blob: 5e7c13aa66d0d59b9eb8e8cee23902dcf325b2b3 [file] [log] [blame]
xjb04a4022021-11-25 15:01:52 +08001/*
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#include <linux/rcupdate.h>
52
53#include <trace/events/sunrpc.h>
54
55#include "sunrpc.h"
56
57/*
58 * Local variables
59 */
60
61#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
62# define RPCDBG_FACILITY RPCDBG_XPRT
63#endif
64
65/*
66 * Local functions
67 */
68static void xprt_init(struct rpc_xprt *xprt, struct net *net);
69static __be32 xprt_alloc_xid(struct rpc_xprt *xprt);
70static void xprt_connect_status(struct rpc_task *task);
71static int __xprt_get_cong(struct rpc_xprt *, struct rpc_task *);
72static void __xprt_put_cong(struct rpc_xprt *, struct rpc_rqst *);
73static void xprt_destroy(struct rpc_xprt *xprt);
74
75static DEFINE_SPINLOCK(xprt_list_lock);
76static LIST_HEAD(xprt_list);
77
78/**
79 * xprt_register_transport - register a transport implementation
80 * @transport: transport to register
81 *
82 * If a transport implementation is loaded as a kernel module, it can
83 * call this interface to make itself known to the RPC client.
84 *
85 * Returns:
86 * 0: transport successfully registered
87 * -EEXIST: transport already registered
88 * -EINVAL: transport module being unloaded
89 */
90int xprt_register_transport(struct xprt_class *transport)
91{
92 struct xprt_class *t;
93 int result;
94
95 result = -EEXIST;
96 spin_lock(&xprt_list_lock);
97 list_for_each_entry(t, &xprt_list, list) {
98 /* don't register the same transport class twice */
99 if (t->ident == transport->ident)
100 goto out;
101 }
102
103 list_add_tail(&transport->list, &xprt_list);
104 printk(KERN_INFO "RPC: Registered %s transport module.\n",
105 transport->name);
106 result = 0;
107
108out:
109 spin_unlock(&xprt_list_lock);
110 return result;
111}
112EXPORT_SYMBOL_GPL(xprt_register_transport);
113
114/**
115 * xprt_unregister_transport - unregister a transport implementation
116 * @transport: transport to unregister
117 *
118 * Returns:
119 * 0: transport successfully unregistered
120 * -ENOENT: transport never registered
121 */
122int xprt_unregister_transport(struct xprt_class *transport)
123{
124 struct xprt_class *t;
125 int result;
126
127 result = 0;
128 spin_lock(&xprt_list_lock);
129 list_for_each_entry(t, &xprt_list, list) {
130 if (t == transport) {
131 printk(KERN_INFO
132 "RPC: Unregistered %s transport module.\n",
133 transport->name);
134 list_del_init(&transport->list);
135 goto out;
136 }
137 }
138 result = -ENOENT;
139
140out:
141 spin_unlock(&xprt_list_lock);
142 return result;
143}
144EXPORT_SYMBOL_GPL(xprt_unregister_transport);
145
146/**
147 * xprt_load_transport - load a transport implementation
148 * @transport_name: transport to load
149 *
150 * Returns:
151 * 0: transport successfully loaded
152 * -ENOENT: transport module not available
153 */
154int xprt_load_transport(const char *transport_name)
155{
156 struct xprt_class *t;
157 int result;
158
159 result = 0;
160 spin_lock(&xprt_list_lock);
161 list_for_each_entry(t, &xprt_list, list) {
162 if (strcmp(t->name, transport_name) == 0) {
163 spin_unlock(&xprt_list_lock);
164 goto out;
165 }
166 }
167 spin_unlock(&xprt_list_lock);
168 result = request_module("xprt%s", transport_name);
169out:
170 return result;
171}
172EXPORT_SYMBOL_GPL(xprt_load_transport);
173
174/**
175 * xprt_reserve_xprt - serialize write access to transports
176 * @task: task that is requesting access to the transport
177 * @xprt: pointer to the target transport
178 *
179 * This prevents mixing the payload of separate requests, and prevents
180 * transport connects from colliding with writes. No congestion control
181 * is provided.
182 */
183int xprt_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
184{
185 struct rpc_rqst *req = task->tk_rqstp;
186 int priority;
187
188 if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) {
189 if (task == xprt->snd_task)
190 return 1;
191 goto out_sleep;
192 }
193 xprt->snd_task = task;
194 if (req != NULL)
195 req->rq_ntrans++;
196
197 return 1;
198
199out_sleep:
200 dprintk("RPC: %5u failed to lock transport %p\n",
201 task->tk_pid, xprt);
202 task->tk_timeout = 0;
203 task->tk_status = -EAGAIN;
204 if (req == NULL)
205 priority = RPC_PRIORITY_LOW;
206 else if (!req->rq_ntrans)
207 priority = RPC_PRIORITY_NORMAL;
208 else
209 priority = RPC_PRIORITY_HIGH;
210 rpc_sleep_on_priority(&xprt->sending, task, NULL, priority);
211 return 0;
212}
213EXPORT_SYMBOL_GPL(xprt_reserve_xprt);
214
215static void xprt_clear_locked(struct rpc_xprt *xprt)
216{
217 xprt->snd_task = NULL;
218 if (!test_bit(XPRT_CLOSE_WAIT, &xprt->state)) {
219 smp_mb__before_atomic();
220 clear_bit(XPRT_LOCKED, &xprt->state);
221 smp_mb__after_atomic();
222 } else
223 queue_work(xprtiod_workqueue, &xprt->task_cleanup);
224}
225
226/*
227 * xprt_reserve_xprt_cong - serialize write access to transports
228 * @task: task that is requesting access to the transport
229 *
230 * Same as xprt_reserve_xprt, but Van Jacobson congestion control is
231 * integrated into the decision of whether a request is allowed to be
232 * woken up and given access to the transport.
