team.c 61.9 KB
Newer Older
1
/*
2
 * drivers/net/team/team.c - Network team device driver
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 * Copyright (c) 2011 Jiri Pirko <jpirko@redhat.com>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 */

#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/rcupdate.h>
#include <linux/errno.h>
#include <linux/ctype.h>
#include <linux/notifier.h>
#include <linux/netdevice.h>
J
Jiri Pirko 已提交
21
#include <linux/netpoll.h>
22
#include <linux/if_vlan.h>
23 24 25 26 27 28 29
#include <linux/if_arp.h>
#include <linux/socket.h>
#include <linux/etherdevice.h>
#include <linux/rtnetlink.h>
#include <net/rtnetlink.h>
#include <net/genetlink.h>
#include <net/netlink.h>
J
Jiri Pirko 已提交
30
#include <net/sch_generic.h>
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
#include <linux/if_team.h>

#define DRV_NAME "team"


/**********
 * Helpers
 **********/

#define team_port_exists(dev) (dev->priv_flags & IFF_TEAM_PORT)

static struct team_port *team_port_get_rcu(const struct net_device *dev)
{
	struct team_port *port = rcu_dereference(dev->rx_handler_data);

	return team_port_exists(dev) ? port : NULL;
}

static struct team_port *team_port_get_rtnl(const struct net_device *dev)
{
	struct team_port *port = rtnl_dereference(dev->rx_handler_data);

	return team_port_exists(dev) ? port : NULL;
}

/*
57
 * Since the ability to change device address for open port device is tested in
58 59
 * team_port_add, this function can be called without control of return value
 */
60 61
static int __set_port_dev_addr(struct net_device *port_dev,
			       const unsigned char *dev_addr)
62 63 64
{
	struct sockaddr addr;

65 66
	memcpy(addr.sa_data, dev_addr, port_dev->addr_len);
	addr.sa_family = port_dev->type;
67 68 69
	return dev_set_mac_address(port_dev, &addr);
}

70
static int team_port_set_orig_dev_addr(struct team_port *port)
71
{
72
	return __set_port_dev_addr(port->dev, port->orig.dev_addr);
73 74
}

75
int team_port_set_team_dev_addr(struct team_port *port)
76
{
77
	return __set_port_dev_addr(port->dev, port->team->dev->dev_addr);
78
}
79
EXPORT_SYMBOL(team_port_set_team_dev_addr);
80

81 82 83 84 85
static void team_refresh_port_linkup(struct team_port *port)
{
	port->linkup = port->user.linkup_enabled ? port->user.linkup :
						   port->state.linkup;
}
86

J
Jiri Pirko 已提交
87

88 89 90 91
/*******************
 * Options handling
 *******************/

92 93
struct team_option_inst { /* One for each option instance */
	struct list_head list;
94
	struct list_head tmp_list;
95
	struct team_option *option;
96
	struct team_option_inst_info info;
97 98 99 100 101 102
	bool changed;
	bool removed;
};

static struct team_option *__team_find_option(struct team *team,
					      const char *opt_name)
J
Jiri Pirko 已提交
103 104 105 106 107 108 109 110 111 112
{
	struct team_option *option;

	list_for_each_entry(option, &team->option_list, list) {
		if (strcmp(option->name, opt_name) == 0)
			return option;
	}
	return NULL;
}

113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
static void __team_option_inst_del(struct team_option_inst *opt_inst)
{
	list_del(&opt_inst->list);
	kfree(opt_inst);
}

static void __team_option_inst_del_option(struct team *team,
					  struct team_option *option)
{
	struct team_option_inst *opt_inst, *tmp;

	list_for_each_entry_safe(opt_inst, tmp, &team->option_inst_list, list) {
		if (opt_inst->option == option)
			__team_option_inst_del(opt_inst);
	}
}

J
Jiri Pirko 已提交
130 131 132 133 134 135
static int __team_option_inst_add(struct team *team, struct team_option *option,
				  struct team_port *port)
{
	struct team_option_inst *opt_inst;
	unsigned int array_size;
	unsigned int i;
136
	int err;
J
Jiri Pirko 已提交
137 138 139 140 141 142 143 144 145 146

	array_size = option->array_size;
	if (!array_size)
		array_size = 1; /* No array but still need one instance */

	for (i = 0; i < array_size; i++) {
		opt_inst = kmalloc(sizeof(*opt_inst), GFP_KERNEL);
		if (!opt_inst)
			return -ENOMEM;
		opt_inst->option = option;
147 148
		opt_inst->info.port = port;
		opt_inst->info.array_index = i;
J
Jiri Pirko 已提交
149 150 151
		opt_inst->changed = true;
		opt_inst->removed = false;
		list_add_tail(&opt_inst->list, &team->option_inst_list);
152 153 154 155 156 157
		if (option->init) {
			err = option->init(team, &opt_inst->info);
			if (err)
				return err;
		}

J
Jiri Pirko 已提交
158 159 160 161
	}
	return 0;
}

162 163 164 165 166 167
static int __team_option_inst_add_option(struct team *team,
					 struct team_option *option)
{
	struct team_port *port;
	int err;

J
Jiri Pirko 已提交
168
	if (!option->per_port) {
169
		err = __team_option_inst_add(team, option, NULL);
J
Jiri Pirko 已提交
170 171 172
		if (err)
			goto inst_del_option;
	}
173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205

	list_for_each_entry(port, &team->port_list, list) {
		err = __team_option_inst_add(team, option, port);
		if (err)
			goto inst_del_option;
	}
	return 0;

inst_del_option:
	__team_option_inst_del_option(team, option);
	return err;
}

static void __team_option_inst_mark_removed_option(struct team *team,
						   struct team_option *option)
{
	struct team_option_inst *opt_inst;

	list_for_each_entry(opt_inst, &team->option_inst_list, list) {
		if (opt_inst->option == option) {
			opt_inst->changed = true;
			opt_inst->removed = true;
		}
	}
}

static void __team_option_inst_del_port(struct team *team,
					struct team_port *port)
{
	struct team_option_inst *opt_inst, *tmp;

	list_for_each_entry_safe(opt_inst, tmp, &team->option_inst_list, list) {
		if (opt_inst->option->per_port &&
206
		    opt_inst->info.port == port)
207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236
			__team_option_inst_del(opt_inst);
	}
}

static int __team_option_inst_add_port(struct team *team,
				       struct team_port *port)
{
	struct team_option *option;
	int err;

	list_for_each_entry(option, &team->option_list, list) {
		if (!option->per_port)
			continue;
		err = __team_option_inst_add(team, option, port);
		if (err)
			goto inst_del_port;
	}
	return 0;

inst_del_port:
	__team_option_inst_del_port(team, port);
	return err;
}

static void __team_option_inst_mark_removed_port(struct team *team,
						 struct team_port *port)
{
	struct team_option_inst *opt_inst;

	list_for_each_entry(opt_inst, &team->option_inst_list, list) {
237
		if (opt_inst->info.port == port) {
238 239 240 241 242 243 244 245 246
			opt_inst->changed = true;
			opt_inst->removed = true;
		}
	}
}

static int __team_options_register(struct team *team,
				   const struct team_option *option,
				   size_t option_count)
247 248
{
	int i;
249
	struct team_option **dst_opts;
J
Jiri Pirko 已提交
250
	int err;
251

252 253 254 255
	dst_opts = kzalloc(sizeof(struct team_option *) * option_count,
			   GFP_KERNEL);
	if (!dst_opts)
		return -ENOMEM;
J
Jiri Pirko 已提交
256 257 258
	for (i = 0; i < option_count; i++, option++) {
		if (__team_find_option(team, option->name)) {
			err = -EEXIST;
259
			goto alloc_rollback;
J
Jiri Pirko 已提交
260
		}
261 262
		dst_opts[i] = kmemdup(option, sizeof(*option), GFP_KERNEL);
		if (!dst_opts[i]) {
J
Jiri Pirko 已提交
263
			err = -ENOMEM;
264
			goto alloc_rollback;
J
Jiri Pirko 已提交
265 266 267
		}
	}

268
	for (i = 0; i < option_count; i++) {
269 270 271
		err = __team_option_inst_add_option(team, dst_opts[i]);
		if (err)
			goto inst_rollback;
J
Jiri Pirko 已提交
272
		list_add_tail(&dst_opts[i]->list, &team->option_list);
273
	}
J
Jiri Pirko 已提交
274

275
	kfree(dst_opts);
J
Jiri Pirko 已提交
276 277
	return 0;

278 279 280 281 282 283 284
inst_rollback:
	for (i--; i >= 0; i--)
		__team_option_inst_del_option(team, dst_opts[i]);

	i = option_count - 1;
alloc_rollback:
	for (i--; i >= 0; i--)
J
Jiri Pirko 已提交
285 286
		kfree(dst_opts[i]);

287
	kfree(dst_opts);
J
Jiri Pirko 已提交
288
	return err;
289
}
J
Jiri Pirko 已提交
290

291 292 293 294 295 296 297 298
static void __team_options_mark_removed(struct team *team,
					const struct team_option *option,
					size_t option_count)
{
	int i;

	for (i = 0; i < option_count; i++, option++) {
		struct team_option *del_opt;
299

300
		del_opt = __team_find_option(team, option->name);
301 302
		if (del_opt)
			__team_option_inst_mark_removed_option(team, del_opt);
303 304
	}
}
305 306

static void __team_options_unregister(struct team *team,
J
Jiri Pirko 已提交
307
				      const struct team_option *option,
308 309 310 311
				      size_t option_count)
{
	int i;

J
Jiri Pirko 已提交
312 313 314 315 316
	for (i = 0; i < option_count; i++, option++) {
		struct team_option *del_opt;

		del_opt = __team_find_option(team, option->name);
		if (del_opt) {
317
			__team_option_inst_del_option(team, del_opt);
J
Jiri Pirko 已提交
318 319 320 321
			list_del(&del_opt->list);
			kfree(del_opt);
		}
	}
322 323
}

324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
static void __team_options_change_check(struct team *team);

int team_options_register(struct team *team,
			  const struct team_option *option,
			  size_t option_count)
{
	int err;

	err = __team_options_register(team, option, option_count);
	if (err)
		return err;
	__team_options_change_check(team);
	return 0;
}
EXPORT_SYMBOL(team_options_register);

J
Jiri Pirko 已提交
340 341
void team_options_unregister(struct team *team,
			     const struct team_option *option,
342 343
			     size_t option_count)
{
344 345
	__team_options_mark_removed(team, option, option_count);
	__team_options_change_check(team);
346 347 348 349
	__team_options_unregister(team, option, option_count);
}
EXPORT_SYMBOL(team_options_unregister);

350 351 352 353
static int team_option_get(struct team *team,
			   struct team_option_inst *opt_inst,
			   struct team_gsetter_ctx *ctx)
{
J
Jiri Pirko 已提交
354 355
	if (!opt_inst->option->getter)
		return -EOPNOTSUPP;
356 357 358 359 360 361
	return opt_inst->option->getter(team, ctx);
}

static int team_option_set(struct team *team,
			   struct team_option_inst *opt_inst,
			   struct team_gsetter_ctx *ctx)
362
{
J
Jiri Pirko 已提交
363 364
	if (!opt_inst->option->setter)
		return -EOPNOTSUPP;
365
	return opt_inst->option->setter(team, ctx);
366 367
}

