xref: /openbmc/linux/net/sunrpc/xprtrdma/transport.c (revision 113094f7)
1 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
2 /*
3  * Copyright (c) 2014-2017 Oracle.  All rights reserved.
4  * Copyright (c) 2003-2007 Network Appliance, Inc. All rights reserved.
5  *
6  * This software is available to you under a choice of one of two
7  * licenses.  You may choose to be licensed under the terms of the GNU
8  * General Public License (GPL) Version 2, available from the file
9  * COPYING in the main directory of this source tree, or the BSD-type
10  * license below:
11  *
12  * Redistribution and use in source and binary forms, with or without
13  * modification, are permitted provided that the following conditions
14  * are met:
15  *
16  *      Redistributions of source code must retain the above copyright
17  *      notice, this list of conditions and the following disclaimer.
18  *
19  *      Redistributions in binary form must reproduce the above
20  *      copyright notice, this list of conditions and the following
21  *      disclaimer in the documentation and/or other materials provided
22  *      with the distribution.
23  *
24  *      Neither the name of the Network Appliance, Inc. nor the names of
25  *      its contributors may be used to endorse or promote products
26  *      derived from this software without specific prior written
27  *      permission.
28  *
29  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
30  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
31  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
32  * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
33  * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
34  * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
35  * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
36  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
37  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
38  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
39  * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
40  */
41 
42 /*
43  * transport.c
44  *
45  * This file contains the top-level implementation of an RPC RDMA
46  * transport.
47  *
48  * Naming convention: functions beginning with xprt_ are part of the
49  * transport switch. All others are RPC RDMA internal.
50  */
51 
52 #include <linux/module.h>
53 #include <linux/slab.h>
54 #include <linux/seq_file.h>
55 #include <linux/smp.h>
56 
57 #include <linux/sunrpc/addr.h>
58 #include <linux/sunrpc/svc_rdma.h>
59 
60 #include "xprt_rdma.h"
61 #include <trace/events/rpcrdma.h>
62 
63 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
64 # define RPCDBG_FACILITY	RPCDBG_TRANS
65 #endif
66 
67 /*
68  * tunables
69  */
70 
71 unsigned int xprt_rdma_slot_table_entries = RPCRDMA_DEF_SLOT_TABLE;
72 unsigned int xprt_rdma_max_inline_read = RPCRDMA_DEF_INLINE;
73 unsigned int xprt_rdma_max_inline_write = RPCRDMA_DEF_INLINE;
74 unsigned int xprt_rdma_memreg_strategy		= RPCRDMA_FRWR;
75 int xprt_rdma_pad_optimize;
76 
77 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
78 
79 static unsigned int min_slot_table_size = RPCRDMA_MIN_SLOT_TABLE;
80 static unsigned int max_slot_table_size = RPCRDMA_MAX_SLOT_TABLE;
81 static unsigned int min_inline_size = RPCRDMA_MIN_INLINE;
82 static unsigned int max_inline_size = RPCRDMA_MAX_INLINE;
83 static unsigned int zero;
84 static unsigned int max_padding = PAGE_SIZE;
85 static unsigned int min_memreg = RPCRDMA_BOUNCEBUFFERS;
86 static unsigned int max_memreg = RPCRDMA_LAST - 1;
87 static unsigned int dummy;
88 
89 static struct ctl_table_header *sunrpc_table_header;
90 
91 static struct ctl_table xr_tunables_table[] = {
92 	{
93 		.procname	= "rdma_slot_table_entries",
94 		.data		= &xprt_rdma_slot_table_entries,
95 		.maxlen		= sizeof(unsigned int),
96 		.mode		= 0644,
97 		.proc_handler	= proc_dointvec_minmax,
98 		.extra1		= &min_slot_table_size,
