1 // SPDX-License-Identifier: GPL-2.0
2 /*
3 * Copyright (c) 2000-2006 Silicon Graphics, Inc.
4 * Copyright (c) 2012-2013 Red Hat, Inc.
5 * All rights reserved.
6 */
7 #include "xfs.h"
8 #include "xfs_shared.h"
9 #include "xfs_fs.h"
10 #include "xfs_format.h"
11 #include "xfs_log_format.h"
12 #include "xfs_trans_resv.h"
13 #include "xfs_bit.h"
14 #include "xfs_mount.h"
15 #include "xfs_dir2.h"
16 #include "xfs_inode.h"
17 #include "xfs_bmap.h"
18 #include "xfs_bmap_btree.h"
19 #include "xfs_quota.h"
20 #include "xfs_symlink.h"
21 #include "xfs_trans_space.h"
22 #include "xfs_trace.h"
23 #include "xfs_trans.h"
24 #include "xfs_ialloc.h"
25 #include "xfs_error.h"
26
27 /* ----- Kernel only functions below ----- */
28 int
xfs_readlink_bmap_ilocked(struct xfs_inode * ip,char * link)29 xfs_readlink_bmap_ilocked(
30 struct xfs_inode *ip,
31 char *link)
32 {
33 struct xfs_mount *mp = ip->i_mount;
34 struct xfs_bmbt_irec mval[XFS_SYMLINK_MAPS];
35 struct xfs_buf *bp;
36 xfs_daddr_t d;
37 char *cur_chunk;
38 int pathlen = ip->i_disk_size;
39 int nmaps = XFS_SYMLINK_MAPS;
40 int byte_cnt;
41 int n;
42 int error = 0;
43 int fsblocks = 0;
44 int offset;
45
46 ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED | XFS_ILOCK_EXCL));
47
48 fsblocks = xfs_symlink_blocks(mp, pathlen);
49 error = xfs_bmapi_read(ip, 0, fsblocks, mval, &nmaps, 0);
50 if (error)
51 goto out;
52
53 offset = 0;
54 for (n = 0; n < nmaps; n++) {
55 d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
56 byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
57
58 error = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0,
59 &bp, &xfs_symlink_buf_ops);
60 if (error)
61 return error;
62 byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
63 if (pathlen < byte_cnt)
64 byte_cnt = pathlen;
65
66 cur_chunk = bp->b_addr;
67 if (xfs_has_crc(mp)) {
68 if (!xfs_symlink_hdr_ok(ip->i_ino, offset,
69 byte_cnt, bp)) {
70 error = -EFSCORRUPTED;
71 xfs_alert(mp,
72 "symlink header does not match required off/len/owner (0x%x/Ox%x,0x%llx)",
73 offset, byte_cnt, ip->i_ino);
74 xfs_buf_relse(bp);
75 goto out;
76
77 }
78
79 cur_chunk += sizeof(struct xfs_dsymlink_hdr);
80 }
81
82 memcpy(link + offset, cur_chunk, byte_cnt);
83
84 pathlen -= byte_cnt;
85 offset += byte_cnt;
86
87 xfs_buf_relse(bp);
88 }
89 ASSERT(pathlen == 0);
90
91 link[ip->i_disk_size] = '\0';
92 error = 0;
93
94 out:
95 return error;
96 }
97
98 int
xfs_readlink(struct xfs_inode * ip,char * link)99 xfs_readlink(
100 struct xfs_inode *ip,
101 char *link)
102 {
103 struct xfs_mount *mp = ip->i_mount;
104 xfs_fsize_t pathlen;
105 int error = -EFSCORRUPTED;
106
107 trace_xfs_readlink(ip);
108
109 if (xfs_is_shutdown(mp))
110 return -EIO;
111
112 xfs_ilock(ip, XFS_ILOCK_SHARED);
113
114 pathlen = ip->i_disk_size;
115 if (!pathlen)
116 goto out;
117
118 if (pathlen < 0 || pathlen > XFS_SYMLINK_MAXLEN) {
119 xfs_alert(mp, "%s: inode (%llu) bad symlink length (%lld)",
120 __func__, (unsigned long long) ip->i_ino,
121 (long long) pathlen);
122 ASSERT(0);
123 goto out;
124 }
125
126 if (ip->i_df.if_format == XFS_DINODE_FMT_LOCAL) {
127 /*
128 * The VFS crashes on a NULL pointer, so return -EFSCORRUPTED
129 * if if_data is junk.
