?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Add linux-next specific files for 20110817
[linux-2.6/next.git] / net / ceph / messenger.c
blobe6d485724dd592c5d0c76c49630cf37b1964df85
1 #include <linux/ceph/ceph_debug.h>
2
3 #include <linux/crc32c.h>
4 #include <linux/ctype.h>
5 #include <linux/highmem.h>
6 #include <linux/inet.h>
7 #include <linux/kthread.h>
8 #include <linux/net.h>
9 #include <linux/slab.h>
10 #include <linux/socket.h>
11 #include <linux/string.h>
12 #include <linux/bio.h>
13 #include <linux/blkdev.h>
14 #include <net/tcp.h>
15
16 #include <linux/ceph/libceph.h>
17 #include <linux/ceph/messenger.h>
18 #include <linux/ceph/decode.h>
19 #include <linux/ceph/pagelist.h>
20
21 /*
22 * Ceph uses the messenger to exchange ceph_msg messages with other
23 * hosts in the system. The messenger provides ordered and reliable
24 * delivery. We tolerate TCP disconnects by reconnecting (with
25 * exponential backoff) in the case of a fault (disconnection, bad
26 * crc, protocol error). Acks allow sent messages to be discarded by
27 * the sender.
28 */
29
30 /* static tag bytes (protocol control messages) */
31 static char tag_msg = CEPH_MSGR_TAG_MSG;
32 static char tag_ack = CEPH_MSGR_TAG_ACK;
33 static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE;
34
35 #ifdef CONFIG_LOCKDEP
36 static struct lock_class_key socket_class;
37 #endif
38
39
40 static void queue_con(struct ceph_connection *con);
41 static void con_work(struct work_struct *);
42 static void ceph_fault(struct ceph_connection *con);
43
44 /*
45 * nicely render a sockaddr as a string.
46 */
47 #define MAX_ADDR_STR 20
48 #define MAX_ADDR_STR_LEN 60
49 static char addr_str[MAX_ADDR_STR][MAX_ADDR_STR_LEN];
50 static DEFINE_SPINLOCK(addr_str_lock);
51 static int last_addr_str;
52
53 const char *ceph_pr_addr(const struct sockaddr_storage *ss)
54 {
55 int i;
56 char *s;
57 struct sockaddr_in *in4 = (void *)ss;
58 struct sockaddr_in6 *in6 = (void *)ss;
59
60 spin_lock(&addr_str_lock);
61 i = last_addr_str++;
62 if (last_addr_str == MAX_ADDR_STR)
63 last_addr_str = 0;
64 spin_unlock(&addr_str_lock);
65 s = addr_str[i];
66
67 switch (ss->ss_family) {
68 case AF_INET:
69 snprintf(s, MAX_ADDR_STR_LEN, "%pI4:%u", &in4->sin_addr,
70 (unsigned int)ntohs(in4->sin_port));
71 break;
72
73 case AF_INET6:
74 snprintf(s, MAX_ADDR_STR_LEN, "[%pI6c]:%u", &in6->sin6_addr,
75 (unsigned int)ntohs(in6->sin6_port));
76 break;
77
78 default:
79 snprintf(s, MAX_ADDR_STR_LEN, "(unknown sockaddr family %d)",
80 (int)ss->ss_family);
81 }
82
83 return s;
84 }
85 EXPORT_SYMBOL(ceph_pr_addr);
86
87 static void encode_my_addr(struct ceph_messenger *msgr)
88 {
89 memcpy(&msgr->my_enc_addr, &msgr->inst.addr, sizeof(msgr->my_enc_addr));
90 ceph_encode_addr(&msgr->my_enc_addr);
91 }
92
93 /*
94 * work queue for all reading and writing to/from the socket.
95 */
96 struct workqueue_struct *ceph_msgr_wq;
97
98 int ceph_msgr_init(void)
99 {
100 ceph_msgr_wq = alloc_workqueue("ceph-msgr", WQ_NON_REENTRANT, 0);
101 if (!ceph_msgr_wq) {
102 pr_err("msgr_init failed to create workqueue\n");
103 return -ENOMEM;
104 }
105 return 0;
106 }
107 EXPORT_SYMBOL(ceph_msgr_init);
108
109 void ceph_msgr_exit(void)
110 {
111 destroy_workqueue(ceph_msgr_wq);
112 }
113 EXPORT_SYMBOL(ceph_msgr_exit);
114
115 void ceph_msgr_flush(void)
116 {
117 flush_workqueue(ceph_msgr_wq);
118 }
119 EXPORT_SYMBOL(ceph_msgr_flush);
120
121
122 /*
123 * socket callback functions
124 */
125
126 /* data available on socket, or listen socket received a connect */
127 static void ceph_data_ready(struct sock *sk, int count_unused)
128 {
129 struct ceph_connection *con =
130 (struct ceph_connection *)sk->sk_user_data;
131 if (sk->sk_state != TCP_CLOSE_WAIT) {
132 dout("ceph_data_ready on %p state = %lu, queueing work\n",
133 con, con->state);
134 queue_con(con);
135 }
136 }
137
138 /* socket has buffer space for writing */
139 static void ceph_write_space(struct sock *sk)
140 {
141 struct ceph_connection *con =
142 (struct ceph_connection *)sk->sk_user_data;
143
144 /* only queue to workqueue if there is data we want to write. */
145 if (test_bit(WRITE_PENDING, &con->state)) {
146 dout("ceph_write_space %p queueing write work\n", con);
147 queue_con(con);
148 } else {
149 dout("ceph_write_space %p nothing to write\n", con);
150 }
151
152 /* since we have our own write_space, clear the SOCK_NOSPACE flag */
153 clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
154 }
155
156 /* socket's state has changed */
157 static void ceph_state_change(struct sock *sk)
158 {
159 struct ceph_connection *con =
160 (struct ceph_connection *)sk->sk_user_data;
161
162 dout("ceph_state_change %p state = %lu sk_state = %u\n",
163 con, con->state, sk->sk_state);
164
165 if (test_bit(CLOSED, &con->state))
166 return;
167
168 switch (sk->sk_state) {
169 case TCP_CLOSE:
170 dout("ceph_state_change TCP_CLOSE\n");
171 case TCP_CLOSE_WAIT:
172 dout("ceph_state_change TCP_CLOSE_WAIT\n");
173 if (test_and_set_bit(SOCK_CLOSED, &con->state) == 0) {
174 if (test_bit(CONNECTING, &con->state))
175 con->error_msg = "connection failed";
176 else
177 con->error_msg = "socket closed";
178 queue_con(con);
179 }
180 break;
181 case TCP_ESTABLISHED:
182 dout("ceph_state_change TCP_ESTABLISHED\n");
183 queue_con(con);
184 break;
185 }
186 }
187
188 /*
189 * set up socket callbacks
190 */
191 static void set_sock_callbacks(struct socket *sock,
192 struct ceph_connection *con)
193 {
194 struct sock *sk = sock->sk;
195 sk->sk_user_data = (void *)con;
196 sk->sk_data_ready = ceph_data_ready;
197 sk->sk_write_space = ceph_write_space;
198 sk->sk_state_change = ceph_state_change;
199 }
200
201
202 /*
203 * socket helpers
204 */
205
206 /*
207 * initiate connection to a remote socket.
208 */
209 static struct socket *ceph_tcp_connect(struct ceph_connection *con)
210 {
211 struct sockaddr_storage *paddr = &con->peer_addr.in_addr;
212 struct socket *sock;
213 int ret;
214
215 BUG_ON(con->sock);
216 ret = sock_create_kern(con->peer_addr.in_addr.ss_family, SOCK_STREAM,
217 IPPROTO_TCP, &sock);
218 if (ret)
219 return ERR_PTR(ret);
220 con->sock = sock;
221 sock->sk->sk_allocation = GFP_NOFS;
222
223 #ifdef CONFIG_LOCKDEP
224 lockdep_set_class(&sock->sk->sk_lock, &socket_class);
225 #endif
226
227 set_sock_callbacks(sock, con);
228
229 dout("connect %s\n", ceph_pr_addr(&con->peer_addr.in_addr));
230
231 ret = sock->ops->connect(sock, (struct sockaddr *)paddr, sizeof(*paddr),
232 O_NONBLOCK);
233 if (ret == -EINPROGRESS) {
234 dout("connect %s EINPROGRESS sk_state = %u\n",
235 ceph_pr_addr(&con->peer_addr.in_addr),
236 sock->sk->sk_state);
237 ret = 0;
238 }
239 if (ret < 0) {
240 pr_err("connect %s error %d\n",
241 ceph_pr_addr(&con->peer_addr.in_addr), ret);
242 sock_release(sock);
243 con->sock = NULL;
244 con->error_msg = "connect error";
245 }
246
247 if (ret < 0)
248 return ERR_PTR(ret);
249 return sock;
250 }
251
252 static int ceph_tcp_recvmsg(struct socket *sock, void *buf, size_t len)
253 {
254 struct kvec iov = {buf, len};
255 struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
256 int r;
257
258 r = kernel_recvmsg(sock, &msg, &iov, 1, len, msg.msg_flags);
259 if (r == -EAGAIN)
260 r = 0;
261 return r;
262 }
263
264 /*
265 * write something. @more is true if caller will be sending more data
266 * shortly.
267 */
268 static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov,
269 size_t kvlen, size_t len, int more)
270 {
271 struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL };
272 int r;
273
274 if (more)
275 msg.msg_flags |= MSG_MORE;
276 else
277 msg.msg_flags |= MSG_EOR; /* superfluous, but what the hell */
278
279 r = kernel_sendmsg(sock, &msg, iov, kvlen, len);
280 if (r == -EAGAIN)
281 r = 0;
282 return r;
283 }
284
285
286 /*
287 * Shutdown/close the socket for the given connection.
288 */
289 static int con_close_socket(struct ceph_connection *con)
290 {
291 int rc;
292
293 dout("con_close_socket on %p sock %p\n", con, con->sock);
294 if (!con->sock)
295 return 0;
296 set_bit(SOCK_CLOSED, &con->state);
297 rc = con->sock->ops->shutdown(con->sock, SHUT_RDWR);
298 sock_release(con->sock);
299 con->sock = NULL;
300 clear_bit(SOCK_CLOSED, &con->state);
301 return rc;
302 }
303
304 /*
305 * Reset a connection. Discard all incoming and outgoing messages
306 * and clear *_seq state.
307 */
308 static void ceph_msg_remove(struct ceph_msg *msg)
309 {
310 list_del_init(&msg->list_head);
311 ceph_msg_put(msg);
312 }
313 static void ceph_msg_remove_list(struct list_head *head)
314 {
315 while (!list_empty(head)) {
316 struct ceph_msg *msg = list_first_entry(head, struct ceph_msg,
317 list_head);
318 ceph_msg_remove(msg);
319 }
320 }
321
322 static void reset_connection(struct ceph_connection *con)
323 {
324 /* reset connection, out_queue, msg_ and connect_seq */
325 /* discard existing out_queue and msg_seq */
326 ceph_msg_remove_list(&con->out_queue);
327 ceph_msg_remove_list(&con->out_sent);
328
329 if (con->in_msg) {
330 ceph_msg_put(con->in_msg);
331 con->in_msg = NULL;
332 }
333
334 con->connect_seq = 0;
335 con->out_seq = 0;
336 if (con->out_msg) {
337 ceph_msg_put(con->out_msg);
338 con->out_msg = NULL;
339 }
340 con->in_seq = 0;
341 con->in_seq_acked = 0;
342 }
343
344 /*
345 * mark a peer down. drop any open connections.