233 */
234int xprt_reserve_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
235{
236 struct rpc_rqst *req = task->tk_rqstp;
237 int priority;
238
239 if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) {
240 if (task == xprt->snd_task)
241 return 1;
242 goto out_sleep;
243 }
244 if (req == NULL) {
245 xprt->snd_task = task;
246 return 1;
247 }
248 if (__xprt_get_cong(xprt, task)) {
249 xprt->snd_task = task;
250 req->rq_ntrans++;
251 return 1;
252 }
253 xprt_clear_locked(xprt);
254out_sleep:
255 if (req)
256 __xprt_put_cong(xprt, req);
257 dprintk("RPC: %5u failed to lock transport %p\n", task->tk_pid, xprt);
258 task->tk_timeout = 0;
259 task->tk_status = -EAGAIN;
260 if (req == NULL)
261 priority = RPC_PRIORITY_LOW;
262 else if (!req->rq_ntrans)
263 priority = RPC_PRIORITY_NORMAL;
264 else
265 priority = RPC_PRIORITY_HIGH;
266 rpc_sleep_on_priority(&xprt->sending, task, NULL, priority);
267 return 0;
268}
269EXPORT_SYMBOL_GPL(xprt_reserve_xprt_cong);
270
271static inline int xprt_lock_write(struct rpc_xprt *xprt, struct rpc_task *task)
272{
273 int retval;
274
275 spin_lock_bh(&xprt->transport_lock);
276 retval = xprt->ops->reserve_xprt(xprt, task);
277 spin_unlock_bh(&xprt->transport_lock);
278 return retval;
279}
280
281static bool __xprt_lock_write_func(struct rpc_task *task, void *data)
282{
283 struct rpc_xprt *xprt = data;
284 struct rpc_rqst *req;
285
286 req = task->tk_rqstp;
287 xprt->snd_task = task;
288 if (req)
289 req->rq_ntrans++;
290 return true;
291}
292
293static void __xprt_lock_write_next(struct rpc_xprt *xprt)
294{
295 if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
296 return;
297
298 if (rpc_wake_up_first_on_wq(xprtiod_workqueue, &xprt->sending,
299 __xprt_lock_write_func, xprt))
300 return;
301 xprt_clear_locked(xprt);
302}
303
304static bool __xprt_lock_write_cong_func(struct rpc_task *task, void *data)
305{
306 struct rpc_xprt *xprt = data;
307 struct rpc_rqst *req;
308
309 req = task->tk_rqstp;
310 if (req == NULL) {
311 xprt->snd_task = task;
312 return true;
313 }
314 if (__xprt_get_cong(xprt, task)) {
315 xprt->snd_task = task;
316 req->rq_ntrans++;
317 return true;
318 }
319 return false;
320}
321
322static void __xprt_lock_write_next_cong(struct rpc_xprt *xprt)
323{
324 if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
325 return;
326 if (RPCXPRT_CONGESTED(xprt))
327 goto out_unlock;
328 if (rpc_wake_up_first_on_wq(xprtiod_workqueue, &xprt->sending,
329 __xprt_lock_write_cong_func, xprt))
330 return;
331out_unlock:
332 xprt_clear_locked(xprt);
333}
334
335static void xprt_task_clear_bytes_sent(struct rpc_task *task)
336{
337 if (task != NULL) {
338 struct rpc_rqst *req = task->tk_rqstp;
339 if (req != NULL)
340 req->rq_bytes_sent = 0;
341 }
342}
343
344/**
345 * xprt_release_xprt - allow other requests to use a transport
346 * @xprt: transport with other tasks potentially waiting
347 * @task: task that is releasing access to the transport
348 *
349 * Note that "task" can be NULL. No congestion control is provided.
350 */
351void xprt_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task)
352{
353 if (xprt->snd_task == task) {
354 xprt_task_clear_bytes_sent(task);
355 xprt_clear_locked(xprt);
356 __xprt_lock_write_next(xprt);
357 }
358}
359EXPORT_SYMBOL_GPL(xprt_release_xprt);
360
361/**
362 * xprt_release_xprt_cong - allow other requests to use a transport
363 * @xprt: transport with other tasks potentially waiting
364 * @task: task that is releasing access to the transport
365 *
366 * Note that "task" can be NULL. Another task is awoken to use the
367 * transport if the transport's congestion window allows it.
368 */
369void xprt_release_xprt_cong(struct rpc_xprt *xprt, struct rpc_task *task)
370{
371 if (xprt->snd_task == task) {
372 xprt_task_clear_bytes_sent(task);
373 xprt_clear_locked(xprt);
374 __xprt_lock_write_next_cong(xprt);
375 }
376}
377EXPORT_SYMBOL_GPL(xprt_release_xprt_cong);
378
379static inline void xprt_release_write(struct rpc_xprt *xprt, struct rpc_task *task)
380{
381 spin_lock_bh(&xprt->transport_lock);
382 xprt->ops->release_xprt(xprt, task);
383 spin_unlock_bh(&xprt->transport_lock);
384}
385
386/*
387 * Van Jacobson congestion avoidance. Check if the congestion window
388 * overflowed. Put the task to sleep if this is the case.
389 */
390static int
391__xprt_get_cong(struct rpc_xprt *xprt, struct rpc_task *task)
392{
393 struct rpc_rqst *req = task->tk_rqstp;
394
395 if (req->rq_cong)
396 return 1;
397 dprintk("RPC: %5u xprt_cwnd_limited cong = %lu cwnd = %lu\n",
398 task->tk_pid, xprt->cong, xprt->cwnd);
399 if (RPCXPRT_CONGESTED(xprt))
400 return 0;
401 req->rq_cong = 1;
402 xprt->cong += RPC_CWNDSCALE;
403 return 1;
404}
405
406/*
407 * Adjust the congestion window, and wake up the next task
408 * that has been sleeping due to congestion
409 */
410static void
411__xprt_put_cong(struct rpc_xprt *xprt, struct rpc_rqst *req)
412{
413 if (!req->rq_cong)
414 return;
415 req->rq_cong = 0;
416 xprt->cong -= RPC_CWNDSCALE;
417 __xprt_lock_write_next_cong(xprt);
418}
419
420/**
421 * xprt_release_rqst_cong - housekeeping when request is complete
422 * @task: RPC request that recently completed
423 *
424 * Useful for transports that require congestion control.
425 */
426void xprt_release_rqst_cong(struct rpc_task *task)
427{
428 struct rpc_rqst *req = task->tk_rqstp;
429
430 __xprt_put_cong(req->rq_xprt, req);
431}
432EXPORT_SYMBOL_GPL(xprt_release_rqst_cong);
433
434/**
435 * xprt_adjust_cwnd - adjust transport congestion window
436 * @xprt: pointer to xprt
437 * @task: recently completed RPC request used to adjust window
438 * @result: result code of completed RPC request
439 *
440 * The transport code maintains an estimate on the maximum number of out-
441 * standing RPC requests, using a smoothed version of the congestion
442 * avoidance implemented in 44BSD. This is basically the Van Jacobson
443 * congestion algorithm: If a retransmit occurs, the congestion window is
444 * halved; otherwise, it is incremented by 1/cwnd when
445 *
446 * - a reply is received and
447 * - a full number of requests are outstanding and
448 * - the congestion window hasn't been updated recently.
