[qemu.git] / net / tap.c

/*
 * QEMU System Emulator
 *
 * Copyright (c) 2003-2008 Fabrice Bellard
 * Copyright (c) 2009 Red Hat, Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */

#include "net/tap.h"

#include "config-host.h"

#include <signal.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <sys/wait.h>
#include <sys/socket.h>
#include <net/if.h>

#include "net.h"
#include "sysemu.h"
#include "qemu-char.h"
#include "qemu-common.h"

#include "net/tap-linux.h"

/* Maximum GSO packet size (64k) plus plenty of room for
 * the ethernet and virtio_net headers
 */
#define TAP_BUFSIZE (4096 + 65536)

typedef struct TAPState {
    VLANClientState nc;
    int fd;
    char down_script[1024];
    char down_script_arg[128];
    uint8_t buf[TAP_BUFSIZE];
    unsigned int read_poll : 1;
    unsigned int write_poll : 1;
    unsigned int has_vnet_hdr : 1;
    unsigned int using_vnet_hdr : 1;
    unsigned int has_ufo: 1;
} TAPState;

static int launch_script(const char *setup_script, const char *ifname, int fd);

static int tap_can_send(void *opaque);
static void tap_send(void *opaque);
static void tap_writable(void *opaque);

static void tap_update_fd_handler(TAPState *s)
{
    qemu_set_fd_handler2(s->fd,
                         s->read_poll  ? tap_can_send : NULL,
                         s->read_poll  ? tap_send     : NULL,
                         s->write_poll ? tap_writable : NULL,
                         s);
}

static void tap_read_poll(TAPState *s, int enable)
{
    s->read_poll = !!enable;
    tap_update_fd_handler(s);
}

static void tap_write_poll(TAPState *s, int enable)
{
    s->write_poll = !!enable;
    tap_update_fd_handler(s);
}

static void tap_writable(void *opaque)
{
    TAPState *s = opaque;

    tap_write_poll(s, 0);

    qemu_flush_queued_packets(&s->nc);
}

static ssize_t tap_write_packet(TAPState *s, const struct iovec *iov, int iovcnt)
{
    ssize_t len;

    do {
        len = writev(s->fd, iov, iovcnt);
    } while (len == -1 && errno == EINTR);

    if (len == -1 && errno == EAGAIN) {
        tap_write_poll(s, 1);
        return 0;
    }

    return len;
}

static ssize_t tap_receive_iov(VLANClientState *nc, const struct iovec *iov,
                               int iovcnt)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
    const struct iovec *iovp = iov;
    struct iovec iov_copy[iovcnt + 1];
    struct virtio_net_hdr hdr = { 0, };

    if (s->has_vnet_hdr && !s->using_vnet_hdr) {
        iov_copy[0].iov_base = &hdr;
        iov_copy[0].iov_len =  sizeof(hdr);
        memcpy(&iov_copy[1], iov, iovcnt * sizeof(*iov));
        iovp = iov_copy;
        iovcnt++;
    }

    return tap_write_packet(s, iovp, iovcnt);
}

static ssize_t tap_receive_raw(VLANClientState *nc, const uint8_t *buf, size_t size)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
    struct iovec iov[2];
    int iovcnt = 0;
    struct virtio_net_hdr hdr = { 0, };

    if (s->has_vnet_hdr) {
        iov[iovcnt].iov_base = &hdr;
        iov[iovcnt].iov_len  = sizeof(hdr);
        iovcnt++;
    }

    iov[iovcnt].iov_base = (char *)buf;
    iov[iovcnt].iov_len  = size;
    iovcnt++;

    return tap_write_packet(s, iov, iovcnt);
}

static ssize_t tap_receive(VLANClientState *nc, const uint8_t *buf, size_t size)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
    struct iovec iov[1];

    if (s->has_vnet_hdr && !s->using_vnet_hdr) {
        return tap_receive_raw(nc, buf, size);
    }

    iov[0].iov_base = (char *)buf;
    iov[0].iov_len  = size;

    return tap_write_packet(s, iov, 1);
}

static int tap_can_send(void *opaque)
{
    TAPState *s = opaque;