346 */
347 void ceph_con_close(struct ceph_connection *con)
348 {
349 dout("con_close %p peer %s\n", con,
350 ceph_pr_addr(&con->peer_addr.in_addr));
351 set_bit(CLOSED, &con->state); /* in case there's queued work */
352 clear_bit(STANDBY, &con->state); /* avoid connect_seq bump */
353 clear_bit(LOSSYTX, &con->state); /* so we retry next connect */
354 clear_bit(KEEPALIVE_PENDING, &con->state);
355 clear_bit(WRITE_PENDING, &con->state);
356 mutex_lock(&con->mutex);
357 reset_connection(con);
358 con->peer_global_seq = 0;
359 cancel_delayed_work(&con->work);
360 mutex_unlock(&con->mutex);
361 queue_con(con);
362 }
363 EXPORT_SYMBOL(ceph_con_close);
364
365 /*
366 * Reopen a closed connection, with a new peer address.
367 */
368 void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr)
369 {
370 dout("con_open %p %s\n", con, ceph_pr_addr(&addr->in_addr));
371 set_bit(OPENING, &con->state);
372 clear_bit(CLOSED, &con->state);
373 memcpy(&con->peer_addr, addr, sizeof(*addr));
374 con->delay = 0; /* reset backoff memory */
375 queue_con(con);
376 }
377 EXPORT_SYMBOL(ceph_con_open);
378
379 /*
380 * return true if this connection ever successfully opened
381 */
382 bool ceph_con_opened(struct ceph_connection *con)
383 {
384 return con->connect_seq > 0;
385 }
386
387 /*
388 * generic get/put
389 */
390 struct ceph_connection *ceph_con_get(struct ceph_connection *con)
391 {
392 dout("con_get %p nref = %d -> %d\n", con,
393 atomic_read(&con->nref), atomic_read(&con->nref) + 1);
394 if (atomic_inc_not_zero(&con->nref))
395 return con;
396 return NULL;
397 }
398
399 void ceph_con_put(struct ceph_connection *con)
400 {
401 dout("con_put %p nref = %d -> %d\n", con,
402 atomic_read(&con->nref), atomic_read(&con->nref) - 1);
403 BUG_ON(atomic_read(&con->nref) == 0);
404 if (atomic_dec_and_test(&con->nref)) {
405 BUG_ON(con->sock);
406 kfree(con);
407 }
408 }
409
410 /*
411 * initialize a new connection.
412 */
413 void ceph_con_init(struct ceph_messenger *msgr, struct ceph_connection *con)
414 {
415 dout("con_init %p\n", con);
416 memset(con, 0, sizeof(*con));
417 atomic_set(&con->nref, 1);
418 con->msgr = msgr;
419 mutex_init(&con->mutex);
420 INIT_LIST_HEAD(&con->out_queue);
421 INIT_LIST_HEAD(&con->out_sent);
422 INIT_DELAYED_WORK(&con->work, con_work);
423 }
424 EXPORT_SYMBOL(ceph_con_init);
425
426
427 /*
428 * We maintain a global counter to order connection attempts. Get
429 * a unique seq greater than @gt.
430 */
431 static u32 get_global_seq(struct ceph_messenger *msgr, u32 gt)
432 {
433 u32 ret;
434
435 spin_lock(&msgr->global_seq_lock);
436 if (msgr->global_seq < gt)
437 msgr->global_seq = gt;
438 ret = ++msgr->global_seq;
439 spin_unlock(&msgr->global_seq_lock);
440 return ret;
441 }
442
443
444 /*
445 * Prepare footer for currently outgoing message, and finish things
446 * off. Assumes out_kvec* are already valid.. we just add on to the end.
447 */
448 static void prepare_write_message_footer(struct ceph_connection *con, int v)
449 {
450 struct ceph_msg *m = con->out_msg;
451
452 dout("prepare_write_message_footer %p\n", con);
453 con->out_kvec_is_msg = true;
454 con->out_kvec[v].iov_base = &m->footer;
455 con->out_kvec[v].iov_len = sizeof(m->footer);
456 con->out_kvec_bytes += sizeof(m->footer);
457 con->out_kvec_left++;
458 con->out_more = m->more_to_follow;
459 con->out_msg_done = true;
460 }
461
462 /*
463 * Prepare headers for the next outgoing message.
464 */
465 static void prepare_write_message(struct ceph_connection *con)
466 {
467 struct ceph_msg *m;
468 int v = 0;
469
470 con->out_kvec_bytes = 0;
471 con->out_kvec_is_msg = true;
472 con->out_msg_done = false;
473
474 /* Sneak an ack in there first? If we can get it into the same
475 * TCP packet that's a good thing. */
476 if (con->in_seq > con->in_seq_acked) {
477 con->in_seq_acked = con->in_seq;
478 con->out_kvec[v].iov_base = &tag_ack;
479 con->out_kvec[v++].iov_len = 1;
480 con->out_temp_ack = cpu_to_le64(con->in_seq_acked);
481 con->out_kvec[v].iov_base = &con->out_temp_ack;
482 con->out_kvec[v++].iov_len = sizeof(con->out_temp_ack);
483 con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack);
484 }
485
486 m = list_first_entry(&con->out_queue,
487 struct ceph_msg, list_head);
488 con->out_msg = m;
489
490 /* put message on sent list */
491 ceph_msg_get(m);
492 list_move_tail(&m->list_head, &con->out_sent);
493
494 /*
495 * only assign outgoing seq # if we haven't sent this message
496 * yet. if it is requeued, resend with it's original seq.
497 */
498 if (m->needs_out_seq) {
499 m->hdr.seq = cpu_to_le64(++con->out_seq);
500 m->needs_out_seq = false;
501 }
502
503 dout("prepare_write_message %p seq %lld type %d len %d+%d+%d %d pgs\n",
504 m, con->out_seq, le16_to_cpu(m->hdr.type),
505 le32_to_cpu(m->hdr.front_len), le32_to_cpu(m->hdr.middle_len),
506 le32_to_cpu(m->hdr.data_len),
507 m->nr_pages);
508 BUG_ON(le32_to_cpu(m->hdr.front_len) != m->front.iov_len);
509
510 /* tag + hdr + front + middle */
511 con->out_kvec[v].iov_base = &tag_msg;
512 con->out_kvec[v++].iov_len = 1;
513 con->out_kvec[v].iov_base = &m->hdr;
514 con->out_kvec[v++].iov_len = sizeof(m->hdr);
515 con->out_kvec[v++] = m->front;
516 if (m->middle)
517 con->out_kvec[v++] = m->middle->vec;
518 con->out_kvec_left = v;
519 con->out_kvec_bytes += 1 + sizeof(m->hdr) + m->front.iov_len +
520 (m->middle ? m->middle->vec.iov_len : 0);
521 con->out_kvec_cur = con->out_kvec;
522
523 /* fill in crc (except data pages), footer */
524 con->out_msg->hdr.crc =
525 cpu_to_le32(crc32c(0, (void *)&m->hdr,
526 sizeof(m->hdr) - sizeof(m->hdr.crc)));
527 con->out_msg->footer.flags = CEPH_MSG_FOOTER_COMPLETE;
528 con->out_msg->footer.front_crc =
529 cpu_to_le32(crc32c(0, m->front.iov_base, m->front.iov_len));
530 if (m->middle)
531 con->out_msg->footer.middle_crc =
532 cpu_to_le32(crc32c(0, m->middle->vec.iov_base,
533 m->middle->vec.iov_len));
534 else
535 con->out_msg->footer.middle_crc = 0;
536 con->out_msg->footer.data_crc = 0;
537 dout("prepare_write_message front_crc %u data_crc %u\n",
538 le32_to_cpu(con->out_msg->footer.front_crc),
539 le32_to_cpu(con->out_msg->footer.middle_crc));
540
541 /* is there a data payload? */
542 if (le32_to_cpu(m->hdr.data_len) > 0) {
543 /* initialize page iterator */
544 con->out_msg_pos.page = 0;
545 if (m->pages)
546 con->out_msg_pos.page_pos = m->page_alignment;
547 else
548 con->out_msg_pos.page_pos = 0;
549 con->out_msg_pos.data_pos = 0;
550 con->out_msg_pos.did_page_crc = 0;
551 con->out_more = 1; /* data + footer will follow */
552 } else {
553 /* no, queue up footer too and be done */
554 prepare_write_message_footer(con, v);
555 }
556
557 set_bit(WRITE_PENDING, &con->state);
558 }
559
560 /*
561 * Prepare an ack.
562 */
563 static void prepare_write_ack(struct ceph_connection *con)
564 {
565 dout("prepare_write_ack %p %llu -> %llu\n", con,
566 con->in_seq_acked, con->in_seq);
567 con->in_seq_acked = con->in_seq;
568
569 con->out_kvec[0].iov_base = &tag_ack;
570 con->out_kvec[0].iov_len = 1;
571 con->out_temp_ack = cpu_to_le64(con->in_seq_acked);
572 con->out_kvec[1].iov_base = &con->out_temp_ack;
573 con->out_kvec[1].iov_len = sizeof(con->out_temp_ack);
574 con->out_kvec_left = 2;
575 con->out_kvec_bytes = 1 + sizeof(con->out_temp_ack);
576 con->out_kvec_cur = con->out_kvec;
577 con->out_more = 1; /* more will follow.. eventually.. */
578 set_bit(WRITE_PENDING, &con->state);
579 }
580
581 /*
582 * Prepare to write keepalive byte.
583 */
584 static void prepare_write_keepalive(struct ceph_connection *con)
585 {
586 dout("prepare_write_keepalive %p\n", con);
587 con->out_kvec[0].iov_base = &tag_keepalive;
588 con->out_kvec[0].iov_len = 1;
589 con->out_kvec_left = 1;
590 con->out_kvec_bytes = 1;
591 con->out_kvec_cur = con->out_kvec;
592 set_bit(WRITE_PENDING, &con->state);
593 }
594
595 /*
596 * Connection negotiation.
597 */
598
599 static int prepare_connect_authorizer(struct ceph_connection *con)
600 {
601 void *auth_buf;
602 int auth_len = 0;
603 int auth_protocol = 0;
604
605 mutex_unlock(&con->mutex);
606 if (con->ops->get_authorizer)
607 con->ops->get_authorizer(con, &auth_buf, &auth_len,
608 &auth_protocol, &con->auth_reply_buf,
609 &con->auth_reply_buf_len,
610 con->auth_retry);
611 mutex_lock(&con->mutex);
612
613 if (test_bit(CLOSED, &con->state) ||
614 test_bit(OPENING, &con->state))
615 return -EAGAIN;
616
617 con->out_connect.authorizer_protocol = cpu_to_le32(auth_protocol);
618 con->out_connect.authorizer_len = cpu_to_le32(auth_len);
619
620 if (auth_len) {
621 con->out_kvec[con->out_kvec_left].iov_base = auth_buf;
622 con->out_kvec[con->out_kvec_left].iov_len = auth_len;
623 con->out_kvec_left++;
624 con->out_kvec_bytes += auth_len;
625 }
626 return 0;
627 }
628
629 /*
630 * We connected to a peer and are saying hello.
