]> git.ipfire.org Git - people/ms/linux.git/blame - fs/ceph/mds_client.c
projects / people / ms / linux.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
Merge branch '5.20/scsi-queue' into 6.0/scsi-fixes
[people/ms/linux.git] / fs / ceph / mds_client.c
CommitLineData
b2441318 1// SPDX-License-Identifier: GPL-2.0
3d14c5d2 2#include <linux/ceph/ceph_debug.h>
2f2dc053 3
496e5955 4#include <linux/fs.h>
2f2dc053 5#include <linux/wait.h>
5a0e3ad6 6#include <linux/slab.h>
54008399 7#include <linux/gfp.h>
2f2dc053 8#include <linux/sched.h>
3d14c5d2
YS		 9#include <linux/debugfs.h>
10#include <linux/seq_file.h>
3e0708b9 11#include <linux/ratelimit.h>
9ba1e224 12#include <linux/bits.h>
70c94820 13#include <linux/ktime.h>
d517b398 14#include <linux/bitmap.h>
2f2dc053 15
2f2dc053 16#include "super.h"
3d14c5d2
YS		 17#include "mds_client.h"
18
1fe60e51 19#include <linux/ceph/ceph_features.h>
3d14c5d2
YS		 20#include <linux/ceph/messenger.h>
21#include <linux/ceph/decode.h>
22#include <linux/ceph/pagelist.h>
23#include <linux/ceph/auth.h>
24#include <linux/ceph/debugfs.h>
2f2dc053 25
81c5a148
YZ		 26#define RECONNECT_MAX_SIZE (INT_MAX - PAGE_SIZE)
27
2f2dc053
SW		 28/*
29 * A cluster of MDS (metadata server) daemons is responsible for
30 * managing the file system namespace (the directory hierarchy and
31 * inodes) and for coordinating shared access to storage. Metadata is
32 * partitioning hierarchically across a number of servers, and that
33 * partition varies over time as the cluster adjusts the distribution
34 * in order to balance load.
35 *
36 * The MDS client is primarily responsible to managing synchronous
37 * metadata requests for operations like open, unlink, and so forth.
38 * If there is a MDS failure, we find out about it when we (possibly
39 * request and) receive a new MDS map, and can resubmit affected
40 * requests.
41 *
42 * For the most part, though, we take advantage of a lossless
43 * communications channel to the MDS, and do not need to worry about
44 * timing out or resubmitting requests.
45 *
46 * We maintain a stateful "session" with each MDS we interact with.
47 * Within each session, we sent periodic heartbeat messages to ensure
48 * any capabilities or leases we have been issues remain valid. If
49 * the session times out and goes stale, our leases and capabilities
50 * are no longer valid.
51 */
52
20cb34ae 53struct ceph_reconnect_state {
81c5a148
YZ		 54 struct ceph_mds_session *session;
55 int nr_caps, nr_realms;
20cb34ae 56 struct ceph_pagelist *pagelist;
121f22a1 57 unsigned msg_version;
81c5a148 58 bool allow_multi;
20cb34ae
SW		 59};
60
2f2dc053
SW		 61static void __wake_requests(struct ceph_mds_client *mdsc,
62 struct list_head *head);
e3ec8d68 63static void ceph_cap_release_work(struct work_struct *work);
37c4efc1 64static void ceph_cap_reclaim_work(struct work_struct *work);
2f2dc053 65
9e32789f 66static const struct ceph_connection_operations mds_con_ops;
2f2dc053
SW		 67
68
69/*
70 * mds reply parsing
71 */
72
b37fe1f9
YZ		 73static int parse_reply_info_quota(void **p, void *end,
74 struct ceph_mds_reply_info_in *info)
75{
76 u8 struct_v, struct_compat;
77 u32 struct_len;
78
79 ceph_decode_8_safe(p, end, struct_v, bad);
80 ceph_decode_8_safe(p, end, struct_compat, bad);
81 /* struct_v is expected to be >= 1. we only
82 * understand encoding with struct_compat == 1. */
83 if (!struct_v || struct_compat != 1)
84 goto bad;
85 ceph_decode_32_safe(p, end, struct_len, bad);
86 ceph_decode_need(p, end, struct_len, bad);
87 end = *p + struct_len;
88 ceph_decode_64_safe(p, end, info->max_bytes, bad);
89 ceph_decode_64_safe(p, end, info->max_files, bad);
90 *p = end;
91 return 0;
92bad:
93 return -EIO;
94}
95
2f2dc053
SW		 96/*
97 * parse individual inode info
98 */
99static int parse_reply_info_in(void **p, void *end,
14303d20 100 struct ceph_mds_reply_info_in *info,
12b4629a 101 u64 features)
2f2dc053 102{
b37fe1f9
YZ		 103 int err = 0;
104 u8 struct_v = 0;
2f2dc053 105
b37fe1f9
YZ		 106 if (features == (u64)-1) {
107 u32 struct_len;
108 u8 struct_compat;
109 ceph_decode_8_safe(p, end, struct_v, bad);
110 ceph_decode_8_safe(p, end, struct_compat, bad);
111 /* struct_v is expected to be >= 1. we only understand
112 * encoding with struct_compat == 1. */
113 if (!struct_v || struct_compat != 1)
114 goto bad;
115 ceph_decode_32_safe(p, end, struct_len, bad);
116 ceph_decode_need(p, end, struct_len, bad);
117 end = *p + struct_len;
118 }
119
120 ceph_decode_need(p, end, sizeof(struct ceph_mds_reply_inode), bad);
2f2dc053
SW		 121 info->in = *p;
122 *p += sizeof(struct ceph_mds_reply_inode) +
123 sizeof(*info->in->fragtree.splits) *
124 le32_to_cpu(info->in->fragtree.nsplits);
125
126 ceph_decode_32_safe(p, end, info->symlink_len, bad);
127 ceph_decode_need(p, end, info->symlink_len, bad);
128 info->symlink = *p;
129 *p += info->symlink_len;
130
23c625ce
ID		 131 ceph_decode_copy_safe(p, end, &info->dir_layout,
132 sizeof(info->dir_layout), bad);
2f2dc053
SW		 133 ceph_decode_32_safe(p, end, info->xattr_len, bad);
134 ceph_decode_need(p, end, info->xattr_len, bad);
135 info->xattr_data = *p;
136 *p += info->xattr_len;
fb01d1f8 137
b37fe1f9
YZ		 138 if (features == (u64)-1) {
139 /* inline data */
fb01d1f8
YZ		 140 ceph_decode_64_safe(p, end, info->inline_version, bad);
141 ceph_decode_32_safe(p, end, info->inline_len, bad);
142 ceph_decode_need(p, end, info->inline_len, bad);
143 info->inline_data = *p;
144 *p += info->inline_len;
b37fe1f9
YZ		 145 /* quota */
146 err = parse_reply_info_quota(p, end, info);
147 if (err < 0)
148 goto out_bad;
149 /* pool namespace */
150 ceph_decode_32_safe(p, end, info->pool_ns_len, bad);
151 if (info->pool_ns_len > 0) {
152 ceph_decode_need(p, end, info->pool_ns_len, bad);
153 info->pool_ns_data = *p;
154 *p += info->pool_ns_len;
155 }
245ce991
JL		 156
157 /* btime */
158 ceph_decode_need(p, end, sizeof(info->btime), bad);
159 ceph_decode_copy(p, &info->btime, sizeof(info->btime));
160
161 /* change attribute */
a35ead31 162 ceph_decode_64_safe(p, end, info->change_attr, bad);
fb01d1f8 163
08796873
YZ		 164 /* dir pin */
165 if (struct_v >= 2) {
166 ceph_decode_32_safe(p, end, info->dir_pin, bad);
167 } else {
168 info->dir_pin = -ENODATA;
169 }
170
193e7b37
DD		 171 /* snapshot birth time, remains zero for v<=2 */
172 if (struct_v >= 3) {
173 ceph_decode_need(p, end, sizeof(info->snap_btime), bad);
174 ceph_decode_copy(p, &info->snap_btime,
175 sizeof(info->snap_btime));
176 } else {
177 memset(&info->snap_btime, 0, sizeof(info->snap_btime));
178 }
179
e7f72952
YC		 180 /* snapshot count, remains zero for v<=3 */
181 if (struct_v >= 4) {
182 ceph_decode_64_safe(p, end, info->rsnaps, bad);
183 } else {
184 info->rsnaps = 0;
185 }
186
b37fe1f9
YZ		 187 *p = end;
188 } else {
189 if (features & CEPH_FEATURE_MDS_INLINE_DATA) {
190 ceph_decode_64_safe(p, end, info->inline_version, bad);
191 ceph_decode_32_safe(p, end, info->inline_len, bad);
192 ceph_decode_need(p, end, info->inline_len, bad);
193 info->inline_data = *p;
194 *p += info->inline_len;
195 } else
196 info->inline_version = CEPH_INLINE_NONE;
197
198 if (features & CEPH_FEATURE_MDS_QUOTA) {
199 err = parse_reply_info_quota(p, end, info);
200 if (err < 0)
201 goto out_bad;
202 } else {
203 info->max_bytes = 0;
204 info->max_files = 0;
205 }
206
207 info->pool_ns_len = 0;
208 info->pool_ns_data = NULL;
209 if (features & CEPH_FEATURE_FS_FILE_LAYOUT_V2) {
210 ceph_decode_32_safe(p, end, info->pool_ns_len, bad);
211 if (info->pool_ns_len > 0) {
212 ceph_decode_need(p, end, info->pool_ns_len, bad);
213 info->pool_ns_data = *p;
214 *p += info->pool_ns_len;
215 }
216 }
08796873 217
245ce991
JL		 218 if (features & CEPH_FEATURE_FS_BTIME) {
219 ceph_decode_need(p, end, sizeof(info->btime), bad);
220 ceph_decode_copy(p, &info->btime, sizeof(info->btime));
a35ead31 221 ceph_decode_64_safe(p, end, info->change_attr, bad);
245ce991
JL		 222 }
223
08796873 224 info->dir_pin = -ENODATA;
e7f72952 225 /* info->snap_btime and info->rsnaps remain zero */
b37fe1f9
YZ		 226 }
227 return 0;
228bad:
229 err = -EIO;
230out_bad:
231 return err;
232}
233
234static int parse_reply_info_dir(void **p, void *end,
235 struct ceph_mds_reply_dirfrag **dirfrag,
236 u64 features)
237{
238 if (features == (u64)-1) {
fb18a575
LH		 239 u8 struct_v, struct_compat;
240 u32 struct_len;
fb18a575
LH		 241 ceph_decode_8_safe(p, end, struct_v, bad);
242 ceph_decode_8_safe(p, end, struct_compat, bad);
b37fe1f9
YZ		 243 /* struct_v is expected to be >= 1. we only understand
244 * encoding whose struct_compat == 1. */
245 if (!struct_v || struct_compat != 1)
fb18a575
LH		 246 goto bad;
247 ceph_decode_32_safe(p, end, struct_len, bad);
248 ceph_decode_need(p, end, struct_len, bad);
b37fe1f9 249 end = *p + struct_len;
fb18a575
LH		 250 }
251
b37fe1f9
YZ		 252 ceph_decode_need(p, end, sizeof(**dirfrag), bad);
253 *dirfrag = *p;
254 *p += sizeof(**dirfrag) + sizeof(u32) * le32_to_cpu((*dirfrag)->ndist);
255 if (unlikely(*p > end))
256 goto bad;
257 if (features == (u64)-1)
258 *p = end;
259 return 0;
260bad:
261 return -EIO;
262}
263
264static int parse_reply_info_lease(void **p, void *end,
265 struct ceph_mds_reply_lease **lease,
266 u64 features)
267{
268 if (features == (u64)-1) {
269 u8 struct_v, struct_compat;
270 u32 struct_len;
271 ceph_decode_8_safe(p, end, struct_v, bad);
272 ceph_decode_8_safe(p, end, struct_compat, bad);
273 /* struct_v is expected to be >= 1. we only understand
274 * encoding whose struct_compat == 1. */
275 if (!struct_v || struct_compat != 1)
276 goto bad;
277 ceph_decode_32_safe(p, end, struct_len, bad);
278 ceph_decode_need(p, end, struct_len, bad);
279 end = *p + struct_len;
5ea5c5e0
YZ		 280 }
281
b37fe1f9
YZ		 282 ceph_decode_need(p, end, sizeof(**lease), bad);
283 *lease = *p;
284 *p += sizeof(**lease);
285 if (features == (u64)-1)
286 *p = end;
2f2dc053
SW		 287 return 0;
288bad:
b37fe1f9 289 return -EIO;
2f2dc053
SW		 290}
291
292/*
293 * parse a normal reply, which may contain a (dir+)dentry and/or a
294 * target inode.
295 */
296static int parse_reply_info_trace(void **p, void *end,
14303d20 297 struct ceph_mds_reply_info_parsed *info,
12b4629a 298 u64 features)
2f2dc053
SW		 299{
300 int err;
301
302 if (info->head->is_dentry) {
14303d20 303 err = parse_reply_info_in(p, end, &info->diri, features);
2f2dc053
SW		 304 if (err < 0)
305 goto out_bad;
306
b37fe1f9
YZ		 307 err = parse_reply_info_dir(p, end, &info->dirfrag, features);
308 if (err < 0)
309 goto out_bad;
2f2dc053
SW		 310
311 ceph_decode_32_safe(p, end, info->dname_len, bad);
312 ceph_decode_need(p, end, info->dname_len, bad);
313 info->dname = *p;
314 *p += info->dname_len;
b37fe1f9
YZ		 315
316 err = parse_reply_info_lease(p, end, &info->dlease, features);
317 if (err < 0)
318 goto out_bad;
2f2dc053
SW		 319 }
320
321 if (info->head->is_target) {
14303d20 322 err = parse_reply_info_in(p, end, &info->targeti, features);
2f2dc053
SW		 323 if (err < 0)
324 goto out_bad;
325 }
326
327 if (unlikely(*p != end))
328 goto bad;
329 return 0;
330
331bad:
332 err = -EIO;
333out_bad:
334 pr_err("problem parsing mds trace %d\n", err);
335 return err;
336}
337
338/*
339 * parse readdir results
340 */
b37fe1f9 341static int parse_reply_info_readdir(void **p, void *end,
14303d20 342 struct ceph_mds_reply_info_parsed *info,
12b4629a 343 u64 features)
2f2dc053
SW		 344{
345 u32 num, i = 0;
346 int err;
347
b37fe1f9
YZ		 348 err = parse_reply_info_dir(p, end, &info->dir_dir, features);
349 if (err < 0)
350 goto out_bad;
2f2dc053
SW		 351
352 ceph_decode_need(p, end, sizeof(num) + 2, bad);
c89136ea 353 num = ceph_decode_32(p);
956d39d6
YZ		 354 {
355 u16 flags = ceph_decode_16(p);
356 info->dir_end = !!(flags & CEPH_READDIR_FRAG_END);
357 info->dir_complete = !!(flags & CEPH_READDIR_FRAG_COMPLETE);
f3c4ebe6 358 info->hash_order = !!(flags & CEPH_READDIR_HASH_ORDER);
79162547 359 info->offset_hash = !!(flags & CEPH_READDIR_OFFSET_HASH);
956d39d6 360 }
2f2dc053
SW		 361 if (num == 0)
362 goto done;
363
2a5beea3
YZ		 364 BUG_ON(!info->dir_entries);
365 if ((unsigned long)(info->dir_entries + num) >
366 (unsigned long)info->dir_entries + info->dir_buf_size) {
54008399
YZ		 367 pr_err("dir contents are larger than expected\n");
368 WARN_ON(1);
369 goto bad;
370 }
2f2dc053 371
54008399 372 info->dir_nr = num;
2f2dc053 373 while (num) {
2a5beea3 374 struct ceph_mds_reply_dir_entry *rde = info->dir_entries + i;
2f2dc053 375 /* dentry */
b37fe1f9 376 ceph_decode_32_safe(p, end, rde->name_len, bad);
2a5beea3
YZ		 377 ceph_decode_need(p, end, rde->name_len, bad);
378 rde->name = *p;
379 *p += rde->name_len;
380 dout("parsed dir dname '%.*s'\n", rde->name_len, rde->name);
2f2dc053 381
b37fe1f9
YZ		 382 /* dentry lease */
383 err = parse_reply_info_lease(p, end, &rde->lease, features);
384 if (err)
385 goto out_bad;
2f2dc053 386 /* inode */
2a5beea3 387 err = parse_reply_info_in(p, end, &rde->inode, features);
2f2dc053
SW		 388 if (err < 0)
389 goto out_bad;
8974eebd
YZ		 390 /* ceph_readdir_prepopulate() will update it */
391 rde->offset = 0;
2f2dc053
SW		 392 i++;
393 num--;
394 }
395
396done:
1d3f8723
JL		 397 /* Skip over any unrecognized fields */
398 *p = end;
2f2dc053
SW		 399 return 0;
400
401bad:
402 err = -EIO;
403out_bad:
404 pr_err("problem parsing dir contents %d\n", err);
405 return err;
406}
407
25933abd
HS		 408/*
409 * parse fcntl F_GETLK results
410 */
411static int parse_reply_info_filelock(void **p, void *end,
14303d20 412 struct ceph_mds_reply_info_parsed *info,
12b4629a 413 u64 features)
25933abd
HS		 414{
415 if (*p + sizeof(*info->filelock_reply) > end)
416 goto bad;
417
418 info->filelock_reply = *p;
25933abd 419
1d3f8723
JL		 420 /* Skip over any unrecognized fields */
421 *p = end;
25933abd 422 return 0;
25933abd
HS		 423bad:
424 return -EIO;
425}
426
d4846487
JL		 427
428#if BITS_PER_LONG == 64
429
430#define DELEGATED_INO_AVAILABLE xa_mk_value(1)
431
432static int ceph_parse_deleg_inos(void **p, void *end,
433 struct ceph_mds_session *s)
434{
435 u32 sets;
436
437 ceph_decode_32_safe(p, end, sets, bad);
438 dout("got %u sets of delegated inodes\n", sets);
439 while (sets--) {
2ecd0edd 440 u64 start, len;
d4846487
JL		 441
442 ceph_decode_64_safe(p, end, start, bad);
443 ceph_decode_64_safe(p, end, len, bad);
d4f6b31d
JL		 444
445 /* Don't accept a delegation of system inodes */
446 if (start < CEPH_INO_SYSTEM_BASE) {
447 pr_warn_ratelimited("ceph: ignoring reserved inode range delegation (start=0x%llx len=0x%llx)\n",
448 start, len);
449 continue;
450 }
d4846487 451 while (len--) {
2ecd0edd 452 int err = xa_insert(&s->s_delegated_inos, start++,
d4846487
JL		 453 DELEGATED_INO_AVAILABLE,
454 GFP_KERNEL);
455 if (!err) {
456 dout("added delegated inode 0x%llx\n",
457 start - 1);
458 } else if (err == -EBUSY) {
4868e537 459 pr_warn("MDS delegated inode 0x%llx more than once.\n",
d4846487
JL		 460 start - 1);
461 } else {
462 return err;
463 }
464 }
465 }
466 return 0;
467bad:
468 return -EIO;
469}
470
471u64 ceph_get_deleg_ino(struct ceph_mds_session *s)
472{
473 unsigned long ino;
474 void *val;
475
476 xa_for_each(&s->s_delegated_inos, ino, val) {
477 val = xa_erase(&s->s_delegated_inos, ino);
478 if (val == DELEGATED_INO_AVAILABLE)
479 return ino;
480 }
481 return 0;
482}
483
484int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino)
485{
486 return xa_insert(&s->s_delegated_inos, ino, DELEGATED_INO_AVAILABLE,
487 GFP_KERNEL);
488}
489#else /* BITS_PER_LONG == 64 */
490/*
491 * FIXME: xarrays can't handle 64-bit indexes on a 32-bit arch. For now, just
492 * ignore delegated_inos on 32 bit arch. Maybe eventually add xarrays for top
493 * and bottom words?
494 */
495static int ceph_parse_deleg_inos(void **p, void *end,
496 struct ceph_mds_session *s)
497{
498 u32 sets;
499
500 ceph_decode_32_safe(p, end, sets, bad);
501 if (sets)
502 ceph_decode_skip_n(p, end, sets * 2 * sizeof(__le64), bad);
503 return 0;
504bad:
505 return -EIO;
506}
507
508u64 ceph_get_deleg_ino(struct ceph_mds_session *s)
509{
510 return 0;
511}
512
513int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino)
514{
515 return 0;
516}
517#endif /* BITS_PER_LONG == 64 */
518
6e8575fa
SL		 519/*
520 * parse create results
521 */
522static int parse_reply_info_create(void **p, void *end,
523 struct ceph_mds_reply_info_parsed *info,
d4846487 524 u64 features, struct ceph_mds_session *s)
6e8575fa 525{
d4846487
JL		 526 int ret;
527
b37fe1f9
YZ		 528 if (features == (u64)-1 ||
529 (features & CEPH_FEATURE_REPLY_CREATE_INODE)) {
6e8575fa 530 if (*p == end) {
d4846487 531 /* Malformed reply? */
6e8575fa 532 info->has_create_ino = false;
d4846487 533 } else if (test_bit(CEPHFS_FEATURE_DELEG_INO, &s->s_features)) {
6e8575fa 534 info->has_create_ino = true;
06a1ad43
JL		 535 /* struct_v, struct_compat, and len */
536 ceph_decode_skip_n(p, end, 2 + sizeof(u32), bad);
d4846487
JL		 537 ceph_decode_64_safe(p, end, info->ino, bad);
538 ret = ceph_parse_deleg_inos(p, end, s);
539 if (ret)
540 return ret;
541 } else {
542 /* legacy */
1d3f8723 543 ceph_decode_64_safe(p, end, info->ino, bad);
d4846487 544 info->has_create_ino = true;
6e8575fa 545 }
1d3f8723
JL		 546 } else {
547 if (*p != end)
548 goto bad;
6e8575fa
SL		 549 }
550
1d3f8723
JL		 551 /* Skip over any unrecognized fields */
552 *p = end;
6e8575fa 553 return 0;
6e8575fa
SL		 554bad:
555 return -EIO;
556}
557
6ddf5f16
MC		 558static int parse_reply_info_getvxattr(void **p, void *end,
559 struct ceph_mds_reply_info_parsed *info,
560 u64 features)
561{
562 u32 value_len;
563
564 ceph_decode_skip_8(p, end, bad); /* skip current version: 1 */
565 ceph_decode_skip_8(p, end, bad); /* skip first version: 1 */
566 ceph_decode_skip_32(p, end, bad); /* skip payload length */
567
568 ceph_decode_32_safe(p, end, value_len, bad);
569
570 if (value_len == end - *p) {
571 info->xattr_info.xattr_value = *p;
572 info->xattr_info.xattr_value_len = value_len;
573 *p = end;
574 return value_len;
575 }
576bad:
577 return -EIO;
578}
579
25933abd
HS		 580/*
581 * parse extra results
582 */
583static int parse_reply_info_extra(void **p, void *end,
14303d20 584 struct ceph_mds_reply_info_parsed *info,
d4846487 585 u64 features, struct ceph_mds_session *s)
25933abd 586{
6df8c9d8
JL		 587 u32 op = le32_to_cpu(info->head->op);
588
589 if (op == CEPH_MDS_OP_GETFILELOCK)
14303d20 590 return parse_reply_info_filelock(p, end, info, features);
6df8c9d8 591 else if (op == CEPH_MDS_OP_READDIR || op == CEPH_MDS_OP_LSSNAP)
b37fe1f9 592 return parse_reply_info_readdir(p, end, info, features);
6df8c9d8 593 else if (op == CEPH_MDS_OP_CREATE)
d4846487 594 return parse_reply_info_create(p, end, info, features, s);
6ddf5f16
MC		 595 else if (op == CEPH_MDS_OP_GETVXATTR)
596 return parse_reply_info_getvxattr(p, end, info, features);
6e8575fa
SL		 597 else
598 return -EIO;
25933abd
HS		 599}
600
2f2dc053
SW		 601/*
602 * parse entire mds reply
603 */
d4846487 604static int parse_reply_info(struct ceph_mds_session *s, struct ceph_msg *msg,
14303d20 605 struct ceph_mds_reply_info_parsed *info,
12b4629a 606 u64 features)
2f2dc053
SW		 607{
608 void *p, *end;
609 u32 len;
610 int err;
611
612 info->head = msg->front.iov_base;
613 p = msg->front.iov_base + sizeof(struct ceph_mds_reply_head);
614 end = p + msg->front.iov_len - sizeof(struct ceph_mds_reply_head);
615
616 /* trace */
617 ceph_decode_32_safe(&p, end, len, bad);
618 if (len > 0) {
32852a81 619 ceph_decode_need(&p, end, len, bad);
14303d20 620 err = parse_reply_info_trace(&p, p+len, info, features);
2f2dc053
SW		 621 if (err < 0)
622 goto out_bad;
623 }
624
25933abd 625 /* extra */
2f2dc053
SW		 626 ceph_decode_32_safe(&p, end, len, bad);
627 if (len > 0) {
32852a81 628 ceph_decode_need(&p, end, len, bad);
d4846487 629 err = parse_reply_info_extra(&p, p+len, info, features, s);
2f2dc053
SW		 630 if (err < 0)
631 goto out_bad;
632 }
633
634 /* snap blob */
635 ceph_decode_32_safe(&p, end, len, bad);
636 info->snapblob_len = len;
637 info->snapblob = p;
638 p += len;
639
640 if (p != end)
641 goto bad;
642 return 0;
643
644bad:
645 err = -EIO;
646out_bad:
647 pr_err("mds parse_reply err %d\n", err);
648 return err;
649}
650
651static void destroy_reply_info(struct ceph_mds_reply_info_parsed *info)
652{
2a5beea3 653 if (!info->dir_entries)
54008399 654 return;
2a5beea3 655 free_pages((unsigned long)info->dir_entries, get_order(info->dir_buf_size));
2f2dc053
SW		 656}
657
4868e537
XL		 658/*
659 * In async unlink case the kclient won't wait for the first reply
660 * from MDS and just drop all the links and unhash the dentry and then
661 * succeeds immediately.
662 *
663 * For any new create/link/rename,etc requests followed by using the
664 * same file names we must wait for the first reply of the inflight
665 * unlink request, or the MDS possibly will fail these following
666 * requests with -EEXIST if the inflight async unlink request was
667 * delayed for some reasons.
668 *
669 * And the worst case is that for the none async openc request it will
670 * successfully open the file if the CDentry hasn't been unlinked yet,
671 * but later the previous delayed async unlink request will remove the
672 * CDenty. That means the just created file is possiblly deleted later
673 * by accident.
674 *
675 * We need to wait for the inflight async unlink requests to finish
676 * when creating new files/directories by using the same file names.
677 */
678int ceph_wait_on_conflict_unlink(struct dentry *dentry)
679{
680 struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
681 struct dentry *pdentry = dentry->d_parent;
682 struct dentry *udentry, *found = NULL;
683 struct ceph_dentry_info *di;
684 struct qstr dname;
685 u32 hash = dentry->d_name.hash;
686 int err;
687
688 dname.name = dentry->d_name.name;
689 dname.len = dentry->d_name.len;
690
691 rcu_read_lock();
692 hash_for_each_possible_rcu(fsc->async_unlink_conflict, di,
693 hnode, hash) {
694 udentry = di->dentry;
695
696 spin_lock(&udentry->d_lock);
697 if (udentry->d_name.hash != hash)
698 goto next;
699 if (unlikely(udentry->d_parent != pdentry))
700 goto next;
701 if (!hash_hashed(&di->hnode))
702 goto next;
703
704 if (!test_bit(CEPH_DENTRY_ASYNC_UNLINK_BIT, &di->flags))
705 pr_warn("%s dentry %p:%pd async unlink bit is not set\n",
706 __func__, dentry, dentry);
707
708 if (!d_same_name(udentry, pdentry, &dname))
709 goto next;
710
711 spin_unlock(&udentry->d_lock);
712 found = dget(udentry);
713 break;
714next:
715 spin_unlock(&udentry->d_lock);
716 }
717 rcu_read_unlock();
718
719 if (likely(!found))
720 return 0;
721
722 dout("%s dentry %p:%pd conflict with old %p:%pd\n", __func__,
723 dentry, dentry, found, found);
724
725 err = wait_on_bit(&di->flags, CEPH_DENTRY_ASYNC_UNLINK_BIT,
726 TASK_KILLABLE);
727 dput(found);
728 return err;
729}
730
2f2dc053
SW		 731
732/*
733 * sessions
734 */
a687ecaf 735const char *ceph_session_state_name(int s)
2f2dc053
SW		 736{
737 switch (s) {
738 case CEPH_MDS_SESSION_NEW: return "new";
739 case CEPH_MDS_SESSION_OPENING: return "opening";
740 case CEPH_MDS_SESSION_OPEN: return "open";
741 case CEPH_MDS_SESSION_HUNG: return "hung";
742 case CEPH_MDS_SESSION_CLOSING: return "closing";
4d681c2f 743 case CEPH_MDS_SESSION_CLOSED: return "closed";
44ca18f2 744 case CEPH_MDS_SESSION_RESTARTING: return "restarting";
2f2dc053 745 case CEPH_MDS_SESSION_RECONNECTING: return "reconnecting";
fcff415c 746 case CEPH_MDS_SESSION_REJECTED: return "rejected";
2f2dc053
SW		 747 default: return "???";
748 }
749}
750
5b3248c6 751struct ceph_mds_session *ceph_get_mds_session(struct ceph_mds_session *s)
2f2dc053 752{
9f358999 753 if (refcount_inc_not_zero(&s->s_ref))
2f2dc053 754 return s;
9f358999 755 return NULL;
2f2dc053
SW		 756}
757
758void ceph_put_mds_session(struct ceph_mds_session *s)
759{
7e65624d
JL		 760 if (IS_ERR_OR_NULL(s))
761 return;
762
3997c01d 763 if (refcount_dec_and_test(&s->s_ref)) {
6c4a1915 764 if (s->s_auth.authorizer)
6c1ea260 765 ceph_auth_destroy_authorizer(s->s_auth.authorizer);
88828190 766 WARN_ON(mutex_is_locked(&s->s_mutex));
d4846487 767 xa_destroy(&s->s_delegated_inos);
2f2dc053 768 kfree(s);
4e7a5dcd 769 }
2f2dc053
SW		 770}
771
772/*
773 * called under mdsc->mutex
774 */
775struct ceph_mds_session *__ceph_lookup_mds_session(struct ceph_mds_client *mdsc,
776 int mds)
777{
d37b1d99 778 if (mds >= mdsc->max_sessions || !mdsc->sessions[mds])
2f2dc053 779 return NULL;
5b3248c6 780 return ceph_get_mds_session(mdsc->sessions[mds]);
2f2dc053
SW		 781}
782
783static bool __have_session(struct ceph_mds_client *mdsc, int mds)
784{
98cfda81 785 if (mds >= mdsc->max_sessions || !mdsc->sessions[mds])
2f2dc053 786 return false;
98cfda81
CX		 787 else
788 return true;
2f2dc053
SW		 789}
790
2600d2dd
SW		 791static int __verify_registered_session(struct ceph_mds_client *mdsc,
792 struct ceph_mds_session *s)
793{
794 if (s->s_mds >= mdsc->max_sessions ||
795 mdsc->sessions[s->s_mds] != s)
796 return -ENOENT;
797 return 0;
798}
799
2f2dc053
SW		 800/*
801 * create+register a new session for given mds.
