DPDK logo

Elixir Cross Referencer

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
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
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
/*-
 *   BSD LICENSE
 *
 *   Copyright 2015 6WIND S.A.
 *   Copyright 2015 Mellanox.
 *
 *   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.
 *     * Neither the name of 6WIND S.A. nor the names of its
 *       contributors may be used to endorse or promote products derived
 *       from this software without specific prior written permission.
 *
 *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
 *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
 *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#include <stddef.h>
#include <unistd.h>
#include <string.h>
#include <assert.h>
#include <stdint.h>
#include <stdlib.h>
#include <net/if.h>

/* Verbs header. */
/* ISO C doesn't support unnamed structs/unions, disabling -pedantic. */
#ifdef PEDANTIC
#pragma GCC diagnostic ignored "-pedantic"
#endif
#include <infiniband/verbs.h>
#ifdef PEDANTIC
#pragma GCC diagnostic error "-pedantic"
#endif

/* DPDK headers don't like -pedantic. */
#ifdef PEDANTIC
#pragma GCC diagnostic ignored "-pedantic"
#endif
#include <rte_malloc.h>
#include <rte_ethdev.h>
#include <rte_pci.h>
#include <rte_common.h>
#ifdef PEDANTIC
#pragma GCC diagnostic error "-pedantic"
#endif

#include "mlx5.h"
#include "mlx5_utils.h"
#include "mlx5_rxtx.h"
#include "mlx5_autoconf.h"
#include "mlx5_defs.h"

/**
 * DPDK callback to close the device.
 *
 * Destroy all queues and objects, free memory.
 *
 * @param dev
 *   Pointer to Ethernet device structure.
 */
static void
mlx5_dev_close(struct rte_eth_dev *dev)
{
	struct priv *priv = dev->data->dev_private;
	void *tmp;
	unsigned int i;

	priv_lock(priv);
	DEBUG("%p: closing device \"%s\"",
	      (void *)dev,
	      ((priv->ctx != NULL) ? priv->ctx->device->name : ""));
	/* In case mlx5_dev_stop() has not been called. */
	priv_dev_interrupt_handler_uninstall(priv, dev);
	priv_allmulticast_disable(priv);
	priv_promiscuous_disable(priv);
	priv_mac_addrs_disable(priv);
	priv_destroy_hash_rxqs(priv);
	/* Prevent crashes when queues are still in use. */
	dev->rx_pkt_burst = removed_rx_burst;
	dev->tx_pkt_burst = removed_tx_burst;
	if (priv->rxqs != NULL) {
		/* XXX race condition if mlx5_rx_burst() is still running. */
		usleep(1000);
		for (i = 0; (i != priv->rxqs_n); ++i) {
			tmp = (*priv->rxqs)[i];
			if (tmp == NULL)
				continue;
			(*priv->rxqs)[i] = NULL;
			rxq_cleanup(tmp);
			rte_free(tmp);
		}
		priv->rxqs_n = 0;
		priv->rxqs = NULL;
	}
	if (priv->txqs != NULL) {
		/* XXX race condition if mlx5_tx_burst() is still running. */
		usleep(1000);
		for (i = 0; (i != priv->txqs_n); ++i) {
			tmp = (*priv->txqs)[i];
			if (tmp == NULL)
				continue;
			(*priv->txqs)[i] = NULL;
			txq_cleanup(tmp);
			rte_free(tmp);
		}
		priv->txqs_n = 0;
		priv->txqs = NULL;
	}
	if (priv->pd != NULL) {
		assert(priv->ctx != NULL);
		claim_zero(ibv_dealloc_pd(priv->pd));
		claim_zero(ibv_close_device(priv->ctx));
	} else
		assert(priv->ctx == NULL);
	if (priv->rss_conf != NULL) {
		for (i = 0; (i != hash_rxq_init_n); ++i)
			rte_free((*priv->rss_conf)[i]);
		rte_free(priv->rss_conf);
	}
	if (priv->reta_idx != NULL)
		rte_free(priv->reta_idx);
	priv_unlock(priv);
	memset(priv, 0, sizeof(*priv));
}