130 */
131 if (XFS_IS_CORRUPT(ip->i_mount, !ip->i_df.if_u1.if_data))
132 goto out;
133
134 memcpy(link, ip->i_df.if_u1.if_data, pathlen + 1);
135 error = 0;
136 } else {
137 error = xfs_readlink_bmap_ilocked(ip, link);
138 }
139
140 out:
141 xfs_iunlock(ip, XFS_ILOCK_SHARED);
142 return error;
143 }
144
145 int
xfs_symlink(struct user_namespace * mnt_userns,struct xfs_inode * dp,struct xfs_name * link_name,const char * target_path,umode_t mode,struct xfs_inode ** ipp)146 xfs_symlink(
147 struct user_namespace *mnt_userns,
148 struct xfs_inode *dp,
149 struct xfs_name *link_name,
150 const char *target_path,
151 umode_t mode,
152 struct xfs_inode **ipp)
153 {
154 struct xfs_mount *mp = dp->i_mount;
155 struct xfs_trans *tp = NULL;
156 struct xfs_inode *ip = NULL;
157 int error = 0;
158 int pathlen;
159 bool unlock_dp_on_error = false;
160 xfs_fileoff_t first_fsb;
161 xfs_filblks_t fs_blocks;
162 int nmaps;
163 struct xfs_bmbt_irec mval[XFS_SYMLINK_MAPS];
164 xfs_daddr_t d;
165 const char *cur_chunk;
166 int byte_cnt;
167 int n;
168 struct xfs_buf *bp;
169 prid_t prid;
170 struct xfs_dquot *udqp = NULL;
171 struct xfs_dquot *gdqp = NULL;
172 struct xfs_dquot *pdqp = NULL;
173 uint resblks;
174 xfs_ino_t ino;
175
176 *ipp = NULL;
177
178 trace_xfs_symlink(dp, link_name);
179
180 if (xfs_is_shutdown(mp))
181 return -EIO;
182
183 /*
184 * Check component lengths of the target path name.
185 */
186 pathlen = strlen(target_path);
187 if (pathlen >= XFS_SYMLINK_MAXLEN) /* total string too long */
188 return -ENAMETOOLONG;
189 ASSERT(pathlen > 0);
190
191 prid = xfs_get_initial_prid(dp);
192
193 /*
194 * Make sure that we have allocated dquot(s) on disk.
195 */
196 error = xfs_qm_vop_dqalloc(dp, mapped_fsuid(mnt_userns, &init_user_ns),
197 mapped_fsgid(mnt_userns, &init_user_ns), prid,
198 XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT,
199 &udqp, &gdqp, &pdqp);
200 if (error)
201 return error;
202
203 /*
204 * The symlink will fit into the inode data fork?
205 * There can't be any attributes so we get the whole variable part.
206 */
207 if (pathlen <= XFS_LITINO(mp))
208 fs_blocks = 0;
209 else
210 fs_blocks = xfs_symlink_blocks(mp, pathlen);
211 resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks);
212
213 error = xfs_trans_alloc_icreate(mp, &M_RES(mp)->tr_symlink, udqp, gdqp,
214 pdqp, resblks, &tp);
215 if (error)
216 goto out_release_dquots;
217
218 xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT);
219 unlock_dp_on_error = true;
220
221 /*
222 * Check whether the directory allows new symlinks or not.
223 */
224 if (dp->i_diflags & XFS_DIFLAG_NOSYMLINKS) {
225 error = -EPERM;
226 goto out_trans_cancel;
227 }
228
229 error = xfs_iext_count_may_overflow(dp, XFS_DATA_FORK,
230 XFS_IEXT_DIR_MANIP_CNT(mp));
231 if (error)
232 goto out_trans_cancel;
233
234 /*
235 * Allocate an inode for the symlink.
236 */
237 error = xfs_dialloc(&tp, dp->i_ino, S_IFLNK, &ino);
238 if (!error)
239 error = xfs_init_new_inode(mnt_userns, tp, dp, ino,
240 S_IFLNK | (mode & ~S_IFMT), 1, 0, prid,
241 false, &ip);
242 if (error)
243 goto out_trans_cancel;
244
245 /*
246 * Now we join the directory inode to the transaction. We do not do it
247 * earlier because xfs_dir_ialloc might commit the previous transaction
248 * (and release all the locks). An error from here on will result in
249 * the transaction cancel unlocking dp so don't do it explicitly in the
250 * error path.
251 */
252 xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL);
253 unlock_dp_on_error = false;
254
255 /*
256 * Also attach the dquot(s) to it, if applicable.
257 */
258 xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp);
259
260 resblks -= XFS_IALLOC_SPACE_RES(mp);
261 /*
262 * If the symlink will fit into the inode, write it inline.
