cm.c 70.4 KB
Newer Older
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
/*
 * Copyright (c) 2009-2010 Chelsio, 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
 * OpenIB.org BSD 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.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */
#include <linux/module.h>
#include <linux/list.h>
#include <linux/workqueue.h>
#include <linux/skbuff.h>
#include <linux/timer.h>
#include <linux/notifier.h>
#include <linux/inetdevice.h>
#include <linux/ip.h>
#include <linux/tcp.h>

#include <net/neighbour.h>
#include <net/netevent.h>
#include <net/route.h>

#include "iw_cxgb4.h"

static char *states[] = {
	"idle",
	"listen",
	"connecting",
	"mpa_wait_req",
	"mpa_req_sent",
	"mpa_req_rcvd",
	"mpa_rep_sent",
	"fpdu_mode",
	"aborting",
	"closing",
	"moribund",
	"dead",
	NULL,
};

S
Steve Wise 已提交
64
static int dack_mode = 1;
65
module_param(dack_mode, int, 0644);
S
Steve Wise 已提交
66
MODULE_PARM_DESC(dack_mode, "Delayed ack mode (default=1)");
67

68 69 70 71
int c4iw_max_read_depth = 8;
module_param(c4iw_max_read_depth, int, 0644);
MODULE_PARM_DESC(c4iw_max_read_depth, "Per-connection max ORD/IRD (default=8)");

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
static int enable_tcp_timestamps;
module_param(enable_tcp_timestamps, int, 0644);
MODULE_PARM_DESC(enable_tcp_timestamps, "Enable tcp timestamps (default=0)");

static int enable_tcp_sack;
module_param(enable_tcp_sack, int, 0644);
MODULE_PARM_DESC(enable_tcp_sack, "Enable tcp SACK (default=0)");

static int enable_tcp_window_scaling = 1;
module_param(enable_tcp_window_scaling, int, 0644);
MODULE_PARM_DESC(enable_tcp_window_scaling,
		 "Enable tcp window scaling (default=1)");

int c4iw_debug;
module_param(c4iw_debug, int, 0644);
MODULE_PARM_DESC(c4iw_debug, "Enable debug logging (default=0)");

static int peer2peer;
module_param(peer2peer, int, 0644);
MODULE_PARM_DESC(peer2peer, "Support peer2peer ULPs (default=0)");

static int p2p_type = FW_RI_INIT_P2PTYPE_READ_REQ;
module_param(p2p_type, int, 0644);
MODULE_PARM_DESC(p2p_type, "RDMAP opcode to use for the RTR message: "
			   "1=RDMA_READ 0=RDMA_WRITE (default 1)");

static int ep_timeout_secs = 60;
module_param(ep_timeout_secs, int, 0644);
MODULE_PARM_DESC(ep_timeout_secs, "CM Endpoint operation timeout "
				   "in seconds (default=60)");

static int mpa_rev = 1;
module_param(mpa_rev, int, 0644);
MODULE_PARM_DESC(mpa_rev, "MPA Revision, 0 supports amso1100, "
106 107
		"1 is RFC0544 spec compliant, 2 is IETF MPA Peer Connect Draft"
		" compliant (default=1)");
108 109 110 111 112 113 114 115 116 117 118 119 120

static int markers_enabled;
module_param(markers_enabled, int, 0644);
MODULE_PARM_DESC(markers_enabled, "Enable MPA MARKERS (default(0)=disabled)");

static int crc_enabled = 1;
module_param(crc_enabled, int, 0644);
MODULE_PARM_DESC(crc_enabled, "Enable MPA CRC (default(1)=enabled)");

static int rcv_win = 256 * 1024;
module_param(rcv_win, int, 0644);
MODULE_PARM_DESC(rcv_win, "TCP receive window in bytes (default=256KB)");

121
static int snd_win = 128 * 1024;
122
module_param(snd_win, int, 0644);
123
MODULE_PARM_DESC(snd_win, "TCP send window in bytes (default=128KB)");
124 125 126 127 128 129 130 131 132

static struct workqueue_struct *workq;

static struct sk_buff_head rxq;

static struct sk_buff *get_skb(struct sk_buff *skb, int len, gfp_t gfp);
static void ep_timeout(unsigned long arg);
static void connect_reply_upcall(struct c4iw_ep *ep, int status);

133 134 135
static LIST_HEAD(timeout_list);
static spinlock_t timeout_lock;

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
static void start_ep_timer(struct c4iw_ep *ep)
{
	PDBG("%s ep %p\n", __func__, ep);
	if (timer_pending(&ep->timer)) {
		PDBG("%s stopped / restarted timer ep %p\n", __func__, ep);
		del_timer_sync(&ep->timer);
	} else
		c4iw_get_ep(&ep->com);
	ep->timer.expires = jiffies + ep_timeout_secs * HZ;
	ep->timer.data = (unsigned long)ep;
	ep->timer.function = ep_timeout;
	add_timer(&ep->timer);
}

static void stop_ep_timer(struct c4iw_ep *ep)
{
	PDBG("%s ep %p\n", __func__, ep);
	if (!timer_pending(&ep->timer)) {
		printk(KERN_ERR "%s timer stopped when its not running! "
		       "ep %p state %u\n", __func__, ep, ep->com.state);
		WARN_ON(1);
		return;
	}
	del_timer_sync(&ep->timer);
	c4iw_put_ep(&ep->com);
}

static int c4iw_l2t_send(struct c4iw_rdev *rdev, struct sk_buff *skb,
		  struct l2t_entry *l2e)
{
	int	error = 0;

	if (c4iw_fatal_error(rdev)) {
		kfree_skb(skb);
		PDBG("%s - device in error state - dropping\n", __func__);
		return -EIO;
	}
	error = cxgb4_l2t_send(rdev->lldi.ports[0], skb, l2e);
	if (error < 0)
		kfree_skb(skb);
176
	return error < 0 ? error : 0;
177 178 179 180 181 182 183 184 185 186 187 188 189 190
}

int c4iw_ofld_send(struct c4iw_rdev *rdev, struct sk_buff *skb)
{
	int	error = 0;

	if (c4iw_fatal_error(rdev)) {
		kfree_skb(skb);
		PDBG("%s - device in error state - dropping\n", __func__);
		return -EIO;
	}
	error = cxgb4_ofld_send(rdev->lldi.ports[0], skb);
	if (error < 0)
		kfree_skb(skb);
191
	return error < 0 ? error : 0;
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
}

static void release_tid(struct c4iw_rdev *rdev, u32 hwtid, struct sk_buff *skb)
{
	struct cpl_tid_release *req;

	skb = get_skb(skb, sizeof *req, GFP_KERNEL);
	if (!skb)
		return;
	req = (struct cpl_tid_release *) skb_put(skb, sizeof(*req));
	INIT_TP_WR(req, hwtid);
	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_TID_RELEASE, hwtid));
	set_wr_txq(skb, CPL_PRIORITY_SETUP, 0);
	c4iw_ofld_send(rdev, skb);
	return;
}

static void set_emss(struct c4iw_ep *ep, u16 opt)
{
	ep->emss = ep->com.dev->rdev.lldi.mtus[GET_TCPOPT_MSS(opt)] - 40;
	ep->mss = ep->emss;
	if (GET_TCPOPT_TSTAMP(opt))
		ep->emss -= 12;
	if (ep->emss < 128)
		ep->emss = 128;
	PDBG("%s mss_idx %u mss %u emss=%u\n", __func__, GET_TCPOPT_MSS(opt),
	     ep->mss, ep->emss);
}

static enum c4iw_ep_state state_read(struct c4iw_ep_common *epc)
{
	enum c4iw_ep_state state;

225
	mutex_lock(&epc->mutex);
226
	state = epc->state;
227
	mutex_unlock(&epc->mutex);
228 229 230 231 232 233 234 235 236 237
	return state;
}

static void __state_set(struct c4iw_ep_common *epc, enum c4iw_ep_state new)
{
	epc->state = new;
}

static void state_set(struct c4iw_ep_common *epc, enum c4iw_ep_state new)
{
238
	mutex_lock(&epc->mutex);
239 240
	PDBG("%s - %s -> %s\n", __func__, states[epc->state], states[new]);
	__state_set(epc, new);
241
	mutex_unlock(&epc->mutex);
242 243 244 245 246 247 248 249 250 251
	return;
}

static void *alloc_ep(int size, gfp_t gfp)
{
	struct c4iw_ep_common *epc;

	epc = kzalloc(size, gfp);
	if (epc) {
		kref_init(&epc->kref);
252
		mutex_init(&epc->mutex);
253
		c4iw_init_wr_wait(&epc->wr_wait);
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
	}
	PDBG("%s alloc ep %p\n", __func__, epc);
	return epc;
}

void _c4iw_free_ep(struct kref *kref)
{
	struct c4iw_ep *ep;

	ep = container_of(kref, struct c4iw_ep, com.kref);
	PDBG("%s ep %p state %s\n", __func__, ep, states[state_read(&ep->com)]);
	if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) {
		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid);
		dst_release(ep->dst);
		cxgb4_l2t_release(ep->l2t);
	}
	kfree(ep);
}

static void release_ep_resources(struct c4iw_ep *ep)
{
	set_bit(RELEASE_RESOURCES, &ep->com.flags);
	c4iw_put_ep(&ep->com);
}

static int status2errno(int status)
{
	switch (status) {
	case CPL_ERR_NONE:
		return 0;
	case CPL_ERR_CONN_RESET:
		return -ECONNRESET;
	case CPL_ERR_ARP_MISS:
		return -EHOSTUNREACH;
	case CPL_ERR_CONN_TIMEDOUT:
		return -ETIMEDOUT;
	case CPL_ERR_TCAM_FULL:
		return -ENOMEM;
	case CPL_ERR_CONN_EXIST:
		return -EADDRINUSE;
	default:
		return -EIO;
	}
}

/*
 * Try and reuse skbs already allocated...
 */
static struct sk_buff *get_skb(struct sk_buff *skb, int len, gfp_t gfp)
{
	if (skb && !skb_is_nonlinear(skb) && !skb_cloned(skb)) {
		skb_trim(skb, 0);
		skb_get(skb);
		skb_reset_transport_header(skb);
	} else {
		skb = alloc_skb(len, gfp);
	}
	return skb;
}

static struct rtable *find_route(struct c4iw_dev *dev, __be32 local_ip,
				 __be32 peer_ip, __be16 local_port,
				 __be16 peer_port, u8 tos)
{
	struct rtable *rt;
319
	struct flowi4 fl4;
320

321
	rt = ip_route_output_ports(&init_net, &fl4, NULL, peer_ip, local_ip,
322 323
				   peer_port, local_port, IPPROTO_TCP,
				   tos, 0);
324
	if (IS_ERR(rt))
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
		return NULL;
	return rt;
}

static void arp_failure_discard(void *handle, struct sk_buff *skb)
{
	PDBG("%s c4iw_dev %p\n", __func__, handle);
	kfree_skb(skb);
}

/*
 * Handle an ARP failure for an active open.
 */
static void act_open_req_arp_failure(void *handle, struct sk_buff *skb)
{
	printk(KERN_ERR MOD "ARP failure duing connect\n");
	kfree_skb(skb);
}

/*
 * Handle an ARP failure for a CPL_ABORT_REQ.  Change it into a no RST variant
 * and send it along.
 */
static void abort_arp_failure(void *handle, struct sk_buff *skb)
{
	struct c4iw_rdev *rdev = handle;
	struct cpl_abort_req *req = cplhdr(skb);

