1 /* $NetBSD: ip_flow.c,v 1.23 2002/03/08 20:48:43 thorpej Exp $ */ 2 3 /*- 4 * Copyright (c) 1998 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by the 3am Software Foundry ("3am"). It was developed by Matt Thomas. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 3. All advertising materials mentioning features or use of this software 19 * must display the following acknowledgement: 20 * This product includes software developed by the NetBSD 21 * Foundation, Inc. and its contributors. 22 * 4. Neither the name of The NetBSD Foundation nor the names of its 23 * contributors may be used to endorse or promote products derived 24 * from this software without specific prior written permission. 25 * 26 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 27 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 28 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 29 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 30 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 31 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 32 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 33 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 34 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 35 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 36 * POSSIBILITY OF SUCH DAMAGE. 37 */ 38 39 #include <sys/cdefs.h> 40 __KERNEL_RCSID(0, "$NetBSD: ip_flow.c,v 1.23 2002/03/08 20:48:43 thorpej Exp $"); 41 42 #include <sys/param.h> 43 #include <sys/systm.h> 44 #include <sys/malloc.h> 45 #include <sys/mbuf.h> 46 #include <sys/domain.h> 47 #include <sys/protosw.h> 48 #include <sys/socket.h> 49 #include <sys/socketvar.h> 50 #include <sys/errno.h> 51 #include <sys/time.h> 52 #include <sys/kernel.h> 53 #include <sys/pool.h> 54 #include <sys/sysctl.h> 55 56 #include <net/if.h> 57 #include <net/if_dl.h> 58 #include <net/route.h> 59 #include <net/pfil.h> 60 61 #include <netinet/in.h> 62 #include <netinet/in_systm.h> 63 #include <netinet/ip.h> 64 #include <netinet/in_pcb.h> 65 #include <netinet/in_var.h> 66 #include <netinet/ip_var.h> 67 68 struct pool ipflow_pool; 69 70 LIST_HEAD(ipflowhead, ipflow); 71 72 #define IPFLOW_TIMER (5 * PR_SLOWHZ) 73 #define IPFLOW_HASHSIZE (1 << IPFLOW_HASHBITS) 74 75 static struct ipflowhead ipflowtable[IPFLOW_HASHSIZE]; 76 static struct ipflowhead ipflowlist; 77 static int ipflow_inuse; 78 79 #define IPFLOW_INSERT(bucket, ipf) \ 80 do { \ 81 LIST_INSERT_HEAD((bucket), (ipf), ipf_hash); \ 82 LIST_INSERT_HEAD(&ipflowlist, (ipf), ipf_list); \ 83 } while (0) 84 85 #define IPFLOW_REMOVE(ipf) \ 86 do { \ 87 LIST_REMOVE((ipf), ipf_hash); \ 88 LIST_REMOVE((ipf), ipf_list); \ 89 } while (0) 90 91 #ifndef IPFLOW_MAX 92 #define IPFLOW_MAX 256 93 #endif 94 int ip_maxflows = IPFLOW_MAX; 95 96 static unsigned 97 ipflow_hash( 98 struct in_addr dst, 99 struct in_addr src, 100 unsigned tos) 101 { 102 unsigned hash = tos; 103 int idx; 104 for (idx = 0; idx < 32; idx += IPFLOW_HASHBITS) 105 hash += (dst.s_addr >> (32 - idx)) + (src.s_addr >> idx); 106 return hash & (IPFLOW_HASHSIZE-1); 107 } 108 109 static struct ipflow * 110 ipflow_lookup( 111 const struct ip *ip) 112 { 113 unsigned hash; 114 struct ipflow *ipf; 115 116 hash = ipflow_hash(ip->ip_dst, ip->ip_src, ip->ip_tos); 117 118 ipf = LIST_FIRST(&ipflowtable[hash]); 119 while (ipf != NULL) { 120 if (ip->ip_dst.s_addr == ipf->ipf_dst.s_addr 121 && ip->ip_src.s_addr == ipf->ipf_src.s_addr 122 && ip->ip_tos == ipf->ipf_tos) 123 break; 124 ipf = LIST_NEXT(ipf, ipf_hash); 125 } 126 return ipf; 127 } 128 129 void 130 ipflow_init() 131 { 132 int i; 133 134 pool_init(&ipflow_pool, sizeof(struct ipflow), 0, 0, 0, "ipflowpl", 135 NULL); 136 137 LIST_INIT(&ipflowlist); 138 for (i = 0; i < IPFLOW_HASHSIZE; i++) 139 LIST_INIT(&ipflowtable[i]); 140 } 141 142 int 143 ipflow_fastforward( 144 struct mbuf *m) 145 { 146 struct ip *ip; 147 struct ipflow *ipf; 148 struct rtentry *rt; 149 struct sockaddr *dst; 150 int error; 151 int iplen; 152 153 /* 154 * Are we forwarding packets? Big enough for an IP packet? 