static const struct eth_dev_ops mlx5_dev_ops = {
	.dev_configure = mlx5_dev_configure,
	.dev_start = mlx5_dev_start,
	.dev_stop = mlx5_dev_stop,
	.dev_close = mlx5_dev_close,
	.promiscuous_enable = mlx5_promiscuous_enable,
	.promiscuous_disable = mlx5_promiscuous_disable,
	.allmulticast_enable = mlx5_allmulticast_enable,
	.allmulticast_disable = mlx5_allmulticast_disable,
	.link_update = mlx5_link_update,
	.stats_get = mlx5_stats_get,
	.stats_reset = mlx5_stats_reset,
	.dev_infos_get = mlx5_dev_infos_get,
	.vlan_filter_set = mlx5_vlan_filter_set,
	.rx_queue_setup = mlx5_rx_queue_setup,
	.tx_queue_setup = mlx5_tx_queue_setup,
	.rx_queue_release = mlx5_rx_queue_release,
	.tx_queue_release = mlx5_tx_queue_release,
	.flow_ctrl_get = mlx5_dev_get_flow_ctrl,
	.flow_ctrl_set = mlx5_dev_set_flow_ctrl,
	.mac_addr_remove = mlx5_mac_addr_remove,
	.mac_addr_add = mlx5_mac_addr_add,
	.mtu_set = mlx5_dev_set_mtu,
	.reta_update = mlx5_dev_rss_reta_update,
	.reta_query = mlx5_dev_rss_reta_query,
	.rss_hash_update = mlx5_rss_hash_update,
	.rss_hash_conf_get = mlx5_rss_hash_conf_get,
};

static struct {
	struct rte_pci_addr pci_addr; /* associated PCI address */
	uint32_t ports; /* physical ports bitfield. */
} mlx5_dev[32];

/**
 * Get device index in mlx5_dev[] from PCI bus address.
 *
 * @param[in] pci_addr
 *   PCI bus address to look for.
 *
 * @return
 *   mlx5_dev[] index on success, -1 on failure.
 */
static int
mlx5_dev_idx(struct rte_pci_addr *pci_addr)
{
	unsigned int i;
	int ret = -1;

	assert(pci_addr != NULL);
	for (i = 0; (i != RTE_DIM(mlx5_dev)); ++i) {
		if ((mlx5_dev[i].pci_addr.domain == pci_addr->domain) &&
		    (mlx5_dev[i].pci_addr.bus == pci_addr->bus) &&
		    (mlx5_dev[i].pci_addr.devid == pci_addr->devid) &&
		    (mlx5_dev[i].pci_addr.function == pci_addr->function))
			return i;
		if ((mlx5_dev[i].ports == 0) && (ret == -1))
			ret = i;
	}
	return ret;
}

static struct eth_driver mlx5_driver;

/**
 * DPDK callback to register a PCI device.
 *
 * This function creates an Ethernet device for each port of a given
 * PCI device.
 *
 * @param[in] pci_drv
 *   PCI driver structure (mlx5_driver).
 * @param[in] pci_dev
 *   PCI device information.
 *
 * @return
 *   0 on success, negative errno value on failure.
 */
static int
mlx5_pci_devinit(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev)
{
	struct ibv_device **list;
	struct ibv_device *ibv_dev;
	int err = 0;
	struct ibv_context *attr_ctx = NULL;
	struct ibv_device_attr device_attr;
	unsigned int vf;
	int idx;
	int i;

	(void)pci_drv;
	assert(pci_drv == &mlx5_driver.pci_drv);
	/* Get mlx5_dev[] index. */
	idx = mlx5_dev_idx(&pci_dev->addr);
	if (idx == -1) {
		ERROR("this driver cannot support any more adapters");
		return -ENOMEM;
	}
	DEBUG("using driver device index %d", idx);

	/* Save PCI address. */
	mlx5_dev[idx].pci_addr = pci_dev->addr;
	list = ibv_get_device_list(&i);
	if (list == NULL) {
		assert(errno);
		if (errno == ENOSYS) {
			WARN("cannot list devices, is ib_uverbs loaded?");
			return 0;
		}
		return -errno;
	}
	assert(i >= 0);
	/*
	 * For each listed device, check related sysfs entry against
	 * the provided PCI ID.
	 */
	while (i != 0) {
		struct rte_pci_addr pci_addr;

