,
4139 .writepages = cifs_writepages,
4140 .write_begin = cifs_write_begin,
4141 .write_end = cifs_write_end,
4142 .set_page_dirty = __set_page_dirty_nobuffers,
4143 .releasepage = cifs_release_page,
4144 .direct_IO = cifs_direct_io,
4145 .invalidatepage = cifs_invalidate_page,
4146 .launder_page = cifs_launder_page,
4147};
4148
4149/*
4150 * cifs_readpages requires the server to support a buffer large enough to
4151 * contain the header plus one complete page of data. Otherwise, we need
4152 * to leave cifs_readpages out of the address space operations.
4153 */
4154const struct address_space_operations cifs_addr_ops_smallbuf = {
4155 .readpage = cifs_readpage,
4156 .writepage = cifs_writepage,
4157 .writepages = cifs_writepages,
4158 .write_begin = cifs_write_begin,
4159 .write_end = cifs_write_end,
4160 .set_page_dirty = __set_page_dirty_nobuffers,
4161 .releasepage = cifs_release_page,
4162 .invalidatepage = cifs_invalidate_page,
4163 .launder_page = cifs_launder_page,
4164};