Copied libratbox and related stuff from shadowircd upstream.

[irc/rqf/shadowircd.git] / libratbox / src / commio.c

/*
 *  ircd-ratbox: A slightly useful ircd.
 *  commio.c: Network/file related functions
 *
 *  Copyright (C) 1990 Jarkko Oikarinen and University of Oulu, Co Center
 *  Copyright (C) 1996-2002 Hybrid Development Team
 *  Copyright (C) 2002-2005 ircd-ratbox development team
 *
 *  This program is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation; either version 2 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with this program; if not, write to the Free Software
 *  Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301
 *  USA
 *
 *  $Id: commio.c 26096 2008-09-20 01:05:42Z androsyn $
 */
#include <libratbox_config.h>
#include <ratbox_lib.h>
#include <commio-int.h>
#include <commio-ssl.h>
#include <event-int.h>
#ifdef HAVE_SYS_UIO_H
#include <sys/uio.h>
#endif
#define HAVE_SSL 1

#ifndef MSG_NOSIGNAL
#define MSG_NOSIGNAL 0
#endif


struct timeout_data
{
        rb_fde_t *F;
        rb_dlink_node node;
        time_t timeout;
        PF *timeout_handler;
        void *timeout_data;
};

rb_dlink_list *rb_fd_table;
static rb_bh *fd_heap;

static rb_dlink_list timeout_list;
static rb_dlink_list closed_list;

static struct ev_entry *rb_timeout_ev;


static const char *rb_err_str[] = { "Comm OK", "Error during bind()",
        "Error during DNS lookup", "connect timeout",
        "Error during connect()",
        "Comm Error"
};

/* Highest FD and number of open FDs .. */
static int number_fd = 0;
int rb_maxconnections = 0;

static PF rb_connect_timeout;
static PF rb_connect_tryconnect;
#ifdef RB_IPV6
static void mangle_mapped_sockaddr(struct sockaddr *in);
#endif

#ifndef HAVE_SOCKETPAIR
static int rb_inet_socketpair(int d, int type, int protocol, int sv[2]);
static int rb_inet_socketpair_udp(rb_fde_t **newF1, rb_fde_t **newF2);
#endif

static inline rb_fde_t *
add_fd(int fd)
{
        rb_fde_t *F = rb_find_fd(fd);

        /* look up to see if we have it already */
        if(F != NULL)
                return F;

        F = rb_bh_alloc(fd_heap);
        F->fd = fd;
        rb_dlinkAdd(F, &F->node, &rb_fd_table[rb_hash_fd(fd)]);
        return (F);
}

static inline void
remove_fd(rb_fde_t *F)
{
        if(F == NULL || !IsFDOpen(F))
                return;

        rb_dlinkMoveNode(&F->node, &rb_fd_table[rb_hash_fd(F->fd)], &closed_list);
}

static void
free_fds(void)
{
        rb_fde_t *F;
        rb_dlink_node *ptr, *next;
        RB_DLINK_FOREACH_SAFE(ptr, next, closed_list.head)
        {
                F = ptr->data;
                rb_dlinkDelete(ptr, &closed_list);
                rb_bh_free(fd_heap, F);
        }
}

/* 32bit solaris is kinda slow and stdio only supports fds < 256
 * so we got to do this crap below.
 * (BTW Fuck you Sun, I hate your guts and I hope you go bankrupt soon)
 */

#if defined (__SVR4) && defined (__sun)
static void
rb_fd_hack(int *fd)
{
        int newfd;
        if(*fd > 256 || *fd < 0)
                return;
        if((newfd = fcntl(*fd, F_DUPFD, 256)) != -1)
        {
                close(*fd);
                *fd = newfd;
        }
        return;
}
#else
#define rb_fd_hack(fd)
#endif


/* close_all_connections() can be used *before* the system come up! */

static void
rb_close_all(void)
{
#ifndef _WIN32
        int i;

        /* XXX someone tell me why we care about 4 fd's ? */
        /* XXX btw, fd 3 is used for profiler ! */
        for(i = 3; i < rb_maxconnections; ++i)
        {
                close(i);
        }
#endif
}

/*
 * get_sockerr - get the error value from the socket or the current errno
 *
 * Get the *real* error from the socket (well try to anyway..).
 * This may only work when SO_DEBUG is enabled but its worth the
 * gamble anyway.
 */
int
rb_get_sockerr(rb_fde_t *F)
{
        int errtmp;
        int err = 0;
        rb_socklen_t len = sizeof(err);

        if(!(F->type & RB_FD_SOCKET))
                return errno;

        rb_get_errno();
        errtmp = errno;

#ifdef SO_ERROR
        if(F != NULL
           && !getsockopt(rb_get_fd(F), SOL_SOCKET, SO_ERROR, (char *)&err, (rb_socklen_t *) & len))
        {
                if(err)
                        errtmp = err;
        }
        errno = errtmp;
#endif
        return errtmp;
}

/*
 * rb_getmaxconnect - return the max number of connections allowed
 */
int
rb_getmaxconnect(void)
{
        return (rb_maxconnections);
}

/*
 * set_sock_buffers - set send and receive buffers for socket
 * 
 * inputs       - fd file descriptor
 *              - size to set
 * output       - returns true (1) if successful, false (0) otherwise
 * side effects -
 */
int
rb_set_buffers(rb_fde_t *F, int size)
{
        if(F == NULL)
                return 0;
        if(setsockopt
           (F->fd, SOL_SOCKET, SO_RCVBUF, (char *)&size, sizeof(size))
           || setsockopt(F->fd, SOL_SOCKET, SO_SNDBUF, (char *)&size, sizeof(size)))
                return 0;
        return 1;
}

/*
 * set_non_blocking - Set the client connection into non-blocking mode. 
 *
 * inputs       - fd to set into non blocking mode
 * output       - 1 if successful 0 if not
 * side effects - use POSIX compliant non blocking and
 *                be done with it.
 */
int
rb_set_nb(rb_fde_t *F)
{
        int nonb = 0;
        int res;
        int fd;
        if(F == NULL)
                return 0;
        fd = F->fd;

        if((res = rb_setup_fd(F)))
                return res;
#ifdef O_NONBLOCK
        nonb |= O_NONBLOCK;
        res = fcntl(fd, F_GETFL, 0);
        if(-1 == res || fcntl(fd, F_SETFL, res | nonb) == -1)
                return 0;
#else
        nonb = 1;
        res = 0;
        if(ioctl(fd, FIONBIO, (char *)&nonb) == -1)
                return 0;
#endif

        return 1;
}

/*
 * rb_settimeout() - set the socket timeout
 *
 * Set the timeout for the fd
 */
void
rb_settimeout(rb_fde_t *F, time_t timeout, PF * callback, void *cbdata)
{
        struct timeout_data *td;

        if(F == NULL)
                return;

        lrb_assert(IsFDOpen(F));
        td = F->timeout;
        if(callback == NULL)    /* user wants to remove */
        {
                if(td == NULL)
                        return;
                rb_dlinkDelete(&td->node, &timeout_list);
                rb_free(td);
                F->timeout = NULL;
                if(rb_dlink_list_length(&timeout_list) == 0)
                {
                        rb_event_delete(rb_timeout_ev);
                        rb_timeout_ev = NULL;
                }
                return;
        }

        if(F->timeout == NULL)
                td = F->timeout = rb_malloc(sizeof(struct timeout_data));

        td->F = F;
        td->timeout = rb_current_time() + timeout;
        td->timeout_handler = callback;
        td->timeout_data = cbdata;
        rb_dlinkAdd(td, &td->node, &timeout_list);
        if(rb_timeout_ev == NULL)
        {
                rb_timeout_ev = rb_event_add("rb_checktimeouts", rb_checktimeouts, NULL, 5);
        }
}

