en_netdev.c 69.7 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
/*
 * Copyright (c) 2007 Mellanox Technologies. 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/etherdevice.h>
#include <linux/tcp.h>
#include <linux/if_vlan.h>
#include <linux/delay.h>
38
#include <linux/slab.h>
39 40
#include <linux/hash.h>
#include <net/ip.h>
41
#include <net/busy_poll.h>
O
Or Gerlitz 已提交
42
#include <net/vxlan.h>
43 44 45 46 47 48 49 50 51

#include <linux/mlx4/driver.h>
#include <linux/mlx4/device.h>
#include <linux/mlx4/cmd.h>
#include <linux/mlx4/cq.h>

#include "mlx4_en.h"
#include "en_port.h"

52
int mlx4_en_setup_tc(struct net_device *dev, u8 up)
53
{
54 55
	struct mlx4_en_priv *priv = netdev_priv(dev);
	int i;
56
	unsigned int offset = 0;
57 58

	if (up && up != MLX4_EN_NUM_UP)
59 60
		return -EINVAL;

61 62 63 64
	netdev_set_num_tc(dev, up);

	/* Partition Tx queues evenly amongst UP's */
	for (i = 0; i < up; i++) {
65 66
		netdev_set_tc_queue(dev, i, priv->num_tx_rings_p_up, offset);
		offset += priv->num_tx_rings_p_up;
67 68
	}

69 70 71
	return 0;
}

72
#ifdef CONFIG_NET_RX_BUSY_POLL
73 74 75 76 77 78
/* must be called with local_bh_disable()d */
static int mlx4_en_low_latency_recv(struct napi_struct *napi)
{
	struct mlx4_en_cq *cq = container_of(napi, struct mlx4_en_cq, napi);
	struct net_device *dev = cq->dev;
	struct mlx4_en_priv *priv = netdev_priv(dev);
79
	struct mlx4_en_rx_ring *rx_ring = priv->rx_ring[cq->ring];
80 81 82 83 84 85 86 87 88
	int done;

	if (!priv->port_up)
		return LL_FLUSH_FAILED;

	if (!mlx4_en_cq_lock_poll(cq))
		return LL_FLUSH_BUSY;

	done = mlx4_en_process_rx_cq(dev, cq, 4);
89 90 91 92
	if (likely(done))
		rx_ring->cleaned += done;
	else
		rx_ring->misses++;
93 94 95 96 97

	mlx4_en_cq_unlock_poll(cq);

	return done;
}
98
#endif	/* CONFIG_NET_RX_BUSY_POLL */
99

100 101 102 103 104 105
#ifdef CONFIG_RFS_ACCEL

struct mlx4_en_filter {
	struct list_head next;
	struct work_struct work;

E
Eyal Perry 已提交
106
	u8     ip_proto;
107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
	__be32 src_ip;
	__be32 dst_ip;
	__be16 src_port;
	__be16 dst_port;

	int rxq_index;
	struct mlx4_en_priv *priv;
	u32 flow_id;			/* RFS infrastructure id */
	int id;				/* mlx4_en driver id */
	u64 reg_id;			/* Flow steering API id */
	u8 activated;			/* Used to prevent expiry before filter
					 * is attached
					 */
	struct hlist_node filter_chain;
};

static void mlx4_en_filter_rfs_expire(struct mlx4_en_priv *priv);

E
Eyal Perry 已提交
125 126 127 128 129 130 131 132 133 134 135 136
static enum mlx4_net_trans_rule_id mlx4_ip_proto_to_trans_rule_id(u8 ip_proto)
{
	switch (ip_proto) {
	case IPPROTO_UDP:
		return MLX4_NET_TRANS_RULE_ID_UDP;
	case IPPROTO_TCP:
		return MLX4_NET_TRANS_RULE_ID_TCP;
	default:
		return -EPROTONOSUPPORT;
	}
};

137 138 139 140 141 142
static void mlx4_en_filter_work(struct work_struct *work)
{
	struct mlx4_en_filter *filter = container_of(work,
						     struct mlx4_en_filter,
						     work);
	struct mlx4_en_priv *priv = filter->priv;
E
Eyal Perry 已提交
143 144
	struct mlx4_spec_list spec_tcp_udp = {
		.id = mlx4_ip_proto_to_trans_rule_id(filter->ip_proto),
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
		{
			.tcp_udp = {
				.dst_port = filter->dst_port,
				.dst_port_msk = (__force __be16)-1,
				.src_port = filter->src_port,
				.src_port_msk = (__force __be16)-1,
			},
		},
	};
	struct mlx4_spec_list spec_ip = {
		.id = MLX4_NET_TRANS_RULE_ID_IPV4,
		{
			.ipv4 = {
				.dst_ip = filter->dst_ip,
				.dst_ip_msk = (__force __be32)-1,
				.src_ip = filter->src_ip,
				.src_ip_msk = (__force __be32)-1,
			},
		},
	};
	struct mlx4_spec_list spec_eth = {
		.id = MLX4_NET_TRANS_RULE_ID_ETH,
	};
	struct mlx4_net_trans_rule rule = {
		.list = LIST_HEAD_INIT(rule.list),
		.queue_mode = MLX4_NET_TRANS_Q_LIFO,
		.exclusive = 1,
		.allow_loopback = 1,
173
		.promisc_mode = MLX4_FS_REGULAR,
174 175 176 177 178 179
		.port = priv->port,
		.priority = MLX4_DOMAIN_RFS,
	};
	int rc;
	__be64 mac_mask = cpu_to_be64(MLX4_MAC_MASK << 16);

E
Eyal Perry 已提交
180 181 182 183 184
	if (spec_tcp_udp.id < 0) {
		en_warn(priv, "RFS: ignoring unsupported ip protocol (%d)\n",
			filter->ip_proto);
		goto ignore;
	}
185 186
	list_add_tail(&spec_eth.list, &rule.list);
	list_add_tail(&spec_ip.list, &rule.list);
E
Eyal Perry 已提交
187
	list_add_tail(&spec_tcp_udp.list, &rule.list);
188 189

	rule.qpn = priv->rss_map.qps[filter->rxq_index].qpn;
190
	memcpy(spec_eth.eth.dst_mac, priv->dev->dev_addr, ETH_ALEN);
191 192 193 194 195 196 197 198 199 200 201 202 203 204
	memcpy(spec_eth.eth.dst_mac_msk, &mac_mask, ETH_ALEN);

	filter->activated = 0;

	if (filter->reg_id) {
		rc = mlx4_flow_detach(priv->mdev->dev, filter->reg_id);
		if (rc && rc != -ENOENT)
			en_err(priv, "Error detaching flow. rc = %d\n", rc);
	}

	rc = mlx4_flow_attach(priv->mdev->dev, &rule, &filter->reg_id);
	if (rc)
		en_err(priv, "Error attaching flow. err = %d\n", rc);

E
Eyal Perry 已提交
205
ignore:
206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228
	mlx4_en_filter_rfs_expire(priv);

	filter->activated = 1;
}

static inline struct hlist_head *
filter_hash_bucket(struct mlx4_en_priv *priv, __be32 src_ip, __be32 dst_ip,
		   __be16 src_port, __be16 dst_port)
{
	unsigned long l;
	int bucket_idx;

	l = (__force unsigned long)src_port |
	    ((__force unsigned long)dst_port << 2);
	l ^= (__force unsigned long)(src_ip ^ dst_ip);

	bucket_idx = hash_long(l, MLX4_EN_FILTER_HASH_SHIFT);

	return &priv->filter_hash[bucket_idx];
}

static struct mlx4_en_filter *
mlx4_en_filter_alloc(struct mlx4_en_priv *priv, int rxq_index, __be32 src_ip,
E
Eyal Perry 已提交
229 230
		     __be32 dst_ip, u8 ip_proto, __be16 src_port,
		     __be16 dst_port, u32 flow_id)
231 232 233 234 235 236 237 238 239 240 241 242 243
{
	struct mlx4_en_filter *filter = NULL;

	filter = kzalloc(sizeof(struct mlx4_en_filter), GFP_ATOMIC);
	if (!filter)
		return NULL;

	filter->priv = priv;
	filter->rxq_index = rxq_index;
	INIT_WORK(&filter->work, mlx4_en_filter_work);

	filter->src_ip = src_ip;
	filter->dst_ip = dst_ip;
E
Eyal Perry 已提交
244
	filter->ip_proto = ip_proto;
245 246 247 248 249
	filter->src_port = src_port;
	filter->dst_port = dst_port;

	filter->flow_id = flow_id;

250
	filter->id = priv->last_filter_id++ % RPS_NO_FILTER;
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275

	list_add_tail(&filter->next, &priv->filters);
	hlist_add_head(&filter->filter_chain,
		       filter_hash_bucket(priv, src_ip, dst_ip, src_port,
					  dst_port));

	return filter;
}

static void mlx4_en_filter_free(struct mlx4_en_filter *filter)
{
	struct mlx4_en_priv *priv = filter->priv;
	int rc;

	list_del(&filter->next);

	rc = mlx4_flow_detach(priv->mdev->dev, filter->reg_id);
	if (rc && rc != -ENOENT)
		en_err(priv, "Error detaching flow. rc = %d\n", rc);

	kfree(filter);
}

static inline struct mlx4_en_filter *
mlx4_en_filter_find(struct mlx4_en_priv *priv, __be32 src_ip, __be32 dst_ip,
E
Eyal Perry 已提交
276
		    u8 ip_proto, __be16 src_port, __be16 dst_port)
277 278 279 280
{
	struct mlx4_en_filter *filter;
	struct mlx4_en_filter *ret = NULL;

281
	hlist_for_each_entry(filter,
282 283 284 285 286
			     filter_hash_bucket(priv, src_ip, dst_ip,
						src_port, dst_port),
			     filter_chain) {
		if (filter->src_ip == src_ip &&
		    filter->dst_ip == dst_ip &&
E
Eyal Perry 已提交
287
		    filter->ip_proto == ip_proto &&
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305
		    filter->src_port == src_port &&
		    filter->dst_port == dst_port) {
			ret = filter;
			break;
		}
	}

	return ret;
}

static int
mlx4_en_filter_rfs(struct net_device *net_dev, const struct sk_buff *skb,
		   u16 rxq_index, u32 flow_id)
{
	struct mlx4_en_priv *priv = netdev_priv(net_dev);
	struct mlx4_en_filter *filter;
	const struct iphdr *ip;
	const __be16 *ports;
E
Eyal Perry 已提交
306
	u8 ip_proto;
307 308 309 310 311 312 313 314 315 316 317 318 319 320
	__be32 src_ip;
	__be32 dst_ip;
	__be16 src_port;
	__be16 dst_port;
	int nhoff = skb_network_offset(skb);
	int ret = 0;

	if (skb->protocol != htons(ETH_P_IP))
		return -EPROTONOSUPPORT;

	ip = (const struct iphdr *)(skb->data + nhoff);
	if (ip_is_fragment(ip))
		return -EPROTONOSUPPORT;

E
Eyal Perry 已提交
321 322
	if ((ip->protocol != IPPROTO_TCP) && (ip->protocol != IPPROTO_UDP))
		return -EPROTONOSUPPORT;
323 324
	ports = (const __be16 *)(skb->data + nhoff + 4 * ip->ihl);

E
Eyal Perry 已提交
325
	ip_proto = ip->protocol;
326 327 328 329 330 331
	src_ip = ip->saddr;
	dst_ip = ip->daddr;
	src_port = ports[0];
	dst_port = ports[1];

	spin_lock_bh(&priv->filters_lock);
E
Eyal Perry 已提交
332 333
	filter = mlx4_en_filter_find(priv, src_ip, dst_ip, ip_proto,
				     src_port, dst_port);
334 335 336 337 338 339 340
	if (filter) {
		if (filter->rxq_index == rxq_index)
			goto out;

		filter->rxq_index = rxq_index;
	} else {
		filter = mlx4_en_filter_alloc(priv, rxq_index,
E
Eyal Perry 已提交
341
					      src_ip, dst_ip, ip_proto,
342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358
					      src_port, dst_port, flow_id);
		if (!filter) {
			ret = -ENOMEM;
			goto err;
		}
	}

	queue_work(priv->mdev->workqueue, &filter->work);

out:
	ret = filter->id;
err:
	spin_unlock_bh(&priv->filters_lock);

	return ret;
}

359
void mlx4_en_cleanup_filters(struct mlx4_en_priv *priv)
360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
{
	struct mlx4_en_filter *filter, *tmp;
	LIST_HEAD(del_list);

	spin_lock_bh(&priv->filters_lock);
	list_for_each_entry_safe(filter, tmp, &priv->filters, next) {
		list_move(&filter->next, &del_list);
		hlist_del(&filter->filter_chain);
	}
	spin_unlock_bh(&priv->filters_lock);

	list_for_each_entry_safe(filter, tmp, &del_list, next) {
		cancel_work_sync(&filter->work);
		mlx4_en_filter_free(filter);
	}
}

static void mlx4_en_filter_rfs_expire(struct mlx4_en_priv *priv)
{
	struct mlx4_en_filter *filter = NULL, *tmp, *last_filter = NULL;
	LIST_HEAD(del_list);
	int i = 0;

	spin_lock_bh(&priv->filters_lock);
	list_for_each_entry_safe(filter, tmp, &priv->filters, next) {
		if (i > MLX4_EN_FILTER_EXPIRY_QUOTA)
			break;

		if (filter->activated &&
		    !work_pending(&filter->work) &&
		    rps_may_expire_flow(priv->dev,
					filter->rxq_index, filter->flow_id,
					filter->id)) {
			list_move(&filter->next, &del_list);
			hlist_del(&filter->filter_chain);
		} else
			last_filter = filter;

		i++;
	}

	if (last_filter && (&last_filter->next != priv->filters.next))
		list_move(&priv->filters, &last_filter->next);

	spin_unlock_bh(&priv->filters_lock);

	list_for_each_entry_safe(filter, tmp, &del_list, next)
		mlx4_en_filter_free(filter);
}
#endif

