eswitch.c 59.7 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
/*
 * Copyright (c) 2015, Mellanox Technologies. All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the
 * OpenIB.org BSD license below:
 *
 *     Redistribution and use in source and binary forms, with or
 *     without modification, are permitted provided that the following
 *     conditions are met:
 *
 *      - Redistributions of source code must retain the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer.
 *
 *      - Redistributions in binary form must reproduce the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer in the documentation and/or other materials
 *        provided with the distribution.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */

#include <linux/etherdevice.h>
#include <linux/mlx5/driver.h>
#include <linux/mlx5/mlx5_ifc.h>
#include <linux/mlx5/vport.h>
37
#include <linux/mlx5/fs.h>
38
39
40
#include "mlx5_core.h"
#include "eswitch.h"

41
42
#define UPLINK_VPORT 0xFFFF

43
44
45
46
47
48
enum {
	MLX5_ACTION_NONE = 0,
	MLX5_ACTION_ADD  = 1,
	MLX5_ACTION_DEL  = 2,
};

49
50
/* E-Switch UC L2 table hash node */
struct esw_uc_addr {
51
52
53
54
55
	struct l2addr_node node;
	u32                table_index;
	u32                vport;
};

56
57
58
/* E-Switch MC FDB table hash node */
struct esw_mc_addr { /* SRIOV only */
	struct l2addr_node     node;
Mark Bloch's avatar
Mark Bloch committed
59
	struct mlx5_flow_handle *uplink_rule; /* Forward to uplink rule */
60
61
62
63
64
65
66
67
	u32                    refcnt;
};

/* Vport UC/MC hash node */
struct vport_addr {
	struct l2addr_node     node;
	u8                     action;
	u32                    vport;
Mark Bloch's avatar
Mark Bloch committed
68
	struct mlx5_flow_handle *flow_rule; /* SRIOV only */
69
70
	/* A flag indicating that mac was added due to mc promiscuous vport */
	bool mc_promisc;
71
72
73
74
75
};

enum {
	UC_ADDR_CHANGE = BIT(0),
	MC_ADDR_CHANGE = BIT(1),
76
	PROMISC_CHANGE = BIT(3),
77
78
};

79
80
/* Vport context events */
#define SRIOV_VPORT_EVENTS (UC_ADDR_CHANGE | \
81
82
			    MC_ADDR_CHANGE | \
			    PROMISC_CHANGE)
83
84

static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport,
85
86
					u32 events_mask)
{
87
88
	int in[MLX5_ST_SZ_DW(modify_nic_vport_context_in)]   = {0};
	int out[MLX5_ST_SZ_DW(modify_nic_vport_context_out)] = {0};
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
	void *nic_vport_ctx;

	MLX5_SET(modify_nic_vport_context_in, in,
		 opcode, MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT);
	MLX5_SET(modify_nic_vport_context_in, in, field_select.change_event, 1);
	MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport);
	if (vport)
		MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1);
	nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in,
				     in, nic_vport_context);

	MLX5_SET(nic_vport_context, nic_vport_ctx, arm_change_event, 1);

	if (events_mask & UC_ADDR_CHANGE)
		MLX5_SET(nic_vport_context, nic_vport_ctx,
			 event_on_uc_address_change, 1);
	if (events_mask & MC_ADDR_CHANGE)
		MLX5_SET(nic_vport_context, nic_vport_ctx,
			 event_on_mc_address_change, 1);
108
109
110
	if (events_mask & PROMISC_CHANGE)
		MLX5_SET(nic_vport_context, nic_vport_ctx,
			 event_on_promisc_change, 1);
111

112
	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
113
114
}

115
116
117
118
/* E-Switch vport context HW commands */
static int modify_esw_vport_context_cmd(struct mlx5_core_dev *dev, u16 vport,
					void *in, int inlen)
{
119
	u32 out[MLX5_ST_SZ_DW(modify_esw_vport_context_out)] = {0};
120

121
122
	MLX5_SET(modify_esw_vport_context_in, in, opcode,
		 MLX5_CMD_OP_MODIFY_ESW_VPORT_CONTEXT);
123
124
125
	MLX5_SET(modify_esw_vport_context_in, in, vport_number, vport);
	if (vport)
		MLX5_SET(modify_esw_vport_context_in, in, other_vport, 1);
126
	return mlx5_cmd_exec(dev, in, inlen, out, sizeof(out));
127
128
129
}

static int modify_esw_vport_cvlan(struct mlx5_core_dev *dev, u32 vport,
130
				  u16 vlan, u8 qos, u8 set_flags)
131
{
132
	u32 in[MLX5_ST_SZ_DW(modify_esw_vport_context_in)] = {0};
133
134
135
136
137

	if (!MLX5_CAP_ESW(dev, vport_cvlan_strip) ||
	    !MLX5_CAP_ESW(dev, vport_cvlan_insert_if_not_exist))
		return -ENOTSUPP;

