ib_addr.h 10.1 KB
Newer Older
1 2 3 4
/*
 * Copyright (c) 2005 Voltaire Inc.  All rights reserved.
 * Copyright (c) 2005 Intel Corporation.  All rights reserved.
 *
Sean Hefty's avatar
Sean Hefty committed
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
 * 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.
32 33 34 35 36 37 38
 */

#if !defined(IB_ADDR_H)
#define IB_ADDR_H

#include <linux/in.h>
#include <linux/in6.h>
39
#include <linux/if_arp.h>
40
#include <linux/netdevice.h>
41
#include <linux/inetdevice.h>
42
#include <linux/socket.h>
43
#include <linux/if_vlan.h>
44 45 46
#include <net/ipv6.h>
#include <net/if_inet6.h>
#include <net/ip.h>
47
#include <rdma/ib_verbs.h>
48
#include <rdma/ib_pack.h>
49
#include <net/ipv6.h>
50
#include <net/net_namespace.h>
51

52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
struct rdma_addr_client {
	atomic_t refcount;
	struct completion comp;
};

/**
 * rdma_addr_register_client - Register an address client.
 */
void rdma_addr_register_client(struct rdma_addr_client *client);

/**
 * rdma_addr_unregister_client - Deregister an address client.
 * @client: Client object to deregister.
 */
void rdma_addr_unregister_client(struct rdma_addr_client *client);

68 69 70 71 72 73 74 75 76 77
/**
 * struct rdma_dev_addr - Contains resolved RDMA hardware addresses
 * @src_dev_addr:	Source MAC address.
 * @dst_dev_addr:	Destination MAC address.
 * @broadcast:		Broadcast address of the device.
 * @dev_type:		The interface hardware type of the device.
 * @bound_dev_if:	An optional device interface index.
 * @transport:		The transport type used.
 * @net:		Network namespace containing the bound_dev_if net_dev.
 */
78 79 80 81
struct rdma_dev_addr {
	unsigned char src_dev_addr[MAX_ADDR_LEN];
	unsigned char dst_dev_addr[MAX_ADDR_LEN];
	unsigned char broadcast[MAX_ADDR_LEN];
82
	unsigned short dev_type;
83
	int bound_dev_if;
84
	enum rdma_transport_type transport;
85
	struct net *net;
86
	enum rdma_network_type network;
87
	int hoplimit;
88 89 90 91 92
};

/**
 * rdma_translate_ip - Translate a local IP address to an RDMA hardware
 *   address.
93 94
 *
 * The dev_addr->net field must be initialized.
95
 */
96 97
int rdma_translate_ip(const struct sockaddr *addr,
		      struct rdma_dev_addr *dev_addr, u16 *vlan_id);
98 99 100 101

/**
 * rdma_resolve_ip - Resolve source and destination IP addresses to
 *   RDMA hardware addresses.
102
 * @client: Address client associated with request.
103 104 105 106 107 108
 * @src_addr: An optional source address to use in the resolution.  If a
 *   source address is not provided, a usable address will be returned via
 *   the callback.
 * @dst_addr: The destination address to resolve.
 * @addr: A reference to a data location that will receive the resolved
 *   addresses.  The data location must remain valid until the callback has
109
 *   been invoked. The net field of the addr struct must be valid.
110 111 112 113 114
 * @timeout_ms: Amount of time to wait for the address resolution to complete.
 * @callback: Call invoked once address resolution has completed, timed out,
 *   or been canceled.  A status of 0 indicates success.
 * @context: User-specified context associated with the call.
 */
115 116
int rdma_resolve_ip(struct rdma_addr_client *client,
		    struct sockaddr *src_addr, struct sockaddr *dst_addr,
117 118 119 120 121
		    struct rdma_dev_addr *addr, int timeout_ms,
		    void (*callback)(int status, struct sockaddr *src_addr,
				     struct rdma_dev_addr *addr, void *context),
		    void *context);

122 123 124 125
int rdma_resolve_ip_route(struct sockaddr *src_addr,
			  const struct sockaddr *dst_addr,
			  struct rdma_dev_addr *addr);

126 127
void rdma_addr_cancel(struct rdma_dev_addr *addr);

Tom Tucker's avatar
Tom Tucker committed
128 129 130
int rdma_copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev,
	      const unsigned char *dst_dev_addr);

