Series comparison

-[Qemu-devel] [PULL 0/6] Net patches
+[Qemu-devel] [PULL V2 00/26] Net patches
-The following changes since commit e607bbee553cfe73072870cef458cfa4e78133e2:
+The following changes since commit 77f7c747193662edfadeeb3118d63eed0eac51a6:
-  Merge remote-tracking branch 'remotes/edgar/tags/edgar/xilinx-next-2018-01-26.for-upstream' into staging (2018-01-26 14:24:25 +0000)
+  Merge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2018-10-17' into staging (2018-10-18 13:40:19 +0100)
 are available in the git repository at:
   https://github.com/jasowang/qemu.git tags/net-pull-request
-for you to fetch changes up to bf4835a4d5338bb7424827715df22570a8adc67c:
+for you to fetch changes up to 37a4442a76d010f5d957e3ee09dfb23364281b37:
-  MAINTAINERS: update Dmitry Fleytman email (2018-01-29 16:05:38 +0800)
+  qemu-options: Fix bad "macaddr" property in the documentation (2018-10-19 11:15:04 +0800)
 ----------------------------------------------------------------
 ----------------------------------------------------------------
-Mao Zhongyi (2):
+Jason Wang (5):
-      colo: modified the payload compare function
+      ne2000: fix possible out of bound access in ne2000_receive
-      colo: compare the packet based on the tcp sequence number
+      rtl8139: fix possible out of bound access
       pcnet: fix possible buffer overflow
       net: ignore packet size greater than INT_MAX
       e1000: indicate dropped packets in HW counters
-Philippe Mathieu-Daudé (1):
+Thomas Huth (1):
-      MAINTAINERS: update Dmitry Fleytman email
+      qemu-options: Fix bad "macaddr" property in the documentation
-Thomas Huth (3):
+Zhang Chen (15):
-      net: Allow hubports to connect to other netdevs
+      filter-rewriter: Add TCP state machine and fix memory leak in connection_track_table
-      net: Allow netdevs to be used with 'hostfwd_add' and 'hostfwd_remove'
+      colo-compare: implement the process of checkpoint
-      qemu-doc: Get rid of "vlan=X" example in the documentation
+      colo-compare: use notifier to notify packets comparing result
       COLO: integrate colo compare with colo frame
       COLO: Add block replication into colo process
       COLO: Remove colo_state migration struct
       COLO: Load dirty pages into SVM's RAM cache firstly
       ram/COLO: Record the dirty pages that SVM received
       COLO: Flush memory data from ram cache
       qapi/migration.json: Rename COLO unknown mode to none mode.
       qapi: Add new command to query colo status
       savevm: split the process of different stages for loadvm/savevm
       filter: Add handle_event method for NetFilterClass
       filter-rewriter: handle checkpoint and failover event
       docs: Add COLO status diagram to COLO-FT.txt
- MAINTAINERS        |   8 +-
+liujunjie (1):
- hmp-commands.hx    |   4 +-
+      clean up callback when del virtqueue
- net/colo-compare.c | 411 +++++++++++++++++++++++++++++++++--------------------
- net/colo.c         |   9 ++
+zhanghailiang (4):
- net/colo.h         |  15 ++
+      qmp event: Add COLO_EXIT event to notify users while exited COLO
- net/hub.c          |  27 +++-
+      COLO: flush host dirty ram from cache
- net/hub.h          |   3 +-
+      COLO: notify net filters about checkpoint/failover event
- net/net.c          |   2 +-
+      COLO: quick failover process by kick COLO thread
- net/slirp.c        |  33 +++--
- net/trace-events   |   2 +-
+ docs/COLO-FT.txt          |  34 ++++++++
- qapi/net.json      |   4 +-
+ hw/net/e1000.c            |  16 +++-
- qemu-options.hx    |  12 +-
+ hw/net/ne2000.c           |   4 +-
-files changed, 347 insertions(+), 183 deletions(-)
+ hw/net/pcnet.c            |   4 +-
  hw/net/rtl8139.c          |   8 +-