J
Jiri Pirko 已提交
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383
void team_option_inst_set_change(struct team_option_inst_info *opt_inst_info)
{
	struct team_option_inst *opt_inst;

	opt_inst = container_of(opt_inst_info, struct team_option_inst, info);
	opt_inst->changed = true;
}
EXPORT_SYMBOL(team_option_inst_set_change);

void team_options_change_check(struct team *team)
{
	__team_options_change_check(team);
}
EXPORT_SYMBOL(team_options_change_check);


384 385 386 387 388 389 390
/****************
 * Mode handling
 ****************/

static LIST_HEAD(mode_list);
static DEFINE_SPINLOCK(mode_list_lock);

J
Jiri Pirko 已提交
391 392 393 394 395 396
struct team_mode_item {
	struct list_head list;
	const struct team_mode *mode;
};

static struct team_mode_item *__find_mode(const char *kind)
397
{
J
Jiri Pirko 已提交
398
	struct team_mode_item *mitem;
399

J
Jiri Pirko 已提交
400 401 402
	list_for_each_entry(mitem, &mode_list, list) {
		if (strcmp(mitem->mode->kind, kind) == 0)
			return mitem;
403 404 405 406 407 408 409 410 411 412 413 414 415 416
	}
	return NULL;
}

static bool is_good_mode_name(const char *name)
{
	while (*name != '\0') {
		if (!isalpha(*name) && !isdigit(*name) && *name != '_')
			return false;
		name++;
	}
	return true;
}

J
Jiri Pirko 已提交
417
int team_mode_register(const struct team_mode *mode)
418 419
{
	int err = 0;
J
Jiri Pirko 已提交
420
	struct team_mode_item *mitem;
421 422 423 424

	if (!is_good_mode_name(mode->kind) ||
	    mode->priv_size > TEAM_MODE_PRIV_SIZE)
		return -EINVAL;
J
Jiri Pirko 已提交
425 426 427 428 429

	mitem = kmalloc(sizeof(*mitem), GFP_KERNEL);
	if (!mitem)
		return -ENOMEM;

430 431 432
	spin_lock(&mode_list_lock);
	if (__find_mode(mode->kind)) {
		err = -EEXIST;
J
Jiri Pirko 已提交
433
		kfree(mitem);
434 435
		goto unlock;
	}
J
Jiri Pirko 已提交
436 437
	mitem->mode = mode;
	list_add_tail(&mitem->list, &mode_list);
438 439 440 441 442 443
unlock:
	spin_unlock(&mode_list_lock);
	return err;
}
EXPORT_SYMBOL(team_mode_register);

J
Jiri Pirko 已提交
444
void team_mode_unregister(const struct team_mode *mode)
445
{
J
Jiri Pirko 已提交
446 447
	struct team_mode_item *mitem;

448
	spin_lock(&mode_list_lock);
J
Jiri Pirko 已提交
449 450 451 452 453
	mitem = __find_mode(mode->kind);
	if (mitem) {
		list_del_init(&mitem->list);
		kfree(mitem);
	}
454 455 456 457
	spin_unlock(&mode_list_lock);
}
EXPORT_SYMBOL(team_mode_unregister);

J
Jiri Pirko 已提交
458
static const struct team_mode *team_mode_get(const char *kind)
459
{
J
Jiri Pirko 已提交
460 461
	struct team_mode_item *mitem;
	const struct team_mode *mode = NULL;
462 463

	spin_lock(&mode_list_lock);
J
Jiri Pirko 已提交
464 465
	mitem = __find_mode(kind);
	if (!mitem) {
466 467 468
		spin_unlock(&mode_list_lock);
		request_module("team-mode-%s", kind);
		spin_lock(&mode_list_lock);
J
Jiri Pirko 已提交
469
		mitem = __find_mode(kind);
470
	}
J
Jiri Pirko 已提交
471 472
	if (mitem) {
		mode = mitem->mode;
473 474
		if (!try_module_get(mode->owner))
			mode = NULL;
J
Jiri Pirko 已提交
475
	}
476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498

	spin_unlock(&mode_list_lock);
	return mode;
}

static void team_mode_put(const struct team_mode *mode)
{
	module_put(mode->owner);
}

static bool team_dummy_transmit(struct team *team, struct sk_buff *skb)
{
	dev_kfree_skb_any(skb);
	return false;
}

rx_handler_result_t team_dummy_receive(struct team *team,
				       struct team_port *port,
				       struct sk_buff *skb)
{
	return RX_HANDLER_ANOTHER;
}

499 500 501 502 503 504 505 506 507 508 509 510 511 512
static const struct team_mode __team_no_mode = {
	.kind		= "*NOMODE*",
};

static bool team_is_mode_set(struct team *team)
{
	return team->mode != &__team_no_mode;
}

static void team_set_no_mode(struct team *team)
{
	team->mode = &__team_no_mode;
}

513
static void __team_adjust_ops(struct team *team, int en_port_count)
514 515 516 517 518 519
{
	/*
	 * To avoid checks in rx/tx skb paths, ensure here that non-null and
	 * correct ops are always set.
	 */

520 521
	if (!en_port_count || !team_is_mode_set(team) ||
	    !team->mode->ops->transmit)
522 523 524 525
		team->ops.transmit = team_dummy_transmit;
	else
		team->ops.transmit = team->mode->ops->transmit;

526 527
	if (!en_port_count || !team_is_mode_set(team) ||
	    !team->mode->ops->receive)
528 529 530 531 532
		team->ops.receive = team_dummy_receive;
	else
		team->ops.receive = team->mode->ops->receive;
}

533 534 535 536 537
static void team_adjust_ops(struct team *team)
{
	__team_adjust_ops(team, team->en_port_count);
}

538 539 540 541 542 543 544 545 546
/*
 * We can benefit from the fact that it's ensured no port is present
 * at the time of mode change. Therefore no packets are in fly so there's no
 * need to set mode operations in any special way.
 */
static int __team_change_mode(struct team *team,
			      const struct team_mode *new_mode)
{
	/* Check if mode was previously set and do cleanup if so */
547
	if (team_is_mode_set(team)) {
548 549 550 551 552 553 554 555 556
		void (*exit_op)(struct team *team) = team->ops.exit;

		/* Clear ops area so no callback is called any longer */
		memset(&team->ops, 0, sizeof(struct team_mode_ops));
		team_adjust_ops(team);

		if (exit_op)
			exit_op(team);
		team_mode_put(team->mode);
557
		team_set_no_mode(team);
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
		/* zero private data area */
		memset(&team->mode_priv, 0,
		       sizeof(struct team) - offsetof(struct team, mode_priv));
	}

	if (!new_mode)
		return 0;

	if (new_mode->ops->init) {
		int err;

		err = new_mode->ops->init(team);
		if (err)
			return err;
	}

	team->mode = new_mode;
	memcpy(&team->ops, new_mode->ops, sizeof(struct team_mode_ops));
	team_adjust_ops(team);

	return 0;
}

static int team_change_mode(struct team *team, const char *kind)
{
J
Jiri Pirko 已提交
583
	const struct team_mode *new_mode;
584 585 586 587 588 589 590 591
	struct net_device *dev = team->dev;
	int err;

	if (!list_empty(&team->port_list)) {
		netdev_err(dev, "No ports can be present during mode change\n");
		return -EBUSY;
	}

592
	if (team_is_mode_set(team) && strcmp(team->mode->kind, kind) == 0) {
593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634
		netdev_err(dev, "Unable to change to the same mode the team is in\n");
		return -EINVAL;
	}

	new_mode = team_mode_get(kind);
	if (!new_mode) {
		netdev_err(dev, "Mode \"%s\" not found\n", kind);
		return -EINVAL;
	}

	err = __team_change_mode(team, new_mode);
	if (err) {
		netdev_err(dev, "Failed to change to mode \"%s\"\n", kind);
		team_mode_put(new_mode);
		return err;
	}

	netdev_info(dev, "Mode changed to \"%s\"\n", kind);
	return 0;
}


/************************
 * Rx path frame handler
 ************************/

/* note: already called with rcu_read_lock */
static rx_handler_result_t team_handle_frame(struct sk_buff **pskb)
{
	struct sk_buff *skb = *pskb;
	struct team_port *port;
	struct team *team;
	rx_handler_result_t res;

	skb = skb_share_check(skb, GFP_ATOMIC);
	if (!skb)
		return RX_HANDLER_CONSUMED;

	*pskb = skb;

	port = team_port_get_rcu(skb->dev);
	team = port->team;
J
Jiri Pirko 已提交
635 636 637 638 639 640
	if (!team_port_enabled(port)) {
		/* allow exact match delivery for disabled ports */
		res = RX_HANDLER_EXACT;
	} else {
		res = team->ops.receive(team, port, skb);
	}
641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660
	if (res == RX_HANDLER_ANOTHER) {
		struct team_pcpu_stats *pcpu_stats;

		pcpu_stats = this_cpu_ptr(team->pcpu_stats);
		u64_stats_update_begin(&pcpu_stats->syncp);
		pcpu_stats->rx_packets++;
		pcpu_stats->rx_bytes += skb->len;
		if (skb->pkt_type == PACKET_MULTICAST)
			pcpu_stats->rx_multicast++;
		u64_stats_update_end(&pcpu_stats->syncp);

		skb->dev = team->dev;
	} else {
		this_cpu_inc(team->pcpu_stats->rx_dropped);
	}

	return res;
}


661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776
/*************************************
 * Multiqueue Tx port select override
 *************************************/

static int team_queue_override_init(struct team *team)
{
	struct list_head *listarr;
	unsigned int queue_cnt = team->dev->num_tx_queues - 1;
	unsigned int i;

	if (!queue_cnt)
		return 0;
	listarr = kmalloc(sizeof(struct list_head) * queue_cnt, GFP_KERNEL);
	if (!listarr)
		return -ENOMEM;
	team->qom_lists = listarr;
	for (i = 0; i < queue_cnt; i++)
		INIT_LIST_HEAD(listarr++);
	return 0;
}

static void team_queue_override_fini(struct team *team)
{
	kfree(team->qom_lists);
}

static struct list_head *__team_get_qom_list(struct team *team, u16 queue_id)
{
	return &team->qom_lists[queue_id - 1];
}

/*
 * note: already called with rcu_read_lock
 */
static bool team_queue_override_transmit(struct team *team, struct sk_buff *skb)
{
	struct list_head *qom_list;
	struct team_port *port;

	if (!team->queue_override_enabled || !skb->queue_mapping)
		return false;
	qom_list = __team_get_qom_list(team, skb->queue_mapping);
	list_for_each_entry_rcu(port, qom_list, qom_list) {
		if (!team_dev_queue_xmit(team, port, skb))
			return true;
	}
	return false;
}

static void __team_queue_override_port_del(struct team *team,
					   struct team_port *port)
{
	list_del_rcu(&port->qom_list);
	synchronize_rcu();
	INIT_LIST_HEAD(&port->qom_list);
}

static bool team_queue_override_port_has_gt_prio_than(struct team_port *port,
						      struct team_port *cur)
{
	if (port->priority < cur->priority)
		return true;
	if (port->priority > cur->priority)
		return false;
	if (port->index < cur->index)
		return true;
	return false;
}

static void __team_queue_override_port_add(struct team *team,
					   struct team_port *port)
{
	struct team_port *cur;
	struct list_head *qom_list;
	struct list_head *node;

	if (!port->queue_id || !team_port_enabled(port))
		return;

	qom_list = __team_get_qom_list(team, port->queue_id);
	node = qom_list;
	list_for_each_entry(cur, qom_list, qom_list) {
		if (team_queue_override_port_has_gt_prio_than(port, cur))
			break;
		node = &cur->qom_list;
	}
	list_add_tail_rcu(&port->qom_list, node);
}

static void __team_queue_override_enabled_check(struct team *team)
{
	struct team_port *port;
	bool enabled = false;

	list_for_each_entry(port, &team->port_list, list) {
		if (!list_empty(&port->qom_list)) {
			enabled = true;
			break;
		}
	}
	if (enabled == team->queue_override_enabled)
		return;
	netdev_dbg(team->dev, "%s queue override\n",
		   enabled ? "Enabling" : "Disabling");
	team->queue_override_enabled = enabled;
}

static void team_queue_override_port_refresh(struct team *team,
					     struct team_port *port)
{
	__team_queue_override_port_del(team, port);
	__team_queue_override_port_add(team, port);
	__team_queue_override_enabled_check(team);
}


