1 // SPDX-License-Identifier: GPL-2.0-or-later
2 /* AF_RXRPC sendmsg() implementation.
4 * Copyright (C) 2007, 2016 Red Hat, Inc. All Rights Reserved.
5 * Written by David Howells (dhowells@redhat.com)
8 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
10 #include <linux/net.h>
11 #include <linux/gfp.h>
12 #include <linux/skbuff.h>
13 #include <linux/export.h>
14 #include <linux/sched/signal.h>
17 #include <net/af_rxrpc.h>
18 #include "ar-internal.h"
21 * Return true if there's sufficient Tx queue space.
23 static bool rxrpc_check_tx_space(struct rxrpc_call *call, rxrpc_seq_t *_tx_win)
26 *_tx_win = call->tx_bottom;
27 return call->tx_prepared - call->tx_bottom < 256;
31 * Wait for space to appear in the Tx queue or a signal to occur.
33 static int rxrpc_wait_for_tx_window_intr(struct rxrpc_sock *rx,
34 struct rxrpc_call *call,
38 set_current_state(TASK_INTERRUPTIBLE);
39 if (rxrpc_check_tx_space(call, NULL))
42 if (call->state >= RXRPC_CALL_COMPLETE)
45 if (signal_pending(current))
46 return sock_intr_errno(*timeo);
48 trace_rxrpc_txqueue(call, rxrpc_txqueue_wait);
49 *timeo = schedule_timeout(*timeo);
54 * Wait for space to appear in the Tx queue uninterruptibly, but with
55 * a timeout of 2*RTT if no progress was made and a signal occurred.
57 static int rxrpc_wait_for_tx_window_waitall(struct rxrpc_sock *rx,
58 struct rxrpc_call *call)
60 rxrpc_seq_t tx_start, tx_win;
61 signed long rtt, timeout;
63 rtt = READ_ONCE(call->peer->srtt_us) >> 3;
64 rtt = usecs_to_jiffies(rtt) * 2;
69 tx_start = smp_load_acquire(&call->acks_hard_ack);
72 set_current_state(TASK_UNINTERRUPTIBLE);
74 if (rxrpc_check_tx_space(call, &tx_win))
77 if (call->state >= RXRPC_CALL_COMPLETE)
81 tx_win == tx_start && signal_pending(current))
84 if (tx_win != tx_start) {
89 trace_rxrpc_txqueue(call, rxrpc_txqueue_wait);
90 timeout = schedule_timeout(timeout);
95 * Wait for space to appear in the Tx queue uninterruptibly.
97 static int rxrpc_wait_for_tx_window_nonintr(struct rxrpc_sock *rx,
98 struct rxrpc_call *call,
102 set_current_state(TASK_UNINTERRUPTIBLE);
103 if (rxrpc_check_tx_space(call, NULL))
106 if (call->state >= RXRPC_CALL_COMPLETE)
109 trace_rxrpc_txqueue(call, rxrpc_txqueue_wait);
110 *timeo = schedule_timeout(*timeo);
115 * wait for space to appear in the transmit/ACK window
116 * - caller holds the socket locked
118 static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx,
119 struct rxrpc_call *call,
123 DECLARE_WAITQUEUE(myself, current);
126 _enter(",{%u,%u,%u,%u}",
127 call->tx_bottom, call->acks_hard_ack, call->tx_top, call->tx_winsize);
129 add_wait_queue(&call->waitq, &myself);
131 switch (call->interruptibility) {
132 case RXRPC_INTERRUPTIBLE:
134 ret = rxrpc_wait_for_tx_window_waitall(rx, call);
136 ret = rxrpc_wait_for_tx_window_intr(rx, call, timeo);
138 case RXRPC_PREINTERRUPTIBLE:
139 case RXRPC_UNINTERRUPTIBLE:
141 ret = rxrpc_wait_for_tx_window_nonintr(rx, call, timeo);
145 remove_wait_queue(&call->waitq, &myself);
146 set_current_state(TASK_RUNNING);
147 _leave(" = %d", ret);
152 * Notify the owner of the call that the transmit phase is ended and the last
153 * packet has been queued.
155 static void rxrpc_notify_end_tx(struct rxrpc_sock *rx, struct rxrpc_call *call,
156 rxrpc_notify_end_tx_t notify_end_tx)
159 notify_end_tx(&rx->sk, call, call->user_call_ID);
163 * Queue a DATA packet for transmission, set the resend timeout and send
164 * the packet immediately. Returns the error from rxrpc_send_data_packet()
165 * in case the caller wants to do something with it.
