en_tc.c 15.1 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
/*
 * Copyright (c) 2016, 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.
 */

33
34
35
#include <net/flow_dissector.h>
#include <net/pkt_cls.h>
#include <net/tc_act/tc_gact.h>
36
#include <net/tc_act/tc_skbedit.h>
37
38
39
#include <linux/mlx5/fs.h>
#include <linux/mlx5/device.h>
#include <linux/rhashtable.h>
40
41
#include <net/switchdev.h>
#include <net/tc_act/tc_mirred.h>
42
43
#include "en.h"
#include "en_tc.h"
44
#include "eswitch.h"
45
46
47
48
49
50
51

struct mlx5e_tc_flow {
	struct rhash_head	node;
	u64			cookie;
	struct mlx5_flow_rule	*rule;
};

52
53
#define MLX5E_TC_TABLE_NUM_ENTRIES 1024
#define MLX5E_TC_TABLE_NUM_GROUPS 4
54

55
56
57
static struct mlx5_flow_rule *mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv,
						    struct mlx5_flow_spec *spec,
						    u32 action, u32 flow_tag)
58
{
59
60
61
	struct mlx5_core_dev *dev = priv->mdev;
	struct mlx5_flow_destination dest = { 0 };
	struct mlx5_fc *counter = NULL;
62
63
64
	struct mlx5_flow_rule *rule;
	bool table_created = false;

65
66
67
	if (action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
		dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
		dest.ft = priv->fs.vlan.ft.t;
68
	} else if (action & MLX5_FLOW_CONTEXT_ACTION_COUNT) {
69
70
71
72
73
74
75
76
		counter = mlx5_fc_create(dev, true);
		if (IS_ERR(counter))
			return ERR_CAST(counter);

		dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
		dest.counter = counter;
	}

77
78
79
80
81
82
	if (IS_ERR_OR_NULL(priv->fs.tc.t)) {
		priv->fs.tc.t =
			mlx5_create_auto_grouped_flow_table(priv->fs.ns,
							    MLX5E_TC_PRIO,
							    MLX5E_TC_TABLE_NUM_ENTRIES,
							    MLX5E_TC_TABLE_NUM_GROUPS,
83
							    0);
84
		if (IS_ERR(priv->fs.tc.t)) {
85
86
			netdev_err(priv->netdev,
				   "Failed to create tc offload table\n");
87
88
			rule = ERR_CAST(priv->fs.tc.t);
			goto err_create_ft;
89
90
91
92
93
		}

		table_created = true;
	}

94
95
	spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
	rule = mlx5_add_flow_rule(priv->fs.tc.t, spec,
96
				  action, flow_tag,
97
98
99
100
101
102
				  &dest);

	if (IS_ERR(rule))
		goto err_add_rule;

	return rule;
103

104
105
err_add_rule:
	if (table_created) {
106
107
		mlx5_destroy_flow_table(priv->fs.tc.t);
		priv->fs.tc.t = NULL;
108
	}
109
110
err_create_ft:
	mlx5_fc_destroy(dev, counter);
111
112
113
114

	return rule;
}

115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
static struct mlx5_flow_rule *mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv,
						    struct mlx5_flow_spec *spec,
						    u32 action, u32 dst_vport)
{
	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
	struct mlx5_eswitch_rep *rep = priv->ppriv;
	u32 src_vport;

	if (rep->vport) /* set source vport for the flow */
		src_vport = rep->vport;
	else
		src_vport = FDB_UPLINK_VPORT;

	return mlx5_eswitch_add_offloaded_rule(esw, spec, action, src_vport, dst_vport);
}

131
132
133
static void mlx5e_tc_del_flow(struct mlx5e_priv *priv,
			      struct mlx5_flow_rule *rule)
{
134
135
136
137
	struct mlx5_fc *counter = NULL;

	counter = mlx5_flow_rule_counter(rule);

138
139
	mlx5_del_flow_rule(rule);

140
141
	mlx5_fc_destroy(priv->mdev, counter);

142
	if (!mlx5e_tc_num_filters(priv) && (priv->fs.tc.t)) {
143
144
		mlx5_destroy_flow_table(priv->fs.tc.t);
		priv->fs.tc.t = NULL;
145
146
147
	}
}

