net/sunrpc/xprtrdma/transport.c

   1 /*
   2  * Copyright (c) 2003-2007 Network Appliance, Inc. 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 BSD-type
   8  * license below:
   9  *
  10  * Redistribution and use in source and binary forms, with or without
  11  * modification, are permitted provided that the following conditions
  12  * are met:
  13  *
  14  *      Redistributions of source code must retain the above copyright
  15  *      notice, this list of conditions and the following disclaimer.
  16  *
  17  *      Redistributions in binary form must reproduce the above
  18  *      copyright notice, this list of conditions and the following
  19  *      disclaimer in the documentation and/or other materials provided
  20  *      with the distribution.
  21  *
  22  *      Neither the name of the Network Appliance, Inc. nor the names of
  23  *      its contributors may be used to endorse or promote products
  24  *      derived from this software without specific prior written
  25  *      permission.
  26  *
  27  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  28  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  29  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
  30  * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
  31  * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  32  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  33  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  34  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  35  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  36  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  37  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  38  */
  39
  40 /*
  41  * transport.c
  42  *
  43  * This file contains the top-level implementation of an RPC RDMA
  44  * transport.
  45  *
  46  * Naming convention: functions beginning with xprt_ are part of the
  47  * transport switch. All others are RPC RDMA internal.
  48  */
  49
  50 #include <linux/module.h>
  51 #include <linux/slab.h>
  52 #include <linux/seq_file.h>
  53 #include <linux/sunrpc/addr.h>
  54
  55 #include "xprt_rdma.h"
  56
  57 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
  58 # define RPCDBG_FACILITY        RPCDBG_TRANS
  59 #endif
  60
  61 /*
  62  * tunables
  63  */
  64
  65 static unsigned int xprt_rdma_slot_table_entries = RPCRDMA_DEF_SLOT_TABLE;
  66 unsigned int xprt_rdma_max_inline_read = RPCRDMA_DEF_INLINE;
  67 static unsigned int xprt_rdma_max_inline_write = RPCRDMA_DEF_INLINE;
  68 static unsigned int xprt_rdma_inline_write_padding;
  69 unsigned int xprt_rdma_memreg_strategy          = RPCRDMA_FRMR;
  70 int xprt_rdma_pad_optimize;
  71
  72 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
  73
  74 static unsigned int min_slot_table_size = RPCRDMA_MIN_SLOT_TABLE;
  75 static unsigned int max_slot_table_size = RPCRDMA_MAX_SLOT_TABLE;
  76 static unsigned int min_inline_size = RPCRDMA_MIN_INLINE;
  77 static unsigned int max_inline_size = RPCRDMA_MAX_INLINE;
  78 static unsigned int zero;
  79 static unsigned int max_padding = PAGE_SIZE;
  80 static unsigned int min_memreg = RPCRDMA_BOUNCEBUFFERS;
  81 static unsigned int max_memreg = RPCRDMA_LAST - 1;
  82
  83 static struct ctl_table_header *sunrpc_table_header;
  84
  85 static struct ctl_table xr_tunables_table[] = {
  86         {
  87                 .procname       = "rdma_slot_table_entries",
  88                 .data           = &xprt_rdma_slot_table_entries,
  89                 .maxlen         = sizeof(unsigned int),
  90                 .mode           = 0644,
  91                 .proc_handler   = proc_dointvec_minmax,
  92                 .extra1         = &min_slot_table_size,
  93                 .extra2         = &max_slot_table_size
  94         },
  95         {
  96                 .procname       = "rdma_max_inline_read",
  97                 .data           = &xprt_rdma_max_inline_read,
  98                 .maxlen         = sizeof(unsigned int),
  99                 .mode           = 0644,
 100                 .proc_handler   = proc_dointvec_minmax,
 101                 .extra1         = &min_inline_size,
 102                 .extra2         = &max_inline_size,
 103         },
 104         {
 105                 .procname       = "rdma_max_inline_write",
 106                 .data           = &xprt_rdma_max_inline_write,
 107                 .maxlen         = sizeof(unsigned int),
 108                 .mode           = 0644,
 109                 .proc_handler   = proc_dointvec_minmax,
 110                 .extra1         = &min_inline_size,
