kernel/net/sunrpc/xprt.c

   1 /*
   2  *  linux/net/sunrpc/xprt.c
   3  *
   4  *  This is a generic RPC call interface supporting congestion avoidance,
   5  *  and asynchronous calls.
   6  *
   7  *  The interface works like this:
   8  *
   9  *  -   When a process places a call, it allocates a request slot if
  10  *      one is available. Otherwise, it sleeps on the backlog queue
  11  *      (xprt_reserve).
  12  *  -   Next, the caller puts together the RPC message, stuffs it into
  13  *      the request struct, and calls xprt_transmit().
  14  *  -   xprt_transmit sends the message and installs the caller on the
  15  *      transport's wait list. At the same time, if a reply is expected,
  16  *      it installs a timer that is run after the packet's timeout has
  17  *      expired.
  18  *  -   When a packet arrives, the data_ready handler walks the list of
  19  *      pending requests for that transport. If a matching XID is found, the
  20  *      caller is woken up, and the timer removed.
  21  *  -   When no reply arrives within the timeout interval, the timer is
  22  *      fired by the kernel and runs xprt_timer(). It either adjusts the
  23  *      timeout values (minor timeout) or wakes up the caller with a status
  24  *      of -ETIMEDOUT.
  25  *  -   When the caller receives a notification from RPC that a reply arrived,
  26  *      it should release the RPC slot, and process the reply.
  27  *      If the call timed out, it may choose to retry the operation by
  28  *      adjusting the initial timeout value, and simply calling rpc_call
  29  *      again.
  30  *
  31  *  Support for async RPC is done through a set of RPC-specific scheduling
  32  *  primitives that `transparently' work for processes as well as async
  33  *  tasks that rely on callbacks.
  34  *
  35  *  Copyright (C) 1995-1997, Olaf Kirch <okir@monad.swb.de>
  36  *
  37  *  Transport switch API copyright (C) 2005, Chuck Lever <cel@netapp.com>
  38  */
  39
  40 #include <linux/module.h>
  41
  42 #include <linux/types.h>
  43 #include <linux/interrupt.h>
  44 #include <linux/workqueue.h>
  45 #include <linux/net.h>
  46 #include <linux/ktime.h>
  47
  48 #include <linux/sunrpc/clnt.h>
  49 #include <linux/sunrpc/metrics.h>
  50 #include <linux/sunrpc/bc_xprt.h>
  51
  52 #include <trace/events/sunrpc.h>
  53
  54 #include "sunrpc.h"
  55
  56 /*
  57  * Local variables
  58  */
  59
  60 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
  61 # define RPCDBG_FACILITY        RPCDBG_XPRT
  62 #endif
  63
  64 /*
  65  * Local functions
  66  */
  67 static void      xprt_init(struct rpc_xprt *xprt, struct net *net);
  68 static void     xprt_request_init(struct rpc_task *, struct rpc_xprt *);
  69 static void     xprt_connect_status(struct rpc_task *task);
  70 static int      __xprt_get_cong(struct rpc_xprt *, struct rpc_task *);
  71 static void      xprt_destroy(struct rpc_xprt *xprt);
  72
  73 static DEFINE_SPINLOCK(xprt_list_lock);
  74 static LIST_HEAD(xprt_list);
  75
  76 /**
  77  * xprt_register_transport - register a transport implementation
  78  * @transport: transport to register
  79  *
  80  * If a transport implementation is loaded as a kernel module, it can
  81  * call this interface to make itself known to the RPC client.
  82  *
  83  * Returns:
  84  * 0:           transport successfully registered
  85  * -EEXIST:     transport already registered
  86  * -EINVAL:     transport module being unloaded
  87  */
  88 int xprt_register_transport(struct xprt_class *transport)
  89 {
  90         struct xprt_class *t;
  91         int result;
  92
  93         result = -EEXIST;
  94         spin_lock(&xprt_list_lock);
  95         list_for_each_entry(t, &xprt_list, list) {
  96                 /* don't register the same transport class twice */
  97                 if (t->ident == transport->ident)
  98                         goto out;
  99         }
 100
 101         list_add_tail(&transport->list, &xprt_list);
 102         printk(KERN_INFO "RPC: Registered %s transport module.\n",
 103                transport->name);
 104         result = 0;
 105
 106 out:
 107         spin_unlock(&xprt_list_lock);
 108         return result;
 109 }
 110 EXPORT_SYMBOL_GPL(xprt_register_transport);
 111
 112 /**
 113  * xprt_unregister_transport - unregister a transport implementation
 114  * @transport: transport to unregister
 115  *
 116  * Returns:
 117  * 0:           transport successfully unregistered
 118  * -ENOENT:     transport never registered
 119  */
 120 int xprt_unregister_transport(struct xprt_class *transport)
 121 {
 122         struct xprt_class *t;
 123         int result;
 124
 125         result = 0;
 126         spin_lock(&xprt_list_lock);
 127         list_for_each_entry(t, &xprt_list, list) {
 128                 if (t == transport) {
 129                         printk(KERN_INFO
 130                                 "RPC: Unregistered %s transport module.\n",
 131                                 transport->name);
 132                         list_del_init(&transport->list);
 133                         goto out;
 134                 }
 135         }
 136         result = -ENOENT;
 137
 138 out:
 139         spin_unlock(&xprt_list_lock);
 140         return result;
 141 }
 142 EXPORT_SYMBOL_GPL(xprt_unregister_transport);
 143
 144 /**
 145  * xprt_load_transport - load a transport implementation
 146  * @transport_name: transport to load
 147  *
 148  * Returns:
 149  * 0:           transport successfully loaded
 150  * -ENOENT:     transport module not available
 151  */
 152 int xprt_load_transport(const char *transport_name)
 153 {
 154         struct xprt_class *t;
 155         int result;
 156
 157         result = 0;
 158         spin_lock(&xprt_list_lock);
 159         list_for_each_entry(t, &xprt_list, list) {
 160                 if (strcmp(t->name, transport_name) == 0) {
 161                         spin_unlock(&xprt_list_lock);
 162                         goto out;
 163                 }
 164         }
 165         spin_unlock(&xprt_list_lock);
 166         result = request_module("xprt%s", transport_name);
 167 out:
 168         return result;
 169 }
 170 EXPORT_SYMBOL_GPL(xprt_load_transport);
 171
 172 /**
 173  * xprt_reserve_xprt - serialize write access to transports
 174  * @task: task that is requesting access to the transport
 175  * @xprt: pointer to the target transport
 176  *
 177  * This prevents mixing the payload of separate requests, and prevents
 178  * transport connects from colliding with writes.  No congestion control
 179  * is provided.