	PDBG("%s rdev %p\n", __func__, rdev);
	req->cmd = CPL_ABORT_NO_RST;
	c4iw_ofld_send(rdev, skb);
}

static void send_flowc(struct c4iw_ep *ep, struct sk_buff *skb)
{
	unsigned int flowclen = 80;
	struct fw_flowc_wr *flowc;
	int i;

	skb = get_skb(skb, flowclen, GFP_KERNEL);
	flowc = (struct fw_flowc_wr *)__skb_put(skb, flowclen);

	flowc->op_to_nparams = cpu_to_be32(FW_WR_OP(FW_FLOWC_WR) |
					   FW_FLOWC_WR_NPARAMS(8));
	flowc->flowid_len16 = cpu_to_be32(FW_WR_LEN16(DIV_ROUND_UP(flowclen,
					  16)) | FW_WR_FLOWID(ep->hwtid));

	flowc->mnemval[0].mnemonic = FW_FLOWC_MNEM_PFNVFN;
373
	flowc->mnemval[0].val = cpu_to_be32(PCI_FUNC(ep->com.dev->rdev.lldi.pdev->devfn) << 8);
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
	flowc->mnemval[1].mnemonic = FW_FLOWC_MNEM_CH;
	flowc->mnemval[1].val = cpu_to_be32(ep->tx_chan);
	flowc->mnemval[2].mnemonic = FW_FLOWC_MNEM_PORT;
	flowc->mnemval[2].val = cpu_to_be32(ep->tx_chan);
	flowc->mnemval[3].mnemonic = FW_FLOWC_MNEM_IQID;
	flowc->mnemval[3].val = cpu_to_be32(ep->rss_qid);
	flowc->mnemval[4].mnemonic = FW_FLOWC_MNEM_SNDNXT;
	flowc->mnemval[4].val = cpu_to_be32(ep->snd_seq);
	flowc->mnemval[5].mnemonic = FW_FLOWC_MNEM_RCVNXT;
	flowc->mnemval[5].val = cpu_to_be32(ep->rcv_seq);
	flowc->mnemval[6].mnemonic = FW_FLOWC_MNEM_SNDBUF;
	flowc->mnemval[6].val = cpu_to_be32(snd_win);
	flowc->mnemval[7].mnemonic = FW_FLOWC_MNEM_MSS;
	flowc->mnemval[7].val = cpu_to_be32(ep->emss);
	/* Pad WR to 16 byte boundary */
	flowc->mnemval[8].mnemonic = 0;
	flowc->mnemval[8].val = 0;
	for (i = 0; i < 9; i++) {
		flowc->mnemval[i].r4[0] = 0;
		flowc->mnemval[i].r4[1] = 0;
		flowc->mnemval[i].r4[2] = 0;
	}

	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);
	c4iw_ofld_send(&ep->com.dev->rdev, skb);
}

static int send_halfclose(struct c4iw_ep *ep, gfp_t gfp)
{
	struct cpl_close_con_req *req;
	struct sk_buff *skb;
	int wrlen = roundup(sizeof *req, 16);

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	skb = get_skb(NULL, wrlen, gfp);
	if (!skb) {
		printk(KERN_ERR MOD "%s - failed to alloc skb\n", __func__);
		return -ENOMEM;
	}
	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);
	t4_set_arp_err_handler(skb, NULL, arp_failure_discard);
	req = (struct cpl_close_con_req *) skb_put(skb, wrlen);
	memset(req, 0, wrlen);
	INIT_TP_WR(req, ep->hwtid);
	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_CLOSE_CON_REQ,
						    ep->hwtid));
	return c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
}

static int send_abort(struct c4iw_ep *ep, struct sk_buff *skb, gfp_t gfp)
{
	struct cpl_abort_req *req;
	int wrlen = roundup(sizeof *req, 16);

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	skb = get_skb(skb, wrlen, gfp);
	if (!skb) {
		printk(KERN_ERR MOD "%s - failed to alloc skb.\n",
		       __func__);
		return -ENOMEM;
	}
	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);
	t4_set_arp_err_handler(skb, &ep->com.dev->rdev, abort_arp_failure);
	req = (struct cpl_abort_req *) skb_put(skb, wrlen);
	memset(req, 0, wrlen);
	INIT_TP_WR(req, ep->hwtid);
	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_ABORT_REQ, ep->hwtid));
	req->cmd = CPL_ABORT_SEND_RST;
	return c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
}

static int send_connect(struct c4iw_ep *ep)
{
	struct cpl_act_open_req *req;
	struct sk_buff *skb;
	u64 opt0;
	u32 opt2;
	unsigned int mtu_idx;
	int wscale;
	int wrlen = roundup(sizeof *req, 16);

	PDBG("%s ep %p atid %u\n", __func__, ep, ep->atid);

	skb = get_skb(NULL, wrlen, GFP_KERNEL);
	if (!skb) {
		printk(KERN_ERR MOD "%s - failed to alloc skb.\n",
		       __func__);
		return -ENOMEM;
	}
S
Steve Wise 已提交
463
	set_wr_txq(skb, CPL_PRIORITY_SETUP, ep->ctrlq_idx);
464 465 466 467

	cxgb4_best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx);
	wscale = compute_wscale(rcv_win);
	opt0 = KEEP_ALIVE(1) |
468
	       DELACK(1) |
469 470 471 472 473 474
	       WND_SCALE(wscale) |
	       MSS_IDX(mtu_idx) |
	       L2T_IDX(ep->l2t->idx) |
	       TX_CHAN(ep->tx_chan) |
	       SMAC_SEL(ep->smac_idx) |
	       DSCP(ep->tos) |
S
Steve Wise 已提交
475
	       ULP_MODE(ULP_MODE_TCPDDP) |
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
	       RCV_BUFSIZ(rcv_win>>10);
	opt2 = RX_CHANNEL(0) |
	       RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);
	if (enable_tcp_timestamps)
		opt2 |= TSTAMPS_EN(1);
	if (enable_tcp_sack)
		opt2 |= SACK_EN(1);
	if (wscale && enable_tcp_window_scaling)
		opt2 |= WND_SCALE_EN(1);
	t4_set_arp_err_handler(skb, NULL, act_open_req_arp_failure);

	req = (struct cpl_act_open_req *) skb_put(skb, wrlen);
	INIT_TP_WR(req, 0);
	OPCODE_TID(req) = cpu_to_be32(
		MK_OPCODE_TID(CPL_ACT_OPEN_REQ, ((ep->rss_qid<<14)|ep->atid)));
	req->local_port = ep->com.local_addr.sin_port;
	req->peer_port = ep->com.remote_addr.sin_port;
	req->local_ip = ep->com.local_addr.sin_addr.s_addr;
	req->peer_ip = ep->com.remote_addr.sin_addr.s_addr;
	req->opt0 = cpu_to_be64(opt0);
	req->params = 0;
	req->opt2 = cpu_to_be32(opt2);
	return c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
}

501 502
static void send_mpa_req(struct c4iw_ep *ep, struct sk_buff *skb,
		u8 mpa_rev_to_use)
503 504 505 506
{
	int mpalen, wrlen;
	struct fw_ofld_tx_data_wr *req;
	struct mpa_message *mpa;
507
	struct mpa_v2_conn_params mpa_v2_params;
508 509 510 511 512 513

	PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen);

	BUG_ON(skb_cloned(skb));

	mpalen = sizeof(*mpa) + ep->plen;
514 515
	if (mpa_rev_to_use == 2)
		mpalen += sizeof(struct mpa_v2_conn_params);
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
	wrlen = roundup(mpalen + sizeof *req, 16);
	skb = get_skb(skb, wrlen, GFP_KERNEL);
	if (!skb) {
		connect_reply_upcall(ep, -ENOMEM);
		return;
	}
	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);

	req = (struct fw_ofld_tx_data_wr *)skb_put(skb, wrlen);
	memset(req, 0, wrlen);
	req->op_to_immdlen = cpu_to_be32(
		FW_WR_OP(FW_OFLD_TX_DATA_WR) |
		FW_WR_COMPL(1) |
		FW_WR_IMMDLEN(mpalen));
	req->flowid_len16 = cpu_to_be32(
		FW_WR_FLOWID(ep->hwtid) |
		FW_WR_LEN16(wrlen >> 4));
	req->plen = cpu_to_be32(mpalen);
	req->tunnel_to_proxy = cpu_to_be32(
		FW_OFLD_TX_DATA_WR_FLUSH(1) |
		FW_OFLD_TX_DATA_WR_SHOVE(1));

	mpa = (struct mpa_message *)(req + 1);
	memcpy(mpa->key, MPA_KEY_REQ, sizeof(mpa->key));
	mpa->flags = (crc_enabled ? MPA_CRC : 0) |
541 542
		     (markers_enabled ? MPA_MARKERS : 0) |
		     (mpa_rev_to_use == 2 ? MPA_ENHANCED_RDMA_CONN : 0);
543
	mpa->private_data_size = htons(ep->plen);
544
	mpa->revision = mpa_rev_to_use;
545
	if (mpa_rev_to_use == 1) {
546
		ep->tried_with_mpa_v1 = 1;
547 548
		ep->retry_with_mpa_v1 = 0;
	}
549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566

	if (mpa_rev_to_use == 2) {
		mpa->private_data_size +=
			htons(sizeof(struct mpa_v2_conn_params));
		mpa_v2_params.ird = htons((u16)ep->ird);
		mpa_v2_params.ord = htons((u16)ep->ord);

		if (peer2peer) {
			mpa_v2_params.ird |= htons(MPA_V2_PEER2PEER_MODEL);
			if (p2p_type == FW_RI_INIT_P2PTYPE_RDMA_WRITE)
				mpa_v2_params.ord |=
					htons(MPA_V2_RDMA_WRITE_RTR);
			else if (p2p_type == FW_RI_INIT_P2PTYPE_READ_REQ)
				mpa_v2_params.ord |=
					htons(MPA_V2_RDMA_READ_RTR);
		}
		memcpy(mpa->private_data, &mpa_v2_params,
		       sizeof(struct mpa_v2_conn_params));
567

568 569 570 571 572 573 574 575
		if (ep->plen)
			memcpy(mpa->private_data +
			       sizeof(struct mpa_v2_conn_params),
			       ep->mpa_pkt + sizeof(*mpa), ep->plen);
	} else
		if (ep->plen)
			memcpy(mpa->private_data,
					ep->mpa_pkt + sizeof(*mpa), ep->plen);
576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598

	/*
	 * Reference the mpa skb.  This ensures the data area
	 * will remain in memory until the hw acks the tx.
	 * Function fw4_ack() will deref it.
	 */
	skb_get(skb);
	t4_set_arp_err_handler(skb, NULL, arp_failure_discard);
	BUG_ON(ep->mpa_skb);
	ep->mpa_skb = skb;
	c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
	start_ep_timer(ep);
	state_set(&ep->com, MPA_REQ_SENT);
	ep->mpa_attr.initiator = 1;
	return;
}

static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen)
{
	int mpalen, wrlen;
	struct fw_ofld_tx_data_wr *req;
	struct mpa_message *mpa;
	struct sk_buff *skb;
599
	struct mpa_v2_conn_params mpa_v2_params;
600 601 602 603

	PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen);

	mpalen = sizeof(*mpa) + plen;
604 605
	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn)
		mpalen += sizeof(struct mpa_v2_conn_params);
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
	wrlen = roundup(mpalen + sizeof *req, 16);

	skb = get_skb(NULL, wrlen, GFP_KERNEL);
	if (!skb) {
		printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__);
		return -ENOMEM;
	}
	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);

	req = (struct fw_ofld_tx_data_wr *)skb_put(skb, wrlen);
	memset(req, 0, wrlen);
	req->op_to_immdlen = cpu_to_be32(
		FW_WR_OP(FW_OFLD_TX_DATA_WR) |
		FW_WR_COMPL(1) |
		FW_WR_IMMDLEN(mpalen));
	req->flowid_len16 = cpu_to_be32(
		FW_WR_FLOWID(ep->hwtid) |
		FW_WR_LEN16(wrlen >> 4));
	req->plen = cpu_to_be32(mpalen);
	req->tunnel_to_proxy = cpu_to_be32(
		FW_OFLD_TX_DATA_WR_FLUSH(1) |
		FW_OFLD_TX_DATA_WR_SHOVE(1));

	mpa = (struct mpa_message *)(req + 1);
	memset(mpa, 0, sizeof(*mpa));
	memcpy(mpa->key, MPA_KEY_REP, sizeof(mpa->key));
	mpa->flags = MPA_REJECT;
	mpa->revision = mpa_rev;
	mpa->private_data_size = htons(plen);
635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657

	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) {
		mpa->flags |= MPA_ENHANCED_RDMA_CONN;
		mpa->private_data_size +=
			htons(sizeof(struct mpa_v2_conn_params));
		mpa_v2_params.ird = htons(((u16)ep->ird) |
					  (peer2peer ? MPA_V2_PEER2PEER_MODEL :
					   0));
		mpa_v2_params.ord = htons(((u16)ep->ord) | (peer2peer ?
					  (p2p_type ==
					   FW_RI_INIT_P2PTYPE_RDMA_WRITE ?
					   MPA_V2_RDMA_WRITE_RTR : p2p_type ==
					   FW_RI_INIT_P2PTYPE_READ_REQ ?
					   MPA_V2_RDMA_READ_RTR : 0) : 0));
		memcpy(mpa->private_data, &mpa_v2_params,
		       sizeof(struct mpa_v2_conn_params));

		if (ep->plen)
			memcpy(mpa->private_data +
			       sizeof(struct mpa_v2_conn_params), pdata, plen);
	} else
		if (plen)
			memcpy(mpa->private_data, pdata, plen);
658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677

