svc_rdma_recvfrom.c 17.8 KB
Newer Older
T
Tom Tucker 已提交
1
/*
S
Steve Wise 已提交
2
 * Copyright (c) 2014 Open Grid Computing, Inc. All rights reserved.
T
Tom Tucker 已提交
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
 * 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>
 */

#include <linux/sunrpc/debug.h>
#include <linux/sunrpc/rpc_rdma.h>
#include <linux/spinlock.h>
46
#include <linux/highmem.h>
T
Tom Tucker 已提交
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
#include <asm/unaligned.h>
#include <rdma/ib_verbs.h>
#include <rdma/rdma_cm.h>
#include <linux/sunrpc/svc_rdma.h>

#define RPCDBG_FACILITY	RPCDBG_SVCXPRT

/*
 * Replace the pages in the rq_argpages array with the pages from the SGE in
 * the RDMA_RECV completion. The SGL should contain full pages up until the
 * last one.
 */
static void rdma_build_arg_xdr(struct svc_rqst *rqstp,
			       struct svc_rdma_op_ctxt *ctxt,
			       u32 byte_count)
{
	struct page *page;
	u32 bc;
	int sge_no;

	/* Swap the page in the SGE with the page in argpages */
	page = ctxt->pages[0];
	put_page(rqstp->rq_pages[0]);
	rqstp->rq_pages[0] = page;

	/* Set up the XDR head */
	rqstp->rq_arg.head[0].iov_base = page_address(page);
S
Steve Wise 已提交
74 75
	rqstp->rq_arg.head[0].iov_len =
		min_t(size_t, byte_count, ctxt->sge[0].length);
T
Tom Tucker 已提交
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
	rqstp->rq_arg.len = byte_count;
	rqstp->rq_arg.buflen = byte_count;

	/* Compute bytes past head in the SGL */
	bc = byte_count - rqstp->rq_arg.head[0].iov_len;

	/* If data remains, store it in the pagelist */
	rqstp->rq_arg.page_len = bc;
	rqstp->rq_arg.page_base = 0;
	rqstp->rq_arg.pages = &rqstp->rq_pages[1];
	sge_no = 1;
	while (bc && sge_no < ctxt->count) {
		page = ctxt->pages[sge_no];
		put_page(rqstp->rq_pages[sge_no]);
		rqstp->rq_pages[sge_no] = page;
S
Steve Wise 已提交
91
		bc -= min_t(u32, bc, ctxt->sge[sge_no].length);
T
Tom Tucker 已提交
92 93 94 95
		rqstp->rq_arg.buflen += ctxt->sge[sge_no].length;
		sge_no++;
	}
	rqstp->rq_respages = &rqstp->rq_pages[sge_no];
T
Tom Tucker 已提交
96
	rqstp->rq_next_page = rqstp->rq_respages + 1;
T
Tom Tucker 已提交
97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118

	/* We should never run out of SGE because the limit is defined to
	 * support the max allowed RPC data length
	 */
	BUG_ON(bc && (sge_no == ctxt->count));
	BUG_ON((rqstp->rq_arg.head[0].iov_len + rqstp->rq_arg.page_len)
	       != byte_count);
	BUG_ON(rqstp->rq_arg.len != byte_count);

	/* If not all pages were used from the SGL, free the remaining ones */
	bc = sge_no;
	while (sge_no < ctxt->count) {
		page = ctxt->pages[sge_no++];
		put_page(page);
	}
	ctxt->count = bc;

	/* Set up tail */
	rqstp->rq_arg.tail[0].iov_base = NULL;
	rqstp->rq_arg.tail[0].iov_len = 0;
}