  hw/net/trace-events       |   3 +
  hw/virtio/virtio.c        |   2 +
  include/exec/ram_addr.h   |   1 +
  include/migration/colo.h  |  11 ++-
  include/net/filter.h      |   5 ++
  migration/Makefile.objs   |   2 +-
  migration/colo-comm.c     |  76 -----------------
  migration/colo-failover.c |   2 +-
  migration/colo.c          | 212 +++++++++++++++++++++++++++++++++++++++++++---
  migration/migration.c     |  46 ++++++++--
  migration/ram.c           | 166 +++++++++++++++++++++++++++++++++++-
  migration/ram.h           |   4 +
  migration/savevm.c        |  53 ++++++++++--
  migration/savevm.h        |   5 ++
  migration/trace-events    |   3 +
  net/colo-compare.c        | 115 ++++++++++++++++++++++---
  net/colo-compare.h        |  24 ++++++
  net/colo.c                |  10 ++-
  net/colo.h                |  11 +--
  net/filter-rewriter.c     | 166 +++++++++++++++++++++++++++++++++---
  net/filter.c              |  17 ++++
  net/net.c                 |  26 +++++-
  qapi/migration.json       |  80 +++++++++++++++--
  qemu-options.hx           |   2 +-
  vl.c                      |   2 -
 files changed, 958 insertions(+), 152 deletions(-)
  delete mode 100644 migration/colo-comm.c
  create mode 100644 net/colo-compare.h

-New patch
+[Qemu-devel] [PULL V2 01/26] filter-rewriter: Add TCP state machine and fix memory leak in connection_track_table
+From: Zhang Chen <zhangckid@gmail.com>
 We add almost full TCP state machine in filter-rewriter, except
 TCPS_LISTEN and some simplify in VM active close FIN states.
 The reason for this simplify job is because guest kernel will track
 the TCP status and wait 2MSL time too, if client resend the FIN packet,
 guest will resend the last ACK, so we needn't wait 2MSL time in filter-rewriter.
 After a net connection is closed, we didn't clear its related resources
 in connection_track_table, which will lead to memory leak.
 Let's track the state of net connection, if it is closed, its related
 resources will be cleared up.
 Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
 Signed-off-by: Zhang Chen <zhangckid@gmail.com>
 Signed-off-by: Zhang Chen <chen.zhang@intel.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
  net/colo.c            |   2 +-
  net/colo.h            |   9 ++---
  net/filter-rewriter.c | 109 +++++++++++++++++++++++++++++++++++++++++++++-----
 files changed, 104 insertions(+), 16 deletions(-)
 diff --git a/net/colo.c b/net/colo.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/colo.c
 +++ b/net/colo.c
@@ -XXX,XX +XXX,XX @@ Connection *connection_new(ConnectionKey *key)
      conn->ip_proto = key->ip_proto;
      conn->processing = false;
      conn->offset = 0;
 -    conn->syn_flag = 0;
 +    conn->tcp_state = TCPS_CLOSED;
      conn->pack = 0;
      conn->sack = 0;
      g_queue_init(&conn->primary_list);
 diff --git a/net/colo.h b/net/colo.h
 index XXXXXXX..XXXXXXX 100644
 --- a/net/colo.h
 +++ b/net/colo.h
@@ -XXX,XX +XXX,XX @@
  #include "slirp/slirp.h"
  #include "qemu/jhash.h"
  #include "qemu/timer.h"
 +#include "slirp/tcp.h"
  #define HASHTABLE_MAX_SIZE 16384
@@ -XXX,XX +XXX,XX @@ typedef struct Connection {
      uint32_t sack;
      /* offset = secondary_seq - primary_seq */
      tcp_seq  offset;
 -    /*
 -     * we use this flag update offset func
 -     * run once in independent tcp connection
 -     */
 -    int syn_flag;
 +
 +    int tcp_state; /* TCP FSM state */
 +    tcp_seq fin_ack_seq; /* the seq of 'fin=1,ack=1' */
  } Connection;
  uint32_t connection_key_hash(const void *opaque);
 diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/filter-rewriter.c
 +++ b/net/filter-rewriter.c
@@ -XXX,XX +XXX,XX @@ static int is_tcp_packet(Packet *pkt)
  }