148
static int parse_cls_flower(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec,
149
150
			    struct tc_cls_flower_offload *f)
{
151
152
153
154
	void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
				       outer_headers);
	void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
				       outer_headers);
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
	u16 addr_type = 0;
	u8 ip_proto = 0;

	if (f->dissector->used_keys &
	    ~(BIT(FLOW_DISSECTOR_KEY_CONTROL) |
	      BIT(FLOW_DISSECTOR_KEY_BASIC) |
	      BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS) |
	      BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
	      BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
	      BIT(FLOW_DISSECTOR_KEY_PORTS))) {
		netdev_warn(priv->netdev, "Unsupported key used: 0x%x\n",
			    f->dissector->used_keys);
		return -EOPNOTSUPP;
	}

	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_CONTROL)) {
		struct flow_dissector_key_control *key =
			skb_flow_dissector_target(f->dissector,
173
						  FLOW_DISSECTOR_KEY_CONTROL,
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
						  f->key);
		addr_type = key->addr_type;
	}

	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_BASIC)) {
		struct flow_dissector_key_basic *key =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_BASIC,
						  f->key);
		struct flow_dissector_key_basic *mask =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_BASIC,
						  f->mask);
		ip_proto = key->ip_proto;

		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype,
			 ntohs(mask->n_proto));
		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype,
			 ntohs(key->n_proto));

		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_protocol,
			 mask->ip_proto);
		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_protocol,
			 key->ip_proto);
	}

	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
		struct flow_dissector_key_eth_addrs *key =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_ETH_ADDRS,
						  f->key);
		struct flow_dissector_key_eth_addrs *mask =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_ETH_ADDRS,
						  f->mask);

		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
					     dmac_47_16),
				mask->dst);
		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
					     dmac_47_16),
				key->dst);

		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
					     smac_47_16),
				mask->src);
		ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
					     smac_47_16),
				key->src);
	}

	if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
		struct flow_dissector_key_ipv4_addrs *key =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_IPV4_ADDRS,
						  f->key);
		struct flow_dissector_key_ipv4_addrs *mask =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_IPV4_ADDRS,
						  f->mask);

		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
				    src_ipv4_src_ipv6.ipv4_layout.ipv4),
		       &mask->src, sizeof(mask->src));
		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
				    src_ipv4_src_ipv6.ipv4_layout.ipv4),
		       &key->src, sizeof(key->src));
		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
				    dst_ipv4_dst_ipv6.ipv4_layout.ipv4),
		       &mask->dst, sizeof(mask->dst));
		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
				    dst_ipv4_dst_ipv6.ipv4_layout.ipv4),
		       &key->dst, sizeof(key->dst));
	}

	if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
		struct flow_dissector_key_ipv6_addrs *key =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_IPV6_ADDRS,
						  f->key);
		struct flow_dissector_key_ipv6_addrs *mask =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_IPV6_ADDRS,
						  f->mask);

		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
				    src_ipv4_src_ipv6.ipv6_layout.ipv6),
		       &mask->src, sizeof(mask->src));
		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
				    src_ipv4_src_ipv6.ipv6_layout.ipv6),
		       &key->src, sizeof(key->src));

		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
				    dst_ipv4_dst_ipv6.ipv6_layout.ipv6),
		       &mask->dst, sizeof(mask->dst));
		memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
				    dst_ipv4_dst_ipv6.ipv6_layout.ipv6),
		       &key->dst, sizeof(key->dst));
	}

	if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_PORTS)) {
		struct flow_dissector_key_ports *key =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_PORTS,
						  f->key);
		struct flow_dissector_key_ports *mask =
			skb_flow_dissector_target(f->dissector,
						  FLOW_DISSECTOR_KEY_PORTS,
						  f->mask);
		switch (ip_proto) {
		case IPPROTO_TCP:
			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
				 tcp_sport, ntohs(mask->src));
			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
				 tcp_sport, ntohs(key->src));

			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
				 tcp_dport, ntohs(mask->dst));
			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
				 tcp_dport, ntohs(key->dst));
			break;

		case IPPROTO_UDP:
			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
				 udp_sport, ntohs(mask->src));
			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
				 udp_sport, ntohs(key->src));