631 */
632 static void prepare_write_banner(struct ceph_messenger *msgr,
633 struct ceph_connection *con)
634 {
635 int len = strlen(CEPH_BANNER);
636
637 con->out_kvec[0].iov_base = CEPH_BANNER;
638 con->out_kvec[0].iov_len = len;
639 con->out_kvec[1].iov_base = &msgr->my_enc_addr;
640 con->out_kvec[1].iov_len = sizeof(msgr->my_enc_addr);
641 con->out_kvec_left = 2;
642 con->out_kvec_bytes = len + sizeof(msgr->my_enc_addr);
643 con->out_kvec_cur = con->out_kvec;
644 con->out_more = 0;
645 set_bit(WRITE_PENDING, &con->state);
646 }
647
648 static int prepare_write_connect(struct ceph_messenger *msgr,
649 struct ceph_connection *con,
650 int after_banner)
651 {
652 unsigned global_seq = get_global_seq(con->msgr, 0);
653 int proto;
654
655 switch (con->peer_name.type) {
656 case CEPH_ENTITY_TYPE_MON:
657 proto = CEPH_MONC_PROTOCOL;
658 break;
659 case CEPH_ENTITY_TYPE_OSD:
660 proto = CEPH_OSDC_PROTOCOL;
661 break;
662 case CEPH_ENTITY_TYPE_MDS:
663 proto = CEPH_MDSC_PROTOCOL;
664 break;
665 default:
666 BUG();
667 }
668
669 dout("prepare_write_connect %p cseq=%d gseq=%d proto=%d\n", con,
670 con->connect_seq, global_seq, proto);
671
672 con->out_connect.features = cpu_to_le64(msgr->supported_features);
673 con->out_connect.host_type = cpu_to_le32(CEPH_ENTITY_TYPE_CLIENT);
674 con->out_connect.connect_seq = cpu_to_le32(con->connect_seq);
675 con->out_connect.global_seq = cpu_to_le32(global_seq);
676 con->out_connect.protocol_version = cpu_to_le32(proto);
677 con->out_connect.flags = 0;
678
679 if (!after_banner) {
680 con->out_kvec_left = 0;
681 con->out_kvec_bytes = 0;
682 }
683 con->out_kvec[con->out_kvec_left].iov_base = &con->out_connect;
684 con->out_kvec[con->out_kvec_left].iov_len = sizeof(con->out_connect);
685 con->out_kvec_left++;
686 con->out_kvec_bytes += sizeof(con->out_connect);
687 con->out_kvec_cur = con->out_kvec;
688 con->out_more = 0;
689 set_bit(WRITE_PENDING, &con->state);
690
691 return prepare_connect_authorizer(con);
692 }
693
694
695 /*
696 * write as much of pending kvecs to the socket as we can.
697 * 1 -> done
698 * 0 -> socket full, but more to do
699 * <0 -> error
700 */
701 static int write_partial_kvec(struct ceph_connection *con)
702 {
703 int ret;
704
705 dout("write_partial_kvec %p %d left\n", con, con->out_kvec_bytes);
706 while (con->out_kvec_bytes > 0) {
707 ret = ceph_tcp_sendmsg(con->sock, con->out_kvec_cur,
708 con->out_kvec_left, con->out_kvec_bytes,
709 con->out_more);
710 if (ret <= 0)
711 goto out;
712 con->out_kvec_bytes -= ret;
713 if (con->out_kvec_bytes == 0)
714 break; /* done */
715 while (ret > 0) {
716 if (ret >= con->out_kvec_cur->iov_len) {
717 ret -= con->out_kvec_cur->iov_len;
718 con->out_kvec_cur++;
719 con->out_kvec_left--;
720 } else {
721 con->out_kvec_cur->iov_len -= ret;
722 con->out_kvec_cur->iov_base += ret;
723 ret = 0;
724 break;
725 }
726 }
727 }
728 con->out_kvec_left = 0;
729 con->out_kvec_is_msg = false;
730 ret = 1;
731 out:
732 dout("write_partial_kvec %p %d left in %d kvecs ret = %d\n", con,
733 con->out_kvec_bytes, con->out_kvec_left, ret);
734 return ret; /* done! */
735 }
736
737 #ifdef CONFIG_BLOCK
738 static void init_bio_iter(struct bio *bio, struct bio **iter, int *seg)
739 {
740 if (!bio) {
741 *iter = NULL;
742 *seg = 0;
743 return;
744 }
745 *iter = bio;
746 *seg = bio->bi_idx;
747 }
748
749 static void iter_bio_next(struct bio **bio_iter, int *seg)
750 {
751 if (*bio_iter == NULL)
752 return;
753
754 BUG_ON(*seg >= (*bio_iter)->bi_vcnt);
755
756 (*seg)++;
757 if (*seg == (*bio_iter)->bi_vcnt)
758 init_bio_iter((*bio_iter)->bi_next, bio_iter, seg);
759 }
760 #endif
761
762 /*
763 * Write as much message data payload as we can. If we finish, queue
764 * up the footer.
765 * 1 -> done, footer is now queued in out_kvec[].
766 * 0 -> socket full, but more to do
767 * <0 -> error
768 */
769 static int write_partial_msg_pages(struct ceph_connection *con)
770 {
771 struct ceph_msg *msg = con->out_msg;
772 unsigned data_len = le32_to_cpu(msg->hdr.data_len);
773 size_t len;
774 int crc = con->msgr->nocrc;
775 int ret;
776 int total_max_write;
777 int in_trail = 0;
778 size_t trail_len = (msg->trail ? msg->trail->length : 0);
779
780 dout("write_partial_msg_pages %p msg %p page %d/%d offset %d\n",
781 con, con->out_msg, con->out_msg_pos.page, con->out_msg->nr_pages,
782 con->out_msg_pos.page_pos);
783
784 #ifdef CONFIG_BLOCK
785 if (msg->bio && !msg->bio_iter)
786 init_bio_iter(msg->bio, &msg->bio_iter, &msg->bio_seg);
787 #endif
788
789 while (data_len > con->out_msg_pos.data_pos) {
790 struct page *page = NULL;
791 void *kaddr = NULL;
792 int max_write = PAGE_SIZE;
793 int page_shift = 0;
794
795 total_max_write = data_len - trail_len -
796 con->out_msg_pos.data_pos;
797
798 /*
799 * if we are calculating the data crc (the default), we need
800 * to map the page. if our pages[] has been revoked, use the
801 * zero page.
802 */
803
804 /* have we reached the trail part of the data? */
805 if (con->out_msg_pos.data_pos >= data_len - trail_len) {
806 in_trail = 1;
807
808 total_max_write = data_len - con->out_msg_pos.data_pos;
809
810 page = list_first_entry(&msg->trail->head,
811 struct page, lru);
812 if (crc)
813 kaddr = kmap(page);
814 max_write = PAGE_SIZE;
815 } else if (msg->pages) {
816 page = msg->pages[con->out_msg_pos.page];
817 if (crc)
818 kaddr = kmap(page);
819 } else if (msg->pagelist) {
820 page = list_first_entry(&msg->pagelist->head,
821 struct page, lru);
822 if (crc)
823 kaddr = kmap(page);
824 #ifdef CONFIG_BLOCK
825 } else if (msg->bio) {
826 struct bio_vec *bv;
827
828 bv = bio_iovec_idx(msg->bio_iter, msg->bio_seg);
829 page = bv->bv_page;
830 page_shift = bv->bv_offset;
831 if (crc)
832 kaddr = kmap(page) + page_shift;
833 max_write = bv->bv_len;
834 #endif
835 } else {
836 page = con->msgr->zero_page;
837 if (crc)
838 kaddr = page_address(con->msgr->zero_page);
839 }
840 len = min_t(int, max_write - con->out_msg_pos.page_pos,
841 total_max_write);
842
843 if (crc && !con->out_msg_pos.did_page_crc) {
844 void *base = kaddr + con->out_msg_pos.page_pos;
845 u32 tmpcrc = le32_to_cpu(con->out_msg->footer.data_crc);
846
847 BUG_ON(kaddr == NULL);
848 con->out_msg->footer.data_crc =
849 cpu_to_le32(crc32c(tmpcrc, base, len));
850 con->out_msg_pos.did_page_crc = 1;
851 }
852 ret = kernel_sendpage(con->sock, page,
853 con->out_msg_pos.page_pos + page_shift,
854 len,
855 MSG_DONTWAIT | MSG_NOSIGNAL |
856 MSG_MORE);
857
858 if (crc &&
859 (msg->pages || msg->pagelist || msg->bio || in_trail))
860 kunmap(page);
861
862 if (ret == -EAGAIN)
863 ret = 0;
864 if (ret <= 0)
865 goto out;
866
867 con->out_msg_pos.data_pos += ret;
868 con->out_msg_pos.page_pos += ret;
869 if (ret == len) {
870 con->out_msg_pos.page_pos = 0;
871 con->out_msg_pos.page++;
872 con->out_msg_pos.did_page_crc = 0;
873 if (in_trail)
874 list_move_tail(&page->lru,
875 &msg->trail->head);
876 else if (msg->pagelist)
877 list_move_tail(&page->lru,
878 &msg->pagelist->head);
879 #ifdef CONFIG_BLOCK
880 else if (msg->bio)
881 iter_bio_next(&msg->bio_iter, &msg->bio_seg);
882 #endif
883 }
884 }
885
886 dout("write_partial_msg_pages %p msg %p done\n", con, msg);
887
888 /* prepare and queue up footer, too */
889 if (!crc)
890 con->out_msg->footer.flags |= CEPH_MSG_FOOTER_NOCRC;
891 con->out_kvec_bytes = 0;
892 con->out_kvec_left = 0;
893 con->out_kvec_cur = con->out_kvec;
894 prepare_write_message_footer(con, 0);
895 ret = 1;
896 out:
897 return ret;
898 }
899
900 /*
901 * write some zeros
902 */
903 static int write_partial_skip(struct ceph_connection *con)
904 {
905 int ret;
906
907 while (con->out_skip > 0) {
908 struct kvec iov = {
909 .iov_base = page_address(con->msgr->zero_page),
910 .iov_len = min(con->out_skip, (int)PAGE_CACHE_SIZE)
911 };
912
913 ret = ceph_tcp_sendmsg(con->sock, &iov, 1, iov.iov_len, 1);
914 if (ret <= 0)
915 goto out;
916 con->out_skip -= ret;
917 }
918 ret = 1;
919 out:
920 return ret;
921 }
922
923 /*
924 * Prepare to read connection handshake, or an ack.
925 */
926 static void prepare_read_banner(struct ceph_connection *con)
927 {
928 dout("prepare_read_banner %p\n", con);
929 con->in_base_pos = 0;
930 }
931
932 static void prepare_read_connect(struct ceph_connection *con)
933 {
934 dout("prepare_read_connect %p\n", con);
935 con->in_base_pos = 0;
936 }
937
938 static void prepare_read_ack(struct ceph_connection *con)
939 {
940 dout("prepare_read_ack %p\n", con);
941 con->in_base_pos = 0;
942 }
943
944 static void prepare_read_tag(struct ceph_connection *con)
945 {
946 dout("prepare_read_tag %p\n", con);
947 con->in_base_pos = 0;
948 con->in_tag = CEPH_MSGR_TAG_READY;
949 }
950
951 /*
952 * Prepare to read a message.