S
Steve Wise 已提交
119
static int rdma_read_max_sge(struct svcxprt_rdma *xprt, int sge_count)
T
Tom Tucker 已提交
120
{
S
Steve Wise 已提交
121 122 123 124 125 126
	if (rdma_node_get_transport(xprt->sc_cm_id->device->node_type) ==
	     RDMA_TRANSPORT_IWARP)
		return 1;
	else
		return min_t(int, sge_count, xprt->sc_max_sge);
}
T
Tom Tucker 已提交
127

S
Steve Wise 已提交
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
typedef int (*rdma_reader_fn)(struct svcxprt_rdma *xprt,
			      struct svc_rqst *rqstp,
			      struct svc_rdma_op_ctxt *head,
			      int *page_no,
			      u32 *page_offset,
			      u32 rs_handle,
			      u32 rs_length,
			      u64 rs_offset,
			      int last);

/* Issue an RDMA_READ using the local lkey to map the data sink */
static int rdma_read_chunk_lcl(struct svcxprt_rdma *xprt,
			       struct svc_rqst *rqstp,
			       struct svc_rdma_op_ctxt *head,
			       int *page_no,
			       u32 *page_offset,
			       u32 rs_handle,
			       u32 rs_length,
			       u64 rs_offset,
			       int last)
{
	struct ib_send_wr read_wr;
	int pages_needed = PAGE_ALIGN(*page_offset + rs_length) >> PAGE_SHIFT;
	struct svc_rdma_op_ctxt *ctxt = svc_rdma_get_context(xprt);
	int ret, read, pno;
	u32 pg_off = *page_offset;
	u32 pg_no = *page_no;

	ctxt->direction = DMA_FROM_DEVICE;
	ctxt->read_hdr = head;
	pages_needed =
		min_t(int, pages_needed, rdma_read_max_sge(xprt, pages_needed));
	read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);

	for (pno = 0; pno < pages_needed; pno++) {
		int len = min_t(int, rs_length, PAGE_SIZE - pg_off);

		head->arg.pages[pg_no] = rqstp->rq_arg.pages[pg_no];
		head->arg.page_len += len;
		head->arg.len += len;
		if (!pg_off)
			head->count++;
		rqstp->rq_respages = &rqstp->rq_arg.pages[pg_no+1];
T
Tom Tucker 已提交
171
		rqstp->rq_next_page = rqstp->rq_respages + 1;
S
Steve Wise 已提交
172 173 174 175 176 177 178 179 180 181
		ctxt->sge[pno].addr =
			ib_dma_map_page(xprt->sc_cm_id->device,
					head->arg.pages[pg_no], pg_off,
					PAGE_SIZE - pg_off,
					DMA_FROM_DEVICE);
		ret = ib_dma_mapping_error(xprt->sc_cm_id->device,
					   ctxt->sge[pno].addr);
		if (ret)
			goto err;
		atomic_inc(&xprt->sc_dma_used);
T
Tom Tucker 已提交
182

S
Steve Wise 已提交
183 184 185 186 187 188 189 190 191 192
		/* The lkey here is either a local dma lkey or a dma_mr lkey */
		ctxt->sge[pno].lkey = xprt->sc_dma_lkey;
		ctxt->sge[pno].length = len;
		ctxt->count++;

		/* adjust offset and wrap to next page if needed */
		pg_off += len;
		if (pg_off == PAGE_SIZE) {
			pg_off = 0;
			pg_no++;
T
Tom Tucker 已提交
193
		}
S
Steve Wise 已提交
194
		rs_length -= len;
T
Tom Tucker 已提交
195
	}
S
Steve Wise 已提交
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

	if (last && rs_length == 0)
		set_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);
	else
		clear_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);

	memset(&read_wr, 0, sizeof(read_wr));
	read_wr.wr_id = (unsigned long)ctxt;
	read_wr.opcode = IB_WR_RDMA_READ;
	ctxt->wr_op = read_wr.opcode;
	read_wr.send_flags = IB_SEND_SIGNALED;
	read_wr.wr.rdma.rkey = rs_handle;
	read_wr.wr.rdma.remote_addr = rs_offset;
	read_wr.sg_list = ctxt->sge;
	read_wr.num_sge = pages_needed;

	ret = svc_rdma_send(xprt, &read_wr);
	if (ret) {
		pr_err("svcrdma: Error %d posting RDMA_READ\n", ret);
		set_bit(XPT_CLOSE, &xprt->sc_xprt.xpt_flags);
		goto err;
	}