    return qemu_can_send_packet(&s->nc);
}

#ifndef __sun__
ssize_t tap_read_packet(int tapfd, uint8_t *buf, int maxlen)
{
    return read(tapfd, buf, maxlen);
}
#endif

static void tap_send_completed(VLANClientState *nc, ssize_t len)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
    tap_read_poll(s, 1);
}

static void tap_send(void *opaque)
{
    TAPState *s = opaque;
    int size;

    do {
        uint8_t *buf = s->buf;

        size = tap_read_packet(s->fd, s->buf, sizeof(s->buf));
        if (size <= 0) {
            break;
        }

        if (s->has_vnet_hdr && !s->using_vnet_hdr) {
            buf  += sizeof(struct virtio_net_hdr);
            size -= sizeof(struct virtio_net_hdr);
        }

        size = qemu_send_packet_async(&s->nc, buf, size, tap_send_completed);
        if (size == 0) {
            tap_read_poll(s, 0);
        }
    } while (size > 0 && qemu_can_send_packet(&s->nc));
}

int tap_has_ufo(VLANClientState *nc)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);

    assert(nc->info->type == NET_CLIENT_TYPE_TAP);

    return s->has_ufo;
}

int tap_has_vnet_hdr(VLANClientState *nc)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);

    assert(nc->info->type == NET_CLIENT_TYPE_TAP);

    return s->has_vnet_hdr;
}

void tap_using_vnet_hdr(VLANClientState *nc, int using_vnet_hdr)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);

    using_vnet_hdr = using_vnet_hdr != 0;

    assert(nc->info->type == NET_CLIENT_TYPE_TAP);
    assert(s->has_vnet_hdr == using_vnet_hdr);

    s->using_vnet_hdr = using_vnet_hdr;
}

void tap_set_offload(VLANClientState *nc, int csum, int tso4,
                     int tso6, int ecn, int ufo)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);

    return tap_fd_set_offload(s->fd, csum, tso4, tso6, ecn, ufo);
}

static void tap_cleanup(VLANClientState *nc)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);

    qemu_purge_queued_packets(nc);

    if (s->down_script[0])
        launch_script(s->down_script, s->down_script_arg, s->fd);

    tap_read_poll(s, 0);
    tap_write_poll(s, 0);
    close(s->fd);
}

/* fd support */

static NetClientInfo net_tap_info = {
    .type = NET_CLIENT_TYPE_TAP,
    .size = sizeof(TAPState),
    .receive = tap_receive,
    .receive_raw = tap_receive_raw,
    .receive_iov = tap_receive_iov,
    .cleanup = tap_cleanup,
};

static TAPState *net_tap_fd_init(VLANState *vlan,
                                 const char *model,
                                 const char *name,
                                 int fd,
                                 int vnet_hdr)
{
    VLANClientState *nc;
    TAPState *s;

    nc = qemu_new_net_client(&net_tap_info, vlan, NULL, model, name);

    s = DO_UPCAST(TAPState, nc, nc);

    s->fd = fd;
    s->has_vnet_hdr = vnet_hdr != 0;
    s->using_vnet_hdr = 0;
    s->has_ufo = tap_probe_has_ufo(s->fd);
    tap_set_offload(&s->nc, 0, 0, 0, 0, 0);
    tap_read_poll(s, 1);
    return s;
}

static int launch_script(const char *setup_script, const char *ifname, int fd)
{
    sigset_t oldmask, mask;
    int pid, status;
    char *args[3];
    char **parg;

    sigemptyset(&mask);
    sigaddset(&mask, SIGCHLD);
    sigprocmask(SIG_BLOCK, &mask, &oldmask);