	/*
	 * Reference the mpa skb again.  This ensures the data area
	 * will remain in memory until the hw acks the tx.
	 * Function fw4_ack() will deref it.
	 */
	skb_get(skb);
	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);
	t4_set_arp_err_handler(skb, NULL, arp_failure_discard);
	BUG_ON(ep->mpa_skb);
	ep->mpa_skb = skb;
	return c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
}

static int send_mpa_reply(struct c4iw_ep *ep, const void *pdata, u8 plen)
{
	int mpalen, wrlen;
	struct fw_ofld_tx_data_wr *req;
	struct mpa_message *mpa;
	struct sk_buff *skb;
678
	struct mpa_v2_conn_params mpa_v2_params;
679 680 681 682

	PDBG("%s ep %p tid %u pd_len %d\n", __func__, ep, ep->hwtid, ep->plen);

	mpalen = sizeof(*mpa) + plen;
683 684
	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn)
		mpalen += sizeof(struct mpa_v2_conn_params);
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
	wrlen = roundup(mpalen + sizeof *req, 16);

	skb = get_skb(NULL, wrlen, GFP_KERNEL);
	if (!skb) {
		printk(KERN_ERR MOD "%s - cannot alloc skb!\n", __func__);
		return -ENOMEM;
	}
	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);

	req = (struct fw_ofld_tx_data_wr *) skb_put(skb, wrlen);
	memset(req, 0, wrlen);
	req->op_to_immdlen = cpu_to_be32(
		FW_WR_OP(FW_OFLD_TX_DATA_WR) |
		FW_WR_COMPL(1) |
		FW_WR_IMMDLEN(mpalen));
	req->flowid_len16 = cpu_to_be32(
		FW_WR_FLOWID(ep->hwtid) |
		FW_WR_LEN16(wrlen >> 4));
	req->plen = cpu_to_be32(mpalen);
	req->tunnel_to_proxy = cpu_to_be32(
		FW_OFLD_TX_DATA_WR_FLUSH(1) |
		FW_OFLD_TX_DATA_WR_SHOVE(1));

	mpa = (struct mpa_message *)(req + 1);
	memset(mpa, 0, sizeof(*mpa));
	memcpy(mpa->key, MPA_KEY_REP, sizeof(mpa->key));
	mpa->flags = (ep->mpa_attr.crc_enabled ? MPA_CRC : 0) |
		     (markers_enabled ? MPA_MARKERS : 0);
713
	mpa->revision = ep->mpa_attr.version;
714
	mpa->private_data_size = htons(plen);
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

	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) {
		mpa->flags |= MPA_ENHANCED_RDMA_CONN;
		mpa->private_data_size +=
			htons(sizeof(struct mpa_v2_conn_params));
		mpa_v2_params.ird = htons((u16)ep->ird);
		mpa_v2_params.ord = htons((u16)ep->ord);
		if (peer2peer && (ep->mpa_attr.p2p_type !=
					FW_RI_INIT_P2PTYPE_DISABLED)) {
			mpa_v2_params.ird |= htons(MPA_V2_PEER2PEER_MODEL);

			if (p2p_type == FW_RI_INIT_P2PTYPE_RDMA_WRITE)
				mpa_v2_params.ord |=
					htons(MPA_V2_RDMA_WRITE_RTR);
			else if (p2p_type == FW_RI_INIT_P2PTYPE_READ_REQ)
				mpa_v2_params.ord |=
					htons(MPA_V2_RDMA_READ_RTR);
		}

		memcpy(mpa->private_data, &mpa_v2_params,
		       sizeof(struct mpa_v2_conn_params));

		if (ep->plen)
			memcpy(mpa->private_data +
			       sizeof(struct mpa_v2_conn_params), pdata, plen);
	} else
		if (plen)
			memcpy(mpa->private_data, pdata, plen);
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

	/*
	 * Reference the mpa skb.  This ensures the data area
	 * will remain in memory until the hw acks the tx.
	 * Function fw4_ack() will deref it.
	 */
	skb_get(skb);
	t4_set_arp_err_handler(skb, NULL, arp_failure_discard);
	ep->mpa_skb = skb;
	state_set(&ep->com, MPA_REP_SENT);
	return c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
}

static int act_establish(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *ep;
	struct cpl_act_establish *req = cplhdr(skb);
	unsigned int tid = GET_TID(req);
	unsigned int atid = GET_TID_TID(ntohl(req->tos_atid));
	struct tid_info *t = dev->rdev.lldi.tids;

	ep = lookup_atid(t, atid);

	PDBG("%s ep %p tid %u snd_isn %u rcv_isn %u\n", __func__, ep, tid,
	     be32_to_cpu(req->snd_isn), be32_to_cpu(req->rcv_isn));

	dst_confirm(ep->dst);

	/* setup the hwtid for this connection */
	ep->hwtid = tid;
	cxgb4_insert_tid(t, ep, tid);

	ep->snd_seq = be32_to_cpu(req->snd_isn);
	ep->rcv_seq = be32_to_cpu(req->rcv_isn);

	set_emss(ep, ntohs(req->tcp_opt));

	/* dealloc the atid */
	cxgb4_free_atid(t, atid);

	/* start MPA negotiation */
	send_flowc(ep, NULL);
785 786 787 788
	if (ep->retry_with_mpa_v1)
		send_mpa_req(ep, skb, 1);
	else
		send_mpa_req(ep, skb, mpa_rev);
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

	return 0;
}

static void close_complete_upcall(struct c4iw_ep *ep)
{
	struct iw_cm_event event;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	memset(&event, 0, sizeof(event));
	event.event = IW_CM_EVENT_CLOSE;
	if (ep->com.cm_id) {
		PDBG("close complete delivered ep %p cm_id %p tid %u\n",
		     ep, ep->com.cm_id, ep->hwtid);
		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
		ep->com.cm_id->rem_ref(ep->com.cm_id);
		ep->com.cm_id = NULL;
		ep->com.qp = NULL;
	}
}

static int abort_connection(struct c4iw_ep *ep, struct sk_buff *skb, gfp_t gfp)
{
	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	close_complete_upcall(ep);
	state_set(&ep->com, ABORTING);
	return send_abort(ep, skb, gfp);
}

static void peer_close_upcall(struct c4iw_ep *ep)
{
	struct iw_cm_event event;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	memset(&event, 0, sizeof(event));
	event.event = IW_CM_EVENT_DISCONNECT;
	if (ep->com.cm_id) {
		PDBG("peer close delivered ep %p cm_id %p tid %u\n",
		     ep, ep->com.cm_id, ep->hwtid);
		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
	}
}

static void peer_abort_upcall(struct c4iw_ep *ep)
{
	struct iw_cm_event event;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	memset(&event, 0, sizeof(event));
	event.event = IW_CM_EVENT_CLOSE;
	event.status = -ECONNRESET;
	if (ep->com.cm_id) {
		PDBG("abort delivered ep %p cm_id %p tid %u\n", ep,
		     ep->com.cm_id, ep->hwtid);
		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
		ep->com.cm_id->rem_ref(ep->com.cm_id);
		ep->com.cm_id = NULL;
		ep->com.qp = NULL;
	}
}

static void connect_reply_upcall(struct c4iw_ep *ep, int status)
{
	struct iw_cm_event event;

	PDBG("%s ep %p tid %u status %d\n", __func__, ep, ep->hwtid, status);
	memset(&event, 0, sizeof(event));
	event.event = IW_CM_EVENT_CONNECT_REPLY;
	event.status = status;
	event.local_addr = ep->com.local_addr;
	event.remote_addr = ep->com.remote_addr;

	if ((status == 0) || (status == -ECONNREFUSED)) {
862 863 864 865 866 867 868 869 870 871 872 873 874
		if (!ep->tried_with_mpa_v1) {
			/* this means MPA_v2 is used */
			event.private_data_len = ep->plen -
				sizeof(struct mpa_v2_conn_params);
			event.private_data = ep->mpa_pkt +
				sizeof(struct mpa_message) +
				sizeof(struct mpa_v2_conn_params);
		} else {
			/* this means MPA_v1 is used */
			event.private_data_len = ep->plen;
			event.private_data = ep->mpa_pkt +
				sizeof(struct mpa_message);
		}
875
	}
876 877 878 879 880

	PDBG("%s ep %p tid %u status %d\n", __func__, ep,
	     ep->hwtid, status);
	ep->com.cm_id->event_handler(ep->com.cm_id, &event);

881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897
	if (status < 0) {
		ep->com.cm_id->rem_ref(ep->com.cm_id);
		ep->com.cm_id = NULL;
		ep->com.qp = NULL;
	}
}

static void connect_request_upcall(struct c4iw_ep *ep)
{
	struct iw_cm_event event;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	memset(&event, 0, sizeof(event));
	event.event = IW_CM_EVENT_CONNECT_REQUEST;
	event.local_addr = ep->com.local_addr;
	event.remote_addr = ep->com.remote_addr;
	event.provider_data = ep;
898 899 900 901 902 903 904 905 906 907 908 909 910 911 912
	if (!ep->tried_with_mpa_v1) {
		/* this means MPA_v2 is used */
		event.ord = ep->ord;
		event.ird = ep->ird;
		event.private_data_len = ep->plen -
			sizeof(struct mpa_v2_conn_params);
		event.private_data = ep->mpa_pkt + sizeof(struct mpa_message) +
			sizeof(struct mpa_v2_conn_params);
	} else {
		/* this means MPA_v1 is used. Send max supported */
		event.ord = c4iw_max_read_depth;
		event.ird = c4iw_max_read_depth;
		event.private_data_len = ep->plen;
		event.private_data = ep->mpa_pkt + sizeof(struct mpa_message);
	}
913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929
	if (state_read(&ep->parent_ep->com) != DEAD) {
		c4iw_get_ep(&ep->com);
		ep->parent_ep->com.cm_id->event_handler(
						ep->parent_ep->com.cm_id,
						&event);
	}
	c4iw_put_ep(&ep->parent_ep->com);
	ep->parent_ep = NULL;
}

static void established_upcall(struct c4iw_ep *ep)
{
	struct iw_cm_event event;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	memset(&event, 0, sizeof(event));
	event.event = IW_CM_EVENT_ESTABLISHED;
930 931
	event.ird = ep->ird;
	event.ord = ep->ord;
932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955
	if (ep->com.cm_id) {
		PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
	}
}

static int update_rx_credits(struct c4iw_ep *ep, u32 credits)
{
	struct cpl_rx_data_ack *req;
	struct sk_buff *skb;
	int wrlen = roundup(sizeof *req, 16);

	PDBG("%s ep %p tid %u credits %u\n", __func__, ep, ep->hwtid, credits);
	skb = get_skb(NULL, wrlen, GFP_KERNEL);
	if (!skb) {
		printk(KERN_ERR MOD "update_rx_credits - cannot alloc skb!\n");
		return 0;
	}

	req = (struct cpl_rx_data_ack *) skb_put(skb, wrlen);
	memset(req, 0, wrlen);
	INIT_TP_WR(req, ep->hwtid);
	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_RX_DATA_ACK,
						    ep->hwtid));
956 957 958
	req->credit_dack = cpu_to_be32(credits | RX_FORCE_ACK(1) |
				       F_RX_DACK_CHANGE |
				       V_RX_DACK_MODE(dack_mode));
S
Steve Wise 已提交
959
	set_wr_txq(skb, CPL_PRIORITY_ACK, ep->ctrlq_idx);
960 961 962 963 964 965 966
	c4iw_ofld_send(&ep->com.dev->rdev, skb);
	return credits;
}

static void process_mpa_reply(struct c4iw_ep *ep, struct sk_buff *skb)
{
	struct mpa_message *mpa;
967
	struct mpa_v2_conn_params *mpa_v2_params;
968
	u16 plen;
969 970
	u16 resp_ird, resp_ord;
	u8 rtr_mismatch = 0, insuff_ird = 0;
971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009
	struct c4iw_qp_attributes attrs;
	enum c4iw_qp_attr_mask mask;
	int err;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);

