virnetdev.c 85.7 KB
Newer Older
1
/*
2
 * Copyright (C) 2007-2016 Red Hat, Inc.
3 4 5 6 7 8 9 10 11 12 13 14
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * This library 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
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
15
 * License along with this library.  If not, see
O
Osier Yang 已提交
16
 * <http://www.gnu.org/licenses/>.
17 18 19 20 21 22 23 24 25
 *
 * Authors:
 *     Mark McLoughlin <markmc@redhat.com>
 *     Daniel P. Berrange <berrange@redhat.com>
 */

#include <config.h>

#include "virnetdev.h"
26
#include "virnetlink.h"
27
#include "virmacaddr.h"
28
#include "virfile.h"
29
#include "virerror.h"
30
#include "vircommand.h"
31
#include "viralloc.h"
32
#include "virpci.h"
33
#include "virlog.h"
34 35
#include "virstring.h"
#include "virutil.h"
36
#include "virjson.h"
37 38

#include <sys/ioctl.h>
39
#include <net/if.h>
40
#include <fcntl.h>
41

42 43 44
#ifdef __linux__
# include <linux/sockios.h>
# include <linux/if_vlan.h>
45
# define VIR_NETDEV_FAMILY AF_UNIX
46 47 48
#elif defined(HAVE_STRUCT_IFREQ) && defined(AF_LOCAL)
# define VIR_NETDEV_FAMILY AF_LOCAL
#else
E
Eric Blake 已提交
49
# undef HAVE_STRUCT_IFREQ
50 51
#endif

J
Ján Tomko 已提交
52
#if defined(SIOCETHTOOL) && defined(HAVE_STRUCT_IFREQ)
53
# include <linux/types.h>
J
Ján Tomko 已提交
54 55 56
# include <linux/ethtool.h>
#endif

57 58 59 60 61
#if HAVE_DECL_LINK_ADDR
# include <sys/sockio.h>
# include <net/if_dl.h>
#endif

P
Pavel Hrdina 已提交
62 63 64 65
#ifndef IFNAMSIZ
# define IFNAMSIZ 16
#endif

66 67
#define VIR_FROM_THIS VIR_FROM_NONE

68 69
VIR_LOG_INIT("util.netdev");

70 71 72 73 74 75
#define PROC_NET_DEV_MCAST "/proc/net/dev_mcast"
#define MAX_MCAST_SIZE 50*14336
#define VIR_MCAST_NAME_LEN (IFNAMSIZ + 1)
#define VIR_MCAST_TOKEN_DELIMS " \n"
#define VIR_MCAST_ADDR_LEN (VIR_MAC_HEXLEN + 1)

76 77 78 79 80 81 82 83 84 85 86 87
#if defined(SIOCSIFFLAGS) && defined(HAVE_STRUCT_IFREQ)
# define VIR_IFF_UP IFF_UP
# define VIR_IFF_PROMISC IFF_PROMISC
# define VIR_IFF_MULTICAST IFF_MULTICAST
# define VIR_IFF_ALLMULTI IFF_ALLMULTI
#else
# define VIR_IFF_UP 0
# define VIR_IFF_PROMISC 0
# define VIR_IFF_MULTICAST 0
# define VIR_IFF_ALLMULTI 0
#endif

88 89 90 91 92 93 94 95 96 97
#define RESOURCE_FILE_LEN 4096
#if HAVE_DECL_ETHTOOL_GFEATURES
# define TX_UDP_TNL 25
# define GFEATURES_SIZE 2
# define FEATURE_WORD(blocks, index, field)  ((blocks)[(index) / 32U].field)
# define FEATURE_FIELD_FLAG(index)      (1U << (index) % 32U)
# define FEATURE_BIT_IS_SET(blocks, index, field)        \
    (FEATURE_WORD(blocks, index, field) & FEATURE_FIELD_FLAG(index))
#endif

J
John Ferlan 已提交
98 99 100 101 102 103 104 105 106 107
typedef enum {
    VIR_MCAST_TYPE_INDEX_TOKEN,
    VIR_MCAST_TYPE_NAME_TOKEN,
    VIR_MCAST_TYPE_USERS_TOKEN,
    VIR_MCAST_TYPE_GLOBAL_TOKEN,
    VIR_MCAST_TYPE_ADDR_TOKEN,

    VIR_MCAST_TYPE_LAST
} virMCastType;

108 109 110
typedef struct _virNetDevMcastEntry virNetDevMcastEntry;
typedef virNetDevMcastEntry *virNetDevMcastEntryPtr;
struct _virNetDevMcastEntry  {
J
John Ferlan 已提交
111
        int idx;
112 113 114 115 116 117 118 119 120 121 122 123 124
        char name[VIR_MCAST_NAME_LEN];
        int users;
        bool global;
        virMacAddr macaddr;
};

typedef struct _virNetDevMcastList virNetDevMcastList;
typedef virNetDevMcastList *virNetDevMcastListPtr;
struct _virNetDevMcastList {
    size_t nentries;
    virNetDevMcastEntryPtr *entries;
};

125
#if defined(HAVE_STRUCT_IFREQ)
126 127 128 129 130 131 132
static int virNetDevSetupControlFull(const char *ifname,
                                     struct ifreq *ifr,
                                     int domain,
                                     int type)
{
    int fd;

133 134
    if (ifr && ifname) {
        memset(ifr, 0, sizeof(*ifr));
135

136 137 138 139 140 141
        if (virStrcpyStatic(ifr->ifr_name, ifname) == NULL) {
            virReportSystemError(ERANGE,
                                 _("Network interface name '%s' is too long"),
                                 ifname);
            return -1;
        }
142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160
    }

    if ((fd = socket(domain, type, 0)) < 0) {
        virReportSystemError(errno, "%s",
                             _("Cannot open network interface control socket"));
        return -1;
    }

    if (virSetInherit(fd, false) < 0) {
        virReportSystemError(errno, "%s",
                             _("Cannot set close-on-exec flag for socket"));
        VIR_FORCE_CLOSE(fd);
        return -1;
    }

    return fd;
}


161 162 163
int
virNetDevSetupControl(const char *ifname,
                      struct ifreq *ifr)
164
{
165
    return virNetDevSetupControlFull(ifname, ifr, VIR_NETDEV_FAMILY, SOCK_DGRAM);
166
}
167 168 169 170 171 172 173 174 175 176 177
#else /* !HAVE_STRUCT_IFREQ */
int
virNetDevSetupControl(const char *ifname ATTRIBUTE_UNUSED,
                      void *ifr ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Network device configuration is not supported "
                           "on this platform"));
    return -1;
}
#endif /* HAVE_STRUCT_IFREQ */
178 179


180
#if defined(SIOCGIFFLAGS) && defined(HAVE_STRUCT_IFREQ)
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
/**
 * virNetDevExists:
 * @ifname
 *
 * Check if the network device @ifname exists
 *
 * Returns 1 if it exists, 0 if it does not, -1 on error
 */
int virNetDevExists(const char *ifname)
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    if (ioctl(fd, SIOCGIFFLAGS, &ifr)) {
199
        if (errno == ENODEV || errno == ENXIO)
200 201 202 203 204 205 206 207 208
            ret = 0;
        else
            virReportSystemError(errno,
                                 _("Unable to check interface flags for %s"), ifname);
        goto cleanup;
    }

    ret = 1;

209
 cleanup:
210 211 212 213 214 215 216 217 218 219 220 221 222
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
int virNetDevExists(const char *ifname)
{
    virReportSystemError(ENOSYS,
                         _("Unable to check interface %s"), ifname);
    return -1;
}
#endif


223 224
#if defined(SIOCGIFHWADDR) && defined(SIOCSIFHWADDR) && \
    defined(HAVE_STRUCT_IFREQ)
225
/**
226
 * virNetDevSetMACInternal:
227
 * @ifname: interface name to set MTU for
228
 * @macaddr: MAC address
229 230
 * @quiet: true if a failure to set MAC address with errno == EADDRNOTAVAIL
 *         should be silent (still returns error, but without log)
231
 *
232
 * This function sets the @macaddr for a given interface @ifname.
233 234 235
 *
 * Returns 0 in case of success or -1 on failure
 */
236 237 238 239
static int
virNetDevSetMACInternal(const char *ifname,
                        const virMacAddr *macaddr,
                        bool quiet)
240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    /* To fill ifr.ifr_hdaddr.sa_family field */
    if (ioctl(fd, SIOCGIFHWADDR, &ifr) < 0) {
        virReportSystemError(errno,
                             _("Cannot get interface MAC on '%s'"),
                             ifname);
        goto cleanup;
    }

256
    virMacAddrGetRaw(macaddr, (unsigned char *)ifr.ifr_hwaddr.sa_data);
257 258

    if (ioctl(fd, SIOCSIFHWADDR, &ifr) < 0) {
259 260
        char macstr[VIR_MAC_STRING_BUFLEN];

261 262 263
        if (quiet && errno == EADDRNOTAVAIL)
            goto cleanup;

264
        virReportSystemError(errno,
265 266
                             _("Cannot set interface MAC to %s on '%s'"),
                             virMacAddrFormat(macaddr, macstr), ifname);
267 268 269 270 271
        goto cleanup;
    }

    ret = 0;

272
 cleanup:
273 274 275
    VIR_FORCE_CLOSE(fd);
    return ret;
}
276 277 278


#elif defined(SIOCSIFLLADDR) && defined(HAVE_STRUCT_IFREQ) &&   \
279
    HAVE_DECL_LINK_ADDR
280 281 282 283 284 285


static int
virNetDevSetMACInternal(const char *ifname,
                        const virMacAddr *macaddr,
                        bool quiet)
286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303
{
        struct ifreq ifr;
        struct sockaddr_dl sdl;
        char mac[VIR_MAC_STRING_BUFLEN + 1] = ":";
        int s;
        int ret = -1;

        if ((s = virNetDevSetupControl(ifname, &ifr)) < 0)
            return -1;

        virMacAddrFormat(macaddr, mac + 1);
        sdl.sdl_len = sizeof(sdl);
        link_addr(mac, &sdl);

        memcpy(ifr.ifr_addr.sa_data, sdl.sdl_data, VIR_MAC_BUFLEN);
        ifr.ifr_addr.sa_len = VIR_MAC_BUFLEN;

        if (ioctl(s, SIOCSIFLLADDR, &ifr) < 0) {
304 305 306
            if (quiet && errno == EADDRNOTAVAIL)
                goto cleanup;

307
            virReportSystemError(errno,
308 309
                                 _("Cannot set interface MAC to %s on '%s'"),
                                 mac + 1, ifname);
310 311 312 313
            goto cleanup;
        }

        ret = 0;
314
 cleanup:
315 316 317 318
        VIR_FORCE_CLOSE(s);

        return ret;
}
319 320


321
#else
322 323 324 325 326 327


static int
virNetDevSetMACInternal(const char *ifname,
                        const virMacAddr *macaddr ATTRIBUTE_UNUSED,
                        bool quiet ATTRIBUTE_UNUSED)
328 329 330 331 332 333
{
    virReportSystemError(ENOSYS,
                         _("Cannot set interface MAC on '%s'"),
                         ifname);
    return -1;
}
334 335


336 337 338
#endif


339 340 341 342 343 344 345 346
int
virNetDevSetMAC(const char *ifname,
                const virMacAddr *macaddr)
{
    return virNetDevSetMACInternal(ifname, macaddr, false);
}


347
#if defined(SIOCGIFHWADDR) && defined(HAVE_STRUCT_IFREQ)
348 349 350
/**
 * virNetDevGetMAC:
 * @ifname: interface name to set MTU for
351
 * @macaddr: MAC address
352 353 354 355 356 357
 *
 * This function gets the @macaddr for a given interface @ifname.
 *
 * Returns 0 in case of success or -1 on failure
 */
int virNetDevGetMAC(const char *ifname,
358
                    virMacAddrPtr macaddr)
359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    if (ioctl(fd, SIOCGIFHWADDR, &ifr) < 0) {
        virReportSystemError(errno,
                             _("Cannot get interface MAC on '%s'"),
                             ifname);
        goto cleanup;
    }

374
    virMacAddrSetRaw(macaddr, (unsigned char *)ifr.ifr_hwaddr.sa_data);
375 376 377

    ret = 0;

378
 cleanup:
379 380 381 382 383
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
int virNetDevGetMAC(const char *ifname,
384
                    virMacAddrPtr macaddr ATTRIBUTE_UNUSED)
385 386 387 388 389 390 391 392 393
{
    virReportSystemError(ENOSYS,
                         _("Cannot get interface MAC on '%s'"),
                         ifname);
    return -1;
}
#endif


394
#if defined(SIOCGIFMTU) && defined(HAVE_STRUCT_IFREQ)
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
/**
 * virNetDevGetMTU:
 * @ifname: interface name get MTU for
 *
 * This function gets the @mtu value set for a given interface @ifname.
 *
 * Returns the MTU value in case of success, or -1 on failure.
 */
int virNetDevGetMTU(const char *ifname)
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    if (ioctl(fd, SIOCGIFMTU, &ifr) < 0) {
        virReportSystemError(errno,
                             _("Cannot get interface MTU on '%s'"),
                             ifname);
        goto cleanup;
    }

    ret = ifr.ifr_mtu;