167 static void rxrpc_queue_packet(struct rxrpc_sock *rx, struct rxrpc_call *call,
168 struct rxrpc_txbuf *txb,
169 rxrpc_notify_end_tx_t notify_end_tx)
172 rxrpc_seq_t seq = txb->seq;
173 bool last = test_bit(RXRPC_TXBUF_LAST, &txb->flags), poke;
175 rxrpc_inc_stat(call->rxnet, stat_tx_data);
177 ASSERTCMP(txb->seq, ==, call->tx_prepared + 1);
179 /* We have to set the timestamp before queueing as the retransmit
180 * algorithm can see the packet as soon as we queue it.
182 txb->last_sent = ktime_get_real();
185 trace_rxrpc_txqueue(call, rxrpc_txqueue_queue_last);
187 trace_rxrpc_txqueue(call, rxrpc_txqueue_queue);
189 /* Add the packet to the call's output buffer */
190 spin_lock(&call->tx_lock);
191 poke = list_empty(&call->tx_sendmsg);
192 list_add_tail(&txb->call_link, &call->tx_sendmsg);
193 call->tx_prepared = seq;
194 spin_unlock(&call->tx_lock);
196 if (last || call->state == RXRPC_CALL_SERVER_ACK_REQUEST) {
197 _debug("________awaiting reply/ACK__________");
198 write_lock(&call->state_lock);
199 switch (call->state) {
200 case RXRPC_CALL_CLIENT_SEND_REQUEST:
201 call->state = RXRPC_CALL_CLIENT_AWAIT_REPLY;
202 rxrpc_notify_end_tx(rx, call, notify_end_tx);
204 case RXRPC_CALL_SERVER_ACK_REQUEST:
205 call->state = RXRPC_CALL_SERVER_SEND_REPLY;
207 WRITE_ONCE(call->delay_ack_at, now + MAX_JIFFY_OFFSET);
208 if (call->ackr_reason == RXRPC_ACK_DELAY)
209 call->ackr_reason = 0;
210 trace_rxrpc_timer(call, rxrpc_timer_init_for_send_reply, now);
214 case RXRPC_CALL_SERVER_SEND_REPLY:
215 call->state = RXRPC_CALL_SERVER_AWAIT_ACK;
216 rxrpc_notify_end_tx(rx, call, notify_end_tx);
221 write_unlock(&call->state_lock);
225 rxrpc_poke_call(call, rxrpc_call_poke_start);
229 * send data through a socket
230 * - must be called in process context
231 * - The caller holds the call user access mutex, but not the socket lock.
233 static int rxrpc_send_data(struct rxrpc_sock *rx,
234 struct rxrpc_call *call,
235 struct msghdr *msg, size_t len,
236 rxrpc_notify_end_tx_t notify_end_tx,
239 struct rxrpc_txbuf *txb;
240 struct sock *sk = &rx->sk;
241 enum rxrpc_call_state state;
243 bool more = msg->msg_flags & MSG_MORE;
246 timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
248 /* this should be in poll */
249 sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk);
253 if (sk->sk_shutdown & SEND_SHUTDOWN)
255 state = READ_ONCE(call->state);
257 if (state >= RXRPC_CALL_COMPLETE)
260 if (state != RXRPC_CALL_CLIENT_SEND_REQUEST &&
261 state != RXRPC_CALL_SERVER_ACK_REQUEST &&
262 state != RXRPC_CALL_SERVER_SEND_REPLY)
266 if (call->tx_total_len != -1) {
267 if (len - copied > call->tx_total_len)
269 if (!more && len - copied != call->tx_total_len)
273 txb = call->tx_pending;
274 call->tx_pending = NULL;
276 rxrpc_see_txbuf(txb, rxrpc_txbuf_see_send_more);
280 size_t remain, bufsize, chunk, offset;
284 if (!rxrpc_check_tx_space(call, NULL))
287 /* Work out the maximum size of a packet. Assume that
288 * the security header is going to be in the padded
289 * region (enc blocksize), but the trailer is not.