155 */ 156 if (!ipforwarding || ipflow_inuse == 0 || m->m_len < sizeof(struct ip)) 157 return 0; 158 159 /* 160 * Was packet received as a link-level multicast or broadcast? 161 * If so, don't try to fast forward.. 162 */ 163 if ((m->m_flags & (M_BCAST|M_MCAST)) != 0) 164 return 0; 165 166 /* 167 * IP header with no option and valid version and length 168 */ 169 ip = mtod(m, struct ip *); 170 iplen = ntohs(ip->ip_len); 171 if (ip->ip_v != IPVERSION || ip->ip_hl != (sizeof(struct ip) >> 2) || 172 iplen < sizeof(struct ip) || iplen > m->m_pkthdr.len) 173 return 0; 174 /* 175 * Find a flow. 176 */ 177 if ((ipf = ipflow_lookup(ip)) == NULL) 178 return 0; 179 180 /* 181 * Verify the IP header checksum. 182 */ 183 switch (m->m_pkthdr.csum_flags & 184 ((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_IPv4) | 185 M_CSUM_IPv4_BAD)) { 186 case M_CSUM_IPv4|M_CSUM_IPv4_BAD: 187 return (0); 188 189 case M_CSUM_IPv4: 190 /* Checksum was okay. */ 191 break; 192 193 default: 194 /* Must compute it ourselves. */ 195 if (in_cksum(m, sizeof(struct ip)) != 0) 196 return (0); 197 break; 198 } 199 200 /* 201 * Route and interface still up? 202 */ 203 rt = ipf->ipf_ro.ro_rt; 204 if ((rt->rt_flags & RTF_UP) == 0 || 205 (rt->rt_ifp->if_flags & IFF_UP) == 0) 206 return 0; 207 208 /* 209 * Packet size OK? TTL? 210 */ 211 if (m->m_pkthdr.len > rt->rt_ifp->if_mtu || ip->ip_ttl <= IPTTLDEC) 212 return 0; 213 214 /* 215 * Clear any in-bound checksum flags for this packet. 216 */ 217 m->m_pkthdr.csum_flags = 0; 218 219 /* 220 * Everything checks out and so we can forward this packet. 221 * Modify the TTL and incrementally change the checksum. 222 * 223 * This method of adding the checksum works on either endian CPU. 224 * If htons() is inlined, all the arithmetic is folded; otherwise 225 * the htons()s are combined by CSE due to the __const__ attribute. 226 * 227 * Don't bother using HW checksumming here -- the incremental 228 * update is pretty fast. 229 */ 230 ip->ip_ttl -= IPTTLDEC; 231 if (ip->ip_sum >= (u_int16_t) ~htons(IPTTLDEC << 8)) 232 ip->ip_sum -= ~htons(IPTTLDEC << 8); 233 else 234 ip->ip_sum += htons(IPTTLDEC << 8); 235 236 /* 237 * Trim the packet in case it's too long.. 238 */ 239 if (m->m_pkthdr.len > iplen) { 240 if (m->m_len == m->m_pkthdr.len) { 241 m->m_len = iplen; 242 m->m_pkthdr.len = iplen; 243 } else 244 m_adj(m, iplen - m->m_pkthdr.len); 245 } 246 247 /* 248 * Send the packet on it's way. All we can get back is ENOBUFS 249 */ 250 ipf->ipf_uses++; 251 PRT_SLOW_ARM(ipf->ipf_timer, IPFLOW_TIMER); 252 253 if (rt->rt_flags & RTF_GATEWAY) 254 dst = rt->rt_gateway; 255 else 256 dst = &ipf->ipf_ro.ro_dst; 257 258 if ((error = (*rt->rt_ifp->if_output)(rt->rt_ifp, m, dst, rt)) != 0) { 259 if (error == ENOBUFS) 260 ipf->ipf_dropped++; 261 else 262 ipf->ipf_errors++; 263 } 264 return 1; 265 } 266 267 static void 268 ipflow_addstats( 269 struct ipflow *ipf) 270 { 271 ipf->ipf_ro.ro_rt->rt_use += ipf->ipf_uses; 272 ipstat.ips_cantforward += ipf->ipf_errors + ipf->ipf_dropped; 273 ipstat.ips_forward += ipf->ipf_uses; 274 ipstat.ips_fastforward += ipf->ipf_uses; 275 } 276 277 static void 278 ipflow_free( 279 struct ipflow *ipf) 280 { 281 int s; 282 /* 283 * Remove the flow from the hash table (at elevated IPL). 