net/rds/send.c

   1 /*
   2  * Copyright (c) 2006, 2018 Oracle and/or its affiliates. All rights reserved.
   3  *
   4  * This software is available to you under a choice of one of two
   5  * licenses.  You may choose to be licensed under the terms of the GNU
   6  * General Public License (GPL) Version 2, available from the file
   7  * COPYING in the main directory of this source tree, or the
   8  * OpenIB.org BSD license below:
   9  *
  10  *     Redistribution and use in source and binary forms, with or
  11  *     without modification, are permitted provided that the following
  12  *     conditions are met:
  13  *
  14  *      - Redistributions of source code must retain the above
  15  *        copyright notice, this list of conditions and the following
  16  *        disclaimer.
  17  *
  18  *      - Redistributions in binary form must reproduce the above
  19  *        copyright notice, this list of conditions and the following
  20  *        disclaimer in the documentation and/or other materials
  21  *        provided with the distribution.
  22  *
  23  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  24  * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
  25  * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
  26  * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
  27  * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
  28  * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
  29  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  30  * SOFTWARE.
  31  *
  32  */
  33 #include <linux/kernel.h>
  34 #include <linux/moduleparam.h>
  35 #include <linux/gfp.h>
  36 #include <net/sock.h>
  37 #include <linux/in.h>
  38 #include <linux/list.h>
  39 #include <linux/ratelimit.h>
  40 #include <linux/export.h>
  41 #include <linux/sizes.h>
  42
  43 #include "rds.h"
  44
  45 /* When transmitting messages in rds_send_xmit, we need to emerge from
  46  * time to time and briefly release the CPU. Otherwise the softlock watchdog
  47  * will kick our shin.
  48  * Also, it seems fairer to not let one busy connection stall all the
  49  * others.
  50  *
  51  * send_batch_count is the number of times we'll loop in send_xmit. Setting
  52  * it to 0 will restore the old behavior (where we looped until we had
  53  * drained the queue).
  54  */
  55 static int send_batch_count = SZ_1K;
  56 module_param(send_batch_count, int, 0444);
  57 MODULE_PARM_DESC(send_batch_count, " batch factor when working the send queue");
  58
  59 static void rds_send_remove_from_sock(struct list_head *messages, int status);
  60
  61 /*
  62  * Reset the send state.  Callers must ensure that this doesn't race with
  63  * rds_send_xmit().
  64  */
  65 void rds_send_path_reset(struct rds_conn_path *cp)
  66 {
  67         struct rds_message *rm, *tmp;
  68         unsigned long flags;
  69
  70         if (cp->cp_xmit_rm) {
  71                 rm = cp->cp_xmit_rm;
  72                 cp->cp_xmit_rm = NULL;
  73                 /* Tell the user the RDMA op is no longer mapped by the
  74                  * transport. This isn't entirely true (it's flushed out
  75                  * independently) but as the connection is down, there's
  76                  * no ongoing RDMA to/from that memory */
  77                 rds_message_unmapped(rm);
  78                 rds_message_put(rm);
  79         }
  80
  81         cp->cp_xmit_sg = 0;
  82         cp->cp_xmit_hdr_off = 0;
  83         cp->cp_xmit_data_off = 0;
  84         cp->cp_xmit_atomic_sent = 0;
  85         cp->cp_xmit_rdma_sent = 0;
  86         cp->cp_xmit_data_sent = 0;
  87
  88         cp->cp_conn->c_map_queued = 0;
  89
  90         cp->cp_unacked_packets = rds_sysctl_max_unacked_packets;
  91         cp->cp_unacked_bytes = rds_sysctl_max_unacked_bytes;
  92
  93         /* Mark messages as retransmissions, and move them to the send q */
  94         spin_lock_irqsave(&cp->cp_lock, flags);
  95         list_for_each_entry_safe(rm, tmp, &cp->cp_retrans, m_conn_item) {
  96                 set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags);
  97                 set_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags);
  98         }
  99         list_splice_init(&cp->cp_retrans, &cp->cp_send_queue);
 100         spin_unlock_irqrestore(&cp->cp_lock, flags);
 101 }
 102 EXPORT_SYMBOL_GPL(rds_send_path_reset);
 103
 104 static int acquire_in_xmit(struct rds_conn_path *cp)
 105 {
 106         return test_and_set_bit(RDS_IN_XMIT, &cp->cp_flags) == 0;
 107 }
 108
 109 static void release_in_xmit(struct rds_conn_path *cp)
 110 {
 111         clear_bit(RDS_IN_XMIT, &cp->cp_flags);
 112         smp_mb__after_atomic();
 113         /*
 114          * We don't use wait_on_bit()/wake_up_bit() because our waking is in a
 115          * hot path and finding waiters is very rare.  We don't want to walk
 116          * the system-wide hashed waitqueue buckets in the fast path only to
 117          * almost never find waiters.
 118          */
 119         if (waitqueue_active(&cp->cp_waitq))
 120                 wake_up_all(&cp->cp_waitq);
 121 }
 122
 123 /*
 124  * We're making the conscious trade-off here to only send one message
 125  * down the connection at a time.
 126  *   Pro:
 127  *      - tx queueing is a simple fifo list
 128  *      - reassembly is optional and easily done by transports per conn
 129  *      - no per flow rx lookup at all, straight to the socket
 130  *      - less per-frag memory and wire overhead
 131  *   Con:
 132  *      - queued acks can be delayed behind large messages
 133  *   Depends:
 134  *      - small message latency is higher behind queued large messages
 135  *      - large message latency isn't starved by intervening small sends
 136  */
 137 int rds_send_xmit(struct rds_conn_path *cp)
 138 {
 139         struct rds_connection *conn = cp->cp_conn;
 140         struct rds_message *rm;
 141         unsigned long flags;
 142         unsigned int tmp;
 143         struct scatterlist *sg;
 144         int ret = 0;
 145         LIST_HEAD(to_be_dropped);
 146         int batch_count;
 147         unsigned long send_gen = 0;
 148
 149 restart:
 150         batch_count = 0;
 151
 152         /*
 153          * sendmsg calls here after having queued its message on the send
 154          * queue.  We only have one task feeding the connection at a time.  If
 155          * another thread is already feeding the queue then we back off.  This
 156          * avoids blocking the caller and trading per-connection data between
 157          * caches per message.
 158          */
 159         if (!acquire_in_xmit(cp)) {
 160                 rds_stats_inc(s_send_lock_contention);
 161                 ret = -ENOMEM;
 162                 goto out;
 163         }
 164
 165         if (rds_destroy_pending(cp->cp_conn)) {
 166                 release_in_xmit(cp);
 167                 ret = -ENETUNREACH; /* dont requeue send work */
 168                 goto out;
 169         }
 170
 171         /*
 172          * we record the send generation after doing the xmit acquire.
 173          * if someone else manages to jump in and do some work, we'll use
 174          * this to avoid a goto restart farther down.
 175          *
 176          * The acquire_in_xmit() check above ensures that only one
 177          * caller can increment c_send_gen at any time.
 178          */
 179         send_gen = READ_ONCE(cp->cp_send_gen) + 1;
 180         WRITE_ONCE(cp->cp_send_gen, send_gen);
 181
 182         /*
 183          * rds_conn_shutdown() sets the conn state and then tests RDS_IN_XMIT,
 184          * we do the opposite to avoid races.