802 * called under mdsc->mutex.
803 */
804static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
805 int mds)
806{
807 struct ceph_mds_session *s;
808
b38c9eb4 809 if (mds >= mdsc->mdsmap->possible_max_rank)
c338c07c
NY		 810 return ERR_PTR(-EINVAL);
811
2f2dc053 812 s = kzalloc(sizeof(*s), GFP_NOFS);
4736b009
DC		 813 if (!s)
814 return ERR_PTR(-ENOMEM);
47474d0b
CX		 815
816 if (mds >= mdsc->max_sessions) {
817 int newmax = 1 << get_count_order(mds + 1);
818 struct ceph_mds_session **sa;
819
820 dout("%s: realloc to %d\n", __func__, newmax);
821 sa = kcalloc(newmax, sizeof(void *), GFP_NOFS);
822 if (!sa)
823 goto fail_realloc;
824 if (mdsc->sessions) {
825 memcpy(sa, mdsc->sessions,
826 mdsc->max_sessions * sizeof(void *));
827 kfree(mdsc->sessions);
828 }
829 mdsc->sessions = sa;
830 mdsc->max_sessions = newmax;
831 }
832
833 dout("%s: mds%d\n", __func__, mds);
2f2dc053
SW		 834 s->s_mdsc = mdsc;
835 s->s_mds = mds;
836 s->s_state = CEPH_MDS_SESSION_NEW;
2f2dc053
SW		 837 mutex_init(&s->s_mutex);
838
b7a9e5dd 839 ceph_con_init(&s->s_con, s, &mds_con_ops, &mdsc->fsc->client->msgr);
2f2dc053 840
52d60f8e 841 atomic_set(&s->s_cap_gen, 1);
1ce208a6 842 s->s_cap_ttl = jiffies - 1;
d8fb02ab
AE		 843
844 spin_lock_init(&s->s_cap_lock);
2f2dc053 845 INIT_LIST_HEAD(&s->s_caps);
3997c01d 846 refcount_set(&s->s_ref, 1);
2f2dc053
SW		 847 INIT_LIST_HEAD(&s->s_waiting);
848 INIT_LIST_HEAD(&s->s_unsafe);
d4846487 849 xa_init(&s->s_delegated_inos);
2f2dc053 850 INIT_LIST_HEAD(&s->s_cap_releases);
e3ec8d68
YZ		 851 INIT_WORK(&s->s_cap_release_work, ceph_cap_release_work);
852
1cf03a68 853 INIT_LIST_HEAD(&s->s_cap_dirty);
2f2dc053 854 INIT_LIST_HEAD(&s->s_cap_flushing);
2f2dc053 855
2f2dc053 856 mdsc->sessions[mds] = s;
86d8f67b 857 atomic_inc(&mdsc->num_sessions);
3997c01d 858 refcount_inc(&s->s_ref); /* one ref to sessions[], one to caller */
42ce56e5 859
b7a9e5dd
SW		 860 ceph_con_open(&s->s_con, CEPH_ENTITY_TYPE_MDS, mds,
861 ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
42ce56e5 862
2f2dc053 863 return s;
42ce56e5
SW		 864
865fail_realloc:
866 kfree(s);
867 return ERR_PTR(-ENOMEM);
2f2dc053
SW		 868}
869
870/*
871 * called under mdsc->mutex
872 */
2600d2dd 873static void __unregister_session(struct ceph_mds_client *mdsc,
42ce56e5 874 struct ceph_mds_session *s)
2f2dc053 875{
2600d2dd
SW		 876 dout("__unregister_session mds%d %p\n", s->s_mds, s);
877 BUG_ON(mdsc->sessions[s->s_mds] != s);
42ce56e5
SW		 878 mdsc->sessions[s->s_mds] = NULL;
879 ceph_con_close(&s->s_con);
880 ceph_put_mds_session(s);
86d8f67b 881 atomic_dec(&mdsc->num_sessions);
2f2dc053
SW		 882}
883
884/*
885 * drop session refs in request.
886 *
887 * should be last request ref, or hold mdsc->mutex
888 */
889static void put_request_session(struct ceph_mds_request *req)
890{
891 if (req->r_session) {
892 ceph_put_mds_session(req->r_session);
893 req->r_session = NULL;
894 }
895}
896
59b312f3
XL		 897void ceph_mdsc_iterate_sessions(struct ceph_mds_client *mdsc,
898 void (*cb)(struct ceph_mds_session *),
899 bool check_state)
900{
901 int mds;
902
903 mutex_lock(&mdsc->mutex);
904 for (mds = 0; mds < mdsc->max_sessions; ++mds) {
905 struct ceph_mds_session *s;
906
907 s = __ceph_lookup_mds_session(mdsc, mds);
908 if (!s)
909 continue;
910
911 if (check_state && !check_session_state(s)) {
912 ceph_put_mds_session(s);
913 continue;
914 }
915
916 mutex_unlock(&mdsc->mutex);
917 cb(s);
918 ceph_put_mds_session(s);
919 mutex_lock(&mdsc->mutex);
920 }
921 mutex_unlock(&mdsc->mutex);
922}
923
153c8e6b 924void ceph_mdsc_release_request(struct kref *kref)
2f2dc053 925{
153c8e6b
SW		 926 struct ceph_mds_request *req = container_of(kref,
927 struct ceph_mds_request,
928 r_kref);
e64f44a8 929 ceph_mdsc_release_dir_caps_no_check(req);
54008399 930 destroy_reply_info(&req->r_reply_info);
153c8e6b
SW		 931 if (req->r_request)
932 ceph_msg_put(req->r_request);
54008399 933 if (req->r_reply)
153c8e6b 934 ceph_msg_put(req->r_reply);
153c8e6b 935 if (req->r_inode) {
41b02e1f 936 ceph_put_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
23c2c76e 937 iput(req->r_inode);
153c8e6b 938 }
9c1c2b35 939 if (req->r_parent) {
3dd69aab 940 ceph_put_cap_refs(ceph_inode(req->r_parent), CEPH_CAP_PIN);
23c2c76e 941 iput(req->r_parent);
9c1c2b35 942 }
23c2c76e 943 iput(req->r_target_inode);
153c8e6b
SW		 944 if (req->r_dentry)
945 dput(req->r_dentry);
844d87c3
SW		 946 if (req->r_old_dentry)
947 dput(req->r_old_dentry);
948 if (req->r_old_dentry_dir) {
41b02e1f
SW		 949 /*
950 * track (and drop pins for) r_old_dentry_dir
951 * separately, since r_old_dentry's d_parent may have
952 * changed between the dir mutex being dropped and
953 * this request being freed.
954 */
955 ceph_put_cap_refs(ceph_inode(req->r_old_dentry_dir),
956 CEPH_CAP_PIN);
23c2c76e 957 iput(req->r_old_dentry_dir);
2f2dc053 958 }
153c8e6b
SW		 959 kfree(req->r_path1);
960 kfree(req->r_path2);
7fe0cdeb 961 put_cred(req->r_cred);
25e6bae3
YZ		 962 if (req->r_pagelist)
963 ceph_pagelist_release(req->r_pagelist);
153c8e6b 964 put_request_session(req);
37151668 965 ceph_unreserve_caps(req->r_mdsc, &req->r_caps_reservation);
428138c9 966 WARN_ON_ONCE(!list_empty(&req->r_wait));
058daab7 967 kmem_cache_free(ceph_mds_request_cachep, req);
2f2dc053
SW		 968}
969
fcd00b68
ID		 970DEFINE_RB_FUNCS(request, struct ceph_mds_request, r_tid, r_node)
971
2f2dc053
SW		 972/*
973 * lookup session, bump ref if found.
974 *
975 * called under mdsc->mutex.
976 */
fcd00b68
ID		 977static struct ceph_mds_request *
978lookup_get_request(struct ceph_mds_client *mdsc, u64 tid)
2f2dc053
SW		 979{
980 struct ceph_mds_request *req;
44ca18f2 981
fcd00b68
ID		 982 req = lookup_request(&mdsc->request_tree, tid);
983 if (req)
984 ceph_mdsc_get_request(req);
44ca18f2 985
fcd00b68 986 return req;
2f2dc053
SW		 987}
988
989/*
990 * Register an in-flight request, and assign a tid. Link to directory
991 * are modifying (if any).
992 *
993 * Called under mdsc->mutex.
994 */
995static void __register_request(struct ceph_mds_client *mdsc,
996 struct ceph_mds_request *req,
997 struct inode *dir)
998{
e30ee581
ZZ		 999 int ret = 0;
1000
2f2dc053 1001 req->r_tid = ++mdsc->last_tid;
e30ee581
ZZ		 1002 if (req->r_num_caps) {
1003 ret = ceph_reserve_caps(mdsc, &req->r_caps_reservation,
1004 req->r_num_caps);
1005 if (ret < 0) {
1006 pr_err("__register_request %p "
1007 "failed to reserve caps: %d\n", req, ret);
1008 /* set req->r_err to fail early from __do_request */
1009 req->r_err = ret;
1010 return;
1011 }
1012 }
2f2dc053
SW		 1013 dout("__register_request %p tid %lld\n", req, req->r_tid);
1014 ceph_mdsc_get_request(req);
fcd00b68 1015 insert_request(&mdsc->request_tree, req);
2f2dc053 1016
7fe0cdeb 1017 req->r_cred = get_current_cred();
cb4276cc 1018
e8a7b8b1
YZ		 1019 if (mdsc->oldest_tid == 0 && req->r_op != CEPH_MDS_OP_SETFILELOCK)
1020 mdsc->oldest_tid = req->r_tid;
1021
2f2dc053 1022 if (dir) {
3db0a2fc
JL		 1023 struct ceph_inode_info *ci = ceph_inode(dir);
1024
3b663780 1025 ihold(dir);
2f2dc053 1026 req->r_unsafe_dir = dir;
3db0a2fc
JL		 1027 spin_lock(&ci->i_unsafe_lock);
1028 list_add_tail(&req->r_unsafe_dir_item, &ci->i_unsafe_dirops);
1029 spin_unlock(&ci->i_unsafe_lock);
2f2dc053
SW		 1030 }
1031}
1032
1033static void __unregister_request(struct ceph_mds_client *mdsc,
1034 struct ceph_mds_request *req)
1035{
1036 dout("__unregister_request %p tid %lld\n", req, req->r_tid);
e8a7b8b1 1037
df963ea8
JL		 1038 /* Never leave an unregistered request on an unsafe list! */
1039 list_del_init(&req->r_unsafe_item);
1040
e8a7b8b1
YZ		 1041 if (req->r_tid == mdsc->oldest_tid) {
1042 struct rb_node *p = rb_next(&req->r_node);
1043 mdsc->oldest_tid = 0;
1044 while (p) {
1045 struct ceph_mds_request *next_req =
1046 rb_entry(p, struct ceph_mds_request, r_node);
1047 if (next_req->r_op != CEPH_MDS_OP_SETFILELOCK) {
1048 mdsc->oldest_tid = next_req->r_tid;
1049 break;
1050 }
1051 p = rb_next(p);
1052 }
1053 }
1054
fcd00b68 1055 erase_request(&mdsc->request_tree, req);
2f2dc053 1056
3db0a2fc 1057 if (req->r_unsafe_dir) {
2f2dc053 1058 struct ceph_inode_info *ci = ceph_inode(req->r_unsafe_dir);
2f2dc053
SW		 1059 spin_lock(&ci->i_unsafe_lock);
1060 list_del_init(&req->r_unsafe_dir_item);
1061 spin_unlock(&ci->i_unsafe_lock);
4c06ace8 1062 }
bc2de10d
JL		 1063 if (req->r_target_inode &&
1064 test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
68cd5b4b
YZ		 1065 struct ceph_inode_info *ci = ceph_inode(req->r_target_inode);
1066 spin_lock(&ci->i_unsafe_lock);
1067 list_del_init(&req->r_unsafe_target_item);
1068 spin_unlock(&ci->i_unsafe_lock);
1069 }
3b663780 1070
4c06ace8 1071 if (req->r_unsafe_dir) {
23c2c76e 1072 iput(req->r_unsafe_dir);
3b663780 1073 req->r_unsafe_dir = NULL;
2f2dc053 1074 }
94aa8ae1 1075
fc55d2c9
YZ		 1076 complete_all(&req->r_safe_completion);
1077
94aa8ae1 1078 ceph_mdsc_put_request(req);
2f2dc053
SW		 1079}
1080
30c71233
JL		 1081/*
1082 * Walk back up the dentry tree until we hit a dentry representing a
1083 * non-snapshot inode. We do this using the rcu_read_lock (which must be held
1084 * when calling this) to ensure that the objects won't disappear while we're
1085 * working with them. Once we hit a candidate dentry, we attempt to take a
1086 * reference to it, and return that as the result.
1087 */
f1075480
DC		 1088static struct inode *get_nonsnap_parent(struct dentry *dentry)
1089{
1090 struct inode *inode = NULL;
30c71233
JL		 1091
1092 while (dentry && !IS_ROOT(dentry)) {
1093 inode = d_inode_rcu(dentry);
1094 if (!inode || ceph_snap(inode) == CEPH_NOSNAP)
1095 break;
1096 dentry = dentry->d_parent;
1097 }
1098 if (inode)
1099 inode = igrab(inode);
1100 return inode;
1101}
1102
2f2dc053
SW		 1103/*
1104 * Choose mds to send request to next. If there is a hint set in the
1105 * request (e.g., due to a prior forward hint from the mds), use that.
1106 * Otherwise, consult frag tree and/or caps to identify the
1107 * appropriate mds. If all else fails, choose randomly.
1108 *
1109 * Called under mdsc->mutex.
1110 */
1111static int __choose_mds(struct ceph_mds_client *mdsc,
c4853e97
XL		 1112 struct ceph_mds_request *req,
1113 bool *random)
2f2dc053
SW		 1114{
1115 struct inode *inode;
1116 struct ceph_inode_info *ci;
1117 struct ceph_cap *cap;
1118 int mode = req->r_direct_mode;
1119 int mds = -1;
1120 u32 hash = req->r_direct_hash;
bc2de10d 1121 bool is_hash = test_bit(CEPH_MDS_R_DIRECT_IS_HASH, &req->r_req_flags);
2f2dc053 1122
c4853e97
XL		 1123 if (random)
1124 *random = false;
1125
2f2dc053
SW		 1126 /*
1127 * is there a specific mds we should try? ignore hint if we have
1128 * no session and the mds is not up (active or recovering).
1129 */
1130 if (req->r_resend_mds >= 0 &&
1131 (__have_session(mdsc, req->r_resend_mds) ||
1132 ceph_mdsmap_get_state(mdsc->mdsmap, req->r_resend_mds) > 0)) {
3c802092 1133 dout("%s using resend_mds mds%d\n", __func__,
2f2dc053
SW		 1134 req->r_resend_mds);
1135 return req->r_resend_mds;
1136 }
1137
1138 if (mode == USE_RANDOM_MDS)
1139 goto random;
1140
1141 inode = NULL;
1142 if (req->r_inode) {
5d37ca14
YZ		 1143 if (ceph_snap(req->r_inode) != CEPH_SNAPDIR) {
1144 inode = req->r_inode;
1145 ihold(inode);
1146 } else {
38f340cc
YZ		 1147 /* req->r_dentry is non-null for LSSNAP request */
1148 rcu_read_lock();
1149 inode = get_nonsnap_parent(req->r_dentry);
1150 rcu_read_unlock();
3c802092 1151 dout("%s using snapdir's parent %p\n", __func__, inode);
5d37ca14 1152 }
38f340cc 1153 } else if (req->r_dentry) {
d79698da 1154 /* ignore race with rename; old or new d_parent is okay */
30c71233
JL		 1155 struct dentry *parent;
1156 struct inode *dir;
1157
1158 rcu_read_lock();
41883ba8 1159 parent = READ_ONCE(req->r_dentry->d_parent);
3dd69aab 1160 dir = req->r_parent ? : d_inode_rcu(parent);
eb6bb1c5 1161
30c71233
JL		 1162 if (!dir || dir->i_sb != mdsc->fsc->sb) {
1163 /* not this fs or parent went negative */
2b0143b5 1164 inode = d_inode(req->r_dentry);
30c71233
JL		 1165 if (inode)
1166 ihold(inode);
eb6bb1c5
SW		 1167 } else if (ceph_snap(dir) != CEPH_NOSNAP) {
1168 /* direct snapped/virtual snapdir requests
1169 * based on parent dir inode */
30c71233 1170 inode = get_nonsnap_parent(parent);
3c802092 1171 dout("%s using nonsnap parent %p\n", __func__, inode);
ca18bede 1172 } else {
eb6bb1c5 1173 /* dentry target */
2b0143b5 1174 inode = d_inode(req->r_dentry);
ca18bede
YZ		 1175 if (!inode || mode == USE_AUTH_MDS) {
1176 /* dir + name */
30c71233 1177 inode = igrab(dir);
ca18bede
YZ		 1178 hash = ceph_dentry_hash(dir, req->r_dentry);
1179 is_hash = true;
30c71233
JL		 1180 } else {
1181 ihold(inode);
ca18bede 1182 }
2f2dc053 1183 }
30c71233 1184 rcu_read_unlock();
2f2dc053 1185 }
eb6bb1c5 1186
3c802092
XL		 1187 dout("%s %p is_hash=%d (0x%x) mode %d\n", __func__, inode, (int)is_hash,
1188 hash, mode);
2f2dc053
SW		 1189 if (!inode)
1190 goto random;
1191 ci = ceph_inode(inode);
1192
1193 if (is_hash && S_ISDIR(inode->i_mode)) {
1194 struct ceph_inode_frag frag;
1195 int found;
1196
1197 ceph_choose_frag(ci, hash, &frag, &found);
1198 if (found) {
1199 if (mode == USE_ANY_MDS && frag.ndist > 0) {
1200 u8 r;
1201
1202 /* choose a random replica */
1203 get_random_bytes(&r, 1);
1204 r %= frag.ndist;
1205 mds = frag.dist[r];
3c802092
XL		 1206 dout("%s %p %llx.%llx frag %u mds%d (%d/%d)\n",
1207 __func__, inode, ceph_vinop(inode),
1208 frag.frag, mds, (int)r, frag.ndist);
d66bbd44 1209 if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
5d47648f
XL		 1210 CEPH_MDS_STATE_ACTIVE &&
1211 !ceph_mdsmap_is_laggy(mdsc->mdsmap, mds))
30c71233 1212 goto out;
2f2dc053
SW		 1213 }
1214
1215 /* since this file/dir wasn't known to be
1216 * replicated, then we want to look for the
1217 * authoritative mds. */
2f2dc053
SW		 1218 if (frag.mds >= 0) {
1219 /* choose auth mds */
1220 mds = frag.mds;
3c802092
XL		 1221 dout("%s %p %llx.%llx frag %u mds%d (auth)\n",
1222 __func__, inode, ceph_vinop(inode),
1223 frag.frag, mds);
d66bbd44 1224 if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
5d47648f 1225 CEPH_MDS_STATE_ACTIVE) {
224c7b67 1226 if (!ceph_mdsmap_is_laggy(mdsc->mdsmap,
5d47648f
XL		 1227 mds))
1228 goto out;
1229 }
2f2dc053 1230 }
5d47648f 1231 mode = USE_AUTH_MDS;
2f2dc053
SW		 1232 }
1233 }
1234
be655596 1235 spin_lock(&ci->i_ceph_lock);
2f2dc053
SW		 1236 cap = NULL;
1237 if (mode == USE_AUTH_MDS)
1238 cap = ci->i_auth_cap;
1239 if (!cap && !RB_EMPTY_ROOT(&ci->i_caps))
1240 cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node);
1241 if (!cap) {
be655596 1242 spin_unlock(&ci->i_ceph_lock);
23c2c76e 1243 iput(inode);
2f2dc053
SW		 1244 goto random;
1245 }
1246 mds = cap->session->s_mds;
3c802092 1247 dout("%s %p %llx.%llx mds%d (%scap %p)\n", __func__,
2f2dc053
SW		 1248 inode, ceph_vinop(inode), mds,
1249 cap == ci->i_auth_cap ? "auth " : "", cap);
be655596 1250 spin_unlock(&ci->i_ceph_lock);
30c71233 1251out:
23c2c76e 1252 iput(inode);
2f2dc053
SW		 1253 return mds;
1254
1255random:
c4853e97
XL		 1256 if (random)
1257 *random = true;
1258
2f2dc053 1259 mds = ceph_mdsmap_get_random_mds(mdsc->mdsmap);
3c802092 1260 dout("%s chose random mds%d\n", __func__, mds);
2f2dc053
SW		 1261 return mds;
1262}
1263
1264
1265/*
1266 * session messages
1267 */
fba97e80 1268struct ceph_msg *ceph_create_session_msg(u32 op, u64 seq)
2f2dc053
SW		 1269{
1270 struct ceph_msg *msg;
1271 struct ceph_mds_session_head *h;
1272
b61c2763
SW		 1273 msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h), GFP_NOFS,
1274 false);
a79832f2 1275 if (!msg) {
fba97e80
XL		 1276 pr_err("ENOMEM creating session %s msg\n",
1277 ceph_session_op_name(op));
a79832f2 1278 return NULL;
2f2dc053
SW		 1279 }
1280 h = msg->front.iov_base;
1281 h->op = cpu_to_le32(op);
1282 h->seq = cpu_to_le64(seq);
dbd0c8bf
JS		 1283
1284 return msg;
1285}
1286
9ba1e224
XL		 1287static const unsigned char feature_bits[] = CEPHFS_FEATURES_CLIENT_SUPPORTED;
1288#define FEATURE_BYTES(c) (DIV_ROUND_UP((size_t)feature_bits[c - 1] + 1, 64) * 8)
b682c6d4 1289static int encode_supported_features(void **p, void *end)
342ce182 1290{
9ba1e224 1291 static const size_t count = ARRAY_SIZE(feature_bits);
342ce182
YZ		 1292
1293 if (count > 0) {
1294 size_t i;
9ba1e224 1295 size_t size = FEATURE_BYTES(count);
fea013e0 1296 unsigned long bit;
342ce182 1297
b682c6d4
XL		 1298 if (WARN_ON_ONCE(*p + 4 + size > end))
1299 return -ERANGE;
1300
342ce182
YZ		 1301 ceph_encode_32(p, size);
1302 memset(*p, 0, size);
fea013e0
LH		 1303 for (i = 0; i < count; i++) {
1304 bit = feature_bits[i];
1305 ((unsigned char *)(*p))[bit / 8] |= BIT(bit % 8);
1306 }
342ce182
YZ		 1307 *p += size;
1308 } else {
b682c6d4
XL		 1309 if (WARN_ON_ONCE(*p + 4 > end))
1310 return -ERANGE;
1311
342ce182
YZ		 1312 ceph_encode_32(p, 0);
1313 }
b682c6d4
XL		 1314
1315 return 0;
342ce182
YZ		 1316}
1317
3b4168dd
XL		 1318static const unsigned char metric_bits[] = CEPHFS_METRIC_SPEC_CLIENT_SUPPORTED;
1319#define METRIC_BYTES(cnt) (DIV_ROUND_UP((size_t)metric_bits[cnt - 1] + 1, 64) * 8)
1320static int encode_metric_spec(void **p, void *end)
1321{
1322 static const size_t count = ARRAY_SIZE(metric_bits);
1323
1324 /* header */
1325 if (WARN_ON_ONCE(*p + 2 > end))
1326 return -ERANGE;
1327
1328 ceph_encode_8(p, 1); /* version */
1329 ceph_encode_8(p, 1); /* compat */
1330
1331 if (count > 0) {
1332 size_t i;
1333 size_t size = METRIC_BYTES(count);
1334
1335 if (WARN_ON_ONCE(*p + 4 + 4 + size > end))
1336 return -ERANGE;
1337
1338 /* metric spec info length */
1339 ceph_encode_32(p, 4 + size);
1340
1341 /* metric spec */
1342 ceph_encode_32(p, size);
1343 memset(*p, 0, size);
1344 for (i = 0; i < count; i++)
1345 ((unsigned char *)(*p))[i / 8] |= BIT(metric_bits[i] % 8);
1346 *p += size;
1347 } else {
1348 if (WARN_ON_ONCE(*p + 4 + 4 > end))
1349 return -ERANGE;
1350
1351 /* metric spec info length */
1352 ceph_encode_32(p, 4);
1353 /* metric spec */
1354 ceph_encode_32(p, 0);
1355 }
1356
1357 return 0;
1358}
1359
dbd0c8bf
JS		 1360/*
1361 * session message, specialization for CEPH_SESSION_REQUEST_OPEN
1362 * to include additional client metadata fields.
1363 */
1364static struct ceph_msg *create_session_open_msg(struct ceph_mds_client *mdsc, u64 seq)
1365{
1366 struct ceph_msg *msg;
1367 struct ceph_mds_session_head *h;
4a756db2 1368 int i;
342ce182 1369 int extra_bytes = 0;
dbd0c8bf
JS		 1370 int metadata_key_count = 0;
1371 struct ceph_options *opt = mdsc->fsc->client->options;
3f384954 1372 struct ceph_mount_options *fsopt = mdsc->fsc->mount_options;
9ba1e224 1373 size_t size, count;
342ce182 1374 void *p, *end;
b682c6d4 1375 int ret;
dbd0c8bf 1376
a6a5ce4f 1377 const char* metadata[][2] = {
717e6f28
YZ		 1378 {"hostname", mdsc->nodename},
1379 {"kernel_version", init_utsname()->release},
3f384954
YZ		 1380 {"entity_id", opt->name ? : ""},
1381 {"root", fsopt->server_path ? : "/"},
dbd0c8bf
JS		 1382 {NULL, NULL}
1383 };
1384
1385 /* Calculate serialized length of metadata */
342ce182 1386 extra_bytes = 4; /* map length */
d37b1d99 1387 for (i = 0; metadata[i][0]; ++i) {
342ce182 1388 extra_bytes += 8 + strlen(metadata[i][0]) +
dbd0c8bf
JS		 1389 strlen(metadata[i][1]);
1390 metadata_key_count++;
1391 }
9ba1e224 1392
342ce182 1393 /* supported feature */
9ba1e224
XL		 1394 size = 0;
1395 count = ARRAY_SIZE(feature_bits);
1396 if (count > 0)
1397 size = FEATURE_BYTES(count);
1398 extra_bytes += 4 + size;
dbd0c8bf 1399
3b4168dd
XL		 1400 /* metric spec */
1401 size = 0;
1402 count = ARRAY_SIZE(metric_bits);
1403 if (count > 0)
1404 size = METRIC_BYTES(count);
1405 extra_bytes += 2 + 4 + 4 + size;
1406
dbd0c8bf 1407 /* Allocate the message */
342ce182 1408 msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h) + extra_bytes,
dbd0c8bf
JS		 1409 GFP_NOFS, false);
1410 if (!msg) {
fba97e80 1411 pr_err("ENOMEM creating session open msg\n");
b682c6d4 1412 return ERR_PTR(-ENOMEM);
dbd0c8bf 1413 }
342ce182
YZ		 1414 p = msg->front.iov_base;
1415 end = p + msg->front.iov_len;
1416
1417 h = p;
dbd0c8bf
JS		 1418 h->op = cpu_to_le32(CEPH_SESSION_REQUEST_OPEN);
1419 h->seq = cpu_to_le64(seq);
1420
1421 /*
1422 * Serialize client metadata into waiting buffer space, using
1423 * the format that userspace expects for map<string, string>
7cfa0313 1424 *
3b4168dd 1425 * ClientSession messages with metadata are v4
dbd0c8bf 1426 */
3b4168dd 1427 msg->hdr.version = cpu_to_le16(4);
7cfa0313 1428 msg->hdr.compat_version = cpu_to_le16(1);
dbd0c8bf
JS		 1429
1430 /* The write pointer, following the session_head structure */
342ce182 1431 p += sizeof(*h);
dbd0c8bf
JS		 1432
1433 /* Number of entries in the map */
1434 ceph_encode_32(&p, metadata_key_count);
1435
1436 /* Two length-prefixed strings for each entry in the map */
d37b1d99 1437 for (i = 0; metadata[i][0]; ++i) {
dbd0c8bf
JS		 1438 size_t const key_len = strlen(metadata[i][0]);
1439 size_t const val_len = strlen(metadata[i][1]);
1440
1441 ceph_encode_32(&p, key_len);
1442 memcpy(p, metadata[i][0], key_len);
1443 p += key_len;
1444 ceph_encode_32(&p, val_len);
1445 memcpy(p, metadata[i][1], val_len);
1446 p += val_len;
1447 }
1448
b682c6d4
XL		 1449 ret = encode_supported_features(&p, end);
1450 if (ret) {
1451 pr_err("encode_supported_features failed!\n");
1452 ceph_msg_put(msg);
1453 return ERR_PTR(ret);
1454 }
1455
3b4168dd
XL		 1456 ret = encode_metric_spec(&p, end);
1457 if (ret) {
1458 pr_err("encode_metric_spec failed!\n");
1459 ceph_msg_put(msg);
1460 return ERR_PTR(ret);
1461 }
1462
342ce182
YZ		 1463 msg->front.iov_len = p - msg->front.iov_base;
1464 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
1465
2f2dc053
SW		 1466 return msg;
1467}
1468
1469/*
1470 * send session open request.