 180  */
 181 int xprt_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
 182 {
 183         struct rpc_rqst *req = task->tk_rqstp;
 184         int priority;
 185
 186         if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) {
 187                 if (task == xprt->snd_task)
 188                         return 1;
 189                 goto out_sleep;
 190         }
 191         xprt->snd_task = task;
 192         if (req != NULL)
 193                 req->rq_ntrans++;
 194
 195         return 1;
 196
 197 out_sleep:
 198         dprintk("RPC: %5u failed to lock transport %p\n",
 199                         task->tk_pid, xprt);
 200         task->tk_timeout = 0;
 201         task->tk_status = -EAGAIN;
 202         if (req == NULL)
 203                 priority = RPC_PRIORITY_LOW;
 204         else if (!req->rq_ntrans)
 205                 priority = RPC_PRIORITY_NORMAL;
 206         else
 207                 priority = RPC_PRIORITY_HIGH;
 208         rpc_sleep_on_priority(&xprt->sending, task, NULL, priority);
 209         return 0;
 210 }
 211 EXPORT_SYMBOL_GPL(xprt_reserve_xprt);
 212
 213 static void xprt_clear_locked(struct rpc_xprt *xprt)
 214 {
 215         xprt->snd_task = NULL;
 216         if (!test_bit(XPRT_CLOSE_WAIT, &xprt->state)) {
 217                 smp_mb__before_atomic();
 218                 clear_bit(XPRT_LOCKED, &xprt->state);
 219                 smp_mb__after_atomic();
 220         } else
 221                 queue_work(rpciod_workqueue, &xprt->task_cleanup);
 222 }
 223
 224 /*
 225  * xprt_reserve_xprt_cong - serialize write access to transports
 226  * @task: task that is requesting access to the transport
 227  *
 228  * Same as xprt_reserve_xprt, but Van Jacobson congestion control is
 229  * integrated into the decision of whether a request is allowed to be
 230  * woken up and given access to the transport.
 231  */
 232 int xprt_reserve_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
 233 {
 234         struct rpc_rqst *req = task->tk_rqstp;
 235         int priority;
 236
 237         if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) {
 238                 if (task == xprt->snd_task)
 239                         return 1;
 240                 goto out_sleep;
 241         }
 242         if (req == NULL) {
 243                 xprt->snd_task = task;
 244                 return 1;
 245         }
 246         if (__xprt_get_cong(xprt, task)) {
 247                 xprt->snd_task = task;
 248                 req->rq_ntrans++;
 249                 return 1;
 250         }
 251         xprt_clear_locked(xprt);
 252 out_sleep:
 253         dprintk("RPC: %5u failed to lock transport %p\n", task->tk_pid, xprt);
 254         task->tk_timeout = 0;
 255         task->tk_status = -EAGAIN;
 256         if (req == NULL)
 257                 priority = RPC_PRIORITY_LOW;
 258         else if (!req->rq_ntrans)
 259                 priority = RPC_PRIORITY_NORMAL;
 260         else
 261                 priority = RPC_PRIORITY_HIGH;
 262         rpc_sleep_on_priority(&xprt->sending, task, NULL, priority);
 263         return 0;
 264 }
 265 EXPORT_SYMBOL_GPL(xprt_reserve_xprt_cong);
 266
 267 static inline int xprt_lock_write(struct rpc_xprt *xprt, struct rpc_task *task)
 268 {
 269         int retval;
 270
 271         spin_lock_bh(&xprt->transport_lock);
 272         retval = xprt->ops->reserve_xprt(xprt, task);
 273         spin_unlock_bh(&xprt->transport_lock);
 274         return retval;
 275 }
 276
 277 static bool __xprt_lock_write_func(struct rpc_task *task, void *data)
 278 {
 279         struct rpc_xprt *xprt = data;
 280         struct rpc_rqst *req;
 281
 282         req = task->tk_rqstp;
 283         xprt->snd_task = task;
 284         if (req)
 285                 req->rq_ntrans++;
 286         return true;
 287 }
 288
 289 static void __xprt_lock_write_next(struct rpc_xprt *xprt)
 290 {
 291         if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
 292                 return;
 293
 294         if (rpc_wake_up_first(&xprt->sending, __xprt_lock_write_func, xprt))
 295                 return;
 296         xprt_clear_locked(xprt);
 297 }
 298
 299 static bool __xprt_lock_write_cong_func(struct rpc_task *task, void *data)
 300 {
 301         struct rpc_xprt *xprt = data;
 302         struct rpc_rqst *req;
 303
 304         req = task->tk_rqstp;
 305         if (req == NULL) {
 306                 xprt->snd_task = task;
 307                 return true;
 308         }
 309         if (__xprt_get_cong(xprt, task)) {
 310                 xprt->snd_task = task;
 311                 req->rq_ntrans++;
 312                 return true;
 313         }
 314         return false;
 315 }
 316
 317 static void __xprt_lock_write_next_cong(struct rpc_xprt *xprt)
 318 {
 319         if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
 320                 return;
 321         if (RPCXPRT_CONGESTED(xprt))
 322                 goto out_unlock;
 323         if (rpc_wake_up_first(&xprt->sending, __xprt_lock_write_cong_func, xprt))
 324                 return;
 325 out_unlock:
 326         xprt_clear_locked(xprt);
 327 }
 328
 329 static void xprt_task_clear_bytes_sent(struct rpc_task *task)
 330 {
 331         if (task != NULL) {
 332                 struct rpc_rqst *req = task->tk_rqstp;
 333                 if (req != NULL)
 334                         req->rq_bytes_sent = 0;
 335         }
 336 }
 337
 338 /**
 339  * xprt_release_xprt - allow other requests to use a transport
 340  * @xprt: transport with other tasks potentially waiting
 341  * @task: task that is releasing access to the transport
 342  *
 343  * Note that "task" can be NULL.  No congestion control is provided.
 344  */
 345 void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
 346 {
 347         if (xprt->snd_task == task) {
 348                 xprt_task_clear_bytes_sent(task);
 349                 xprt_clear_locked(xprt);
 350                 __xprt_lock_write_next(xprt);
 351         }
 352 }
 353 EXPORT_SYMBOL_GPL(xprt_release_xprt);
 354
 355 /**
 356  * xprt_release_xprt_cong - allow other requests to use a transport
 357  * @xprt: transport with other tasks potentially waiting
 358  * @task: task that is releasing access to the transport
 359  *
 360  * Note that "task" can be NULL.  Another task is awoken to use the
 361  * transport if the transport's congestion window allows it.
 362  */
 363 void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
 364 {
 365         if (xprt->snd_task == task) {
 366                 xprt_task_clear_bytes_sent(task);
 367                 xprt_clear_locked(xprt);
 368                 __xprt_lock_write_next_cong(xprt);
 369         }
 370 }
 371 EXPORT_SYMBOL_GPL(xprt_release_xprt_cong);
 372
 373 static inline void xprt_release_write(struct rpc_xprt *xprt, struct rpc_task *task)
 374 {
 375         spin_lock_bh(&xprt->transport_lock);
 376         xprt->ops->release_xprt(xprt, task);
 377         spin_unlock_bh(&xprt->transport_lock);
 378 }
 379
 380 /*
 381  * Van Jacobson congestion avoidance. Check if the congestion window
 382  * overflowed. Put the task to sleep if this is the case.
