Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

[dragonfly.git] / sys / netinet / ip_demux.c

/*
 * Copyright (c) 2003, 2004 Jeffrey M. Hsu.  All rights reserved.
 * Copyright (c) 2003, 2004 The DragonFly Project.  All rights reserved.
 *
 * This code is derived from software contributed to The DragonFly Project
 * by Jeffrey M. Hsu.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. 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.
 * 3. Neither the name of The DragonFly Project 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 HOLDERS 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.
 *
 * $DragonFly: src/sys/netinet/ip_demux.c,v 1.45 2008/11/11 10:46:58 sephe Exp $
 */

#include "opt_inet.h"
#include "opt_rss.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/thread.h>
#include <sys/sysctl.h>
#include <sys/globaldata.h>

#include <net/if.h>
#include <net/netisr.h>
#ifdef RSS
#include <net/toeplitz2.h>
#endif

#include <netinet/in_systm.h>
#include <netinet/in.h>
#include <netinet/in_var.h>
#include <netinet/in_pcb.h>
#include <netinet/ip.h>
#include <netinet/ip_var.h>
#include <netinet/tcp.h>
#include <netinet/tcpip.h>
#include <netinet/tcp_var.h>
#include <netinet/udp.h>
#include <netinet/udp_var.h>

extern struct thread netisr_cpu[];
extern int udp_mpsafe_thread;

static struct thread tcp_thread[MAXCPU];
static struct thread udp_thread[MAXCPU];

#ifndef RSS

static __inline int
INP_MPORT_HASH(in_addr_t faddr, in_addr_t laddr,
               in_port_t fport, in_port_t lport)
{
        /*
         * Use low order bytes.
         */

#if (BYTE_ORDER == LITTLE_ENDIAN)
        KASSERT(ncpus2 < 256, ("need different hash function"));  /* XXX JH */
        return (((faddr >> 24) ^ (fport >> 8) ^ (laddr >> 24) ^ (lport >> 8)) &
                ncpus2_mask);
#else
        return ((faddr ^ fport ^ laddr ^ lport) & ncpus2_mask);
#endif
}

#endif  /* !RSS */

static __inline int
INP_MPORT_HASH_UDP(in_addr_t faddr, in_addr_t laddr,
                   in_port_t fport, in_port_t lport)
{
#ifndef RSS
        return INP_MPORT_HASH(faddr, laddr, fport, lport);
#else
        return toeplitz_hash(toeplitz_rawhash_addr(faddr, laddr));
#endif
}

static __inline int
INP_MPORT_HASH_TCP(in_addr_t faddr, in_addr_t laddr,
                   in_port_t fport, in_port_t lport)
{
#ifndef RSS
        return INP_MPORT_HASH(faddr, laddr, fport, lport);
#else
        return toeplitz_hash(
               toeplitz_rawhash_addrport(faddr, laddr, fport, lport));
#endif
}

/*
 * If the packet is a valid IP datagram, upon returning of this function
 * following things are promised:
 *
 * o  IP header (including any possible IP options) is in one mbuf (m_len).
 * o  IP header length is not less than the minimum (sizeof(struct ip)).
 * o  IP total length is not less than IP header length.
 * o  IP datagram resides completely in the mbuf chain,
 *    i.e. pkthdr.len >= IP total length.
 *
 * If the packet is a UDP datagram,
 * o  IP header (including any possible IP options) and UDP header are in
 *    one mbuf (m_len).
 * o  IP total length is not less than (IP header length + UDP header length).
 *
 * If the packet is a TCP segment,
 * o  IP header (including any possible IP options) and TCP header (including
 *    any possible TCP options) are in one mbuf (m_len).
 * o  TCP header length is not less than the minimum (sizeof(struct tcphdr)).
 * o  IP total length is not less than (IP header length + TCP header length).
 */
boolean_t
ip_lengthcheck(struct mbuf **mp)
{
        struct mbuf *m = *mp;
        struct ip *ip;
        int iphlen, iplen;
        struct tcphdr *th;
        int thoff;                              /* TCP data offset */

        /* The packet must be at least the size of an IP header. */
        if (m->m_pkthdr.len < sizeof(struct ip)) {
                ipstat.ips_tooshort++;
                goto fail;
        }

        /* The fixed IP header must reside completely in the first mbuf. */
        if (m->m_len < sizeof(struct ip)) {
                m = m_pullup(m, sizeof(struct ip));
                if (m == NULL) {
                        ipstat.ips_toosmall++;
                        goto fail;
                }
        }

        ip = mtod(m, struct ip *);