777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792
/****************
 * Port handling
 ****************/

static bool team_port_find(const struct team *team,
			   const struct team_port *port)
{
	struct team_port *cur;

	list_for_each_entry(cur, &team->port_list, list)
		if (cur == port)
			return true;
	return false;
}

/*
J
Jiri Pirko 已提交
793 794 795 796
 * Enable/disable port by adding to enabled port hashlist and setting
 * port->index (Might be racy so reader could see incorrect ifindex when
 * processing a flying packet, but that is not a problem). Write guarded
 * by team->lock.
797
 */
J
Jiri Pirko 已提交
798 799
static void team_port_enable(struct team *team,
			     struct team_port *port)
800
{
J
Jiri Pirko 已提交
801 802 803
	if (team_port_enabled(port))
		return;
	port->index = team->en_port_count++;
804 805
	hlist_add_head_rcu(&port->hlist,
			   team_port_index_hash(team, port->index));
806
	team_adjust_ops(team);
807
	team_queue_override_port_refresh(team, port);
808 809
	if (team->ops.port_enabled)
		team->ops.port_enabled(team, port);
810 811 812 813 814 815 816
}

static void __reconstruct_port_hlist(struct team *team, int rm_index)
{
	int i;
	struct team_port *port;

J
Jiri Pirko 已提交
817
	for (i = rm_index + 1; i < team->en_port_count; i++) {
818 819 820 821 822 823 824 825
		port = team_get_port_by_index(team, i);
		hlist_del_rcu(&port->hlist);
		port->index--;
		hlist_add_head_rcu(&port->hlist,
				   team_port_index_hash(team, port->index));
	}
}

J
Jiri Pirko 已提交
826 827
static void team_port_disable(struct team *team,
			      struct team_port *port)
828
{
J
Jiri Pirko 已提交
829 830
	if (!team_port_enabled(port))
		return;
831 832
	if (team->ops.port_disabled)
		team->ops.port_disabled(team, port);
833
	hlist_del_rcu(&port->hlist);
834
	__reconstruct_port_hlist(team, port->index);
J
Jiri Pirko 已提交
835
	port->index = -1;
836
	team_queue_override_port_refresh(team, port);
837 838 839 840 841 842 843
	__team_adjust_ops(team, team->en_port_count - 1);
	/*
	 * Wait until readers see adjusted ops. This ensures that
	 * readers never see team->en_port_count == 0
	 */
	synchronize_rcu();
	team->en_port_count--;
844 845 846 847 848 849 850 851 852 853 854
}

#define TEAM_VLAN_FEATURES (NETIF_F_ALL_CSUM | NETIF_F_SG | \
			    NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \
			    NETIF_F_HIGHDMA | NETIF_F_LRO)

static void __team_compute_features(struct team *team)
{
	struct team_port *port;
	u32 vlan_features = TEAM_VLAN_FEATURES;
	unsigned short max_hard_header_len = ETH_HLEN;
855
	unsigned int flags, dst_release_flag = IFF_XMIT_DST_RELEASE;
856 857 858 859 860 861

	list_for_each_entry(port, &team->port_list, list) {
		vlan_features = netdev_increment_features(vlan_features,
					port->dev->vlan_features,
					TEAM_VLAN_FEATURES);

862
		dst_release_flag &= port->dev->priv_flags;
863 864 865 866 867 868 869
		if (port->dev->hard_header_len > max_hard_header_len)
			max_hard_header_len = port->dev->hard_header_len;
	}

	team->dev->vlan_features = vlan_features;
	team->dev->hard_header_len = max_hard_header_len;

870 871 872
	flags = team->dev->priv_flags & ~IFF_XMIT_DST_RELEASE;
	team->dev->priv_flags = flags | dst_release_flag;

873 874 875 876 877
	netdev_change_features(team->dev);
}

static void team_compute_features(struct team *team)
{
878
	mutex_lock(&team->lock);
879
	__team_compute_features(team);
880
	mutex_unlock(&team->lock);
881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914
}

static int team_port_enter(struct team *team, struct team_port *port)
{
	int err = 0;

	dev_hold(team->dev);
	port->dev->priv_flags |= IFF_TEAM_PORT;
	if (team->ops.port_enter) {
		err = team->ops.port_enter(team, port);
		if (err) {
			netdev_err(team->dev, "Device %s failed to enter team mode\n",
				   port->dev->name);
			goto err_port_enter;
		}
	}

	return 0;

err_port_enter:
	port->dev->priv_flags &= ~IFF_TEAM_PORT;
	dev_put(team->dev);

	return err;
}

static void team_port_leave(struct team *team, struct team_port *port)
{
	if (team->ops.port_leave)
		team->ops.port_leave(team, port);
	port->dev->priv_flags &= ~IFF_TEAM_PORT;
	dev_put(team->dev);
}

J
Jiri Pirko 已提交
915
#ifdef CONFIG_NET_POLL_CONTROLLER
916 917
static int team_port_enable_netpoll(struct team *team, struct team_port *port,
				    gfp_t gfp)
J
Jiri Pirko 已提交
918 919 920 921
{
	struct netpoll *np;
	int err;

922
	np = kzalloc(sizeof(*np), gfp);
J
Jiri Pirko 已提交
923 924 925
	if (!np)
		return -ENOMEM;

926
	err = __netpoll_setup(np, port->dev, gfp);
J
Jiri Pirko 已提交
927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954
	if (err) {
		kfree(np);
		return err;
	}
	port->np = np;
	return err;
}

static void team_port_disable_netpoll(struct team_port *port)
{
	struct netpoll *np = port->np;

	if (!np)
		return;
	port->np = NULL;

	/* Wait for transmitting packets to finish before freeing. */
	synchronize_rcu_bh();
	__netpoll_cleanup(np);
	kfree(np);
}

static struct netpoll_info *team_netpoll_info(struct team *team)
{
	return team->dev->npinfo;
}

#else
955 956
static int team_port_enable_netpoll(struct team *team, struct team_port *port,
				    gfp_t gfp)
J
Jiri Pirko 已提交
957 958 959 960 961 962 963 964 965 966 967 968
{
	return 0;
}
static void team_port_disable_netpoll(struct team_port *port)
{
}
static struct netpoll_info *team_netpoll_info(struct team *team)
{
	return NULL;
}
#endif

J
Jiri Pirko 已提交
969
static void __team_port_change_port_added(struct team_port *port, bool linkup);
970 971
static int team_dev_type_check_change(struct net_device *dev,
				      struct net_device *port_dev);
972 973 974 975 976 977 978 979

static int team_port_add(struct team *team, struct net_device *port_dev)
{
	struct net_device *dev = team->dev;
	struct team_port *port;
	char *portname = port_dev->name;
	int err;

980 981
	if (port_dev->flags & IFF_LOOPBACK) {
		netdev_err(dev, "Device %s is loopback device. Loopback devices can't be added as a team port\n",
982 983 984 985 986 987 988 989 990 991
			   portname);
		return -EINVAL;
	}

	if (team_port_exists(port_dev)) {
		netdev_err(dev, "Device %s is already a port "
				"of a team device\n", portname);
		return -EBUSY;
	}

992 993 994 995 996 997 998
	if (port_dev->features & NETIF_F_VLAN_CHALLENGED &&
	    vlan_uses_dev(dev)) {
		netdev_err(dev, "Device %s is VLAN challenged and team device has VLAN set up\n",
			   portname);
		return -EPERM;
	}

999 1000 1001 1002
	err = team_dev_type_check_change(dev, port_dev);
	if (err)
		return err;

1003 1004 1005 1006 1007 1008
	if (port_dev->flags & IFF_UP) {
		netdev_err(dev, "Device %s is up. Set it down before adding it as a team port\n",
			   portname);
		return -EBUSY;
	}

J
Jiri Pirko 已提交
1009 1010
	port = kzalloc(sizeof(struct team_port) + team->mode->port_priv_size,
		       GFP_KERNEL);
1011 1012 1013 1014 1015
	if (!port)
		return -ENOMEM;

	port->dev = port_dev;
	port->team = team;
1016
	INIT_LIST_HEAD(&port->qom_list);
1017 1018 1019 1020 1021 1022 1023 1024

	port->orig.mtu = port_dev->mtu;
	err = dev_set_mtu(port_dev, dev->mtu);
	if (err) {
		netdev_dbg(dev, "Error %d calling dev_set_mtu\n", err);
		goto err_set_mtu;
	}

1025
	memcpy(port->orig.dev_addr, port_dev->dev_addr, port_dev->addr_len);
1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040

	err = team_port_enter(team, port);
	if (err) {
		netdev_err(dev, "Device %s failed to enter team mode\n",
			   portname);
		goto err_port_enter;
	}

	err = dev_open(port_dev);
	if (err) {
		netdev_dbg(dev, "Device %s opening failed\n",
			   portname);
		goto err_dev_open;
	}

J
Jiri Pirko 已提交
1041 1042 1043 1044 1045 1046 1047
	err = vlan_vids_add_by_dev(port_dev, dev);
	if (err) {
		netdev_err(dev, "Failed to add vlan ids to device %s\n",
				portname);
		goto err_vids_add;
	}

J
Jiri Pirko 已提交
1048
	if (team_netpoll_info(team)) {
1049
		err = team_port_enable_netpoll(team, port, GFP_KERNEL);
J
Jiri Pirko 已提交
1050 1051 1052 1053 1054 1055 1056
		if (err) {
			netdev_err(dev, "Failed to enable netpoll on device %s\n",
				   portname);
			goto err_enable_netpoll;
		}
	}

1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070
	err = netdev_set_master(port_dev, dev);
	if (err) {
		netdev_err(dev, "Device %s failed to set master\n", portname);
		goto err_set_master;
	}

	err = netdev_rx_handler_register(port_dev, team_handle_frame,
					 port);
	if (err) {
		netdev_err(dev, "Device %s failed to register rx_handler\n",
			   portname);
		goto err_handler_register;
	}

1071
	err = __team_option_inst_add_port(team, port);
1072 1073 1074 1075 1076 1077
	if (err) {
		netdev_err(dev, "Device %s failed to add per-port options\n",
			   portname);
		goto err_option_port_add;
	}

J
Jiri Pirko 已提交
1078 1079 1080
	port->index = -1;
	team_port_enable(team, port);
	list_add_tail_rcu(&port->list, &team->port_list);
1081
	__team_compute_features(team);
J
Jiri Pirko 已提交
1082
	__team_port_change_port_added(port, !!netif_carrier_ok(port_dev));
1083
	__team_options_change_check(team);
1084 1085 1086 1087 1088

	netdev_info(dev, "Port device %s added\n", portname);

	return 0;