1471 *
1472 * called under mdsc->mutex
1473 */
1474static int __open_session(struct ceph_mds_client *mdsc,
1475 struct ceph_mds_session *session)
1476{
1477 struct ceph_msg *msg;
1478 int mstate;
1479 int mds = session->s_mds;
2f2dc053
SW		 1480
1481 /* wait for mds to go active? */
1482 mstate = ceph_mdsmap_get_state(mdsc->mdsmap, mds);
1483 dout("open_session to mds%d (%s)\n", mds,
1484 ceph_mds_state_name(mstate));
1485 session->s_state = CEPH_MDS_SESSION_OPENING;
1486 session->s_renew_requested = jiffies;
1487
1488 /* send connect message */
dbd0c8bf 1489 msg = create_session_open_msg(mdsc, session->s_seq);
b682c6d4
XL		 1490 if (IS_ERR(msg))
1491 return PTR_ERR(msg);
2f2dc053 1492 ceph_con_send(&session->s_con, msg);
2f2dc053
SW		 1493 return 0;
1494}
1495
ed0552a1
SW		 1496/*
1497 * open sessions for any export targets for the given mds
1498 *
1499 * called under mdsc->mutex
1500 */
5d72d13c
YZ		 1501static struct ceph_mds_session *
1502__open_export_target_session(struct ceph_mds_client *mdsc, int target)
1503{
1504 struct ceph_mds_session *session;
b682c6d4 1505 int ret;
5d72d13c
YZ		 1506
1507 session = __ceph_lookup_mds_session(mdsc, target);
1508 if (!session) {
1509 session = register_session(mdsc, target);
1510 if (IS_ERR(session))
1511 return session;
1512 }
1513 if (session->s_state == CEPH_MDS_SESSION_NEW ||
b682c6d4
XL		 1514 session->s_state == CEPH_MDS_SESSION_CLOSING) {
1515 ret = __open_session(mdsc, session);
1516 if (ret)
1517 return ERR_PTR(ret);
1518 }
5d72d13c
YZ		 1519
1520 return session;
1521}
1522
1523struct ceph_mds_session *
1524ceph_mdsc_open_export_target_session(struct ceph_mds_client *mdsc, int target)
1525{
1526 struct ceph_mds_session *session;
1527
1528 dout("open_export_target_session to mds%d\n", target);
1529
1530 mutex_lock(&mdsc->mutex);
1531 session = __open_export_target_session(mdsc, target);
1532 mutex_unlock(&mdsc->mutex);
1533
1534 return session;
1535}
1536
ed0552a1
SW		 1537static void __open_export_target_sessions(struct ceph_mds_client *mdsc,
1538 struct ceph_mds_session *session)
1539{
1540 struct ceph_mds_info *mi;
1541 struct ceph_mds_session *ts;
1542 int i, mds = session->s_mds;
ed0552a1 1543
b38c9eb4 1544 if (mds >= mdsc->mdsmap->possible_max_rank)
ed0552a1 1545 return;
5d72d13c 1546
ed0552a1
SW		 1547 mi = &mdsc->mdsmap->m_info[mds];
1548 dout("open_export_target_sessions for mds%d (%d targets)\n",
1549 session->s_mds, mi->num_export_targets);
1550
1551 for (i = 0; i < mi->num_export_targets; i++) {
5d72d13c 1552 ts = __open_export_target_session(mdsc, mi->export_targets[i]);
7e65624d 1553 ceph_put_mds_session(ts);
ed0552a1
SW		 1554 }
1555}
1556
154f42c2
SW		 1557void ceph_mdsc_open_export_target_sessions(struct ceph_mds_client *mdsc,
1558 struct ceph_mds_session *session)
1559{
1560 mutex_lock(&mdsc->mutex);
1561 __open_export_target_sessions(mdsc, session);
1562 mutex_unlock(&mdsc->mutex);
1563}
1564
2f2dc053
SW		 1565/*
1566 * session caps
1567 */
1568
c8a96a31
JL		 1569static void detach_cap_releases(struct ceph_mds_session *session,
1570 struct list_head *target)
2f2dc053 1571{
c8a96a31
JL		 1572 lockdep_assert_held(&session->s_cap_lock);
1573
1574 list_splice_init(&session->s_cap_releases, target);
745a8e3b 1575 session->s_num_cap_releases = 0;
c8a96a31
JL		 1576 dout("dispose_cap_releases mds%d\n", session->s_mds);
1577}
2f2dc053 1578
c8a96a31
JL		 1579static void dispose_cap_releases(struct ceph_mds_client *mdsc,
1580 struct list_head *dispose)
1581{
1582 while (!list_empty(dispose)) {
745a8e3b
YZ		 1583 struct ceph_cap *cap;
1584 /* zero out the in-progress message */
c8a96a31 1585 cap = list_first_entry(dispose, struct ceph_cap, session_caps);
745a8e3b
YZ		 1586 list_del(&cap->session_caps);
1587 ceph_put_cap(mdsc, cap);
2f2dc053 1588 }
2f2dc053
SW		 1589}
1590
1c841a96
YZ		 1591static void cleanup_session_requests(struct ceph_mds_client *mdsc,
1592 struct ceph_mds_session *session)
1593{
1594 struct ceph_mds_request *req;
1595 struct rb_node *p;
1596
1597 dout("cleanup_session_requests mds%d\n", session->s_mds);
1598 mutex_lock(&mdsc->mutex);
1599 while (!list_empty(&session->s_unsafe)) {
1600 req = list_first_entry(&session->s_unsafe,
1601 struct ceph_mds_request, r_unsafe_item);
3e0708b9
YZ		 1602 pr_warn_ratelimited(" dropping unsafe request %llu\n",
1603 req->r_tid);
1bd85aa6
JL		 1604 if (req->r_target_inode)
1605 mapping_set_error(req->r_target_inode->i_mapping, -EIO);
1606 if (req->r_unsafe_dir)
1607 mapping_set_error(req->r_unsafe_dir->i_mapping, -EIO);
1c841a96
YZ		 1608 __unregister_request(mdsc, req);
1609 }
1610 /* zero r_attempts, so kick_requests() will re-send requests */
1611 p = rb_first(&mdsc->request_tree);
1612 while (p) {
1613 req = rb_entry(p, struct ceph_mds_request, r_node);
1614 p = rb_next(p);
1615 if (req->r_session &&
1616 req->r_session->s_mds == session->s_mds)
1617 req->r_attempts = 0;
1618 }
1619 mutex_unlock(&mdsc->mutex);
1620}
1621
2f2dc053 1622/*
f818a736
SW		 1623 * Helper to safely iterate over all caps associated with a session, with
1624 * special care taken to handle a racing __ceph_remove_cap().
2f2dc053 1625 *
f818a736 1626 * Caller must hold session s_mutex.
2f2dc053 1627 */
f5d77269
JL		 1628int ceph_iterate_session_caps(struct ceph_mds_session *session,
1629 int (*cb)(struct inode *, struct ceph_cap *,
1630 void *), void *arg)
2f2dc053 1631{
7c1332b8
SW		 1632 struct list_head *p;
1633 struct ceph_cap *cap;
1634 struct inode *inode, *last_inode = NULL;
1635 struct ceph_cap *old_cap = NULL;
2f2dc053
SW		 1636 int ret;
1637
1638 dout("iterate_session_caps %p mds%d\n", session, session->s_mds);
1639 spin_lock(&session->s_cap_lock);
7c1332b8
SW		 1640 p = session->s_caps.next;
1641 while (p != &session->s_caps) {
1642 cap = list_entry(p, struct ceph_cap, session_caps);
874c8ca1 1643 inode = igrab(&cap->ci->netfs.inode);
7c1332b8
SW		 1644 if (!inode) {
1645 p = p->next;
2f2dc053 1646 continue;
7c1332b8
SW		 1647 }
1648 session->s_cap_iterator = cap;
2f2dc053 1649 spin_unlock(&session->s_cap_lock);
7c1332b8
SW		 1650
1651 if (last_inode) {
23c2c76e 1652 iput(last_inode);
7c1332b8
SW		 1653 last_inode = NULL;
1654 }
1655 if (old_cap) {
37151668 1656 ceph_put_cap(session->s_mdsc, old_cap);
7c1332b8
SW		 1657 old_cap = NULL;
1658 }
1659
2f2dc053 1660 ret = cb(inode, cap, arg);
7c1332b8
SW		 1661 last_inode = inode;
1662
2f2dc053 1663 spin_lock(&session->s_cap_lock);
7c1332b8 1664 p = p->next;
d37b1d99 1665 if (!cap->ci) {
7c1332b8
SW		 1666 dout("iterate_session_caps finishing cap %p removal\n",
1667 cap);
1668 BUG_ON(cap->session != session);
745a8e3b 1669 cap->session = NULL;
7c1332b8
SW		 1670 list_del_init(&cap->session_caps);
1671 session->s_nr_caps--;
4f1d756d 1672 atomic64_dec(&session->s_mdsc->metric.total_caps);
e3ec8d68
YZ		 1673 if (cap->queue_release)
1674 __ceph_queue_cap_release(session, cap);
1675 else
745a8e3b 1676 old_cap = cap; /* put_cap it w/o locks held */
7c1332b8 1677 }
5dacf091
SW		 1678 if (ret < 0)
1679 goto out;
2f2dc053 1680 }
5dacf091
SW		 1681 ret = 0;
1682out:
7c1332b8 1683 session->s_cap_iterator = NULL;
2f2dc053 1684 spin_unlock(&session->s_cap_lock);
7c1332b8 1685
23c2c76e 1686 iput(last_inode);
7c1332b8 1687 if (old_cap)
37151668 1688 ceph_put_cap(session->s_mdsc, old_cap);
7c1332b8 1689
5dacf091 1690 return ret;
2f2dc053
SW		 1691}
1692
1693static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
6c99f254 1694 void *arg)
2f2dc053
SW		 1695{
1696 struct ceph_inode_info *ci = ceph_inode(inode);
6c93df5d 1697 bool invalidate = false;
36e6da98 1698 int iputs;
6c99f254 1699
2f2dc053 1700 dout("removing cap %p, ci is %p, inode is %p\n",
874c8ca1 1701 cap, ci, &ci->netfs.inode);
be655596 1702 spin_lock(&ci->i_ceph_lock);
36e6da98 1703 iputs = ceph_purge_inode_cap(inode, cap, &invalidate);
be655596 1704 spin_unlock(&ci->i_ceph_lock);
77310320
YZ		 1705
1706 wake_up_all(&ci->i_cap_wq);
6c93df5d
YZ		 1707 if (invalidate)
1708 ceph_queue_invalidate(inode);
36e6da98 1709 while (iputs--)
a6d37ccd 1710 iput(inode);
2f2dc053
SW		 1711 return 0;
1712}
1713
1714/*
1715 * caller must hold session s_mutex
1716 */
1717static void remove_session_caps(struct ceph_mds_session *session)
1718{
6c93df5d
YZ		 1719 struct ceph_fs_client *fsc = session->s_mdsc->fsc;
1720 struct super_block *sb = fsc->sb;
c8a96a31
JL		 1721 LIST_HEAD(dispose);
1722
2f2dc053 1723 dout("remove_session_caps on %p\n", session);
f5d77269 1724 ceph_iterate_session_caps(session, remove_session_caps_cb, fsc);
6f60f889 1725
c8799fc4
YZ		 1726 wake_up_all(&fsc->mdsc->cap_flushing_wq);
1727
6f60f889
YZ		 1728 spin_lock(&session->s_cap_lock);
1729 if (session->s_nr_caps > 0) {
6f60f889
YZ		 1730 struct inode *inode;
1731 struct ceph_cap *cap, *prev = NULL;
1732 struct ceph_vino vino;
1733 /*
1734 * iterate_session_caps() skips inodes that are being
1735 * deleted, we need to wait until deletions are complete.
1736 * __wait_on_freeing_inode() is designed for the job,
1737 * but it is not exported, so use lookup inode function
1738 * to access it.
1739 */
1740 while (!list_empty(&session->s_caps)) {
1741 cap = list_entry(session->s_caps.next,
1742 struct ceph_cap, session_caps);
1743 if (cap == prev)
1744 break;
1745 prev = cap;
1746 vino = cap->ci->i_vino;
1747 spin_unlock(&session->s_cap_lock);
1748
ed284c49 1749 inode = ceph_find_inode(sb, vino);
23c2c76e 1750 iput(inode);
6f60f889
YZ		 1751
1752 spin_lock(&session->s_cap_lock);
1753 }
1754 }
745a8e3b
YZ		 1755
1756 // drop cap expires and unlock s_cap_lock
c8a96a31 1757 detach_cap_releases(session, &dispose);
6f60f889 1758
2f2dc053 1759 BUG_ON(session->s_nr_caps > 0);
6c99f254 1760 BUG_ON(!list_empty(&session->s_cap_flushing));
c8a96a31
JL		 1761 spin_unlock(&session->s_cap_lock);
1762 dispose_cap_releases(session->s_mdsc, &dispose);
2f2dc053
SW		 1763}
1764
d2f8bb27
YZ		 1765enum {
1766 RECONNECT,
1767 RENEWCAPS,
1768 FORCE_RO,
1769};
1770
2f2dc053
SW		 1771/*
1772 * wake up any threads waiting on this session's caps. if the cap is
1773 * old (didn't get renewed on the client reconnect), remove it now.
1774 *
1775 * caller must hold s_mutex.
1776 */
1777static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap,
1778 void *arg)
1779{
0dc2570f 1780 struct ceph_inode_info *ci = ceph_inode(inode);
d2f8bb27 1781 unsigned long ev = (unsigned long)arg;
0dc2570f 1782
d2f8bb27 1783 if (ev == RECONNECT) {
be655596 1784 spin_lock(&ci->i_ceph_lock);
0dc2570f
SW		 1785 ci->i_wanted_max_size = 0;
1786 ci->i_requested_max_size = 0;
be655596 1787 spin_unlock(&ci->i_ceph_lock);
d2f8bb27 1788 } else if (ev == RENEWCAPS) {
52d60f8e 1789 if (cap->cap_gen < atomic_read(&cap->session->s_cap_gen)) {
d2f8bb27
YZ		 1790 /* mds did not re-issue stale cap */
1791 spin_lock(&ci->i_ceph_lock);
1792 cap->issued = cap->implemented = CEPH_CAP_PIN;
d2f8bb27
YZ		 1793 spin_unlock(&ci->i_ceph_lock);
1794 }
1795 } else if (ev == FORCE_RO) {
0dc2570f 1796 }
e5360309 1797 wake_up_all(&ci->i_cap_wq);
2f2dc053
SW		 1798 return 0;
1799}
1800
d2f8bb27 1801static void wake_up_session_caps(struct ceph_mds_session *session, int ev)
2f2dc053
SW		 1802{
1803 dout("wake_up_session_caps %p mds%d\n", session, session->s_mds);
f5d77269
JL		 1804 ceph_iterate_session_caps(session, wake_up_session_cb,
1805 (void *)(unsigned long)ev);
2f2dc053
SW		 1806}
1807
1808/*
1809 * Send periodic message to MDS renewing all currently held caps. The
1810 * ack will reset the expiration for all caps from this session.
1811 *
1812 * caller holds s_mutex
1813 */
1814static int send_renew_caps(struct ceph_mds_client *mdsc,
1815 struct ceph_mds_session *session)
1816{
1817 struct ceph_msg *msg;
1818 int state;
1819
1820 if (time_after_eq(jiffies, session->s_cap_ttl) &&
1821 time_after_eq(session->s_cap_ttl, session->s_renew_requested))
1822 pr_info("mds%d caps stale\n", session->s_mds);
e4cb4cb8 1823 session->s_renew_requested = jiffies;
2f2dc053
SW		 1824
1825 /* do not try to renew caps until a recovering mds has reconnected
1826 * with its clients. */
1827 state = ceph_mdsmap_get_state(mdsc->mdsmap, session->s_mds);
1828 if (state < CEPH_MDS_STATE_RECONNECT) {
1829 dout("send_renew_caps ignoring mds%d (%s)\n",
1830 session->s_mds, ceph_mds_state_name(state));
1831 return 0;
1832 }
1833
1834 dout("send_renew_caps to mds%d (%s)\n", session->s_mds,
1835 ceph_mds_state_name(state));
fba97e80
XL		 1836 msg = ceph_create_session_msg(CEPH_SESSION_REQUEST_RENEWCAPS,
1837 ++session->s_renew_seq);
a79832f2
SW		 1838 if (!msg)
1839 return -ENOMEM;
2f2dc053
SW		 1840 ceph_con_send(&session->s_con, msg);
1841 return 0;
1842}
1843
186e4f7a
YZ		 1844static int send_flushmsg_ack(struct ceph_mds_client *mdsc,
1845 struct ceph_mds_session *session, u64 seq)
1846{
1847 struct ceph_msg *msg;
1848
1849 dout("send_flushmsg_ack to mds%d (%s)s seq %lld\n",
a687ecaf 1850 session->s_mds, ceph_session_state_name(session->s_state), seq);
fba97e80 1851 msg = ceph_create_session_msg(CEPH_SESSION_FLUSHMSG_ACK, seq);
186e4f7a
YZ		 1852 if (!msg)
1853 return -ENOMEM;
1854 ceph_con_send(&session->s_con, msg);
1855 return 0;
1856}
1857
1858
2f2dc053
SW		 1859/*
1860 * Note new cap ttl, and any transition from stale -> not stale (fresh?).
0dc2570f
SW		 1861 *
1862 * Called under session->s_mutex
2f2dc053
SW		 1863 */
1864static void renewed_caps(struct ceph_mds_client *mdsc,
1865 struct ceph_mds_session *session, int is_renew)
1866{
1867 int was_stale;
1868 int wake = 0;
1869
1870 spin_lock(&session->s_cap_lock);
1ce208a6 1871 was_stale = is_renew && time_after_eq(jiffies, session->s_cap_ttl);
2f2dc053
SW		 1872
1873 session->s_cap_ttl = session->s_renew_requested +
1874 mdsc->mdsmap->m_session_timeout*HZ;
1875
1876 if (was_stale) {
1877 if (time_before(jiffies, session->s_cap_ttl)) {
1878 pr_info("mds%d caps renewed\n", session->s_mds);
1879 wake = 1;
1880 } else {
1881 pr_info("mds%d caps still stale\n", session->s_mds);
1882 }
1883 }
1884 dout("renewed_caps mds%d ttl now %lu, was %s, now %s\n",
1885 session->s_mds, session->s_cap_ttl, was_stale ? "stale" : "fresh",
1886 time_before(jiffies, session->s_cap_ttl) ? "stale" : "fresh");
1887 spin_unlock(&session->s_cap_lock);
1888
1889 if (wake)
d2f8bb27 1890 wake_up_session_caps(session, RENEWCAPS);
2f2dc053
SW		 1891}
1892
1893/*
1894 * send a session close request
1895 */
3e699bd8 1896static int request_close_session(struct ceph_mds_session *session)
2f2dc053
SW		 1897{
1898 struct ceph_msg *msg;
2f2dc053
SW		 1899
1900 dout("request_close_session mds%d state %s seq %lld\n",
a687ecaf 1901 session->s_mds, ceph_session_state_name(session->s_state),
2f2dc053 1902 session->s_seq);
fba97e80
XL		 1903 msg = ceph_create_session_msg(CEPH_SESSION_REQUEST_CLOSE,
1904 session->s_seq);
a79832f2
SW		 1905 if (!msg)
1906 return -ENOMEM;
1907 ceph_con_send(&session->s_con, msg);
fcff415c 1908 return 1;
2f2dc053
SW		 1909}
1910
1911/*
1912 * Called with s_mutex held.
1913 */
1914static int __close_session(struct ceph_mds_client *mdsc,
1915 struct ceph_mds_session *session)
1916{
1917 if (session->s_state >= CEPH_MDS_SESSION_CLOSING)
1918 return 0;
1919 session->s_state = CEPH_MDS_SESSION_CLOSING;
3e699bd8 1920 return request_close_session(session);
2f2dc053
SW		 1921}
1922
040d7860
YZ		 1923static bool drop_negative_children(struct dentry *dentry)
1924{
1925 struct dentry *child;
1926 bool all_negative = true;
1927
1928 if (!d_is_dir(dentry))
1929 goto out;
1930
1931 spin_lock(&dentry->d_lock);
1932 list_for_each_entry(child, &dentry->d_subdirs, d_child) {
1933 if (d_really_is_positive(child)) {
1934 all_negative = false;
1935 break;
1936 }
1937 }
1938 spin_unlock(&dentry->d_lock);
1939
1940 if (all_negative)
1941 shrink_dcache_parent(dentry);
1942out:
1943 return all_negative;
1944}
1945
2f2dc053
SW		 1946/*
1947 * Trim old(er) caps.
1948 *
1949 * Because we can't cache an inode without one or more caps, we do
1950 * this indirectly: if a cap is unused, we prune its aliases, at which
1951 * point the inode will hopefully get dropped to.
1952 *
1953 * Yes, this is a bit sloppy. Our only real goal here is to respond to
1954 * memory pressure from the MDS, though, so it needn't be perfect.
1955 */
1956static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg)
1957{
533a2818 1958 int *remaining = arg;
2f2dc053 1959 struct ceph_inode_info *ci = ceph_inode(inode);
979abfdd 1960 int used, wanted, oissued, mine;
2f2dc053 1961
533a2818 1962 if (*remaining <= 0)
2f2dc053
SW		 1963 return -1;
1964
be655596 1965 spin_lock(&ci->i_ceph_lock);
2f2dc053
SW		 1966 mine = cap->issued | cap->implemented;
1967 used = __ceph_caps_used(ci);
979abfdd 1968 wanted = __ceph_caps_file_wanted(ci);
2f2dc053
SW		 1969 oissued = __ceph_caps_issued_other(ci, cap);
1970
979abfdd 1971 dout("trim_caps_cb %p cap %p mine %s oissued %s used %s wanted %s\n",
2f2dc053 1972 inode, cap, ceph_cap_string(mine), ceph_cap_string(oissued),
979abfdd
YZ		 1973 ceph_cap_string(used), ceph_cap_string(wanted));
1974 if (cap == ci->i_auth_cap) {
622f3e25
YZ		 1975 if (ci->i_dirty_caps || ci->i_flushing_caps ||
1976 !list_empty(&ci->i_cap_snaps))
979abfdd
YZ		 1977 goto out;
1978 if ((used | wanted) & CEPH_CAP_ANY_WR)
1979 goto out;
89aa5930
YZ		 1980 /* Note: it's possible that i_filelock_ref becomes non-zero
1981 * after dropping auth caps. It doesn't hurt because reply
1982 * of lock mds request will re-add auth caps. */
1983 if (atomic_read(&ci->i_filelock_ref) > 0)
1984 goto out;
979abfdd 1985 }
5e804ac4
YZ		 1986 /* The inode has cached pages, but it's no longer used.
1987 * we can safely drop it */
525d15e8
YZ		 1988 if (S_ISREG(inode->i_mode) &&
1989 wanted == 0 && used == CEPH_CAP_FILE_CACHE &&
5e804ac4
YZ		 1990 !(oissued & CEPH_CAP_FILE_CACHE)) {
1991 used = 0;
1992 oissued = 0;
1993 }
979abfdd 1994 if ((used | wanted) & ~oissued & mine)
2f2dc053
SW		 1995 goto out; /* we need these caps */
1996
2f2dc053
SW		 1997 if (oissued) {
1998 /* we aren't the only cap.. just remove us */
a76d0a9c 1999 ceph_remove_cap(cap, true);
533a2818 2000 (*remaining)--;
2f2dc053 2001 } else {
040d7860 2002 struct dentry *dentry;
5e804ac4 2003 /* try dropping referring dentries */
be655596 2004 spin_unlock(&ci->i_ceph_lock);
040d7860
YZ		 2005 dentry = d_find_any_alias(inode);
2006 if (dentry && drop_negative_children(dentry)) {
2007 int count;
2008 dput(dentry);
2009 d_prune_aliases(inode);
2010 count = atomic_read(&inode->i_count);
2011 if (count == 1)
533a2818 2012 (*remaining)--;
040d7860
YZ		 2013 dout("trim_caps_cb %p cap %p pruned, count now %d\n",
2014 inode, cap, count);
2015 } else {
2016 dput(dentry);
2017 }
2f2dc053
SW		 2018 return 0;
2019 }
2020
2021out:
be655596 2022 spin_unlock(&ci->i_ceph_lock);
2f2dc053
SW		 2023 return 0;
2024}
2025
2026/*
2027 * Trim session cap count down to some max number.
2028 */
e30ee581
ZZ		 2029int ceph_trim_caps(struct ceph_mds_client *mdsc,
2030 struct ceph_mds_session *session,
2031 int max_caps)
2f2dc053
SW		 2032{
2033 int trim_caps = session->s_nr_caps - max_caps;
2034
2035 dout("trim_caps mds%d start: %d / %d, trim %d\n",
2036 session->s_mds, session->s_nr_caps, max_caps, trim_caps);
2037 if (trim_caps > 0) {
533a2818
JL		 2038 int remaining = trim_caps;
2039
2040 ceph_iterate_session_caps(session, trim_caps_cb, &remaining);
2f2dc053
SW		 2041 dout("trim_caps mds%d done: %d / %d, trimmed %d\n",
2042 session->s_mds, session->s_nr_caps, max_caps,
533a2818 2043 trim_caps - remaining);
2f2dc053 2044 }
a56371d9 2045
e3ec8d68 2046 ceph_flush_cap_releases(mdsc, session);
2f2dc053
SW		 2047 return 0;
2048}
2049
8310b089
YZ		 2050static int check_caps_flush(struct ceph_mds_client *mdsc,
2051 u64 want_flush_tid)
2052{
8310b089
YZ		 2053 int ret = 1;
2054
2055 spin_lock(&mdsc->cap_dirty_lock);
e4500b5e
YZ		 2056 if (!list_empty(&mdsc->cap_flush_list)) {
2057 struct ceph_cap_flush *cf =
2058 list_first_entry(&mdsc->cap_flush_list,
2059 struct ceph_cap_flush, g_list);
2060 if (cf->tid <= want_flush_tid) {
2061 dout("check_caps_flush still flushing tid "
2062 "%llu <= %llu\n", cf->tid, want_flush_tid);
2063 ret = 0;
2064 }
8310b089
YZ		 2065 }
2066 spin_unlock(&mdsc->cap_dirty_lock);
2067 return ret;
d3383a8e
YZ		 2068}
2069
2f2dc053
SW		 2070/*
2071 * flush all dirty inode data to disk.