 185          */
 186         if (!rds_conn_path_up(cp)) {
 187                 release_in_xmit(cp);
 188                 ret = 0;
 189                 goto out;
 190         }
 191
 192         if (conn->c_trans->xmit_path_prepare)
 193                 conn->c_trans->xmit_path_prepare(cp);
 194
 195         /*
 196          * spin trying to push headers and data down the connection until
 197          * the connection doesn't make forward progress.
 198          */
 199         while (1) {
 200
 201                 rm = cp->cp_xmit_rm;
 202
 203                 /*
 204                  * If between sending messages, we can send a pending congestion
 205                  * map update.
 206                  */
 207                 if (!rm && test_and_clear_bit(0, &conn->c_map_queued)) {
 208                         rm = rds_cong_update_alloc(conn);
 209                         if (IS_ERR(rm)) {
 210                                 ret = PTR_ERR(rm);
 211                                 break;
 212                         }
 213                         rm->data.op_active = 1;
 214                         rm->m_inc.i_conn_path = cp;
 215                         rm->m_inc.i_conn = cp->cp_conn;
 216
 217                         cp->cp_xmit_rm = rm;
 218                 }
 219
 220                 /*
 221                  * If not already working on one, grab the next message.
 222                  *
 223                  * cp_xmit_rm holds a ref while we're sending this message down
 224                  * the connction.  We can use this ref while holding the
 225                  * send_sem.. rds_send_reset() is serialized with it.
 226                  */
 227                 if (!rm) {
 228                         unsigned int len;
 229
 230                         batch_count++;
 231
 232                         /* we want to process as big a batch as we can, but
 233                          * we also want to avoid softlockups.  If we've been
 234                          * through a lot of messages, lets back off and see
 235                          * if anyone else jumps in
 236                          */
 237                         if (batch_count >= send_batch_count)
 238                                 goto over_batch;
 239
 240                         spin_lock_irqsave(&cp->cp_lock, flags);
 241
 242                         if (!list_empty(&cp->cp_send_queue)) {
 243                                 rm = list_entry(cp->cp_send_queue.next,
 244                                                 struct rds_message,
 245                                                 m_conn_item);
 246                                 rds_message_addref(rm);
 247
 248                                 /*
 249                                  * Move the message from the send queue to the retransmit
 250                                  * list right away.
 251                                  */
 252                                 list_move_tail(&rm->m_conn_item,
 253                                                &cp->cp_retrans);
 254                         }
 255
 256                         spin_unlock_irqrestore(&cp->cp_lock, flags);
 257
 258                         if (!rm)
 259                                 break;
 260
 261                         /* Unfortunately, the way Infiniband deals with
 262                          * RDMA to a bad MR key is by moving the entire
 263                          * queue pair to error state. We cold possibly
 264                          * recover from that, but right now we drop the
 265                          * connection.
 266                          * Therefore, we never retransmit messages with RDMA ops.
 267                          */
 268                         if (test_bit(RDS_MSG_FLUSH, &rm->m_flags) ||
 269                             (rm->rdma.op_active &&
 270                             test_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags))) {
 271                                 spin_lock_irqsave(&cp->cp_lock, flags);
 272                                 if (test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags))
 273                                         list_move(&rm->m_conn_item, &to_be_dropped);
 274                                 spin_unlock_irqrestore(&cp->cp_lock, flags);
 275                                 continue;
 276                         }
 277
 278                         /* Require an ACK every once in a while */
 279                         len = ntohl(rm->m_inc.i_hdr.h_len);
 280                         if (cp->cp_unacked_packets == 0 ||
 281                             cp->cp_unacked_bytes < len) {
 282                                 set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags);
 283
 284                                 cp->cp_unacked_packets =
 285                                         rds_sysctl_max_unacked_packets;
 286                                 cp->cp_unacked_bytes =
 287                                         rds_sysctl_max_unacked_bytes;
 288                                 rds_stats_inc(s_send_ack_required);
 289                         } else {
 290                                 cp->cp_unacked_bytes -= len;
 291                                 cp->cp_unacked_packets--;
 292                         }
 293
 294                         cp->cp_xmit_rm = rm;
 295                 }
 296
 297                 /* The transport either sends the whole rdma or none of it */
 298                 if (rm->rdma.op_active && !cp->cp_xmit_rdma_sent) {
 299                         rm->m_final_op = &rm->rdma;
 300                         /* The transport owns the mapped memory for now.
 301                          * You can't unmap it while it's on the send queue
 302                          */
 303                         set_bit(RDS_MSG_MAPPED, &rm->m_flags);
 304                         ret = conn->c_trans->xmit_rdma(conn, &rm->rdma);
 305                         if (ret) {
 306                                 clear_bit(RDS_MSG_MAPPED, &rm->m_flags);
 307                                 wake_up_interruptible(&rm->m_flush_wait);
 308                                 break;
 309                         }
 310                         cp->cp_xmit_rdma_sent = 1;
 311
 312                 }
 313
 314                 if (rm->atomic.op_active && !cp->cp_xmit_atomic_sent) {
 315                         rm->m_final_op = &rm->atomic;
 316                         /* The transport owns the mapped memory for now.
 317                          * You can't unmap it while it's on the send queue
 318                          */
 319                         set_bit(RDS_MSG_MAPPED, &rm->m_flags);
 320                         ret = conn->c_trans->xmit_atomic(conn, &rm->atomic);
 321                         if (ret) {
 322                                 clear_bit(RDS_MSG_MAPPED, &rm->m_flags);
 323                                 wake_up_interruptible(&rm->m_flush_wait);
 324                                 break;
 325                         }
 326                         cp->cp_xmit_atomic_sent = 1;
 327
 328                 }
 329
 330                 /*
 331                  * A number of cases require an RDS header to be sent
 332                  * even if there is no data.
 333                  * We permit 0-byte sends; rds-ping depends on this.
 334                  * However, if there are exclusively attached silent ops,
 335                  * we skip the hdr/data send, to enable silent operation.
 336                  */
 337                 if (rm->data.op_nents == 0) {
 338                         int ops_present;
 339                         int all_ops_are_silent = 1;
 340
 341                         ops_present = (rm->atomic.op_active || rm->rdma.op_active);
 342                         if (rm->atomic.op_active && !rm->atomic.op_silent)
 343                                 all_ops_are_silent = 0;
 344                         if (rm->rdma.op_active && !rm->rdma.op_silent)
 345                                 all_ops_are_silent = 0;
 346
 347                         if (ops_present && all_ops_are_silent
 348                             && !rm->m_rdma_cookie)
 349                                 rm->data.op_active = 0;
 350                 }
 351
 352                 if (rm->data.op_active && !cp->cp_xmit_data_sent) {
 353                         rm->m_final_op = &rm->data;
 354
 355                         ret = conn->c_trans->xmit(conn, rm,
 356                                                   cp->cp_xmit_hdr_off,
 357                                                   cp->cp_xmit_sg,
 358                                                   cp->cp_xmit_data_off);
 359                         if (ret <= 0)
 360                                 break;
 361
 362                         if (cp->cp_xmit_hdr_off < sizeof(struct rds_header)) {
 363                                 tmp = min_t(int, ret,
 364                                             sizeof(struct rds_header) -
 365                                             cp->cp_xmit_hdr_off);
 366                                 cp->cp_xmit_hdr_off += tmp;
 367                                 ret -= tmp;
 368                         }
 369
 370                         sg = &rm->data.op_sg[cp->cp_xmit_sg];
 371                         while (ret) {
 372                                 tmp = min_t(int, ret, sg->length -
 373                                                       cp->cp_xmit_data_off);
 374                                 cp->cp_xmit_data_off += tmp;
 375                                 ret -= tmp;
 376                                 if (cp->cp_xmit_data_off == sg->length) {
 377                                         cp->cp_xmit_data_off = 0;
 378                                         sg++;
 379                                         cp->cp_xmit_sg++;
 380                                         BUG_ON(ret != 0 && cp->cp_xmit_sg ==
 381                                                rm->data.op_nents);
 382                                 }
 383                         }
 384
 385                         if (cp->cp_xmit_hdr_off == sizeof(struct rds_header) &&
 386                             (cp->cp_xmit_sg == rm->data.op_nents))
 387                                 cp->cp_xmit_data_sent = 1;
 388                 }
 389
 390                 /*
 391                  * A rm will only take multiple times through this loop
 392                  * if there is a data op. Thus, if the data is sent (or there was
 393                  * none), then we're done with the rm.