        /* Bound check the packet's stated IP header length. */
        iphlen = ip->ip_hl << 2;
        if (iphlen < sizeof(struct ip)) {       /* minimum header length */
                ipstat.ips_badhlen++;
                goto fail;
        }

        /* The full IP header must reside completely in the one mbuf. */
        if (m->m_len < iphlen) {
                m = m_pullup(m, iphlen);
                if (m == NULL) {
                        ipstat.ips_badhlen++;
                        goto fail;
                }
                ip = mtod(m, struct ip *);
        }

        iplen = ntohs(ip->ip_len);

        /*
         * Check that the amount of data in the buffers is as
         * at least much as the IP header would have us expect.
         */
        if (m->m_pkthdr.len < iplen) {
                ipstat.ips_tooshort++;
                goto fail;
        }

        /*
         * Fragments other than the first fragment don't have much
         * length information.
         */
        if (ntohs(ip->ip_off) & IP_OFFMASK)
                goto ipcheckonly;

        /*
         * The TCP/IP or UDP/IP header must be entirely contained within
         * the first fragment of a packet.  Packet filters will break if they
         * aren't.
         *
         * Since the packet will be trimmed to ip_len we must also make sure
         * the potentially trimmed down length is still sufficient to hold
         * the header(s).
         */
        switch (ip->ip_p) {
        case IPPROTO_TCP:
                if (iplen < iphlen + sizeof(struct tcphdr)) {
                        ++tcpstat.tcps_rcvshort;
                        goto fail;
                }
                if (m->m_len < iphlen + sizeof(struct tcphdr)) {
                        m = m_pullup(m, iphlen + sizeof(struct tcphdr));
                        if (m == NULL) {
                                tcpstat.tcps_rcvshort++;
                                goto fail;
                        }
                        ip = mtod(m, struct ip *);
                }
                th = (struct tcphdr *)((caddr_t)ip + iphlen);
                thoff = th->th_off << 2;
                if (thoff < sizeof(struct tcphdr) ||
                    thoff + iphlen > ntohs(ip->ip_len)) {
                        tcpstat.tcps_rcvbadoff++;
                        goto fail;
                }
                if (m->m_len < iphlen + thoff) {
                        m = m_pullup(m, iphlen + thoff);
                        if (m == NULL) {
                                tcpstat.tcps_rcvshort++;
                                goto fail;
                        }
                }
                break;
        case IPPROTO_UDP:
                if (iplen < iphlen + sizeof(struct udphdr)) {
                        ++udpstat.udps_hdrops;
                        goto fail;
                }
                if (m->m_len < iphlen + sizeof(struct udphdr)) {
                        m = m_pullup(m, iphlen + sizeof(struct udphdr));
                        if (m == NULL) {
                                udpstat.udps_hdrops++;
                                goto fail;
                        }
                }
                break;
        default:
ipcheckonly:
                if (iplen < iphlen) {
                        ++ipstat.ips_badlen;
                        goto fail;
                }
                break;
        }

        m->m_flags |= M_LENCHECKED;
        *mp = m;
        return TRUE;

fail:
        if (m != NULL)
                m_freem(m);
        *mp = NULL;
        return FALSE;
}

/*
 * Map a packet to a protocol processing thread and return the thread's port.
 * If an error occurs, the passed mbuf will be freed, *mptr will be set
 * to NULL, and NULL will be returned.  If no error occurs, the passed mbuf
 * may be modified and a port pointer will be returned.
 */
lwkt_port_t
ip_mport(struct mbuf **mptr, int dir)
{
        struct ip *ip;
        int iphlen;
        struct tcphdr *th;
        struct udphdr *uh;
        struct mbuf *m;
        int thoff;                              /* TCP data offset */
        lwkt_port_t port;
        int cpu;

        if (!ip_lengthcheck(mptr))
                return (NULL);

        m = *mptr;
        ip = mtod(m, struct ip *);
        iphlen = ip->ip_hl << 2;

        /*
         * XXX generic packet handling defrag on CPU 0 for now.
         */
        if (ntohs(ip->ip_off) & (IP_MF | IP_OFFMASK)) {
                cpu = 0;
                port = &netisr_cpu[cpu].td_msgport;
                goto back;
        }

        switch (ip->ip_p) {
        case IPPROTO_TCP:
                th = (struct tcphdr *)((caddr_t)ip + iphlen);
                thoff = th->th_off << 2;
                cpu = INP_MPORT_HASH_TCP(ip->ip_src.s_addr, ip->ip_dst.s_addr,
                    th->th_sport, th->th_dport);
                port = &tcp_thread[cpu].td_msgport;
                break;

        case IPPROTO_UDP:
                uh = (struct udphdr *)((caddr_t)ip + iphlen);