421
 cleanup:
422 423 424 425 426 427 428 429 430 431 432 433 434 435
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
int virNetDevGetMTU(const char *ifname)
{
    virReportSystemError(ENOSYS,
                         _("Cannot get interface MTU on '%s'"),
                         ifname);
    return -1;
}
#endif


436
#if defined(SIOCSIFMTU) && defined(HAVE_STRUCT_IFREQ)
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
/**
 * virNetDevSetMTU:
 * @ifname: interface name to set MTU for
 * @mtu: MTU value
 *
 * This function sets the @mtu for a given interface @ifname.  Typically
 * used on a tap device to set up for Jumbo Frames.
 *
 * Returns 0 in case of success, or -1 on failure
 */
int virNetDevSetMTU(const char *ifname, int mtu)
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    ifr.ifr_mtu = mtu;

    if (ioctl(fd, SIOCSIFMTU, &ifr) < 0) {
        virReportSystemError(errno,
                             _("Cannot set interface MTU on '%s'"),
                             ifname);
        goto cleanup;
    }

    ret = 0;

467
 cleanup:
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
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
int virNetDevSetMTU(const char *ifname, int mtu ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS,
                         _("Cannot set interface MTU on '%s'"),
                         ifname);
    return -1;
}
#endif


/**
 * virNetDevSetMTUFromDevice:
 * @ifname: name of the interface whose MTU we want to set
 * @otherifname: name of the interface whose MTU we want to copy
 *
 * Sets the interface mtu to the same MTU as another interface
 *
 * Returns 0 in case of success, or -1 on failure
 */
int virNetDevSetMTUFromDevice(const char *ifname,
                              const char *otherifname)
{
    int mtu = virNetDevGetMTU(otherifname);

    if (mtu < 0)
        return -1;

    return virNetDevSetMTU(ifname, mtu);
}


503 504 505 506 507 508 509 510 511 512 513
/**
 * virNetDevSetNamespace:
 * @ifname: name of device
 * @pidInNs: PID of process in target net namespace
 *
 * Moves the given device into the target net namespace specified by the given
 * pid using this command:
 *     ip link set @iface netns @pidInNs
 *
 * Returns 0 on success or -1 in case of error
 */
514
int virNetDevSetNamespace(const char *ifname, pid_t pidInNs)
515
{
516
    int ret = -1;
517
    char *pid = NULL;
518 519 520 521
    char *phy = NULL;
    char *phy_path = NULL;
    int len;

522
    if (virAsprintf(&pid, "%lld", (long long) pidInNs) == -1)
523 524
        return -1;

525 526 527 528
    /* The 802.11 wireless devices only move together with their PHY. */
    if (virNetDevSysfsFile(&phy_path, ifname, "phy80211/name") < 0)
        goto cleanup;

529 530 531 532 533 534 535 536 537
    if ((len = virFileReadAllQuiet(phy_path, 1024, &phy)) <= 0) {
        /* Not a wireless device. */
        const char *argv[] = {
            "ip", "link", "set", ifname, "netns", NULL, NULL
        };

        argv[5] = pid;
        if (virRun(argv, NULL) < 0)
            goto cleanup;
538

539 540 541 542
    } else {
        const char *argv[] = {
            "iw", "phy", NULL, "set", "netns", NULL, NULL
        };
543

544 545 546 547 548 549 550 551
        /* Remove a line break. */
        phy[len - 1] = '\0';

        argv[2] = phy;
        argv[5] = pid;
        if (virRun(argv, NULL) < 0)
            goto cleanup;
    }
552 553 554 555 556

    ret = 0;
 cleanup:
    VIR_FREE(phy_path);
    VIR_FREE(phy);
557
    VIR_FREE(pid);
558
    return ret;
559 560
}

561
#if defined(SIOCSIFNAME) && defined(HAVE_STRUCT_IFREQ)
562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579
/**
 * virNetDevSetName:
 * @ifname: name of device
 * @newifname: new name of @ifname
 *
 * Changes the name of the given device.
 *
 * Returns 0 on success, -1 on error
 */
int virNetDevSetName(const char* ifname, const char *newifname)
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

580
# ifdef HAVE_STRUCT_IFREQ_IFR_NEWNAME
581 582 583 584 585 586
    if (virStrcpyStatic(ifr.ifr_newname, newifname) == NULL) {
        virReportSystemError(ERANGE,
                             _("Network interface name '%s' is too long"),
                             newifname);
        goto cleanup;
    }
587 588 589
# else
    ifr.ifr_data = (caddr_t)newifname;
# endif
590 591 592 593 594 595 596 597 598 599

    if (ioctl(fd, SIOCSIFNAME, &ifr)) {
        virReportSystemError(errno,
                             _("Unable to rename '%s' to '%s'"),
                             ifname, newifname);
        goto cleanup;
    }

    ret = 0;

600
 cleanup:
601 602 603 604 605 606 607 608 609 610 611 612 613 614
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
int virNetDevSetName(const char* ifname, const char *newifname)
{
    virReportSystemError(ENOSYS,
                         _("Cannot rename interface '%s' to '%s' on this platform"),
                         ifname, newifname);
    return -1;
}
#endif


615
#if defined(SIOCSIFFLAGS) && defined(HAVE_STRUCT_IFREQ)
616 617
static int
virNetDevSetIFFlag(const char *ifname, int flag, bool val)
618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;
    int ifflags;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    if (ioctl(fd, SIOCGIFFLAGS, &ifr) < 0) {
        virReportSystemError(errno,
                             _("Cannot get interface flags on '%s'"),
                             ifname);
        goto cleanup;
    }

634 635
    if (val)
        ifflags = ifr.ifr_flags | flag;
636
    else
637
        ifflags = ifr.ifr_flags & ~flag;
638 639 640 641 642 643 644 645 646 647 648 649 650

    if (ifr.ifr_flags != ifflags) {
        ifr.ifr_flags = ifflags;
        if (ioctl(fd, SIOCSIFFLAGS, &ifr) < 0) {
            virReportSystemError(errno,
                                 _("Cannot set interface flags on '%s'"),
                                 ifname);
            goto cleanup;
        }
    }

    ret = 0;

651
 cleanup:
652 653 654 655
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
656 657 658 659
static int
virNetDevSetIFFlag(const char *ifname,
                   int flag ATTRIBUTE_UNUSED,
                   bool val ATTRIBUTE_UNUSED)
660 661 662 663 664 665 666 667 668
{
    virReportSystemError(ENOSYS,
                         _("Cannot set interface flags on '%s'"),
                         ifname);
    return -1;
}
#endif


669

670
/**
671
 * virNetDevSetOnline:
672
 * @ifname: the interface name
673
 * @online: true for up, false for down
674
 *
675
 * Function to control if an interface is activated (up, true) or not (down, false)
676
 *
677 678
 * Returns 0 in case of success or -1 on error.
 */
679 680 681
int
virNetDevSetOnline(const char *ifname,
                   bool online)
682 683
{

684
    return virNetDevSetIFFlag(ifname, VIR_IFF_UP, online);
685 686 687 688 689 690 691 692 693 694 695 696 697
}

/**
 * virNetDevSetPromiscuous:
 * @ifname: the interface name
 * @promiscuous: true for receive all packets, false for do not receive
 *               all packets
 *
 * Function to control if an interface is to receive all
 * packets (receive all, true) or not (do not receive all, false)
 *
 * Returns 0 in case of success or -1 on error.
 */
698 699 700
int
virNetDevSetPromiscuous(const char *ifname,
                        bool promiscuous)
701
{
702
    return virNetDevSetIFFlag(ifname, VIR_IFF_PROMISC, promiscuous);
703 704 705 706 707 708 709 710 711 712 713 714 715
}

/**
 * virNetDevSetRcvMulti:
 * @ifname: the interface name
 * @:receive true for receive multicast packets, false for do not receive
 *           multicast packets
 *
 * Function to control if an interface is to receive multicast
 * packets in which it is interested (receive, true)
 * or not (do not receive, false)
 *
 * Returns 0 in case of success or -1 on error.
716
 */
717 718 719
int
virNetDevSetRcvMulti(const char *ifname,
                     bool receive)
720
{
721
    return virNetDevSetIFFlag(ifname, VIR_IFF_MULTICAST, receive);
722 723 724 725 726 727 728 729 730 731 732 733
}

/**
 * virNetDevSetRcvAllMulti:
 * @ifname: the interface name
 * @:receive true for receive all packets, false for do not receive all packets
 *
 * Function to control if an interface is to receive all multicast
 * packets (receive, true) or not (do not receive, false)
 *
 * Returns 0 in case of success or -1 on error.
 */
734 735 736
int
virNetDevSetRcvAllMulti(const char *ifname,
                        bool receive)
737
{
738
    return virNetDevSetIFFlag(ifname, VIR_IFF_ALLMULTI, receive);
739 740 741 742
}


#if defined(SIOCGIFFLAGS) && defined(HAVE_STRUCT_IFREQ)
743 744
static int
virNetDevGetIFFlag(const char *ifname, int flag, bool *val)
745 746 747 748 749 750 751 752 753 754 755 756 757 758 759
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    if (ioctl(fd, SIOCGIFFLAGS, &ifr) < 0) {
        virReportSystemError(errno,
                             _("Cannot get interface flags on '%s'"),
                             ifname);
        goto cleanup;
    }

760
    *val = (ifr.ifr_flags & flag) ? true : false;
761 762
    ret = 0;

763
 cleanup:
764 765 766 767
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
768 769 770 771
static int
virNetDevGetIFFlag(const char *ifname,
                   int flag ATTRIBUTE_UNUSED,
                   bool *val ATTRIBUTE_UNUSED)
772 773 774 775 776 777 778 779 780
{
    virReportSystemError(ENOSYS,
                         _("Cannot get interface flags on '%s'"),
                         ifname);
    return -1;
}
#endif


781 782 783 784 785 786 787 788 789
/**
 * virNetDevGetOnline:
 * @ifname: the interface name
 * @online: where to store the status
 *
 * Function to query if an interface is activated (true) or not (false)
 *
 * Returns 0 in case of success or an errno code in case of failure.
 */
790 791 792
int
virNetDevGetOnline(const char *ifname,
                   bool *online)
793
{
794
    return virNetDevGetIFFlag(ifname, VIR_IFF_UP, online);
795 796 797 798 799 800 801 802 803 804 805 806
}

/**
 * virNetDevIsPromiscuous:
 * @ifname: the interface name
 * @promiscuous: where to store the status
 *
 * Function to query if an interface is receiving all packets (true) or
 * not (false)
 *
 * Returns 0 in case of success or an errno code in case of failure.
 */
807 808 809
int
virNetDevGetPromiscuous(const char *ifname,
                        bool *promiscuous)
810
{
811
    return virNetDevGetIFFlag(ifname, VIR_IFF_PROMISC, promiscuous);
812 813 814 815 816 817 818 819 820 821 822 823
}

/**
 * virNetDevIsRcvMulti:
 * @ifname: the interface name
 * @receive where to store the status
 *
 * Function to query whether an interface is receiving multicast packets (true)
 * in which it is interested, or not (false)
 *
 * Returns 0 in case of success or -1 on error.
 */
824 825 826
int
virNetDevGetRcvMulti(const char *ifname,
                     bool *receive)
827
{
828
    return virNetDevGetIFFlag(ifname, VIR_IFF_MULTICAST, receive);
829 830 831 832 833 834 835 836 837 838 839 840
}

/**
 * virNetDevIsRcvAllMulti:
 * @ifname: the interface name
 * @:receive where to store the status
 *
 * Function to query whether an interface is receiving all multicast
 * packets (receiving, true) or not (is not receiving, false)
 *
 * Returns 0 in case of success or -1 on error.
 */
841 842 843
int
virNetDevGetRcvAllMulti(const char *ifname,
                        bool *receive)
844
{
845
    return virNetDevGetIFFlag(ifname, VIR_IFF_ALLMULTI, receive);
846 847
}

848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866
char *virNetDevGetName(int ifindex)
{
    char name[IFNAMSIZ];
    char *ifname = NULL;

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

    if (!if_indextoname(ifindex, name)) {
        virReportSystemError(errno,
                             _("Failed to convert interface index %d to a name"),
                             ifindex);
        goto cleanup;
    }

   ignore_value(VIR_STRDUP(ifname, name));

 cleanup:
     return ifname;
}
867

868 869 870 871 872 873 874 875 876
/**
 * virNetDevGetIndex:
 * @ifname : Name of the interface whose index is to be found
 * @ifindex: Pointer to int where the index will be written into
 *
 * Get the index of an interface given its name.
 *
 * Returns 0 on success, -1 on failure
 */
877
#if defined(SIOCGIFINDEX) && defined(HAVE_STRUCT_IFREQ)
878 879 880 881
int virNetDevGetIndex(const char *ifname, int *ifindex)
{
    int ret = -1;
    struct ifreq ifreq;
882
    int fd = socket(VIR_NETDEV_FAMILY, SOCK_DGRAM, 0);
883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905

    if (fd < 0) {
        virReportSystemError(errno, "%s",
                             _("Unable to open control socket"));
        return -1;
    }

