migration/colo.c

   1 /*
   2  * COarse-grain LOck-stepping Virtual Machines for Non-stop Service (COLO)
   3  * (a.k.a. Fault Tolerance or Continuous Replication)
   4  *
   5  * Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
   6  * Copyright (c) 2016 FUJITSU LIMITED
   7  * Copyright (c) 2016 Intel Corporation
   8  *
   9  * This work is licensed under the terms of the GNU GPL, version 2 or
  10  * later.  See the COPYING file in the top-level directory.
  11  */
  12
  13 #include "qemu/osdep.h"
  14 #include "sysemu/sysemu.h"
  15 #include "qapi/error.h"
  16 #include "qapi/qapi-commands-migration.h"
  17 #include "migration.h"
  18 #include "qemu-file.h"
  19 #include "savevm.h"
  20 #include "migration/colo.h"
  21 #include "block.h"
  22 #include "io/channel-buffer.h"
  23 #include "trace.h"
  24 #include "qemu/error-report.h"
  25 #include "qemu/main-loop.h"
  26 #include "qemu/rcu.h"
  27 #include "migration/failover.h"
  28 #include "migration/ram.h"
  29 #ifdef CONFIG_REPLICATION
  30 #include "block/replication.h"
  31 #endif
  32 #include "net/colo-compare.h"
  33 #include "net/colo.h"
  34 #include "block/block.h"
  35 #include "qapi/qapi-events-migration.h"
  36 #include "qapi/qmp/qerror.h"
  37 #include "sysemu/cpus.h"
  38 #include "sysemu/runstate.h"
  39 #include "net/filter.h"
  40
  41 static bool vmstate_loading;
  42 static Notifier packets_compare_notifier;
  43
  44 /* User need to know colo mode after COLO failover */
  45 static COLOMode last_colo_mode;
  46
  47 #define COLO_BUFFER_BASE_SIZE (4 * 1024 * 1024)
  48
  49 bool migration_in_colo_state(void)
  50 {
  51     MigrationState *s = migrate_get_current();
  52
  53     return (s->state == MIGRATION_STATUS_COLO);
  54 }
  55
  56 bool migration_incoming_in_colo_state(void)
  57 {
  58     MigrationIncomingState *mis = migration_incoming_get_current();
  59
  60     return mis && (mis->state == MIGRATION_STATUS_COLO);
  61 }
  62
  63 static bool colo_runstate_is_stopped(void)
  64 {
  65     return runstate_check(RUN_STATE_COLO) || !runstate_is_running();
  66 }
  67
  68 static void secondary_vm_do_failover(void)
  69 {
  70 /* COLO needs enable block-replication */
  71 #ifdef CONFIG_REPLICATION
  72     int old_state;
  73     MigrationIncomingState *mis = migration_incoming_get_current();
  74     Error *local_err = NULL;
  75
  76     /* Can not do failover during the process of VM's loading VMstate, Or
  77      * it will break the secondary VM.
  78      */
  79     if (vmstate_loading) {
  80         old_state = failover_set_state(FAILOVER_STATUS_ACTIVE,
  81                         FAILOVER_STATUS_RELAUNCH);
  82         if (old_state != FAILOVER_STATUS_ACTIVE) {
  83             error_report("Unknown error while do failover for secondary VM,"
  84                          "old_state: %s", FailoverStatus_str(old_state));
  85         }
  86         return;
  87     }
  88
  89     migrate_set_state(&mis->state, MIGRATION_STATUS_COLO,
  90                       MIGRATION_STATUS_COMPLETED);
  91
  92     replication_stop_all(true, &local_err);
  93     if (local_err) {
  94         error_report_err(local_err);
  95         local_err = NULL;
  96     }
  97
  98     /* Notify all filters of all NIC to do checkpoint */
  99     colo_notify_filters_event(COLO_EVENT_FAILOVER, &local_err);
 100     if (local_err) {
 101         error_report_err(local_err);
 102     }
 103
 104     if (!autostart) {
 105         error_report("\"-S\" qemu option will be ignored in secondary side");
 106         /* recover runstate to normal migration finish state */
 107         autostart = true;
 108     }
 109     /*
 110      * Make sure COLO incoming thread not block in recv or send,
 111      * If mis->from_src_file and mis->to_src_file use the same fd,
 112      * The second shutdown() will return -1, we ignore this value,
 113      * It is harmless.