		--i;
		DEBUG("checking device \"%s\"", list[i]->name);
		if (mlx5_ibv_device_to_pci_addr(list[i], &pci_addr))
			continue;
		if ((pci_dev->addr.domain != pci_addr.domain) ||
		    (pci_dev->addr.bus != pci_addr.bus) ||
		    (pci_dev->addr.devid != pci_addr.devid) ||
		    (pci_dev->addr.function != pci_addr.function))
			continue;
		vf = ((pci_dev->id.device_id ==
		       PCI_DEVICE_ID_MELLANOX_CONNECTX4VF) ||
		      (pci_dev->id.device_id ==
		       PCI_DEVICE_ID_MELLANOX_CONNECTX4LXVF));
		INFO("PCI information matches, using device \"%s\" (VF: %s)",
		     list[i]->name, (vf ? "true" : "false"));
		attr_ctx = ibv_open_device(list[i]);
		err = errno;
		break;
	}
	if (attr_ctx == NULL) {
		ibv_free_device_list(list);
		switch (err) {
		case 0:
			WARN("cannot access device, is mlx5_ib loaded?");
			return 0;
		case EINVAL:
			WARN("cannot use device, are drivers up to date?");
			return 0;
		}
		assert(err > 0);
		return -err;
	}
	ibv_dev = list[i];

	DEBUG("device opened");
	if (ibv_query_device(attr_ctx, &device_attr))
		goto error;
	INFO("%u port(s) detected", device_attr.phys_port_cnt);

	for (i = 0; i < device_attr.phys_port_cnt; i++) {
		uint32_t port = i + 1; /* ports are indexed from one */
		uint32_t test = (1 << i);
		struct ibv_context *ctx = NULL;
		struct ibv_port_attr port_attr;
		struct ibv_pd *pd = NULL;
		struct priv *priv = NULL;
		struct rte_eth_dev *eth_dev;
#ifdef HAVE_EXP_QUERY_DEVICE
		struct ibv_exp_device_attr exp_device_attr;
#endif /* HAVE_EXP_QUERY_DEVICE */
		struct ether_addr mac;

#ifdef HAVE_EXP_QUERY_DEVICE
		exp_device_attr.comp_mask =
			IBV_EXP_DEVICE_ATTR_EXP_CAP_FLAGS |
			IBV_EXP_DEVICE_ATTR_RX_HASH;
#endif /* HAVE_EXP_QUERY_DEVICE */

		DEBUG("using port %u (%08" PRIx32 ")", port, test);

		ctx = ibv_open_device(ibv_dev);
		if (ctx == NULL)
			goto port_error;

		/* Check port status. */
		err = ibv_query_port(ctx, port, &port_attr);
		if (err) {
			ERROR("port query failed: %s", strerror(err));
			goto port_error;
		}
		if (port_attr.state != IBV_PORT_ACTIVE)
			DEBUG("port %d is not active: \"%s\" (%d)",
			      port, ibv_port_state_str(port_attr.state),
			      port_attr.state);

		/* Allocate protection domain. */
		pd = ibv_alloc_pd(ctx);
		if (pd == NULL) {
			ERROR("PD allocation failure");
			err = ENOMEM;
			goto port_error;
		}

		mlx5_dev[idx].ports |= test;

		/* from rte_ethdev.c */
		priv = rte_zmalloc("ethdev private structure",
				   sizeof(*priv),
				   RTE_CACHE_LINE_SIZE);
		if (priv == NULL) {
			ERROR("priv allocation failure");
			err = ENOMEM;
			goto port_error;
		}

		priv->ctx = ctx;
		priv->device_attr = device_attr;
		priv->port = port;
		priv->pd = pd;
		priv->mtu = ETHER_MTU;
#ifdef HAVE_EXP_QUERY_DEVICE
		if (ibv_exp_query_device(ctx, &exp_device_attr)) {
			ERROR("ibv_exp_query_device() failed");
			goto port_error;
		}