291 remain = more ? INT_MAX : msg_data_left(msg);
292 ret = call->conn->security->how_much_data(call, remain,
293 &bufsize, &chunk, &offset);
297 _debug("SIZE: %zu/%zu @%zu", chunk, bufsize, offset);
299 /* create a buffer that we can retain until it's ACK'd */
301 txb = rxrpc_alloc_txbuf(call, RXRPC_PACKET_TYPE_DATA,
306 txb->offset = offset;
307 txb->space -= offset;
308 txb->space = min_t(size_t, chunk, txb->space);
313 /* append next segment of data to the current buffer */
314 if (msg_data_left(msg) > 0) {
315 size_t copy = min_t(size_t, txb->space, msg_data_left(msg));
317 _debug("add %zu", copy);
318 if (!copy_from_iter_full(txb->data + txb->offset, copy,
326 if (call->tx_total_len != -1)
327 call->tx_total_len -= copy;
330 /* check for the far side aborting the call or a network error
332 if (call->state == RXRPC_CALL_COMPLETE)
333 goto call_terminated;
335 /* add the packet to the send queue if it's now full */
337 (msg_data_left(msg) == 0 && !more)) {
338 if (msg_data_left(msg) == 0 && !more) {
339 txb->wire.flags |= RXRPC_LAST_PACKET;
340 __set_bit(RXRPC_TXBUF_LAST, &txb->flags);
342 else if (call->tx_top - call->acks_hard_ack <
344 txb->wire.flags |= RXRPC_MORE_PACKETS;
346 ret = call->security->secure_packet(call, txb);
350 rxrpc_queue_packet(rx, call, txb, notify_end_tx);
353 } while (msg_data_left(msg) > 0);
357 if (READ_ONCE(call->state) == RXRPC_CALL_COMPLETE) {
358 read_lock(&call->state_lock);
361 read_unlock(&call->state_lock);
364 call->tx_pending = txb;
365 _leave(" = %d", ret);
369 rxrpc_put_txbuf(txb, rxrpc_txbuf_put_send_aborted);
370 _leave(" = %d", call->error);
384 if (msg->msg_flags & MSG_DONTWAIT)
386 mutex_unlock(&call->user_mutex);
387 *_dropped_lock = true;
388 ret = rxrpc_wait_for_tx_window(rx, call, &timeo,
389 msg->msg_flags & MSG_WAITALL);
392 if (call->interruptibility == RXRPC_INTERRUPTIBLE) {
393 if (mutex_lock_interruptible(&call->user_mutex) < 0) {
394 ret = sock_intr_errno(timeo);
398 mutex_lock(&call->user_mutex);
400 *_dropped_lock = false;
405 * extract control messages from the sendmsg() control buffer
407 static int rxrpc_sendmsg_cmsg(struct msghdr *msg, struct rxrpc_send_params *p)
409 struct cmsghdr *cmsg;
410 bool got_user_ID = false;
413 if (msg->msg_controllen == 0)
416 for_each_cmsghdr(cmsg, msg) {
417 if (!CMSG_OK(msg, cmsg))
420 len = cmsg->cmsg_len - sizeof(struct cmsghdr);
421 _debug("CMSG %d, %d, %d",
422 cmsg->cmsg_level, cmsg->cmsg_type, len);
424 if (cmsg->cmsg_level != SOL_RXRPC)
427 switch (cmsg->cmsg_type) {
428 case RXRPC_USER_CALL_ID:
429 if (msg->msg_flags & MSG_CMSG_COMPAT) {
430 if (len != sizeof(u32))
432 p->call.user_call_ID = *(u32 *)CMSG_DATA(cmsg);
434 if (len != sizeof(unsigned long))
436 p->call.user_call_ID = *(unsigned long *)
443 if (p->command != RXRPC_CMD_SEND_DATA)
445 p->command = RXRPC_CMD_SEND_ABORT;
446 if (len != sizeof(p->abort_code))
448 p->abort_code = *(unsigned int *)CMSG_DATA(cmsg);
449 if (p->abort_code == 0)
453 case RXRPC_CHARGE_ACCEPT:
454 if (p->command != RXRPC_CMD_SEND_DATA)
456 p->command = RXRPC_CMD_CHARGE_ACCEPT;
461 case RXRPC_EXCLUSIVE_CALL:
467 case RXRPC_UPGRADE_SERVICE:
473 case RXRPC_TX_LENGTH:
474 if (p->call.tx_total_len != -1 || len != sizeof(__s64))
476 p->call.tx_total_len = *(__s64 *)CMSG_DATA(cmsg);
477 if (p->call.tx_total_len < 0)
481 case RXRPC_SET_CALL_TIMEOUT:
482 if (len & 3 || len < 4 || len > 12)
484 memcpy(&p->call.timeouts, CMSG_DATA(cmsg), len);
485 p->call.nr_timeouts = len / 4;
486 if (p->call.timeouts.hard > INT_MAX / HZ)
488 if (p->call.nr_timeouts >= 2 && p->call.timeouts.idle > 60 * 60 * 1000)
490 if (p->call.nr_timeouts >= 3 && p->call.timeouts.normal > 60 * 60 * 1000)
501 if (p->call.tx_total_len != -1 && p->command != RXRPC_CMD_SEND_DATA)
508 * Create a new client call for sendmsg().