 114      */
 115     if (mis->from_src_file) {
 116         qemu_file_shutdown(mis->from_src_file);
 117     }
 118     if (mis->to_src_file) {
 119         qemu_file_shutdown(mis->to_src_file);
 120     }
 121
 122     old_state = failover_set_state(FAILOVER_STATUS_ACTIVE,
 123                                    FAILOVER_STATUS_COMPLETED);
 124     if (old_state != FAILOVER_STATUS_ACTIVE) {
 125         error_report("Incorrect state (%s) while doing failover for "
 126                      "secondary VM", FailoverStatus_str(old_state));
 127         return;
 128     }
 129     /* Notify COLO incoming thread that failover work is finished */
 130     qemu_sem_post(&mis->colo_incoming_sem);
 131
 132     /* For Secondary VM, jump to incoming co */
 133     if (mis->migration_incoming_co) {
 134         qemu_coroutine_enter(mis->migration_incoming_co);
 135     }
 136 #else
 137     abort();
 138 #endif
 139 }
 140
 141 static void primary_vm_do_failover(void)
 142 {
 143 #ifdef CONFIG_REPLICATION
 144     MigrationState *s = migrate_get_current();
 145     int old_state;
 146     Error *local_err = NULL;
 147
 148     migrate_set_state(&s->state, MIGRATION_STATUS_COLO,
 149                       MIGRATION_STATUS_COMPLETED);
 150     /*
 151      * kick COLO thread which might wait at
 152      * qemu_sem_wait(&s->colo_checkpoint_sem).
 153      */
 154     colo_checkpoint_notify(s);
 155
 156     /*
 157      * Wake up COLO thread which may blocked in recv() or send(),
 158      * The s->rp_state.from_dst_file and s->to_dst_file may use the
 159      * same fd, but we still shutdown the fd for twice, it is harmless.
 160      */
 161     if (s->to_dst_file) {
 162         qemu_file_shutdown(s->to_dst_file);
 163     }
 164     if (s->rp_state.from_dst_file) {
 165         qemu_file_shutdown(s->rp_state.from_dst_file);
 166     }
 167
 168     old_state = failover_set_state(FAILOVER_STATUS_ACTIVE,
 169                                    FAILOVER_STATUS_COMPLETED);
 170     if (old_state != FAILOVER_STATUS_ACTIVE) {
 171         error_report("Incorrect state (%s) while doing failover for Primary VM",
 172                      FailoverStatus_str(old_state));
 173         return;
 174     }
 175
 176     replication_stop_all(true, &local_err);
 177     if (local_err) {
 178         error_report_err(local_err);
 179         local_err = NULL;
 180     }
 181
 182     /* Notify COLO thread that failover work is finished */
 183     qemu_sem_post(&s->colo_exit_sem);
 184 #else
 185     abort();
 186 #endif
 187 }
 188
 189 COLOMode get_colo_mode(void)
 190 {
 191     if (migration_in_colo_state()) {
 192         return COLO_MODE_PRIMARY;
 193     } else if (migration_incoming_in_colo_state()) {
 194         return COLO_MODE_SECONDARY;
 195     } else {
 196         return COLO_MODE_NONE;
 197     }
 198 }
 199
 200 void colo_do_failover(void)
 201 {
 202     /* Make sure VM stopped while failover happened. */
 203     if (!colo_runstate_is_stopped()) {
 204         vm_stop_force_state(RUN_STATE_COLO);
 205     }
 206
 207     switch (last_colo_mode = get_colo_mode()) {
 208     case COLO_MODE_PRIMARY:
 209         primary_vm_do_failover();
 210         break;
 211     case COLO_MODE_SECONDARY:
 212         secondary_vm_do_failover();
 213         break;
 214     default:
 215         error_report("colo_do_failover failed because the colo mode"
 216                      " could not be obtained");