	/*
	 * Stop mpa timer.  If it expired, then the state has
	 * changed and we bail since ep_timeout already aborted
	 * the connection.
	 */
	stop_ep_timer(ep);
	if (state_read(&ep->com) != MPA_REQ_SENT)
		return;

	/*
	 * If we get more than the supported amount of private data
	 * then we must fail this connection.
	 */
	if (ep->mpa_pkt_len + skb->len > sizeof(ep->mpa_pkt)) {
		err = -EINVAL;
		goto err;
	}

	/*
	 * copy the new data into our accumulation buffer.
	 */
	skb_copy_from_linear_data(skb, &(ep->mpa_pkt[ep->mpa_pkt_len]),
				  skb->len);
	ep->mpa_pkt_len += skb->len;

	/*
	 * if we don't even have the mpa message, then bail.
	 */
	if (ep->mpa_pkt_len < sizeof(*mpa))
		return;
	mpa = (struct mpa_message *) ep->mpa_pkt;

	/* Validate MPA header. */
1010 1011 1012
	if (mpa->revision > mpa_rev) {
		printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d,"
		       " Received = %d\n", __func__, mpa_rev, mpa->revision);
1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061
		err = -EPROTO;
		goto err;
	}
	if (memcmp(mpa->key, MPA_KEY_REP, sizeof(mpa->key))) {
		err = -EPROTO;
		goto err;
	}

	plen = ntohs(mpa->private_data_size);

	/*
	 * Fail if there's too much private data.
	 */
	if (plen > MPA_MAX_PRIVATE_DATA) {
		err = -EPROTO;
		goto err;
	}

	/*
	 * If plen does not account for pkt size
	 */
	if (ep->mpa_pkt_len > (sizeof(*mpa) + plen)) {
		err = -EPROTO;
		goto err;
	}

	ep->plen = (u8) plen;

	/*
	 * If we don't have all the pdata yet, then bail.
	 * We'll continue process when more data arrives.
	 */
	if (ep->mpa_pkt_len < (sizeof(*mpa) + plen))
		return;

	if (mpa->flags & MPA_REJECT) {
		err = -ECONNREFUSED;
		goto err;
	}

	/*
	 * If we get here we have accumulated the entire mpa
	 * start reply message including private data. And
	 * the MPA header is valid.
	 */
	state_set(&ep->com, FPDU_MODE);
	ep->mpa_attr.crc_enabled = (mpa->flags & MPA_CRC) | crc_enabled ? 1 : 0;
	ep->mpa_attr.recv_marker_enabled = markers_enabled;
	ep->mpa_attr.xmit_marker_enabled = mpa->flags & MPA_MARKERS ? 1 : 0;
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103
	ep->mpa_attr.version = mpa->revision;
	ep->mpa_attr.p2p_type = FW_RI_INIT_P2PTYPE_DISABLED;

	if (mpa->revision == 2) {
		ep->mpa_attr.enhanced_rdma_conn =
			mpa->flags & MPA_ENHANCED_RDMA_CONN ? 1 : 0;
		if (ep->mpa_attr.enhanced_rdma_conn) {
			mpa_v2_params = (struct mpa_v2_conn_params *)
				(ep->mpa_pkt + sizeof(*mpa));
			resp_ird = ntohs(mpa_v2_params->ird) &
				MPA_V2_IRD_ORD_MASK;
			resp_ord = ntohs(mpa_v2_params->ord) &
				MPA_V2_IRD_ORD_MASK;

			/*
			 * This is a double-check. Ideally, below checks are
			 * not required since ird/ord stuff has been taken
			 * care of in c4iw_accept_cr
			 */
			if ((ep->ird < resp_ord) || (ep->ord > resp_ird)) {
				err = -ENOMEM;
				ep->ird = resp_ord;
				ep->ord = resp_ird;
				insuff_ird = 1;
			}

			if (ntohs(mpa_v2_params->ird) &
					MPA_V2_PEER2PEER_MODEL) {
				if (ntohs(mpa_v2_params->ord) &
						MPA_V2_RDMA_WRITE_RTR)
					ep->mpa_attr.p2p_type =
						FW_RI_INIT_P2PTYPE_RDMA_WRITE;
				else if (ntohs(mpa_v2_params->ord) &
						MPA_V2_RDMA_READ_RTR)
					ep->mpa_attr.p2p_type =
						FW_RI_INIT_P2PTYPE_READ_REQ;
			}
		}
	} else if (mpa->revision == 1)
		if (peer2peer)
			ep->mpa_attr.p2p_type = p2p_type;

1104
	PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, "
1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116
	     "xmit_marker_enabled=%d, version=%d p2p_type=%d local-p2p_type = "
	     "%d\n", __func__, ep->mpa_attr.crc_enabled,
	     ep->mpa_attr.recv_marker_enabled,
	     ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version,
	     ep->mpa_attr.p2p_type, p2p_type);

	/*
	 * If responder's RTR does not match with that of initiator, assign
	 * FW_RI_INIT_P2PTYPE_DISABLED in mpa attributes so that RTR is not
	 * generated when moving QP to RTS state.
	 * A TERM message will be sent after QP has moved to RTS state
	 */
1117
	if ((ep->mpa_attr.version == 2) && peer2peer &&
1118 1119 1120 1121
			(ep->mpa_attr.p2p_type != p2p_type)) {
		ep->mpa_attr.p2p_type = FW_RI_INIT_P2PTYPE_DISABLED;
		rtr_mismatch = 1;
	}
1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137

	attrs.mpa_attr = ep->mpa_attr;
	attrs.max_ird = ep->ird;
	attrs.max_ord = ep->ord;
	attrs.llp_stream_handle = ep;
	attrs.next_state = C4IW_QP_STATE_RTS;

	mask = C4IW_QP_ATTR_NEXT_STATE |
	    C4IW_QP_ATTR_LLP_STREAM_HANDLE | C4IW_QP_ATTR_MPA_ATTR |
	    C4IW_QP_ATTR_MAX_IRD | C4IW_QP_ATTR_MAX_ORD;

	/* bind QP and TID with INIT_WR */
	err = c4iw_modify_qp(ep->com.qp->rhp,
			     ep->com.qp, mask, &attrs, 1);
	if (err)
		goto err;
1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170

	/*
	 * If responder's RTR requirement did not match with what initiator
	 * supports, generate TERM message
	 */
	if (rtr_mismatch) {
		printk(KERN_ERR "%s: RTR mismatch, sending TERM\n", __func__);
		attrs.layer_etype = LAYER_MPA | DDP_LLP;
		attrs.ecode = MPA_NOMATCH_RTR;
		attrs.next_state = C4IW_QP_STATE_TERMINATE;
		err = c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
				C4IW_QP_ATTR_NEXT_STATE, &attrs, 0);
		err = -ENOMEM;
		goto out;
	}

	/*
	 * Generate TERM if initiator IRD is not sufficient for responder
	 * provided ORD. Currently, we do the same behaviour even when
	 * responder provided IRD is also not sufficient as regards to
	 * initiator ORD.
	 */
	if (insuff_ird) {
		printk(KERN_ERR "%s: Insufficient IRD, sending TERM\n",
				__func__);
		attrs.layer_etype = LAYER_MPA | DDP_LLP;
		attrs.ecode = MPA_INSUFF_IRD;
		attrs.next_state = C4IW_QP_STATE_TERMINATE;
		err = c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
				C4IW_QP_ATTR_NEXT_STATE, &attrs, 0);
		err = -ENOMEM;
		goto out;
	}
1171 1172
	goto out;
err:
1173 1174
	state_set(&ep->com, ABORTING);
	send_abort(ep, skb, GFP_KERNEL);
1175 1176 1177 1178 1179 1180 1181 1182
out:
	connect_reply_upcall(ep, err);
	return;
}

static void process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
{
	struct mpa_message *mpa;
1183
	struct mpa_v2_conn_params *mpa_v2_params;
1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
	u16 plen;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);

	if (state_read(&ep->com) != MPA_REQ_WAIT)
		return;

	/*
	 * If we get more than the supported amount of private data
	 * then we must fail this connection.
	 */
	if (ep->mpa_pkt_len + skb->len > sizeof(ep->mpa_pkt)) {
		stop_ep_timer(ep);
		abort_connection(ep, skb, GFP_KERNEL);
		return;
	}

	PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__);

	/*
	 * Copy the new data into our accumulation buffer.
	 */
	skb_copy_from_linear_data(skb, &(ep->mpa_pkt[ep->mpa_pkt_len]),
				  skb->len);
	ep->mpa_pkt_len += skb->len;

	/*
	 * If we don't even have the mpa message, then bail.
	 * We'll continue process when more data arrives.
	 */
	if (ep->mpa_pkt_len < sizeof(*mpa))
		return;

	PDBG("%s enter (%s line %u)\n", __func__, __FILE__, __LINE__);
	stop_ep_timer(ep);
	mpa = (struct mpa_message *) ep->mpa_pkt;

	/*
	 * Validate MPA Header.
	 */
1224 1225 1226
	if (mpa->revision > mpa_rev) {
		printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d,"
		       " Received = %d\n", __func__, mpa_rev, mpa->revision);
1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268
		abort_connection(ep, skb, GFP_KERNEL);
		return;
	}

	if (memcmp(mpa->key, MPA_KEY_REQ, sizeof(mpa->key))) {
		abort_connection(ep, skb, GFP_KERNEL);
		return;
	}

	plen = ntohs(mpa->private_data_size);

	/*
	 * Fail if there's too much private data.
	 */
	if (plen > MPA_MAX_PRIVATE_DATA) {
		abort_connection(ep, skb, GFP_KERNEL);
		return;
	}

	/*
	 * If plen does not account for pkt size
	 */
	if (ep->mpa_pkt_len > (sizeof(*mpa) + plen)) {
		abort_connection(ep, skb, GFP_KERNEL);
		return;
	}
	ep->plen = (u8) plen;

	/*
	 * If we don't have all the pdata yet, then bail.
	 */
	if (ep->mpa_pkt_len < (sizeof(*mpa) + plen))
		return;

	/*
	 * If we get here we have accumulated the entire mpa
	 * start reply message including private data.
	 */
	ep->mpa_attr.initiator = 0;
	ep->mpa_attr.crc_enabled = (mpa->flags & MPA_CRC) | crc_enabled ? 1 : 0;
	ep->mpa_attr.recv_marker_enabled = markers_enabled;
	ep->mpa_attr.xmit_marker_enabled = mpa->flags & MPA_MARKERS ? 1 : 0;
1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299
	ep->mpa_attr.version = mpa->revision;
	if (mpa->revision == 1)
		ep->tried_with_mpa_v1 = 1;
	ep->mpa_attr.p2p_type = FW_RI_INIT_P2PTYPE_DISABLED;

	if (mpa->revision == 2) {
		ep->mpa_attr.enhanced_rdma_conn =
			mpa->flags & MPA_ENHANCED_RDMA_CONN ? 1 : 0;
		if (ep->mpa_attr.enhanced_rdma_conn) {
			mpa_v2_params = (struct mpa_v2_conn_params *)
				(ep->mpa_pkt + sizeof(*mpa));
			ep->ird = ntohs(mpa_v2_params->ird) &
				MPA_V2_IRD_ORD_MASK;
			ep->ord = ntohs(mpa_v2_params->ord) &
				MPA_V2_IRD_ORD_MASK;
			if (ntohs(mpa_v2_params->ird) & MPA_V2_PEER2PEER_MODEL)
				if (peer2peer) {
					if (ntohs(mpa_v2_params->ord) &
							MPA_V2_RDMA_WRITE_RTR)
						ep->mpa_attr.p2p_type =
						FW_RI_INIT_P2PTYPE_RDMA_WRITE;
					else if (ntohs(mpa_v2_params->ord) &
							MPA_V2_RDMA_READ_RTR)
						ep->mpa_attr.p2p_type =
						FW_RI_INIT_P2PTYPE_READ_REQ;
				}
		}
	} else if (mpa->revision == 1)
		if (peer2peer)
			ep->mpa_attr.p2p_type = p2p_type;

1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365
	PDBG("%s - crc_enabled=%d, recv_marker_enabled=%d, "
	     "xmit_marker_enabled=%d, version=%d p2p_type=%d\n", __func__,
	     ep->mpa_attr.crc_enabled, ep->mpa_attr.recv_marker_enabled,
	     ep->mpa_attr.xmit_marker_enabled, ep->mpa_attr.version,
	     ep->mpa_attr.p2p_type);

	state_set(&ep->com, MPA_REQ_RCVD);