953 */
954 static int prepare_read_message(struct ceph_connection *con)
955 {
956 dout("prepare_read_message %p\n", con);
957 BUG_ON(con->in_msg != NULL);
958 con->in_base_pos = 0;
959 con->in_front_crc = con->in_middle_crc = con->in_data_crc = 0;
960 return 0;
961 }
962
963
964 static int read_partial(struct ceph_connection *con,
965 int *to, int size, void *object)
966 {
967 *to += size;
968 while (con->in_base_pos < *to) {
969 int left = *to - con->in_base_pos;
970 int have = size - left;
971 int ret = ceph_tcp_recvmsg(con->sock, object + have, left);
972 if (ret <= 0)
973 return ret;
974 con->in_base_pos += ret;
975 }
976 return 1;
977 }
978
979
980 /*
981 * Read all or part of the connect-side handshake on a new connection
982 */
983 static int read_partial_banner(struct ceph_connection *con)
984 {
985 int ret, to = 0;
986
987 dout("read_partial_banner %p at %d\n", con, con->in_base_pos);
988
989 /* peer's banner */
990 ret = read_partial(con, &to, strlen(CEPH_BANNER), con->in_banner);
991 if (ret <= 0)
992 goto out;
993 ret = read_partial(con, &to, sizeof(con->actual_peer_addr),
994 &con->actual_peer_addr);
995 if (ret <= 0)
996 goto out;
997 ret = read_partial(con, &to, sizeof(con->peer_addr_for_me),
998 &con->peer_addr_for_me);
999 if (ret <= 0)
1000 goto out;
1001 out:
1002 return ret;
1003 }
1004
1005 static int read_partial_connect(struct ceph_connection *con)
1006 {
1007 int ret, to = 0;
1008
1009 dout("read_partial_connect %p at %d\n", con, con->in_base_pos);
1010
1011 ret = read_partial(con, &to, sizeof(con->in_reply), &con->in_reply);
1012 if (ret <= 0)
1013 goto out;
1014 ret = read_partial(con, &to, le32_to_cpu(con->in_reply.authorizer_len),
1015 con->auth_reply_buf);
1016 if (ret <= 0)
1017 goto out;
1018
1019 dout("read_partial_connect %p tag %d, con_seq = %u, g_seq = %u\n",
1020 con, (int)con->in_reply.tag,
1021 le32_to_cpu(con->in_reply.connect_seq),
1022 le32_to_cpu(con->in_reply.global_seq));
1023 out:
1024 return ret;
1025
1026 }
1027
1028 /*
1029 * Verify the hello banner looks okay.
1030 */
1031 static int verify_hello(struct ceph_connection *con)
1032 {
1033 if (memcmp(con->in_banner, CEPH_BANNER, strlen(CEPH_BANNER))) {
1034 pr_err("connect to %s got bad banner\n",
1035 ceph_pr_addr(&con->peer_addr.in_addr));
1036 con->error_msg = "protocol error, bad banner";
1037 return -1;
1038 }
1039 return 0;
1040 }
1041
1042 static bool addr_is_blank(struct sockaddr_storage *ss)
1043 {
1044 switch (ss->ss_family) {
1045 case AF_INET:
1046 return ((struct sockaddr_in *)ss)->sin_addr.s_addr == 0;
1047 case AF_INET6:
1048 return
1049 ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[0] == 0 &&
1050 ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[1] == 0 &&
1051 ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[2] == 0 &&
1052 ((struct sockaddr_in6 *)ss)->sin6_addr.s6_addr32[3] == 0;
1053 }
1054 return false;
1055 }
1056
1057 static int addr_port(struct sockaddr_storage *ss)
1058 {
1059 switch (ss->ss_family) {
1060 case AF_INET:
1061 return ntohs(((struct sockaddr_in *)ss)->sin_port);
1062 case AF_INET6:
1063 return ntohs(((struct sockaddr_in6 *)ss)->sin6_port);
1064 }
1065 return 0;
1066 }
1067
1068 static void addr_set_port(struct sockaddr_storage *ss, int p)
1069 {
1070 switch (ss->ss_family) {
1071 case AF_INET:
1072 ((struct sockaddr_in *)ss)->sin_port = htons(p);
1073 break;
1074 case AF_INET6:
1075 ((struct sockaddr_in6 *)ss)->sin6_port = htons(p);
1076 break;
1077 }
1078 }
1079
1080 /*
1081 * Parse an ip[:port] list into an addr array. Use the default
1082 * monitor port if a port isn't specified.
1083 */
1084 int ceph_parse_ips(const char *c, const char *end,
1085 struct ceph_entity_addr *addr,
1086 int max_count, int *count)
1087 {
1088 int i;
1089 const char *p = c;
1090
1091 dout("parse_ips on '%.*s'\n", (int)(end-c), c);
1092 for (i = 0; i < max_count; i++) {
1093 const char *ipend;
1094 struct sockaddr_storage *ss = &addr[i].in_addr;
1095 struct sockaddr_in *in4 = (void *)ss;
1096 struct sockaddr_in6 *in6 = (void *)ss;
1097 int port;
1098 char delim = ',';
1099
1100 if (*p == '[') {
1101 delim = ']';
1102 p++;
1103 }
1104
1105 memset(ss, 0, sizeof(*ss));
1106 if (in4_pton(p, end - p, (u8 *)&in4->sin_addr.s_addr,
1107 delim, &ipend))
1108 ss->ss_family = AF_INET;
1109 else if (in6_pton(p, end - p, (u8 *)&in6->sin6_addr.s6_addr,
1110 delim, &ipend))
1111 ss->ss_family = AF_INET6;
1112 else
1113 goto bad;
1114 p = ipend;
1115
1116 if (delim == ']') {
1117 if (*p != ']') {
1118 dout("missing matching ']'\n");
1119 goto bad;
1120 }
1121 p++;
1122 }
1123
1124 /* port? */
1125 if (p < end && *p == ':') {
1126 port = 0;
1127 p++;
1128 while (p < end && *p >= '0' && *p <= '9') {
1129 port = (port * 10) + (*p - '0');
1130 p++;
1131 }
1132 if (port > 65535 || port == 0)
1133 goto bad;
1134 } else {
1135 port = CEPH_MON_PORT;
1136 }
1137
1138 addr_set_port(ss, port);
1139
1140 dout("parse_ips got %s\n", ceph_pr_addr(ss));
1141
1142 if (p == end)
1143 break;
1144 if (*p != ',')
1145 goto bad;
1146 p++;
1147 }
1148
1149 if (p != end)
1150 goto bad;
1151
1152 if (count)
1153 *count = i + 1;
1154 return 0;
1155
1156 bad:
1157 pr_err("parse_ips bad ip '%.*s'\n", (int)(end - c), c);
1158 return -EINVAL;
1159 }
1160 EXPORT_SYMBOL(ceph_parse_ips);
1161
1162 static int process_banner(struct ceph_connection *con)
1163 {
1164 dout("process_banner on %p\n", con);
1165
1166 if (verify_hello(con) < 0)
1167 return -1;
1168
1169 ceph_decode_addr(&con->actual_peer_addr);
1170 ceph_decode_addr(&con->peer_addr_for_me);
1171
1172 /*
1173 * Make sure the other end is who we wanted. note that the other
1174 * end may not yet know their ip address, so if it's 0.0.0.0, give
1175 * them the benefit of the doubt.
1176 */
1177 if (memcmp(&con->peer_addr, &con->actual_peer_addr,
1178 sizeof(con->peer_addr)) != 0 &&
1179 !(addr_is_blank(&con->actual_peer_addr.in_addr) &&
1180 con->actual_peer_addr.nonce == con->peer_addr.nonce)) {
1181 pr_warning("wrong peer, want %s/%d, got %s/%d\n",
1182 ceph_pr_addr(&con->peer_addr.in_addr),
1183 (int)le32_to_cpu(con->peer_addr.nonce),
1184 ceph_pr_addr(&con->actual_peer_addr.in_addr),
1185 (int)le32_to_cpu(con->actual_peer_addr.nonce));
1186 con->error_msg = "wrong peer at address";
1187 return -1;
1188 }
1189
1190 /*
1191 * did we learn our address?
1192 */
1193 if (addr_is_blank(&con->msgr->inst.addr.in_addr)) {
1194 int port = addr_port(&con->msgr->inst.addr.in_addr);
1195
1196 memcpy(&con->msgr->inst.addr.in_addr,
1197 &con->peer_addr_for_me.in_addr,
1198 sizeof(con->peer_addr_for_me.in_addr));
1199 addr_set_port(&con->msgr->inst.addr.in_addr, port);
1200 encode_my_addr(con->msgr);
1201 dout("process_banner learned my addr is %s\n",
1202 ceph_pr_addr(&con->msgr->inst.addr.in_addr));
1203 }
1204
1205 set_bit(NEGOTIATING, &con->state);
1206 prepare_read_connect(con);
1207 return 0;
1208 }
1209
1210 static void fail_protocol(struct ceph_connection *con)
1211 {
1212 reset_connection(con);
1213 set_bit(CLOSED, &con->state); /* in case there's queued work */
1214
1215 mutex_unlock(&con->mutex);
1216 if (con->ops->bad_proto)
1217 con->ops->bad_proto(con);
1218 mutex_lock(&con->mutex);
1219 }
1220
1221 static int process_connect(struct ceph_connection *con)
1222 {
1223 u64 sup_feat = con->msgr->supported_features;
1224 u64 req_feat = con->msgr->required_features;
1225 u64 server_feat = le64_to_cpu(con->in_reply.features);
1226 int ret;
1227
1228 dout("process_connect on %p tag %d\n", con, (int)con->in_tag);
1229
1230 switch (con->in_reply.tag) {
1231 case CEPH_MSGR_TAG_FEATURES:
1232 pr_err("%s%lld %s feature set mismatch,"
1233 " my %llx < server's %llx, missing %llx\n",
1234 ENTITY_NAME(con->peer_name),
1235 ceph_pr_addr(&con->peer_addr.in_addr),
1236 sup_feat, server_feat, server_feat & ~sup_feat);
1237 con->error_msg = "missing required protocol features";
1238 fail_protocol(con);
1239 return -1;
1240
1241 case CEPH_MSGR_TAG_BADPROTOVER:
1242 pr_err("%s%lld %s protocol version mismatch,"
1243 " my %d != server's %d\n",
1244 ENTITY_NAME(con->peer_name),
1245 ceph_pr_addr(&con->peer_addr.in_addr),
1246 le32_to_cpu(con->out_connect.protocol_version),
1247 le32_to_cpu(con->in_reply.protocol_version));
1248 con->error_msg = "protocol version mismatch";
1249 fail_protocol(con);
1250 return -1;
1251
1252 case CEPH_MSGR_TAG_BADAUTHORIZER:
1253 con->auth_retry++;
1254 dout("process_connect %p got BADAUTHORIZER attempt %d\n", con,
1255 con->auth_retry);
1256 if (con->auth_retry == 2) {
1257 con->error_msg = "connect authorization failure";
1258 return -1;
1259 }
1260 con->auth_retry = 1;
1261 ret = prepare_write_connect(con->msgr, con, 0);
1262 if (ret < 0)
1263 return ret;
1264 prepare_read_connect(con);
1265 break;
1266
1267 case CEPH_MSGR_TAG_RESETSESSION:
1268 /*
1269 * If we connected with a large connect_seq but the peer
1270 * has no record of a session with us (no connection, or
1271 * connect_seq == 0), they will send RESETSESION to indicate
1272 * that they must have reset their session, and may have
1273 * dropped messages.