			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
				 udp_dport, ntohs(mask->dst));
			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
				 udp_dport, ntohs(key->dst));
			break;
		default:
			netdev_err(priv->netdev,
				   "Only UDP and TCP transport are supported\n");
			return -EINVAL;
		}
	}

	return 0;
}

317
318
static int parse_tc_nic_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
				u32 *action, u32 *flow_tag)
319
320
{
	const struct tc_action *a;
321
	LIST_HEAD(actions);
322
323
324
325
326
327
328

	if (tc_no_actions(exts))
		return -EINVAL;

	*flow_tag = MLX5_FS_DEFAULT_FLOW_TAG;
	*action = 0;

329
330
	tcf_exts_to_list(exts, &actions);
	list_for_each_entry(a, &actions, list) {
331
332
333
334
335
336
		/* Only support a single action per rule */
		if (*action)
			return -EINVAL;

		if (is_tcf_gact_shot(a)) {
			*action |= MLX5_FLOW_CONTEXT_ACTION_DROP;
337
338
339
			if (MLX5_CAP_FLOWTABLE(priv->mdev,
					       flow_table_properties_nic_receive.flow_counter))
				*action |= MLX5_FLOW_CONTEXT_ACTION_COUNT;
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
			continue;
		}

		if (is_tcf_skbedit_mark(a)) {
			u32 mark = tcf_skbedit_mark(a);

			if (mark & ~MLX5E_TC_FLOW_ID_MASK) {
				netdev_warn(priv->netdev, "Bad flow mark - only 16 bit is supported: 0x%x\n",
					    mark);
				return -EINVAL;
			}

			*flow_tag = mark;
			*action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
			continue;
		}

		return -EINVAL;
	}

	return 0;
}

363
364
365
366
static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
				u32 *action, u32 *dest_vport)
{
	const struct tc_action *a;
367
	LIST_HEAD(actions);
368
369
370
371
372
373

	if (tc_no_actions(exts))
		return -EINVAL;

	*action = 0;

374
375
	tcf_exts_to_list(exts, &actions);
	list_for_each_entry(a, &actions, list) {
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
		/* Only support a single action per rule */
		if (*action)
			return -EINVAL;

		if (is_tcf_gact_shot(a)) {
			*action = MLX5_FLOW_CONTEXT_ACTION_DROP |
				  MLX5_FLOW_CONTEXT_ACTION_COUNT;
			continue;
		}

		if (is_tcf_mirred_redirect(a)) {
			int ifindex = tcf_mirred_ifindex(a);
			struct net_device *out_dev;
			struct mlx5e_priv *out_priv;
			struct mlx5_eswitch_rep *out_rep;

			out_dev = __dev_get_by_index(dev_net(priv->netdev), ifindex);

			if (!switchdev_port_same_parent_id(priv->netdev, out_dev)) {
				pr_err("devices %s %s not on same switch HW, can't offload forwarding\n",
				       priv->netdev->name, out_dev->name);
				return -EINVAL;
			}

			out_priv = netdev_priv(out_dev);
			out_rep  = out_priv->ppriv;
			if (out_rep->vport == 0)
				*dest_vport = FDB_UPLINK_VPORT;
			else
				*dest_vport = out_rep->vport;
			*action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
			continue;
		}

		return -EINVAL;
	}
	return 0;
}

415
416
417
int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol,
			   struct tc_cls_flower_offload *f)
{
418
	struct mlx5e_tc_table *tc = &priv->fs.tc;
419
	int err = 0;
420
	u32 flow_tag, action, dest_vport = 0;
421
	struct mlx5e_tc_flow *flow;
422
	struct mlx5_flow_spec *spec;
423
	struct mlx5_flow_rule *old = NULL;
424
	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
425
426
427
428
429
430
431
432

	flow = rhashtable_lookup_fast(&tc->ht, &f->cookie,
				      tc->ht_params);
	if (flow)
		old = flow->rule;
	else
		flow = kzalloc(sizeof(*flow), GFP_KERNEL);

433
434
	spec = mlx5_vzalloc(sizeof(*spec));
	if (!spec || !flow) {
435
436
437
438
439
440
		err = -ENOMEM;
		goto err_free;
	}

	flow->cookie = f->cookie;

441
	err = parse_cls_flower(priv, spec, f);
442
443
444
	if (err < 0)
		goto err_free;

