u-boot.git - Unnamed repository; edit this file 'description' to name the repository.

diff options

Diffstat (limited to 'include/netdev.h')

0 files changed, 0 insertions, 0 deletions

ef='#n40'>40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 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 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 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 1276 1277 1278 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 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 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 1375 1376 1377 1378 1379 1380

/* libguestfs
 * Copyright (C) 2009-2011 Red Hat Inc.
 *
 * 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 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
 * License along with this library; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#include <config.h>

#define _BSD_SOURCE /* for mkdtemp, usleep */

#include <stdio.h>
#include <stdlib.h>
#include <stdarg.h>
#include <stddef.h>
#include <stdint.h>
#include <inttypes.h>
#include <unistd.h>
#include <string.h>
#include <fcntl.h>
#include <time.h>
#include <sys/stat.h>
#include <sys/select.h>
#include <dirent.h>
#include <signal.h>
#include <assert.h>

#include <rpc/types.h>
#include <rpc/xdr.h>

#ifdef HAVE_ERRNO_H
#include <errno.h>
#endif

#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif

#ifdef HAVE_SYS_WAIT_H
#include <sys/wait.h>
#endif

#ifdef HAVE_SYS_SOCKET_H
#include <sys/socket.h>
#endif

#ifdef HAVE_SYS_UN_H
#include <sys/un.h>
#endif

#include <arpa/inet.h>
#include <netinet/in.h>

#include "c-ctype.h"
#include "ignore-value.h"
#include "glthread/lock.h"

#include "guestfs.h"
#include "guestfs-internal.h"
#include "guestfs-internal-actions.h"
#include "guestfs_protocol.h"

static int launch_appliance (guestfs_h *g);
static int64_t timeval_diff (const struct timeval *x, const struct timeval *y);
static void print_qemu_command_line (guestfs_h *g, char **argv);
static int connect_unix_socket (guestfs_h *g, const char *sock);
static int qemu_supports (guestfs_h *g, const char *option);
static char *qemu_drive_param (guestfs_h *g, const struct drive *drv);

#if 0
static int qemu_supports_re (guestfs_h *g, const pcre *option_regex);

static void compile_regexps (void) __attribute__((constructor));
static void free_regexps (void) __attribute__((destructor));

static void
compile_regexps (void)
{
  const char *err;
  int offset;

#define COMPILE(re,pattern,options)                                     \
  do {                                                                  \
    re = pcre_compile ((pattern), (options), &err, &offset, NULL);      \
    if (re == NULL) {                                                   \
      ignore_value (write (2, err, strlen (err)));                      \
      abort ();                                                         \
    }                                                                   \
  } while (0)
}

static void
free_regexps (void)
{
}
#endif

/* Functions to add a string to the current command line. */
static void
alloc_cmdline (guestfs_h *g)
{
  if (g->cmdline == NULL) {
    /* g->cmdline[0] is reserved for argv[0], set in guestfs_launch. */
    g->cmdline_size = 1;
    g->cmdline = safe_malloc (g, sizeof (char *));
    g->cmdline[0] = NULL;
  }
}

static void
incr_cmdline_size (guestfs_h *g)
{
  alloc_cmdline (g);
  g->cmdline_size++;
  g->cmdline = safe_realloc (g, g->cmdline, sizeof (char *) * g->cmdline_size);
}

static int
add_cmdline (guestfs_h *g, const char *str)
{
  if (g->state != CONFIG) {
    error (g,
        _("command line cannot be altered after qemu subprocess launched"));
    return -1;
  }

  incr_cmdline_size (g);
  g->cmdline[g->cmdline_size-1] = safe_strdup (g, str);
  return 0;
}

struct drive **
guestfs___checkpoint_drives (guestfs_h *g)
{
  struct drive **i = &g->drives;
  while (*i != NULL) i = &((*i)->next);
  return i;
}

void
guestfs___rollback_drives (guestfs_h *g, struct drive **i)
{
  guestfs___free_drives(i);
}