1089 1090 1091
err_option_port_add:
	netdev_rx_handler_unregister(port_dev);

1092 1093 1094 1095
err_handler_register:
	netdev_set_master(port_dev, NULL);

err_set_master:
J
Jiri Pirko 已提交
1096 1097 1098
	team_port_disable_netpoll(port);

err_enable_netpoll:
J
Jiri Pirko 已提交
1099 1100 1101
	vlan_vids_del_by_dev(port_dev, dev);

err_vids_add:
1102 1103 1104 1105
	dev_close(port_dev);

err_dev_open:
	team_port_leave(team, port);
1106
	team_port_set_orig_dev_addr(port);
1107 1108 1109 1110 1111 1112 1113 1114 1115 1116

err_port_enter:
	dev_set_mtu(port_dev, port->orig.mtu);

err_set_mtu:
	kfree(port);

	return err;
}

J
Jiri Pirko 已提交
1117 1118
static void __team_port_change_port_removed(struct team_port *port);

1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
static int team_port_del(struct team *team, struct net_device *port_dev)
{
	struct net_device *dev = team->dev;
	struct team_port *port;
	char *portname = port_dev->name;

	port = team_port_get_rtnl(port_dev);
	if (!port || !team_port_find(team, port)) {
		netdev_err(dev, "Device %s does not act as a port of this team\n",
			   portname);
		return -ENOENT;
	}

1132 1133 1134
	__team_option_inst_mark_removed_port(team, port);
	__team_options_change_check(team);
	__team_option_inst_del_port(team, port);
J
Jiri Pirko 已提交
1135
	__team_port_change_port_removed(port);
J
Jiri Pirko 已提交
1136 1137
	team_port_disable(team, port);
	list_del_rcu(&port->list);
1138 1139
	netdev_rx_handler_unregister(port_dev);
	netdev_set_master(port_dev, NULL);
J
Jiri Pirko 已提交
1140
	team_port_disable_netpoll(port);
J
Jiri Pirko 已提交
1141
	vlan_vids_del_by_dev(port_dev, dev);
1142 1143
	dev_close(port_dev);
	team_port_leave(team, port);
1144
	team_port_set_orig_dev_addr(port);
1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
	dev_set_mtu(port_dev, port->orig.mtu);
	synchronize_rcu();
	kfree(port);
	netdev_info(dev, "Port device %s removed\n", portname);
	__team_compute_features(team);

	return 0;
}


/*****************
 * Net device ops
 *****************/

1159
static int team_mode_option_get(struct team *team, struct team_gsetter_ctx *ctx)
1160
{
1161
	ctx->data.str_val = team->mode->kind;
1162 1163 1164
	return 0;
}

1165
static int team_mode_option_set(struct team *team, struct team_gsetter_ctx *ctx)
1166
{
1167
	return team_change_mode(team, ctx->data.str_val);
1168 1169
}

1170 1171 1172
static int team_port_en_option_get(struct team *team,
				   struct team_gsetter_ctx *ctx)
{
1173 1174 1175
	struct team_port *port = ctx->info->port;

	ctx->data.bool_val = team_port_enabled(port);
1176 1177 1178 1179 1180 1181
	return 0;
}

static int team_port_en_option_set(struct team *team,
				   struct team_gsetter_ctx *ctx)
{
1182 1183
	struct team_port *port = ctx->info->port;

1184
	if (ctx->data.bool_val)
1185
		team_port_enable(team, port);
1186
	else
1187
		team_port_disable(team, port);
1188 1189 1190
	return 0;
}

1191 1192 1193
static int team_user_linkup_option_get(struct team *team,
				       struct team_gsetter_ctx *ctx)
{
1194 1195 1196
	struct team_port *port = ctx->info->port;

	ctx->data.bool_val = port->user.linkup;
1197 1198 1199 1200 1201 1202
	return 0;
}

static int team_user_linkup_option_set(struct team *team,
				       struct team_gsetter_ctx *ctx)
{
1203 1204 1205 1206
	struct team_port *port = ctx->info->port;

	port->user.linkup = ctx->data.bool_val;
	team_refresh_port_linkup(port);
1207 1208 1209 1210 1211 1212
	return 0;
}

static int team_user_linkup_en_option_get(struct team *team,
					  struct team_gsetter_ctx *ctx)
{
1213
	struct team_port *port = ctx->info->port;
1214 1215 1216 1217 1218 1219 1220 1221

	ctx->data.bool_val = port->user.linkup_enabled;
	return 0;
}

static int team_user_linkup_en_option_set(struct team *team,
					  struct team_gsetter_ctx *ctx)
{
1222
	struct team_port *port = ctx->info->port;
1223 1224

	port->user.linkup_enabled = ctx->data.bool_val;
1225
	team_refresh_port_linkup(port);
1226 1227 1228
	return 0;
}

J
Jiri Pirko 已提交
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243
static int team_priority_option_get(struct team *team,
				    struct team_gsetter_ctx *ctx)
{
	struct team_port *port = ctx->info->port;

	ctx->data.s32_val = port->priority;
	return 0;
}

static int team_priority_option_set(struct team *team,
				    struct team_gsetter_ctx *ctx)
{
	struct team_port *port = ctx->info->port;

	port->priority = ctx->data.s32_val;
1244 1245 1246 1247 1248 1249 1250 1251 1252 1253
	team_queue_override_port_refresh(team, port);
	return 0;
}

static int team_queue_id_option_get(struct team *team,
				    struct team_gsetter_ctx *ctx)
{
	struct team_port *port = ctx->info->port;

	ctx->data.u32_val = port->queue_id;
J
Jiri Pirko 已提交
1254 1255 1256
	return 0;
}

1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271
static int team_queue_id_option_set(struct team *team,
				    struct team_gsetter_ctx *ctx)
{
	struct team_port *port = ctx->info->port;

	if (port->queue_id == ctx->data.u32_val)
		return 0;
	if (ctx->data.u32_val >= team->dev->real_num_tx_queues)
		return -EINVAL;
	port->queue_id = ctx->data.u32_val;
	team_queue_override_port_refresh(team, port);
	return 0;
}


J
Jiri Pirko 已提交
1272
static const struct team_option team_options[] = {
1273 1274 1275 1276 1277 1278
	{
		.name = "mode",
		.type = TEAM_OPTION_TYPE_STRING,
		.getter = team_mode_option_get,
		.setter = team_mode_option_set,
	},
1279 1280 1281 1282 1283 1284 1285
	{
		.name = "enabled",
		.type = TEAM_OPTION_TYPE_BOOL,
		.per_port = true,
		.getter = team_port_en_option_get,
		.setter = team_port_en_option_set,
	},
1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299
	{
		.name = "user_linkup",
		.type = TEAM_OPTION_TYPE_BOOL,
		.per_port = true,
		.getter = team_user_linkup_option_get,
		.setter = team_user_linkup_option_set,
	},
	{
		.name = "user_linkup_enabled",
		.type = TEAM_OPTION_TYPE_BOOL,
		.per_port = true,
		.getter = team_user_linkup_en_option_get,
		.setter = team_user_linkup_en_option_set,
	},
J
Jiri Pirko 已提交
1300 1301 1302 1303 1304 1305 1306
	{
		.name = "priority",
		.type = TEAM_OPTION_TYPE_S32,
		.per_port = true,
		.getter = team_priority_option_get,
		.setter = team_priority_option_set,
	},
1307 1308 1309 1310 1311 1312 1313
	{
		.name = "queue_id",
		.type = TEAM_OPTION_TYPE_U32,
		.per_port = true,
		.getter = team_queue_id_option_get,
		.setter = team_queue_id_option_set,
	},
1314 1315
};

J
Jiri Pirko 已提交
1316 1317
static struct lock_class_key team_netdev_xmit_lock_key;
static struct lock_class_key team_netdev_addr_lock_key;
1318
static struct lock_class_key team_tx_busylock_key;
J
Jiri Pirko 已提交
1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330

static void team_set_lockdep_class_one(struct net_device *dev,
				       struct netdev_queue *txq,
				       void *unused)
{
	lockdep_set_class(&txq->_xmit_lock, &team_netdev_xmit_lock_key);
}

static void team_set_lockdep_class(struct net_device *dev)
{
	lockdep_set_class(&dev->addr_list_lock, &team_netdev_addr_lock_key);
	netdev_for_each_tx_queue(dev, team_set_lockdep_class_one, NULL);
1331
	dev->qdisc_tx_busylock = &team_tx_busylock_key;
J
Jiri Pirko 已提交
1332 1333
}

1334 1335 1336 1337
static int team_init(struct net_device *dev)
{
	struct team *team = netdev_priv(dev);
	int i;
J
Jiri Pirko 已提交
1338
	int err;
1339 1340

	team->dev = dev;
1341
	mutex_init(&team->lock);
1342
	team_set_no_mode(team);
1343 1344 1345 1346 1347 1348

	team->pcpu_stats = alloc_percpu(struct team_pcpu_stats);
	if (!team->pcpu_stats)
		return -ENOMEM;

	for (i = 0; i < TEAM_PORT_HASHENTRIES; i++)
J
Jiri Pirko 已提交
1349
		INIT_HLIST_HEAD(&team->en_port_hlist[i]);
1350
	INIT_LIST_HEAD(&team->port_list);
1351 1352 1353
	err = team_queue_override_init(team);
	if (err)
		goto err_team_queue_override_init;
1354 1355 1356 1357

	team_adjust_ops(team);

	INIT_LIST_HEAD(&team->option_list);
1358
	INIT_LIST_HEAD(&team->option_inst_list);
J
Jiri Pirko 已提交
1359 1360 1361
	err = team_options_register(team, team_options, ARRAY_SIZE(team_options));
	if (err)
		goto err_options_register;
1362 1363
	netif_carrier_off(dev);

J
Jiri Pirko 已提交
1364 1365
	team_set_lockdep_class(dev);

1366
	return 0;
J
Jiri Pirko 已提交
1367 1368

err_options_register:
1369 1370
	team_queue_override_fini(team);
err_team_queue_override_init:
J
Jiri Pirko 已提交
1371 1372 1373
	free_percpu(team->pcpu_stats);

	return err;
1374 1375 1376 1377 1378 1379 1380 1381
}

static void team_uninit(struct net_device *dev)
{
	struct team *team = netdev_priv(dev);
	struct team_port *port;
	struct team_port *tmp;

1382
	mutex_lock(&team->lock);
1383 1384 1385 1386 1387
	list_for_each_entry_safe(port, tmp, &team->port_list, list)
		team_port_del(team, port->dev);

	__team_change_mode(team, NULL); /* cleanup */
	__team_options_unregister(team, team_options, ARRAY_SIZE(team_options));
1388
	team_queue_override_fini(team);
1389
	mutex_unlock(&team->lock);
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
}

static void team_destructor(struct net_device *dev)
{
	struct team *team = netdev_priv(dev);

	free_percpu(team->pcpu_stats);
	free_netdev(dev);
}

static int team_open(struct net_device *dev)
{
	netif_carrier_on(dev);
	return 0;
}

static int team_close(struct net_device *dev)
{
	netif_carrier_off(dev);
	return 0;
}