/*
 * rb_checktimeouts() - check the socket timeouts
 *
 * All this routine does is call the given callback/cbdata, without closing
 * down the file descriptor. When close handlers have been implemented,
 * this will happen.
 */
void
rb_checktimeouts(void *notused)
{
        rb_dlink_node *ptr, *next;
        struct timeout_data *td;
        rb_fde_t *F;
        PF *hdl;
        void *data;

        RB_DLINK_FOREACH_SAFE(ptr, next, timeout_list.head)
        {
                td = ptr->data;
                F = td->F;
                if(F == NULL || !IsFDOpen(F))
                        continue;

                if(td->timeout < rb_current_time())
                {
                        hdl = td->timeout_handler;
                        data = td->timeout_data;
                        rb_dlinkDelete(&td->node, &timeout_list);
                        F->timeout = NULL;
                        rb_free(td);
                        hdl(F, data);
                }
        }
}

static void
rb_accept_tryaccept(rb_fde_t *F, void *data)
{
        struct rb_sockaddr_storage st;
        rb_fde_t *new_F;
        rb_socklen_t addrlen = sizeof(st);
        int new_fd;

        while(1)
        {
                new_fd = accept(F->fd, (struct sockaddr *)&st, &addrlen);
                rb_get_errno();
                if(new_fd < 0)
                {
                        rb_setselect(F, RB_SELECT_ACCEPT, rb_accept_tryaccept, NULL);
                        return;
                }

                rb_fd_hack(&new_fd);

                new_F = rb_open(new_fd, RB_FD_SOCKET, "Incoming Connection");

                if(new_F == NULL)
                {
                        rb_lib_log
                                ("rb_accept: new_F == NULL on incoming connection. Closing new_fd == %d\n",
                                 new_fd);
                        close(new_fd);
                        continue;
                }

                if(rb_unlikely(!rb_set_nb(new_F)))
                {
                        rb_get_errno();
                        rb_lib_log("rb_accept: Couldn't set FD %d non blocking!", new_F->fd);
                        rb_close(new_F);
                }

#ifdef RB_IPV6
                mangle_mapped_sockaddr((struct sockaddr *)&st);
#endif

                if(F->accept->precb != NULL)
                {
                        if(!F->accept->precb(new_F, (struct sockaddr *)&st, addrlen, F->accept->data))  /* pre-callback decided to drop it */
                                continue;
                }
#ifdef HAVE_SSL
                if(F->type & RB_FD_SSL)
                {
                        rb_ssl_accept_setup(F, new_F, (struct sockaddr *)&st, addrlen);
                }
                else
#endif /* HAVE_SSL */
                {
                        F->accept->callback(new_F, RB_OK, (struct sockaddr *)&st, addrlen,
                                            F->accept->data);
                }
        }

}

/* try to accept a TCP connection */
void
rb_accept_tcp(rb_fde_t *F, ACPRE * precb, ACCB * callback, void *data)
{
        if(F == NULL)
                return;
        lrb_assert(callback);

        F->accept = rb_malloc(sizeof(struct acceptdata));
        F->accept->callback = callback;
        F->accept->data = data;
        F->accept->precb = precb;
        rb_accept_tryaccept(F, NULL);
}

/*
 * void rb_connect_tcp(int fd, struct sockaddr *dest,
 *                       struct sockaddr *clocal, int socklen,
 *                       CNCB *callback, void *data, int timeout)
 * Input: An fd to connect with, a host and port to connect to,
 *        a local sockaddr to connect from + length(or NULL to use the
 *        default), a callback, the data to pass into the callback, the
 *        address family.
 * Output: None.
 * Side-effects: A non-blocking connection to the host is started, and
 *               if necessary, set up for selection. The callback given
 *               may be called now, or it may be called later.
 */
void
rb_connect_tcp(rb_fde_t *F, struct sockaddr *dest,
               struct sockaddr *clocal, int socklen, CNCB * callback, void *data, int timeout)
{
        if(F == NULL)
                return;

        lrb_assert(callback);
        F->connect = rb_malloc(sizeof(struct conndata));
        F->connect->callback = callback;
        F->connect->data = data;

        memcpy(&F->connect->hostaddr, dest, sizeof(F->connect->hostaddr));

        /* Note that we're using a passed sockaddr here. This is because
         * generally you'll be bind()ing to a sockaddr grabbed from
         * getsockname(), so this makes things easier.
         * XXX If NULL is passed as local, we should later on bind() to the
         * virtual host IP, for completeness.
         *   -- adrian
         */
        if((clocal != NULL) && (bind(F->fd, clocal, socklen) < 0))
        {
                /* Failure, call the callback with RB_ERR_BIND */
                rb_connect_callback(F, RB_ERR_BIND);
                /* ... and quit */
                return;
        }

        /* We have a valid IP, so we just call tryconnect */
        /* Make sure we actually set the timeout here .. */
        rb_settimeout(F, timeout, rb_connect_timeout, NULL);
        rb_connect_tryconnect(F, NULL);
}


/*
 * rb_connect_callback() - call the callback, and continue with life
 */
void
rb_connect_callback(rb_fde_t *F, int status)
{
        CNCB *hdl;
        void *data;
        int errtmp = errno;     /* save errno as rb_settimeout clobbers it sometimes */

        /* This check is gross..but probably necessary */
        if(F == NULL || F->connect == NULL || F->connect->callback == NULL)
                return;
        /* Clear the connect flag + handler */
        hdl = F->connect->callback;
        data = F->connect->data;
        F->connect->callback = NULL;


        /* Clear the timeout handler */
        rb_settimeout(F, 0, NULL, NULL);
        errno = errtmp;
        /* Call the handler */
        hdl(F, status, data);
}


/*
 * rb_connect_timeout() - this gets called when the socket connection
 * times out. This *only* can be called once connect() is initially
 * called ..
 */
static void
rb_connect_timeout(rb_fde_t *F, void *notused)
{
        /* error! */
        rb_connect_callback(F, RB_ERR_TIMEOUT);
}

/* static void rb_connect_tryconnect(int fd, void *notused)
 * Input: The fd, the handler data(unused).
 * Output: None.
 * Side-effects: Try and connect with pending connect data for the FD. If
 *               we succeed or get a fatal error, call the callback.
 *               Otherwise, it is still blocking or something, so register
 *               to select for a write event on this FD.
 */
static void
rb_connect_tryconnect(rb_fde_t *F, void *notused)
{
        int retval;

        if(F == NULL || F->connect == NULL || F->connect->callback == NULL)
                return;
        /* Try the connect() */
        retval = connect(F->fd,
                         (struct sockaddr *)&F->connect->hostaddr,
                         GET_SS_LEN(&F->connect->hostaddr));
        /* Error? */
        if(retval < 0)
        {
                /*
                 * If we get EISCONN, then we've already connect()ed the socket,
                 * which is a good thing.
                 *   -- adrian
                 */
                rb_get_errno();
                if(errno == EISCONN)
                        rb_connect_callback(F, RB_OK);
                else if(rb_ignore_errno(errno))
                        /* Ignore error? Reschedule */
                        rb_setselect(F, RB_SELECT_CONNECT, rb_connect_tryconnect, NULL);
                else
                        /* Error? Fail with RB_ERR_CONNECT */
                        rb_connect_callback(F, RB_ERR_CONNECT);
                return;
        }
        /* If we get here, we've suceeded, so call with RB_OK */
        rb_connect_callback(F, RB_OK);
}


int
rb_connect_sockaddr(rb_fde_t *F, struct sockaddr *addr, int len)
{
        if(F == NULL)
                return 0;

        memcpy(addr, &F->connect->hostaddr, len);
        return 1;
}

/*
 * rb_error_str() - return an error string for the given error condition
 */
const char *
rb_errstr(int error)
{
        if(error < 0 || error >= RB_ERR_MAX)
                return "Invalid error number!";
        return rb_err_str[error];
}


int
rb_socketpair(int family, int sock_type, int proto, rb_fde_t **F1, rb_fde_t **F2, const char *note)
{
        int nfd[2];
        if(number_fd >= rb_maxconnections)
        {
                errno = ENFILE;
                return -1;
        }