 217     }
 218 }
 219
 220 #ifdef CONFIG_REPLICATION
 221 void qmp_xen_set_replication(bool enable, bool primary,
 222                              bool has_failover, bool failover,
 223                              Error **errp)
 224 {
 225     ReplicationMode mode = primary ?
 226                            REPLICATION_MODE_PRIMARY :
 227                            REPLICATION_MODE_SECONDARY;
 228
 229     if (has_failover && enable) {
 230         error_setg(errp, "Parameter 'failover' is only for"
 231                    " stopping replication");
 232         return;
 233     }
 234
 235     if (enable) {
 236         replication_start_all(mode, errp);
 237     } else {
 238         if (!has_failover) {
 239             failover = NULL;
 240         }
 241         replication_stop_all(failover, failover ? NULL : errp);
 242     }
 243 }
 244
 245 ReplicationStatus *qmp_query_xen_replication_status(Error **errp)
 246 {
 247     Error *err = NULL;
 248     ReplicationStatus *s = g_new0(ReplicationStatus, 1);
 249
 250     replication_get_error_all(&err);
 251     if (err) {
 252         s->error = true;
 253         s->has_desc = true;
 254         s->desc = g_strdup(error_get_pretty(err));
 255     } else {
 256         s->error = false;
 257     }
 258
 259     error_free(err);
 260     return s;
 261 }
 262
 263 void qmp_xen_colo_do_checkpoint(Error **errp)
 264 {
 265     Error *err = NULL;
 266
 267     replication_do_checkpoint_all(&err);
 268     if (err) {
 269         error_propagate(errp, err);
 270         return;
 271     }
 272     /* Notify all filters of all NIC to do checkpoint */
 273     colo_notify_filters_event(COLO_EVENT_CHECKPOINT, errp);
 274 }
 275 #endif
 276
 277 COLOStatus *qmp_query_colo_status(Error **errp)
 278 {
 279     COLOStatus *s = g_new0(COLOStatus, 1);
 280
 281     s->mode = get_colo_mode();
 282     s->last_mode = last_colo_mode;
 283
 284     switch (failover_get_state()) {
 285     case FAILOVER_STATUS_NONE:
 286         s->reason = COLO_EXIT_REASON_NONE;
 287         break;
 288     case FAILOVER_STATUS_COMPLETED:
 289         s->reason = COLO_EXIT_REASON_REQUEST;
 290         break;
 291     default:
 292         if (migration_in_colo_state()) {
 293             s->reason = COLO_EXIT_REASON_PROCESSING;
 294         } else {
 295             s->reason = COLO_EXIT_REASON_ERROR;
 296         }
 297     }
 298
 299     return s;
 300 }
 301
 302 static void colo_send_message(QEMUFile *f, COLOMessage msg,
 303                               Error **errp)
 304 {
 305     int ret;
 306
 307     if (msg >= COLO_MESSAGE__MAX) {
 308         error_setg(errp, "%s: Invalid message", __func__);
 309         return;
 310     }
 311     qemu_put_be32(f, msg);
 312     qemu_fflush(f);
 313
 314     ret = qemu_file_get_error(f);
 315     if (ret < 0) {
 316         error_setg_errno(errp, -ret, "Can't send COLO message");
 317     }
 318     trace_colo_send_message(COLOMessage_str(msg));
 319 }
 320
 321 static void colo_send_message_value(QEMUFile *f, COLOMessage msg,
 322                                     uint64_t value, Error **errp)
 323 {
 324     Error *local_err = NULL;
 325     int ret;
 326
 327     colo_send_message(f, msg, &local_err);
 328     if (local_err) {
 329         error_propagate(errp, local_err);
 330         return;
 331     }
 332     qemu_put_be64(f, value);
 333     qemu_fflush(f);
 334
 335     ret = qemu_file_get_error(f);
 336     if (ret < 0) {
 337         error_setg_errno(errp, -ret, "Failed to send value for message:%s",
 338                          COLOMessage_str(msg));
 339     }
 340 }
 341
 342 static COLOMessage colo_receive_message(QEMUFile *f, Error **errp)
 343 {
 344     COLOMessage msg;
 345     int ret;
 346
 347     msg = qemu_get_be32(f);
 348     ret = qemu_file_get_error(f);
 349     if (ret < 0) {
 350         error_setg_errno(errp, -ret, "Can't receive COLO message");
 351         return msg;
 352     }
 353     if (msg >= COLO_MESSAGE__MAX) {