131
int rdma_addr_size(struct sockaddr *addr);
132 133
int rdma_addr_size_in6(struct sockaddr_in6 *addr);
int rdma_addr_size_kss(struct __kernel_sockaddr_storage *addr);
134

135
int rdma_addr_find_smac_by_sgid(union ib_gid *sgid, u8 *smac, u16 *vlan_id);
136 137
int rdma_addr_find_l2_eth_by_grh(const union ib_gid *sgid,
				 const union ib_gid *dgid,
138 139
				 u8 *smac, u16 *vlan_id, int *if_index,
				 int *hoplimit);
140

141 142 143 144 145 146 147 148 149 150 151
static inline u16 ib_addr_get_pkey(struct rdma_dev_addr *dev_addr)
{
	return ((u16)dev_addr->broadcast[8] << 8) | (u16)dev_addr->broadcast[9];
}

static inline void ib_addr_set_pkey(struct rdma_dev_addr *dev_addr, u16 pkey)
{
	dev_addr->broadcast[8] = pkey >> 8;
	dev_addr->broadcast[9] = (unsigned char) pkey;
}

152 153 154 155 156 157
static inline void ib_addr_get_mgid(struct rdma_dev_addr *dev_addr,
				    union ib_gid *gid)
{
	memcpy(gid, dev_addr->broadcast + 4, sizeof *gid);
}

158
static inline int rdma_addr_gid_offset(struct rdma_dev_addr *dev_addr)
159
{
160
	return dev_addr->dev_type == ARPHRD_INFINIBAND ? 4 : 0;
161 162
}

163
static inline u16 rdma_vlan_dev_vlan_id(const struct net_device *dev)
164
{
165
	return is_vlan_dev(dev) ? vlan_dev_vlan_id(dev) : 0xffff;
166 167
}

168 169 170 171 172 173 174 175 176
static inline int rdma_ip2gid(struct sockaddr *addr, union ib_gid *gid)
{
	switch (addr->sa_family) {
	case AF_INET:
		ipv6_addr_set_v4mapped(((struct sockaddr_in *)
					addr)->sin_addr.s_addr,
				       (struct in6_addr *)gid);
		break;
	case AF_INET6:
177 178
		*(struct in6_addr *)&gid->raw =
			((struct sockaddr_in6 *)addr)->sin6_addr;
179 180 181 182 183 184 185 186
		break;
	default:
		return -EINVAL;
	}
	return 0;
}

/* Important - sockaddr should be a union of sockaddr_in and sockaddr_in6 */
187
static inline void rdma_gid2ip(struct sockaddr *out, const union ib_gid *gid)
188 189 190 191 192 193 194 195 196 197 198 199 200 201
{
	if (ipv6_addr_v4mapped((struct in6_addr *)gid)) {
		struct sockaddr_in *out_in = (struct sockaddr_in *)out;
		memset(out_in, 0, sizeof(*out_in));
		out_in->sin_family = AF_INET;
		memcpy(&out_in->sin_addr.s_addr, gid->raw + 12, 4);
	} else {
		struct sockaddr_in6 *out_in = (struct sockaddr_in6 *)out;
		memset(out_in, 0, sizeof(*out_in));
		out_in->sin6_family = AF_INET6;
		memcpy(&out_in->sin6_addr.s6_addr, gid->raw, 16);
	}
}

202 203 204
static inline void iboe_addr_get_sgid(struct rdma_dev_addr *dev_addr,
				      union ib_gid *gid)
{
205
	struct net_device *dev;
206
	struct in_device *ip4;
207 208 209

	dev = dev_get_by_index(&init_net, dev_addr->bound_dev_if);
	if (dev) {
210
		ip4 = in_dev_get(dev);
211
		if (ip4 && ip4->ifa_list && ip4->ifa_list->ifa_address)
212 213
			ipv6_addr_set_v4mapped(ip4->ifa_list->ifa_address,
					       (struct in6_addr *)gid);
214 215

		if (ip4)
216
			in_dev_put(ip4);
217

218 219
		dev_put(dev);
	}
220 221
}

222
static inline void rdma_addr_get_sgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
223
{
224 225 226 227 228 229
	if (dev_addr->transport == RDMA_TRANSPORT_IB &&
	    dev_addr->dev_type != ARPHRD_INFINIBAND)
		iboe_addr_get_sgid(dev_addr, gid);
	else
		memcpy(gid, dev_addr->src_dev_addr +
		       rdma_addr_gid_offset(dev_addr), sizeof *gid);
230 231
}