449 */
450void xprt_adjust_cwnd(struct rpc_xprt *xprt, struct rpc_task *task, int result)
451{
452 struct rpc_rqst *req = task->tk_rqstp;
453 unsigned long cwnd = xprt->cwnd;
454
455 if (result >= 0 && cwnd <= xprt->cong) {
456 /* The (cwnd >> 1) term makes sure
457 * the result gets rounded properly. */
458 cwnd += (RPC_CWNDSCALE * RPC_CWNDSCALE + (cwnd >> 1)) / cwnd;
459 if (cwnd > RPC_MAXCWND(xprt))
460 cwnd = RPC_MAXCWND(xprt);
461 __xprt_lock_write_next_cong(xprt);
462 } else if (result == -ETIMEDOUT) {
463 cwnd >>= 1;
464 if (cwnd < RPC_CWNDSCALE)
465 cwnd = RPC_CWNDSCALE;
466 }
467 dprintk("RPC: cong %ld, cwnd was %ld, now %ld\n",
468 xprt->cong, xprt->cwnd, cwnd);
469 xprt->cwnd = cwnd;
470 __xprt_put_cong(xprt, req);
471}
472EXPORT_SYMBOL_GPL(xprt_adjust_cwnd);
473
474/**
475 * xprt_wake_pending_tasks - wake all tasks on a transport's pending queue
476 * @xprt: transport with waiting tasks
477 * @status: result code to plant in each task before waking it
478 *
479 */
480void xprt_wake_pending_tasks(struct rpc_xprt *xprt, int status)
481{
482 if (status < 0)
483 rpc_wake_up_status(&xprt->pending, status);
484 else
485 rpc_wake_up(&xprt->pending);
486}
487EXPORT_SYMBOL_GPL(xprt_wake_pending_tasks);
488
489/**
490 * xprt_wait_for_buffer_space - wait for transport output buffer to clear
491 * @task: task to be put to sleep
492 * @action: function pointer to be executed after wait
493 *
494 * Note that we only set the timer for the case of RPC_IS_SOFT(), since
495 * we don't in general want to force a socket disconnection due to
496 * an incomplete RPC call transmission.
497 */
498void xprt_wait_for_buffer_space(struct rpc_task *task, rpc_action action)
499{
500 struct rpc_rqst *req = task->tk_rqstp;
501 struct rpc_xprt *xprt = req->rq_xprt;
502
503 task->tk_timeout = RPC_IS_SOFT(task) ? req->rq_timeout : 0;
504 rpc_sleep_on(&xprt->pending, task, action);
505}
506EXPORT_SYMBOL_GPL(xprt_wait_for_buffer_space);
507
508/**
509 * xprt_write_space - wake the task waiting for transport output buffer space
510 * @xprt: transport with waiting tasks
511 *
512 * Can be called in a soft IRQ context, so xprt_write_space never sleeps.
513 */
514void xprt_write_space(struct rpc_xprt *xprt)
515{
516 spin_lock_bh(&xprt->transport_lock);
517 if (xprt->snd_task) {
518 dprintk("RPC: write space: waking waiting task on "
519 "xprt %p\n", xprt);
520 rpc_wake_up_queued_task_on_wq(xprtiod_workqueue,
521 &xprt->pending, xprt->snd_task);
522 }
523 spin_unlock_bh(&xprt->transport_lock);
524}
525EXPORT_SYMBOL_GPL(xprt_write_space);
526
527/**
528 * xprt_set_retrans_timeout_def - set a request's retransmit timeout
529 * @task: task whose timeout is to be set
530 *
531 * Set a request's retransmit timeout based on the transport's
532 * default timeout parameters. Used by transports that don't adjust
533 * the retransmit timeout based on round-trip time estimation.
534 */
535void xprt_set_retrans_timeout_def(struct rpc_task *task)
536{
537 task->tk_timeout = task->tk_rqstp->rq_timeout;
538}
539EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_def);
540
541/**
542 * xprt_set_retrans_timeout_rtt - set a request's retransmit timeout
543 * @task: task whose timeout is to be set
544 *
545 * Set a request's retransmit timeout using the RTT estimator.
546 */
547void xprt_set_retrans_timeout_rtt(struct rpc_task *task)
548{
549 int timer = task->tk_msg.rpc_proc->p_timer;
550 struct rpc_clnt *clnt = task->tk_client;
551 struct rpc_rtt *rtt = clnt->cl_rtt;
552 struct rpc_rqst *req = task->tk_rqstp;
553 unsigned long max_timeout = clnt->cl_timeout->to_maxval;
554
555 task->tk_timeout = rpc_calc_rto(rtt, timer);
556 task->tk_timeout <<= rpc_ntimeo(rtt, timer) + req->rq_retries;
557 if (task->tk_timeout > max_timeout || task->tk_timeout == 0)
558 task->tk_timeout = max_timeout;
559}
560EXPORT_SYMBOL_GPL(xprt_set_retrans_timeout_rtt);
561
562static void xprt_reset_majortimeo(struct rpc_rqst *req)
563{
564 const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
565
566 req->rq_majortimeo = req->rq_timeout;
567 if (to->to_exponential)
568 req->rq_majortimeo <<= to->to_retries;
569 else
570 req->rq_majortimeo += to->to_increment * to->to_retries;
571 if (req->rq_majortimeo > to->to_maxval || req->rq_majortimeo == 0)
572 req->rq_majortimeo = to->to_maxval;
573 req->rq_majortimeo += jiffies;
574}
575
576/**
577 * xprt_adjust_timeout - adjust timeout values for next retransmit
578 * @req: RPC request containing parameters to use for the adjustment
579 *
580 */
581int xprt_adjust_timeout(struct rpc_rqst *req)
582{
583 struct rpc_xprt *xprt = req->rq_xprt;
584 const struct rpc_timeout *to = req->rq_task->tk_client->cl_timeout;
585 int status = 0;
586
587 if (time_before(jiffies, req->rq_majortimeo)) {
588 if (to->to_exponential)
589 req->rq_timeout <<= 1;
590 else
591 req->rq_timeout += to->to_increment;
592 if (to->to_maxval && req->rq_timeout >= to->to_maxval)
593 req->rq_timeout = to->to_maxval;
594 req->rq_retries++;
595 } else {
596 req->rq_timeout = to->to_initval;
597 req->rq_retries = 0;
598 xprt_reset_majortimeo(req);
599 /* Reset the RTT counters == "slow start" */
600 spin_lock_bh(&xprt->transport_lock);
601 rpc_init_rtt(req->rq_task->tk_client->cl_rtt, to->to_initval);
602 spin_unlock_bh(&xprt->transport_lock);
603 status = -ETIMEDOUT;
604 }
605
606 if (req->rq_timeout == 0) {
607 printk(KERN_WARNING "xprt_adjust_timeout: rq_timeout = 0!\n");
608 req->rq_timeout = 5 * HZ;
609 }
610 return status;
611}
612
613static void xprt_autoclose(struct work_struct *work)
614{
615 struct rpc_xprt *xprt =
616 container_of(work, struct rpc_xprt, task_cleanup);
617
618 clear_bit(XPRT_CLOSE_WAIT, &xprt->state);
619 xprt->ops->close(xprt);
620 xprt_release_write(xprt, NULL);
621 wake_up_bit(&xprt->state, XPRT_LOCKED);
622}
623
624/**
625 * xprt_disconnect_done - mark a transport as disconnected
626 * @xprt: transport to flag for disconnect
627 *
628 */
629void xprt_disconnect_done(struct rpc_xprt *xprt)
630{
631 dprintk("RPC: disconnected transport %p\n", xprt);
632 spin_lock_bh(&xprt->transport_lock);
633 xprt_clear_connected(xprt);
634 xprt_wake_pending_tasks(xprt, -EAGAIN);
635 spin_unlock_bh(&xprt->transport_lock);
636}
637EXPORT_SYMBOL_GPL(xprt_disconnect_done);
638
639/**
640 * xprt_force_disconnect - force a transport to disconnect
641 * @xprt: transport to disconnect
642 *
643 */
644void xprt_force_disconnect(struct rpc_xprt *xprt)
645{
646 /* Don't race with the test_bit() in xprt_clear_locked() */
647 spin_lock_bh(&xprt->transport_lock);
648 set_bit(XPRT_CLOSE_WAIT, &xprt->state);
649 /* Try to schedule an autoclose RPC call */
650 if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
651 queue_work(xprtiod_workqueue, &xprt->task_cleanup);
652 xprt_wake_pending_tasks(xprt, -EAGAIN);
653 spin_unlock_bh(&xprt->transport_lock);
654}
655EXPORT_SYMBOL_GPL(xprt_force_disconnect);
656
657/**
658 * xprt_conditional_disconnect - force a transport to disconnect
659 * @xprt: transport to disconnect
660 * @cookie: 'connection cookie'
661 *
662 * This attempts to break the connection if and only if 'cookie' matches
663 * the current transport 'connection cookie'. It ensures that we don't
664 * try to break the connection more than once when we need to retransmit
665 * a batch of RPC requests.