 354         error_setg(errp, "%s: Invalid message", __func__);
 355         return msg;
 356     }
 357     trace_colo_receive_message(COLOMessage_str(msg));
 358     return msg;
 359 }
 360
 361 static void colo_receive_check_message(QEMUFile *f, COLOMessage expect_msg,
 362                                        Error **errp)
 363 {
 364     COLOMessage msg;
 365     Error *local_err = NULL;
 366
 367     msg = colo_receive_message(f, &local_err);
 368     if (local_err) {
 369         error_propagate(errp, local_err);
 370         return;
 371     }
 372     if (msg != expect_msg) {
 373         error_setg(errp, "Unexpected COLO message %d, expected %d",
 374                           msg, expect_msg);
 375     }
 376 }
 377
 378 static uint64_t colo_receive_message_value(QEMUFile *f, uint32_t expect_msg,
 379                                            Error **errp)
 380 {
 381     Error *local_err = NULL;
 382     uint64_t value;
 383     int ret;
 384
 385     colo_receive_check_message(f, expect_msg, &local_err);
 386     if (local_err) {
 387         error_propagate(errp, local_err);
 388         return 0;
 389     }
 390
 391     value = qemu_get_be64(f);
 392     ret = qemu_file_get_error(f);
 393     if (ret < 0) {
 394         error_setg_errno(errp, -ret, "Failed to get value for COLO message: %s",
 395                          COLOMessage_str(expect_msg));
 396     }
 397     return value;
 398 }
 399
 400 static int colo_do_checkpoint_transaction(MigrationState *s,
 401                                           QIOChannelBuffer *bioc,
 402                                           QEMUFile *fb)
 403 {
 404     Error *local_err = NULL;
 405     int ret = -1;
 406
 407     colo_send_message(s->to_dst_file, COLO_MESSAGE_CHECKPOINT_REQUEST,
 408                       &local_err);
 409     if (local_err) {
 410         goto out;
 411     }
 412
 413     colo_receive_check_message(s->rp_state.from_dst_file,
 414                     COLO_MESSAGE_CHECKPOINT_REPLY, &local_err);
 415     if (local_err) {
 416         goto out;
 417     }
 418     /* Reset channel-buffer directly */
 419     qio_channel_io_seek(QIO_CHANNEL(bioc), 0, 0, NULL);
 420     bioc->usage = 0;
 421
 422     qemu_mutex_lock_iothread();
 423     if (failover_get_state() != FAILOVER_STATUS_NONE) {
 424         qemu_mutex_unlock_iothread();
 425         goto out;
 426     }
 427     vm_stop_force_state(RUN_STATE_COLO);
 428     qemu_mutex_unlock_iothread();
 429     trace_colo_vm_state_change("run", "stop");
 430     /*
 431      * Failover request bh could be called after vm_stop_force_state(),
 432      * So we need check failover_request_is_active() again.
 433      */
 434     if (failover_get_state() != FAILOVER_STATUS_NONE) {
 435         goto out;
 436     }
 437     qemu_mutex_lock_iothread();
 438
 439 #ifdef CONFIG_REPLICATION
 440     replication_do_checkpoint_all(&local_err);
 441     if (local_err) {
 442         qemu_mutex_unlock_iothread();
 443         goto out;
 444     }
 445 #else
 446         abort();
 447 #endif
 448
 449     colo_send_message(s->to_dst_file, COLO_MESSAGE_VMSTATE_SEND, &local_err);
 450     if (local_err) {
 451         qemu_mutex_unlock_iothread();
 452         goto out;
 453     }
 454     /* Note: device state is saved into buffer */
 455     ret = qemu_save_device_state(fb);
 456
 457     qemu_mutex_unlock_iothread();
 458     if (ret < 0) {
 459         goto out;
 460     }
 461
 462     if (migrate_auto_converge()) {
 463         mig_throttle_counter_reset();
 464     }
 465     /*
 466      * Only save VM's live state, which not including device state.
 467      * TODO: We may need a timeout mechanism to prevent COLO process
 468      * to be blocked here.
 469      */
 470     qemu_savevm_live_state(s->to_dst_file);
 471
 472     qemu_fflush(fb);
 473
 474     /*
 475      * We need the size of the VMstate data in Secondary side,
 476      * With which we can decide how much data should be read.