 394                  */
 395                 if (!rm->data.op_active || cp->cp_xmit_data_sent) {
 396                         cp->cp_xmit_rm = NULL;
 397                         cp->cp_xmit_sg = 0;
 398                         cp->cp_xmit_hdr_off = 0;
 399                         cp->cp_xmit_data_off = 0;
 400                         cp->cp_xmit_rdma_sent = 0;
 401                         cp->cp_xmit_atomic_sent = 0;
 402                         cp->cp_xmit_data_sent = 0;
 403
 404                         rds_message_put(rm);
 405                 }
 406         }
 407
 408 over_batch:
 409         if (conn->c_trans->xmit_path_complete)
 410                 conn->c_trans->xmit_path_complete(cp);
 411         release_in_xmit(cp);
 412
 413         /* Nuke any messages we decided not to retransmit. */
 414         if (!list_empty(&to_be_dropped)) {
 415                 /* irqs on here, so we can put(), unlike above */
 416                 list_for_each_entry(rm, &to_be_dropped, m_conn_item)
 417                         rds_message_put(rm);
 418                 rds_send_remove_from_sock(&to_be_dropped, RDS_RDMA_DROPPED);
 419         }
 420
 421         /*
 422          * Other senders can queue a message after we last test the send queue
 423          * but before we clear RDS_IN_XMIT.  In that case they'd back off and
 424          * not try and send their newly queued message.  We need to check the
 425          * send queue after having cleared RDS_IN_XMIT so that their message
 426          * doesn't get stuck on the send queue.
 427          *
 428          * If the transport cannot continue (i.e ret != 0), then it must
 429          * call us when more room is available, such as from the tx
 430          * completion handler.
 431          *
 432          * We have an extra generation check here so that if someone manages
 433          * to jump in after our release_in_xmit, we'll see that they have done
 434          * some work and we will skip our goto
 435          */
 436         if (ret == 0) {
 437                 bool raced;
 438
 439                 smp_mb();
 440                 raced = send_gen != READ_ONCE(cp->cp_send_gen);
 441
 442                 if ((test_bit(0, &conn->c_map_queued) ||
 443                     !list_empty(&cp->cp_send_queue)) && !raced) {
 444                         if (batch_count < send_batch_count)
 445                                 goto restart;
 446                         rcu_read_lock();
 447                         if (rds_destroy_pending(cp->cp_conn))
 448                                 ret = -ENETUNREACH;
 449                         else
 450                                 queue_delayed_work(rds_wq, &cp->cp_send_w, 1);
 451                         rcu_read_unlock();
 452                 } else if (raced) {
 453                         rds_stats_inc(s_send_lock_queue_raced);
 454                 }
 455         }
 456 out:
 457         return ret;
 458 }
 459 EXPORT_SYMBOL_GPL(rds_send_xmit);
 460
 461 static void rds_send_sndbuf_remove(struct rds_sock *rs, struct rds_message *rm)
 462 {
 463         u32 len = be32_to_cpu(rm->m_inc.i_hdr.h_len);
 464
 465         assert_spin_locked(&rs->rs_lock);
 466
 467         BUG_ON(rs->rs_snd_bytes < len);
 468         rs->rs_snd_bytes -= len;
 469
 470         if (rs->rs_snd_bytes == 0)
 471                 rds_stats_inc(s_send_queue_empty);
 472 }
 473
 474 static inline int rds_send_is_acked(struct rds_message *rm, u64 ack,
 475                                     is_acked_func is_acked)
 476 {
 477         if (is_acked)
 478                 return is_acked(rm, ack);
 479         return be64_to_cpu(rm->m_inc.i_hdr.h_sequence) <= ack;
 480 }
 481
 482 /*
 483  * This is pretty similar to what happens below in the ACK
 484  * handling code - except that we call here as soon as we get
 485  * the IB send completion on the RDMA op and the accompanying
 486  * message.
 487  */
 488 void rds_rdma_send_complete(struct rds_message *rm, int status)
 489 {
 490         struct rds_sock *rs = NULL;
 491         struct rm_rdma_op *ro;
 492         struct rds_notifier *notifier;
 493         unsigned long flags;
 494         unsigned int notify = 0;
 495
 496         spin_lock_irqsave(&rm->m_rs_lock, flags);
 497
 498         notify =  rm->rdma.op_notify | rm->data.op_notify;
 499         ro = &rm->rdma;
 500         if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags) &&
 501             ro->op_active && notify && ro->op_notifier) {
 502                 notifier = ro->op_notifier;
 503                 rs = rm->m_rs;
 504                 sock_hold(rds_rs_to_sk(rs));
 505
 506                 notifier->n_status = status;
 507                 spin_lock(&rs->rs_lock);
 508                 list_add_tail(&notifier->n_list, &rs->rs_notify_queue);
 509                 spin_unlock(&rs->rs_lock);
 510
 511                 ro->op_notifier = NULL;
 512         }
 513
 514         spin_unlock_irqrestore(&rm->m_rs_lock, flags);
 515
 516         if (rs) {
 517                 rds_wake_sk_sleep(rs);
 518                 sock_put(rds_rs_to_sk(rs));
 519         }
 520 }
 521 EXPORT_SYMBOL_GPL(rds_rdma_send_complete);
 522
 523 /*
 524  * Just like above, except looks at atomic op
 525  */
 526 void rds_atomic_send_complete(struct rds_message *rm, int status)
 527 {
 528         struct rds_sock *rs = NULL;
 529         struct rm_atomic_op *ao;
 530         struct rds_notifier *notifier;
 531         unsigned long flags;
 532
 533         spin_lock_irqsave(&rm->m_rs_lock, flags);
 534
 535         ao = &rm->atomic;
 536         if (test_bit(RDS_MSG_ON_SOCK, &rm->m_flags)
 537             && ao->op_active && ao->op_notify && ao->op_notifier) {
 538                 notifier = ao->op_notifier;
 539                 rs = rm->m_rs;
 540                 sock_hold(rds_rs_to_sk(rs));
 541
 542                 notifier->n_status = status;
 543                 spin_lock(&rs->rs_lock);
 544                 list_add_tail(&notifier->n_list, &rs->rs_notify_queue);
 545                 spin_unlock(&rs->rs_lock);
 546
 547                 ao->op_notifier = NULL;
 548         }
 549
 550         spin_unlock_irqrestore(&rm->m_rs_lock, flags);
 551
 552         if (rs) {
 553                 rds_wake_sk_sleep(rs);
 554                 sock_put(rds_rs_to_sk(rs));
 555         }
 556 }
 557 EXPORT_SYMBOL_GPL(rds_atomic_send_complete);
 558
 559 /*
 560  * This is the same as rds_rdma_send_complete except we
 561  * don't do any locking - we have all the ingredients (message,
 562  * socket, socket lock) and can just move the notifier.