 111                 .extra2         = &max_inline_size,
 112         },
 113         {
 114                 .procname       = "rdma_inline_write_padding",
 115                 .data           = &xprt_rdma_inline_write_padding,
 116                 .maxlen         = sizeof(unsigned int),
 117                 .mode           = 0644,
 118                 .proc_handler   = proc_dointvec_minmax,
 119                 .extra1         = &zero,
 120                 .extra2         = &max_padding,
 121         },
 122         {
 123                 .procname       = "rdma_memreg_strategy",
 124                 .data           = &xprt_rdma_memreg_strategy,
 125                 .maxlen         = sizeof(unsigned int),
 126                 .mode           = 0644,
 127                 .proc_handler   = proc_dointvec_minmax,
 128                 .extra1         = &min_memreg,
 129                 .extra2         = &max_memreg,
 130         },
 131         {
 132                 .procname       = "rdma_pad_optimize",
 133                 .data           = &xprt_rdma_pad_optimize,
 134                 .maxlen         = sizeof(unsigned int),
 135                 .mode           = 0644,
 136                 .proc_handler   = proc_dointvec,
 137         },
 138         { },
 139 };
 140
 141 static struct ctl_table sunrpc_table[] = {
 142         {
 143                 .procname       = "sunrpc",
 144                 .mode           = 0555,
 145                 .child          = xr_tunables_table
 146         },
 147         { },
 148 };
 149
 150 #endif
 151
 152 static struct rpc_xprt_ops xprt_rdma_procs;     /*forward reference */
 153
 154 static void
 155 xprt_rdma_format_addresses4(struct rpc_xprt *xprt, struct sockaddr *sap)
 156 {
 157         struct sockaddr_in *sin = (struct sockaddr_in *)sap;
 158         char buf[20];
 159
 160         snprintf(buf, sizeof(buf), "%08x", ntohl(sin->sin_addr.s_addr));
 161         xprt->address_strings[RPC_DISPLAY_HEX_ADDR] = kstrdup(buf, GFP_KERNEL);
 162
 163         xprt->address_strings[RPC_DISPLAY_NETID] = RPCBIND_NETID_RDMA;
 164 }
 165
 166 static void
 167 xprt_rdma_format_addresses6(struct rpc_xprt *xprt, struct sockaddr *sap)
 168 {
 169         struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
 170         char buf[40];
 171
 172         snprintf(buf, sizeof(buf), "%pi6", &sin6->sin6_addr);
 173         xprt->address_strings[RPC_DISPLAY_HEX_ADDR] = kstrdup(buf, GFP_KERNEL);
 174
 175         xprt->address_strings[RPC_DISPLAY_NETID] = RPCBIND_NETID_RDMA6;
 176 }
 177
 178 void
 179 xprt_rdma_format_addresses(struct rpc_xprt *xprt, struct sockaddr *sap)
 180 {
 181         char buf[128];
 182
 183         switch (sap->sa_family) {
 184         case AF_INET:
 185                 xprt_rdma_format_addresses4(xprt, sap);
 186                 break;
 187         case AF_INET6:
 188                 xprt_rdma_format_addresses6(xprt, sap);
 189                 break;
 190         default:
 191                 pr_err("rpcrdma: Unrecognized address family\n");
 192                 return;
 193         }
 194
 195         (void)rpc_ntop(sap, buf, sizeof(buf));
 196         xprt->address_strings[RPC_DISPLAY_ADDR] = kstrdup(buf, GFP_KERNEL);
 197
 198         snprintf(buf, sizeof(buf), "%u", rpc_get_port(sap));
 199         xprt->address_strings[RPC_DISPLAY_PORT] = kstrdup(buf, GFP_KERNEL);
 200
 201         snprintf(buf, sizeof(buf), "%4hx", rpc_get_port(sap));
 202         xprt->address_strings[RPC_DISPLAY_HEX_PORT] = kstrdup(buf, GFP_KERNEL);
 203
 204         xprt->address_strings[RPC_DISPLAY_PROTO] = "rdma";
 205 }
 206
 207 void
 208 xprt_rdma_free_addresses(struct rpc_xprt *xprt)
 209 {
 210         unsigned int i;
 211
 212         for (i = 0; i < RPC_DISPLAY_MAX; i++)
 213                 switch (i) {
 214                 case RPC_DISPLAY_PROTO:
 215                 case RPC_DISPLAY_NETID:
 216                         continue;
 217                 default:
 218                         kfree(xprt->address_strings[i]);
 219                 }
 220 }
 221
 222 void
 223 rpcrdma_conn_func(struct rpcrdma_ep *ep)
 224 {
 225         schedule_delayed_work(&ep->rep_connect_worker, 0);
 226 }
 227
 228 void
 229 rpcrdma_connect_worker(struct work_struct *work)
 230 {
 231         struct rpcrdma_ep *ep =
 232                 container_of(work, struct rpcrdma_ep, rep_connect_worker.work);
 233         struct rpcrdma_xprt *r_xprt =
 234                 container_of(ep, struct rpcrdma_xprt, rx_ep);
 235         struct rpc_xprt *xprt = &r_xprt->rx_xprt;
 236
 237         spin_lock_bh(&xprt->transport_lock);