 477      */
 478     colo_send_message_value(s->to_dst_file, COLO_MESSAGE_VMSTATE_SIZE,
 479                             bioc->usage, &local_err);
 480     if (local_err) {
 481         goto out;
 482     }
 483
 484     qemu_put_buffer(s->to_dst_file, bioc->data, bioc->usage);
 485     qemu_fflush(s->to_dst_file);
 486     ret = qemu_file_get_error(s->to_dst_file);
 487     if (ret < 0) {
 488         goto out;
 489     }
 490
 491     colo_receive_check_message(s->rp_state.from_dst_file,
 492                        COLO_MESSAGE_VMSTATE_RECEIVED, &local_err);
 493     if (local_err) {
 494         goto out;
 495     }
 496
 497     qemu_event_reset(&s->colo_checkpoint_event);
 498     colo_notify_compares_event(NULL, COLO_EVENT_CHECKPOINT, &local_err);
 499     if (local_err) {
 500         goto out;
 501     }
 502
 503     colo_receive_check_message(s->rp_state.from_dst_file,
 504                        COLO_MESSAGE_VMSTATE_LOADED, &local_err);
 505     if (local_err) {
 506         goto out;
 507     }
 508
 509     ret = 0;
 510
 511     qemu_mutex_lock_iothread();
 512     vm_start();
 513     qemu_mutex_unlock_iothread();
 514     trace_colo_vm_state_change("stop", "run");
 515
 516 out:
 517     if (local_err) {
 518         error_report_err(local_err);
 519     }
 520     return ret;
 521 }
 522
 523 static void colo_compare_notify_checkpoint(Notifier *notifier, void *data)
 524 {
 525     colo_checkpoint_notify(data);
 526 }
 527
 528 static void colo_process_checkpoint(MigrationState *s)
 529 {
 530     QIOChannelBuffer *bioc;
 531     QEMUFile *fb = NULL;
 532     Error *local_err = NULL;
 533     int ret;
 534
 535     if (get_colo_mode() != COLO_MODE_PRIMARY) {
 536         error_report("COLO mode must be COLO_MODE_PRIMARY");
 537         return;
 538     }
 539
 540     failover_init_state();
 541
 542     s->rp_state.from_dst_file = qemu_file_get_return_path(s->to_dst_file);
 543     if (!s->rp_state.from_dst_file) {
 544         error_report("Open QEMUFile from_dst_file failed");
 545         goto out;
 546     }
 547
 548     packets_compare_notifier.notify = colo_compare_notify_checkpoint;
 549     colo_compare_register_notifier(&packets_compare_notifier);
 550
 551     /*
 552      * Wait for Secondary finish loading VM states and enter COLO
 553      * restore.
 554      */
 555     colo_receive_check_message(s->rp_state.from_dst_file,
 556                        COLO_MESSAGE_CHECKPOINT_READY, &local_err);
 557     if (local_err) {
 558         goto out;
 559     }
 560     bioc = qio_channel_buffer_new(COLO_BUFFER_BASE_SIZE);
 561     fb = qemu_file_new_output(QIO_CHANNEL(bioc));
 562     object_unref(OBJECT(bioc));
 563
 564     qemu_mutex_lock_iothread();
 565 #ifdef CONFIG_REPLICATION
 566     replication_start_all(REPLICATION_MODE_PRIMARY, &local_err);
 567     if (local_err) {
 568         qemu_mutex_unlock_iothread();
 569         goto out;
 570     }
 571 #else
 572         abort();
 573 #endif
 574
 575     vm_start();
 576     qemu_mutex_unlock_iothread();
 577     trace_colo_vm_state_change("stop", "run");
 578
 579     timer_mod(s->colo_delay_timer, qemu_clock_get_ms(QEMU_CLOCK_HOST) +
 580               s->parameters.x_checkpoint_delay);
 581
 582     while (s->state == MIGRATION_STATUS_COLO) {
 583         if (failover_get_state() != FAILOVER_STATUS_NONE) {
 584             error_report("failover request");
 585             goto out;
 586         }
 587
 588         qemu_event_wait(&s->colo_checkpoint_event);
 589
 590         if (s->state != MIGRATION_STATUS_COLO) {
 591             goto out;
 592         }
 593         ret = colo_do_checkpoint_transaction(s, bioc, fb);
 594         if (ret < 0) {
 595             goto out;
 596         }
 597     }
 598
 599 out:
 600     /* Throw the unreported error message after exited from loop */
 601     if (local_err) {
 602         error_report_err(local_err);
 603     }
 604
 605     if (fb) {
 606         qemu_fclose(fb);
 607     }
 608
 609     /*
 610      * There are only two reasons we can get here, some error happened
 611      * or the user triggered failover.