411 412
static int mlx4_en_vlan_rx_add_vid(struct net_device *dev,
				   __be16 proto, u16 vid)
413 414 415 416
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	int err;
E
Eli Cohen 已提交
417
	int idx;
418

J
Jiri Pirko 已提交
419
	en_dbg(HW, priv, "adding VLAN:%d\n", vid);
420

J
Jiri Pirko 已提交
421
	set_bit(vid, priv->active_vlans);
422 423 424 425

	/* Add VID to port VLAN filter */
	mutex_lock(&mdev->state_lock);
	if (mdev->device_up && priv->port_up) {
J
Jiri Pirko 已提交
426
		err = mlx4_SET_VLAN_FLTR(mdev->dev, priv);
427
		if (err)
428
			en_err(priv, "Failed configuring VLAN filter\n");
429
	}
E
Eli Cohen 已提交
430
	if (mlx4_register_vlan(mdev->dev, priv->port, vid, &idx))
431
		en_dbg(HW, priv, "failed adding vlan %d\n", vid);
432
	mutex_unlock(&mdev->state_lock);
E
Eli Cohen 已提交
433

434
	return 0;
435 436
}

437 438
static int mlx4_en_vlan_rx_kill_vid(struct net_device *dev,
				    __be16 proto, u16 vid)
439 440 441 442 443
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	int err;

J
Jiri Pirko 已提交
444
	en_dbg(HW, priv, "Killing VID:%d\n", vid);
445

J
Jiri Pirko 已提交
446
	clear_bit(vid, priv->active_vlans);
447 448 449

	/* Remove VID from port VLAN filter */
	mutex_lock(&mdev->state_lock);
450
	mlx4_unregister_vlan(mdev->dev, priv->port, vid);
E
Eli Cohen 已提交
451

452
	if (mdev->device_up && priv->port_up) {
J
Jiri Pirko 已提交
453
		err = mlx4_SET_VLAN_FLTR(mdev->dev, priv);
454
		if (err)
455
			en_err(priv, "Failed configuring VLAN filter\n");
456 457
	}
	mutex_unlock(&mdev->state_lock);
458 459

	return 0;
460 461
}

462 463
static void mlx4_en_u64_to_mac(unsigned char dst_mac[ETH_ALEN + 2], u64 src_mac)
{
464 465
	int i;
	for (i = ETH_ALEN - 1; i >= 0; --i) {
466 467 468 469 470 471
		dst_mac[i] = src_mac & 0xff;
		src_mac >>= 8;
	}
	memset(&dst_mac[ETH_ALEN], 0, 2);
}

472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518

static int mlx4_en_tunnel_steer_add(struct mlx4_en_priv *priv, unsigned char *addr,
				    int qpn, u64 *reg_id)
{
	int err;
	struct mlx4_spec_list spec_eth_outer = { {NULL} };
	struct mlx4_spec_list spec_vxlan     = { {NULL} };
	struct mlx4_spec_list spec_eth_inner = { {NULL} };

	struct mlx4_net_trans_rule rule = {
		.queue_mode = MLX4_NET_TRANS_Q_FIFO,
		.exclusive = 0,
		.allow_loopback = 1,
		.promisc_mode = MLX4_FS_REGULAR,
		.priority = MLX4_DOMAIN_NIC,
	};

	__be64 mac_mask = cpu_to_be64(MLX4_MAC_MASK << 16);

	if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN)
		return 0; /* do nothing */

	rule.port = priv->port;
	rule.qpn = qpn;
	INIT_LIST_HEAD(&rule.list);

	spec_eth_outer.id = MLX4_NET_TRANS_RULE_ID_ETH;
	memcpy(spec_eth_outer.eth.dst_mac, addr, ETH_ALEN);
	memcpy(spec_eth_outer.eth.dst_mac_msk, &mac_mask, ETH_ALEN);

	spec_vxlan.id = MLX4_NET_TRANS_RULE_ID_VXLAN;    /* any vxlan header */
	spec_eth_inner.id = MLX4_NET_TRANS_RULE_ID_ETH;	 /* any inner eth header */

	list_add_tail(&spec_eth_outer.list, &rule.list);
	list_add_tail(&spec_vxlan.list,     &rule.list);
	list_add_tail(&spec_eth_inner.list, &rule.list);

	err = mlx4_flow_attach(priv->mdev->dev, &rule, reg_id);
	if (err) {
		en_err(priv, "failed to add vxlan steering rule, err %d\n", err);
		return err;
	}
	en_dbg(DRV, priv, "added vxlan steering rule, mac %pM reg_id %llx\n", addr, *reg_id);
	return 0;
}


519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
static int mlx4_en_uc_steer_add(struct mlx4_en_priv *priv,
				unsigned char *mac, int *qpn, u64 *reg_id)
{
	struct mlx4_en_dev *mdev = priv->mdev;
	struct mlx4_dev *dev = mdev->dev;
	int err;

	switch (dev->caps.steering_mode) {
	case MLX4_STEERING_MODE_B0: {
		struct mlx4_qp qp;
		u8 gid[16] = {0};

		qp.qpn = *qpn;
		memcpy(&gid[10], mac, ETH_ALEN);
		gid[5] = priv->port;

		err = mlx4_unicast_attach(dev, &qp, gid, 0, MLX4_PROT_ETH);
		break;
	}
	case MLX4_STEERING_MODE_DEVICE_MANAGED: {
		struct mlx4_spec_list spec_eth = { {NULL} };
		__be64 mac_mask = cpu_to_be64(MLX4_MAC_MASK << 16);

		struct mlx4_net_trans_rule rule = {
			.queue_mode = MLX4_NET_TRANS_Q_FIFO,
			.exclusive = 0,
			.allow_loopback = 1,
546
			.promisc_mode = MLX4_FS_REGULAR,
547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606
			.priority = MLX4_DOMAIN_NIC,
		};

		rule.port = priv->port;
		rule.qpn = *qpn;
		INIT_LIST_HEAD(&rule.list);

		spec_eth.id = MLX4_NET_TRANS_RULE_ID_ETH;
		memcpy(spec_eth.eth.dst_mac, mac, ETH_ALEN);
		memcpy(spec_eth.eth.dst_mac_msk, &mac_mask, ETH_ALEN);
		list_add_tail(&spec_eth.list, &rule.list);

		err = mlx4_flow_attach(dev, &rule, reg_id);
		break;
	}
	default:
		return -EINVAL;
	}
	if (err)
		en_warn(priv, "Failed Attaching Unicast\n");

	return err;
}

static void mlx4_en_uc_steer_release(struct mlx4_en_priv *priv,
				     unsigned char *mac, int qpn, u64 reg_id)
{
	struct mlx4_en_dev *mdev = priv->mdev;
	struct mlx4_dev *dev = mdev->dev;

	switch (dev->caps.steering_mode) {
	case MLX4_STEERING_MODE_B0: {
		struct mlx4_qp qp;
		u8 gid[16] = {0};

		qp.qpn = qpn;
		memcpy(&gid[10], mac, ETH_ALEN);
		gid[5] = priv->port;

		mlx4_unicast_detach(dev, &qp, gid, MLX4_PROT_ETH);
		break;
	}
	case MLX4_STEERING_MODE_DEVICE_MANAGED: {
		mlx4_flow_detach(dev, reg_id);
		break;
	}
	default:
		en_err(priv, "Invalid steering mode.\n");
	}
}

static int mlx4_en_get_qp(struct mlx4_en_priv *priv)
{
	struct mlx4_en_dev *mdev = priv->mdev;
	struct mlx4_dev *dev = mdev->dev;
	struct mlx4_mac_entry *entry;
	int index = 0;
	int err = 0;
	u64 reg_id;
	int *qpn = &priv->base_qpn;
607
	u64 mac = mlx4_mac_to_u64(priv->dev->dev_addr);
608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635

	en_dbg(DRV, priv, "Registering MAC: %pM for adding\n",
	       priv->dev->dev_addr);
	index = mlx4_register_mac(dev, priv->port, mac);
	if (index < 0) {
		err = index;
		en_err(priv, "Failed adding MAC: %pM\n",
		       priv->dev->dev_addr);
		return err;
	}

	if (dev->caps.steering_mode == MLX4_STEERING_MODE_A0) {
		int base_qpn = mlx4_get_base_qpn(dev, priv->port);
		*qpn = base_qpn + index;
		return 0;
	}

	err = mlx4_qp_reserve_range(dev, 1, 1, qpn);
	en_dbg(DRV, priv, "Reserved qp %d\n", *qpn);
	if (err) {
		en_err(priv, "Failed to reserve qp for mac registration\n");
		goto qp_err;
	}

	err = mlx4_en_uc_steer_add(priv, priv->dev->dev_addr, qpn, &reg_id);
	if (err)
		goto steer_err;

636 637 638
	err = mlx4_en_tunnel_steer_add(priv, priv->dev->dev_addr, *qpn,
				       &priv->tunnel_reg_id);
	if (err)
639 640
		goto tunnel_err;

641 642 643 644 645 646 647 648
	entry = kmalloc(sizeof(*entry), GFP_KERNEL);
	if (!entry) {
		err = -ENOMEM;
		goto alloc_err;
	}
	memcpy(entry->mac, priv->dev->dev_addr, sizeof(entry->mac));
	entry->reg_id = reg_id;

649 650
	hlist_add_head_rcu(&entry->hlist,
			   &priv->mac_hash[entry->mac[MLX4_EN_MAC_HASH_IDX]]);
651

652
	return 0;
653 654

alloc_err:
655 656 657
	if (priv->tunnel_reg_id)
		mlx4_flow_detach(priv->mdev->dev, priv->tunnel_reg_id);
tunnel_err:
658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
	mlx4_en_uc_steer_release(priv, priv->dev->dev_addr, *qpn, reg_id);

steer_err:
	mlx4_qp_release_range(dev, *qpn, 1);

qp_err:
	mlx4_unregister_mac(dev, priv->port, mac);
	return err;
}

static void mlx4_en_put_qp(struct mlx4_en_priv *priv)
{
	struct mlx4_en_dev *mdev = priv->mdev;
	struct mlx4_dev *dev = mdev->dev;
	int qpn = priv->base_qpn;
673
	u64 mac;
674

675
	if (dev->caps.steering_mode == MLX4_STEERING_MODE_A0) {
676
		mac = mlx4_mac_to_u64(priv->dev->dev_addr);
677 678 679 680
		en_dbg(DRV, priv, "Registering MAC: %pM for deleting\n",
		       priv->dev->dev_addr);
		mlx4_unregister_mac(dev, priv->port, mac);
	} else {
681
		struct mlx4_mac_entry *entry;
682
		struct hlist_node *tmp;
683
		struct hlist_head *bucket;
684
		unsigned int i;
685

686 687 688
		for (i = 0; i < MLX4_EN_MAC_HASH_SIZE; ++i) {
			bucket = &priv->mac_hash[i];
			hlist_for_each_entry_safe(entry, tmp, bucket, hlist) {
689
				mac = mlx4_mac_to_u64(entry->mac);
690 691
				en_dbg(DRV, priv, "Registering MAC: %pM for deleting\n",
				       entry->mac);
692 693 694
				mlx4_en_uc_steer_release(priv, entry->mac,
							 qpn, entry->reg_id);

695
				mlx4_unregister_mac(dev, priv->port, mac);
696 697 698
				hlist_del_rcu(&entry->hlist);
				kfree_rcu(entry, rcu);
			}
699
		}
700

701 702 703 704 705
		if (priv->tunnel_reg_id) {
			mlx4_flow_detach(priv->mdev->dev, priv->tunnel_reg_id);
			priv->tunnel_reg_id = 0;
		}