99 		.extra2		= &max_slot_table_size
100 	},
101 	{
102 		.procname	= "rdma_max_inline_read",
103 		.data		= &xprt_rdma_max_inline_read,
104 		.maxlen		= sizeof(unsigned int),
105 		.mode		= 0644,
106 		.proc_handler	= proc_dointvec_minmax,
107 		.extra1		= &min_inline_size,
108 		.extra2		= &max_inline_size,
109 	},
110 	{
111 		.procname	= "rdma_max_inline_write",
112 		.data		= &xprt_rdma_max_inline_write,
113 		.maxlen		= sizeof(unsigned int),
114 		.mode		= 0644,
115 		.proc_handler	= proc_dointvec_minmax,
116 		.extra1		= &min_inline_size,
117 		.extra2		= &max_inline_size,
118 	},
119 	{
120 		.procname	= "rdma_inline_write_padding",
121 		.data		= &dummy,
122 		.maxlen		= sizeof(unsigned int),
123 		.mode		= 0644,
124 		.proc_handler	= proc_dointvec_minmax,
125 		.extra1		= &zero,
126 		.extra2		= &max_padding,
127 	},
128 	{
129 		.procname	= "rdma_memreg_strategy",
130 		.data		= &xprt_rdma_memreg_strategy,
131 		.maxlen		= sizeof(unsigned int),
132 		.mode		= 0644,
133 		.proc_handler	= proc_dointvec_minmax,
134 		.extra1		= &min_memreg,
135 		.extra2		= &max_memreg,
136 	},
137 	{
138 		.procname	= "rdma_pad_optimize",
139 		.data		= &xprt_rdma_pad_optimize,
140 		.maxlen		= sizeof(unsigned int),
141 		.mode		= 0644,
142 		.proc_handler	= proc_dointvec,
143 	},
144 	{ },
145 };
146 
147 static struct ctl_table sunrpc_table[] = {
148 	{
149 		.procname	= "sunrpc",
150 		.mode		= 0555,
151 		.child		= xr_tunables_table
152 	},
153 	{ },
154 };
155 
156 #endif
157 
158 static const struct rpc_xprt_ops xprt_rdma_procs;
159 
160 static void
161 xprt_rdma_format_addresses4(struct rpc_xprt *xprt, struct sockaddr *sap)
162 {
163 	struct sockaddr_in *sin = (struct sockaddr_in *)sap;
164 	char buf[20];
165 
166 	snprintf(buf, sizeof(buf), "%08x", ntohl(sin->sin_addr.s_addr));
167 	xprt->address_strings[RPC_DISPLAY_HEX_ADDR] = kstrdup(buf, GFP_KERNEL);
168 
169 	xprt->address_strings[RPC_DISPLAY_NETID] = RPCBIND_NETID_RDMA;
170 }
171 
172 static void
173 xprt_rdma_format_addresses6(struct rpc_xprt *xprt, struct sockaddr *sap)
174 {
175 	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
176 	char buf[40];
177 
178 	snprintf(buf, sizeof(buf), "%pi6", &sin6->sin6_addr);
179 	xprt->address_strings[RPC_DISPLAY_HEX_ADDR] = kstrdup(buf, GFP_KERNEL);
180 
181 	xprt->address_strings[RPC_DISPLAY_NETID] = RPCBIND_NETID_RDMA6;
182 }
183 
184 void
185 xprt_rdma_format_addresses(struct rpc_xprt *xprt, struct sockaddr *sap)
186 {
187 	char buf[128];
188 
189 	switch (sap->sa_family) {
190 	case AF_INET:
191 		xprt_rdma_format_addresses4(xprt, sap);
192 		break;
193 	case AF_INET6:
194 		xprt_rdma_format_addresses6(xprt, sap);
195 		break;
196 	default:
197 		pr_err("rpcrdma: Unrecognized address family\n");
198 		return;
199 	}
200 
201 	(void)rpc_ntop(sap, buf, sizeof(buf));
202 	xprt->address_strings[RPC_DISPLAY_ADDR] = kstrdup(buf, GFP_KERNEL);
203 
204 	snprintf(buf, sizeof(buf), "%u", rpc_get_port(sap));
205 	xprt->address_strings[RPC_DISPLAY_PORT] = kstrdup(buf, GFP_KERNEL);
206 
207 	snprintf(buf, sizeof(buf), "%4hx", rpc_get_port(sap));
208 	xprt->address_strings[RPC_DISPLAY_HEX_PORT] = kstrdup(buf, GFP_KERNEL);
209 
210 	xprt->address_strings[RPC_DISPLAY_PROTO] = "rdma";
211 }
212 
213 void
214 xprt_rdma_free_addresses(struct rpc_xprt *xprt)
215 {
216 	unsigned int i;
217 
218 	for (i = 0; i < RPC_DISPLAY_MAX; i++)
219 		switch (i) {
220 		case RPC_DISPLAY_PROTO:
221 		case RPC_DISPLAY_NETID:
222 			continue;
223 		default:
224 			kfree(xprt->address_strings[i]);
225 		}
226 }
227 
228 /**
229  * xprt_rdma_connect_worker - establish connection in the background
230  * @work: worker thread context
231  *
232  * Requester holds the xprt's send lock to prevent activity on this
233  * transport while a fresh connection is being established. RPC tasks
234  * sleep on the xprt's pending queue waiting for connect to complete.