 238         if (++xprt->connect_cookie == 0)        /* maintain a reserved value */
 239                 ++xprt->connect_cookie;
 240         if (ep->rep_connected > 0) {
 241                 if (!xprt_test_and_set_connected(xprt))
 242                         xprt_wake_pending_tasks(xprt, 0);
 243         } else {
 244                 if (xprt_test_and_clear_connected(xprt))
 245                         xprt_wake_pending_tasks(xprt, -ENOTCONN);
 246         }
 247         spin_unlock_bh(&xprt->transport_lock);
 248 }
 249
 250 static void
 251 xprt_rdma_connect_worker(struct work_struct *work)
 252 {
 253         struct rpcrdma_xprt *r_xprt = container_of(work, struct rpcrdma_xprt,
 254                                                    rx_connect_worker.work);
 255         struct rpc_xprt *xprt = &r_xprt->rx_xprt;
 256         int rc = 0;
 257
 258         xprt_clear_connected(xprt);
 259
 260         dprintk("RPC:       %s: %sconnect\n", __func__,
 261                         r_xprt->rx_ep.rep_connected != 0 ? "re" : "");
 262         rc = rpcrdma_ep_connect(&r_xprt->rx_ep, &r_xprt->rx_ia);
 263         if (rc)
 264                 xprt_wake_pending_tasks(xprt, rc);
 265
 266         dprintk("RPC:       %s: exit\n", __func__);
 267         xprt_clear_connecting(xprt);
 268 }
 269
 270 static void
 271 xprt_rdma_inject_disconnect(struct rpc_xprt *xprt)
 272 {
 273         struct rpcrdma_xprt *r_xprt = container_of(xprt, struct rpcrdma_xprt,
 274                                                    rx_xprt);
 275
 276         pr_info("rpcrdma: injecting transport disconnect on xprt=%p\n", xprt);
 277         rdma_disconnect(r_xprt->rx_ia.ri_id);
 278 }
 279
 280 /*
 281  * xprt_rdma_destroy
 282  *
 283  * Destroy the xprt.
 284  * Free all memory associated with the object, including its own.
 285  * NOTE: none of the *destroy methods free memory for their top-level
 286  * objects, even though they may have allocated it (they do free
 287  * private memory). It's up to the caller to handle it. In this
 288  * case (RDMA transport), all structure memory is inlined with the
 289  * struct rpcrdma_xprt.
 290  */
 291 static void
 292 xprt_rdma_destroy(struct rpc_xprt *xprt)
 293 {
 294         struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 295
 296         dprintk("RPC:       %s: called\n", __func__);
 297
 298         cancel_delayed_work_sync(&r_xprt->rx_connect_worker);
 299
 300         xprt_clear_connected(xprt);
 301
 302         rpcrdma_ep_destroy(&r_xprt->rx_ep, &r_xprt->rx_ia);
 303         rpcrdma_buffer_destroy(&r_xprt->rx_buf);
 304         rpcrdma_ia_close(&r_xprt->rx_ia);
 305
 306         xprt_rdma_free_addresses(xprt);
 307
 308         xprt_free(xprt);
 309
 310         dprintk("RPC:       %s: returning\n", __func__);
 311
 312         module_put(THIS_MODULE);
 313 }
 314
 315 static const struct rpc_timeout xprt_rdma_default_timeout = {
 316         .to_initval = 60 * HZ,
 317         .to_maxval = 60 * HZ,
 318 };
 319
 320 /**
 321  * xprt_setup_rdma - Set up transport to use RDMA
 322  *
 323  * @args: rpc transport arguments
 324  */
 325 static struct rpc_xprt *
 326 xprt_setup_rdma(struct xprt_create *args)
 327 {
 328         struct rpcrdma_create_data_internal cdata;
 329         struct rpc_xprt *xprt;
 330         struct rpcrdma_xprt *new_xprt;
 331         struct rpcrdma_ep *new_ep;
 332         struct sockaddr *sap;
 333         int rc;
 334
 335         if (args->addrlen > sizeof(xprt->addr)) {
 336                 dprintk("RPC:       %s: address too large\n", __func__);
 337                 return ERR_PTR(-EBADF);
 338         }
 339
 340         xprt = xprt_alloc(args->net, sizeof(struct rpcrdma_xprt),
 341                         xprt_rdma_slot_table_entries,
 342                         xprt_rdma_slot_table_entries);
 343         if (xprt == NULL) {
 344                 dprintk("RPC:       %s: couldn't allocate rpcrdma_xprt\n",
 345                         __func__);
 346                 return ERR_PTR(-ENOMEM);
 347         }
 348
 349         /* 60 second timeout, no retries */
 350         xprt->timeout = &xprt_rdma_default_timeout;
 351         xprt->bind_timeout = RPCRDMA_BIND_TO;
 352         xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
 353         xprt->idle_timeout = RPCRDMA_IDLE_DISC_TO;
 354
 355         xprt->resvport = 0;             /* privileged port not needed */
 356         xprt->tsh_size = 0;             /* RPC-RDMA handles framing */
 357         xprt->ops = &xprt_rdma_procs;
 358
 359         /*
 360          * Set up RDMA-specific connect data.