/*
 * note: already called with rcu_read_lock
 */
static netdev_tx_t team_xmit(struct sk_buff *skb, struct net_device *dev)
{
	struct team *team = netdev_priv(dev);
1418
	bool tx_success;
1419 1420
	unsigned int len = skb->len;

1421 1422 1423
	tx_success = team_queue_override_transmit(team, skb);
	if (!tx_success)
		tx_success = team->ops.transmit(team, skb);
1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438
	if (tx_success) {
		struct team_pcpu_stats *pcpu_stats;

		pcpu_stats = this_cpu_ptr(team->pcpu_stats);
		u64_stats_update_begin(&pcpu_stats->syncp);
		pcpu_stats->tx_packets++;
		pcpu_stats->tx_bytes += len;
		u64_stats_update_end(&pcpu_stats->syncp);
	} else {
		this_cpu_inc(team->pcpu_stats->tx_dropped);
	}

	return NETDEV_TX_OK;
}

J
Jiri Pirko 已提交
1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461
static u16 team_select_queue(struct net_device *dev, struct sk_buff *skb)
{
	/*
	 * This helper function exists to help dev_pick_tx get the correct
	 * destination queue.  Using a helper function skips a call to
	 * skb_tx_hash and will put the skbs in the queue we expect on their
	 * way down to the team driver.
	 */
	u16 txq = skb_rx_queue_recorded(skb) ? skb_get_rx_queue(skb) : 0;

	/*
	 * Save the original txq to restore before passing to the driver
	 */
	qdisc_skb_cb(skb)->slave_dev_queue_mapping = skb->queue_mapping;

	if (unlikely(txq >= dev->real_num_tx_queues)) {
		do {
			txq -= dev->real_num_tx_queues;
		} while (txq >= dev->real_num_tx_queues);
	}
	return txq;
}

1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496
static void team_change_rx_flags(struct net_device *dev, int change)
{
	struct team *team = netdev_priv(dev);
	struct team_port *port;
	int inc;

	rcu_read_lock();
	list_for_each_entry_rcu(port, &team->port_list, list) {
		if (change & IFF_PROMISC) {
			inc = dev->flags & IFF_PROMISC ? 1 : -1;
			dev_set_promiscuity(port->dev, inc);
		}
		if (change & IFF_ALLMULTI) {
			inc = dev->flags & IFF_ALLMULTI ? 1 : -1;
			dev_set_allmulti(port->dev, inc);
		}
	}
	rcu_read_unlock();
}

static void team_set_rx_mode(struct net_device *dev)
{
	struct team *team = netdev_priv(dev);
	struct team_port *port;

	rcu_read_lock();
	list_for_each_entry_rcu(port, &team->port_list, list) {
		dev_uc_sync(port->dev, dev);
		dev_mc_sync(port->dev, dev);
	}
	rcu_read_unlock();
}

static int team_set_mac_address(struct net_device *dev, void *p)
{
1497
	struct sockaddr *addr = p;
1498 1499 1500
	struct team *team = netdev_priv(dev);
	struct team_port *port;

1501 1502 1503 1504
	if (dev->type == ARPHRD_ETHER && !is_valid_ether_addr(addr->sa_data))
		return -EADDRNOTAVAIL;
	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
	dev->addr_assign_type &= ~NET_ADDR_RANDOM;
1505 1506
	rcu_read_lock();
	list_for_each_entry_rcu(port, &team->port_list, list)
1507 1508
		if (team->ops.port_change_dev_addr)
			team->ops.port_change_dev_addr(team, port);
1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522
	rcu_read_unlock();
	return 0;
}

static int team_change_mtu(struct net_device *dev, int new_mtu)
{
	struct team *team = netdev_priv(dev);
	struct team_port *port;
	int err;

	/*
	 * Alhough this is reader, it's guarded by team lock. It's not possible
	 * to traverse list in reverse under rcu_read_lock
	 */
1523
	mutex_lock(&team->lock);
1524 1525 1526 1527 1528 1529 1530 1531
	list_for_each_entry(port, &team->port_list, list) {
		err = dev_set_mtu(port->dev, new_mtu);
		if (err) {
			netdev_err(dev, "Device %s failed to change mtu",
				   port->dev->name);
			goto unwind;
		}
	}
1532
	mutex_unlock(&team->lock);
1533 1534 1535 1536 1537 1538 1539 1540

	dev->mtu = new_mtu;

	return 0;

unwind:
	list_for_each_entry_continue_reverse(port, &team->port_list, list)
		dev_set_mtu(port->dev, dev->mtu);
1541
	mutex_unlock(&team->lock);
1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583

	return err;
}

static struct rtnl_link_stats64 *
team_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
{
	struct team *team = netdev_priv(dev);
	struct team_pcpu_stats *p;
	u64 rx_packets, rx_bytes, rx_multicast, tx_packets, tx_bytes;
	u32 rx_dropped = 0, tx_dropped = 0;
	unsigned int start;
	int i;

	for_each_possible_cpu(i) {
		p = per_cpu_ptr(team->pcpu_stats, i);
		do {
			start = u64_stats_fetch_begin_bh(&p->syncp);
			rx_packets	= p->rx_packets;
			rx_bytes	= p->rx_bytes;
			rx_multicast	= p->rx_multicast;
			tx_packets	= p->tx_packets;
			tx_bytes	= p->tx_bytes;
		} while (u64_stats_fetch_retry_bh(&p->syncp, start));

		stats->rx_packets	+= rx_packets;
		stats->rx_bytes		+= rx_bytes;
		stats->multicast	+= rx_multicast;
		stats->tx_packets	+= tx_packets;
		stats->tx_bytes		+= tx_bytes;
		/*
		 * rx_dropped & tx_dropped are u32, updated
		 * without syncp protection.
		 */
		rx_dropped	+= p->rx_dropped;
		tx_dropped	+= p->tx_dropped;
	}
	stats->rx_dropped	= rx_dropped;
	stats->tx_dropped	= tx_dropped;
	return stats;
}

1584
static int team_vlan_rx_add_vid(struct net_device *dev, uint16_t vid)
1585 1586 1587
{
	struct team *team = netdev_priv(dev);
	struct team_port *port;
1588
	int err;
1589

1590 1591 1592 1593 1594 1595 1596 1597 1598
	/*
	 * Alhough this is reader, it's guarded by team lock. It's not possible
	 * to traverse list in reverse under rcu_read_lock
	 */
	mutex_lock(&team->lock);
	list_for_each_entry(port, &team->port_list, list) {
		err = vlan_vid_add(port->dev, vid);
		if (err)
			goto unwind;
1599
	}
1600
	mutex_unlock(&team->lock);
1601 1602

	return 0;
1603 1604 1605 1606 1607 1608 1609

unwind:
	list_for_each_entry_continue_reverse(port, &team->port_list, list)
		vlan_vid_del(port->dev, vid);
	mutex_unlock(&team->lock);

	return err;
1610 1611
}

1612
static int team_vlan_rx_kill_vid(struct net_device *dev, uint16_t vid)
1613 1614 1615 1616 1617
{
	struct team *team = netdev_priv(dev);
	struct team_port *port;

	rcu_read_lock();
1618 1619
	list_for_each_entry_rcu(port, &team->port_list, list)
		vlan_vid_del(port->dev, vid);
1620
	rcu_read_unlock();
1621 1622

	return 0;
1623 1624
}

J
Jiri Pirko 已提交
1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647
#ifdef CONFIG_NET_POLL_CONTROLLER
static void team_poll_controller(struct net_device *dev)
{
}

static void __team_netpoll_cleanup(struct team *team)
{
	struct team_port *port;

	list_for_each_entry(port, &team->port_list, list)
		team_port_disable_netpoll(port);
}

static void team_netpoll_cleanup(struct net_device *dev)
{
	struct team *team = netdev_priv(dev);

	mutex_lock(&team->lock);
	__team_netpoll_cleanup(team);
	mutex_unlock(&team->lock);
}

static int team_netpoll_setup(struct net_device *dev,
1648
			      struct netpoll_info *npifo, gfp_t gfp)
J
Jiri Pirko 已提交
1649 1650 1651
{
	struct team *team = netdev_priv(dev);
	struct team_port *port;
1652
	int err = 0;
J
Jiri Pirko 已提交
1653 1654 1655

	mutex_lock(&team->lock);
	list_for_each_entry(port, &team->port_list, list) {
1656
		err = team_port_enable_netpoll(team, port, gfp);
J
Jiri Pirko 已提交
1657 1658 1659 1660 1661 1662 1663 1664 1665 1666
		if (err) {
			__team_netpoll_cleanup(team);
			break;
		}
	}
	mutex_unlock(&team->lock);
	return err;
}
#endif

1667 1668 1669 1670 1671
static int team_add_slave(struct net_device *dev, struct net_device *port_dev)
{
	struct team *team = netdev_priv(dev);
	int err;

1672
	mutex_lock(&team->lock);
1673
	err = team_port_add(team, port_dev);
1674
	mutex_unlock(&team->lock);
1675 1676 1677 1678 1679 1680 1681 1682
	return err;
}

static int team_del_slave(struct net_device *dev, struct net_device *port_dev)
{
	struct team *team = netdev_priv(dev);
	int err;

1683
	mutex_lock(&team->lock);
1684
	err = team_port_del(team, port_dev);
1685
	mutex_unlock(&team->lock);
1686 1687 1688
	return err;
}

J
Jiri Pirko 已提交
1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709
static netdev_features_t team_fix_features(struct net_device *dev,
					   netdev_features_t features)
{
	struct team_port *port;
	struct team *team = netdev_priv(dev);
	netdev_features_t mask;

	mask = features;
	features &= ~NETIF_F_ONE_FOR_ALL;
	features |= NETIF_F_ALL_FOR_ALL;

	rcu_read_lock();
	list_for_each_entry_rcu(port, &team->port_list, list) {
		features = netdev_increment_features(features,
						     port->dev->features,
						     mask);
	}
	rcu_read_unlock();
	return features;
}

1710 1711 1712 1713 1714 1715
static const struct net_device_ops team_netdev_ops = {
	.ndo_init		= team_init,
	.ndo_uninit		= team_uninit,
	.ndo_open		= team_open,
	.ndo_stop		= team_close,
	.ndo_start_xmit		= team_xmit,
J
Jiri Pirko 已提交
1716
	.ndo_select_queue	= team_select_queue,
1717 1718 1719 1720 1721 1722 1723
	.ndo_change_rx_flags	= team_change_rx_flags,
	.ndo_set_rx_mode	= team_set_rx_mode,
	.ndo_set_mac_address	= team_set_mac_address,
	.ndo_change_mtu		= team_change_mtu,
	.ndo_get_stats64	= team_get_stats64,
	.ndo_vlan_rx_add_vid	= team_vlan_rx_add_vid,
	.ndo_vlan_rx_kill_vid	= team_vlan_rx_kill_vid,
J
Jiri Pirko 已提交
1724 1725 1726 1727 1728
#ifdef CONFIG_NET_POLL_CONTROLLER
	.ndo_poll_controller	= team_poll_controller,
	.ndo_netpoll_setup	= team_netpoll_setup,
	.ndo_netpoll_cleanup	= team_netpoll_cleanup,
#endif
1729 1730
	.ndo_add_slave		= team_add_slave,
	.ndo_del_slave		= team_del_slave,
J
Jiri Pirko 已提交
1731
	.ndo_fix_features	= team_fix_features,
1732 1733 1734 1735 1736 1737 1738
};