138
139
140
141
	esw_debug(dev, "Set Vport[%d] VLAN %d qos %d set=%x\n",
		  vport, vlan, qos, set_flags);

	if (set_flags & SET_VLAN_STRIP)
142
143
		MLX5_SET(modify_esw_vport_context_in, in,
			 esw_vport_context.vport_cvlan_strip, 1);
144
145

	if (set_flags & SET_VLAN_INSERT) {
146
147
148
		/* insert only if no vlan in packet */
		MLX5_SET(modify_esw_vport_context_in, in,
			 esw_vport_context.vport_cvlan_insert, 1);
149

150
151
152
153
154
155
156
157
158
159
160
161
162
163
		MLX5_SET(modify_esw_vport_context_in, in,
			 esw_vport_context.cvlan_pcp, qos);
		MLX5_SET(modify_esw_vport_context_in, in,
			 esw_vport_context.cvlan_id, vlan);
	}

	MLX5_SET(modify_esw_vport_context_in, in,
		 field_select.vport_cvlan_strip, 1);
	MLX5_SET(modify_esw_vport_context_in, in,
		 field_select.vport_cvlan_insert, 1);

	return modify_esw_vport_context_cmd(dev, vport, in, sizeof(in));
}

164
165
166
167
/* HW L2 Table (MPFS) management */
static int set_l2_table_entry_cmd(struct mlx5_core_dev *dev, u32 index,
				  u8 *mac, u8 vlan_valid, u16 vlan)
{
168
169
	u32 in[MLX5_ST_SZ_DW(set_l2_table_entry_in)]   = {0};
	u32 out[MLX5_ST_SZ_DW(set_l2_table_entry_out)] = {0};
170
171
172
173
174
175
176
177
178
179
180
	u8 *in_mac_addr;

	MLX5_SET(set_l2_table_entry_in, in, opcode,
		 MLX5_CMD_OP_SET_L2_TABLE_ENTRY);
	MLX5_SET(set_l2_table_entry_in, in, table_index, index);
	MLX5_SET(set_l2_table_entry_in, in, vlan_valid, vlan_valid);
	MLX5_SET(set_l2_table_entry_in, in, vlan, vlan);

	in_mac_addr = MLX5_ADDR_OF(set_l2_table_entry_in, in, mac_address);
	ether_addr_copy(&in_mac_addr[2], mac);

181
	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
182
183
184
185
}

static int del_l2_table_entry_cmd(struct mlx5_core_dev *dev, u32 index)
{
186
187
	u32 in[MLX5_ST_SZ_DW(delete_l2_table_entry_in)]   = {0};
	u32 out[MLX5_ST_SZ_DW(delete_l2_table_entry_out)] = {0};
188
189
190
191

	MLX5_SET(delete_l2_table_entry_in, in, opcode,
		 MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY);
	MLX5_SET(delete_l2_table_entry_in, in, table_index, index);
192
	return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
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
}

static int alloc_l2_table_index(struct mlx5_l2_table *l2_table, u32 *ix)
{
	int err = 0;

	*ix = find_first_zero_bit(l2_table->bitmap, l2_table->size);
	if (*ix >= l2_table->size)
		err = -ENOSPC;
	else
		__set_bit(*ix, l2_table->bitmap);

	return err;
}

static void free_l2_table_index(struct mlx5_l2_table *l2_table, u32 ix)
{
	__clear_bit(ix, l2_table->bitmap);
}

static int set_l2_table_entry(struct mlx5_core_dev *dev, u8 *mac,
			      u8 vlan_valid, u16 vlan,
			      u32 *index)
{
	struct mlx5_l2_table *l2_table = &dev->priv.eswitch->l2_table;
	int err;

	err = alloc_l2_table_index(l2_table, index);
	if (err)
		return err;

	err = set_l2_table_entry_cmd(dev, *index, mac, vlan_valid, vlan);
	if (err)
		free_l2_table_index(l2_table, *index);

	return err;
}

static void del_l2_table_entry(struct mlx5_core_dev *dev, u32 index)
{
	struct mlx5_l2_table *l2_table = &dev->priv.eswitch->l2_table;

	del_l2_table_entry_cmd(dev, index);
	free_l2_table_index(l2_table, index);
}

239
/* E-Switch FDB */
Mark Bloch's avatar
Mark Bloch committed
240
static struct mlx5_flow_handle *
241
__esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule,
242
			 u8 mac_c[ETH_ALEN], u8 mac_v[ETH_ALEN])
243
{
244
245
	int match_header = (is_zero_ether_addr(mac_c) ? 0 :
			    MLX5_MATCH_OUTER_HEADERS);
Mark Bloch's avatar
Mark Bloch committed
246
	struct mlx5_flow_handle *flow_rule = NULL;
247
	struct mlx5_flow_act flow_act = {0};
248
	struct mlx5_flow_destination dest;
249
	struct mlx5_flow_spec *spec;
250
251
	void *mv_misc = NULL;
	void *mc_misc = NULL;
252
253
	u8 *dmac_v = NULL;
	u8 *dmac_c = NULL;
254