509 * - Called with the socket lock held, which it must release.
510 * - If it returns a call, the call's lock will need releasing by the caller.
512 static struct rxrpc_call *
513 rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
514 struct rxrpc_send_params *p)
515 __releases(&rx->sk.sk_lock.slock)
516 __acquires(&call->user_mutex)
518 struct rxrpc_conn_parameters cp;
519 struct rxrpc_call *call;
522 DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, msg->msg_name);
526 if (!msg->msg_name) {
527 release_sock(&rx->sk);
528 return ERR_PTR(-EDESTADDRREQ);
532 if (key && !rx->key->payload.data[0])
535 memset(&cp, 0, sizeof(cp));
536 cp.local = rx->local;
538 cp.security_level = rx->min_sec_level;
539 cp.exclusive = rx->exclusive | p->exclusive;
540 cp.upgrade = p->upgrade;
541 cp.service_id = srx->srx_service;
542 call = rxrpc_new_client_call(rx, &cp, srx, &p->call, GFP_KERNEL,
543 atomic_inc_return(&rxrpc_debug_id));
544 /* The socket is now unlocked */
546 rxrpc_put_peer(cp.peer, rxrpc_peer_put_discard_tmp);
547 _leave(" = %p\n", call);
552 * send a message forming part of a client call through an RxRPC socket
553 * - caller holds the socket locked
554 * - the socket may be either a client socket or a server socket
556 int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len)
557 __releases(&rx->sk.sk_lock.slock)
559 enum rxrpc_call_state state;
560 struct rxrpc_call *call;
561 unsigned long now, j;
562 bool dropped_lock = false;
565 struct rxrpc_send_params p = {
566 .call.tx_total_len = -1,
567 .call.user_call_ID = 0,
568 .call.nr_timeouts = 0,
569 .call.interruptibility = RXRPC_INTERRUPTIBLE,
571 .command = RXRPC_CMD_SEND_DATA,
578 ret = rxrpc_sendmsg_cmsg(msg, &p);
580 goto error_release_sock;
582 if (p.command == RXRPC_CMD_CHARGE_ACCEPT) {
584 if (rx->sk.sk_state != RXRPC_SERVER_LISTENING)
585 goto error_release_sock;
586 ret = rxrpc_user_charge_accept(rx, p.call.user_call_ID);
587 goto error_release_sock;
590 call = rxrpc_find_call_by_user_ID(rx, p.call.user_call_ID);
593 if (p.command != RXRPC_CMD_SEND_DATA)
594 goto error_release_sock;
595 call = rxrpc_new_client_call_for_sendmsg(rx, msg, &p);
596 /* The socket is now unlocked... */
598 return PTR_ERR(call);
599 /* ... and we have the call lock. */
601 if (READ_ONCE(call->state) == RXRPC_CALL_COMPLETE)
604 switch (READ_ONCE(call->state)) {
605 case RXRPC_CALL_UNINITIALISED:
606 case RXRPC_CALL_CLIENT_AWAIT_CONN:
607 case RXRPC_CALL_SERVER_PREALLOC:
608 case RXRPC_CALL_SERVER_SECURING:
609 rxrpc_put_call(call, rxrpc_call_put_sendmsg);
611 goto error_release_sock;
616 ret = mutex_lock_interruptible(&call->user_mutex);
617 release_sock(&rx->sk);
623 if (p.call.tx_total_len != -1) {
625 if (call->tx_total_len != -1 ||
629 call->tx_total_len = p.call.tx_total_len;
633 switch (p.call.nr_timeouts) {
635 j = msecs_to_jiffies(p.call.timeouts.normal);
636 if (p.call.timeouts.normal > 0 && j == 0)
638 WRITE_ONCE(call->next_rx_timo, j);
641 j = msecs_to_jiffies(p.call.timeouts.idle);
642 if (p.call.timeouts.idle > 0 && j == 0)
644 WRITE_ONCE(call->next_req_timo, j);
647 if (p.call.timeouts.hard > 0) {
648 j = msecs_to_jiffies(p.call.timeouts.hard);
651 WRITE_ONCE(call->expect_term_by, j);
652 rxrpc_reduce_call_timer(call, j, now,
653 rxrpc_timer_set_for_hard);
658 state = READ_ONCE(call->state);
659 _debug("CALL %d USR %lx ST %d on CONN %p",
660 call->debug_id, call->user_call_ID, state, call->conn);
662 if (state >= RXRPC_CALL_COMPLETE) {
663 /* it's too late for this call */
665 } else if (p.command == RXRPC_CMD_SEND_ABORT) {
667 if (rxrpc_abort_call("CMD", call, 0, p.abort_code, -ECONNABORTED))
668 ret = rxrpc_send_abort_packet(call);
669 } else if (p.command != RXRPC_CMD_SEND_DATA) {
672 ret = rxrpc_send_data(rx, call, msg, len, NULL, &dropped_lock);
677 mutex_unlock(&call->user_mutex);
679 rxrpc_put_call(call, rxrpc_call_put_sendmsg);
680 _leave(" = %d", ret);
684 release_sock(&rx->sk);
689 * rxrpc_kernel_send_data - Allow a kernel service to send data on a call
690 * @sock: The socket the call is on
691 * @call: The call to send data through
692 * @msg: The data to send
693 * @len: The amount of data to send
694 * @notify_end_tx: Notification that the last packet is queued.