706 707 708 709
		en_dbg(DRV, priv, "Releasing qp: port %d, qpn %d\n",
		       priv->port, qpn);
		mlx4_qp_release_range(dev, qpn, 1);
		priv->flags &= ~MLX4_EN_FLAG_FORCE_PROMISC;
710 711 712 713
	}
}

static int mlx4_en_replace_mac(struct mlx4_en_priv *priv, int qpn,
714
			       unsigned char *new_mac, unsigned char *prev_mac)
715 716 717 718
{
	struct mlx4_en_dev *mdev = priv->mdev;
	struct mlx4_dev *dev = mdev->dev;
	int err = 0;
719
	u64 new_mac_u64 = mlx4_mac_to_u64(new_mac);
720 721

	if (dev->caps.steering_mode != MLX4_STEERING_MODE_A0) {
722 723 724
		struct hlist_head *bucket;
		unsigned int mac_hash;
		struct mlx4_mac_entry *entry;
725
		struct hlist_node *tmp;
726
		u64 prev_mac_u64 = mlx4_mac_to_u64(prev_mac);
727 728

		bucket = &priv->mac_hash[prev_mac[MLX4_EN_MAC_HASH_IDX]];
729
		hlist_for_each_entry_safe(entry, tmp, bucket, hlist) {
730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745
			if (ether_addr_equal_64bits(entry->mac, prev_mac)) {
				mlx4_en_uc_steer_release(priv, entry->mac,
							 qpn, entry->reg_id);
				mlx4_unregister_mac(dev, priv->port,
						    prev_mac_u64);
				hlist_del_rcu(&entry->hlist);
				synchronize_rcu();
				memcpy(entry->mac, new_mac, ETH_ALEN);
				entry->reg_id = 0;
				mac_hash = new_mac[MLX4_EN_MAC_HASH_IDX];
				hlist_add_head_rcu(&entry->hlist,
						   &priv->mac_hash[mac_hash]);
				mlx4_register_mac(dev, priv->port, new_mac_u64);
				err = mlx4_en_uc_steer_add(priv, new_mac,
							   &qpn,
							   &entry->reg_id);
746 747 748 749 750 751 752 753
				if (err)
					return err;
				if (priv->tunnel_reg_id) {
					mlx4_flow_detach(priv->mdev->dev, priv->tunnel_reg_id);
					priv->tunnel_reg_id = 0;
				}
				err = mlx4_en_tunnel_steer_add(priv, new_mac, qpn,
							       &priv->tunnel_reg_id);
754 755 756 757
				return err;
			}
		}
		return -EINVAL;
758 759 760 761 762
	}

	return __mlx4_replace_mac(dev, priv->port, qpn, new_mac_u64);
}

763
static int mlx4_en_do_set_mac(struct mlx4_en_priv *priv)
764 765 766 767 768
{
	int err = 0;

	if (priv->port_up) {
		/* Remove old MAC and insert the new one */
769
		err = mlx4_en_replace_mac(priv, priv->base_qpn,
770
					  priv->dev->dev_addr, priv->prev_mac);
771
		if (err)
772
			en_err(priv, "Failed changing HW MAC address\n");
773 774
		memcpy(priv->prev_mac, priv->dev->dev_addr,
		       sizeof(priv->prev_mac));
775
	} else
776
		en_dbg(HW, priv, "Port is down while registering mac, exiting...\n");
777

778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794
	return err;
}

static int mlx4_en_set_mac(struct net_device *dev, void *addr)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	struct sockaddr *saddr = addr;
	int err;

	if (!is_valid_ether_addr(saddr->sa_data))
		return -EADDRNOTAVAIL;

	memcpy(dev->dev_addr, saddr->sa_data, ETH_ALEN);

	mutex_lock(&mdev->state_lock);
	err = mlx4_en_do_set_mac(priv);
795
	mutex_unlock(&mdev->state_lock);
796 797

	return err;
798 799 800 801 802
}

static void mlx4_en_clear_list(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
803
	struct mlx4_en_mc_list *tmp, *mc_to_del;
804

805 806 807 808
	list_for_each_entry_safe(mc_to_del, tmp, &priv->mc_list, list) {
		list_del(&mc_to_del->list);
		kfree(mc_to_del);
	}
809 810 811 812 813
}

static void mlx4_en_cache_mclist(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
814
	struct netdev_hw_addr *ha;
815
	struct mlx4_en_mc_list *tmp;
816

817
	mlx4_en_clear_list(dev);
818 819 820 821 822 823 824 825 826
	netdev_for_each_mc_addr(ha, dev) {
		tmp = kzalloc(sizeof(struct mlx4_en_mc_list), GFP_ATOMIC);
		if (!tmp) {
			mlx4_en_clear_list(dev);
			return;
		}
		memcpy(tmp->addr, ha->addr, ETH_ALEN);
		list_add_tail(&tmp->list, &priv->mc_list);
	}
827 828
}

829 830 831 832 833 834 835 836 837 838 839 840 841
static void update_mclist_flags(struct mlx4_en_priv *priv,
				struct list_head *dst,
				struct list_head *src)
{
	struct mlx4_en_mc_list *dst_tmp, *src_tmp, *new_mc;
	bool found;

	/* Find all the entries that should be removed from dst,
	 * These are the entries that are not found in src
	 */
	list_for_each_entry(dst_tmp, dst, list) {
		found = false;
		list_for_each_entry(src_tmp, src, list) {
842
			if (ether_addr_equal(dst_tmp->addr, src_tmp->addr)) {
843 844 845 846 847 848 849 850 851 852 853 854 855 856
				found = true;
				break;
			}
		}
		if (!found)
			dst_tmp->action = MCLIST_REM;
	}

	/* Add entries that exist in src but not in dst
	 * mark them as need to add
	 */
	list_for_each_entry(src_tmp, src, list) {
		found = false;
		list_for_each_entry(dst_tmp, dst, list) {
857
			if (ether_addr_equal(dst_tmp->addr, src_tmp->addr)) {
858 859 860 861 862 863
				dst_tmp->action = MCLIST_NONE;
				found = true;
				break;
			}
		}
		if (!found) {
864 865
			new_mc = kmemdup(src_tmp,
					 sizeof(struct mlx4_en_mc_list),
866
					 GFP_KERNEL);
867
			if (!new_mc)
868
				return;
869

870 871 872 873 874
			new_mc->action = MCLIST_ADD;
			list_add_tail(&new_mc->list, dst);
		}
	}
}
875

876
static void mlx4_en_set_rx_mode(struct net_device *dev)
877 878 879 880 881 882
{
	struct mlx4_en_priv *priv = netdev_priv(dev);

	if (!priv->port_up)
		return;

883
	queue_work(priv->mdev->workqueue, &priv->rx_mode_task);
884 885
}

886 887
static void mlx4_en_set_promisc_mode(struct mlx4_en_priv *priv,
				     struct mlx4_en_dev *mdev)
888
{
889
	int err = 0;
890

891
	if (!(priv->flags & MLX4_EN_FLAG_PROMISC)) {
892
		if (netif_msg_rx_status(priv))
893 894
			en_warn(priv, "Entering promiscuous mode\n");
		priv->flags |= MLX4_EN_FLAG_PROMISC;
895

896
		/* Enable promiscouos mode */
897
		switch (mdev->dev->caps.steering_mode) {
898
		case MLX4_STEERING_MODE_DEVICE_MANAGED:
899 900 901
			err = mlx4_flow_steer_promisc_add(mdev->dev,
							  priv->port,
							  priv->base_qpn,
902
							  MLX4_FS_ALL_DEFAULT);
903
			if (err)
904 905
				en_err(priv, "Failed enabling promiscuous mode\n");
			priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
906 907
			break;

908
		case MLX4_STEERING_MODE_B0:
909 910 911
			err = mlx4_unicast_promisc_add(mdev->dev,
						       priv->base_qpn,
						       priv->port);
912
			if (err)
913 914 915 916 917 918 919 920 921
				en_err(priv, "Failed enabling unicast promiscuous mode\n");

			/* Add the default qp number as multicast
			 * promisc
			 */
			if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
				err = mlx4_multicast_promisc_add(mdev->dev,
								 priv->base_qpn,
								 priv->port);
922
				if (err)
923 924
					en_err(priv, "Failed enabling multicast promiscuous mode\n");
				priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
925 926
			}
			break;
927

928 929 930
		case MLX4_STEERING_MODE_A0:
			err = mlx4_SET_PORT_qpn_calc(mdev->dev,
						     priv->port,
931 932
						     priv->base_qpn,
						     1);
933
			if (err)
934
				en_err(priv, "Failed enabling promiscuous mode\n");
935
			break;
936 937
		}

938 939 940 941 942 943 944
		/* Disable port multicast filter (unconditionally) */
		err = mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0,
					  0, MLX4_MCAST_DISABLE);
		if (err)
			en_err(priv, "Failed disabling multicast filter\n");

		/* Disable port VLAN filter */
J
Jiri Pirko 已提交
945
		err = mlx4_SET_VLAN_FLTR(mdev->dev, priv);
946
		if (err)
947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964
			en_err(priv, "Failed disabling VLAN filter\n");
	}
}

static void mlx4_en_clear_promisc_mode(struct mlx4_en_priv *priv,
				       struct mlx4_en_dev *mdev)
{
	int err = 0;

	if (netif_msg_rx_status(priv))
		en_warn(priv, "Leaving promiscuous mode\n");
	priv->flags &= ~MLX4_EN_FLAG_PROMISC;

	/* Disable promiscouos mode */
	switch (mdev->dev->caps.steering_mode) {
	case MLX4_STEERING_MODE_DEVICE_MANAGED:
		err = mlx4_flow_steer_promisc_remove(mdev->dev,
						     priv->port,
965
						     MLX4_FS_ALL_DEFAULT);
966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994
		if (err)
			en_err(priv, "Failed disabling promiscuous mode\n");
		priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
		break;

	case MLX4_STEERING_MODE_B0:
		err = mlx4_unicast_promisc_remove(mdev->dev,
						  priv->base_qpn,
						  priv->port);
		if (err)
			en_err(priv, "Failed disabling unicast promiscuous mode\n");
		/* Disable Multicast promisc */
		if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
			err = mlx4_multicast_promisc_remove(mdev->dev,
							    priv->base_qpn,
							    priv->port);
			if (err)
				en_err(priv, "Failed disabling multicast promiscuous mode\n");
			priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
		}
		break;

	case MLX4_STEERING_MODE_A0:
		err = mlx4_SET_PORT_qpn_calc(mdev->dev,
					     priv->port,
					     priv->base_qpn, 0);
		if (err)
			en_err(priv, "Failed disabling promiscuous mode\n");
		break;
995 996
	}

997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011
	/* Enable port VLAN filter */
	err = mlx4_SET_VLAN_FLTR(mdev->dev, priv);
	if (err)
		en_err(priv, "Failed enabling VLAN filter\n");
}

static void mlx4_en_do_multicast(struct mlx4_en_priv *priv,
				 struct net_device *dev,
				 struct mlx4_en_dev *mdev)
{
	struct mlx4_en_mc_list *mclist, *tmp;
	u64 mcast_addr = 0;
	u8 mc_list[16] = {0};
	int err = 0;

1012 1013 1014 1015 1016
	/* Enable/disable the multicast filter according to IFF_ALLMULTI */
	if (dev->flags & IFF_ALLMULTI) {
		err = mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0,
					  0, MLX4_MCAST_DISABLE);
		if (err)
1017
			en_err(priv, "Failed disabling multicast filter\n");
1018 1019 1020

		/* Add the default qp number as multicast promisc */
		if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
1021
			switch (mdev->dev->caps.steering_mode) {
1022 1023 1024 1025
			case MLX4_STEERING_MODE_DEVICE_MANAGED:
				err = mlx4_flow_steer_promisc_add(mdev->dev,
								  priv->port,
								  priv->base_qpn,
1026
								  MLX4_FS_MC_DEFAULT);
1027 1028
				break;

1029 1030 1031 1032 1033 1034 1035 1036 1037
			case MLX4_STEERING_MODE_B0:
				err = mlx4_multicast_promisc_add(mdev->dev,
								 priv->base_qpn,
								 priv->port);
				break;

			case MLX4_STEERING_MODE_A0:
				break;
			}
1038 1039 1040 1041
			if (err)
				en_err(priv, "Failed entering multicast promisc mode\n");
			priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
		}
1042
	} else {
1043 1044
		/* Disable Multicast promisc */
		if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
1045
			switch (mdev->dev->caps.steering_mode) {
1046 1047 1048
			case MLX4_STEERING_MODE_DEVICE_MANAGED:
				err = mlx4_flow_steer_promisc_remove(mdev->dev,
								     priv->port,
1049
								     MLX4_FS_MC_DEFAULT);
1050 1051
				break;

1052 1053 1054 1055 1056 1057 1058 1059 1060
			case MLX4_STEERING_MODE_B0:
				err = mlx4_multicast_promisc_remove(mdev->dev,
								    priv->base_qpn,
								    priv->port);
				break;

			case MLX4_STEERING_MODE_A0:
				break;
			}
1061
			if (err)
L
Lucas De Marchi 已提交
1062
				en_err(priv, "Failed disabling multicast promiscuous mode\n");
1063 1064
			priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
		}
1065

1066 1067 1068
		err = mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0,
					  0, MLX4_MCAST_DISABLE);
		if (err)
1069
			en_err(priv, "Failed disabling multicast filter\n");
1070 1071 1072 1073 1074 1075 1076

		/* Flush mcast filter and init it with broadcast address */
		mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, ETH_BCAST,
				    1, MLX4_MCAST_CONFIG);