#ifdef HAVE_SOCKETPAIR
        if(socketpair(family, sock_type, proto, nfd))
#else
        if(sock_type == SOCK_DGRAM)
        {
                return rb_inet_socketpair_udp(F1, F2);
        }

        if(rb_inet_socketpair(AF_INET, sock_type, proto, nfd))
#endif
                return -1;

        rb_fd_hack(&nfd[0]);
        rb_fd_hack(&nfd[1]);

        *F1 = rb_open(nfd[0], RB_FD_SOCKET, note);
        *F2 = rb_open(nfd[1], RB_FD_SOCKET, note);

        if(*F1 == NULL)
        {
                if(*F2 != NULL)
                        rb_close(*F2);
                return -1;
        }

        if(*F2 == NULL)
        {
                rb_close(*F1);
                return -1;
        }

        /* Set the socket non-blocking, and other wonderful bits */
        if(rb_unlikely(!rb_set_nb(*F1)))
        {
                rb_lib_log("rb_open: Couldn't set FD %d non blocking: %s", nfd[0], strerror(errno));
                rb_close(*F1);
                rb_close(*F2);
                return -1;
        }

        if(rb_unlikely(!rb_set_nb(*F2)))
        {
                rb_lib_log("rb_open: Couldn't set FD %d non blocking: %s", nfd[1], strerror(errno));
                rb_close(*F1);
                rb_close(*F2);
                return -1;
        }

        return 0;
}


int
rb_pipe(rb_fde_t **F1, rb_fde_t **F2, const char *desc)
{
#ifndef _WIN32
        int fd[2];
        if(number_fd >= rb_maxconnections)
        {
                errno = ENFILE;
                return -1;
        }
        if(pipe(fd) == -1)
                return -1;
        rb_fd_hack(&fd[0]);
        rb_fd_hack(&fd[1]);
        *F1 = rb_open(fd[0], RB_FD_PIPE, desc);
        *F2 = rb_open(fd[1], RB_FD_PIPE, desc);

        if(rb_unlikely(!rb_set_nb(*F1)))
        {
                rb_lib_log("rb_open: Couldn't set FD %d non blocking: %s", fd[0], strerror(errno));
                rb_close(*F1);
                rb_close(*F2);
                return -1;
        }

        if(rb_unlikely(!rb_set_nb(*F2)))
        {
                rb_lib_log("rb_open: Couldn't set FD %d non blocking: %s", fd[1], strerror(errno));
                rb_close(*F1);
                rb_close(*F2);
                return -1;
        }


        return 0;
#else
        /* Its not a pipe..but its selectable.  I'll take dirty hacks
         * for $500 Alex.
         */
        return rb_socketpair(AF_INET, SOCK_STREAM, 0, F1, F2, desc);
#endif
}

/*
 * rb_socket() - open a socket
 *
 * This is a highly highly cut down version of squid's rb_open() which
 * for the most part emulates socket(), *EXCEPT* it fails if we're about
 * to run out of file descriptors.
 */
rb_fde_t *
rb_socket(int family, int sock_type, int proto, const char *note)
{
        rb_fde_t *F;
        int fd;
        /* First, make sure we aren't going to run out of file descriptors */
        if(rb_unlikely(number_fd >= rb_maxconnections))
        {
                errno = ENFILE;
                return NULL;
        }

        /*
         * Next, we try to open the socket. We *should* drop the reserved FD
         * limit if/when we get an error, but we can deal with that later.
         * XXX !!! -- adrian
         */
        fd = socket(family, sock_type, proto);
        rb_fd_hack(&fd);
        if(rb_unlikely(fd < 0))
                return NULL;    /* errno will be passed through, yay.. */

#if defined(RB_IPV6) && defined(IPV6_V6ONLY)
        /* 
         * Make sure we can take both IPv4 and IPv6 connections
         * on an AF_INET6 socket
         */
        if(family == AF_INET6)
        {
                int off = 1;
                if(setsockopt(fd, IPPROTO_IPV6, IPV6_V6ONLY, &off, sizeof(off)) == -1)
                {
                        rb_lib_log("rb_socket: Could not set IPV6_V6ONLY option to 1 on FD %d: %s",
                                   fd, strerror(errno));
                        close(fd);
                        return NULL;
                }
        }
#endif

        F = rb_open(fd, RB_FD_SOCKET, note);
        if(F == NULL)
        {
                rb_lib_log("rb_socket: rb_open returns NULL on FD %d: %s, closing fd", fd,
                           strerror(errno));
                close(fd);
                return NULL;
        }
        /* Set the socket non-blocking, and other wonderful bits */
        if(rb_unlikely(!rb_set_nb(F)))
        {
                rb_lib_log("rb_open: Couldn't set FD %d non blocking: %s", fd, strerror(errno));
                rb_close(F);
                return NULL;
        }

        return F;
}

/*
 * If a sockaddr_storage is AF_INET6 but is a mapped IPv4
 * socket manged the sockaddr.
 */
#ifdef RB_IPV6
static void
mangle_mapped_sockaddr(struct sockaddr *in)
{
        struct sockaddr_in6 *in6 = (struct sockaddr_in6 *)in;

        if(in->sa_family == AF_INET)
                return;

        if(in->sa_family == AF_INET6 && IN6_IS_ADDR_V4MAPPED(&in6->sin6_addr))
        {
                struct sockaddr_in in4;
                memset(&in4, 0, sizeof(struct sockaddr_in));
                in4.sin_family = AF_INET;
                in4.sin_port = in6->sin6_port;
                in4.sin_addr.s_addr = ((uint32_t *)&in6->sin6_addr)[3];
                memcpy(in, &in4, sizeof(struct sockaddr_in));
        }
        return;
}
#endif

/*
 * rb_listen() - listen on a port
 */
int
rb_listen(rb_fde_t *F, int backlog)
{
        F->type = RB_FD_SOCKET | RB_FD_LISTEN;
        /* Currently just a simple wrapper for the sake of being complete */
        return listen(F->fd, backlog);
}

void
rb_fdlist_init(int closeall, int maxfds, size_t heapsize)
{
        static int initialized = 0;
#ifdef _WIN32
        WSADATA wsaData;
        int err;
        int vers = MAKEWORD(2, 0);

        err = WSAStartup(vers, &wsaData);
        if(err != 0)
        {
                rb_lib_die("WSAStartup failed");
        }

#endif
        if(!initialized)
        {
                rb_maxconnections = maxfds;
                if(closeall)
                        rb_close_all();
                /* Since we're doing this once .. */
                initialized = 1;
        }
        fd_heap = rb_bh_create(sizeof(rb_fde_t), heapsize, "librb_fd_heap");

}


/* Called to open a given filedescriptor */
rb_fde_t *
rb_open(int fd, uint8_t type, const char *desc)
{
        rb_fde_t *F;
        lrb_assert(fd >= 0);