 361          */
 362
 363         sap = (struct sockaddr *)&cdata.addr;
 364         memcpy(sap, args->dstaddr, args->addrlen);
 365
 366         /* Ensure xprt->addr holds valid server TCP (not RDMA)
 367          * address, for any side protocols which peek at it */
 368         xprt->prot = IPPROTO_TCP;
 369         xprt->addrlen = args->addrlen;
 370         memcpy(&xprt->addr, sap, xprt->addrlen);
 371
 372         if (rpc_get_port(sap))
 373                 xprt_set_bound(xprt);
 374
 375         cdata.max_requests = xprt->max_reqs;
 376
 377         cdata.rsize = RPCRDMA_MAX_SEGS * PAGE_SIZE; /* RDMA write max */
 378         cdata.wsize = RPCRDMA_MAX_SEGS * PAGE_SIZE; /* RDMA read max */
 379
 380         cdata.inline_wsize = xprt_rdma_max_inline_write;
 381         if (cdata.inline_wsize > cdata.wsize)
 382                 cdata.inline_wsize = cdata.wsize;
 383
 384         cdata.inline_rsize = xprt_rdma_max_inline_read;
 385         if (cdata.inline_rsize > cdata.rsize)
 386                 cdata.inline_rsize = cdata.rsize;
 387
 388         cdata.padding = xprt_rdma_inline_write_padding;
 389
 390         /*
 391          * Create new transport instance, which includes initialized
 392          *  o ia
 393          *  o endpoint
 394          *  o buffers
 395          */
 396
 397         new_xprt = rpcx_to_rdmax(xprt);
 398
 399         rc = rpcrdma_ia_open(new_xprt, sap);
 400         if (rc)
 401                 goto out1;
 402
 403         /*
 404          * initialize and create ep
 405          */
 406         new_xprt->rx_data = cdata;
 407         new_ep = &new_xprt->rx_ep;
 408         new_ep->rep_remote_addr = cdata.addr;
 409
 410         rc = rpcrdma_ep_create(&new_xprt->rx_ep,
 411                                 &new_xprt->rx_ia, &new_xprt->rx_data);
 412         if (rc)
 413                 goto out2;
 414
 415         /*
 416          * Allocate pre-registered send and receive buffers for headers and
 417          * any inline data. Also specify any padding which will be provided
 418          * from a preregistered zero buffer.
 419          */
 420         rc = rpcrdma_buffer_create(new_xprt);
 421         if (rc)
 422                 goto out3;
 423
 424         /*
 425          * Register a callback for connection events. This is necessary because
 426          * connection loss notification is async. We also catch connection loss
 427          * when reaping receives.
 428          */
 429         INIT_DELAYED_WORK(&new_xprt->rx_connect_worker,
 430                           xprt_rdma_connect_worker);
 431
 432         xprt_rdma_format_addresses(xprt, sap);
 433         xprt->max_payload = new_xprt->rx_ia.ri_ops->ro_maxpages(new_xprt);
 434         if (xprt->max_payload == 0)
 435                 goto out4;
 436         xprt->max_payload <<= PAGE_SHIFT;
 437         dprintk("RPC:       %s: transport data payload maximum: %zu bytes\n",
 438                 __func__, xprt->max_payload);
 439
 440         if (!try_module_get(THIS_MODULE))
 441                 goto out4;
 442
 443         dprintk("RPC:       %s: %s:%s\n", __func__,
 444                 xprt->address_strings[RPC_DISPLAY_ADDR],
 445                 xprt->address_strings[RPC_DISPLAY_PORT]);
 446         return xprt;
 447
 448 out4:
 449         xprt_rdma_free_addresses(xprt);
 450         rc = -EINVAL;
 451 out3:
 452         rpcrdma_ep_destroy(new_ep, &new_xprt->rx_ia);
 453 out2:
 454         rpcrdma_ia_close(&new_xprt->rx_ia);
 455 out1:
 456         xprt_free(xprt);
 457         return ERR_PTR(rc);
 458 }
 459
 460 /**
 461  * xprt_rdma_close - Close down RDMA connection
 462  * @xprt: generic transport to be closed
 463  *
 464  * Called during transport shutdown reconnect, or device
 465  * removal. Caller holds the transport's write lock.