		/* Update multicast list - we cache all addresses so they won't
		 * change while HW is updated holding the command semaphor */
1077
		netif_addr_lock_bh(dev);
1078
		mlx4_en_cache_mclist(dev);
1079
		netif_addr_unlock_bh(dev);
1080
		list_for_each_entry(mclist, &priv->mc_list, list) {
1081
			mcast_addr = mlx4_mac_to_u64(mclist->addr);
1082 1083 1084 1085 1086 1087
			mlx4_SET_MCAST_FLTR(mdev->dev, priv->port,
					    mcast_addr, 0, MLX4_MCAST_CONFIG);
		}
		err = mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0,
					  0, MLX4_MCAST_ENABLE);
		if (err)
1088
			en_err(priv, "Failed enabling multicast filter\n");
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098

		update_mclist_flags(priv, &priv->curr_list, &priv->mc_list);
		list_for_each_entry_safe(mclist, tmp, &priv->curr_list, list) {
			if (mclist->action == MCLIST_REM) {
				/* detach this address and delete from list */
				memcpy(&mc_list[10], mclist->addr, ETH_ALEN);
				mc_list[5] = priv->port;
				err = mlx4_multicast_detach(mdev->dev,
							    &priv->rss_map.indir_qp,
							    mc_list,
1099 1100
							    MLX4_PROT_ETH,
							    mclist->reg_id);
1101 1102 1103
				if (err)
					en_err(priv, "Fail to detach multicast address\n");

1104 1105 1106 1107 1108 1109
				if (mclist->tunnel_reg_id) {
					err = mlx4_flow_detach(priv->mdev->dev, mclist->tunnel_reg_id);
					if (err)
						en_err(priv, "Failed to detach multicast address\n");
				}

1110 1111 1112
				/* remove from list */
				list_del(&mclist->list);
				kfree(mclist);
1113
			} else if (mclist->action == MCLIST_ADD) {
1114 1115
				/* attach the address */
				memcpy(&mc_list[10], mclist->addr, ETH_ALEN);
1116
				/* needed for B0 steering support */
1117 1118 1119
				mc_list[5] = priv->port;
				err = mlx4_multicast_attach(mdev->dev,
							    &priv->rss_map.indir_qp,
1120 1121 1122 1123
							    mc_list,
							    priv->port, 0,
							    MLX4_PROT_ETH,
							    &mclist->reg_id);
1124 1125 1126
				if (err)
					en_err(priv, "Fail to attach multicast address\n");

1127 1128 1129 1130
				err = mlx4_en_tunnel_steer_add(priv, &mc_list[10], priv->base_qpn,
							       &mclist->tunnel_reg_id);
				if (err)
					en_err(priv, "Failed to attach multicast address\n");
1131 1132
			}
		}
1133
	}
1134 1135
}

1136 1137 1138 1139 1140 1141
static void mlx4_en_do_uc_filter(struct mlx4_en_priv *priv,
				 struct net_device *dev,
				 struct mlx4_en_dev *mdev)
{
	struct netdev_hw_addr *ha;
	struct mlx4_mac_entry *entry;
1142
	struct hlist_node *tmp;
1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157
	bool found;
	u64 mac;
	int err = 0;
	struct hlist_head *bucket;
	unsigned int i;
	int removed = 0;
	u32 prev_flags;

	/* Note that we do not need to protect our mac_hash traversal with rcu,
	 * since all modification code is protected by mdev->state_lock
	 */

	/* find what to remove */
	for (i = 0; i < MLX4_EN_MAC_HASH_SIZE; ++i) {
		bucket = &priv->mac_hash[i];
1158
		hlist_for_each_entry_safe(entry, tmp, bucket, hlist) {
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
			found = false;
			netdev_for_each_uc_addr(ha, dev) {
				if (ether_addr_equal_64bits(entry->mac,
							    ha->addr)) {
					found = true;
					break;
				}
			}

			/* MAC address of the port is not in uc list */
			if (ether_addr_equal_64bits(entry->mac, dev->dev_addr))
				found = true;

			if (!found) {
1173
				mac = mlx4_mac_to_u64(entry->mac);
1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
				mlx4_en_uc_steer_release(priv, entry->mac,
							 priv->base_qpn,
							 entry->reg_id);
				mlx4_unregister_mac(mdev->dev, priv->port, mac);

				hlist_del_rcu(&entry->hlist);
				kfree_rcu(entry, rcu);
				en_dbg(DRV, priv, "Removed MAC %pM on port:%d\n",
				       entry->mac, priv->port);
				++removed;
			}
		}
	}

	/* if we didn't remove anything, there is no use in trying to add
	 * again once we are in a forced promisc mode state
	 */
	if ((priv->flags & MLX4_EN_FLAG_FORCE_PROMISC) && 0 == removed)
		return;

	prev_flags = priv->flags;
	priv->flags &= ~MLX4_EN_FLAG_FORCE_PROMISC;

	/* find what to add */
	netdev_for_each_uc_addr(ha, dev) {
		found = false;
		bucket = &priv->mac_hash[ha->addr[MLX4_EN_MAC_HASH_IDX]];
1201
		hlist_for_each_entry(entry, bucket, hlist) {
1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215
			if (ether_addr_equal_64bits(entry->mac, ha->addr)) {
				found = true;
				break;
			}
		}

		if (!found) {
			entry = kmalloc(sizeof(*entry), GFP_KERNEL);
			if (!entry) {
				en_err(priv, "Failed adding MAC %pM on port:%d (out of memory)\n",
				       ha->addr, priv->port);
				priv->flags |= MLX4_EN_FLAG_FORCE_PROMISC;
				break;
			}
1216
			mac = mlx4_mac_to_u64(ha->addr);
1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 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
			memcpy(entry->mac, ha->addr, ETH_ALEN);
			err = mlx4_register_mac(mdev->dev, priv->port, mac);
			if (err < 0) {
				en_err(priv, "Failed registering MAC %pM on port %d: %d\n",
				       ha->addr, priv->port, err);
				kfree(entry);
				priv->flags |= MLX4_EN_FLAG_FORCE_PROMISC;
				break;
			}
			err = mlx4_en_uc_steer_add(priv, ha->addr,
						   &priv->base_qpn,
						   &entry->reg_id);
			if (err) {
				en_err(priv, "Failed adding MAC %pM on port %d: %d\n",
				       ha->addr, priv->port, err);
				mlx4_unregister_mac(mdev->dev, priv->port, mac);
				kfree(entry);
				priv->flags |= MLX4_EN_FLAG_FORCE_PROMISC;
				break;
			} else {
				unsigned int mac_hash;
				en_dbg(DRV, priv, "Added MAC %pM on port:%d\n",
				       ha->addr, priv->port);
				mac_hash = ha->addr[MLX4_EN_MAC_HASH_IDX];
				bucket = &priv->mac_hash[mac_hash];
				hlist_add_head_rcu(&entry->hlist, bucket);
			}
		}
	}

	if (priv->flags & MLX4_EN_FLAG_FORCE_PROMISC) {
		en_warn(priv, "Forcing promiscuous mode on port:%d\n",
			priv->port);
	} else if (prev_flags & MLX4_EN_FLAG_FORCE_PROMISC) {
		en_warn(priv, "Stop forcing promiscuous mode on port:%d\n",
			priv->port);
	}
}

1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282
static void mlx4_en_do_set_rx_mode(struct work_struct *work)
{
	struct mlx4_en_priv *priv = container_of(work, struct mlx4_en_priv,
						 rx_mode_task);
	struct mlx4_en_dev *mdev = priv->mdev;
	struct net_device *dev = priv->dev;

	mutex_lock(&mdev->state_lock);
	if (!mdev->device_up) {
		en_dbg(HW, priv, "Card is not up, ignoring rx mode change.\n");
		goto out;
	}
	if (!priv->port_up) {
		en_dbg(HW, priv, "Port is down, ignoring rx mode change.\n");
		goto out;
	}

	if (!netif_carrier_ok(dev)) {
		if (!mlx4_en_QUERY_PORT(mdev, priv->port)) {
			if (priv->port_state.link_state) {
				priv->last_link_state = MLX4_DEV_EVENT_PORT_UP;
				netif_carrier_on(dev);
				en_dbg(LINK, priv, "Link Up\n");
			}
		}
	}

1283 1284 1285
	if (dev->priv_flags & IFF_UNICAST_FLT)
		mlx4_en_do_uc_filter(priv, dev, mdev);

1286
	/* Promsicuous mode: disable all filters */
1287 1288
	if ((dev->flags & IFF_PROMISC) ||
	    (priv->flags & MLX4_EN_FLAG_FORCE_PROMISC)) {
1289 1290 1291 1292 1293 1294 1295 1296 1297
		mlx4_en_set_promisc_mode(priv, mdev);
		goto out;
	}

	/* Not in promiscuous mode */
	if (priv->flags & MLX4_EN_FLAG_PROMISC)
		mlx4_en_clear_promisc_mode(priv, mdev);

	mlx4_en_do_multicast(priv, dev, mdev);
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310
out:
	mutex_unlock(&mdev->state_lock);
}

#ifdef CONFIG_NET_POLL_CONTROLLER
static void mlx4_en_netpoll(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_cq *cq;
	unsigned long flags;
	int i;

	for (i = 0; i < priv->rx_ring_num; i++) {
1311
		cq = priv->rx_cq[i];
1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323
		spin_lock_irqsave(&cq->lock, flags);
		napi_synchronize(&cq->napi);
		mlx4_en_process_rx_cq(dev, cq, 0);
		spin_unlock_irqrestore(&cq->lock, flags);
	}
}
#endif

static void mlx4_en_tx_timeout(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
1324
	int i;
1325 1326

	if (netif_msg_timer(priv))
1327
		en_warn(priv, "Tx timeout called on port:%d\n", priv->port);
1328

1329 1330 1331 1332
	for (i = 0; i < priv->tx_ring_num; i++) {
		if (!netif_tx_queue_stopped(netdev_get_tx_queue(dev, i)))
			continue;
		en_warn(priv, "TX timeout on queue: %d, QP: 0x%x, CQ: 0x%x, Cons: 0x%x, Prod: 0x%x\n",
1333 1334
			i, priv->tx_ring[i]->qpn, priv->tx_ring[i]->cqn,
			priv->tx_ring[i]->cons, priv->tx_ring[i]->prod);
1335 1336
	}

1337
	priv->port_stats.tx_timeout++;
1338
	en_dbg(DRV, priv, "Scheduling watchdog\n");
1339
	queue_work(mdev->workqueue, &priv->watchdog_task);
1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359
}


static struct net_device_stats *mlx4_en_get_stats(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);

	spin_lock_bh(&priv->stats_lock);
	memcpy(&priv->ret_stats, &priv->stats, sizeof(priv->stats));
	spin_unlock_bh(&priv->stats_lock);

	return &priv->ret_stats;
}

static void mlx4_en_set_default_moderation(struct mlx4_en_priv *priv)
{
	struct mlx4_en_cq *cq;
	int i;

	/* If we haven't received a specific coalescing setting
1360
	 * (module param), we set the moderation parameters as follows:
1361
	 * - moder_cnt is set to the number of mtu sized packets to
E
Eric Dumazet 已提交
1362
	 *   satisfy our coalescing target.
1363 1364
	 * - moder_time is set to a fixed value.
	 */
1365
	priv->rx_frames = MLX4_EN_RX_COAL_TARGET;
1366
	priv->rx_usecs = MLX4_EN_RX_COAL_TIME;
1367 1368
	priv->tx_frames = MLX4_EN_TX_COAL_PKTS;
	priv->tx_usecs = MLX4_EN_TX_COAL_TIME;
1369 1370
	en_dbg(INTR, priv, "Default coalesing params for mtu:%d - rx_frames:%d rx_usecs:%d\n",
	       priv->dev->mtu, priv->rx_frames, priv->rx_usecs);
1371 1372 1373

	/* Setup cq moderation params */
	for (i = 0; i < priv->rx_ring_num; i++) {
1374
		cq = priv->rx_cq[i];
1375 1376
		cq->moder_cnt = priv->rx_frames;
		cq->moder_time = priv->rx_usecs;
1377 1378 1379
		priv->last_moder_time[i] = MLX4_EN_AUTO_CONF;
		priv->last_moder_packets[i] = 0;
		priv->last_moder_bytes[i] = 0;
1380 1381 1382
	}

	for (i = 0; i < priv->tx_ring_num; i++) {
1383
		cq = priv->tx_cq[i];
1384 1385
		cq->moder_cnt = priv->tx_frames;
		cq->moder_time = priv->tx_usecs;
1386 1387 1388 1389 1390 1391 1392 1393
	}