263 */
264 if (pathlen <= XFS_IFORK_DSIZE(ip)) {
265 xfs_init_local_fork(ip, XFS_DATA_FORK, target_path, pathlen);
266
267 ip->i_disk_size = pathlen;
268 ip->i_df.if_format = XFS_DINODE_FMT_LOCAL;
269 xfs_trans_log_inode(tp, ip, XFS_ILOG_DDATA | XFS_ILOG_CORE);
270 } else {
271 int offset;
272
273 first_fsb = 0;
274 nmaps = XFS_SYMLINK_MAPS;
275
276 error = xfs_bmapi_write(tp, ip, first_fsb, fs_blocks,
277 XFS_BMAPI_METADATA, resblks, mval, &nmaps);
278 if (error)
279 goto out_trans_cancel;
280
281 resblks -= fs_blocks;
282 ip->i_disk_size = pathlen;
283 xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
284
285 cur_chunk = target_path;
286 offset = 0;
287 for (n = 0; n < nmaps; n++) {
288 char *buf;
289
290 d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
291 byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
292 error = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
293 BTOBB(byte_cnt), 0, &bp);
294 if (error)
295 goto out_trans_cancel;
296 bp->b_ops = &xfs_symlink_buf_ops;
297
298 byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
299 byte_cnt = min(byte_cnt, pathlen);
300
301 buf = bp->b_addr;
302 buf += xfs_symlink_hdr_set(mp, ip->i_ino, offset,
303 byte_cnt, bp);
304
305 memcpy(buf, cur_chunk, byte_cnt);
306
307 cur_chunk += byte_cnt;
308 pathlen -= byte_cnt;
309 offset += byte_cnt;
310
311 xfs_trans_buf_set_type(tp, bp, XFS_BLFT_SYMLINK_BUF);
312 xfs_trans_log_buf(tp, bp, 0, (buf + byte_cnt - 1) -
313 (char *)bp->b_addr);
314 }
315 ASSERT(pathlen == 0);
316 }
317 i_size_write(VFS_I(ip), ip->i_disk_size);
318
319 /*
320 * Create the directory entry for the symlink.
321 */
322 error = xfs_dir_createname(tp, dp, link_name, ip->i_ino, resblks);
323 if (error)
324 goto out_trans_cancel;
325 xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
326 xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
327
328 /*
329 * If this is a synchronous mount, make sure that the
330 * symlink transaction goes to disk before returning to
331 * the user.
332 */
333 if (xfs_has_wsync(mp) || xfs_has_dirsync(mp))
334 xfs_trans_set_sync(tp);
335
336 error = xfs_trans_commit(tp);
337 if (error)
338 goto out_release_inode;
339
340 xfs_qm_dqrele(udqp);
341 xfs_qm_dqrele(gdqp);
342 xfs_qm_dqrele(pdqp);
343
344 *ipp = ip;
345 return 0;
346
347 out_trans_cancel:
348 xfs_trans_cancel(tp);
349 out_release_inode:
350 /*
351 * Wait until after the current transaction is aborted to finish the
352 * setup of the inode and release the inode. This prevents recursive
353 * transactions and deadlocks from xfs_inactive.
354 */
355 if (ip) {
356 xfs_finish_inode_setup(ip);
357 xfs_irele(ip);
358 }
359 out_release_dquots:
360 xfs_qm_dqrele(udqp);
361 xfs_qm_dqrele(gdqp);
362 xfs_qm_dqrele(pdqp);
363
364 if (unlock_dp_on_error)
365 xfs_iunlock(dp, XFS_ILOCK_EXCL);
366 return error;
367 }
368
369 /*
370 * Free a symlink that has blocks associated with it.
371 *
372 * Note: zero length symlinks are not allowed to exist. When we set the size to
373 * zero, also change it to a regular file so that it does not get written to
374 * disk as a zero length symlink. The inode is on the unlinked list already, so
375 * userspace cannot find this inode anymore, so this change is not user visible
376 * but allows us to catch corrupt zero-length symlinks in the verifiers.
377 */
378 STATIC int
xfs_inactive_symlink_rmt(struct xfs_inode * ip)379 xfs_inactive_symlink_rmt(
380 struct xfs_inode *ip)
381 {
382 struct xfs_buf *bp;
383 int done;
384 int error;
385 int i;
386 xfs_mount_t *mp;
387 xfs_bmbt_irec_t mval[XFS_SYMLINK_MAPS];
388 int nmaps;
389 int size;
390 xfs_trans_t *tp;
391
392 mp = ip->i_mount;
393 ASSERT(!xfs_need_iread_extents(&ip->i_df));
394 /*
395 * We're freeing a symlink that has some
396 * blocks allocated to it. Free the
397 * blocks here. We know that we've got
398 * either 1 or 2 extents and that we can
399 * free them all in one bunmapi call.