/***********************
 * rt netlink interface
 ***********************/

1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777
static void team_setup_by_port(struct net_device *dev,
			       struct net_device *port_dev)
{
	dev->header_ops	= port_dev->header_ops;
	dev->type = port_dev->type;
	dev->hard_header_len = port_dev->hard_header_len;
	dev->addr_len = port_dev->addr_len;
	dev->mtu = port_dev->mtu;
	memcpy(dev->broadcast, port_dev->broadcast, port_dev->addr_len);
	memcpy(dev->dev_addr, port_dev->dev_addr, port_dev->addr_len);
	dev->addr_assign_type &= ~NET_ADDR_RANDOM;
}

static int team_dev_type_check_change(struct net_device *dev,
				      struct net_device *port_dev)
{
	struct team *team = netdev_priv(dev);
	char *portname = port_dev->name;
	int err;

	if (dev->type == port_dev->type)
		return 0;
	if (!list_empty(&team->port_list)) {
		netdev_err(dev, "Device %s is of different type\n", portname);
		return -EBUSY;
	}
	err = call_netdevice_notifiers(NETDEV_PRE_TYPE_CHANGE, dev);
	err = notifier_to_errno(err);
	if (err) {
		netdev_err(dev, "Refused to change device type\n");
		return err;
	}
	dev_uc_flush(dev);
	dev_mc_flush(dev);
	team_setup_by_port(dev, port_dev);
	call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev);
	return 0;
}

1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792
static void team_setup(struct net_device *dev)
{
	ether_setup(dev);

	dev->netdev_ops = &team_netdev_ops;
	dev->destructor	= team_destructor;
	dev->tx_queue_len = 0;
	dev->flags |= IFF_MULTICAST;
	dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);

	/*
	 * Indicate we support unicast address filtering. That way core won't
	 * bring us to promisc mode in case a unicast addr is added.
	 * Let this up to underlay drivers.
	 */
1793
	dev->priv_flags |= IFF_UNICAST_FLT | IFF_LIVE_ADDR_CHANGE;
1794 1795 1796

	dev->features |= NETIF_F_LLTX;
	dev->features |= NETIF_F_GRO;
J
Jiri Pirko 已提交
1797 1798
	dev->hw_features = TEAM_VLAN_FEATURES |
			   NETIF_F_HW_VLAN_TX |
1799 1800 1801
			   NETIF_F_HW_VLAN_RX |
			   NETIF_F_HW_VLAN_FILTER;

J
Jiri Pirko 已提交
1802
	dev->hw_features &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_HW_CSUM);
1803 1804 1805 1806 1807 1808 1809 1810 1811
	dev->features |= dev->hw_features;
}

static int team_newlink(struct net *src_net, struct net_device *dev,
			struct nlattr *tb[], struct nlattr *data[])
{
	int err;

	if (tb[IFLA_ADDRESS] == NULL)
1812
		eth_hw_addr_random(dev);
1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831

	err = register_netdevice(dev);
	if (err)
		return err;

	return 0;
}

static int team_validate(struct nlattr *tb[], struct nlattr *data[])
{
	if (tb[IFLA_ADDRESS]) {
		if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
			return -EINVAL;
		if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
			return -EADDRNOTAVAIL;
	}
	return 0;
}

J
Jiri Pirko 已提交
1832 1833 1834 1835 1836 1837 1838 1839 1840 1841
static unsigned int team_get_num_tx_queues(void)
{
	return TEAM_DEFAULT_NUM_TX_QUEUES;
}

static unsigned int team_get_num_rx_queues(void)
{
	return TEAM_DEFAULT_NUM_RX_QUEUES;
}

1842
static struct rtnl_link_ops team_link_ops __read_mostly = {
J
Jiri Pirko 已提交
1843 1844 1845 1846 1847 1848 1849
	.kind			= DRV_NAME,
	.priv_size		= sizeof(struct team),
	.setup			= team_setup,
	.newlink		= team_newlink,
	.validate		= team_validate,
	.get_num_tx_queues	= team_get_num_tx_queues,
	.get_num_rx_queues	= team_get_num_rx_queues,
1850 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
};


/***********************************
 * Generic netlink custom interface
 ***********************************/

static struct genl_family team_nl_family = {
	.id		= GENL_ID_GENERATE,
	.name		= TEAM_GENL_NAME,
	.version	= TEAM_GENL_VERSION,
	.maxattr	= TEAM_ATTR_MAX,
	.netnsok	= true,
};

static const struct nla_policy team_nl_policy[TEAM_ATTR_MAX + 1] = {
	[TEAM_ATTR_UNSPEC]			= { .type = NLA_UNSPEC, },
	[TEAM_ATTR_TEAM_IFINDEX]		= { .type = NLA_U32 },
	[TEAM_ATTR_LIST_OPTION]			= { .type = NLA_NESTED },
	[TEAM_ATTR_LIST_PORT]			= { .type = NLA_NESTED },
};

static const struct nla_policy
team_nl_option_policy[TEAM_ATTR_OPTION_MAX + 1] = {
	[TEAM_ATTR_OPTION_UNSPEC]		= { .type = NLA_UNSPEC, },
	[TEAM_ATTR_OPTION_NAME] = {
		.type = NLA_STRING,
		.len = TEAM_STRING_MAX_LEN,
	},
	[TEAM_ATTR_OPTION_CHANGED]		= { .type = NLA_FLAG },
	[TEAM_ATTR_OPTION_TYPE]			= { .type = NLA_U8 },
J
Jiri Pirko 已提交
1881
	[TEAM_ATTR_OPTION_DATA]			= { .type = NLA_BINARY },
1882 1883 1884 1885 1886 1887 1888 1889
};

static int team_nl_cmd_noop(struct sk_buff *skb, struct genl_info *info)
{
	struct sk_buff *msg;
	void *hdr;
	int err;

1890
	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
1891 1892 1893
	if (!msg)
		return -ENOMEM;

1894
	hdr = genlmsg_put(msg, info->snd_portid, info->snd_seq,
1895
			  &team_nl_family, 0, TEAM_CMD_NOOP);
W
Wei Yongjun 已提交
1896 1897
	if (!hdr) {
		err = -EMSGSIZE;
1898 1899 1900 1901 1902
		goto err_msg_put;
	}

	genlmsg_end(msg, hdr);

1903
	return genlmsg_unicast(genl_info_net(info), msg, info->snd_portid);
1904 1905 1906 1907 1908 1909 1910 1911 1912

err_msg_put:
	nlmsg_free(msg);

	return err;
}

/*
 * Netlink cmd functions should be locked by following two functions.
1913
 * Since dev gets held here, that ensures dev won't disappear in between.
1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925
 */
static struct team *team_nl_team_get(struct genl_info *info)
{
	struct net *net = genl_info_net(info);
	int ifindex;
	struct net_device *dev;
	struct team *team;

	if (!info->attrs[TEAM_ATTR_TEAM_IFINDEX])
		return NULL;

	ifindex = nla_get_u32(info->attrs[TEAM_ATTR_TEAM_IFINDEX]);
1926
	dev = dev_get_by_index(net, ifindex);
1927
	if (!dev || dev->netdev_ops != &team_netdev_ops) {
1928 1929
		if (dev)
			dev_put(dev);
1930 1931 1932 1933
		return NULL;
	}

	team = netdev_priv(dev);
1934
	mutex_lock(&team->lock);
1935 1936 1937 1938 1939
	return team;
}

static void team_nl_team_put(struct team *team)
{
1940
	mutex_unlock(&team->lock);
1941
	dev_put(team->dev);
1942 1943 1944 1945 1946 1947 1948 1949 1950 1951
}

static int team_nl_send_generic(struct genl_info *info, struct team *team,
				int (*fill_func)(struct sk_buff *skb,
						 struct genl_info *info,
						 int flags, struct team *team))
{
	struct sk_buff *skb;
	int err;

1952
	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
1953 1954 1955 1956 1957 1958 1959
	if (!skb)
		return -ENOMEM;

	err = fill_func(skb, info, NLM_F_ACK, team);
	if (err < 0)
		goto err_fill;

1960
	err = genlmsg_unicast(genl_info_net(info), skb, info->snd_portid);
1961 1962 1963 1964 1965 1966 1967
	return err;

err_fill:
	nlmsg_free(skb);
	return err;
}

1968
typedef int team_nl_send_func_t(struct sk_buff *skb,
1969
				struct team *team, u32 portid);
1970

1971
static int team_nl_send_unicast(struct sk_buff *skb, struct team *team, u32 portid)
1972
{
1973
	return genlmsg_unicast(dev_net(team->dev), skb, portid);
1974 1975
}

1976 1977 1978 1979 1980
static int team_nl_fill_one_option_get(struct sk_buff *skb, struct team *team,
				       struct team_option_inst *opt_inst)
{
	struct nlattr *option_item;
	struct team_option *option = opt_inst->option;
1981
	struct team_option_inst_info *opt_inst_info = &opt_inst->info;
1982 1983 1984
	struct team_gsetter_ctx ctx;
	int err;

1985 1986 1987 1988 1989
	ctx.info = opt_inst_info;
	err = team_option_get(team, opt_inst, &ctx);
	if (err)
		return err;

1990 1991
	option_item = nla_nest_start(skb, TEAM_ATTR_ITEM_OPTION);
	if (!option_item)
1992
		return -EMSGSIZE;
1993

1994 1995
	if (nla_put_string(skb, TEAM_ATTR_OPTION_NAME, option->name))
		goto nest_cancel;
1996 1997 1998
	if (opt_inst_info->port &&
	    nla_put_u32(skb, TEAM_ATTR_OPTION_PORT_IFINDEX,
			opt_inst_info->port->dev->ifindex))
1999
		goto nest_cancel;
2000 2001 2002
	if (opt_inst->option->array_size &&
	    nla_put_u32(skb, TEAM_ATTR_OPTION_ARRAY_INDEX,
			opt_inst_info->array_index))
2003
		goto nest_cancel;
2004 2005 2006 2007

	switch (option->type) {
	case TEAM_OPTION_TYPE_U32:
		if (nla_put_u8(skb, TEAM_ATTR_OPTION_TYPE, NLA_U32))
2008
			goto nest_cancel;
2009
		if (nla_put_u32(skb, TEAM_ATTR_OPTION_DATA, ctx.data.u32_val))
2010
			goto nest_cancel;
2011 2012 2013
		break;
	case TEAM_OPTION_TYPE_STRING:
		if (nla_put_u8(skb, TEAM_ATTR_OPTION_TYPE, NLA_STRING))
2014
			goto nest_cancel;
2015 2016
		if (nla_put_string(skb, TEAM_ATTR_OPTION_DATA,
				   ctx.data.str_val))
2017
			goto nest_cancel;
2018 2019 2020
		break;
	case TEAM_OPTION_TYPE_BINARY:
		if (nla_put_u8(skb, TEAM_ATTR_OPTION_TYPE, NLA_BINARY))
2021
			goto nest_cancel;
2022 2023
		if (nla_put(skb, TEAM_ATTR_OPTION_DATA, ctx.data.bin_val.len,
			    ctx.data.bin_val.ptr))
2024
			goto nest_cancel;
2025 2026 2027
		break;
	case TEAM_OPTION_TYPE_BOOL:
		if (nla_put_u8(skb, TEAM_ATTR_OPTION_TYPE, NLA_FLAG))
2028
			goto nest_cancel;
2029 2030
		if (ctx.data.bool_val &&
		    nla_put_flag(skb, TEAM_ATTR_OPTION_DATA))
2031
			goto nest_cancel;
2032
		break;
2033 2034 2035 2036 2037 2038
	case TEAM_OPTION_TYPE_S32:
		if (nla_put_u8(skb, TEAM_ATTR_OPTION_TYPE, NLA_S32))
			goto nest_cancel;
		if (nla_put_s32(skb, TEAM_ATTR_OPTION_DATA, ctx.data.s32_val))
			goto nest_cancel;
		break;
2039 2040 2041
	default:
		BUG();
	}
2042 2043 2044 2045 2046 2047 2048
	if (opt_inst->removed && nla_put_flag(skb, TEAM_ATTR_OPTION_REMOVED))
		goto nest_cancel;
	if (opt_inst->changed) {
		if (nla_put_flag(skb, TEAM_ATTR_OPTION_CHANGED))
			goto nest_cancel;
		opt_inst->changed = false;
	}
2049 2050 2051
	nla_nest_end(skb, option_item);
	return 0;