666 *
667 */
668void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie)
669{
670 /* Don't race with the test_bit() in xprt_clear_locked() */
671 spin_lock_bh(&xprt->transport_lock);
672 if (cookie != xprt->connect_cookie)
673 goto out;
674 if (test_bit(XPRT_CLOSING, &xprt->state))
675 goto out;
676 set_bit(XPRT_CLOSE_WAIT, &xprt->state);
677 /* Try to schedule an autoclose RPC call */
678 if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
679 queue_work(xprtiod_workqueue, &xprt->task_cleanup);
680 xprt_wake_pending_tasks(xprt, -EAGAIN);
681out:
682 spin_unlock_bh(&xprt->transport_lock);
683}
684
685static bool
686xprt_has_timer(const struct rpc_xprt *xprt)
687{
688 return xprt->idle_timeout != 0;
689}
690
691static void
692xprt_schedule_autodisconnect(struct rpc_xprt *xprt)
693 __must_hold(&xprt->transport_lock)
694{
695 if (list_empty(&xprt->recv) && xprt_has_timer(xprt))
696 mod_timer(&xprt->timer, xprt->last_used + xprt->idle_timeout);
697}
698
699static void
700xprt_init_autodisconnect(struct timer_list *t)
701{
702 struct rpc_xprt *xprt = from_timer(xprt, t, timer);
703
704 spin_lock(&xprt->transport_lock);
705 if (!list_empty(&xprt->recv))
706 goto out_abort;
707 /* Reset xprt->last_used to avoid connect/autodisconnect cycling */
708 xprt->last_used = jiffies;
709 if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
710 goto out_abort;
711 spin_unlock(&xprt->transport_lock);
712 queue_work(xprtiod_workqueue, &xprt->task_cleanup);
713 return;
714out_abort:
715 spin_unlock(&xprt->transport_lock);
716}
717
718bool xprt_lock_connect(struct rpc_xprt *xprt,
719 struct rpc_task *task,
720 void *cookie)
721{
722 bool ret = false;
723
724 spin_lock_bh(&xprt->transport_lock);
725 if (!test_bit(XPRT_LOCKED, &xprt->state))
726 goto out;
727 if (xprt->snd_task != task)
728 goto out;
729 xprt_task_clear_bytes_sent(task);
730 xprt->snd_task = cookie;
731 ret = true;
732out:
733 spin_unlock_bh(&xprt->transport_lock);
734 return ret;
735}
736
737void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie)
738{
739 spin_lock_bh(&xprt->transport_lock);
740 if (xprt->snd_task != cookie)
741 goto out;
742 if (!test_bit(XPRT_LOCKED, &xprt->state))
743 goto out;
744 xprt->snd_task =NULL;
745 xprt->ops->release_xprt(xprt, NULL);
746 xprt_schedule_autodisconnect(xprt);
747out:
748 spin_unlock_bh(&xprt->transport_lock);
749 wake_up_bit(&xprt->state, XPRT_LOCKED);
750}
751
752/**
753 * xprt_connect - schedule a transport connect operation
754 * @task: RPC task that is requesting the connect
755 *
756 */
757void xprt_connect(struct rpc_task *task)
758{
759 struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt;
760
761 dprintk("RPC: %5u xprt_connect xprt %p %s connected\n", task->tk_pid,
762 xprt, (xprt_connected(xprt) ? "is" : "is not"));
763
764 if (!xprt_bound(xprt)) {
765 task->tk_status = -EAGAIN;
766 return;
767 }
768 if (!xprt_lock_write(xprt, task))
769 return;
770
771 if (test_and_clear_bit(XPRT_CLOSE_WAIT, &xprt->state))
772 xprt->ops->close(xprt);
773
774 if (!xprt_connected(xprt)) {
775 task->tk_rqstp->rq_bytes_sent = 0;
776 task->tk_timeout = task->tk_rqstp->rq_timeout;
777 task->tk_rqstp->rq_connect_cookie = xprt->connect_cookie;
778 rpc_sleep_on(&xprt->pending, task, xprt_connect_status);
779
780 if (test_bit(XPRT_CLOSING, &xprt->state))
781 return;
782 if (xprt_test_and_set_connecting(xprt))
783 return;
784 /* Race breaker */
785 if (!xprt_connected(xprt)) {
786 xprt->stat.connect_start = jiffies;
787 xprt->ops->connect(xprt, task);
788 } else {
789 xprt_clear_connecting(xprt);
790 task->tk_status = 0;
791 rpc_wake_up_queued_task(&xprt->pending, task);
792 }
793 }
794 xprt_release_write(xprt, task);
795}
796
797static void xprt_connect_status(struct rpc_task *task)
798{
799 switch (task->tk_status) {
800 case 0:
801 dprintk("RPC: %5u xprt_connect_status: connection established\n",
802 task->tk_pid);
803 break;
804 case -ECONNREFUSED:
805 case -ECONNRESET:
806 case -ECONNABORTED:
807 case -ENETUNREACH:
808 case -EHOSTUNREACH:
809 case -EPIPE:
810 case -EAGAIN:
811 dprintk("RPC: %5u xprt_connect_status: retrying\n", task->tk_pid);
812 break;
813 case -ETIMEDOUT:
814 dprintk("RPC: %5u xprt_connect_status: connect attempt timed "
815 "out\n", task->tk_pid);
816 break;
817 default:
818 dprintk("RPC: %5u xprt_connect_status: error %d connecting to "
819 "server %s\n", task->tk_pid, -task->tk_status,
820 task->tk_rqstp->rq_xprt->servername);
821 task->tk_status = -EIO;
822 }
823}
824
825/**
826 * xprt_lookup_rqst - find an RPC request corresponding to an XID
827 * @xprt: transport on which the original request was transmitted
828 * @xid: RPC XID of incoming reply
829 *
830 * Caller holds xprt->recv_lock.