  /* handle tcp packet from primary guest */
 -static int handle_primary_tcp_pkt(NetFilterState *nf,
 +static int handle_primary_tcp_pkt(RewriterState *rf,
                                    Connection *conn,
 -                                  Packet *pkt)
 +                                  Packet *pkt, ConnectionKey *key)
  {
      struct tcphdr *tcp_pkt;
@@ -XXX,XX +XXX,XX @@ static int handle_primary_tcp_pkt(NetFilterState *nf,
          trace_colo_filter_rewriter_conn_offset(conn->offset);
      }
 +    if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == (TH_ACK | TH_SYN)) &&
 +        conn->tcp_state == TCPS_SYN_SENT) {
 +        conn->tcp_state = TCPS_ESTABLISHED;
 +    }
 +
      if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_SYN)) {
          /*
           * we use this flag update offset func
           * run once in independent tcp connection
           */
 -        conn->syn_flag = 1;
 +        conn->tcp_state = TCPS_SYN_RECEIVED;
      }
      if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_ACK)) {
 -        if (conn->syn_flag) {
 +        if (conn->tcp_state == TCPS_SYN_RECEIVED) {
              /*
               * offset = secondary_seq - primary seq
               * ack packet sent by guest from primary node,
               * so we use th_ack - 1 get primary_seq
               */
              conn->offset -= (ntohl(tcp_pkt->th_ack) - 1);
 -            conn->syn_flag = 0;
 +            conn->tcp_state = TCPS_ESTABLISHED;
          }
          if (conn->offset) {
              /* handle packets to the secondary from the primary */
@@ -XXX,XX +XXX,XX @@ static int handle_primary_tcp_pkt(NetFilterState *nf,
              net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_hdr_len,
                                     pkt->size - pkt->vnet_hdr_len);
          }
 +
 +        /*
 +         * Passive close step 3
 +         */
 +        if ((conn->tcp_state == TCPS_LAST_ACK) &&
 +            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
 +            conn->tcp_state = TCPS_CLOSED;
 +            g_hash_table_remove(rf->connection_track_table, key);
 +        }
 +    }
 +
 +    if ((tcp_pkt->th_flags & TH_FIN) == TH_FIN) {
 +        /*
 +         * Passive close.
 +         * Step 1:
 +         * The *server* side of this connect is VM, *client* tries to close
 +         * the connection. We will into CLOSE_WAIT status.
 +         *
 +         * Step 2:
 +         * In this step we will into LAST_ACK status.
 +         *
 +         * We got 'fin=1, ack=1' packet from server side, we need to
 +         * record the seq of 'fin=1, ack=1' packet.
 +         *
 +         * Step 3:
 +         * We got 'ack=1' packets from client side, it acks 'fin=1, ack=1'
 +         * packet from server side. From this point, we can ensure that there
 +         * will be no packets in the connection, except that, some errors
 +         * happen between the path of 'filter object' and vNIC, if this rare
 +         * case really happen, we can still create a new connection,
 +         * So it is safe to remove the connection from connection_track_table.
 +         *
 +         */
 +        if (conn->tcp_state == TCPS_ESTABLISHED) {
 +            conn->tcp_state = TCPS_CLOSE_WAIT;
 +        }
 +
 +        /*
 +         * Active close step 2.
 +         */
 +        if (conn->tcp_state == TCPS_FIN_WAIT_1) {
 +            conn->tcp_state = TCPS_TIME_WAIT;
 +            /*
 +             * For simplify implementation, we needn't wait 2MSL time
 +             * in filter rewriter. Because guest kernel will track the
 +             * TCP status and wait 2MSL time, if client resend the FIN
 +             * packet, guest will apply the last ACK too.