	/* drive upcall */
	connect_request_upcall(ep);
	return;
}

static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *ep;
	struct cpl_rx_data *hdr = cplhdr(skb);
	unsigned int dlen = ntohs(hdr->len);
	unsigned int tid = GET_TID(hdr);
	struct tid_info *t = dev->rdev.lldi.tids;

	ep = lookup_tid(t, tid);
	PDBG("%s ep %p tid %u dlen %u\n", __func__, ep, ep->hwtid, dlen);
	skb_pull(skb, sizeof(*hdr));
	skb_trim(skb, dlen);

	ep->rcv_seq += dlen;
	BUG_ON(ep->rcv_seq != (ntohl(hdr->seq) + dlen));

	/* update RX credits */
	update_rx_credits(ep, dlen);

	switch (state_read(&ep->com)) {
	case MPA_REQ_SENT:
		process_mpa_reply(ep, skb);
		break;
	case MPA_REQ_WAIT:
		process_mpa_request(ep, skb);
		break;
	case MPA_REP_SENT:
		break;
	default:
		printk(KERN_ERR MOD "%s Unexpected streaming data."
		       " ep %p state %d tid %u\n",
		       __func__, ep, state_read(&ep->com), ep->hwtid);

		/*
		 * The ep will timeout and inform the ULP of the failure.
		 * See ep_timeout().
		 */
		break;
	}
	return 0;
}

static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *ep;
	struct cpl_abort_rpl_rss *rpl = cplhdr(skb);
	int release = 0;
	unsigned int tid = GET_TID(rpl);
	struct tid_info *t = dev->rdev.lldi.tids;

	ep = lookup_tid(t, tid);
	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	BUG_ON(!ep);
1366
	mutex_lock(&ep->com.mutex);
1367 1368 1369 1370 1371 1372 1373 1374 1375 1376
	switch (ep->com.state) {
	case ABORTING:
		__state_set(&ep->com, DEAD);
		release = 1;
		break;
	default:
		printk(KERN_ERR "%s ep %p state %d\n",
		     __func__, ep, ep->com.state);
		break;
	}
1377
	mutex_unlock(&ep->com.mutex);
1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439

	if (release)
		release_ep_resources(ep);
	return 0;
}

/*
 * Return whether a failed active open has allocated a TID
 */
static inline int act_open_has_tid(int status)
{
	return status != CPL_ERR_TCAM_FULL && status != CPL_ERR_CONN_EXIST &&
	       status != CPL_ERR_ARP_MISS;
}

static int act_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *ep;
	struct cpl_act_open_rpl *rpl = cplhdr(skb);
	unsigned int atid = GET_TID_TID(GET_AOPEN_ATID(
					ntohl(rpl->atid_status)));
	struct tid_info *t = dev->rdev.lldi.tids;
	int status = GET_AOPEN_STATUS(ntohl(rpl->atid_status));

	ep = lookup_atid(t, atid);

	PDBG("%s ep %p atid %u status %u errno %d\n", __func__, ep, atid,
	     status, status2errno(status));

	if (status == CPL_ERR_RTX_NEG_ADVICE) {
		printk(KERN_WARNING MOD "Connection problems for atid %u\n",
			atid);
		return 0;
	}

	connect_reply_upcall(ep, status2errno(status));
	state_set(&ep->com, DEAD);

	if (status && act_open_has_tid(status))
		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, GET_TID(rpl));

	cxgb4_free_atid(t, atid);
	dst_release(ep->dst);
	cxgb4_l2t_release(ep->l2t);
	c4iw_put_ep(&ep->com);

	return 0;
}

static int pass_open_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_pass_open_rpl *rpl = cplhdr(skb);
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int stid = GET_TID(rpl);
	struct c4iw_listen_ep *ep = lookup_stid(t, stid);

	if (!ep) {
		printk(KERN_ERR MOD "stid %d lookup failure!\n", stid);
		return 0;
	}
	PDBG("%s ep %p status %d error %d\n", __func__, ep,
	     rpl->status, status2errno(rpl->status));
1440
	c4iw_wake_up(&ep->com.wr_wait, status2errno(rpl->status));
1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473

	return 0;
}

static int listen_stop(struct c4iw_listen_ep *ep)
{
	struct sk_buff *skb;
	struct cpl_close_listsvr_req *req;

	PDBG("%s ep %p\n", __func__, ep);
	skb = get_skb(NULL, sizeof(*req), GFP_KERNEL);
	if (!skb) {
		printk(KERN_ERR MOD "%s - failed to alloc skb\n", __func__);
		return -ENOMEM;
	}
	req = (struct cpl_close_listsvr_req *) skb_put(skb, sizeof(*req));
	INIT_TP_WR(req, 0);
	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_CLOSE_LISTSRV_REQ,
						    ep->stid));
	req->reply_ctrl = cpu_to_be16(
			  QUEUENO(ep->com.dev->rdev.lldi.rxq_ids[0]));
	set_wr_txq(skb, CPL_PRIORITY_SETUP, 0);
	return c4iw_ofld_send(&ep->com.dev->rdev, skb);
}

static int close_listsrv_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_close_listsvr_rpl *rpl = cplhdr(skb);
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int stid = GET_TID(rpl);
	struct c4iw_listen_ep *ep = lookup_stid(t, stid);

	PDBG("%s ep %p\n", __func__, ep);
1474
	c4iw_wake_up(&ep->com.wr_wait, status2errno(rpl->status));
1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493
	return 0;
}

static void accept_cr(struct c4iw_ep *ep, __be32 peer_ip, struct sk_buff *skb,
		      struct cpl_pass_accept_req *req)
{
	struct cpl_pass_accept_rpl *rpl;
	unsigned int mtu_idx;
	u64 opt0;
	u32 opt2;
	int wscale;

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	BUG_ON(skb_cloned(skb));
	skb_trim(skb, sizeof(*rpl));
	skb_get(skb);
	cxgb4_best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx);
	wscale = compute_wscale(rcv_win);
	opt0 = KEEP_ALIVE(1) |
1494
	       DELACK(1) |
1495 1496 1497 1498 1499 1500
	       WND_SCALE(wscale) |
	       MSS_IDX(mtu_idx) |
	       L2T_IDX(ep->l2t->idx) |
	       TX_CHAN(ep->tx_chan) |
	       SMAC_SEL(ep->smac_idx) |
	       DSCP(ep->tos) |
S
Steve Wise 已提交
1501
	       ULP_MODE(ULP_MODE_TCPDDP) |
1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518
	       RCV_BUFSIZ(rcv_win>>10);
	opt2 = RX_CHANNEL(0) |
	       RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);

	if (enable_tcp_timestamps && req->tcpopt.tstamp)
		opt2 |= TSTAMPS_EN(1);
	if (enable_tcp_sack && req->tcpopt.sack)
		opt2 |= SACK_EN(1);
	if (wscale && enable_tcp_window_scaling)
		opt2 |= WND_SCALE_EN(1);

	rpl = cplhdr(skb);
	INIT_TP_WR(rpl, ep->hwtid);
	OPCODE_TID(rpl) = cpu_to_be32(MK_OPCODE_TID(CPL_PASS_ACCEPT_RPL,
				      ep->hwtid));
	rpl->opt0 = cpu_to_be64(opt0);
	rpl->opt2 = cpu_to_be32(opt2);
S
Steve Wise 已提交
1519
	set_wr_txq(skb, CPL_PRIORITY_SETUP, ep->ctrlq_idx);
1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558
	c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);

	return;
}

static void reject_cr(struct c4iw_dev *dev, u32 hwtid, __be32 peer_ip,
		      struct sk_buff *skb)
{
	PDBG("%s c4iw_dev %p tid %u peer_ip %x\n", __func__, dev, hwtid,
	     peer_ip);
	BUG_ON(skb_cloned(skb));
	skb_trim(skb, sizeof(struct cpl_tid_release));
	skb_get(skb);
	release_tid(&dev->rdev, hwtid, skb);
	return;
}

static void get_4tuple(struct cpl_pass_accept_req *req,
		       __be32 *local_ip, __be32 *peer_ip,
		       __be16 *local_port, __be16 *peer_port)
{
	int eth_len = G_ETH_HDR_LEN(be32_to_cpu(req->hdr_len));
	int ip_len = G_IP_HDR_LEN(be32_to_cpu(req->hdr_len));
	struct iphdr *ip = (struct iphdr *)((u8 *)(req + 1) + eth_len);
	struct tcphdr *tcp = (struct tcphdr *)
			     ((u8 *)(req + 1) + eth_len + ip_len);

	PDBG("%s saddr 0x%x daddr 0x%x sport %u dport %u\n", __func__,
	     ntohl(ip->saddr), ntohl(ip->daddr), ntohs(tcp->source),
	     ntohs(tcp->dest));

	*peer_ip = ip->saddr;
	*local_ip = ip->daddr;
	*peer_port = tcp->source;
	*local_port = tcp->dest;

	return;
}

1559 1560 1561 1562 1563 1564
static int import_ep(struct c4iw_ep *ep, __be32 peer_ip, struct dst_entry *dst,
		     struct c4iw_dev *cdev, bool clear_mpa_v1)
{
	struct neighbour *n;
	int err, step;

1565
	n = dst_neigh_lookup(dst, &peer_ip);
1566
	if (!n)
1567 1568 1569
		return -ENODEV;

	rcu_read_lock();
1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595
	err = -ENOMEM;
	if (n->dev->flags & IFF_LOOPBACK) {
		struct net_device *pdev;

		pdev = ip_dev_find(&init_net, peer_ip);
		ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t,
					n, pdev, 0);
		if (!ep->l2t)
			goto out;
		ep->mtu = pdev->mtu;
		ep->tx_chan = cxgb4_port_chan(pdev);
		ep->smac_idx = (cxgb4_port_viid(pdev) & 0x7F) << 1;
		step = cdev->rdev.lldi.ntxq /
			cdev->rdev.lldi.nchan;
		ep->txq_idx = cxgb4_port_idx(pdev) * step;
		step = cdev->rdev.lldi.nrxq /
			cdev->rdev.lldi.nchan;
		ep->ctrlq_idx = cxgb4_port_idx(pdev);
		ep->rss_qid = cdev->rdev.lldi.rxq_ids[
			cxgb4_port_idx(pdev) * step];
		dev_put(pdev);
	} else {
		ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t,
					n, n->dev, 0);
		if (!ep->l2t)
			goto out;
1596
		ep->mtu = dst_mtu(dst);
1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616
		ep->tx_chan = cxgb4_port_chan(n->dev);
		ep->smac_idx = (cxgb4_port_viid(n->dev) & 0x7F) << 1;
		step = cdev->rdev.lldi.ntxq /
			cdev->rdev.lldi.nchan;
		ep->txq_idx = cxgb4_port_idx(n->dev) * step;
		ep->ctrlq_idx = cxgb4_port_idx(n->dev);
		step = cdev->rdev.lldi.nrxq /
			cdev->rdev.lldi.nchan;
		ep->rss_qid = cdev->rdev.lldi.rxq_ids[
			cxgb4_port_idx(n->dev) * step];

		if (clear_mpa_v1) {
			ep->retry_with_mpa_v1 = 0;
			ep->tried_with_mpa_v1 = 0;
		}
	}
	err = 0;
out:
	rcu_read_unlock();

1617 1618
	neigh_release(n);

1619 1620 1621
	return err;
}

1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632
static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *child_ep, *parent_ep;
	struct cpl_pass_accept_req *req = cplhdr(skb);
	unsigned int stid = GET_POPEN_TID(ntohl(req->tos_stid));
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int hwtid = GET_TID(req);
	struct dst_entry *dst;
	struct rtable *rt;
	__be32 local_ip, peer_ip;
	__be16 local_port, peer_port;
1633
	int err;
1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653

	parent_ep = lookup_stid(t, stid);
	PDBG("%s parent ep %p tid %u\n", __func__, parent_ep, hwtid);

	get_4tuple(req, &local_ip, &peer_ip, &local_port, &peer_port);

	if (state_read(&parent_ep->com) != LISTEN) {
		printk(KERN_ERR "%s - listening ep not in LISTEN\n",
		       __func__);
		goto reject;
	}