2072 *
8310b089 2073 * returns true if we've flushed through want_flush_tid
2f2dc053 2074 */
affbc19a 2075static void wait_caps_flush(struct ceph_mds_client *mdsc,
0e294387 2076 u64 want_flush_tid)
2f2dc053 2077{
0e294387 2078 dout("check_caps_flush want %llu\n", want_flush_tid);
8310b089
YZ		 2079
2080 wait_event(mdsc->cap_flushing_wq,
2081 check_caps_flush(mdsc, want_flush_tid));
2082
2083 dout("check_caps_flush ok, flushed thru %llu\n", want_flush_tid);
2f2dc053
SW		 2084}
2085
2086/*
2087 * called under s_mutex
2088 */
e3ec8d68
YZ		 2089static void ceph_send_cap_releases(struct ceph_mds_client *mdsc,
2090 struct ceph_mds_session *session)
2f2dc053 2091{
745a8e3b
YZ		 2092 struct ceph_msg *msg = NULL;
2093 struct ceph_mds_cap_release *head;
2094 struct ceph_mds_cap_item *item;
92475f05 2095 struct ceph_osd_client *osdc = &mdsc->fsc->client->osdc;
745a8e3b
YZ		 2096 struct ceph_cap *cap;
2097 LIST_HEAD(tmp_list);
2098 int num_cap_releases;
92475f05
JL		 2099 __le32 barrier, *cap_barrier;
2100
2101 down_read(&osdc->lock);
2102 barrier = cpu_to_le32(osdc->epoch_barrier);
2103 up_read(&osdc->lock);
2f2dc053 2104
0f8605f2 2105 spin_lock(&session->s_cap_lock);
745a8e3b
YZ		 2106again:
2107 list_splice_init(&session->s_cap_releases, &tmp_list);
2108 num_cap_releases = session->s_num_cap_releases;
2109 session->s_num_cap_releases = 0;
2f2dc053 2110 spin_unlock(&session->s_cap_lock);
e01a5946 2111
745a8e3b
YZ		 2112 while (!list_empty(&tmp_list)) {
2113 if (!msg) {
2114 msg = ceph_msg_new(CEPH_MSG_CLIENT_CAPRELEASE,
09cbfeaf 2115 PAGE_SIZE, GFP_NOFS, false);
745a8e3b
YZ		 2116 if (!msg)
2117 goto out_err;
2118 head = msg->front.iov_base;
2119 head->num = cpu_to_le32(0);
2120 msg->front.iov_len = sizeof(*head);
92475f05
JL		 2121
2122 msg->hdr.version = cpu_to_le16(2);
2123 msg->hdr.compat_version = cpu_to_le16(1);
745a8e3b 2124 }
92475f05 2125
745a8e3b
YZ		 2126 cap = list_first_entry(&tmp_list, struct ceph_cap,
2127 session_caps);
2128 list_del(&cap->session_caps);
2129 num_cap_releases--;
e01a5946 2130
00bd8edb 2131 head = msg->front.iov_base;
4198aba4
JL		 2132 put_unaligned_le32(get_unaligned_le32(&head->num) + 1,
2133 &head->num);
745a8e3b
YZ		 2134 item = msg->front.iov_base + msg->front.iov_len;
2135 item->ino = cpu_to_le64(cap->cap_ino);
2136 item->cap_id = cpu_to_le64(cap->cap_id);
2137 item->migrate_seq = cpu_to_le32(cap->mseq);
2138 item->seq = cpu_to_le32(cap->issue_seq);
2139 msg->front.iov_len += sizeof(*item);
2140
2141 ceph_put_cap(mdsc, cap);
2142
2143 if (le32_to_cpu(head->num) == CEPH_CAPS_PER_RELEASE) {
92475f05
JL		 2144 // Append cap_barrier field
2145 cap_barrier = msg->front.iov_base + msg->front.iov_len;
2146 *cap_barrier = barrier;
2147 msg->front.iov_len += sizeof(*cap_barrier);
2148
745a8e3b
YZ		 2149 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
2150 dout("send_cap_releases mds%d %p\n", session->s_mds, msg);
2151 ceph_con_send(&session->s_con, msg);
2152 msg = NULL;
2153 }
00bd8edb 2154 }
e01a5946 2155
745a8e3b 2156 BUG_ON(num_cap_releases != 0);
e01a5946 2157
745a8e3b
YZ		 2158 spin_lock(&session->s_cap_lock);
2159 if (!list_empty(&session->s_cap_releases))
2160 goto again;
2161 spin_unlock(&session->s_cap_lock);
2162
2163 if (msg) {
92475f05
JL		 2164 // Append cap_barrier field
2165 cap_barrier = msg->front.iov_base + msg->front.iov_len;
2166 *cap_barrier = barrier;
2167 msg->front.iov_len += sizeof(*cap_barrier);
2168
745a8e3b
YZ		 2169 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
2170 dout("send_cap_releases mds%d %p\n", session->s_mds, msg);
2171 ceph_con_send(&session->s_con, msg);
e01a5946 2172 }
745a8e3b
YZ		 2173 return;
2174out_err:
2175 pr_err("send_cap_releases mds%d, failed to allocate message\n",
2176 session->s_mds);
2177 spin_lock(&session->s_cap_lock);
2178 list_splice(&tmp_list, &session->s_cap_releases);
2179 session->s_num_cap_releases += num_cap_releases;
2180 spin_unlock(&session->s_cap_lock);
e01a5946
SW		 2181}
2182
e3ec8d68
YZ		 2183static void ceph_cap_release_work(struct work_struct *work)
2184{
2185 struct ceph_mds_session *session =
2186 container_of(work, struct ceph_mds_session, s_cap_release_work);
2187
2188 mutex_lock(&session->s_mutex);
2189 if (session->s_state == CEPH_MDS_SESSION_OPEN ||
2190 session->s_state == CEPH_MDS_SESSION_HUNG)
2191 ceph_send_cap_releases(session->s_mdsc, session);
2192 mutex_unlock(&session->s_mutex);
2193 ceph_put_mds_session(session);
2194}
2195
2196void ceph_flush_cap_releases(struct ceph_mds_client *mdsc,
2197 struct ceph_mds_session *session)
2198{
2199 if (mdsc->stopping)
2200 return;
2201
5b3248c6 2202 ceph_get_mds_session(session);
e3ec8d68
YZ		 2203 if (queue_work(mdsc->fsc->cap_wq,
2204 &session->s_cap_release_work)) {
2205 dout("cap release work queued\n");
2206 } else {
2207 ceph_put_mds_session(session);
2208 dout("failed to queue cap release work\n");
2209 }
2210}
2211
2212/*
2213 * caller holds session->s_cap_lock
2214 */
2215void __ceph_queue_cap_release(struct ceph_mds_session *session,
2216 struct ceph_cap *cap)
2217{
2218 list_add_tail(&cap->session_caps, &session->s_cap_releases);
2219 session->s_num_cap_releases++;
2220
2221 if (!(session->s_num_cap_releases % CEPH_CAPS_PER_RELEASE))
2222 ceph_flush_cap_releases(session->s_mdsc, session);
2223}
2224
37c4efc1
YZ		 2225static void ceph_cap_reclaim_work(struct work_struct *work)
2226{
2227 struct ceph_mds_client *mdsc =
2228 container_of(work, struct ceph_mds_client, cap_reclaim_work);
2229 int ret = ceph_trim_dentries(mdsc);
2230 if (ret == -EAGAIN)
2231 ceph_queue_cap_reclaim_work(mdsc);
2232}
2233
2234void ceph_queue_cap_reclaim_work(struct ceph_mds_client *mdsc)
2235{
2236 if (mdsc->stopping)
2237 return;
2238
2239 if (queue_work(mdsc->fsc->cap_wq, &mdsc->cap_reclaim_work)) {
2240 dout("caps reclaim work queued\n");
2241 } else {
2242 dout("failed to queue caps release work\n");
2243 }
2244}
2245
fe33032d
YZ		 2246void ceph_reclaim_caps_nr(struct ceph_mds_client *mdsc, int nr)
2247{
2248 int val;
2249 if (!nr)
2250 return;
2251 val = atomic_add_return(nr, &mdsc->cap_reclaim_pending);
bba1560b 2252 if ((val % CEPH_CAPS_PER_RELEASE) < nr) {
fe33032d
YZ		 2253 atomic_set(&mdsc->cap_reclaim_pending, 0);
2254 ceph_queue_cap_reclaim_work(mdsc);
2255 }
2256}
2257
2f2dc053
SW		 2258/*
2259 * requests
2260 */
2261
54008399
YZ		 2262int ceph_alloc_readdir_reply_buffer(struct ceph_mds_request *req,
2263 struct inode *dir)
2264{
2265 struct ceph_inode_info *ci = ceph_inode(dir);
2266 struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
2267 struct ceph_mount_options *opt = req->r_mdsc->fsc->mount_options;
2a5beea3 2268 size_t size = sizeof(struct ceph_mds_reply_dir_entry);
ad8c28a9
JL		 2269 unsigned int num_entries;
2270 int order;
54008399
YZ		 2271
2272 spin_lock(&ci->i_ceph_lock);
2273 num_entries = ci->i_files + ci->i_subdirs;
2274 spin_unlock(&ci->i_ceph_lock);
ad8c28a9 2275 num_entries = max(num_entries, 1U);
54008399
YZ		 2276 num_entries = min(num_entries, opt->max_readdir);
2277
2278 order = get_order(size * num_entries);
2279 while (order >= 0) {
2a5beea3 2280 rinfo->dir_entries = (void*)__get_free_pages(GFP_KERNEL |
2941bf53
XL		 2281 __GFP_NOWARN |
2282 __GFP_ZERO,
2a5beea3
YZ		 2283 order);
2284 if (rinfo->dir_entries)
54008399
YZ		 2285 break;
2286 order--;
2287 }
2a5beea3 2288 if (!rinfo->dir_entries)
54008399
YZ		 2289 return -ENOMEM;
2290
2291 num_entries = (PAGE_SIZE << order) / size;
2292 num_entries = min(num_entries, opt->max_readdir);
2293
2294 rinfo->dir_buf_size = PAGE_SIZE << order;
2295 req->r_num_caps = num_entries + 1;
2296 req->r_args.readdir.max_entries = cpu_to_le32(num_entries);
2297 req->r_args.readdir.max_bytes = cpu_to_le32(opt->max_readdir_bytes);
2298 return 0;
2299}
2300
2f2dc053
SW		 2301/*
2302 * Create an mds request.
2303 */
2304struct ceph_mds_request *
2305ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
2306{
058daab7 2307 struct ceph_mds_request *req;
2f2dc053 2308
058daab7 2309 req = kmem_cache_zalloc(ceph_mds_request_cachep, GFP_NOFS);
2f2dc053
SW		 2310 if (!req)
2311 return ERR_PTR(-ENOMEM);
2312
b4556396 2313 mutex_init(&req->r_fill_mutex);
37151668 2314 req->r_mdsc = mdsc;
2f2dc053 2315 req->r_started = jiffies;
70c94820 2316 req->r_start_latency = ktime_get();
2f2dc053
SW		 2317 req->r_resend_mds = -1;
2318 INIT_LIST_HEAD(&req->r_unsafe_dir_item);
68cd5b4b 2319 INIT_LIST_HEAD(&req->r_unsafe_target_item);
2f2dc053 2320 req->r_fmode = -1;
153c8e6b 2321 kref_init(&req->r_kref);
fcd00b68 2322 RB_CLEAR_NODE(&req->r_node);
2f2dc053
SW		 2323 INIT_LIST_HEAD(&req->r_wait);
2324 init_completion(&req->r_completion);
2325 init_completion(&req->r_safe_completion);
2326 INIT_LIST_HEAD(&req->r_unsafe_item);
2327
668c9a61 2328 ktime_get_coarse_real_ts64(&req->r_stamp);
b8e69066 2329
2f2dc053
SW		 2330 req->r_op = op;
2331 req->r_direct_mode = mode;
2332 return req;
2333}
2334
2335/*
44ca18f2 2336 * return oldest (lowest) request, tid in request tree, 0 if none.
2f2dc053
SW		 2337 *
2338 * called under mdsc->mutex.
2339 */
44ca18f2
SW		 2340static struct ceph_mds_request *__get_oldest_req(struct ceph_mds_client *mdsc)
2341{
2342 if (RB_EMPTY_ROOT(&mdsc->request_tree))
2343 return NULL;
2344 return rb_entry(rb_first(&mdsc->request_tree),
2345 struct ceph_mds_request, r_node);
2346}
2347
e8a7b8b1 2348static inline u64 __get_oldest_tid(struct ceph_mds_client *mdsc)
2f2dc053 2349{
e8a7b8b1 2350 return mdsc->oldest_tid;
2f2dc053
SW		 2351}
2352
2353/*
2354 * Build a dentry's path. Allocate on heap; caller must kfree. Based
2355 * on build_path_from_dentry in fs/cifs/dir.c.
2356 *
2357 * If @stop_on_nosnap, generate path relative to the first non-snapped
2358 * inode.
2359 *
2360 * Encode hidden .snap dirs as a double /, i.e.
2361 * foo/.snap/bar -> foo//bar
2362 */
69a10fb3 2363char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *pbase,
2f2dc053
SW		 2364 int stop_on_nosnap)
2365{
2366 struct dentry *temp;
2367 char *path;
f77f21bb 2368 int pos;
1b71fe2e 2369 unsigned seq;
69a10fb3 2370 u64 base;
2f2dc053 2371
d37b1d99 2372 if (!dentry)
2f2dc053
SW		 2373 return ERR_PTR(-EINVAL);
2374
f77f21bb 2375 path = __getname();
d37b1d99 2376 if (!path)
2f2dc053 2377 return ERR_PTR(-ENOMEM);
f77f21bb
JL		 2378retry:
2379 pos = PATH_MAX - 1;
2380 path[pos] = '\0';
2381
2382 seq = read_seqbegin(&rename_lock);
1b71fe2e 2383 rcu_read_lock();
f77f21bb
JL		 2384 temp = dentry;
2385 for (;;) {
1b71fe2e 2386 struct inode *inode;
2f2dc053 2387
1b71fe2e 2388 spin_lock(&temp->d_lock);
2b0143b5 2389 inode = d_inode(temp);
2f2dc053 2390 if (inode && ceph_snap(inode) == CEPH_SNAPDIR) {
104648ad 2391 dout("build_path path+%d: %p SNAPDIR\n",
2f2dc053 2392 pos, temp);
d6b8bd67 2393 } else if (stop_on_nosnap && inode && dentry != temp &&
2f2dc053 2394 ceph_snap(inode) == CEPH_NOSNAP) {
9d5a09e6 2395 spin_unlock(&temp->d_lock);
d6b8bd67 2396 pos++; /* get rid of any prepended '/' */
2f2dc053
SW		 2397 break;
2398 } else {
2399 pos -= temp->d_name.len;
1b71fe2e
AV		 2400 if (pos < 0) {
2401 spin_unlock(&temp->d_lock);
2f2dc053 2402 break;
1b71fe2e 2403 }
f77f21bb 2404 memcpy(path + pos, temp->d_name.name, temp->d_name.len);
2f2dc053 2405 }
1b71fe2e 2406 spin_unlock(&temp->d_lock);
41883ba8 2407 temp = READ_ONCE(temp->d_parent);
f77f21bb
JL		 2408
2409 /* Are we at the root? */
2410 if (IS_ROOT(temp))
2411 break;
2412
2413 /* Are we out of buffer? */
2414 if (--pos < 0)
2415 break;
2416
2417 path[pos] = '/';
2f2dc053 2418 }
69a10fb3 2419 base = ceph_ino(d_inode(temp));
1b71fe2e 2420 rcu_read_unlock();
f5946bcc
JL		 2421
2422 if (read_seqretry(&rename_lock, seq))
2423 goto retry;
2424
2425 if (pos < 0) {
2426 /*
2427 * A rename didn't occur, but somehow we didn't end up where
2428 * we thought we would. Throw a warning and try again.
2429 */
2430 pr_warn("build_path did not end path lookup where "
2431 "expected, pos is %d\n", pos);
2f2dc053
SW		 2432 goto retry;
2433 }
2434
69a10fb3 2435 *pbase = base;
f77f21bb 2436 *plen = PATH_MAX - 1 - pos;
104648ad 2437 dout("build_path on %p %d built %llx '%.*s'\n",
f77f21bb
JL		 2438 dentry, d_count(dentry), base, *plen, path + pos);
2439 return path + pos;
2f2dc053
SW		 2440}
2441
fd36a717 2442static int build_dentry_path(struct dentry *dentry, struct inode *dir,
2f2dc053 2443 const char **ppath, int *ppathlen, u64 *pino,
1bcb3440 2444 bool *pfreepath, bool parent_locked)
2f2dc053
SW		 2445{
2446 char *path;
2447
c6b0b656 2448 rcu_read_lock();
fd36a717
JL		 2449 if (!dir)
2450 dir = d_inode_rcu(dentry->d_parent);
964fff74 2451 if (dir && parent_locked && ceph_snap(dir) == CEPH_NOSNAP) {
c6b0b656
JL		 2452 *pino = ceph_ino(dir);
2453 rcu_read_unlock();
964fff74
JL		 2454 *ppath = dentry->d_name.name;
2455 *ppathlen = dentry->d_name.len;
2f2dc053
SW		 2456 return 0;
2457 }
c6b0b656 2458 rcu_read_unlock();
2f2dc053
SW		 2459 path = ceph_mdsc_build_path(dentry, ppathlen, pino, 1);
2460 if (IS_ERR(path))
2461 return PTR_ERR(path);
2462 *ppath = path;
1bcb3440 2463 *pfreepath = true;
2f2dc053
SW		 2464 return 0;
2465}
2466
2467static int build_inode_path(struct inode *inode,
2468 const char **ppath, int *ppathlen, u64 *pino,
1bcb3440 2469 bool *pfreepath)
2f2dc053
SW		 2470{
2471 struct dentry *dentry;
2472 char *path;
2473
2474 if (ceph_snap(inode) == CEPH_NOSNAP) {
2475 *pino = ceph_ino(inode);
2476 *ppathlen = 0;
2477 return 0;
2478 }
2479 dentry = d_find_alias(inode);
2480 path = ceph_mdsc_build_path(dentry, ppathlen, pino, 1);
2481 dput(dentry);
2482 if (IS_ERR(path))
2483 return PTR_ERR(path);
2484 *ppath = path;
1bcb3440 2485 *pfreepath = true;
2f2dc053
SW		 2486 return 0;
2487}
2488
2489/*
2490 * request arguments may be specified via an inode *, a dentry *, or
2491 * an explicit ino+path.
2492 */
2493static int set_request_path_attr(struct inode *rinode, struct dentry *rdentry,
fd36a717
JL		 2494 struct inode *rdiri, const char *rpath,
2495 u64 rino, const char **ppath, int *pathlen,
1bcb3440 2496 u64 *ino, bool *freepath, bool parent_locked)
2f2dc053
SW		 2497{
2498 int r = 0;
2499
2500 if (rinode) {
2501 r = build_inode_path(rinode, ppath, pathlen, ino, freepath);
2502 dout(" inode %p %llx.%llx\n", rinode, ceph_ino(rinode),
2503 ceph_snap(rinode));
2504 } else if (rdentry) {
fd36a717 2505 r = build_dentry_path(rdentry, rdiri, ppath, pathlen, ino,
1bcb3440 2506 freepath, parent_locked);
2f2dc053
SW		 2507 dout(" dentry %p %llx/%.*s\n", rdentry, *ino, *pathlen,
2508 *ppath);
795858db 2509 } else if (rpath || rino) {
2f2dc053
SW		 2510 *ino = rino;
2511 *ppath = rpath;
b000056a 2512 *pathlen = rpath ? strlen(rpath) : 0;
2f2dc053
SW		 2513 dout(" path %.*s\n", *pathlen, rpath);
2514 }
2515
2516 return r;
2517}
2518
60267ba3
ID		 2519static void encode_timestamp_and_gids(void **p,
2520 const struct ceph_mds_request *req)
2521{
2522 struct ceph_timespec ts;
2523 int i;
2524
2525 ceph_encode_timespec64(&ts, &req->r_stamp);
2526 ceph_encode_copy(p, &ts, sizeof(ts));
2527
2528 /* gid_list */
2529 ceph_encode_32(p, req->r_cred->group_info->ngroups);
2530 for (i = 0; i < req->r_cred->group_info->ngroups; i++)
2531 ceph_encode_64(p, from_kgid(&init_user_ns,
2532 req->r_cred->group_info->gid[i]));
2533}
2534
2f2dc053
SW		 2535/*
2536 * called under mdsc->mutex
2537 */
4f1ddb1e 2538static struct ceph_msg *create_request_message(struct ceph_mds_session *session,
2f2dc053 2539 struct ceph_mds_request *req,
4f1ddb1e 2540 bool drop_cap_releases)
2f2dc053 2541{
4f1ddb1e
JL		 2542 int mds = session->s_mds;
2543 struct ceph_mds_client *mdsc = session->s_mdsc;
2f2dc053 2544 struct ceph_msg *msg;
4f1ddb1e 2545 struct ceph_mds_request_head_old *head;
2f2dc053
SW		 2546 const char *path1 = NULL;
2547 const char *path2 = NULL;
2548 u64 ino1 = 0, ino2 = 0;
2549 int pathlen1 = 0, pathlen2 = 0;
1bcb3440 2550 bool freepath1 = false, freepath2 = false;
60267ba3 2551 int len;
2f2dc053
SW		 2552 u16 releases;
2553 void *p, *end;
2554 int ret;
4f1ddb1e 2555 bool legacy = !(session->s_con.peer_features & CEPH_FEATURE_FS_BTIME);
2f2dc053
SW		 2556
2557 ret = set_request_path_attr(req->r_inode, req->r_dentry,
3dd69aab 2558 req->r_parent, req->r_path1, req->r_ino1.ino,
1bcb3440
JL		 2559 &path1, &pathlen1, &ino1, &freepath1,
2560 test_bit(CEPH_MDS_R_PARENT_LOCKED,
2561 &req->r_req_flags));
2f2dc053
SW		 2562 if (ret < 0) {
2563 msg = ERR_PTR(ret);
2564 goto out;
2565 }
2566
1bcb3440 2567 /* If r_old_dentry is set, then assume that its parent is locked */
2f2dc053 2568 ret = set_request_path_attr(NULL, req->r_old_dentry,
fd36a717 2569 req->r_old_dentry_dir,
2f2dc053 2570 req->r_path2, req->r_ino2.ino,
1bcb3440 2571 &path2, &pathlen2, &ino2, &freepath2, true);
2f2dc053
SW		 2572 if (ret < 0) {
2573 msg = ERR_PTR(ret);
2574 goto out_free1;
2575 }
2576
60267ba3 2577 len = legacy ? sizeof(*head) : sizeof(struct ceph_mds_request_head);
4f1ddb1e 2578 len += pathlen1 + pathlen2 + 2*(1 + sizeof(u32) + sizeof(u64)) +
777d738a 2579 sizeof(struct ceph_timespec);
60267ba3 2580 len += sizeof(u32) + (sizeof(u64) * req->r_cred->group_info->ngroups);
2f2dc053
SW		 2581
2582 /* calculate (max) length for cap releases */
2583 len += sizeof(struct ceph_mds_request_release) *
2584 (!!req->r_inode_drop + !!req->r_dentry_drop +
2585 !!req->r_old_inode_drop + !!req->r_old_dentry_drop);
4f1ddb1e 2586
2f2dc053 2587 if (req->r_dentry_drop)
c1dfc277 2588 len += pathlen1;
2f2dc053 2589 if (req->r_old_dentry_drop)
c1dfc277 2590 len += pathlen2;
2f2dc053 2591
0d9c1ab3 2592 msg = ceph_msg_new2(CEPH_MSG_CLIENT_REQUEST, len, 1, GFP_NOFS, false);
a79832f2
SW		 2593 if (!msg) {
2594 msg = ERR_PTR(-ENOMEM);
2f2dc053 2595 goto out_free2;
a79832f2 2596 }
2f2dc053 2597
6df058c0
SW		 2598 msg->hdr.tid = cpu_to_le64(req->r_tid);
2599
4f1ddb1e 2600 /*
60267ba3 2601 * The old ceph_mds_request_head didn't contain a version field, and
4f1ddb1e
JL		 2602 * one was added when we moved the message version from 3->4.
2603 */
2604 if (legacy) {
2605 msg->hdr.version = cpu_to_le16(3);
2606 head = msg->front.iov_base;
2607 p = msg->front.iov_base + sizeof(*head);
2608 } else {
2609 struct ceph_mds_request_head *new_head = msg->front.iov_base;
2610
2611 msg->hdr.version = cpu_to_le16(4);
2612 new_head->version = cpu_to_le16(CEPH_MDS_REQUEST_HEAD_VERSION);
2613 head = (struct ceph_mds_request_head_old *)&new_head->oldest_client_tid;
2614 p = msg->front.iov_base + sizeof(*new_head);
2615 }
2616
2f2dc053
SW		 2617 end = msg->front.iov_base + msg->front.iov_len;
2618
2619 head->mdsmap_epoch = cpu_to_le32(mdsc->mdsmap->m_epoch);
2620 head->op = cpu_to_le32(req->r_op);
7fe0cdeb
JL		 2621 head->caller_uid = cpu_to_le32(from_kuid(&init_user_ns,
2622 req->r_cred->fsuid));
2623 head->caller_gid = cpu_to_le32(from_kgid(&init_user_ns,
2624 req->r_cred->fsgid));
6deb8008 2625 head->ino = cpu_to_le64(req->r_deleg_ino);
2f2dc053
SW		 2626 head->args = req->r_args;
2627
2628 ceph_encode_filepath(&p, end, ino1, path1);
2629 ceph_encode_filepath(&p, end, ino2, path2);
2630
e979cf50
SW		 2631 /* make note of release offset, in case we need to replay */
2632 req->r_request_release_offset = p - msg->front.iov_base;
2633
2f2dc053
SW		 2634 /* cap releases */
2635 releases = 0;
2636 if (req->r_inode_drop)
2637 releases += ceph_encode_inode_release(&p,
2b0143b5 2638 req->r_inode ? req->r_inode : d_inode(req->r_dentry),
719a2514
YZ		 2639 mds, req->r_inode_drop, req->r_inode_unless,
2640 req->r_op == CEPH_MDS_OP_READDIR);
2f2dc053
SW		 2641 if (req->r_dentry_drop)
2642 releases += ceph_encode_dentry_release(&p, req->r_dentry,
3dd69aab 2643 req->r_parent, mds, req->r_dentry_drop,
ca6c8ae0 2644 req->r_dentry_unless);
2f2dc053
SW		 2645 if (req->r_old_dentry_drop)
2646 releases += ceph_encode_dentry_release(&p, req->r_old_dentry,
ca6c8ae0
JL		 2647 req->r_old_dentry_dir, mds,
2648 req->r_old_dentry_drop,
2649 req->r_old_dentry_unless);
2f2dc053
SW		 2650 if (req->r_old_inode_drop)
2651 releases += ceph_encode_inode_release(&p,
2b0143b5 2652 d_inode(req->r_old_dentry),
2f2dc053 2653 mds, req->r_old_inode_drop, req->r_old_inode_unless, 0);
6e6f0923
YZ		 2654
2655 if (drop_cap_releases) {
2656 releases = 0;
2657 p = msg->front.iov_base + req->r_request_release_offset;
2658 }
2659
2f2dc053
SW		 2660 head->num_releases = cpu_to_le16(releases);
2661
60267ba3 2662 encode_timestamp_and_gids(&p, req);
4f1ddb1e 2663
b682c6d4
XL		 2664 if (WARN_ON_ONCE(p > end)) {
2665 ceph_msg_put(msg);
2666 msg = ERR_PTR(-ERANGE);
2667 goto out_free2;
2668 }
2669
2f2dc053
SW		 2670 msg->front.iov_len = p - msg->front.iov_base;
2671 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
2672
25e6bae3
YZ		 2673 if (req->r_pagelist) {
2674 struct ceph_pagelist *pagelist = req->r_pagelist;
25e6bae3
YZ		 2675 ceph_msg_data_add_pagelist(msg, pagelist);
2676 msg->hdr.data_len = cpu_to_le32(pagelist->length);
2677 } else {
2678 msg->hdr.data_len = 0;
ebf18f47 2679 }
02afca6c 2680
2f2dc053
SW		 2681 msg->hdr.data_off = cpu_to_le16(0);
2682
2683out_free2:
2684 if (freepath2)
f77f21bb 2685 ceph_mdsc_free_path((char *)path2, pathlen2);
2f2dc053
SW		 2686out_free1:
2687 if (freepath1)
f77f21bb 2688 ceph_mdsc_free_path((char *)path1, pathlen1);
2f2dc053
SW		 2689out:
2690 return msg;
2691}
2692
2693/*
2694 * called under mdsc->mutex if error, under no mutex if
2695 * success.
2696 */
2697static void complete_request(struct ceph_mds_client *mdsc,
2698 struct ceph_mds_request *req)
2699{
70c94820
XL		 2700 req->r_end_latency = ktime_get();
2701
2f2dc053
SW		 2702 if (req->r_callback)
2703 req->r_callback(mdsc, req);
111c7081 2704 complete_all(&req->r_completion);
2f2dc053
SW		 2705}
2706
4f1ddb1e
JL		 2707static struct ceph_mds_request_head_old *
2708find_old_request_head(void *p, u64 features)
2709{
2710 bool legacy = !(features & CEPH_FEATURE_FS_BTIME);
2711 struct ceph_mds_request_head *new_head;
2712
2713 if (legacy)
2714 return (struct ceph_mds_request_head_old *)p;
2715 new_head = (struct ceph_mds_request_head *)p;
2716 return (struct ceph_mds_request_head_old *)&new_head->oldest_client_tid;
2717}
2718
2f2dc053
SW		 2719/*
2720 * called under mdsc->mutex
2721 */
396bd62c 2722static int __prepare_send_request(struct ceph_mds_session *session,
2f2dc053 2723 struct ceph_mds_request *req,
396bd62c 2724 bool drop_cap_releases)
2f2dc053 2725{
396bd62c
JL		 2726 int mds = session->s_mds;
2727 struct ceph_mds_client *mdsc = session->s_mdsc;
4f1ddb1e 2728 struct ceph_mds_request_head_old *rhead;
2f2dc053 2729 struct ceph_msg *msg;
546a5d61
XL		 2730 int flags = 0, max_retry;
2731
2732 /*
2733 * The type of 'r_attempts' in kernel 'ceph_mds_request'
2734 * is 'int', while in 'ceph_mds_request_head' the type of
2735 * 'num_retry' is '__u8'. So in case the request retries
2736 * exceeding 256 times, the MDS will receive a incorrect
2737 * retry seq.
2738 *
2739 * In this case it's ususally a bug in MDS and continue
2740 * retrying the request makes no sense.
2741 *
2742 * In future this could be fixed in ceph code, so avoid
2743 * using the hardcode here.
2744 */
2745 max_retry = sizeof_field(struct ceph_mds_request_head, num_retry);
2746 max_retry = 1 << (max_retry * BITS_PER_BYTE);
2747 if (req->r_attempts >= max_retry) {
2748 pr_warn_ratelimited("%s request tid %llu seq overflow\n",
2749 __func__, req->r_tid);
2750 return -EMULTIHOP;
2751 }
2f2dc053 2752
2f2dc053 2753 req->r_attempts++;
e55b71f8
GF		 2754 if (req->r_inode) {
2755 struct ceph_cap *cap =
2756 ceph_get_cap_for_mds(ceph_inode(req->r_inode), mds);
2757
2758 if (cap)
2759 req->r_sent_on_mseq = cap->mseq;
2760 else
2761 req->r_sent_on_mseq = -1;
2762 }
546a5d61 2763 dout("%s %p tid %lld %s (attempt %d)\n", __func__, req,
2f2dc053
SW		 2764 req->r_tid, ceph_mds_op_name(req->r_op), req->r_attempts);
2765
bc2de10d 2766 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
c5c9a0bf 2767 void *p;
4f1ddb1e 2768
01a92f17
SW		 2769 /*
2770 * Replay. Do not regenerate message (and rebuild
2771 * paths, etc.); just use the original message.
2772 * Rebuilding paths will break for renames because
2773 * d_move mangles the src name.
2774 */
2775 msg = req->r_request;
4f1ddb1e
JL		 2776 rhead = find_old_request_head(msg->front.iov_base,
2777 session->s_con.peer_features);
01a92f17
SW		 2778
2779 flags = le32_to_cpu(rhead->flags);
2780 flags |= CEPH_MDS_FLAG_REPLAY;
2781 rhead->flags = cpu_to_le32(flags);
2782
2783 if (req->r_target_inode)
2784 rhead->ino = cpu_to_le64(ceph_ino(req->r_target_inode));
2785
2786 rhead->num_retry = req->r_attempts - 1;
e979cf50
SW		 2787
2788 /* remove cap/dentry releases from message */
2789 rhead->num_releases = 0;
c5c9a0bf 2790
c5c9a0bf 2791 p = msg->front.iov_base + req->r_request_release_offset;
60267ba3 2792 encode_timestamp_and_gids(&p, req);
c5c9a0bf
YZ		 2793
2794 msg->front.iov_len = p - msg->front.iov_base;
2795 msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
01a92f17
SW		 2796 return 0;
2797 }
2798
2f2dc053
SW		 2799 if (req->r_request) {
2800 ceph_msg_put(req->r_request);
2801 req->r_request = NULL;
2802 }
4f1ddb1e 2803 msg = create_request_message(session, req, drop_cap_releases);
2f2dc053 2804 if (IS_ERR(msg)) {
e1518c7c 2805 req->r_err = PTR_ERR(msg);
a79832f2 2806 return PTR_ERR(msg);
2f2dc053
SW		 2807 }
2808 req->r_request = msg;
2809
4f1ddb1e
JL		 2810 rhead = find_old_request_head(msg->front.iov_base,
2811 session->s_con.peer_features);
2f2dc053 2812 rhead->oldest_client_tid = cpu_to_le64(__get_oldest_tid(mdsc));
bc2de10d 2813 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags))
2f2dc053 2814 flags |= CEPH_MDS_FLAG_REPLAY;
3bb48b41
JL		 2815 if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags))
2816 flags |= CEPH_MDS_FLAG_ASYNC;
3dd69aab 2817 if (req->r_parent)
2f2dc053
SW		 2818 flags |= CEPH_MDS_FLAG_WANT_DENTRY;
2819 rhead->flags = cpu_to_le32(flags);
2820 rhead->num_fwd = req->r_num_fwd;
2821 rhead->num_retry = req->r_attempts - 1;
2822
3dd69aab 2823 dout(" r_parent = %p\n", req->r_parent);
2f2dc053
SW		 2824 return 0;
2825}
2826
9cf54563
XL		 2827/*
2828 * called under mdsc->mutex
2829 */
396bd62c 2830static int __send_request(struct ceph_mds_session *session,
9cf54563
XL		 2831 struct ceph_mds_request *req,
2832 bool drop_cap_releases)
2833{
2834 int err;
2835
396bd62c 2836 err = __prepare_send_request(session, req, drop_cap_releases);
9cf54563
XL		 2837 if (!err) {
2838 ceph_msg_get(req->r_request);
2839 ceph_con_send(&session->s_con, req->r_request);
2840 }
2841
2842 return err;
2843}
2844
2f2dc053
SW		 2845/*
2846 * send request, or put it on the appropriate wait list.