235  */
236 static void
237 xprt_rdma_connect_worker(struct work_struct *work)
238 {
239 	struct rpcrdma_xprt *r_xprt = container_of(work, struct rpcrdma_xprt,
240 						   rx_connect_worker.work);
241 	struct rpc_xprt *xprt = &r_xprt->rx_xprt;
242 	int rc;
243 
244 	rc = rpcrdma_ep_connect(&r_xprt->rx_ep, &r_xprt->rx_ia);
245 	xprt_clear_connecting(xprt);
246 	if (r_xprt->rx_ep.rep_connected > 0) {
247 		if (!xprt_test_and_set_connected(xprt)) {
248 			xprt->stat.connect_count++;
249 			xprt->stat.connect_time += (long)jiffies -
250 						   xprt->stat.connect_start;
251 			xprt_wake_pending_tasks(xprt, -EAGAIN);
252 		}
253 	} else {
254 		if (xprt_test_and_clear_connected(xprt))
255 			xprt_wake_pending_tasks(xprt, rc);
256 	}
257 }
258 
259 /**
260  * xprt_rdma_inject_disconnect - inject a connection fault
261  * @xprt: transport context
262  *
263  * If @xprt is connected, disconnect it to simulate spurious connection
264  * loss.
265  */
266 static void
267 xprt_rdma_inject_disconnect(struct rpc_xprt *xprt)
268 {
269 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
270 
271 	trace_xprtrdma_op_inject_dsc(r_xprt);
272 	rdma_disconnect(r_xprt->rx_ia.ri_id);
273 }
274 
275 /**
276  * xprt_rdma_destroy - Full tear down of transport
277  * @xprt: doomed transport context
278  *
279  * Caller guarantees there will be no more calls to us with
280  * this @xprt.
281  */
282 static void
283 xprt_rdma_destroy(struct rpc_xprt *xprt)
284 {
285 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
286 
287 	trace_xprtrdma_op_destroy(r_xprt);
288 
289 	cancel_delayed_work_sync(&r_xprt->rx_connect_worker);
290 
291 	rpcrdma_ep_destroy(r_xprt);
292 	rpcrdma_buffer_destroy(&r_xprt->rx_buf);
293 	rpcrdma_ia_close(&r_xprt->rx_ia);
294 
295 	xprt_rdma_free_addresses(xprt);
296 	xprt_free(xprt);
297 
298 	module_put(THIS_MODULE);
299 }
300 
301 static const struct rpc_timeout xprt_rdma_default_timeout = {
302 	.to_initval = 60 * HZ,
303 	.to_maxval = 60 * HZ,
304 };
305 
306 /**
307  * xprt_setup_rdma - Set up transport to use RDMA
308  *
309  * @args: rpc transport arguments
310  */
311 static struct rpc_xprt *
312 xprt_setup_rdma(struct xprt_create *args)
313 {
314 	struct rpc_xprt *xprt;
315 	struct rpcrdma_xprt *new_xprt;
316 	struct sockaddr *sap;
317 	int rc;
318 
319 	if (args->addrlen > sizeof(xprt->addr))
320 		return ERR_PTR(-EBADF);
321 
322 	xprt = xprt_alloc(args->net, sizeof(struct rpcrdma_xprt), 0, 0);
323 	if (!xprt)
324 		return ERR_PTR(-ENOMEM);
325 
326 	/* 60 second timeout, no retries */
327 	xprt->timeout = &xprt_rdma_default_timeout;
328 	xprt->bind_timeout = RPCRDMA_BIND_TO;
329 	xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
330 	xprt->idle_timeout = RPCRDMA_IDLE_DISC_TO;
331 
332 	xprt->resvport = 0;		/* privileged port not needed */
333 	xprt->ops = &xprt_rdma_procs;
334 
335 	/*
336 	 * Set up RDMA-specific connect data.