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

    if (virStrncpy(ifreq.ifr_name, ifname, strlen(ifname),
                   sizeof(ifreq.ifr_name)) == NULL) {
        virReportSystemError(ERANGE,
                             _("invalid interface name %s"),
                             ifname);
        goto cleanup;
    }

    if (ioctl(fd, SIOCGIFINDEX, &ifreq) < 0) {
        virReportSystemError(errno,
                             _("Unable to get index for interface %s"), ifname);
        goto cleanup;
    }

906 907 908
# ifdef HAVE_STRUCT_IFREQ_IFR_INDEX
    *ifindex = ifreq.ifr_index;
# else
909
    *ifindex = ifreq.ifr_ifindex;
910
# endif
911 912
    ret = 0;

913
 cleanup:
914 915 916
    VIR_FORCE_CLOSE(fd);
    return ret;
}
917
#else /* ! SIOCGIFINDEX */
918 919 920 921 922 923 924
int virNetDevGetIndex(const char *ifname ATTRIBUTE_UNUSED,
                      int *ifindex ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to get interface index on this platform"));
    return -1;
}
925
#endif /* ! SIOCGIFINDEX */
926 927


928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965
#if defined(__linux__) && defined(HAVE_LIBNL)
/**
 * virNetDevGetMaster:
 * @ifname: name of interface we're interested in
 * @master: used to return a string containing the name of @ifname's "master"
 *          (this is the bridge or bond device that this device is attached to)
 *
 * Returns 0 on success, -1 on failure (if @ifname has no master
 * @master will be NULL, but return value will still be 0 (success)).
 */
int
virNetDevGetMaster(const char *ifname, char **master)
{
    int ret = -1;
    void *nlData = NULL;
    struct nlattr *tb[IFLA_MAX + 1] = {NULL, };

    *master = NULL;

    if (virNetlinkDumpLink(ifname, -1, &nlData, tb, 0, 0) < 0)
        goto cleanup;

    if (tb[IFLA_MASTER]) {
        if (!(*master = virNetDevGetName(*(int *)RTA_DATA(tb[IFLA_MASTER]))))
            goto cleanup;
    }

    ret = 0;
 cleanup:
    VIR_FREE(nlData);
    return ret;
}


#else


int
R
Roman Bogorodskiy 已提交
966 967
virNetDevGetMaster(const char *ifname ATTRIBUTE_UNUSED,
                   char **master ATTRIBUTE_UNUSED)
968 969 970 971 972 973 974 975 976 977
{
    virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                   _("Unable to get device master from netlink on this platform"));
    return -1;
}


#endif /* defined(__linux__) && defined(HAVE_LIBNL) */


978
#if defined(SIOCGIFVLAN) && defined(HAVE_STRUCT_IFREQ) && HAVE_DECL_GET_VLAN_VID_CMD
979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013
int virNetDevGetVLanID(const char *ifname, int *vlanid)
{
    struct vlan_ioctl_args vlanargs = {
      .cmd = GET_VLAN_VID_CMD,
    };
    int ret = -1;
    int fd = socket(PF_PACKET, SOCK_DGRAM, 0);

    if (fd < 0) {
        virReportSystemError(errno, "%s",
                             _("Unable to open control socket"));
        return -1;
    }

    if (virStrcpyStatic(vlanargs.device1, ifname) == NULL) {
        virReportSystemError(ERANGE,
                             _("invalid interface name %s"),
                             ifname);
        goto cleanup;
    }

    if (ioctl(fd, SIOCGIFVLAN, &vlanargs) != 0) {
        virReportSystemError(errno,
                             _("Unable to get VLAN for interface %s"), ifname);
        goto cleanup;
    }

    *vlanid = vlanargs.u.VID;
    ret = 0;

 cleanup:
    VIR_FORCE_CLOSE(fd);

    return ret;
}
1014
#else /* ! SIOCGIFVLAN */
1015 1016 1017 1018 1019 1020 1021
int virNetDevGetVLanID(const char *ifname ATTRIBUTE_UNUSED,
                       int *vlanid ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to get VLAN on this platform"));
    return -1;
}
1022
#endif /* ! SIOCGIFVLAN */
1023 1024


1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036
/**
 * virNetDevValidateConfig:
 * @ifname: Name of the interface
 * @macaddr: expected MAC address of the interface; not checked if NULL
 * @ifindex: expected index of the interface; not checked if '-1'
 *
 * Determine whether a given interface is still available. If so,
 * it must have the given MAC address and if an interface index is
 * passed, it must also match the interface index.
 *
 * Returns 1 if the config matches, 0 if the config does not match, or interface does not exist, -1 on error
 */
1037
#if defined(SIOCGIFHWADDR) && defined(HAVE_STRUCT_IFREQ)
1038
int virNetDevValidateConfig(const char *ifname,
1039
                            const virMacAddr *macaddr, int ifindex)
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;
    int idx;
    int rc;

    if ((rc = virNetDevExists(ifname)) < 0)
        return -1;
    if (rc == 0) {
        ret = 0;
        goto cleanup;
    }

    if (macaddr != NULL) {
        if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
            return -1;

        if (ioctl(fd, SIOCGIFHWADDR, &ifr) < 0) {
            if (errno == ENODEV) {
                ret = 0;
                goto cleanup;
            }
            virReportSystemError(errno,
Y
Yuri Chornoivan 已提交
1064
                                 _("could not get MAC address of interface %s"),
1065 1066 1067 1068
                                 ifname);
            goto cleanup;
        }

1069 1070
        if (virMacAddrCmpRaw(macaddr,
                             (unsigned char *)ifr.ifr_hwaddr.sa_data) != 0) {
1071 1072 1073 1074 1075 1076 1077 1078
            ret = 0;
            goto cleanup;
        }
    }

    if (ifindex != -1) {
        if (virNetDevGetIndex(ifname, &idx) < 0)
            goto cleanup;
1079
        if (idx != ifindex) {
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090
            ret = 0;
            goto cleanup;
        }
    }

    ret = 1;

 cleanup:
    VIR_FORCE_CLOSE(fd);
    return ret;
}
1091
#else
1092
int virNetDevValidateConfig(const char *ifname ATTRIBUTE_UNUSED,
1093
                            const virMacAddr *macaddr ATTRIBUTE_UNUSED,
1094 1095 1096 1097 1098 1099
                            int ifindex ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to check interface config on this platform"));
    return -1;
}
1100
#endif
1101 1102 1103 1104


#ifdef __linux__

M
Michal Privoznik 已提交
1105
int
1106
virNetDevSysfsFile(char **pf_sysfs_device_link, const char *ifname,
M
Michal Privoznik 已提交
1107
                   const char *file)
1108 1109
{

1110
    if (virAsprintf(pf_sysfs_device_link, SYSFS_NET_DIR "%s/%s", ifname, file) < 0)
1111 1112
        return -1;
    return 0;
1113 1114 1115 1116 1117 1118 1119
}

static int
virNetDevSysfsDeviceFile(char **pf_sysfs_device_link, const char *ifname,
                     const char *file)
{

1120
    if (virAsprintf(pf_sysfs_device_link, SYSFS_NET_DIR "%s/device/%s", ifname,
1121 1122 1123
                    file) < 0)
        return -1;
    return 0;
1124 1125
}

1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151

static virPCIDevicePtr
virNetDevGetPCIDevice(const char *devName)
{
    char *vfSysfsDevicePath = NULL;
    virPCIDeviceAddressPtr vfPCIAddr = NULL;
    virPCIDevicePtr vfPCIDevice = NULL;

    if (virNetDevSysfsFile(&vfSysfsDevicePath, devName, "device") < 0)
        goto cleanup;

    vfPCIAddr = virPCIGetDeviceAddressFromSysfsLink(vfSysfsDevicePath);
    if (!vfPCIAddr)
        goto cleanup;

    vfPCIDevice = virPCIDeviceNew(vfPCIAddr->domain, vfPCIAddr->bus,
                                  vfPCIAddr->slot, vfPCIAddr->function);

 cleanup:
    VIR_FREE(vfSysfsDevicePath);
    VIR_FREE(vfPCIAddr);

    return vfPCIDevice;
}


1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164
/**
 * virNetDevGetVirtualFunctions:
 *
 * @pfname : name of the physical function interface name
 * @vfname: array that will hold the interface names of the virtual_functions
 * @n_vfname: pointer to the number of virtual functions
 *
 * Returns 0 on success and -1 on failure
 */

int
virNetDevGetVirtualFunctions(const char *pfname,
                             char ***vfname,
1165
                             virPCIDeviceAddressPtr **virt_fns,
1166 1167
                             size_t *n_vfname,
                             unsigned int *max_vfs)
1168
{
1169 1170
    int ret = -1;
    size_t i;
1171 1172
    char *pf_sysfs_device_link = NULL;
    char *pci_sysfs_device_link = NULL;
1173
    char *pciConfigAddr = NULL;
1174

1175 1176
    *virt_fns = NULL;
    *n_vfname = 0;
1177
    *max_vfs = 0;
1178

1179 1180 1181
    if (virNetDevSysfsFile(&pf_sysfs_device_link, pfname, "device") < 0)
        return ret;

1182
    if (virPCIGetVirtualFunctions(pf_sysfs_device_link, virt_fns,
1183
                                  n_vfname, max_vfs) < 0)
1184 1185
        goto cleanup;

1186
    if (VIR_ALLOC_N(*vfname, *n_vfname) < 0)
1187 1188
        goto cleanup;

E
Eric Blake 已提交
1189
    for (i = 0; i < *n_vfname; i++) {
1190 1191 1192 1193 1194
        if (virPCIGetAddrString((*virt_fns)[i]->domain,
                                (*virt_fns)[i]->bus,
                                (*virt_fns)[i]->slot,
                                (*virt_fns)[i]->function,
                                &pciConfigAddr) < 0) {
1195 1196 1197 1198
            virReportSystemError(ENOSYS, "%s",
                                 _("Failed to get PCI Config Address String"));
            goto cleanup;
        }
1199
        if (virPCIGetSysfsFile(pciConfigAddr, &pci_sysfs_device_link) < 0) {
1200 1201 1202 1203 1204
            virReportSystemError(ENOSYS, "%s",
                                 _("Failed to get PCI SYSFS file"));
            goto cleanup;
        }

1205
        if (virPCIGetNetName(pci_sysfs_device_link, &((*vfname)[i])) < 0)
1206 1207 1208
            goto cleanup;

        if (!(*vfname)[i])
1209
            VIR_INFO("VF does not have an interface name");
1210 1211 1212 1213
    }

    ret = 0;

1214
 cleanup:
1215
    if (ret < 0) {
1216
        VIR_FREE(*vfname);
1217 1218
        VIR_FREE(*virt_fns);
    }
1219 1220 1221 1222 1223 1224
    VIR_FREE(pf_sysfs_device_link);
    VIR_FREE(pci_sysfs_device_link);
    VIR_FREE(pciConfigAddr);
    return ret;
}

1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242
/**
 * virNetDevIsVirtualFunction:
 * @ifname : name of the interface
 *
 * Checks if an interface is a SRIOV virtual function.
 *
 * Returns 1 if interface is SRIOV virtual function, 0 if not and -1 if error
 *
 */
int
virNetDevIsVirtualFunction(const char *ifname)
{
    char *if_sysfs_device_link = NULL;
    int ret = -1;

    if (virNetDevSysfsFile(&if_sysfs_device_link, ifname, "device") < 0)
        return ret;

1243
    ret = virPCIIsVirtualFunction(if_sysfs_device_link);
1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275

    VIR_FREE(if_sysfs_device_link);

    return ret;
}

/**
 * virNetDevGetVirtualFunctionIndex
 *
 * @pfname : name of the physical function interface name
 * @vfname : name of the virtual function interface name
 * @vf_index : Pointer to int. Contains vf index of interface upon successful
 *             return
 *
 * Returns 0 on success, -1 on failure
 *
 */
int
virNetDevGetVirtualFunctionIndex(const char *pfname, const char *vfname,
                                 int *vf_index)
{
    char *pf_sysfs_device_link = NULL, *vf_sysfs_device_link = NULL;
    int ret = -1;

    if (virNetDevSysfsFile(&pf_sysfs_device_link, pfname, "device") < 0)
        return ret;

    if (virNetDevSysfsFile(&vf_sysfs_device_link, vfname, "device") < 0) {
        VIR_FREE(pf_sysfs_device_link);
        return ret;
    }

1276 1277 1278
    ret = virPCIGetVirtualFunctionIndex(pf_sysfs_device_link,
                                        vf_sysfs_device_link,
                                        vf_index);
1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304