 612      */
 613     switch (failover_get_state()) {
 614     case FAILOVER_STATUS_COMPLETED:
 615         qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
 616                                   COLO_EXIT_REASON_REQUEST);
 617         break;
 618     default:
 619         qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
 620                                   COLO_EXIT_REASON_ERROR);
 621     }
 622
 623     /* Hope this not to be too long to wait here */
 624     qemu_sem_wait(&s->colo_exit_sem);
 625     qemu_sem_destroy(&s->colo_exit_sem);
 626
 627     /*
 628      * It is safe to unregister notifier after failover finished.
 629      * Besides, colo_delay_timer and colo_checkpoint_sem can't be
 630      * released before unregister notifier, or there will be use-after-free
 631      * error.
 632      */
 633     colo_compare_unregister_notifier(&packets_compare_notifier);
 634     timer_free(s->colo_delay_timer);
 635     qemu_event_destroy(&s->colo_checkpoint_event);
 636
 637     /*
 638      * Must be called after failover BH is completed,
 639      * Or the failover BH may shutdown the wrong fd that
 640      * re-used by other threads after we release here.
 641      */
 642     if (s->rp_state.from_dst_file) {
 643         qemu_fclose(s->rp_state.from_dst_file);
 644         s->rp_state.from_dst_file = NULL;
 645     }
 646 }
 647
 648 void colo_checkpoint_notify(void *opaque)
 649 {
 650     MigrationState *s = opaque;
 651     int64_t next_notify_time;
 652
 653     qemu_event_set(&s->colo_checkpoint_event);
 654     s->colo_checkpoint_time = qemu_clock_get_ms(QEMU_CLOCK_HOST);
 655     next_notify_time = s->colo_checkpoint_time +
 656                     s->parameters.x_checkpoint_delay;
 657     timer_mod(s->colo_delay_timer, next_notify_time);
 658 }
 659
 660 void migrate_start_colo_process(MigrationState *s)
 661 {
 662     qemu_mutex_unlock_iothread();
 663     qemu_event_init(&s->colo_checkpoint_event, false);
 664     s->colo_delay_timer =  timer_new_ms(QEMU_CLOCK_HOST,
 665                                 colo_checkpoint_notify, s);
 666
 667     qemu_sem_init(&s->colo_exit_sem, 0);
 668     colo_process_checkpoint(s);
 669     qemu_mutex_lock_iothread();
 670 }
 671
 672 static void colo_incoming_process_checkpoint(MigrationIncomingState *mis,
 673                       QEMUFile *fb, QIOChannelBuffer *bioc, Error **errp)
 674 {
 675     uint64_t total_size;
 676     uint64_t value;
 677     Error *local_err = NULL;
 678     int ret;
 679
 680     qemu_mutex_lock_iothread();
 681     vm_stop_force_state(RUN_STATE_COLO);
 682     qemu_mutex_unlock_iothread();
 683     trace_colo_vm_state_change("run", "stop");
 684
 685     /* FIXME: This is unnecessary for periodic checkpoint mode */
 686     colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_REPLY,
 687                  &local_err);
 688     if (local_err) {
 689         error_propagate(errp, local_err);
 690         return;
 691     }
 692
 693     colo_receive_check_message(mis->from_src_file,
 694                        COLO_MESSAGE_VMSTATE_SEND, &local_err);
 695     if (local_err) {
 696         error_propagate(errp, local_err);
 697         return;
 698     }
 699
 700     qemu_mutex_lock_iothread();
 701     cpu_synchronize_all_states();
 702     ret = qemu_loadvm_state_main(mis->from_src_file, mis);
 703     qemu_mutex_unlock_iothread();
 704
 705     if (ret < 0) {
 706         error_setg(errp, "Load VM's live state (ram) error");
 707         return;
 708     }
 709
 710     value = colo_receive_message_value(mis->from_src_file,
 711                              COLO_MESSAGE_VMSTATE_SIZE, &local_err);
 712     if (local_err) {
 713         error_propagate(errp, local_err);
 714         return;
 715     }
 716
 717     /*
 718      * Read VM device state data into channel buffer,
 719      * It's better to re-use the memory allocated.