	/* Reset auto-moderation params */
	priv->pkt_rate_low = MLX4_EN_RX_RATE_LOW;
	priv->rx_usecs_low = MLX4_EN_RX_COAL_TIME_LOW;
	priv->pkt_rate_high = MLX4_EN_RX_RATE_HIGH;
	priv->rx_usecs_high = MLX4_EN_RX_COAL_TIME_HIGH;
	priv->sample_interval = MLX4_EN_SAMPLE_INTERVAL;
1394
	priv->adaptive_rx_coal = 1;
1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409
	priv->last_moder_jiffies = 0;
	priv->last_moder_tx_packets = 0;
}

static void mlx4_en_auto_moderation(struct mlx4_en_priv *priv)
{
	unsigned long period = (unsigned long) (jiffies - priv->last_moder_jiffies);
	struct mlx4_en_cq *cq;
	unsigned long packets;
	unsigned long rate;
	unsigned long avg_pkt_size;
	unsigned long rx_packets;
	unsigned long rx_bytes;
	unsigned long rx_pkt_diff;
	int moder_time;
1410
	int ring, err;
1411 1412 1413 1414

	if (!priv->adaptive_rx_coal || period < priv->sample_interval * HZ)
		return;

1415 1416
	for (ring = 0; ring < priv->rx_ring_num; ring++) {
		spin_lock_bh(&priv->stats_lock);
1417 1418
		rx_packets = priv->rx_ring[ring]->packets;
		rx_bytes = priv->rx_ring[ring]->bytes;
1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431
		spin_unlock_bh(&priv->stats_lock);

		rx_pkt_diff = ((unsigned long) (rx_packets -
				priv->last_moder_packets[ring]));
		packets = rx_pkt_diff;
		rate = packets * HZ / period;
		avg_pkt_size = packets ? ((unsigned long) (rx_bytes -
				priv->last_moder_bytes[ring])) / packets : 0;

		/* Apply auto-moderation only when packet rate
		 * exceeds a rate that it matters */
		if (rate > (MLX4_EN_RX_RATE_THRESH / priv->rx_ring_num) &&
		    avg_pkt_size > MLX4_EN_AVG_PKT_SMALL) {
1432 1433 1434 1435 1436 1437 1438 1439 1440
			if (rate < priv->pkt_rate_low)
				moder_time = priv->rx_usecs_low;
			else if (rate > priv->pkt_rate_high)
				moder_time = priv->rx_usecs_high;
			else
				moder_time = (rate - priv->pkt_rate_low) *
					(priv->rx_usecs_high - priv->rx_usecs_low) /
					(priv->pkt_rate_high - priv->pkt_rate_low) +
					priv->rx_usecs_low;
1441 1442
		} else {
			moder_time = priv->rx_usecs_low;
1443 1444
		}

1445 1446
		if (moder_time != priv->last_moder_time[ring]) {
			priv->last_moder_time[ring] = moder_time;
1447
			cq = priv->rx_cq[ring];
1448
			cq->moder_time = moder_time;
1449
			cq->moder_cnt = priv->rx_frames;
1450
			err = mlx4_en_set_cq_moder(priv, cq);
1451
			if (err)
1452 1453
				en_err(priv, "Failed modifying moderation for cq:%d\n",
				       ring);
1454
		}
1455 1456
		priv->last_moder_packets[ring] = rx_packets;
		priv->last_moder_bytes[ring] = rx_bytes;
1457 1458 1459 1460 1461 1462 1463
	}

	priv->last_moder_jiffies = jiffies;
}

static void mlx4_en_do_get_stats(struct work_struct *work)
{
1464
	struct delayed_work *delay = to_delayed_work(work);
1465 1466 1467 1468 1469 1470 1471
	struct mlx4_en_priv *priv = container_of(delay, struct mlx4_en_priv,
						 stats_task);
	struct mlx4_en_dev *mdev = priv->mdev;
	int err;

	mutex_lock(&mdev->state_lock);
	if (mdev->device_up) {
1472 1473 1474 1475
		if (priv->port_up) {
			err = mlx4_en_DUMP_ETH_STATS(mdev, priv->port, 0);
			if (err)
				en_dbg(HW, priv, "Could not update stats\n");
1476

1477
			mlx4_en_auto_moderation(priv);
1478
		}
1479 1480 1481

		queue_delayed_work(mdev->workqueue, &priv->stats_task, STATS_DELAY);
	}
1482
	if (mdev->mac_removed[MLX4_MAX_PORTS + 1 - priv->port]) {
1483
		mlx4_en_do_set_mac(priv);
1484 1485
		mdev->mac_removed[MLX4_MAX_PORTS + 1 - priv->port] = 0;
	}
1486 1487 1488
	mutex_unlock(&mdev->state_lock);
}

A
Amir Vadai 已提交
1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500
/* mlx4_en_service_task - Run service task for tasks that needed to be done
 * periodically
 */
static void mlx4_en_service_task(struct work_struct *work)
{
	struct delayed_work *delay = to_delayed_work(work);
	struct mlx4_en_priv *priv = container_of(delay, struct mlx4_en_priv,
						 service_task);
	struct mlx4_en_dev *mdev = priv->mdev;

	mutex_lock(&mdev->state_lock);
	if (mdev->device_up) {
1501 1502
		if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS)
			mlx4_en_ptp_overflow_check(mdev);
A
Amir Vadai 已提交
1503 1504 1505 1506 1507 1508 1509

		queue_delayed_work(mdev->workqueue, &priv->service_task,
				   SERVICE_TASK_DELAY);
	}
	mutex_unlock(&mdev->state_lock);
}

1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521
static void mlx4_en_linkstate(struct work_struct *work)
{
	struct mlx4_en_priv *priv = container_of(work, struct mlx4_en_priv,
						 linkstate_task);
	struct mlx4_en_dev *mdev = priv->mdev;
	int linkstate = priv->link_state;

	mutex_lock(&mdev->state_lock);
	/* If observable port state changed set carrier state and
	 * report to system log */
	if (priv->last_link_state != linkstate) {
		if (linkstate == MLX4_DEV_EVENT_PORT_DOWN) {
1522
			en_info(priv, "Link Down\n");
1523 1524
			netif_carrier_off(priv->dev);
		} else {
1525
			en_info(priv, "Link Up\n");
1526 1527 1528 1529 1530 1531 1532 1533
			netif_carrier_on(priv->dev);
		}
	}
	priv->last_link_state = linkstate;
	mutex_unlock(&mdev->state_lock);
}


1534
int mlx4_en_start_port(struct net_device *dev)
1535 1536 1537 1538 1539 1540 1541 1542 1543 1544
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	struct mlx4_en_cq *cq;
	struct mlx4_en_tx_ring *tx_ring;
	int rx_index = 0;
	int tx_index = 0;
	int err = 0;
	int i;
	int j;
1545
	u8 mc_list[16] = {0};
1546 1547

	if (priv->port_up) {
1548
		en_dbg(DRV, priv, "start port called while port already up\n");
1549 1550 1551
		return 0;
	}

1552 1553
	INIT_LIST_HEAD(&priv->mc_list);
	INIT_LIST_HEAD(&priv->curr_list);
1554 1555 1556
	INIT_LIST_HEAD(&priv->ethtool_list);
	memset(&priv->ethtool_rules[0], 0,
	       sizeof(struct ethtool_flow_id) * MAX_NUM_OF_FS_RULES);
1557

1558 1559 1560
	/* Calculate Rx buf size */
	dev->mtu = min(dev->mtu, priv->max_mtu);
	mlx4_en_calc_rx_buf(dev);
1561
	en_dbg(DRV, priv, "Rx buf size:%d\n", priv->rx_skb_size);
1562

1563
	/* Configure rx cq's and rings */
1564 1565
	err = mlx4_en_activate_rx_rings(priv);
	if (err) {
1566
		en_err(priv, "Failed to activate RX rings\n");
1567 1568
		return err;
	}
1569
	for (i = 0; i < priv->rx_ring_num; i++) {
1570
		cq = priv->rx_cq[i];
1571

1572 1573
		mlx4_en_cq_init_lock(cq);

1574
		err = mlx4_en_activate_cq(priv, cq, i);
1575
		if (err) {
1576
			en_err(priv, "Failed activating Rx CQ\n");
1577
			goto cq_err;
1578 1579 1580 1581 1582
		}
		for (j = 0; j < cq->size; j++)
			cq->buf[j].owner_sr_opcode = MLX4_CQE_OWNER_MASK;
		err = mlx4_en_set_cq_moder(priv, cq);
		if (err) {
1583
			en_err(priv, "Failed setting cq moderation parameters");
1584 1585 1586 1587
			mlx4_en_deactivate_cq(priv, cq);
			goto cq_err;
		}
		mlx4_en_arm_cq(priv, cq);
1588
		priv->rx_ring[i]->cqn = cq->mcq.cqn;
1589 1590 1591
		++rx_index;
	}

1592 1593
	/* Set qp number */
	en_dbg(DRV, priv, "Getting qp number for port %d\n", priv->port);
1594
	err = mlx4_en_get_qp(priv);
1595
	if (err) {
1596
		en_err(priv, "Failed getting eth qp\n");
1597 1598 1599 1600
		goto cq_err;
	}
	mdev->mac_removed[priv->port] = 0;

1601 1602
	err = mlx4_en_config_rss_steer(priv);
	if (err) {
1603
		en_err(priv, "Failed configuring rss steering\n");
1604
		goto mac_err;
1605 1606
	}

1607 1608 1609 1610
	err = mlx4_en_create_drop_qp(priv);
	if (err)
		goto rss_err;

1611 1612 1613
	/* Configure tx cq's and rings */
	for (i = 0; i < priv->tx_ring_num; i++) {
		/* Configure cq */
1614
		cq = priv->tx_cq[i];
1615
		err = mlx4_en_activate_cq(priv, cq, i);
1616
		if (err) {
1617
			en_err(priv, "Failed allocating Tx CQ\n");
1618 1619 1620 1621
			goto tx_err;
		}
		err = mlx4_en_set_cq_moder(priv, cq);
		if (err) {
1622
			en_err(priv, "Failed setting cq moderation parameters");
1623 1624 1625
			mlx4_en_deactivate_cq(priv, cq);
			goto tx_err;
		}
1626
		en_dbg(DRV, priv, "Resetting index of collapsed CQ:%d to -1\n", i);
1627 1628 1629
		cq->buf->wqe_index = cpu_to_be16(0xffff);

		/* Configure ring */
1630
		tx_ring = priv->tx_ring[i];
1631
		err = mlx4_en_activate_tx_ring(priv, tx_ring, cq->mcq.cqn,
1632
			i / priv->num_tx_rings_p_up);
1633
		if (err) {
1634
			en_err(priv, "Failed allocating Tx ring\n");
1635 1636 1637
			mlx4_en_deactivate_cq(priv, cq);
			goto tx_err;
		}
1638
		tx_ring->tx_queue = netdev_get_tx_queue(dev, i);
1639 1640 1641 1642

		/* Arm CQ for TX completions */
		mlx4_en_arm_cq(priv, cq);

1643 1644 1645 1646 1647 1648 1649 1650 1651
		/* Set initial ownership of all Tx TXBBs to SW (1) */
		for (j = 0; j < tx_ring->buf_size; j += STAMP_STRIDE)
			*((u32 *) (tx_ring->buf + j)) = 0xffffffff;
		++tx_index;
	}

	/* Configure port */
	err = mlx4_SET_PORT_general(mdev->dev, priv->port,
				    priv->rx_skb_size + ETH_FCS_LEN,
1652 1653 1654 1655
				    priv->prof->tx_pause,
				    priv->prof->tx_ppp,
				    priv->prof->rx_pause,
				    priv->prof->rx_ppp);
1656
	if (err) {
1657 1658
		en_err(priv, "Failed setting port general configurations for port %d, with error %d\n",
		       priv->port, err);
1659 1660 1661 1662 1663
		goto tx_err;
	}
	/* Set default qp number */
	err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port, priv->base_qpn, 0);
	if (err) {
1664
		en_err(priv, "Failed setting default qp numbers\n");
1665 1666 1667
		goto tx_err;
	}

1668
	if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
O
Or Gerlitz 已提交
1669
		err = mlx4_SET_PORT_VXLAN(mdev->dev, priv->port, VXLAN_STEER_BY_OUTER_MAC, 1);
1670 1671 1672 1673 1674 1675 1676
		if (err) {
			en_err(priv, "Failed setting port L2 tunnel configuration, err %d\n",
			       err);
			goto tx_err;
		}
	}

1677
	/* Init port */
1678
	en_dbg(HW, priv, "Initializing port\n");
1679 1680
	err = mlx4_INIT_PORT(mdev->dev, priv->port);
	if (err) {
1681
		en_err(priv, "Failed Initializing port\n");
1682
		goto tx_err;
1683 1684
	}

1685 1686
	/* Attach rx QP to bradcast address */
	memset(&mc_list[10], 0xff, ETH_ALEN);
1687
	mc_list[5] = priv->port; /* needed for B0 steering support */
1688
	if (mlx4_multicast_attach(mdev->dev, &priv->rss_map.indir_qp, mc_list,
1689 1690
				  priv->port, 0, MLX4_PROT_ETH,
				  &priv->broadcast_id))
1691 1692
		mlx4_warn(mdev, "Failed Attaching Broadcast\n");

H
Herbert Xu 已提交
1693 1694 1695
	/* Must redo promiscuous mode setup. */
	priv->flags &= ~(MLX4_EN_FLAG_PROMISC | MLX4_EN_FLAG_MC_PROMISC);