445
446
447
448
449
450
451
452
453
454
455
	if (esw && esw->mode == SRIOV_OFFLOADS) {
		err = parse_tc_fdb_actions(priv, f->exts, &action, &dest_vport);
		if (err < 0)
			goto err_free;
		flow->rule = mlx5e_tc_add_fdb_flow(priv, spec, action, dest_vport);
	} else {
		err = parse_tc_nic_actions(priv, f->exts, &action, &flow_tag);
		if (err < 0)
			goto err_free;
		flow->rule = mlx5e_tc_add_nic_flow(priv, spec, action, flow_tag);
	}
456
457
458

	if (IS_ERR(flow->rule)) {
		err = PTR_ERR(flow->rule);
459
		goto err_free;
460
461
	}

462
463
464
465
466
	err = rhashtable_insert_fast(&tc->ht, &flow->node,
				     tc->ht_params);
	if (err)
		goto err_del_rule;

467
468
469
470
471
	if (old)
		mlx5e_tc_del_flow(priv, old);

	goto out;

472
473
err_del_rule:
	mlx5_del_flow_rule(flow->rule);
474
475
476
477
478

err_free:
	if (!old)
		kfree(flow);
out:
479
	kvfree(spec);
480
481
482
483
484
485
486
	return err;
}

int mlx5e_delete_flower(struct mlx5e_priv *priv,
			struct tc_cls_flower_offload *f)
{
	struct mlx5e_tc_flow *flow;
487
	struct mlx5e_tc_table *tc = &priv->fs.tc;
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502

	flow = rhashtable_lookup_fast(&tc->ht, &f->cookie,
				      tc->ht_params);
	if (!flow)
		return -EINVAL;

	rhashtable_remove_fast(&tc->ht, &flow->node, tc->ht_params);

	mlx5e_tc_del_flow(priv, flow->rule);

	kfree(flow);

	return 0;
}

503
504
505
506
507
508
509
int mlx5e_stats_flower(struct mlx5e_priv *priv,
		       struct tc_cls_flower_offload *f)
{
	struct mlx5e_tc_table *tc = &priv->fs.tc;
	struct mlx5e_tc_flow *flow;
	struct tc_action *a;
	struct mlx5_fc *counter;
510
	LIST_HEAD(actions);
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
	u64 bytes;
	u64 packets;
	u64 lastuse;

	flow = rhashtable_lookup_fast(&tc->ht, &f->cookie,
				      tc->ht_params);
	if (!flow)
		return -EINVAL;

	counter = mlx5_flow_rule_counter(flow->rule);
	if (!counter)
		return 0;

	mlx5_fc_query_cached(counter, &bytes, &packets, &lastuse);

526
527
	tcf_exts_to_list(f->exts, &actions);
	list_for_each_entry(a, &actions, list)
528
529
530
531
532
		tcf_action_stats_update(a, bytes, packets, lastuse);

	return 0;
}

533
534
535
536
537
538
539
540
541
static const struct rhashtable_params mlx5e_tc_flow_ht_params = {
	.head_offset = offsetof(struct mlx5e_tc_flow, node),
	.key_offset = offsetof(struct mlx5e_tc_flow, cookie),
	.key_len = sizeof(((struct mlx5e_tc_flow *)0)->cookie),
	.automatic_shrinking = true,
};

int mlx5e_tc_init(struct mlx5e_priv *priv)
{
542
	struct mlx5e_tc_table *tc = &priv->fs.tc;
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558

	tc->ht_params = mlx5e_tc_flow_ht_params;
	return rhashtable_init(&tc->ht, &tc->ht_params);
}

static void _mlx5e_tc_del_flow(void *ptr, void *arg)
{
	struct mlx5e_tc_flow *flow = ptr;
	struct mlx5e_priv *priv = arg;

	mlx5e_tc_del_flow(priv, flow->rule);
	kfree(flow);
}

void mlx5e_tc_cleanup(struct mlx5e_priv *priv)
{
559
	struct mlx5e_tc_table *tc = &priv->fs.tc;
560
561
562

	rhashtable_free_and_destroy(&tc->ht, _mlx5e_tc_del_flow, priv);

563
564
565
	if (!IS_ERR_OR_NULL(tc->t)) {
		mlx5_destroy_flow_table(tc->t);
		tc->t = NULL;
566
567
	}
}