/* Internal command to return the command line. */
char **
guestfs__debug_cmdline (guestfs_h *g)
{
  size_t i;
  char **r;

  alloc_cmdline (g);

  r = safe_malloc (g, sizeof (char *) * (g->cmdline_size + 1));
  r[0] = safe_strdup (g, g->qemu); /* g->cmdline[0] is always NULL */

  for (i = 1; i < g->cmdline_size; ++i)
    r[i] = safe_strdup (g, g->cmdline[i]);

  r[g->cmdline_size] = NULL;

  return r;                     /* caller frees */
}

/* Internal command to return the list of drives. */
char **
guestfs__debug_drives (guestfs_h *g)
{
  size_t i, count;
  char **ret;
  struct drive *drv;

  for (count = 0, drv = g->drives; drv; count++, drv = drv->next)
    ;

  ret = safe_malloc (g, sizeof (char *) * (count + 1));

  for (i = 0, drv = g->drives; drv; i++, drv = drv->next)
    ret[i] = qemu_drive_param (g, drv);

  ret[count] = NULL;

  return ret;                   /* caller frees */
}

int
guestfs__config (guestfs_h *g,
                 const char *qemu_param, const char *qemu_value)
{
  if (qemu_param[0] != '-') {
    error (g, _("guestfs_config: parameter must begin with '-' character"));
    return -1;
  }

  /* A bit fascist, but the user will probably break the extra
   * parameters that we add if they try to set any of these.
   */
  if (STREQ (qemu_param, "-kernel") ||
      STREQ (qemu_param, "-initrd") ||
      STREQ (qemu_param, "-nographic") ||
      STREQ (qemu_param, "-serial") ||
      STREQ (qemu_param, "-full-screen") ||
      STREQ (qemu_param, "-std-vga") ||
      STREQ (qemu_param, "-vnc")) {
    error (g, _("guestfs_config: parameter '%s' isn't allowed"), qemu_param);
    return -1;
  }

  if (add_cmdline (g, qemu_param) != 0) return -1;

  if (qemu_value != NULL) {
    if (add_cmdline (g, qemu_value) != 0) return -1;
  }

  return 0;
}

/* cache=off improves reliability in the event of a host crash.
 *
 * However this option causes qemu to try to open the file with
 * O_DIRECT.  This fails on some filesystem types (notably tmpfs).
 * So we check if we can open the file with or without O_DIRECT,
 * and use cache=off (or not) accordingly.
 *
 * NB: This function is only called on the !readonly path.  We must
 * try to open with O_RDWR to test that the file is readable and
 * writable here.
 */
static int
test_cache_off (guestfs_h *g, const char *filename)
{
  int fd = open (filename, O_RDWR|O_DIRECT);
  if (fd >= 0) {
    close (fd);
    return 1;
  }

  fd = open (filename, O_RDWR);
  if (fd >= 0) {
    close (fd);
    return 0;
  }

  perrorf (g, "%s", filename);
  return -1;
}

/* Check string parameter matches ^[-_[:alnum:]]+$ (in C locale). */
static int
valid_format_iface (const char *str)
{
  size_t len = strlen (str);

  if (len == 0)
    return 0;

  while (len > 0) {
    char c = *str++;
    len--;
    if (c != '-' && c != '_' && !c_isalnum (c))
      return 0;
  }
  return 1;
}

int
guestfs__add_drive_opts (guestfs_h *g, const char *filename,
                         const struct guestfs_add_drive_opts_argv *optargs)
{
  int readonly;
  char *format;
  char *iface;
  char *name;
  int use_cache_off;

  if (strchr (filename, ',') != NULL) {
    error (g, _("filename cannot contain ',' (comma) character"));
    return -1;
  }

  readonly = optargs->bitmask & GUESTFS_ADD_DRIVE_OPTS_READONLY_BITMASK
             ? optargs->readonly : 0;
  format = optargs->bitmask & GUESTFS_ADD_DRIVE_OPTS_FORMAT_BITMASK
           ? safe_strdup (g, optargs->format) : NULL;
  iface = optargs->bitmask & GUESTFS_ADD_DRIVE_OPTS_IFACE_BITMASK
          ? safe_strdup (g, optargs->iface) : safe_strdup (g, DRIVE_IF);
  name = optargs->bitmask & GUESTFS_ADD_DRIVE_OPTS_NAME_BITMASK
          ? safe_strdup (g, optargs->name) : NULL;