        F = add_fd(fd);

        lrb_assert(!IsFDOpen(F));
        if(rb_unlikely(IsFDOpen(F)))
        {
                const char *fdesc;
                if(F != NULL && F->desc != NULL)
                        fdesc = F->desc;
                else
                        fdesc = "NULL";
                rb_lib_log("Trying to rb_open an already open FD: %d desc: %s", fd, fdesc);
                return NULL;
        }
        F->fd = fd;
        F->type = type;
        SetFDOpen(F);

        if(desc != NULL)
                F->desc = rb_strndup(desc, FD_DESC_SZ);
        number_fd++;
        return F;
}


/* Called to close a given filedescriptor */
void
rb_close(rb_fde_t *F)
{
        int type, fd;

        if(F == NULL)
                return;

        fd = F->fd;
        type = F->type;
        lrb_assert(IsFDOpen(F));

        lrb_assert(!(type & RB_FD_FILE));
        if(rb_unlikely(type & RB_FD_FILE))
        {
                lrb_assert(F->read_handler == NULL);
                lrb_assert(F->write_handler == NULL);
        }
        rb_setselect(F, RB_SELECT_WRITE | RB_SELECT_READ, NULL, NULL);
        rb_settimeout(F, 0, NULL, NULL);
        rb_free(F->accept);
        rb_free(F->connect);
        rb_free(F->desc);
#ifdef HAVE_SSL
        if(type & RB_FD_SSL)
        {
                rb_ssl_shutdown(F);
        }
#endif /* HAVE_SSL */
        if(IsFDOpen(F))
        {
                remove_fd(F);
                ClearFDOpen(F);
        }

        number_fd--;

#ifdef _WIN32
        if(type & (RB_FD_SOCKET | RB_FD_PIPE))
        {
                closesocket(fd);
                return;
        }
        else
#endif
                close(fd);
}


/*
 * rb_dump_fd() - dump the list of active filedescriptors
 */
void
rb_dump_fd(DUMPCB * cb, void *data)
{
        static const char *empty = "";
        rb_dlink_node *ptr;
        rb_dlink_list *bucket;
        rb_fde_t *F;
        unsigned int i;

        for(i = 0; i < RB_FD_HASH_SIZE; i++)
        {
                bucket = &rb_fd_table[i];

                if(rb_dlink_list_length(bucket) <= 0)
                        continue;

                RB_DLINK_FOREACH(ptr, bucket->head)
                {
                        F = ptr->data;
                        if(F == NULL || !IsFDOpen(F))
                                continue;

                        cb(F->fd, F->desc ? F->desc : empty, data);
                }
        }
}

/*
 * rb_note() - set the fd note
 *
 * Note: must be careful not to overflow rb_fd_table[fd].desc when
 *       calling.
 */
void
rb_note(rb_fde_t *F, const char *string)
{
        if(F == NULL)
                return;

        rb_free(F->desc);
        F->desc = rb_strndup(string, FD_DESC_SZ);
}

void
rb_set_type(rb_fde_t *F, uint8_t type)
{
        /* if the caller is calling this, lets assume they have a clue */
        F->type = type;
        return;
}

uint8_t
rb_get_type(rb_fde_t *F)
{
        return F->type;
}

int
rb_fd_ssl(rb_fde_t *F)
{
        if(F == NULL)
                return 0;
        if(F->type & RB_FD_SSL)
                return 1;
        return 0;
}

int
rb_get_fd(rb_fde_t *F)
{
        if(F == NULL)
                return -1;
        return (F->fd);
}

rb_fde_t *
rb_get_fde(int fd)
{
        return rb_find_fd(fd);
}

ssize_t
rb_read(rb_fde_t *F, void *buf, int count)
{
        ssize_t ret;
        if(F == NULL)
                return 0;

        /* This needs to be *before* RB_FD_SOCKET otherwise you'll process 
         * an SSL socket as a regular socket 
         */
#ifdef HAVE_SSL
        if(F->type & RB_FD_SSL)
        {
                return rb_ssl_read(F, buf, count);
        }
#endif
        if(F->type & RB_FD_SOCKET)
        {
                ret = recv(F->fd, buf, count, 0);
                if(ret < 0)
                {
                        rb_get_errno();
                }
                return ret;
        }


        /* default case */
        return read(F->fd, buf, count);
}


ssize_t
rb_write(rb_fde_t *F, const void *buf, int count)
{
        ssize_t ret;
        if(F == NULL)
                return 0;

#ifdef HAVE_SSL
        if(F->type & RB_FD_SSL)
        {
                return rb_ssl_write(F, buf, count);
        }
#endif
        if(F->type & RB_FD_SOCKET)
        {
                ret = send(F->fd, buf, count, MSG_NOSIGNAL);
                if(ret < 0)
                {
                        rb_get_errno();
                }
                return ret;
        }

        return write(F->fd, buf, count);
}

#if defined(HAVE_SSL) || defined(WIN32) || !defined(HAVE_WRITEV)
static ssize_t
rb_fake_writev(rb_fde_t *F, const struct rb_iovec *vp, size_t vpcount)
{
        ssize_t count = 0;

        while(vpcount-- > 0)
        {
                ssize_t written = rb_write(F, vp->iov_base, vp->iov_len);

                if(written <= 0)
                {
                        if(count > 0)
                                return count;
                        else
                                return written;
                }
                count += written;
                vp++;
        }
        return (count);
}
#endif

#if defined(WIN32) || !defined(HAVE_WRITEV)
ssize_t
rb_writev(rb_fde_t *F, struct rb_iovec * vecount, int count)
{
        return rb_fake_writev(F, vecount, count);
}

#else
ssize_t
rb_writev(rb_fde_t *F, struct rb_iovec * vector, int count)
{
        if(F == NULL)
        {
                errno = EBADF;
                return -1;
        }
#ifdef HAVE_SSL
        if(F->type & RB_FD_SSL)
        {
                return rb_fake_writev(F, vector, count);
        }
#endif /* HAVE_SSL */
#ifdef HAVE_SENDMSG
        if(F->type & RB_FD_SOCKET)
        {
                struct msghdr msg;
                memset(&msg, 0, sizeof(msg));
                msg.msg_iov = (struct iovec *)vector;
                msg.msg_iovlen = count;
                return sendmsg(F->fd, &msg, MSG_NOSIGNAL);
        }
#endif /* HAVE_SENDMSG */
        return writev(F->fd, (struct iovec *)vector, count);

}
#endif


/* 
 * From: Thomas Helvey <tomh@inxpress.net>
 */
static const char *IpQuadTab[] = {
        "0", "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"
};

/*
 * inetntoa - in_addr to string
 *      changed name to remove collision possibility and
 *      so behaviour is guaranteed to take a pointer arg.
 *      -avalon 23/11/92
 *  inet_ntoa --  returned the dotted notation of a given
 *      internet number
 *      argv 11/90).
 *  inet_ntoa --  its broken on some Ultrix/Dynix too. -avalon
 */

static const char *
inetntoa(const char *in)
{
        static char buf[16];
        char *bufptr = buf;
        const unsigned char *a = (const unsigned char *)in;
        const char *n;

        n = IpQuadTab[*a++];
        while(*n)
                *bufptr++ = *n++;
        *bufptr++ = '.';
        n = IpQuadTab[*a++];
        while(*n)
                *bufptr++ = *n++;
        *bufptr++ = '.';
        n = IpQuadTab[*a++];
        while(*n)
                *bufptr++ = *n++;
        *bufptr++ = '.';
        n = IpQuadTab[*a];
        while(*n)
                *bufptr++ = *n++;
        *bufptr = '\0';
        return buf;
}


/*
 * Copyright (c) 1996-1999 by Internet Software Consortium.
 *
 * Permission to use, copy, modify, and distribute this software for any
 * purpose with or without fee is hereby granted, provided that the above
 * copyright notice and this permission notice appear in all copies.
 *
 * THE SOFTWARE IS PROVIDED "AS IS" AND INTERNET SOFTWARE CONSORTIUM DISCLAIMS
 * ALL WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES
 * OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL INTERNET SOFTWARE
 * CONSORTIUM BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL
 * DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
 * PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
 * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
 * SOFTWARE.
 */