1696
	/* Schedule multicast task to populate multicast list */
1697
	queue_work(mdev->workqueue, &priv->rx_mode_task);
1698

1699 1700
	mlx4_set_stats_bitmap(mdev->dev, &priv->stats_bitmap);

1701
#ifdef CONFIG_MLX4_EN_VXLAN
O
Or Gerlitz 已提交
1702 1703
	if (priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_VXLAN_OFFLOADS)
		vxlan_get_rx_port(dev);
1704
#endif
1705
	priv->port_up = true;
1706
	netif_tx_start_all_queues(dev);
1707 1708
	netif_device_attach(dev);

1709 1710 1711 1712
	return 0;

tx_err:
	while (tx_index--) {
1713 1714
		mlx4_en_deactivate_tx_ring(priv, priv->tx_ring[tx_index]);
		mlx4_en_deactivate_cq(priv, priv->tx_cq[tx_index]);
1715
	}
1716 1717
	mlx4_en_destroy_drop_qp(priv);
rss_err:
1718
	mlx4_en_release_rss_steer(priv);
1719
mac_err:
1720
	mlx4_en_put_qp(priv);
1721 1722
cq_err:
	while (rx_index--)
1723
		mlx4_en_deactivate_cq(priv, priv->rx_cq[rx_index]);
1724
	for (i = 0; i < priv->rx_ring_num; i++)
1725
		mlx4_en_deactivate_rx_ring(priv, priv->rx_ring[i]);
1726 1727 1728 1729 1730

	return err; /* need to close devices */
}


1731
void mlx4_en_stop_port(struct net_device *dev, int detach)
1732 1733 1734
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
1735
	struct mlx4_en_mc_list *mclist, *tmp;
1736
	struct ethtool_flow_id *flow, *tmp_flow;
1737
	int i;
1738
	u8 mc_list[16] = {0};
1739 1740

	if (!priv->port_up) {
1741
		en_dbg(DRV, priv, "stop port called while port already down\n");
1742 1743 1744
		return;
	}

1745 1746 1747
	/* close port*/
	mlx4_CLOSE_PORT(mdev->dev, priv->port);

1748 1749
	/* Synchronize with tx routine */
	netif_tx_lock_bh(dev);
1750 1751
	if (detach)
		netif_device_detach(dev);
1752
	netif_tx_stop_all_queues(dev);
1753 1754
	netif_tx_unlock_bh(dev);

1755 1756
	netif_tx_disable(dev);

1757
	/* Set port as not active */
1758
	priv->port_up = false;
1759

1760 1761 1762 1763 1764 1765 1766
	/* Promsicuous mode */
	if (mdev->dev->caps.steering_mode ==
	    MLX4_STEERING_MODE_DEVICE_MANAGED) {
		priv->flags &= ~(MLX4_EN_FLAG_PROMISC |
				 MLX4_EN_FLAG_MC_PROMISC);
		mlx4_flow_steer_promisc_remove(mdev->dev,
					       priv->port,
1767
					       MLX4_FS_ALL_DEFAULT);
1768 1769
		mlx4_flow_steer_promisc_remove(mdev->dev,
					       priv->port,
1770
					       MLX4_FS_MC_DEFAULT);
1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785
	} else if (priv->flags & MLX4_EN_FLAG_PROMISC) {
		priv->flags &= ~MLX4_EN_FLAG_PROMISC;

		/* Disable promiscouos mode */
		mlx4_unicast_promisc_remove(mdev->dev, priv->base_qpn,
					    priv->port);

		/* Disable Multicast promisc */
		if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
			mlx4_multicast_promisc_remove(mdev->dev, priv->base_qpn,
						      priv->port);
			priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
		}
	}

1786 1787
	/* Detach All multicasts */
	memset(&mc_list[10], 0xff, ETH_ALEN);
1788
	mc_list[5] = priv->port; /* needed for B0 steering support */
1789
	mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp, mc_list,
1790
			      MLX4_PROT_ETH, priv->broadcast_id);
1791 1792
	list_for_each_entry(mclist, &priv->curr_list, list) {
		memcpy(&mc_list[10], mclist->addr, ETH_ALEN);
1793 1794
		mc_list[5] = priv->port;
		mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp,
1795
				      mc_list, MLX4_PROT_ETH, mclist->reg_id);
1796 1797
		if (mclist->tunnel_reg_id)
			mlx4_flow_detach(mdev->dev, mclist->tunnel_reg_id);
1798 1799
	}
	mlx4_en_clear_list(dev);
1800 1801 1802 1803 1804
	list_for_each_entry_safe(mclist, tmp, &priv->curr_list, list) {
		list_del(&mclist->list);
		kfree(mclist);
	}

1805 1806 1807
	/* Flush multicast filter */
	mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0, 1, MLX4_MCAST_CONFIG);

1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818
	/* Remove flow steering rules for the port*/
	if (mdev->dev->caps.steering_mode ==
	    MLX4_STEERING_MODE_DEVICE_MANAGED) {
		ASSERT_RTNL();
		list_for_each_entry_safe(flow, tmp_flow,
					 &priv->ethtool_list, list) {
			mlx4_flow_detach(mdev->dev, flow->id);
			list_del(&flow->list);
		}
	}

1819 1820
	mlx4_en_destroy_drop_qp(priv);

1821 1822
	/* Free TX Rings */
	for (i = 0; i < priv->tx_ring_num; i++) {
1823 1824
		mlx4_en_deactivate_tx_ring(priv, priv->tx_ring[i]);
		mlx4_en_deactivate_cq(priv, priv->tx_cq[i]);
1825 1826 1827 1828
	}
	msleep(10);

	for (i = 0; i < priv->tx_ring_num; i++)
1829
		mlx4_en_free_tx_buf(dev, priv->tx_ring[i]);
1830 1831 1832 1833

	/* Free RSS qps */
	mlx4_en_release_rss_steer(priv);

1834
	/* Unregister Mac address for the port */
1835
	mlx4_en_put_qp(priv);
1836
	if (!(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_REASSIGN_MAC_EN))
1837
		mdev->mac_removed[priv->port] = 1;
1838

1839 1840
	/* Free RX Rings */
	for (i = 0; i < priv->rx_ring_num; i++) {
1841
		struct mlx4_en_cq *cq = priv->rx_cq[i];
1842 1843 1844 1845 1846 1847 1848 1849 1850

		local_bh_disable();
		while (!mlx4_en_cq_lock_napi(cq)) {
			pr_info("CQ %d locked\n", i);
			mdelay(1);
		}
		local_bh_enable();

		while (test_bit(NAPI_STATE_SCHED, &cq->napi.state))
1851
			msleep(1);
1852
		mlx4_en_deactivate_rx_ring(priv, priv->rx_ring[i]);
1853
		mlx4_en_deactivate_cq(priv, cq);
1854 1855 1856 1857 1858 1859 1860 1861 1862 1863
	}
}

static void mlx4_en_restart(struct work_struct *work)
{
	struct mlx4_en_priv *priv = container_of(work, struct mlx4_en_priv,
						 watchdog_task);
	struct mlx4_en_dev *mdev = priv->mdev;
	struct net_device *dev = priv->dev;

1864
	en_dbg(DRV, priv, "Watchdog task called for port %d\n", priv->port);
1865 1866 1867

	mutex_lock(&mdev->state_lock);
	if (priv->port_up) {
1868
		mlx4_en_stop_port(dev, 1);
1869
		if (mlx4_en_start_port(dev))
1870
			en_err(priv, "Failed restarting port %d\n", priv->port);
1871 1872
	}
	mutex_unlock(&mdev->state_lock);
1873 1874
}

1875
static void mlx4_en_clear_stats(struct net_device *dev)
1876 1877 1878 1879 1880 1881
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	int i;

	if (mlx4_en_DUMP_ETH_STATS(mdev, priv->port, 1))
1882
		en_dbg(HW, priv, "Failed dumping statistics\n");
1883 1884 1885

	memset(&priv->stats, 0, sizeof(priv->stats));
	memset(&priv->pstats, 0, sizeof(priv->pstats));
1886 1887
	memset(&priv->pkstats, 0, sizeof(priv->pkstats));
	memset(&priv->port_stats, 0, sizeof(priv->port_stats));
1888 1889

	for (i = 0; i < priv->tx_ring_num; i++) {
1890 1891 1892
		priv->tx_ring[i]->bytes = 0;
		priv->tx_ring[i]->packets = 0;
		priv->tx_ring[i]->tx_csum = 0;
1893 1894
	}
	for (i = 0; i < priv->rx_ring_num; i++) {
1895 1896 1897 1898
		priv->rx_ring[i]->bytes = 0;
		priv->rx_ring[i]->packets = 0;
		priv->rx_ring[i]->csum_ok = 0;
		priv->rx_ring[i]->csum_none = 0;
1899
	}
1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917
}

static int mlx4_en_open(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	int err = 0;

	mutex_lock(&mdev->state_lock);

	if (!mdev->device_up) {
		en_err(priv, "Cannot open - device down/disabled\n");
		err = -EBUSY;
		goto out;
	}

	/* Reset HW statistics and SW counters */
	mlx4_en_clear_stats(dev);
1918 1919 1920

	err = mlx4_en_start_port(dev);
	if (err)
1921
		en_err(priv, "Failed starting port:%d\n", priv->port);
1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933

out:
	mutex_unlock(&mdev->state_lock);
	return err;
}


static int mlx4_en_close(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;

1934
	en_dbg(IFDOWN, priv, "Close port called\n");
1935 1936 1937

	mutex_lock(&mdev->state_lock);

1938
	mlx4_en_stop_port(dev, 0);
1939 1940 1941 1942 1943 1944
	netif_carrier_off(dev);

	mutex_unlock(&mdev->state_lock);
	return 0;
}

1945
void mlx4_en_free_resources(struct mlx4_en_priv *priv)
1946 1947 1948
{
	int i;

1949 1950 1951 1952 1953
#ifdef CONFIG_RFS_ACCEL
	free_irq_cpu_rmap(priv->dev->rx_cpu_rmap);
	priv->dev->rx_cpu_rmap = NULL;
#endif

1954
	for (i = 0; i < priv->tx_ring_num; i++) {
1955
		if (priv->tx_ring && priv->tx_ring[i])
1956
			mlx4_en_destroy_tx_ring(priv, &priv->tx_ring[i]);
1957
		if (priv->tx_cq && priv->tx_cq[i])
1958
			mlx4_en_destroy_cq(priv, &priv->tx_cq[i]);
1959 1960 1961
	}

	for (i = 0; i < priv->rx_ring_num; i++) {
1962
		if (priv->rx_ring[i])
1963 1964
			mlx4_en_destroy_rx_ring(priv, &priv->rx_ring[i],
				priv->prof->rx_ring_size, priv->stride);
1965
		if (priv->rx_cq[i])
1966
			mlx4_en_destroy_cq(priv, &priv->rx_cq[i]);
1967
	}
1968 1969 1970 1971 1972

	if (priv->base_tx_qpn) {
		mlx4_qp_release_range(priv->mdev->dev, priv->base_tx_qpn, priv->tx_ring_num);
		priv->base_tx_qpn = 0;
	}
1973 1974
}

1975
int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
1976 1977 1978
{
	struct mlx4_en_port_profile *prof = priv->prof;
	int i;
1979
	int err;
1980
	int node;
1981

1982
	err = mlx4_qp_reserve_range(priv->mdev->dev, priv->tx_ring_num, 256, &priv->base_tx_qpn);
1983 1984 1985 1986
	if (err) {
		en_err(priv, "failed reserving range for TX rings\n");
		return err;
	}
1987 1988 1989

	/* Create tx Rings */
	for (i = 0; i < priv->tx_ring_num; i++) {
1990
		node = cpu_to_node(i % num_online_cpus());
1991
		if (mlx4_en_create_cq(priv, &priv->tx_cq[i],
1992
				      prof->tx_ring_size, i, TX, node))
1993 1994
			goto err;

1995 1996 1997 1998
		if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i],
					   priv->base_tx_qpn + i,
					   prof->tx_ring_size, TXBB_SIZE,
					   node, i))
1999 2000 2001 2002 2003
			goto err;
	}

	/* Create rx Rings */
	for (i = 0; i < priv->rx_ring_num; i++) {
2004
		node = cpu_to_node(i % num_online_cpus());
2005
		if (mlx4_en_create_cq(priv, &priv->rx_cq[i],
2006
				      prof->rx_ring_size, i, RX, node))
2007 2008 2009
			goto err;

		if (mlx4_en_create_rx_ring(priv, &priv->rx_ring[i],
2010 2011
					   prof->rx_ring_size, priv->stride,
					   node))
2012 2013 2014
			goto err;
	}

2015
#ifdef CONFIG_RFS_ACCEL
2016 2017 2018 2019 2020
	if (priv->mdev->dev->caps.comp_pool) {
		priv->dev->rx_cpu_rmap = alloc_irq_cpu_rmap(priv->mdev->dev->caps.comp_pool);
		if (!priv->dev->rx_cpu_rmap)
			goto err;
	}
2021 2022
#endif