 563  */
 564 static inline void
 565 __rds_send_complete(struct rds_sock *rs, struct rds_message *rm, int status)
 566 {
 567         struct rm_rdma_op *ro;
 568         struct rm_atomic_op *ao;
 569
 570         ro = &rm->rdma;
 571         if (ro->op_active && ro->op_notify && ro->op_notifier) {
 572                 ro->op_notifier->n_status = status;
 573                 list_add_tail(&ro->op_notifier->n_list, &rs->rs_notify_queue);
 574                 ro->op_notifier = NULL;
 575         }
 576
 577         ao = &rm->atomic;
 578         if (ao->op_active && ao->op_notify && ao->op_notifier) {
 579                 ao->op_notifier->n_status = status;
 580                 list_add_tail(&ao->op_notifier->n_list, &rs->rs_notify_queue);
 581                 ao->op_notifier = NULL;
 582         }
 583
 584         /* No need to wake the app - caller does this */
 585 }
 586
 587 /*
 588  * This removes messages from the socket's list if they're on it.  The list
 589  * argument must be private to the caller, we must be able to modify it
 590  * without locks.  The messages must have a reference held for their
 591  * position on the list.  This function will drop that reference after
 592  * removing the messages from the 'messages' list regardless of if it found
 593  * the messages on the socket list or not.
 594  */
 595 static void rds_send_remove_from_sock(struct list_head *messages, int status)
 596 {
 597         unsigned long flags;
 598         struct rds_sock *rs = NULL;
 599         struct rds_message *rm;
 600
 601         while (!list_empty(messages)) {
 602                 int was_on_sock = 0;
 603
 604                 rm = list_entry(messages->next, struct rds_message,
 605                                 m_conn_item);
 606                 list_del_init(&rm->m_conn_item);
 607
 608                 /*
 609                  * If we see this flag cleared then we're *sure* that someone
 610                  * else beat us to removing it from the sock.  If we race
 611                  * with their flag update we'll get the lock and then really
 612                  * see that the flag has been cleared.
 613                  *
 614                  * The message spinlock makes sure nobody clears rm->m_rs
 615                  * while we're messing with it. It does not prevent the
 616                  * message from being removed from the socket, though.
 617                  */
 618                 spin_lock_irqsave(&rm->m_rs_lock, flags);
 619                 if (!test_bit(RDS_MSG_ON_SOCK, &rm->m_flags))
 620                         goto unlock_and_drop;
 621
 622                 if (rs != rm->m_rs) {
 623                         if (rs) {
 624                                 rds_wake_sk_sleep(rs);
 625                                 sock_put(rds_rs_to_sk(rs));
 626                         }
 627                         rs = rm->m_rs;
 628                         if (rs)
 629                                 sock_hold(rds_rs_to_sk(rs));
 630                 }
 631                 if (!rs)
 632                         goto unlock_and_drop;
 633                 spin_lock(&rs->rs_lock);
 634
 635                 if (test_and_clear_bit(RDS_MSG_ON_SOCK, &rm->m_flags)) {
 636                         struct rm_rdma_op *ro = &rm->rdma;
 637                         struct rds_notifier *notifier;
 638
 639                         list_del_init(&rm->m_sock_item);
 640                         rds_send_sndbuf_remove(rs, rm);
 641
 642                         if (ro->op_active && ro->op_notifier &&
 643                                (ro->op_notify || (ro->op_recverr && status))) {
 644                                 notifier = ro->op_notifier;
 645                                 list_add_tail(&notifier->n_list,
 646                                                 &rs->rs_notify_queue);
 647                                 if (!notifier->n_status)
 648                                         notifier->n_status = status;
 649                                 rm->rdma.op_notifier = NULL;
 650                         }
 651                         was_on_sock = 1;
 652                 }
 653                 spin_unlock(&rs->rs_lock);
 654
 655 unlock_and_drop:
 656                 spin_unlock_irqrestore(&rm->m_rs_lock, flags);
 657                 rds_message_put(rm);
 658                 if (was_on_sock)
 659                         rds_message_put(rm);
 660         }
 661
 662         if (rs) {
 663                 rds_wake_sk_sleep(rs);
 664                 sock_put(rds_rs_to_sk(rs));
 665         }
 666 }
 667
 668 /*
 669  * Transports call here when they've determined that the receiver queued
 670  * messages up to, and including, the given sequence number.  Messages are
 671  * moved to the retrans queue when rds_send_xmit picks them off the send
 672  * queue. This means that in the TCP case, the message may not have been
 673  * assigned the m_ack_seq yet - but that's fine as long as tcp_is_acked
 674  * checks the RDS_MSG_HAS_ACK_SEQ bit.
 675  */
 676 void rds_send_path_drop_acked(struct rds_conn_path *cp, u64 ack,
 677                               is_acked_func is_acked)
 678 {
 679         struct rds_message *rm, *tmp;
 680         unsigned long flags;
 681         LIST_HEAD(list);
 682
 683         spin_lock_irqsave(&cp->cp_lock, flags);
 684
 685         list_for_each_entry_safe(rm, tmp, &cp->cp_retrans, m_conn_item) {
 686                 if (!rds_send_is_acked(rm, ack, is_acked))
 687                         break;
 688
 689                 list_move(&rm->m_conn_item, &list);
 690                 clear_bit(RDS_MSG_ON_CONN, &rm->m_flags);
 691         }
 692
 693         /* order flag updates with spin locks */
 694         if (!list_empty(&list))
 695                 smp_mb__after_atomic();
 696
 697         spin_unlock_irqrestore(&cp->cp_lock, flags);
 698
 699         /* now remove the messages from the sock list as needed */
 700         rds_send_remove_from_sock(&list, RDS_RDMA_SUCCESS);
 701 }
 702 EXPORT_SYMBOL_GPL(rds_send_path_drop_acked);
 703
 704 void rds_send_drop_acked(struct rds_connection *conn, u64 ack,
 705                          is_acked_func is_acked)
 706 {
 707         WARN_ON(conn->c_trans->t_mp_capable);
 708         rds_send_path_drop_acked(&conn->c_path[0], ack, is_acked);
 709 }
 710 EXPORT_SYMBOL_GPL(rds_send_drop_acked);
 711
 712 void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in6 *dest)
 713 {
 714         struct rds_message *rm, *tmp;
 715         struct rds_connection *conn;
 716         struct rds_conn_path *cp;
 717         unsigned long flags;
 718         LIST_HEAD(list);
 719
 720         /* get all the messages we're dropping under the rs lock */
 721         spin_lock_irqsave(&rs->rs_lock, flags);
 722
 723         list_for_each_entry_safe(rm, tmp, &rs->rs_send_queue, m_sock_item) {
 724                 if (dest &&
 725                     (!ipv6_addr_equal(&dest->sin6_addr, &rm->m_daddr) ||
 726                      dest->sin6_port != rm->m_inc.i_hdr.h_dport))
 727                         continue;
 728
 729                 list_move(&rm->m_sock_item, &list);
 730                 rds_send_sndbuf_remove(rs, rm);
 731                 clear_bit(RDS_MSG_ON_SOCK, &rm->m_flags);
 732         }
 733
 734         /* order flag updates with the rs lock */
 735         smp_mb__after_atomic();
 736
 737         spin_unlock_irqrestore(&rs->rs_lock, flags);
 738
 739         if (list_empty(&list))
 740                 return;
 741
 742         /* Remove the messages from the conn */
 743         list_for_each_entry(rm, &list, m_sock_item) {
 744
 745                 conn = rm->m_inc.i_conn;
 746                 if (conn->c_trans->t_mp_capable)
 747                         cp = rm->m_inc.i_conn_path;
 748                 else
 749                         cp = &conn->c_path[0];
 750
 751                 spin_lock_irqsave(&cp->cp_lock, flags);
 752                 /*
 753                  * Maybe someone else beat us to removing rm from the conn.