		priv->hw_csum =
			((exp_device_attr.exp_device_cap_flags &
			  IBV_EXP_DEVICE_RX_CSUM_TCP_UDP_PKT) &&
			 (exp_device_attr.exp_device_cap_flags &
			  IBV_EXP_DEVICE_RX_CSUM_IP_PKT));
		DEBUG("checksum offloading is %ssupported",
		      (priv->hw_csum ? "" : "not "));

		priv->hw_csum_l2tun = !!(exp_device_attr.exp_device_cap_flags &
					 IBV_EXP_DEVICE_VXLAN_SUPPORT);
		DEBUG("L2 tunnel checksum offloads are %ssupported",
		      (priv->hw_csum_l2tun ? "" : "not "));

		priv->ind_table_max_size = exp_device_attr.rx_hash_caps.max_rwq_indirection_table_size;
		/* Remove this check once DPDK supports larger/variable
		 * indirection tables. */
		if (priv->ind_table_max_size > (unsigned int)RSS_INDIRECTION_TABLE_SIZE)
			priv->ind_table_max_size = RSS_INDIRECTION_TABLE_SIZE;
		DEBUG("maximum RX indirection table size is %u",
		      priv->ind_table_max_size);

#else /* HAVE_EXP_QUERY_DEVICE */
		priv->ind_table_max_size = RSS_INDIRECTION_TABLE_SIZE;
#endif /* HAVE_EXP_QUERY_DEVICE */

		priv->vf = vf;
		/* Allocate and register default RSS hash keys. */
		priv->rss_conf = rte_calloc(__func__, hash_rxq_init_n,
					    sizeof((*priv->rss_conf)[0]), 0);
		if (priv->rss_conf == NULL) {
			err = ENOMEM;
			goto port_error;
		}
		err = rss_hash_rss_conf_new_key(priv,
						rss_hash_default_key,
						rss_hash_default_key_len,
						ETH_RSS_PROTO_MASK);
		if (err)
			goto port_error;
		/* Configure the first MAC address by default. */
		if (priv_get_mac(priv, &mac.addr_bytes)) {
			ERROR("cannot get MAC address, is mlx5_en loaded?"
			      " (errno: %s)", strerror(errno));
			goto port_error;
		}
		INFO("port %u MAC address is %02x:%02x:%02x:%02x:%02x:%02x",
		     priv->port,
		     mac.addr_bytes[0], mac.addr_bytes[1],
		     mac.addr_bytes[2], mac.addr_bytes[3],
		     mac.addr_bytes[4], mac.addr_bytes[5]);
		/* Register MAC and broadcast addresses. */
		claim_zero(priv_mac_addr_add(priv, 0,
					     (const uint8_t (*)[ETHER_ADDR_LEN])
					     mac.addr_bytes));
		claim_zero(priv_mac_addr_add(priv, (RTE_DIM(priv->mac) - 1),
					     &(const uint8_t [ETHER_ADDR_LEN])
					     { "\xff\xff\xff\xff\xff\xff" }));
#ifndef NDEBUG
		{
			char ifname[IF_NAMESIZE];

			if (priv_get_ifname(priv, &ifname) == 0)
				DEBUG("port %u ifname is \"%s\"",
				      priv->port, ifname);
			else
				DEBUG("port %u ifname is unknown", priv->port);
		}
#endif
		/* Get actual MTU if possible. */
		priv_get_mtu(priv, &priv->mtu);
		DEBUG("port %u MTU is %u", priv->port, priv->mtu);

		/* from rte_ethdev.c */
		{
			char name[RTE_ETH_NAME_MAX_LEN];

			snprintf(name, sizeof(name), "%s port %u",
				 ibv_get_device_name(ibv_dev), port);
			eth_dev = rte_eth_dev_allocate(name, RTE_ETH_DEV_PCI);
		}
		if (eth_dev == NULL) {
			ERROR("can not allocate rte ethdev");
			err = ENOMEM;
			goto port_error;
		}

		eth_dev->data->dev_private = priv;
		eth_dev->pci_dev = pci_dev;

		rte_eth_copy_pci_info(eth_dev, pci_dev);

		eth_dev->driver = &mlx5_driver;
		eth_dev->data->rx_mbuf_alloc_failed = 0;
		eth_dev->data->mtu = ETHER_MTU;

		priv->dev = eth_dev;
		eth_dev->dev_ops = &mlx5_dev_ops;
		eth_dev->data->mac_addrs = priv->mac;
		TAILQ_INIT(&eth_dev->link_intr_cbs);