337 	 */
338 	sap = args->dstaddr;
339 
340 	/* Ensure xprt->addr holds valid server TCP (not RDMA)
341 	 * address, for any side protocols which peek at it */
342 	xprt->prot = IPPROTO_TCP;
343 	xprt->addrlen = args->addrlen;
344 	memcpy(&xprt->addr, sap, xprt->addrlen);
345 
346 	if (rpc_get_port(sap))
347 		xprt_set_bound(xprt);
348 	xprt_rdma_format_addresses(xprt, sap);
349 
350 	new_xprt = rpcx_to_rdmax(xprt);
351 	rc = rpcrdma_ia_open(new_xprt);
352 	if (rc)
353 		goto out1;
354 
355 	rc = rpcrdma_ep_create(new_xprt);
356 	if (rc)
357 		goto out2;
358 
359 	rc = rpcrdma_buffer_create(new_xprt);
360 	if (rc)
361 		goto out3;
362 
363 	INIT_DELAYED_WORK(&new_xprt->rx_connect_worker,
364 			  xprt_rdma_connect_worker);
365 
366 	xprt->max_payload = frwr_maxpages(new_xprt);
367 	if (xprt->max_payload == 0)
368 		goto out4;
369 	xprt->max_payload <<= PAGE_SHIFT;
370 	dprintk("RPC:       %s: transport data payload maximum: %zu bytes\n",
371 		__func__, xprt->max_payload);
372 
373 	if (!try_module_get(THIS_MODULE))
374 		goto out4;
375 
376 	dprintk("RPC:       %s: %s:%s\n", __func__,
377 		xprt->address_strings[RPC_DISPLAY_ADDR],
378 		xprt->address_strings[RPC_DISPLAY_PORT]);
379 	trace_xprtrdma_create(new_xprt);
380 	return xprt;
381 
382 out4:
383 	rpcrdma_buffer_destroy(&new_xprt->rx_buf);
384 	rc = -ENODEV;
385 out3:
386 	rpcrdma_ep_destroy(new_xprt);
387 out2:
388 	rpcrdma_ia_close(&new_xprt->rx_ia);
389 out1:
390 	trace_xprtrdma_op_destroy(new_xprt);
391 	xprt_rdma_free_addresses(xprt);
392 	xprt_free(xprt);
393 	return ERR_PTR(rc);
394 }
395 
396 /**
397  * xprt_rdma_close - close a transport connection
398  * @xprt: transport context
399  *
400  * Called during autoclose or device removal.
401  *
402  * Caller holds @xprt's send lock to prevent activity on this
403  * transport while the connection is torn down.
