net/sunrpc/xprt.c

   1 // SPDX-License-Identifier: GPL-2.0-only
   2 /*
   3  *  linux/net/sunrpc/xprt.c
   4  *
   5  *  This is a generic RPC call interface supporting congestion avoidance,
   6  *  and asynchronous calls.
   7  *
   8  *  The interface works like this:
   9  *
  10  *  -   When a process places a call, it allocates a request slot if
  11  *      one is available. Otherwise, it sleeps on the backlog queue
  12  *      (xprt_reserve).
  13  *  -   Next, the caller puts together the RPC message, stuffs it into
  14  *      the request struct, and calls xprt_transmit().
  15  *  -   xprt_transmit sends the message and installs the caller on the
  16  *      transport's wait list. At the same time, if a reply is expected,
  17  *      it installs a timer that is run after the packet's timeout has
  18  *      expired.
  19  *  -   When a packet arrives, the data_ready handler walks the list of
  20  *      pending requests for that transport. If a matching XID is found, the
  21  *      caller is woken up, and the timer removed.
  22  *  -   When no reply arrives within the timeout interval, the timer is
  23  *      fired by the kernel and runs xprt_timer(). It either adjusts the
  24  *      timeout values (minor timeout) or wakes up the caller with a status
  25  *      of -ETIMEDOUT.
  26  *  -   When the caller receives a notification from RPC that a reply arrived,
  27  *      it should release the RPC slot, and process the reply.
  28  *      If the call timed out, it may choose to retry the operation by
  29  *      adjusting the initial timeout value, and simply calling rpc_call
  30  *      again.
  31  *
  32  *  Support for async RPC is done through a set of RPC-specific scheduling
  33  *  primitives that `transparently' work for processes as well as async
  34  *  tasks that rely on callbacks.
  35  *
  36  *  Copyright (C) 1995-1997, Olaf Kirch <okir@monad.swb.de>
  37  *
  38  *  Transport switch API copyright (C) 2005, Chuck Lever <cel@netapp.com>
  39  */
  40
  41 #include <linux/module.h>
  42
  43 #include <linux/types.h>
  44 #include <linux/interrupt.h>
  45 #include <linux/workqueue.h>
  46 #include <linux/net.h>
  47 #include <linux/ktime.h>
  48
  49 #include <linux/sunrpc/clnt.h>
  50 #include <linux/sunrpc/metrics.h>
  51 #include <linux/sunrpc/bc_xprt.h>
  52 #include <linux/rcupdate.h>
  53 #include <linux/sched/mm.h>
  54
  55 #include <trace/events/sunrpc.h>
  56
  57 #include "sunrpc.h"
  58
  59 /*
  60  * Local variables
  61  */
  62
  63 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
  64 # define RPCDBG_FACILITY        RPCDBG_XPRT
  65 #endif
  66
  67 /*
  68  * Local functions
  69  */
  70 static void      xprt_init(struct rpc_xprt *xprt, struct net *net);
  71 static __be32   xprt_alloc_xid(struct rpc_xprt *xprt);
  72 static void      xprt_destroy(struct rpc_xprt *xprt);
  73
  74 static DEFINE_SPINLOCK(xprt_list_lock);
  75 static LIST_HEAD(xprt_list);
  76
  77 static unsigned long xprt_request_timeout(const struct rpc_rqst *req)
  78 {
  79         unsigned long timeout = jiffies + req->rq_timeout;
  80
  81         if (time_before(timeout, req->rq_majortimeo))
  82                 return timeout;
  83         return req->rq_majortimeo;
  84 }
  85
  86 /**
  87  * xprt_register_transport - register a transport implementation
  88  * @transport: transport to register
  89  *
  90  * If a transport implementation is loaded as a kernel module, it can
  91  * call this interface to make itself known to the RPC client.
  92  *
  93  * Returns:
  94  * 0:           transport successfully registered
  95  * -EEXIST:     transport already registered
  96  * -EINVAL:     transport module being unloaded
  97  */
  98 int xprt_register_transport(struct xprt_class *transport)
  99 {
 100         struct xprt_class *t;
 101         int result;
 102
 103         result = -EEXIST;
 104         spin_lock(&xprt_list_lock);
 105         list_for_each_entry(t, &xprt_list, list) {
 106                 /* don't register the same transport class twice */
 107                 if (t->ident == transport->ident)
 108                         goto out;
 109         }
 110
 111         list_add_tail(&transport->list, &xprt_list);
 112         printk(KERN_INFO "RPC: Registered %s transport module.\n",
 113                transport->name);
 114         result = 0;
 115
 116 out:
 117         spin_unlock(&xprt_list_lock);
 118         return result;
 119 }
 120 EXPORT_SYMBOL_GPL(xprt_register_transport);
 121
 122 /**
 123  * xprt_unregister_transport - unregister a transport implementation
 124  * @transport: transport to unregister
 125  *
 126  * Returns:
 127  * 0:           transport successfully unregistered
 128  * -ENOENT:     transport never registered
 129  */
 130 int xprt_unregister_transport(struct xprt_class *transport)
 131 {
 132         struct xprt_class *t;
 133         int result;
 134
 135         result = 0;
 136         spin_lock(&xprt_list_lock);
 137         list_for_each_entry(t, &xprt_list, list) {
 138                 if (t == transport) {
 139                         printk(KERN_INFO
 140                                 "RPC: Unregistered %s transport module.\n",
 141                                 transport->name);
 142                         list_del_init(&transport->list);
 143                         goto out;
 144                 }
 145         }
 146         result = -ENOENT;
 147
 148 out:
 149         spin_unlock(&xprt_list_lock);
 150         return result;
 151 }
 152 EXPORT_SYMBOL_GPL(xprt_unregister_transport);
 153
 154 /**
 155  * xprt_load_transport - load a transport implementation
 156  * @transport_name: transport to load
 157  *
 158  * Returns:
 159  * 0:           transport successfully loaded
 160  * -ENOENT:     transport module not available
 161  */
 162 int xprt_load_transport(const char *transport_name)
 163 {
 164         struct xprt_class *t;
 165         int result;
 166
 167         result = 0;
 168         spin_lock(&xprt_list_lock);
 169         list_for_each_entry(t, &xprt_list, list) {
 170                 if (strcmp(t->name, transport_name) == 0) {
 171                         spin_unlock(&xprt_list_lock);
 172                         goto out;
 173                 }
 174         }
 175         spin_unlock(&xprt_list_lock);
 176         result = request_module("xprt%s", transport_name);
 177 out:
 178         return result;
 179 }
 180 EXPORT_SYMBOL_GPL(xprt_load_transport);
 181
 182 static void xprt_clear_locked(struct rpc_xprt *xprt)
 183 {
 184         xprt->snd_task = NULL;
 185         if (!test_bit(XPRT_CLOSE_WAIT, &xprt->state)) {
 186                 smp_mb__before_atomic();
 187                 clear_bit(XPRT_LOCKED, &xprt->state);
 188                 smp_mb__after_atomic();
 189         } else
 190                 queue_work(xprtiod_workqueue, &xprt->task_cleanup);
 191 }
 192
 193 /**
 194  * xprt_reserve_xprt - serialize write access to transports
 195  * @task: task that is requesting access to the transport
 196  * @xprt: pointer to the target transport
 197  *
 198  * This prevents mixing the payload of separate requests, and prevents
 199  * transport connects from colliding with writes.  No congestion control
 200  * is provided.
 201  */
 202 int xprt_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
 203 {
 204         struct rpc_rqst *req = task->tk_rqstp;
 205
 206         if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) {
 207                 if (task == xprt->snd_task)
 208                         goto out_locked;
 209                 goto out_sleep;
 210         }
 211         if (test_bit(XPRT_WRITE_SPACE, &xprt->state))
 212                 goto out_unlock;
 213         xprt->snd_task = task;
 214
 215 out_locked:
 216         trace_xprt_reserve_xprt(xprt, task);
 217         return 1;
 218
 219 out_unlock:
 220         xprt_clear_locked(xprt);
 221 out_sleep:
 222         task->tk_status = -EAGAIN;
 223         if  (RPC_IS_SOFT(task))
 224                 rpc_sleep_on_timeout(&xprt->sending, task, NULL,
 225                                 xprt_request_timeout(req));
 226         else
 227                 rpc_sleep_on(&xprt->sending, task, NULL);
 228         return 0;
 229 }
 230 EXPORT_SYMBOL_GPL(xprt_reserve_xprt);
 231
 232 static bool
 233 xprt_need_congestion_window_wait(struct rpc_xprt *xprt)
 234 {
 235         return test_bit(XPRT_CWND_WAIT, &xprt->state);
 236 }
 237
 238 static void
 239 xprt_set_congestion_window_wait(struct rpc_xprt *xprt)
 240 {
 241         if (!list_empty(&xprt->xmit_queue)) {
 242                 /* Peek at head of queue to see if it can make progress */
 243                 if (list_first_entry(&xprt->xmit_queue, struct rpc_rqst,
 244                                         rq_xmit)->rq_cong)
 245                         return;
 246         }
 247         set_bit(XPRT_CWND_WAIT, &xprt->state);
 248 }
 249
 250 static void
 251 xprt_test_and_clear_congestion_window_wait(struct rpc_xprt *xprt)
 252 {
 253         if (!RPCXPRT_CONGESTED(xprt))
 254                 clear_bit(XPRT_CWND_WAIT, &xprt->state);
 255 }
 256
 257 /*
 258  * xprt_reserve_xprt_cong - serialize write access to transports
 259  * @task: task that is requesting access to the transport
 260  *
 261  * Same as xprt_reserve_xprt, but Van Jacobson congestion control is
 262  * integrated into the decision of whether a request is allowed to be
 263  * woken up and given access to the transport.
 264  * Note that the lock is only granted if we know there are free slots.