  if (format && !valid_format_iface (format)) {
    error (g, _("%s parameter is empty or contains disallowed characters"),
           "format");
    free (format);
    free (iface);
    free (name);
    return -1;
  }
  if (!valid_format_iface (iface)) {
    error (g, _("%s parameter is empty or contains disallowed characters"),
           "iface");
    free (format);
    free (iface);
    free (name);
    return -1;
  }

  /* For writable files, see if we can use cache=off.  This also
   * checks for the existence of the file.  For readonly we have
   * to do the check explicitly.
   */
  use_cache_off = readonly ? 0 : test_cache_off (g, filename);
  if (use_cache_off == -1) {
    free (format);
    free (iface);
    free (name);
    return -1;
  }

  if (readonly) {
    if (access (filename, R_OK) == -1) {
      perrorf (g, "%s", filename);
      free (format);
      free (iface);
      free (name);
      return -1;
    }
  }

  struct drive **i = &(g->drives);
  while (*i != NULL) i = &((*i)->next);

  *i = safe_malloc (g, sizeof (struct drive));
  (*i)->next = NULL;
  (*i)->path = safe_strdup (g, filename);
  (*i)->readonly = readonly;
  (*i)->format = format;
  (*i)->iface = iface;
  (*i)->name = name;
  (*i)->use_cache_off = use_cache_off;

  return 0;
}

int
guestfs__add_drive (guestfs_h *g, const char *filename)
{
  struct guestfs_add_drive_opts_argv optargs = {
    .bitmask = 0,
  };

  return guestfs__add_drive_opts (g, filename, &optargs);
}

int
guestfs__add_drive_ro (guestfs_h *g, const char *filename)
{
  struct guestfs_add_drive_opts_argv optargs = {
    .bitmask = GUESTFS_ADD_DRIVE_OPTS_READONLY_BITMASK,
    .readonly = 1,
  };

  return guestfs__add_drive_opts (g, filename, &optargs);
}

int
guestfs__add_drive_with_if (guestfs_h *g, const char *filename,
                            const char *iface)
{
  struct guestfs_add_drive_opts_argv optargs = {
    .bitmask = GUESTFS_ADD_DRIVE_OPTS_IFACE_BITMASK,
    .iface = iface,
  };

  return guestfs__add_drive_opts (g, filename, &optargs);
}

int
guestfs__add_drive_ro_with_if (guestfs_h *g, const char *filename,
                               const char *iface)
{
  struct guestfs_add_drive_opts_argv optargs = {
    .bitmask = GUESTFS_ADD_DRIVE_OPTS_IFACE_BITMASK
             | GUESTFS_ADD_DRIVE_OPTS_READONLY_BITMASK,
    .iface = iface,
    .readonly = 1,
  };

  return guestfs__add_drive_opts (g, filename, &optargs);
}

int
guestfs__add_cdrom (guestfs_h *g, const char *filename)
{
  if (strchr (filename, ',') != NULL) {
    error (g, _("filename cannot contain ',' (comma) character"));
    return -1;
  }

  if (access (filename, F_OK) == -1) {
    perrorf (g, "%s", filename);
    return -1;
  }

  return guestfs__config (g, "-cdrom", filename);
}

static int is_openable (guestfs_h *g, const char *path, int flags);

int
guestfs__launch (guestfs_h *g)
{
  /* Configured? */
  if (g->state != CONFIG) {
    error (g, _("the libguestfs handle has already been launched"));
    return -1;
  }

  TRACE0 (launch_start);

  /* Make the temporary directory. */
  if (!g->tmpdir) {
    TMP_TEMPLATE_ON_STACK (dir_template);
    g->tmpdir = safe_strdup (g, dir_template);
    if (mkdtemp (g->tmpdir) == NULL) {
      perrorf (g, _("%s: cannot create temporary directory"), dir_template);
      return -1;
    }
  }