2052 2053 2054 2055 2056 2057
nest_cancel:
	nla_nest_cancel(skb, option_item);
	return -EMSGSIZE;
}

static int __send_and_alloc_skb(struct sk_buff **pskb,
2058
				struct team *team, u32 portid,
2059 2060 2061 2062 2063
				team_nl_send_func_t *send_func)
{
	int err;

	if (*pskb) {
2064
		err = send_func(*pskb, team, portid);
2065 2066 2067
		if (err)
			return err;
	}
2068
	*pskb = genlmsg_new(GENLMSG_DEFAULT_SIZE, GFP_KERNEL);
2069 2070 2071
	if (!*pskb)
		return -ENOMEM;
	return 0;
2072 2073
}

2074
static int team_nl_send_options_get(struct team *team, u32 portid, u32 seq,
2075
				    int flags, team_nl_send_func_t *send_func,
2076
				    struct list_head *sel_opt_inst_list)
2077 2078
{
	struct nlattr *option_list;
2079
	struct nlmsghdr *nlh;
2080
	void *hdr;
2081 2082
	struct team_option_inst *opt_inst;
	int err;
2083 2084 2085
	struct sk_buff *skb = NULL;
	bool incomplete;
	int i;
2086

2087 2088 2089 2090
	opt_inst = list_first_entry(sel_opt_inst_list,
				    struct team_option_inst, tmp_list);

start_again:
2091
	err = __send_and_alloc_skb(&skb, team, portid, send_func);
2092 2093 2094
	if (err)
		return err;

2095
	hdr = genlmsg_put(skb, portid, seq, &team_nl_family, flags | NLM_F_MULTI,
2096
			  TEAM_CMD_OPTIONS_GET);
W
Wei Yongjun 已提交
2097 2098
	if (!hdr)
		return -EMSGSIZE;
2099

D
David S. Miller 已提交
2100 2101
	if (nla_put_u32(skb, TEAM_ATTR_TEAM_IFINDEX, team->dev->ifindex))
		goto nla_put_failure;
2102 2103
	option_list = nla_nest_start(skb, TEAM_ATTR_LIST_OPTION);
	if (!option_list)
2104
		goto nla_put_failure;
2105

2106 2107 2108
	i = 0;
	incomplete = false;
	list_for_each_entry_from(opt_inst, sel_opt_inst_list, tmp_list) {
2109
		err = team_nl_fill_one_option_get(skb, team, opt_inst);
2110 2111 2112 2113 2114 2115 2116
		if (err) {
			if (err == -EMSGSIZE) {
				if (!i)
					goto errout;
				incomplete = true;
				break;
			}
2117
			goto errout;
2118 2119
		}
		i++;
2120 2121 2122
	}

	nla_nest_end(skb, option_list);
2123 2124 2125 2126 2127
	genlmsg_end(skb, hdr);
	if (incomplete)
		goto start_again;

send_done:
2128
	nlh = nlmsg_put(skb, portid, seq, NLMSG_DONE, 0, flags | NLM_F_MULTI);
2129
	if (!nlh) {
2130
		err = __send_and_alloc_skb(&skb, team, portid, send_func);
2131 2132 2133 2134 2135
		if (err)
			goto errout;
		goto send_done;
	}

2136
	return send_func(skb, team, portid);
2137 2138

nla_put_failure:
2139 2140
	err = -EMSGSIZE;
errout:
2141
	genlmsg_cancel(skb, hdr);
2142
	nlmsg_free(skb);
2143
	return err;
2144 2145 2146 2147 2148
}

static int team_nl_cmd_options_get(struct sk_buff *skb, struct genl_info *info)
{
	struct team *team;
2149
	struct team_option_inst *opt_inst;
2150
	int err;
2151
	LIST_HEAD(sel_opt_inst_list);
2152 2153 2154 2155 2156

	team = team_nl_team_get(info);
	if (!team)
		return -EINVAL;

2157 2158
	list_for_each_entry(opt_inst, &team->option_inst_list, list)
		list_add_tail(&opt_inst->tmp_list, &sel_opt_inst_list);
2159
	err = team_nl_send_options_get(team, info->snd_portid, info->snd_seq,
2160 2161
				       NLM_F_ACK, team_nl_send_unicast,
				       &sel_opt_inst_list);
2162 2163 2164 2165 2166 2167

	team_nl_team_put(team);

	return err;
}

2168 2169 2170
static int team_nl_send_event_options_get(struct team *team,
					  struct list_head *sel_opt_inst_list);

2171 2172 2173 2174 2175 2176
static int team_nl_cmd_options_set(struct sk_buff *skb, struct genl_info *info)
{
	struct team *team;
	int err = 0;
	int i;
	struct nlattr *nl_option;
2177
	LIST_HEAD(opt_inst_list);
2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189

	team = team_nl_team_get(info);
	if (!team)
		return -EINVAL;

	err = -EINVAL;
	if (!info->attrs[TEAM_ATTR_LIST_OPTION]) {
		err = -EINVAL;
		goto team_put;
	}

	nla_for_each_nested(nl_option, info->attrs[TEAM_ATTR_LIST_OPTION], i) {
2190
		struct nlattr *opt_attrs[TEAM_ATTR_OPTION_MAX + 1];
J
Jiri Pirko 已提交
2191
		struct nlattr *attr;
J
Jiri Pirko 已提交
2192
		struct nlattr *attr_data;
2193
		enum team_option_type opt_type;
2194
		int opt_port_ifindex = 0; /* != 0 for per-port options */
J
Jiri Pirko 已提交
2195 2196
		u32 opt_array_index = 0;
		bool opt_is_array = false;
2197
		struct team_option_inst *opt_inst;
2198 2199 2200 2201 2202 2203 2204
		char *opt_name;
		bool opt_found = false;

		if (nla_type(nl_option) != TEAM_ATTR_ITEM_OPTION) {
			err = -EINVAL;
			goto team_put;
		}
2205
		err = nla_parse_nested(opt_attrs, TEAM_ATTR_OPTION_MAX,
2206 2207 2208
				       nl_option, team_nl_option_policy);
		if (err)
			goto team_put;
2209
		if (!opt_attrs[TEAM_ATTR_OPTION_NAME] ||
J
Jiri Pirko 已提交
2210
		    !opt_attrs[TEAM_ATTR_OPTION_TYPE]) {
2211 2212 2213
			err = -EINVAL;
			goto team_put;
		}
2214
		switch (nla_get_u8(opt_attrs[TEAM_ATTR_OPTION_TYPE])) {
2215 2216 2217 2218 2219 2220
		case NLA_U32:
			opt_type = TEAM_OPTION_TYPE_U32;
			break;
		case NLA_STRING:
			opt_type = TEAM_OPTION_TYPE_STRING;
			break;
J
Jiri Pirko 已提交
2221 2222 2223
		case NLA_BINARY:
			opt_type = TEAM_OPTION_TYPE_BINARY;
			break;
J
Jiri Pirko 已提交
2224 2225 2226
		case NLA_FLAG:
			opt_type = TEAM_OPTION_TYPE_BOOL;
			break;
2227 2228 2229
		case NLA_S32:
			opt_type = TEAM_OPTION_TYPE_S32;
			break;
2230 2231 2232 2233
		default:
			goto team_put;
		}

J
Jiri Pirko 已提交
2234 2235 2236 2237 2238 2239
		attr_data = opt_attrs[TEAM_ATTR_OPTION_DATA];
		if (opt_type != TEAM_OPTION_TYPE_BOOL && !attr_data) {
			err = -EINVAL;
			goto team_put;
		}

2240
		opt_name = nla_data(opt_attrs[TEAM_ATTR_OPTION_NAME]);
J
Jiri Pirko 已提交
2241 2242 2243 2244 2245 2246 2247 2248 2249
		attr = opt_attrs[TEAM_ATTR_OPTION_PORT_IFINDEX];
		if (attr)
			opt_port_ifindex = nla_get_u32(attr);

		attr = opt_attrs[TEAM_ATTR_OPTION_ARRAY_INDEX];
		if (attr) {
			opt_is_array = true;
			opt_array_index = nla_get_u32(attr);
		}
2250 2251 2252 2253

		list_for_each_entry(opt_inst, &team->option_inst_list, list) {
			struct team_option *option = opt_inst->option;
			struct team_gsetter_ctx ctx;
2254
			struct team_option_inst_info *opt_inst_info;
2255
			int tmp_ifindex;
2256

2257 2258 2259
			opt_inst_info = &opt_inst->info;
			tmp_ifindex = opt_inst_info->port ?
				      opt_inst_info->port->dev->ifindex : 0;
2260
			if (option->type != opt_type ||
2261
			    strcmp(option->name, opt_name) ||
J
Jiri Pirko 已提交
2262 2263
			    tmp_ifindex != opt_port_ifindex ||
			    (option->array_size && !opt_is_array) ||
2264
			    opt_inst_info->array_index != opt_array_index)
2265 2266
				continue;
			opt_found = true;
2267
			ctx.info = opt_inst_info;
2268 2269
			switch (opt_type) {
			case TEAM_OPTION_TYPE_U32:
J
Jiri Pirko 已提交
2270
				ctx.data.u32_val = nla_get_u32(attr_data);
2271 2272
				break;
			case TEAM_OPTION_TYPE_STRING:
J
Jiri Pirko 已提交
2273
				if (nla_len(attr_data) > TEAM_STRING_MAX_LEN) {
J
Jiri Pirko 已提交
2274 2275 2276
					err = -EINVAL;
					goto team_put;
				}
J
Jiri Pirko 已提交
2277
				ctx.data.str_val = nla_data(attr_data);
2278
				break;
J
Jiri Pirko 已提交
2279
			case TEAM_OPTION_TYPE_BINARY:
J
Jiri Pirko 已提交
2280 2281 2282 2283 2284
				ctx.data.bin_val.len = nla_len(attr_data);
				ctx.data.bin_val.ptr = nla_data(attr_data);
				break;
			case TEAM_OPTION_TYPE_BOOL:
				ctx.data.bool_val = attr_data ? true : false;
J
Jiri Pirko 已提交
2285
				break;
2286 2287 2288
			case TEAM_OPTION_TYPE_S32:
				ctx.data.s32_val = nla_get_s32(attr_data);
				break;
2289 2290 2291
			default:
				BUG();
			}
2292
			err = team_option_set(team, opt_inst, &ctx);
2293 2294
			if (err)
				goto team_put;
2295 2296
			opt_inst->changed = true;
			list_add(&opt_inst->tmp_list, &opt_inst_list);
2297 2298 2299 2300 2301 2302 2303
		}
		if (!opt_found) {
			err = -ENOENT;
			goto team_put;
		}
	}