#define SPRINTF(x) ((size_t)rb_sprintf x)

/*
 * WARNING: Don't even consider trying to compile this on a system where
 * sizeof(int) < 4.  sizeof(int) > 4 is fine; all the world's not a VAX.
 */

static const char *inet_ntop4(const unsigned char *src, char *dst, unsigned int size);
#ifdef RB_IPV6
static const char *inet_ntop6(const unsigned char *src, char *dst, unsigned int size);
#endif

/* const char *
 * inet_ntop4(src, dst, size)
 *      format an IPv4 address
 * return:
 *      `dst' (as a const)
 * notes:
 *      (1) uses no statics
 *      (2) takes a unsigned char* not an in_addr as input
 * author:
 *      Paul Vixie, 1996.
 */
static const char *
inet_ntop4(const unsigned char *src, char *dst, unsigned int size)
{
        if(size < 16)
                return NULL;
        return strcpy(dst, inetntoa((const char *)src));
}

/* const char *
 * inet_ntop6(src, dst, size)
 *      convert IPv6 binary address into presentation (printable) format
 * author:
 *      Paul Vixie, 1996.
 */
#ifdef RB_IPV6
static const char *
inet_ntop6(const unsigned char *src, char *dst, unsigned int size)
{
        /*
         * Note that int32_t and int16_t need only be "at least" large enough
         * to contain a value of the specified size.  On some systems, like
         * Crays, there is no such thing as an integer variable with 16 bits.
         * Keep this in mind if you think this function should have been coded
         * to use pointer overlays.  All the world's not a VAX.
         */
        char tmp[sizeof "ffff:ffff:ffff:ffff:ffff:ffff:255.255.255.255"], *tp;
        struct
        {
                int base, len;
        }
        best, cur;
        unsigned int words[IN6ADDRSZ / INT16SZ];
        int i;

        /*
         * Preprocess:
         *      Copy the input (bytewise) array into a wordwise array.
         *      Find the longest run of 0x00's in src[] for :: shorthanding.
         */
        memset(words, '\0', sizeof words);
        for(i = 0; i < IN6ADDRSZ; i += 2)
                words[i / 2] = (src[i] << 8) | src[i + 1];
        best.base = -1;
        best.len = 0;
        cur.base = -1;
        cur.len = 0;
        for(i = 0; i < (IN6ADDRSZ / INT16SZ); i++)
        {
                if(words[i] == 0)
                {
                        if(cur.base == -1)
                                cur.base = i, cur.len = 1;
                        else
                                cur.len++;
                }
                else
                {
                        if(cur.base != -1)
                        {
                                if(best.base == -1 || cur.len > best.len)
                                        best = cur;
                                cur.base = -1;
                        }
                }
        }
        if(cur.base != -1)
        {
                if(best.base == -1 || cur.len > best.len)
                        best = cur;
        }
        if(best.base != -1 && best.len < 2)
                best.base = -1;

        /*
         * Format the result.
         */
        tp = tmp;
        for(i = 0; i < (IN6ADDRSZ / INT16SZ); i++)
        {
                /* Are we inside the best run of 0x00's? */
                if(best.base != -1 && i >= best.base && i < (best.base + best.len))
                {
                        if(i == best.base)
                        {
                                if(i == 0)
                                        *tp++ = '0';
                                *tp++ = ':';
                        }
                        continue;
                }
                /* Are we following an initial run of 0x00s or any real hex? */
                if(i != 0)
                        *tp++ = ':';
                /* Is this address an encapsulated IPv4? */
                if(i == 6 && best.base == 0 &&
                   (best.len == 6 || (best.len == 5 && words[5] == 0xffff)))
                {
                        if(!inet_ntop4(src + 12, tp, sizeof tmp - (tp - tmp)))
                                return (NULL);
                        tp += strlen(tp);
                        break;
                }
                tp += SPRINTF((tp, "%x", words[i]));
        }
        /* Was it a trailing run of 0x00's? */
        if(best.base != -1 && (best.base + best.len) == (IN6ADDRSZ / INT16SZ))
                *tp++ = ':';
        *tp++ = '\0';

        /*
         * Check for overflow, copy, and we're done.
         */

        if((unsigned int)(tp - tmp) > size)
        {
                return (NULL);
        }
        return strcpy(dst, tmp);
}
#endif

int
rb_inet_pton_sock(const char *src, struct sockaddr *dst)
{
        if(rb_inet_pton(AF_INET, src, &((struct sockaddr_in *)dst)->sin_addr))
        {
                ((struct sockaddr_in *)dst)->sin_port = 0;
                ((struct sockaddr_in *)dst)->sin_family = AF_INET;
                SET_SS_LEN(dst, sizeof(struct sockaddr_in));
                return 1;
        }
#ifdef RB_IPV6
        else if(rb_inet_pton(AF_INET6, src, &((struct sockaddr_in6 *)dst)->sin6_addr))
        {
                ((struct sockaddr_in6 *)dst)->sin6_port = 0;
                ((struct sockaddr_in6 *)dst)->sin6_family = AF_INET6;
                SET_SS_LEN(dst, sizeof(struct sockaddr_in6));
                return 1;
        }
#endif
        return 0;
}

const char *
rb_inet_ntop_sock(struct sockaddr *src, char *dst, unsigned int size)
{
        switch (src->sa_family)
        {
        case AF_INET:
                return (rb_inet_ntop(AF_INET, &((struct sockaddr_in *)src)->sin_addr, dst, size));
                break;
#ifdef RB_IPV6
        case AF_INET6:
                return (rb_inet_ntop
                        (AF_INET6, &((struct sockaddr_in6 *)src)->sin6_addr, dst, size));
                break;
#endif
        default:
                return NULL;
                break;
        }
}

/* char *
 * rb_inet_ntop(af, src, dst, size)
 *      convert a network format address to presentation format.
 * return:
 *      pointer to presentation format address (`dst'), or NULL (see errno).
 * author:
 *      Paul Vixie, 1996.
 */
const char *
rb_inet_ntop(int af, const void *src, char *dst, unsigned int size)
{
        switch (af)
        {
        case AF_INET:
                return (inet_ntop4(src, dst, size));
#ifdef RB_IPV6
        case AF_INET6:
                if(IN6_IS_ADDR_V4MAPPED((const struct in6_addr *)src) ||
                   IN6_IS_ADDR_V4COMPAT((const struct in6_addr *)src))
                        return (inet_ntop4
                                ((const unsigned char *)&((const struct in6_addr *)src)->
                                 s6_addr[12], dst, size));
                else
                        return (inet_ntop6(src, dst, size));


#endif
        default:
                return (NULL);
        }
        /* NOTREACHED */
}

/*
 * WARNING: Don't even consider trying to compile this on a system where
 * sizeof(int) < 4.  sizeof(int) > 4 is fine; all the world's not a VAX.
 */

/* int
 * rb_inet_pton(af, src, dst)
 *      convert from presentation format (which usually means ASCII printable)
 *      to network format (which is usually some kind of binary format).
 * return:
 *      1 if the address was valid for the specified address family
 *      0 if the address wasn't valid (`dst' is untouched in this case)
 *      -1 if some other error occurred (`dst' is untouched in this case, too)
 * author:
 *      Paul Vixie, 1996.
 */