1274 */
1275 dout("process_connect got RESET peer seq %u\n",
1276 le32_to_cpu(con->in_connect.connect_seq));
1277 pr_err("%s%lld %s connection reset\n",
1278 ENTITY_NAME(con->peer_name),
1279 ceph_pr_addr(&con->peer_addr.in_addr));
1280 reset_connection(con);
1281 prepare_write_connect(con->msgr, con, 0);
1282 prepare_read_connect(con);
1283
1284 /* Tell ceph about it. */
1285 mutex_unlock(&con->mutex);
1286 pr_info("reset on %s%lld\n", ENTITY_NAME(con->peer_name));
1287 if (con->ops->peer_reset)
1288 con->ops->peer_reset(con);
1289 mutex_lock(&con->mutex);
1290 if (test_bit(CLOSED, &con->state) ||
1291 test_bit(OPENING, &con->state))
1292 return -EAGAIN;
1293 break;
1294
1295 case CEPH_MSGR_TAG_RETRY_SESSION:
1296 /*
1297 * If we sent a smaller connect_seq than the peer has, try
1298 * again with a larger value.
1299 */
1300 dout("process_connect got RETRY my seq = %u, peer_seq = %u\n",
1301 le32_to_cpu(con->out_connect.connect_seq),
1302 le32_to_cpu(con->in_connect.connect_seq));
1303 con->connect_seq = le32_to_cpu(con->in_connect.connect_seq);
1304 prepare_write_connect(con->msgr, con, 0);
1305 prepare_read_connect(con);
1306 break;
1307
1308 case CEPH_MSGR_TAG_RETRY_GLOBAL:
1309 /*
1310 * If we sent a smaller global_seq than the peer has, try
1311 * again with a larger value.
1312 */
1313 dout("process_connect got RETRY_GLOBAL my %u peer_gseq %u\n",
1314 con->peer_global_seq,
1315 le32_to_cpu(con->in_connect.global_seq));
1316 get_global_seq(con->msgr,
1317 le32_to_cpu(con->in_connect.global_seq));
1318 prepare_write_connect(con->msgr, con, 0);
1319 prepare_read_connect(con);
1320 break;
1321
1322 case CEPH_MSGR_TAG_READY:
1323 if (req_feat & ~server_feat) {
1324 pr_err("%s%lld %s protocol feature mismatch,"
1325 " my required %llx > server's %llx, need %llx\n",
1326 ENTITY_NAME(con->peer_name),
1327 ceph_pr_addr(&con->peer_addr.in_addr),
1328 req_feat, server_feat, req_feat & ~server_feat);
1329 con->error_msg = "missing required protocol features";
1330 fail_protocol(con);
1331 return -1;
1332 }
1333 clear_bit(CONNECTING, &con->state);
1334 con->peer_global_seq = le32_to_cpu(con->in_reply.global_seq);
1335 con->connect_seq++;
1336 con->peer_features = server_feat;
1337 dout("process_connect got READY gseq %d cseq %d (%d)\n",
1338 con->peer_global_seq,
1339 le32_to_cpu(con->in_reply.connect_seq),
1340 con->connect_seq);
1341 WARN_ON(con->connect_seq !=
1342 le32_to_cpu(con->in_reply.connect_seq));
1343
1344 if (con->in_reply.flags & CEPH_MSG_CONNECT_LOSSY)
1345 set_bit(LOSSYTX, &con->state);
1346
1347 prepare_read_tag(con);
1348 break;
1349
1350 case CEPH_MSGR_TAG_WAIT:
1351 /*
1352 * If there is a connection race (we are opening
1353 * connections to each other), one of us may just have
1354 * to WAIT. This shouldn't happen if we are the
1355 * client.
1356 */
1357 pr_err("process_connect got WAIT as client\n");
1358 con->error_msg = "protocol error, got WAIT as client";
1359 return -1;
1360
1361 default:
1362 pr_err("connect protocol error, will retry\n");
1363 con->error_msg = "protocol error, garbage tag during connect";
1364 return -1;
1365 }
1366 return 0;
1367 }
1368
1369
1370 /*
1371 * read (part of) an ack
1372 */
1373 static int read_partial_ack(struct ceph_connection *con)
1374 {
1375 int to = 0;
1376
1377 return read_partial(con, &to, sizeof(con->in_temp_ack),
1378 &con->in_temp_ack);
1379 }
1380
1381
1382 /*
1383 * We can finally discard anything that's been acked.
1384 */
1385 static void process_ack(struct ceph_connection *con)
1386 {
1387 struct ceph_msg *m;
1388 u64 ack = le64_to_cpu(con->in_temp_ack);
1389 u64 seq;
1390
1391 while (!list_empty(&con->out_sent)) {
1392 m = list_first_entry(&con->out_sent, struct ceph_msg,
1393 list_head);
1394 seq = le64_to_cpu(m->hdr.seq);
1395 if (seq > ack)
1396 break;
1397 dout("got ack for seq %llu type %d at %p\n", seq,
1398 le16_to_cpu(m->hdr.type), m);
1399 m->ack_stamp = jiffies;
1400 ceph_msg_remove(m);
1401 }
1402 prepare_read_tag(con);
1403 }
1404
1405
1406
1407
1408 static int read_partial_message_section(struct ceph_connection *con,
1409 struct kvec *section,
1410 unsigned int sec_len, u32 *crc)
1411 {
1412 int ret, left;
1413
1414 BUG_ON(!section);
1415
1416 while (section->iov_len < sec_len) {
1417 BUG_ON(section->iov_base == NULL);
1418 left = sec_len - section->iov_len;
1419 ret = ceph_tcp_recvmsg(con->sock, (char *)section->iov_base +
1420 section->iov_len, left);
1421 if (ret <= 0)
1422 return ret;
1423 section->iov_len += ret;
1424 if (section->iov_len == sec_len)
1425 *crc = crc32c(0, section->iov_base,
1426 section->iov_len);
1427 }
1428
1429 return 1;
1430 }
1431
1432 static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
1433 struct ceph_msg_header *hdr,
1434 int *skip);
1435
1436
1437 static int read_partial_message_pages(struct ceph_connection *con,
1438 struct page **pages,
1439 unsigned data_len, int datacrc)
1440 {
1441 void *p;
1442 int ret;
1443 int left;
1444
1445 left = min((int)(data_len - con->in_msg_pos.data_pos),
1446 (int)(PAGE_SIZE - con->in_msg_pos.page_pos));
1447 /* (page) data */
1448 BUG_ON(pages == NULL);
1449 p = kmap(pages[con->in_msg_pos.page]);
1450 ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
1451 left);
1452 if (ret > 0 && datacrc)
1453 con->in_data_crc =
1454 crc32c(con->in_data_crc,
1455 p + con->in_msg_pos.page_pos, ret);
1456 kunmap(pages[con->in_msg_pos.page]);
1457 if (ret <= 0)
1458 return ret;
1459 con->in_msg_pos.data_pos += ret;
1460 con->in_msg_pos.page_pos += ret;
1461 if (con->in_msg_pos.page_pos == PAGE_SIZE) {
1462 con->in_msg_pos.page_pos = 0;
1463 con->in_msg_pos.page++;
1464 }
1465
1466 return ret;
1467 }
1468
1469 #ifdef CONFIG_BLOCK
1470 static int read_partial_message_bio(struct ceph_connection *con,
1471 struct bio **bio_iter, int *bio_seg,
1472 unsigned data_len, int datacrc)
1473 {
1474 struct bio_vec *bv = bio_iovec_idx(*bio_iter, *bio_seg);
1475 void *p;
1476 int ret, left;
1477
1478 if (IS_ERR(bv))
1479 return PTR_ERR(bv);
1480
1481 left = min((int)(data_len - con->in_msg_pos.data_pos),
1482 (int)(bv->bv_len - con->in_msg_pos.page_pos));
1483
1484 p = kmap(bv->bv_page) + bv->bv_offset;
1485
1486 ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
1487 left);
1488 if (ret > 0 && datacrc)
1489 con->in_data_crc =
1490 crc32c(con->in_data_crc,
1491 p + con->in_msg_pos.page_pos, ret);
1492 kunmap(bv->bv_page);
1493 if (ret <= 0)
1494 return ret;
1495 con->in_msg_pos.data_pos += ret;
1496 con->in_msg_pos.page_pos += ret;
1497 if (con->in_msg_pos.page_pos == bv->bv_len) {
1498 con->in_msg_pos.page_pos = 0;
1499 iter_bio_next(bio_iter, bio_seg);
1500 }
1501
1502 return ret;
1503 }
1504 #endif
1505
1506 /*
1507 * read (part of) a message.