255
256
	if (rx_rule)
		match_header |= MLX5_MATCH_MISC_PARAMETERS;
257
258
259

	spec = mlx5_vzalloc(sizeof(*spec));
	if (!spec) {
260
		esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n");
261
		return NULL;
262
	}
263
	dmac_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
264
			      outer_headers.dmac_47_16);
265
	dmac_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
266
267
			      outer_headers.dmac_47_16);

268
	if (match_header & MLX5_MATCH_OUTER_HEADERS) {
269
270
271
		ether_addr_copy(dmac_v, mac_v);
		ether_addr_copy(dmac_c, mac_c);
	}
272

273
	if (match_header & MLX5_MATCH_MISC_PARAMETERS) {
274
275
276
277
		mv_misc  = MLX5_ADDR_OF(fte_match_param, spec->match_value,
					misc_parameters);
		mc_misc  = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
					misc_parameters);
278
279
280
281
		MLX5_SET(fte_match_set_misc, mv_misc, source_port, UPLINK_VPORT);
		MLX5_SET_TO_ONES(fte_match_set_misc, mc_misc, source_port);
	}

282
283
284
285
286
287
	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	dest.vport_num = vport;

	esw_debug(esw->dev,
		  "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n",
		  dmac_v, dmac_c, vport);
288
	spec->match_criteria_enable = match_header;
289
	flow_act.action =  MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
290
	flow_rule =
Mark Bloch's avatar
Mark Bloch committed
291
		mlx5_add_flow_rules(esw->fdb_table.fdb, spec,
292
				    &flow_act, &dest, 1);
293
	if (IS_ERR(flow_rule)) {
294
295
		esw_warn(esw->dev,
			 "FDB: Failed to add flow rule: dmac_v(%pM) dmac_c(%pM) -> vport(%d), err(%ld)\n",
296
297
298
			 dmac_v, dmac_c, vport, PTR_ERR(flow_rule));
		flow_rule = NULL;
	}
299
300

	kvfree(spec);
301
302
303
	return flow_rule;
}

Mark Bloch's avatar
Mark Bloch committed
304
static struct mlx5_flow_handle *
305
306
307
308
309
esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u8 mac[ETH_ALEN], u32 vport)
{
	u8 mac_c[ETH_ALEN];

	eth_broadcast_addr(mac_c);
310
311
312
	return __esw_fdb_set_vport_rule(esw, vport, false, mac_c, mac);
}

Mark Bloch's avatar
Mark Bloch committed
313
static struct mlx5_flow_handle *
314
315
316
317
318
319
320
321
322
323
324
325
esw_fdb_set_vport_allmulti_rule(struct mlx5_eswitch *esw, u32 vport)
{
	u8 mac_c[ETH_ALEN];
	u8 mac_v[ETH_ALEN];

	eth_zero_addr(mac_c);
	eth_zero_addr(mac_v);
	mac_c[0] = 0x01;
	mac_v[0] = 0x01;
	return __esw_fdb_set_vport_rule(esw, vport, false, mac_c, mac_v);
}

Mark Bloch's avatar
Mark Bloch committed
326
static struct mlx5_flow_handle *
327
328
329
330
331
332
333
334
esw_fdb_set_vport_promisc_rule(struct mlx5_eswitch *esw, u32 vport)
{
	u8 mac_c[ETH_ALEN];
	u8 mac_v[ETH_ALEN];

	eth_zero_addr(mac_c);
	eth_zero_addr(mac_v);
	return __esw_fdb_set_vport_rule(esw, vport, true, mac_c, mac_v);
335
336
}

337
static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw, int nvports)
338
{
339
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
340
	struct mlx5_core_dev *dev = esw->dev;
341
	struct mlx5_flow_namespace *root_ns;
342
	struct mlx5_flow_table *fdb;
343
344
345
346
	struct mlx5_flow_group *g;
	void *match_criteria;
	int table_size;
	u32 *flow_group_in;
347
	u8 *dmac;
348
	int err = 0;
349
350
351
352

	esw_debug(dev, "Create FDB log_max_size(%d)\n",
		  MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));

353
354
355
356
357
	root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB);
	if (!root_ns) {
		esw_warn(dev, "Failed to get FDB flow namespace\n");
		return -ENOMEM;
	}
358

359
360
361
362
363
364
	flow_group_in = mlx5_vzalloc(inlen);
	if (!flow_group_in)
		return -ENOMEM;
	memset(flow_group_in, 0, inlen);

	table_size = BIT(MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size));
365
	fdb = mlx5_create_flow_table(root_ns, 0, table_size, 0, 0);
366
	if (IS_ERR(fdb)) {
367
368
369
370
		err = PTR_ERR(fdb);
		esw_warn(dev, "Failed to create FDB Table err %d\n", err);
		goto out;
	}
371
	esw->fdb_table.fdb = fdb;
372