 466  */
 467 static void
 468 xprt_rdma_close(struct rpc_xprt *xprt)
 469 {
 470         struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 471         struct rpcrdma_ep *ep = &r_xprt->rx_ep;
 472         struct rpcrdma_ia *ia = &r_xprt->rx_ia;
 473
 474         dprintk("RPC:       %s: closing xprt %p\n", __func__, xprt);
 475
 476         if (test_and_clear_bit(RPCRDMA_IAF_REMOVING, &ia->ri_flags)) {
 477                 xprt_clear_connected(xprt);
 478                 rpcrdma_ia_remove(ia);
 479                 return;
 480         }
 481         if (ep->rep_connected == -ENODEV)
 482                 return;
 483         if (ep->rep_connected > 0)
 484                 xprt->reestablish_timeout = 0;
 485         xprt_disconnect_done(xprt);
 486         rpcrdma_ep_disconnect(ep, ia);
 487 }
 488
 489 static void
 490 xprt_rdma_set_port(struct rpc_xprt *xprt, u16 port)
 491 {
 492         struct sockaddr_in *sap;
 493
 494         sap = (struct sockaddr_in *)&xprt->addr;
 495         sap->sin_port = htons(port);
 496         sap = (struct sockaddr_in *)&rpcx_to_rdmad(xprt).addr;
 497         sap->sin_port = htons(port);
 498         dprintk("RPC:       %s: %u\n", __func__, port);
 499 }
 500
 501 /**
 502  * xprt_rdma_timer - invoked when an RPC times out
 503  * @xprt: controlling RPC transport
 504  * @task: RPC task that timed out
 505  *
 506  * Invoked when the transport is still connected, but an RPC
 507  * retransmit timeout occurs.
 508  *
 509  * Since RDMA connections don't have a keep-alive, forcibly
 510  * disconnect and retry to connect. This drives full
 511  * detection of the network path, and retransmissions of
 512  * all pending RPCs.
 513  */
 514 static void
 515 xprt_rdma_timer(struct rpc_xprt *xprt, struct rpc_task *task)
 516 {
 517         dprintk("RPC: %5u %s: xprt = %p\n", task->tk_pid, __func__, xprt);
 518
 519         xprt_force_disconnect(xprt);
 520 }
 521
 522 static void
 523 xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task)
 524 {
 525         struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 526
 527         if (r_xprt->rx_ep.rep_connected != 0) {
 528                 /* Reconnect */
 529                 schedule_delayed_work(&r_xprt->rx_connect_worker,
 530                                       xprt->reestablish_timeout);
 531                 xprt->reestablish_timeout <<= 1;
 532                 if (xprt->reestablish_timeout > RPCRDMA_MAX_REEST_TO)
 533                         xprt->reestablish_timeout = RPCRDMA_MAX_REEST_TO;
 534                 else if (xprt->reestablish_timeout < RPCRDMA_INIT_REEST_TO)
 535                         xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
 536         } else {
 537                 schedule_delayed_work(&r_xprt->rx_connect_worker, 0);
 538                 if (!RPC_IS_ASYNC(task))
 539                         flush_delayed_work(&r_xprt->rx_connect_worker);
 540         }
 541 }
 542
 543 /* Allocate a fixed-size buffer in which to construct and send the
 544  * RPC-over-RDMA header for this request.
 545  */
 546 static bool
 547 rpcrdma_get_rdmabuf(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req,
 548                     gfp_t flags)
 549 {
 550         size_t size = RPCRDMA_HDRBUF_SIZE;
 551         struct rpcrdma_regbuf *rb;
 552
 553         if (req->rl_rdmabuf)
 554                 return true;
 555
 556         rb = rpcrdma_alloc_regbuf(size, DMA_TO_DEVICE, flags);
 557         if (IS_ERR(rb))
 558                 return false;
 559
 560         r_xprt->rx_stats.hardway_register_count += size;
 561         req->rl_rdmabuf = rb;
 562         return true;
 563 }
 564
 565 static bool
 566 rpcrdma_get_sendbuf(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req,
 567                     size_t size, gfp_t flags)
 568 {
 569         struct rpcrdma_regbuf *rb;
 570
 571         if (req->rl_sendbuf && rdmab_length(req->rl_sendbuf) >= size)
 572                 return true;
 573
 574         rb = rpcrdma_alloc_regbuf(size, DMA_TO_DEVICE, flags);
 575         if (IS_ERR(rb))
 576                 return false;
 577
 578         rpcrdma_free_regbuf(req->rl_sendbuf);
 579         r_xprt->rx_stats.hardway_register_count += size;
 580         req->rl_sendbuf = rb;
 581         return true;
 582 }
 583
 584 /* The rq_rcv_buf is used only if a Reply chunk is necessary.