696 * Allow a kernel service to send data on a call. The call must be in an state
697 * appropriate to sending data. No control data should be supplied in @msg,
698 * nor should an address be supplied. MSG_MORE should be flagged if there's
699 * more data to come, otherwise this data will end the transmission phase.
701 int rxrpc_kernel_send_data(struct socket *sock, struct rxrpc_call *call,
702 struct msghdr *msg, size_t len,
703 rxrpc_notify_end_tx_t notify_end_tx)
705 bool dropped_lock = false;
708 _enter("{%d,%s},", call->debug_id, rxrpc_call_states[call->state]);
710 ASSERTCMP(msg->msg_name, ==, NULL);
711 ASSERTCMP(msg->msg_control, ==, NULL);
713 mutex_lock(&call->user_mutex);
715 _debug("CALL %d USR %lx ST %d on CONN %p",
716 call->debug_id, call->user_call_ID, call->state, call->conn);
718 switch (READ_ONCE(call->state)) {
719 case RXRPC_CALL_CLIENT_SEND_REQUEST:
720 case RXRPC_CALL_SERVER_ACK_REQUEST:
721 case RXRPC_CALL_SERVER_SEND_REPLY:
722 ret = rxrpc_send_data(rxrpc_sk(sock->sk), call, msg, len,
723 notify_end_tx, &dropped_lock);
725 case RXRPC_CALL_COMPLETE:
726 read_lock(&call->state_lock);
728 read_unlock(&call->state_lock);
731 /* Request phase complete for this client call */
732 trace_rxrpc_rx_eproto(call, 0, tracepoint_string("late_send"));
738 mutex_unlock(&call->user_mutex);
739 _leave(" = %d", ret);
742 EXPORT_SYMBOL(rxrpc_kernel_send_data);
745 * rxrpc_kernel_abort_call - Allow a kernel service to abort a call
746 * @sock: The socket the call is on
747 * @call: The call to be aborted
748 * @abort_code: The abort code to stick into the ABORT packet
749 * @error: Local error value
750 * @why: 3-char string indicating why.
752 * Allow a kernel service to abort a call, if it's still in an abortable state
753 * and return true if the call was aborted, false if it was already complete.
755 bool rxrpc_kernel_abort_call(struct socket *sock, struct rxrpc_call *call,
756 u32 abort_code, int error, const char *why)
760 _enter("{%d},%d,%d,%s", call->debug_id, abort_code, error, why);
762 mutex_lock(&call->user_mutex);
764 aborted = rxrpc_abort_call(why, call, 0, abort_code, error);
766 rxrpc_send_abort_packet(call);
768 mutex_unlock(&call->user_mutex);
771 EXPORT_SYMBOL(rxrpc_kernel_abort_call);
774 * rxrpc_kernel_set_tx_length - Set the total Tx length on a call
775 * @sock: The socket the call is on
776 * @call: The call to be informed
777 * @tx_total_len: The amount of data to be transmitted for this call
779 * Allow a kernel service to set the total transmit length on a call. This
780 * allows buffer-to-packet encrypt-and-copy to be performed.
782 * This function is primarily for use for setting the reply length since the
783 * request length can be set when beginning the call.
785 void rxrpc_kernel_set_tx_length(struct socket *sock, struct rxrpc_call *call,
788 WARN_ON(call->tx_total_len != -1);
789 call->tx_total_len = tx_total_len;
791 EXPORT_SYMBOL(rxrpc_kernel_set_tx_length);