2847 */
d5548492 2848static void __do_request(struct ceph_mds_client *mdsc,
2f2dc053
SW		 2849 struct ceph_mds_request *req)
2850{
2851 struct ceph_mds_session *session = NULL;
2852 int mds = -1;
48fec5d0 2853 int err = 0;
c4853e97 2854 bool random;
2f2dc053 2855
bc2de10d
JL		 2856 if (req->r_err || test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags)) {
2857 if (test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags))
eb1b8af3 2858 __unregister_request(mdsc, req);
d5548492 2859 return;
eb1b8af3 2860 }
2f2dc053
SW		 2861
2862 if (req->r_timeout &&
2863 time_after_eq(jiffies, req->r_started + req->r_timeout)) {
2864 dout("do_request timed out\n");
8ccf7fcc 2865 err = -ETIMEDOUT;
2f2dc053
SW		 2866 goto finish;
2867 }
52953d55 2868 if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_SHUTDOWN) {
48fec5d0
YZ		 2869 dout("do_request forced umount\n");
2870 err = -EIO;
2871 goto finish;
2872 }
52953d55 2873 if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_MOUNTING) {
e9e427f0
YZ		 2874 if (mdsc->mdsmap_err) {
2875 err = mdsc->mdsmap_err;
2876 dout("do_request mdsmap err %d\n", err);
2877 goto finish;
2878 }
cc8e8342
YZ		 2879 if (mdsc->mdsmap->m_epoch == 0) {
2880 dout("do_request no mdsmap, waiting for map\n");
2881 list_add(&req->r_wait, &mdsc->waiting_for_map);
d5548492 2882 return;
cc8e8342 2883 }
e9e427f0
YZ		 2884 if (!(mdsc->fsc->mount_options->flags &
2885 CEPH_MOUNT_OPT_MOUNTWAIT) &&
2886 !ceph_mdsmap_is_cluster_available(mdsc->mdsmap)) {
97820058 2887 err = -EHOSTUNREACH;
e9e427f0
YZ		 2888 goto finish;
2889 }
2890 }
2f2dc053 2891
dc69e2e9
SW		 2892 put_request_session(req);
2893
c4853e97 2894 mds = __choose_mds(mdsc, req, &random);
2f2dc053
SW		 2895 if (mds < 0 ||
2896 ceph_mdsmap_get_state(mdsc->mdsmap, mds) < CEPH_MDS_STATE_ACTIVE) {
3bb48b41
JL		 2897 if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags)) {
2898 err = -EJUKEBOX;
2899 goto finish;
2900 }
2f2dc053
SW		 2901 dout("do_request no mds or not active, waiting for map\n");
2902 list_add(&req->r_wait, &mdsc->waiting_for_map);
d5548492 2903 return;
2f2dc053
SW		 2904 }
2905
2906 /* get, open session */
2907 session = __ceph_lookup_mds_session(mdsc, mds);
9c423956 2908 if (!session) {
2f2dc053 2909 session = register_session(mdsc, mds);
9c423956
SW		 2910 if (IS_ERR(session)) {
2911 err = PTR_ERR(session);
2912 goto finish;
2913 }
2914 }
5b3248c6 2915 req->r_session = ceph_get_mds_session(session);
dc69e2e9 2916
2f2dc053 2917 dout("do_request mds%d session %p state %s\n", mds, session,
a687ecaf 2918 ceph_session_state_name(session->s_state));
2f2dc053
SW		 2919 if (session->s_state != CEPH_MDS_SESSION_OPEN &&
2920 session->s_state != CEPH_MDS_SESSION_HUNG) {
3bb48b41
JL		 2921 /*
2922 * We cannot queue async requests since the caps and delegated
2923 * inodes are bound to the session. Just return -EJUKEBOX and
2924 * let the caller retry a sync request in that case.
2925 */
2926 if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags)) {
2927 err = -EJUKEBOX;
2928 goto out_session;
2929 }
4ae3713f
JL		 2930
2931 /*
2932 * If the session has been REJECTED, then return a hard error,
2933 * unless it's a CLEANRECOVER mount, in which case we'll queue
2934 * it to the mdsc queue.
2935 */
2936 if (session->s_state == CEPH_MDS_SESSION_REJECTED) {
2937 if (ceph_test_mount_opt(mdsc->fsc, CLEANRECOVER))
2938 list_add(&req->r_wait, &mdsc->waiting_for_map);
2939 else
2940 err = -EACCES;
2941 goto out_session;
2942 }
2943
2f2dc053 2944 if (session->s_state == CEPH_MDS_SESSION_NEW ||
c4853e97 2945 session->s_state == CEPH_MDS_SESSION_CLOSING) {
b682c6d4
XL		 2946 err = __open_session(mdsc, session);
2947 if (err)
2948 goto out_session;
c4853e97
XL		 2949 /* retry the same mds later */
2950 if (random)
2951 req->r_resend_mds = mds;
2952 }
2f2dc053
SW		 2953 list_add(&req->r_wait, &session->s_waiting);
2954 goto out_session;
2955 }
2956
2957 /* send request */
2f2dc053
SW		 2958 req->r_resend_mds = -1; /* forget any previous mds hint */
2959
2960 if (req->r_request_started == 0) /* note request start time */
2961 req->r_request_started = jiffies;
2962
00061645
XL		 2963 /*
2964 * For async create we will choose the auth MDS of frag in parent
2965 * directory to send the request and ususally this works fine, but
2966 * if the migrated the dirtory to another MDS before it could handle
2967 * it the request will be forwarded.
2968 *
2969 * And then the auth cap will be changed.
2970 */
2971 if (test_bit(CEPH_MDS_R_ASYNC, &req->r_req_flags) && req->r_num_fwd) {
2972 struct ceph_dentry_info *di = ceph_dentry(req->r_dentry);
2973 struct ceph_inode_info *ci;
2974 struct ceph_cap *cap;
2975
2976 /*
2977 * The request maybe handled very fast and the new inode
2978 * hasn't been linked to the dentry yet. We need to wait
2979 * for the ceph_finish_async_create(), which shouldn't be
2980 * stuck too long or fail in thoery, to finish when forwarding
2981 * the request.
2982 */
2983 if (!d_inode(req->r_dentry)) {
2984 err = wait_on_bit(&di->flags, CEPH_DENTRY_ASYNC_CREATE_BIT,
2985 TASK_KILLABLE);
2986 if (err) {
2987 mutex_lock(&req->r_fill_mutex);
2988 set_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags);
2989 mutex_unlock(&req->r_fill_mutex);
2990 goto out_session;
2991 }
2992 }
2993
2994 ci = ceph_inode(d_inode(req->r_dentry));
2995
2996 spin_lock(&ci->i_ceph_lock);
2997 cap = ci->i_auth_cap;
2998 if (ci->i_ceph_flags & CEPH_I_ASYNC_CREATE && mds != cap->mds) {
2999 dout("do_request session changed for auth cap %d -> %d\n",
3000 cap->session->s_mds, session->s_mds);
3001
3002 /* Remove the auth cap from old session */
3003 spin_lock(&cap->session->s_cap_lock);
3004 cap->session->s_nr_caps--;
3005 list_del_init(&cap->session_caps);
3006 spin_unlock(&cap->session->s_cap_lock);
3007
3008 /* Add the auth cap to the new session */
3009 cap->mds = mds;
3010 cap->session = session;
3011 spin_lock(&session->s_cap_lock);
3012 session->s_nr_caps++;
3013 list_add_tail(&cap->session_caps, &session->s_caps);
3014 spin_unlock(&session->s_cap_lock);
3015
3016 change_auth_cap_ses(ci, session);
3017 }
3018 spin_unlock(&ci->i_ceph_lock);
3019 }
3020
396bd62c 3021 err = __send_request(session, req, false);
2f2dc053
SW		 3022
3023out_session:
3024 ceph_put_mds_session(session);
48fec5d0
YZ		 3025finish:
3026 if (err) {
3027 dout("__do_request early error %d\n", err);
3028 req->r_err = err;
3029 complete_request(mdsc, req);
3030 __unregister_request(mdsc, req);
3031 }
d5548492 3032 return;
2f2dc053
SW		 3033}
3034
3035/*
3036 * called under mdsc->mutex
3037 */
3038static void __wake_requests(struct ceph_mds_client *mdsc,
3039 struct list_head *head)
3040{
ed75ec2c
YZ		 3041 struct ceph_mds_request *req;
3042 LIST_HEAD(tmp_list);
3043
3044 list_splice_init(head, &tmp_list);
2f2dc053 3045
ed75ec2c
YZ		 3046 while (!list_empty(&tmp_list)) {
3047 req = list_entry(tmp_list.next,
3048 struct ceph_mds_request, r_wait);
2f2dc053 3049 list_del_init(&req->r_wait);
7971bd92 3050 dout(" wake request %p tid %llu\n", req, req->r_tid);
2f2dc053
SW		 3051 __do_request(mdsc, req);
3052 }
3053}
3054
3055/*
3056 * Wake up threads with requests pending for @mds, so that they can
29790f26 3057 * resubmit their requests to a possibly different mds.
2f2dc053 3058 */
29790f26 3059static void kick_requests(struct ceph_mds_client *mdsc, int mds)
2f2dc053 3060{
44ca18f2 3061 struct ceph_mds_request *req;
282c1052 3062 struct rb_node *p = rb_first(&mdsc->request_tree);
2f2dc053
SW		 3063
3064 dout("kick_requests mds%d\n", mds);
282c1052 3065 while (p) {
44ca18f2 3066 req = rb_entry(p, struct ceph_mds_request, r_node);
282c1052 3067 p = rb_next(p);
bc2de10d 3068 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags))
44ca18f2 3069 continue;
3de22be6
YZ		 3070 if (req->r_attempts > 0)
3071 continue; /* only new requests */
44ca18f2
SW		 3072 if (req->r_session &&
3073 req->r_session->s_mds == mds) {
3074 dout(" kicking tid %llu\n", req->r_tid);
03974e81 3075 list_del_init(&req->r_wait);
44ca18f2 3076 __do_request(mdsc, req);
2f2dc053
SW		 3077 }
3078 }
3079}
3080
86bda539 3081int ceph_mdsc_submit_request(struct ceph_mds_client *mdsc, struct inode *dir,
2f2dc053
SW		 3082 struct ceph_mds_request *req)
3083{
891f3f5a 3084 int err = 0;
86bda539
JL		 3085
3086 /* take CAP_PIN refs for r_inode, r_parent, r_old_dentry */
3087 if (req->r_inode)
3088 ceph_get_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
9c1c2b35 3089 if (req->r_parent) {
719a2514
YZ		 3090 struct ceph_inode_info *ci = ceph_inode(req->r_parent);
3091 int fmode = (req->r_op & CEPH_MDS_OP_WRITE) ?
3092 CEPH_FILE_MODE_WR : CEPH_FILE_MODE_RD;
3093 spin_lock(&ci->i_ceph_lock);
3094 ceph_take_cap_refs(ci, CEPH_CAP_PIN, false);
3095 __ceph_touch_fmode(ci, mdsc, fmode);
3096 spin_unlock(&ci->i_ceph_lock);
9c1c2b35 3097 }
86bda539
JL		 3098 if (req->r_old_dentry_dir)
3099 ceph_get_cap_refs(ceph_inode(req->r_old_dentry_dir),
3100 CEPH_CAP_PIN);
3101
891f3f5a
JL		 3102 if (req->r_inode) {
3103 err = ceph_wait_on_async_create(req->r_inode);
3104 if (err) {
3105 dout("%s: wait for async create returned: %d\n",
3106 __func__, err);
3107 return err;
3108 }
3109 }
3110
3111 if (!err && req->r_old_inode) {
3112 err = ceph_wait_on_async_create(req->r_old_inode);
3113 if (err) {
3114 dout("%s: wait for async create returned: %d\n",
3115 __func__, err);
3116 return err;
3117 }
3118 }
3119
86bda539 3120 dout("submit_request on %p for inode %p\n", req, dir);
2f2dc053 3121 mutex_lock(&mdsc->mutex);
86bda539 3122 __register_request(mdsc, req, dir);
2f2dc053 3123 __do_request(mdsc, req);
86bda539 3124 err = req->r_err;
2f2dc053 3125 mutex_unlock(&mdsc->mutex);
86bda539 3126 return err;
2f2dc053
SW		 3127}
3128
9eaa7b79
JL		 3129int ceph_mdsc_wait_request(struct ceph_mds_client *mdsc,
3130 struct ceph_mds_request *req,
3131 ceph_mds_request_wait_callback_t wait_func)
2f2dc053
SW		 3132{
3133 int err;
3134
e1518c7c 3135 /* wait */
e1518c7c 3136 dout("do_request waiting\n");
9eaa7b79
JL		 3137 if (wait_func) {
3138 err = wait_func(mdsc, req);
e1518c7c 3139 } else {
5be73034
ID		 3140 long timeleft = wait_for_completion_killable_timeout(
3141 &req->r_completion,
3142 ceph_timeout_jiffies(req->r_timeout));
3143 if (timeleft > 0)
3144 err = 0;
3145 else if (!timeleft)
8ccf7fcc 3146 err = -ETIMEDOUT; /* timed out */
5be73034
ID		 3147 else
3148 err = timeleft; /* killed */
e1518c7c
SW		 3149 }
3150 dout("do_request waited, got %d\n", err);
3151 mutex_lock(&mdsc->mutex);
5b1daecd 3152
e1518c7c 3153 /* only abort if we didn't race with a real reply */
bc2de10d 3154 if (test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags)) {
e1518c7c
SW		 3155 err = le32_to_cpu(req->r_reply_info.head->result);
3156 } else if (err < 0) {
3157 dout("aborted request %lld with %d\n", req->r_tid, err);
b4556396
SW		 3158
3159 /*
3160 * ensure we aren't running concurrently with
3161 * ceph_fill_trace or ceph_readdir_prepopulate, which
3162 * rely on locks (dir mutex) held by our caller.
3163 */
3164 mutex_lock(&req->r_fill_mutex);
e1518c7c 3165 req->r_err = err;
bc2de10d 3166 set_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags);
b4556396 3167 mutex_unlock(&req->r_fill_mutex);
5b1daecd 3168
3dd69aab 3169 if (req->r_parent &&
167c9e35
SW		 3170 (req->r_op & CEPH_MDS_OP_WRITE))
3171 ceph_invalidate_dir_request(req);
2f2dc053 3172 } else {
e1518c7c 3173 err = req->r_err;
2f2dc053 3174 }
2f2dc053 3175
e1518c7c 3176 mutex_unlock(&mdsc->mutex);
8340f22c
JL		 3177 return err;
3178}
3179
3180/*
3181 * Synchrously perform an mds request. Take care of all of the
3182 * session setup, forwarding, retry details.
3183 */
3184int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
3185 struct inode *dir,
3186 struct ceph_mds_request *req)
3187{
3188 int err;
3189
3190 dout("do_request on %p\n", req);
3191
3192 /* issue */
3193 err = ceph_mdsc_submit_request(mdsc, dir, req);
3194 if (!err)
9eaa7b79 3195 err = ceph_mdsc_wait_request(mdsc, req, NULL);
2f2dc053
SW		 3196 dout("do_request %p done, result %d\n", req, err);
3197 return err;
3198}
3199
167c9e35 3200/*
2f276c51 3201 * Invalidate dir's completeness, dentry lease state on an aborted MDS
167c9e35
SW		 3202 * namespace request.
3203 */
3204void ceph_invalidate_dir_request(struct ceph_mds_request *req)
3205{
8d8f371c
YZ		 3206 struct inode *dir = req->r_parent;
3207 struct inode *old_dir = req->r_old_dentry_dir;
167c9e35 3208
8d8f371c 3209 dout("invalidate_dir_request %p %p (complete, lease(s))\n", dir, old_dir);
167c9e35 3210
8d8f371c
YZ		 3211 ceph_dir_clear_complete(dir);
3212 if (old_dir)
3213 ceph_dir_clear_complete(old_dir);
167c9e35
SW		 3214 if (req->r_dentry)
3215 ceph_invalidate_dentry_lease(req->r_dentry);
3216 if (req->r_old_dentry)
3217 ceph_invalidate_dentry_lease(req->r_old_dentry);
3218}
3219
2f2dc053
SW		 3220/*
3221 * Handle mds reply.
3222 *
3223 * We take the session mutex and parse and process the reply immediately.
3224 * This preserves the logical ordering of replies, capabilities, etc., sent
3225 * by the MDS as they are applied to our local cache.
3226 */
3227static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
3228{
3229 struct ceph_mds_client *mdsc = session->s_mdsc;
3230 struct ceph_mds_request *req;
3231 struct ceph_mds_reply_head *head = msg->front.iov_base;
3232 struct ceph_mds_reply_info_parsed *rinfo; /* parsed reply info */
982d6011 3233 struct ceph_snap_realm *realm;
2f2dc053
SW		 3234 u64 tid;
3235 int err, result;
2600d2dd 3236 int mds = session->s_mds;
2f2dc053 3237
2f2dc053
SW		 3238 if (msg->front.iov_len < sizeof(*head)) {
3239 pr_err("mdsc_handle_reply got corrupt (short) reply\n");
9ec7cab1 3240 ceph_msg_dump(msg);
2f2dc053
SW		 3241 return;
3242 }
3243
3244 /* get request, session */
6df058c0 3245 tid = le64_to_cpu(msg->hdr.tid);
2f2dc053 3246 mutex_lock(&mdsc->mutex);
fcd00b68 3247 req = lookup_get_request(mdsc, tid);
2f2dc053
SW		 3248 if (!req) {
3249 dout("handle_reply on unknown tid %llu\n", tid);
3250 mutex_unlock(&mdsc->mutex);
3251 return;
3252 }
3253 dout("handle_reply %p\n", req);
2f2dc053
SW		 3254
3255 /* correct session? */
d96d6049 3256 if (req->r_session != session) {
2f2dc053
SW		 3257 pr_err("mdsc_handle_reply got %llu on session mds%d"
3258 " not mds%d\n", tid, session->s_mds,
3259 req->r_session ? req->r_session->s_mds : -1);
3260 mutex_unlock(&mdsc->mutex);
3261 goto out;
3262 }
3263
3264 /* dup? */
bc2de10d
JL		 3265 if ((test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags) && !head->safe) ||
3266 (test_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags) && head->safe)) {
f3ae1b97 3267 pr_warn("got a dup %s reply on %llu from mds%d\n",
2f2dc053
SW		 3268 head->safe ? "safe" : "unsafe", tid, mds);
3269 mutex_unlock(&mdsc->mutex);
3270 goto out;
3271 }
bc2de10d 3272 if (test_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags)) {
f3ae1b97 3273 pr_warn("got unsafe after safe on %llu from mds%d\n",
85792d0d
SW		 3274 tid, mds);
3275 mutex_unlock(&mdsc->mutex);
3276 goto out;
3277 }
2f2dc053
SW		 3278
3279 result = le32_to_cpu(head->result);
3280
2f2dc053 3281 if (head->safe) {
bc2de10d 3282 set_bit(CEPH_MDS_R_GOT_SAFE, &req->r_req_flags);
2f2dc053 3283 __unregister_request(mdsc, req);
2f2dc053 3284
07edc057
XL		 3285 /* last request during umount? */
3286 if (mdsc->stopping && !__get_oldest_req(mdsc))
3287 complete_all(&mdsc->safe_umount_waiters);
3288
bc2de10d 3289 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
2f2dc053
SW		 3290 /*
3291 * We already handled the unsafe response, now do the
3292 * cleanup. No need to examine the response; the MDS
3293 * doesn't include any result info in the safe
3294 * response. And even if it did, there is nothing
3295 * useful we could do with a revised return value.
3296 */
3297 dout("got safe reply %llu, mds%d\n", tid, mds);
2f2dc053 3298
2f2dc053
SW		 3299 mutex_unlock(&mdsc->mutex);
3300 goto out;
3301 }
e1518c7c 3302 } else {
bc2de10d 3303 set_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags);
2f2dc053
SW		 3304 list_add_tail(&req->r_unsafe_item, &req->r_session->s_unsafe);
3305 }
3306
3307 dout("handle_reply tid %lld result %d\n", tid, result);
3308 rinfo = &req->r_reply_info;
b37fe1f9 3309 if (test_bit(CEPHFS_FEATURE_REPLY_ENCODING, &session->s_features))
d4846487 3310 err = parse_reply_info(session, msg, rinfo, (u64)-1);
b37fe1f9 3311 else
d4846487 3312 err = parse_reply_info(session, msg, rinfo, session->s_con.peer_features);
2f2dc053
SW		 3313 mutex_unlock(&mdsc->mutex);
3314
bca9fc14
JL		 3315 /* Must find target inode outside of mutexes to avoid deadlocks */
3316 if ((err >= 0) && rinfo->head->is_target) {
3317 struct inode *in;
3318 struct ceph_vino tvino = {
3319 .ino = le64_to_cpu(rinfo->targeti.in->ino),
3320 .snap = le64_to_cpu(rinfo->targeti.in->snapid)
3321 };
3322
3323 in = ceph_get_inode(mdsc->fsc->sb, tvino);
3324 if (IS_ERR(in)) {
3325 err = PTR_ERR(in);
3326 mutex_lock(&session->s_mutex);
3327 goto out_err;
3328 }
3329 req->r_target_inode = in;
3330 }
3331
2f2dc053
SW		 3332 mutex_lock(&session->s_mutex);
3333 if (err < 0) {
25933abd 3334 pr_err("mdsc_handle_reply got corrupt reply mds%d(tid:%lld)\n", mds, tid);
9ec7cab1 3335 ceph_msg_dump(msg);
2f2dc053
SW		 3336 goto out_err;
3337 }
3338
3339 /* snap trace */
982d6011 3340 realm = NULL;
2f2dc053
SW		 3341 if (rinfo->snapblob_len) {
3342 down_write(&mdsc->snap_rwsem);
3343 ceph_update_snap_trace(mdsc, rinfo->snapblob,
982d6011
YZ		 3344 rinfo->snapblob + rinfo->snapblob_len,
3345 le32_to_cpu(head->op) == CEPH_MDS_OP_RMSNAP,
3346 &realm);
2f2dc053
SW		 3347 downgrade_write(&mdsc->snap_rwsem);
3348 } else {
3349 down_read(&mdsc->snap_rwsem);
3350 }
3351
3352 /* insert trace into our cache */
b4556396 3353 mutex_lock(&req->r_fill_mutex);
315f2408 3354 current->journal_info = req;
f5a03b08 3355 err = ceph_fill_trace(mdsc->fsc->sb, req);
2f2dc053 3356 if (err == 0) {
6e8575fa 3357 if (result == 0 && (req->r_op == CEPH_MDS_OP_READDIR ||
81c6aea5 3358 req->r_op == CEPH_MDS_OP_LSSNAP))
2f2dc053 3359 ceph_readdir_prepopulate(req, req->r_session);
2f2dc053 3360 }
315f2408 3361 current->journal_info = NULL;
b4556396 3362 mutex_unlock(&req->r_fill_mutex);
2f2dc053
SW		 3363
3364 up_read(&mdsc->snap_rwsem);
982d6011
YZ		 3365 if (realm)
3366 ceph_put_snap_realm(mdsc, realm);
68cd5b4b 3367
fe33032d
YZ		 3368 if (err == 0) {
3369 if (req->r_target_inode &&
3370 test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags)) {
3371 struct ceph_inode_info *ci =
3372 ceph_inode(req->r_target_inode);
3373 spin_lock(&ci->i_unsafe_lock);
3374 list_add_tail(&req->r_unsafe_target_item,
3375 &ci->i_unsafe_iops);
3376 spin_unlock(&ci->i_unsafe_lock);
3377 }
3378
3379 ceph_unreserve_caps(mdsc, &req->r_caps_reservation);
68cd5b4b 3380 }
2f2dc053 3381out_err:
e1518c7c 3382 mutex_lock(&mdsc->mutex);
bc2de10d 3383 if (!test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags)) {
e1518c7c
SW		 3384 if (err) {
3385 req->r_err = err;
3386 } else {
5fdb1389 3387 req->r_reply = ceph_msg_get(msg);
bc2de10d 3388 set_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags);
e1518c7c 3389 }
2f2dc053 3390 } else {
e1518c7c 3391 dout("reply arrived after request %lld was aborted\n", tid);
2f2dc053 3392 }
e1518c7c 3393 mutex_unlock(&mdsc->mutex);
2f2dc053 3394
2f2dc053
SW		 3395 mutex_unlock(&session->s_mutex);
3396
3397 /* kick calling process */
3398 complete_request(mdsc, req);
70c94820 3399
8ae99ae2 3400 ceph_update_metadata_metrics(&mdsc->metric, req->r_start_latency,
70c94820 3401 req->r_end_latency, err);
2f2dc053
SW		 3402out:
3403 ceph_mdsc_put_request(req);
3404 return;
3405}
3406
3407
3408
3409/*
3410 * handle mds notification that our request has been forwarded.
3411 */
2600d2dd
SW		 3412static void handle_forward(struct ceph_mds_client *mdsc,
3413 struct ceph_mds_session *session,
3414 struct ceph_msg *msg)
2f2dc053
SW		 3415{
3416 struct ceph_mds_request *req;
a1ea787c 3417 u64 tid = le64_to_cpu(msg->hdr.tid);
2f2dc053
SW		 3418 u32 next_mds;
3419 u32 fwd_seq;
2f2dc053
SW		 3420 int err = -EINVAL;
3421 void *p = msg->front.iov_base;
3422 void *end = p + msg->front.iov_len;
1980b1bf 3423 bool aborted = false;
2f2dc053 3424
a1ea787c 3425 ceph_decode_need(&p, end, 2*sizeof(u32), bad);
c89136ea
SW		 3426 next_mds = ceph_decode_32(&p);
3427 fwd_seq = ceph_decode_32(&p);
2f2dc053
SW		 3428
3429 mutex_lock(&mdsc->mutex);
fcd00b68 3430 req = lookup_get_request(mdsc, tid);
2f2dc053 3431 if (!req) {
1980b1bf 3432 mutex_unlock(&mdsc->mutex);
2a8e5e36 3433 dout("forward tid %llu to mds%d - req dne\n", tid, next_mds);
1980b1bf 3434 return; /* dup reply? */
2f2dc053
SW		 3435 }
3436
bc2de10d 3437 if (test_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags)) {
2a8e5e36
SW		 3438 dout("forward tid %llu aborted, unregistering\n", tid);
3439 __unregister_request(mdsc, req);
3440 } else if (fwd_seq <= req->r_num_fwd) {
1980b1bf
XL		 3441 /*
3442 * The type of 'num_fwd' in ceph 'MClientRequestForward'
3443 * is 'int32_t', while in 'ceph_mds_request_head' the
3444 * type is '__u8'. So in case the request bounces between
3445 * MDSes exceeding 256 times, the client will get stuck.
3446 *
3447 * In this case it's ususally a bug in MDS and continue
3448 * bouncing the request makes no sense.
3449 *
3450 * In future this could be fixed in ceph code, so avoid
3451 * using the hardcode here.
3452 */
3453 int max = sizeof_field(struct ceph_mds_request_head, num_fwd);
3454 max = 1 << (max * BITS_PER_BYTE);
3455 if (req->r_num_fwd >= max) {
3456 mutex_lock(&req->r_fill_mutex);
3457 req->r_err = -EMULTIHOP;
3458 set_bit(CEPH_MDS_R_ABORTED, &req->r_req_flags);
3459 mutex_unlock(&req->r_fill_mutex);
3460 aborted = true;
3461 pr_warn_ratelimited("forward tid %llu seq overflow\n",
3462 tid);
3463 } else {
3464 dout("forward tid %llu to mds%d - old seq %d <= %d\n",
3465 tid, next_mds, req->r_num_fwd, fwd_seq);
3466 }
2f2dc053
SW		 3467 } else {
3468 /* resend. forward race not possible; mds would drop */
2a8e5e36
SW		 3469 dout("forward tid %llu to mds%d (we resend)\n", tid, next_mds);
3470 BUG_ON(req->r_err);
bc2de10d 3471 BUG_ON(test_bit(CEPH_MDS_R_GOT_RESULT, &req->r_req_flags));
3de22be6 3472 req->r_attempts = 0;
2f2dc053
SW		 3473 req->r_num_fwd = fwd_seq;
3474 req->r_resend_mds = next_mds;
3475 put_request_session(req);
3476 __do_request(mdsc, req);
3477 }
2f2dc053 3478 mutex_unlock(&mdsc->mutex);
1980b1bf
XL		 3479
3480 /* kick calling process */
3481 if (aborted)
3482 complete_request(mdsc, req);
3483 ceph_mdsc_put_request(req);
2f2dc053
SW		 3484 return;
3485
3486bad:
3487 pr_err("mdsc_handle_forward decode error err=%d\n", err);
3488}
3489
131d7eb4 3490static int __decode_session_metadata(void **p, void *end,
0b98acd6 3491 bool *blocklisted)
84bf3950
YZ		 3492{
3493 /* map<string,string> */
3494 u32 n;
131d7eb4 3495 bool err_str;
84bf3950
YZ		 3496 ceph_decode_32_safe(p, end, n, bad);
3497 while (n-- > 0) {
3498 u32 len;
3499 ceph_decode_32_safe(p, end, len, bad);
3500 ceph_decode_need(p, end, len, bad);
131d7eb4 3501 err_str = !strncmp(*p, "error_string", len);
84bf3950
YZ		 3502 *p += len;
3503 ceph_decode_32_safe(p, end, len, bad);
3504 ceph_decode_need(p, end, len, bad);
4bb926e8
ID		 3505 /*
3506 * Match "blocklisted (blacklisted)" from newer MDSes,
3507 * or "blacklisted" from older MDSes.