 585  * The decision to use a Reply chunk is made later in
 586  * rpcrdma_marshal_req. This buffer is registered at that time.
 587  *
 588  * Otherwise, the associated RPC Reply arrives in a separate
 589  * Receive buffer, arbitrarily chosen by the HCA. The buffer
 590  * allocated here for the RPC Reply is not utilized in that
 591  * case. See rpcrdma_inline_fixup.
 592  *
 593  * A regbuf is used here to remember the buffer size.
 594  */
 595 static bool
 596 rpcrdma_get_recvbuf(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req,
 597                     size_t size, gfp_t flags)
 598 {
 599         struct rpcrdma_regbuf *rb;
 600
 601         if (req->rl_recvbuf && rdmab_length(req->rl_recvbuf) >= size)
 602                 return true;
 603
 604         rb = rpcrdma_alloc_regbuf(size, DMA_NONE, flags);
 605         if (IS_ERR(rb))
 606                 return false;
 607
 608         rpcrdma_free_regbuf(req->rl_recvbuf);
 609         r_xprt->rx_stats.hardway_register_count += size;
 610         req->rl_recvbuf = rb;
 611         return true;
 612 }
 613
 614 /**
 615  * xprt_rdma_allocate - allocate transport resources for an RPC
 616  * @task: RPC task
 617  *
 618  * Return values:
 619  *        0:    Success; rq_buffer points to RPC buffer to use
 620  *   ENOMEM:    Out of memory, call again later
 621  *      EIO:    A permanent error occurred, do not retry
 622  *
 623  * The RDMA allocate/free functions need the task structure as a place
 624  * to hide the struct rpcrdma_req, which is necessary for the actual
 625  * send/recv sequence.
 626  *
 627  * xprt_rdma_allocate provides buffers that are already mapped for
 628  * DMA, and a local DMA lkey is provided for each.
 629  */
 630 static int
 631 xprt_rdma_allocate(struct rpc_task *task)
 632 {
 633         struct rpc_rqst *rqst = task->tk_rqstp;
 634         struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt);
 635         struct rpcrdma_req *req;
 636         gfp_t flags;
 637
 638         req = rpcrdma_buffer_get(&r_xprt->rx_buf);
 639         if (req == NULL)
 640                 return -ENOMEM;
 641
 642         flags = RPCRDMA_DEF_GFP;
 643         if (RPC_IS_SWAPPER(task))
 644                 flags = __GFP_MEMALLOC | GFP_NOWAIT | __GFP_NOWARN;
 645
 646         if (!rpcrdma_get_rdmabuf(r_xprt, req, flags))
 647                 goto out_fail;
 648         if (!rpcrdma_get_sendbuf(r_xprt, req, rqst->rq_callsize, flags))
 649                 goto out_fail;
 650         if (!rpcrdma_get_recvbuf(r_xprt, req, rqst->rq_rcvsize, flags))
 651                 goto out_fail;
 652
 653         dprintk("RPC: %5u %s: send size = %zd, recv size = %zd, req = %p\n",
 654                 task->tk_pid, __func__, rqst->rq_callsize,
 655                 rqst->rq_rcvsize, req);
 656
 657         req->rl_connect_cookie = 0;     /* our reserved value */
 658         rpcrdma_set_xprtdata(rqst, req);
 659         rqst->rq_buffer = req->rl_sendbuf->rg_base;
 660         rqst->rq_rbuffer = req->rl_recvbuf->rg_base;
 661         return 0;
 662
 663 out_fail:
 664         rpcrdma_buffer_put(req);
 665         return -ENOMEM;
 666 }
 667
 668 /**
 669  * xprt_rdma_free - release resources allocated by xprt_rdma_allocate
 670  * @task: RPC task
 671  *
 672  * Caller guarantees rqst->rq_buffer is non-NULL.
 673  */
 674 static void
 675 xprt_rdma_free(struct rpc_task *task)
 676 {
 677         struct rpc_rqst *rqst = task->tk_rqstp;
 678         struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt);
 679         struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
 680         struct rpcrdma_ia *ia = &r_xprt->rx_ia;
 681
 682         if (req->rl_backchannel)
 683                 return;
 684
 685         dprintk("RPC:       %s: called on 0x%p\n", __func__, req->rl_reply);
 686
 687         if (unlikely(!list_empty(&req->rl_registered)))
 688                 ia->ri_ops->ro_unmap_safe(r_xprt, req, !RPC_IS_ASYNC(task));
 689         rpcrdma_unmap_sges(ia, req);
 690         rpcrdma_buffer_put(req);
 691 }
 692
 693 /**
 694  * xprt_rdma_send_request - marshal and send an RPC request
 695  * @task: RPC task with an RPC message in rq_snd_buf
 696  *
 697  * Caller holds the transport's write lock.