 754                  * If we race with their flag update we'll get the lock and
 755                  * then really see that the flag has been cleared.
 756                  */
 757                 if (!test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags)) {
 758                         spin_unlock_irqrestore(&cp->cp_lock, flags);
 759                         continue;
 760                 }
 761                 list_del_init(&rm->m_conn_item);
 762                 spin_unlock_irqrestore(&cp->cp_lock, flags);
 763
 764                 /*
 765                  * Couldn't grab m_rs_lock in top loop (lock ordering),
 766                  * but we can now.
 767                  */
 768                 spin_lock_irqsave(&rm->m_rs_lock, flags);
 769
 770                 spin_lock(&rs->rs_lock);
 771                 __rds_send_complete(rs, rm, RDS_RDMA_CANCELED);
 772                 spin_unlock(&rs->rs_lock);
 773
 774                 spin_unlock_irqrestore(&rm->m_rs_lock, flags);
 775
 776                 rds_message_put(rm);
 777         }
 778
 779         rds_wake_sk_sleep(rs);
 780
 781         while (!list_empty(&list)) {
 782                 rm = list_entry(list.next, struct rds_message, m_sock_item);
 783                 list_del_init(&rm->m_sock_item);
 784                 rds_message_wait(rm);
 785
 786                 /* just in case the code above skipped this message
 787                  * because RDS_MSG_ON_CONN wasn't set, run it again here
 788                  * taking m_rs_lock is the only thing that keeps us
 789                  * from racing with ack processing.
 790                  */
 791                 spin_lock_irqsave(&rm->m_rs_lock, flags);
 792
 793                 spin_lock(&rs->rs_lock);
 794                 __rds_send_complete(rs, rm, RDS_RDMA_CANCELED);
 795                 spin_unlock(&rs->rs_lock);
 796
 797                 spin_unlock_irqrestore(&rm->m_rs_lock, flags);
 798
 799                 rds_message_put(rm);
 800         }
 801 }
 802
 803 /*
 804  * we only want this to fire once so we use the callers 'queued'.  It's
 805  * possible that another thread can race with us and remove the
 806  * message from the flow with RDS_CANCEL_SENT_TO.
 807  */
 808 static int rds_send_queue_rm(struct rds_sock *rs, struct rds_connection *conn,
 809                              struct rds_conn_path *cp,
 810                              struct rds_message *rm, __be16 sport,
 811                              __be16 dport, int *queued)
 812 {
 813         unsigned long flags;
 814         u32 len;
 815
 816         if (*queued)
 817                 goto out;
 818
 819         len = be32_to_cpu(rm->m_inc.i_hdr.h_len);
 820
 821         /* this is the only place which holds both the socket's rs_lock
 822          * and the connection's c_lock */
 823         spin_lock_irqsave(&rs->rs_lock, flags);
 824
 825         /*
 826          * If there is a little space in sndbuf, we don't queue anything,
 827          * and userspace gets -EAGAIN. But poll() indicates there's send
 828          * room. This can lead to bad behavior (spinning) if snd_bytes isn't
 829          * freed up by incoming acks. So we check the *old* value of
 830          * rs_snd_bytes here to allow the last msg to exceed the buffer,
 831          * and poll() now knows no more data can be sent.
 832          */
 833         if (rs->rs_snd_bytes < rds_sk_sndbuf(rs)) {
 834                 rs->rs_snd_bytes += len;
 835
 836                 /* let recv side know we are close to send space exhaustion.
 837                  * This is probably not the optimal way to do it, as this
 838                  * means we set the flag on *all* messages as soon as our
 839                  * throughput hits a certain threshold.
 840                  */
 841                 if (rs->rs_snd_bytes >= rds_sk_sndbuf(rs) / 2)
 842                         set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags);
 843
 844                 list_add_tail(&rm->m_sock_item, &rs->rs_send_queue);
 845                 set_bit(RDS_MSG_ON_SOCK, &rm->m_flags);
 846                 rds_message_addref(rm);
 847                 sock_hold(rds_rs_to_sk(rs));
 848                 rm->m_rs = rs;
 849
 850                 /* The code ordering is a little weird, but we're
 851                    trying to minimize the time we hold c_lock */
 852                 rds_message_populate_header(&rm->m_inc.i_hdr, sport, dport, 0);
 853                 rm->m_inc.i_conn = conn;
 854                 rm->m_inc.i_conn_path = cp;
 855                 rds_message_addref(rm);
 856
 857                 spin_lock(&cp->cp_lock);
 858                 rm->m_inc.i_hdr.h_sequence = cpu_to_be64(cp->cp_next_tx_seq++);
 859                 list_add_tail(&rm->m_conn_item, &cp->cp_send_queue);
 860                 set_bit(RDS_MSG_ON_CONN, &rm->m_flags);
 861                 spin_unlock(&cp->cp_lock);
 862
 863                 rdsdebug("queued msg %p len %d, rs %p bytes %d seq %llu\n",
 864                          rm, len, rs, rs->rs_snd_bytes,
 865                          (unsigned long long)be64_to_cpu(rm->m_inc.i_hdr.h_sequence));
 866
 867                 *queued = 1;
 868         }
 869
 870         spin_unlock_irqrestore(&rs->rs_lock, flags);
 871 out:
 872         return *queued;
 873 }
 874
 875 /*
 876  * rds_message is getting to be quite complicated, and we'd like to allocate
 877  * it all in one go. This figures out how big it needs to be up front.