    /* try to launch network script */
    pid = fork();
    if (pid == 0) {
        int open_max = sysconf(_SC_OPEN_MAX), i;

        for (i = 0; i < open_max; i++) {
            if (i != STDIN_FILENO &&
                i != STDOUT_FILENO &&
                i != STDERR_FILENO &&
                i != fd) {
                close(i);
            }
        }
        parg = args;
        *parg++ = (char *)setup_script;
        *parg++ = (char *)ifname;
        *parg++ = NULL;
        execv(setup_script, args);
        _exit(1);
    } else if (pid > 0) {
        while (waitpid(pid, &status, 0) != pid) {
            /* loop */
        }
        sigprocmask(SIG_SETMASK, &oldmask, NULL);

        if (WIFEXITED(status) && WEXITSTATUS(status) == 0) {
            return 0;
        }
    }
    fprintf(stderr, "%s: could not launch network script\n", setup_script);
    return -1;
}

static int net_tap_init(QemuOpts *opts, int *vnet_hdr)
{
    int fd, vnet_hdr_required;
    char ifname[128] = {0,};
    const char *setup_script;

    if (qemu_opt_get(opts, "ifname")) {
        pstrcpy(ifname, sizeof(ifname), qemu_opt_get(opts, "ifname"));
    }

    *vnet_hdr = qemu_opt_get_bool(opts, "vnet_hdr", 1);
    if (qemu_opt_get(opts, "vnet_hdr")) {
        vnet_hdr_required = *vnet_hdr;
    } else {
        vnet_hdr_required = 0;
    }

    TFR(fd = tap_open(ifname, sizeof(ifname), vnet_hdr, vnet_hdr_required));
    if (fd < 0) {
        return -1;
    }

    setup_script = qemu_opt_get(opts, "script");
    if (setup_script &&
        setup_script[0] != '\0' &&
        strcmp(setup_script, "no") != 0 &&
        launch_script(setup_script, ifname, fd)) {
        close(fd);
        return -1;
    }

    qemu_opt_set(opts, "ifname", ifname);

    return fd;
}

int net_init_tap(QemuOpts *opts, Monitor *mon, const char *name, VLANState *vlan)
{
    TAPState *s;
    int fd, vnet_hdr;

    if (qemu_opt_get(opts, "fd")) {
        if (qemu_opt_get(opts, "ifname") ||
            qemu_opt_get(opts, "script") ||
            qemu_opt_get(opts, "downscript") ||
            qemu_opt_get(opts, "vnet_hdr")) {
            qemu_error("ifname=, script=, downscript= and vnet_hdr= is invalid with fd=\n");
            return -1;
        }

        fd = net_handle_fd_param(mon, qemu_opt_get(opts, "fd"));
        if (fd == -1) {
            return -1;
        }

        fcntl(fd, F_SETFL, O_NONBLOCK);

        vnet_hdr = tap_probe_vnet_hdr(fd);
    } else {
        if (!qemu_opt_get(opts, "script")) {
            qemu_opt_set(opts, "script", DEFAULT_NETWORK_SCRIPT);
        }

        if (!qemu_opt_get(opts, "downscript")) {
            qemu_opt_set(opts, "downscript", DEFAULT_NETWORK_DOWN_SCRIPT);
        }

        fd = net_tap_init(opts, &vnet_hdr);
        if (fd == -1) {
            return -1;
        }
    }

    s = net_tap_fd_init(vlan, "tap", name, fd, vnet_hdr);
    if (!s) {
        close(fd);
        return -1;
    }

    if (tap_set_sndbuf(s->fd, opts) < 0) {
        return -1;
    }

    if (qemu_opt_get(opts, "fd")) {
        snprintf(s->nc.info_str, sizeof(s->nc.info_str), "fd=%d", fd);
    } else {
        const char *ifname, *script, *downscript;

        ifname     = qemu_opt_get(opts, "ifname");
        script     = qemu_opt_get(opts, "script");
        downscript = qemu_opt_get(opts, "downscript");

        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
                 "ifname=%s,script=%s,downscript=%s",
                 ifname, script, downscript);

        if (strcmp(downscript, "no") != 0) {
            snprintf(s->down_script, sizeof(s->down_script), "%s", downscript);
            snprintf(s->down_script_arg, sizeof(s->down_script_arg), "%s", ifname);
        }
    }