831 */
832struct rpc_rqst *xprt_lookup_rqst(struct rpc_xprt *xprt, __be32 xid)
833{
834 struct rpc_rqst *entry;
835
836 list_for_each_entry(entry, &xprt->recv, rq_list)
837 if (entry->rq_xid == xid) {
838 trace_xprt_lookup_rqst(xprt, xid, 0);
839 entry->rq_rtt = ktime_sub(ktime_get(), entry->rq_xtime);
840 return entry;
841 }
842
843 dprintk("RPC: xprt_lookup_rqst did not find xid %08x\n",
844 ntohl(xid));
845 trace_xprt_lookup_rqst(xprt, xid, -ENOENT);
846 xprt->stat.bad_xids++;
847 return NULL;
848}
849EXPORT_SYMBOL_GPL(xprt_lookup_rqst);
850
851/**
852 * xprt_pin_rqst - Pin a request on the transport receive list
853 * @req: Request to pin
854 *
855 * Caller must ensure this is atomic with the call to xprt_lookup_rqst()
856 * so should be holding the xprt transport lock.
857 */
858void xprt_pin_rqst(struct rpc_rqst *req)
859{
860 set_bit(RPC_TASK_MSG_RECV, &req->rq_task->tk_runstate);
861}
862EXPORT_SYMBOL_GPL(xprt_pin_rqst);
863
864/**
865 * xprt_unpin_rqst - Unpin a request on the transport receive list
866 * @req: Request to pin
867 *
868 * Caller should be holding the xprt transport lock.
869 */
870void xprt_unpin_rqst(struct rpc_rqst *req)
871{
872 struct rpc_task *task = req->rq_task;
873
874 clear_bit(RPC_TASK_MSG_RECV, &task->tk_runstate);
875 if (test_bit(RPC_TASK_MSG_RECV_WAIT, &task->tk_runstate))
876 wake_up_bit(&task->tk_runstate, RPC_TASK_MSG_RECV);
877}
878EXPORT_SYMBOL_GPL(xprt_unpin_rqst);
879
880static void xprt_wait_on_pinned_rqst(struct rpc_rqst *req)
881__must_hold(&req->rq_xprt->recv_lock)
882{
883 struct rpc_task *task = req->rq_task;
884
885 if (task && test_bit(RPC_TASK_MSG_RECV, &task->tk_runstate)) {
886 spin_unlock(&req->rq_xprt->recv_lock);
887 set_bit(RPC_TASK_MSG_RECV_WAIT, &task->tk_runstate);
888 wait_on_bit(&task->tk_runstate, RPC_TASK_MSG_RECV,
889 TASK_UNINTERRUPTIBLE);
890 clear_bit(RPC_TASK_MSG_RECV_WAIT, &task->tk_runstate);
891 spin_lock(&req->rq_xprt->recv_lock);
892 }
893}
894
895/**
896 * xprt_update_rtt - Update RPC RTT statistics
897 * @task: RPC request that recently completed
898 *
899 * Caller holds xprt->recv_lock.
900 */
901void xprt_update_rtt(struct rpc_task *task)
902{
903 struct rpc_rqst *req = task->tk_rqstp;
904 struct rpc_rtt *rtt = task->tk_client->cl_rtt;
905 unsigned int timer = task->tk_msg.rpc_proc->p_timer;
906 long m = usecs_to_jiffies(ktime_to_us(req->rq_rtt));
907
908 if (timer) {
909 if (req->rq_ntrans == 1)
910 rpc_update_rtt(rtt, timer, m);
911 rpc_set_timeo(rtt, timer, req->rq_ntrans - 1);
912 }
913}
914EXPORT_SYMBOL_GPL(xprt_update_rtt);
915
916/**
917 * xprt_complete_rqst - called when reply processing is complete
918 * @task: RPC request that recently completed
919 * @copied: actual number of bytes received from the transport
920 *
921 * Caller holds xprt->recv_lock.
922 */
923void xprt_complete_rqst(struct rpc_task *task, int copied)
924{
925 struct rpc_rqst *req = task->tk_rqstp;
926 struct rpc_xprt *xprt = req->rq_xprt;
927
928 dprintk("RPC: %5u xid %08x complete (%d bytes received)\n",
929 task->tk_pid, ntohl(req->rq_xid), copied);
930 trace_xprt_complete_rqst(xprt, req->rq_xid, copied);
931
932 xprt->stat.recvs++;
933
934 list_del_init(&req->rq_list);
935 req->rq_private_buf.len = copied;
936 /* Ensure all writes are done before we update */
937 /* req->rq_reply_bytes_recvd */
938 smp_wmb();
939 req->rq_reply_bytes_recvd = copied;
940 rpc_wake_up_queued_task(&xprt->pending, task);
941}
942EXPORT_SYMBOL_GPL(xprt_complete_rqst);
943
944static void xprt_timer(struct rpc_task *task)
945{
946 struct rpc_rqst *req = task->tk_rqstp;
947 struct rpc_xprt *xprt = req->rq_xprt;
948
949 if (task->tk_status != -ETIMEDOUT)
950 return;
951
952 trace_xprt_timer(xprt, req->rq_xid, task->tk_status);
953 if (!req->rq_reply_bytes_recvd) {
954 if (xprt->ops->timer)
955 xprt->ops->timer(xprt, task);
956 } else
957 task->tk_status = 0;
958}
959
960/**
961 * xprt_prepare_transmit - reserve the transport before sending a request
962 * @task: RPC task about to send a request
963 *
964 */
965bool xprt_prepare_transmit(struct rpc_task *task)
966{
967 struct rpc_rqst *req = task->tk_rqstp;
968 struct rpc_xprt *xprt = req->rq_xprt;
969 bool ret = false;
970
971 dprintk("RPC: %5u xprt_prepare_transmit\n", task->tk_pid);
972
973 spin_lock_bh(&xprt->transport_lock);
974 if (!req->rq_bytes_sent) {
975 if (req->rq_reply_bytes_recvd) {
976 task->tk_status = req->rq_reply_bytes_recvd;
977 goto out_unlock;
978 }
979 if ((task->tk_flags & RPC_TASK_NO_RETRANS_TIMEOUT)
980 && xprt_connected(xprt)
981 && req->rq_connect_cookie == xprt->connect_cookie) {
982 xprt->ops->set_retrans_timeout(task);
983 rpc_sleep_on(&xprt->pending, task, xprt_timer);
984 goto out_unlock;
985 }
986 }
987 if (!xprt->ops->reserve_xprt(xprt, task)) {
988 task->tk_status = -EAGAIN;
989 goto out_unlock;
990 }
991 ret = true;
992out_unlock:
993 spin_unlock_bh(&xprt->transport_lock);
994 return ret;
995}
996
997void xprt_end_transmit(struct rpc_task *task)
998{
999 xprt_release_write(task->tk_rqstp->rq_xprt, task);
1000}
1001
1002/**
1003 * xprt_transmit - send an RPC request on a transport
1004 * @task: controlling RPC task
1005 *
1006 * We have to copy the iovec because sendmsg fiddles with its contents.