    VIR_FREE(pf_sysfs_device_link);
    VIR_FREE(vf_sysfs_device_link);

    return ret;
}

/**
 * virNetDevGetPhysicalFunction
 *
 * @ifname : name of the physical function interface name
 * @pfname : Contains sriov physical function for interface ifname
 *           upon successful return
 *
 * Returns 0 on success, -1 on failure
 *
 */
int
virNetDevGetPhysicalFunction(const char *ifname, char **pfname)
{
    char *physfn_sysfs_path = NULL;
    int ret = -1;

    if (virNetDevSysfsDeviceFile(&physfn_sysfs_path, ifname, "physfn") < 0)
        return ret;

1305 1306
    if (virPCIGetNetName(physfn_sysfs_path, pfname) < 0)
        goto cleanup;
1307

1308 1309 1310 1311 1312 1313 1314 1315 1316
    if (!*pfname) {
        /* this shouldn't be possible. A VF can't exist unless its
         * PF device is bound to a network driver
         */
        virReportError(VIR_ERR_INTERNAL_ERROR,
                       _("The PF device for VF %s has no network device name"),
                       ifname);
        goto cleanup;
    }
1317

1318 1319 1320
    ret = 0;
 cleanup:
    VIR_FREE(physfn_sysfs_path);
1321 1322
    return ret;
}
1323

1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374

/**
 * virNetDevPFGetVF:
 *
 * @pfname: netdev name of the physical function (PF)
 * @vf: virtual function (VF) number for the device of interest
 * @vfname: name of the physical function interface name
 *
 * Finds the netdev name of VF# @vf of SRIOV PF @pfname, and puts it
 * in @vfname. The caller must free @vfname when it's finished with
 * it.
 *
 * Returns 0 on success, -1 on failure
 *
 * NB: if the VF has no netdev name, that is *not* considered an
 * error; *vfname simply gets a NULL and the return value is 0
 * (success).
 */
int
virNetDevPFGetVF(const char *pfname, int vf, char **vfname)
{
    char *virtfnName = NULL;
    char *virtfnSysfsPath = NULL;
    int ret = -1;

    if (virAsprintf(&virtfnName, "virtfn%d", vf) < 0)
        goto cleanup;

    /* this provides the path to the VF's directory in sysfs,
     * e.g. "/sys/class/net/enp2s0f0/virtfn3"
     */
    if (virNetDevSysfsDeviceFile(&virtfnSysfsPath, pfname, virtfnName) < 0)
        goto cleanup;

    /* and this gets the netdev name associated with it, which is a
     * directory entry in [virtfnSysfsPath]/net,
     * e.g. "/sys/class/net/enp2s0f0/virtfn3/net/enp2s11f4" - in this
     * example the VF for enp2s0f0 vf#3 is "enp2s11f4". (If the VF
     * isn't bound to a netdev driver, it won't have a netdev name,
     * and vfname will be NULL).
     */
    ret = virPCIGetNetName(virtfnSysfsPath, vfname);

 cleanup:
    VIR_FREE(virtfnName);
    VIR_FREE(virtfnSysfsPath);

    return ret;
}


1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401
/**
 * virNetDevGetVirtualFunctionInfo:
 * @vfname: name of the virtual function interface
 * @pfname: name of the physical function
 * @vf: vf index
 *
 * Returns 0 on success, -errno on failure.
 *
 */
int
virNetDevGetVirtualFunctionInfo(const char *vfname, char **pfname,
                                int *vf)
{
    char *pf_sysfs_path = NULL, *vf_sysfs_path = NULL;
    int ret = -1;

    *pfname = NULL;

    if (virNetDevGetPhysicalFunction(vfname, pfname) < 0)
        return ret;

    if (virNetDevSysfsFile(&pf_sysfs_path, *pfname, "device") < 0)
        goto cleanup;

    if (virNetDevSysfsFile(&vf_sysfs_path, vfname, "device") < 0)
        goto cleanup;

1402
    ret = virPCIGetVirtualFunctionIndex(pf_sysfs_path, vf_sysfs_path, vf);
1403

1404
 cleanup:
1405 1406 1407 1408 1409 1410 1411 1412 1413 1414
    if (ret < 0)
        VIR_FREE(*pfname);

    VIR_FREE(vf_sysfs_path);
    VIR_FREE(pf_sysfs_path);

    return ret;
}

#else /* !__linux__ */
1415 1416 1417
int
virNetDevGetVirtualFunctions(const char *pfname ATTRIBUTE_UNUSED,
                             char ***vfname ATTRIBUTE_UNUSED,
1418
                             virPCIDeviceAddressPtr **virt_fns ATTRIBUTE_UNUSED,
1419 1420
                             size_t *n_vfname ATTRIBUTE_UNUSED,
                             unsigned int *max_vfs ATTRIBUTE_UNUSED)
1421 1422
{
    virReportSystemError(ENOSYS, "%s",
1423
                         _("Unable to get virtual functions on this platform"));
1424 1425 1426
    return -1;
}

1427 1428 1429 1430
int
virNetDevIsVirtualFunction(const char *ifname ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
Y
Yuri Chornoivan 已提交
1431
                         _("Unable to check virtual function status on this platform"));
1432 1433 1434 1435 1436
    return -1;
}

int
virNetDevGetVirtualFunctionIndex(const char *pfname ATTRIBUTE_UNUSED,
R
Roman Bogorodskiy 已提交
1437 1438
                                 const char *vfname ATTRIBUTE_UNUSED,
                                 int *vf_index ATTRIBUTE_UNUSED)
1439 1440
{
    virReportSystemError(ENOSYS, "%s",
Y
Yuri Chornoivan 已提交
1441
                         _("Unable to get virtual function index on this platform"));
1442 1443 1444 1445 1446 1447 1448 1449
    return -1;
}

int
virNetDevGetPhysicalFunction(const char *ifname ATTRIBUTE_UNUSED,
                             char **pfname ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
Y
Yuri Chornoivan 已提交
1450
                         _("Unable to get physical function status on this platform"));
1451 1452
    return -1;
}
1453

1454 1455 1456
int
virNetDevPFGetVF(const char *pfname ATTRIBUTE_UNUSED,
                 int vf ATTRIBUTE_UNUSED,
R
Roman Bogorodskiy 已提交
1457
                 char **vfname ATTRIBUTE_UNUSED)
1458 1459 1460 1461 1462 1463
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to get virtual function name on this platform"));
    return -1;
}

1464 1465 1466 1467 1468 1469 1470 1471 1472
int
virNetDevGetVirtualFunctionInfo(const char *vfname ATTRIBUTE_UNUSED,
                                char **pfname ATTRIBUTE_UNUSED,
                                int *vf ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to get virtual function info on this platform"));
    return -1;
}
1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483

int
virNetDevSysfsFile(char **pf_sysfs_device_link ATTRIBUTE_UNUSED,
                   const char *ifname ATTRIBUTE_UNUSED,
                   const char *file ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to get sysfs info on this platform"));
    return -1;
}

1484

1485
#endif /* !__linux__ */
1486
#if defined(__linux__) && defined(HAVE_LIBNL) && defined(IFLA_VF_MAX)
1487

R
Roman Bogorodskiy 已提交
1488

1489 1490 1491 1492 1493 1494 1495 1496
static virMacAddr zeroMAC = { .addr = { 0, 0, 0, 0, 0, 0 } };

/* if a net driver doesn't allow setting MAC to all 0, try setting
 * to this (the only bit that is set is the "locally administered" bit")
 */
static virMacAddr altZeroMAC = { .addr = { 2, 0, 0, 0, 0, 0 } };


1497 1498 1499 1500 1501 1502 1503 1504 1505
static struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = {
    [IFLA_VF_MAC]       = { .type = NLA_UNSPEC,
                            .maxlen = sizeof(struct ifla_vf_mac) },
    [IFLA_VF_VLAN]      = { .type = NLA_UNSPEC,
                            .maxlen = sizeof(struct ifla_vf_vlan) },
};


static int
1506
virNetDevSetVfConfig(const char *ifname, int vf,
1507 1508
                     const virMacAddr *macaddr, int vlanid,
                     bool *allowRetry)
1509 1510
{
    int rc = -1;
1511
    struct nlmsghdr *resp = NULL;
1512 1513 1514 1515 1516 1517
    struct nlmsgerr *err;
    unsigned int recvbuflen = 0;
    struct nl_msg *nl_msg;
    struct nlattr *vfinfolist, *vfinfo;
    struct ifinfomsg ifinfo = {
        .ifi_family = AF_UNSPEC,
1518
        .ifi_index  = -1,
1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549
    };

    if (!macaddr && vlanid < 0)
        return -1;

    nl_msg = nlmsg_alloc_simple(RTM_SETLINK, NLM_F_REQUEST);
    if (!nl_msg) {
        virReportOOMError();
        return rc;
    }

    if (nlmsg_append(nl_msg,  &ifinfo, sizeof(ifinfo), NLMSG_ALIGNTO) < 0)
        goto buffer_too_small;

    if (ifname &&
        nla_put(nl_msg, IFLA_IFNAME, strlen(ifname)+1, ifname) < 0)
        goto buffer_too_small;


    if (!(vfinfolist = nla_nest_start(nl_msg, IFLA_VFINFO_LIST)))
        goto buffer_too_small;

    if (!(vfinfo = nla_nest_start(nl_msg, IFLA_VF_INFO)))
        goto buffer_too_small;

    if (macaddr) {
        struct ifla_vf_mac ifla_vf_mac = {
             .vf = vf,
             .mac = { 0, },
        };

1550
        virMacAddrGetRaw(macaddr, ifla_vf_mac.mac);
1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571

        if (nla_put(nl_msg, IFLA_VF_MAC, sizeof(ifla_vf_mac),
                    &ifla_vf_mac) < 0)
            goto buffer_too_small;
    }

    if (vlanid >= 0) {
        struct ifla_vf_vlan ifla_vf_vlan = {
             .vf = vf,
             .vlan = vlanid,
             .qos = 0,
        };

        if (nla_put(nl_msg, IFLA_VF_VLAN, sizeof(ifla_vf_vlan),
                    &ifla_vf_vlan) < 0)
            goto buffer_too_small;
    }

    nla_nest_end(nl_msg, vfinfo);
    nla_nest_end(nl_msg, vfinfolist);

1572
    if (virNetlinkCommand(nl_msg, &resp, &recvbuflen, 0, 0,
1573
                          NETLINK_ROUTE, 0) < 0)
1574 1575
        goto cleanup;

1576
    if (recvbuflen < NLMSG_LENGTH(0) || resp == NULL)
1577 1578 1579 1580 1581 1582 1583 1584
        goto malformed_resp;

    switch (resp->nlmsg_type) {
    case NLMSG_ERROR:
        err = (struct nlmsgerr *)NLMSG_DATA(resp);
        if (resp->nlmsg_len < NLMSG_LENGTH(sizeof(*err)))
            goto malformed_resp;

1585 1586 1587 1588 1589 1590 1591 1592 1593
        /* if allowRetry is true and the error was EINVAL, then
         * silently return a failure so the caller can retry with a
         * different MAC address
         */
        if (err->error == -EINVAL && *allowRetry &&
            macaddr && !virMacAddrCmp(macaddr, &zeroMAC)) {
            goto cleanup;
        } else if (err->error) {
            /* other errors are permanent */
1594 1595
            char macstr[VIR_MAC_STRING_BUFLEN];

1596
            virReportSystemError(-err->error,
1597
                                 _("Cannot set interface MAC/vlanid to %s/%d "
1598
                                   "for ifname %s vf %d"),
1599 1600 1601 1602 1603
                                 (macaddr
                                  ? virMacAddrFormat(macaddr, macstr)
                                  : "(unchanged)"),
                                 vlanid,
                                 ifname ? ifname : "(unspecified)",
1604
                                 vf);
1605
            *allowRetry = false; /* no use retrying */
1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617
            goto cleanup;
        }
        break;

    case NLMSG_DONE:
        break;

    default:
        goto malformed_resp;
    }

    rc = 0;
1618
 cleanup:
1619
    nlmsg_free(nl_msg);
1620
    VIR_FREE(resp);
1621 1622
    return rc;

1623
 malformed_resp:
1624
    virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
1625
                   _("malformed netlink response message"));
1626
    goto cleanup;
1627

1628
 buffer_too_small:
1629
    virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
1630
                   _("allocated netlink buffer is too small"));
1631
    goto cleanup;
1632 1633 1634
}

static int
1635
virNetDevParseVfConfig(struct nlattr **tb, int32_t vf, virMacAddrPtr mac,
1636 1637 1638
                       int *vlanid)
{
    int rc = -1;
1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649
    struct ifla_vf_mac *vf_mac;
    struct ifla_vf_vlan *vf_vlan;
    struct nlattr *tb_vf_info = {NULL, };
    struct nlattr *tb_vf[IFLA_VF_MAX+1];
    int rem;

    if (!tb[IFLA_VFINFO_LIST]) {
        virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                       _("missing IFLA_VF_INFO in netlink response"));
        goto cleanup;
    }
1650