  /* Allow anyone to read the temporary directory.  The socket in this
   * directory won't be readable but anyone can see it exists if they
   * want. (RHBZ#610880).
   */
  if (chmod (g->tmpdir, 0755) == -1)
    warning (g, "chmod: %s: %m (ignored)", g->tmpdir);

  /* Launch the appliance or attach to an existing daemon. */
  switch (g->attach_method) {
  case ATTACH_METHOD_APPLIANCE:
    return launch_appliance (g);

  case ATTACH_METHOD_UNIX:
    return connect_unix_socket (g, g->attach_method_arg);

  default:
    abort ();
  }
}

static int
launch_appliance (guestfs_h *g)
{
  int r;
  int wfd[2], rfd[2];
  char guestfsd_sock[256];
  struct sockaddr_un addr;

  /* At present you must add drives before starting the appliance.  In
   * future when we enable hotplugging you won't need to do this.
   */
  if (!g->drives) {
    error (g, _("you must call guestfs_add_drive before guestfs_launch"));
    return -1;
  }

  /* Start the clock ... */
  gettimeofday (&g->launch_t, NULL);
  guestfs___launch_send_progress (g, 0);

  TRACE0 (launch_build_appliance_start);

  /* Locate and/or build the appliance. */
  char *kernel = NULL, *initrd = NULL, *appliance = NULL;
  if (guestfs___build_appliance (g, &kernel, &initrd, &appliance) == -1)
    return -1;

  TRACE0 (launch_build_appliance_end);

  guestfs___launch_send_progress (g, 3);

  if (g->verbose)
    guestfs___print_timestamped_message (g, "begin testing qemu features");

  /* Get qemu help text and version. */
  if (qemu_supports (g, NULL) == -1)
    goto cleanup0;

  /* Using virtio-serial, we need to create a local Unix domain socket
   * for qemu to connect to.
   */
  snprintf (guestfsd_sock, sizeof guestfsd_sock, "%s/guestfsd.sock", g->tmpdir);
  unlink (guestfsd_sock);

  g->sock = socket (AF_UNIX, SOCK_STREAM, 0);
  if (g->sock == -1) {
    perrorf (g, "socket");
    goto cleanup0;
  }

  if (fcntl (g->sock, F_SETFL, O_NONBLOCK) == -1) {
    perrorf (g, "fcntl");
    goto cleanup0;
  }

  addr.sun_family = AF_UNIX;
  strncpy (addr.sun_path, guestfsd_sock, UNIX_PATH_MAX);
  addr.sun_path[UNIX_PATH_MAX-1] = '\0';

  if (bind (g->sock, &addr, sizeof addr) == -1) {
    perrorf (g, "bind");
    goto cleanup0;
  }

  if (listen (g->sock, 1) == -1) {
    perrorf (g, "listen");
    goto cleanup0;
  }

  if (!g->direct) {
    if (pipe (wfd) == -1 || pipe (rfd) == -1) {
      perrorf (g, "pipe");
      goto cleanup0;
    }
  }

  if (g->verbose)
    guestfs___print_timestamped_message (g, "finished testing qemu features");

  r = fork ();
  if (r == -1) {
    perrorf (g, "fork");
    if (!g->direct) {
      close (wfd[0]);
      close (wfd[1]);
      close (rfd[0]);
      close (rfd[1]);
    }
    goto cleanup0;
  }

  if (r == 0) {			/* Child (qemu). */
    char buf[256];

    /* Set up the full command line.  Do this in the subprocess so we
     * don't need to worry about cleaning up.
     */

    /* Set g->cmdline[0] to the name of the qemu process.  However
     * it is possible that no g->cmdline has been allocated yet so
     * we must do that first.
     */
    alloc_cmdline (g);
    g->cmdline[0] = g->qemu;