 +             */
 +            conn->tcp_state = TCPS_CLOSED;
 +            g_hash_table_remove(rf->connection_track_table, key);
 +        }
      }
      return 0;
  }
  /* handle tcp packet from secondary guest */
 -static int handle_secondary_tcp_pkt(NetFilterState *nf,
 +static int handle_secondary_tcp_pkt(RewriterState *rf,
                                      Connection *conn,
 -                                    Packet *pkt)
 +                                    Packet *pkt, ConnectionKey *key)
  {
      struct tcphdr *tcp_pkt;
@@ -XXX,XX +XXX,XX @@ static int handle_secondary_tcp_pkt(NetFilterState *nf,
          trace_colo_filter_rewriter_conn_offset(conn->offset);
      }
 -    if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == (TH_ACK | TH_SYN))) {
 +    if (conn->tcp_state == TCPS_SYN_RECEIVED &&
 +        ((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == (TH_ACK | TH_SYN))) {
          /*
           * save offset = secondary_seq and then
           * in handle_primary_tcp_pkt make offset
@@ -XXX,XX +XXX,XX @@ static int handle_secondary_tcp_pkt(NetFilterState *nf,
          conn->offset = ntohl(tcp_pkt->th_seq);
      }
 +    /* VM active connect */
 +    if (conn->tcp_state == TCPS_CLOSED &&
 +        ((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_SYN)) {
 +        conn->tcp_state = TCPS_SYN_SENT;
 +    }
 +
      if ((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_ACK) {
          /* Only need to adjust seq while offset is Non-zero */
          if (conn->offset) {
@@ -XXX,XX +XXX,XX @@ static int handle_secondary_tcp_pkt(NetFilterState *nf,
          }
      }
 +    /*
 +     * Passive close step 2:
 +     */
 +    if (conn->tcp_state == TCPS_CLOSE_WAIT &&
 +        (tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
 +        conn->fin_ack_seq = ntohl(tcp_pkt->th_seq);
 +        conn->tcp_state = TCPS_LAST_ACK;
 +    }
 +
 +    /*
 +     * Active close
 +     *
 +     * Step 1:
 +     * The *server* side of this connect is VM, *server* tries to close
 +     * the connection.
 +     *
 +     * Step 2:
 +     * We will into CLOSE_WAIT status.
 +     * We simplify the TCPS_FIN_WAIT_2, TCPS_TIME_WAIT and
 +     * CLOSING status.
 +     */
 +    if (conn->tcp_state == TCPS_ESTABLISHED &&
 +        (tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == TH_FIN) {
 +        conn->tcp_state = TCPS_FIN_WAIT_1;
 +    }
 +
      return 0;
  }
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
          if (sender == nf->netdev) {
              /* NET_FILTER_DIRECTION_TX */
 -            if (!handle_primary_tcp_pkt(nf, conn, pkt)) {
 +            if (!handle_primary_tcp_pkt(s, conn, pkt, &key)) {
                  qemu_net_queue_send(s->incoming_queue, sender, 0,
                  (const uint8_t *)pkt->data, pkt->size, NULL);
                  packet_destroy(pkt, NULL);
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
              }
          } else {
              /* NET_FILTER_DIRECTION_RX */
 -            if (!handle_secondary_tcp_pkt(nf, conn, pkt)) {
 +            if (!handle_secondary_tcp_pkt(s, conn, pkt, &key)) {
                  qemu_net_queue_send(s->incoming_queue, sender, 0,
                  (const uint8_t *)pkt->data, pkt->size, NULL);
                  packet_destroy(pkt, NULL);
 --
 .5.0

-New patch
+[Qemu-devel] [PULL V2 02/26] colo-compare: implement the process of checkpoint
+From: Zhang Chen <zhangckid@gmail.com>
 While do checkpoint, we need to flush all the unhandled packets,
 By using the filter notifier mechanism, we can easily to notify
 every compare object to do this process, which runs inside
 of compare threads as a coroutine.
 Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
 Signed-off-by: Zhang Chen <zhangckid@gmail.com>
 Signed-off-by: Zhang Chen <chen.zhang@intel.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
  include/migration/colo.h |  6 ++++
  net/colo-compare.c       | 78 ++++++++++++++++++++++++++++++++++++++++++++++++
  net/colo-compare.h       | 22 ++++++++++++++
 files changed, 106 insertions(+)
  create mode 100644 net/colo-compare.h
 diff --git a/include/migration/colo.h b/include/migration/colo.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/migration/colo.h
 +++ b/include/migration/colo.h
@@ -XXX,XX +XXX,XX @@
  #include "qemu-common.h"
  #include "qapi/qapi-types-migration.h"
 +enum colo_event {
 +    COLO_EVENT_NONE,
 +    COLO_EVENT_CHECKPOINT,
 +    COLO_EVENT_FAILOVER,
 +};
 +
  void colo_info_init(void);
  void migrate_start_colo_process(MigrationState *s);
 diff --git a/net/colo-compare.c b/net/colo-compare.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/colo-compare.c
 +++ b/net/colo-compare.c
@@ -XXX,XX +XXX,XX @@
  #include "qemu/sockets.h"
  #include "colo.h"
  #include "sysemu/iothread.h"
 +#include "net/colo-compare.h"
 +#include "migration/colo.h"
  #define TYPE_COLO_COMPARE "colo-compare"
  #define COLO_COMPARE(obj) \
      OBJECT_CHECK(CompareState, (obj), TYPE_COLO_COMPARE)
 +static QTAILQ_HEAD(, CompareState) net_compares =
 +       QTAILQ_HEAD_INITIALIZER(net_compares);
 +
  #define COMPARE_READ_LEN_MAX NET_BUFSIZE
  #define MAX_QUEUE_SIZE 1024
@@ -XXX,XX +XXX,XX @@
  /* TODO: Should be configurable */
  #define REGULAR_PACKET_CHECK_MS 3000
 +static QemuMutex event_mtx;
 +static QemuCond event_complete_cond;
 +static int event_unhandled_count;
 +
  /*
   *  + CompareState ++
   *  |               |
@@ -XXX,XX +XXX,XX @@ typedef struct CompareState {
      IOThread *iothread;
      GMainContext *worker_context;
      QEMUTimer *packet_check_timer;
 +
 +    QEMUBH *event_bh;
 +    enum colo_event event;
 +
 +    QTAILQ_ENTRY(CompareState) next;
  } CompareState;
  typedef struct CompareClass {
@@ -XXX,XX +XXX,XX @@ static void check_old_packet_regular(void *opaque)
                  REGULAR_PACKET_CHECK_MS);
  }
 +/* Public API, Used for COLO frame to notify compare event */
 +void colo_notify_compares_event(void *opaque, int event, Error **errp)
 +{
 +    CompareState *s;
 +
 +    qemu_mutex_lock(&event_mtx);
 +    QTAILQ_FOREACH(s, &net_compares, next) {
 +        s->event = event;
 +        qemu_bh_schedule(s->event_bh);
 +        event_unhandled_count++;
 +    }
 +    /* Wait all compare threads to finish handling this event */
 +    while (event_unhandled_count > 0) {
 +        qemu_cond_wait(&event_complete_cond, &event_mtx);
 +    }
 +
 +    qemu_mutex_unlock(&event_mtx);
 +}
 +
  static void colo_compare_timer_init(CompareState *s)
  {
      AioContext *ctx = iothread_get_aio_context(s->iothread);
@@ -XXX,XX +XXX,XX @@ static void colo_compare_timer_del(CompareState *s)
      }
   }
 +static void colo_flush_packets(void *opaque, void *user_data);
 +
 +static void colo_compare_handle_event(void *opaque)
 +{
 +    CompareState *s = opaque;
 +
 +    switch (s->event) {
 +    case COLO_EVENT_CHECKPOINT:
 +        g_queue_foreach(&s->conn_list, colo_flush_packets, s);
 +        break;
 +    case COLO_EVENT_FAILOVER:
 +        break;
 +    default:
 +        break;
 +    }
 +
 +    assert(event_unhandled_count > 0);
 +
 +    qemu_mutex_lock(&event_mtx);
 +    event_unhandled_count--;
 +    qemu_cond_broadcast(&event_complete_cond);