3508 */
131d7eb4 3509 if (err_str && strnstr(*p, "blacklisted", len))
0b98acd6 3510 *blocklisted = true;
84bf3950
YZ		 3511 *p += len;
3512 }
3513 return 0;
3514bad:
3515 return -1;
3516}
3517
2f2dc053
SW		 3518/*
3519 * handle a mds session control message
3520 */
3521static void handle_session(struct ceph_mds_session *session,
3522 struct ceph_msg *msg)
3523{
3524 struct ceph_mds_client *mdsc = session->s_mdsc;
84bf3950
YZ		 3525 int mds = session->s_mds;
3526 int msg_version = le16_to_cpu(msg->hdr.version);
3527 void *p = msg->front.iov_base;
3528 void *end = p + msg->front.iov_len;
3529 struct ceph_mds_session_head *h;
2f2dc053 3530 u32 op;
0fa82633 3531 u64 seq, features = 0;
2f2dc053 3532 int wake = 0;
0b98acd6 3533 bool blocklisted = false;
2f2dc053 3534
2f2dc053 3535 /* decode */
84bf3950
YZ		 3536 ceph_decode_need(&p, end, sizeof(*h), bad);
3537 h = p;
3538 p += sizeof(*h);
3539
2f2dc053
SW		 3540 op = le32_to_cpu(h->op);
3541 seq = le64_to_cpu(h->seq);
3542
84bf3950
YZ		 3543 if (msg_version >= 3) {
3544 u32 len;
e1c9788c
KH		 3545 /* version >= 2 and < 5, decode metadata, skip otherwise
3546 * as it's handled via flags.
3547 */
3548 if (msg_version >= 5)
3549 ceph_decode_skip_map(&p, end, string, string, bad);
3550 else if (__decode_session_metadata(&p, end, &blocklisted) < 0)
84bf3950 3551 goto bad;
e1c9788c 3552
84bf3950
YZ		 3553 /* version >= 3, feature bits */
3554 ceph_decode_32_safe(&p, end, len, bad);
02e37571
JL		 3555 if (len) {
3556 ceph_decode_64_safe(&p, end, features, bad);
3557 p += len - sizeof(features);
3558 }
84bf3950
YZ		 3559 }
3560
e1c9788c 3561 if (msg_version >= 5) {
ea16567f
LH		 3562 u32 flags, len;
3563
3564 /* version >= 4 */
3565 ceph_decode_skip_16(&p, end, bad); /* struct_v, struct_cv */
3566 ceph_decode_32_safe(&p, end, len, bad); /* len */
3567 ceph_decode_skip_n(&p, end, len, bad); /* metric_spec */
3568
e1c9788c 3569 /* version >= 5, flags */
ea16567f 3570 ceph_decode_32_safe(&p, end, flags, bad);
e1c9788c 3571 if (flags & CEPH_SESSION_BLOCKLISTED) {
ea16567f 3572 pr_warn("mds%d session blocklisted\n", session->s_mds);
e1c9788c
KH		 3573 blocklisted = true;
3574 }
3575 }
3576
2f2dc053 3577 mutex_lock(&mdsc->mutex);
0a07fc8c 3578 if (op == CEPH_SESSION_CLOSE) {
5b3248c6 3579 ceph_get_mds_session(session);
2600d2dd 3580 __unregister_session(mdsc, session);
0a07fc8c 3581 }
2f2dc053
SW		 3582 /* FIXME: this ttl calculation is generous */
3583 session->s_ttl = jiffies + HZ*mdsc->mdsmap->m_session_autoclose;
3584 mutex_unlock(&mdsc->mutex);
3585
3586 mutex_lock(&session->s_mutex);
3587
3588 dout("handle_session mds%d %s %p state %s seq %llu\n",
3589 mds, ceph_session_op_name(op), session,
a687ecaf 3590 ceph_session_state_name(session->s_state), seq);
2f2dc053
SW		 3591
3592 if (session->s_state == CEPH_MDS_SESSION_HUNG) {
3593 session->s_state = CEPH_MDS_SESSION_OPEN;
3594 pr_info("mds%d came back\n", session->s_mds);
3595 }
3596
3597 switch (op) {
3598 case CEPH_SESSION_OPEN:
29790f26
SW		 3599 if (session->s_state == CEPH_MDS_SESSION_RECONNECTING)
3600 pr_info("mds%d reconnect success\n", session->s_mds);
300e42a2
XL		 3601
3602 if (session->s_state == CEPH_MDS_SESSION_OPEN) {
3603 pr_notice("mds%d is already opened\n", session->s_mds);
3604 } else {
3605 session->s_state = CEPH_MDS_SESSION_OPEN;
3606 session->s_features = features;
3607 renewed_caps(mdsc, session, 0);
3608 if (test_bit(CEPHFS_FEATURE_METRIC_COLLECT,
3609 &session->s_features))
3610 metric_schedule_delayed(&mdsc->metric);
3611 }
3612
3613 /*
3614 * The connection maybe broken and the session in client
3615 * side has been reinitialized, need to update the seq
3616 * anyway.
3617 */
3618 if (!session->s_seq && seq)
3619 session->s_seq = seq;
3620
2f2dc053
SW		 3621 wake = 1;
3622 if (mdsc->stopping)
3623 __close_session(mdsc, session);
3624 break;
3625
3626 case CEPH_SESSION_RENEWCAPS:
3627 if (session->s_renew_seq == seq)
3628 renewed_caps(mdsc, session, 1);
3629 break;
3630
3631 case CEPH_SESSION_CLOSE:
29790f26
SW		 3632 if (session->s_state == CEPH_MDS_SESSION_RECONNECTING)
3633 pr_info("mds%d reconnect denied\n", session->s_mds);
4d681c2f 3634 session->s_state = CEPH_MDS_SESSION_CLOSED;
1c841a96 3635 cleanup_session_requests(mdsc, session);
2f2dc053 3636 remove_session_caps(session);
656e4382 3637 wake = 2; /* for good measure */
f3c60c59 3638 wake_up_all(&mdsc->session_close_wq);
2f2dc053
SW		 3639 break;
3640
3641 case CEPH_SESSION_STALE:
3642 pr_info("mds%d caps went stale, renewing\n",
3643 session->s_mds);
52d60f8e 3644 atomic_inc(&session->s_cap_gen);
1ce208a6 3645 session->s_cap_ttl = jiffies - 1;
2f2dc053
SW		 3646 send_renew_caps(mdsc, session);
3647 break;
3648
3649 case CEPH_SESSION_RECALL_STATE:
e30ee581 3650 ceph_trim_caps(mdsc, session, le32_to_cpu(h->max_caps));
2f2dc053
SW		 3651 break;
3652
186e4f7a
YZ		 3653 case CEPH_SESSION_FLUSHMSG:
3654 send_flushmsg_ack(mdsc, session, seq);
3655 break;
3656
03f4fcb0
YZ		 3657 case CEPH_SESSION_FORCE_RO:
3658 dout("force_session_readonly %p\n", session);
3659 spin_lock(&session->s_cap_lock);
3660 session->s_readonly = true;
3661 spin_unlock(&session->s_cap_lock);
d2f8bb27 3662 wake_up_session_caps(session, FORCE_RO);
03f4fcb0
YZ		 3663 break;
3664
fcff415c
YZ		 3665 case CEPH_SESSION_REJECT:
3666 WARN_ON(session->s_state != CEPH_MDS_SESSION_OPENING);
3667 pr_info("mds%d rejected session\n", session->s_mds);
3668 session->s_state = CEPH_MDS_SESSION_REJECTED;
3669 cleanup_session_requests(mdsc, session);
3670 remove_session_caps(session);
0b98acd6
ID		 3671 if (blocklisted)
3672 mdsc->fsc->blocklisted = true;
fcff415c
YZ		 3673 wake = 2; /* for good measure */
3674 break;
3675
2f2dc053
SW		 3676 default:
3677 pr_err("mdsc_handle_session bad op %d mds%d\n", op, mds);
3678 WARN_ON(1);
3679 }
3680
3681 mutex_unlock(&session->s_mutex);
3682 if (wake) {
3683 mutex_lock(&mdsc->mutex);
3684 __wake_requests(mdsc, &session->s_waiting);
656e4382
YZ		 3685 if (wake == 2)
3686 kick_requests(mdsc, mds);
2f2dc053
SW		 3687 mutex_unlock(&mdsc->mutex);
3688 }
0a07fc8c
YZ		 3689 if (op == CEPH_SESSION_CLOSE)
3690 ceph_put_mds_session(session);
2f2dc053
SW		 3691 return;
3692
3693bad:
3694 pr_err("mdsc_handle_session corrupt message mds%d len %d\n", mds,
3695 (int)msg->front.iov_len);
9ec7cab1 3696 ceph_msg_dump(msg);
2f2dc053
SW		 3697 return;
3698}
3699
a25949b9
JL		 3700void ceph_mdsc_release_dir_caps(struct ceph_mds_request *req)
3701{
3702 int dcaps;
3703
3704 dcaps = xchg(&req->r_dir_caps, 0);
3705 if (dcaps) {
3706 dout("releasing r_dir_caps=%s\n", ceph_cap_string(dcaps));
3707 ceph_put_cap_refs(ceph_inode(req->r_parent), dcaps);
3708 }
3709}
3710
e64f44a8
XL		 3711void ceph_mdsc_release_dir_caps_no_check(struct ceph_mds_request *req)
3712{
3713 int dcaps;
3714
3715 dcaps = xchg(&req->r_dir_caps, 0);
3716 if (dcaps) {
3717 dout("releasing r_dir_caps=%s\n", ceph_cap_string(dcaps));
3718 ceph_put_cap_refs_no_check_caps(ceph_inode(req->r_parent),
3719 dcaps);
3720 }
3721}
3722
2f2dc053
SW		 3723/*
3724 * called under session->mutex.
3725 */
3726static void replay_unsafe_requests(struct ceph_mds_client *mdsc,
3727 struct ceph_mds_session *session)
3728{
3729 struct ceph_mds_request *req, *nreq;
3de22be6 3730 struct rb_node *p;
2f2dc053
SW		 3731
3732 dout("replay_unsafe_requests mds%d\n", session->s_mds);
3733
3734 mutex_lock(&mdsc->mutex);
9cf54563 3735 list_for_each_entry_safe(req, nreq, &session->s_unsafe, r_unsafe_item)
396bd62c 3736 __send_request(session, req, true);
3de22be6
YZ		 3737
3738 /*
3739 * also re-send old requests when MDS enters reconnect stage. So that MDS
3740 * can process completed request in clientreplay stage.
3741 */
3742 p = rb_first(&mdsc->request_tree);
3743 while (p) {
3744 req = rb_entry(p, struct ceph_mds_request, r_node);
3745 p = rb_next(p);
bc2de10d 3746 if (test_bit(CEPH_MDS_R_GOT_UNSAFE, &req->r_req_flags))
3de22be6
YZ		 3747 continue;
3748 if (req->r_attempts == 0)
3749 continue; /* only old requests */
a25949b9
JL		 3750 if (!req->r_session)
3751 continue;
3752 if (req->r_session->s_mds != session->s_mds)
3753 continue;
3754
e64f44a8 3755 ceph_mdsc_release_dir_caps_no_check(req);
a25949b9 3756
396bd62c 3757 __send_request(session, req, true);
3de22be6 3758 }
2f2dc053
SW		 3759 mutex_unlock(&mdsc->mutex);
3760}
3761
81c5a148
YZ		 3762static int send_reconnect_partial(struct ceph_reconnect_state *recon_state)
3763{
3764 struct ceph_msg *reply;
3765 struct ceph_pagelist *_pagelist;
3766 struct page *page;
3767 __le32 *addr;
3768 int err = -ENOMEM;
3769
3770 if (!recon_state->allow_multi)
3771 return -ENOSPC;
3772
3773 /* can't handle message that contains both caps and realm */
3774 BUG_ON(!recon_state->nr_caps == !recon_state->nr_realms);
3775
3776 /* pre-allocate new pagelist */
3777 _pagelist = ceph_pagelist_alloc(GFP_NOFS);
3778 if (!_pagelist)
3779 return -ENOMEM;
3780
3781 reply = ceph_msg_new2(CEPH_MSG_CLIENT_RECONNECT, 0, 1, GFP_NOFS, false);
3782 if (!reply)
3783 goto fail_msg;
3784
3785 /* placeholder for nr_caps */
3786 err = ceph_pagelist_encode_32(_pagelist, 0);
3787 if (err < 0)
3788 goto fail;
3789
3790 if (recon_state->nr_caps) {
3791 /* currently encoding caps */
3792 err = ceph_pagelist_encode_32(recon_state->pagelist, 0);
3793 if (err)
3794 goto fail;
3795 } else {
3796 /* placeholder for nr_realms (currently encoding relams) */
3797 err = ceph_pagelist_encode_32(_pagelist, 0);
3798 if (err < 0)
3799 goto fail;
3800 }
3801
3802 err = ceph_pagelist_encode_8(recon_state->pagelist, 1);
3803 if (err)
3804 goto fail;
3805
3806 page = list_first_entry(&recon_state->pagelist->head, struct page, lru);
3807 addr = kmap_atomic(page);
3808 if (recon_state->nr_caps) {
3809 /* currently encoding caps */
3810 *addr = cpu_to_le32(recon_state->nr_caps);
3811 } else {
3812 /* currently encoding relams */
3813 *(addr + 1) = cpu_to_le32(recon_state->nr_realms);
3814 }
3815 kunmap_atomic(addr);
3816
3817 reply->hdr.version = cpu_to_le16(5);
3818 reply->hdr.compat_version = cpu_to_le16(4);
3819
3820 reply->hdr.data_len = cpu_to_le32(recon_state->pagelist->length);
3821 ceph_msg_data_add_pagelist(reply, recon_state->pagelist);
3822
3823 ceph_con_send(&recon_state->session->s_con, reply);
3824 ceph_pagelist_release(recon_state->pagelist);
3825
3826 recon_state->pagelist = _pagelist;
3827 recon_state->nr_caps = 0;
3828 recon_state->nr_realms = 0;
3829 recon_state->msg_version = 5;
3830 return 0;
3831fail:
3832 ceph_msg_put(reply);
3833fail_msg:
3834 ceph_pagelist_release(_pagelist);
3835 return err;
3836}
3837
a33f6432
YZ		 3838static struct dentry* d_find_primary(struct inode *inode)
3839{
3840 struct dentry *alias, *dn = NULL;
3841
3842 if (hlist_empty(&inode->i_dentry))
3843 return NULL;
3844
3845 spin_lock(&inode->i_lock);
3846 if (hlist_empty(&inode->i_dentry))
3847 goto out_unlock;
3848
3849 if (S_ISDIR(inode->i_mode)) {
3850 alias = hlist_entry(inode->i_dentry.first, struct dentry, d_u.d_alias);
3851 if (!IS_ROOT(alias))
3852 dn = dget(alias);
3853 goto out_unlock;
3854 }
3855
3856 hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) {
3857 spin_lock(&alias->d_lock);
3858 if (!d_unhashed(alias) &&
3859 (ceph_dentry(alias)->flags & CEPH_DENTRY_PRIMARY_LINK)) {
3860 dn = dget_dlock(alias);
3861 }
3862 spin_unlock(&alias->d_lock);
3863 if (dn)
3864 break;
3865 }
3866out_unlock:
3867 spin_unlock(&inode->i_lock);
3868 return dn;
3869}
3870
2f2dc053
SW		 3871/*
3872 * Encode information about a cap for a reconnect with the MDS.
3873 */
a25949b9 3874static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap,
2f2dc053
SW		 3875 void *arg)
3876{
20cb34ae
SW		 3877 union {
3878 struct ceph_mds_cap_reconnect v2;
3879 struct ceph_mds_cap_reconnect_v1 v1;
3880 } rec;
b3f8d68f 3881 struct ceph_inode_info *ci = cap->ci;
20cb34ae
SW		 3882 struct ceph_reconnect_state *recon_state = arg;
3883 struct ceph_pagelist *pagelist = recon_state->pagelist;
a33f6432
YZ		 3884 struct dentry *dentry;
3885 char *path;
ee2a095d 3886 int pathlen = 0, err;
a33f6432 3887 u64 pathbase;
3469ed0d 3888 u64 snap_follows;
2f2dc053 3889
2f2dc053
SW		 3890 dout(" adding %p ino %llx.%llx cap %p %lld %s\n",
3891 inode, ceph_vinop(inode), cap, cap->cap_id,
3892 ceph_cap_string(cap->issued));
2f2dc053 3893
a33f6432
YZ		 3894 dentry = d_find_primary(inode);
3895 if (dentry) {
3896 /* set pathbase to parent dir when msg_version >= 2 */
3897 path = ceph_mdsc_build_path(dentry, &pathlen, &pathbase,
3898 recon_state->msg_version >= 2);
3899 dput(dentry);
3900 if (IS_ERR(path)) {
3901 err = PTR_ERR(path);
3902 goto out_err;
3903 }
3904 } else {
3905 path = NULL;
a33f6432
YZ		 3906 pathbase = 0;
3907 }
3908
be655596 3909 spin_lock(&ci->i_ceph_lock);
2f2dc053
SW		 3910 cap->seq = 0; /* reset cap seq */
3911 cap->issue_seq = 0; /* and issue_seq */
667ca05c 3912 cap->mseq = 0; /* and migrate_seq */
52d60f8e 3913 cap->cap_gen = atomic_read(&cap->session->s_cap_gen);
20cb34ae 3914
a25949b9 3915 /* These are lost when the session goes away */
785892fe
JL		 3916 if (S_ISDIR(inode->i_mode)) {
3917 if (cap->issued & CEPH_CAP_DIR_CREATE) {
3918 ceph_put_string(rcu_dereference_raw(ci->i_cached_layout.pool_ns));
3919 memset(&ci->i_cached_layout, 0, sizeof(ci->i_cached_layout));
3920 }
a25949b9 3921 cap->issued &= ~CEPH_CAP_ANY_DIR_OPS;
785892fe 3922 }
a25949b9 3923
121f22a1 3924 if (recon_state->msg_version >= 2) {
20cb34ae
SW		 3925 rec.v2.cap_id = cpu_to_le64(cap->cap_id);
3926 rec.v2.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
3927 rec.v2.issued = cpu_to_le32(cap->issued);
3928 rec.v2.snaprealm = cpu_to_le64(ci->i_snap_realm->ino);
a33f6432 3929 rec.v2.pathbase = cpu_to_le64(pathbase);
ec1dff25
JL		 3930 rec.v2.flock_len = (__force __le32)
3931 ((ci->i_ceph_flags & CEPH_I_ERROR_FILELOCK) ? 0 : 1);
20cb34ae
SW		 3932 } else {
3933 rec.v1.cap_id = cpu_to_le64(cap->cap_id);
3934 rec.v1.wanted = cpu_to_le32(__ceph_caps_wanted(ci));
3935 rec.v1.issued = cpu_to_le32(cap->issued);
2d6795fb 3936 rec.v1.size = cpu_to_le64(i_size_read(inode));
9bbeab41
AB		 3937 ceph_encode_timespec64(&rec.v1.mtime, &inode->i_mtime);
3938 ceph_encode_timespec64(&rec.v1.atime, &inode->i_atime);
20cb34ae 3939 rec.v1.snaprealm = cpu_to_le64(ci->i_snap_realm->ino);
a33f6432 3940 rec.v1.pathbase = cpu_to_le64(pathbase);
20cb34ae 3941 }
3469ed0d
YZ		 3942
3943 if (list_empty(&ci->i_cap_snaps)) {
92776fd2 3944 snap_follows = ci->i_head_snapc ? ci->i_head_snapc->seq : 0;
3469ed0d
YZ		 3945 } else {
3946 struct ceph_cap_snap *capsnap =
3947 list_first_entry(&ci->i_cap_snaps,
3948 struct ceph_cap_snap, ci_item);
3949 snap_follows = capsnap->follows;
20cb34ae 3950 }
be655596 3951 spin_unlock(&ci->i_ceph_lock);
2f2dc053 3952
121f22a1 3953 if (recon_state->msg_version >= 2) {
40819f6f 3954 int num_fcntl_locks, num_flock_locks;
4deb14a2 3955 struct ceph_filelock *flocks = NULL;
81c5a148 3956 size_t struct_len, total_len = sizeof(u64);
121f22a1 3957 u8 struct_v = 0;
39be95e9
JS		 3958
3959encode_again:
b3f8d68f
YZ		 3960 if (rec.v2.flock_len) {
3961 ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks);
3962 } else {
3963 num_fcntl_locks = 0;
3964 num_flock_locks = 0;
3965 }
4deb14a2 3966 if (num_fcntl_locks + num_flock_locks > 0) {
6da2ec56
KC		 3967 flocks = kmalloc_array(num_fcntl_locks + num_flock_locks,
3968 sizeof(struct ceph_filelock),
3969 GFP_NOFS);
4deb14a2
YZ		 3970 if (!flocks) {
3971 err = -ENOMEM;
5ccedf1c 3972 goto out_err;
4deb14a2
YZ		 3973 }
3974 err = ceph_encode_locks_to_buffer(inode, flocks,
3975 num_fcntl_locks,
3976 num_flock_locks);
3977 if (err) {
3978 kfree(flocks);
3979 flocks = NULL;
3980 if (err == -ENOSPC)
3981 goto encode_again;
5ccedf1c 3982 goto out_err;
4deb14a2
YZ		 3983 }
3984 } else {
39be95e9 3985 kfree(flocks);
4deb14a2 3986 flocks = NULL;
39be95e9 3987 }
121f22a1
YZ		 3988
3989 if (recon_state->msg_version >= 3) {
3990 /* version, compat_version and struct_len */
81c5a148 3991 total_len += 2 * sizeof(u8) + sizeof(u32);
3469ed0d 3992 struct_v = 2;
121f22a1 3993 }
39be95e9
JS		 3994 /*
3995 * number of encoded locks is stable, so copy to pagelist
3996 */
121f22a1
YZ		 3997 struct_len = 2 * sizeof(u32) +
3998 (num_fcntl_locks + num_flock_locks) *
3999 sizeof(struct ceph_filelock);
4000 rec.v2.flock_len = cpu_to_le32(struct_len);
4001
a33f6432 4002 struct_len += sizeof(u32) + pathlen + sizeof(rec.v2);
121f22a1 4003
3469ed0d
YZ		 4004 if (struct_v >= 2)
4005 struct_len += sizeof(u64); /* snap_follows */
4006
121f22a1 4007 total_len += struct_len;
81c5a148
YZ		 4008
4009 if (pagelist->length + total_len > RECONNECT_MAX_SIZE) {
4010 err = send_reconnect_partial(recon_state);
4011 if (err)
4012 goto out_freeflocks;
4013 pagelist = recon_state->pagelist;
5ccedf1c 4014 }
121f22a1 4015
81c5a148
YZ		 4016 err = ceph_pagelist_reserve(pagelist, total_len);
4017 if (err)
4018 goto out_freeflocks;
4019
4020 ceph_pagelist_encode_64(pagelist, ceph_ino(inode));
5ccedf1c
YZ		 4021 if (recon_state->msg_version >= 3) {
4022 ceph_pagelist_encode_8(pagelist, struct_v);
4023 ceph_pagelist_encode_8(pagelist, 1);
4024 ceph_pagelist_encode_32(pagelist, struct_len);
121f22a1 4025 }
a33f6432 4026 ceph_pagelist_encode_string(pagelist, path, pathlen);
5ccedf1c
YZ		 4027 ceph_pagelist_append(pagelist, &rec, sizeof(rec.v2));
4028 ceph_locks_to_pagelist(flocks, pagelist,
4029 num_fcntl_locks, num_flock_locks);
4030 if (struct_v >= 2)
4031 ceph_pagelist_encode_64(pagelist, snap_follows);
81c5a148 4032out_freeflocks:
39be95e9 4033 kfree(flocks);
3612abbd 4034 } else {
5ccedf1c 4035 err = ceph_pagelist_reserve(pagelist,
81c5a148
YZ		 4036 sizeof(u64) + sizeof(u32) +
4037 pathlen + sizeof(rec.v1));
a33f6432
YZ		 4038 if (err)
4039 goto out_err;
5ccedf1c 4040
81c5a148 4041 ceph_pagelist_encode_64(pagelist, ceph_ino(inode));
5ccedf1c
YZ		 4042 ceph_pagelist_encode_string(pagelist, path, pathlen);
4043 ceph_pagelist_append(pagelist, &rec, sizeof(rec.v1));
40819f6f 4044 }
44c99757 4045
5ccedf1c 4046out_err:
a33f6432
YZ		 4047 ceph_mdsc_free_path(path, pathlen);
4048 if (!err)
81c5a148
YZ		 4049 recon_state->nr_caps++;
4050 return err;
4051}
4052
4053static int encode_snap_realms(struct ceph_mds_client *mdsc,
4054 struct ceph_reconnect_state *recon_state)
4055{
4056 struct rb_node *p;
4057 struct ceph_pagelist *pagelist = recon_state->pagelist;
4058 int err = 0;
4059
4060 if (recon_state->msg_version >= 4) {
4061 err = ceph_pagelist_encode_32(pagelist, mdsc->num_snap_realms);
4062 if (err < 0)
4063 goto fail;
4064 }
4065
4066 /*
4067 * snaprealms. we provide mds with the ino, seq (version), and
4068 * parent for all of our realms. If the mds has any newer info,
4069 * it will tell us.
4070 */
4071 for (p = rb_first(&mdsc->snap_realms); p; p = rb_next(p)) {
4072 struct ceph_snap_realm *realm =
4073 rb_entry(p, struct ceph_snap_realm, node);
4074 struct ceph_mds_snaprealm_reconnect sr_rec;
4075
4076 if (recon_state->msg_version >= 4) {
4077 size_t need = sizeof(u8) * 2 + sizeof(u32) +
4078 sizeof(sr_rec);
4079
4080 if (pagelist->length + need > RECONNECT_MAX_SIZE) {
4081 err = send_reconnect_partial(recon_state);
4082 if (err)
4083 goto fail;
4084 pagelist = recon_state->pagelist;
4085 }
4086
4087 err = ceph_pagelist_reserve(pagelist, need);
4088 if (err)
4089 goto fail;
4090
4091 ceph_pagelist_encode_8(pagelist, 1);
4092 ceph_pagelist_encode_8(pagelist, 1);
4093 ceph_pagelist_encode_32(pagelist, sizeof(sr_rec));
4094 }
4095
4096 dout(" adding snap realm %llx seq %lld parent %llx\n",
4097 realm->ino, realm->seq, realm->parent_ino);
4098 sr_rec.ino = cpu_to_le64(realm->ino);
4099 sr_rec.seq = cpu_to_le64(realm->seq);
4100 sr_rec.parent = cpu_to_le64(realm->parent_ino);
4101
4102 err = ceph_pagelist_append(pagelist, &sr_rec, sizeof(sr_rec));
4103 if (err)
4104 goto fail;
4105
4106 recon_state->nr_realms++;
4107 }
4108fail:
93cea5be 4109 return err;
2f2dc053
SW		 4110}
4111
4112
4113/*
4114 * If an MDS fails and recovers, clients need to reconnect in order to
4115 * reestablish shared state. This includes all caps issued through
4116 * this session _and_ the snap_realm hierarchy. Because it's not
4117 * clear which snap realms the mds cares about, we send everything we
4118 * know about.. that ensures we'll then get any new info the
4119 * recovering MDS might have.
4120 *
4121 * This is a relatively heavyweight operation, but it's rare.
2f2dc053 4122 */
34b6c855
SW		 4123static void send_mds_reconnect(struct ceph_mds_client *mdsc,
4124 struct ceph_mds_session *session)
2f2dc053 4125{
2f2dc053 4126 struct ceph_msg *reply;
34b6c855 4127 int mds = session->s_mds;
9abf82b8 4128 int err = -ENOMEM;
81c5a148
YZ		 4129 struct ceph_reconnect_state recon_state = {
4130 .session = session,
4131 };
c8a96a31 4132 LIST_HEAD(dispose);
2f2dc053 4133
34b6c855 4134 pr_info("mds%d reconnect start\n", mds);
2f2dc053 4135
81c5a148
YZ		 4136 recon_state.pagelist = ceph_pagelist_alloc(GFP_NOFS);
4137 if (!recon_state.pagelist)
93cea5be 4138 goto fail_nopagelist;
93cea5be 4139
0d9c1ab3 4140 reply = ceph_msg_new2(CEPH_MSG_CLIENT_RECONNECT, 0, 1, GFP_NOFS, false);
a79832f2 4141 if (!reply)
93cea5be 4142 goto fail_nomsg;
93cea5be 4143
d4846487
JL		 4144 xa_destroy(&session->s_delegated_inos);
4145
34b6c855
SW		 4146 mutex_lock(&session->s_mutex);
4147 session->s_state = CEPH_MDS_SESSION_RECONNECTING;
4148 session->s_seq = 0;
2f2dc053 4149
2f2dc053 4150 dout("session %p state %s\n", session,
a687ecaf 4151 ceph_session_state_name(session->s_state));
2f2dc053 4152
52d60f8e 4153 atomic_inc(&session->s_cap_gen);
99a9c273
YZ		 4154
4155 spin_lock(&session->s_cap_lock);
03f4fcb0
YZ		 4156 /* don't know if session is readonly */
4157 session->s_readonly = 0;
99a9c273
YZ		 4158 /*
4159 * notify __ceph_remove_cap() that we are composing cap reconnect.
4160 * If a cap get released before being added to the cap reconnect,
4161 * __ceph_remove_cap() should skip queuing cap release.