1508 */
1509 static int read_partial_message(struct ceph_connection *con)
1510 {
1511 struct ceph_msg *m = con->in_msg;
1512 int ret;
1513 int to, left;
1514 unsigned front_len, middle_len, data_len;
1515 int datacrc = con->msgr->nocrc;
1516 int skip;
1517 u64 seq;
1518
1519 dout("read_partial_message con %p msg %p\n", con, m);
1520
1521 /* header */
1522 while (con->in_base_pos < sizeof(con->in_hdr)) {
1523 left = sizeof(con->in_hdr) - con->in_base_pos;
1524 ret = ceph_tcp_recvmsg(con->sock,
1525 (char *)&con->in_hdr + con->in_base_pos,
1526 left);
1527 if (ret <= 0)
1528 return ret;
1529 con->in_base_pos += ret;
1530 if (con->in_base_pos == sizeof(con->in_hdr)) {
1531 u32 crc = crc32c(0, (void *)&con->in_hdr,
1532 sizeof(con->in_hdr) - sizeof(con->in_hdr.crc));
1533 if (crc != le32_to_cpu(con->in_hdr.crc)) {
1534 pr_err("read_partial_message bad hdr "
1535 " crc %u != expected %u\n",
1536 crc, con->in_hdr.crc);
1537 return -EBADMSG;
1538 }
1539 }
1540 }
1541 front_len = le32_to_cpu(con->in_hdr.front_len);
1542 if (front_len > CEPH_MSG_MAX_FRONT_LEN)
1543 return -EIO;
1544 middle_len = le32_to_cpu(con->in_hdr.middle_len);
1545 if (middle_len > CEPH_MSG_MAX_DATA_LEN)
1546 return -EIO;
1547 data_len = le32_to_cpu(con->in_hdr.data_len);
1548 if (data_len > CEPH_MSG_MAX_DATA_LEN)
1549 return -EIO;
1550
1551 /* verify seq# */
1552 seq = le64_to_cpu(con->in_hdr.seq);
1553 if ((s64)seq - (s64)con->in_seq < 1) {
1554 pr_info("skipping %s%lld %s seq %lld expected %lld\n",
1555 ENTITY_NAME(con->peer_name),
1556 ceph_pr_addr(&con->peer_addr.in_addr),
1557 seq, con->in_seq + 1);
1558 con->in_base_pos = -front_len - middle_len - data_len -
1559 sizeof(m->footer);
1560 con->in_tag = CEPH_MSGR_TAG_READY;
1561 return 0;
1562 } else if ((s64)seq - (s64)con->in_seq > 1) {
1563 pr_err("read_partial_message bad seq %lld expected %lld\n",
1564 seq, con->in_seq + 1);
1565 con->error_msg = "bad message sequence # for incoming message";
1566 return -EBADMSG;
1567 }
1568
1569 /* allocate message? */
1570 if (!con->in_msg) {
1571 dout("got hdr type %d front %d data %d\n", con->in_hdr.type,
1572 con->in_hdr.front_len, con->in_hdr.data_len);
1573 skip = 0;
1574 con->in_msg = ceph_alloc_msg(con, &con->in_hdr, &skip);
1575 if (skip) {
1576 /* skip this message */
1577 dout("alloc_msg said skip message\n");
1578 BUG_ON(con->in_msg);
1579 con->in_base_pos = -front_len - middle_len - data_len -
1580 sizeof(m->footer);
1581 con->in_tag = CEPH_MSGR_TAG_READY;
1582 con->in_seq++;
1583 return 0;
1584 }
1585 if (!con->in_msg) {
1586 con->error_msg =
1587 "error allocating memory for incoming message";
1588 return -ENOMEM;
1589 }
1590 m = con->in_msg;
1591 m->front.iov_len = 0; /* haven't read it yet */
1592 if (m->middle)
1593 m->middle->vec.iov_len = 0;
1594
1595 con->in_msg_pos.page = 0;
1596 if (m->pages)
1597 con->in_msg_pos.page_pos = m->page_alignment;
1598 else
1599 con->in_msg_pos.page_pos = 0;
1600 con->in_msg_pos.data_pos = 0;
1601 }
1602
1603 /* front */
1604 ret = read_partial_message_section(con, &m->front, front_len,
1605 &con->in_front_crc);
1606 if (ret <= 0)
1607 return ret;
1608
1609 /* middle */
1610 if (m->middle) {
1611 ret = read_partial_message_section(con, &m->middle->vec,
1612 middle_len,
1613 &con->in_middle_crc);
1614 if (ret <= 0)
1615 return ret;
1616 }
1617 #ifdef CONFIG_BLOCK
1618 if (m->bio && !m->bio_iter)
1619 init_bio_iter(m->bio, &m->bio_iter, &m->bio_seg);
1620 #endif
1621
1622 /* (page) data */
1623 while (con->in_msg_pos.data_pos < data_len) {
1624 if (m->pages) {
1625 ret = read_partial_message_pages(con, m->pages,
1626 data_len, datacrc);
1627 if (ret <= 0)
1628 return ret;
1629 #ifdef CONFIG_BLOCK
1630 } else if (m->bio) {
1631
1632 ret = read_partial_message_bio(con,
1633 &m->bio_iter, &m->bio_seg,
1634 data_len, datacrc);
1635 if (ret <= 0)
1636 return ret;
1637 #endif
1638 } else {
1639 BUG_ON(1);
1640 }
1641 }
1642
1643 /* footer */
1644 to = sizeof(m->hdr) + sizeof(m->footer);
1645 while (con->in_base_pos < to) {
1646 left = to - con->in_base_pos;
1647 ret = ceph_tcp_recvmsg(con->sock, (char *)&m->footer +
1648 (con->in_base_pos - sizeof(m->hdr)),
1649 left);
1650 if (ret <= 0)
1651 return ret;
1652 con->in_base_pos += ret;
1653 }
1654 dout("read_partial_message got msg %p %d (%u) + %d (%u) + %d (%u)\n",
1655 m, front_len, m->footer.front_crc, middle_len,
1656 m->footer.middle_crc, data_len, m->footer.data_crc);
1657
1658 /* crc ok? */
1659 if (con->in_front_crc != le32_to_cpu(m->footer.front_crc)) {
1660 pr_err("read_partial_message %p front crc %u != exp. %u\n",
1661 m, con->in_front_crc, m->footer.front_crc);
1662 return -EBADMSG;
1663 }
1664 if (con->in_middle_crc != le32_to_cpu(m->footer.middle_crc)) {
1665 pr_err("read_partial_message %p middle crc %u != exp %u\n",
1666 m, con->in_middle_crc, m->footer.middle_crc);
1667 return -EBADMSG;
1668 }
1669 if (datacrc &&
1670 (m->footer.flags & CEPH_MSG_FOOTER_NOCRC) == 0 &&
1671 con->in_data_crc != le32_to_cpu(m->footer.data_crc)) {
1672 pr_err("read_partial_message %p data crc %u != exp. %u\n", m,
1673 con->in_data_crc, le32_to_cpu(m->footer.data_crc));
1674 return -EBADMSG;
1675 }
1676
1677 return 1; /* done! */
1678 }
1679
1680 /*
1681 * Process message. This happens in the worker thread. The callback should
1682 * be careful not to do anything that waits on other incoming messages or it
1683 * may deadlock.
1684 */
1685 static void process_message(struct ceph_connection *con)
1686 {
1687 struct ceph_msg *msg;
1688
1689 msg = con->in_msg;
1690 con->in_msg = NULL;
1691
1692 /* if first message, set peer_name */
1693 if (con->peer_name.type == 0)
1694 con->peer_name = msg->hdr.src;
1695
1696 con->in_seq++;
1697 mutex_unlock(&con->mutex);
1698
1699 dout("===== %p %llu from %s%lld %d=%s len %d+%d (%u %u %u) =====\n",
1700 msg, le64_to_cpu(msg->hdr.seq),
1701 ENTITY_NAME(msg->hdr.src),
1702 le16_to_cpu(msg->hdr.type),
1703 ceph_msg_type_name(le16_to_cpu(msg->hdr.type)),
1704 le32_to_cpu(msg->hdr.front_len),
1705 le32_to_cpu(msg->hdr.data_len),
1706 con->in_front_crc, con->in_middle_crc, con->in_data_crc);
1707 con->ops->dispatch(con, msg);
1708
1709 mutex_lock(&con->mutex);
1710 prepare_read_tag(con);
1711 }
1712
1713
1714 /*
1715 * Write something to the socket. Called in a worker thread when the
1716 * socket appears to be writeable and we have something ready to send.
1717 */
1718 static int try_write(struct ceph_connection *con)
1719 {
1720 struct ceph_messenger *msgr = con->msgr;
1721 int ret = 1;
1722
1723 dout("try_write start %p state %lu nref %d\n", con, con->state,
1724 atomic_read(&con->nref));
1725
1726 more:
1727 dout("try_write out_kvec_bytes %d\n", con->out_kvec_bytes);
1728
1729 /* open the socket first? */
1730 if (con->sock == NULL) {
1731 prepare_write_banner(msgr, con);
1732 prepare_write_connect(msgr, con, 1);
1733 prepare_read_banner(con);
1734 set_bit(CONNECTING, &con->state);
1735 clear_bit(NEGOTIATING, &con->state);
1736
1737 BUG_ON(con->in_msg);
1738 con->in_tag = CEPH_MSGR_TAG_READY;
1739 dout("try_write initiating connect on %p new state %lu\n",
1740 con, con->state);
1741 con->sock = ceph_tcp_connect(con);
1742 if (IS_ERR(con->sock)) {
1743 con->sock = NULL;
1744 con->error_msg = "connect error";
1745 ret = -1;
1746 goto out;
1747 }
1748 }
1749
1750 more_kvec:
1751 /* kvec data queued? */
1752 if (con->out_skip) {
1753 ret = write_partial_skip(con);
1754 if (ret <= 0)
1755 goto out;
1756 }
1757 if (con->out_kvec_left) {
1758 ret = write_partial_kvec(con);
1759 if (ret <= 0)
1760 goto out;
1761 }
1762
1763 /* msg pages? */
1764 if (con->out_msg) {
1765 if (con->out_msg_done) {
1766 ceph_msg_put(con->out_msg);
1767 con->out_msg = NULL; /* we're done with this one */
1768 goto do_next;
1769 }
1770
1771 ret = write_partial_msg_pages(con);
1772 if (ret == 1)
1773 goto more_kvec; /* we need to send the footer, too! */
1774 if (ret == 0)
1775 goto out;
1776 if (ret < 0) {
1777 dout("try_write write_partial_msg_pages err %d\n",
1778 ret);
1779 goto out;
1780 }
1781 }
1782
1783 do_next:
1784 if (!test_bit(CONNECTING, &con->state)) {
1785 /* is anything else pending? */
1786 if (!list_empty(&con->out_queue)) {
1787 prepare_write_message(con);
1788 goto more;
1789 }
1790 if (con->in_seq > con->in_seq_acked) {
1791 prepare_write_ack(con);
1792 goto more;
1793 }
1794 if (test_and_clear_bit(KEEPALIVE_PENDING, &con->state)) {
1795 prepare_write_keepalive(con);
1796 goto more;
1797 }
1798 }
1799
1800 /* Nothing to do! */
1801 clear_bit(WRITE_PENDING, &con->state);
1802 dout("try_write nothing else to write.\n");
1803 ret = 0;
1804 out:
1805 dout("try_write done on %p ret %d\n", con, ret);
1806 return ret;
1807 }
1808
1809
1810
1811 /*
1812 * Read what we can from the socket.
1813 */
1814 static int try_read(struct ceph_connection *con)
1815 {
1816 int ret = -1;
1817
1818 if (!con->sock)
1819 return 0;
1820
1821 if (test_bit(STANDBY, &con->state))
1822 return 0;
1823
1824 dout("try_read start on %p\n", con);
1825
1826 more:
1827 dout("try_read tag %d in_base_pos %d\n", (int)con->in_tag,
1828 con->in_base_pos);
1829
1830 /*
1831 * process_connect and process_message drop and re-take
1832 * con->mutex. make sure we handle a racing close or reopen.
1833 */
1834 if (test_bit(CLOSED, &con->state) ||
1835 test_bit(OPENING, &con->state)) {
1836 ret = -EAGAIN;
1837 goto out;
1838 }
1839
1840 if (test_bit(CONNECTING, &con->state)) {
1841 if (!test_bit(NEGOTIATING, &con->state)) {
1842 dout("try_read connecting\n");
1843 ret = read_partial_banner(con);
1844 if (ret <= 0)
1845 goto out;
1846 ret = process_banner(con);
1847 if (ret < 0)
1848 goto out;
1849 }
1850 ret = read_partial_connect(con);
1851 if (ret <= 0)
1852 goto out;
1853 ret = process_connect(con);
1854 if (ret < 0)
1855 goto out;
1856 goto more;
1857 }
1858
1859 if (con->in_base_pos < 0) {
1860 /*
1861 * skipping + discarding content.
1862 *
1863 * FIXME: there must be a better way to do this!