 +    qemu_mutex_unlock(&event_mtx);
 +}
 +
  static void colo_compare_iothread(CompareState *s)
  {
      object_ref(OBJECT(s->iothread));
@@ -XXX,XX +XXX,XX @@ static void colo_compare_iothread(CompareState *s)
                               s, s->worker_context, true);
      colo_compare_timer_init(s);
 +    s->event_bh = qemu_bh_new(colo_compare_handle_event, s);
  }
  static char *compare_get_pri_indev(Object *obj, Error **errp)
@@ -XXX,XX +XXX,XX @@ static void colo_compare_complete(UserCreatable *uc, Error **errp)
      net_socket_rs_init(&s->pri_rs, compare_pri_rs_finalize, s->vnet_hdr);
      net_socket_rs_init(&s->sec_rs, compare_sec_rs_finalize, s->vnet_hdr);
 +    QTAILQ_INSERT_TAIL(&net_compares, s, next);
 +
      g_queue_init(&s->conn_list);
 +    qemu_mutex_init(&event_mtx);
 +    qemu_cond_init(&event_complete_cond);
 +
      s->connection_track_table = g_hash_table_new_full(connection_key_hash,
                                                        connection_key_equal,
                                                        g_free,
@@ -XXX,XX +XXX,XX @@ static void colo_compare_init(Object *obj)
  static void colo_compare_finalize(Object *obj)
  {
      CompareState *s = COLO_COMPARE(obj);
 +    CompareState *tmp = NULL;
      qemu_chr_fe_deinit(&s->chr_pri_in, false);
      qemu_chr_fe_deinit(&s->chr_sec_in, false);
@@ -XXX,XX +XXX,XX @@ static void colo_compare_finalize(Object *obj)
      if (s->iothread) {
          colo_compare_timer_del(s);
      }
 +
 +    qemu_bh_delete(s->event_bh);
 +
 +    QTAILQ_FOREACH(tmp, &net_compares, next) {
 +        if (tmp == s) {
 +            QTAILQ_REMOVE(&net_compares, s, next);
 +            break;
 +        }
 +    }
 +
      /* Release all unhandled packets after compare thead exited */
      g_queue_foreach(&s->conn_list, colo_flush_packets, s);
@@ -XXX,XX +XXX,XX @@ static void colo_compare_finalize(Object *obj)
      if (s->iothread) {
          object_unref(OBJECT(s->iothread));
      }
 +
 +    qemu_mutex_destroy(&event_mtx);
 +    qemu_cond_destroy(&event_complete_cond);
 +
      g_free(s->pri_indev);
      g_free(s->sec_indev);
      g_free(s->outdev);
 diff --git a/net/colo-compare.h b/net/colo-compare.h
 new file mode 100644
 index XXXXXXX..XXXXXXX
 --- /dev/null
 +++ b/net/colo-compare.h
@@ -XXX,XX +XXX,XX @@
 +/*
 + * COarse-grain LOck-stepping Virtual Machines for Non-stop Service (COLO)
 + * (a.k.a. Fault Tolerance or Continuous Replication)
 + *
 + * Copyright (c) 2017 HUAWEI TECHNOLOGIES CO., LTD.
 + * Copyright (c) 2017 FUJITSU LIMITED
 + * Copyright (c) 2017 Intel Corporation
 + *
 + * Authors:
 + *    zhanghailiang <zhang.zhanghailiang@huawei.com>
 + *    Zhang Chen <zhangckid@gmail.com>
 + *
 + * This work is licensed under the terms of the GNU GPL, version 2 or
 + * later.  See the COPYING file in the top-level directory.
 + */
 +
 +#ifndef QEMU_COLO_COMPARE_H
 +#define QEMU_COLO_COMPARE_H
 +
 +void colo_notify_compares_event(void *opaque, int event, Error **errp);
 +
 +#endif /* QEMU_COLO_COMPARE_H */
 --
 .5.0

-[Qemu-devel] [PULL 1/6] colo: modified the payload compare function
+[Qemu-devel] [PULL V2 03/26] colo-compare: use notifier to notify packets comparing result
-From: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>
+From: Zhang Chen <zhangckid@gmail.com>
-Modified the function colo_packet_compare_common to prepare for the
+It's a good idea to use notifier to notify COLO frame of
-tcp packet comparison in the next patch.
+inconsistent packets comparing.