1651 1652 1653
    nla_for_each_nested(tb_vf_info, tb[IFLA_VFINFO_LIST], rem) {
        if (nla_type(tb_vf_info) != IFLA_VF_INFO)
            continue;
1654

1655 1656 1657 1658 1659 1660
        if (nla_parse_nested(tb_vf, IFLA_VF_MAX, tb_vf_info,
                             ifla_vf_policy)) {
            virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                           _("error parsing IFLA_VF_INFO"));
            goto cleanup;
        }
1661

1662
        if (mac && tb[IFLA_VF_MAC]) {
1663 1664 1665
            vf_mac = RTA_DATA(tb_vf[IFLA_VF_MAC]);
            if (vf_mac && vf_mac->vf == vf)  {
                virMacAddrSetRaw(mac, vf_mac->mac);
J
Ján Tomko 已提交
1666
                rc = 0;
1667
            }
1668 1669
        }

1670
        if (vlanid && tb[IFLA_VF_VLAN]) {
1671 1672 1673
            vf_vlan = RTA_DATA(tb_vf[IFLA_VF_VLAN]);
            if (vf_vlan && vf_vlan->vf == vf)  {
                *vlanid = vf_vlan->vlan;
J
Ján Tomko 已提交
1674
                rc = 0;
1675 1676
            }
        }
J
Ján Tomko 已提交
1677 1678 1679

        if (rc == 0)
            break;
1680
    }
J
Ján Tomko 已提交
1681 1682 1683 1684
    if (rc < 0)
        virReportError(VIR_ERR_INTERNAL_ERROR,
                       _("couldn't find IFLA_VF_INFO for VF %d "
                         "in netlink response"), vf);
1685
 cleanup:
1686 1687 1688 1689
    return rc;
}

static int
1690
virNetDevGetVfConfig(const char *ifname, int vf, virMacAddrPtr mac,
1691 1692 1693
                     int *vlanid)
{
    int rc = -1;
1694
    void *nlData = NULL;
1695 1696 1697
    struct nlattr *tb[IFLA_MAX + 1] = {NULL, };
    int ifindex = -1;

1698
    rc = virNetlinkDumpLink(ifname, ifindex, &nlData, tb, 0, 0);
1699
    if (rc < 0)
1700
        goto cleanup;
1701 1702 1703

    rc = virNetDevParseVfConfig(tb, vf, mac, vlanid);

1704 1705
 cleanup:
    VIR_FREE(nlData);
1706 1707 1708 1709
    return rc;
}


1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098
# define VIR_NETDEV_KEYNAME_ADMIN_MAC "adminMac"
# define VIR_NETDEV_KEYNAME_VLAN_TAG "vlanTag"
# define VIR_NETDEV_KEYNAME_MAC "mac"

/**
 * virNetDevSaveNetConfig:
 * @linkdev: name of the interface
 * @vf: vf index if linkdev is a pf
 * @stateDir: directory to store old net config
 * @saveVlan: false if we shouldn't attempt to save vlan tag info
 *            (eg for interfaces using 802.1Qbg, since it handles
 *            vlan tags internally)
 *
 * Save current MAC address and (if linkdev itself is a VF, or if @vf
 * >= 0) the "admin MAC address" and vlan tag the device described by
 * @linkdev:@vf to @stateDir. (the "admin MAC address" is stored in
 * the PF, and is what the VF MAC will be initialized to the next time
 * its driver is reloaded (either on host or guest).
 *
 * File Format:
 *
 * The file is in json format and will contain 1 or more of the
 * following values:
 *
 *      "mac"      - VF MAC address (or missing if VF has no host net driver)
 *      "vlanTag"  - a single vlan tag id
 *      "adminMac" - admin MAC address (stored in the PF)
 *
 * For example:
 *
 *    {"mac": "9A:11:22:33:44:55",
 *     "vlanTag": "42",
 *     "adminMac": "00:00:00:00:00:00"
 *    }
 *
 * File Name:
 *
 * If the device is a VF and we're allowed to save vlan tag info, the
 * file will be named ${pfDevName_vf#{vf} (e.g. "enp2s0f0_vf5") and
 * will contain at least "adminMac" and "vlanTag" (if the device was bound
 * to a net driver on the host prior to use, it will also have "mac"..
 * If the device isn't a VF, or we're not allowed to save vlan tag
 * info, the file will be named ${linkdev} (e.g. "enp3s0f0") and will
 * contain just linkdev's MAC address.
 *
 * Returns 0 on success, -1 on failure
 *
 */
int
virNetDevSaveNetConfig(const char *linkdev, int vf,
                       const char *stateDir,
                       bool saveVlan)
{
    int ret = -1;
    const char *pfDevName = NULL;
    char *pfDevOrig = NULL;
    char *vfDevOrig = NULL;
    virMacAddr oldMAC;
    char MACStr[VIR_MAC_STRING_BUFLEN];
    int oldVlanTag = -1;
    char *filePath = NULL;
    char *fileStr = NULL;
    virJSONValuePtr configJSON = NULL;

    if (vf >= 0) {
        /* linkdev is the PF */
        pfDevName = linkdev;

        /* linkdev should get the VF's netdev name (or NULL if none) */
        if (virNetDevPFGetVF(pfDevName, vf, &vfDevOrig) < 0)
            goto cleanup;

        linkdev = vfDevOrig;

    } else if (saveVlan && virNetDevIsVirtualFunction(linkdev) == 1) {
        /* when vf is -1, linkdev might be a standard netdevice (not
         * SRIOV), or it might be an SRIOV VF. If it's a VF, normalize
         * it to PF + VFname
         */

        if (virNetDevGetPhysicalFunction(linkdev, &pfDevOrig) < 0)
            goto cleanup;

        pfDevName = pfDevOrig;

        if (virNetDevGetVirtualFunctionIndex(pfDevName, linkdev, &vf) < 0)
            goto cleanup;
    }

    if (!(configJSON = virJSONValueNewObject()))
        goto cleanup;

    /* if there is a PF, it's now in pfDevName, and linkdev is either
     * the VF's name, or NULL (if the VF isn't bound to a net driver
     * on the host)
     */

    if (pfDevName) {
        if (virAsprintf(&filePath, "%s/%s_vf%d", stateDir, pfDevName, vf) < 0)
            goto cleanup;

        /* get admin MAC and vlan tag */
        if (virNetDevGetVfConfig(pfDevName, vf, &oldMAC,
                                 saveVlan ? &oldVlanTag : NULL) < 0) {
            goto cleanup;
        }

        if (virJSONValueObjectAppendString(configJSON,
                                           VIR_NETDEV_KEYNAME_ADMIN_MAC,
                                           virMacAddrFormat(&oldMAC, MACStr)) < 0 ||
            virJSONValueObjectAppendNumberInt(configJSON,
                                              VIR_NETDEV_KEYNAME_VLAN_TAG,
                                              oldVlanTag) < 0) {
            goto cleanup;
        }

    } else {
        if (virAsprintf(&filePath, "%s/%s", stateDir, linkdev) < 0)
            goto cleanup;
    }

    if (linkdev) {
        if (virNetDevGetMAC(linkdev, &oldMAC) < 0)
            goto cleanup;

        /* for interfaces with no pfDevName (i.e. not a VF, this will
         * be the only value in the file.
         */
        if (virJSONValueObjectAppendString(configJSON, VIR_NETDEV_KEYNAME_MAC,
                                           virMacAddrFormat(&oldMAC, MACStr)) < 0)
           goto cleanup;
    }

    if (!(fileStr = virJSONValueToString(configJSON, true)))
        goto cleanup;

    if (virFileWriteStr(filePath, fileStr, O_CREAT|O_TRUNC|O_WRONLY) < 0) {
        virReportSystemError(errno, _("Unable to preserve mac/vlan tag "
                                      "for device = %s, vf = %d"), linkdev, vf);
        goto cleanup;
    }

    ret = 0;
 cleanup:
    VIR_FREE(pfDevOrig);
    VIR_FREE(vfDevOrig);
    VIR_FREE(filePath);
    VIR_FREE(fileStr);
    virJSONValueFree(configJSON);
    return ret;
}


/**
 * virNetDevReadNetConfig:
 * @linkdev: name of the interface
 * @vf: vf index if linkdev is a pf
 * @stateDir: directory where net config is stored
 * @adminMAC: returns admin MAC to store in the PF (if this is a VF)
 * @MAC: returns MAC to set on device immediately
 *
 * Read saved MAC address and (if linkdev itself is a VF, or if @vf >=
 * 0) "admin MAC address" and vlan tag of the device described by
 * @linkdev:@vf from a file in @stateDir. (see virNetDevSaveNetConfig
 * for details of file name and format).
 *
 * Returns 0 on success, -1 on failure.
 *
 * The caller MUST free adminMAC, vlan, and MAC when it is finished
 * with them (they will be NULL if they weren't found in the file)
 *
 */
int
virNetDevReadNetConfig(const char *linkdev, int vf,
                       const char *stateDir,
                       virMacAddrPtr *adminMAC,
                       virNetDevVlanPtr *vlan,
                       virMacAddrPtr *MAC)
{
    int ret = -1;
    const char *pfDevName = NULL;
    char *pfDevOrig = NULL;
    char *vfDevOrig = NULL;
    char *filePath = NULL;
    char *fileStr = NULL;
    virJSONValuePtr configJSON = NULL;
    const char *MACStr = NULL;
    const char *adminMACStr = NULL;
    int vlanTag = -1;

    *adminMAC = NULL;
    *vlan = NULL;
    *MAC = NULL;

    if (vf >= 0) {
        /* linkdev is the PF */
        pfDevName = linkdev;

        /* linkdev should get the VF's netdev name (or NULL if none) */
        if (virNetDevPFGetVF(pfDevName, vf, &vfDevOrig) < 0)
            goto cleanup;

        linkdev = vfDevOrig;

    } else if (virNetDevIsVirtualFunction(linkdev) == 1) {
        /* when vf is -1, linkdev might be a standard netdevice (not
         * SRIOV), or it might be an SRIOV VF. If it's a VF, normalize
         * it to PF + VFname
         */

        if (virNetDevGetPhysicalFunction(linkdev, &pfDevOrig) < 0)
            goto cleanup;

        pfDevName = pfDevOrig;

        if (virNetDevGetVirtualFunctionIndex(pfDevName, linkdev, &vf) < 0)
            goto cleanup;
    }

    /* if there is a PF, it's now in pfDevName, and linkdev is either
     * the VF's name, or NULL (if the VF isn't bound to a net driver
     * on the host)
     */

    if (pfDevName) {
        if (virAsprintf(&filePath, "%s/%s_vf%d", stateDir, pfDevName, vf) < 0)
            goto cleanup;

        if (linkdev && !virFileExists(filePath)) {
            /* the device may have been stored in a file named for the
             * VF due to saveVlan == false (or an older version of
             * libvirt), so reset filePath so we'll try the other
             * filename before failing.
             */
            VIR_FREE(filePath);
            pfDevName = NULL;
        }
    }

    if (!pfDevName) {
        if (virAsprintf(&filePath, "%s/%s", stateDir, linkdev) < 0)
            goto cleanup;
    }

    if (virFileReadAll(filePath, 128, &fileStr) < 0)
        goto cleanup;

    if (strchr("0123456789abcdefABCDEF", fileStr[0])) {
        const char *vlanStr = NULL;

        /* old version of file - just two lines of text. Line 1 is the
         * MAC address (or if line 2 is present, line 1 is adminMAC),
         * and line 2 (if present) is the vlan tag
         */

        if ((vlanStr = strchr(fileStr, '\n'))) {
            char *endptr;

            /* if there are 2 lines, the first is adminMAC */
            adminMACStr = fileStr;
            vlanStr++;

            if ((virStrToLong_i(vlanStr, &endptr, 10, &vlanTag) < 0) ||
                (endptr && *endptr != '\n' && *endptr != 0)) {
                virReportError(VIR_ERR_INTERNAL_ERROR,
                               _("cannot parse vlan tag '%s' from file '%s'"),
                               vlanStr, filePath);
                goto cleanup;
            }
        } else {
            /* if there is only one line, it is MAC */
            MACStr = fileStr;
        }
    } else {
        /* if it doesn't start with a hex digit, it is a modern
         * version of the config file - JSON format as described in
         * preamble to virNetDevSaveNetConfig()
         */
        if (!(configJSON = virJSONValueFromString(fileStr))) {
            virReportError(VIR_ERR_INTERNAL_ERROR,
                           _("invalid json in net device saved "
                             "config file '%s': '%.60s'"),
                           filePath, fileStr);
            goto cleanup;
        }