/* int
 * inet_pton4(src, dst)
 *      like inet_aton() but without all the hexadecimal and shorthand.
 * return:
 *      1 if `src' is a valid dotted quad, else 0.
 * notice:
 *      does not touch `dst' unless it's returning 1.
 * author:
 *      Paul Vixie, 1996.
 */
static int
inet_pton4(const char *src, unsigned char *dst)
{
        int saw_digit, octets, ch;
        unsigned char tmp[INADDRSZ], *tp;

        saw_digit = 0;
        octets = 0;
        *(tp = tmp) = 0;
        while((ch = *src++) != '\0')
        {

                if(ch >= '0' && ch <= '9')
                {
                        unsigned int new = *tp * 10 + (ch - '0');

                        if(new > 255)
                                return (0);
                        *tp = new;
                        if(!saw_digit)
                        {
                                if(++octets > 4)
                                        return (0);
                                saw_digit = 1;
                        }
                }
                else if(ch == '.' && saw_digit)
                {
                        if(octets == 4)
                                return (0);
                        *++tp = 0;
                        saw_digit = 0;
                }
                else
                        return (0);
        }
        if(octets < 4)
                return (0);
        memcpy(dst, tmp, INADDRSZ);
        return (1);
}

#ifdef RB_IPV6
/* int
 * inet_pton6(src, dst)
 *      convert presentation level address to network order binary form.
 * return:
 *      1 if `src' is a valid [RFC1884 2.2] address, else 0.
 * notice:
 *      (1) does not touch `dst' unless it's returning 1.
 *      (2) :: in a full address is silently ignored.
 * credit:
 *      inspired by Mark Andrews.
 * author:
 *      Paul Vixie, 1996.
 */

static int
inet_pton6(const char *src, unsigned char *dst)
{
        static const char xdigits[] = "0123456789abcdef";
        unsigned char tmp[IN6ADDRSZ], *tp, *endp, *colonp;
        const char *curtok;
        int ch, saw_xdigit;
        unsigned int val;

        tp = memset(tmp, '\0', IN6ADDRSZ);
        endp = tp + IN6ADDRSZ;
        colonp = NULL;
        /* Leading :: requires some special handling. */
        if(*src == ':')
                if(*++src != ':')
                        return (0);
        curtok = src;
        saw_xdigit = 0;
        val = 0;
        while((ch = tolower(*src++)) != '\0')
        {
                const char *pch;

                pch = strchr(xdigits, ch);
                if(pch != NULL)
                {
                        val <<= 4;
                        val |= (pch - xdigits);
                        if(val > 0xffff)
                                return (0);
                        saw_xdigit = 1;
                        continue;
                }
                if(ch == ':')
                {
                        curtok = src;
                        if(!saw_xdigit)
                        {
                                if(colonp)
                                        return (0);
                                colonp = tp;
                                continue;
                        }
                        else if(*src == '\0')
                        {
                                return (0);
                        }
                        if(tp + INT16SZ > endp)
                                return (0);
                        *tp++ = (unsigned char)(val >> 8) & 0xff;
                        *tp++ = (unsigned char)val & 0xff;
                        saw_xdigit = 0;
                        val = 0;
                        continue;
                }
                if(*src != '\0' && ch == '.')
                {
                        if(((tp + INADDRSZ) <= endp) && inet_pton4(curtok, tp) > 0)
                        {
                                tp += INADDRSZ;
                                saw_xdigit = 0;
                                break;  /* '\0' was seen by inet_pton4(). */
                        }
                }
                else
                        continue;
                return (0);
        }
        if(saw_xdigit)
        {
                if(tp + INT16SZ > endp)
                        return (0);
                *tp++ = (unsigned char)(val >> 8) & 0xff;
                *tp++ = (unsigned char)val & 0xff;
        }
        if(colonp != NULL)
        {
                /*
                 * Since some memmove()'s erroneously fail to handle
                 * overlapping regions, we'll do the shift by hand.
                 */
                const int n = tp - colonp;
                int i;

                if(tp == endp)
                        return (0);
                for(i = 1; i <= n; i++)
                {
                        endp[-i] = colonp[n - i];
                        colonp[n - i] = 0;
                }
                tp = endp;
        }
        if(tp != endp)
                return (0);
        memcpy(dst, tmp, IN6ADDRSZ);
        return (1);
}
#endif
int
rb_inet_pton(int af, const char *src, void *dst)
{
        switch (af)
        {
        case AF_INET:
                return (inet_pton4(src, dst));
#ifdef RB_IPV6
        case AF_INET6:
                /* Somebody might have passed as an IPv4 address this is sick but it works */
                if(inet_pton4(src, dst))
                {
                        char tmp[HOSTIPLEN];
                        rb_sprintf(tmp, "::ffff:%s", src);
                        return (inet_pton6(tmp, dst));
                }
                else
                        return (inet_pton6(src, dst));
#endif
        default:
                return (-1);
        }
        /* NOTREACHED */
}


#ifndef HAVE_SOCKETPAIR

/* mostly based on perl's emulation of socketpair udp */
static int
rb_inet_socketpair_udp(rb_fde_t **newF1, rb_fde_t **newF2)
{
        struct sockaddr_in addr[2];
        rb_socklen_t size = sizeof(struct sockaddr_in);
        rb_fde_t *F[2];
        unsigned int fd[2];
        int i, got;
        unsigned short port;

        memset(&addr, 0, sizeof(addr));

        for(i = 0; i < 2; i++)
        {
                F[i] = rb_socket(AF_INET, SOCK_DGRAM, 0, "udp socketpair");
                if(F[i] == NULL)
                        goto failed;
                addr[i].sin_family = AF_INET;
                addr[i].sin_addr.s_addr = htonl(INADDR_LOOPBACK);
                addr[i].sin_port = 0;
                if(bind(rb_get_fd(F[i]), (struct sockaddr *)&addr[i], sizeof(struct sockaddr_in)))
                        goto failed;
                fd[i] = rb_get_fd(F[i]);
        }

        for(i = 0; i < 2; i++)
        {
                if(getsockname(fd[i], (struct sockaddr *)&addr[i], &size))
                        goto failed;
                if(size != sizeof(struct sockaddr_in))
                        goto failed;
                if(connect(fd[!i], (struct sockaddr *)&addr[i], sizeof(struct sockaddr_in)) == -1)
                        goto failed;
        }

        for(i = 0; i < 2; i++)
        {
                port = addr[i].sin_port;
                got = rb_write(F[i], &port, sizeof(port));
                if(got != sizeof(port))
                {
                        if(got == -1)
                                goto failed;
                        goto abort_failed;
                }
        }


        struct timeval wait = { 0, 100000 };

        int max = fd[1] > fd[0] ? fd[1] : fd[0];
        fd_set rset;
        FD_ZERO(&rset);
        FD_SET(fd[0], &rset);
        FD_SET(fd[1], &rset);
        got = select(max + 1, &rset, NULL, NULL, &wait);
        if(got != 2 || !FD_ISSET(fd[0], &rset) || !FD_ISSET(fd[1], &rset))
        {
                if(got == -1)
                        goto failed;
                goto abort_failed;
        }

        struct sockaddr_in readfrom;
        unsigned short buf[2];
        for(i = 0; i < 2; i++)
        {
#ifdef MSG_DONTWAIT
                int flag = MSG_DONTWAIT
#else
                int flag = 0;
#endif
                got = recvfrom(rb_get_fd(F[i]), (char *)&buf, sizeof(buf), flag,
                               (struct sockaddr *)&readfrom, &size);
                if(got == -1)
                        goto failed;
                if(got != sizeof(port)
                   || size != sizeof(struct sockaddr_in)
                   || buf[0] != (unsigned short)addr[!i].sin_port
                   || readfrom.sin_family != addr[!i].sin_family
                   || readfrom.sin_addr.s_addr != addr[!i].sin_addr.s_addr
                   || readfrom.sin_port != addr[!i].sin_port)
                        goto abort_failed;
        }