 878  */
 879 static int rds_rm_size(struct msghdr *msg, int num_sgs)
 880 {
 881         struct cmsghdr *cmsg;
 882         int size = 0;
 883         int cmsg_groups = 0;
 884         int retval;
 885         bool zcopy_cookie = false;
 886
 887         for_each_cmsghdr(cmsg, msg) {
 888                 if (!CMSG_OK(msg, cmsg))
 889                         return -EINVAL;
 890
 891                 if (cmsg->cmsg_level != SOL_RDS)
 892                         continue;
 893
 894                 switch (cmsg->cmsg_type) {
 895                 case RDS_CMSG_RDMA_ARGS:
 896                         cmsg_groups |= 1;
 897                         retval = rds_rdma_extra_size(CMSG_DATA(cmsg));
 898                         if (retval < 0)
 899                                 return retval;
 900                         size += retval;
 901
 902                         break;
 903
 904                 case RDS_CMSG_ZCOPY_COOKIE:
 905                         zcopy_cookie = true;
 906                         /* fall through */
 907
 908                 case RDS_CMSG_RDMA_DEST:
 909                 case RDS_CMSG_RDMA_MAP:
 910                         cmsg_groups |= 2;
 911                         /* these are valid but do no add any size */
 912                         break;
 913
 914                 case RDS_CMSG_ATOMIC_CSWP:
 915                 case RDS_CMSG_ATOMIC_FADD:
 916                 case RDS_CMSG_MASKED_ATOMIC_CSWP:
 917                 case RDS_CMSG_MASKED_ATOMIC_FADD:
 918                         cmsg_groups |= 1;
 919                         size += sizeof(struct scatterlist);
 920                         break;
 921
 922                 default:
 923                         return -EINVAL;
 924                 }
 925
 926         }
 927
 928         if ((msg->msg_flags & MSG_ZEROCOPY) && !zcopy_cookie)
 929                 return -EINVAL;
 930
 931         size += num_sgs * sizeof(struct scatterlist);
 932
 933         /* Ensure (DEST, MAP) are never used with (ARGS, ATOMIC) */
 934         if (cmsg_groups == 3)
 935                 return -EINVAL;
 936
 937         return size;
 938 }
 939
 940 static int rds_cmsg_zcopy(struct rds_sock *rs, struct rds_message *rm,
 941                           struct cmsghdr *cmsg)
 942 {
 943         u32 *cookie;
 944
 945         if (cmsg->cmsg_len < CMSG_LEN(sizeof(*cookie)) ||
 946             !rm->data.op_mmp_znotifier)
 947                 return -EINVAL;
 948         cookie = CMSG_DATA(cmsg);
 949         rm->data.op_mmp_znotifier->z_cookie = *cookie;
 950         return 0;
 951 }
 952
 953 static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm,
 954                          struct msghdr *msg, int *allocated_mr)
 955 {
 956         struct cmsghdr *cmsg;
 957         int ret = 0;
 958
 959         for_each_cmsghdr(cmsg, msg) {
 960                 if (!CMSG_OK(msg, cmsg))
 961                         return -EINVAL;
 962
 963                 if (cmsg->cmsg_level != SOL_RDS)
 964                         continue;
 965
 966                 /* As a side effect, RDMA_DEST and RDMA_MAP will set
 967                  * rm->rdma.m_rdma_cookie and rm->rdma.m_rdma_mr.
 968                  */
 969                 switch (cmsg->cmsg_type) {
 970                 case RDS_CMSG_RDMA_ARGS:
 971                         ret = rds_cmsg_rdma_args(rs, rm, cmsg);
 972                         break;
 973
 974                 case RDS_CMSG_RDMA_DEST:
 975                         ret = rds_cmsg_rdma_dest(rs, rm, cmsg);
 976                         break;
 977
 978                 case RDS_CMSG_RDMA_MAP:
 979                         ret = rds_cmsg_rdma_map(rs, rm, cmsg);
 980                         if (!ret)
 981                                 *allocated_mr = 1;
 982                         else if (ret == -ENODEV)
 983                                 /* Accommodate the get_mr() case which can fail
 984                                  * if connection isn't established yet.
 985                                  */
 986                                 ret = -EAGAIN;
 987                         break;
 988                 case RDS_CMSG_ATOMIC_CSWP:
 989                 case RDS_CMSG_ATOMIC_FADD:
 990                 case RDS_CMSG_MASKED_ATOMIC_CSWP:
 991                 case RDS_CMSG_MASKED_ATOMIC_FADD:
 992                         ret = rds_cmsg_atomic(rs, rm, cmsg);
 993                         break;
 994
 995                 case RDS_CMSG_ZCOPY_COOKIE:
 996                         ret = rds_cmsg_zcopy(rs, rm, cmsg);
 997                         break;
 998
 999                 default:
1000                         return -EINVAL;
1001                 }
1002
1003                 if (ret)
1004                         break;
1005         }
1006
1007         return ret;
1008 }
1009
1010 static int rds_send_mprds_hash(struct rds_sock *rs,
1011                                struct rds_connection *conn, int nonblock)
1012 {
1013         int hash;
1014
1015         if (conn->c_npaths == 0)
1016                 hash = RDS_MPATH_HASH(rs, RDS_MPATH_WORKERS);
1017         else
1018                 hash = RDS_MPATH_HASH(rs, conn->c_npaths);
1019         if (conn->c_npaths == 0 && hash != 0) {
1020                 rds_send_ping(conn, 0);
1021
1022                 /* The underlying connection is not up yet.  Need to wait
1023                  * until it is up to be sure that the non-zero c_path can be
1024                  * used.  But if we are interrupted, we have to use the zero
1025                  * c_path in case the connection ends up being non-MP capable.
1026                  */
1027                 if (conn->c_npaths == 0) {
1028                         /* Cannot wait for the connection be made, so just use
1029                          * the base c_path.
1030                          */
1031                         if (nonblock)
1032                                 return 0;
1033                         if (wait_event_interruptible(conn->c_hs_waitq,
1034                                                      conn->c_npaths != 0))
1035                                 hash = 0;
1036                 }
1037                 if (conn->c_npaths == 1)
1038                         hash = 0;
1039         }
1040         return hash;
1041 }
1042
1043 static int rds_rdma_bytes(struct msghdr *msg, size_t *rdma_bytes)
1044 {
1045         struct rds_rdma_args *args;
1046         struct cmsghdr *cmsg;
1047
1048         for_each_cmsghdr(cmsg, msg) {
1049                 if (!CMSG_OK(msg, cmsg))
1050                         return -EINVAL;
1051
1052                 if (cmsg->cmsg_level != SOL_RDS)
1053                         continue;
1054
1055                 if (cmsg->cmsg_type == RDS_CMSG_RDMA_ARGS) {
1056                         if (cmsg->cmsg_len <
1057                             CMSG_LEN(sizeof(struct rds_rdma_args)))
1058                                 return -EINVAL;
1059                         args = CMSG_DATA(cmsg);
1060                         *rdma_bytes += args->remote_vec.bytes;
1061                 }
1062         }
1063         return 0;
1064 }
1065
1066 int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len)
1067 {
1068         struct sock *sk = sock->sk;
1069         struct rds_sock *rs = rds_sk_to_rs(sk);
1070         DECLARE_SOCKADDR(struct sockaddr_in6 *, sin6, msg->msg_name);
1071         DECLARE_SOCKADDR(struct sockaddr_in *, usin, msg->msg_name);
1072         __be16 dport;
1073         struct rds_message *rm = NULL;
1074         struct rds_connection *conn;
1075         int ret = 0;
1076         int queued = 0, allocated_mr = 0;
1077         int nonblock = msg->msg_flags & MSG_DONTWAIT;
1078         long timeo = sock_sndtimeo(sk, nonblock);
1079         struct rds_conn_path *cpath;
1080         struct in6_addr daddr;
1081         __u32 scope_id = 0;
1082         size_t total_payload_len = payload_len, rdma_payload_len = 0;
1083         bool zcopy = ((msg->msg_flags & MSG_ZEROCOPY) &&
1084                       sock_flag(rds_rs_to_sk(rs), SOCK_ZEROCOPY));
1085         int num_sgs = ceil(payload_len, PAGE_SIZE);
1086         int namelen;
1087
1088         /* Mirror Linux UDP mirror of BSD error message compatibility */
1089         /* XXX: Perhaps MSG_MORE someday */
1090         if (msg->msg_flags & ~(MSG_DONTWAIT | MSG_CMSG_COMPAT | MSG_ZEROCOPY)) {
1091                 ret = -EOPNOTSUPP;
1092                 goto out;
1093         }
1094
1095         namelen = msg->msg_namelen;
1096         if (namelen != 0) {
1097                 if (namelen < sizeof(*usin)) {
1098                         ret = -EINVAL;
1099                         goto out;
1100                 }
1101                 switch (usin->sin_family) {
1102                 case AF_INET:
1103                         if (usin->sin_addr.s_addr == htonl(INADDR_ANY) ||
1104                             usin->sin_addr.s_addr == htonl(INADDR_BROADCAST) ||
1105                             IN_MULTICAST(ntohl(usin->sin_addr.s_addr))) {
1106                                 ret = -EINVAL;
1107                                 goto out;
1108                         }
1109                         ipv6_addr_set_v4mapped(usin->sin_addr.s_addr, &daddr);
1110                         dport = usin->sin_port;
1111                         break;
1112
1113 #if IS_ENABLED(CONFIG_IPV6)
1114                 case AF_INET6: {
1115                         int addr_type;
1116
1117                         if (namelen < sizeof(*sin6)) {
1118                                 ret = -EINVAL;
1119                                 goto out;
1120                         }
1121                         addr_type = ipv6_addr_type(&sin6->sin6_addr);
1122                         if (!(addr_type & IPV6_ADDR_UNICAST)) {
1123                                 __be32 addr4;
1124
1125                                 if (!(addr_type & IPV6_ADDR_MAPPED)) {
1126                                         ret = -EINVAL;
1127                                         goto out;
1128                                 }
1129
1130                                 /* It is a mapped address.  Need to do some
1131                                  * sanity checks.