 720      * Here we need to handle the channel buffer directly.
 721      */
 722     if (value > bioc->capacity) {
 723         bioc->capacity = value;
 724         bioc->data = g_realloc(bioc->data, bioc->capacity);
 725     }
 726     total_size = qemu_get_buffer(mis->from_src_file, bioc->data, value);
 727     if (total_size != value) {
 728         error_setg(errp, "Got %" PRIu64 " VMState data, less than expected"
 729                     " %" PRIu64, total_size, value);
 730         return;
 731     }
 732     bioc->usage = total_size;
 733     qio_channel_io_seek(QIO_CHANNEL(bioc), 0, 0, NULL);
 734
 735     colo_send_message(mis->to_src_file, COLO_MESSAGE_VMSTATE_RECEIVED,
 736                  &local_err);
 737     if (local_err) {
 738         error_propagate(errp, local_err);
 739         return;
 740     }
 741
 742     qemu_mutex_lock_iothread();
 743     vmstate_loading = true;
 744     colo_flush_ram_cache();
 745     ret = qemu_load_device_state(fb);
 746     if (ret < 0) {
 747         error_setg(errp, "COLO: load device state failed");
 748         vmstate_loading = false;
 749         qemu_mutex_unlock_iothread();
 750         return;
 751     }
 752
 753 #ifdef CONFIG_REPLICATION
 754     replication_get_error_all(&local_err);
 755     if (local_err) {
 756         error_propagate(errp, local_err);
 757         vmstate_loading = false;
 758         qemu_mutex_unlock_iothread();
 759         return;
 760     }
 761
 762     /* discard colo disk buffer */
 763     replication_do_checkpoint_all(&local_err);
 764     if (local_err) {
 765         error_propagate(errp, local_err);
 766         vmstate_loading = false;
 767         qemu_mutex_unlock_iothread();
 768         return;
 769     }
 770 #else
 771     abort();
 772 #endif
 773     /* Notify all filters of all NIC to do checkpoint */
 774     colo_notify_filters_event(COLO_EVENT_CHECKPOINT, &local_err);
 775
 776     if (local_err) {
 777         error_propagate(errp, local_err);
 778         vmstate_loading = false;
 779         qemu_mutex_unlock_iothread();
 780         return;
 781     }
 782
 783     vmstate_loading = false;
 784     vm_start();
 785     qemu_mutex_unlock_iothread();
 786     trace_colo_vm_state_change("stop", "run");
 787
 788     if (failover_get_state() == FAILOVER_STATUS_RELAUNCH) {
 789         return;
 790     }
 791
 792     colo_send_message(mis->to_src_file, COLO_MESSAGE_VMSTATE_LOADED,
 793                  &local_err);
 794     error_propagate(errp, local_err);
 795 }
 796
 797 static void colo_wait_handle_message(MigrationIncomingState *mis,
 798                 QEMUFile *fb, QIOChannelBuffer *bioc, Error **errp)
 799 {
 800     COLOMessage msg;
 801     Error *local_err = NULL;
 802
 803     msg = colo_receive_message(mis->from_src_file, &local_err);
 804     if (local_err) {
 805         error_propagate(errp, local_err);
 806         return;
 807     }
 808
 809     switch (msg) {
 810     case COLO_MESSAGE_CHECKPOINT_REQUEST:
 811         colo_incoming_process_checkpoint(mis, fb, bioc, errp);
 812         break;
 813     default:
 814         error_setg(errp, "Got unknown COLO message: %d", msg);
 815         break;
 816     }
 817 }
 818
 819 void colo_shutdown(void)
 820 {
 821     MigrationIncomingState *mis = NULL;
 822     MigrationState *s = NULL;
 823
 824     switch (get_colo_mode()) {
 825     case COLO_MODE_PRIMARY:
 826         s = migrate_get_current();
 827         qemu_event_set(&s->colo_checkpoint_event);
 828         qemu_sem_post(&s->colo_exit_sem);
 829         break;
 830     case COLO_MODE_SECONDARY:
 831         mis = migration_incoming_get_current();
 832         qemu_sem_post(&mis->colo_incoming_sem);