    /* CVE-2011-4127 mitigation: Disable SCSI ioctls on virtio-blk
     * devices.  The -global option must exist, but you can pass any
     * strings to it so we don't need to check for the specific virtio
     * feature.
     */
    if (qemu_supports (g, "-global")) {
      add_cmdline (g, "-global");
      add_cmdline (g, "virtio-blk-pci.scsi=off");
    }

    /* Add drives */
    struct drive *drv = g->drives;
    while (drv != NULL) {
      /* Construct the final -drive parameter. */
      char *buf = qemu_drive_param (g, drv);

      add_cmdline (g, "-drive");
      add_cmdline (g, buf);
      free (buf);

      drv = drv->next;
    }

    if (qemu_supports (g, "-nodefconfig"))
      add_cmdline (g, "-nodefconfig");

    /* The qemu -machine option (added 2010-12) is a bit more sane
     * since it falls back through various different acceleration
     * modes, so try that first (thanks Markus Armbruster).
     */
    if (qemu_supports (g, "-machine")) {
      add_cmdline (g, "-machine");
#if QEMU_MACHINE_TYPE_IS_BROKEN
      /* Workaround for qemu 0.15: We have to add the '[type=]pc'
       * since there is no default.  This is not a permanent solution
       * because this only works on PC-like hardware.  Other platforms
       * like ppc would need a different machine type.
       *
       * This bug is fixed in qemu commit 2645c6dcaf6ea2a51a, and was
       * not a problem in qemu < 0.15.
       */
      add_cmdline (g, "pc,accel=kvm:tcg");
#else
      add_cmdline (g, "accel=kvm:tcg");
#endif
    } else {
      /* qemu sometimes needs this option to enable hardware
       * virtualization, but some versions of 'qemu-kvm' will use KVM
       * regardless (even where this option appears in the help text).
       * It is rumoured that there are versions of qemu where supplying
       * this option when hardware virtualization is not available will
       * cause qemu to fail, so we we have to check at least that
       * /dev/kvm is openable.  That's not reliable, since /dev/kvm
       * might be openable by qemu but not by us (think: SELinux) in
       * which case the user would not get hardware virtualization,
       * although at least shouldn't fail.  A giant clusterfuck with the
       * qemu command line, again.
       */
      if (qemu_supports (g, "-enable-kvm") &&
          is_openable (g, "/dev/kvm", O_RDWR))
        add_cmdline (g, "-enable-kvm");
    }

    /* Newer versions of qemu (from around 2009/12) changed the
     * behaviour of monitors so that an implicit '-monitor stdio' is
     * assumed if we are in -nographic mode and there is no other
     * -monitor option.  Only a single stdio device is allowed, so
     * this broke the '-serial stdio' option.  There is a new flag
     * called -nodefaults which gets rid of all this default crud, so
     * let's use that to avoid this and any future surprises.
     */
    if (qemu_supports (g, "-nodefaults"))
      add_cmdline (g, "-nodefaults");

    add_cmdline (g, "-nographic");

    if (g->smp > 1) {
      snprintf (buf, sizeof buf, "%d", g->smp);
      add_cmdline (g, "-smp");
      add_cmdline (g, buf);
    }

    snprintf (buf, sizeof buf, "%d", g->memsize);
    add_cmdline (g, "-m");
    add_cmdline (g, buf);

    /* Force exit instead of reboot on panic */
    add_cmdline (g, "-no-reboot");

    /* These options recommended by KVM developers to improve reliability. */
    if (qemu_supports (g, "-no-hpet"))
      add_cmdline (g, "-no-hpet");

    if (qemu_supports (g, "-rtc-td-hack"))
      add_cmdline (g, "-rtc-td-hack");

    /* Create the virtio serial bus. */
    add_cmdline (g, "-device");
    add_cmdline (g, "virtio-serial");

#if 0
    /* Use virtio-console (a variant form of virtio-serial) for the
     * guest's serial console.
     */
    add_cmdline (g, "-chardev");
    add_cmdline (g, "stdio,id=console");
    add_cmdline (g, "-device");
    add_cmdline (g, "virtconsole,chardev=console,name=org.libguestfs.console.0");
#else
    /* When the above works ...  until then: */
    add_cmdline (g, "-serial");
    add_cmdline (g, "stdio");
#endif