1864 */
1865 static char buf[1024];
1866 int skip = min(1024, -con->in_base_pos);
1867 dout("skipping %d / %d bytes\n", skip, -con->in_base_pos);
1868 ret = ceph_tcp_recvmsg(con->sock, buf, skip);
1869 if (ret <= 0)
1870 goto out;
1871 con->in_base_pos += ret;
1872 if (con->in_base_pos)
1873 goto more;
1874 }
1875 if (con->in_tag == CEPH_MSGR_TAG_READY) {
1876 /*
1877 * what's next?
1878 */
1879 ret = ceph_tcp_recvmsg(con->sock, &con->in_tag, 1);
1880 if (ret <= 0)
1881 goto out;
1882 dout("try_read got tag %d\n", (int)con->in_tag);
1883 switch (con->in_tag) {
1884 case CEPH_MSGR_TAG_MSG:
1885 prepare_read_message(con);
1886 break;
1887 case CEPH_MSGR_TAG_ACK:
1888 prepare_read_ack(con);
1889 break;
1890 case CEPH_MSGR_TAG_CLOSE:
1891 set_bit(CLOSED, &con->state); /* fixme */
1892 goto out;
1893 default:
1894 goto bad_tag;
1895 }
1896 }
1897 if (con->in_tag == CEPH_MSGR_TAG_MSG) {
1898 ret = read_partial_message(con);
1899 if (ret <= 0) {
1900 switch (ret) {
1901 case -EBADMSG:
1902 con->error_msg = "bad crc";
1903 ret = -EIO;
1904 break;
1905 case -EIO:
1906 con->error_msg = "io error";
1907 break;
1908 }
1909 goto out;
1910 }
1911 if (con->in_tag == CEPH_MSGR_TAG_READY)
1912 goto more;
1913 process_message(con);
1914 goto more;
1915 }
1916 if (con->in_tag == CEPH_MSGR_TAG_ACK) {
1917 ret = read_partial_ack(con);
1918 if (ret <= 0)
1919 goto out;
1920 process_ack(con);
1921 goto more;
1922 }
1923
1924 out:
1925 dout("try_read done on %p ret %d\n", con, ret);
1926 return ret;
1927
1928 bad_tag:
1929 pr_err("try_read bad con->in_tag = %d\n", (int)con->in_tag);
1930 con->error_msg = "protocol error, garbage tag";
1931 ret = -1;
1932 goto out;
1933 }
1934
1935
1936 /*
1937 * Atomically queue work on a connection. Bump @con reference to
1938 * avoid races with connection teardown.
1939 */
1940 static void queue_con(struct ceph_connection *con)
1941 {
1942 if (test_bit(DEAD, &con->state)) {
1943 dout("queue_con %p ignoring: DEAD\n",
1944 con);
1945 return;
1946 }
1947
1948 if (!con->ops->get(con)) {
1949 dout("queue_con %p ref count 0\n", con);
1950 return;
1951 }
1952
1953 if (!queue_delayed_work(ceph_msgr_wq, &con->work, 0)) {
1954 dout("queue_con %p - already queued\n", con);
1955 con->ops->put(con);
1956 } else {
1957 dout("queue_con %p\n", con);
1958 }
1959 }
1960
1961 /*
1962 * Do some work on a connection. Drop a connection ref when we're done.
1963 */
1964 static void con_work(struct work_struct *work)
1965 {
1966 struct ceph_connection *con = container_of(work, struct ceph_connection,
1967 work.work);
1968 int ret;
1969
1970 mutex_lock(&con->mutex);
1971 restart:
1972 if (test_and_clear_bit(BACKOFF, &con->state)) {
1973 dout("con_work %p backing off\n", con);
1974 if (queue_delayed_work(ceph_msgr_wq, &con->work,
1975 round_jiffies_relative(con->delay))) {
1976 dout("con_work %p backoff %lu\n", con, con->delay);
1977 mutex_unlock(&con->mutex);
1978 return;
1979 } else {
1980 con->ops->put(con);
1981 dout("con_work %p FAILED to back off %lu\n", con,
1982 con->delay);
1983 }
1984 }
1985
1986 if (test_bit(STANDBY, &con->state)) {
1987 dout("con_work %p STANDBY\n", con);
1988 goto done;
1989 }
1990 if (test_bit(CLOSED, &con->state)) { /* e.g. if we are replaced */
1991 dout("con_work CLOSED\n");
1992 con_close_socket(con);
1993 goto done;
1994 }
1995 if (test_and_clear_bit(OPENING, &con->state)) {
1996 /* reopen w/ new peer */
1997 dout("con_work OPENING\n");
1998 con_close_socket(con);
1999 }
2000
2001 if (test_and_clear_bit(SOCK_CLOSED, &con->state))
2002 goto fault;
2003
2004 ret = try_read(con);
2005 if (ret == -EAGAIN)
2006 goto restart;
2007 if (ret < 0)
2008 goto fault;
2009
2010 ret = try_write(con);
2011 if (ret == -EAGAIN)
2012 goto restart;
2013 if (ret < 0)
2014 goto fault;
2015
2016 done:
2017 mutex_unlock(&con->mutex);
2018 done_unlocked:
2019 con->ops->put(con);
2020 return;
2021
2022 fault:
2023 mutex_unlock(&con->mutex);
2024 ceph_fault(con); /* error/fault path */
2025 goto done_unlocked;
2026 }
2027
2028
2029 /*
2030 * Generic error/fault handler. A retry mechanism is used with
2031 * exponential backoff
2032 */
2033 static void ceph_fault(struct ceph_connection *con)
2034 {
2035 pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
2036 ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg);
2037 dout("fault %p state %lu to peer %s\n",
2038 con, con->state, ceph_pr_addr(&con->peer_addr.in_addr));
2039
2040 if (test_bit(LOSSYTX, &con->state)) {
2041 dout("fault on LOSSYTX channel\n");
2042 goto out;
2043 }
2044
2045 mutex_lock(&con->mutex);
2046 if (test_bit(CLOSED, &con->state))
2047 goto out_unlock;
2048
2049 con_close_socket(con);
2050
2051 if (con->in_msg) {
2052 ceph_msg_put(con->in_msg);
2053 con->in_msg = NULL;
2054 }
2055
2056 /* Requeue anything that hasn't been acked */
2057 list_splice_init(&con->out_sent, &con->out_queue);
2058
2059 /* If there are no messages queued or keepalive pending, place
2060 * the connection in a STANDBY state */
2061 if (list_empty(&con->out_queue) &&
2062 !test_bit(KEEPALIVE_PENDING, &con->state)) {
2063 dout("fault %p setting STANDBY clearing WRITE_PENDING\n", con);
2064 clear_bit(WRITE_PENDING, &con->state);
2065 set_bit(STANDBY, &con->state);
2066 } else {
2067 /* retry after a delay. */
2068 if (con->delay == 0)
2069 con->delay = BASE_DELAY_INTERVAL;
2070 else if (con->delay < MAX_DELAY_INTERVAL)
2071 con->delay *= 2;
2072 con->ops->get(con);
2073 if (queue_delayed_work(ceph_msgr_wq, &con->work,
2074 round_jiffies_relative(con->delay))) {
2075 dout("fault queued %p delay %lu\n", con, con->delay);
2076 } else {
2077 con->ops->put(con);
2078 dout("fault failed to queue %p delay %lu, backoff\n",
2079 con, con->delay);
2080 /*
2081 * In many cases we see a socket state change
2082 * while con_work is running and end up
2083 * queuing (non-delayed) work, such that we
2084 * can't backoff with a delay. Set a flag so
2085 * that when con_work restarts we schedule the
2086 * delay then.
2087 */
2088 set_bit(BACKOFF, &con->state);
2089 }
2090 }
2091
2092 out_unlock:
2093 mutex_unlock(&con->mutex);
2094 out:
2095 /*
2096 * in case we faulted due to authentication, invalidate our
2097 * current tickets so that we can get new ones.
2098 */
2099 if (con->auth_retry && con->ops->invalidate_authorizer) {
2100 dout("calling invalidate_authorizer()\n");
2101 con->ops->invalidate_authorizer(con);
2102 }
2103
2104 if (con->ops->fault)
2105 con->ops->fault(con);
2106 }
2107
2108
2109
2110 /*
2111 * create a new messenger instance
2112 */
2113 struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr,
2114 u32 supported_features,
2115 u32 required_features)
2116 {
2117 struct ceph_messenger *msgr;
2118
2119 msgr = kzalloc(sizeof(*msgr), GFP_KERNEL);
2120 if (msgr == NULL)
2121 return ERR_PTR(-ENOMEM);
2122
2123 msgr->supported_features = supported_features;
2124 msgr->required_features = required_features;
2125
2126 spin_lock_init(&msgr->global_seq_lock);
2127
2128 /* the zero page is needed if a request is "canceled" while the message
2129 * is being written over the socket */
2130 msgr->zero_page = __page_cache_alloc(GFP_KERNEL | __GFP_ZERO);
2131 if (!msgr->zero_page) {
2132 kfree(msgr);
2133 return ERR_PTR(-ENOMEM);
2134 }
2135 kmap(msgr->zero_page);
2136
2137 if (myaddr)
2138 msgr->inst.addr = *myaddr;
2139
2140 /* select a random nonce */
2141 msgr->inst.addr.type = 0;
2142 get_random_bytes(&msgr->inst.addr.nonce, sizeof(msgr->inst.addr.nonce));
2143 encode_my_addr(msgr);
2144
2145 dout("messenger_create %p\n", msgr);
2146 return msgr;
2147 }
2148 EXPORT_SYMBOL(ceph_messenger_create);
2149
2150 void ceph_messenger_destroy(struct ceph_messenger *msgr)
2151 {
2152 dout("destroy %p\n", msgr);
2153 kunmap(msgr->zero_page);
2154 __free_page(msgr->zero_page);
2155 kfree(msgr);
2156 dout("destroyed messenger %p\n", msgr);
2157 }
2158 EXPORT_SYMBOL(ceph_messenger_destroy);
2159
2160 static void clear_standby(struct ceph_connection *con)
2161 {
2162 /* come back from STANDBY? */
2163 if (test_and_clear_bit(STANDBY, &con->state)) {
2164 mutex_lock(&con->mutex);
2165 dout("clear_standby %p and ++connect_seq\n", con);
2166 con->connect_seq++;
2167 WARN_ON(test_bit(WRITE_PENDING, &con->state));
2168 WARN_ON(test_bit(KEEPALIVE_PENDING, &con->state));
2169 mutex_unlock(&con->mutex);
2170 }
2171 }
2172
2173 /*
2174 * Queue up an outgoing message on the given connection.