	/* Find output route */
	rt = find_route(dev, local_ip, peer_ip, local_port, peer_port,
			GET_POPEN_TOS(ntohl(req->tos_stid)));
	if (!rt) {
		printk(KERN_ERR MOD "%s - failed to find dst entry!\n",
		       __func__);
		goto reject;
	}
1654
	dst = &rt->dst;
1655 1656 1657 1658

	child_ep = alloc_ep(sizeof(*child_ep), GFP_KERNEL);
	if (!child_ep) {
		printk(KERN_ERR MOD "%s - failed to allocate ep entry!\n",
1659 1660 1661 1662 1663
		       __func__);
		dst_release(dst);
		goto reject;
	}

1664 1665 1666
	err = import_ep(child_ep, peer_ip, dst, dev, false);
	if (err) {
		printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n",
1667 1668
		       __func__);
		dst_release(dst);
1669
		kfree(child_ep);
1670 1671
		goto reject;
	}
1672

1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688
	state_set(&child_ep->com, CONNECTING);
	child_ep->com.dev = dev;
	child_ep->com.cm_id = NULL;
	child_ep->com.local_addr.sin_family = PF_INET;
	child_ep->com.local_addr.sin_port = local_port;
	child_ep->com.local_addr.sin_addr.s_addr = local_ip;
	child_ep->com.remote_addr.sin_family = PF_INET;
	child_ep->com.remote_addr.sin_port = peer_port;
	child_ep->com.remote_addr.sin_addr.s_addr = peer_ip;
	c4iw_get_ep(&parent_ep->com);
	child_ep->parent_ep = parent_ep;
	child_ep->tos = GET_POPEN_TOS(ntohl(req->tos_stid));
	child_ep->dst = dst;
	child_ep->hwtid = hwtid;

	PDBG("%s tx_chan %u smac_idx %u rss_qid %u\n", __func__,
1689
	     child_ep->tx_chan, child_ep->smac_idx, child_ep->rss_qid);
1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731

	init_timer(&child_ep->timer);
	cxgb4_insert_tid(t, child_ep, hwtid);
	accept_cr(child_ep, peer_ip, skb, req);
	goto out;
reject:
	reject_cr(dev, hwtid, peer_ip, skb);
out:
	return 0;
}

static int pass_establish(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *ep;
	struct cpl_pass_establish *req = cplhdr(skb);
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int tid = GET_TID(req);

	ep = lookup_tid(t, tid);
	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	ep->snd_seq = be32_to_cpu(req->snd_isn);
	ep->rcv_seq = be32_to_cpu(req->rcv_isn);

	set_emss(ep, ntohs(req->tcp_opt));

	dst_confirm(ep->dst);
	state_set(&ep->com, MPA_REQ_WAIT);
	start_ep_timer(ep);
	send_flowc(ep, skb);

	return 0;
}

static int peer_close(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_peer_close *hdr = cplhdr(skb);
	struct c4iw_ep *ep;
	struct c4iw_qp_attributes attrs;
	int disconnect = 1;
	int release = 0;
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int tid = GET_TID(hdr);
S
Steve Wise 已提交
1732
	int ret;
1733 1734 1735 1736 1737

	ep = lookup_tid(t, tid);
	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	dst_confirm(ep->dst);

1738
	mutex_lock(&ep->com.mutex);
1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756
	switch (ep->com.state) {
	case MPA_REQ_WAIT:
		__state_set(&ep->com, CLOSING);
		break;
	case MPA_REQ_SENT:
		__state_set(&ep->com, CLOSING);
		connect_reply_upcall(ep, -ECONNRESET);
		break;
	case MPA_REQ_RCVD:

		/*
		 * We're gonna mark this puppy DEAD, but keep
		 * the reference on it until the ULP accepts or
		 * rejects the CR. Also wake up anyone waiting
		 * in rdma connection migration (see c4iw_accept_cr()).
		 */
		__state_set(&ep->com, CLOSING);
		PDBG("waking up ep %p tid %u\n", ep, ep->hwtid);
1757
		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
1758 1759 1760 1761
		break;
	case MPA_REP_SENT:
		__state_set(&ep->com, CLOSING);
		PDBG("waking up ep %p tid %u\n", ep, ep->hwtid);
1762
		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
1763 1764
		break;
	case FPDU_MODE:
1765
		start_ep_timer(ep);
1766
		__state_set(&ep->com, CLOSING);
1767
		attrs.next_state = C4IW_QP_STATE_CLOSING;
S
Steve Wise 已提交
1768
		ret = c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
1769
				       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
S
Steve Wise 已提交
1770 1771 1772 1773
		if (ret != -ECONNRESET) {
			peer_close_upcall(ep);
			disconnect = 1;
		}
1774 1775 1776 1777 1778 1779 1780 1781 1782
		break;
	case ABORTING:
		disconnect = 0;
		break;
	case CLOSING:
		__state_set(&ep->com, MORIBUND);
		disconnect = 0;
		break;
	case MORIBUND:
1783
		stop_ep_timer(ep);
1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799
		if (ep->com.cm_id && ep->com.qp) {
			attrs.next_state = C4IW_QP_STATE_IDLE;
			c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
				       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
		}
		close_complete_upcall(ep);
		__state_set(&ep->com, DEAD);
		release = 1;
		disconnect = 0;
		break;
	case DEAD:
		disconnect = 0;
		break;
	default:
		BUG_ON(1);
	}
1800
	mutex_unlock(&ep->com.mutex);
1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816
	if (disconnect)
		c4iw_ep_disconnect(ep, 0, GFP_KERNEL);
	if (release)
		release_ep_resources(ep);
	return 0;
}

/*
 * Returns whether an ABORT_REQ_RSS message is a negative advice.
 */
static int is_neg_adv_abort(unsigned int status)
{
	return status == CPL_ERR_RTX_NEG_ADVICE ||
	       status == CPL_ERR_PERSIST_NEG_ADVICE;
}

1817 1818 1819
static int c4iw_reconnect(struct c4iw_ep *ep)
{
	struct rtable *rt;
1820
	int err = 0;
1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847

	PDBG("%s qp %p cm_id %p\n", __func__, ep->com.qp, ep->com.cm_id);
	init_timer(&ep->timer);

	/*
	 * Allocate an active TID to initiate a TCP connection.
	 */
	ep->atid = cxgb4_alloc_atid(ep->com.dev->rdev.lldi.tids, ep);
	if (ep->atid == -1) {
		printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__);
		err = -ENOMEM;
		goto fail2;
	}

	/* find a route */
	rt = find_route(ep->com.dev,
			ep->com.cm_id->local_addr.sin_addr.s_addr,
			ep->com.cm_id->remote_addr.sin_addr.s_addr,
			ep->com.cm_id->local_addr.sin_port,
			ep->com.cm_id->remote_addr.sin_port, 0);
	if (!rt) {
		printk(KERN_ERR MOD "%s - cannot find route.\n", __func__);
		err = -EHOSTUNREACH;
		goto fail3;
	}
	ep->dst = &rt->dst;

1848 1849 1850
	err = import_ep(ep, ep->com.cm_id->remote_addr.sin_addr.s_addr,
			ep->dst, ep->com.dev, false);
	if (err) {
1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884
		printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__);
		goto fail4;
	}

	PDBG("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n",
	     __func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid,
	     ep->l2t->idx);

	state_set(&ep->com, CONNECTING);
	ep->tos = 0;

	/* send connect request to rnic */
	err = send_connect(ep);
	if (!err)
		goto out;

	cxgb4_l2t_release(ep->l2t);
fail4:
	dst_release(ep->dst);
fail3:
	cxgb4_free_atid(ep->com.dev->rdev.lldi.tids, ep->atid);
fail2:
	/*
	 * remember to send notification to upper layer.
	 * We are in here so the upper layer is not aware that this is
	 * re-connect attempt and so, upper layer is still waiting for
	 * response of 1st connect request.
	 */
	connect_reply_upcall(ep, -ECONNRESET);
	c4iw_put_ep(&ep->com);
out:
	return err;
}

1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904
static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_abort_req_rss *req = cplhdr(skb);
	struct c4iw_ep *ep;
	struct cpl_abort_rpl *rpl;
	struct sk_buff *rpl_skb;
	struct c4iw_qp_attributes attrs;
	int ret;
	int release = 0;
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int tid = GET_TID(req);

	ep = lookup_tid(t, tid);
	if (is_neg_adv_abort(req->status)) {
		PDBG("%s neg_adv_abort ep %p tid %u\n", __func__, ep,
		     ep->hwtid);
		return 0;
	}
	PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid,
	     ep->com.state);
1905 1906 1907

	/*
	 * Wake up any threads in rdma_init() or rdma_fini().
1908 1909
	 * However, this is not needed if com state is just
	 * MPA_REQ_SENT
1910
	 */
1911 1912
	if (ep->com.state != MPA_REQ_SENT)
		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
1913 1914

	mutex_lock(&ep->com.mutex);
1915 1916 1917 1918
	switch (ep->com.state) {
	case CONNECTING:
		break;
	case MPA_REQ_WAIT:
1919
		stop_ep_timer(ep);
1920 1921
		break;
	case MPA_REQ_SENT:
1922
		stop_ep_timer(ep);
1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937
		if (mpa_rev == 2 && ep->tried_with_mpa_v1)
			connect_reply_upcall(ep, -ECONNRESET);
		else {
			/*
			 * we just don't send notification upwards because we
			 * want to retry with mpa_v1 without upper layers even
			 * knowing it.
			 *
			 * do some housekeeping so as to re-initiate the
			 * connection
			 */
			PDBG("%s: mpa_rev=%d. Retrying with mpav1\n", __func__,
			     mpa_rev);
			ep->retry_with_mpa_v1 = 1;
		}
1938 1939 1940 1941 1942 1943 1944
		break;
	case MPA_REP_SENT:
		break;
	case MPA_REQ_RCVD:
		break;
	case MORIBUND:
	case CLOSING:
1945
		stop_ep_timer(ep);
1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963
		/*FALLTHROUGH*/
	case FPDU_MODE:
		if (ep->com.cm_id && ep->com.qp) {
			attrs.next_state = C4IW_QP_STATE_ERROR;
			ret = c4iw_modify_qp(ep->com.qp->rhp,
				     ep->com.qp, C4IW_QP_ATTR_NEXT_STATE,
				     &attrs, 1);
			if (ret)
				printk(KERN_ERR MOD
				       "%s - qp <- error failed!\n",
				       __func__);
		}
		peer_abort_upcall(ep);
		break;
	case ABORTING:
		break;
	case DEAD:
		PDBG("%s PEER_ABORT IN DEAD STATE!!!!\n", __func__);
1964
		mutex_unlock(&ep->com.mutex);
1965 1966 1967 1968 1969 1970 1971 1972
		return 0;
	default:
		BUG_ON(1);
		break;
	}
	dst_confirm(ep->dst);
	if (ep->com.state != ABORTING) {
		__state_set(&ep->com, DEAD);
1973 1974 1975
		/* we don't release if we want to retry with mpa_v1 */
		if (!ep->retry_with_mpa_v1)
			release = 1;
1976
	}
1977
	mutex_unlock(&ep->com.mutex);
1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994

	rpl_skb = get_skb(skb, sizeof(*rpl), GFP_KERNEL);
	if (!rpl_skb) {
		printk(KERN_ERR MOD "%s - cannot allocate skb!\n",
		       __func__);
		release = 1;
		goto out;
	}
	set_wr_txq(skb, CPL_PRIORITY_DATA, ep->txq_idx);
	rpl = (struct cpl_abort_rpl *) skb_put(rpl_skb, sizeof(*rpl));
	INIT_TP_WR(rpl, ep->hwtid);
	OPCODE_TID(rpl) = cpu_to_be32(MK_OPCODE_TID(CPL_ABORT_RPL, ep->hwtid));
	rpl->cmd = CPL_ABORT_NO_RST;
	c4iw_ofld_send(&ep->com.dev->rdev, rpl_skb);
out:
	if (release)
		release_ep_resources(ep);
1995 1996 1997 1998 1999 2000 2001 2002 2003

	/* retry with mpa-v1 */
	if (ep && ep->retry_with_mpa_v1) {
		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid);
		dst_release(ep->dst);
		cxgb4_l2t_release(ep->l2t);
		c4iw_reconnect(ep);
	}

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
	return 0;
}

static int close_con_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *ep;
	struct c4iw_qp_attributes attrs;
	struct cpl_close_con_rpl *rpl = cplhdr(skb);
	int release = 0;
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int tid = GET_TID(rpl);

	ep = lookup_tid(t, tid);