1132                                  */
1133                                 addr4 = sin6->sin6_addr.s6_addr32[3];
1134                                 if (addr4 == htonl(INADDR_ANY) ||
1135                                     addr4 == htonl(INADDR_BROADCAST) ||
1136                                     IN_MULTICAST(ntohl(addr4))) {
1137                                         ret = -EINVAL;
1138                                         goto out;
1139                                 }
1140                         }
1141                         if (addr_type & IPV6_ADDR_LINKLOCAL) {
1142                                 if (sin6->sin6_scope_id == 0) {
1143                                         ret = -EINVAL;
1144                                         goto out;
1145                                 }
1146                                 scope_id = sin6->sin6_scope_id;
1147                         }
1148
1149                         daddr = sin6->sin6_addr;
1150                         dport = sin6->sin6_port;
1151                         break;
1152                 }
1153 #endif
1154
1155                 default:
1156                         ret = -EINVAL;
1157                         goto out;
1158                 }
1159         } else {
1160                 /* We only care about consistency with ->connect() */
1161                 lock_sock(sk);
1162                 daddr = rs->rs_conn_addr;
1163                 dport = rs->rs_conn_port;
1164                 scope_id = rs->rs_bound_scope_id;
1165                 release_sock(sk);
1166         }
1167
1168         lock_sock(sk);
1169         if (ipv6_addr_any(&rs->rs_bound_addr) || ipv6_addr_any(&daddr)) {
1170                 release_sock(sk);
1171                 ret = -ENOTCONN;
1172                 goto out;
1173         } else if (namelen != 0) {
1174                 /* Cannot send to an IPv4 address using an IPv6 source
1175                  * address and cannot send to an IPv6 address using an
1176                  * IPv4 source address.
1177                  */
1178                 if (ipv6_addr_v4mapped(&daddr) ^
1179                     ipv6_addr_v4mapped(&rs->rs_bound_addr)) {
1180                         release_sock(sk);
1181                         ret = -EOPNOTSUPP;
1182                         goto out;
1183                 }
1184                 /* If the socket is already bound to a link local address,
1185                  * it can only send to peers on the same link.  But allow
1186                  * communicating beween link local and non-link local address.
1187                  */
1188                 if (scope_id != rs->rs_bound_scope_id) {
1189                         if (!scope_id) {
1190                                 scope_id = rs->rs_bound_scope_id;
1191                         } else if (rs->rs_bound_scope_id) {
1192                                 release_sock(sk);
1193                                 ret = -EINVAL;
1194                                 goto out;
1195                         }
1196                 }
1197         }
1198         release_sock(sk);
1199
1200         ret = rds_rdma_bytes(msg, &rdma_payload_len);
1201         if (ret)
1202                 goto out;
1203
1204         total_payload_len += rdma_payload_len;
1205         if (max_t(size_t, payload_len, rdma_payload_len) > RDS_MAX_MSG_SIZE) {
1206                 ret = -EMSGSIZE;
1207                 goto out;
1208         }
1209
1210         if (payload_len > rds_sk_sndbuf(rs)) {
1211                 ret = -EMSGSIZE;
1212                 goto out;
1213         }
1214
1215         if (zcopy) {
1216                 if (rs->rs_transport->t_type != RDS_TRANS_TCP) {
1217                         ret = -EOPNOTSUPP;
1218                         goto out;
1219                 }
1220                 num_sgs = iov_iter_npages(&msg->msg_iter, INT_MAX);
1221         }
1222         /* size of rm including all sgs */
1223         ret = rds_rm_size(msg, num_sgs);
1224         if (ret < 0)
1225                 goto out;
1226
1227         rm = rds_message_alloc(ret, GFP_KERNEL);
1228         if (!rm) {
1229                 ret = -ENOMEM;
1230                 goto out;
1231         }
1232
1233         /* Attach data to the rm */
1234         if (payload_len) {
1235                 rm->data.op_sg = rds_message_alloc_sgs(rm, num_sgs);
1236                 if (!rm->data.op_sg) {
1237                         ret = -ENOMEM;
1238                         goto out;
1239                 }
1240                 ret = rds_message_copy_from_user(rm, &msg->msg_iter, zcopy);
1241                 if (ret)
1242                         goto out;
1243         }
1244         rm->data.op_active = 1;
1245
1246         rm->m_daddr = daddr;
1247
1248         /* rds_conn_create has a spinlock that runs with IRQ off.