404  */
405 void xprt_rdma_close(struct rpc_xprt *xprt)
406 {
407 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
408 	struct rpcrdma_ep *ep = &r_xprt->rx_ep;
409 	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
410 
411 	might_sleep();
412 
413 	trace_xprtrdma_op_close(r_xprt);
414 
415 	/* Prevent marshaling and sending of new requests */
416 	xprt_clear_connected(xprt);
417 
418 	if (test_and_clear_bit(RPCRDMA_IAF_REMOVING, &ia->ri_flags)) {
419 		rpcrdma_ia_remove(ia);
420 		goto out;
421 	}
422 
423 	if (ep->rep_connected == -ENODEV)
424 		return;
425 	if (ep->rep_connected > 0)
426 		xprt->reestablish_timeout = 0;
427 	rpcrdma_ep_disconnect(ep, ia);
428 
429 	/* Prepare @xprt for the next connection by reinitializing
430 	 * its credit grant to one (see RFC 8166, Section 3.3.3).
431 	 */
432 	r_xprt->rx_buf.rb_credits = 1;
433 	xprt->cwnd = RPC_CWNDSHIFT;
434 
435 out:
436 	++xprt->connect_cookie;
437 	xprt_disconnect_done(xprt);
438 }
439 
440 /**
441  * xprt_rdma_set_port - update server port with rpcbind result
442  * @xprt: controlling RPC transport
443  * @port: new port value
444  *
445  * Transport connect status is unchanged.
446  */
447 static void
448 xprt_rdma_set_port(struct rpc_xprt *xprt, u16 port)
449 {
450 	struct sockaddr *sap = (struct sockaddr *)&xprt->addr;
451 	char buf[8];
452 
453 	dprintk("RPC:       %s: setting port for xprt %p (%s:%s) to %u\n",
454 		__func__, xprt,
455 		xprt->address_strings[RPC_DISPLAY_ADDR],
456 		xprt->address_strings[RPC_DISPLAY_PORT],
457 		port);
458 
459 	rpc_set_port(sap, port);
460 
461 	kfree(xprt->address_strings[RPC_DISPLAY_PORT]);
462 	snprintf(buf, sizeof(buf), "%u", port);
463 	xprt->address_strings[RPC_DISPLAY_PORT] = kstrdup(buf, GFP_KERNEL);
464 
465 	kfree(xprt->address_strings[RPC_DISPLAY_HEX_PORT]);
466 	snprintf(buf, sizeof(buf), "%4hx", port);
467 	xprt->address_strings[RPC_DISPLAY_HEX_PORT] = kstrdup(buf, GFP_KERNEL);
468 }
469 
470 /**
471  * xprt_rdma_timer - invoked when an RPC times out
472  * @xprt: controlling RPC transport
473  * @task: RPC task that timed out
474  *
475  * Invoked when the transport is still connected, but an RPC
476  * retransmit timeout occurs.
477  *
478  * Since RDMA connections don't have a keep-alive, forcibly
479  * disconnect and retry to connect. This drives full
480  * detection of the network path, and retransmissions of
481  * all pending RPCs.
482  */
483 static void
484 xprt_rdma_timer(struct rpc_xprt *xprt, struct rpc_task *task)
485 {
486 	xprt_force_disconnect(xprt);
487 }
488 
489 /**
490  * xprt_rdma_connect - try to establish a transport connection
491  * @xprt: transport state
492  * @task: RPC scheduler context
493  *
494  */
495 static void
496 xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task)
497 {
498 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
499 
500 	trace_xprtrdma_op_connect(r_xprt);
501 	if (r_xprt->rx_ep.rep_connected != 0) {
502 		/* Reconnect */
503 		schedule_delayed_work(&r_xprt->rx_connect_worker,
504 				      xprt->reestablish_timeout);
505 		xprt->reestablish_timeout <<= 1;
506 		if (xprt->reestablish_timeout > RPCRDMA_MAX_REEST_TO)
507 			xprt->reestablish_timeout = RPCRDMA_MAX_REEST_TO;
508 		else if (xprt->reestablish_timeout < RPCRDMA_INIT_REEST_TO)
509 			xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
510 	} else {
511 		schedule_delayed_work(&r_xprt->rx_connect_worker, 0);
512 		if (!RPC_IS_ASYNC(task))