-Cc: Zhang Chen <zhangckid@gmail.com>
-Cc: Li Zhijian <lizhijian@cn.fujitsu.com>
-Cc: Jason Wang <jasowang@redhat.com>
-Signed-off-by: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>
-Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
 Signed-off-by: Zhang Chen <zhangckid@gmail.com>
-Reviewed-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
 Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
- net/colo-compare.c | 88 +++++++++++++++++++++++++++---------------------------
+ net/colo-compare.c | 37 ++++++++++++++++++++++++++-----------
-file changed, 44 insertions(+), 44 deletions(-)
+ net/colo-compare.h |  2 ++
 files changed, 28 insertions(+), 11 deletions(-)
 diff --git a/net/colo-compare.c b/net/colo-compare.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/colo-compare.c
 +++ b/net/colo-compare.c
-@@ -XXX,XX +XXX,XX @@ static int packet_enqueue(CompareState *s, int mode, Connection **con)
+@@ -XXX,XX +XXX,XX @@
-  * return:    0  means packet same
+ #include "sysemu/iothread.h"
-  *            > 0 || < 0 means packet different
+ #include "net/colo-compare.h"
-  */
+ #include "migration/colo.h"
--static int colo_packet_compare_common(Packet *ppkt,
++#include "migration/migration.h"
--                                      Packet *spkt,
--                                      int poffset,
+ #define TYPE_COLO_COMPARE "colo-compare"
--                                      int soffset)
+ #define COLO_COMPARE(obj) \
-+static int colo_compare_packet_payload(Packet *ppkt,
+@@ -XXX,XX +XXX,XX @@
-+                                       Packet *spkt,
+ static QTAILQ_HEAD(, CompareState) net_compares =
-+                                       uint16_t poffset,
+        QTAILQ_HEAD_INITIALIZER(net_compares);
-+                                       uint16_t soffset,
-+                                       uint16_t len)
++static NotifierList colo_compare_notifiers =
 +    NOTIFIER_LIST_INITIALIZER(colo_compare_notifiers);
 +
+ #define COMPARE_READ_LEN_MAX NET_BUFSIZE
+ #define MAX_QUEUE_SIZE 1024
+@@ -XXX,XX +XXX,XX @@ static bool colo_mark_tcp_pkt(Packet *ppkt, Packet *spkt,
+     return false;
+ }
++static void colo_compare_inconsistency_notify(void)
++{
++    notifier_list_notify(&colo_compare_notifiers,
++                migrate_get_current());
++}
++
+ static void colo_compare_tcp(CompareState *s, Connection *conn)
  {
-     if (trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
+     Packet *ppkt = NULL, *spkt = NULL;
-         char pri_ip_src[20], pri_ip_dst[20], sec_ip_src[20], sec_ip_dst[20];
+@@ -XXX,XX +XXX,XX @@ sec:
-@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_common(Packet *ppkt,
+         qemu_hexdump((char *)spkt->data, stderr,
-                                    sec_ip_src, sec_ip_dst);
+                      "colo-compare spkt", spkt->size);
 -        /*
 -         * colo_compare_inconsistent_notify();
 -         * TODO: notice to checkpoint();
 -         */
 +        colo_compare_inconsistency_notify();
      }
--    poffset = ppkt->vnet_hdr_len + poffset;
--    soffset = ppkt->vnet_hdr_len + soffset;
--
--    if (ppkt->size - poffset == spkt->size - soffset) {
--        return memcmp(ppkt->data + poffset,
--                      spkt->data + soffset,
--                      spkt->size - soffset);
--    } else {
--        trace_colo_compare_main("Net packet size are not the same");
--        return -1;
--    }
-+    return memcmp(ppkt->data + poffset, spkt->data + soffset, len);
  }
- /*
+@@ -XXX,XX +XXX,XX @@ static int colo_old_packet_check_one(Packet *pkt, int64_t *check_time)
-@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_tcp(Packet *spkt, Packet *ppkt)
+     }
-      * the secondary guest's timestamp. COLO just focus on payload,
+ }
-      * so we just need skip this field.
-      */
++void colo_compare_register_notifier(Notifier *notify)
--    if (ptcp->th_off > 5) {
++{
--        ptrdiff_t ptcp_offset, stcp_offset;
++    notifier_list_add(&colo_compare_notifiers, notify);
++}
--        ptcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
++
--                      + (ptcp->th_off * 4) - ppkt->vnet_hdr_len;
++void colo_compare_unregister_notifier(Notifier *notify)
--        stcp_offset = spkt->transport_header - (uint8_t *)spkt->data
++{
--                      + (stcp->th_off * 4) - spkt->vnet_hdr_len;
++    notifier_remove(notify);
-+    ptrdiff_t ptcp_offset, stcp_offset;
++}
++
  static int colo_old_packet_check_one_conn(Connection *conn,
 -                                          void *user_data)
 +                                           void *user_data)
  {
      GList *result = NULL;
      int64_t check_time = REGULAR_PACKET_CHECK_MS;
@@ -XXX,XX +XXX,XX @@ static int colo_old_packet_check_one_conn(Connection *conn,
      if (result) {
          /* Do checkpoint will flush old packet */
 -        /*
--         * When network is busy, some tcp options(like sack) will unpredictable
+-         * TODO: Notify colo frame to do checkpoint.