1249          * Caching the conn in the socket helps a lot. */
1250         if (rs->rs_conn && ipv6_addr_equal(&rs->rs_conn->c_faddr, &daddr))
1251                 conn = rs->rs_conn;
1252         else {
1253                 conn = rds_conn_create_outgoing(sock_net(sock->sk),
1254                                                 &rs->rs_bound_addr, &daddr,
1255                                                 rs->rs_transport,
1256                                                 sock->sk->sk_allocation,
1257                                                 scope_id);
1258                 if (IS_ERR(conn)) {
1259                         ret = PTR_ERR(conn);
1260                         goto out;
1261                 }
1262                 rs->rs_conn = conn;
1263         }
1264
1265         if (conn->c_trans->t_mp_capable)
1266                 cpath = &conn->c_path[rds_send_mprds_hash(rs, conn, nonblock)];
1267         else
1268                 cpath = &conn->c_path[0];
1269
1270         rm->m_conn_path = cpath;
1271
1272         /* Parse any control messages the user may have included. */
1273         ret = rds_cmsg_send(rs, rm, msg, &allocated_mr);
1274         if (ret) {
1275                 /* Trigger connection so that its ready for the next retry */
1276                 if (ret ==  -EAGAIN)
1277                         rds_conn_connect_if_down(conn);
1278                 goto out;
1279         }
1280
1281         if (rm->rdma.op_active && !conn->c_trans->xmit_rdma) {
1282                 printk_ratelimited(KERN_NOTICE "rdma_op %p conn xmit_rdma %p\n",
1283                                &rm->rdma, conn->c_trans->xmit_rdma);
1284                 ret = -EOPNOTSUPP;
1285                 goto out;
1286         }
1287
1288         if (rm->atomic.op_active && !conn->c_trans->xmit_atomic) {
1289                 printk_ratelimited(KERN_NOTICE "atomic_op %p conn xmit_atomic %p\n",
1290                                &rm->atomic, conn->c_trans->xmit_atomic);
1291                 ret = -EOPNOTSUPP;
1292                 goto out;
1293         }
1294
1295         if (rds_destroy_pending(conn)) {
1296                 ret = -EAGAIN;
1297                 goto out;
1298         }
1299
1300         rds_conn_path_connect_if_down(cpath);
1301
1302         ret = rds_cong_wait(conn->c_fcong, dport, nonblock, rs);
1303         if (ret) {
1304                 rs->rs_seen_congestion = 1;
1305                 goto out;
1306         }
1307         while (!rds_send_queue_rm(rs, conn, cpath, rm, rs->rs_bound_port,
1308                                   dport, &queued)) {
1309                 rds_stats_inc(s_send_queue_full);
1310
1311                 if (nonblock) {
1312                         ret = -EAGAIN;
1313                         goto out;
1314                 }
1315
1316                 timeo = wait_event_interruptible_timeout(*sk_sleep(sk),
1317                                         rds_send_queue_rm(rs, conn, cpath, rm,
1318                                                           rs->rs_bound_port,
1319                                                           dport,
1320                                                           &queued),
1321                                         timeo);
1322                 rdsdebug("sendmsg woke queued %d timeo %ld\n", queued, timeo);
1323                 if (timeo > 0 || timeo == MAX_SCHEDULE_TIMEOUT)
1324                         continue;
1325
1326                 ret = timeo;
1327                 if (ret == 0)
1328                         ret = -ETIMEDOUT;
1329                 goto out;
1330         }
1331
1332         /*
1333          * By now we've committed to the send.  We reuse rds_send_worker()
1334          * to retry sends in the rds thread if the transport asks us to.
1335          */
1336         rds_stats_inc(s_send_queued);
1337
1338         ret = rds_send_xmit(cpath);
1339         if (ret == -ENOMEM || ret == -EAGAIN) {
1340                 ret = 0;
1341                 rcu_read_lock();
1342                 if (rds_destroy_pending(cpath->cp_conn))
1343                         ret = -ENETUNREACH;
1344                 else
1345                         queue_delayed_work(rds_wq, &cpath->cp_send_w, 1);
1346                 rcu_read_unlock();
1347         }
1348         if (ret)
1349                 goto out;
1350         rds_message_put(rm);
1351         return payload_len;
1352
1353 out:
1354         /* If the user included a RDMA_MAP cmsg, we allocated a MR on the fly.
1355          * If the sendmsg goes through, we keep the MR. If it fails with EAGAIN
1356          * or in any other way, we need to destroy the MR again */
1357         if (allocated_mr)
1358                 rds_rdma_unuse(rs, rds_rdma_cookie_key(rm->m_rdma_cookie), 1);
1359
1360         if (rm)
1361                 rds_message_put(rm);
1362         return ret;
1363 }
1364
1365 /*
1366  * send out a probe. Can be shared by rds_send_ping,
1367  * rds_send_pong, rds_send_hb.
1368  * rds_send_hb should use h_flags
1369  *   RDS_FLAG_HB_PING|RDS_FLAG_ACK_REQUIRED
1370  * or
1371  *   RDS_FLAG_HB_PONG|RDS_FLAG_ACK_REQUIRED
1372  */
1373 static int
1374 rds_send_probe(struct rds_conn_path *cp, __be16 sport,
1375                __be16 dport, u8 h_flags)
1376 {
1377         struct rds_message *rm;
1378         unsigned long flags;
1379         int ret = 0;
1380
1381         rm = rds_message_alloc(0, GFP_ATOMIC);
1382         if (!rm) {
1383                 ret = -ENOMEM;
1384                 goto out;
1385         }
1386
1387         rm->m_daddr = cp->cp_conn->c_faddr;
1388         rm->data.op_active = 1;
1389
1390         rds_conn_path_connect_if_down(cp);
1391
1392         ret = rds_cong_wait(cp->cp_conn->c_fcong, dport, 1, NULL);
1393         if (ret)
1394                 goto out;
1395
1396         spin_lock_irqsave(&cp->cp_lock, flags);
1397         list_add_tail(&rm->m_conn_item, &cp->cp_send_queue);
1398         set_bit(RDS_MSG_ON_CONN, &rm->m_flags);
1399         rds_message_addref(rm);
1400         rm->m_inc.i_conn = cp->cp_conn;
1401         rm->m_inc.i_conn_path = cp;
1402
1403         rds_message_populate_header(&rm->m_inc.i_hdr, sport, dport,
1404                                     cp->cp_next_tx_seq);
1405         rm->m_inc.i_hdr.h_flags |= h_flags;
1406         cp->cp_next_tx_seq++;
1407
1408         if (RDS_HS_PROBE(be16_to_cpu(sport), be16_to_cpu(dport)) &&
1409             cp->cp_conn->c_trans->t_mp_capable) {
1410                 u16 npaths = cpu_to_be16(RDS_MPATH_WORKERS);
1411                 u32 my_gen_num = cpu_to_be32(cp->cp_conn->c_my_gen_num);
1412
1413                 rds_message_add_extension(&rm->m_inc.i_hdr,
1414                                           RDS_EXTHDR_NPATHS, &npaths,
1415                                           sizeof(npaths));
1416                 rds_message_add_extension(&rm->m_inc.i_hdr,
1417                                           RDS_EXTHDR_GEN_NUM,
1418                                           &my_gen_num,
1419                                           sizeof(u32));
1420         }
1421         spin_unlock_irqrestore(&cp->cp_lock, flags);
1422
1423         rds_stats_inc(s_send_queued);
1424         rds_stats_inc(s_send_pong);
1425
1426         /* schedule the send work on rds_wq */
1427         rcu_read_lock();
1428         if (!rds_destroy_pending(cp->cp_conn))
1429                 queue_delayed_work(rds_wq, &cp->cp_send_w, 1);
1430         rcu_read_unlock();
1431
1432         rds_message_put(rm);
1433         return 0;
1434
1435 out:
1436         if (rm)
1437                 rds_message_put(rm);
1438         return ret;
1439 }
1440
1441 int
1442 rds_send_pong(struct rds_conn_path *cp, __be16 dport)
1443 {
1444         return rds_send_probe(cp, 0, dport, 0);
1445 }
1446
1447 void
1448 rds_send_ping(struct rds_connection *conn, int cp_index)
1449 {
1450         unsigned long flags;
1451         struct rds_conn_path *cp = &conn->c_path[cp_index];
1452
1453         spin_lock_irqsave(&cp->cp_lock, flags);
1454         if (conn->c_ping_triggered) {
1455                 spin_unlock_irqrestore(&cp->cp_lock, flags);
1456                 return;
1457         }
1458         conn->c_ping_triggered = 1;
1459         spin_unlock_irqrestore(&cp->cp_lock, flags);
1460         rds_send_probe(cp, cpu_to_be16(RDS_FLAG_PROBE_PORT), 0, 0);
1461 }
1462 EXPORT_SYMBOL_GPL(rds_send_ping);