 265  */
 266 int xprt_reserve_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
 267 {
 268         struct rpc_rqst *req = task->tk_rqstp;
 269
 270         if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) {
 271                 if (task == xprt->snd_task)
 272                         goto out_locked;
 273                 goto out_sleep;
 274         }
 275         if (req == NULL) {
 276                 xprt->snd_task = task;
 277                 goto out_locked;
 278         }
 279         if (test_bit(XPRT_WRITE_SPACE, &xprt->state))
 280                 goto out_unlock;
 281         if (!xprt_need_congestion_window_wait(xprt)) {
 282                 xprt->snd_task = task;
 283                 goto out_locked;
 284         }
 285 out_unlock:
 286         xprt_clear_locked(xprt);
 287 out_sleep:
 288         task->tk_status = -EAGAIN;
 289         if (RPC_IS_SOFT(task))
 290                 rpc_sleep_on_timeout(&xprt->sending, task, NULL,
 291                                 xprt_request_timeout(req));
 292         else
 293                 rpc_sleep_on(&xprt->sending, task, NULL);
 294         return 0;
 295 out_locked:
 296         trace_xprt_reserve_cong(xprt, task);
 297         return 1;
 298 }
 299 EXPORT_SYMBOL_GPL(xprt_reserve_xprt_cong);
 300
 301 static inline int xprt_lock_write(struct rpc_xprt *xprt, struct rpc_task *task)
 302 {
 303         int retval;
 304
 305         if (test_bit(XPRT_LOCKED, &xprt->state) && xprt->snd_task == task)
 306                 return 1;
 307         spin_lock(&xprt->transport_lock);
 308         retval = xprt->ops->reserve_xprt(xprt, task);
 309         spin_unlock(&xprt->transport_lock);
 310         return retval;
 311 }
 312
 313 static bool __xprt_lock_write_func(struct rpc_task *task, void *data)
 314 {
 315         struct rpc_xprt *xprt = data;
 316
 317         xprt->snd_task = task;
 318         return true;
 319 }
 320
 321 static void __xprt_lock_write_next(struct rpc_xprt *xprt)
 322 {
 323         if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
 324                 return;
 325         if (test_bit(XPRT_WRITE_SPACE, &xprt->state))
 326                 goto out_unlock;
 327         if (rpc_wake_up_first_on_wq(xprtiod_workqueue, &xprt->sending,
 328                                 __xprt_lock_write_func, xprt))
 329                 return;
 330 out_unlock:
 331         xprt_clear_locked(xprt);
 332 }
 333
 334 static void __xprt_lock_write_next_cong(struct rpc_xprt *xprt)
 335 {
 336         if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
 337                 return;
 338         if (test_bit(XPRT_WRITE_SPACE, &xprt->state))
 339                 goto out_unlock;
 340         if (xprt_need_congestion_window_wait(xprt))
 341                 goto out_unlock;
 342         if (rpc_wake_up_first_on_wq(xprtiod_workqueue, &xprt->sending,
 343                                 __xprt_lock_write_func, xprt))
 344                 return;
 345 out_unlock:
 346         xprt_clear_locked(xprt);
 347 }
 348
 349 /**
 350  * xprt_release_xprt - allow other requests to use a transport
 351  * @xprt: transport with other tasks potentially waiting
 352  * @task: task that is releasing access to the transport
 353  *
 354  * Note that "task" can be NULL.  No congestion control is provided.
 355  */
 356 void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
 357 {
 358         if (xprt->snd_task == task) {
 359                 xprt_clear_locked(xprt);
 360                 __xprt_lock_write_next(xprt);
 361         }
 362         trace_xprt_release_xprt(xprt, task);
 363 }
 364 EXPORT_SYMBOL_GPL(xprt_release_xprt);
 365
 366 /**
 367  * xprt_release_xprt_cong - allow other requests to use a transport
 368  * @xprt: transport with other tasks potentially waiting
 369  * @task: task that is releasing access to the transport
 370  *
 371  * Note that "task" can be NULL.  Another task is awoken to use the
 372  * transport if the transport's congestion window allows it.
 373  */
 374 void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
 375 {
 376         if (xprt->snd_task == task) {
 377                 xprt_clear_locked(xprt);
 378                 __xprt_lock_write_next_cong(xprt);
 379         }
 380         trace_xprt_release_cong(xprt, task);
 381 }
 382 EXPORT_SYMBOL_GPL(xprt_release_xprt_cong);
 383
 384 static inline void xprt_release_write(struct rpc_xprt *xprt, struct rpc_task *task)
 385 {
 386         if (xprt->snd_task != task)
 387                 return;
 388         spin_lock(&xprt->transport_lock);
 389         xprt->ops->release_xprt(xprt, task);
 390         spin_unlock(&xprt->transport_lock);
 391 }
 392
 393 /*
 394  * Van Jacobson congestion avoidance. Check if the congestion window
 395  * overflowed. Put the task to sleep if this is the case.
 396  */
 397 static int
 398 __xprt_get_cong(struct rpc_xprt *xprt, struct rpc_rqst *req)
 399 {
 400         if (req->rq_cong)
 401                 return 1;
 402         trace_xprt_get_cong(xprt, req->rq_task);
 403         if (RPCXPRT_CONGESTED(xprt)) {
 404                 xprt_set_congestion_window_wait(xprt);
 405                 return 0;
 406         }
 407         req->rq_cong = 1;
 408         xprt->cong += RPC_CWNDSCALE;
 409         return 1;
 410 }
 411
 412 /*
 413  * Adjust the congestion window, and wake up the next task
 414  * that has been sleeping due to congestion
 415  */
 416 static void
 417 __xprt_put_cong(struct rpc_xprt *xprt, struct rpc_rqst *req)
 418 {
 419         if (!req->rq_cong)
 420                 return;
 421         req->rq_cong = 0;
 422         xprt->cong -= RPC_CWNDSCALE;
 423         xprt_test_and_clear_congestion_window_wait(xprt);
 424         trace_xprt_put_cong(xprt, req->rq_task);
 425         __xprt_lock_write_next_cong(xprt);
 426 }
 427
 428 /**
 429  * xprt_request_get_cong - Request congestion control credits
 430  * @xprt: pointer to transport
 431  * @req: pointer to RPC request
 432  *
 433  * Useful for transports that require congestion control.
 434  */
 435 bool
 436 xprt_request_get_cong(struct rpc_xprt *xprt, struct rpc_rqst *req)
 437 {
 438         bool ret = false;
 439
 440         if (req->rq_cong)
 441                 return true;
 442         spin_lock(&xprt->transport_lock);
 443         ret = __xprt_get_cong(xprt, req) != 0;
 444         spin_unlock(&xprt->transport_lock);
 445         return ret;
 446 }
 447 EXPORT_SYMBOL_GPL(xprt_request_get_cong);
 448
 449 /**
 450  * xprt_release_rqst_cong - housekeeping when request is complete
 451  * @task: RPC request that recently completed
 452  *
 453  * Useful for transports that require congestion control.
 454  */
 455 void xprt_release_rqst_cong(struct rpc_task *task)
 456 {
 457         struct rpc_rqst *req = task->tk_rqstp;
 458
 459         __xprt_put_cong(req->rq_xprt, req);
 460 }
 461 EXPORT_SYMBOL_GPL(xprt_release_rqst_cong);
 462
 463 static void xprt_clear_congestion_window_wait_locked(struct rpc_xprt *xprt)
 464 {
 465         if (test_and_clear_bit(XPRT_CWND_WAIT, &xprt->state))
 466                 __xprt_lock_write_next_cong(xprt);
 467 }
 468
 469 /*
 470  * Clear the congestion window wait flag and wake up the next
 471  * entry on xprt->sending
 472  */
 473 static void
 474 xprt_clear_congestion_window_wait(struct rpc_xprt *xprt)
 475 {
 476         if (test_and_clear_bit(XPRT_CWND_WAIT, &xprt->state)) {
 477                 spin_lock(&xprt->transport_lock);
 478                 __xprt_lock_write_next_cong(xprt);
 479                 spin_unlock(&xprt->transport_lock);
 480         }
 481 }
 482
 483 /**
 484  * xprt_adjust_cwnd - adjust transport congestion window
 485  * @xprt: pointer to xprt
 486  * @task: recently completed RPC request used to adjust window
 487  * @result: result code of completed RPC request
 488  *
 489  * The transport code maintains an estimate on the maximum number of out-
 490  * standing RPC requests, using a smoothed version of the congestion
 491  * avoidance implemented in 44BSD. This is basically the Van Jacobson
 492  * congestion algorithm: If a retransmit occurs, the congestion window is
 493  * halved; otherwise, it is incremented by 1/cwnd when
 494  *
 495  *      -       a reply is received and
 496  *      -       a full number of requests are outstanding and
 497  *      -       the congestion window hasn't been updated recently.
