summaryrefslogtreecommitdiff
path: root/examples/tep_termination/vxlan.c
blob: 6545b8ae4f908dff78fd053fcdb277bdf13b8cd2 (plain)
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
/* SPDX-License-Identifier: BSD-3-Clause
 * Copyright(c) 2010-2015 Intel Corporation
 */

#include <stdint.h>
#include <rte_mbuf.h>
#include <rte_hash_crc.h>
#include <rte_byteorder.h>
#include <rte_udp.h>
#include <rte_tcp.h>
#include <rte_sctp.h>

#include "main.h"
#include "vxlan.h"

static uint16_t
get_psd_sum(void *l3_hdr, uint16_t ethertype, uint64_t ol_flags)
{
	if (ethertype == ETHER_TYPE_IPv4)
		return rte_ipv4_phdr_cksum(l3_hdr, ol_flags);
	else /* assume ethertype == ETHER_TYPE_IPv6 */
		return rte_ipv6_phdr_cksum(l3_hdr, ol_flags);
}

/**
 * Parse an ethernet header to fill the ethertype, outer_l2_len, outer_l3_len and
 * ipproto. This function is able to recognize IPv4/IPv6 with one optional vlan
 * header.
 */
static void
parse_ethernet(struct rte_ether_hdr *eth_hdr, union tunnel_offload_info *info,
		uint8_t *l4_proto)
{
	struct ipv4_hdr *ipv4_hdr;
	struct ipv6_hdr *ipv6_hdr;
	uint16_t ethertype;

	info->outer_l2_len = sizeof(struct rte_ether_hdr);
	ethertype = rte_be_to_cpu_16(eth_hdr->ether_type);

	if (ethertype == ETHER_TYPE_VLAN) {
		struct rte_vlan_hdr *vlan_hdr =
			(struct rte_vlan_hdr *)(eth_hdr + 1);
		info->outer_l2_len  += sizeof(struct rte_vlan_hdr);
		ethertype = rte_be_to_cpu_16(vlan_hdr->eth_proto);
	}

	switch (ethertype) {
	case ETHER_TYPE_IPv4:
		ipv4_hdr = (struct ipv4_hdr *)
			((char *)eth_hdr + info->outer_l2_len);
		info->outer_l3_len = sizeof(struct ipv4_hdr);
		*l4_proto = ipv4_hdr->next_proto_id;
		break;
	case ETHER_TYPE_IPv6:
		ipv6_hdr = (struct ipv6_hdr *)
			((char *)eth_hdr + info->outer_l2_len);
		info->outer_l3_len = sizeof(struct ipv6_hdr);
		*l4_proto = ipv6_hdr->proto;
		break;
	default:
		info->outer_l3_len = 0;
		*l4_proto = 0;
		break;
	}
}

/**
 * Calculate the checksum of a packet in hardware
 */
static uint64_t
process_inner_cksums(struct rte_ether_hdr *eth_hdr,
		union tunnel_offload_info *info)
{
	void *l3_hdr = NULL;
	uint8_t l4_proto;
	uint16_t ethertype;
	struct ipv4_hdr *ipv4_hdr;
	struct ipv6_hdr *ipv6_hdr;
	struct udp_hdr *udp_hdr;
	struct tcp_hdr *tcp_hdr;
	struct sctp_hdr *sctp_hdr;
	uint64_t ol_flags = 0;

	info->l2_len = sizeof(struct rte_ether_hdr);
	ethertype = rte_be_to_cpu_16(eth_hdr->ether_type);

	if (ethertype == ETHER_TYPE_VLAN) {
		struct rte_vlan_hdr *vlan_hdr =
			(struct rte_vlan_hdr *)(eth_hdr + 1);
		info->l2_len  += sizeof(struct rte_vlan_hdr);
		ethertype = rte_be_to_cpu_16(vlan_hdr->eth_proto);
	}

	l3_hdr = (char *)eth_hdr + info->l2_len;

	if (ethertype == ETHER_TYPE_IPv4) {
		ipv4_hdr = (struct ipv4_hdr *)l3_hdr;
		ipv4_hdr->hdr_checksum = 0;
		ol_flags |= PKT_TX_IPV4;
		ol_flags |= PKT_TX_IP_CKSUM;
		info->l3_len = sizeof(struct ipv4_hdr);
		l4_proto = ipv4_hdr->next_proto_id;
	} else if (ethertype == ETHER_TYPE_IPv6) {
		ipv6_hdr = (struct ipv6_hdr *)l3_hdr;
		info->l3_len = sizeof(struct ipv6_hdr);
		l4_proto = ipv6_hdr->proto;
		ol_flags |= PKT_TX_IPV6;
	} else
		return 0; /* packet type not supported, nothing to do */

	if (l4_proto == IPPROTO_UDP) {
		udp_hdr = (struct udp_hdr *)((char *)l3_hdr + info->l3_len);
		ol_flags |= PKT_TX_UDP_CKSUM;
		udp_hdr->dgram_cksum = get_psd_sum(l3_hdr,
				ethertype, ol_flags);
	} else if (l4_proto == IPPROTO_TCP) {
		tcp_hdr = (struct tcp_hdr *)((char *)l3_hdr + info->l3_len);
		/* Put PKT_TX_TCP_SEG bit setting before get_psd_sum(), because
		 * it depends on PKT_TX_TCP_SEG to calculate pseudo-header
		 * checksum.
		 */
		if (tso_segsz != 0) {
			ol_flags |= PKT_TX_TCP_SEG;
			info->tso_segsz = tso_segsz;
			info->l4_len = (tcp_hdr->data_off & 0xf0) >> 2;
		}
		ol_flags |= PKT_TX_TCP_CKSUM;
		tcp_hdr->cksum = get_psd_sum(l3_hdr, ethertype, ol_flags);