373
	/* Addresses group : Full match unicast/multicast addresses */
374
375
376
377
378
	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_OUTER_HEADERS);
	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
	dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, outer_headers.dmac_47_16);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
379
380
	/* Preserve 2 entries for allmulti and promisc rules*/
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 3);
381
382
	eth_broadcast_addr(dmac);
	g = mlx5_create_flow_group(fdb, flow_group_in);
383
	if (IS_ERR(g)) {
384
385
386
387
		err = PTR_ERR(g);
		esw_warn(dev, "Failed to create flow group err(%d)\n", err);
		goto out;
	}
388
	esw->fdb_table.legacy.addr_grp = g;
389
390
391
392
393
394
395
396
397

	/* Allmulti group : One rule that forwards any mcast traffic */
	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_OUTER_HEADERS);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, table_size - 2);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 2);
	eth_zero_addr(dmac);
	dmac[0] = 0x01;
	g = mlx5_create_flow_group(fdb, flow_group_in);
398
	if (IS_ERR(g)) {
399
400
401
402
		err = PTR_ERR(g);
		esw_warn(dev, "Failed to create allmulti flow group err(%d)\n", err);
		goto out;
	}
403
	esw->fdb_table.legacy.allmulti_grp = g;
404
405
406
407
408
409
410
411
412
413
414

	/* Promiscuous group :
	 * One rule that forward all unmatched traffic from previous groups
	 */
	eth_zero_addr(dmac);
	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_MISC_PARAMETERS);
	MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_port);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, table_size - 1);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 1);
	g = mlx5_create_flow_group(fdb, flow_group_in);
415
	if (IS_ERR(g)) {
416
417
418
419
		err = PTR_ERR(g);
		esw_warn(dev, "Failed to create promisc flow group err(%d)\n", err);
		goto out;
	}
420
	esw->fdb_table.legacy.promisc_grp = g;
421

422
out:
423
	if (err) {
424
425
426
		if (!IS_ERR_OR_NULL(esw->fdb_table.legacy.allmulti_grp)) {
			mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp);
			esw->fdb_table.legacy.allmulti_grp = NULL;
427
		}
428
429
430
		if (!IS_ERR_OR_NULL(esw->fdb_table.legacy.addr_grp)) {
			mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp);
			esw->fdb_table.legacy.addr_grp = NULL;
431
432
433
434
435
436
437
		}
		if (!IS_ERR_OR_NULL(esw->fdb_table.fdb)) {
			mlx5_destroy_flow_table(esw->fdb_table.fdb);
			esw->fdb_table.fdb = NULL;
		}
	}

438
	kvfree(flow_group_in);
439
	return err;
440
441
}

442
static void esw_destroy_legacy_fdb_table(struct mlx5_eswitch *esw)
443
444
445
446
{
	if (!esw->fdb_table.fdb)
		return;

447
	esw_debug(esw->dev, "Destroy FDB Table\n");
448
449
450
	mlx5_destroy_flow_group(esw->fdb_table.legacy.promisc_grp);
	mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp);
	mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp);
451
452
	mlx5_destroy_flow_table(esw->fdb_table.fdb);
	esw->fdb_table.fdb = NULL;
453
454
455
	esw->fdb_table.legacy.addr_grp = NULL;
	esw->fdb_table.legacy.allmulti_grp = NULL;
	esw->fdb_table.legacy.promisc_grp = NULL;
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
}

/* E-Switch vport UC/MC lists management */
typedef int (*vport_addr_action)(struct mlx5_eswitch *esw,
				 struct vport_addr *vaddr);

static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
{
	struct hlist_head *hash = esw->l2_table.l2_hash;
	struct esw_uc_addr *esw_uc;
	u8 *mac = vaddr->node.addr;
	u32 vport = vaddr->vport;
	int err;

	esw_uc = l2addr_hash_find(hash, mac, struct esw_uc_addr);
	if (esw_uc) {
472
473
		esw_warn(esw->dev,
			 "Failed to set L2 mac(%pM) for vport(%d), mac is already in use by vport(%d)\n",
474
			 mac, vport, esw_uc->vport);
475
476
477
		return -EEXIST;
	}

478
479
	esw_uc = l2addr_hash_add(hash, mac, struct esw_uc_addr, GFP_KERNEL);
	if (!esw_uc)
480
		return -ENOMEM;
481
	esw_uc->vport = vport;
482

483
	err = set_l2_table_entry(esw->dev, mac, 0, 0, &esw_uc->table_index);
484
	if (err)
485
486
		goto abort;

487
488
	/* SRIOV is enabled: Forward UC MAC to vport */
	if (esw->fdb_table.fdb && esw->mode == SRIOV_LEGACY)
489
490
491
492
493
494
495
		vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport);

	esw_debug(esw->dev, "\tADDED UC MAC: vport[%d] %pM index:%d fr(%p)\n",
		  vport, mac, esw_uc->table_index, vaddr->flow_rule);
	return err;
abort:
	l2addr_hash_del(esw_uc);
496
497
498
	return err;
}