 383  */
 384 static int
 385 __xprt_get_cong(struct rpc_xprt *xprt, struct rpc_task *task)
 386 {
 387         struct rpc_rqst *req = task->tk_rqstp;
 388
 389         if (req->rq_cong)
 390                 return 1;
 391         dprintk("RPC: %5u xprt_cwnd_limited cong = %lu cwnd = %lu\n",
 392                         task->tk_pid, xprt->cong, xprt->cwnd);
 393         if (RPCXPRT_CONGESTED(xprt))
 394                 return 0;
 395         req->rq_cong = 1;
 396         xprt->cong += RPC_CWNDSCALE;
 397         return 1;
 398 }
 399
 400 /*
 401  * Adjust the congestion window, and wake up the next task
 402  * that has been sleeping due to congestion
 403  */
 404 static void
 405 __xprt_put_cong(struct rpc_xprt *xprt, struct rpc_rqst *req)
 406 {
 407         if (!req->rq_cong)
 408                 return;
 409         req->rq_cong = 0;
 410         xprt->cong -= RPC_CWNDSCALE;
 411         __xprt_lock_write_next_cong(xprt);
 412 }
 413
 414 /**
 415  * xprt_release_rqst_cong - housekeeping when request is complete
 416  * @task: RPC request that recently completed
 417  *
 418  * Useful for transports that require congestion control.
 419  */
 420 void xprt_release_rqst_cong(struct rpc_task *task)
 421 {
 422         struct rpc_rqst *req = task->tk_rqstp;
 423
 424         __xprt_put_cong(req->rq_xprt, req);
 425 }
 426 EXPORT_SYMBOL_GPL(xprt_release_rqst_cong);
 427
 428 /**
 429  * xprt_adjust_cwnd - adjust transport congestion window
 430  * @xprt: pointer to xprt
 431  * @task: recently completed RPC request used to adjust window
 432  * @result: result code of completed RPC request
 433  *
 434  * The transport code maintains an estimate on the maximum number of out-
 435  * standing RPC requests, using a smoothed version of the congestion
 436  * avoidance implemented in 44BSD. This is basically the Van Jacobson
 437  * congestion algorithm: If a retransmit occurs, the congestion window is
 438  * halved; otherwise, it is incremented by 1/cwnd when
 439  *
 440  *      -       a reply is received and
 441  *      -       a full number of requests are outstanding and
 442  *      -       the congestion window hasn't been updated recently.
 443  */
 444 void xprt_adjust_cwnd(struct rpc_xprt *xprt, struct rpc_task *task, int result)
 445 {
 446         struct rpc_rqst *req = task->tk_rqstp;
 447         unsigned long cwnd = xprt->cwnd;
 448
 449         if (result >= 0 && cwnd <= xprt->cong) {
 450                 /* The (cwnd >> 1) term makes sure
 451                  * the result gets rounded properly. */
 452                 cwnd += (RPC_CWNDSCALE * RPC_CWNDSCALE + (cwnd >> 1)) / cwnd;
 453                 if (cwnd > RPC_MAXCWND(xprt))
 454                         cwnd = RPC_MAXCWND(xprt);
 455                 __xprt_lock_write_next_cong(xprt);
 456         } else if (result == -ETIMEDOUT) {
 457                 cwnd >>= 1;
 458                 if (cwnd < RPC_CWNDSCALE)
 459                         cwnd = RPC_CWNDSCALE;
 460         }
 461         dprintk("RPC:       cong %ld, cwnd was %ld, now %ld\n",
 462                         xprt->cong, xprt->cwnd, cwnd);
 463         xprt->cwnd = cwnd;
 464         __xprt_put_cong(xprt, req);
 465 }
 466 EXPORT_SYMBOL_GPL(xprt_adjust_cwnd);
 467
 468 /**
 469  * xprt_wake_pending_tasks - wake all tasks on a transport's pending queue
 470  * @xprt: transport with waiting tasks
 471  * @status: result code to plant in each task before waking it
 472  *
 473  */
 474 void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status)
 475 {
 476         if (status < 0)
 477                 rpc_wake_up_status(&xprt->pending, status);
 478         else
 479                 rpc_wake_up(&xprt->pending);
 480 }
 481 EXPORT_SYMBOL_GPL(xprt_wake_pending_tasks);
 482
 483 /**
 484  * xprt_wait_for_buffer_space - wait for transport output buffer to clear
 485  * @task: task to be put to sleep
 486  * @action: function pointer to be executed after wait
 487  *
 488  * Note that we only set the timer for the case of RPC_IS_SOFT(), since
 489  * we don't in general want to force a socket disconnection due to
 490  * an incomplete RPC call transmission.
 491  */
 492 void xprt_wait_for_buffer_space(struct rpc_task *task, rpc_action action)
 493 {
 494         struct rpc_rqst *req = task->tk_rqstp;
 495         struct rpc_xprt *xprt = req->rq_xprt;
 496
 497         task->tk_timeout = RPC_IS_SOFT(task) ? req->rq_timeout : 0;
 498         rpc_sleep_on(&xprt->pending, task, action);
 499 }
 500 EXPORT_SYMBOL_GPL(xprt_wait_for_buffer_space);
 501
 502 /**
 503  * xprt_write_space - wake the task waiting for transport output buffer space
 504  * @xprt: transport with waiting tasks
 505  *
 506  * Can be called in a soft IRQ context, so xprt_write_space never sleeps.
 507  */
 508 void xprt_write_space(struct rpc_xprt *xprt)
 509 {
 510         spin_lock_bh(&xprt->transport_lock);
 511         if (xprt->snd_task) {
 512                 dprintk("RPC:       write space: waking waiting task on "
 513                                 "xprt %p\n", xprt);
 514                 rpc_wake_up_queued_task(&xprt->pending, xprt->snd_task);
 515         }
 516         spin_unlock_bh(&xprt->transport_lock);
 517 }
 518 EXPORT_SYMBOL_GPL(xprt_write_space);
 519
 520 /**
 521  * xprt_set_retrans_timeout_def - set a request's retransmit timeout
 522  * @task: task whose timeout is to be set
 523  *
 524  * Set a request's retransmit timeout based on the transport's
 525  * default timeout parameters.  Used by transports that don't adjust
 526  * the retransmit timeout based on round-trip time estimation.
 527  */
 528 void xprt_set_retrans_timeout_def(struct rpc_task *task)
 529 {
 530         task->tk_timeout = task->tk_rqstp->rq_timeout;
 531 }
 532 EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_def);
 533
 534 /**
 535  * xprt_set_retrans_timeout_rtt - set a request's retransmit timeout
 536  * @task: task whose timeout is to be set
 537  *
 538  * Set a request's retransmit timeout using the RTT estimator.