2023 2024 2025
	return 0;

err:
2026
	en_err(priv, "Failed to allocate NIC resources\n");
2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040
	for (i = 0; i < priv->rx_ring_num; i++) {
		if (priv->rx_ring[i])
			mlx4_en_destroy_rx_ring(priv, &priv->rx_ring[i],
						prof->rx_ring_size,
						priv->stride);
		if (priv->rx_cq[i])
			mlx4_en_destroy_cq(priv, &priv->rx_cq[i]);
	}
	for (i = 0; i < priv->tx_ring_num; i++) {
		if (priv->tx_ring[i])
			mlx4_en_destroy_tx_ring(priv, &priv->tx_ring[i]);
		if (priv->tx_cq[i])
			mlx4_en_destroy_cq(priv, &priv->tx_cq[i]);
	}
2041 2042 2043 2044 2045 2046 2047 2048 2049
	return -ENOMEM;
}


void mlx4_en_destroy_netdev(struct net_device *dev)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;

2050
	en_dbg(DRV, priv, "Destroying netdev on port:%d\n", priv->port);
2051 2052 2053 2054 2055 2056 2057 2058 2059

	/* Unregister device - this will close the port if it was up */
	if (priv->registered)
		unregister_netdev(dev);

	if (priv->allocated)
		mlx4_free_hwq_res(mdev->dev, &priv->res, MLX4_EN_PAGE_SIZE);

	cancel_delayed_work(&priv->stats_task);
A
Amir Vadai 已提交
2060
	cancel_delayed_work(&priv->service_task);
2061 2062 2063 2064 2065 2066 2067 2068
	/* flush any pending task for this netdev */
	flush_workqueue(mdev->workqueue);

	/* Detach the netdev so tasks would not attempt to access it */
	mutex_lock(&mdev->state_lock);
	mdev->pndev[priv->port] = NULL;
	mutex_unlock(&mdev->state_lock);

2069
	mlx4_en_free_resources(priv);
A
Amir Vadai 已提交
2070

2071 2072 2073
	kfree(priv->tx_ring);
	kfree(priv->tx_cq);

2074 2075 2076 2077 2078 2079 2080 2081 2082
	free_netdev(dev);
}

static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	int err = 0;

2083
	en_dbg(DRV, priv, "Change MTU called - current:%d new:%d\n",
2084 2085 2086
		 dev->mtu, new_mtu);

	if ((new_mtu < MLX4_EN_MIN_MTU) || (new_mtu > priv->max_mtu)) {
2087
		en_err(priv, "Bad MTU size:%d.\n", new_mtu);
2088 2089 2090 2091 2092 2093 2094 2095 2096
		return -EPERM;
	}
	dev->mtu = new_mtu;

	if (netif_running(dev)) {
		mutex_lock(&mdev->state_lock);
		if (!mdev->device_up) {
			/* NIC is probably restarting - let watchdog task reset
			 * the port */
2097
			en_dbg(DRV, priv, "Change MTU called with card down!?\n");
2098
		} else {
2099
			mlx4_en_stop_port(dev, 1);
2100 2101
			err = mlx4_en_start_port(dev);
			if (err) {
2102
				en_err(priv, "Failed restarting port:%d\n",
2103 2104 2105 2106 2107 2108 2109 2110 2111
					 priv->port);
				queue_work(mdev->workqueue, &priv->watchdog_task);
			}
		}
		mutex_unlock(&mdev->state_lock);
	}
	return 0;
}

2112
static int mlx4_en_hwtstamp_set(struct net_device *dev, struct ifreq *ifr)
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 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = priv->mdev;
	struct hwtstamp_config config;

	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
		return -EFAULT;

	/* reserved for future extensions */
	if (config.flags)
		return -EINVAL;

	/* device doesn't support time stamping */
	if (!(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS))
		return -EINVAL;

	/* TX HW timestamp */
	switch (config.tx_type) {
	case HWTSTAMP_TX_OFF:
	case HWTSTAMP_TX_ON:
		break;
	default:
		return -ERANGE;
	}

	/* RX HW timestamp */
	switch (config.rx_filter) {
	case HWTSTAMP_FILTER_NONE:
		break;
	case HWTSTAMP_FILTER_ALL:
	case HWTSTAMP_FILTER_SOME:
	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
	case HWTSTAMP_FILTER_PTP_V2_EVENT:
	case HWTSTAMP_FILTER_PTP_V2_SYNC:
	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
		config.rx_filter = HWTSTAMP_FILTER_ALL;
		break;
	default:
		return -ERANGE;
	}

	if (mlx4_en_timestamp_config(dev, config.tx_type, config.rx_filter)) {
		config.tx_type = HWTSTAMP_TX_OFF;
		config.rx_filter = HWTSTAMP_FILTER_NONE;
	}

	return copy_to_user(ifr->ifr_data, &config,
			    sizeof(config)) ? -EFAULT : 0;
}

2171 2172 2173 2174 2175 2176 2177 2178
static int mlx4_en_hwtstamp_get(struct net_device *dev, struct ifreq *ifr)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);

	return copy_to_user(ifr->ifr_data, &priv->hwtstamp_config,
			    sizeof(priv->hwtstamp_config)) ? -EFAULT : 0;
}

2179 2180 2181 2182
static int mlx4_en_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
	switch (cmd) {
	case SIOCSHWTSTAMP:
2183 2184 2185
		return mlx4_en_hwtstamp_set(dev, ifr);
	case SIOCGHWTSTAMP:
		return mlx4_en_hwtstamp_get(dev, ifr);
2186 2187 2188 2189 2190
	default:
		return -EOPNOTSUPP;
	}
}

A
Amir Vadai 已提交
2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201
static int mlx4_en_set_features(struct net_device *netdev,
		netdev_features_t features)
{
	struct mlx4_en_priv *priv = netdev_priv(netdev);

	if (features & NETIF_F_LOOPBACK)
		priv->ctrl_flags |= cpu_to_be32(MLX4_WQE_CTRL_FORCE_LOOPBACK);
	else
		priv->ctrl_flags &=
			cpu_to_be32(~MLX4_WQE_CTRL_FORCE_LOOPBACK);

2202 2203
	mlx4_en_update_loopback_state(netdev, features);

A
Amir Vadai 已提交
2204 2205 2206 2207
	return 0;

}

2208 2209 2210 2211
static int mlx4_en_set_vf_mac(struct net_device *dev, int queue, u8 *mac)
{
	struct mlx4_en_priv *en_priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = en_priv->mdev;
2212
	u64 mac_u64 = mlx4_mac_to_u64(mac);
2213 2214 2215 2216 2217 2218 2219

	if (!is_valid_ether_addr(mac))
		return -EINVAL;

	return mlx4_set_vf_mac(mdev->dev, en_priv->port, queue, mac_u64);
}

2220 2221 2222 2223 2224 2225 2226 2227
static int mlx4_en_set_vf_vlan(struct net_device *dev, int vf, u16 vlan, u8 qos)
{
	struct mlx4_en_priv *en_priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = en_priv->mdev;

	return mlx4_set_vf_vlan(mdev->dev, en_priv->port, vf, vlan, qos);
}

2228 2229 2230 2231 2232 2233 2234 2235
static int mlx4_en_set_vf_spoofchk(struct net_device *dev, int vf, bool setting)
{
	struct mlx4_en_priv *en_priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = en_priv->mdev;

	return mlx4_set_vf_spoofchk(mdev->dev, en_priv->port, vf, setting);
}

2236 2237 2238 2239 2240 2241 2242
static int mlx4_en_get_vf_config(struct net_device *dev, int vf, struct ifla_vf_info *ivf)
{
	struct mlx4_en_priv *en_priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = en_priv->mdev;

	return mlx4_get_vf_config(mdev->dev, en_priv->port, vf, ivf);
}
2243

2244 2245 2246 2247 2248 2249 2250
static int mlx4_en_set_vf_link_state(struct net_device *dev, int vf, int link_state)
{
	struct mlx4_en_priv *en_priv = netdev_priv(dev);
	struct mlx4_en_dev *mdev = en_priv->mdev;

	return mlx4_set_vf_link_state(mdev->dev, en_priv->port, vf, link_state);
}
2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271

#define PORT_ID_BYTE_LEN 8
static int mlx4_en_get_phys_port_id(struct net_device *dev,
				    struct netdev_phys_port_id *ppid)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	struct mlx4_dev *mdev = priv->mdev->dev;
	int i;
	u64 phys_port_id = mdev->caps.phys_port_id[priv->port];

	if (!phys_port_id)
		return -EOPNOTSUPP;

	ppid->id_len = sizeof(phys_port_id);
	for (i = PORT_ID_BYTE_LEN - 1; i >= 0; --i) {
		ppid->id[i] =  phys_port_id & 0xff;
		phys_port_id >>= 8;
	}
	return 0;
}

2272
#ifdef CONFIG_MLX4_EN_VXLAN
O
Or Gerlitz 已提交
2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 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
static void mlx4_en_add_vxlan_offloads(struct work_struct *work)
{
	int ret;
	struct mlx4_en_priv *priv = container_of(work, struct mlx4_en_priv,
						 vxlan_add_task);

	ret = mlx4_config_vxlan_port(priv->mdev->dev, priv->vxlan_port);
	if (ret)
		goto out;

	ret = mlx4_SET_PORT_VXLAN(priv->mdev->dev, priv->port,
				  VXLAN_STEER_BY_OUTER_MAC, 1);
out:
	if (ret)
		en_err(priv, "failed setting L2 tunnel configuration ret %d\n", ret);
}

static void mlx4_en_del_vxlan_offloads(struct work_struct *work)
{
	int ret;
	struct mlx4_en_priv *priv = container_of(work, struct mlx4_en_priv,
						 vxlan_del_task);

	ret = mlx4_SET_PORT_VXLAN(priv->mdev->dev, priv->port,
				  VXLAN_STEER_BY_OUTER_MAC, 0);
	if (ret)
		en_err(priv, "failed setting L2 tunnel configuration ret %d\n", ret);

	priv->vxlan_port = 0;
}

static void mlx4_en_add_vxlan_port(struct  net_device *dev,
				   sa_family_t sa_family, __be16 port)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	__be16 current_port;

	if (!(priv->mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_VXLAN_OFFLOADS))
		return;

	if (sa_family == AF_INET6)
		return;

	current_port = priv->vxlan_port;
	if (current_port && current_port != port) {
		en_warn(priv, "vxlan port %d configured, can't add port %d\n",
			ntohs(current_port), ntohs(port));
		return;
	}

	priv->vxlan_port = port;
	queue_work(priv->mdev->workqueue, &priv->vxlan_add_task);
}

static void mlx4_en_del_vxlan_port(struct  net_device *dev,
				   sa_family_t sa_family, __be16 port)
{
	struct mlx4_en_priv *priv = netdev_priv(dev);
	__be16 current_port;

	if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN)
		return;

	if (sa_family == AF_INET6)
		return;

	current_port = priv->vxlan_port;
	if (current_port != port) {
		en_dbg(DRV, priv, "vxlan port %d isn't configured, ignoring\n", ntohs(port));
		return;
	}

	queue_work(priv->mdev->workqueue, &priv->vxlan_del_task);
}
2347
#endif
O
Or Gerlitz 已提交
2348

2349 2350 2351 2352
static const struct net_device_ops mlx4_netdev_ops = {
	.ndo_open		= mlx4_en_open,
	.ndo_stop		= mlx4_en_close,
	.ndo_start_xmit		= mlx4_en_xmit,
Y
Yevgeny Petrilin 已提交
2353
	.ndo_select_queue	= mlx4_en_select_queue,
2354
	.ndo_get_stats		= mlx4_en_get_stats,
2355
	.ndo_set_rx_mode	= mlx4_en_set_rx_mode,
2356
	.ndo_set_mac_address	= mlx4_en_set_mac,
2357
	.ndo_validate_addr	= eth_validate_addr,
2358
	.ndo_change_mtu		= mlx4_en_change_mtu,
2359
	.ndo_do_ioctl		= mlx4_en_ioctl,
2360 2361 2362 2363 2364 2365
	.ndo_tx_timeout		= mlx4_en_tx_timeout,
	.ndo_vlan_rx_add_vid	= mlx4_en_vlan_rx_add_vid,
	.ndo_vlan_rx_kill_vid	= mlx4_en_vlan_rx_kill_vid,
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= mlx4_en_netpoll,
#endif
A
Amir Vadai 已提交
2366
	.ndo_set_features	= mlx4_en_set_features,
2367
	.ndo_setup_tc		= mlx4_en_setup_tc,
2368 2369 2370
#ifdef CONFIG_RFS_ACCEL
	.ndo_rx_flow_steer	= mlx4_en_filter_rfs,
#endif
2371
#ifdef CONFIG_NET_RX_BUSY_POLL
2372
	.ndo_busy_poll		= mlx4_en_low_latency_recv,
2373
#endif
2374
	.ndo_get_phys_port_id	= mlx4_en_get_phys_port_id,
2375
#ifdef CONFIG_MLX4_EN_VXLAN
O
Or Gerlitz 已提交
2376 2377
	.ndo_add_vxlan_port	= mlx4_en_add_vxlan_port,
	.ndo_del_vxlan_port	= mlx4_en_del_vxlan_port,
2378
#endif
2379 2380
};