    if (vlan) {
        vlan->nb_host_devs++;
    }

    return 0;
}
Commit	Line	Data
5281d757 MM	1	/*
	2	* QEMU System Emulator
	3	*
	4	* Copyright (c) 2003-2008 Fabrice Bellard
	5	* Copyright (c) 2009 Red Hat, Inc.
	6	*
	7	* Permission is hereby granted, free of charge, to any person obtaining a copy
	8	* of this software and associated documentation files (the "Software"), to deal
	9	* in the Software without restriction, including without limitation the rights
	10	* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
	11	* copies of the Software, and to permit persons to whom the Software is
	12	* furnished to do so, subject to the following conditions:
	13	*
	14	* The above copyright notice and this permission notice shall be included in
	15	* all copies or substantial portions of the Software.
	16	*
	17	* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
	18	* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
	19	* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
	20	* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
	21	* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
	22	* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
	23	* THE SOFTWARE.
	24	*/
	25
	26	#include "net/tap.h"
	27
	28	#include "config-host.h"
	29
	30	#include <signal.h>
	31	#include <sys/ioctl.h>
	32	#include <sys/stat.h>
	33	#include <sys/wait.h>
71f4effc	34	#include <sys/socket.h>
5281d757 MM	35	#include <net/if.h>
	36
	37	#include "net.h"
	38	#include "sysemu.h"
	39	#include "qemu-char.h"
	40	#include "qemu-common.h"
	41
5281d757	42	#include "net/tap-linux.h"
5281d757	43
5281d757 MM	44	/* Maximum GSO packet size (64k) plus plenty of room for
	45	* the ethernet and virtio_net headers
	46	*/
	47	#define TAP_BUFSIZE (4096 + 65536)
	48
	49	typedef struct TAPState {
3e35ba93	50	VLANClientState nc;
5281d757 MM	51	int fd;
	52	char down_script[1024];
	53	char down_script_arg[128];
	54	uint8_t buf[TAP_BUFSIZE];
	55	unsigned int read_poll : 1;
	56	unsigned int write_poll : 1;
	57	unsigned int has_vnet_hdr : 1;
	58	unsigned int using_vnet_hdr : 1;
	59	unsigned int has_ufo: 1;
	60	} TAPState;
	61
	62	static int launch_script(const char setup_script, const char ifname, int fd);
	63
	64	static int tap_can_send(void *opaque);
	65	static void tap_send(void *opaque);
	66	static void tap_writable(void *opaque);
	67
	68	static void tap_update_fd_handler(TAPState *s)
	69	{
	70	qemu_set_fd_handler2(s->fd,
	71	s->read_poll ? tap_can_send : NULL,
	72	s->read_poll ? tap_send : NULL,
	73	s->write_poll ? tap_writable : NULL,
	74	s);
	75	}
	76
	77	static void tap_read_poll(TAPState *s, int enable)
	78	{
	79	s->read_poll = !!enable;
	80	tap_update_fd_handler(s);
	81	}
	82
	83	static void tap_write_poll(TAPState *s, int enable)
	84	{
	85	s->write_poll = !!enable;
	86	tap_update_fd_handler(s);
	87	}
	88
	89	static void tap_writable(void *opaque)
	90	{
	91	TAPState *s = opaque;
	92
	93	tap_write_poll(s, 0);
	94
3e35ba93	95	qemu_flush_queued_packets(&s->nc);
5281d757 MM	96	}
	97
	98	static ssize_t tap_write_packet(TAPState s, const struct iovec iov, int iovcnt)
	99	{
	100	ssize_t len;
	101
	102	do {
	103	len = writev(s->fd, iov, iovcnt);