        *newF1 = F[0];
        *newF2 = F[1];
        return 0;

#ifdef _WIN32
#define ECONNABORTED WSAECONNABORTED
#endif

      abort_failed:
        rb_get_errno();
        errno = ECONNABORTED;
      failed:
        if(errno != ECONNABORTED)
                rb_get_errno();
        int o_errno = errno;
        if(F[0] != NULL)
                rb_close(F[0]);
        if(F[1] != NULL)
                rb_close(F[1]);
        errno = o_errno;
        return -1;
}


int
rb_inet_socketpair(int family, int type, int protocol, int fd[2])
{
        int listener = -1;
        int connector = -1;
        int acceptor = -1;
        struct sockaddr_in listen_addr;
        struct sockaddr_in connect_addr;
        rb_socklen_t size;

        if(protocol || family != AF_INET)
        {
                errno = EAFNOSUPPORT;
                return -1;
        }
        if(!fd)
        {
                errno = EINVAL;
                return -1;
        }

        listener = socket(AF_INET, type, 0);
        if(listener == -1)
                return -1;
        memset(&listen_addr, 0, sizeof(listen_addr));
        listen_addr.sin_family = AF_INET;
        listen_addr.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
        listen_addr.sin_port = 0;       /* kernel choses port.  */
        if(bind(listener, (struct sockaddr *)&listen_addr, sizeof(listen_addr)) == -1)
                goto tidy_up_and_fail;
        if(listen(listener, 1) == -1)
                goto tidy_up_and_fail;

        connector = socket(AF_INET, type, 0);
        if(connector == -1)
                goto tidy_up_and_fail;
        /* We want to find out the port number to connect to.  */
        size = sizeof(connect_addr);
        if(getsockname(listener, (struct sockaddr *)&connect_addr, &size) == -1)
                goto tidy_up_and_fail;
        if(size != sizeof(connect_addr))
                goto abort_tidy_up_and_fail;
        if(connect(connector, (struct sockaddr *)&connect_addr, sizeof(connect_addr)) == -1)
                goto tidy_up_and_fail;

        size = sizeof(listen_addr);
        acceptor = accept(listener, (struct sockaddr *)&listen_addr, &size);
        if(acceptor == -1)
                goto tidy_up_and_fail;
        if(size != sizeof(listen_addr))
                goto abort_tidy_up_and_fail;
        close(listener);
        /* Now check we are talking to ourself by matching port and host on the
           two sockets.  */
        if(getsockname(connector, (struct sockaddr *)&connect_addr, &size) == -1)
                goto tidy_up_and_fail;
        if(size != sizeof(connect_addr)
           || listen_addr.sin_family != connect_addr.sin_family
           || listen_addr.sin_addr.s_addr != connect_addr.sin_addr.s_addr
           || listen_addr.sin_port != connect_addr.sin_port)
        {
                goto abort_tidy_up_and_fail;
        }
        fd[0] = connector;
        fd[1] = acceptor;
        return 0;

      abort_tidy_up_and_fail:
        errno = EINVAL;         /* I hope this is portable and appropriate.  */

      tidy_up_and_fail:
        {
                int save_errno = errno;
                if(listener != -1)
                        close(listener);
                if(connector != -1)
                        close(connector);
                if(acceptor != -1)
                        close(acceptor);
                errno = save_errno;
                return -1;
        }
}

#endif


static void (*setselect_handler) (rb_fde_t *, unsigned int, PF *, void *);
static int (*select_handler) (long);
static int (*setup_fd_handler) (rb_fde_t *);
static int (*io_sched_event) (struct ev_entry *, int);
static void (*io_unsched_event) (struct ev_entry *);
static int (*io_supports_event) (void);
static void (*io_init_event) (void);
static char iotype[25];

const char *
rb_get_iotype(void)
{
        return iotype;
}

static int
rb_unsupported_event(void)
{
        return 0;
}

static int
try_kqueue(void)
{
        if(!rb_init_netio_kqueue())
        {
                setselect_handler = rb_setselect_kqueue;
                select_handler = rb_select_kqueue;
                setup_fd_handler = rb_setup_fd_kqueue;
                io_sched_event = rb_kqueue_sched_event;
                io_unsched_event = rb_kqueue_unsched_event;
                io_init_event = rb_kqueue_init_event;
                io_supports_event = rb_kqueue_supports_event;
                rb_strlcpy(iotype, "kqueue", sizeof(iotype));
                return 0;
        }
        return -1;
}

static int
try_epoll(void)
{
        if(!rb_init_netio_epoll())
        {
                setselect_handler = rb_setselect_epoll;
                select_handler = rb_select_epoll;
                setup_fd_handler = rb_setup_fd_epoll;
                io_sched_event = rb_epoll_sched_event;
                io_unsched_event = rb_epoll_unsched_event;
                io_supports_event = rb_epoll_supports_event;
                io_init_event = rb_epoll_init_event;
                rb_strlcpy(iotype, "epoll", sizeof(iotype));
                return 0;
        }
        return -1;
}

static int
try_ports(void)
{
        if(!rb_init_netio_ports())
        {
                setselect_handler = rb_setselect_ports;
                select_handler = rb_select_ports;
                setup_fd_handler = rb_setup_fd_ports;
                io_sched_event = NULL;
                io_unsched_event = NULL;
                io_init_event = NULL;
                io_supports_event = rb_unsupported_event;
                rb_strlcpy(iotype, "ports", sizeof(iotype));
                return 0;
        }
        return -1;
}

static int
try_devpoll(void)
{
        if(!rb_init_netio_devpoll())
        {
                setselect_handler = rb_setselect_devpoll;
                select_handler = rb_select_devpoll;
                setup_fd_handler = rb_setup_fd_devpoll;
                io_sched_event = NULL;
                io_unsched_event = NULL;
                io_init_event = NULL;
                io_supports_event = rb_unsupported_event;
                rb_strlcpy(iotype, "devpoll", sizeof(iotype));
                return 0;
        }
        return -1;
}

static int
try_sigio(void)
{
        if(!rb_init_netio_sigio())
        {
                setselect_handler = rb_setselect_sigio;
                select_handler = rb_select_sigio;
                setup_fd_handler = rb_setup_fd_sigio;
                io_sched_event = rb_sigio_sched_event;
                io_unsched_event = rb_sigio_unsched_event;
                io_supports_event = rb_sigio_supports_event;
                io_init_event = rb_sigio_init_event;

                rb_strlcpy(iotype, "sigio", sizeof(iotype));
                return 0;
        }
        return -1;
}

static int
try_poll(void)
{
        if(!rb_init_netio_poll())
        {
                setselect_handler = rb_setselect_poll;
                select_handler = rb_select_poll;
                setup_fd_handler = rb_setup_fd_poll;
                io_sched_event = NULL;
                io_unsched_event = NULL;
                io_init_event = NULL;
                io_supports_event = rb_unsupported_event;
                rb_strlcpy(iotype, "poll", sizeof(iotype));
                return 0;
        }
        return -1;
}

static int
try_win32(void)
{
        if(!rb_init_netio_win32())
        {
                setselect_handler = rb_setselect_win32;
                select_handler = rb_select_win32;
                setup_fd_handler = rb_setup_fd_win32;
                io_sched_event = NULL;
                io_unsched_event = NULL;
                io_init_event = NULL;
                io_supports_event = rb_unsupported_event;
                rb_strlcpy(iotype, "win32", sizeof(iotype));
                return 0;
        }
        return -1;
}