        MACStr = virJSONValueObjectGetString(configJSON,
                                             VIR_NETDEV_KEYNAME_MAC);
        adminMACStr = virJSONValueObjectGetString(configJSON,
                                                  VIR_NETDEV_KEYNAME_ADMIN_MAC);
        ignore_value(virJSONValueObjectGetNumberInt(configJSON,
                                                    VIR_NETDEV_KEYNAME_VLAN_TAG,
                                                    &vlanTag));

        if (!(MACStr || adminMACStr)) {
            virReportError(VIR_ERR_INTERNAL_ERROR,
                           _("network device saved config file '%s' "
                             "has unexpected contents, missing both "
                             "'MAC' and 'adminMAC': '%.60s'"),
                           filePath, fileStr);
            goto cleanup;
        }
    }

    if (MACStr) {
        if (VIR_ALLOC(*MAC) < 0)
            goto cleanup;

        if (virMacAddrParse(MACStr, *MAC) < 0) {
            virReportError(VIR_ERR_INTERNAL_ERROR,
                           _("cannot parse MAC address '%s' from file '%s'"),
                           MACStr, filePath);
            goto cleanup;
        }
    }

    if (adminMACStr) {
        if (VIR_ALLOC(*adminMAC) < 0)
            goto cleanup;

        if (virMacAddrParse(adminMACStr, *adminMAC) < 0) {
            virReportError(VIR_ERR_INTERNAL_ERROR,
                           _("cannot parse MAC address '%s' from file '%s'"),
                           adminMACStr, filePath);
            goto cleanup;
        }
    }

    if (vlanTag != -1) {
        /* construct a simple virNetDevVlan object with a single tag */
        if (VIR_ALLOC(*vlan) < 0)
            goto cleanup;
        if (VIR_ALLOC((*vlan)->tag) < 0)
            goto cleanup;
        (*vlan)->nTags = 1;
        (*vlan)->tag[0] = vlanTag;
    }

    /* we won't need the file again */
    ignore_value(unlink(filePath));

    ret = 0;
 cleanup:
    if (ret < 0) {
        VIR_FREE(*adminMAC);
        VIR_FREE(*MAC);
        VIR_FREE(*vlan);
    }

    VIR_FREE(pfDevOrig);
    VIR_FREE(vfDevOrig);
    VIR_FREE(filePath);
    VIR_FREE(fileStr);
    virJSONValueFree(configJSON);
    return ret;
}


/**
 * virNetDevSetNetConfig:
 * @linkdev: name of the interface
 * @vf: vf index if linkdev is a PF
 * @adminMAC: new admin MAC address (will be stored in PF and
 *            used for next initialization of VF driver)
 * @vlan: new vlan tag info (or NULL)
 * @MAC: new MAC address to set on the device immediately
 * @setVlan: true to enable setting vlan tag (even if @vlan is NULL,
 *           the interface vlan tag will be set to 0).
 *
 *
 * Set new MAC address and (optionally) admin MAC and vlan tag of
 * @linkdev VF# @vf.
 *
 * Returns 0 on success, -1 on failure
 *
 */
int
virNetDevSetNetConfig(const char *linkdev, int vf,
                      const virMacAddr *adminMAC,
                      virNetDevVlanPtr vlan,
                      const virMacAddr *MAC,
                      bool setVlan)
{
    int ret = -1;
    char MACStr[VIR_MAC_STRING_BUFLEN];
    const char *pfDevName = NULL;
    char *pfDevOrig = NULL;
    char *vfDevOrig = NULL;
    int vlanTag = -1;
2099
    virPCIDevicePtr vfPCIDevice = NULL;
2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194

    if (vf >= 0) {
        /* linkdev is the PF */
        pfDevName = linkdev;

        /* linkdev should get the VF's netdev name (or NULL if none) */
        if (virNetDevPFGetVF(pfDevName, vf, &vfDevOrig) < 0)
            goto cleanup;

        linkdev = vfDevOrig;

    } else if (virNetDevIsVirtualFunction(linkdev) == 1) {
        /* when vf is -1, linkdev might be a standard netdevice (not
         * SRIOV), or it might be an SRIOV VF. If it's a VF, normalize
         * it to PF + VFname
         */

        if (virNetDevGetPhysicalFunction(linkdev, &pfDevOrig) < 0)
            goto cleanup;

        pfDevName = pfDevOrig;

        if (virNetDevGetVirtualFunctionIndex(pfDevName, linkdev, &vf) < 0)
            goto cleanup;
    }


    if (!pfDevName) {
        /* if it's not SRIOV, then we can't set the admin MAC address
         * or vlan tag
         */
        if (adminMAC) {
            virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
                           _("admin MAC can only be set for SR-IOV VFs, but "
                             "%s is not a VF"), linkdev);
            goto cleanup;
        }

        if (vlan) {
            virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
                           _("vlan can only be set for SR-IOV VFs, but "
                             "%s is not a VF"), linkdev);
            goto cleanup;
        }

    } else {
        bool pfIsOnline;

        /* Assure that PF is online before trying to use it to set
         * anything up for this VF. It *should* be online already,
         * but if it isn't online the changes made to the VF via the
         * PF won't take effect, yet there will be no error
         * reported. In the case that the PF isn't online, we need to
         * fail and report the error, rather than automatically
         * setting it online, since setting an unconfigured interface
         * online automatically turns on IPv6 autoconfig, which may
         * not be what the admin expects, so we require them to
         * explicitly enable the PF in the host system network config.
         */
        if (virNetDevGetOnline(pfDevName, &pfIsOnline) < 0)
            goto cleanup;

        if (!pfIsOnline) {
            virReportError(VIR_ERR_INTERNAL_ERROR,
                           _("Unable to configure VF %d of PF '%s' "
                             "because the PF is not online. Please "
                             "change host network config to put the "
                             "PF online."),
                           vf, pfDevName);
            goto cleanup;
        }

        if (vlan) {
            if (vlan->nTags != 1 || vlan->trunk) {
                virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
                               _("vlan trunking is not supported "
                                 "by SR-IOV network devices"));
                goto cleanup;
            }

            if (!setVlan) {
                virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
                               _("vlan tag set for interface %s but "
                                 "caller requested it not be set"));
                goto cleanup;
            }

            vlanTag = vlan->tag[0];

        } else if (setVlan) {
            vlanTag = 0; /* assure any existing vlan tag is reset */
        }
    }

    if (MAC) {
2195 2196
        int setMACrc;

2197 2198 2199 2200 2201 2202 2203 2204
        if (!linkdev) {
            virReportError(VIR_ERR_INTERNAL_ERROR,
                           _("VF %d of PF '%s' is not bound to a net driver, "
                             "so its MAC address cannot be set to %s"),
                           vf, pfDevName, virMacAddrFormat(MAC, MACStr));
            goto cleanup;
        }

2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221
        setMACrc = virNetDevSetMACInternal(linkdev, MAC, !!pfDevOrig);
        if (setMACrc < 0) {
            bool allowRetry = false;
            int retries = 100;

            /* if pfDevOrig == NULL, this isn't a VF, so we've failed */
            if (!pfDevOrig || errno != EADDRNOTAVAIL)
                goto cleanup;

            /* Otherwise this is a VF, and virNetDevSetMAC failed with
             * EADDRNOTAVAIL, which could be due to the
             * "administratively set" flag being set in the PF for
             * this VF.  When this happens, we can attempt to use an
             * alternate method to set the VF MAC: first set it into
             * the admin MAC for this VF in the PF, then unbind/rebind
             * the VF from its net driver. This causes the VF's MAC to
             * be initialized to whatever was stored in the admin MAC.
2222
             */
2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244

            if (virNetDevSetVfConfig(pfDevName, vf,
                                     MAC, vlanTag, &allowRetry) < 0) {
                goto cleanup;
            }

            /* admin MAC is set, now we need to construct a virPCIDevice
             * object so we can call virPCIDeviceRebind()
             */
            if (!(vfPCIDevice = virNetDevGetPCIDevice(linkdev)))
                goto cleanup;

            /* Rebind the device. This should set the proper MAC address */
            if (virPCIDeviceRebind(vfPCIDevice) < 0)
                goto cleanup;

            /* Wait until virNetDevGetIndex for the VF netdev returns success.
             * This indicates that the device is ready to be used. If we don't
             * wait, then upcoming operations on the VF may fail.
             */
            while (retries-- > 0 && !virNetDevExists(linkdev))
               usleep(1000);
2245
        }
2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259

        if (pfDevOrig && setMACrc == 0) {
            /* if pfDevOrig is set, it that the caller was *really*
             * only interested in setting the MAC of the VF itself,
             * *not* the admin MAC via the PF. In those cases, the
             * adminMAC was only provided in case we need to set the
             * VF's MAC by temporarily unbinding/rebinding the VF's
             * net driver with the admin MAC set to the desired MAC,
             * and then want to restore the admin MAC to its original
             * setting when we're finished. We would only need to do
             * that if the virNetDevSetMAC() above had failed; since
             * setMACrc == 0, we know it didn't fail and we don't need
             * to set the adminMAC, so we are NULLing it out here to
             * avoid that below.
2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278

             * (NB: since setting the admin MAC sets the
             * "administratively set" flag for the VF in the PF's
             * driver, which prevents any future changes to the VF's
             * MAC address, we want to avoid setting the admin MAC as
             * much as possible.)
             */
            adminMAC = NULL;
        }
    }

    if (adminMAC || vlanTag >= 0) {
        /* Set vlanTag and admin MAC using an RTM_SETLINK request sent to
         * PFdevname+VF#, if mac != NULL this will set the "admin MAC" via
         * the PF, *not* the actual VF MAC - the admin MAC only takes
         * effect the next time the VF's driver is initialized (either in
         * guest or host). if there is a vlanTag to set, it will take
         * effect immediately though.
         */
2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296
        bool allowRetry = true;

        if (virNetDevSetVfConfig(pfDevName, vf,
                                 adminMAC, vlanTag, &allowRetry) < 0) {
            /* allowRetry will still be true if the failure was due to
             * trying to set the MAC address to all 0. In that case,
             * we can retry with "altZeroMAC", which is just an all-0 MAC
             * with the "locally administered" bit set.
             */
            if (!allowRetry)
                goto cleanup;

            allowRetry = false;
            if (virNetDevSetVfConfig(pfDevName, vf,
                                     &altZeroMAC, vlanTag, &allowRetry) < 0) {
                goto cleanup;
            }
        }
2297 2298 2299 2300 2301 2302
    }

    ret = 0;
 cleanup:
    VIR_FREE(pfDevOrig);
    VIR_FREE(vfDevOrig);
2303
    virPCIDeviceFree(vfPCIDevice);
2304 2305 2306 2307 2308
    return ret;
}


#else /* defined(__linux__) && defined(HAVE_LIBNL)  && defined(IFLA_VF_MAX) */
2309

2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351

int
virNetDevSaveNetConfig(const char *linkdev ATTRIBUTE_UNUSED,
                       int vf ATTRIBUTE_UNUSED,
                       const char *stateDir ATTRIBUTE_UNUSED,
                       bool saveVlan ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to save net device config on this platform"));
    return -1;
}


int
virNetDevReadNetConfig(const char *linkdev ATTRIBUTE_UNUSED,
                       int vf ATTRIBUTE_UNUSED,
                       const char *stateDir ATTRIBUTE_UNUSED,
                       virMacAddrPtr *adminMAC ATTRIBUTE_UNUSED,
                       virNetDevVLanPtr *vlan ATTRIBUTE_UNUSED,
                       virMacAddrPtr *MAC ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to read net device config on this platform"));
    return -1;
}


int
virNetDevSetNetConfig(const char *linkdev ATTRIBUTE_UNUSED,
                      int vf ATTRIBUTE_UNUSED,
                      const virMacAddr *adminMAC ATTRIBUTE_UNUSED,
                      virNetDevVlanPtr vlan ATTRIBUTE_UNUSED,
                      const virMacAddr *MAC ATTRIBUTE_UNUSED,
                      bool setVlan ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to set net device config on this platform"));
    return -1;
}


#endif /* defined(__linux__) && defined(HAVE_LIBNL) && defined(IFLA_VF_MAX) */
2352

2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375
VIR_ENUM_IMPL(virNetDevIfState,
              VIR_NETDEV_IF_STATE_LAST,
              "" /* value of zero means no state */,
              "unknown", "notpresent",
              "down", "lowerlayerdown",
              "testing", "dormant", "up")

VIR_ENUM_IMPL(virNetDevFeature,
              VIR_NET_DEV_FEAT_LAST,
              "rx",
              "tx",
              "sg",
              "tso",
              "gso",
              "gro",
              "lro",
              "rxvlan",
              "txvlan",
              "ntuple",
              "rxhash",
              "rdma",
              "txudptnl")