 833         break;
 834     default:
 835         break;
 836     }
 837 }
 838
 839 void *colo_process_incoming_thread(void *opaque)
 840 {
 841     MigrationIncomingState *mis = opaque;
 842     QEMUFile *fb = NULL;
 843     QIOChannelBuffer *bioc = NULL; /* Cache incoming device state */
 844     Error *local_err = NULL;
 845
 846     rcu_register_thread();
 847     qemu_sem_init(&mis->colo_incoming_sem, 0);
 848
 849     migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
 850                       MIGRATION_STATUS_COLO);
 851
 852     if (get_colo_mode() != COLO_MODE_SECONDARY) {
 853         error_report("COLO mode must be COLO_MODE_SECONDARY");
 854         return NULL;
 855     }
 856
 857     failover_init_state();
 858
 859     mis->to_src_file = qemu_file_get_return_path(mis->from_src_file);
 860     if (!mis->to_src_file) {
 861         error_report("COLO incoming thread: Open QEMUFile to_src_file failed");
 862         goto out;
 863     }
 864     /*
 865      * Note: the communication between Primary side and Secondary side
 866      * should be sequential, we set the fd to unblocked in migration incoming
 867      * coroutine, and here we are in the COLO incoming thread, so it is ok to
 868      * set the fd back to blocked.
 869      */
 870     qemu_file_set_blocking(mis->from_src_file, true);
 871
 872     colo_incoming_start_dirty_log();
 873
 874     bioc = qio_channel_buffer_new(COLO_BUFFER_BASE_SIZE);
 875     fb = qemu_file_new_input(QIO_CHANNEL(bioc));
 876     object_unref(OBJECT(bioc));
 877
 878     qemu_mutex_lock_iothread();
 879 #ifdef CONFIG_REPLICATION
 880     replication_start_all(REPLICATION_MODE_SECONDARY, &local_err);
 881     if (local_err) {
 882         qemu_mutex_unlock_iothread();
 883         goto out;
 884     }
 885 #else
 886         abort();
 887 #endif
 888     vm_start();
 889     qemu_mutex_unlock_iothread();
 890     trace_colo_vm_state_change("stop", "run");
 891
 892     colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_READY,
 893                       &local_err);
 894     if (local_err) {
 895         goto out;
 896     }
 897
 898     while (mis->state == MIGRATION_STATUS_COLO) {
 899         colo_wait_handle_message(mis, fb, bioc, &local_err);
 900         if (local_err) {
 901             error_report_err(local_err);
 902             break;
 903         }
 904
 905         if (failover_get_state() == FAILOVER_STATUS_RELAUNCH) {
 906             failover_set_state(FAILOVER_STATUS_RELAUNCH,
 907                             FAILOVER_STATUS_NONE);
 908             failover_request_active(NULL);
 909             break;
 910         }
 911
 912         if (failover_get_state() != FAILOVER_STATUS_NONE) {
 913             error_report("failover request");
 914             break;
 915         }
 916     }
 917
 918 out:
 919     /*
 920      * There are only two reasons we can get here, some error happened
 921      * or the user triggered failover.
 922      */
 923     switch (failover_get_state()) {
 924     case FAILOVER_STATUS_COMPLETED:
 925         qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
 926                                   COLO_EXIT_REASON_REQUEST);
 927         break;
 928     default:
 929         qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
 930                                   COLO_EXIT_REASON_ERROR);
 931     }
 932
 933     if (fb) {
 934         qemu_fclose(fb);
 935     }
 936
 937     /* Hope this not to be too long to loop here */
 938     qemu_sem_wait(&mis->colo_incoming_sem);
 939     qemu_sem_destroy(&mis->colo_incoming_sem);
 940
 941     rcu_unregister_thread();
 942     return NULL;
 943 }