499
static int esw_del_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
500
{
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
	struct hlist_head *hash = esw->l2_table.l2_hash;
	struct esw_uc_addr *esw_uc;
	u8 *mac = vaddr->node.addr;
	u32 vport = vaddr->vport;

	esw_uc = l2addr_hash_find(hash, mac, struct esw_uc_addr);
	if (!esw_uc || esw_uc->vport != vport) {
		esw_debug(esw->dev,
			  "MAC(%pM) doesn't belong to vport (%d)\n",
			  mac, vport);
		return -EINVAL;
	}
	esw_debug(esw->dev, "\tDELETE UC MAC: vport[%d] %pM index:%d fr(%p)\n",
		  vport, mac, esw_uc->table_index, vaddr->flow_rule);

	del_l2_table_entry(esw->dev, esw_uc->table_index);

	if (vaddr->flow_rule)
Mark Bloch's avatar
Mark Bloch committed
519
		mlx5_del_flow_rules(vaddr->flow_rule);
520
521
522
523
524
525
	vaddr->flow_rule = NULL;

	l2addr_hash_del(esw_uc);
	return 0;
}

526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
static void update_allmulti_vports(struct mlx5_eswitch *esw,
				   struct vport_addr *vaddr,
				   struct esw_mc_addr *esw_mc)
{
	u8 *mac = vaddr->node.addr;
	u32 vport_idx = 0;

	for (vport_idx = 0; vport_idx < esw->total_vports; vport_idx++) {
		struct mlx5_vport *vport = &esw->vports[vport_idx];
		struct hlist_head *vport_hash = vport->mc_list;
		struct vport_addr *iter_vaddr =
					l2addr_hash_find(vport_hash,
							 mac,
							 struct vport_addr);
		if (IS_ERR_OR_NULL(vport->allmulti_rule) ||
		    vaddr->vport == vport_idx)
			continue;
		switch (vaddr->action) {
		case MLX5_ACTION_ADD:
			if (iter_vaddr)
				continue;
			iter_vaddr = l2addr_hash_add(vport_hash, mac,
						     struct vport_addr,
						     GFP_KERNEL);
			if (!iter_vaddr) {
				esw_warn(esw->dev,
					 "ALL-MULTI: Failed to add MAC(%pM) to vport[%d] DB\n",
					 mac, vport_idx);
				continue;
			}
			iter_vaddr->vport = vport_idx;
			iter_vaddr->flow_rule =
					esw_fdb_set_vport_rule(esw,
							       mac,
							       vport_idx);
561
			iter_vaddr->mc_promisc = true;
562
563
564
565
			break;
		case MLX5_ACTION_DEL:
			if (!iter_vaddr)
				continue;
Mark Bloch's avatar
Mark Bloch committed
566
			mlx5_del_flow_rules(iter_vaddr->flow_rule);
567
568
569
570
571
572
			l2addr_hash_del(iter_vaddr);
			break;
		}
	}
}

573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
static int esw_add_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
{
	struct hlist_head *hash = esw->mc_table;
	struct esw_mc_addr *esw_mc;
	u8 *mac = vaddr->node.addr;
	u32 vport = vaddr->vport;

	if (!esw->fdb_table.fdb)
		return 0;

	esw_mc = l2addr_hash_find(hash, mac, struct esw_mc_addr);
	if (esw_mc)
		goto add;

	esw_mc = l2addr_hash_add(hash, mac, struct esw_mc_addr, GFP_KERNEL);
	if (!esw_mc)
		return -ENOMEM;

	esw_mc->uplink_rule = /* Forward MC MAC to Uplink */
		esw_fdb_set_vport_rule(esw, mac, UPLINK_VPORT);
593
594
595
596

	/* Add this multicast mac to all the mc promiscuous vports */
	update_allmulti_vports(esw, vaddr, esw_mc);

597
add:
598
599
600
601
602
603
	/* If the multicast mac is added as a result of mc promiscuous vport,
	 * don't increment the multicast ref count
	 */
	if (!vaddr->mc_promisc)
		esw_mc->refcnt++;

604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
	/* Forward MC MAC to vport */
	vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport);
	esw_debug(esw->dev,
		  "\tADDED MC MAC: vport[%d] %pM fr(%p) refcnt(%d) uplinkfr(%p)\n",
		  vport, mac, vaddr->flow_rule,
		  esw_mc->refcnt, esw_mc->uplink_rule);
	return 0;
}

static int esw_del_mc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr)
{
	struct hlist_head *hash = esw->mc_table;
	struct esw_mc_addr *esw_mc;
	u8 *mac = vaddr->node.addr;
	u32 vport = vaddr->vport;
619