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	BUG_ON(!ep);

	/* The cm_id may be null if we failed to connect */
2022
	mutex_lock(&ep->com.mutex);
2023 2024 2025 2026 2027
	switch (ep->com.state) {
	case CLOSING:
		__state_set(&ep->com, MORIBUND);
		break;
	case MORIBUND:
2028
		stop_ep_timer(ep);
2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046
		if ((ep->com.cm_id) && (ep->com.qp)) {
			attrs.next_state = C4IW_QP_STATE_IDLE;
			c4iw_modify_qp(ep->com.qp->rhp,
					     ep->com.qp,
					     C4IW_QP_ATTR_NEXT_STATE,
					     &attrs, 1);
		}
		close_complete_upcall(ep);
		__state_set(&ep->com, DEAD);
		release = 1;
		break;
	case ABORTING:
	case DEAD:
		break;
	default:
		BUG_ON(1);
		break;
	}
2047
	mutex_unlock(&ep->com.mutex);
2048 2049 2050 2051 2052 2053 2054
	if (release)
		release_ep_resources(ep);
	return 0;
}

static int terminate(struct c4iw_dev *dev, struct sk_buff *skb)
{
2055
	struct cpl_rdma_terminate *rpl = cplhdr(skb);
2056
	struct tid_info *t = dev->rdev.lldi.tids;
2057 2058 2059
	unsigned int tid = GET_TID(rpl);
	struct c4iw_ep *ep;
	struct c4iw_qp_attributes attrs;
2060 2061

	ep = lookup_tid(t, tid);
2062
	BUG_ON(!ep);
2063

2064
	if (ep && ep->com.qp) {
2065 2066 2067 2068 2069 2070
		printk(KERN_WARNING MOD "TERM received tid %u qpid %u\n", tid,
		       ep->com.qp->wq.sq.qid);
		attrs.next_state = C4IW_QP_STATE_TERMINATE;
		c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
			       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
	} else
2071
		printk(KERN_WARNING MOD "TERM received tid %u no ep/qp\n", tid);
2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092

	return 0;
}

/*
 * Upcall from the adapter indicating data has been transmitted.
 * For us its just the single MPA request or reply.  We can now free
 * the skb holding the mpa message.
 */
static int fw4_ack(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct c4iw_ep *ep;
	struct cpl_fw4_ack *hdr = cplhdr(skb);
	u8 credits = hdr->credits;
	unsigned int tid = GET_TID(hdr);
	struct tid_info *t = dev->rdev.lldi.tids;


	ep = lookup_tid(t, tid);
	PDBG("%s ep %p tid %u credits %u\n", __func__, ep, ep->hwtid, credits);
	if (credits == 0) {
2093 2094
		PDBG("%s 0 credit ack ep %p tid %u state %u\n",
		     __func__, ep, ep->hwtid, state_read(&ep->com));
2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147
		return 0;
	}

	dst_confirm(ep->dst);
	if (ep->mpa_skb) {
		PDBG("%s last streaming msg ack ep %p tid %u state %u "
		     "initiator %u freeing skb\n", __func__, ep, ep->hwtid,
		     state_read(&ep->com), ep->mpa_attr.initiator ? 1 : 0);
		kfree_skb(ep->mpa_skb);
		ep->mpa_skb = NULL;
	}
	return 0;
}

int c4iw_reject_cr(struct iw_cm_id *cm_id, const void *pdata, u8 pdata_len)
{
	int err;
	struct c4iw_ep *ep = to_ep(cm_id);
	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);

	if (state_read(&ep->com) == DEAD) {
		c4iw_put_ep(&ep->com);
		return -ECONNRESET;
	}
	BUG_ON(state_read(&ep->com) != MPA_REQ_RCVD);
	if (mpa_rev == 0)
		abort_connection(ep, NULL, GFP_KERNEL);
	else {
		err = send_mpa_reject(ep, pdata, pdata_len);
		err = c4iw_ep_disconnect(ep, 0, GFP_KERNEL);
	}
	c4iw_put_ep(&ep->com);
	return 0;
}

int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
{
	int err;
	struct c4iw_qp_attributes attrs;
	enum c4iw_qp_attr_mask mask;
	struct c4iw_ep *ep = to_ep(cm_id);
	struct c4iw_dev *h = to_c4iw_dev(cm_id->device);
	struct c4iw_qp *qp = get_qhp(h, conn_param->qpn);

	PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid);
	if (state_read(&ep->com) == DEAD) {
		err = -ECONNRESET;
		goto err;
	}

	BUG_ON(state_read(&ep->com) != MPA_REQ_RCVD);
	BUG_ON(!qp);

2148 2149
	if ((conn_param->ord > c4iw_max_read_depth) ||
	    (conn_param->ird > c4iw_max_read_depth)) {
2150 2151 2152 2153 2154
		abort_connection(ep, NULL, GFP_KERNEL);
		err = -EINVAL;
		goto err;
	}

2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173
	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) {
		if (conn_param->ord > ep->ird) {
			ep->ird = conn_param->ird;
			ep->ord = conn_param->ord;
			send_mpa_reject(ep, conn_param->private_data,
					conn_param->private_data_len);
			abort_connection(ep, NULL, GFP_KERNEL);
			err = -ENOMEM;
			goto err;
		}
		if (conn_param->ird > ep->ord) {
			if (!ep->ord)
				conn_param->ird = 1;
			else {
				abort_connection(ep, NULL, GFP_KERNEL);
				err = -ENOMEM;
				goto err;
			}
		}
2174

2175
	}
2176 2177 2178
	ep->ird = conn_param->ird;
	ep->ord = conn_param->ord;

2179 2180 2181
	if (ep->mpa_attr.version != 2)
		if (peer2peer && ep->ird == 0)
			ep->ird = 1;
2182 2183 2184

	PDBG("%s %d ird %d ord %d\n", __func__, __LINE__, ep->ird, ep->ord);

2185 2186 2187 2188
	cm_id->add_ref(cm_id);
	ep->com.cm_id = cm_id;
	ep->com.qp = qp;

2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229
	/* bind QP to EP and move to RTS */
	attrs.mpa_attr = ep->mpa_attr;
	attrs.max_ird = ep->ird;
	attrs.max_ord = ep->ord;
	attrs.llp_stream_handle = ep;
	attrs.next_state = C4IW_QP_STATE_RTS;

	/* bind QP and TID with INIT_WR */
	mask = C4IW_QP_ATTR_NEXT_STATE |
			     C4IW_QP_ATTR_LLP_STREAM_HANDLE |
			     C4IW_QP_ATTR_MPA_ATTR |
			     C4IW_QP_ATTR_MAX_IRD |
			     C4IW_QP_ATTR_MAX_ORD;

	err = c4iw_modify_qp(ep->com.qp->rhp,
			     ep->com.qp, mask, &attrs, 1);
	if (err)
		goto err1;
	err = send_mpa_reply(ep, conn_param->private_data,
			     conn_param->private_data_len);
	if (err)
		goto err1;

	state_set(&ep->com, FPDU_MODE);
	established_upcall(ep);
	c4iw_put_ep(&ep->com);
	return 0;
err1:
	ep->com.cm_id = NULL;
	ep->com.qp = NULL;
	cm_id->rem_ref(cm_id);
err:
	c4iw_put_ep(&ep->com);
	return err;
}

int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
{
	struct c4iw_dev *dev = to_c4iw_dev(cm_id->device);
	struct c4iw_ep *ep;
	struct rtable *rt;
2230
	int err = 0;
2231

2232 2233 2234 2235 2236
	if ((conn_param->ord > c4iw_max_read_depth) ||
	    (conn_param->ird > c4iw_max_read_depth)) {
		err = -EINVAL;
		goto out;
	}
2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288
	ep = alloc_ep(sizeof(*ep), GFP_KERNEL);
	if (!ep) {
		printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__);
		err = -ENOMEM;
		goto out;
	}
	init_timer(&ep->timer);
	ep->plen = conn_param->private_data_len;
	if (ep->plen)
		memcpy(ep->mpa_pkt + sizeof(struct mpa_message),
		       conn_param->private_data, ep->plen);
	ep->ird = conn_param->ird;
	ep->ord = conn_param->ord;

	if (peer2peer && ep->ord == 0)
		ep->ord = 1;

	cm_id->add_ref(cm_id);
	ep->com.dev = dev;
	ep->com.cm_id = cm_id;
	ep->com.qp = get_qhp(dev, conn_param->qpn);
	BUG_ON(!ep->com.qp);
	PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn,
	     ep->com.qp, cm_id);

	/*
	 * Allocate an active TID to initiate a TCP connection.
	 */
	ep->atid = cxgb4_alloc_atid(dev->rdev.lldi.tids, ep);
	if (ep->atid == -1) {
		printk(KERN_ERR MOD "%s - cannot alloc atid.\n", __func__);
		err = -ENOMEM;
		goto fail2;
	}

	PDBG("%s saddr 0x%x sport 0x%x raddr 0x%x rport 0x%x\n", __func__,
	     ntohl(cm_id->local_addr.sin_addr.s_addr),
	     ntohs(cm_id->local_addr.sin_port),
	     ntohl(cm_id->remote_addr.sin_addr.s_addr),
	     ntohs(cm_id->remote_addr.sin_port));

	/* find a route */
	rt = find_route(dev,
			cm_id->local_addr.sin_addr.s_addr,
			cm_id->remote_addr.sin_addr.s_addr,
			cm_id->local_addr.sin_port,
			cm_id->remote_addr.sin_port, 0);
	if (!rt) {
		printk(KERN_ERR MOD "%s - cannot find route.\n", __func__);
		err = -EHOSTUNREACH;
		goto fail3;
	}
2289
	ep->dst = &rt->dst;
2290

2291 2292 2293
	err = import_ep(ep, cm_id->remote_addr.sin_addr.s_addr,
			ep->dst, ep->com.dev, true);
	if (err) {
2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350
		printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__);
		goto fail4;
	}

	PDBG("%s txq_idx %u tx_chan %u smac_idx %u rss_qid %u l2t_idx %u\n",
		__func__, ep->txq_idx, ep->tx_chan, ep->smac_idx, ep->rss_qid,
		ep->l2t->idx);

	state_set(&ep->com, CONNECTING);
	ep->tos = 0;
	ep->com.local_addr = cm_id->local_addr;
	ep->com.remote_addr = cm_id->remote_addr;

	/* send connect request to rnic */
	err = send_connect(ep);
	if (!err)
		goto out;

	cxgb4_l2t_release(ep->l2t);
fail4:
	dst_release(ep->dst);
fail3:
	cxgb4_free_atid(ep->com.dev->rdev.lldi.tids, ep->atid);
fail2:
	cm_id->rem_ref(cm_id);
	c4iw_put_ep(&ep->com);
out:
	return err;
}

int c4iw_create_listen(struct iw_cm_id *cm_id, int backlog)
{
	int err = 0;
	struct c4iw_dev *dev = to_c4iw_dev(cm_id->device);
	struct c4iw_listen_ep *ep;


	might_sleep();

	ep = alloc_ep(sizeof(*ep), GFP_KERNEL);
	if (!ep) {
		printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __func__);
		err = -ENOMEM;
		goto fail1;
	}
	PDBG("%s ep %p\n", __func__, ep);
	cm_id->add_ref(cm_id);
	ep->com.cm_id = cm_id;
	ep->com.dev = dev;
	ep->backlog = backlog;
	ep->com.local_addr = cm_id->local_addr;

	/*
	 * Allocate a server TID.
	 */
	ep->stid = cxgb4_alloc_stid(dev->rdev.lldi.tids, PF_INET, ep);
	if (ep->stid == -1) {
2351
		printk(KERN_ERR MOD "%s - cannot alloc stid.\n", __func__);
2352 2353 2354 2355 2356
		err = -ENOMEM;
		goto fail2;
	}

	state_set(&ep->com, LISTEN);
2357
	c4iw_init_wr_wait(&ep->com.wr_wait);
2358 2359 2360 2361 2362 2363 2364 2365
	err = cxgb4_create_server(ep->com.dev->rdev.lldi.ports[0], ep->stid,
				  ep->com.local_addr.sin_addr.s_addr,
				  ep->com.local_addr.sin_port,
				  ep->com.dev->rdev.lldi.rxq_ids[0]);
	if (err)
		goto fail3;