1007 */
1008void xprt_transmit(struct rpc_task *task)
1009{
1010 struct rpc_rqst *req = task->tk_rqstp;
1011 struct rpc_xprt *xprt = req->rq_xprt;
1012 unsigned int connect_cookie;
1013 int status;
1014
1015 dprintk("RPC: %5u xprt_transmit(%u)\n", task->tk_pid, req->rq_slen);
1016
1017 if (!req->rq_reply_bytes_recvd) {
1018 if (list_empty(&req->rq_list) && rpc_reply_expected(task)) {
1019 /*
1020 * Add to the list only if we're expecting a reply
1021 */
1022 /* Update the softirq receive buffer */
1023 memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
1024 sizeof(req->rq_private_buf));
1025 /* Add request to the receive list */
1026 spin_lock(&xprt->recv_lock);
1027 list_add_tail(&req->rq_list, &xprt->recv);
1028 spin_unlock(&xprt->recv_lock);
1029 xprt_reset_majortimeo(req);
1030 /* Turn off autodisconnect */
1031 del_singleshot_timer_sync(&xprt->timer);
1032 }
1033 } else if (!req->rq_bytes_sent)
1034 return;
1035
1036 connect_cookie = xprt->connect_cookie;
1037 status = xprt->ops->send_request(task);
1038 trace_xprt_transmit(xprt, req->rq_xid, status);
1039 if (status != 0) {
1040 task->tk_status = status;
1041 return;
1042 }
1043 xprt_inject_disconnect(xprt);
1044
1045 dprintk("RPC: %5u xmit complete\n", task->tk_pid);
1046 task->tk_flags |= RPC_TASK_SENT;
1047 spin_lock_bh(&xprt->transport_lock);
1048
1049 xprt->ops->set_retrans_timeout(task);
1050
1051 xprt->stat.sends++;
1052 xprt->stat.req_u += xprt->stat.sends - xprt->stat.recvs;
1053 xprt->stat.bklog_u += xprt->backlog.qlen;
1054 xprt->stat.sending_u += xprt->sending.qlen;
1055 xprt->stat.pending_u += xprt->pending.qlen;
1056 spin_unlock_bh(&xprt->transport_lock);
1057
1058 req->rq_connect_cookie = connect_cookie;
1059 if (rpc_reply_expected(task) && !READ_ONCE(req->rq_reply_bytes_recvd)) {
1060 /*
1061 * Sleep on the pending queue if we're expecting a reply.
1062 * The spinlock ensures atomicity between the test of
1063 * req->rq_reply_bytes_recvd, and the call to rpc_sleep_on().
1064 */
1065 spin_lock(&xprt->recv_lock);
1066 if (!req->rq_reply_bytes_recvd) {
1067 rpc_sleep_on(&xprt->pending, task, xprt_timer);
1068 /*
1069 * Send an extra queue wakeup call if the
1070 * connection was dropped in case the call to
1071 * rpc_sleep_on() raced.
1072 */
1073 if (!xprt_connected(xprt))
1074 xprt_wake_pending_tasks(xprt, -ENOTCONN);
1075 }
1076 spin_unlock(&xprt->recv_lock);
1077 }
1078}
1079
1080static void xprt_add_backlog(struct rpc_xprt *xprt, struct rpc_task *task)
1081{
1082 set_bit(XPRT_CONGESTED, &xprt->state);
1083 rpc_sleep_on(&xprt->backlog, task, NULL);
1084}
1085
1086static void xprt_wake_up_backlog(struct rpc_xprt *xprt)
1087{
1088 if (rpc_wake_up_next(&xprt->backlog) == NULL)
1089 clear_bit(XPRT_CONGESTED, &xprt->state);
1090}
1091
1092static bool xprt_throttle_congested(struct rpc_xprt *xprt, struct rpc_task *task)
1093{
1094 bool ret = false;
1095
1096 if (!test_bit(XPRT_CONGESTED, &xprt->state))
1097 goto out;
1098 spin_lock(&xprt->reserve_lock);
1099 if (test_bit(XPRT_CONGESTED, &xprt->state)) {
1100 rpc_sleep_on(&xprt->backlog, task, NULL);
1101 ret = true;
1102 }
1103 spin_unlock(&xprt->reserve_lock);
1104out:
1105 return ret;
1106}
1107
1108static struct rpc_rqst *xprt_dynamic_alloc_slot(struct rpc_xprt *xprt)
1109{
1110 struct rpc_rqst *req = ERR_PTR(-EAGAIN);
1111
1112 if (xprt->num_reqs >= xprt->max_reqs)
1113 goto out;
1114 ++xprt->num_reqs;
1115 spin_unlock(&xprt->reserve_lock);
1116 req = kzalloc(sizeof(struct rpc_rqst), GFP_NOFS);
1117 spin_lock(&xprt->reserve_lock);
1118 if (req != NULL)
1119 goto out;
1120 --xprt->num_reqs;
1121 req = ERR_PTR(-ENOMEM);
1122out:
1123 return req;
1124}
1125
1126static bool xprt_dynamic_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req)
1127{
1128 if (xprt->num_reqs > xprt->min_reqs) {
1129 --xprt->num_reqs;
1130 kfree(req);
1131 return true;
1132 }
1133 return false;
1134}
1135
1136void xprt_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task)
1137{
1138 struct rpc_rqst *req;
1139
1140 spin_lock(&xprt->reserve_lock);
1141 if (!list_empty(&xprt->free)) {
1142 req = list_entry(xprt->free.next, struct rpc_rqst, rq_list);
1143 list_del(&req->rq_list);
1144 goto out_init_req;
1145 }
1146 req = xprt_dynamic_alloc_slot(xprt);
1147 if (!IS_ERR(req))
1148 goto out_init_req;
1149 switch (PTR_ERR(req)) {
1150 case -ENOMEM:
1151 dprintk("RPC: dynamic allocation of request slot "
1152 "failed! Retrying\n");
1153 task->tk_status = -ENOMEM;
1154 break;
1155 case -EAGAIN:
1156 xprt_add_backlog(xprt, task);
1157 dprintk("RPC: waiting for request slot\n");
1158 /* fall through */
1159 default:
1160 task->tk_status = -EAGAIN;
1161 }
1162 spin_unlock(&xprt->reserve_lock);
1163 return;
1164out_init_req:
1165 xprt->stat.max_slots = max_t(unsigned int, xprt->stat.max_slots,
1166 xprt->num_reqs);
1167 spin_unlock(&xprt->reserve_lock);
1168
1169 task->tk_status = 0;
1170 task->tk_rqstp = req;
1171}
1172EXPORT_SYMBOL_GPL(xprt_alloc_slot);
1173
1174void xprt_lock_and_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task)
1175{
1176 /* Note: grabbing the xprt_lock_write() ensures that we throttle
1177 * new slot allocation if the transport is congested (i.e. when
1178 * reconnecting a stream transport or when out of socket write
1179 * buffer space).