2376 2377 2378
#ifdef __linux__
int
virNetDevGetLinkInfo(const char *ifname,
2379
                     virNetDevIfLinkPtr lnk)
2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408
{
    int ret = -1;
    char *path = NULL;
    char *buf = NULL;
    char *tmp;
    int tmp_state;
    unsigned int tmp_speed;

    if (virNetDevSysfsFile(&path, ifname, "operstate") < 0)
        goto cleanup;

    if (virFileReadAll(path, 1024, &buf) < 0) {
        virReportSystemError(errno,
                             _("unable to read: %s"),
                             path);
        goto cleanup;
    }

    if (!(tmp = strchr(buf, '\n'))) {
        virReportError(VIR_ERR_INTERNAL_ERROR,
                       _("Unable to parse: %s"),
                       buf);
        goto cleanup;
    }

    *tmp = '\0';

    /* We shouldn't allow 0 here, because
     * virInterfaceState enum starts from 1. */
2409
    if ((tmp_state = virNetDevIfStateTypeFromString(buf)) <= 0) {
2410 2411 2412 2413 2414 2415 2416 2417
        virReportError(VIR_ERR_INTERNAL_ERROR,
                       _("Unable to parse: %s"),
                       buf);
        goto cleanup;
    }

    lnk->state = tmp_state;

2418 2419 2420 2421
    /* Shortcut to avoid some kernel issues. If link is not up several drivers
     * report several misleading values. While igb reports 65535, realtek goes
     * with 10. To avoid muddying XML with insane values, don't report link
     * speed if that's the case. */
2422
    if (lnk->state != VIR_NETDEV_IF_STATE_UP) {
2423 2424 2425 2426 2427
        lnk->speed = 0;
        ret = 0;
        goto cleanup;
    }

2428 2429 2430 2431 2432 2433
    VIR_FREE(path);
    VIR_FREE(buf);

    if (virNetDevSysfsFile(&path, ifname, "speed") < 0)
        goto cleanup;

2434
    if (virFileReadAllQuiet(path, 1024, &buf) < 0) {
2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453
        /* Some devices doesn't report speed, in which case we get EINVAL */
        if (errno == EINVAL) {
            ret = 0;
            goto cleanup;
        }
        virReportSystemError(errno,
                             _("unable to read: %s"),
                             path);
        goto cleanup;
    }

    if (virStrToLong_ui(buf, &tmp, 10, &tmp_speed) < 0 ||
        *tmp != '\n') {
        virReportError(VIR_ERR_INTERNAL_ERROR,
                       _("Unable to parse: %s"),
                       buf);
        goto cleanup;
    }

2454
    lnk->speed = tmp_speed;
2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466

    ret = 0;
 cleanup:
    VIR_FREE(buf);
    VIR_FREE(path);
    return ret;
}

#else

int
virNetDevGetLinkInfo(const char *ifname,
2467
                     virNetDevIfLinkPtr lnk)
2468 2469
{
    /* Port me */
2470 2471
    VIR_DEBUG("Getting link info on %s is not implemented on this platform",
              ifname);
2472 2473 2474 2475
    lnk->speed = lnk->state = 0;
    return 0;
}
#endif /* defined(__linux__) */
2476 2477


2478 2479
#if defined(SIOCADDMULTI) && defined(HAVE_STRUCT_IFREQ) && \
    defined(HAVE_STRUCT_IFREQ_IFR_HWADDR)
2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526
/**
 * virNetDevAddMulti:
 * @ifname: interface name to which to add multicast MAC address
 * @macaddr: MAC address
 *
 * This function adds the @macaddr to the multicast list for a given interface
 * @ifname.
 *
 * Returns 0 in case of success or -1 on failure
 */
int virNetDevAddMulti(const char *ifname,
                      virMacAddrPtr macaddr)
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    ifr.ifr_hwaddr.sa_family = AF_UNSPEC;
    virMacAddrGetRaw(macaddr, (unsigned char *)ifr.ifr_hwaddr.sa_data);

    if (ioctl(fd, SIOCADDMULTI, &ifr) < 0) {
        char macstr[VIR_MAC_STRING_BUFLEN];
        virReportSystemError(errno,
                             _("Cannot add multicast MAC %s on '%s' interface"),
                             virMacAddrFormat(macaddr, macstr), ifname);
        goto cleanup;
    }

    ret = 0;
 cleanup:
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
int virNetDevAddMulti(const char *ifname ATTRIBUTE_UNUSED,
                      virMacAddrPtr macaddr ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to add address to interface "
                           "multicast list on this platform"));
    return -1;
}
#endif

2527 2528
#if defined(SIOCDELMULTI) && defined(HAVE_STRUCT_IFREQ) && \
    defined(HAVE_STRUCT_IFREQ_IFR_HWADDR)
2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584
/**
 * virNetDevDelMulti:
 * @ifname: interface name from which to delete the multicast MAC address
 * @macaddr: MAC address
 *
 * This function deletes the @macaddr from the multicast list for a given
 * interface @ifname.
 *
 * Returns 0 in case of success or -1 on failure
 */
int virNetDevDelMulti(const char *ifname,
                      virMacAddrPtr macaddr)
{
    int fd = -1;
    int ret = -1;
    struct ifreq ifr;

    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        return -1;

    ifr.ifr_hwaddr.sa_family = AF_UNSPEC;
    virMacAddrGetRaw(macaddr, (unsigned char *)ifr.ifr_hwaddr.sa_data);

    if (ioctl(fd, SIOCDELMULTI, &ifr) < 0) {
        char macstr[VIR_MAC_STRING_BUFLEN];
        virReportSystemError(errno,
                             _("Cannot add multicast MAC %s on '%s' interface"),
                             virMacAddrFormat(macaddr, macstr), ifname);
        goto cleanup;
    }

    ret = 0;
 cleanup:
    VIR_FORCE_CLOSE(fd);
    return ret;
}
#else
int virNetDevDelMulti(const char *ifname ATTRIBUTE_UNUSED,
                      virMacAddrPtr macaddr ATTRIBUTE_UNUSED)
{
    virReportSystemError(ENOSYS, "%s",
                         _("Unable to delete address from interface "
                           "multicast list on this platform"));
    return -1;
}
#endif

static int virNetDevParseMcast(char *buf, virNetDevMcastEntryPtr mcast)
{
    int ifindex;
    int num;
    char *next;
    char *token;
    char *saveptr;
    char *endptr;

2585
    for (ifindex = VIR_MCAST_TYPE_INDEX_TOKEN, next = buf; ifindex < VIR_MCAST_TYPE_LAST; ifindex++,
2586 2587 2588 2589 2590 2591 2592 2593 2594 2595
         next = NULL) {
        token = strtok_r(next, VIR_MCAST_TOKEN_DELIMS, &saveptr);

        if (token == NULL) {
            virReportSystemError(EINVAL,
                                 _("failed to parse multicast address from '%s'"),
                                 buf);
            return -1;
        }

J
John Ferlan 已提交
2596 2597
        switch ((virMCastType)ifindex) {
            case VIR_MCAST_TYPE_INDEX_TOKEN:
2598 2599 2600 2601 2602 2603 2604
                if (virStrToLong_i(token, &endptr, 10, &num) < 0) {
                    virReportSystemError(EINVAL,
                                         _("Failed to parse interface index from '%s'"),
                                         buf);
                    return -1;

                }
J
John Ferlan 已提交
2605
                mcast->idx = num;
2606
                break;
J
John Ferlan 已提交
2607
            case VIR_MCAST_TYPE_NAME_TOKEN:
2608 2609 2610 2611 2612 2613 2614 2615
                if (virStrncpy(mcast->name, token, strlen(token),
                    VIR_MCAST_NAME_LEN) == NULL) {
                    virReportSystemError(EINVAL,
                                         _("Failed to parse network device name from '%s'"),
                                         buf);
                    return -1;
                }
                break;
J
John Ferlan 已提交
2616
            case VIR_MCAST_TYPE_USERS_TOKEN:
2617 2618 2619 2620 2621 2622 2623 2624 2625
                if (virStrToLong_i(token, &endptr, 10, &num) < 0) {
                    virReportSystemError(EINVAL,
                                         _("Failed to parse users from '%s'"),
                                         buf);
                    return -1;

                }
                mcast->users = num;
                break;
J
John Ferlan 已提交
2626
            case VIR_MCAST_TYPE_GLOBAL_TOKEN:
2627 2628 2629 2630 2631 2632 2633 2634 2635
                if (virStrToLong_i(token, &endptr, 10, &num) < 0) {
                    virReportSystemError(EINVAL,
                                         _("Failed to parse users from '%s'"),
                                         buf);
                    return -1;

                }
                mcast->global = num;
                break;
J
John Ferlan 已提交
2636
            case VIR_MCAST_TYPE_ADDR_TOKEN:
2637 2638 2639 2640 2641 2642 2643
                if (virMacAddrParseHex((const char*)token,
                    &mcast->macaddr) < 0) {
                    virReportSystemError(EINVAL,
                                         _("Failed to parse MAC address from '%s'"),
                                         buf);
                }
                break;
J
John Ferlan 已提交
2644 2645 2646

            /* coverity[dead_error_begin] */
            case VIR_MCAST_TYPE_LAST:
2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704
                break;
        }
    }
    return 0;
}


static void virNetDevMcastListClear(virNetDevMcastListPtr mcast)
{
    size_t i;

    for (i = 0; i < mcast->nentries; i++)
       VIR_FREE(mcast->entries[i]);
    VIR_FREE(mcast->entries);
    mcast->nentries = 0;
}


static int virNetDevGetMcastList(const char *ifname,
                                 virNetDevMcastListPtr mcast)
{
    char *cur = NULL;
    char *buf = NULL;
    char *next = NULL;
    int ret = -1, len;
    virNetDevMcastEntryPtr entry = NULL;

    mcast->entries = NULL;
    mcast->nentries = 0;

    /* Read entire multicast table into memory */
    if ((len = virFileReadAll(PROC_NET_DEV_MCAST, MAX_MCAST_SIZE, &buf)) <= 0)
        goto cleanup;

    cur = buf;
    while (cur) {
        if (!entry && VIR_ALLOC(entry) < 0)
                goto cleanup;

        next = strchr(cur, '\n');
        if (next)
            next++;
        if (virNetDevParseMcast(cur, entry))
            goto cleanup;

        /* Only return global multicast MAC addresses for
         * specified interface */
        if (entry->global && STREQ(ifname, entry->name)) {
            if (VIR_APPEND_ELEMENT(mcast->entries, mcast->nentries, entry))
                 goto cleanup;
        } else {
            memset(entry, 0, sizeof(virNetDevMcastEntry));
        }
        cur = next && ((next - buf) < len) ? next : NULL;
    }

    ret = 0;
 cleanup:
2705
    VIR_FREE(buf);
2706 2707 2708 2709 2710 2711
    VIR_FREE(entry);

    return ret;
}


2712 2713 2714 2715 2716 2717 2718
VIR_ENUM_IMPL(virNetDevRxFilterMode,
              VIR_NETDEV_RX_FILTER_MODE_LAST,
              "none",
              "normal",
              "all");


2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743
static int virNetDevGetMulticastTable(const char *ifname,
                                      virNetDevRxFilterPtr filter)
{
    size_t i;
    int ret = -1;
    virNetDevMcastList mcast;
    filter->multicast.nTable = 0;
    filter->multicast.table = NULL;

    if (virNetDevGetMcastList(ifname, &mcast) < 0)
        goto cleanup;

    if (mcast.nentries > 0) {
        if (VIR_ALLOC_N(filter->multicast.table, mcast.nentries) < 0)
            goto cleanup;

        for (i = 0; i < mcast.nentries; i++) {
            virMacAddrSet(&filter->multicast.table[i],
                          &mcast.entries[i]->macaddr);
        }

        filter->multicast.nTable = mcast.nentries;
    }

    ret = 0;
2744

2745 2746 2747 2748 2749 2750
 cleanup:
    virNetDevMcastListClear(&mcast);
    return ret;
}


2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772
virNetDevRxFilterPtr
virNetDevRxFilterNew(void)
{
    virNetDevRxFilterPtr filter;

    if (VIR_ALLOC(filter) < 0)
        return NULL;
    return filter;
}


void
virNetDevRxFilterFree(virNetDevRxFilterPtr filter)
{
    if (filter) {
        VIR_FREE(filter->name);
        VIR_FREE(filter->unicast.table);
        VIR_FREE(filter->multicast.table);
        VIR_FREE(filter->vlan.table);
        VIR_FREE(filter);
    }
}
2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787


/**
 * virNetDevGetRxFilter:
 * This function supplies the RX filter list for a given device interface
 *
 * @ifname: Name of the interface
 * @filter: The RX filter list
 *
 * Returns 0 or -1 on failure.
 */
int virNetDevGetRxFilter(const char *ifname,
                         virNetDevRxFilterPtr *filter)
{
    int ret = -1;
2788
    bool receive = false;
2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799
    virNetDevRxFilterPtr fil = virNetDevRxFilterNew();

    if (!fil)
        goto cleanup;

    if (virNetDevGetMAC(ifname, &fil->mac))
        goto cleanup;

    if (virNetDevGetMulticastTable(ifname, fil))
        goto cleanup;