2175 */
2176 void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
2177 {
2178 if (test_bit(CLOSED, &con->state)) {
2179 dout("con_send %p closed, dropping %p\n", con, msg);
2180 ceph_msg_put(msg);
2181 return;
2182 }
2183
2184 /* set src+dst */
2185 msg->hdr.src = con->msgr->inst.name;
2186
2187 BUG_ON(msg->front.iov_len != le32_to_cpu(msg->hdr.front_len));
2188
2189 msg->needs_out_seq = true;
2190
2191 /* queue */
2192 mutex_lock(&con->mutex);
2193 BUG_ON(!list_empty(&msg->list_head));
2194 list_add_tail(&msg->list_head, &con->out_queue);
2195 dout("----- %p to %s%lld %d=%s len %d+%d+%d -----\n", msg,
2196 ENTITY_NAME(con->peer_name), le16_to_cpu(msg->hdr.type),
2197 ceph_msg_type_name(le16_to_cpu(msg->hdr.type)),
2198 le32_to_cpu(msg->hdr.front_len),
2199 le32_to_cpu(msg->hdr.middle_len),
2200 le32_to_cpu(msg->hdr.data_len));
2201 mutex_unlock(&con->mutex);
2202
2203 /* if there wasn't anything waiting to send before, queue
2204 * new work */
2205 clear_standby(con);
2206 if (test_and_set_bit(WRITE_PENDING, &con->state) == 0)
2207 queue_con(con);
2208 }
2209 EXPORT_SYMBOL(ceph_con_send);
2210
2211 /*
2212 * Revoke a message that was previously queued for send
2213 */
2214 void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg)
2215 {
2216 mutex_lock(&con->mutex);
2217 if (!list_empty(&msg->list_head)) {
2218 dout("con_revoke %p msg %p - was on queue\n", con, msg);
2219 list_del_init(&msg->list_head);
2220 ceph_msg_put(msg);
2221 msg->hdr.seq = 0;
2222 }
2223 if (con->out_msg == msg) {
2224 dout("con_revoke %p msg %p - was sending\n", con, msg);
2225 con->out_msg = NULL;
2226 if (con->out_kvec_is_msg) {
2227 con->out_skip = con->out_kvec_bytes;
2228 con->out_kvec_is_msg = false;
2229 }
2230 ceph_msg_put(msg);
2231 msg->hdr.seq = 0;
2232 }
2233 mutex_unlock(&con->mutex);
2234 }
2235
2236 /*
2237 * Revoke a message that we may be reading data into
2238 */
2239 void ceph_con_revoke_message(struct ceph_connection *con, struct ceph_msg *msg)
2240 {
2241 mutex_lock(&con->mutex);
2242 if (con->in_msg && con->in_msg == msg) {
2243 unsigned front_len = le32_to_cpu(con->in_hdr.front_len);
2244 unsigned middle_len = le32_to_cpu(con->in_hdr.middle_len);
2245 unsigned data_len = le32_to_cpu(con->in_hdr.data_len);
2246
2247 /* skip rest of message */
2248 dout("con_revoke_pages %p msg %p revoked\n", con, msg);
2249 con->in_base_pos = con->in_base_pos -
2250 sizeof(struct ceph_msg_header) -
2251 front_len -
2252 middle_len -
2253 data_len -
2254 sizeof(struct ceph_msg_footer);
2255 ceph_msg_put(con->in_msg);
2256 con->in_msg = NULL;
2257 con->in_tag = CEPH_MSGR_TAG_READY;
2258 con->in_seq++;
2259 } else {
2260 dout("con_revoke_pages %p msg %p pages %p no-op\n",
2261 con, con->in_msg, msg);
2262 }
2263 mutex_unlock(&con->mutex);
2264 }
2265
2266 /*
2267 * Queue a keepalive byte to ensure the tcp connection is alive.
2268 */
2269 void ceph_con_keepalive(struct ceph_connection *con)
2270 {
2271 dout("con_keepalive %p\n", con);
2272 clear_standby(con);
2273 if (test_and_set_bit(KEEPALIVE_PENDING, &con->state) == 0 &&
2274 test_and_set_bit(WRITE_PENDING, &con->state) == 0)
2275 queue_con(con);
2276 }
2277 EXPORT_SYMBOL(ceph_con_keepalive);
2278
2279
2280 /*
2281 * construct a new message with given type, size
2282 * the new msg has a ref count of 1.
2283 */
2284 struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags,
2285 bool can_fail)
2286 {
2287 struct ceph_msg *m;
2288
2289 m = kmalloc(sizeof(*m), flags);
2290 if (m == NULL)
2291 goto out;
2292 kref_init(&m->kref);
2293 INIT_LIST_HEAD(&m->list_head);
2294
2295 m->hdr.tid = 0;
2296 m->hdr.type = cpu_to_le16(type);
2297 m->hdr.priority = cpu_to_le16(CEPH_MSG_PRIO_DEFAULT);
2298 m->hdr.version = 0;
2299 m->hdr.front_len = cpu_to_le32(front_len);
2300 m->hdr.middle_len = 0;
2301 m->hdr.data_len = 0;
2302 m->hdr.data_off = 0;
2303 m->hdr.reserved = 0;
2304 m->footer.front_crc = 0;
2305 m->footer.middle_crc = 0;
2306 m->footer.data_crc = 0;
2307 m->footer.flags = 0;
2308 m->front_max = front_len;
2309 m->front_is_vmalloc = false;
2310 m->more_to_follow = false;
2311 m->pool = NULL;
2312
2313 /* middle */
2314 m->middle = NULL;
2315
2316 /* data */
2317 m->nr_pages = 0;
2318 m->page_alignment = 0;
2319 m->pages = NULL;
2320 m->pagelist = NULL;
2321 m->bio = NULL;
2322 m->bio_iter = NULL;
2323 m->bio_seg = 0;
2324 m->trail = NULL;
2325
2326 /* front */
2327 if (front_len) {
2328 if (front_len > PAGE_CACHE_SIZE) {
2329 m->front.iov_base = __vmalloc(front_len, flags,
2330 PAGE_KERNEL);
2331 m->front_is_vmalloc = true;
2332 } else {
2333 m->front.iov_base = kmalloc(front_len, flags);
2334 }
2335 if (m->front.iov_base == NULL) {
2336 dout("ceph_msg_new can't allocate %d bytes\n",
2337 front_len);
2338 goto out2;
2339 }
2340 } else {
2341 m->front.iov_base = NULL;
2342 }
2343 m->front.iov_len = front_len;
2344
2345 dout("ceph_msg_new %p front %d\n", m, front_len);
2346 return m;
2347
2348 out2:
2349 ceph_msg_put(m);
2350 out:
2351 if (!can_fail) {
2352 pr_err("msg_new can't create type %d front %d\n", type,
2353 front_len);
2354 WARN_ON(1);
2355 } else {
2356 dout("msg_new can't create type %d front %d\n", type,
2357 front_len);
2358 }
2359 return NULL;
2360 }
2361 EXPORT_SYMBOL(ceph_msg_new);
2362
2363 /*
2364 * Allocate "middle" portion of a message, if it is needed and wasn't
2365 * allocated by alloc_msg. This allows us to read a small fixed-size
2366 * per-type header in the front and then gracefully fail (i.e.,
2367 * propagate the error to the caller based on info in the front) when
2368 * the middle is too large.
2369 */
2370 static int ceph_alloc_middle(struct ceph_connection *con, struct ceph_msg *msg)
2371 {
2372 int type = le16_to_cpu(msg->hdr.type);
2373 int middle_len = le32_to_cpu(msg->hdr.middle_len);
2374
2375 dout("alloc_middle %p type %d %s middle_len %d\n", msg, type,
2376 ceph_msg_type_name(type), middle_len);
2377 BUG_ON(!middle_len);
2378 BUG_ON(msg->middle);
2379
2380 msg->middle = ceph_buffer_new(middle_len, GFP_NOFS);
2381 if (!msg->middle)
2382 return -ENOMEM;
2383 return 0;
2384 }
2385
2386 /*
2387 * Generic message allocator, for incoming messages.
2388 */
2389 static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
2390 struct ceph_msg_header *hdr,
2391 int *skip)
2392 {
2393 int type = le16_to_cpu(hdr->type);
2394 int front_len = le32_to_cpu(hdr->front_len);
2395 int middle_len = le32_to_cpu(hdr->middle_len);
2396 struct ceph_msg *msg = NULL;
2397 int ret;
2398
2399 if (con->ops->alloc_msg) {
2400 mutex_unlock(&con->mutex);
2401 msg = con->ops->alloc_msg(con, hdr, skip);
2402 mutex_lock(&con->mutex);
2403 if (!msg || *skip)
2404 return NULL;
2405 }
2406 if (!msg) {
2407 *skip = 0;
2408 msg = ceph_msg_new(type, front_len, GFP_NOFS, false);
2409 if (!msg) {
2410 pr_err("unable to allocate msg type %d len %d\n",
2411 type, front_len);
2412 return NULL;
2413 }
2414 msg->page_alignment = le16_to_cpu(hdr->data_off);
2415 }
2416 memcpy(&msg->hdr, &con->in_hdr, sizeof(con->in_hdr));
2417
2418 if (middle_len && !msg->middle) {
2419 ret = ceph_alloc_middle(con, msg);
2420 if (ret < 0) {
2421 ceph_msg_put(msg);
2422 return NULL;
2423 }
2424 }
2425
2426 return msg;
2427 }
2428
2429
2430 /*
2431 * Free a generically kmalloc'd message.
2432 */
2433 void ceph_msg_kfree(struct ceph_msg *m)
2434 {
2435 dout("msg_kfree %p\n", m);
2436 if (m->front_is_vmalloc)
2437 vfree(m->front.iov_base);
2438 else
2439 kfree(m->front.iov_base);
2440 kfree(m);
2441 }
2442
2443 /*
2444 * Drop a msg ref. Destroy as needed.
2445 */
2446 void ceph_msg_last_put(struct kref *kref)
2447 {
2448 struct ceph_msg *m = container_of(kref, struct ceph_msg, kref);
2449
2450 dout("ceph_msg_put last one on %p\n", m);
2451 WARN_ON(!list_empty(&m->list_head));
2452
2453 /* drop middle, data, if any */
2454 if (m->middle) {
2455 ceph_buffer_put(m->middle);
2456 m->middle = NULL;
2457 }
2458 m->nr_pages = 0;
2459 m->pages = NULL;
2460
2461 if (m->pagelist) {
2462 ceph_pagelist_release(m->pagelist);
2463 kfree(m->pagelist);
2464 m->pagelist = NULL;
2465 }
2466
2467 m->trail = NULL;
2468
2469 if (m->pool)
2470 ceph_msgpool_put(m->pool, m);
2471 else
2472 ceph_msg_kfree(m);
2473 }
2474 EXPORT_SYMBOL(ceph_msg_last_put);
2475
2476 void ceph_msg_dump(struct ceph_msg *msg)
2477 {
2478 pr_debug("msg_dump %p (front_max %d nr_pages %d)\n", msg,
2479 msg->front_max, msg->nr_pages);
2480 print_hex_dump(KERN_DEBUG, "header: ",
2481 DUMP_PREFIX_OFFSET, 16, 1,
2482 &msg->hdr, sizeof(msg->hdr), true);
2483 print_hex_dump(KERN_DEBUG, " front: ",
2484 DUMP_PREFIX_OFFSET, 16, 1,
2485 msg->front.iov_base, msg->front.iov_len, true);
2486 if (msg->middle)
2487 print_hex_dump(KERN_DEBUG, "middle: ",
2488 DUMP_PREFIX_OFFSET, 16, 1,
2489 msg->middle->vec.iov_base,
2490 msg->middle->vec.iov_len, true);
2491 print_hex_dump(KERN_DEBUG, "footer: ",
2492 DUMP_PREFIX_OFFSET, 16, 1,
2493 &msg->footer, sizeof(msg->footer), true);
2494 }
2495 EXPORT_SYMBOL(ceph_msg_dump);
linux-next