513 			flush_delayed_work(&r_xprt->rx_connect_worker);
514 	}
515 }
516 
517 /**
518  * xprt_rdma_alloc_slot - allocate an rpc_rqst
519  * @xprt: controlling RPC transport
520  * @task: RPC task requesting a fresh rpc_rqst
521  *
522  * tk_status values:
523  *	%0 if task->tk_rqstp points to a fresh rpc_rqst
524  *	%-EAGAIN if no rpc_rqst is available; queued on backlog
525  */
526 static void
527 xprt_rdma_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task)
528 {
529 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
530 	struct rpcrdma_req *req;
531 
532 	req = rpcrdma_buffer_get(&r_xprt->rx_buf);
533 	if (!req)
534 		goto out_sleep;
535 	task->tk_rqstp = &req->rl_slot;
536 	task->tk_status = 0;
537 	return;
538 
539 out_sleep:
540 	rpc_sleep_on(&xprt->backlog, task, NULL);
541 	task->tk_status = -EAGAIN;
542 }
543 
544 /**
545  * xprt_rdma_free_slot - release an rpc_rqst
546  * @xprt: controlling RPC transport
547  * @rqst: rpc_rqst to release
548  *
549  */
550 static void
551 xprt_rdma_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *rqst)
552 {
553 	memset(rqst, 0, sizeof(*rqst));
554 	rpcrdma_buffer_put(rpcr_to_rdmar(rqst));
555 	rpc_wake_up_next(&xprt->backlog);
556 }
557 
558 static bool rpcrdma_check_regbuf(struct rpcrdma_xprt *r_xprt,
559 				 struct rpcrdma_regbuf *rb, size_t size,
560 				 gfp_t flags)
561 {
562 	if (unlikely(rdmab_length(rb) < size)) {
563 		if (!rpcrdma_regbuf_realloc(rb, size, flags))
564 			return false;
565 		r_xprt->rx_stats.hardway_register_count += size;
566 	}
567 	return true;
568 }
569 
570 /**
571  * xprt_rdma_allocate - allocate transport resources for an RPC
572  * @task: RPC task
573  *
574  * Return values:
575  *        0:	Success; rq_buffer points to RPC buffer to use
576  *   ENOMEM:	Out of memory, call again later
577  *      EIO:	A permanent error occurred, do not retry
578  */
579 static int
580 xprt_rdma_allocate(struct rpc_task *task)
581 {
582 	struct rpc_rqst *rqst = task->tk_rqstp;
583 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt);
584 	struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
585 	gfp_t flags;
586 
587 	flags = RPCRDMA_DEF_GFP;
588 	if (RPC_IS_SWAPPER(task))
589 		flags = __GFP_MEMALLOC | GFP_NOWAIT | __GFP_NOWARN;
590 
591 	if (!rpcrdma_check_regbuf(r_xprt, req->rl_sendbuf, rqst->rq_callsize,
592 				  flags))
593 		goto out_fail;
594 	if (!rpcrdma_check_regbuf(r_xprt, req->rl_recvbuf, rqst->rq_rcvsize,
595 				  flags))
596 		goto out_fail;
597 
598 	rqst->rq_buffer = rdmab_data(req->rl_sendbuf);
599 	rqst->rq_rbuffer = rdmab_data(req->rl_recvbuf);
600 	trace_xprtrdma_op_allocate(task, req);
601 	return 0;
602 
603 out_fail:
604 	trace_xprtrdma_op_allocate(task, NULL);
605 	return -ENOMEM;
606 }
607 
608 /**
609  * xprt_rdma_free - release resources allocated by xprt_rdma_allocate
610  * @task: RPC task
611  *
612  * Caller guarantees rqst->rq_buffer is non-NULL.
613  */
614 static void
615 xprt_rdma_free(struct rpc_task *task)
616 {
617 	struct rpc_rqst *rqst = task->tk_rqstp;
618 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt);
619 	struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
620 
621 	if (test_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags))
622 		rpcrdma_release_rqst(r_xprt, req);
623 	trace_xprtrdma_op_free(task, req);
624 }
625 
626 /**
627  * xprt_rdma_send_request - marshal and send an RPC request
628  * @rqst: RPC message in rq_snd_buf
629  *
630  * Caller holds the transport's write lock.