2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817
    if (virNetDevGetPromiscuous(ifname, &fil->promiscuous))
        goto cleanup;

    if (virNetDevGetRcvAllMulti(ifname, &receive))
        goto cleanup;

    if (receive) {
        fil->multicast.mode = VIR_NETDEV_RX_FILTER_MODE_ALL;
    } else {
        if (virNetDevGetRcvMulti(ifname, &receive))
            goto cleanup;

        if (receive)
            fil->multicast.mode = VIR_NETDEV_RX_FILTER_MODE_NORMAL;
        else
            fil->multicast.mode = VIR_NETDEV_RX_FILTER_MODE_NONE;
    }

2818 2819 2820 2821 2822 2823 2824 2825 2826 2827
    ret = 0;
 cleanup:
    if (ret < 0) {
        virNetDevRxFilterFree(fil);
        fil = NULL;
    }

    *filter = fil;
    return ret;
}
2828

R
Roman Bogorodskiy 已提交
2829
#if defined(SIOCETHTOOL) && defined(HAVE_STRUCT_IFREQ)
2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852

/**
 * virNetDevRDMAFeature
 * This function checks for the availability of RDMA feature
 * and add it to bitmap
 *
 * @ifname: name of the interface
 * @out: add RDMA feature if exist to bitmap
 *
 * Returns 0 on success, -1 on failure.
 */
static int
virNetDevRDMAFeature(const char *ifname,
                     virBitmapPtr *out)
{
    char *eth_devpath = NULL;
    char *ib_devpath = NULL;
    char *eth_res_buf = NULL;
    char *ib_res_buf = NULL;
    DIR *dirp = NULL;
    struct dirent *dp;
    int ret = -1;

2853 2854 2855
    if (!virFileExists(SYSFS_INFINIBAND_DIR))
        return 0;

J
Ján Tomko 已提交
2856
    if (virDirOpen(&dirp, SYSFS_INFINIBAND_DIR) < 0)
2857
        return -1;
2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880

    if (virAsprintf(&eth_devpath, SYSFS_NET_DIR "%s/device/resource", ifname) < 0)
        goto cleanup;
    if (!virFileExists(eth_devpath))
        goto cleanup;
    if (virFileReadAll(eth_devpath, RESOURCE_FILE_LEN, &eth_res_buf) < 0)
        goto cleanup;

    while (virDirRead(dirp, &dp, SYSFS_INFINIBAND_DIR) > 0) {
        if (virAsprintf(&ib_devpath, SYSFS_INFINIBAND_DIR "%s/device/resource",
                        dp->d_name) < 0)
            continue;
        if (virFileReadAll(ib_devpath, RESOURCE_FILE_LEN, &ib_res_buf) > 0 &&
            STREQ(eth_res_buf, ib_res_buf)) {
            ignore_value(virBitmapSetBit(*out, VIR_NET_DEV_FEAT_RDMA));
            break;
        }
        VIR_FREE(ib_devpath);
        VIR_FREE(ib_res_buf);
    }
    ret = 0;

 cleanup:
J
Ján Tomko 已提交
2881
    VIR_DIR_CLOSE(dirp);
2882 2883 2884 2885 2886 2887 2888
    VIR_FREE(eth_devpath);
    VIR_FREE(ib_devpath);
    VIR_FREE(eth_res_buf);
    VIR_FREE(ib_res_buf);
    return ret;
}

2889 2890

/**
2891 2892
 * virNetDevSendEthtoolIoctl
 * This function sends ethtool ioctl request
2893
 *
2894 2895
 * @fd: socket to operate on
 * @ifr: struct ifreq with the command
2896 2897 2898 2899
 *
 * Returns 0 on success, -1 on failure.
 */
static int
2900
virNetDevSendEthtoolIoctl(int fd, struct ifreq *ifr)
2901 2902
{
    int ret = -1;
2903

2904
    ret = ioctl(fd, SIOCETHTOOL, ifr);
2905 2906
    if (ret != 0) {
        switch (errno) {
J
Ján Tomko 已提交
2907 2908 2909 2910 2911 2912 2913 2914 2915
        case EINVAL: /* kernel doesn't support SIOCETHTOOL */
            VIR_DEBUG("ethtool ioctl: invalid request");
            break;
        case EOPNOTSUPP: /* kernel doesn't support specific feature */
            VIR_DEBUG("ethtool ioctl: request not supported");
            break;
        default:
            virReportSystemError(errno, "%s", _("ethtool ioctl error"));
            break;
2916 2917
        }
    }
2918

2919 2920
    return ret;
}
2921

2922 2923 2924 2925 2926
struct virNetDevEthtoolFeatureCmd {
    const int cmd;
    const virNetDevFeature feat;
};

2927 2928

/**
2929 2930 2931
 * virNetDevFeatureAvailable
 * This function checks for the availability of a network device feature
 *
2932 2933
 * @fd: socket to operate on
 * @ifr: struct ifreq with the command
2934 2935
 * @cmd: reference to an ethtool command structure
 *
2936
 * Returns true if the feature is available, false otherwise.
2937
 */
2938
static bool
2939
virNetDevFeatureAvailable(int fd, struct ifreq *ifr, struct ethtool_value *cmd)
2940
{
2941 2942
    ifr->ifr_data = (void*)cmd;
    if (virNetDevSendEthtoolIoctl(fd, ifr) == 0 &&
2943 2944 2945
        cmd->data > 0)
        return true;
    return false;
2946 2947 2948
}


2949 2950
static void
virNetDevGetEthtoolFeatures(virBitmapPtr bitmap,
2951 2952
                            int fd,
                            struct ifreq *ifr)
2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991
{
    size_t i;
    struct ethtool_value cmd = { 0 };

    /* legacy ethtool getters */
    struct virNetDevEthtoolFeatureCmd ethtool_cmds[] = {
        {ETHTOOL_GRXCSUM, VIR_NET_DEV_FEAT_GRXCSUM},
        {ETHTOOL_GTXCSUM, VIR_NET_DEV_FEAT_GTXCSUM},
        {ETHTOOL_GSG, VIR_NET_DEV_FEAT_GSG},
        {ETHTOOL_GTSO, VIR_NET_DEV_FEAT_GTSO},
# if HAVE_DECL_ETHTOOL_GGSO
        {ETHTOOL_GGSO, VIR_NET_DEV_FEAT_GGSO},
# endif
# if HAVE_DECL_ETHTOOL_GGRO
        {ETHTOOL_GGRO, VIR_NET_DEV_FEAT_GGRO},
# endif
    };

# if HAVE_DECL_ETHTOOL_GFLAGS
    /* ethtool masks */
    struct virNetDevEthtoolFeatureCmd flags[] = {
#  if HAVE_DECL_ETH_FLAG_LRO
        {ETH_FLAG_LRO, VIR_NET_DEV_FEAT_LRO},
#  endif
#  if HAVE_DECL_ETH_FLAG_TXVLAN
        {ETH_FLAG_RXVLAN, VIR_NET_DEV_FEAT_RXVLAN},
        {ETH_FLAG_TXVLAN, VIR_NET_DEV_FEAT_TXVLAN},
#  endif
#  if HAVE_DECL_ETH_FLAG_NTUBLE
        {ETH_FLAG_NTUPLE, VIR_NET_DEV_FEAT_NTUPLE},
#  endif
#  if HAVE_DECL_ETH_FLAG_RXHASH
        {ETH_FLAG_RXHASH, VIR_NET_DEV_FEAT_RXHASH},
#  endif
    };
# endif

    for (i = 0; i < ARRAY_CARDINALITY(ethtool_cmds); i++) {
        cmd.cmd = ethtool_cmds[i].cmd;
2992
        if (virNetDevFeatureAvailable(fd, ifr, &cmd))
2993 2994 2995 2996 2997
            ignore_value(virBitmapSetBit(bitmap, ethtool_cmds[i].feat));
    }

# if HAVE_DECL_ETHTOOL_GFLAGS
    cmd.cmd = ETHTOOL_GFLAGS;
2998
    if (virNetDevFeatureAvailable(fd, ifr, &cmd)) {
2999 3000 3001 3002 3003 3004 3005 3006 3007
        for (i = 0; i < ARRAY_CARDINALITY(flags); i++) {
            if (cmd.data & flags[i].cmd)
                ignore_value(virBitmapSetBit(bitmap, flags[i].feat));
        }
    }
# endif
}


3008 3009 3010 3011 3012
# if HAVE_DECL_ETHTOOL_GFEATURES
/**
 * virNetDevGFeatureAvailable
 * This function checks for the availability of a network device gfeature
 *
3013 3014 3015
 * @fd: socket to operate on
 * @ifr: struct ifreq with the command
 * @cmd: reference to an ethtool command structure
3016
 *
3017
 * Returns true if the feature is available, false otherwise.
3018
 */
3019
static bool
3020 3021 3022
virNetDevGFeatureAvailable(int fd,
                           struct ifreq *ifr,
                           struct ethtool_gfeatures *cmd)
3023
{
3024 3025
    ifr->ifr_data = (void*)cmd;
    if (virNetDevSendEthtoolIoctl(fd, ifr) == 0)
3026 3027
        return !!FEATURE_BIT_IS_SET(cmd->features, TX_UDP_TNL, active);
    return false;
3028
}
3029 3030 3031 3032


static int
virNetDevGetEthtoolGFeatures(virBitmapPtr bitmap,
3033 3034
                             int fd,
                             struct ifreq *ifr)
3035 3036 3037 3038 3039 3040 3041 3042 3043
{
    struct ethtool_gfeatures *g_cmd;

    if (VIR_ALLOC_VAR(g_cmd,
                      struct ethtool_get_features_block, GFEATURES_SIZE) < 0)
        return -1;

    g_cmd->cmd = ETHTOOL_GFEATURES;
    g_cmd->size = GFEATURES_SIZE;
3044
    if (virNetDevGFeatureAvailable(fd, ifr, g_cmd))
3045 3046 3047 3048 3049 3050 3051
        ignore_value(virBitmapSetBit(bitmap, VIR_NET_DEV_FEAT_TXUDPTNL));
    VIR_FREE(g_cmd);
    return 0;
}
# else
static int
virNetDevGetEthtoolGFeatures(virBitmapPtr bitmap ATTRIBUTE_UNUSED,
3052
                             int fd ATTRIBUTE_UNUSED,
3053
                             struct ifreq *ifr ATTRIBUTE_UNUSED)
3054 3055 3056
{
    return 0;
}
3057 3058 3059
# endif


3060 3061 3062 3063 3064
/**
 * virNetDevGetFeatures:
 * This function gets the nic offloads features available for ifname
 *
 * @ifname: name of the interface
3065
 * @out: bitmap of the available virNetDevFeature feature bits
3066
 *
3067 3068
 * Returns 0 on success or if called from session mode, -1 on failure.
 * If called from session mode, an empty bitmap is returned.
3069 3070 3071 3072 3073
 */
int
virNetDevGetFeatures(const char *ifname,
                     virBitmapPtr *out)
{
3074 3075 3076 3077
    struct ifreq ifr;
    int ret = -1;
    int fd = -1;

3078
    if (!(*out = virBitmapNew(VIR_NET_DEV_FEAT_LAST)))
3079
        return -1;
3080

3081 3082
    if ((fd = virNetDevSetupControl(ifname, &ifr)) < 0)
        goto cleanup;
3083

3084 3085 3086 3087
    virNetDevGetEthtoolFeatures(*out, fd, &ifr);

    if (virNetDevGetEthtoolGFeatures(*out, fd, &ifr) < 0)
        goto cleanup;
3088

3089
    if (virNetDevRDMAFeature(ifname, out) < 0)
3090 3091 3092 3093 3094 3095
        goto cleanup;

    ret = 0;
 cleanup:
    VIR_FORCE_CLOSE(fd);
    return ret;
3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106
}
#else
int
virNetDevGetFeatures(const char *ifname ATTRIBUTE_UNUSED,
                     virBitmapPtr *out ATTRIBUTE_UNUSED)
{
    VIR_DEBUG("Getting network device features on %s is not implemented on this platform",
              ifname);
    return 0;
}
#endif
3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122


/**
 * virNetDevRunEthernetScript:
 * @ifname: the interface name
 * @script: the script name
 *
 * This function executes script for new tap device created by libvirt.
 * Returns 0 in case of success or -1 on failure
 */
int
virNetDevRunEthernetScript(const char *ifname, const char *script)
{
    virCommandPtr cmd;
    int ret;

3123 3124 3125 3126
    /* Not a bug! Previously we did accept script="" as a NO-OP. */
    if (STREQ(script, ""))
        return 0;

3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139
    cmd = virCommandNew(script);
    virCommandAddArgFormat(cmd, "%s", ifname);
    virCommandClearCaps(cmd);
#ifdef CAP_NET_ADMIN
    virCommandAllowCap(cmd, CAP_NET_ADMIN);
#endif
    virCommandAddEnvPassCommon(cmd);

    ret = virCommandRun(cmd, NULL);

    virCommandFree(cmd);
    return ret;
}