    /* Set up virtio-serial for the communications channel. */
    add_cmdline (g, "-chardev");
    snprintf (buf, sizeof buf, "socket,path=%s,id=channel0", guestfsd_sock);
    add_cmdline (g, buf);
    add_cmdline (g, "-device");
    add_cmdline (g, "virtserialport,chardev=channel0,name=org.libguestfs.channel.0");

    /* Enable user networking. */
    if (g->enable_network) {
      add_cmdline (g, "-netdev");
      add_cmdline (g, "user,id=usernet,net=169.254.0.0/16");
      add_cmdline (g, "-device");
      add_cmdline (g, NET_IF ",netdev=usernet");
    }

#define LINUX_CMDLINE							\
    "panic=1 "         /* force kernel to panic if daemon exits */	\
    "console=ttyS0 "   /* serial console */				\
    "udevtimeout=300 " /* good for very slow systems (RHBZ#480319) */	\
    "no_timer_check "  /* fix for RHBZ#502058 */                        \
    "acpi=off "        /* we don't need ACPI, turn it off */		\
    "printk.time=1 "   /* display timestamp before kernel messages */   \
    "cgroup_disable=memory " /* saves us about 5 MB of RAM */

    /* Linux kernel command line. */
    snprintf (buf, sizeof buf,
              LINUX_CMDLINE
              "%s "             /* (selinux) */
              "%s "             /* (verbose) */
              "TERM=%s "        /* (TERM environment variable) */
              "%s",             /* (append) */
              g->selinux ? "selinux=1 enforcing=0" : "selinux=0",
              g->verbose ? "guestfs_verbose=1" : "",
              getenv ("TERM") ? : "linux",
              g->append ? g->append : "");

    add_cmdline (g, "-kernel");
    add_cmdline (g, kernel);
    add_cmdline (g, "-initrd");
    add_cmdline (g, initrd);
    add_cmdline (g, "-append");
    add_cmdline (g, buf);

    /* Add the ext2 appliance drive (last of all). */
    if (appliance) {
      const char *cachemode = "";
      if (qemu_supports (g, "cache=")) {
        if (qemu_supports (g, "unsafe"))
          cachemode = ",cache=unsafe";
        else if (qemu_supports (g, "writeback"))
          cachemode = ",cache=writeback";
      }

      char buf2[PATH_MAX + 64];
      add_cmdline (g, "-drive");
      snprintf (buf2, sizeof buf2, "file=%s,snapshot=on,if=" DRIVE_IF "%s",
                appliance, cachemode);
      add_cmdline (g, buf2);
    }

    /* Finish off the command line. */
    incr_cmdline_size (g);
    g->cmdline[g->cmdline_size-1] = NULL;

    if (!g->direct) {
      /* Set up stdin, stdout, stderr. */
      close (0);
      close (1);
      close (wfd[1]);
      close (rfd[0]);

      /* Stdin. */
      if (dup (wfd[0]) == -1) {
      dup_failed:
        perror ("dup failed");
        _exit (EXIT_FAILURE);
      }
      /* Stdout. */
      if (dup (rfd[1]) == -1)
        goto dup_failed;

      /* Particularly since qemu 0.15, qemu spews all sorts of debug
       * information on stderr.  It is useful to both capture this and
       * not confuse casual users, so send stderr to the pipe as well.
       */
      close (2);
      if (dup (rfd[1]) == -1)
        goto dup_failed;

      close (wfd[0]);
      close (rfd[1]);
    }

    /* Dump the command line (after setting up stderr above). */
    if (g->verbose)
      print_qemu_command_line (g, g->cmdline);

    /* Put qemu in a new process group. */
    if (g->pgroup)
      setpgid (0, 0);

    setenv ("LC_ALL", "C", 1);

    TRACE0 (launch_run_qemu);

    execv (g->qemu, g->cmdline); /* Run qemu. */
    perror (g->qemu);
    _exit (EXIT_FAILURE);
  }