284 * Once it's off the list, we can deal with it at normal 285 * network IPL. 286 */ 287 s = splnet(); 288 IPFLOW_REMOVE(ipf); 289 splx(s); 290 ipflow_addstats(ipf); 291 RTFREE(ipf->ipf_ro.ro_rt); 292 ipflow_inuse--; 293 pool_put(&ipflow_pool, ipf); 294 } 295 296 struct ipflow * 297 ipflow_reap( 298 int just_one) 299 { 300 while (just_one || ipflow_inuse > ip_maxflows) { 301 struct ipflow *ipf, *maybe_ipf = NULL; 302 int s; 303 304 ipf = LIST_FIRST(&ipflowlist); 305 while (ipf != NULL) { 306 /* 307 * If this no longer points to a valid route 308 * reclaim it. 309 */ 310 if ((ipf->ipf_ro.ro_rt->rt_flags & RTF_UP) == 0) 311 goto done; 312 /* 313 * choose the one that's been least recently 314 * used or has had the least uses in the 315 * last 1.5 intervals. 316 */ 317 if (maybe_ipf == NULL || 318 ipf->ipf_timer < maybe_ipf->ipf_timer || 319 (ipf->ipf_timer == maybe_ipf->ipf_timer && 320 ipf->ipf_last_uses + ipf->ipf_uses < 321 maybe_ipf->ipf_last_uses + 322 maybe_ipf->ipf_uses)) 323 maybe_ipf = ipf; 324 ipf = LIST_NEXT(ipf, ipf_list); 325 } 326 ipf = maybe_ipf; 327 done: 328 /* 329 * Remove the entry from the flow table. 330 */ 331 s = splnet(); 332 IPFLOW_REMOVE(ipf); 333 splx(s); 334 ipflow_addstats(ipf); 335 RTFREE(ipf->ipf_ro.ro_rt); 336 if (just_one) 337 return ipf; 338 pool_put(&ipflow_pool, ipf); 339 ipflow_inuse--; 340 } 341 return NULL; 342 } 343 344 void 345 ipflow_slowtimo( 346 void) 347 { 348 struct ipflow *ipf, *next_ipf; 349 350 ipf = LIST_FIRST(&ipflowlist); 351 while (ipf != NULL) { 352 next_ipf = LIST_NEXT(ipf, ipf_list); 353 if (PRT_SLOW_ISEXPIRED(ipf->ipf_timer)) { 354 ipflow_free(ipf); 355 } else { 356 ipf->ipf_last_uses = ipf->ipf_uses; 357 ipf->ipf_ro.ro_rt->rt_use += ipf->ipf_uses; 358 ipstat.ips_forward += ipf->ipf_uses; 359 ipstat.ips_fastforward += ipf->ipf_uses; 360 ipf->ipf_uses = 0; 361 } 362 ipf = next_ipf; 363 } 364 } 365 366 void 367 ipflow_create( 368 const struct route *ro, 369 struct mbuf *m) 370 { 371 const struct ip *const ip = mtod(m, struct ip *); 372 struct ipflow *ipf; 373 unsigned hash; 374 int s; 375 376 /* 377 * Don't create cache entries for ICMP messages. 378 */ 379 if (ip_maxflows == 0 || ip->ip_p == IPPROTO_ICMP) 380 return; 381 /* 382 * See if an existing flow struct exists. If so remove it from it's 383 * list and free the old route. If not, try to malloc a new one 384 * (if we aren't at our limit). 385 */ 386 ipf = ipflow_lookup(ip); 387 if (ipf == NULL) { 388 if (ipflow_inuse >= ip_maxflows) { 389 ipf = ipflow_reap(1); 390 } else { 391 ipf = pool_get(&ipflow_pool, PR_NOWAIT); 392 if (ipf == NULL) 393 return; 394 ipflow_inuse++; 395 } 396 bzero((caddr_t) ipf, sizeof(*ipf)); 397 } else { 398 s = splnet(); 399 IPFLOW_REMOVE(ipf); 400 splx(s); 401 ipflow_addstats(ipf); 402 RTFREE(ipf->ipf_ro.ro_rt); 403 ipf->ipf_uses = ipf->ipf_last_uses = 0; 404 ipf->ipf_errors = ipf->ipf_dropped = 0; 405 } 406 407 /* 408 * Fill in the updated information. 409 */ 410 ipf->ipf_ro = *ro; 411 ro->ro_rt->rt_refcnt++; 412 ipf->ipf_dst = ip->ip_dst; 413 ipf->ipf_src = ip->ip_src; 414 ipf->ipf_tos = ip->ip_tos; 415 PRT_SLOW_ARM(ipf->ipf_timer, IPFLOW_TIMER); 416 ipf->ipf_start = time.tv_sec; 417 /* 418 * Insert into the approriate bucket of the flow table. 419 */ 420 hash = ipflow_hash(ip->ip_dst, ip->ip_src, ip->ip_tos); 421 s = splnet(); 422 IPFLOW_INSERT(&ipflowtable[hash], ipf); 423 splx(s); 424 } 425