		/* Bring Ethernet device up. */
		DEBUG("forcing Ethernet interface up");
		priv_set_flags(priv, ~IFF_UP, IFF_UP);
		continue;

port_error:
		rte_free(priv->rss_conf);
		rte_free(priv);
		if (pd)
			claim_zero(ibv_dealloc_pd(pd));
		if (ctx)
			claim_zero(ibv_close_device(ctx));
		break;
	}

	/*
	 * XXX if something went wrong in the loop above, there is a resource
	 * leak (ctx, pd, priv, dpdk ethdev) but we can do nothing about it as
	 * long as the dpdk does not provide a way to deallocate a ethdev and a
	 * way to enumerate the registered ethdevs to free the previous ones.
	 */

	/* no port found, complain */
	if (!mlx5_dev[idx].ports) {
		err = ENODEV;
		goto error;
	}

error:
	if (attr_ctx)
		claim_zero(ibv_close_device(attr_ctx));
	if (list)
		ibv_free_device_list(list);
	assert(err >= 0);
	return -err;
}

static const struct rte_pci_id mlx5_pci_id_map[] = {
	{
		.vendor_id = PCI_VENDOR_ID_MELLANOX,
		.device_id = PCI_DEVICE_ID_MELLANOX_CONNECTX4,
		.subsystem_vendor_id = PCI_ANY_ID,
		.subsystem_device_id = PCI_ANY_ID
	},
	{
		.vendor_id = PCI_VENDOR_ID_MELLANOX,
		.device_id = PCI_DEVICE_ID_MELLANOX_CONNECTX4VF,
		.subsystem_vendor_id = PCI_ANY_ID,
		.subsystem_device_id = PCI_ANY_ID
	},
	{
		.vendor_id = PCI_VENDOR_ID_MELLANOX,
		.device_id = PCI_DEVICE_ID_MELLANOX_CONNECTX4LX,
		.subsystem_vendor_id = PCI_ANY_ID,
		.subsystem_device_id = PCI_ANY_ID
	},
	{
		.vendor_id = PCI_VENDOR_ID_MELLANOX,
		.device_id = PCI_DEVICE_ID_MELLANOX_CONNECTX4LXVF,
		.subsystem_vendor_id = PCI_ANY_ID,
		.subsystem_device_id = PCI_ANY_ID
	},
	{
		.vendor_id = 0
	}
};

static struct eth_driver mlx5_driver = {
	.pci_drv = {
		.name = MLX5_DRIVER_NAME,
		.id_table = mlx5_pci_id_map,
		.devinit = mlx5_pci_devinit,
		.drv_flags = RTE_PCI_DRV_INTR_LSC,
	},
	.dev_private_size = sizeof(struct priv)
};

/**
 * Driver initialization routine.
 */
static int
rte_mlx5_pmd_init(const char *name, const char *args)
{
	(void)name;
	(void)args;
	/*
	 * RDMAV_HUGEPAGES_SAFE tells ibv_fork_init() we intend to use
	 * huge pages. Calling ibv_fork_init() during init allows
	 * applications to use fork() safely for purposes other than
	 * using this PMD, which is not supported in forked processes.
	 */
	setenv("RDMAV_HUGEPAGES_SAFE", "1", 1);
	ibv_fork_init();
	rte_eal_pci_register(&mlx5_driver.pci_drv);
	return 0;
}

static struct rte_driver rte_mlx5_driver = {
	.type = PMD_PDEV,
	.name = MLX5_DRIVER_NAME,
	.init = rte_mlx5_pmd_init,
};

PMD_REGISTER_DRIVER(rte_mlx5_driver)