400 */
401 ASSERT(ip->i_df.if_nextents > 0 && ip->i_df.if_nextents <= 2);
402
403 error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate, 0, 0, 0, &tp);
404 if (error)
405 return error;
406
407 xfs_ilock(ip, XFS_ILOCK_EXCL);
408 xfs_trans_ijoin(tp, ip, 0);
409
410 /*
411 * Lock the inode, fix the size, turn it into a regular file and join it
412 * to the transaction. Hold it so in the normal path, we still have it
413 * locked for the second transaction. In the error paths we need it
414 * held so the cancel won't rele it, see below.
415 */
416 size = (int)ip->i_disk_size;
417 ip->i_disk_size = 0;
418 VFS_I(ip)->i_mode = (VFS_I(ip)->i_mode & ~S_IFMT) | S_IFREG;
419 xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
420 /*
421 * Find the block(s) so we can inval and unmap them.
422 */
423 done = 0;
424 nmaps = ARRAY_SIZE(mval);
425 error = xfs_bmapi_read(ip, 0, xfs_symlink_blocks(mp, size),
426 mval, &nmaps, 0);
427 if (error)
428 goto error_trans_cancel;
429 /*
430 * Invalidate the block(s). No validation is done.
431 */
432 for (i = 0; i < nmaps; i++) {
433 error = xfs_trans_get_buf(tp, mp->m_ddev_targp,
434 XFS_FSB_TO_DADDR(mp, mval[i].br_startblock),
435 XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0,
436 &bp);
437 if (error)
438 goto error_trans_cancel;
439 xfs_trans_binval(tp, bp);
440 }
441 /*
442 * Unmap the dead block(s) to the dfops.
443 */
444 error = xfs_bunmapi(tp, ip, 0, size, 0, nmaps, &done);
445 if (error)
446 goto error_trans_cancel;
447 ASSERT(done);
448
449 /*
450 * Commit the transaction. This first logs the EFI and the inode, then
451 * rolls and commits the transaction that frees the extents.
452 */
453 xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
454 error = xfs_trans_commit(tp);
455 if (error) {
456 ASSERT(xfs_is_shutdown(mp));
457 goto error_unlock;
458 }
459
460 /*
461 * Remove the memory for extent descriptions (just bookkeeping).
462 */
463 if (ip->i_df.if_bytes)
464 xfs_idata_realloc(ip, -ip->i_df.if_bytes, XFS_DATA_FORK);
465 ASSERT(ip->i_df.if_bytes == 0);
466
467 xfs_iunlock(ip, XFS_ILOCK_EXCL);
468 return 0;
469
470 error_trans_cancel:
471 xfs_trans_cancel(tp);
472 error_unlock:
473 xfs_iunlock(ip, XFS_ILOCK_EXCL);
474 return error;
475 }
476
477 /*
478 * xfs_inactive_symlink - free a symlink
479 */
480 int
xfs_inactive_symlink(struct xfs_inode * ip)481 xfs_inactive_symlink(
482 struct xfs_inode *ip)
483 {
484 struct xfs_mount *mp = ip->i_mount;
485 int pathlen;
486
487 trace_xfs_inactive_symlink(ip);
488
489 if (xfs_is_shutdown(mp))
490 return -EIO;
491
492 xfs_ilock(ip, XFS_ILOCK_EXCL);
493 pathlen = (int)ip->i_disk_size;
494 ASSERT(pathlen);
495
496 if (pathlen <= 0 || pathlen > XFS_SYMLINK_MAXLEN) {
497 xfs_alert(mp, "%s: inode (0x%llx) bad symlink length (%d)",
498 __func__, (unsigned long long)ip->i_ino, pathlen);
499 xfs_iunlock(ip, XFS_ILOCK_EXCL);
500 ASSERT(0);
501 return -EFSCORRUPTED;
502 }
503
504 /*
505 * Inline fork state gets removed by xfs_difree() so we have nothing to
506 * do here in that case.
507 */
508 if (ip->i_df.if_format == XFS_DINODE_FMT_LOCAL) {
509 xfs_iunlock(ip, XFS_ILOCK_EXCL);
510 return 0;
511 }
512
513 xfs_iunlock(ip, XFS_ILOCK_EXCL);
514
515 /* remove the remote symlink */
516 return xfs_inactive_symlink_rmt(ip);
517 }
518