summaryrefslogtreecommitdiff
path: root/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
blob: 232860ea683bfa865a3d1dea62f19a59d30e6087 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
/*
 * Copyright (c) 2016-2018 Oracle. All rights reserved.
 * Copyright (c) 2014 Open Grid Computing, Inc. All rights reserved.
 * Copyright (c) 2005-2006 Network Appliance, Inc. All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the BSD-type
 * license below:
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 *
 *      Redistributions of source code must retain the above copyright
 *      notice, this list of conditions and the following disclaimer.
 *
 *      Redistributions in binary form must reproduce the above
 *      copyright notice, this list of conditions and the following
 *      disclaimer in the documentation and/or other materials provided
 *      with the distribution.
 *
 *      Neither the name of the Network Appliance, Inc. nor the names of
 *      its contributors may be used to endorse or promote products
 *      derived from this software without specific prior written
 *      permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 *
 * Author: Tom Tucker <tom@opengridcomputing.com>
 */

/* Operation
 *
 * The main entry point is svc_rdma_recvfrom. This is called from
 * svc_recv when the transport indicates there is incoming data to
 * be read. "Data Ready" is signaled when an RDMA Receive completes,
 * or when a set of RDMA Reads complete.
 *
 * An svc_rqst is passed in. This structure contains an array of
 * free pages (rq_pages) that will contain the incoming RPC message.
 *
 * Short messages are moved directly into svc_rqst::rq_arg, and
 * the RPC Call is ready to be processed by the Upper Layer.
 * svc_rdma_recvfrom returns the length of the RPC Call message,
 * completing the reception of the RPC Call.
 *
 * However, when an incoming message has Read chunks,
 * svc_rdma_recvfrom must post RDMA Reads to pull the RPC Call's
 * data payload from the client. svc_rdma_recvfrom sets up the
 * RDMA Reads using pages in svc_rqst::rq_pages, which are
 * transferred to an svc_rdma_recv_ctxt for the duration of the
 * I/O. svc_rdma_recvfrom then returns zero, since the RPC message
 * is still not yet ready.
 *
 * When the Read chunk payloads have become available on the
 * server, "Data Ready" is raised again, and svc_recv calls
 * svc_rdma_recvfrom again. This second call may use a different
 * svc_rqst than the first one, thus any information that needs
 * to be preserved across these two calls is kept in an
 * svc_rdma_recv_ctxt.
 *
 * The second call to svc_rdma_recvfrom performs final assembly
 * of the RPC Call message, using the RDMA Read sink pages kept in
 * the svc_rdma_recv_ctxt. The xdr_buf is copied from the
 * svc_rdma_recv_ctxt to the second svc_rqst. The second call returns
 * the length of the completed RPC Call message.
 *
 * Page Management
 *
 * Pages under I/O must be transferred from the first svc_rqst to an
 * svc_rdma_recv_ctxt before the first svc_rdma_recvfrom call returns.
 *
 * The first svc_rqst supplies pages for RDMA Reads. These are moved
 * from rqstp::rq_pages into ctxt::pages. The consumed elements of
 * the rq_pages array are set to NULL and refilled with the first
 * svc_rdma_recvfrom call returns.
 *
 * During the second svc_rdma_recvfrom call, RDMA Read sink pages
 * are transferred from the svc_rdma_recv_ctxt to the second svc_rqst.
 */

#include <linux/slab.h>
#include <linux/spinlock.h>
#include <asm/unaligned.h>
#include <rdma/ib_verbs.h>
#include <rdma/rdma_cm.h>

#include <linux/sunrpc/xdr.h>
#include <linux/sunrpc/debug.h>
#include <linux/sunrpc/rpc_rdma.h>
#include <linux/sunrpc/svc_rdma.h>

#include "xprt_rdma.h"
#include <trace/events/rpcrdma.h>

static void svc_rdma_wc_receive(struct ib_cq *cq, struct ib_wc *wc);

static inline struct svc_rdma_recv_ctxt *
svc_rdma_next_recv_ctxt(struct list_head *list)
{
	return list_first_entry_or_null(list, struct svc_rdma_recv_ctxt,
					rc_list);
}

static void svc_rdma_recv_cid_init(struct svcxprt_rdma *rdma,
				   struct rpc_rdma_cid *cid)
{
	cid->ci_queue_id = rdma->sc_rq_cq->res.id;
	cid->ci_completion_id = atomic_inc_return(&rdma->sc_completion_ids);
}

static struct svc_rdma_recv_ctxt *
svc_rdma_recv_ctxt_alloc(struct svcxprt_rdma *rdma)
{
	struct svc_rdma_recv_ctxt *ctxt;
	dma_addr_t addr;
	void *buffer;

	ctxt = kmalloc(sizeof(*ctxt), GFP_KERNEL);
	if (!ctxt)
		goto fail0;
	buffer = kmalloc(rdma->sc_max_req_size, GFP_KERNEL);
	if (!buffer)
		goto fail1;
	addr = ib_dma_map_single(rdma->sc_pd->device, buffer,
				 rdma->sc_max_req_size, DMA_FROM_DEVICE);
	if (ib_dma_mapping_error(rdma->sc_pd->device, addr))
		goto fail2;

	svc_rdma_recv_cid_init(rdma, &ctxt->rc_cid);
	pcl_init(&ctxt->rc_call_pcl);
	pcl_init(&ctxt->rc_read_pcl);
	pcl_init(&ctxt->rc_write_pcl);
	pcl_init(&ctxt->rc_reply_pcl);

	ctxt->rc_recv_wr.next = NULL;
	ctxt->rc_recv_wr.wr_cqe = &ctxt->rc_cqe;
	ctxt->rc_recv_wr.sg_list = &ctxt->rc_recv_sge;
	ctxt->rc_recv_wr.num_sge = 1;
	ctxt->rc_cqe.done = svc_rdma_wc_receive;
	ctxt->rc_recv_sge.addr = addr;
	ctxt->rc_recv_sge.length = rdma->sc_max_req_size;
	ctxt->rc_recv_sge.lkey = rdma->sc_pd->local_dma_lkey;
	ctxt->rc_recv_buf = buffer;
	ctxt->rc_temp = false;
	return ctxt;

fail2:
	kfree(buffer);
fail1:
	kfree(ctxt);
fail0:
	return NULL;
}

static void svc_rdma_recv_ctxt_destroy(struct svcxprt_rdma *rdma,
				       struct svc_rdma_recv_ctxt *ctxt)
{
	ib_dma_unmap_single(rdma->sc_pd->device, ctxt->rc_recv_sge.addr,
			    ctxt->rc_recv_sge.length, DMA_FROM_DEVICE);
	kfree(ctxt->rc_recv_buf);
	kfree(ctxt);
}

/**
 * svc_rdma_recv_ctxts_destroy - Release all recv_ctxt's for an xprt
 * @rdma: svcxprt_rdma being torn down
 *
 */
void svc_rdma_recv_ctxts_destroy(struct svcxprt_rdma *rdma)
{
	struct svc_rdma_recv_ctxt *ctxt;
	struct llist_node *node;

	while ((node = llist_del_first(&rdma->sc_recv_ctxts))) {
		ctxt = llist_entry(node, struct svc_rdma_recv_ctxt, rc_node);
		svc_rdma_recv_ctxt_destroy(rdma, ctxt);
	}
}

/**
 * svc_rdma_recv_ctxt_get - Allocate a recv_ctxt
 * @rdma: controlling svcxprt_rdma
 *
 * Returns a recv_ctxt or (rarely) NULL if none are available.
 */
struct svc_rdma_recv_ctxt *svc_rdma_recv_ctxt_get(struct svcxprt_rdma *rdma)
{
	struct svc_rdma_recv_ctxt *ctxt;
	struct llist_node *node;

	node = llist_del_first(&rdma->sc_recv_ctxts);
	if (!node)
		goto out_empty;
	ctxt = llist_entry(node, struct svc_rdma_recv_ctxt, rc_node);

out:
	ctxt->rc_page_count = 0;
	return ctxt;

out_empty:
	ctxt = svc_rdma_recv_ctxt_alloc(rdma);
	if (!ctxt)
		return NULL;
	goto out;
}

/**
 * svc_rdma_recv_ctxt_put - Return recv_ctxt to free list
 * @rdma: controlling svcxprt_rdma
 * @ctxt: object to return to the free list
 *
 */
void svc_rdma_recv_ctxt_put(struct svcxprt_rdma *rdma,
			    struct svc_rdma_recv_ctxt *ctxt)
{
	pcl_free(&ctxt->rc_call_pcl);
	pcl_free(&ctxt->rc_read_pcl);
	pcl_free(&ctxt->rc_write_pcl);
	pcl_free(&ctxt->rc_reply_pcl);

	if (!ctxt->rc_temp)
		llist_add(&ctxt->rc_node, &rdma->sc_recv_ctxts);
	else
		svc_rdma_recv_ctxt_destroy(rdma, ctxt);
}

/**
 * svc_rdma_release_rqst - Release transport-specific per-rqst resources
 * @rqstp: svc_rqst being released
 *
 * Ensure that the recv_ctxt is released whether or not a Reply
 * was sent. For example, the client could close the connection,
 * or svc_process could drop an RPC, before the Reply is sent.
 */
void svc_rdma_release_rqst(struct svc_rqst *rqstp)
{
	struct svc_rdma_recv_ctxt *ctxt = rqstp->rq_xprt_ctxt;
	struct svc_xprt *xprt = rqstp->rq_xprt;
	struct svcxprt_rdma *rdma =
		container_of(xprt, struct svcxprt_rdma, sc_xprt);

	rqstp->rq_xprt_ctxt = NULL;
	if (ctxt)
		svc_rdma_recv_ctxt_put(rdma, ctxt);
}

static bool svc_rdma_refresh_recvs(struct svcxprt_rdma *rdma,
				   unsigned int wanted, bool temp)
{
	const struct ib_recv_wr *bad_wr = NULL;
	struct svc_rdma_recv_ctxt *ctxt;
	struct ib_recv_wr *recv_chain;
	int ret;

	if (test_bit(XPT_CLOSE, &rdma->sc_xprt.xpt_flags))
		return false;

	recv_chain = NULL;
	while (wanted--) {
		ctxt = svc_rdma_recv_ctxt_get(rdma);
		if (!ctxt)
			break;

		trace_svcrdma_post_recv(ctxt);
		ctxt->rc_temp = temp;
		ctxt->rc_recv_wr.next = recv_chain;
		recv_chain = &ctxt->rc_recv_wr;
		rdma->sc_pending_recvs++;
	}
	if (!recv_chain)
		return false;

	ret = ib_post_recv(rdma->sc_qp, recv_chain, &bad_wr);
	if (ret)
		goto err_free;
	return true;

err_free:
	trace_svcrdma_rq_post_err(rdma, ret);
	while (bad_wr) {
		ctxt = container_of(bad_wr, struct svc_rdma_recv_ctxt,
				    rc_recv_wr);
		bad_wr = bad_wr->next;
		svc_rdma_recv_ctxt_put(rdma, ctxt);
	}
	/* Since we're destroying the xprt, no need to reset
	 * sc_pending_recvs. */
	return false;
}

/**
 * svc_rdma_post_recvs - Post initial set of Recv WRs
 * @rdma: fresh svcxprt_rdma
 *
 * Returns true if successful, otherwise false.
 */
bool svc_rdma_post_recvs(struct svcxprt_rdma *rdma)
{
	return svc_rdma_refresh_recvs(rdma, rdma->sc_max_requests, true);
}

/**
 * svc_rdma_wc_receive - Invoked by RDMA provider for each polled Receive WC
 * @cq: Completion Queue context
 * @wc: Work Completion object
 *
 */
static void svc_rdma_wc_receive(struct ib_cq *cq, struct ib_wc *wc)
{
	struct svcxprt_rdma *rdma = cq->cq_context;
	struct ib_cqe *cqe = wc->wr_cqe;
	struct svc_rdma_recv_ctxt *ctxt;

	rdma->sc_pending_recvs--;

	/* WARNING: Only wc->wr_cqe and wc->status are reliable */
	ctxt = container_of(cqe, struct svc_rdma_recv_ctxt, rc_cqe);

	trace_svcrdma_wc_receive(wc, &ctxt->rc_cid);
	if (wc->status != IB_WC_SUCCESS)
		goto flushed;

	/* If receive posting fails, the connection is about to be
	 * lost anyway. The server will not be able to send a reply
	 * for this RPC, and the client will retransmit this RPC
	 * anyway when it reconnects.
	 *
	 * Therefore we drop the Receive, even if status was SUCCESS
	 * to reduce the likelihood of replayed requests once the
	 * client reconnects.
	 */
	if (rdma->sc_pending_recvs < rdma->sc_max_requests)
		if (!svc_rdma_refresh_recvs(rdma, rdma->sc_recv_batch, false))
			goto flushed;

	/* All wc fields are now known to be valid */
	ctxt->rc_byte_len = wc->byte_len;

	spin_lock(&rdma->sc_rq_dto_lock);
	list_add_tail(&ctxt->rc_list, &rdma->sc_rq_dto_q);
	/* Note the unlock pairs with the smp_rmb in svc_xprt_ready: */
	set_bit(XPT_DATA, &rdma->sc_xprt.xpt_flags);
	spin_unlock(&rdma->sc_rq_dto_lock);
	if (!test_bit(RDMAXPRT_CONN_PENDING, &rdma->sc_flags))
		svc_xprt_enqueue(&rdma->sc_xprt);
	return;

flushed:
	svc_rdma_recv_ctxt_put(rdma, ctxt);
	svc_xprt_deferred_close(&rdma->sc_xprt);
}

/**
 * svc_rdma_flush_recv_queues - Drain pending Receive work
 * @rdma: svcxprt_rdma being shut down
 *
 */
void svc_rdma_flush_recv_queues(struct svcxprt_rdma *rdma)
{
	struct svc_rdma_recv_ctxt *ctxt;

	while ((ctxt = svc_rdma_next_recv_ctxt(&rdma->sc_rq_dto_q))) {
		list_del(&ctxt->rc_list);
		svc_rdma_recv_ctxt_put(rdma, ctxt);
	}
}

static void svc_rdma_build_arg_xdr(struct svc_rqst *rqstp,
				   struct svc_rdma_recv_ctxt *ctxt)
{
	struct xdr_buf *arg = &rqstp->rq_arg;

	arg->head[0].iov_base = ctxt->rc_recv_buf;
	arg->head[0].iov_len = ctxt->rc_byte_len;
	arg->tail[0].iov_base = NULL;
	arg->tail[0].iov_len = 0;
	arg->page_len = 0;
	arg->page_base = 0;
	arg->buflen = ctxt->rc_byte_len;
	arg->len = ctxt->rc_byte_len;
}

/**
 * xdr_count_read_segments - Count number of Read segments in Read list
 * @rctxt: Ingress receive context
 * @p: Start of an un-decoded Read list
 *
 * Before allocating anything, ensure the ingress Read list is safe
 * to use.
 *
 * The segment count is limited to how many segments can fit in the
 * transport header without overflowing the buffer. That's about 40
 * Read segments for a 1KB inline threshold.
 *
 * Return values:
 *   %true: Read list is valid. @rctxt's xdr_stream is updated to point
 *	    to the first byte past the Read list. rc_read_pcl and
 *	    rc_call_pcl cl_count fields are set to the number of
 *	    Read segments in the list.
 *  %false: Read list is corrupt. @rctxt's xdr_stream is left in an
 *	    unknown state.
 */
static bool xdr_count_read_segments(struct svc_rdma_recv_ctxt *rctxt, __be32 *p)
{
	rctxt->rc_call_pcl.cl_count = 0;
	rctxt->rc_read_pcl.cl_count = 0;
	while (xdr_item_is_present(p)) {
		u32 position, handle, length;
		u64 offset;

		p = xdr_inline_decode(&rctxt->rc_stream,
				      rpcrdma_readseg_maxsz * sizeof(*p));
		if (!p)
			return false;

		xdr_decode_read_segment(p, &position, &handle,
					    &length, &offset);
		if (position) {
			if (position & 3)
				return false;
			++rctxt->rc_read_pcl.cl_count;
		} else {
			++rctxt->rc_call_pcl.cl_count;
		}

		p = xdr_inline_decode(&rctxt->rc_stream, sizeof(*p));
		if (!p)
			return false;
	}
	return true;
}

/* Sanity check the Read list.
 *
 * Sanity checks:
 * - Read list does not overflow Receive buffer.
 * - Chunk size limited by largest NFS data payload.
 *
 * Return values:
 *   %true: Read list is valid. @rctxt's xdr_stream is updated
 *	    to point to the first byte past the Read list.
 *  %false: Read list is corrupt. @rctxt's xdr_stream is left
 *	    in an unknown state.
 */
static bool xdr_check_read_list(struct svc_rdma_recv_ctxt *rctxt)
{
	__be32 *p;

	p = xdr_inline_decode(&rctxt->rc_stream, sizeof(*p));
	if (!p)
		return false;
	if (!xdr_count_read_segments(rctxt, p))
		return false;
	if (!pcl_alloc_call(rctxt, p))
		return false;
	return pcl_alloc_read(rctxt, p);
}

static bool xdr_check_write_chunk(struct svc_rdma_recv_ctxt *rctxt)
{
	u32 segcount;
	__be32 *p;

	if (xdr_stream_decode_u32(&rctxt->rc_stream, &segcount))
		return false;

	/* A bogus segcount causes this buffer overflow check to fail. */
	p = xdr_inline_decode(&rctxt->rc_stream,
			      segcount * rpcrdma_segment_maxsz * sizeof(*p));
	return p != NULL;
}

/**
 * xdr_count_write_chunks - Count number of Write chunks in Write list
 * @rctxt: Received header and decoding state
 * @p: start of an un-decoded Write list
 *
 * Before allocating anything, ensure the ingress Write list is
 * safe to use.
 *
 * Return values:
 *       %true: Write list is valid. @rctxt's xdr_stream is updated
 *		to point to the first byte past the Write list, and
 *		the number of Write chunks is in rc_write_pcl.cl_count.
 *      %false: Write list is corrupt. @rctxt's xdr_stream is left
 *		in an indeterminate state.
 */
static bool xdr_count_write_chunks(struct svc_rdma_recv_ctxt *rctxt, __be32 *p)
{
	rctxt->rc_write_pcl.cl_count = 0;
	while (xdr_item_is_present(p)) {
		if (!xdr_check_write_chunk(rctxt))
			return false;
		++rctxt->rc_write_pcl.cl_count;
		p = xdr_inline_decode(&rctxt->rc_stream, sizeof(*p));
		if (!p)
			return false;
	}
	return true;
}

/* Sanity check the Write list.
 *
 * Implementation limits:
 * - This implementation currently supports only one Write chunk.
 *
 * Sanity checks:
 * - Write list does not overflow Receive buffer.
 * - Chunk size limited by largest NFS data payload.
 *
 * Return values:
 *       %true: Write list is valid. @rctxt's xdr_stream is updated
 *		to point to the first byte past the Write list.
 *      %false: Write list is corrupt. @rctxt's xdr_stream is left
 *		in an unknown state.
 */
static bool xdr_check_write_list(struct svc_rdma_recv_ctxt *rctxt)
{
	__be32 *p;

	p = xdr_inline_decode(&rctxt->rc_stream, sizeof(*p));
	if (!p)
		return false;
	if (!xdr_count_write_chunks(rctxt, p))
		return false;
	if (!pcl_alloc_write(rctxt, &rctxt->rc_write_pcl, p))
		return false;

	rctxt->rc_cur_result_payload = pcl_first_chunk(&rctxt->rc_write_pcl);
	return true;
}

/* Sanity check the Reply chunk.
 *
 * Sanity checks:
 * - Reply chunk does not overflow Receive buffer.
 * - Chunk size limited by largest NFS data payload.
 *
 * Return values:
 *       %true: Reply chunk is valid. @rctxt's xdr_stream is updated
 *		to point to the first byte past the Reply chunk.
 *      %false: Reply chunk is corrupt. @rctxt's xdr_stream is left
 *		in an unknown state.
 */
static bool xdr_check_reply_chunk(struct svc_rdma_recv_ctxt *rctxt)
{
	__be32 *p;

	p = xdr_inline_decode(&rctxt->rc_stream, sizeof(*p));
	if (!p)
		return false;

	if (!xdr_item_is_present(p))
		return true;
	if (!xdr_check_write_chunk(rctxt))
		return false;

	rctxt->rc_reply_pcl.cl_count = 1;
	return pcl_alloc_write(rctxt, &rctxt->rc_reply_pcl, p);
}

/* RPC-over-RDMA Version One private extension: Remote Invalidation.
 * Responder's choice: requester signals it can handle Send With
 * Invalidate, and responder chooses one R_key to invalidate.
 *
 * If there is exactly one distinct R_key in the received transport
 * header, set rc_inv_rkey to that R_key. Otherwise, set it to zero.
 */
static void svc_rdma_get_inv_rkey(struct svcxprt_rdma *rdma,
				  struct svc_rdma_recv_ctxt *ctxt)
{
	struct svc_rdma_segment *segment;
	struct svc_rdma_chunk *chunk;
	u32 inv_rkey;

	ctxt->rc_inv_rkey = 0;

	if (!rdma->sc_snd_w_inv)
		return;

	inv_rkey = 0;
	pcl_for_each_chunk(chunk, &ctxt->rc_call_pcl) {
		pcl_for_each_segment(segment, chunk) {
			if (inv_rkey == 0)
				inv_rkey = segment->rs_handle;
			else if (inv_rkey != segment->rs_handle)
				return;
		}
	}
	pcl_for_each_chunk(chunk, &ctxt->rc_read_pcl) {
		pcl_for_each_segment(segment, chunk) {
			if (inv_rkey == 0)
				inv_rkey = segment->rs_handle;
			else if (inv_rkey != segment->rs_handle)
				return;
		}
	}
	pcl_for_each_chunk(chunk, &ctxt->rc_write_pcl) {
		pcl_for_each_segment(segment, chunk) {
			if (inv_rkey == 0)
				inv_rkey = segment->rs_handle;
			else if (inv_rkey != segment->rs_handle)
				return;
		}
	}
	pcl_for_each_chunk(chunk, &ctxt->rc_reply_pcl) {
		pcl_for_each_segment(segment, chunk) {
			if (inv_rkey == 0)
				inv_rkey = segment->rs_handle;
			else if (inv_rkey != segment->rs_handle)
				return;
		}
	}
	ctxt->rc_inv_rkey = inv_rkey;
}

/**
 * svc_rdma_xdr_decode_req - Decode the transport header
 * @rq_arg: xdr_buf containing ingress RPC/RDMA message
 * @rctxt: state of decoding
 *
 * On entry, xdr->head[0].iov_base points to first byte of the
 * RPC-over-RDMA transport header.
 *
 * On successful exit, head[0] points to first byte past the
 * RPC-over-RDMA header. For RDMA_MSG, this is the RPC message.
 *
 * The length of the RPC-over-RDMA header is returned.
 *
 * Assumptions:
 * - The transport header is entirely contained in the head iovec.
 */
static int svc_rdma_xdr_decode_req(struct xdr_buf *rq_arg,
				   struct svc_rdma_recv_ctxt *rctxt)
{
	__be32 *p, *rdma_argp;
	unsigned int hdr_len;

	rdma_argp = rq_arg->head[0].iov_base;
	xdr_init_decode(&rctxt->rc_stream, rq_arg, rdma_argp, NULL);

	p = xdr_inline_decode(&rctxt->rc_stream,
			      rpcrdma_fixed_maxsz * sizeof(*p));
	if (unlikely(!p))
		goto out_short;
	p++;
	if (*p != rpcrdma_version)
		goto out_version;
	p += 2;
	rctxt->rc_msgtype = *p;
	switch (rctxt->rc_msgtype) {
	case rdma_msg:
		break;
	case rdma_nomsg:
		break;
	case rdma_done:
		goto out_drop;
	case rdma_error:
		goto out_drop;
	default:
		goto out_proc;
	}

	if (!xdr_check_read_list(rctxt))
		goto out_inval;
	if (!xdr_check_write_list(rctxt))
		goto out_inval;
	if (!xdr_check_reply_chunk(rctxt))
		goto out_inval;

	rq_arg->head[0].iov_base = rctxt->rc_stream.p;
	hdr_len = xdr_stream_pos(&rctxt->rc_stream);
	rq_arg->head[0].iov_len -= hdr_len;
	rq_arg->len -= hdr_len;
	trace_svcrdma_decode_rqst(rctxt, rdma_argp, hdr_len);
	return hdr_len;

out_short:
	trace_svcrdma_decode_short_err(rctxt, rq_arg->len);
	return -EINVAL;

out_version:
	trace_svcrdma_decode_badvers_err(rctxt, rdma_argp);
	return -EPROTONOSUPPORT;

out_drop:
	trace_svcrdma_decode_drop_err(rctxt, rdma_argp);
	return 0;

out_proc:
	trace_svcrdma_decode_badproc_err(rctxt, rdma_argp);
	return -EINVAL;

out_inval:
	trace_svcrdma_decode_parse_err(rctxt, rdma_argp);
	return -EINVAL;
}

static void svc_rdma_send_error(struct svcxprt_rdma *rdma,
				struct svc_rdma_recv_ctxt *rctxt,
				int status)
{
	struct svc_rdma_send_ctxt *sctxt;

	sctxt = svc_rdma_send_ctxt_get(rdma);
	if (!sctxt)
		return;
	svc_rdma_send_error_msg(rdma, sctxt, rctxt, status);
}

/* By convention, backchannel calls arrive via rdma_msg type
 * messages, and never populate the chunk lists. This makes
 * the RPC/RDMA header small and fixed in size, so it is
 * straightforward to check the RPC header's direction field.
 */
static bool svc_rdma_is_reverse_direction_reply(struct svc_xprt *xprt,
						struct svc_rdma_recv_ctxt *rctxt)
{
	__be32 *p = rctxt->rc_recv_buf;

	if (!xprt->xpt_bc_xprt)
		return false;

	if (rctxt->rc_msgtype != rdma_msg)
		return false;

	if (!pcl_is_empty(&rctxt->rc_call_pcl))
		return false;
	if (!pcl_is_empty(&rctxt->rc_read_pcl))
		return false;
	if (!pcl_is_empty(&rctxt->rc_write_pcl))
		return false;
	if (!pcl_is_empty(&rctxt->rc_reply_pcl))
		return false;

	/* RPC call direction */
	if (*(p + 8) == cpu_to_be32(RPC_CALL))
		return false;

	return true;
}

/**
 * svc_rdma_recvfrom - Receive an RPC call
 * @rqstp: request structure into which to receive an RPC Call
 *
 * Returns:
 *	The positive number of bytes in the RPC Call message,
 *	%0 if there were no Calls ready to return,
 *	%-EINVAL if the Read chunk data is too large,
 *	%-ENOMEM if rdma_rw context pool was exhausted,
 *	%-ENOTCONN if posting failed (connection is lost),
 *	%-EIO if rdma_rw initialization failed (DMA mapping, etc).
 *
 * Called in a loop when XPT_DATA is set. XPT_DATA is cleared only
 * when there are no remaining ctxt's to process.
 *
 * The next ctxt is removed from the "receive" lists.
 *
 * - If the ctxt completes a Read, then finish assembling the Call
 *   message and return the number of bytes in the message.
 *
 * - If the ctxt completes a Receive, then construct the Call
 *   message from the contents of the Receive buffer.
 *
 *   - If there are no Read chunks in this message, then finish
 *     assembling the Call message and return the number of bytes
 *     in the message.
 *
 *   - If there are Read chunks in this message, post Read WRs to
 *     pull that payload and return 0.
 */
int svc_rdma_recvfrom(struct svc_rqst *rqstp)
{
	struct svc_xprt *xprt = rqstp->rq_xprt;
	struct svcxprt_rdma *rdma_xprt =
		container_of(xprt, struct svcxprt_rdma, sc_xprt);
	struct svc_rdma_recv_ctxt *ctxt;
	int ret;

	rqstp->rq_xprt_ctxt = NULL;

	spin_lock(&rdma_xprt->sc_rq_dto_lock);
	ctxt = svc_rdma_next_recv_ctxt(&rdma_xprt->sc_rq_dto_q);
	if (!ctxt) {
		/* No new incoming requests, terminate the loop */
		clear_bit(XPT_DATA, &xprt->xpt_flags);
		spin_unlock(&rdma_xprt->sc_rq_dto_lock);
		svc_xprt_received(xprt);
		return 0;
	}
	list_del(&ctxt->rc_list);
	spin_unlock(&rdma_xprt->sc_rq_dto_lock);
	percpu_counter_inc(&svcrdma_stat_recv);

	/* Unblock the transport for the next receive */
	svc_xprt_received(xprt);

	ib_dma_sync_single_for_cpu(rdma_xprt->sc_pd->device,
				   ctxt->rc_recv_sge.addr, ctxt->rc_byte_len,
				   DMA_FROM_DEVICE);
	svc_rdma_build_arg_xdr(rqstp, ctxt);

	/* Prevent svc_xprt_release from releasing pages in rq_pages
	 * if we return 0 or an error.
	 */
	rqstp->rq_respages = rqstp->rq_pages;
	rqstp->rq_next_page = rqstp->rq_respages;

	ret = svc_rdma_xdr_decode_req(&rqstp->rq_arg, ctxt);
	if (ret < 0)
		goto out_err;
	if (ret == 0)
		goto out_drop;
	rqstp->rq_xprt_hlen = ret;

	if (svc_rdma_is_reverse_direction_reply(xprt, ctxt))
		goto out_backchannel;

	svc_rdma_get_inv_rkey(rdma_xprt, ctxt);

	if (!pcl_is_empty(&ctxt->rc_read_pcl) ||
	    !pcl_is_empty(&ctxt->rc_call_pcl)) {
		ret = svc_rdma_process_read_list(rdma_xprt, rqstp, ctxt);
		if (ret < 0)
			goto out_readfail;
	}

	rqstp->rq_xprt_ctxt = ctxt;
	rqstp->rq_prot = IPPROTO_MAX;
	svc_xprt_copy_addrs(rqstp, xprt);
	return rqstp->rq_arg.len;

out_err:
	svc_rdma_send_error(rdma_xprt, ctxt, ret);
	svc_rdma_recv_ctxt_put(rdma_xprt, ctxt);
	return 0;

out_readfail:
	if (ret == -EINVAL)
		svc_rdma_send_error(rdma_xprt, ctxt, ret);
	svc_rdma_recv_ctxt_put(rdma_xprt, ctxt);
	return ret;

out_backchannel:
	svc_rdma_handle_bc_reply(rqstp, ctxt);
out_drop:
	svc_rdma_recv_ctxt_put(rdma_xprt, ctxt);
	return 0;
}