631  *
632  * Returns:
633  *	%0 if the RPC message has been sent
634  *	%-ENOTCONN if the caller should reconnect and call again
635  *	%-EAGAIN if the caller should call again
636  *	%-ENOBUFS if the caller should call again after a delay
637  *	%-EMSGSIZE if encoding ran out of buffer space. The request
638  *		was not sent. Do not try to send this message again.
639  *	%-EIO if an I/O error occurred. The request was not sent.
640  *		Do not try to send this message again.
641  */
642 static int
643 xprt_rdma_send_request(struct rpc_rqst *rqst)
644 {
645 	struct rpc_xprt *xprt = rqst->rq_xprt;
646 	struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
647 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
648 	int rc = 0;
649 
650 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
651 	if (unlikely(!rqst->rq_buffer))
652 		return xprt_rdma_bc_send_reply(rqst);
653 #endif	/* CONFIG_SUNRPC_BACKCHANNEL */
654 
655 	if (!xprt_connected(xprt))
656 		return -ENOTCONN;
657 
658 	if (!xprt_request_get_cong(xprt, rqst))
659 		return -EBADSLT;
660 
661 	rc = rpcrdma_marshal_req(r_xprt, rqst);
662 	if (rc < 0)
663 		goto failed_marshal;
664 
665 	/* Must suppress retransmit to maintain credits */
666 	if (rqst->rq_connect_cookie == xprt->connect_cookie)
667 		goto drop_connection;
668 	rqst->rq_xtime = ktime_get();
669 
670 	__set_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags);
671 	if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req))
672 		goto drop_connection;
673 
674 	rqst->rq_xmit_bytes_sent += rqst->rq_snd_buf.len;
675 
676 	/* An RPC with no reply will throw off credit accounting,
677 	 * so drop the connection to reset the credit grant.
678 	 */
679 	if (!rpc_reply_expected(rqst->rq_task))
680 		goto drop_connection;
681 	return 0;
682 
683 failed_marshal:
684 	if (rc != -ENOTCONN)
685 		return rc;
686 drop_connection:
687 	xprt_rdma_close(xprt);
688 	return -ENOTCONN;
689 }
690 
691 void xprt_rdma_print_stats(struct rpc_xprt *xprt, struct seq_file *seq)
692 {
693 	struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
694 	long idle_time = 0;
695 
696 	if (xprt_connected(xprt))
697 		idle_time = (long)(jiffies - xprt->last_used) / HZ;
698 
699 	seq_puts(seq, "\txprt:\trdma ");
700 	seq_printf(seq, "%u %lu %lu %lu %ld %lu %lu %lu %llu %llu ",
701 		   0,	/* need a local port? */
702 		   xprt->stat.bind_count,
703 		   xprt->stat.connect_count,
704 		   xprt->stat.connect_time / HZ,
705 		   idle_time,
706 		   xprt->stat.sends,
707 		   xprt->stat.recvs,
708 		   xprt->stat.bad_xids,
709 		   xprt->stat.req_u,
710 		   xprt->stat.bklog_u);
711 	seq_printf(seq, "%lu %lu %lu %llu %llu %llu %llu %lu %lu %lu %lu ",
712 		   r_xprt->rx_stats.read_chunk_count,
713 		   r_xprt->rx_stats.write_chunk_count,
714 		   r_xprt->rx_stats.reply_chunk_count,
715 		   r_xprt->rx_stats.total_rdma_request,
716 		   r_xprt->rx_stats.total_rdma_reply,
717 		   r_xprt->rx_stats.pullup_copy_count,
718 		   r_xprt->rx_stats.fixup_copy_count,