static int
try_select(void)
{
        if(!rb_init_netio_select())
        {
                setselect_handler = rb_setselect_select;
                select_handler = rb_select_select;
                setup_fd_handler = rb_setup_fd_select;
                io_sched_event = NULL;
                io_unsched_event = NULL;
                io_init_event = NULL;
                io_supports_event = rb_unsupported_event;
                rb_strlcpy(iotype, "select", sizeof(iotype));
                return 0;
        }
        return -1;
}


int
rb_io_sched_event(struct ev_entry *ev, int when)
{
        if(ev == NULL || io_supports_event == NULL || io_sched_event == NULL
           || !io_supports_event())
                return 0;
        return io_sched_event(ev, when);
}

void
rb_io_unsched_event(struct ev_entry *ev)
{
        if(ev == NULL || io_supports_event == NULL || io_unsched_event == NULL
           || !io_supports_event())
                return;
        io_unsched_event(ev);
}

int
rb_io_supports_event(void)
{
        if(io_supports_event == NULL)
                return 0;
        return io_supports_event();
}

void
rb_io_init_event(void)
{
        io_init_event();
        rb_event_io_register_all();
}

void
rb_init_netio(void)
{
        char *ioenv = getenv("LIBRB_USE_IOTYPE");
        rb_fd_table = rb_malloc(RB_FD_HASH_SIZE * sizeof(rb_dlink_list));
        rb_init_ssl();

        if(ioenv != NULL)
        {
                if(!strcmp("epoll", ioenv))
                {
                        if(!try_epoll())
                                return;
                }
                else if(!strcmp("kqueue", ioenv))
                {
                        if(!try_kqueue())
                                return;
                }
                else if(!strcmp("ports", ioenv))
                {
                        if(!try_ports())
                                return;
                }
                else if(!strcmp("poll", ioenv))
                {
                        if(!try_poll())
                                return;
                }
                else if(!strcmp("devpoll", ioenv))
                {
                        if(!try_devpoll())
                                return;
                }
                else if(!strcmp("sigio", ioenv))
                {
                        if(!try_sigio())
                                return;
                }
                else if(!strcmp("select", ioenv))
                {
                        if(!try_select())
                                return;
                }
                if(!strcmp("win32", ioenv))
                {
                        if(!try_win32())
                                return;
                }

        }

        if(!try_kqueue())
                return;
        if(!try_epoll())
                return;
        if(!try_ports())
                return;
        if(!try_devpoll())
                return;
        if(!try_sigio())
                return;
        if(!try_poll())
                return;
        if(!try_win32())
                return;
        if(!try_select())
                return;

        rb_lib_log("rb_init_netio: Could not find any io handlers...giving up");

        abort();
}

void
rb_setselect(rb_fde_t *F, unsigned int type, PF * handler, void *client_data)
{
        setselect_handler(F, type, handler, client_data);
}

int
rb_select(unsigned long timeout)
{
        int ret = select_handler(timeout);
        free_fds();
        return ret;
}

int
rb_setup_fd(rb_fde_t *F)
{
        return setup_fd_handler(F);
}



int
rb_ignore_errno(int error)
{
        switch (error)
        {
#ifdef EINPROGRESS
        case EINPROGRESS:
#endif
#if defined EWOULDBLOCK
        case EWOULDBLOCK:
#endif
#if defined(EAGAIN) && (EWOULDBLOCK != EAGAIN)
        case EAGAIN:
#endif
#ifdef EINTR
        case EINTR:
#endif
#ifdef ERESTART
        case ERESTART:
#endif
#ifdef ENOBUFS
        case ENOBUFS:
#endif
                return 1;
        default:
                break;
        }
        return 0;
}


#if defined(HAVE_SENDMSG) && !defined(WIN32)
int
rb_recv_fd_buf(rb_fde_t *F, void *data, size_t datasize, rb_fde_t **xF, int nfds)
{
        struct msghdr msg;
        struct cmsghdr *cmsg;
        struct iovec iov[1];
        struct stat st;
        uint8_t stype = RB_FD_UNKNOWN;
        const char *desc;
        int fd, len, x, rfds;

        int control_len = CMSG_SPACE(sizeof(int) * nfds);

        iov[0].iov_base = data;
        iov[0].iov_len = datasize;

        msg.msg_name = NULL;
        msg.msg_namelen = 0;
        msg.msg_iov = iov;
        msg.msg_iovlen = 1;
        msg.msg_flags = 0;
        cmsg = alloca(control_len);
        msg.msg_control = cmsg;
        msg.msg_controllen = control_len;

        if((len = recvmsg(rb_get_fd(F), &msg, 0)) <= 0)
                return len;

        if(msg.msg_controllen > 0 && msg.msg_control != NULL
           && (cmsg = CMSG_FIRSTHDR(&msg)) != NULL)
        {
                rfds = (msg.msg_controllen - sizeof(struct cmsghdr)) / sizeof(int);

                for(x = 0; x < nfds && x < rfds; x++)
                {
                        fd = ((int *)CMSG_DATA(cmsg))[x];
                        stype = RB_FD_UNKNOWN;
                        desc = "remote unknown";
                        if(!fstat(fd, &st))
                        {
                                if(S_ISSOCK(st.st_mode))
                                {
                                        stype = RB_FD_SOCKET;
                                        desc = "remote socket";
                                }
                                else if(S_ISFIFO(st.st_mode))
                                {
                                        stype = RB_FD_PIPE;
                                        desc = "remote pipe";
                                }
                                else if(S_ISREG(st.st_mode))
                                {
                                        stype = RB_FD_FILE;
                                        desc = "remote file";
                                }
                        }
                        xF[x] = rb_open(fd, stype, desc);
                }
        }
        else
                *xF = NULL;
        return len;
}


int
rb_send_fd_buf(rb_fde_t *xF, rb_fde_t **F, int count, void *data, size_t datasize, pid_t pid)
{
        int n;
        struct msghdr msg;
        struct cmsghdr *cmsg;
        struct iovec iov[1];
        char empty = '0';
        char *buf;

        memset(&msg, 0, sizeof(&msg));
        if(datasize == 0)
        {
                iov[0].iov_base = &empty;
                iov[0].iov_len = 1;
        }
        else
        {
                iov[0].iov_base = data;
                iov[0].iov_len = datasize;
        }
        msg.msg_iov = iov;
        msg.msg_iovlen = 1;
        msg.msg_name = NULL;
        msg.msg_namelen = 0;
        msg.msg_flags = 0;
        msg.msg_control = NULL;
        msg.msg_controllen = 0;

        if(count > 0)
        {
                int i;
                int len = CMSG_SPACE(sizeof(int) * count);
                buf = alloca(len);

                msg.msg_control = buf;
                msg.msg_controllen = len;
                cmsg = CMSG_FIRSTHDR(&msg);
                cmsg->cmsg_level = SOL_SOCKET;
                cmsg->cmsg_type = SCM_RIGHTS;
                cmsg->cmsg_len = CMSG_LEN(sizeof(int) * count);

                for(i = 0; i < count; i++)
                {
                        ((int *)CMSG_DATA(cmsg))[i] = rb_get_fd(F[i]);
                }
                msg.msg_controllen = cmsg->cmsg_len;
        }
        n = sendmsg(rb_get_fd(xF), &msg, MSG_NOSIGNAL);
        return n;
}
#else
#ifndef _WIN32
int
rb_recv_fd_buf(rb_fde_t *F, void *data, size_t datasize, rb_fde_t **xF, int nfds)
{
        errno = ENOSYS;
        return -1;
}

int
rb_send_fd_buf(rb_fde_t *xF, rb_fde_t **F, int count, void *data, size_t datasize, pid_t pid)
{
        errno = ENOSYS;
        return -1;
}
#endif
#endif