 498  */
 499 void xprt_adjust_cwnd(struct rpc_xprt *xprt, struct rpc_task *task, int result)
 500 {
 501         struct rpc_rqst *req = task->tk_rqstp;
 502         unsigned long cwnd = xprt->cwnd;
 503
 504         if (result >= 0 && cwnd <= xprt->cong) {
 505                 /* The (cwnd >> 1) term makes sure
 506                  * the result gets rounded properly. */
 507                 cwnd += (RPC_CWNDSCALE * RPC_CWNDSCALE + (cwnd >> 1)) / cwnd;
 508                 if (cwnd > RPC_MAXCWND(xprt))
 509                         cwnd = RPC_MAXCWND(xprt);
 510                 __xprt_lock_write_next_cong(xprt);
 511         } else if (result == -ETIMEDOUT) {
 512                 cwnd >>= 1;
 513                 if (cwnd < RPC_CWNDSCALE)
 514                         cwnd = RPC_CWNDSCALE;
 515         }
 516         dprintk("RPC:       cong %ld, cwnd was %ld, now %ld\n",
 517                         xprt->cong, xprt->cwnd, cwnd);
 518         xprt->cwnd = cwnd;
 519         __xprt_put_cong(xprt, req);
 520 }
 521 EXPORT_SYMBOL_GPL(xprt_adjust_cwnd);
 522
 523 /**
 524  * xprt_wake_pending_tasks - wake all tasks on a transport's pending queue
 525  * @xprt: transport with waiting tasks
 526  * @status: result code to plant in each task before waking it
 527  *
 528  */
 529 void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status)
 530 {
 531         if (status < 0)
 532                 rpc_wake_up_status(&xprt->pending, status);
 533         else
 534                 rpc_wake_up(&xprt->pending);
 535 }
 536 EXPORT_SYMBOL_GPL(xprt_wake_pending_tasks);
 537
 538 /**
 539  * xprt_wait_for_buffer_space - wait for transport output buffer to clear
 540  * @xprt: transport
 541  *
 542  * Note that we only set the timer for the case of RPC_IS_SOFT(), since
 543  * we don't in general want to force a socket disconnection due to
 544  * an incomplete RPC call transmission.
 545  */
 546 void xprt_wait_for_buffer_space(struct rpc_xprt *xprt)
 547 {
 548         set_bit(XPRT_WRITE_SPACE, &xprt->state);
 549 }
 550 EXPORT_SYMBOL_GPL(xprt_wait_for_buffer_space);
 551
 552 static bool
 553 xprt_clear_write_space_locked(struct rpc_xprt *xprt)
 554 {
 555         if (test_and_clear_bit(XPRT_WRITE_SPACE, &xprt->state)) {
 556                 __xprt_lock_write_next(xprt);
 557                 dprintk("RPC:       write space: waking waiting task on "
 558                                 "xprt %p\n", xprt);
 559                 return true;
 560         }
 561         return false;
 562 }
 563
 564 /**
 565  * xprt_write_space - wake the task waiting for transport output buffer space
 566  * @xprt: transport with waiting tasks
 567  *
 568  * Can be called in a soft IRQ context, so xprt_write_space never sleeps.
 569  */
 570 bool xprt_write_space(struct rpc_xprt *xprt)
 571 {
 572         bool ret;
 573
 574         if (!test_bit(XPRT_WRITE_SPACE, &xprt->state))
 575                 return false;
 576         spin_lock(&xprt->transport_lock);
 577         ret = xprt_clear_write_space_locked(xprt);
 578         spin_unlock(&xprt->transport_lock);
 579         return ret;
 580 }
 581 EXPORT_SYMBOL_GPL(xprt_write_space);
 582
 583 static unsigned long xprt_abs_ktime_to_jiffies(ktime_t abstime)
 584 {
 585         s64 delta = ktime_to_ns(ktime_get() - abstime);
 586         return likely(delta >= 0) ?
 587                 jiffies - nsecs_to_jiffies(delta) :
 588                 jiffies + nsecs_to_jiffies(-delta);
 589 }
 590
 591 static unsigned long xprt_calc_majortimeo(struct rpc_rqst *req)
 592 {
 593         const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
 594         unsigned long majortimeo = req->rq_timeout;
 595
 596         if (to->to_exponential)
 597                 majortimeo <<= to->to_retries;
 598         else
 599                 majortimeo += to->to_increment * to->to_retries;
 600         if (majortimeo > to->to_maxval || majortimeo == 0)
 601                 majortimeo = to->to_maxval;
 602         return majortimeo;
 603 }
 604
 605 static void xprt_reset_majortimeo(struct rpc_rqst *req)
 606 {
 607         req->rq_majortimeo += xprt_calc_majortimeo(req);
 608 }
 609
 610 static void xprt_reset_minortimeo(struct rpc_rqst *req)
 611 {
 612         req->rq_minortimeo += req->rq_timeout;
 613 }
 614
 615 static void xprt_init_majortimeo(struct rpc_task *task, struct rpc_rqst *req)
 616 {
 617         unsigned long time_init;
 618         struct rpc_xprt *xprt = req->rq_xprt;
 619
 620         if (likely(xprt && xprt_connected(xprt)))
 621                 time_init = jiffies;
 622         else
 623                 time_init = xprt_abs_ktime_to_jiffies(task->tk_start);
 624         req->rq_timeout = task->tk_client->cl_timeout->to_initval;
 625         req->rq_majortimeo = time_init + xprt_calc_majortimeo(req);
 626         req->rq_minortimeo = time_init + req->rq_timeout;
 627 }
 628
 629 /**
 630  * xprt_adjust_timeout - adjust timeout values for next retransmit
 631  * @req: RPC request containing parameters to use for the adjustment
 632  *
 633  */
 634 int xprt_adjust_timeout(struct rpc_rqst *req)
 635 {
 636         struct rpc_xprt *xprt = req->rq_xprt;
 637         const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
 638         int status = 0;
 639
 640         if (time_before(jiffies, req->rq_minortimeo))
 641                 return status;
 642         if (time_before(jiffies, req->rq_majortimeo)) {
 643                 if (to->to_exponential)
 644                         req->rq_timeout <<= 1;
 645                 else
 646                         req->rq_timeout += to->to_increment;
 647                 if (to->to_maxval && req->rq_timeout >= to->to_maxval)
 648                         req->rq_timeout = to->to_maxval;
 649                 req->rq_retries++;
 650         } else {
 651                 req->rq_timeout = to->to_initval;
 652                 req->rq_retries = 0;
 653                 xprt_reset_majortimeo(req);
 654                 /* Reset the RTT counters == "slow start" */
 655                 spin_lock(&xprt->transport_lock);
 656                 rpc_init_rtt(req->rq_task->tk_client->cl_rtt, to->to_initval);
 657                 spin_unlock(&xprt->transport_lock);
 658                 status = -ETIMEDOUT;
 659         }
 660         xprt_reset_minortimeo(req);
 661
 662         if (req->rq_timeout == 0) {
 663                 printk(KERN_WARNING "xprt_adjust_timeout: rq_timeout = 0!\n");
 664                 req->rq_timeout = 5 * HZ;
 665         }
 666         return status;
 667 }
 668
 669 static void xprt_autoclose(struct work_struct *work)
 670 {
 671         struct rpc_xprt *xprt =
 672                 container_of(work, struct rpc_xprt, task_cleanup);
 673         unsigned int pflags = memalloc_nofs_save();
 674
 675         trace_xprt_disconnect_auto(xprt);
 676         clear_bit(XPRT_CLOSE_WAIT, &xprt->state);
 677         xprt->ops->close(xprt);
 678         xprt_release_write(xprt, NULL);
 679         wake_up_bit(&xprt->state, XPRT_LOCKED);
 680         memalloc_nofs_restore(pflags);
 681 }
 682
 683 /**
 684  * xprt_disconnect_done - mark a transport as disconnected
 685  * @xprt: transport to flag for disconnect
 686  *
 687  */
 688 void xprt_disconnect_done(struct rpc_xprt *xprt)
 689 {
 690         trace_xprt_disconnect_done(xprt);
 691         spin_lock(&xprt->transport_lock);
 692         xprt_clear_connected(xprt);
 693         xprt_clear_write_space_locked(xprt);
 694         xprt_clear_congestion_window_wait_locked(xprt);
 695         xprt_wake_pending_tasks(xprt, -ENOTCONN);
 696         spin_unlock(&xprt->transport_lock);
 697 }
 698 EXPORT_SYMBOL_GPL(xprt_disconnect_done);
 699
 700 /**
 701  * xprt_force_disconnect - force a transport to disconnect
 702  * @xprt: transport to disconnect
 703  *
 704  */
 705 void xprt_force_disconnect(struct rpc_xprt *xprt)
 706 {
 707         trace_xprt_disconnect_force(xprt);
 708
 709         /* Don't race with the test_bit() in xprt_clear_locked() */
 710         spin_lock(&xprt->transport_lock);
 711         set_bit(XPRT_CLOSE_WAIT, &xprt->state);
 712         /* Try to schedule an autoclose RPC call */
 713         if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
 714                 queue_work(xprtiod_workqueue, &xprt->task_cleanup);
 715         else if (xprt->snd_task)
 716                 rpc_wake_up_queued_task_set_status(&xprt->pending,
 717                                 xprt->snd_task, -ENOTCONN);
 718         spin_unlock(&xprt->transport_lock);
 719 }
 720 EXPORT_SYMBOL_GPL(xprt_force_disconnect);
 721
 722 static unsigned int
 723 xprt_connect_cookie(struct rpc_xprt *xprt)
 724 {
 725         return READ_ONCE(xprt->connect_cookie);
 726 }
 727
 728 static bool
 729 xprt_request_retransmit_after_disconnect(struct rpc_task *task)
 730 {
 731         struct rpc_rqst *req = task->tk_rqstp;
 732         struct rpc_xprt *xprt = req->rq_xprt;
 733
 734         return req->rq_connect_cookie != xprt_connect_cookie(xprt) ||
 735                 !xprt_connected(xprt);
 736 }
 737
 738 /**
 739  * xprt_conditional_disconnect - force a transport to disconnect
 740  * @xprt: transport to disconnect
 741  * @cookie: 'connection cookie'
 742  *
 743  * This attempts to break the connection if and only if 'cookie' matches
 744  * the current transport 'connection cookie'. It ensures that we don't
 745  * try to break the connection more than once when we need to retransmit
 746  * a batch of RPC requests.