2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394
static const struct net_device_ops mlx4_netdev_ops_master = {
	.ndo_open		= mlx4_en_open,
	.ndo_stop		= mlx4_en_close,
	.ndo_start_xmit		= mlx4_en_xmit,
	.ndo_select_queue	= mlx4_en_select_queue,
	.ndo_get_stats		= mlx4_en_get_stats,
	.ndo_set_rx_mode	= mlx4_en_set_rx_mode,
	.ndo_set_mac_address	= mlx4_en_set_mac,
	.ndo_validate_addr	= eth_validate_addr,
	.ndo_change_mtu		= mlx4_en_change_mtu,
	.ndo_tx_timeout		= mlx4_en_tx_timeout,
	.ndo_vlan_rx_add_vid	= mlx4_en_vlan_rx_add_vid,
	.ndo_vlan_rx_kill_vid	= mlx4_en_vlan_rx_kill_vid,
	.ndo_set_vf_mac		= mlx4_en_set_vf_mac,
2395
	.ndo_set_vf_vlan	= mlx4_en_set_vf_vlan,
2396
	.ndo_set_vf_spoofchk	= mlx4_en_set_vf_spoofchk,
2397
	.ndo_set_vf_link_state	= mlx4_en_set_vf_link_state,
2398
	.ndo_get_vf_config	= mlx4_en_get_vf_config,
2399 2400 2401 2402 2403 2404 2405 2406
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= mlx4_en_netpoll,
#endif
	.ndo_set_features	= mlx4_en_set_features,
	.ndo_setup_tc		= mlx4_en_setup_tc,
#ifdef CONFIG_RFS_ACCEL
	.ndo_rx_flow_steer	= mlx4_en_filter_rfs,
#endif
2407
	.ndo_get_phys_port_id	= mlx4_en_get_phys_port_id,
2408 2409
};

2410 2411 2412 2413 2414
int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
			struct mlx4_en_port_profile *prof)
{
	struct net_device *dev;
	struct mlx4_en_priv *priv;
2415
	int i;
2416
	int err;
2417
	u64 mac_u64;
2418

2419
	dev = alloc_etherdev_mqs(sizeof(struct mlx4_en_priv),
2420
				 MAX_TX_RINGS, MAX_RX_RINGS);
2421
	if (dev == NULL)
2422 2423
		return -ENOMEM;

2424 2425 2426
	netif_set_real_num_tx_queues(dev, prof->tx_ring_num);
	netif_set_real_num_rx_queues(dev, prof->rx_ring_num);

2427
	SET_NETDEV_DEV(dev, &mdev->dev->pdev->dev);
2428
	dev->dev_port = port - 1;
2429 2430 2431 2432 2433 2434 2435 2436 2437

	/*
	 * Initialize driver private data
	 */

	priv = netdev_priv(dev);
	memset(priv, 0, sizeof(struct mlx4_en_priv));
	priv->dev = dev;
	priv->mdev = mdev;
2438
	priv->ddev = &mdev->pdev->dev;
2439 2440 2441 2442
	priv->prof = prof;
	priv->port = port;
	priv->port_up = false;
	priv->flags = prof->flags;
A
Amir Vadai 已提交
2443 2444
	priv->ctrl_flags = cpu_to_be32(MLX4_WQE_CTRL_CQ_UPDATE |
			MLX4_WQE_CTRL_SOLICITED);
2445
	priv->num_tx_rings_p_up = mdev->profile.num_tx_rings_p_up;
2446
	priv->tx_ring_num = prof->tx_ring_num;
2447

2448
	priv->tx_ring = kzalloc(sizeof(struct mlx4_en_tx_ring *) * MAX_TX_RINGS,
2449
				GFP_KERNEL);
2450 2451 2452 2453
	if (!priv->tx_ring) {
		err = -ENOMEM;
		goto out;
	}
2454
	priv->tx_cq = kzalloc(sizeof(struct mlx4_en_cq *) * MAX_TX_RINGS,
2455
			      GFP_KERNEL);
2456 2457 2458 2459
	if (!priv->tx_cq) {
		err = -ENOMEM;
		goto out;
	}
2460
	priv->rx_ring_num = prof->rx_ring_num;
O
Or Gerlitz 已提交
2461
	priv->cqe_factor = (mdev->dev->caps.cqe_size == 64) ? 1 : 0;
2462 2463 2464
	priv->mac_index = -1;
	priv->msg_enable = MLX4_EN_MSG_LEVEL;
	spin_lock_init(&priv->stats_lock);
2465
	INIT_WORK(&priv->rx_mode_task, mlx4_en_do_set_rx_mode);
2466 2467 2468
	INIT_WORK(&priv->watchdog_task, mlx4_en_restart);
	INIT_WORK(&priv->linkstate_task, mlx4_en_linkstate);
	INIT_DELAYED_WORK(&priv->stats_task, mlx4_en_do_get_stats);
A
Amir Vadai 已提交
2469
	INIT_DELAYED_WORK(&priv->service_task, mlx4_en_service_task);
2470
#ifdef CONFIG_MLX4_EN_VXLAN
O
Or Gerlitz 已提交
2471 2472
	INIT_WORK(&priv->vxlan_add_task, mlx4_en_add_vxlan_offloads);
	INIT_WORK(&priv->vxlan_del_task, mlx4_en_del_vxlan_offloads);
2473
#endif
A
Amir Vadai 已提交
2474
#ifdef CONFIG_MLX4_EN_DCB
2475 2476 2477 2478 2479 2480 2481 2482
	if (!mlx4_is_slave(priv->mdev->dev)) {
		if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_SET_ETH_SCHED) {
			dev->dcbnl_ops = &mlx4_en_dcbnl_ops;
		} else {
			en_info(priv, "enabling only PFC DCB ops\n");
			dev->dcbnl_ops = &mlx4_en_dcbnl_pfc_ops;
		}
	}
A
Amir Vadai 已提交
2483
#endif
2484

2485 2486
	for (i = 0; i < MLX4_EN_MAC_HASH_SIZE; ++i)
		INIT_HLIST_HEAD(&priv->mac_hash[i]);
2487

2488 2489
	/* Query for default mac and max mtu */
	priv->max_mtu = mdev->dev->caps.eth_mtu_cap[priv->port];
2490 2491 2492 2493 2494

	/* Set default MAC */
	dev->addr_len = ETH_ALEN;
	mlx4_en_u64_to_mac(dev->dev_addr, mdev->dev->caps.def_mac[priv->port]);
	if (!is_valid_ether_addr(dev->dev_addr)) {
2495 2496 2497
		if (mlx4_is_slave(priv->mdev->dev)) {
			eth_hw_addr_random(dev);
			en_warn(priv, "Assigned random MAC address %pM\n", dev->dev_addr);
2498
			mac_u64 = mlx4_mac_to_u64(dev->dev_addr);
2499 2500 2501 2502 2503 2504 2505
			mdev->dev->caps.def_mac[priv->port] = mac_u64;
		} else {
			en_err(priv, "Port: %d, invalid mac burned: %pM, quiting\n",
			       priv->port, dev->dev_addr);
			err = -EINVAL;
			goto out;
		}
2506 2507
	}

2508 2509
	memcpy(priv->prev_mac, dev->dev_addr, sizeof(priv->prev_mac));

2510 2511 2512 2513 2514 2515
	priv->stride = roundup_pow_of_two(sizeof(struct mlx4_en_rx_desc) +
					  DS_SIZE * MLX4_EN_MAX_RX_FRAGS);
	err = mlx4_en_alloc_resources(priv);
	if (err)
		goto out;

2516 2517 2518 2519 2520
#ifdef CONFIG_RFS_ACCEL
	INIT_LIST_HEAD(&priv->filters);
	spin_lock_init(&priv->filters_lock);
#endif

2521 2522 2523 2524 2525
	/* Initialize time stamping config */
	priv->hwtstamp_config.flags = 0;
	priv->hwtstamp_config.tx_type = HWTSTAMP_TX_OFF;
	priv->hwtstamp_config.rx_filter = HWTSTAMP_FILTER_NONE;

2526 2527 2528 2529
	/* Allocate page for receive rings */
	err = mlx4_alloc_hwq_res(mdev->dev, &priv->res,
				MLX4_EN_PAGE_SIZE, MLX4_EN_PAGE_SIZE);
	if (err) {
2530
		en_err(priv, "Failed to allocate page for rx qps\n");
2531 2532 2533 2534 2535 2536 2537
		goto out;
	}
	priv->allocated = 1;

	/*
	 * Initialize netdev entry points
	 */
2538 2539 2540 2541
	if (mlx4_is_master(priv->mdev->dev))
		dev->netdev_ops = &mlx4_netdev_ops_master;
	else
		dev->netdev_ops = &mlx4_netdev_ops;
2542
	dev->watchdog_timeo = MLX4_EN_WATCHDOG_TIMEOUT;
2543 2544
	netif_set_real_num_tx_queues(dev, priv->tx_ring_num);
	netif_set_real_num_rx_queues(dev, priv->rx_ring_num);
2545

2546 2547 2548 2549 2550
	SET_ETHTOOL_OPS(dev, &mlx4_en_ethtool_ops);

	/*
	 * Set driver features
	 */
2551 2552 2553 2554 2555 2556
	dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
	if (mdev->LSO_support)
		dev->hw_features |= NETIF_F_TSO | NETIF_F_TSO6;

	dev->vlan_features = dev->hw_features;

Y
Yevgeny Petrilin 已提交
2557
	dev->hw_features |= NETIF_F_RXCSUM | NETIF_F_RXHASH;
2558
	dev->features = dev->hw_features | NETIF_F_HIGHDMA |
2559 2560
			NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX |
			NETIF_F_HW_VLAN_CTAG_FILTER;
A
Amir Vadai 已提交
2561
	dev->hw_features |= NETIF_F_LOOPBACK;
2562

2563 2564 2565 2566
	if (mdev->dev->caps.steering_mode ==
	    MLX4_STEERING_MODE_DEVICE_MANAGED)
		dev->hw_features |= NETIF_F_NTUPLE;

2567 2568 2569
	if (mdev->dev->caps.steering_mode != MLX4_STEERING_MODE_A0)
		dev->priv_flags |= IFF_UNICAST_FLT;

2570 2571 2572 2573 2574 2575 2576
	if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
		dev->hw_enc_features |= NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
					NETIF_F_TSO | NETIF_F_GSO_UDP_TUNNEL;
		dev->hw_features |= NETIF_F_GSO_UDP_TUNNEL;
		dev->features    |= NETIF_F_GSO_UDP_TUNNEL;
	}

2577 2578 2579
	mdev->pndev[port] = dev;

	netif_carrier_off(dev);
2580 2581
	mlx4_en_set_default_moderation(priv);

2582 2583
	err = register_netdev(dev);
	if (err) {
2584
		en_err(priv, "Netdev registration failed for port %d\n", port);
2585 2586
		goto out;
	}
2587
	priv->registered = 1;
2588 2589 2590 2591

	en_warn(priv, "Using %d TX rings\n", prof->tx_ring_num);
	en_warn(priv, "Using %d RX rings\n", prof->rx_ring_num);

2592 2593
	mlx4_en_update_loopback_state(priv->dev, priv->dev->features);

2594
	/* Configure port */
2595
	mlx4_en_calc_rx_buf(dev);
2596
	err = mlx4_SET_PORT_general(mdev->dev, priv->port,
2597 2598 2599
				    priv->rx_skb_size + ETH_FCS_LEN,
				    prof->tx_pause, prof->tx_ppp,
				    prof->rx_pause, prof->rx_ppp);
2600 2601 2602 2603 2604 2605
	if (err) {
		en_err(priv, "Failed setting port general configurations "
		       "for port %d, with error %d\n", priv->port, err);
		goto out;
	}

2606
	if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
O
Or Gerlitz 已提交
2607
		err = mlx4_SET_PORT_VXLAN(mdev->dev, priv->port, VXLAN_STEER_BY_OUTER_MAC, 1);
2608 2609 2610 2611 2612 2613 2614
		if (err) {
			en_err(priv, "Failed setting port L2 tunnel configuration, err %d\n",
			       err);
			goto out;
		}
	}

2615 2616 2617 2618 2619 2620 2621
	/* Init port */
	en_warn(priv, "Initializing port\n");
	err = mlx4_INIT_PORT(mdev->dev, priv->port);
	if (err) {
		en_err(priv, "Failed Initializing port\n");
		goto out;
	}
2622
	queue_delayed_work(mdev->workqueue, &priv->stats_task, STATS_DELAY);
2623 2624 2625 2626 2627

	if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS)
		queue_delayed_work(mdev->workqueue, &priv->service_task,
				   SERVICE_TASK_DELAY);

2628 2629 2630 2631 2632 2633 2634
	return 0;

out:
	mlx4_en_destroy_netdev(dev);
	return err;
}