	/* return current location in page array */
	*page_no = pg_no;
	*page_offset = pg_off;
	ret = read;
	atomic_inc(&rdma_stat_read);
	return ret;
 err:
	svc_rdma_unmap_dma(ctxt);
	svc_rdma_put_context(ctxt, 0);
	return ret;
T
Tom Tucker 已提交
229 230
}

S
Steve Wise 已提交
231 232
/* Issue an RDMA_READ using an FRMR to map the data sink */
static int rdma_read_chunk_frmr(struct svcxprt_rdma *xprt,
233 234
				struct svc_rqst *rqstp,
				struct svc_rdma_op_ctxt *head,
S
Steve Wise 已提交
235 236 237 238 239 240
				int *page_no,
				u32 *page_offset,
				u32 rs_handle,
				u32 rs_length,
				u64 rs_offset,
				int last)
241
{
S
Steve Wise 已提交
242 243 244 245 246 247 248 249 250 251
	struct ib_send_wr read_wr;
	struct ib_send_wr inv_wr;
	struct ib_send_wr fastreg_wr;
	u8 key;
	int pages_needed = PAGE_ALIGN(*page_offset + rs_length) >> PAGE_SHIFT;
	struct svc_rdma_op_ctxt *ctxt = svc_rdma_get_context(xprt);
	struct svc_rdma_fastreg_mr *frmr = svc_rdma_get_frmr(xprt);
	int ret, read, pno;
	u32 pg_off = *page_offset;
	u32 pg_no = *page_no;
252 253 254 255

	if (IS_ERR(frmr))
		return -ENOMEM;

S
Steve Wise 已提交
256 257 258 259
	ctxt->direction = DMA_FROM_DEVICE;
	ctxt->frmr = frmr;
	pages_needed = min_t(int, pages_needed, xprt->sc_frmr_pg_list_len);
	read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);
260

S
Steve Wise 已提交
261
	frmr->kva = page_address(rqstp->rq_arg.pages[pg_no]);
262 263
	frmr->direction = DMA_FROM_DEVICE;
	frmr->access_flags = (IB_ACCESS_LOCAL_WRITE|IB_ACCESS_REMOTE_WRITE);
S
Steve Wise 已提交
264 265 266 267 268 269 270 271 272 273 274 275 276 277
	frmr->map_len = pages_needed << PAGE_SHIFT;
	frmr->page_list_len = pages_needed;

	for (pno = 0; pno < pages_needed; pno++) {
		int len = min_t(int, rs_length, PAGE_SIZE - pg_off);

		head->arg.pages[pg_no] = rqstp->rq_arg.pages[pg_no];
		head->arg.page_len += len;
		head->arg.len += len;
		if (!pg_off)
			head->count++;
		rqstp->rq_respages = &rqstp->rq_arg.pages[pg_no+1];
		rqstp->rq_next_page = rqstp->rq_respages + 1;
		frmr->page_list->page_list[pno] =
278
			ib_dma_map_page(xprt->sc_cm_id->device,
S
Steve Wise 已提交
279
					head->arg.pages[pg_no], 0,
280
					PAGE_SIZE, DMA_FROM_DEVICE);
S
Steve Wise 已提交
281 282 283 284
		ret = ib_dma_mapping_error(xprt->sc_cm_id->device,
					   frmr->page_list->page_list[pno]);
		if (ret)
			goto err;
285 286
		atomic_inc(&xprt->sc_dma_used);