 747  *
 748  */
 749 void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie)
 750 {
 751         /* Don't race with the test_bit() in xprt_clear_locked() */
 752         spin_lock(&xprt->transport_lock);
 753         if (cookie != xprt->connect_cookie)
 754                 goto out;
 755         if (test_bit(XPRT_CLOSING, &xprt->state))
 756                 goto out;
 757         set_bit(XPRT_CLOSE_WAIT, &xprt->state);
 758         /* Try to schedule an autoclose RPC call */
 759         if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
 760                 queue_work(xprtiod_workqueue, &xprt->task_cleanup);
 761         xprt_wake_pending_tasks(xprt, -EAGAIN);
 762 out:
 763         spin_unlock(&xprt->transport_lock);
 764 }
 765
 766 static bool
 767 xprt_has_timer(const struct rpc_xprt *xprt)
 768 {
 769         return xprt->idle_timeout != 0;
 770 }
 771
 772 static void
 773 xprt_schedule_autodisconnect(struct rpc_xprt *xprt)
 774         __must_hold(&xprt->transport_lock)
 775 {
 776         xprt->last_used = jiffies;
 777         if (RB_EMPTY_ROOT(&xprt->recv_queue) && xprt_has_timer(xprt))
 778                 mod_timer(&xprt->timer, xprt->last_used + xprt->idle_timeout);
 779 }
 780
 781 static void
 782 xprt_init_autodisconnect(struct timer_list *t)
 783 {
 784         struct rpc_xprt *xprt = from_timer(xprt, t, timer);
 785
 786         if (!RB_EMPTY_ROOT(&xprt->recv_queue))
 787                 return;
 788         /* Reset xprt->last_used to avoid connect/autodisconnect cycling */
 789         xprt->last_used = jiffies;
 790         if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
 791                 return;
 792         queue_work(xprtiod_workqueue, &xprt->task_cleanup);
 793 }
 794
 795 bool xprt_lock_connect(struct rpc_xprt *xprt,
 796                 struct rpc_task *task,
 797                 void *cookie)
 798 {
 799         bool ret = false;
 800
 801         spin_lock(&xprt->transport_lock);
 802         if (!test_bit(XPRT_LOCKED, &xprt->state))
 803                 goto out;
 804         if (xprt->snd_task != task)
 805                 goto out;
 806         xprt->snd_task = cookie;
 807         ret = true;
 808 out:
 809         spin_unlock(&xprt->transport_lock);
 810         return ret;
 811 }
 812
 813 void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie)
 814 {
 815         spin_lock(&xprt->transport_lock);
 816         if (xprt->snd_task != cookie)
 817                 goto out;
 818         if (!test_bit(XPRT_LOCKED, &xprt->state))
 819                 goto out;
 820         xprt->snd_task =NULL;
 821         xprt->ops->release_xprt(xprt, NULL);
 822         xprt_schedule_autodisconnect(xprt);
 823 out:
 824         spin_unlock(&xprt->transport_lock);
 825         wake_up_bit(&xprt->state, XPRT_LOCKED);
 826 }
 827
 828 /**
 829  * xprt_connect - schedule a transport connect operation
 830  * @task: RPC task that is requesting the connect
 831  *
 832  */
 833 void xprt_connect(struct rpc_task *task)
 834 {
 835         struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt;
 836
 837         trace_xprt_connect(xprt);
 838
 839         if (!xprt_bound(xprt)) {
 840                 task->tk_status = -EAGAIN;
 841                 return;
 842         }
 843         if (!xprt_lock_write(xprt, task))
 844                 return;
 845
 846         if (test_and_clear_bit(XPRT_CLOSE_WAIT, &xprt->state)) {
 847                 trace_xprt_disconnect_cleanup(xprt);
 848                 xprt->ops->close(xprt);
 849         }
 850
 851         if (!xprt_connected(xprt)) {
 852                 task->tk_rqstp->rq_connect_cookie = xprt->connect_cookie;
 853                 rpc_sleep_on_timeout(&xprt->pending, task, NULL,
 854                                 xprt_request_timeout(task->tk_rqstp));
 855
 856                 if (test_bit(XPRT_CLOSING, &xprt->state))
 857                         return;
 858                 if (xprt_test_and_set_connecting(xprt))
 859                         return;
 860                 /* Race breaker */
 861                 if (!xprt_connected(xprt)) {
 862                         xprt->stat.connect_start = jiffies;
 863                         xprt->ops->connect(xprt, task);
 864                 } else {
 865                         xprt_clear_connecting(xprt);
 866                         task->tk_status = 0;
 867                         rpc_wake_up_queued_task(&xprt->pending, task);
 868                 }
 869         }
 870         xprt_release_write(xprt, task);
 871 }
 872
 873 /**
 874  * xprt_reconnect_delay - compute the wait before scheduling a connect
 875  * @xprt: transport instance
 876  *
 877  */
 878 unsigned long xprt_reconnect_delay(const struct rpc_xprt *xprt)
 879 {
 880         unsigned long start, now = jiffies;
 881
 882         start = xprt->stat.connect_start + xprt->reestablish_timeout;
 883         if (time_after(start, now))
 884                 return start - now;
 885         return 0;
 886 }
 887 EXPORT_SYMBOL_GPL(xprt_reconnect_delay);
 888
 889 /**
 890  * xprt_reconnect_backoff - compute the new re-establish timeout
 891  * @xprt: transport instance
 892  * @init_to: initial reestablish timeout
 893  *
 894  */
 895 void xprt_reconnect_backoff(struct rpc_xprt *xprt, unsigned long init_to)
 896 {
 897         xprt->reestablish_timeout <<= 1;
 898         if (xprt->reestablish_timeout > xprt->max_reconnect_timeout)
 899                 xprt->reestablish_timeout = xprt->max_reconnect_timeout;
 900         if (xprt->reestablish_timeout < init_to)
 901                 xprt->reestablish_timeout = init_to;
 902 }
 903 EXPORT_SYMBOL_GPL(xprt_reconnect_backoff);
 904
 905 enum xprt_xid_rb_cmp {
 906         XID_RB_EQUAL,
 907         XID_RB_LEFT,
 908         XID_RB_RIGHT,
 909 };
 910 static enum xprt_xid_rb_cmp
 911 xprt_xid_cmp(__be32 xid1, __be32 xid2)
 912 {
 913         if (xid1 == xid2)
 914                 return XID_RB_EQUAL;
 915         if ((__force u32)xid1 < (__force u32)xid2)
 916                 return XID_RB_LEFT;
 917         return XID_RB_RIGHT;
 918 }
 919
 920 static struct rpc_rqst *
 921 xprt_request_rb_find(struct rpc_xprt *xprt, __be32 xid)
 922 {
 923         struct rb_node *n = xprt->recv_queue.rb_node;
 924         struct rpc_rqst *req;
 925
 926         while (n != NULL) {
 927                 req = rb_entry(n, struct rpc_rqst, rq_recv);
 928                 switch (xprt_xid_cmp(xid, req->rq_xid)) {
 929                 case XID_RB_LEFT:
 930                         n = n->rb_left;
 931                         break;
 932                 case XID_RB_RIGHT:
 933                         n = n->rb_right;
 934                         break;
 935                 case XID_RB_EQUAL:
 936                         return req;
 937                 }
 938         }
 939         return NULL;
 940 }
 941
 942 static void
 943 xprt_request_rb_insert(struct rpc_xprt *xprt, struct rpc_rqst *new)
 944 {
 945         struct rb_node **p = &xprt->recv_queue.rb_node;
 946         struct rb_node *n = NULL;
 947         struct rpc_rqst *req;
 948
 949         while (*p != NULL) {
 950                 n = *p;
 951                 req = rb_entry(n, struct rpc_rqst, rq_recv);
 952                 switch(xprt_xid_cmp(new->rq_xid, req->rq_xid)) {
 953                 case XID_RB_LEFT:
 954                         p = &n->rb_left;
 955                         break;
 956                 case XID_RB_RIGHT:
 957                         p = &n->rb_right;
 958                         break;
 959                 case XID_RB_EQUAL:
 960                         WARN_ON_ONCE(new != req);
 961                         return;
 962                 }
 963         }
 964         rb_link_node(&new->rq_recv, n, p);
 965         rb_insert_color(&new->rq_recv, &xprt->recv_queue);
 966 }
 967
 968 static void
 969 xprt_request_rb_remove(struct rpc_xprt *xprt, struct rpc_rqst *req)
 970 {
 971         rb_erase(&req->rq_recv, &xprt->recv_queue);
 972 }
 973
 974 /**
 975  * xprt_lookup_rqst - find an RPC request corresponding to an XID
 976  * @xprt: transport on which the original request was transmitted
 977  * @xid: RPC XID of incoming reply
 978  *
 979  * Caller holds xprt->queue_lock.
 980  */
 981 struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid)
 982 {
 983         struct rpc_rqst *entry;
 984
 985         entry = xprt_request_rb_find(xprt, xid);
 986         if (entry != NULL) {
 987                 trace_xprt_lookup_rqst(xprt, xid, 0);
 988                 entry->rq_rtt = ktime_sub(ktime_get(), entry->rq_xtime);
 989                 return entry;
 990         }
 991
 992         dprintk("RPC:       xprt_lookup_rqst did not find xid %08x\n",
 993                         ntohl(xid));
 994         trace_xprt_lookup_rqst(xprt, xid, -ENOENT);
 995         xprt->stat.bad_xids++;
 996         return NULL;
 997 }
 998 EXPORT_SYMBOL_GPL(xprt_lookup_rqst);
 999
1000 static bool
1001 xprt_is_pinned_rqst(struct rpc_rqst *req)
1002 {
1003         return atomic_read(&req->rq_pin) != 0;
1004 }
1005
1006 /**
1007  * xprt_pin_rqst - Pin a request on the transport receive list
1008  * @req: Request to pin
1009  *
1010  * Caller must ensure this is atomic with the call to xprt_lookup_rqst()
1011  * so should be holding xprt->queue_lock.
1012  */
1013 void xprt_pin_rqst(struct rpc_rqst *req)
1014 {
1015         atomic_inc(&req->rq_pin);
1016 }
1017 EXPORT_SYMBOL_GPL(xprt_pin_rqst);
1018
1019 /**
1020  * xprt_unpin_rqst - Unpin a request on the transport receive list
1021  * @req: Request to pin
1022  *
1023  * Caller should be holding xprt->queue_lock.
1024  */
1025 void xprt_unpin_rqst(struct rpc_rqst *req)
1026 {
1027         if (!test_bit(RPC_TASK_MSG_PIN_WAIT, &req->rq_task->tk_runstate)) {
1028                 atomic_dec(&req->rq_pin);
1029                 return;
1030         }
1031         if (atomic_dec_and_test(&req->rq_pin))
1032                 wake_up_var(&req->rq_pin);
1033 }
1034 EXPORT_SYMBOL_GPL(xprt_unpin_rqst);
1035
1036 static void xprt_wait_on_pinned_rqst(struct rpc_rqst *req)
1037 {
1038         wait_var_event(&req->rq_pin, !xprt_is_pinned_rqst(req));
1039 }
1040
1041 static bool
1042 xprt_request_data_received(struct rpc_task *task)
1043 {
1044         return !test_bit(RPC_TASK_NEED_RECV, &task->tk_runstate) &&
1045                 READ_ONCE(task->tk_rqstp->rq_reply_bytes_recvd) != 0;
1046 }
1047
1048 static bool
1049 xprt_request_need_enqueue_receive(struct rpc_task *task, struct rpc_rqst *req)
1050 {
1051         return !test_bit(RPC_TASK_NEED_RECV, &task->tk_runstate) &&
1052                 READ_ONCE(task->tk_rqstp->rq_reply_bytes_recvd) == 0;
1053 }
1054
1055 /**
1056  * xprt_request_enqueue_receive - Add an request to the receive queue
1057  * @task: RPC task
1058  *
1059  */
1060 void
1061 xprt_request_enqueue_receive(struct rpc_task *task)
1062 {
1063         struct rpc_rqst *req = task->tk_rqstp;
1064         struct rpc_xprt *xprt = req->rq_xprt;
1065
1066         if (!xprt_request_need_enqueue_receive(task, req))
1067                 return;
1068
1069         xprt_request_prepare(task->tk_rqstp);
1070         spin_lock(&xprt->queue_lock);
1071
1072         /* Update the softirq receive buffer */
1073         memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
1074                         sizeof(req->rq_private_buf));
1075
1076         /* Add request to the receive list */
1077         xprt_request_rb_insert(xprt, req);
1078         set_bit(RPC_TASK_NEED_RECV, &task->tk_runstate);
1079         spin_unlock(&xprt->queue_lock);
1080
1081         /* Turn off autodisconnect */
1082         del_singleshot_timer_sync(&xprt->timer);
1083 }
1084
1085 /**
1086  * xprt_request_dequeue_receive_locked - Remove a request from the receive queue
1087  * @task: RPC task
1088  *
1089  * Caller must hold xprt->queue_lock.
1090  */
1091 static void
1092 xprt_request_dequeue_receive_locked(struct rpc_task *task)
1093 {
1094         struct rpc_rqst *req = task->tk_rqstp;
1095
1096         if (test_and_clear_bit(RPC_TASK_NEED_RECV, &task->tk_runstate))
1097                 xprt_request_rb_remove(req->rq_xprt, req);
1098 }
1099
1100 /**
1101  * xprt_update_rtt - Update RPC RTT statistics
1102  * @task: RPC request that recently completed
1103  *
1104  * Caller holds xprt->queue_lock.
1105  */
1106 void xprt_update_rtt(struct rpc_task *task)
1107 {
1108         struct rpc_rqst *req = task->tk_rqstp;
1109         struct rpc_rtt *rtt = task->tk_client->cl_rtt;
1110         unsigned int timer = task->tk_msg.rpc_proc->p_timer;
1111         long m = usecs_to_jiffies(ktime_to_us(req->rq_rtt));
1112
1113         if (timer) {
1114                 if (req->rq_ntrans == 1)
1115                         rpc_update_rtt(rtt, timer, m);
1116                 rpc_set_timeo(rtt, timer, req->rq_ntrans - 1);
1117         }
1118 }
1119 EXPORT_SYMBOL_GPL(xprt_update_rtt);
1120
1121 /**
1122  * xprt_complete_rqst - called when reply processing is complete
1123  * @task: RPC request that recently completed
1124  * @copied: actual number of bytes received from the transport
1125  *
1126  * Caller holds xprt->queue_lock.
1127  */
1128 void xprt_complete_rqst(struct rpc_task *task, int copied)
1129 {
1130         struct rpc_rqst *req = task->tk_rqstp;
1131         struct rpc_xprt *xprt = req->rq_xprt;
1132
1133         xprt->stat.recvs++;
1134
1135         req->rq_private_buf.len = copied;
1136         /* Ensure all writes are done before we update */
1137         /* req->rq_reply_bytes_recvd */
1138         smp_wmb();
1139         req->rq_reply_bytes_recvd = copied;
1140         xprt_request_dequeue_receive_locked(task);
1141         rpc_wake_up_queued_task(&xprt->pending, task);
1142 }
1143 EXPORT_SYMBOL_GPL(xprt_complete_rqst);
1144
1145 static void xprt_timer(struct rpc_task *task)
1146 {
1147         struct rpc_rqst *req = task->tk_rqstp;
1148         struct rpc_xprt *xprt = req->rq_xprt;
1149
1150         if (task->tk_status != -ETIMEDOUT)
1151                 return;
1152
1153         trace_xprt_timer(xprt, req->rq_xid, task->tk_status);
1154         if (!req->rq_reply_bytes_recvd) {
1155                 if (xprt->ops->timer)
1156                         xprt->ops->timer(xprt, task);
1157         } else
1158                 task->tk_status = 0;
1159 }
1160
1161 /**
1162  * xprt_wait_for_reply_request_def - wait for reply
1163  * @task: pointer to rpc_task
1164  *
1165  * Set a request's retransmit timeout based on the transport's
1166  * default timeout parameters.  Used by transports that don't adjust
1167  * the retransmit timeout based on round-trip time estimation,
1168  * and put the task to sleep on the pending queue.
1169  */
1170 void xprt_wait_for_reply_request_def(struct rpc_task *task)
1171 {
1172         struct rpc_rqst *req = task->tk_rqstp;
1173
1174         rpc_sleep_on_timeout(&req->rq_xprt->pending, task, xprt_timer,
1175                         xprt_request_timeout(req));
1176 }
1177 EXPORT_SYMBOL_GPL(xprt_wait_for_reply_request_def);
1178
1179 /**
1180  * xprt_wait_for_reply_request_rtt - wait for reply using RTT estimator
1181  * @task: pointer to rpc_task
1182  *
1183  * Set a request's retransmit timeout using the RTT estimator,
1184  * and put the task to sleep on the pending queue.
1185  */
1186 void xprt_wait_for_reply_request_rtt(struct rpc_task *task)
1187 {
1188         int timer = task->tk_msg.rpc_proc->p_timer;
1189         struct rpc_clnt *clnt = task->tk_client;
1190         struct rpc_rtt *rtt = clnt->cl_rtt;
1191         struct rpc_rqst *req = task->tk_rqstp;
1192         unsigned long max_timeout = clnt->cl_timeout->to_maxval;
1193         unsigned long timeout;
1194
1195         timeout = rpc_calc_rto(rtt, timer);
1196         timeout <<= rpc_ntimeo(rtt, timer) + req->rq_retries;
1197         if (timeout > max_timeout || timeout == 0)
1198                 timeout = max_timeout;
1199         rpc_sleep_on_timeout(&req->rq_xprt->pending, task, xprt_timer,
1200                         jiffies + timeout);
1201 }
1202 EXPORT_SYMBOL_GPL(xprt_wait_for_reply_request_rtt);
1203
1204 /**
1205  * xprt_request_wait_receive - wait for the reply to an RPC request
1206  * @task: RPC task about to send a request
1207  *
1208  */
1209 void xprt_request_wait_receive(struct rpc_task *task)
1210 {
1211         struct rpc_rqst *req = task->tk_rqstp;
1212         struct rpc_xprt *xprt = req->rq_xprt;
1213
1214         if (!test_bit(RPC_TASK_NEED_RECV, &task->tk_runstate))
1215                 return;
1216         /*
1217          * Sleep on the pending queue if we're expecting a reply.
1218          * The spinlock ensures atomicity between the test of
1219          * req->rq_reply_bytes_recvd, and the call to rpc_sleep_on().
1220          */
1221         spin_lock(&xprt->queue_lock);
1222         if (test_bit(RPC_TASK_NEED_RECV, &task->tk_runstate)) {
1223                 xprt->ops->wait_for_reply_request(task);
1224                 /*
1225                  * Send an extra queue wakeup call if the
1226                  * connection was dropped in case the call to
1227                  * rpc_sleep_on() raced.
1228                  */
1229                 if (xprt_request_retransmit_after_disconnect(task))
1230                         rpc_wake_up_queued_task_set_status(&xprt->pending,
1231                                         task, -ENOTCONN);
1232         }
1233         spin_unlock(&xprt->queue_lock);
1234 }
1235
1236 static bool
1237 xprt_request_need_enqueue_transmit(struct rpc_task *task, struct rpc_rqst *req)
1238 {
1239         return !test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate);
1240 }
1241
1242 /**
1243  * xprt_request_enqueue_transmit - queue a task for transmission
1244  * @task: pointer to rpc_task
1245  *
1246  * Add a task to the transmission queue.
1247  */
1248 void
1249 xprt_request_enqueue_transmit(struct rpc_task *task)
1250 {
1251         struct rpc_rqst *pos, *req = task->tk_rqstp;
1252         struct rpc_xprt *xprt = req->rq_xprt;
1253
1254         if (xprt_request_need_enqueue_transmit(task, req)) {
1255                 req->rq_bytes_sent = 0;
1256                 spin_lock(&xprt->queue_lock);
1257                 /*
1258                  * Requests that carry congestion control credits are added
1259                  * to the head of the list to avoid starvation issues.
1260                  */
1261                 if (req->rq_cong) {
1262                         xprt_clear_congestion_window_wait(xprt);
1263                         list_for_each_entry(pos, &xprt->xmit_queue, rq_xmit) {
1264                                 if (pos->rq_cong)
1265                                         continue;
1266                                 /* Note: req is added _before_ pos */
1267                                 list_add_tail(&req->rq_xmit, &pos->rq_xmit);
1268                                 INIT_LIST_HEAD(&req->rq_xmit2);
1269                                 goto out;
1270                         }
1271                 } else if (RPC_IS_SWAPPER(task)) {
1272                         list_for_each_entry(pos, &xprt->xmit_queue, rq_xmit) {
1273                                 if (pos->rq_cong || pos->rq_bytes_sent)
1274                                         continue;
1275                                 if (RPC_IS_SWAPPER(pos->rq_task))
1276                                         continue;
1277                                 /* Note: req is added _before_ pos */
1278                                 list_add_tail(&req->rq_xmit, &pos->rq_xmit);
1279                                 INIT_LIST_HEAD(&req->rq_xmit2);
1280                                 goto out;
1281                         }
1282                 } else if (!req->rq_seqno) {
1283                         list_for_each_entry(pos, &xprt->xmit_queue, rq_xmit) {
1284                                 if (pos->rq_task->tk_owner != task->tk_owner)
1285                                         continue;
1286                                 list_add_tail(&req->rq_xmit2, &pos->rq_xmit2);
1287                                 INIT_LIST_HEAD(&req->rq_xmit);
1288                                 goto out;
1289                         }
1290                 }
1291                 list_add_tail(&req->rq_xmit, &xprt->xmit_queue);
1292                 INIT_LIST_HEAD(&req->rq_xmit2);
1293 out:
1294                 set_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate);
1295                 spin_unlock(&xprt->queue_lock);
1296         }
1297 }
1298
1299 /**
1300  * xprt_request_dequeue_transmit_locked - remove a task from the transmission queue
1301  * @task: pointer to rpc_task
1302  *
1303  * Remove a task from the transmission queue
1304  * Caller must hold xprt->queue_lock
1305  */
1306 static void
1307 xprt_request_dequeue_transmit_locked(struct rpc_task *task)
1308 {
1309         struct rpc_rqst *req = task->tk_rqstp;
1310
1311         if (!test_and_clear_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate))
1312                 return;
1313         if (!list_empty(&req->rq_xmit)) {
1314                 list_del(&req->rq_xmit);
1315                 if (!list_empty(&req->rq_xmit2)) {
1316                         struct rpc_rqst *next = list_first_entry(&req->rq_xmit2,
1317                                         struct rpc_rqst, rq_xmit2);
1318                         list_del(&req->rq_xmit2);
1319                         list_add_tail(&next->rq_xmit, &next->rq_xprt->xmit_queue);
1320                 }
1321         } else
1322                 list_del(&req->rq_xmit2);
1323 }
1324
1325 /**
1326  * xprt_request_dequeue_transmit - remove a task from the transmission queue
1327  * @task: pointer to rpc_task
1328  *
1329  * Remove a task from the transmission queue
1330  */
1331 static void
1332 xprt_request_dequeue_transmit(struct rpc_task *task)
1333 {
1334         struct rpc_rqst *req = task->tk_rqstp;
1335         struct rpc_xprt *xprt = req->rq_xprt;
1336
1337         spin_lock(&xprt->queue_lock);
1338         xprt_request_dequeue_transmit_locked(task);
1339         spin_unlock(&xprt->queue_lock);
1340 }
1341
1342 /**
1343  * xprt_request_dequeue_xprt - remove a task from the transmit+receive queue
1344  * @task: pointer to rpc_task
1345  *
1346  * Remove a task from the transmit and receive queues, and ensure that
1347  * it is not pinned by the receive work item.
1348  */
1349 void
1350 xprt_request_dequeue_xprt(struct rpc_task *task)
1351 {
1352         struct rpc_rqst *req = task->tk_rqstp;
1353         struct rpc_xprt *xprt = req->rq_xprt;
1354
1355         if (test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate) ||
1356             test_bit(RPC_TASK_NEED_RECV, &task->tk_runstate) ||
1357             xprt_is_pinned_rqst(req)) {
1358                 spin_lock(&xprt->queue_lock);
1359                 xprt_request_dequeue_transmit_locked(task);
1360                 xprt_request_dequeue_receive_locked(task);
1361                 while (xprt_is_pinned_rqst(req)) {
1362                         set_bit(RPC_TASK_MSG_PIN_WAIT, &task->tk_runstate);
1363                         spin_unlock(&xprt->queue_lock);
1364                         xprt_wait_on_pinned_rqst(req);
1365                         spin_lock(&xprt->queue_lock);
1366                         clear_bit(RPC_TASK_MSG_PIN_WAIT, &task->tk_runstate);
1367                 }
1368                 spin_unlock(&xprt->queue_lock);
1369         }
1370 }
1371
1372 /**
1373  * xprt_request_prepare - prepare an encoded request for transport
1374  * @req: pointer to rpc_rqst
1375  *
1376  * Calls into the transport layer to do whatever is needed to prepare
1377  * the request for transmission or receive.
1378  */
1379 void
1380 xprt_request_prepare(struct rpc_rqst *req)
1381 {
1382         struct rpc_xprt *xprt = req->rq_xprt;
1383
1384         if (xprt->ops->prepare_request)
1385                 xprt->ops->prepare_request(req);
1386 }
1387
1388 /**
1389  * xprt_request_need_retransmit - Test if a task needs retransmission
1390  * @task: pointer to rpc_task
1391  *
1392  * Test for whether a connection breakage requires the task to retransmit
1393  */
1394 bool
1395 xprt_request_need_retransmit(struct rpc_task *task)
1396 {
1397         return xprt_request_retransmit_after_disconnect(task);
1398 }
1399
1400 /**
1401  * xprt_prepare_transmit - reserve the transport before sending a request
1402  * @task: RPC task about to send a request
1403  *
1404  */
1405 bool xprt_prepare_transmit(struct rpc_task *task)
1406 {
1407         struct rpc_rqst *req = task->tk_rqstp;
1408         struct rpc_xprt *xprt = req->rq_xprt;
1409
1410         if (!xprt_lock_write(xprt, task)) {
1411                 trace_xprt_transmit_queued(xprt, task);
1412
1413                 /* Race breaker: someone may have transmitted us */
1414                 if (!test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate))
1415                         rpc_wake_up_queued_task_set_status(&xprt->sending,
1416                                         task, 0);
1417                 return false;
1418
1419         }
1420         return true;
1421 }
1422
1423 void xprt_end_transmit(struct rpc_task *task)
1424 {
1425         xprt_release_write(task->tk_rqstp->rq_xprt, task);
1426 }
1427
1428 /**
1429  * xprt_request_transmit - send an RPC request on a transport
1430  * @req: pointer to request to transmit
1431  * @snd_task: RPC task that owns the transport lock
1432  *
1433  * This performs the transmission of a single request.
1434  * Note that if the request is not the same as snd_task, then it
1435  * does need to be pinned.
1436  * Returns '0' on success.
1437  */
1438 static int
1439 xprt_request_transmit(struct rpc_rqst *req, struct rpc_task *snd_task)
1440 {
1441         struct rpc_xprt *xprt = req->rq_xprt;
1442         struct rpc_task *task = req->rq_task;
1443         unsigned int connect_cookie;
1444         int is_retrans = RPC_WAS_SENT(task);
1445         int status;
1446
1447         if (!req->rq_bytes_sent) {
1448                 if (xprt_request_data_received(task)) {
1449                         status = 0;
1450                         goto out_dequeue;
1451                 }
1452                 /* Verify that our message lies in the RPCSEC_GSS window */
1453                 if (rpcauth_xmit_need_reencode(task)) {
1454                         status = -EBADMSG;
1455                         goto out_dequeue;
1456                 }
1457                 if (RPC_SIGNALLED(task)) {
1458                         status = -ERESTARTSYS;
1459                         goto out_dequeue;
1460                 }
1461         }
1462
1463         /*
1464          * Update req->rq_ntrans before transmitting to avoid races with
1465          * xprt_update_rtt(), which needs to know that it is recording a
1466          * reply to the first transmission.
1467          */
1468         req->rq_ntrans++;
1469
1470         trace_rpc_xdr_sendto(task, &req->rq_snd_buf);
1471         connect_cookie = xprt->connect_cookie;
1472         status = xprt->ops->send_request(req);
1473         if (status != 0) {
1474                 req->rq_ntrans--;
1475                 trace_xprt_transmit(req, status);
1476                 return status;
1477         }
1478
1479         if (is_retrans)
1480                 task->tk_client->cl_stats->rpcretrans++;
1481
1482         xprt_inject_disconnect(xprt);
1483
1484         task->tk_flags |= RPC_TASK_SENT;
1485         spin_lock(&xprt->transport_lock);
1486
1487         xprt->stat.sends++;
1488         xprt->stat.req_u += xprt->stat.sends - xprt->stat.recvs;
1489         xprt->stat.bklog_u += xprt->backlog.qlen;
1490         xprt->stat.sending_u += xprt->sending.qlen;
1491         xprt->stat.pending_u += xprt->pending.qlen;
1492         spin_unlock(&xprt->transport_lock);
1493
1494         req->rq_connect_cookie = connect_cookie;
1495 out_dequeue:
1496         trace_xprt_transmit(req, status);
1497         xprt_request_dequeue_transmit(task);
1498         rpc_wake_up_queued_task_set_status(&xprt->sending, task, status);
1499         return status;
1500 }
1501
1502 /**
1503  * xprt_transmit - send an RPC request on a transport
1504  * @task: controlling RPC task
1505  *
1506  * Attempts to drain the transmit queue. On exit, either the transport
1507  * signalled an error that needs to be handled before transmission can
1508  * resume, or @task finished transmitting, and detected that it already
1509  * received a reply.
1510  */
1511 void
1512 xprt_transmit(struct rpc_task *task)
1513 {
1514         struct rpc_rqst *next, *req = task->tk_rqstp;
1515         struct rpc_xprt *xprt = req->rq_xprt;
1516         int counter, status;
1517
1518         spin_lock(&xprt->queue_lock);
1519         counter = 0;
1520         while (!list_empty(&xprt->xmit_queue)) {
1521                 if (++counter == 20)
1522                         break;
1523                 next = list_first_entry(&xprt->xmit_queue,
1524                                 struct rpc_rqst, rq_xmit);
1525                 xprt_pin_rqst(next);
1526                 spin_unlock(&xprt->queue_lock);
1527                 status = xprt_request_transmit(next, task);
1528                 if (status == -EBADMSG && next != req)
1529                         status = 0;
1530                 spin_lock(&xprt->queue_lock);
1531                 xprt_unpin_rqst(next);
1532                 if (status == 0) {
1533                         if (!xprt_request_data_received(task) ||
1534                             test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate))
1535                                 continue;
1536                 } else if (test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate))
1537                         task->tk_status = status;
1538                 break;
1539         }
1540         spin_unlock(&xprt->queue_lock);
1541 }
1542
1543 static void xprt_add_backlog(struct rpc_xprt *xprt, struct rpc_task *task)
1544 {
1545         set_bit(XPRT_CONGESTED, &xprt->state);
1546         rpc_sleep_on(&xprt->backlog, task, NULL);
1547 }
1548
1549 static void xprt_wake_up_backlog(struct rpc_xprt *xprt)
1550 {
1551         if (rpc_wake_up_next(&xprt->backlog) == NULL)
1552                 clear_bit(XPRT_CONGESTED, &xprt->state);
1553 }
1554
1555 static bool xprt_throttle_congested(struct rpc_xprt *xprt, struct rpc_task *task)
1556 {
1557         bool ret = false;
1558
1559         if (!test_bit(XPRT_CONGESTED, &xprt->state))
1560                 goto out;
1561         spin_lock(&xprt->reserve_lock);
1562         if (test_bit(XPRT_CONGESTED, &xprt->state)) {
1563                 rpc_sleep_on(&xprt->backlog, task, NULL);
1564                 ret = true;
1565         }
1566         spin_unlock(&xprt->reserve_lock);
1567 out:
1568         return ret;
1569 }
1570
1571 static struct rpc_rqst *xprt_dynamic_alloc_slot(struct rpc_xprt *xprt)
1572 {
1573         struct rpc_rqst *req = ERR_PTR(-EAGAIN);
1574
1575         if (xprt->num_reqs >= xprt->max_reqs)
1576                 goto out;
1577         ++xprt->num_reqs;
1578         spin_unlock(&xprt->reserve_lock);
1579         req = kzalloc(sizeof(struct rpc_rqst), GFP_NOFS);
1580         spin_lock(&xprt->reserve_lock);
1581         if (req != NULL)
1582                 goto out;
1583         --xprt->num_reqs;
1584         req = ERR_PTR(-ENOMEM);
1585 out:
1586         return req;
1587 }
1588
1589 static bool xprt_dynamic_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req)
1590 {
1591         if (xprt->num_reqs > xprt->min_reqs) {
1592                 --xprt->num_reqs;
1593                 kfree(req);
1594                 return true;
1595         }
1596         return false;
1597 }
1598
1599 void xprt_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task)
1600 {
1601         struct rpc_rqst *req;
1602
1603         spin_lock(&xprt->reserve_lock);
1604         if (!list_empty(&xprt->free)) {
1605                 req = list_entry(xprt->free.next, struct rpc_rqst, rq_list);
1606                 list_del(&req->rq_list);
1607                 goto out_init_req;
1608         }
1609         req = xprt_dynamic_alloc_slot(xprt);
1610         if (!IS_ERR(req))
1611                 goto out_init_req;
1612         switch (PTR_ERR(req)) {
1613         case -ENOMEM:
1614                 dprintk("RPC:       dynamic allocation of request slot "
1615                                 "failed! Retrying\n");
1616                 task->tk_status = -ENOMEM;
1617                 break;
1618         case -EAGAIN:
1619                 xprt_add_backlog(xprt, task);
1620                 dprintk("RPC:       waiting for request slot\n");
1621                 fallthrough;
1622         default:
1623                 task->tk_status = -EAGAIN;
1624         }
1625         spin_unlock(&xprt->reserve_lock);
1626         return;
1627 out_init_req:
1628         xprt->stat.max_slots = max_t(unsigned int, xprt->stat.max_slots,
1629                                      xprt->num_reqs);
1630         spin_unlock(&xprt->reserve_lock);
1631
1632         task->tk_status = 0;
1633         task->tk_rqstp = req;
1634 }
1635 EXPORT_SYMBOL_GPL(xprt_alloc_slot);
1636
1637 void xprt_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req)
1638 {
1639         spin_lock(&xprt->reserve_lock);
1640         if (!xprt_dynamic_free_slot(xprt, req)) {
1641                 memset(req, 0, sizeof(*req));   /* mark unused */
1642                 list_add(&req->rq_list, &xprt->free);
1643         }
1644         xprt_wake_up_backlog(xprt);
1645         spin_unlock(&xprt->reserve_lock);
1646 }
1647 EXPORT_SYMBOL_GPL(xprt_free_slot);
1648
1649 static void xprt_free_all_slots(struct rpc_xprt *xprt)
1650 {
1651         struct rpc_rqst *req;
1652         while (!list_empty(&xprt->free)) {
1653                 req = list_first_entry(&xprt->free, struct rpc_rqst, rq_list);
1654                 list_del(&req->rq_list);
1655                 kfree(req);
1656         }
1657 }
1658
1659 struct rpc_xprt *xprt_alloc(struct net *net, size_t size,
1660                 unsigned int num_prealloc,
1661                 unsigned int max_alloc)
1662 {
1663         struct rpc_xprt *xprt;
1664         struct rpc_rqst *req;
1665         int i;
1666
1667         xprt = kzalloc(size, GFP_KERNEL);
1668         if (xprt == NULL)
1669                 goto out;
1670
1671         xprt_init(xprt, net);
1672
1673         for (i = 0; i < num_prealloc; i++) {
1674                 req = kzalloc(sizeof(struct rpc_rqst), GFP_KERNEL);
1675                 if (!req)
1676                         goto out_free;
1677                 list_add(&req->rq_list, &xprt->free);
1678         }
1679         if (max_alloc > num_prealloc)
1680                 xprt->max_reqs = max_alloc;
1681         else
1682                 xprt->max_reqs = num_prealloc;
1683         xprt->min_reqs = num_prealloc;
1684         xprt->num_reqs = num_prealloc;
1685
1686         return xprt;
1687
1688 out_free:
1689         xprt_free(xprt);
1690 out:
1691         return NULL;
1692 }
1693 EXPORT_SYMBOL_GPL(xprt_alloc);
1694
1695 void xprt_free(struct rpc_xprt *xprt)
1696 {
1697         put_net(xprt->xprt_net);
1698         xprt_free_all_slots(xprt);
1699         kfree_rcu(xprt, rcu);
1700 }
1701 EXPORT_SYMBOL_GPL(xprt_free);
1702
1703 static void
1704 xprt_init_connect_cookie(struct rpc_rqst *req, struct rpc_xprt *xprt)
1705 {
1706         req->rq_connect_cookie = xprt_connect_cookie(xprt) - 1;
1707 }
1708
1709 static __be32
1710 xprt_alloc_xid(struct rpc_xprt *xprt)
1711 {
1712         __be32 xid;
1713
1714         spin_lock(&xprt->reserve_lock);
1715         xid = (__force __be32)xprt->xid++;
1716         spin_unlock(&xprt->reserve_lock);
1717         return xid;
1718 }
1719
1720 static void
1721 xprt_init_xid(struct rpc_xprt *xprt)
1722 {
1723         xprt->xid = prandom_u32();
1724 }
1725
1726 static void
1727 xprt_request_init(struct rpc_task *task)
1728 {
1729         struct rpc_xprt *xprt = task->tk_xprt;
1730         struct rpc_rqst *req = task->tk_rqstp;
1731
1732         req->rq_task    = task;
1733         req->rq_xprt    = xprt;
1734         req->rq_buffer  = NULL;
1735         req->rq_xid     = xprt_alloc_xid(xprt);
1736         xprt_init_connect_cookie(req, xprt);
1737         req->rq_snd_buf.len = 0;
1738         req->rq_snd_buf.buflen = 0;
1739         req->rq_rcv_buf.len = 0;
1740         req->rq_rcv_buf.buflen = 0;
1741         req->rq_snd_buf.bvec = NULL;
1742         req->rq_rcv_buf.bvec = NULL;
1743         req->rq_release_snd_buf = NULL;
1744         xprt_init_majortimeo(task, req);
1745
1746         trace_xprt_reserve(req);
1747 }
1748
1749 static void
1750 xprt_do_reserve(struct rpc_xprt *xprt, struct rpc_task *task)
1751 {
1752         xprt->ops->alloc_slot(xprt, task);
1753         if (task->tk_rqstp != NULL)
1754                 xprt_request_init(task);
1755 }
1756
1757 /**
1758  * xprt_reserve - allocate an RPC request slot
1759  * @task: RPC task requesting a slot allocation
1760  *
1761  * If the transport is marked as being congested, or if no more
1762  * slots are available, place the task on the transport's
1763  * backlog queue.
1764  */
1765 void xprt_reserve(struct rpc_task *task)
1766 {
1767         struct rpc_xprt *xprt = task->tk_xprt;
1768
1769         task->tk_status = 0;
1770         if (task->tk_rqstp != NULL)
1771                 return;
1772
1773         task->tk_status = -EAGAIN;
1774         if (!xprt_throttle_congested(xprt, task))
1775                 xprt_do_reserve(xprt, task);
1776 }
1777
1778 /**
1779  * xprt_retry_reserve - allocate an RPC request slot
1780  * @task: RPC task requesting a slot allocation
1781  *
1782  * If no more slots are available, place the task on the transport's
1783  * backlog queue.
1784  * Note that the only difference with xprt_reserve is that we now
1785  * ignore the value of the XPRT_CONGESTED flag.
1786  */
1787 void xprt_retry_reserve(struct rpc_task *task)
1788 {
1789         struct rpc_xprt *xprt = task->tk_xprt;
1790
1791         task->tk_status = 0;
1792         if (task->tk_rqstp != NULL)
1793                 return;
1794
1795         task->tk_status = -EAGAIN;
1796         xprt_do_reserve(xprt, task);
1797 }
1798
1799 /**
1800  * xprt_release - release an RPC request slot
1801  * @task: task which is finished with the slot
1802  *
1803  */
1804 void xprt_release(struct rpc_task *task)
1805 {
1806         struct rpc_xprt *xprt;
1807         struct rpc_rqst *req = task->tk_rqstp;
1808
1809         if (req == NULL) {
1810                 if (task->tk_client) {
1811                         xprt = task->tk_xprt;
1812                         xprt_release_write(xprt, task);
1813                 }
1814                 return;
1815         }
1816
1817         xprt = req->rq_xprt;
1818         xprt_request_dequeue_xprt(task);
1819         spin_lock(&xprt->transport_lock);
1820         xprt->ops->release_xprt(xprt, task);
1821         if (xprt->ops->release_request)
1822                 xprt->ops->release_request(task);
1823         xprt_schedule_autodisconnect(xprt);
1824         spin_unlock(&xprt->transport_lock);
1825         if (req->rq_buffer)
1826                 xprt->ops->buf_free(task);
1827         xprt_inject_disconnect(xprt);
1828         xdr_free_bvec(&req->rq_rcv_buf);
1829         xdr_free_bvec(&req->rq_snd_buf);
1830         if (req->rq_cred != NULL)
1831                 put_rpccred(req->rq_cred);
1832         task->tk_rqstp = NULL;
1833         if (req->rq_release_snd_buf)
1834                 req->rq_release_snd_buf(req);
1835
1836         if (likely(!bc_prealloc(req)))
1837                 xprt->ops->free_slot(xprt, req);
1838         else
1839                 xprt_free_bc_request(req);
1840 }
1841
1842 #ifdef CONFIG_SUNRPC_BACKCHANNEL
1843 void
1844 xprt_init_bc_request(struct rpc_rqst *req, struct rpc_task *task)
1845 {
1846         struct xdr_buf *xbufp = &req->rq_snd_buf;
1847
1848         task->tk_rqstp = req;
1849         req->rq_task = task;
1850         xprt_init_connect_cookie(req, req->rq_xprt);
1851         /*
1852          * Set up the xdr_buf length.
1853          * This also indicates that the buffer is XDR encoded already.
1854          */
1855         xbufp->len = xbufp->head[0].iov_len + xbufp->page_len +
1856                 xbufp->tail[0].iov_len;
1857 }
1858 #endif
1859
1860 static void xprt_init(struct rpc_xprt *xprt, struct net *net)
1861 {
1862         kref_init(&xprt->kref);
1863
1864         spin_lock_init(&xprt->transport_lock);
1865         spin_lock_init(&xprt->reserve_lock);
1866         spin_lock_init(&xprt->queue_lock);
1867
1868         INIT_LIST_HEAD(&xprt->free);
1869         xprt->recv_queue = RB_ROOT;
1870         INIT_LIST_HEAD(&xprt->xmit_queue);
1871 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
1872         spin_lock_init(&xprt->bc_pa_lock);
1873         INIT_LIST_HEAD(&xprt->bc_pa_list);
1874 #endif /* CONFIG_SUNRPC_BACKCHANNEL */
1875         INIT_LIST_HEAD(&xprt->xprt_switch);
1876
1877         xprt->last_used = jiffies;
1878         xprt->cwnd = RPC_INITCWND;
1879         xprt->bind_index = 0;
1880
1881         rpc_init_wait_queue(&xprt->binding, "xprt_binding");
1882         rpc_init_wait_queue(&xprt->pending, "xprt_pending");
1883         rpc_init_wait_queue(&xprt->sending, "xprt_sending");
1884         rpc_init_priority_wait_queue(&xprt->backlog, "xprt_backlog");
1885
1886         xprt_init_xid(xprt);
1887
1888         xprt->xprt_net = get_net(net);
1889 }
1890
1891 /**
1892  * xprt_create_transport - create an RPC transport
1893  * @args: rpc transport creation arguments
1894  *
1895  */
1896 struct rpc_xprt *xprt_create_transport(struct xprt_create *args)
1897 {
1898         struct rpc_xprt *xprt;
1899         struct xprt_class *t;
1900
1901         spin_lock(&xprt_list_lock);
1902         list_for_each_entry(t, &xprt_list, list) {
1903                 if (t->ident == args->ident) {
1904                         spin_unlock(&xprt_list_lock);
1905                         goto found;
1906                 }
1907         }
1908         spin_unlock(&xprt_list_lock);
1909         dprintk("RPC: transport (%d) not supported\n", args->ident);
1910         return ERR_PTR(-EIO);
1911
1912 found:
1913         xprt = t->setup(args);
1914         if (IS_ERR(xprt))
1915                 goto out;
1916         if (args->flags & XPRT_CREATE_NO_IDLE_TIMEOUT)
1917                 xprt->idle_timeout = 0;
1918         INIT_WORK(&xprt->task_cleanup, xprt_autoclose);
1919         if (xprt_has_timer(xprt))
1920                 timer_setup(&xprt->timer, xprt_init_autodisconnect, 0);
1921         else
1922                 timer_setup(&xprt->timer, NULL, 0);
1923
1924         if (strlen(args->servername) > RPC_MAXNETNAMELEN) {
1925                 xprt_destroy(xprt);
1926                 return ERR_PTR(-EINVAL);
1927         }
1928         xprt->servername = kstrdup(args->servername, GFP_KERNEL);
1929         if (xprt->servername == NULL) {
1930                 xprt_destroy(xprt);
1931                 return ERR_PTR(-ENOMEM);
1932         }
1933
1934         rpc_xprt_debugfs_register(xprt);
1935
1936         trace_xprt_create(xprt);
1937 out:
1938         return xprt;
1939 }
1940
1941 static void xprt_destroy_cb(struct work_struct *work)
1942 {
1943         struct rpc_xprt *xprt =
1944                 container_of(work, struct rpc_xprt, task_cleanup);
1945
1946         trace_xprt_destroy(xprt);
1947
1948         rpc_xprt_debugfs_unregister(xprt);
1949         rpc_destroy_wait_queue(&xprt->binding);
1950         rpc_destroy_wait_queue(&xprt->pending);
1951         rpc_destroy_wait_queue(&xprt->sending);
1952         rpc_destroy_wait_queue(&xprt->backlog);
1953         kfree(xprt->servername);
1954         /*
1955          * Destroy any existing back channel
1956          */
1957         xprt_destroy_backchannel(xprt, UINT_MAX);
1958
1959         /*
1960          * Tear down transport state and free the rpc_xprt
1961          */
1962         xprt->ops->destroy(xprt);
1963 }
1964
1965 /**
1966  * xprt_destroy - destroy an RPC transport, killing off all requests.
1967  * @xprt: transport to destroy
1968  *
1969  */
1970 static void xprt_destroy(struct rpc_xprt *xprt)
1971 {
1972         /*
1973          * Exclude transport connect/disconnect handlers and autoclose
1974          */
1975         wait_on_bit_lock(&xprt->state, XPRT_LOCKED, TASK_UNINTERRUPTIBLE);
1976
1977         del_timer_sync(&xprt->timer);
1978
1979         /*
1980          * Destroy sockets etc from the system workqueue so they can
1981          * safely flush receive work running on rpciod.
1982          */
1983         INIT_WORK(&xprt->task_cleanup, xprt_destroy_cb);
1984         schedule_work(&xprt->task_cleanup);
1985 }
1986
1987 static void xprt_destroy_kref(struct kref *kref)
1988 {
1989         xprt_destroy(container_of(kref, struct rpc_xprt, kref));
1990 }
1991
1992 /**
1993  * xprt_get - return a reference to an RPC transport.
1994  * @xprt: pointer to the transport
1995  *
1996  */
1997 struct rpc_xprt *xprt_get(struct rpc_xprt *xprt)
1998 {
1999         if (xprt != NULL && kref_get_unless_zero(&xprt->kref))
2000                 return xprt;
2001         return NULL;
2002 }
2003 EXPORT_SYMBOL_GPL(xprt_get);
2004
2005 /**
2006  * xprt_put - release a reference to an RPC transport.
2007  * @xprt: pointer to the transport
2008  *
2009  */
2010 void xprt_put(struct rpc_xprt *xprt)
2011 {
2012         if (xprt != NULL)
2013                 kref_put(&xprt->kref, xprt_destroy_kref);
2014 }
2015 EXPORT_SYMBOL_GPL(xprt_put);