719 		   r_xprt->rx_stats.hardway_register_count,
720 		   r_xprt->rx_stats.failed_marshal_count,
721 		   r_xprt->rx_stats.bad_reply_count,
722 		   r_xprt->rx_stats.nomsg_call_count);
723 	seq_printf(seq, "%lu %lu %lu %lu %lu %lu\n",
724 		   r_xprt->rx_stats.mrs_recycled,
725 		   r_xprt->rx_stats.mrs_orphaned,
726 		   r_xprt->rx_stats.mrs_allocated,
727 		   r_xprt->rx_stats.local_inv_needed,
728 		   r_xprt->rx_stats.empty_sendctx_q,
729 		   r_xprt->rx_stats.reply_waits_for_send);
730 }
731 
732 static int
733 xprt_rdma_enable_swap(struct rpc_xprt *xprt)
734 {
735 	return 0;
736 }
737 
738 static void
739 xprt_rdma_disable_swap(struct rpc_xprt *xprt)
740 {
741 }
742 
743 /*
744  * Plumbing for rpc transport switch and kernel module
745  */
746 
747 static const struct rpc_xprt_ops xprt_rdma_procs = {
748 	.reserve_xprt		= xprt_reserve_xprt_cong,
749 	.release_xprt		= xprt_release_xprt_cong, /* sunrpc/xprt.c */
750 	.alloc_slot		= xprt_rdma_alloc_slot,
751 	.free_slot		= xprt_rdma_free_slot,
752 	.release_request	= xprt_release_rqst_cong,       /* ditto */
753 	.wait_for_reply_request	= xprt_wait_for_reply_request_def, /* ditto */
754 	.timer			= xprt_rdma_timer,
755 	.rpcbind		= rpcb_getport_async,	/* sunrpc/rpcb_clnt.c */
756 	.set_port		= xprt_rdma_set_port,
757 	.connect		= xprt_rdma_connect,
758 	.buf_alloc		= xprt_rdma_allocate,
759 	.buf_free		= xprt_rdma_free,
760 	.send_request		= xprt_rdma_send_request,
761 	.close			= xprt_rdma_close,
762 	.destroy		= xprt_rdma_destroy,
763 	.print_stats		= xprt_rdma_print_stats,
764 	.enable_swap		= xprt_rdma_enable_swap,
765 	.disable_swap		= xprt_rdma_disable_swap,
766 	.inject_disconnect	= xprt_rdma_inject_disconnect,
767 #if defined(CONFIG_SUNRPC_BACKCHANNEL)
768 	.bc_setup		= xprt_rdma_bc_setup,
769 	.bc_maxpayload		= xprt_rdma_bc_maxpayload,
770 	.bc_free_rqst		= xprt_rdma_bc_free_rqst,
771 	.bc_destroy		= xprt_rdma_bc_destroy,
772 #endif
773 };
774 
775 static struct xprt_class xprt_rdma = {
776 	.list			= LIST_HEAD_INIT(xprt_rdma.list),
777 	.name			= "rdma",
778 	.owner			= THIS_MODULE,
779 	.ident			= XPRT_TRANSPORT_RDMA,
780 	.setup			= xprt_setup_rdma,
781 };
782 
783 void xprt_rdma_cleanup(void)
784 {
785 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
786 	if (sunrpc_table_header) {
787 		unregister_sysctl_table(sunrpc_table_header);
788 		sunrpc_table_header = NULL;
789 	}
790 #endif
791 
792 	xprt_unregister_transport(&xprt_rdma);
793 	xprt_unregister_transport(&xprt_rdma_bc);
794 }
795 
796 int xprt_rdma_init(void)
797 {
798 	int rc;
799 
800 	rc = xprt_register_transport(&xprt_rdma);
801 	if (rc)
802 		return rc;
803 
804 	rc = xprt_register_transport(&xprt_rdma_bc);
805 	if (rc) {
806 		xprt_unregister_transport(&xprt_rdma);
807 		return rc;
808 	}
809 
810 #if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
811 	if (!sunrpc_table_header)
812 		sunrpc_table_header = register_sysctl_table(sunrpc_table);
813 #endif
814 	return 0;
815 }
816