620
621
622
623
624
625
626
	if (!esw->fdb_table.fdb)
		return 0;

	esw_mc = l2addr_hash_find(hash, mac, struct esw_mc_addr);
	if (!esw_mc) {
		esw_warn(esw->dev,
			 "Failed to find eswitch MC addr for MAC(%pM) vport(%d)",
627
628
629
			 mac, vport);
		return -EINVAL;
	}
630
631
632
633
634
635
	esw_debug(esw->dev,
		  "\tDELETE MC MAC: vport[%d] %pM fr(%p) refcnt(%d) uplinkfr(%p)\n",
		  vport, mac, vaddr->flow_rule, esw_mc->refcnt,
		  esw_mc->uplink_rule);

	if (vaddr->flow_rule)
Mark Bloch's avatar
Mark Bloch committed
636
		mlx5_del_flow_rules(vaddr->flow_rule);
637
638
	vaddr->flow_rule = NULL;

639
640
641
642
	/* If the multicast mac is added as a result of mc promiscuous vport,
	 * don't decrement the multicast ref count.
	 */
	if (vaddr->mc_promisc || (--esw_mc->refcnt > 0))
643
		return 0;
644

645
646
647
	/* Remove this multicast mac from all the mc promiscuous vports */
	update_allmulti_vports(esw, vaddr, esw_mc);

648
	if (esw_mc->uplink_rule)
Mark Bloch's avatar
Mark Bloch committed
649
		mlx5_del_flow_rules(esw_mc->uplink_rule);
650
651

	l2addr_hash_del(esw_mc);
652
653
654
	return 0;
}

655
656
657
/* Apply vport UC/MC list to HW l2 table and FDB table */
static void esw_apply_vport_addr_list(struct mlx5_eswitch *esw,
				      u32 vport_num, int list_type)
658
659
{
	struct mlx5_vport *vport = &esw->vports[vport_num];
660
661
662
663
	bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
	vport_addr_action vport_addr_add;
	vport_addr_action vport_addr_del;
	struct vport_addr *addr;
664
665
666
667
668
	struct l2addr_node *node;
	struct hlist_head *hash;
	struct hlist_node *tmp;
	int hi;

669
670
671
672
673
674
	vport_addr_add = is_uc ? esw_add_uc_addr :
				 esw_add_mc_addr;
	vport_addr_del = is_uc ? esw_del_uc_addr :
				 esw_del_mc_addr;

	hash = is_uc ? vport->uc_list : vport->mc_list;
675
	for_each_l2hash_node(node, tmp, hash, hi) {
676
		addr = container_of(node, struct vport_addr, node);
677
678
		switch (addr->action) {
		case MLX5_ACTION_ADD:
679
			vport_addr_add(esw, addr);
680
681
682
			addr->action = MLX5_ACTION_NONE;
			break;
		case MLX5_ACTION_DEL:
683
			vport_addr_del(esw, addr);
684
685
686
687
688
689
			l2addr_hash_del(addr);
			break;
		}
	}
}

690
691
692
/* Sync vport UC/MC list from vport context */
static void esw_update_vport_addr_list(struct mlx5_eswitch *esw,
				       u32 vport_num, int list_type)
693
694
{
	struct mlx5_vport *vport = &esw->vports[vport_num];
695
	bool is_uc = list_type == MLX5_NVPRT_LIST_TYPE_UC;
696
	u8 (*mac_list)[ETH_ALEN];
697
698
	struct l2addr_node *node;
	struct vport_addr *addr;
699
700
701
702
703
704
705
	struct hlist_head *hash;
	struct hlist_node *tmp;
	int size;
	int err;
	int hi;
	int i;

706
707
	size = is_uc ? MLX5_MAX_UC_PER_VPORT(esw->dev) :
		       MLX5_MAX_MC_PER_VPORT(esw->dev);
708
709
710
711
712

	mac_list = kcalloc(size, ETH_ALEN, GFP_KERNEL);
	if (!mac_list)
		return;

713
	hash = is_uc ? vport->uc_list : vport->mc_list;
714
715

	for_each_l2hash_node(node, tmp, hash, hi) {
716
		addr = container_of(node, struct vport_addr, node);
717
718
719
		addr->action = MLX5_ACTION_DEL;
	}

720
721
722
	if (!vport->enabled)
		goto out;

723
	err = mlx5_query_nic_vport_mac_list(esw->dev, vport_num, list_type,
724
725
					    mac_list, &size);
	if (err)
726
		goto out;
727
728
	esw_debug(esw->dev, "vport[%d] context update %s list size (%d)\n",
		  vport_num, is_uc ? "UC" : "MC", size);
729
730

	for (i = 0; i < size; i++) {
731
		if (is_uc && !is_valid_ether_addr(mac_list[i]))
732
733
			continue;