S
Steve Wise 已提交
287 288 289 290 291 292 293
		/* adjust offset and wrap to next page if needed */
		pg_off += len;
		if (pg_off == PAGE_SIZE) {
			pg_off = 0;
			pg_no++;
		}
		rs_length -= len;
294 295
	}

S
Steve Wise 已提交
296 297 298 299
	if (last && rs_length == 0)
		set_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);
	else
		clear_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);
T
Tom Tucker 已提交
300

S
Steve Wise 已提交
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
	/* Bump the key */
	key = (u8)(frmr->mr->lkey & 0x000000FF);
	ib_update_fast_reg_key(frmr->mr, ++key);

	ctxt->sge[0].addr = (unsigned long)frmr->kva + *page_offset;
	ctxt->sge[0].lkey = frmr->mr->lkey;
	ctxt->sge[0].length = read;
	ctxt->count = 1;
	ctxt->read_hdr = head;

	/* Prepare FASTREG WR */
	memset(&fastreg_wr, 0, sizeof(fastreg_wr));
	fastreg_wr.opcode = IB_WR_FAST_REG_MR;
	fastreg_wr.send_flags = IB_SEND_SIGNALED;
	fastreg_wr.wr.fast_reg.iova_start = (unsigned long)frmr->kva;
	fastreg_wr.wr.fast_reg.page_list = frmr->page_list;
	fastreg_wr.wr.fast_reg.page_list_len = frmr->page_list_len;
	fastreg_wr.wr.fast_reg.page_shift = PAGE_SHIFT;
	fastreg_wr.wr.fast_reg.length = frmr->map_len;
	fastreg_wr.wr.fast_reg.access_flags = frmr->access_flags;
	fastreg_wr.wr.fast_reg.rkey = frmr->mr->lkey;
	fastreg_wr.next = &read_wr;

	/* Prepare RDMA_READ */
	memset(&read_wr, 0, sizeof(read_wr));
	read_wr.send_flags = IB_SEND_SIGNALED;
	read_wr.wr.rdma.rkey = rs_handle;
	read_wr.wr.rdma.remote_addr = rs_offset;
	read_wr.sg_list = ctxt->sge;
	read_wr.num_sge = 1;
	if (xprt->sc_dev_caps & SVCRDMA_DEVCAP_READ_W_INV) {
		read_wr.opcode = IB_WR_RDMA_READ_WITH_INV;
		read_wr.wr_id = (unsigned long)ctxt;
		read_wr.ex.invalidate_rkey = ctxt->frmr->mr->lkey;
	} else {
		read_wr.opcode = IB_WR_RDMA_READ;
		read_wr.next = &inv_wr;
		/* Prepare invalidate */
		memset(&inv_wr, 0, sizeof(inv_wr));
		inv_wr.wr_id = (unsigned long)ctxt;
		inv_wr.opcode = IB_WR_LOCAL_INV;
342
		inv_wr.send_flags = IB_SEND_SIGNALED | IB_SEND_FENCE;
S
Steve Wise 已提交
343 344 345 346 347 348 349 350 351 352
		inv_wr.ex.invalidate_rkey = frmr->mr->lkey;
	}
	ctxt->wr_op = read_wr.opcode;

	/* Post the chain */
	ret = svc_rdma_send(xprt, &fastreg_wr);
	if (ret) {
		pr_err("svcrdma: Error %d posting RDMA_READ\n", ret);
		set_bit(XPT_CLOSE, &xprt->sc_xprt.xpt_flags);
		goto err;
T
Tom Tucker 已提交
353 354
	}

S
Steve Wise 已提交
355 356 357 358 359 360 361 362 363 364 365
	/* return current location in page array */
	*page_no = pg_no;
	*page_offset = pg_off;
	ret = read;
	atomic_inc(&rdma_stat_read);
	return ret;
 err:
	svc_rdma_unmap_dma(ctxt);
	svc_rdma_put_context(ctxt, 0);
	svc_rdma_put_frmr(xprt, frmr);
	return ret;
T
Tom Tucker 已提交
366 367
}