4162 */
4163 session->s_cap_reconnect = 1;
e01a5946 4164 /* drop old cap expires; we're about to reestablish that state */
c8a96a31
JL		 4165 detach_cap_releases(session, &dispose);
4166 spin_unlock(&session->s_cap_lock);
4167 dispose_cap_releases(mdsc, &dispose);
e01a5946 4168
5d23371f 4169 /* trim unused caps to reduce MDS's cache rejoin time */
c0bd50e2
YZ		 4170 if (mdsc->fsc->sb->s_root)
4171 shrink_dcache_parent(mdsc->fsc->sb->s_root);
5d23371f
YZ		 4172
4173 ceph_con_close(&session->s_con);
4174 ceph_con_open(&session->s_con,
4175 CEPH_ENTITY_TYPE_MDS, mds,
4176 ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
4177
4178 /* replay unsafe requests */
4179 replay_unsafe_requests(mdsc, session);
4180
81c5a148
YZ		 4181 ceph_early_kick_flushing_caps(mdsc, session);
4182
5d23371f
YZ		 4183 down_read(&mdsc->snap_rwsem);
4184
81c5a148
YZ		 4185 /* placeholder for nr_caps */
4186 err = ceph_pagelist_encode_32(recon_state.pagelist, 0);
93cea5be
SW		 4187 if (err)
4188 goto fail;
20cb34ae 4189
81c5a148 4190 if (test_bit(CEPHFS_FEATURE_MULTI_RECONNECT, &session->s_features)) {
121f22a1 4191 recon_state.msg_version = 3;
81c5a148
YZ		 4192 recon_state.allow_multi = true;
4193 } else if (session->s_con.peer_features & CEPH_FEATURE_MDSENC) {
4194 recon_state.msg_version = 3;
4195 } else {
23c625ce 4196 recon_state.msg_version = 2;
81c5a148
YZ		 4197 }
4198 /* trsaverse this session's caps */
a25949b9 4199 err = ceph_iterate_session_caps(session, reconnect_caps_cb, &recon_state);
2f2dc053 4200
99a9c273
YZ		 4201 spin_lock(&session->s_cap_lock);
4202 session->s_cap_reconnect = 0;
4203 spin_unlock(&session->s_cap_lock);
4204
81c5a148
YZ		 4205 if (err < 0)
4206 goto fail;
2f2dc053 4207
81c5a148
YZ		 4208 /* check if all realms can be encoded into current message */
4209 if (mdsc->num_snap_realms) {
4210 size_t total_len =
4211 recon_state.pagelist->length +
4212 mdsc->num_snap_realms *
4213 sizeof(struct ceph_mds_snaprealm_reconnect);
4214 if (recon_state.msg_version >= 4) {
4215 /* number of realms */
4216 total_len += sizeof(u32);
4217 /* version, compat_version and struct_len */
4218 total_len += mdsc->num_snap_realms *
4219 (2 * sizeof(u8) + sizeof(u32));
4220 }
4221 if (total_len > RECONNECT_MAX_SIZE) {
4222 if (!recon_state.allow_multi) {
4223 err = -ENOSPC;
4224 goto fail;
4225 }
4226 if (recon_state.nr_caps) {
4227 err = send_reconnect_partial(&recon_state);
4228 if (err)
4229 goto fail;
4230 }
4231 recon_state.msg_version = 5;
4232 }
2f2dc053 4233 }
2f2dc053 4234
81c5a148
YZ		 4235 err = encode_snap_realms(mdsc, &recon_state);
4236 if (err < 0)
4237 goto fail;
4238
4239 if (recon_state.msg_version >= 5) {
4240 err = ceph_pagelist_encode_8(recon_state.pagelist, 0);
4241 if (err < 0)
4242 goto fail;
4243 }
44c99757 4244
81c5a148
YZ		 4245 if (recon_state.nr_caps || recon_state.nr_realms) {
4246 struct page *page =
4247 list_first_entry(&recon_state.pagelist->head,
4248 struct page, lru);
44c99757 4249 __le32 *addr = kmap_atomic(page);
81c5a148
YZ		 4250 if (recon_state.nr_caps) {
4251 WARN_ON(recon_state.nr_realms != mdsc->num_snap_realms);
4252 *addr = cpu_to_le32(recon_state.nr_caps);
4253 } else if (recon_state.msg_version >= 4) {
4254 *(addr + 1) = cpu_to_le32(recon_state.nr_realms);
4255 }
44c99757 4256 kunmap_atomic(addr);
ebf18f47 4257 }
44c99757 4258
81c5a148
YZ		 4259 reply->hdr.version = cpu_to_le16(recon_state.msg_version);
4260 if (recon_state.msg_version >= 4)
4261 reply->hdr.compat_version = cpu_to_le16(4);
e548e9b9 4262
81c5a148
YZ		 4263 reply->hdr.data_len = cpu_to_le32(recon_state.pagelist->length);
4264 ceph_msg_data_add_pagelist(reply, recon_state.pagelist);
e548e9b9 4265
2f2dc053
SW		 4266 ceph_con_send(&session->s_con, reply);
4267
9abf82b8
SW		 4268 mutex_unlock(&session->s_mutex);
4269
4270 mutex_lock(&mdsc->mutex);
4271 __wake_requests(mdsc, &session->s_waiting);
4272 mutex_unlock(&mdsc->mutex);
4273
2f2dc053 4274 up_read(&mdsc->snap_rwsem);
81c5a148 4275 ceph_pagelist_release(recon_state.pagelist);
2f2dc053
SW		 4276 return;
4277
93cea5be 4278fail:
2f2dc053 4279 ceph_msg_put(reply);
9abf82b8
SW		 4280 up_read(&mdsc->snap_rwsem);
4281 mutex_unlock(&session->s_mutex);
93cea5be 4282fail_nomsg:
81c5a148 4283 ceph_pagelist_release(recon_state.pagelist);
93cea5be 4284fail_nopagelist:
9abf82b8 4285 pr_err("error %d preparing reconnect for mds%d\n", err, mds);
9abf82b8 4286 return;
2f2dc053
SW		 4287}
4288
4289
4290/*
4291 * compare old and new mdsmaps, kicking requests
4292 * and closing out old connections as necessary
4293 *
4294 * called under mdsc->mutex.
4295 */
4296static void check_new_map(struct ceph_mds_client *mdsc,
4297 struct ceph_mdsmap *newmap,
4298 struct ceph_mdsmap *oldmap)
4299{
d517b398 4300 int i, j, err;
2f2dc053
SW		 4301 int oldstate, newstate;
4302 struct ceph_mds_session *s;
d517b398 4303 unsigned long targets[DIV_ROUND_UP(CEPH_MAX_MDS, sizeof(unsigned long))] = {0};
2f2dc053
SW		 4304
4305 dout("check_new_map new %u old %u\n",
4306 newmap->m_epoch, oldmap->m_epoch);
4307
d517b398
XL		 4308 if (newmap->m_info) {
4309 for (i = 0; i < newmap->possible_max_rank; i++) {
4310 for (j = 0; j < newmap->m_info[i].num_export_targets; j++)
4311 set_bit(newmap->m_info[i].export_targets[j], targets);
4312 }
4313 }
4314
b38c9eb4 4315 for (i = 0; i < oldmap->possible_max_rank && i < mdsc->max_sessions; i++) {
d37b1d99 4316 if (!mdsc->sessions[i])
2f2dc053
SW		 4317 continue;
4318 s = mdsc->sessions[i];
4319 oldstate = ceph_mdsmap_get_state(oldmap, i);
4320 newstate = ceph_mdsmap_get_state(newmap, i);
4321
0deb01c9 4322 dout("check_new_map mds%d state %s%s -> %s%s (session %s)\n",
2f2dc053 4323 i, ceph_mds_state_name(oldstate),
0deb01c9 4324 ceph_mdsmap_is_laggy(oldmap, i) ? " (laggy)" : "",
2f2dc053 4325 ceph_mds_state_name(newstate),
0deb01c9 4326 ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "",
a687ecaf 4327 ceph_session_state_name(s->s_state));
2f2dc053 4328
b38c9eb4 4329 if (i >= newmap->possible_max_rank) {
6f0f597b 4330 /* force close session for stopped mds */
5b3248c6 4331 ceph_get_mds_session(s);
6f0f597b
YZ		 4332 __unregister_session(mdsc, s);
4333 __wake_requests(mdsc, &s->s_waiting);
4334 mutex_unlock(&mdsc->mutex);
2827528d 4335
6f0f597b
YZ		 4336 mutex_lock(&s->s_mutex);
4337 cleanup_session_requests(mdsc, s);
4338 remove_session_caps(s);
4339 mutex_unlock(&s->s_mutex);
2827528d 4340
6f0f597b 4341 ceph_put_mds_session(s);
2827528d 4342
6f0f597b
YZ		 4343 mutex_lock(&mdsc->mutex);
4344 kick_requests(mdsc, i);
4345 continue;
4346 }
4347
4348 if (memcmp(ceph_mdsmap_get_addr(oldmap, i),
4349 ceph_mdsmap_get_addr(newmap, i),
4350 sizeof(struct ceph_entity_addr))) {
4351 /* just close it */
4352 mutex_unlock(&mdsc->mutex);
4353 mutex_lock(&s->s_mutex);
4354 mutex_lock(&mdsc->mutex);
4355 ceph_con_close(&s->s_con);
4356 mutex_unlock(&s->s_mutex);
4357 s->s_state = CEPH_MDS_SESSION_RESTARTING;
2f2dc053
SW		 4358 } else if (oldstate == newstate) {
4359 continue; /* nothing new with this mds */
4360 }
4361
4362 /*
4363 * send reconnect?
4364 */
4365 if (s->s_state == CEPH_MDS_SESSION_RESTARTING &&
34b6c855
SW		 4366 newstate >= CEPH_MDS_STATE_RECONNECT) {
4367 mutex_unlock(&mdsc->mutex);
d517b398 4368 clear_bit(i, targets);
34b6c855
SW		 4369 send_mds_reconnect(mdsc, s);
4370 mutex_lock(&mdsc->mutex);
4371 }
2f2dc053
SW		 4372
4373 /*
29790f26 4374 * kick request on any mds that has gone active.
2f2dc053
SW		 4375 */
4376 if (oldstate < CEPH_MDS_STATE_ACTIVE &&
4377 newstate >= CEPH_MDS_STATE_ACTIVE) {
29790f26
SW		 4378 if (oldstate != CEPH_MDS_STATE_CREATING &&
4379 oldstate != CEPH_MDS_STATE_STARTING)
4380 pr_info("mds%d recovery completed\n", s->s_mds);
4381 kick_requests(mdsc, i);
ea8412b2 4382 mutex_unlock(&mdsc->mutex);
829ad4db 4383 mutex_lock(&s->s_mutex);
ea8412b2 4384 mutex_lock(&mdsc->mutex);
2f2dc053 4385 ceph_kick_flushing_caps(mdsc, s);
829ad4db 4386 mutex_unlock(&s->s_mutex);
d2f8bb27 4387 wake_up_session_caps(s, RECONNECT);
2f2dc053
SW		 4388 }
4389 }
cb170a22 4390
d517b398
XL		 4391 /*
4392 * Only open and reconnect sessions that don't exist yet.
4393 */
4394 for (i = 0; i < newmap->possible_max_rank; i++) {
4395 /*
4396 * In case the import MDS is crashed just after
4397 * the EImportStart journal is flushed, so when
4398 * a standby MDS takes over it and is replaying
4399 * the EImportStart journal the new MDS daemon
4400 * will wait the client to reconnect it, but the
4401 * client may never register/open the session yet.
4402 *
4403 * Will try to reconnect that MDS daemon if the
4404 * rank number is in the export targets array and
4405 * is the up:reconnect state.
4406 */
4407 newstate = ceph_mdsmap_get_state(newmap, i);
4408 if (!test_bit(i, targets) || newstate != CEPH_MDS_STATE_RECONNECT)
4409 continue;
4410
4411 /*
4412 * The session maybe registered and opened by some
4413 * requests which were choosing random MDSes during
4414 * the mdsc->mutex's unlock/lock gap below in rare
4415 * case. But the related MDS daemon will just queue
4416 * that requests and be still waiting for the client's
4417 * reconnection request in up:reconnect state.
4418 */
4419 s = __ceph_lookup_mds_session(mdsc, i);
4420 if (likely(!s)) {
4421 s = __open_export_target_session(mdsc, i);
4422 if (IS_ERR(s)) {
4423 err = PTR_ERR(s);
4424 pr_err("failed to open export target session, err %d\n",
4425 err);
4426 continue;
4427 }
4428 }
4429 dout("send reconnect to export target mds.%d\n", i);
4430 mutex_unlock(&mdsc->mutex);
4431 send_mds_reconnect(mdsc, s);
4432 ceph_put_mds_session(s);
4433 mutex_lock(&mdsc->mutex);
4434 }
4435
b38c9eb4 4436 for (i = 0; i < newmap->possible_max_rank && i < mdsc->max_sessions; i++) {
cb170a22
SW		 4437 s = mdsc->sessions[i];
4438 if (!s)
4439 continue;
4440 if (!ceph_mdsmap_is_laggy(newmap, i))
4441 continue;
4442 if (s->s_state == CEPH_MDS_SESSION_OPEN ||
4443 s->s_state == CEPH_MDS_SESSION_HUNG ||
4444 s->s_state == CEPH_MDS_SESSION_CLOSING) {
4445 dout(" connecting to export targets of laggy mds%d\n",
4446 i);
4447 __open_export_target_sessions(mdsc, s);
4448 }
4449 }
2f2dc053
SW		 4450}
4451
4452
4453
4454/*
4455 * leases
4456 */
4457
4458/*
4459 * caller must hold session s_mutex, dentry->d_lock
4460 */
4461void __ceph_mdsc_drop_dentry_lease(struct dentry *dentry)
4462{
4463 struct ceph_dentry_info *di = ceph_dentry(dentry);
4464
4465 ceph_put_mds_session(di->lease_session);
4466 di->lease_session = NULL;
4467}
4468
2600d2dd
SW		 4469static void handle_lease(struct ceph_mds_client *mdsc,
4470 struct ceph_mds_session *session,
4471 struct ceph_msg *msg)
2f2dc053 4472{
3d14c5d2 4473 struct super_block *sb = mdsc->fsc->sb;
2f2dc053 4474 struct inode *inode;
2f2dc053
SW		 4475 struct dentry *parent, *dentry;
4476 struct ceph_dentry_info *di;
2600d2dd 4477 int mds = session->s_mds;
2f2dc053 4478 struct ceph_mds_lease *h = msg->front.iov_base;
1e5ea23d 4479 u32 seq;
2f2dc053 4480 struct ceph_vino vino;
2f2dc053
SW		 4481 struct qstr dname;
4482 int release = 0;
4483
2f2dc053
SW		 4484 dout("handle_lease from mds%d\n", mds);
4485
4486 /* decode */
4487 if (msg->front.iov_len < sizeof(*h) + sizeof(u32))
4488 goto bad;
4489 vino.ino = le64_to_cpu(h->ino);
4490 vino.snap = CEPH_NOSNAP;
1e5ea23d 4491 seq = le32_to_cpu(h->seq);
0fcf6c02
YZ		 4492 dname.len = get_unaligned_le32(h + 1);
4493 if (msg->front.iov_len < sizeof(*h) + sizeof(u32) + dname.len)
2f2dc053 4494 goto bad;
0fcf6c02 4495 dname.name = (void *)(h + 1) + sizeof(u32);
2f2dc053 4496
2f2dc053
SW		 4497 /* lookup inode */
4498 inode = ceph_find_inode(sb, vino);
2f90b852
SW		 4499 dout("handle_lease %s, ino %llx %p %.*s\n",
4500 ceph_lease_op_name(h->action), vino.ino, inode,
1e5ea23d 4501 dname.len, dname.name);
6cd3bcad
YZ		 4502
4503 mutex_lock(&session->s_mutex);
62575e27 4504 inc_session_sequence(session);
6cd3bcad 4505
d37b1d99 4506 if (!inode) {
2f2dc053
SW		 4507 dout("handle_lease no inode %llx\n", vino.ino);
4508 goto release;
4509 }
2f2dc053
SW		 4510
4511 /* dentry */
4512 parent = d_find_alias(inode);
4513 if (!parent) {
4514 dout("no parent dentry on inode %p\n", inode);
4515 WARN_ON(1);
4516 goto release; /* hrm... */
4517 }
8387ff25 4518 dname.hash = full_name_hash(parent, dname.name, dname.len);
2f2dc053
SW		 4519 dentry = d_lookup(parent, &dname);
4520 dput(parent);
4521 if (!dentry)
4522 goto release;
4523
4524 spin_lock(&dentry->d_lock);
4525 di = ceph_dentry(dentry);
4526 switch (h->action) {
4527 case CEPH_MDS_LEASE_REVOKE:
3d8eb7a9 4528 if (di->lease_session == session) {
1e5ea23d
SW		 4529 if (ceph_seq_cmp(di->lease_seq, seq) > 0)
4530 h->seq = cpu_to_le32(di->lease_seq);
2f2dc053
SW		 4531 __ceph_mdsc_drop_dentry_lease(dentry);
4532 }
4533 release = 1;
4534 break;
4535
4536 case CEPH_MDS_LEASE_RENEW:
3d8eb7a9 4537 if (di->lease_session == session &&
52d60f8e 4538 di->lease_gen == atomic_read(&session->s_cap_gen) &&
2f2dc053
SW		 4539 di->lease_renew_from &&
4540 di->lease_renew_after == 0) {
4541 unsigned long duration =
3563dbdd 4542 msecs_to_jiffies(le32_to_cpu(h->duration_ms));
2f2dc053 4543
1e5ea23d 4544 di->lease_seq = seq;
9b16f03c 4545 di->time = di->lease_renew_from + duration;
2f2dc053
SW		 4546 di->lease_renew_after = di->lease_renew_from +
4547 (duration >> 1);
4548 di->lease_renew_from = 0;
4549 }
4550 break;
4551 }
4552 spin_unlock(&dentry->d_lock);
4553 dput(dentry);
4554
4555 if (!release)
4556 goto out;
4557
4558release:
4559 /* let's just reuse the same message */
4560 h->action = CEPH_MDS_LEASE_REVOKE_ACK;
4561 ceph_msg_get(msg);
4562 ceph_con_send(&session->s_con, msg);
4563
4564out:
2f2dc053 4565 mutex_unlock(&session->s_mutex);
23c2c76e 4566 iput(inode);
2f2dc053
SW		 4567 return;
4568
4569bad:
4570 pr_err("corrupt lease message\n");
9ec7cab1 4571 ceph_msg_dump(msg);
2f2dc053
SW		 4572}
4573
4574void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
2f2dc053
SW		 4575 struct dentry *dentry, char action,
4576 u32 seq)
4577{
4578 struct ceph_msg *msg;
4579 struct ceph_mds_lease *lease;
8f2a98ef
YZ		 4580 struct inode *dir;
4581 int len = sizeof(*lease) + sizeof(u32) + NAME_MAX;
2f2dc053 4582
8f2a98ef
YZ		 4583 dout("lease_send_msg identry %p %s to mds%d\n",
4584 dentry, ceph_lease_op_name(action), session->s_mds);
2f2dc053 4585
b61c2763 4586 msg = ceph_msg_new(CEPH_MSG_CLIENT_LEASE, len, GFP_NOFS, false);
a79832f2 4587 if (!msg)
2f2dc053
SW		 4588 return;
4589 lease = msg->front.iov_base;
4590 lease->action = action;
2f2dc053 4591 lease->seq = cpu_to_le32(seq);
2f2dc053 4592
8f2a98ef
YZ		 4593 spin_lock(&dentry->d_lock);
4594 dir = d_inode(dentry->d_parent);
4595 lease->ino = cpu_to_le64(ceph_ino(dir));
4596 lease->first = lease->last = cpu_to_le64(ceph_snap(dir));
4597
4598 put_unaligned_le32(dentry->d_name.len, lease + 1);
4599 memcpy((void *)(lease + 1) + 4,
4600 dentry->d_name.name, dentry->d_name.len);
4601 spin_unlock(&dentry->d_lock);
2f2dc053
SW		 4602
4603 ceph_con_send(&session->s_con, msg);
4604}
4605
2f2dc053 4606/*
59b312f3 4607 * lock unlock the session, to wait ongoing session activities
2f2dc053 4608 */
59b312f3 4609static void lock_unlock_session(struct ceph_mds_session *s)
2f2dc053 4610{
59b312f3
XL		 4611 mutex_lock(&s->s_mutex);
4612 mutex_unlock(&s->s_mutex);
2f2dc053
SW		 4613}
4614
131d7eb4
YZ		 4615static void maybe_recover_session(struct ceph_mds_client *mdsc)
4616{
4617 struct ceph_fs_client *fsc = mdsc->fsc;
4618
4619 if (!ceph_test_mount_opt(fsc, CLEANRECOVER))
4620 return;
2f2dc053 4621
131d7eb4
YZ		 4622 if (READ_ONCE(fsc->mount_state) != CEPH_MOUNT_MOUNTED)
4623 return;
4624
0b98acd6 4625 if (!READ_ONCE(fsc->blocklisted))
131d7eb4
YZ		 4626 return;
4627
0b98acd6 4628 pr_info("auto reconnect after blocklisted\n");
131d7eb4
YZ		 4629 ceph_force_reconnect(fsc->sb);
4630}
2f2dc053 4631
3e699bd8
XL		 4632bool check_session_state(struct ceph_mds_session *s)
4633{
62575e27
JL		 4634 switch (s->s_state) {
4635 case CEPH_MDS_SESSION_OPEN:
4636 if (s->s_ttl && time_after(jiffies, s->s_ttl)) {
3e699bd8
XL		 4637 s->s_state = CEPH_MDS_SESSION_HUNG;
4638 pr_info("mds%d hung\n", s->s_mds);
4639 }
62575e27
JL		 4640 break;
4641 case CEPH_MDS_SESSION_CLOSING:
62575e27
JL		 4642 case CEPH_MDS_SESSION_NEW:
4643 case CEPH_MDS_SESSION_RESTARTING:
4644 case CEPH_MDS_SESSION_CLOSED:
4645 case CEPH_MDS_SESSION_REJECTED:
3e699bd8 4646 return false;
62575e27 4647 }
3e699bd8
XL		 4648
4649 return true;
4650}
4651
62575e27
JL		 4652/*
4653 * If the sequence is incremented while we're waiting on a REQUEST_CLOSE reply,
4654 * then we need to retransmit that request.
4655 */
4656void inc_session_sequence(struct ceph_mds_session *s)
4657{
4658 lockdep_assert_held(&s->s_mutex);
4659
4660 s->s_seq++;
4661
4662 if (s->s_state == CEPH_MDS_SESSION_CLOSING) {
4663 int ret;
4664
4665 dout("resending session close request for mds%d\n", s->s_mds);
4666 ret = request_close_session(s);
4667 if (ret < 0)
4668 pr_err("unable to close session to mds%d: %d\n",
4669 s->s_mds, ret);
4670 }
4671}
4672
2f2dc053 4673/*
bf2ba432
LH		 4674 * delayed work -- periodically trim expired leases, renew caps with mds. If
4675 * the @delay parameter is set to 0 or if it's more than 5 secs, the default
4676 * workqueue delay value of 5 secs will be used.
2f2dc053 4677 */
bf2ba432 4678static void schedule_delayed(struct ceph_mds_client *mdsc, unsigned long delay)
2f2dc053 4679{
bf2ba432
LH		 4680 unsigned long max_delay = HZ * 5;
4681
4682 /* 5 secs default delay */
4683 if (!delay || (delay > max_delay))
4684 delay = max_delay;
4685 schedule_delayed_work(&mdsc->delayed_work,
4686 round_jiffies_relative(delay));
2f2dc053
SW		 4687}
4688
4689static void delayed_work(struct work_struct *work)
4690{
2f2dc053
SW		 4691 struct ceph_mds_client *mdsc =
4692 container_of(work, struct ceph_mds_client, delayed_work.work);
bf2ba432 4693 unsigned long delay;
2f2dc053
SW		 4694 int renew_interval;
4695 int renew_caps;
bf2ba432 4696 int i;
2f2dc053
SW		 4697
4698 dout("mdsc delayed_work\n");
75c9627e 4699
fa996773
XL		 4700 if (mdsc->stopping)
4701 return;
4702
2f2dc053
SW		 4703 mutex_lock(&mdsc->mutex);
4704 renew_interval = mdsc->mdsmap->m_session_timeout >> 2;
4705 renew_caps = time_after_eq(jiffies, HZ*renew_interval +
4706 mdsc->last_renew_caps);
4707 if (renew_caps)
4708 mdsc->last_renew_caps = jiffies;
4709
4710 for (i = 0; i < mdsc->max_sessions; i++) {
4711 struct ceph_mds_session *s = __ceph_lookup_mds_session(mdsc, i);
d37b1d99 4712 if (!s)
2f2dc053 4713 continue;
3e699bd8
XL		 4714
4715 if (!check_session_state(s)) {
2f2dc053
SW		 4716 ceph_put_mds_session(s);
4717 continue;
4718 }
4719 mutex_unlock(&mdsc->mutex);
4720
4721 mutex_lock(&s->s_mutex);
4722 if (renew_caps)
4723 send_renew_caps(mdsc, s);
4724 else
4725 ceph_con_keepalive(&s->s_con);
aab53dd9
SW		 4726 if (s->s_state == CEPH_MDS_SESSION_OPEN ||
4727 s->s_state == CEPH_MDS_SESSION_HUNG)
3d7ded4d 4728 ceph_send_cap_releases(mdsc, s);
2f2dc053
SW		 4729 mutex_unlock(&s->s_mutex);
4730 ceph_put_mds_session(s);
4731
4732 mutex_lock(&mdsc->mutex);
4733 }
4734 mutex_unlock(&mdsc->mutex);
4735
bf2ba432 4736 delay = ceph_check_delayed_caps(mdsc);
37c4efc1
YZ		 4737
4738 ceph_queue_cap_reclaim_work(mdsc);
4739
4740 ceph_trim_snapid_map(mdsc);
4741
131d7eb4
YZ		 4742 maybe_recover_session(mdsc);
4743
bf2ba432 4744 schedule_delayed(mdsc, delay);
2f2dc053
SW		 4745}
4746
3d14c5d2 4747int ceph_mdsc_init(struct ceph_fs_client *fsc)
2f2dc053 4748
2f2dc053 4749{
3d14c5d2 4750 struct ceph_mds_client *mdsc;
f9009efa 4751 int err;
3d14c5d2
YS		 4752
4753 mdsc = kzalloc(sizeof(struct ceph_mds_client), GFP_NOFS);
4754 if (!mdsc)
4755 return -ENOMEM;
4756 mdsc->fsc = fsc;
2f2dc053
SW		 4757 mutex_init(&mdsc->mutex);
4758 mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
d37b1d99 4759 if (!mdsc->mdsmap) {
f9009efa
XL		 4760 err = -ENOMEM;
4761 goto err_mdsc;
fb3101b6 4762 }
2d06eeb8 4763
2f2dc053 4764 init_completion(&mdsc->safe_umount_waiters);
f3c60c59 4765 init_waitqueue_head(&mdsc->session_close_wq);
2f2dc053 4766 INIT_LIST_HEAD(&mdsc->waiting_for_map);
0c44a8e0
LH		 4767 mdsc->quotarealms_inodes = RB_ROOT;
4768 mutex_init(&mdsc->quotarealms_inodes_mutex);
2f2dc053 4769 init_rwsem(&mdsc->snap_rwsem);
a105f00c 4770 mdsc->snap_realms = RB_ROOT;
2f2dc053
SW		 4771 INIT_LIST_HEAD(&mdsc->snap_empty);
4772 spin_lock_init(&mdsc->snap_empty_lock);
44ca18f2 4773 mdsc->request_tree = RB_ROOT;
2f2dc053
SW		 4774 INIT_DELAYED_WORK(&mdsc->delayed_work, delayed_work);
4775 mdsc->last_renew_caps = jiffies;
4776 INIT_LIST_HEAD(&mdsc->cap_delay_list);
3a3430af 4777 INIT_LIST_HEAD(&mdsc->cap_wait_list);
2f2dc053
SW		 4778 spin_lock_init(&mdsc->cap_delay_lock);
4779 INIT_LIST_HEAD(&mdsc->snap_flush_list);
4780 spin_lock_init(&mdsc->snap_flush_lock);
553adfd9 4781 mdsc->last_cap_flush_tid = 1;
e4500b5e 4782 INIT_LIST_HEAD(&mdsc->cap_flush_list);
db354052 4783 INIT_LIST_HEAD(&mdsc->cap_dirty_migrating);
2f2dc053
SW		 4784 spin_lock_init(&mdsc->cap_dirty_lock);
4785 init_waitqueue_head(&mdsc->cap_flushing_wq);
37c4efc1 4786 INIT_WORK(&mdsc->cap_reclaim_work, ceph_cap_reclaim_work);
f9009efa
XL		 4787 err = ceph_metric_init(&mdsc->metric);
4788 if (err)
4789 goto err_mdsmap;
37c4efc1
YZ		 4790
4791 spin_lock_init(&mdsc->dentry_list_lock);
4792 INIT_LIST_HEAD(&mdsc->dentry_leases);
4793 INIT_LIST_HEAD(&mdsc->dentry_dir_leases);
2d06eeb8 4794
37151668 4795 ceph_caps_init(mdsc);
fe33032d 4796 ceph_adjust_caps_max_min(mdsc, fsc->mount_options);
37151668 4797
75c9627e
YZ		 4798 spin_lock_init(&mdsc->snapid_map_lock);
4799 mdsc->snapid_map_tree = RB_ROOT;
4800 INIT_LIST_HEAD(&mdsc->snapid_map_lru);
4801
10183a69
YZ		 4802 init_rwsem(&mdsc->pool_perm_rwsem);
4803 mdsc->pool_perm_tree = RB_ROOT;
4804
dfeb84d4
YZ		 4805 strscpy(mdsc->nodename, utsname()->nodename,
4806 sizeof(mdsc->nodename));
a7caa88f
XL		 4807
4808 fsc->mdsc = mdsc;
5f44f142 4809 return 0;
f9009efa
XL		 4810
4811err_mdsmap:
4812 kfree(mdsc->mdsmap);
4813err_mdsc:
4814 kfree(mdsc);
4815 return err;
2f2dc053
SW		 4816}
4817
4818/*
4819 * Wait for safe replies on open mds requests. If we time out, drop
4820 * all requests from the tree to avoid dangling dentry refs.
4821 */
4822static void wait_requests(struct ceph_mds_client *mdsc)
4823{
a319bf56 4824 struct ceph_options *opts = mdsc->fsc->client->options;
2f2dc053 4825 struct ceph_mds_request *req;
2f2dc053
SW		 4826
4827 mutex_lock(&mdsc->mutex);
44ca18f2 4828 if (__get_oldest_req(mdsc)) {
2f2dc053 4829 mutex_unlock(&mdsc->mutex);
44ca18f2 4830
2f2dc053
SW		 4831 dout("wait_requests waiting for requests\n");
4832 wait_for_completion_timeout(&mdsc->safe_umount_waiters,
a319bf56 4833 ceph_timeout_jiffies(opts->mount_timeout));
2f2dc053
SW		 4834
4835 /* tear down remaining requests */
44ca18f2
SW		 4836 mutex_lock(&mdsc->mutex);
4837 while ((req = __get_oldest_req(mdsc))) {
2f2dc053
SW		 4838 dout("wait_requests timed out on tid %llu\n",
4839 req->r_tid);
428138c9 4840 list_del_init(&req->r_wait);
44ca18f2 4841 __unregister_request(mdsc, req);
2f2dc053
SW		 4842 }
4843 }
4844 mutex_unlock(&mdsc->mutex);
4845 dout("wait_requests done\n");
4846}
4847
d095559c
XL		 4848void send_flush_mdlog(struct ceph_mds_session *s)
4849{
4850 struct ceph_msg *msg;
4851
4852 /*
4853 * Pre-luminous MDS crashes when it sees an unknown session request
4854 */
4855 if (!CEPH_HAVE_FEATURE(s->s_con.peer_features, SERVER_LUMINOUS))
4856 return;
4857
4858 mutex_lock(&s->s_mutex);
4859 dout("request mdlog flush to mds%d (%s)s seq %lld\n", s->s_mds,
4860 ceph_session_state_name(s->s_state), s->s_seq);
4861 msg = ceph_create_session_msg(CEPH_SESSION_REQUEST_FLUSH_MDLOG,
4862 s->s_seq);
4863 if (!msg) {
4864 pr_err("failed to request mdlog flush to mds%d (%s) seq %lld\n",
4865 s->s_mds, ceph_session_state_name(s->s_state), s->s_seq);
4866 } else {
4867 ceph_con_send(&s->s_con, msg);
4868 }
4869 mutex_unlock(&s->s_mutex);
4870}
4871
2f2dc053
SW		 4872/*
4873 * called before mount is ro, and before dentries are torn down.