734
735
736
737
		if (!is_uc && !is_multicast_ether_addr(mac_list[i]))
			continue;

		addr = l2addr_hash_find(hash, mac_list[i], struct vport_addr);
738
739
		if (addr) {
			addr->action = MLX5_ACTION_NONE;
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
			/* If this mac was previously added because of allmulti
			 * promiscuous rx mode, its now converted to be original
			 * vport mac.
			 */
			if (addr->mc_promisc) {
				struct esw_mc_addr *esw_mc =
					l2addr_hash_find(esw->mc_table,
							 mac_list[i],
							 struct esw_mc_addr);
				if (!esw_mc) {
					esw_warn(esw->dev,
						 "Failed to MAC(%pM) in mcast DB\n",
						 mac_list[i]);
					continue;
				}
				esw_mc->refcnt++;
				addr->mc_promisc = false;
			}
758
759
760
			continue;
		}

761
		addr = l2addr_hash_add(hash, mac_list[i], struct vport_addr,
762
763
764
765
766
767
768
				       GFP_KERNEL);
		if (!addr) {
			esw_warn(esw->dev,
				 "Failed to add MAC(%pM) to vport[%d] DB\n",
				 mac_list[i], vport_num);
			continue;
		}
769
		addr->vport = vport_num;
770
771
		addr->action = MLX5_ACTION_ADD;
	}
772
out:
773
774
775
	kfree(mac_list);
}

776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
/* Sync vport UC/MC list from vport context
 * Must be called after esw_update_vport_addr_list
 */
static void esw_update_vport_mc_promisc(struct mlx5_eswitch *esw, u32 vport_num)
{
	struct mlx5_vport *vport = &esw->vports[vport_num];
	struct l2addr_node *node;
	struct vport_addr *addr;
	struct hlist_head *hash;
	struct hlist_node *tmp;
	int hi;

	hash = vport->mc_list;

	for_each_l2hash_node(node, tmp, esw->mc_table, hi) {
		u8 *mac = node->addr;

		addr = l2addr_hash_find(hash, mac, struct vport_addr);
		if (addr) {
			if (addr->action == MLX5_ACTION_DEL)
				addr->action = MLX5_ACTION_NONE;
			continue;
		}
		addr = l2addr_hash_add(hash, mac, struct vport_addr,
				       GFP_KERNEL);
		if (!addr) {
			esw_warn(esw->dev,
				 "Failed to add allmulti MAC(%pM) to vport[%d] DB\n",
				 mac, vport_num);
			continue;
		}
		addr->vport = vport_num;
		addr->action = MLX5_ACTION_ADD;
		addr->mc_promisc = true;
	}
}

/* Apply vport rx mode to HW FDB table */
static void esw_apply_vport_rx_mode(struct mlx5_eswitch *esw, u32 vport_num,
				    bool promisc, bool mc_promisc)
{
	struct esw_mc_addr *allmulti_addr = esw->mc_promisc;
	struct mlx5_vport *vport = &esw->vports[vport_num];

	if (IS_ERR_OR_NULL(vport->allmulti_rule) != mc_promisc)
		goto promisc;

	if (mc_promisc) {
		vport->allmulti_rule =
				esw_fdb_set_vport_allmulti_rule(esw, vport_num);
		if (!allmulti_addr->uplink_rule)
			allmulti_addr->uplink_rule =
				esw_fdb_set_vport_allmulti_rule(esw,
								UPLINK_VPORT);
		allmulti_addr->refcnt++;
	} else if (vport->allmulti_rule) {
Mark Bloch's avatar
Mark Bloch committed
832
		mlx5_del_flow_rules(vport->allmulti_rule);
833
834
835
836
837
838
		vport->allmulti_rule = NULL;

		if (--allmulti_addr->refcnt > 0)
			goto promisc;

		if (allmulti_addr->uplink_rule)
Mark Bloch's avatar
Mark Bloch committed
839
			mlx5_del_flow_rules(allmulti_addr->uplink_rule);
840
841
842
843
844
845
846
847
848
849
850
		allmulti_addr->uplink_rule = NULL;
	}

promisc:
	if (IS_ERR_OR_NULL(vport->promisc_rule) != promisc)
		return;

	if (promisc) {
		vport->promisc_rule = esw_fdb_set_vport_promisc_rule(esw,
								     vport_num);
	} else if (vport->promisc_rule) {
Mark Bloch's avatar
Mark Bloch committed
851
		mlx5_del_flow_rules(vport->promisc_rule);
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
		vport->promisc_rule = NULL;
	}
}

/* Sync vport rx mode from vport context */
static void esw_update_vport_rx_mode(struct mlx5_eswitch *esw, u32 vport_num)
{
	struct mlx5_vport *vport = &esw->vports[vport_num];
	int promisc_all = 0;
	int promisc_uc = 0;
	int promisc_mc = 0;
	int err;

	err = mlx5_query_nic_vport_promisc(esw->dev,
					   vport_num,
					   &promisc_uc,
					   &promisc_mc,
					   &promisc_all);
	if (err)
		return;
	esw_debug(esw->dev, "vport[%d] context update rx mode promisc_all=%d, all_multi=%d\n",
		  vport_num, promisc_all, promisc_mc);