 539  */
 540 void xprt_set_retrans_timeout_rtt(struct rpc_task *task)
 541 {
 542         int timer = task->tk_msg.rpc_proc->p_timer;
 543         struct rpc_clnt *clnt = task->tk_client;
 544         struct rpc_rtt *rtt = clnt->cl_rtt;
 545         struct rpc_rqst *req = task->tk_rqstp;
 546         unsigned long max_timeout = clnt->cl_timeout->to_maxval;
 547
 548         task->tk_timeout = rpc_calc_rto(rtt, timer);
 549         task->tk_timeout <<= rpc_ntimeo(rtt, timer) + req->rq_retries;
 550         if (task->tk_timeout > max_timeout || task->tk_timeout == 0)
 551                 task->tk_timeout = max_timeout;
 552 }
 553 EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_rtt);
 554
 555 static void xprt_reset_majortimeo(struct rpc_rqst *req)
 556 {
 557         const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
 558
 559         req->rq_majortimeo = req->rq_timeout;
 560         if (to->to_exponential)
 561                 req->rq_majortimeo <<= to->to_retries;
 562         else
 563                 req->rq_majortimeo += to->to_increment * to->to_retries;
 564         if (req->rq_majortimeo > to->to_maxval || req->rq_majortimeo == 0)
 565                 req->rq_majortimeo = to->to_maxval;
 566         req->rq_majortimeo += jiffies;
 567 }
 568
 569 /**
 570  * xprt_adjust_timeout - adjust timeout values for next retransmit
 571  * @req: RPC request containing parameters to use for the adjustment
 572  *
 573  */
 574 int xprt_adjust_timeout(struct rpc_rqst *req)
 575 {
 576         struct rpc_xprt *xprt = req->rq_xprt;
 577         const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
 578         int status = 0;
 579
 580         if (time_before(jiffies, req->rq_majortimeo)) {
 581                 if (to->to_exponential)
 582                         req->rq_timeout <<= 1;
 583                 else
 584                         req->rq_timeout += to->to_increment;
 585                 if (to->to_maxval && req->rq_timeout >= to->to_maxval)
 586                         req->rq_timeout = to->to_maxval;
 587                 req->rq_retries++;
 588         } else {
 589                 req->rq_timeout = to->to_initval;
 590                 req->rq_retries = 0;
 591                 xprt_reset_majortimeo(req);
 592                 /* Reset the RTT counters == "slow start" */
 593                 spin_lock_bh(&xprt->transport_lock);
 594                 rpc_init_rtt(req->rq_task->tk_client->cl_rtt, to->to_initval);
 595                 spin_unlock_bh(&xprt->transport_lock);
 596                 status = -ETIMEDOUT;
 597         }
 598
 599         if (req->rq_timeout == 0) {
 600                 printk(KERN_WARNING "xprt_adjust_timeout: rq_timeout = 0!\n");
 601                 req->rq_timeout = 5 * HZ;
 602         }
 603         return status;
 604 }
 605
 606 static void xprt_autoclose(struct work_struct *work)
 607 {
 608         struct rpc_xprt *xprt =
 609                 container_of(work, struct rpc_xprt, task_cleanup);
 610
 611         xprt->ops->close(xprt);
 612         clear_bit(XPRT_CLOSE_WAIT, &xprt->state);
 613         xprt_release_write(xprt, NULL);
 614         wake_up_bit(&xprt->state, XPRT_LOCKED);
 615 }
 616
 617 /**
 618  * xprt_disconnect_done - mark a transport as disconnected
 619  * @xprt: transport to flag for disconnect
 620  *
 621  */
 622 void xprt_disconnect_done(struct rpc_xprt *xprt)
 623 {
 624         dprintk("RPC:       disconnected transport %p\n", xprt);
 625         spin_lock_bh(&xprt->transport_lock);
 626         xprt_clear_connected(xprt);
 627         xprt_wake_pending_tasks(xprt, -EAGAIN);
 628         spin_unlock_bh(&xprt->transport_lock);
 629 }
 630 EXPORT_SYMBOL_GPL(xprt_disconnect_done);
 631
 632 /**
 633  * xprt_force_disconnect - force a transport to disconnect
 634  * @xprt: transport to disconnect
 635  *
 636  */
 637 void xprt_force_disconnect(struct rpc_xprt *xprt)
 638 {
 639         /* Don't race with the test_bit() in xprt_clear_locked() */
 640         spin_lock_bh(&xprt->transport_lock);
 641         set_bit(XPRT_CLOSE_WAIT, &xprt->state);
 642         /* Try to schedule an autoclose RPC call */
 643         if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
 644                 queue_work(rpciod_workqueue, &xprt->task_cleanup);
 645         xprt_wake_pending_tasks(xprt, -EAGAIN);
 646         spin_unlock_bh(&xprt->transport_lock);
 647 }
 648
 649 /**
 650  * xprt_conditional_disconnect - force a transport to disconnect
 651  * @xprt: transport to disconnect
 652  * @cookie: 'connection cookie'
 653  *
 654  * This attempts to break the connection if and only if 'cookie' matches
 655  * the current transport 'connection cookie'. It ensures that we don't
 656  * try to break the connection more than once when we need to retransmit
 657  * a batch of RPC requests.
 658  *
 659  */
 660 void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie)
 661 {
 662         /* Don't race with the test_bit() in xprt_clear_locked() */
 663         spin_lock_bh(&xprt->transport_lock);
 664         if (cookie != xprt->connect_cookie)
 665                 goto out;
 666         if (test_bit(XPRT_CLOSING, &xprt->state) || !xprt_connected(xprt))
 667                 goto out;
 668         set_bit(XPRT_CLOSE_WAIT, &xprt->state);
 669         /* Try to schedule an autoclose RPC call */
 670         if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
 671                 queue_work(rpciod_workqueue, &xprt->task_cleanup);
 672         xprt_wake_pending_tasks(xprt, -EAGAIN);
 673 out:
 674         spin_unlock_bh(&xprt->transport_lock);
 675 }
 676
 677 static void
 678 xprt_init_autodisconnect(unsigned long data)
 679 {
 680         struct rpc_xprt *xprt = (struct rpc_xprt *)data;
 681
 682         spin_lock(&xprt->transport_lock);
 683         if (!list_empty(&xprt->recv))
 684                 goto out_abort;
 685         if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
 686                 goto out_abort;
 687         spin_unlock(&xprt->transport_lock);
 688         queue_work(rpciod_workqueue, &xprt->task_cleanup);
 689         return;
 690 out_abort:
 691         spin_unlock(&xprt->transport_lock);
 692 }
 693
 694 bool xprt_lock_connect(struct rpc_xprt *xprt,
 695                 struct rpc_task *task,
 696                 void *cookie)
 697 {
 698         bool ret = false;
 699
 700         spin_lock_bh(&xprt->transport_lock);
 701         if (!test_bit(XPRT_LOCKED, &xprt->state))
 702                 goto out;
 703         if (xprt->snd_task != task)
 704                 goto out;
 705         xprt_task_clear_bytes_sent(task);
 706         xprt->snd_task = cookie;
 707         ret = true;
 708 out:
 709         spin_unlock_bh(&xprt->transport_lock);
 710         return ret;
 711 }
 712
 713 void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie)
 714 {
 715         spin_lock_bh(&xprt->transport_lock);
 716         if (xprt->snd_task != cookie)
 717                 goto out;
 718         if (!test_bit(XPRT_LOCKED, &xprt->state))
 719                 goto out;
 720         xprt->snd_task =NULL;
 721         xprt->ops->release_xprt(xprt, NULL);
 722 out:
 723         spin_unlock_bh(&xprt->transport_lock);
 724         wake_up_bit(&xprt->state, XPRT_LOCKED);
 725 }
 726
 727 /**
 728  * xprt_connect - schedule a transport connect operation
 729  * @task: RPC task that is requesting the connect
 730  *
 731  */
 732 void xprt_connect(struct rpc_task *task)
 733 {
 734         struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt;
 735
 736         dprintk("RPC: %5u xprt_connect xprt %p %s connected\n", task->tk_pid,
 737                         xprt, (xprt_connected(xprt) ? "is" : "is not"));
 738
 739         if (!xprt_bound(xprt)) {
 740                 task->tk_status = -EAGAIN;
 741                 return;
 742         }
 743         if (!xprt_lock_write(xprt, task))
 744                 return;
 745
 746         if (test_and_clear_bit(XPRT_CLOSE_WAIT, &xprt->state))
 747                 xprt->ops->close(xprt);
 748
 749         if (!xprt_connected(xprt)) {
 750                 task->tk_rqstp->rq_bytes_sent = 0;
 751                 task->tk_timeout = task->tk_rqstp->rq_timeout;
 752                 rpc_sleep_on(&xprt->pending, task, xprt_connect_status);
 753
 754                 if (test_bit(XPRT_CLOSING, &xprt->state))
 755                         return;
 756                 if (xprt_test_and_set_connecting(xprt))
 757                         return;
 758                 xprt->stat.connect_start = jiffies;
 759                 xprt->ops->connect(xprt, task);
 760         }
 761         xprt_release_write(xprt, task);
 762 }
 763
 764 static void xprt_connect_status(struct rpc_task *task)
 765 {
 766         struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt;
 767
 768         if (task->tk_status == 0) {
 769                 xprt->stat.connect_count++;
 770                 xprt->stat.connect_time += (long)jiffies - xprt->stat.connect_start;
 771                 dprintk("RPC: %5u xprt_connect_status: connection established\n",
 772                                 task->tk_pid);
 773                 return;
 774         }
 775
 776         switch (task->tk_status) {
 777         case -ECONNREFUSED:
 778         case -ECONNRESET:
 779         case -ECONNABORTED:
 780         case -ENETUNREACH:
 781         case -EHOSTUNREACH:
 782         case -EPIPE:
 783         case -EAGAIN:
 784                 dprintk("RPC: %5u xprt_connect_status: retrying\n", task->tk_pid);
 785                 break;
 786         case -ETIMEDOUT:
 787                 dprintk("RPC: %5u xprt_connect_status: connect attempt timed "
 788                                 "out\n", task->tk_pid);
 789                 break;
 790         default:
 791                 dprintk("RPC: %5u xprt_connect_status: error %d connecting to "
 792                                 "server %s\n", task->tk_pid, -task->tk_status,
 793                                 xprt->servername);
 794                 task->tk_status = -EIO;
 795         }
 796 }
 797
 798 /**
 799  * xprt_lookup_rqst - find an RPC request corresponding to an XID
 800  * @xprt: transport on which the original request was transmitted
 801  * @xid: RPC XID of incoming reply
 802  *
 803  */
 804 struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid)
 805 {
 806         struct rpc_rqst *entry;
 807
 808         list_for_each_entry(entry, &xprt->recv, rq_list)
 809                 if (entry->rq_xid == xid) {
 810                         trace_xprt_lookup_rqst(xprt, xid, 0);
 811                         return entry;
 812                 }
 813
 814         dprintk("RPC:       xprt_lookup_rqst did not find xid %08x\n",
 815                         ntohl(xid));
 816         trace_xprt_lookup_rqst(xprt, xid, -ENOENT);
 817         xprt->stat.bad_xids++;
 818         return NULL;
 819 }
 820 EXPORT_SYMBOL_GPL(xprt_lookup_rqst);
 821
 822 static void xprt_update_rtt(struct rpc_task *task)
 823 {
 824         struct rpc_rqst *req = task->tk_rqstp;
 825         struct rpc_rtt *rtt = task->tk_client->cl_rtt;
 826         unsigned int timer = task->tk_msg.rpc_proc->p_timer;
 827         long m = usecs_to_jiffies(ktime_to_us(req->rq_rtt));
 828
 829         if (timer) {
 830                 if (req->rq_ntrans == 1)
 831                         rpc_update_rtt(rtt, timer, m);
 832                 rpc_set_timeo(rtt, timer, req->rq_ntrans - 1);
 833         }
 834 }
 835
 836 /**
 837  * xprt_complete_rqst - called when reply processing is complete
 838  * @task: RPC request that recently completed
 839  * @copied: actual number of bytes received from the transport
 840  *
 841  * Caller holds transport lock.
 842  */
 843 void xprt_complete_rqst(struct rpc_task *task, int copied)
 844 {
 845         struct rpc_rqst *req = task->tk_rqstp;
 846         struct rpc_xprt *xprt = req->rq_xprt;
 847
 848         dprintk("RPC: %5u xid %08x complete (%d bytes received)\n",
 849                         task->tk_pid, ntohl(req->rq_xid), copied);
 850         trace_xprt_complete_rqst(xprt, req->rq_xid, copied);
 851
 852         xprt->stat.recvs++;
 853         req->rq_rtt = ktime_sub(ktime_get(), req->rq_xtime);
 854         if (xprt->ops->timer != NULL)
 855                 xprt_update_rtt(task);
 856
 857         list_del_init(&req->rq_list);
 858         req->rq_private_buf.len = copied;
 859         /* Ensure all writes are done before we update */
 860         /* req->rq_reply_bytes_recvd */
 861         smp_wmb();
 862         req->rq_reply_bytes_recvd = copied;
 863         rpc_wake_up_queued_task(&xprt->pending, task);
 864 }
 865 EXPORT_SYMBOL_GPL(xprt_complete_rqst);
 866
 867 static void xprt_timer(struct rpc_task *task)
 868 {
 869         struct rpc_rqst *req = task->tk_rqstp;
 870         struct rpc_xprt *xprt = req->rq_xprt;
 871
 872         if (task->tk_status != -ETIMEDOUT)
 873                 return;
 874         dprintk("RPC: %5u xprt_timer\n", task->tk_pid);
 875
 876         spin_lock_bh(&xprt->transport_lock);
 877         if (!req->rq_reply_bytes_recvd) {
 878                 if (xprt->ops->timer)
 879                         xprt->ops->timer(xprt, task);
 880         } else
 881                 task->tk_status = 0;
 882         spin_unlock_bh(&xprt->transport_lock);
 883 }
 884
 885 static inline int xprt_has_timer(struct rpc_xprt *xprt)
 886 {
 887         return xprt->idle_timeout != 0;
 888 }
 889
 890 /**
 891  * xprt_prepare_transmit - reserve the transport before sending a request
 892  * @task: RPC task about to send a request
 893  *
 894  */
 895 bool xprt_prepare_transmit(struct rpc_task *task)
 896 {
 897         struct rpc_rqst *req = task->tk_rqstp;
 898         struct rpc_xprt *xprt = req->rq_xprt;
 899         bool ret = false;
 900
 901         dprintk("RPC: %5u xprt_prepare_transmit\n", task->tk_pid);
 902
 903         spin_lock_bh(&xprt->transport_lock);
 904         if (!req->rq_bytes_sent) {
 905                 if (req->rq_reply_bytes_recvd) {
 906                         task->tk_status = req->rq_reply_bytes_recvd;
 907                         goto out_unlock;
 908                 }
 909                 if ((task->tk_flags & RPC_TASK_NO_RETRANS_TIMEOUT)
 910                     && xprt_connected(xprt)
 911                     && req->rq_connect_cookie == xprt->connect_cookie) {
 912                         xprt->ops->set_retrans_timeout(task);
 913                         rpc_sleep_on(&xprt->pending, task, xprt_timer);
 914                         goto out_unlock;
 915                 }
 916         }
 917         if (!xprt->ops->reserve_xprt(xprt, task)) {
 918                 task->tk_status = -EAGAIN;
 919                 goto out_unlock;
 920         }
 921         ret = true;
 922 out_unlock:
 923         spin_unlock_bh(&xprt->transport_lock);
 924         return ret;
 925 }
 926
 927 void xprt_end_transmit(struct rpc_task *task)
 928 {
 929         xprt_release_write(task->tk_rqstp->rq_xprt, task);
 930 }
 931
 932 /**
 933  * xprt_transmit - send an RPC request on a transport
 934  * @task: controlling RPC task
 935  *
 936  * We have to copy the iovec because sendmsg fiddles with its contents.
 937  */
 938 void xprt_transmit(struct rpc_task *task)
 939 {
 940         struct rpc_rqst *req = task->tk_rqstp;
 941         struct rpc_xprt *xprt = req->rq_xprt;
 942         int status, numreqs;
 943
 944         dprintk("RPC: %5u xprt_transmit(%u)\n", task->tk_pid, req->rq_slen);
 945
 946         if (!req->rq_reply_bytes_recvd) {
 947                 if (list_empty(&req->rq_list) && rpc_reply_expected(task)) {
 948                         /*
 949                          * Add to the list only if we're expecting a reply
 950                          */
 951                         spin_lock_bh(&xprt->transport_lock);
 952                         /* Update the softirq receive buffer */
 953                         memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
 954                                         sizeof(req->rq_private_buf));
 955                         /* Add request to the receive list */
 956                         list_add_tail(&req->rq_list, &xprt->recv);
 957                         spin_unlock_bh(&xprt->transport_lock);
 958                         xprt_reset_majortimeo(req);
 959                         /* Turn off autodisconnect */
 960                         del_singleshot_timer_sync(&xprt->timer);
 961                 }
 962         } else if (!req->rq_bytes_sent)
 963                 return;
 964
 965         req->rq_xtime = ktime_get();
 966         status = xprt->ops->send_request(task);
 967         trace_xprt_transmit(xprt, req->rq_xid, status);
 968         if (status != 0) {
 969                 task->tk_status = status;
 970                 return;
 971         }
 972
 973         dprintk("RPC: %5u xmit complete\n", task->tk_pid);
 974         task->tk_flags |= RPC_TASK_SENT;
 975         spin_lock_bh(&xprt->transport_lock);
 976
 977         xprt->ops->set_retrans_timeout(task);
 978
 979         numreqs = atomic_read(&xprt->num_reqs);
 980         if (numreqs > xprt->stat.max_slots)
 981                 xprt->stat.max_slots = numreqs;
 982         xprt->stat.sends++;
 983         xprt->stat.req_u += xprt->stat.sends - xprt->stat.recvs;
 984         xprt->stat.bklog_u += xprt->backlog.qlen;
 985         xprt->stat.sending_u += xprt->sending.qlen;
 986         xprt->stat.pending_u += xprt->pending.qlen;
 987
 988         /* Don't race with disconnect */
 989         if (!xprt_connected(xprt))
 990                 task->tk_status = -ENOTCONN;
 991         else {
 992                 /*
 993                  * Sleep on the pending queue since
 994                  * we're expecting a reply.
 995                  */
 996                 if (!req->rq_reply_bytes_recvd && rpc_reply_expected(task))
 997                         rpc_sleep_on(&xprt->pending, task, xprt_timer);
 998                 req->rq_connect_cookie = xprt->connect_cookie;
 999         }
1000         spin_unlock_bh(&xprt->transport_lock);
1001 }
1002
1003 static void xprt_add_backlog(struct rpc_xprt *xprt, struct rpc_task *task)
1004 {
1005         set_bit(XPRT_CONGESTED, &xprt->state);
1006         rpc_sleep_on(&xprt->backlog, task, NULL);
1007 }
1008
1009 static void xprt_wake_up_backlog(struct rpc_xprt *xprt)
1010 {
1011         if (rpc_wake_up_next(&xprt->backlog) == NULL)
1012                 clear_bit(XPRT_CONGESTED, &xprt->state);
1013 }
1014
1015 static bool xprt_throttle_congested(struct rpc_xprt *xprt, struct rpc_task *task)
1016 {
1017         bool ret = false;
1018
1019         if (!test_bit(XPRT_CONGESTED, &xprt->state))
1020                 goto out;
1021         spin_lock(&xprt->reserve_lock);
1022         if (test_bit(XPRT_CONGESTED, &xprt->state)) {
1023                 rpc_sleep_on(&xprt->backlog, task, NULL);
1024                 ret = true;
1025         }
1026         spin_unlock(&xprt->reserve_lock);
1027 out:
1028         return ret;
1029 }
1030
1031 static struct rpc_rqst *xprt_dynamic_alloc_slot(struct rpc_xprt *xprt, gfp_t gfp_flags)
1032 {
1033         struct rpc_rqst *req = ERR_PTR(-EAGAIN);
1034
1035         if (!atomic_add_unless(&xprt->num_reqs, 1, xprt->max_reqs))
1036                 goto out;
1037         req = kzalloc(sizeof(struct rpc_rqst), gfp_flags);
1038         if (req != NULL)
1039                 goto out;
1040         atomic_dec(&xprt->num_reqs);
1041         req = ERR_PTR(-ENOMEM);
1042 out:
1043         return req;
1044 }
1045
1046 static bool xprt_dynamic_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req)
1047 {
1048         if (atomic_add_unless(&xprt->num_reqs, -1, xprt->min_reqs)) {
1049                 kfree(req);
1050                 return true;
1051         }
1052         return false;
1053 }
1054
1055 void xprt_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task)
1056 {
1057         struct rpc_rqst *req;
1058
1059         spin_lock(&xprt->reserve_lock);
1060         if (!list_empty(&xprt->free)) {
1061                 req = list_entry(xprt->free.next, struct rpc_rqst, rq_list);
1062                 list_del(&req->rq_list);
1063                 goto out_init_req;
1064         }
1065         req = xprt_dynamic_alloc_slot(xprt, GFP_NOWAIT|__GFP_NOWARN);
1066         if (!IS_ERR(req))
1067                 goto out_init_req;
1068         switch (PTR_ERR(req)) {
1069         case -ENOMEM:
1070                 dprintk("RPC:       dynamic allocation of request slot "
1071                                 "failed! Retrying\n");
1072                 task->tk_status = -ENOMEM;
1073                 break;
1074         case -EAGAIN:
1075                 xprt_add_backlog(xprt, task);
1076                 dprintk("RPC:       waiting for request slot\n");
1077         default:
1078                 task->tk_status = -EAGAIN;
1079         }
1080         spin_unlock(&xprt->reserve_lock);
1081         return;
1082 out_init_req:
1083         task->tk_status = 0;
1084         task->tk_rqstp = req;
1085         xprt_request_init(task, xprt);
1086         spin_unlock(&xprt->reserve_lock);
1087 }
1088 EXPORT_SYMBOL_GPL(xprt_alloc_slot);
1089
1090 void xprt_lock_and_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task)
1091 {
1092         /* Note: grabbing the xprt_lock_write() ensures that we throttle
1093          * new slot allocation if the transport is congested (i.e. when
1094          * reconnecting a stream transport or when out of socket write
1095          * buffer space).
1096          */
1097         if (xprt_lock_write(xprt, task)) {
1098                 xprt_alloc_slot(xprt, task);
1099                 xprt_release_write(xprt, task);
1100         }
1101 }
1102 EXPORT_SYMBOL_GPL(xprt_lock_and_alloc_slot);
1103
1104 static void xprt_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req)
1105 {
1106         spin_lock(&xprt->reserve_lock);
1107         if (!xprt_dynamic_free_slot(xprt, req)) {
1108                 memset(req, 0, sizeof(*req));   /* mark unused */
1109                 list_add(&req->rq_list, &xprt->free);
1110         }
1111         xprt_wake_up_backlog(xprt);
1112         spin_unlock(&xprt->reserve_lock);
1113 }
1114
1115 static void xprt_free_all_slots(struct rpc_xprt *xprt)
1116 {
1117         struct rpc_rqst *req;
1118         while (!list_empty(&xprt->free)) {
1119                 req = list_first_entry(&xprt->free, struct rpc_rqst, rq_list);
1120                 list_del(&req->rq_list);
1121                 kfree(req);
1122         }
1123 }
1124
1125 struct rpc_xprt *xprt_alloc(struct net *net, size_t size,
1126                 unsigned int num_prealloc,
1127                 unsigned int max_alloc)
1128 {
1129         struct rpc_xprt *xprt;
1130         struct rpc_rqst *req;
1131         int i;
1132
1133         xprt = kzalloc(size, GFP_KERNEL);
1134         if (xprt == NULL)
1135                 goto out;
1136
1137         xprt_init(xprt, net);
1138
1139         for (i = 0; i < num_prealloc; i++) {
1140                 req = kzalloc(sizeof(struct rpc_rqst), GFP_KERNEL);
1141                 if (!req)
1142                         goto out_free;
1143                 list_add(&req->rq_list, &xprt->free);
1144         }
1145         if (max_alloc > num_prealloc)
1146                 xprt->max_reqs = max_alloc;
1147         else
1148                 xprt->max_reqs = num_prealloc;
1149         xprt->min_reqs = num_prealloc;
1150         atomic_set(&xprt->num_reqs, num_prealloc);
1151
1152         return xprt;
1153
1154 out_free:
1155         xprt_free(xprt);
1156 out:
1157         return NULL;
1158 }
1159 EXPORT_SYMBOL_GPL(xprt_alloc);
1160
1161 void xprt_free(struct rpc_xprt *xprt)
1162 {
1163         put_net(xprt->xprt_net);
1164         xprt_free_all_slots(xprt);
1165         kfree(xprt);
1166 }
1167 EXPORT_SYMBOL_GPL(xprt_free);
1168
1169 /**
1170  * xprt_reserve - allocate an RPC request slot
1171  * @task: RPC task requesting a slot allocation
1172  *
1173  * If the transport is marked as being congested, or if no more
1174  * slots are available, place the task on the transport's
1175  * backlog queue.
1176  */
1177 void xprt_reserve(struct rpc_task *task)
1178 {
1179         struct rpc_xprt *xprt;
1180
1181         task->tk_status = 0;
1182         if (task->tk_rqstp != NULL)
1183                 return;
1184
1185         task->tk_timeout = 0;
1186         task->tk_status = -EAGAIN;
1187         rcu_read_lock();
1188         xprt = rcu_dereference(task->tk_client->cl_xprt);
1189         if (!xprt_throttle_congested(xprt, task))
1190                 xprt->ops->alloc_slot(xprt, task);
1191         rcu_read_unlock();
1192 }
1193
1194 /**
1195  * xprt_retry_reserve - allocate an RPC request slot
1196  * @task: RPC task requesting a slot allocation
1197  *
1198  * If no more slots are available, place the task on the transport's
1199  * backlog queue.
1200  * Note that the only difference with xprt_reserve is that we now
1201  * ignore the value of the XPRT_CONGESTED flag.