	104	} while (len == -1 && errno == EINTR);
	105
	106	if (len == -1 && errno == EAGAIN) {
	107	tap_write_poll(s, 1);
	108	return 0;
	109	}
	110
	111	return len;
	112	}
	113
3e35ba93	114	static ssize_t tap_receive_iov(VLANClientState nc, const struct iovec iov,
5281d757 MM	115	int iovcnt)
5281d757 MM	116	{
3e35ba93	117	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757 MM	118	const struct iovec *iovp = iov;
	119	struct iovec iov_copy[iovcnt + 1];
	120	struct virtio_net_hdr hdr = { 0, };
	121
	122	if (s->has_vnet_hdr && !s->using_vnet_hdr) {
	123	iov_copy[0].iov_base = &hdr;
	124	iov_copy[0].iov_len = sizeof(hdr);
	125	memcpy(&iov_copy[1], iov, iovcnt * sizeof(*iov));
	126	iovp = iov_copy;
	127	iovcnt++;
	128	}
	129
	130	return tap_write_packet(s, iovp, iovcnt);
	131	}
	132
3e35ba93	133	static ssize_t tap_receive_raw(VLANClientState nc, const uint8_t buf, size_t size)
5281d757	134	{
3e35ba93	135	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757 MM	136	struct iovec iov[2];
	137	int iovcnt = 0;
	138	struct virtio_net_hdr hdr = { 0, };
	139
	140	if (s->has_vnet_hdr) {
	141	iov[iovcnt].iov_base = &hdr;
	142	iov[iovcnt].iov_len = sizeof(hdr);
	143	iovcnt++;
	144	}
	145
	146	iov[iovcnt].iov_base = (char *)buf;
	147	iov[iovcnt].iov_len = size;
	148	iovcnt++;
	149
	150	return tap_write_packet(s, iov, iovcnt);
	151	}
	152
3e35ba93	153	static ssize_t tap_receive(VLANClientState nc, const uint8_t buf, size_t size)
5281d757	154	{
3e35ba93	155	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757 MM	156	struct iovec iov[1];
	157
	158	if (s->has_vnet_hdr && !s->using_vnet_hdr) {
3e35ba93	159	return tap_receive_raw(nc, buf, size);
5281d757 MM	160	}
	161
	162	iov[0].iov_base = (char *)buf;
	163	iov[0].iov_len = size;
	164
	165	return tap_write_packet(s, iov, 1);
	166	}
	167
	168	static int tap_can_send(void *opaque)
	169	{
	170	TAPState *s = opaque;
	171
3e35ba93	172	return qemu_can_send_packet(&s->nc);
5281d757 MM	173	}
5281d757 MM	174
966ea5ec MM	175	#ifndef __sun__
966ea5ec MM	176	ssize_t tap_read_packet(int tapfd, uint8_t *buf, int maxlen)
5281d757 MM	177	{
	178	return read(tapfd, buf, maxlen);
	179	}
	180	#endif
	181
3e35ba93	182	static void tap_send_completed(VLANClientState *nc, ssize_t len)
5281d757	183	{
3e35ba93	184	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757 MM	185	tap_read_poll(s, 1);
	186	}
	187
	188	static void tap_send(void *opaque)
	189	{
	190	TAPState *s = opaque;
	191	int size;
	192
5819c918 MM	193	do {
	194	uint8_t *buf = s->buf;
	195
	196	size = tap_read_packet(s->fd, s->buf, sizeof(s->buf));
	197	if (size <= 0) {
	198	break;
	199	}
	200
	201	if (s->has_vnet_hdr && !s->using_vnet_hdr) {
	202	buf += sizeof(struct virtio_net_hdr);
	203	size -= sizeof(struct virtio_net_hdr);
	204	}
	205
3e35ba93	206	size = qemu_send_packet_async(&s->nc, buf, size, tap_send_completed);
5819c918 MM	207	if (size == 0) {
	208	tap_read_poll(s, 0);
	209	}
3e35ba93	210	} while (size > 0 && qemu_can_send_packet(&s->nc));
5281d757 MM	211	}
5281d757 MM	212
3e35ba93	213	int tap_has_ufo(VLANClientState *nc)
5281d757	214	{
3e35ba93	215	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757	216
665a3b07	217	assert(nc->info->type == NET_CLIENT_TYPE_TAP);
5281d757 MM	218
	219	return s->has_ufo;
	220	}
	221
3e35ba93	222	int tap_has_vnet_hdr(VLANClientState *nc)
5281d757	223	{
3e35ba93	224	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757	225
665a3b07	226	assert(nc->info->type == NET_CLIENT_TYPE_TAP);
5281d757 MM	227
	228	return s->has_vnet_hdr;
	229	}
	230
3e35ba93	231	void tap_using_vnet_hdr(VLANClientState *nc, int using_vnet_hdr)
5281d757	232	{
3e35ba93	233	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757 MM	234
	235	using_vnet_hdr = using_vnet_hdr != 0;
	236
665a3b07	237	assert(nc->info->type == NET_CLIENT_TYPE_TAP);
5281d757 MM	238	assert(s->has_vnet_hdr == using_vnet_hdr);
	239
	240	s->using_vnet_hdr = using_vnet_hdr;
	241	}
	242
3e35ba93	243	void tap_set_offload(VLANClientState *nc, int csum, int tso4,
5281d757 MM	244	int tso6, int ecn, int ufo)
5281d757 MM	245	{
3e35ba93	246	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757	247
1faac1f7	248	return tap_fd_set_offload(s->fd, csum, tso4, tso6, ecn, ufo);
5281d757 MM	249	}
5281d757 MM	250
3e35ba93	251	static void tap_cleanup(VLANClientState *nc)
5281d757	252	{
3e35ba93	253	TAPState *s = DO_UPCAST(TAPState, nc, nc);
5281d757	254
3e35ba93	255	qemu_purge_queued_packets(nc);
5281d757 MM	256
	257	if (s->down_script[0])
	258	launch_script(s->down_script, s->down_script_arg, s->fd);
	259
	260	tap_read_poll(s, 0);
	261	tap_write_poll(s, 0);
	262	close(s->fd);
5281d757 MM	263	}
	264
	265	/* fd support */
	266
3e35ba93 MM	267	static NetClientInfo net_tap_info = {
	268	.type = NET_CLIENT_TYPE_TAP,
	269	.size = sizeof(TAPState),
	270	.receive = tap_receive,
	271	.receive_raw = tap_receive_raw,
	272	.receive_iov = tap_receive_iov,
	273	.cleanup = tap_cleanup,
	274	};
	275
5281d757 MM	276	static TAPState net_tap_fd_init(VLANState vlan,
	277	const char *model,
	278	const char *name,
	279	int fd,
	280	int vnet_hdr)
	281	{
3e35ba93	282	VLANClientState *nc;
5281d757	283	TAPState *s;
5281d757	284
3e35ba93 MM	285	nc = qemu_new_net_client(&net_tap_info, vlan, NULL, model, name);
	286
	287	s = DO_UPCAST(TAPState, nc, nc);
	288
5281d757 MM	289	s->fd = fd;
	290	s->has_vnet_hdr = vnet_hdr != 0;
	291	s->using_vnet_hdr = 0;
9c282718	292	s->has_ufo = tap_probe_has_ufo(s->fd);
3e35ba93	293	tap_set_offload(&s->nc, 0, 0, 0, 0, 0);
5281d757 MM	294	tap_read_poll(s, 1);
	295	return s;
	296	}
	297
5281d757 MM	298	static int launch_script(const char setup_script, const char ifname, int fd)
	299	{
	300	sigset_t oldmask, mask;
	301	int pid, status;
	302	char *args[3];
	303	char **parg;
	304
	305	sigemptyset(&mask);
	306	sigaddset(&mask, SIGCHLD);
	307	sigprocmask(SIG_BLOCK, &mask, &oldmask);
	308
	309	/* try to launch network script */
	310	pid = fork();
	311	if (pid == 0) {
	312	int open_max = sysconf(_SC_OPEN_MAX), i;
	313
	314	for (i = 0; i < open_max; i++) {
	315	if (i != STDIN_FILENO &&
	316	i != STDOUT_FILENO &&
	317	i != STDERR_FILENO &&
	318	i != fd) {
	319	close(i);
	320	}
	321	}
	322	parg = args;
	323	parg++ = (char )setup_script;
	324	parg++ = (char )ifname;
	325	*parg++ = NULL;
	326	execv(setup_script, args);
	327	_exit(1);
	328	} else if (pid > 0) {
	329	while (waitpid(pid, &status, 0) != pid) {
	330	/* loop */
	331	}
	332	sigprocmask(SIG_SETMASK, &oldmask, NULL);
	333
	334	if (WIFEXITED(status) && WEXITSTATUS(status) == 0) {
	335	return 0;
	336	}
	337	}
	338	fprintf(stderr, "%s: could not launch network script\n", setup_script);
	339	return -1;
	340	}
	341
	342	static int net_tap_init(QemuOpts opts, int vnet_hdr)
	343	{
	344	int fd, vnet_hdr_required;
	345	char ifname[128] = {0,};
	346	const char *setup_script;
	347
	348	if (qemu_opt_get(opts, "ifname")) {
	349	pstrcpy(ifname, sizeof(ifname), qemu_opt_get(opts, "ifname"));
	350	}
	351
	352	*vnet_hdr = qemu_opt_get_bool(opts, "vnet_hdr", 1);
	353	if (qemu_opt_get(opts, "vnet_hdr")) {
	354	vnet_hdr_required = *vnet_hdr;
	355	} else {
	356	vnet_hdr_required = 0;
	357	}
	358
	359	TFR(fd = tap_open(ifname, sizeof(ifname), vnet_hdr, vnet_hdr_required));
	360	if (fd < 0) {
	361	return -1;
362	}
363
364	setup_script = qemu_opt_get(opts, "script");
365	if (setup_script &&
366	setup_script[0] != '\0' &&
367	strcmp(setup_script, "no") != 0 &&
368	launch_script(setup_script, ifname, fd)) {
369	close(fd);
370	return -1;
371	}
372
373	qemu_opt_set(opts, "ifname", ifname);
374
375	return fd;
376	}
377
378	int net_init_tap(QemuOpts opts, Monitor mon, const char name, VLANState vlan)
379	{
380	TAPState *s;
381	int fd, vnet_hdr;
382
383	if (qemu_opt_get(opts, "fd")) {
384	if (qemu_opt_get(opts, "ifname") \|\|
385	qemu_opt_get(opts, "script") \|\|
386	qemu_opt_get(opts, "downscript") \|\|
387	qemu_opt_get(opts, "vnet_hdr")) {
388	qemu_error("ifname=, script=, downscript= and vnet_hdr= is invalid with fd=\n");
389	return -1;
390	}
391
392	fd = net_handle_fd_param(mon, qemu_opt_get(opts, "fd"));
393	if (fd == -1) {
394	return -1;
395	}
396
397	fcntl(fd, F_SETFL, O_NONBLOCK);
398
399	vnet_hdr = tap_probe_vnet_hdr(fd);
400	} else {
401	if (!qemu_opt_get(opts, "script")) {
402	qemu_opt_set(opts, "script", DEFAULT_NETWORK_SCRIPT);
403	}
404
405	if (!qemu_opt_get(opts, "downscript")) {
406	qemu_opt_set(opts, "downscript", DEFAULT_NETWORK_DOWN_SCRIPT);
407	}
408
409	fd = net_tap_init(opts, &vnet_hdr);
929fe497 JL	410	if (fd == -1) {
	411	return -1;
	412	}
5281d757 MM	413	}
	414
	415	s = net_tap_fd_init(vlan, "tap", name, fd, vnet_hdr);
	416	if (!s) {
	417	close(fd);
	418	return -1;
	419	}
	420
15ac913b	421	if (tap_set_sndbuf(s->fd, opts) < 0) {
5281d757 MM	422	return -1;
	423	}
	424
	425	if (qemu_opt_get(opts, "fd")) {
3e35ba93	426	snprintf(s->nc.info_str, sizeof(s->nc.info_str), "fd=%d", fd);
5281d757 MM	427	} else {
	428	const char ifname, script, *downscript;
	429
	430	ifname = qemu_opt_get(opts, "ifname");
	431	script = qemu_opt_get(opts, "script");
	432	downscript = qemu_opt_get(opts, "downscript");
	433
3e35ba93	434	snprintf(s->nc.info_str, sizeof(s->nc.info_str),
5281d757 MM	435	"ifname=%s,script=%s,downscript=%s",
	436	ifname, script, downscript);
	437
	438	if (strcmp(downscript, "no") != 0) {
	439	snprintf(s->down_script, sizeof(s->down_script), "%s", downscript);
	440	snprintf(s->down_script_arg, sizeof(s->down_script_arg), "%s", ifname);
	441	}
	442	}
	443
	444	if (vlan) {
	445	vlan->nb_host_devs++;
	446	}
	447
	448	return 0;
	449	}