  /* Parent (library). */
  g->pid = r;

  free (kernel);
  kernel = NULL;
  free (initrd);
  initrd = NULL;
  free (appliance);
  appliance = NULL;

  /* Fork the recovery process off which will kill qemu if the parent
   * process fails to do so (eg. if the parent segfaults).
   */
  g->recoverypid = -1;
  if (g->recovery_proc) {
    r = fork ();
    if (r == 0) {
      pid_t qemu_pid = g->pid;
      pid_t parent_pid = getppid ();

      /* It would be nice to be able to put this in the same process
       * group as qemu (ie. setpgid (0, qemu_pid)).  However this is
       * not possible because we don't have any guarantee here that
       * the qemu process has started yet.
       */
      if (g->pgroup)
        setpgid (0, 0);

      /* Writing to argv is hideously complicated and error prone.  See:
       * http://git.postgresql.org/gitweb/?p=postgresql.git;a=blob;f=src/backend/utils/misc/ps_status.c;hb=HEAD
       */

      /* Loop around waiting for one or both of the other processes to
       * disappear.  It's fair to say this is very hairy.  The PIDs that
       * we are looking at might be reused by another process.  We are
       * effectively polling.  Is the cure worse than the disease?
       */
      for (;;) {
        if (kill (qemu_pid, 0) == -1) /* qemu's gone away, we aren't needed */
          _exit (EXIT_SUCCESS);
        if (kill (parent_pid, 0) == -1) {
          /* Parent's gone away, qemu still around, so kill qemu. */
          kill (qemu_pid, 9);
          _exit (EXIT_SUCCESS);
        }
        sleep (2);
      }
    }

    /* Don't worry, if the fork failed, this will be -1.  The recovery
     * process isn't essential.
     */
    g->recoverypid = r;
  }

  if (!g->direct) {
    /* Close the other ends of the pipe. */
    close (wfd[0]);
    close (rfd[1]);

    if (fcntl (wfd[1], F_SETFL, O_NONBLOCK) == -1 ||
        fcntl (rfd[0], F_SETFL, O_NONBLOCK) == -1) {
      perrorf (g, "fcntl");
      goto cleanup1;
    }

    g->fd[0] = wfd[1];		/* stdin of child */
    g->fd[1] = rfd[0];		/* stdout of child */
  } else {
    g->fd[0] = open ("/dev/null", O_RDWR);
    if (g->fd[0] == -1) {
      perrorf (g, "open /dev/null");
      goto cleanup1;
    }
    g->fd[1] = dup (g->fd[0]);
    if (g->fd[1] == -1) {
      perrorf (g, "dup");
      close (g->fd[0]);
      goto cleanup1;
    }
  }

  g->state = LAUNCHING;

  /* Wait for qemu to start and to connect back to us via
   * virtio-serial and send the GUESTFS_LAUNCH_FLAG message.
   */
  r = guestfs___accept_from_daemon (g);
  if (r == -1)
    goto cleanup1;

  /* NB: We reach here just because qemu has opened the socket.  It
   * does not mean the daemon is up until we read the
   * GUESTFS_LAUNCH_FLAG below.  Failures in qemu startup can still
   * happen even if we reach here, even early failures like not being
   * able to open a drive.
   */

  close (g->sock); /* Close the listening socket. */
  g->sock = r; /* This is the accepted data socket. */

  if (fcntl (g->sock, F_SETFL, O_NONBLOCK) == -1) {
    perrorf (g, "fcntl");
    goto cleanup1;
  }

  uint32_t size;
  void *buf = NULL;
  r = guestfs___recv_from_daemon (g, &size, &buf);
  free (buf);

  if (r == -1) return -1;

  if (size != GUESTFS_LAUNCH_FLAG) {
    error (g, _("guestfs_launch failed, see earlier error messages"));
    goto cleanup1;
  }

  if (g->verbose)
    guestfs___print_timestamped_message (g, "appliance is up");

  /* This is possible in some really strange situations, such as
   * guestfsd starts up OK but then qemu immediately exits.  Check for
   * it because the caller is probably expecting to be able to send
   * commands after this function returns.
   */
  if (g->state != READY) {
    error (g, _("qemu launched and contacted daemon, but state != READY"));
    goto cleanup1;
  }