1202  */
1203 void xprt_retry_reserve(struct rpc_task *task)
1204 {
1205         struct rpc_xprt *xprt;
1206
1207         task->tk_status = 0;
1208         if (task->tk_rqstp != NULL)
1209                 return;
1210
1211         task->tk_timeout = 0;
1212         task->tk_status = -EAGAIN;
1213         rcu_read_lock();
1214         xprt = rcu_dereference(task->tk_client->cl_xprt);
1215         xprt->ops->alloc_slot(xprt, task);
1216         rcu_read_unlock();
1217 }
1218
1219 static inline __be32 xprt_alloc_xid(struct rpc_xprt *xprt)
1220 {
1221         return (__force __be32)xprt->xid++;
1222 }
1223
1224 static inline void xprt_init_xid(struct rpc_xprt *xprt)
1225 {
1226         xprt->xid = prandom_u32();
1227 }
1228
1229 static void xprt_request_init(struct rpc_task *task, struct rpc_xprt *xprt)
1230 {
1231         struct rpc_rqst *req = task->tk_rqstp;
1232
1233         INIT_LIST_HEAD(&req->rq_list);
1234         req->rq_timeout = task->tk_client->cl_timeout->to_initval;
1235         req->rq_task    = task;
1236         req->rq_xprt    = xprt;
1237         req->rq_buffer  = NULL;
1238         req->rq_xid     = xprt_alloc_xid(xprt);
1239         req->rq_connect_cookie = xprt->connect_cookie - 1;
1240         req->rq_bytes_sent = 0;
1241         req->rq_snd_buf.len = 0;
1242         req->rq_snd_buf.buflen = 0;
1243         req->rq_rcv_buf.len = 0;
1244         req->rq_rcv_buf.buflen = 0;
1245         req->rq_release_snd_buf = NULL;
1246         xprt_reset_majortimeo(req);
1247         dprintk("RPC: %5u reserved req %p xid %08x\n", task->tk_pid,
1248                         req, ntohl(req->rq_xid));
1249 }
1250
1251 /**
1252  * xprt_release - release an RPC request slot
1253  * @task: task which is finished with the slot
1254  *
1255  */
1256 void xprt_release(struct rpc_task *task)
1257 {
1258         struct rpc_xprt *xprt;
1259         struct rpc_rqst *req = task->tk_rqstp;
1260
1261         if (req == NULL) {
1262                 if (task->tk_client) {
1263                         rcu_read_lock();
1264                         xprt = rcu_dereference(task->tk_client->cl_xprt);
1265                         if (xprt->snd_task == task)
1266                                 xprt_release_write(xprt, task);
1267                         rcu_read_unlock();
1268                 }
1269                 return;
1270         }
1271
1272         xprt = req->rq_xprt;
1273         if (task->tk_ops->rpc_count_stats != NULL)
1274                 task->tk_ops->rpc_count_stats(task, task->tk_calldata);
1275         else if (task->tk_client)
1276                 rpc_count_iostats(task, task->tk_client->cl_metrics);
1277         spin_lock_bh(&xprt->transport_lock);
1278         xprt->ops->release_xprt(xprt, task);
1279         if (xprt->ops->release_request)
1280                 xprt->ops->release_request(task);
1281         if (!list_empty(&req->rq_list))
1282                 list_del(&req->rq_list);
1283         xprt->last_used = jiffies;
1284         if (list_empty(&xprt->recv) && xprt_has_timer(xprt))
1285                 mod_timer(&xprt->timer,
1286                                 xprt->last_used + xprt->idle_timeout);
1287         spin_unlock_bh(&xprt->transport_lock);
1288         if (req->rq_buffer)
1289                 xprt->ops->buf_free(req->rq_buffer);
1290         if (req->rq_cred != NULL)
1291                 put_rpccred(req->rq_cred);
1292         task->tk_rqstp = NULL;
1293         if (req->rq_release_snd_buf)
1294                 req->rq_release_snd_buf(req);
1295
1296         dprintk("RPC: %5u release request %p\n", task->tk_pid, req);
1297         if (likely(!bc_prealloc(req)))
1298                 xprt_free_slot(xprt, req);
1299         else
1300                 xprt_free_bc_request(req);
1301 }
1302
1303 static void xprt_init(struct rpc_xprt *xprt, struct net *net)
1304 {
1305         atomic_set(&xprt->count, 1);
1306
1307         spin_lock_init(&xprt->transport_lock);
1308         spin_lock_init(&xprt->reserve_lock);
1309
1310         INIT_LIST_HEAD(&xprt->free);
1311         INIT_LIST_HEAD(&xprt->recv);
1312 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
1313         spin_lock_init(&xprt->bc_pa_lock);
1314         INIT_LIST_HEAD(&xprt->bc_pa_list);
1315 #endif /* CONFIG_SUNRPC_BACKCHANNEL */
1316
1317         xprt->last_used = jiffies;
1318         xprt->cwnd = RPC_INITCWND;
1319         xprt->bind_index = 0;
1320
1321         rpc_init_wait_queue(&xprt->binding, "xprt_binding");
1322         rpc_init_wait_queue(&xprt->pending, "xprt_pending");
1323         rpc_init_priority_wait_queue(&xprt->sending, "xprt_sending");
1324         rpc_init_priority_wait_queue(&xprt->backlog, "xprt_backlog");
1325
1326         xprt_init_xid(xprt);
1327
1328         xprt->xprt_net = get_net(net);
1329 }
1330
1331 /**
1332  * xprt_create_transport - create an RPC transport
1333  * @args: rpc transport creation arguments
1334  *
1335  */
1336 struct rpc_xprt *xprt_create_transport(struct xprt_create *args)
1337 {
1338         struct rpc_xprt *xprt;
1339         struct xprt_class *t;
1340
1341         spin_lock(&xprt_list_lock);
1342         list_for_each_entry(t, &xprt_list, list) {
1343                 if (t->ident == args->ident) {
1344                         spin_unlock(&xprt_list_lock);
1345                         goto found;
1346                 }
1347         }
1348         spin_unlock(&xprt_list_lock);
1349         dprintk("RPC: transport (%d) not supported\n", args->ident);
1350         return ERR_PTR(-EIO);
1351
1352 found:
1353         xprt = t->setup(args);
1354         if (IS_ERR(xprt)) {
1355                 dprintk("RPC:       xprt_create_transport: failed, %ld\n",
1356                                 -PTR_ERR(xprt));
1357                 goto out;
1358         }
1359         if (args->flags & XPRT_CREATE_NO_IDLE_TIMEOUT)
1360                 xprt->idle_timeout = 0;
1361         INIT_WORK(&xprt->task_cleanup, xprt_autoclose);
1362         if (xprt_has_timer(xprt))
1363                 setup_timer(&xprt->timer, xprt_init_autodisconnect,
1364                             (unsigned long)xprt);
1365         else
1366                 init_timer(&xprt->timer);
1367
1368         if (strlen(args->servername) > RPC_MAXNETNAMELEN) {
1369                 xprt_destroy(xprt);
1370                 return ERR_PTR(-EINVAL);
1371         }
1372         xprt->servername = kstrdup(args->servername, GFP_KERNEL);
1373         if (xprt->servername == NULL) {
1374                 xprt_destroy(xprt);
1375                 return ERR_PTR(-ENOMEM);
1376         }
1377
1378         rpc_xprt_debugfs_register(xprt);
1379
1380         dprintk("RPC:       created transport %p with %u slots\n", xprt,
1381                         xprt->max_reqs);
1382 out:
1383         return xprt;
1384 }
1385
1386 /**
1387  * xprt_destroy - destroy an RPC transport, killing off all requests.
1388  * @xprt: transport to destroy
1389  *
1390  */
1391 static void xprt_destroy(struct rpc_xprt *xprt)
1392 {
1393         dprintk("RPC:       destroying transport %p\n", xprt);
1394
1395         /* Exclude transport connect/disconnect handlers */
1396         wait_on_bit_lock(&xprt->state, XPRT_LOCKED, TASK_UNINTERRUPTIBLE);
1397
1398         del_timer_sync(&xprt->timer);
1399
1400         rpc_xprt_debugfs_unregister(xprt);
1401         rpc_destroy_wait_queue(&xprt->binding);
1402         rpc_destroy_wait_queue(&xprt->pending);
1403         rpc_destroy_wait_queue(&xprt->sending);
1404         rpc_destroy_wait_queue(&xprt->backlog);
1405         cancel_work_sync(&xprt->task_cleanup);
1406         kfree(xprt->servername);
1407         /*
1408          * Tear down transport state and free the rpc_xprt
1409          */
1410         xprt->ops->destroy(xprt);
1411 }
1412
1413 /**
1414  * xprt_put - release a reference to an RPC transport.
1415  * @xprt: pointer to the transport
1416  *
1417  */
1418 void xprt_put(struct rpc_xprt *xprt)
1419 {
1420         if (atomic_dec_and_test(&xprt->count))
1421                 xprt_destroy(xprt);
1422 }