1180 */
1181 if (xprt_lock_write(xprt, task)) {
1182 xprt_alloc_slot(xprt, task);
1183 xprt_release_write(xprt, task);
1184 }
1185}
1186EXPORT_SYMBOL_GPL(xprt_lock_and_alloc_slot);
1187
1188void xprt_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req)
1189{
1190 spin_lock(&xprt->reserve_lock);
1191 if (!xprt_dynamic_free_slot(xprt, req)) {
1192 memset(req, 0, sizeof(*req)); /* mark unused */
1193 list_add(&req->rq_list, &xprt->free);
1194 }
1195 xprt_wake_up_backlog(xprt);
1196 spin_unlock(&xprt->reserve_lock);
1197}
1198EXPORT_SYMBOL_GPL(xprt_free_slot);
1199
1200static void xprt_free_all_slots(struct rpc_xprt *xprt)
1201{
1202 struct rpc_rqst *req;
1203 while (!list_empty(&xprt->free)) {
1204 req = list_first_entry(&xprt->free, struct rpc_rqst, rq_list);
1205 list_del(&req->rq_list);
1206 kfree(req);
1207 }
1208}
1209
1210struct rpc_xprt *xprt_alloc(struct net *net, size_t size,
1211 unsigned int num_prealloc,
1212 unsigned int max_alloc)
1213{
1214 struct rpc_xprt *xprt;
1215 struct rpc_rqst *req;
1216 int i;
1217
1218 xprt = kzalloc(size, GFP_KERNEL);
1219 if (xprt == NULL)
1220 goto out;
1221
1222 xprt_init(xprt, net);
1223
1224 for (i = 0; i < num_prealloc; i++) {
1225 req = kzalloc(sizeof(struct rpc_rqst), GFP_KERNEL);
1226 if (!req)
1227 goto out_free;
1228 list_add(&req->rq_list, &xprt->free);
1229 }
1230 if (max_alloc > num_prealloc)
1231 xprt->max_reqs = max_alloc;
1232 else
1233 xprt->max_reqs = num_prealloc;
1234 xprt->min_reqs = num_prealloc;
1235 xprt->num_reqs = num_prealloc;
1236
1237 return xprt;
1238
1239out_free:
1240 xprt_free(xprt);
1241out:
1242 return NULL;
1243}
1244EXPORT_SYMBOL_GPL(xprt_alloc);
1245
1246void xprt_free(struct rpc_xprt *xprt)
1247{
1248 put_net(xprt->xprt_net);
1249 xprt_free_all_slots(xprt);
1250 kfree_rcu(xprt, rcu);
1251}
1252EXPORT_SYMBOL_GPL(xprt_free);
1253
1254static __be32
1255xprt_alloc_xid(struct rpc_xprt *xprt)
1256{
1257 __be32 xid;
1258
1259 spin_lock(&xprt->reserve_lock);
1260 xid = (__force __be32)xprt->xid++;
1261 spin_unlock(&xprt->reserve_lock);
1262 return xid;
1263}
1264
1265static void
1266xprt_init_xid(struct rpc_xprt *xprt)
1267{
1268 xprt->xid = prandom_u32();
1269}
1270
1271static void
1272xprt_request_init(struct rpc_task *task)
1273{
1274 struct rpc_xprt *xprt = task->tk_xprt;
1275 struct rpc_rqst *req = task->tk_rqstp;
1276
1277 INIT_LIST_HEAD(&req->rq_list);
1278 req->rq_timeout = task->tk_client->cl_timeout->to_initval;
1279 req->rq_task = task;
1280 req->rq_xprt = xprt;
1281 req->rq_buffer = NULL;
1282 req->rq_xid = xprt_alloc_xid(xprt);
1283 req->rq_connect_cookie = xprt->connect_cookie - 1;
1284 req->rq_bytes_sent = 0;
1285 req->rq_snd_buf.len = 0;
1286 req->rq_snd_buf.buflen = 0;
1287 req->rq_rcv_buf.len = 0;
1288 req->rq_rcv_buf.buflen = 0;
1289 req->rq_release_snd_buf = NULL;
1290 xprt_reset_majortimeo(req);
1291 dprintk("RPC: %5u reserved req %p xid %08x\n", task->tk_pid,
1292 req, ntohl(req->rq_xid));
1293}
1294
1295static void
1296xprt_do_reserve(struct rpc_xprt *xprt, struct rpc_task *task)
1297{
1298 xprt->ops->alloc_slot(xprt, task);
1299 if (task->tk_rqstp != NULL)
1300 xprt_request_init(task);
1301}
1302
1303/**
1304 * xprt_reserve - allocate an RPC request slot
1305 * @task: RPC task requesting a slot allocation
1306 *
1307 * If the transport is marked as being congested, or if no more
1308 * slots are available, place the task on the transport's
1309 * backlog queue.
1310 */
1311void xprt_reserve(struct rpc_task *task)
1312{
1313 struct rpc_xprt *xprt = task->tk_xprt;
1314
1315 task->tk_status = 0;
1316 if (task->tk_rqstp != NULL)
1317 return;
1318
1319 task->tk_timeout = 0;
1320 task->tk_status = -EAGAIN;
1321 if (!xprt_throttle_congested(xprt, task))
1322 xprt_do_reserve(xprt, task);
1323}
1324
1325/**
1326 * xprt_retry_reserve - allocate an RPC request slot
1327 * @task: RPC task requesting a slot allocation
1328 *
1329 * If no more slots are available, place the task on the transport's
1330 * backlog queue.
1331 * Note that the only difference with xprt_reserve is that we now
1332 * ignore the value of the XPRT_CONGESTED flag.