 698  *
 699  * Return values:
 700  *        0:    The request has been sent
 701  * ENOTCONN:    Caller needs to invoke connect logic then call again
 702  *  ENOBUFS:    Call again later to send the request
 703  *      EIO:    A permanent error occurred. The request was not sent,
 704  *              and don't try it again
 705  *
 706  * send_request invokes the meat of RPC RDMA. It must do the following:
 707  *
 708  *  1.  Marshal the RPC request into an RPC RDMA request, which means
 709  *      putting a header in front of data, and creating IOVs for RDMA
 710  *      from those in the request.
 711  *  2.  In marshaling, detect opportunities for RDMA, and use them.
 712  *  3.  Post a recv message to set up asynch completion, then send
 713  *      the request (rpcrdma_ep_post).
 714  *  4.  No partial sends are possible in the RPC-RDMA protocol (as in UDP).
 715  */
 716 static int
 717 xprt_rdma_send_request(struct rpc_task *task)
 718 {
 719         struct rpc_rqst *rqst = task->tk_rqstp;
 720         struct rpc_xprt *xprt = rqst->rq_xprt;
 721         struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
 722         struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 723         int rc = 0;
 724
 725         if (!xprt_connected(xprt))
 726                 goto drop_connection;
 727
 728         /* On retransmit, remove any previously registered chunks */
 729         if (unlikely(!list_empty(&req->rl_registered)))
 730                 r_xprt->rx_ia.ri_ops->ro_unmap_safe(r_xprt, req, false);
 731
 732         rc = rpcrdma_marshal_req(rqst);
 733         if (rc < 0)
 734                 goto failed_marshal;
 735
 736         if (req->rl_reply == NULL)              /* e.g. reconnection */
 737                 rpcrdma_recv_buffer_get(req);
 738
 739         /* Must suppress retransmit to maintain credits */
 740         if (req->rl_connect_cookie == xprt->connect_cookie)
 741                 goto drop_connection;
 742         req->rl_connect_cookie = xprt->connect_cookie;
 743
 744         if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req))
 745                 goto drop_connection;
 746
 747         rqst->rq_xmit_bytes_sent += rqst->rq_snd_buf.len;
 748         rqst->rq_bytes_sent = 0;
 749         return 0;
 750
 751 failed_marshal:
 752         if (rc != -ENOTCONN)
 753                 return rc;
 754 drop_connection:
 755         xprt_disconnect_done(xprt);
 756         return -ENOTCONN;       /* implies disconnect */
 757 }
 758
 759 void xprt_rdma_print_stats(struct rpc_xprt *xprt, struct seq_file *seq)
 760 {
 761         struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
 762         long idle_time = 0;
 763
 764         if (xprt_connected(xprt))
 765                 idle_time = (long)(jiffies - xprt->last_used) / HZ;
 766
 767         seq_puts(seq, "\txprt:\trdma ");
 768         seq_printf(seq, "%u %lu %lu %lu %ld %lu %lu %lu %llu %llu ",
 769                    0,   /* need a local port? */
 770                    xprt->stat.bind_count,
 771                    xprt->stat.connect_count,
 772                    xprt->stat.connect_time,
 773                    idle_time,
 774                    xprt->stat.sends,
 775                    xprt->stat.recvs,
 776                    xprt->stat.bad_xids,
 777                    xprt->stat.req_u,
 778                    xprt->stat.bklog_u);
 779         seq_printf(seq, "%lu %lu %lu %llu %llu %llu %llu %lu %lu %lu %lu ",
 780                    r_xprt->rx_stats.read_chunk_count,
 781                    r_xprt->rx_stats.write_chunk_count,
 782                    r_xprt->rx_stats.reply_chunk_count,
 783                    r_xprt->rx_stats.total_rdma_request,
 784                    r_xprt->rx_stats.total_rdma_reply,
 785                    r_xprt->rx_stats.pullup_copy_count,
 786                    r_xprt->rx_stats.fixup_copy_count,
 787                    r_xprt->rx_stats.hardway_register_count,
 788                    r_xprt->rx_stats.failed_marshal_count,
 789                    r_xprt->rx_stats.bad_reply_count,
 790                    r_xprt->rx_stats.nomsg_call_count);
 791         seq_printf(seq, "%lu %lu %lu %lu\n",
 792                    r_xprt->rx_stats.mrs_recovered,
 793                    r_xprt->rx_stats.mrs_orphaned,
 794                    r_xprt->rx_stats.mrs_allocated,
 795                    r_xprt->rx_stats.local_inv_needed);