S
Steve Wise 已提交
368 369 370 371
static int rdma_read_chunks(struct svcxprt_rdma *xprt,
			    struct rpcrdma_msg *rmsgp,
			    struct svc_rqst *rqstp,
			    struct svc_rdma_op_ctxt *head)
T
Tom Tucker 已提交
372
{
S
Steve Wise 已提交
373
	int page_no, ch_count, ret;
T
Tom Tucker 已提交
374
	struct rpcrdma_read_chunk *ch;
S
Steve Wise 已提交
375 376 377
	u32 page_offset, byte_count;
	u64 rs_offset;
	rdma_reader_fn reader;
T
Tom Tucker 已提交
378 379 380 381 382 383 384

	/* If no read list is present, return 0 */
	ch = svc_rdma_get_read_chunk(rmsgp);
	if (!ch)
		return 0;

	svc_rdma_rcl_chunk_counts(ch, &ch_count, &byte_count);
385 386
	if (ch_count > RPCSVC_MAXPAGES)
		return -EINVAL;
387

S
Steve Wise 已提交
388 389 390 391 392 393 394 395 396 397 398 399
	/* The request is completed when the RDMA_READs complete. The
	 * head context keeps all the pages that comprise the
	 * request.
	 */
	head->arg.head[0] = rqstp->rq_arg.head[0];
	head->arg.tail[0] = rqstp->rq_arg.tail[0];
	head->arg.pages = &head->pages[head->count];
	head->hdr_count = head->count;
	head->arg.page_base = 0;
	head->arg.page_len = 0;
	head->arg.len = rqstp->rq_arg.len;
	head->arg.buflen = rqstp->rq_arg.buflen;
400

S
Steve Wise 已提交
401 402 403
	/* Use FRMR if supported */
	if (xprt->sc_dev_caps & SVCRDMA_DEVCAP_FAST_REG)
		reader = rdma_read_chunk_frmr;
404
	else
S
Steve Wise 已提交
405
		reader = rdma_read_chunk_lcl;
T
Tom Tucker 已提交
406

S
Steve Wise 已提交
407
	page_no = 0; page_offset = 0;
T
Tom Tucker 已提交
408
	for (ch = (struct rpcrdma_read_chunk *)&rmsgp->rm_body.rm_chunks[0];
S
Steve Wise 已提交
409
	     ch->rc_discrim != 0; ch++) {
T
Tom Tucker 已提交
410

411 412
		xdr_decode_hyper((__be32 *)&ch->rc_target.rs_offset,
				 &rs_offset);
S
Steve Wise 已提交
413 414 415 416 417 418 419 420 421 422 423 424 425 426
		byte_count = ntohl(ch->rc_target.rs_length);

		while (byte_count > 0) {
			ret = reader(xprt, rqstp, head,
				     &page_no, &page_offset,
				     ntohl(ch->rc_target.rs_handle),
				     byte_count, rs_offset,
				     ((ch+1)->rc_discrim == 0) /* last */
				     );
			if (ret < 0)
				goto err;
			byte_count -= ret;
			rs_offset += ret;
			head->arg.buflen += ret;
T
Tom Tucker 已提交
427 428
		}
	}
S
Steve Wise 已提交
429 430
	ret = 1;
 err:
T
Tom Tucker 已提交
431
	/* Detach arg pages. svc_recv will replenish them */
S
Steve Wise 已提交
432 433 434
	for (page_no = 0;
	     &rqstp->rq_pages[page_no] < rqstp->rq_respages; page_no++)
		rqstp->rq_pages[page_no] = NULL;
T
Tom Tucker 已提交
435