	/* wait for pass_open_rpl */
2366 2367
	err = c4iw_wait_for_reply(&ep->com.dev->rdev, &ep->com.wr_wait, 0, 0,
				  __func__);
2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390
	if (!err) {
		cm_id->provider_data = ep;
		goto out;
	}
fail3:
	cxgb4_free_stid(ep->com.dev->rdev.lldi.tids, ep->stid, PF_INET);
fail2:
	cm_id->rem_ref(cm_id);
	c4iw_put_ep(&ep->com);
fail1:
out:
	return err;
}

int c4iw_destroy_listen(struct iw_cm_id *cm_id)
{
	int err;
	struct c4iw_listen_ep *ep = to_listen_ep(cm_id);

	PDBG("%s ep %p\n", __func__, ep);

	might_sleep();
	state_set(&ep->com, DEAD);
2391
	c4iw_init_wr_wait(&ep->com.wr_wait);
2392 2393 2394
	err = listen_stop(ep);
	if (err)
		goto done;
2395 2396
	err = c4iw_wait_for_reply(&ep->com.dev->rdev, &ep->com.wr_wait, 0, 0,
				  __func__);
2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410
	cxgb4_free_stid(ep->com.dev->rdev.lldi.tids, ep->stid, PF_INET);
done:
	cm_id->rem_ref(cm_id);
	c4iw_put_ep(&ep->com);
	return err;
}

int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp)
{
	int ret = 0;
	int close = 0;
	int fatal = 0;
	struct c4iw_rdev *rdev;

2411
	mutex_lock(&ep->com.mutex);
2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432

	PDBG("%s ep %p state %s, abrupt %d\n", __func__, ep,
	     states[ep->com.state], abrupt);

	rdev = &ep->com.dev->rdev;
	if (c4iw_fatal_error(rdev)) {
		fatal = 1;
		close_complete_upcall(ep);
		ep->com.state = DEAD;
	}
	switch (ep->com.state) {
	case MPA_REQ_WAIT:
	case MPA_REQ_SENT:
	case MPA_REQ_RCVD:
	case MPA_REP_SENT:
	case FPDU_MODE:
		close = 1;
		if (abrupt)
			ep->com.state = ABORTING;
		else {
			ep->com.state = CLOSING;
2433
			start_ep_timer(ep);
2434 2435 2436 2437 2438 2439 2440
		}
		set_bit(CLOSE_SENT, &ep->com.flags);
		break;
	case CLOSING:
		if (!test_and_set_bit(CLOSE_SENT, &ep->com.flags)) {
			close = 1;
			if (abrupt) {
2441
				stop_ep_timer(ep);
2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458
				ep->com.state = ABORTING;
			} else
				ep->com.state = MORIBUND;
		}
		break;
	case MORIBUND:
	case ABORTING:
	case DEAD:
		PDBG("%s ignoring disconnect ep %p state %u\n",
		     __func__, ep, ep->com.state);
		break;
	default:
		BUG();
		break;
	}

	if (close) {
S
Steve Wise 已提交
2459 2460 2461 2462
		if (abrupt) {
			close_complete_upcall(ep);
			ret = send_abort(ep, NULL, gfp);
		} else
2463 2464 2465 2466
			ret = send_halfclose(ep, gfp);
		if (ret)
			fatal = 1;
	}
S
Steve Wise 已提交
2467
	mutex_unlock(&ep->com.mutex);
2468 2469 2470 2471 2472
	if (fatal)
		release_ep_resources(ep);
	return ret;
}

2473 2474 2475 2476 2477 2478 2479
static int async_event(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_fw6_msg *rpl = cplhdr(skb);
	c4iw_ev_dispatch(dev, (struct t4_cqe *)&rpl->data[0]);
	return 0;
}

2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497
/*
 * These are the real handlers that are called from a
 * work queue.
 */
static c4iw_handler_func work_handlers[NUM_CPL_CMDS] = {
	[CPL_ACT_ESTABLISH] = act_establish,
	[CPL_ACT_OPEN_RPL] = act_open_rpl,
	[CPL_RX_DATA] = rx_data,
	[CPL_ABORT_RPL_RSS] = abort_rpl,
	[CPL_ABORT_RPL] = abort_rpl,
	[CPL_PASS_OPEN_RPL] = pass_open_rpl,
	[CPL_CLOSE_LISTSRV_RPL] = close_listsrv_rpl,
	[CPL_PASS_ACCEPT_REQ] = pass_accept_req,
	[CPL_PASS_ESTABLISH] = pass_establish,
	[CPL_PEER_CLOSE] = peer_close,
	[CPL_ABORT_REQ_RSS] = peer_abort,
	[CPL_CLOSE_CON_RPL] = close_con_rpl,
	[CPL_RDMA_TERMINATE] = terminate,
2498 2499
	[CPL_FW4_ACK] = fw4_ack,
	[CPL_FW6_MSG] = async_event
2500 2501 2502 2503 2504 2505 2506
};

static void process_timeout(struct c4iw_ep *ep)
{
	struct c4iw_qp_attributes attrs;
	int abort = 1;

2507
	mutex_lock(&ep->com.mutex);
2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533
	PDBG("%s ep %p tid %u state %d\n", __func__, ep, ep->hwtid,
	     ep->com.state);
	switch (ep->com.state) {
	case MPA_REQ_SENT:
		__state_set(&ep->com, ABORTING);
		connect_reply_upcall(ep, -ETIMEDOUT);
		break;
	case MPA_REQ_WAIT:
		__state_set(&ep->com, ABORTING);
		break;
	case CLOSING:
	case MORIBUND:
		if (ep->com.cm_id && ep->com.qp) {
			attrs.next_state = C4IW_QP_STATE_ERROR;
			c4iw_modify_qp(ep->com.qp->rhp,
				     ep->com.qp, C4IW_QP_ATTR_NEXT_STATE,
				     &attrs, 1);
		}
		__state_set(&ep->com, ABORTING);
		break;
	default:
		printk(KERN_ERR "%s unexpected state ep %p tid %u state %u\n",
			__func__, ep, ep->hwtid, ep->com.state);
		WARN_ON(1);
		abort = 0;
	}
2534
	mutex_unlock(&ep->com.mutex);
2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561
	if (abort)
		abort_connection(ep, NULL, GFP_KERNEL);
	c4iw_put_ep(&ep->com);
}

static void process_timedout_eps(void)
{
	struct c4iw_ep *ep;

	spin_lock_irq(&timeout_lock);
	while (!list_empty(&timeout_list)) {
		struct list_head *tmp;

		tmp = timeout_list.next;
		list_del(tmp);
		spin_unlock_irq(&timeout_lock);
		ep = list_entry(tmp, struct c4iw_ep, entry);
		process_timeout(ep);
		spin_lock_irq(&timeout_lock);
	}
	spin_unlock_irq(&timeout_lock);
}

static void process_work(struct work_struct *work)
{
	struct sk_buff *skb = NULL;
	struct c4iw_dev *dev;
2562
	struct cpl_act_establish *rpl;
2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590
	unsigned int opcode;
	int ret;

	while ((skb = skb_dequeue(&rxq))) {
		rpl = cplhdr(skb);
		dev = *((struct c4iw_dev **) (skb->cb + sizeof(void *)));
		opcode = rpl->ot.opcode;

		BUG_ON(!work_handlers[opcode]);
		ret = work_handlers[opcode](dev, skb);
		if (!ret)
			kfree_skb(skb);
	}
	process_timedout_eps();
}

static DECLARE_WORK(skb_work, process_work);

static void ep_timeout(unsigned long arg)
{
	struct c4iw_ep *ep = (struct c4iw_ep *)arg;

	spin_lock(&timeout_lock);
	list_add_tail(&ep->entry, &timeout_list);
	spin_unlock(&timeout_lock);
	queue_work(workq, &skb_work);
}

2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617
/*
 * All the CM events are handled on a work queue to have a safe context.
 */
static int sched(struct c4iw_dev *dev, struct sk_buff *skb)
{

	/*
	 * Save dev in the skb->cb area.
	 */
	*((struct c4iw_dev **) (skb->cb + sizeof(void *))) = dev;

	/*
	 * Queue the skb and schedule the worker thread.
	 */
	skb_queue_tail(&rxq, skb);
	queue_work(workq, &skb_work);
	return 0;
}

static int set_tcb_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_set_tcb_rpl *rpl = cplhdr(skb);

	if (rpl->status != CPL_ERR_NONE) {
		printk(KERN_ERR MOD "Unexpected SET_TCB_RPL status %u "
		       "for tid %u\n", rpl->status, GET_TID(rpl));
	}
2618
	kfree_skb(skb);
2619 2620 2621
	return 0;
}

2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632
static int fw6_msg(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_fw6_msg *rpl = cplhdr(skb);
	struct c4iw_wr_wait *wr_waitp;
	int ret;

	PDBG("%s type %u\n", __func__, rpl->type);

	switch (rpl->type) {
	case 1:
		ret = (int)((be64_to_cpu(rpl->data[0]) >> 8) & 0xff);
2633
		wr_waitp = (struct c4iw_wr_wait *)(__force unsigned long) rpl->data[1];
2634
		PDBG("%s wr_waitp %p ret %u\n", __func__, wr_waitp, ret);
2635 2636
		if (wr_waitp)
			c4iw_wake_up(wr_waitp, ret ? -ret : 0);
2637
		kfree_skb(skb);
2638 2639
		break;
	case 2:
2640
		sched(dev, skb);
2641 2642 2643 2644
		break;
	default:
		printk(KERN_ERR MOD "%s unexpected fw6 msg type %u\n", __func__,
		       rpl->type);
2645
		kfree_skb(skb);
2646 2647 2648 2649 2650
		break;
	}
	return 0;
}

S
Steve Wise 已提交
2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669
static int peer_abort_intr(struct c4iw_dev *dev, struct sk_buff *skb)
{
	struct cpl_abort_req_rss *req = cplhdr(skb);
	struct c4iw_ep *ep;
	struct tid_info *t = dev->rdev.lldi.tids;
	unsigned int tid = GET_TID(req);

	ep = lookup_tid(t, tid);
	if (is_neg_adv_abort(req->status)) {
		PDBG("%s neg_adv_abort ep %p tid %u\n", __func__, ep,
		     ep->hwtid);
		kfree_skb(skb);
		return 0;
	}
	PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid,
	     ep->com.state);

	/*
	 * Wake up any threads in rdma_init() or rdma_fini().
2670 2671
	 * However, this is not needed if com state is just
	 * MPA_REQ_SENT
S
Steve Wise 已提交
2672
	 */
2673 2674
	if (ep->com.state != MPA_REQ_SENT)
		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
S
Steve Wise 已提交
2675 2676 2677 2678
	sched(dev, skb);
	return 0;
}

2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694
/*
 * Most upcalls from the T4 Core go to sched() to
 * schedule the processing on a work queue.
 */
c4iw_handler_func c4iw_handlers[NUM_CPL_CMDS] = {
	[CPL_ACT_ESTABLISH] = sched,
	[CPL_ACT_OPEN_RPL] = sched,
	[CPL_RX_DATA] = sched,
	[CPL_ABORT_RPL_RSS] = sched,
	[CPL_ABORT_RPL] = sched,
	[CPL_PASS_OPEN_RPL] = sched,
	[CPL_CLOSE_LISTSRV_RPL] = sched,
	[CPL_PASS_ACCEPT_REQ] = sched,
	[CPL_PASS_ESTABLISH] = sched,
	[CPL_PEER_CLOSE] = sched,
	[CPL_CLOSE_CON_RPL] = sched,
S
Steve Wise 已提交
2695
	[CPL_ABORT_REQ_RSS] = peer_abort_intr,
2696 2697 2698 2699 2700 2701
	[CPL_RDMA_TERMINATE] = sched,
	[CPL_FW4_ACK] = sched,
	[CPL_SET_TCB_RPL] = set_tcb_rpl,
	[CPL_FW6_MSG] = fw6_msg
};

2702 2703
int __init c4iw_cm_init(void)
{
2704
	spin_lock_init(&timeout_lock);
2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715
	skb_queue_head_init(&rxq);

	workq = create_singlethread_workqueue("iw_cxgb4");
	if (!workq)
		return -ENOMEM;

	return 0;
}

void __exit c4iw_cm_term(void)
{
2716
	WARN_ON(!list_empty(&timeout_list));
2717 2718 2719
	flush_workqueue(workq);
	destroy_workqueue(workq);
}