 796 }
 797
 798 static int
 799 xprt_rdma_enable_swap(struct rpc_xprt *xprt)
 800 {
 801         return 0;
 802 }
 803
 804 static void
 805 xprt_rdma_disable_swap(struct rpc_xprt *xprt)
 806 {
 807 }
 808
 809 /*
 810  * Plumbing for rpc transport switch and kernel module
 811  */
 812
 813 static struct rpc_xprt_ops xprt_rdma_procs = {
 814         .reserve_xprt           = xprt_reserve_xprt_cong,
 815         .release_xprt           = xprt_release_xprt_cong, /* sunrpc/xprt.c */
 816         .alloc_slot             = xprt_alloc_slot,
 817         .release_request        = xprt_release_rqst_cong,       /* ditto */
 818         .set_retrans_timeout    = xprt_set_retrans_timeout_def, /* ditto */
 819         .timer                  = xprt_rdma_timer,
 820         .rpcbind                = rpcb_getport_async,   /* sunrpc/rpcb_clnt.c */
 821         .set_port               = xprt_rdma_set_port,
 822         .connect                = xprt_rdma_connect,
 823         .buf_alloc              = xprt_rdma_allocate,
 824         .buf_free               = xprt_rdma_free,
 825         .send_request           = xprt_rdma_send_request,
 826         .close                  = xprt_rdma_close,
 827         .destroy                = xprt_rdma_destroy,
 828         .print_stats            = xprt_rdma_print_stats,
 829         .enable_swap            = xprt_rdma_enable_swap,
 830         .disable_swap           = xprt_rdma_disable_swap,
 831         .inject_disconnect      = xprt_rdma_inject_disconnect,
 832 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
 833         .bc_setup               = xprt_rdma_bc_setup,
 834         .bc_up                  = xprt_rdma_bc_up,
 835         .bc_maxpayload          = xprt_rdma_bc_maxpayload,
 836         .bc_free_rqst           = xprt_rdma_bc_free_rqst,
 837         .bc_destroy             = xprt_rdma_bc_destroy,
 838 #endif
 839 };
 840
 841 static struct xprt_class xprt_rdma = {
 842         .list                   = LIST_HEAD_INIT(xprt_rdma.list),
 843         .name                   = "rdma",
 844         .owner                  = THIS_MODULE,
 845         .ident                  = XPRT_TRANSPORT_RDMA,
 846         .setup                  = xprt_setup_rdma,
 847 };
 848
 849 void xprt_rdma_cleanup(void)
 850 {
 851         int rc;
 852
 853         dprintk("RPCRDMA Module Removed, deregister RPC RDMA transport\n");
 854 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 855         if (sunrpc_table_header) {
 856                 unregister_sysctl_table(sunrpc_table_header);
 857                 sunrpc_table_header = NULL;
 858         }
 859 #endif
 860         rc = xprt_unregister_transport(&xprt_rdma);
 861         if (rc)
 862                 dprintk("RPC:       %s: xprt_unregister returned %i\n",
 863                         __func__, rc);
 864
 865         rpcrdma_destroy_wq();
 866
 867         rc = xprt_unregister_transport(&xprt_rdma_bc);
 868         if (rc)
 869                 dprintk("RPC:       %s: xprt_unregister(bc) returned %i\n",
 870                         __func__, rc);
 871 }
 872
 873 int xprt_rdma_init(void)
 874 {
 875         int rc;
 876
 877         rc = rpcrdma_alloc_wq();
 878         if (rc)
 879                 return rc;
 880
 881         rc = xprt_register_transport(&xprt_rdma);
 882         if (rc) {
 883                 rpcrdma_destroy_wq();
 884                 return rc;
 885         }
 886
 887         rc = xprt_register_transport(&xprt_rdma_bc);
 888         if (rc) {
 889                 xprt_unregister_transport(&xprt_rdma);
 890                 rpcrdma_destroy_wq();
 891                 return rc;
 892         }
 893
 894         dprintk("RPCRDMA Module Init, register RPC RDMA transport\n");
 895
 896         dprintk("Defaults:\n");
 897         dprintk("\tSlots %d\n"
 898                 "\tMaxInlineRead %d\n\tMaxInlineWrite %d\n",
 899                 xprt_rdma_slot_table_entries,
 900                 xprt_rdma_max_inline_read, xprt_rdma_max_inline_write);
 901         dprintk("\tPadding %d\n\tMemreg %d\n",
 902                 xprt_rdma_inline_write_padding, xprt_rdma_memreg_strategy);
 903
 904 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
 905         if (!sunrpc_table_header)
 906                 sunrpc_table_header = register_sysctl_table(sunrpc_table);
 907 #endif
 908         return 0;
 909 }