S
Steve Wise 已提交
436
	return ret;
T
Tom Tucker 已提交
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
/*
 * To avoid a separate RDMA READ just for a handful of zero bytes,
 * RFC 5666 section 3.7 allows the client to omit the XDR zero pad
 * in chunk lists.
 */
static void
rdma_fix_xdr_pad(struct xdr_buf *buf)
{
	unsigned int page_len = buf->page_len;
	unsigned int size = (XDR_QUADLEN(page_len) << 2) - page_len;
	unsigned int offset, pg_no;
	char *p;

	if (size == 0)
		return;

	pg_no = page_len >> PAGE_SHIFT;
	offset = page_len & ~PAGE_MASK;
	p = page_address(buf->pages[pg_no]);
	memset(p + offset, 0, size);

	buf->page_len += size;
	buf->buflen += size;
	buf->len += size;
}

T
Tom Tucker 已提交
465
static int rdma_read_complete(struct svc_rqst *rqstp,
466
			      struct svc_rdma_op_ctxt *head)
T
Tom Tucker 已提交
467 468 469 470 471 472 473 474 475 476 477 478
{
	int page_no;
	int ret;

	BUG_ON(!head);

	/* Copy RPC pages */
	for (page_no = 0; page_no < head->count; page_no++) {
		put_page(rqstp->rq_pages[page_no]);
		rqstp->rq_pages[page_no] = head->pages[page_no];
	}
	/* Point rq_arg.pages past header */
479
	rdma_fix_xdr_pad(&head->arg);
480
	rqstp->rq_arg.pages = &rqstp->rq_pages[head->hdr_count];
T
Tom Tucker 已提交
481 482 483 484 485
	rqstp->rq_arg.page_len = head->arg.page_len;
	rqstp->rq_arg.page_base = head->arg.page_base;

	/* rq_respages starts after the last arg page */
	rqstp->rq_respages = &rqstp->rq_arg.pages[page_no];
T
Tom Tucker 已提交
486
	rqstp->rq_next_page = rqstp->rq_respages + 1;
T
Tom Tucker 已提交
487 488 489 490 491 492 493

	/* Rebuild rq_arg head and tail. */
	rqstp->rq_arg.head[0] = head->arg.head[0];
	rqstp->rq_arg.tail[0] = head->arg.tail[0];
	rqstp->rq_arg.len = head->arg.len;
	rqstp->rq_arg.buflen = head->arg.buflen;

494 495 496
	/* Free the context */
	svc_rdma_put_context(head, 0);

T
Tom Tucker 已提交
497 498
	/* XXX: What should this be? */
	rqstp->rq_prot = IPPROTO_MAX;
499
	svc_xprt_copy_addrs(rqstp, rqstp->rq_xprt);
T
Tom Tucker 已提交
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

	ret = rqstp->rq_arg.head[0].iov_len
		+ rqstp->rq_arg.page_len
		+ rqstp->rq_arg.tail[0].iov_len;
	dprintk("svcrdma: deferred read ret=%d, rq_arg.len =%d, "
		"rq_arg.head[0].iov_base=%p, rq_arg.head[0].iov_len = %zd\n",
		ret, rqstp->rq_arg.len,	rqstp->rq_arg.head[0].iov_base,
		rqstp->rq_arg.head[0].iov_len);

	return ret;
}

/*
 * Set up the rqstp thread context to point to the RQ buffer. If
 * necessary, pull additional data from the client with an RDMA_READ
 * request.
 */
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_op_ctxt *ctxt = NULL;
	struct rpcrdma_msg *rmsgp;
	int ret = 0;
	int len;

	dprintk("svcrdma: rqstp=%p\n", rqstp);