232
static inline void rdma_addr_set_sgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
233
{
234
	memcpy(dev_addr->src_dev_addr + rdma_addr_gid_offset(dev_addr), gid, sizeof *gid);
235 236
}

237
static inline void rdma_addr_get_dgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
238
{
239
	memcpy(gid, dev_addr->dst_dev_addr + rdma_addr_gid_offset(dev_addr), sizeof *gid);
240 241
}

242
static inline void rdma_addr_set_dgid(struct rdma_dev_addr *dev_addr, union ib_gid *gid)
Tom Tucker's avatar
Tom Tucker committed
243
{
244
	memcpy(dev_addr->dst_dev_addr + rdma_addr_gid_offset(dev_addr), gid, sizeof *gid);
Tom Tucker's avatar
Tom Tucker committed
245 246
}

247 248 249
static inline enum ib_mtu iboe_get_mtu(int mtu)
{
	/*
250
	 * Reduce IB headers from effective IBoE MTU.
251
	 */
252 253 254
	mtu = mtu - (IB_GRH_BYTES + IB_UDP_BYTES + IB_BTH_BYTES +
		     IB_EXT_XRC_BYTES + IB_EXT_ATOMICETH_BYTES +
		     IB_ICRC_BYTES);
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271

	if (mtu >= ib_mtu_enum_to_int(IB_MTU_4096))
		return IB_MTU_4096;
	else if (mtu >= ib_mtu_enum_to_int(IB_MTU_2048))
		return IB_MTU_2048;
	else if (mtu >= ib_mtu_enum_to_int(IB_MTU_1024))
		return IB_MTU_1024;
	else if (mtu >= ib_mtu_enum_to_int(IB_MTU_512))
		return IB_MTU_512;
	else if (mtu >= ib_mtu_enum_to_int(IB_MTU_256))
		return IB_MTU_256;
	else
		return 0;
}

static inline int iboe_get_rate(struct net_device *dev)
{
272
	struct ethtool_link_ksettings cmd;
273
	int err;
274

275
	rtnl_lock();
276
	err = __ethtool_get_link_ksettings(dev, &cmd);
277 278
	rtnl_unlock();
	if (err)
279 280
		return IB_RATE_PORT_CURRENT;

281
	if (cmd.base.speed >= 40000)
282
		return IB_RATE_40_GBPS;
283
	else if (cmd.base.speed >= 30000)
284
		return IB_RATE_30_GBPS;
285
	else if (cmd.base.speed >= 20000)
286
		return IB_RATE_20_GBPS;
287
	else if (cmd.base.speed >= 10000)
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
		return IB_RATE_10_GBPS;
	else
		return IB_RATE_PORT_CURRENT;
}

static inline int rdma_link_local_addr(struct in6_addr *addr)
{
	if (addr->s6_addr32[0] == htonl(0xfe800000) &&
	    addr->s6_addr32[1] == 0)
		return 1;

	return 0;
}

static inline void rdma_get_ll_mac(struct in6_addr *addr, u8 *mac)
{
	memcpy(mac, &addr->s6_addr[8], 3);
	memcpy(mac + 3, &addr->s6_addr[13], 3);
	mac[0] ^= 2;
}

static inline int rdma_is_multicast_addr(struct in6_addr *addr)
{
311
	__be32 ipv4_addr;
312 313 314 315

	if (addr->s6_addr[0] == 0xff)
		return 1;

316
	ipv4_addr = addr->s6_addr32[3];
317
	return (ipv6_addr_v4mapped(addr) && ipv4_is_multicast(ipv4_addr));
318 319 320 321 322 323 324 325 326 327 328 329
}

static inline void rdma_get_mcast_mac(struct in6_addr *addr, u8 *mac)
{
	int i;

	mac[0] = 0x33;
	mac[1] = 0x33;
	for (i = 2; i < 6; ++i)
		mac[i] = addr->s6_addr[i + 10];
}

330 331 332 333 334 335 336 337 338 339
static inline u16 rdma_get_vlan_id(union ib_gid *dgid)
{
	u16 vid;

	vid = dgid->raw[11] << 8 | dgid->raw[12];
	return vid < 0x1000 ? vid : 0xffff;
}

static inline struct net_device *rdma_vlan_dev_real_dev(const struct net_device *dev)
{
340
	return is_vlan_dev(dev) ? vlan_dev_real_dev(dev) : NULL;
341 342
}

343
#endif /* IB_ADDR_H */