	} else if (l4_proto == IPPROTO_SCTP) {
		sctp_hdr = (struct sctp_hdr *)((char *)l3_hdr + info->l3_len);
		sctp_hdr->cksum = 0;
		ol_flags |= PKT_TX_SCTP_CKSUM;
	}

	return ol_flags;
}

int
decapsulation(struct rte_mbuf *pkt)
{
	uint8_t l4_proto = 0;
	uint16_t outer_header_len;
	struct udp_hdr *udp_hdr;
	union tunnel_offload_info info = { .data = 0 };
	struct rte_ether_hdr *phdr =
		rte_pktmbuf_mtod(pkt, struct rte_ether_hdr *);

	parse_ethernet(phdr, &info, &l4_proto);

	if (l4_proto != IPPROTO_UDP)
		return -1;

	udp_hdr = (struct udp_hdr *)((char *)phdr +
		info.outer_l2_len + info.outer_l3_len);

	/** check udp destination port, 4789 is the default vxlan port
	 * (rfc7348) or that the rx offload flag is set (i40e only
	 * currently)*/
	if (udp_hdr->dst_port != rte_cpu_to_be_16(DEFAULT_VXLAN_PORT) &&
		(pkt->packet_type & RTE_PTYPE_TUNNEL_MASK) == 0)
		return -1;
	outer_header_len = info.outer_l2_len + info.outer_l3_len
		+ sizeof(struct udp_hdr) + sizeof(struct rte_vxlan_hdr);

	rte_pktmbuf_adj(pkt, outer_header_len);

	return 0;
}

void
encapsulation(struct rte_mbuf *m, uint8_t queue_id)
{
	uint vport_id;
	uint64_t ol_flags = 0;
	uint32_t old_len = m->pkt_len, hash;
	union tunnel_offload_info tx_offload = { .data = 0 };
	struct rte_ether_hdr *phdr =
		rte_pktmbuf_mtod(m, struct rte_ether_hdr *);

	/*Allocate space for new ethernet, IPv4, UDP and VXLAN headers*/
	struct rte_ether_hdr *pneth =
		(struct rte_ether_hdr *) rte_pktmbuf_prepend(m,
		sizeof(struct rte_ether_hdr) + sizeof(struct ipv4_hdr)
		+ sizeof(struct udp_hdr) + sizeof(struct rte_vxlan_hdr));

	struct ipv4_hdr *ip = (struct ipv4_hdr *) &pneth[1];
	struct udp_hdr *udp = (struct udp_hdr *) &ip[1];
	struct rte_vxlan_hdr *vxlan = (struct rte_vxlan_hdr *) &udp[1];

	/* convert TX queue ID to vport ID */
	vport_id = queue_id - 1;

	/* replace original Ethernet header with ours */
	pneth = rte_memcpy(pneth, &app_l2_hdr[vport_id],
		sizeof(struct rte_ether_hdr));

	/* copy in IP header */
	ip = rte_memcpy(ip, &app_ip_hdr[vport_id],
		sizeof(struct ipv4_hdr));
	ip->total_length = rte_cpu_to_be_16(m->pkt_len
				- sizeof(struct rte_ether_hdr));

	/* outer IP checksum */
	ol_flags |= PKT_TX_OUTER_IP_CKSUM;
	ip->hdr_checksum = 0;

	/* inner IP checksum offload */
	if (tx_checksum) {
		ol_flags |= process_inner_cksums(phdr, &tx_offload);
		m->l2_len = tx_offload.l2_len;
		m->l3_len = tx_offload.l3_len;
		m->l4_len = tx_offload.l4_len;
		m->l2_len += ETHER_VXLAN_HLEN;
	}

	m->outer_l2_len = sizeof(struct rte_ether_hdr);
	m->outer_l3_len = sizeof(struct ipv4_hdr);

	ol_flags |= PKT_TX_TUNNEL_VXLAN;

	m->ol_flags |= ol_flags;
	m->tso_segsz = tx_offload.tso_segsz;

	/*VXLAN HEADER*/
	vxlan->vx_flags = rte_cpu_to_be_32(VXLAN_HF_VNI);
	vxlan->vx_vni = rte_cpu_to_be_32(vxdev.out_key << 8);

	/*UDP HEADER*/
	udp->dgram_cksum = 0;
	udp->dgram_len = rte_cpu_to_be_16(old_len
				+ sizeof(struct udp_hdr)
				+ sizeof(struct rte_vxlan_hdr));

	udp->dst_port = rte_cpu_to_be_16(vxdev.dst_port);
	hash = rte_hash_crc(phdr, 2 * ETHER_ADDR_LEN, phdr->ether_type);
	udp->src_port = rte_cpu_to_be_16((((uint64_t) hash * PORT_RANGE) >> 32)
					+ PORT_MIN);

	return;
}