#ifndef RSS
                if (IN_MULTICAST(ntohl(ip->ip_dst.s_addr)) ||
                    (dir == IP_MPORT_IN &&
                     in_broadcast(ip->ip_dst, m->m_pkthdr.rcvif))) {
                        cpu = 0;
                } else
#endif
                {
                        cpu = INP_MPORT_HASH_UDP(ip->ip_src.s_addr,
                            ip->ip_dst.s_addr, uh->uh_sport, uh->uh_dport);
                }
                port = &udp_thread[cpu].td_msgport;
                break;

        default:
                cpu = 0;
                port = &netisr_cpu[cpu].td_msgport;
                break;
        }
back:
        m->m_flags |= M_HASH;
        m->m_pkthdr.hash = cpu;
        return (port);
}

lwkt_port_t
ip_mport_in(struct mbuf **mptr)
{
        return ip_mport(mptr, IP_MPORT_IN);
}

/*
 * Map a packet to a protocol processing thread and return the thread's port.
 * Unlike ip_mport(), the packet content is not accessed.  The packet info
 * (pi) and the hash of the packet (m_pkthdr.hash) is used instead.  NULL is
 * returned if the packet info does not contain enough information.
 *
 * Caller has already made sure that m_pkthdr.hash is valid, i.e. m_flags
 * has M_HASH set.
 */
lwkt_port_t
ip_mport_pktinfo(const struct pktinfo *pi, struct mbuf *m)
{
        lwkt_port_t port;

        KASSERT(m->m_pkthdr.hash < ncpus2,
                ("invalid packet hash %#x\n", m->m_pkthdr.hash));

        /*
         * XXX generic packet handling defrag on CPU 0 for now.
         */
        if (pi->pi_flags & PKTINFO_FLAG_FRAG) {
                m->m_pkthdr.hash = 0;
                return &netisr_cpu[0].td_msgport;
        }

        switch (pi->pi_l3proto) {
        case IPPROTO_TCP:
                port = &tcp_thread[m->m_pkthdr.hash].td_msgport;
                break;

        case IPPROTO_UDP:
                port = &udp_thread[m->m_pkthdr.hash].td_msgport;
                break;

        default:
                port = NULL;
                break;
        }
        return port;
}

/*
 * Map a TCP socket to a protocol processing thread.
 */
lwkt_port_t
tcp_soport(struct socket *so, struct sockaddr *nam __unused,
           struct mbuf **dummy __unused, int req)
{
        struct inpcb *inp;

        /* The following processing all take place on Protocol Thread 0. */
        if (req == PRU_BIND || req == PRU_CONNECT || req == PRU_ATTACH ||
            req == PRU_LISTEN)
                return (&tcp_thread[0].td_msgport);

        inp = so->so_pcb;
        if (!inp)               /* connection reset by peer */
                return (&tcp_thread[0].td_msgport);

        /*
         * Already bound and connected or listening.  For TCP connections,
         * the (faddr, fport, laddr, lport) association cannot change now.
         *
         * Note: T/TCP code needs some reorganization to fit into
         * this model.  XXX JH
         *
         * Rely on type-stable memory and check in protocol handler
         * to fix race condition here w/ deallocation of inp.  XXX JH
         */
        return (&tcp_thread[INP_MPORT_HASH_TCP(inp->inp_faddr.s_addr,
            inp->inp_laddr.s_addr, inp->inp_fport, inp->inp_lport)].td_msgport);
}

/*
 * Used to route icmp messages to the proper protocol thread for ctlinput
 * operation.
 */
lwkt_port_t
tcp_ctlport(int cmd, struct sockaddr *sa, void *vip)
{
        struct ip *ip = vip;
        struct tcphdr *th;
        struct in_addr faddr;
        int cpu;

        faddr = ((struct sockaddr_in *)sa)->sin_addr;
        if (sa->sa_family != AF_INET || faddr.s_addr == INADDR_ANY)
                return(NULL);
        if (ip == NULL || PRC_IS_REDIRECT(cmd) || cmd == PRC_HOSTDEAD) {
                /*
                 * Message will be forwarded to all TCP protocol threads
                 * in following way:
                 *
                 * netisr0 (the msgport we return here)
                 *    |
                 *    |
                 *    | domsg <----------------------------+
                 *    |                                    |
                 *    |                                    | replymsg
                 *    |                                    |
                 *    V   forwardmsg         forwardmsg    |
                 *  tcp0 ------------> tcp1 ------------> tcpN
                 */
                return cpu0_ctlport(cmd, sa, vip);
        } else {
                th = (struct tcphdr *)((caddr_t)ip + (ip->ip_hl << 2));
                cpu = tcp_addrcpu(faddr.s_addr, th->th_dport,
                                  ip->ip_src.s_addr, th->th_sport);
        }
        return(&tcp_thread[cpu].td_msgport);
}

lwkt_port_t
tcp_addrport(in_addr_t faddr, in_port_t fport, in_addr_t laddr, in_port_t lport)
{
        return (&tcp_thread[tcp_addrcpu(faddr, fport,
                                        laddr, lport)].td_msgport);
}