  TRACE0 (launch_end);

  guestfs___launch_send_progress (g, 12);

  return 0;

 cleanup1:
  if (!g->direct) {
    close (wfd[1]);
    close (rfd[0]);
  }
  if (g->pid > 0) kill (g->pid, 9);
  if (g->recoverypid > 0) kill (g->recoverypid, 9);
  if (g->pid > 0) waitpid (g->pid, NULL, 0);
  if (g->recoverypid > 0) waitpid (g->recoverypid, NULL, 0);
  g->fd[0] = -1;
  g->fd[1] = -1;
  g->pid = 0;
  g->recoverypid = 0;
  memset (&g->launch_t, 0, sizeof g->launch_t);

 cleanup0:
  if (g->sock >= 0) {
    close (g->sock);
    g->sock = -1;
  }
  g->state = CONFIG;
  free (kernel);
  free (initrd);
  free (appliance);
  return -1;
}

/* Alternate attach method: instead of launching the appliance,
 * connect to an existing unix socket.
 */
static int
connect_unix_socket (guestfs_h *g, const char *sockpath)
{
  int r;
  struct sockaddr_un addr;

  /* Start the clock ... */
  gettimeofday (&g->launch_t, NULL);

  /* Set these to nothing so we don't try to kill random processes or
   * read from random file descriptors.
   */
  g->pid = 0;
  g->recoverypid = 0;
  g->fd[0] = -1;
  g->fd[1] = -1;

  if (g->verbose)
    guestfs___print_timestamped_message (g, "connecting to %s", sockpath);

  g->sock = socket (AF_UNIX, SOCK_STREAM, 0);
  if (g->sock == -1) {
    perrorf (g, "socket");
    return -1;
  }

  addr.sun_family = AF_UNIX;
  strncpy (addr.sun_path, sockpath, UNIX_PATH_MAX);
  addr.sun_path[UNIX_PATH_MAX-1] = '\0';

  g->state = LAUNCHING;

  if (connect (g->sock, &addr, sizeof addr) == -1) {
    perrorf (g, "bind");
    goto cleanup;
  }

  if (fcntl (g->sock, F_SETFL, O_NONBLOCK) == -1) {
    perrorf (g, "fcntl");
    goto cleanup;
  }

  uint32_t size;
  void *buf = NULL;
  r = guestfs___recv_from_daemon (g, &size, &buf);
  free (buf);

  if (r == -1) return -1;

  if (size != GUESTFS_LAUNCH_FLAG) {
    error (g, _("guestfs_launch failed, unexpected initial message from guestfsd"));
    goto cleanup;
  }

  if (g->verbose)
    guestfs___print_timestamped_message (g, "connected");

  if (g->state != READY) {
    error (g, _("contacted guestfsd, but state != READY"));
    goto cleanup;
  }

  return 0;

 cleanup:
  close (g->sock);
  return -1;
}

/* launch (of the ordinary appliance) generates approximate progress
 * messages.  Currently these are defined as follows:
 *
 *    0 / 12: launch clock starts
 *    3 / 12: appliance created
 *    6 / 12: detected that guest kernel started
 *    9 / 12: detected that /init script is running
 *   12 / 12: launch completed successfully
 *
 * Notes:
 * (1) This is not a documented ABI and the behaviour may be changed
 * or removed in future.
 * (2) Messages are only sent if more than 5 seconds has elapsed
 * since the launch clock started.
 * (3) There is a gross hack in proto.c to make this work.
 */
void
guestfs___launch_send_progress (guestfs_h *g, int perdozen)
{
  struct timeval tv;

  gettimeofday (&tv, NULL);
  if (timeval_diff (&g->launch_t, &tv) >= 5000) {
    guestfs_progress progress_message =
      { .proc = 0, .serial = 0, .position = perdozen, .total = 12 };

    guestfs___progress_message_callback (g, &progress_message);
  }
}


context:
space:
mode: