tap.c 22.7 KB
Newer Older
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
/*
 * 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.
 */

P
Paolo Bonzini 已提交
26
#include "tap_int.h"
27 28 29 30 31 32

#include "config-host.h"

#include <sys/ioctl.h>
#include <sys/stat.h>
#include <sys/wait.h>
A
Alexander Graf 已提交
33
#include <sys/socket.h>
34 35
#include <net/if.h>

P
Paolo Bonzini 已提交
36
#include "net/net.h"
37
#include "clients.h"
38
#include "monitor/monitor.h"
39
#include "sysemu/sysemu.h"
40
#include "qemu-common.h"
41
#include "qemu/error-report.h"
42

P
Paolo Bonzini 已提交
43
#include "net/tap.h"
44

45 46
#include "hw/vhost_net.h"

47 48 49 50 51 52
/* Maximum GSO packet size (64k) plus plenty of room for
 * the ethernet and virtio_net headers
 */
#define TAP_BUFSIZE (4096 + 65536)

typedef struct TAPState {
53
    NetClientState nc;
54 55 56 57
    int fd;
    char down_script[1024];
    char down_script_arg[128];
    uint8_t buf[TAP_BUFSIZE];
58 59 60 61
    bool read_poll;
    bool write_poll;
    bool using_vnet_hdr;
    bool has_ufo;
62
    bool enabled;
63
    VHostNetState *vhost_net;
64
    unsigned host_vnet_hdr_len;
65 66 67 68 69 70 71 72 73 74 75
} 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,
76 77 78
                         s->read_poll && s->enabled ? tap_can_send : NULL,
                         s->read_poll && s->enabled ? tap_send     : NULL,
                         s->write_poll && s->enabled ? tap_writable : NULL,
79 80 81
                         s);
}

82
static void tap_read_poll(TAPState *s, bool enable)
83
{
84
    s->read_poll = enable;
85 86 87
    tap_update_fd_handler(s);
}

88
static void tap_write_poll(TAPState *s, bool enable)
89
{
90
    s->write_poll = enable;
91 92 93 94 95 96 97
    tap_update_fd_handler(s);
}

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

98
    tap_write_poll(s, false);
99

100
    qemu_flush_queued_packets(&s->nc);
101 102 103 104 105 106 107 108 109 110 111
}

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) {
112
        tap_write_poll(s, true);
113 114 115 116 117 118
        return 0;
    }

    return len;
}

119
static ssize_t tap_receive_iov(NetClientState *nc, const struct iovec *iov,
120 121
                               int iovcnt)
{
122
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
123 124
    const struct iovec *iovp = iov;
    struct iovec iov_copy[iovcnt + 1];
125
    struct virtio_net_hdr_mrg_rxbuf hdr = { };
126

127
    if (s->host_vnet_hdr_len && !s->using_vnet_hdr) {
128
        iov_copy[0].iov_base = &hdr;
129
        iov_copy[0].iov_len =  s->host_vnet_hdr_len;
130 131 132 133 134 135 136 137
        memcpy(&iov_copy[1], iov, iovcnt * sizeof(*iov));
        iovp = iov_copy;
        iovcnt++;
    }

    return tap_write_packet(s, iovp, iovcnt);
}

138
static ssize_t tap_receive_raw(NetClientState *nc, const uint8_t *buf, size_t size)
139
{
140
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
141 142
    struct iovec iov[2];
    int iovcnt = 0;
143
    struct virtio_net_hdr_mrg_rxbuf hdr = { };
144

145
    if (s->host_vnet_hdr_len) {
146
        iov[iovcnt].iov_base = &hdr;
147
        iov[iovcnt].iov_len  = s->host_vnet_hdr_len;
148 149 150 151 152 153 154 155 156 157
        iovcnt++;
    }

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

    return tap_write_packet(s, iov, iovcnt);
}

158
static ssize_t tap_receive(NetClientState *nc, const uint8_t *buf, size_t size)
159
{
160
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
161 162
    struct iovec iov[1];

163
    if (s->host_vnet_hdr_len && !s->using_vnet_hdr) {
164
        return tap_receive_raw(nc, buf, size);
165 166 167 168 169 170 171 172 173 174 175 176
    }

    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;

177
    return qemu_can_send_packet(&s->nc);
178 179
}

180 181
#ifndef __sun__
ssize_t tap_read_packet(int tapfd, uint8_t *buf, int maxlen)
182 183 184 185 186
{
    return read(tapfd, buf, maxlen);
}
#endif

187
static void tap_send_completed(NetClientState *nc, ssize_t len)
188
{
189
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
190
    tap_read_poll(s, true);
191 192 193 194 195 196 197
}

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

M
Mark McLoughlin 已提交
198 199 200 201 202 203 204 205
    do {
        uint8_t *buf = s->buf;

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

206 207 208
        if (s->host_vnet_hdr_len && !s->using_vnet_hdr) {
            buf  += s->host_vnet_hdr_len;
            size -= s->host_vnet_hdr_len;
M
Mark McLoughlin 已提交
209 210
        }

211
        size = qemu_send_packet_async(&s->nc, buf, size, tap_send_completed);
M
Mark McLoughlin 已提交
212
        if (size == 0) {
213
            tap_read_poll(s, false);
M
Mark McLoughlin 已提交
214
        }
215
    } while (size > 0 && qemu_can_send_packet(&s->nc));
216 217
}

218
bool tap_has_ufo(NetClientState *nc)
219
{
220
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
221

222
    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP);
223 224 225 226

    return s->has_ufo;
}

227
int tap_has_vnet_hdr(NetClientState *nc)
228
{
229
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
230

231
    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP);
232

233
    return !!s->host_vnet_hdr_len;
234 235
}

236
int tap_has_vnet_hdr_len(NetClientState *nc, int len)
237 238 239
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);

240
    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP);
241 242 243 244

    return tap_probe_vnet_hdr_len(s->fd, len);
}

245
void tap_set_vnet_hdr_len(NetClientState *nc, int len)
246 247 248
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);

249
    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP);
250 251 252 253 254 255 256
    assert(len == sizeof(struct virtio_net_hdr_mrg_rxbuf) ||
           len == sizeof(struct virtio_net_hdr));

    tap_fd_set_vnet_hdr_len(s->fd, len);
    s->host_vnet_hdr_len = len;
}

257
void tap_using_vnet_hdr(NetClientState *nc, bool using_vnet_hdr)
258
{
259
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
260

261
    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP);
262
    assert(!!s->host_vnet_hdr_len == using_vnet_hdr);
263 264 265 266

    s->using_vnet_hdr = using_vnet_hdr;
}

267
void tap_set_offload(NetClientState *nc, int csum, int tso4,
268 269
                     int tso6, int ecn, int ufo)
{
270
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
271 272 273
    if (s->fd < 0) {
        return;
    }
274

275
    tap_fd_set_offload(s->fd, csum, tso4, tso6, ecn, ufo);
276 277
}

278
static void tap_cleanup(NetClientState *nc)
279
{
280
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
281

282 283
    if (s->vhost_net) {
        vhost_net_cleanup(s->vhost_net);
284
        s->vhost_net = NULL;
285 286
    }

287
    qemu_purge_queued_packets(nc);
288 289 290 291

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

292 293
    tap_read_poll(s, false);
    tap_write_poll(s, false);
294
    close(s->fd);
295
    s->fd = -1;
296 297
}

298
static void tap_poll(NetClientState *nc, bool enable)
299 300 301 302 303 304
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
    tap_read_poll(s, enable);
    tap_write_poll(s, enable);
}

305
int tap_get_fd(NetClientState *nc)
306 307
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
308
    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP);
309 310 311
    return s->fd;
}

312 313
/* fd support */

314
static NetClientInfo net_tap_info = {
315
    .type = NET_CLIENT_OPTIONS_KIND_TAP,
316 317 318 319
    .size = sizeof(TAPState),
    .receive = tap_receive,
    .receive_raw = tap_receive_raw,
    .receive_iov = tap_receive_iov,
320
    .poll = tap_poll,
321 322 323
    .cleanup = tap_cleanup,
};

324
static TAPState *net_tap_fd_init(NetClientState *peer,
325 326 327 328 329
                                 const char *model,
                                 const char *name,
                                 int fd,
                                 int vnet_hdr)
{
330
    NetClientState *nc;
331 332
    TAPState *s;

333
    nc = qemu_new_net_client(&net_tap_info, peer, model, name);
334 335 336

    s = DO_UPCAST(TAPState, nc, nc);

337
    s->fd = fd;
338
    s->host_vnet_hdr_len = vnet_hdr ? sizeof(struct virtio_net_hdr) : 0;
339
    s->using_vnet_hdr = false;
340
    s->has_ufo = tap_probe_has_ufo(s->fd);
341
    s->enabled = true;
342
    tap_set_offload(&s->nc, 0, 0, 0, 0, 0);
343 344 345 346 347 348 349
    /*
     * Make sure host header length is set correctly in tap:
     * it might have been modified by another instance of qemu.
     */
    if (tap_probe_vnet_hdr_len(s->fd, s->host_vnet_hdr_len)) {
        tap_fd_set_vnet_hdr_len(s->fd, s->host_vnet_hdr_len);
    }
350
    tap_read_poll(s, true);
351
    s->vhost_net = NULL;
352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376
    return s;
}

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

    /* 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;
377
        *parg = NULL;
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
        execv(setup_script, args);
        _exit(1);
    } else if (pid > 0) {
        while (waitpid(pid, &status, 0) != pid) {
            /* loop */
        }

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

C
Corey Bryant 已提交
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
static int recv_fd(int c)
{
    int fd;
    uint8_t msgbuf[CMSG_SPACE(sizeof(fd))];
    struct msghdr msg = {
        .msg_control = msgbuf,
        .msg_controllen = sizeof(msgbuf),
    };
    struct cmsghdr *cmsg;
    struct iovec iov;
    uint8_t req[1];
    ssize_t len;

    cmsg = CMSG_FIRSTHDR(&msg);
    cmsg->cmsg_level = SOL_SOCKET;
    cmsg->cmsg_type = SCM_RIGHTS;
    cmsg->cmsg_len = CMSG_LEN(sizeof(fd));
    msg.msg_controllen = cmsg->cmsg_len;

    iov.iov_base = req;
    iov.iov_len = sizeof(req);

    msg.msg_iov = &iov;
    msg.msg_iovlen = 1;

    len = recvmsg(c, &msg, 0);
    if (len > 0) {
        memcpy(&fd, CMSG_DATA(cmsg), sizeof(fd));
        return fd;
    }

    return len;
}

static int net_bridge_run_helper(const char *helper, const char *bridge)
{
    sigset_t oldmask, mask;
    int pid, status;
    char *args[5];
    char **parg;
    int sv[2];

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

    if (socketpair(PF_UNIX, SOCK_STREAM, 0, sv) == -1) {
        return -1;
    }

    /* try to launch bridge helper */
    pid = fork();
    if (pid == 0) {
        int open_max = sysconf(_SC_OPEN_MAX), i;
        char fd_buf[6+10];
        char br_buf[6+IFNAMSIZ] = {0};
        char helper_cmd[PATH_MAX + sizeof(fd_buf) + sizeof(br_buf) + 15];

        for (i = 0; i < open_max; i++) {
            if (i != STDIN_FILENO &&
                i != STDOUT_FILENO &&
                i != STDERR_FILENO &&
                i != sv[1]) {
                close(i);
            }
        }

        snprintf(fd_buf, sizeof(fd_buf), "%s%d", "--fd=", sv[1]);

        if (strrchr(helper, ' ') || strrchr(helper, '\t')) {
            /* assume helper is a command */

            if (strstr(helper, "--br=") == NULL) {
                snprintf(br_buf, sizeof(br_buf), "%s%s", "--br=", bridge);
            }

            snprintf(helper_cmd, sizeof(helper_cmd), "%s %s %s %s",
                     helper, "--use-vnet", fd_buf, br_buf);

            parg = args;
            *parg++ = (char *)"sh";
            *parg++ = (char *)"-c";
            *parg++ = helper_cmd;
            *parg++ = NULL;

            execv("/bin/sh", args);
        } else {
            /* assume helper is just the executable path name */

            snprintf(br_buf, sizeof(br_buf), "%s%s", "--br=", bridge);

            parg = args;
            *parg++ = (char *)helper;
            *parg++ = (char *)"--use-vnet";
            *parg++ = fd_buf;
            *parg++ = br_buf;
            *parg++ = NULL;

            execv(helper, args);
        }
        _exit(1);

    } else if (pid > 0) {
        int fd;

        close(sv[1]);

        do {
            fd = recv_fd(sv[0]);
        } while (fd == -1 && errno == EINTR);

        close(sv[0]);

        while (waitpid(pid, &status, 0) != pid) {
            /* loop */
        }
        sigprocmask(SIG_SETMASK, &oldmask, NULL);
        if (fd < 0) {
            fprintf(stderr, "failed to recv file descriptor\n");
            return -1;
        }

        if (WIFEXITED(status) && WEXITSTATUS(status) == 0) {
            return fd;
        }
    }
    fprintf(stderr, "failed to launch bridge helper\n");
    return -1;
}

523
int net_init_bridge(const NetClientOptions *opts, const char *name,
524
                    NetClientState *peer)
C
Corey Bryant 已提交
525
{
526 527 528
    const NetdevBridgeOptions *bridge;
    const char *helper, *br;

C
Corey Bryant 已提交
529 530 531
    TAPState *s;
    int fd, vnet_hdr;

532 533 534 535 536
    assert(opts->kind == NET_CLIENT_OPTIONS_KIND_BRIDGE);
    bridge = opts->bridge;

    helper = bridge->has_helper ? bridge->helper : DEFAULT_BRIDGE_HELPER;
    br     = bridge->has_br     ? bridge->br     : DEFAULT_BRIDGE_INTERFACE;
C
Corey Bryant 已提交
537

538
    fd = net_bridge_run_helper(helper, br);
C
Corey Bryant 已提交
539 540 541 542 543 544 545 546
    if (fd == -1) {
        return -1;
    }

    fcntl(fd, F_SETFL, O_NONBLOCK);

    vnet_hdr = tap_probe_vnet_hdr(fd);

547
    s = net_tap_fd_init(peer, "bridge", name, fd, vnet_hdr);
C
Corey Bryant 已提交
548 549 550 551 552
    if (!s) {
        close(fd);
        return -1;
    }

553 554
    snprintf(s->nc.info_str, sizeof(s->nc.info_str), "helper=%s,br=%s", helper,
             br);
C
Corey Bryant 已提交
555 556 557 558

    return 0;
}

559 560
static int net_tap_init(const NetdevTapOptions *tap, int *vnet_hdr,
                        const char *setup_script, char *ifname,
J
Jason Wang 已提交
561
                        size_t ifname_sz, int mq_required)
562 563 564
{
    int fd, vnet_hdr_required;

565 566
    if (tap->has_vnet_hdr) {
        *vnet_hdr = tap->vnet_hdr;
567 568
        vnet_hdr_required = *vnet_hdr;
    } else {
569
        *vnet_hdr = 1;
570 571 572
        vnet_hdr_required = 0;
    }

J
Jason Wang 已提交
573 574
    TFR(fd = tap_open(ifname, ifname_sz, vnet_hdr, vnet_hdr_required,
                      mq_required));
575 576 577 578 579 580 581 582 583 584 585 586 587 588 589
    if (fd < 0) {
        return -1;
    }

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

    return fd;
}

J
Jason Wang 已提交
590 591
#define MAX_TAP_QUEUES 1024

592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609
static int net_init_tap_one(const NetdevTapOptions *tap, NetClientState *peer,
                            const char *model, const char *name,
                            const char *ifname, const char *script,
                            const char *downscript, const char *vhostfdname,
                            int vnet_hdr, int fd)
{
    TAPState *s;

    s = net_tap_fd_init(peer, model, name, fd, vnet_hdr);
    if (!s) {
        close(fd);
        return -1;
    }

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

J
Jason Wang 已提交
610
    if (tap->has_fd || tap->has_fds) {
611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645
        snprintf(s->nc.info_str, sizeof(s->nc.info_str), "fd=%d", fd);
    } else if (tap->has_helper) {
        snprintf(s->nc.info_str, sizeof(s->nc.info_str), "helper=%s",
                 tap->helper);
    } else {
        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 (tap->has_vhost ? tap->vhost :
        vhostfdname || (tap->has_vhostforce && tap->vhostforce)) {
        int vhostfd;

        if (tap->has_vhostfd) {
            vhostfd = monitor_handle_fd_param(cur_mon, vhostfdname);
            if (vhostfd == -1) {
                return -1;
            }
        } else {
            vhostfd = -1;
        }

        s->vhost_net = vhost_net_init(&s->nc, vhostfd,
                                      tap->has_vhostforce && tap->vhostforce);
        if (!s->vhost_net) {
            error_report("vhost-net requested but could not be initialized");
            return -1;
        }
J
Jason Wang 已提交
646
    } else if (tap->has_vhostfd || tap->has_vhostfds) {
647 648 649 650 651 652 653
        error_report("vhostfd= is not valid without vhost");
        return -1;
    }

    return 0;
}

J
Jason Wang 已提交
654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679
static int get_fds(char *str, char *fds[], int max)
{
    char *ptr = str, *this;
    size_t len = strlen(str);
    int i = 0;

    while (i < max && ptr < str + len) {
        this = strchr(ptr, ':');

        if (this == NULL) {
            fds[i] = g_strdup(ptr);
        } else {
            fds[i] = g_strndup(ptr, this - ptr);
        }

        i++;
        if (this == NULL) {
            break;
        } else {
            ptr = this + 1;
        }
    }

    return i;
}

680
int net_init_tap(const NetClientOptions *opts, const char *name,
681
                 NetClientState *peer)
682
{
683
    const NetdevTapOptions *tap;
J
Jason Wang 已提交
684
    int fd, vnet_hdr = 0, i = 0, queues;
685 686
    /* for the no-fd, no-helper case */
    const char *script = NULL; /* suppress wrong "uninit'd use" gcc warning */
687
    const char *downscript = NULL;
J
Jason Wang 已提交
688
    const char *vhostfdname;
689 690 691 692
    char ifname[128];

    assert(opts->kind == NET_CLIENT_OPTIONS_KIND_TAP);
    tap = opts->tap;
J
Jason Wang 已提交
693 694
    queues = tap->has_queues ? tap->queues : 1;
    vhostfdname = tap->has_vhostfd ? tap->vhostfd : NULL;
695

696 697
    if (tap->has_fd) {
        if (tap->has_ifname || tap->has_script || tap->has_downscript ||
J
Jason Wang 已提交
698 699
            tap->has_vnet_hdr || tap->has_helper || tap->has_queues ||
            tap->has_fds) {
C
Corey Bryant 已提交
700
            error_report("ifname=, script=, downscript=, vnet_hdr=, "
J
Jason Wang 已提交
701
                         "helper=, queues=, and fds= are invalid with fd=");
702 703 704
            return -1;
        }

705
        fd = monitor_handle_fd_param(cur_mon, tap->fd);
706 707 708 709 710 711 712
        if (fd == -1) {
            return -1;
        }

        fcntl(fd, F_SETFL, O_NONBLOCK);

        vnet_hdr = tap_probe_vnet_hdr(fd);
C
Corey Bryant 已提交
713

A
Anthony Liguori 已提交
714
        if (net_init_tap_one(tap, peer, "tap", name, NULL,
J
Jason Wang 已提交
715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748
                             script, downscript,
                             vhostfdname, vnet_hdr, fd)) {
            return -1;
        }
    } else if (tap->has_fds) {
        char *fds[MAX_TAP_QUEUES];
        char *vhost_fds[MAX_TAP_QUEUES];
        int nfds, nvhosts;

        if (tap->has_ifname || tap->has_script || tap->has_downscript ||
            tap->has_vnet_hdr || tap->has_helper || tap->has_queues ||
            tap->has_fd) {
            error_report("ifname=, script=, downscript=, vnet_hdr=, "
                         "helper=, queues=, and fd= are invalid with fds=");
            return -1;
        }

        nfds = get_fds(tap->fds, fds, MAX_TAP_QUEUES);
        if (tap->has_vhostfds) {
            nvhosts = get_fds(tap->vhostfds, vhost_fds, MAX_TAP_QUEUES);
            if (nfds != nvhosts) {
                error_report("The number of fds passed does not match the "
                             "number of vhostfds passed");
                return -1;
            }
        }

        for (i = 0; i < nfds; i++) {
            fd = monitor_handle_fd_param(cur_mon, fds[i]);
            if (fd == -1) {
                return -1;
            }

            fcntl(fd, F_SETFL, O_NONBLOCK);
C
Corey Bryant 已提交
749

J
Jason Wang 已提交
750 751 752 753 754 755 756 757 758 759 760 761 762 763
            if (i == 0) {
                vnet_hdr = tap_probe_vnet_hdr(fd);
            } else if (vnet_hdr != tap_probe_vnet_hdr(fd)) {
                error_report("vnet_hdr not consistent across given tap fds");
                return -1;
            }

            if (net_init_tap_one(tap, peer, "tap", name, ifname,
                                 script, downscript,
                                 tap->has_vhostfds ? vhost_fds[i] : NULL,
                                 vnet_hdr, fd)) {
                return -1;
            }
        }
764 765
    } else if (tap->has_helper) {
        if (tap->has_ifname || tap->has_script || tap->has_downscript ||
J
Jason Wang 已提交
766
            tap->has_vnet_hdr || tap->has_queues || tap->has_fds) {
C
Corey Bryant 已提交
767
            error_report("ifname=, script=, downscript=, and vnet_hdr= "
J
Jason Wang 已提交
768
                         "queues=, and fds= are invalid with helper=");
C
Corey Bryant 已提交
769 770 771
            return -1;
        }

772
        fd = net_bridge_run_helper(tap->helper, DEFAULT_BRIDGE_INTERFACE);
C
Corey Bryant 已提交
773 774 775 776 777 778 779
        if (fd == -1) {
            return -1;
        }

        fcntl(fd, F_SETFL, O_NONBLOCK);
        vnet_hdr = tap_probe_vnet_hdr(fd);

J
Jason Wang 已提交
780 781 782 783 784
        if (net_init_tap_one(tap, peer, "bridge", name, ifname,
                             script, downscript, vhostfdname,
                             vnet_hdr, fd)) {
            return -1;
        }
785
    } else {
786
        script = tap->has_script ? tap->script : DEFAULT_NETWORK_SCRIPT;
787 788
        downscript = tap->has_downscript ? tap->downscript :
            DEFAULT_NETWORK_DOWN_SCRIPT;
J
Jason Wang 已提交
789 790 791 792 793

        if (tap->has_ifname) {
            pstrcpy(ifname, sizeof ifname, tap->ifname);
        } else {
            ifname[0] = '\0';
794
        }
C
Corey Bryant 已提交
795

J
Jason Wang 已提交
796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816
        for (i = 0; i < queues; i++) {
            fd = net_tap_init(tap, &vnet_hdr, i >= 1 ? "no" : script,
                              ifname, sizeof ifname, queues > 1);
            if (fd == -1) {
                return -1;
            }

            if (queues > 1 && i == 0 && !tap->has_ifname) {
                if (tap_fd_get_ifname(fd, ifname)) {
                    error_report("Fail to get ifname");
                    return -1;
                }
            }

            if (net_init_tap_one(tap, peer, "tap", name, ifname,
                                 i >= 1 ? "no" : script,
                                 i >= 1 ? "no" : downscript,
                                 vhostfdname, vnet_hdr, fd)) {
                return -1;
            }
        }
817 818
    }

J
Jason Wang 已提交
819
    return 0;
820
}
821

822
VHostNetState *tap_get_vhost_net(NetClientState *nc)
823 824
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
825
    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_TAP);
826 827
    return s->vhost_net;
}
828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862

int tap_enable(NetClientState *nc)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
    int ret;

    if (s->enabled) {
        return 0;
    } else {
        ret = tap_fd_enable(s->fd);
        if (ret == 0) {
            s->enabled = true;
            tap_update_fd_handler(s);
        }
        return ret;
    }
}

int tap_disable(NetClientState *nc)
{
    TAPState *s = DO_UPCAST(TAPState, nc, nc);
    int ret;

    if (s->enabled == 0) {
        return 0;
    } else {
        ret = tap_fd_disable(s->fd);
        if (ret == 0) {
            qemu_purge_queued_packets(nc);
            s->enabled = false;
            tap_update_fd_handler(s);
        }
        return ret;
    }
}