/*
 * Map a UDP socket to a protocol processing thread.
 */
lwkt_port_t
udp_soport(struct socket *so, struct sockaddr *nam __unused,
           struct mbuf **dummy __unused, int req)
{
        struct inpcb *inp;

        /*
         * The following processing all take place on Protocol Thread 0:
         *   bind()
         *   attach() has a null socket parameter
         *   Fast and slow timeouts pass in null socket parameter
         */
        if (req == PRU_BIND || so == NULL)
                return (&udp_thread[0].td_msgport);

        inp = so->so_pcb;

#ifndef RSS
        if (IN_MULTICAST(ntohl(inp->inp_laddr.s_addr)))
                return (&udp_thread[0].td_msgport);
#endif

        /*
         * Rely on type-stable memory and check in protocol handler
         * to fix race condition here w/ deallocation of inp.  XXX JH
         */

        return (&udp_thread[INP_MPORT_HASH_UDP(inp->inp_faddr.s_addr,
            inp->inp_laddr.s_addr, inp->inp_fport, inp->inp_lport)].td_msgport);
}

/*
 * Used to route icmp messages to the proper protocol thread for ctlinput
 * operation.
 */
lwkt_port_t
udp_ctlport(int cmd, struct sockaddr *sa, void *vip)
{
        struct ip *ip = vip;
        struct udphdr *uh;
        struct in_addr faddr;
        int cpu;

        faddr = ((struct sockaddr_in *)sa)->sin_addr;
        if (sa->sa_family != AF_INET || faddr.s_addr == INADDR_ANY)
                return(NULL);
        if (PRC_IS_REDIRECT(cmd)) {
                /*
                 * See the comment in tcp_ctlport; the only difference
                 * is that message is forwarded to UDP protocol theads.
                 */
                return cpu0_ctlport(cmd, sa, vip);
        } else if (ip == NULL || cmd == PRC_HOSTDEAD) {
                /*
                 * XXX
                 * Once UDP inpcbs are CPU localized, we should do
                 * the same forwarding as PRC_IS_REDIRECT(cmd)
                 */
                cpu = 0;
        } else {
                uh = (struct udphdr *)((caddr_t)ip + (ip->ip_hl << 2));

                cpu = INP_MPORT_HASH_UDP(faddr.s_addr, ip->ip_src.s_addr,
                                         uh->uh_dport, uh->uh_sport);
        }
        return (&udp_thread[cpu].td_msgport);
}

/*
 * Map a network address to a processor.
 */
int
tcp_addrcpu(in_addr_t faddr, in_port_t fport, in_addr_t laddr, in_port_t lport)
{
        return (INP_MPORT_HASH_TCP(faddr, laddr, fport, lport));
}

int
udp_addrcpu(in_addr_t faddr, in_port_t fport, in_addr_t laddr, in_port_t lport)
{
#ifndef RSS
        if (IN_MULTICAST(ntohl(laddr)))
                return (0);
        else
#endif
                return (INP_MPORT_HASH_UDP(faddr, laddr, fport, lport));
}

/*
 * Return LWKT port for cpu.
 */
lwkt_port_t
tcp_cport(int cpu)
{
        return (&tcp_thread[cpu].td_msgport);
}

lwkt_port_t
udp_cport(int cpu)
{
        return (&udp_thread[cpu].td_msgport);
}

void
tcp_thread_init(void)
{
        int cpu;

        for (cpu = 0; cpu < ncpus2; cpu++) {
                lwkt_create(tcpmsg_service_loop, NULL, NULL,
                            &tcp_thread[cpu], TDF_NETWORK | TDF_MPSAFE, cpu,
                            "tcp_thread %d", cpu);
                netmsg_service_port_init(&tcp_thread[cpu].td_msgport);
        }
}

void
udp_thread_init(void)
{
        int cpu;

        for (cpu = 0; cpu < ncpus2; cpu++) {
                lwkt_create(netmsg_service_loop, &udp_mpsafe_thread, NULL,
                            &udp_thread[cpu], TDF_NETWORK | TDF_MPSAFE, cpu,
                            "udp_thread %d", cpu);
                netmsg_service_port_init(&udp_thread[cpu].td_msgport);
        }
}