2304 2305
	err = team_nl_send_event_options_get(team, &opt_inst_list);

2306 2307 2308 2309 2310 2311
team_put:
	team_nl_team_put(team);

	return err;
}

2312
static int team_nl_fill_port_list_get(struct sk_buff *skb,
2313
				      u32 portid, u32 seq, int flags,
2314 2315
				      struct team *team,
				      bool fillall)
2316 2317 2318 2319 2320
{
	struct nlattr *port_list;
	void *hdr;
	struct team_port *port;

2321
	hdr = genlmsg_put(skb, portid, seq, &team_nl_family, flags,
2322
			  TEAM_CMD_PORT_LIST_GET);
W
Wei Yongjun 已提交
2323 2324
	if (!hdr)
		return -EMSGSIZE;
2325

D
David S. Miller 已提交
2326 2327
	if (nla_put_u32(skb, TEAM_ATTR_TEAM_IFINDEX, team->dev->ifindex))
		goto nla_put_failure;
2328 2329
	port_list = nla_nest_start(skb, TEAM_ATTR_LIST_PORT);
	if (!port_list)
2330
		goto nla_put_failure;
2331 2332 2333 2334

	list_for_each_entry(port, &team->port_list, list) {
		struct nlattr *port_item;

2335 2336 2337
		/* Include only changed ports if fill all mode is not on */
		if (!fillall && !port->changed)
			continue;
2338 2339 2340
		port_item = nla_nest_start(skb, TEAM_ATTR_ITEM_PORT);
		if (!port_item)
			goto nla_put_failure;
D
David S. Miller 已提交
2341 2342
		if (nla_put_u32(skb, TEAM_ATTR_PORT_IFINDEX, port->dev->ifindex))
			goto nla_put_failure;
2343
		if (port->changed) {
D
David S. Miller 已提交
2344 2345
			if (nla_put_flag(skb, TEAM_ATTR_PORT_CHANGED))
				goto nla_put_failure;
2346 2347
			port->changed = false;
		}
D
David S. Miller 已提交
2348 2349
		if ((port->removed &&
		     nla_put_flag(skb, TEAM_ATTR_PORT_REMOVED)) ||
2350
		    (port->state.linkup &&
D
David S. Miller 已提交
2351
		     nla_put_flag(skb, TEAM_ATTR_PORT_LINKUP)) ||
2352 2353
		    nla_put_u32(skb, TEAM_ATTR_PORT_SPEED, port->state.speed) ||
		    nla_put_u8(skb, TEAM_ATTR_PORT_DUPLEX, port->state.duplex))
D
David S. Miller 已提交
2354
			goto nla_put_failure;
2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365
		nla_nest_end(skb, port_item);
	}

	nla_nest_end(skb, port_list);
	return genlmsg_end(skb, hdr);

nla_put_failure:
	genlmsg_cancel(skb, hdr);
	return -EMSGSIZE;
}

2366 2367 2368
static int team_nl_fill_port_list_get_all(struct sk_buff *skb,
					  struct genl_info *info, int flags,
					  struct team *team)
2369
{
2370
	return team_nl_fill_port_list_get(skb, info->snd_portid,
2371 2372
					  info->snd_seq, NLM_F_ACK,
					  team, true);
2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384
}

static int team_nl_cmd_port_list_get(struct sk_buff *skb,
				     struct genl_info *info)
{
	struct team *team;
	int err;

	team = team_nl_team_get(info);
	if (!team)
		return -EINVAL;

2385
	err = team_nl_send_generic(info, team, team_nl_fill_port_list_get_all);
2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421

	team_nl_team_put(team);

	return err;
}

static struct genl_ops team_nl_ops[] = {
	{
		.cmd = TEAM_CMD_NOOP,
		.doit = team_nl_cmd_noop,
		.policy = team_nl_policy,
	},
	{
		.cmd = TEAM_CMD_OPTIONS_SET,
		.doit = team_nl_cmd_options_set,
		.policy = team_nl_policy,
		.flags = GENL_ADMIN_PERM,
	},
	{
		.cmd = TEAM_CMD_OPTIONS_GET,
		.doit = team_nl_cmd_options_get,
		.policy = team_nl_policy,
		.flags = GENL_ADMIN_PERM,
	},
	{
		.cmd = TEAM_CMD_PORT_LIST_GET,
		.doit = team_nl_cmd_port_list_get,
		.policy = team_nl_policy,
		.flags = GENL_ADMIN_PERM,
	},
};

static struct genl_multicast_group team_change_event_mcgrp = {
	.name = TEAM_GENL_CHANGE_EVENT_MC_GRP_NAME,
};

2422
static int team_nl_send_multicast(struct sk_buff *skb,
2423
				  struct team *team, u32 portid)
2424 2425 2426 2427 2428
{
	return genlmsg_multicast_netns(dev_net(team->dev), skb, 0,
				       team_change_event_mcgrp.id, GFP_KERNEL);
}

2429 2430
static int team_nl_send_event_options_get(struct team *team,
					  struct list_head *sel_opt_inst_list)
2431
{
2432 2433
	return team_nl_send_options_get(team, 0, 0, 0, team_nl_send_multicast,
					sel_opt_inst_list);
2434 2435
}

2436
static int team_nl_send_event_port_list_get(struct team *team)
2437 2438 2439
{
	struct sk_buff *skb;
	int err;
2440
	struct net *net = dev_net(team->dev);
2441

2442
	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
2443 2444 2445
	if (!skb)
		return -ENOMEM;

2446
	err = team_nl_fill_port_list_get(skb, 0, 0, 0, team, false);
2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489
	if (err < 0)
		goto err_fill;

	err = genlmsg_multicast_netns(net, skb, 0, team_change_event_mcgrp.id,
				      GFP_KERNEL);
	return err;

err_fill:
	nlmsg_free(skb);
	return err;
}

static int team_nl_init(void)
{
	int err;

	err = genl_register_family_with_ops(&team_nl_family, team_nl_ops,
					    ARRAY_SIZE(team_nl_ops));
	if (err)
		return err;

	err = genl_register_mc_group(&team_nl_family, &team_change_event_mcgrp);
	if (err)
		goto err_change_event_grp_reg;

	return 0;

err_change_event_grp_reg:
	genl_unregister_family(&team_nl_family);

	return err;
}

static void team_nl_fini(void)
{
	genl_unregister_family(&team_nl_family);
}


/******************
 * Change checkers
 ******************/

2490
static void __team_options_change_check(struct team *team)
2491 2492
{
	int err;
2493 2494
	struct team_option_inst *opt_inst;
	LIST_HEAD(sel_opt_inst_list);
2495

2496 2497 2498 2499 2500
	list_for_each_entry(opt_inst, &team->option_inst_list, list) {
		if (opt_inst->changed)
			list_add_tail(&opt_inst->tmp_list, &sel_opt_inst_list);
	}
	err = team_nl_send_event_options_get(team, &sel_opt_inst_list);
2501
	if (err && err != -ESRCH)
2502 2503
		netdev_warn(team->dev, "Failed to send options change via netlink (err %d)\n",
			    err);
2504 2505 2506
}

/* rtnl lock is held */
J
Jiri Pirko 已提交
2507 2508

static void __team_port_change_send(struct team_port *port, bool linkup)
2509 2510 2511
{
	int err;

2512
	port->changed = true;
2513 2514
	port->state.linkup = linkup;
	team_refresh_port_linkup(port);
2515 2516 2517 2518 2519
	if (linkup) {
		struct ethtool_cmd ecmd;

		err = __ethtool_get_settings(port->dev, &ecmd);
		if (!err) {
2520 2521
			port->state.speed = ethtool_cmd_speed(&ecmd);
			port->state.duplex = ecmd.duplex;
2522 2523 2524
			goto send_event;
		}
	}
2525 2526
	port->state.speed = 0;
	port->state.duplex = 0;
2527 2528

send_event:
2529
	err = team_nl_send_event_port_list_get(port->team);
2530 2531 2532
	if (err && err != -ESRCH)
		netdev_warn(port->team->dev, "Failed to send port change of device %s via netlink (err %d)\n",
			    port->dev->name, err);
2533 2534 2535

}

J
Jiri Pirko 已提交
2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552
static void __team_port_change_check(struct team_port *port, bool linkup)
{
	if (port->state.linkup != linkup)
		__team_port_change_send(port, linkup);
}

static void __team_port_change_port_added(struct team_port *port, bool linkup)
{
	__team_port_change_send(port, linkup);
}

static void __team_port_change_port_removed(struct team_port *port)
{
	port->removed = true;
	__team_port_change_send(port, false);
}

2553 2554 2555 2556
static void team_port_change_check(struct team_port *port, bool linkup)
{
	struct team *team = port->team;

2557
	mutex_lock(&team->lock);
2558
	__team_port_change_check(port, linkup);
2559
	mutex_unlock(&team->lock);
2560 2561
}

J
Jiri Pirko 已提交
2562

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 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 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651
/************************************
 * Net device notifier event handler
 ************************************/

static int team_device_event(struct notifier_block *unused,
			     unsigned long event, void *ptr)
{
	struct net_device *dev = (struct net_device *) ptr;
	struct team_port *port;

	port = team_port_get_rtnl(dev);
	if (!port)
		return NOTIFY_DONE;

	switch (event) {
	case NETDEV_UP:
		if (netif_carrier_ok(dev))
			team_port_change_check(port, true);
	case NETDEV_DOWN:
		team_port_change_check(port, false);
	case NETDEV_CHANGE:
		if (netif_running(port->dev))
			team_port_change_check(port,
					       !!netif_carrier_ok(port->dev));
		break;
	case NETDEV_UNREGISTER:
		team_del_slave(port->team->dev, dev);
		break;
	case NETDEV_FEAT_CHANGE:
		team_compute_features(port->team);
		break;
	case NETDEV_CHANGEMTU:
		/* Forbid to change mtu of underlaying device */
		return NOTIFY_BAD;
	case NETDEV_PRE_TYPE_CHANGE:
		/* Forbid to change type of underlaying device */
		return NOTIFY_BAD;
	}
	return NOTIFY_DONE;
}

static struct notifier_block team_notifier_block __read_mostly = {
	.notifier_call = team_device_event,
};


/***********************
 * Module init and exit
 ***********************/

static int __init team_module_init(void)
{
	int err;

	register_netdevice_notifier(&team_notifier_block);

	err = rtnl_link_register(&team_link_ops);
	if (err)
		goto err_rtnl_reg;

	err = team_nl_init();
	if (err)
		goto err_nl_init;

	return 0;

err_nl_init:
	rtnl_link_unregister(&team_link_ops);

err_rtnl_reg:
	unregister_netdevice_notifier(&team_notifier_block);

	return err;
}

static void __exit team_module_exit(void)
{
	team_nl_fini();
	rtnl_link_unregister(&team_link_ops);
	unregister_netdevice_notifier(&team_notifier_block);
}

module_init(team_module_init);
module_exit(team_module_exit);

MODULE_LICENSE("GPL v2");
MODULE_AUTHOR("Jiri Pirko <jpirko@redhat.com>");
MODULE_DESCRIPTION("Ethernet team device driver");
MODULE_ALIAS_RTNL_LINK(DRV_NAME);