529
	spin_lock_bh(&rdma_xprt->sc_rq_dto_lock);
T
Tom Tucker 已提交
530 531 532 533 534
	if (!list_empty(&rdma_xprt->sc_read_complete_q)) {
		ctxt = list_entry(rdma_xprt->sc_read_complete_q.next,
				  struct svc_rdma_op_ctxt,
				  dto_q);
		list_del_init(&ctxt->dto_q);
535
		spin_unlock_bh(&rdma_xprt->sc_rq_dto_lock);
T
Tom Tucker 已提交
536
		return rdma_read_complete(rqstp, ctxt);
S
Steve Wise 已提交
537
	} else if (!list_empty(&rdma_xprt->sc_rq_dto_q)) {
T
Tom Tucker 已提交
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
		ctxt = list_entry(rdma_xprt->sc_rq_dto_q.next,
				  struct svc_rdma_op_ctxt,
				  dto_q);
		list_del_init(&ctxt->dto_q);
	} else {
		atomic_inc(&rdma_stat_rq_starve);
		clear_bit(XPT_DATA, &xprt->xpt_flags);
		ctxt = NULL;
	}
	spin_unlock_bh(&rdma_xprt->sc_rq_dto_lock);
	if (!ctxt) {
		/* This is the EAGAIN path. The svc_recv routine will
		 * return -EAGAIN, the nfsd thread will go to call into
		 * svc_recv again and we shouldn't be on the active
		 * transport list
		 */
		if (test_bit(XPT_CLOSE, &xprt->xpt_flags))
			goto close_out;

		goto out;
	}
	dprintk("svcrdma: processing ctxt=%p on xprt=%p, rqstp=%p, status=%d\n",
		ctxt, rdma_xprt, rqstp, ctxt->wc_status);
	BUG_ON(ctxt->wc_status != IB_WC_SUCCESS);
	atomic_inc(&rdma_stat_recv);

	/* Build up the XDR from the receive buffers. */
	rdma_build_arg_xdr(rqstp, ctxt, ctxt->byte_len);

	/* Decode the RDMA header. */
	len = svc_rdma_xdr_decode_req(&rmsgp, rqstp);
	rqstp->rq_xprt_hlen = len;

	/* If the request is invalid, reply with an error */
	if (len < 0) {
		if (len == -ENOSYS)
574
			svc_rdma_send_error(rdma_xprt, rmsgp, ERR_VERS);
T
Tom Tucker 已提交
575 576 577
		goto close_out;
	}

578
	/* Read read-list data. */
S
Steve Wise 已提交
579
	ret = rdma_read_chunks(rdma_xprt, rmsgp, rqstp, ctxt);
580 581
	if (ret > 0) {
		/* read-list posted, defer until data received from client. */
582
		goto defer;
S
Steve Wise 已提交
583
	} else if (ret < 0) {
584 585 586 587
		/* Post of read-list failed, free context. */
		svc_rdma_put_context(ctxt, 1);
		return 0;
	}
T
Tom Tucker 已提交
588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603

	ret = rqstp->rq_arg.head[0].iov_len
		+ rqstp->rq_arg.page_len
		+ rqstp->rq_arg.tail[0].iov_len;
	svc_rdma_put_context(ctxt, 0);
 out:
	dprintk("svcrdma: ret = %d, rq_arg.len =%d, "
		"rq_arg.head[0].iov_base=%p, rq_arg.head[0].iov_len = %zd\n",
		ret, rqstp->rq_arg.len,
		rqstp->rq_arg.head[0].iov_base,
		rqstp->rq_arg.head[0].iov_len);
	rqstp->rq_prot = IPPROTO_MAX;
	svc_xprt_copy_addrs(rqstp, xprt);
	return ret;

 close_out:
604
	if (ctxt)
T
Tom Tucker 已提交
605 606 607 608 609 610 611
		svc_rdma_put_context(ctxt, 1);
	dprintk("svcrdma: transport %p is closing\n", xprt);
	/*
	 * Set the close bit and enqueue it. svc_recv will see the
	 * close bit and call svc_xprt_delete
	 */
	set_bit(XPT_CLOSE, &xprt->xpt_flags);
612
defer:
T
Tom Tucker 已提交
613 614
	return 0;
}