1333 */
1334void xprt_retry_reserve(struct rpc_task *task)
1335{
1336 struct rpc_xprt *xprt = task->tk_xprt;
1337
1338 task->tk_status = 0;
1339 if (task->tk_rqstp != NULL)
1340 return;
1341
1342 task->tk_timeout = 0;
1343 task->tk_status = -EAGAIN;
1344 xprt_do_reserve(xprt, task);
1345}
1346
1347/**
1348 * xprt_release - release an RPC request slot
1349 * @task: task which is finished with the slot
1350 *
1351 */
1352void xprt_release(struct rpc_task *task)
1353{
1354 struct rpc_xprt *xprt;
1355 struct rpc_rqst *req = task->tk_rqstp;
1356
1357 if (req == NULL) {
1358 if (task->tk_client) {
1359 xprt = task->tk_xprt;
1360 if (xprt->snd_task == task)
1361 xprt_release_write(xprt, task);
1362 }
1363 return;
1364 }
1365
1366 xprt = req->rq_xprt;
1367 if (task->tk_ops->rpc_count_stats != NULL)
1368 task->tk_ops->rpc_count_stats(task, task->tk_calldata);
1369 else if (task->tk_client)
1370 rpc_count_iostats(task, task->tk_client->cl_metrics);
1371 spin_lock(&xprt->recv_lock);
1372 if (!list_empty(&req->rq_list)) {
1373 list_del_init(&req->rq_list);
1374 xprt_wait_on_pinned_rqst(req);
1375 }
1376 spin_unlock(&xprt->recv_lock);
1377 spin_lock_bh(&xprt->transport_lock);
1378 xprt->ops->release_xprt(xprt, task);
1379 if (xprt->ops->release_request)
1380 xprt->ops->release_request(task);
1381 xprt->last_used = jiffies;
1382 xprt_schedule_autodisconnect(xprt);
1383 spin_unlock_bh(&xprt->transport_lock);
1384 if (req->rq_buffer)
1385 xprt->ops->buf_free(task);
1386 xprt_inject_disconnect(xprt);
1387 if (req->rq_cred != NULL)
1388 put_rpccred(req->rq_cred);
1389 task->tk_rqstp = NULL;
1390 if (req->rq_release_snd_buf)
1391 req->rq_release_snd_buf(req);
1392
1393 dprintk("RPC: %5u release request %p\n", task->tk_pid, req);
1394 if (likely(!bc_prealloc(req)))
1395 xprt->ops->free_slot(xprt, req);
1396 else
1397 xprt_free_bc_request(req);
1398}
1399
1400static void xprt_init(struct rpc_xprt *xprt, struct net *net)
1401{
1402 kref_init(&xprt->kref);
1403
1404 spin_lock_init(&xprt->transport_lock);
1405 spin_lock_init(&xprt->reserve_lock);
1406 spin_lock_init(&xprt->recv_lock);
1407
1408 INIT_LIST_HEAD(&xprt->free);
1409 INIT_LIST_HEAD(&xprt->recv);
1410#if defined(CONFIG_SUNRPC_BACKCHANNEL)
1411 spin_lock_init(&xprt->bc_pa_lock);
1412 INIT_LIST_HEAD(&xprt->bc_pa_list);
1413#endif /* CONFIG_SUNRPC_BACKCHANNEL */
1414 INIT_LIST_HEAD(&xprt->xprt_switch);
1415
1416 xprt->last_used = jiffies;
1417 xprt->cwnd = RPC_INITCWND;
1418 xprt->bind_index = 0;
1419
1420 rpc_init_wait_queue(&xprt->binding, "xprt_binding");
1421 rpc_init_wait_queue(&xprt->pending, "xprt_pending");
1422 rpc_init_priority_wait_queue(&xprt->sending, "xprt_sending");
1423 rpc_init_priority_wait_queue(&xprt->backlog, "xprt_backlog");
1424
1425 xprt_init_xid(xprt);
1426
1427 xprt->xprt_net = get_net(net);
1428}
1429
1430/**
1431 * xprt_create_transport - create an RPC transport
1432 * @args: rpc transport creation arguments
1433 *
1434 */
1435struct rpc_xprt *xprt_create_transport(struct xprt_create *args)
1436{
1437 struct rpc_xprt *xprt;
1438 struct xprt_class *t;
1439
1440 spin_lock(&xprt_list_lock);
1441 list_for_each_entry(t, &xprt_list, list) {
1442 if (t->ident == args->ident) {
1443 spin_unlock(&xprt_list_lock);
1444 goto found;
1445 }
1446 }
1447 spin_unlock(&xprt_list_lock);
1448 dprintk("RPC: transport (%d) not supported\n", args->ident);
1449 return ERR_PTR(-EIO);
1450
1451found:
1452 xprt = t->setup(args);
1453 if (IS_ERR(xprt)) {
1454 dprintk("RPC: xprt_create_transport: failed, %ld\n",
1455 -PTR_ERR(xprt));
1456 goto out;
1457 }
1458 if (args->flags & XPRT_CREATE_NO_IDLE_TIMEOUT)
1459 xprt->idle_timeout = 0;
1460 INIT_WORK(&xprt->task_cleanup, xprt_autoclose);
1461 if (xprt_has_timer(xprt))
1462 timer_setup(&xprt->timer, xprt_init_autodisconnect, 0);
1463 else
1464 timer_setup(&xprt->timer, NULL, 0);
1465
1466 if (strlen(args->servername) > RPC_MAXNETNAMELEN) {
1467 xprt_destroy(xprt);
1468 return ERR_PTR(-EINVAL);
1469 }
1470 xprt->servername = kstrdup(args->servername, GFP_KERNEL);
1471 if (xprt->servername == NULL) {
1472 xprt_destroy(xprt);
1473 return ERR_PTR(-ENOMEM);
1474 }
1475
1476 rpc_xprt_debugfs_register(xprt);
1477
1478 dprintk("RPC: created transport %p with %u slots\n", xprt,
1479 xprt->max_reqs);
1480out:
1481 return xprt;
1482}
1483
1484static void xprt_destroy_cb(struct work_struct *work)
1485{
1486 struct rpc_xprt *xprt =
1487 container_of(work, struct rpc_xprt, task_cleanup);
1488
1489 rpc_xprt_debugfs_unregister(xprt);
1490 rpc_destroy_wait_queue(&xprt->binding);
1491 rpc_destroy_wait_queue(&xprt->pending);
1492 rpc_destroy_wait_queue(&xprt->sending);
1493 rpc_destroy_wait_queue(&xprt->backlog);
1494 kfree(xprt->servername);
1495 /*
1496 * Tear down transport state and free the rpc_xprt
1497 */
1498 xprt->ops->destroy(xprt);
1499}
1500
1501/**
1502 * xprt_destroy - destroy an RPC transport, killing off all requests.
1503 * @xprt: transport to destroy
1504 *
1505 */
1506static void xprt_destroy(struct rpc_xprt *xprt)
1507{
1508 dprintk("RPC: destroying transport %p\n", xprt);
1509
1510 /*
1511 * Exclude transport connect/disconnect handlers and autoclose
1512 */
1513 wait_on_bit_lock(&xprt->state, XPRT_LOCKED, TASK_UNINTERRUPTIBLE);
1514
1515 del_timer_sync(&xprt->timer);
1516
1517 /*
1518 * Destroy sockets etc from the system workqueue so they can
1519 * safely flush receive work running on rpciod.
1520 */
1521 INIT_WORK(&xprt->task_cleanup, xprt_destroy_cb);
1522 schedule_work(&xprt->task_cleanup);
1523}
1524
1525static void xprt_destroy_kref(struct kref *kref)
1526{
1527 xprt_destroy(container_of(kref, struct rpc_xprt, kref));
1528}
1529
1530/**
1531 * xprt_get - return a reference to an RPC transport.
1532 * @xprt: pointer to the transport
1533 *
1534 */
1535struct rpc_xprt *xprt_get(struct rpc_xprt *xprt)
1536{
1537 if (xprt != NULL && kref_get_unless_zero(&xprt->kref))
1538 return xprt;
1539 return NULL;
1540}
1541EXPORT_SYMBOL_GPL(xprt_get);
1542
1543/**
1544 * xprt_put - release a reference to an RPC transport.
1545 * @xprt: pointer to the transport
1546 *
1547 */
1548void xprt_put(struct rpc_xprt *xprt)
1549{
1550 if (xprt != NULL)
1551 kref_put(&xprt->kref, xprt_destroy_kref);
1552}
1553EXPORT_SYMBOL_GPL(xprt_put);