4874 * (hmm, does this still race with new lookups?)
4875 */
4876void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
4877{
4878 dout("pre_umount\n");
4879 mdsc->stopping = 1;
4880
d095559c 4881 ceph_mdsc_iterate_sessions(mdsc, send_flush_mdlog, true);
59b312f3 4882 ceph_mdsc_iterate_sessions(mdsc, lock_unlock_session, false);
afcdaea3 4883 ceph_flush_dirty_caps(mdsc);
2f2dc053 4884 wait_requests(mdsc);
17c688c3
SW		 4885
4886 /*
4887 * wait for reply handlers to drop their request refs and
4888 * their inode/dcache refs
4889 */
4890 ceph_msgr_flush();
0c44a8e0
LH		 4891
4892 ceph_cleanup_quotarealms_inodes(mdsc);
2f2dc053
SW		 4893}
4894
4895/*
1b2ba3c5 4896 * flush the mdlog and wait for all write mds requests to flush.
2f2dc053 4897 */
1b2ba3c5
XL		 4898static void flush_mdlog_and_wait_mdsc_unsafe_requests(struct ceph_mds_client *mdsc,
4899 u64 want_tid)
2f2dc053 4900{
80fc7314 4901 struct ceph_mds_request *req = NULL, *nextreq;
1b2ba3c5 4902 struct ceph_mds_session *last_session = NULL;
44ca18f2 4903 struct rb_node *n;
2f2dc053
SW		 4904
4905 mutex_lock(&mdsc->mutex);
1b2ba3c5 4906 dout("%s want %lld\n", __func__, want_tid);
80fc7314 4907restart:
44ca18f2
SW		 4908 req = __get_oldest_req(mdsc);
4909 while (req && req->r_tid <= want_tid) {
80fc7314
SW		 4910 /* find next request */
4911 n = rb_next(&req->r_node);
4912 if (n)
4913 nextreq = rb_entry(n, struct ceph_mds_request, r_node);
4914 else
4915 nextreq = NULL;
e8a7b8b1
YZ		 4916 if (req->r_op != CEPH_MDS_OP_SETFILELOCK &&
4917 (req->r_op & CEPH_MDS_OP_WRITE)) {
1b2ba3c5
XL		 4918 struct ceph_mds_session *s = req->r_session;
4919
4920 if (!s) {
4921 req = nextreq;
4922 continue;
4923 }
4924
44ca18f2
SW		 4925 /* write op */
4926 ceph_mdsc_get_request(req);
80fc7314
SW		 4927 if (nextreq)
4928 ceph_mdsc_get_request(nextreq);
1b2ba3c5 4929 s = ceph_get_mds_session(s);
44ca18f2 4930 mutex_unlock(&mdsc->mutex);
1b2ba3c5
XL		 4931
4932 /* send flush mdlog request to MDS */
4933 if (last_session != s) {
4934 send_flush_mdlog(s);
4935 ceph_put_mds_session(last_session);
4936 last_session = s;
4937 } else {
4938 ceph_put_mds_session(s);
4939 }
4940 dout("%s wait on %llu (want %llu)\n", __func__,
44ca18f2
SW		 4941 req->r_tid, want_tid);
4942 wait_for_completion(&req->r_safe_completion);
1b2ba3c5 4943
44ca18f2 4944 mutex_lock(&mdsc->mutex);
44ca18f2 4945 ceph_mdsc_put_request(req);
80fc7314
SW		 4946 if (!nextreq)
4947 break; /* next dne before, so we're done! */
4948 if (RB_EMPTY_NODE(&nextreq->r_node)) {
4949 /* next request was removed from tree */
4950 ceph_mdsc_put_request(nextreq);
4951 goto restart;
4952 }
4953 ceph_mdsc_put_request(nextreq); /* won't go away */
44ca18f2 4954 }
80fc7314 4955 req = nextreq;
2f2dc053
SW		 4956 }
4957 mutex_unlock(&mdsc->mutex);
1b2ba3c5
XL		 4958 ceph_put_mds_session(last_session);
4959 dout("%s done\n", __func__);
2f2dc053
SW		 4960}
4961
4962void ceph_mdsc_sync(struct ceph_mds_client *mdsc)
4963{
0e294387 4964 u64 want_tid, want_flush;
2f2dc053 4965
50c9132d 4966 if (READ_ONCE(mdsc->fsc->mount_state) >= CEPH_MOUNT_SHUTDOWN)
56b7cf95
SW		 4967 return;
4968
2f2dc053
SW		 4969 dout("sync\n");
4970 mutex_lock(&mdsc->mutex);
4971 want_tid = mdsc->last_tid;
2f2dc053 4972 mutex_unlock(&mdsc->mutex);
2f2dc053 4973
afcdaea3 4974 ceph_flush_dirty_caps(mdsc);
d3383a8e 4975 spin_lock(&mdsc->cap_dirty_lock);
8310b089 4976 want_flush = mdsc->last_cap_flush_tid;
c8799fc4
YZ		 4977 if (!list_empty(&mdsc->cap_flush_list)) {
4978 struct ceph_cap_flush *cf =
4979 list_last_entry(&mdsc->cap_flush_list,
4980 struct ceph_cap_flush, g_list);
4981 cf->wake = true;
4982 }
d3383a8e
YZ		 4983 spin_unlock(&mdsc->cap_dirty_lock);
4984
0e294387
YZ		 4985 dout("sync want tid %lld flush_seq %lld\n",
4986 want_tid, want_flush);
2f2dc053 4987
1b2ba3c5 4988 flush_mdlog_and_wait_mdsc_unsafe_requests(mdsc, want_tid);
0e294387 4989 wait_caps_flush(mdsc, want_flush);
2f2dc053
SW		 4990}
4991
f3c60c59
SW		 4992/*
4993 * true if all sessions are closed, or we force unmount
4994 */
fcff415c 4995static bool done_closing_sessions(struct ceph_mds_client *mdsc, int skipped)
f3c60c59 4996{
52953d55 4997 if (READ_ONCE(mdsc->fsc->mount_state) == CEPH_MOUNT_SHUTDOWN)
f3c60c59 4998 return true;
fcff415c 4999 return atomic_read(&mdsc->num_sessions) <= skipped;
f3c60c59 5000}
2f2dc053
SW		 5001
5002/*
5003 * called after sb is ro.
5004 */
5005void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
5006{
a319bf56 5007 struct ceph_options *opts = mdsc->fsc->client->options;
2f2dc053
SW		 5008 struct ceph_mds_session *session;
5009 int i;
fcff415c 5010 int skipped = 0;
2f2dc053
SW		 5011
5012 dout("close_sessions\n");
5013
2f2dc053 5014 /* close sessions */
f3c60c59
SW		 5015 mutex_lock(&mdsc->mutex);
5016 for (i = 0; i < mdsc->max_sessions; i++) {
5017 session = __ceph_lookup_mds_session(mdsc, i);
5018 if (!session)
5019 continue;
2f2dc053 5020 mutex_unlock(&mdsc->mutex);
f3c60c59 5021 mutex_lock(&session->s_mutex);
fcff415c
YZ		 5022 if (__close_session(mdsc, session) <= 0)
5023 skipped++;
f3c60c59
SW		 5024 mutex_unlock(&session->s_mutex);
5025 ceph_put_mds_session(session);
2f2dc053
SW		 5026 mutex_lock(&mdsc->mutex);
5027 }
f3c60c59
SW		 5028 mutex_unlock(&mdsc->mutex);
5029
5030 dout("waiting for sessions to close\n");
fcff415c
YZ		 5031 wait_event_timeout(mdsc->session_close_wq,
5032 done_closing_sessions(mdsc, skipped),
a319bf56 5033 ceph_timeout_jiffies(opts->mount_timeout));
2f2dc053
SW		 5034
5035 /* tear down remaining sessions */
f3c60c59 5036 mutex_lock(&mdsc->mutex);
2f2dc053
SW		 5037 for (i = 0; i < mdsc->max_sessions; i++) {
5038 if (mdsc->sessions[i]) {
5b3248c6 5039 session = ceph_get_mds_session(mdsc->sessions[i]);
2600d2dd 5040 __unregister_session(mdsc, session);
2f2dc053
SW		 5041 mutex_unlock(&mdsc->mutex);
5042 mutex_lock(&session->s_mutex);
5043 remove_session_caps(session);
5044 mutex_unlock(&session->s_mutex);
5045 ceph_put_mds_session(session);
5046 mutex_lock(&mdsc->mutex);
5047 }
5048 }
2f2dc053 5049 WARN_ON(!list_empty(&mdsc->cap_delay_list));
2f2dc053
SW		 5050 mutex_unlock(&mdsc->mutex);
5051
75c9627e 5052 ceph_cleanup_snapid_map(mdsc);
5ed91587 5053 ceph_cleanup_global_and_empty_realms(mdsc);
2f2dc053 5054
37c4efc1 5055 cancel_work_sync(&mdsc->cap_reclaim_work);
2f2dc053
SW		 5056 cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */
5057
5058 dout("stopped\n");
5059}
5060
48fec5d0
YZ		 5061void ceph_mdsc_force_umount(struct ceph_mds_client *mdsc)
5062{
5063 struct ceph_mds_session *session;
5064 int mds;
5065
5066 dout("force umount\n");
5067
5068 mutex_lock(&mdsc->mutex);
5069 for (mds = 0; mds < mdsc->max_sessions; mds++) {
5070 session = __ceph_lookup_mds_session(mdsc, mds);
5071 if (!session)
5072 continue;
d468e729
YZ		 5073
5074 if (session->s_state == CEPH_MDS_SESSION_REJECTED)
5075 __unregister_session(mdsc, session);
5076 __wake_requests(mdsc, &session->s_waiting);
48fec5d0 5077 mutex_unlock(&mdsc->mutex);
d468e729 5078
48fec5d0
YZ		 5079 mutex_lock(&session->s_mutex);
5080 __close_session(mdsc, session);
5081 if (session->s_state == CEPH_MDS_SESSION_CLOSING) {
5082 cleanup_session_requests(mdsc, session);
5083 remove_session_caps(session);
5084 }
5085 mutex_unlock(&session->s_mutex);
5086 ceph_put_mds_session(session);
d468e729 5087
48fec5d0
YZ		 5088 mutex_lock(&mdsc->mutex);
5089 kick_requests(mdsc, mds);
5090 }
5091 __wake_requests(mdsc, &mdsc->waiting_for_map);
5092 mutex_unlock(&mdsc->mutex);
5093}
5094
3d14c5d2 5095static void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
2f2dc053
SW		 5096{
5097 dout("stop\n");
fa996773
XL		 5098 /*
5099 * Make sure the delayed work stopped before releasing
5100 * the resources.
5101 *
5102 * Because the cancel_delayed_work_sync() will only
5103 * guarantee that the work finishes executing. But the
5104 * delayed work will re-arm itself again after that.
5105 */
5106 flush_delayed_work(&mdsc->delayed_work);
5107
2f2dc053
SW		 5108 if (mdsc->mdsmap)
5109 ceph_mdsmap_destroy(mdsc->mdsmap);
5110 kfree(mdsc->sessions);
37151668 5111 ceph_caps_finalize(mdsc);
10183a69 5112 ceph_pool_perm_destroy(mdsc);
2f2dc053
SW		 5113}
5114
3d14c5d2
YS		 5115void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
5116{
5117 struct ceph_mds_client *mdsc = fsc->mdsc;
ef550f6f 5118 dout("mdsc_destroy %p\n", mdsc);
ef550f6f 5119
50c55aec
CX		 5120 if (!mdsc)
5121 return;
5122
ef550f6f
SW		 5123 /* flush out any connection work with references to us */
5124 ceph_msgr_flush();
5125
62a65f36
YZ		 5126 ceph_mdsc_stop(mdsc);
5127
f9009efa
XL		 5128 ceph_metric_destroy(&mdsc->metric);
5129
3d14c5d2
YS		 5130 fsc->mdsc = NULL;
5131 kfree(mdsc);
ef550f6f 5132 dout("mdsc_destroy %p done\n", mdsc);
3d14c5d2
YS		 5133}
5134
430afbad
YZ		 5135void ceph_mdsc_handle_fsmap(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
5136{
5137 struct ceph_fs_client *fsc = mdsc->fsc;
5138 const char *mds_namespace = fsc->mount_options->mds_namespace;
5139 void *p = msg->front.iov_base;
5140 void *end = p + msg->front.iov_len;
5141 u32 epoch;
430afbad
YZ		 5142 u32 num_fs;
5143 u32 mount_fscid = (u32)-1;
430afbad
YZ		 5144 int err = -EINVAL;
5145
5146 ceph_decode_need(&p, end, sizeof(u32), bad);
5147 epoch = ceph_decode_32(&p);
5148
5149 dout("handle_fsmap epoch %u\n", epoch);
5150
06a1ad43
JL		 5151 /* struct_v, struct_cv, map_len, epoch, legacy_client_fscid */
5152 ceph_decode_skip_n(&p, end, 2 + sizeof(u32) * 3, bad);
430afbad 5153
06a1ad43 5154 ceph_decode_32_safe(&p, end, num_fs, bad);
430afbad
YZ		 5155 while (num_fs-- > 0) {
5156 void *info_p, *info_end;
5157 u32 info_len;
430afbad
YZ		 5158 u32 fscid, namelen;
5159
5160 ceph_decode_need(&p, end, 2 + sizeof(u32), bad);
06a1ad43 5161 p += 2; // info_v, info_cv
430afbad
YZ		 5162 info_len = ceph_decode_32(&p);
5163 ceph_decode_need(&p, end, info_len, bad);
5164 info_p = p;
5165 info_end = p + info_len;
5166 p = info_end;
5167
5168 ceph_decode_need(&info_p, info_end, sizeof(u32) * 2, bad);
5169 fscid = ceph_decode_32(&info_p);
5170 namelen = ceph_decode_32(&info_p);
5171 ceph_decode_need(&info_p, info_end, namelen, bad);
5172
5173 if (mds_namespace &&
5174 strlen(mds_namespace) == namelen &&
5175 !strncmp(mds_namespace, (char *)info_p, namelen)) {
5176 mount_fscid = fscid;
5177 break;
5178 }
5179 }
5180
5181 ceph_monc_got_map(&fsc->client->monc, CEPH_SUB_FSMAP, epoch);
5182 if (mount_fscid != (u32)-1) {
5183 fsc->client->monc.fs_cluster_id = mount_fscid;
5184 ceph_monc_want_map(&fsc->client->monc, CEPH_SUB_MDSMAP,
5185 0, true);
5186 ceph_monc_renew_subs(&fsc->client->monc);
5187 } else {
5188 err = -ENOENT;
5189 goto err_out;
5190 }
5191 return;
76bd6ec4 5192
430afbad 5193bad:
631ed4b0
JL		 5194 pr_err("error decoding fsmap %d. Shutting down mount.\n", err);
5195 ceph_umount_begin(mdsc->fsc->sb);
430afbad
YZ		 5196err_out:
5197 mutex_lock(&mdsc->mutex);
76bd6ec4 5198 mdsc->mdsmap_err = err;
430afbad
YZ		 5199 __wake_requests(mdsc, &mdsc->waiting_for_map);
5200 mutex_unlock(&mdsc->mutex);
430afbad 5201}
2f2dc053
SW		 5202
5203/*
5204 * handle mds map update.
5205 */
430afbad 5206void ceph_mdsc_handle_mdsmap(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
2f2dc053
SW		 5207{
5208 u32 epoch;
5209 u32 maplen;
5210 void *p = msg->front.iov_base;
5211 void *end = p + msg->front.iov_len;
5212 struct ceph_mdsmap *newmap, *oldmap;
5213 struct ceph_fsid fsid;
5214 int err = -EINVAL;
5215
5216 ceph_decode_need(&p, end, sizeof(fsid)+2*sizeof(u32), bad);
5217 ceph_decode_copy(&p, &fsid, sizeof(fsid));
3d14c5d2 5218 if (ceph_check_fsid(mdsc->fsc->client, &fsid) < 0)
0743304d 5219 return;
c89136ea
SW		 5220 epoch = ceph_decode_32(&p);
5221 maplen = ceph_decode_32(&p);
2f2dc053
SW		 5222 dout("handle_map epoch %u len %d\n", epoch, (int)maplen);
5223
5224 /* do we need it? */
2f2dc053
SW		 5225 mutex_lock(&mdsc->mutex);
5226 if (mdsc->mdsmap && epoch <= mdsc->mdsmap->m_epoch) {
5227 dout("handle_map epoch %u <= our %u\n",
5228 epoch, mdsc->mdsmap->m_epoch);
5229 mutex_unlock(&mdsc->mutex);
5230 return;
5231 }
5232
cd1a677c 5233 newmap = ceph_mdsmap_decode(&p, end, ceph_msgr2(mdsc->fsc->client));
2f2dc053
SW		 5234 if (IS_ERR(newmap)) {
5235 err = PTR_ERR(newmap);
5236 goto bad_unlock;
5237 }
5238
5239 /* swap into place */
5240 if (mdsc->mdsmap) {
5241 oldmap = mdsc->mdsmap;
5242 mdsc->mdsmap = newmap;
5243 check_new_map(mdsc, newmap, oldmap);
5244 ceph_mdsmap_destroy(oldmap);
5245 } else {
5246 mdsc->mdsmap = newmap; /* first mds map */
5247 }
719784ba
CX		 5248 mdsc->fsc->max_file_size = min((loff_t)mdsc->mdsmap->m_max_file_size,
5249 MAX_LFS_FILESIZE);
2f2dc053
SW		 5250
5251 __wake_requests(mdsc, &mdsc->waiting_for_map);
82dcabad
ID		 5252 ceph_monc_got_map(&mdsc->fsc->client->monc, CEPH_SUB_MDSMAP,
5253 mdsc->mdsmap->m_epoch);
2f2dc053
SW		 5254
5255 mutex_unlock(&mdsc->mutex);
bf2ba432 5256 schedule_delayed(mdsc, 0);
2f2dc053
SW		 5257 return;
5258
5259bad_unlock:
5260 mutex_unlock(&mdsc->mutex);
5261bad:
631ed4b0
JL		 5262 pr_err("error decoding mdsmap %d. Shutting down mount.\n", err);
5263 ceph_umount_begin(mdsc->fsc->sb);
2f2dc053
SW		 5264 return;
5265}
5266
4972cf60 5267static struct ceph_connection *mds_get_con(struct ceph_connection *con)
2f2dc053
SW		 5268{
5269 struct ceph_mds_session *s = con->private;
5270
5b3248c6 5271 if (ceph_get_mds_session(s))
2f2dc053 5272 return con;
2f2dc053
SW		 5273 return NULL;
5274}
5275
4972cf60 5276static void mds_put_con(struct ceph_connection *con)
2f2dc053
SW		 5277{
5278 struct ceph_mds_session *s = con->private;
5279
2f2dc053
SW		 5280 ceph_put_mds_session(s);
5281}
5282
5283/*
5284 * if the client is unresponsive for long enough, the mds will kill
5285 * the session entirely.
5286 */
4972cf60 5287static void mds_peer_reset(struct ceph_connection *con)
2f2dc053
SW		 5288{
5289 struct ceph_mds_session *s = con->private;
7e70f0ed 5290 struct ceph_mds_client *mdsc = s->s_mdsc;
2f2dc053 5291
f3ae1b97 5292 pr_warn("mds%d closed our session\n", s->s_mds);
7e70f0ed 5293 send_mds_reconnect(mdsc, s);
2f2dc053
SW		 5294}
5295
4972cf60 5296static void mds_dispatch(struct ceph_connection *con, struct ceph_msg *msg)
2f2dc053
SW		 5297{
5298 struct ceph_mds_session *s = con->private;
5299 struct ceph_mds_client *mdsc = s->s_mdsc;
5300 int type = le16_to_cpu(msg->hdr.type);
5301
2600d2dd
SW		 5302 mutex_lock(&mdsc->mutex);
5303 if (__verify_registered_session(mdsc, s) < 0) {
5304 mutex_unlock(&mdsc->mutex);
5305 goto out;
5306 }
5307 mutex_unlock(&mdsc->mutex);
5308
2f2dc053
SW		 5309 switch (type) {
5310 case CEPH_MSG_MDS_MAP:
430afbad
YZ		 5311 ceph_mdsc_handle_mdsmap(mdsc, msg);
5312 break;
5313 case CEPH_MSG_FS_MAP_USER:
5314 ceph_mdsc_handle_fsmap(mdsc, msg);
2f2dc053
SW		 5315 break;
5316 case CEPH_MSG_CLIENT_SESSION:
5317 handle_session(s, msg);
5318 break;
5319 case CEPH_MSG_CLIENT_REPLY:
5320 handle_reply(s, msg);
5321 break;
5322 case CEPH_MSG_CLIENT_REQUEST_FORWARD:
2600d2dd 5323 handle_forward(mdsc, s, msg);
2f2dc053
SW		 5324 break;
5325 case CEPH_MSG_CLIENT_CAPS:
5326 ceph_handle_caps(s, msg);
5327 break;
5328 case CEPH_MSG_CLIENT_SNAP:
2600d2dd 5329 ceph_handle_snap(mdsc, s, msg);
2f2dc053
SW		 5330 break;
5331 case CEPH_MSG_CLIENT_LEASE:
2600d2dd 5332 handle_lease(mdsc, s, msg);
2f2dc053 5333 break;
fb18a575
LH		 5334 case CEPH_MSG_CLIENT_QUOTA:
5335 ceph_handle_quota(mdsc, s, msg);
5336 break;
2f2dc053
SW		 5337
5338 default:
5339 pr_err("received unknown message type %d %s\n", type,
5340 ceph_msg_type_name(type));
5341 }
2600d2dd 5342out:
2f2dc053
SW		 5343 ceph_msg_put(msg);
5344}
5345
4e7a5dcd
SW		 5346/*
5347 * authentication
5348 */
a3530df3
AE		 5349
5350/*
5351 * Note: returned pointer is the address of a structure that's
5352 * managed separately. Caller must *not* attempt to free it.
5353 */
4972cf60
ID		 5354static struct ceph_auth_handshake *
5355mds_get_authorizer(struct ceph_connection *con, int *proto, int force_new)
4e7a5dcd
SW		 5356{
5357 struct ceph_mds_session *s = con->private;
5358 struct ceph_mds_client *mdsc = s->s_mdsc;
3d14c5d2 5359 struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
74f1869f 5360 struct ceph_auth_handshake *auth = &s->s_auth;
ce287162 5361 int ret;
4e7a5dcd 5362
ce287162
ID		 5363 ret = __ceph_auth_get_authorizer(ac, auth, CEPH_ENTITY_TYPE_MDS,
5364 force_new, proto, NULL, NULL);
5365 if (ret)
5366 return ERR_PTR(ret);
74f1869f 5367
a3530df3 5368 return auth;
4e7a5dcd
SW		 5369}
5370
4972cf60 5371static int mds_add_authorizer_challenge(struct ceph_connection *con,
6daca13d
ID		 5372 void *challenge_buf, int challenge_buf_len)
5373{
5374 struct ceph_mds_session *s = con->private;
5375 struct ceph_mds_client *mdsc = s->s_mdsc;
5376 struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
5377
5378 return ceph_auth_add_authorizer_challenge(ac, s->s_auth.authorizer,
5379 challenge_buf, challenge_buf_len);
5380}
4e7a5dcd 5381
4972cf60 5382static int mds_verify_authorizer_reply(struct ceph_connection *con)
4e7a5dcd
SW		 5383{
5384 struct ceph_mds_session *s = con->private;
5385 struct ceph_mds_client *mdsc = s->s_mdsc;
3d14c5d2 5386 struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
285ea34f 5387 struct ceph_auth_handshake *auth = &s->s_auth;
4e7a5dcd 5388
285ea34f
ID		 5389 return ceph_auth_verify_authorizer_reply(ac, auth->authorizer,
5390 auth->authorizer_reply_buf, auth->authorizer_reply_buf_len,
5391 NULL, NULL, NULL, NULL);
4e7a5dcd
SW		 5392}
5393
4972cf60 5394static int mds_invalidate_authorizer(struct ceph_connection *con)
9bd2e6f8
SW		 5395{
5396 struct ceph_mds_session *s = con->private;
5397 struct ceph_mds_client *mdsc = s->s_mdsc;
3d14c5d2 5398 struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
9bd2e6f8 5399
27859f97 5400 ceph_auth_invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);
9bd2e6f8 5401
3d14c5d2 5402 return ceph_monc_validate_auth(&mdsc->fsc->client->monc);
9bd2e6f8
SW		 5403}
5404
cd1a677c
ID		 5405static int mds_get_auth_request(struct ceph_connection *con,
5406 void *buf, int *buf_len,
5407 void **authorizer, int *authorizer_len)
5408{
5409 struct ceph_mds_session *s = con->private;
5410 struct ceph_auth_client *ac = s->s_mdsc->fsc->client->monc.auth;
5411 struct ceph_auth_handshake *auth = &s->s_auth;
5412 int ret;
5413
5414 ret = ceph_auth_get_authorizer(ac, auth, CEPH_ENTITY_TYPE_MDS,
5415 buf, buf_len);
5416 if (ret)
5417 return ret;
5418
5419 *authorizer = auth->authorizer_buf;
5420 *authorizer_len = auth->authorizer_buf_len;
5421 return 0;
5422}
5423
5424static int mds_handle_auth_reply_more(struct ceph_connection *con,
5425 void *reply, int reply_len,
5426 void *buf, int *buf_len,
5427 void **authorizer, int *authorizer_len)
5428{
5429 struct ceph_mds_session *s = con->private;
5430 struct ceph_auth_client *ac = s->s_mdsc->fsc->client->monc.auth;
5431 struct ceph_auth_handshake *auth = &s->s_auth;
5432 int ret;
5433
5434 ret = ceph_auth_handle_svc_reply_more(ac, auth, reply, reply_len,
5435 buf, buf_len);
5436 if (ret)
5437 return ret;
5438
5439 *authorizer = auth->authorizer_buf;
5440 *authorizer_len = auth->authorizer_buf_len;
5441 return 0;
5442}
5443
5444static int mds_handle_auth_done(struct ceph_connection *con,
5445 u64 global_id, void *reply, int reply_len,
5446 u8 *session_key, int *session_key_len,
5447 u8 *con_secret, int *con_secret_len)
5448{
5449 struct ceph_mds_session *s = con->private;
5450 struct ceph_auth_client *ac = s->s_mdsc->fsc->client->monc.auth;
5451 struct ceph_auth_handshake *auth = &s->s_auth;
5452
5453 return ceph_auth_handle_svc_reply_done(ac, auth, reply, reply_len,
5454 session_key, session_key_len,
5455 con_secret, con_secret_len);
5456}
5457
5458static int mds_handle_auth_bad_method(struct ceph_connection *con,
5459 int used_proto, int result,
5460 const int *allowed_protos, int proto_cnt,
5461 const int *allowed_modes, int mode_cnt)
5462{
5463 struct ceph_mds_session *s = con->private;
5464 struct ceph_mon_client *monc = &s->s_mdsc->fsc->client->monc;
5465 int ret;
5466
5467 if (ceph_auth_handle_bad_authorizer(monc->auth, CEPH_ENTITY_TYPE_MDS,
5468 used_proto, result,
5469 allowed_protos, proto_cnt,
5470 allowed_modes, mode_cnt)) {
5471 ret = ceph_monc_validate_auth(monc);
5472 if (ret)
5473 return ret;
5474 }
5475
5476 return -EACCES;
5477}
5478
53ded495
AE		 5479static struct ceph_msg *mds_alloc_msg(struct ceph_connection *con,
5480 struct ceph_msg_header *hdr, int *skip)
5481{
5482 struct ceph_msg *msg;
5483 int type = (int) le16_to_cpu(hdr->type);
5484 int front_len = (int) le32_to_cpu(hdr->front_len);
5485
5486 if (con->in_msg)
5487 return con->in_msg;
5488
5489 *skip = 0;
5490 msg = ceph_msg_new(type, front_len, GFP_NOFS, false);
5491 if (!msg) {
5492 pr_err("unable to allocate msg type %d len %d\n",
5493 type, front_len);
5494 return NULL;
5495 }
53ded495
AE		 5496
5497 return msg;
5498}
5499
79dbd1ba 5500static int mds_sign_message(struct ceph_msg *msg)
33d07337 5501{
79dbd1ba 5502 struct ceph_mds_session *s = msg->con->private;
33d07337 5503 struct ceph_auth_handshake *auth = &s->s_auth;
79dbd1ba 5504
33d07337
YZ		 5505 return ceph_auth_sign_message(auth, msg);
5506}
5507
79dbd1ba 5508static int mds_check_message_signature(struct ceph_msg *msg)
33d07337 5509{
79dbd1ba 5510 struct ceph_mds_session *s = msg->con->private;
33d07337 5511 struct ceph_auth_handshake *auth = &s->s_auth;
79dbd1ba 5512
33d07337
YZ		 5513 return ceph_auth_check_message_signature(auth, msg);
5514}
5515
9e32789f 5516static const struct ceph_connection_operations mds_con_ops = {
4972cf60
ID		 5517 .get = mds_get_con,
5518 .put = mds_put_con,
53ded495 5519 .alloc_msg = mds_alloc_msg,
4972cf60
ID		 5520 .dispatch = mds_dispatch,
5521 .peer_reset = mds_peer_reset,
5522 .get_authorizer = mds_get_authorizer,
5523 .add_authorizer_challenge = mds_add_authorizer_challenge,
5524 .verify_authorizer_reply = mds_verify_authorizer_reply,
5525 .invalidate_authorizer = mds_invalidate_authorizer,
79dbd1ba
ID		 5526 .sign_message = mds_sign_message,
5527 .check_message_signature = mds_check_message_signature,
cd1a677c
ID		 5528 .get_auth_request = mds_get_auth_request,
5529 .handle_auth_reply_more = mds_handle_auth_reply_more,
5530 .handle_auth_done = mds_handle_auth_done,
5531 .handle_auth_bad_method = mds_handle_auth_bad_method,
2f2dc053
SW		 5532};
5533
2f2dc053 5534/* eof */
Michael's Linux tree.