875
	if (!vport->info.trusted || !vport->enabled) {
876
877
878
879
880
881
882
883
884
		promisc_uc = 0;
		promisc_mc = 0;
		promisc_all = 0;
	}

	esw_apply_vport_rx_mode(esw, vport_num, promisc_all,
				(promisc_all || promisc_mc));
}

885
static void esw_vport_change_handle_locked(struct mlx5_vport *vport)
886
887
{
	struct mlx5_core_dev *dev = vport->dev;
888
	struct mlx5_eswitch *esw = dev->priv.eswitch;
889
890
891
	u8 mac[ETH_ALEN];

	mlx5_query_nic_vport_mac_address(dev, vport->vport, mac);
892
893
894
895
896
897
898
899
900
	esw_debug(dev, "vport[%d] Context Changed: perm mac: %pM\n",
		  vport->vport, mac);

	if (vport->enabled_events & UC_ADDR_CHANGE) {
		esw_update_vport_addr_list(esw, vport->vport,
					   MLX5_NVPRT_LIST_TYPE_UC);
		esw_apply_vport_addr_list(esw, vport->vport,
					  MLX5_NVPRT_LIST_TYPE_UC);
	}
901

902
903
904
	if (vport->enabled_events & MC_ADDR_CHANGE) {
		esw_update_vport_addr_list(esw, vport->vport,
					   MLX5_NVPRT_LIST_TYPE_MC);
905
906
907
908
909
910
911
912
913
	}

	if (vport->enabled_events & PROMISC_CHANGE) {
		esw_update_vport_rx_mode(esw, vport->vport);
		if (!IS_ERR_OR_NULL(vport->allmulti_rule))
			esw_update_vport_mc_promisc(esw, vport->vport);
	}

	if (vport->enabled_events & (PROMISC_CHANGE | MC_ADDR_CHANGE)) {
914
915
916
		esw_apply_vport_addr_list(esw, vport->vport,
					  MLX5_NVPRT_LIST_TYPE_MC);
	}
917

918
	esw_debug(esw->dev, "vport[%d] Context Changed: Done\n", vport->vport);
919
920
	if (vport->enabled)
		arm_vport_context_events_cmd(dev, vport->vport,
921
					     vport->enabled_events);
922
923
}

924
925
926
927
928
929
930
931
932
933
934
static void esw_vport_change_handler(struct work_struct *work)
{
	struct mlx5_vport *vport =
		container_of(work, struct mlx5_vport, vport_change_handler);
	struct mlx5_eswitch *esw = vport->dev->priv.eswitch;

	mutex_lock(&esw->state_lock);
	esw_vport_change_handle_locked(vport);
	mutex_unlock(&esw->state_lock);
}

935
936
static int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
				       struct mlx5_vport *vport)
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *vlan_grp = NULL;
	struct mlx5_flow_group *drop_grp = NULL;
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_flow_namespace *root_ns;
	struct mlx5_flow_table *acl;
	void *match_criteria;
	u32 *flow_group_in;
	/* The egress acl table contains 2 rules:
	 * 1)Allow traffic with vlan_tag=vst_vlan_id
	 * 2)Drop all other traffic.
	 */
	int table_size = 2;
	int err = 0;

953
954
955
956
957
	if (!MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support))
		return -EOPNOTSUPP;

	if (!IS_ERR_OR_NULL(vport->egress.acl))
		return 0;
958
959
960
961
962
963
964

	esw_debug(dev, "Create vport[%d] egress ACL log_max_size(%d)\n",
		  vport->vport, MLX5_CAP_ESW_EGRESS_ACL(dev, log_max_ft_size));

	root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_EGRESS);
	if (!root_ns) {
		esw_warn(dev, "Failed to get E-Switch egress flow namespace\n");
965
		return -EIO;
966
967
968
969
	}

	flow_group_in = mlx5_vzalloc(inlen);
	if (!flow_group_in)
970
		return -ENOMEM;
971
972

	acl = mlx5_create_vport_flow_table(root_ns, 0, table_size, 0, vport->vport);
973
	if (IS_ERR(acl)) {
974
975
976
977
978
979
980
981
		err = PTR_ERR(acl);
		esw_warn(dev, "Failed to create E-Switch vport[%d] egress flow Table, err(%d)\n",
			 vport->vport, err);
		goto out;
	}

	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
982
	MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag);
983
984
985
986
987
	MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.first_vid);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0);

	vlan_grp = mlx5_create_flow_group(acl, flow_group_in);
988
	if (IS_ERR(vlan_grp)) {
989
990
991
992
993
994
995
996
997
998
		err = PTR_ERR(vlan_grp);
		esw_warn(dev, "Failed to create E-Switch vport[%d] egress allowed vlans flow group, err(%d)\n",
			 vport->vport, err);
		goto out;
	}

	memset(flow_group_in, 0, inlen);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1);
	drop_grp = mlx5_create_flow_group(acl, flow_group_in);
999
	if (IS_ERR(drop_grp)) {
1000
		err = PTR_ERR(drop_grp);