--         * occur in primary side or secondary side. it will make packet size
+-         * colo_compare_inconsistent_notify();
 -         * not same, but the two packet's payload is identical. colo just
 -         * care about packet payload, so we skip the option field.
 -         */
--        res = colo_packet_compare_common(ppkt, spkt, ptcp_offset, stcp_offset);
++        colo_compare_inconsistency_notify();
--    } else if (ptcp->th_sum == stcp->th_sum) {
+         return 0;
 -        res = colo_packet_compare_common(ppkt, spkt, ETH_HLEN, ETH_HLEN);
 +    ptcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
 +                  + (ptcp->th_off << 2) - ppkt->vnet_hdr_len;
 +    stcp_offset = spkt->transport_header - (uint8_t *)spkt->data
 +                  + (stcp->th_off << 2) - spkt->vnet_hdr_len;
 +    if (ppkt->size - ptcp_offset == spkt->size - stcp_offset) {
 +        res = colo_compare_packet_payload(ppkt, spkt,
 +                                          ptcp_offset, stcp_offset,
 +                                          ppkt->size - ptcp_offset);
      } else {
 +        trace_colo_compare_main("TCP: payload size of packets are different");
          res = -1;
      }
-@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_tcp(Packet *spkt, Packet *ppkt)
+@@ -XXX,XX +XXX,XX @@ static void colo_compare_packet(CompareState *s, Connection *conn,
-  */
+             /*
- static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
+              * If one packet arrive late, the secondary_list or
- {
+              * primary_list will be empty, so we can't compare it
--    int ret;
+-             * until next comparison.
--    int network_header_length = ppkt->ip->ip_hl * 4;
++             * until next comparison. If the packets in the list are
-+    uint16_t network_header_length = ppkt->ip->ip_hl << 2;
++             * timeout, it will trigger a checkpoint request.
-+    uint16_t offset = network_header_length + ETH_HLEN + ppkt->vnet_hdr_len;
+              */
+             trace_colo_compare_main("packet different");
-     trace_colo_compare_main("compare udp");
+             g_queue_push_head(&conn->primary_list, pkt);
+-            /* TODO: colo_notify_checkpoint();*/
-@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
++            colo_compare_inconsistency_notify();
-      * other field like TOS,TTL,IP Checksum. we only need to compare
+             break;
       * the ip payload here.
       */
 -    ret = colo_packet_compare_common(ppkt, spkt,
 -                                     network_header_length + ETH_HLEN,
 -                                     network_header_length + ETH_HLEN);
 -
 -    if (ret) {
 +    if (ppkt->size != spkt->size) {
 +        trace_colo_compare_main("UDP: payload size of packets are different");
 +        return -1;
 +    }
 +    if (colo_compare_packet_payload(ppkt, spkt, offset, offset,
 +                                    ppkt->size - offset)) {
          trace_colo_compare_udp_miscompare("primary pkt size", ppkt->size);
          trace_colo_compare_udp_miscompare("Secondary pkt size", spkt->size);
          if (trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
              qemu_hexdump((char *)spkt->data, stderr, "colo-compare sec pkt",
                           spkt->size);
          }
-+        return -1;
-+    } else {
-+        return 0;
      }
--
+diff --git a/net/colo-compare.h b/net/colo-compare.h
--    return ret;
+index XXXXXXX..XXXXXXX 100644
- }
+--- a/net/colo-compare.h
++++ b/net/colo-compare.h
- /*
+@@ -XXX,XX +XXX,XX @@
-@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
+ #define QEMU_COLO_COMPARE_H
-  */
- static int colo_packet_compare_icmp(Packet *spkt, Packet *ppkt)
+ void colo_notify_compares_event(void *opaque, int event, Error **errp);
- {
++void colo_compare_register_notifier(Notifier *notify);
--    int network_header_length = ppkt->ip->ip_hl * 4;
++void colo_compare_unregister_notifier(Notifier *notify);
-+    uint16_t network_header_length = ppkt->ip->ip_hl << 2;
-+    uint16_t offset = network_header_length + ETH_HLEN + ppkt->vnet_hdr_len;
+ #endif /* QEMU_COLO_COMPARE_H */
      trace_colo_compare_main("compare icmp");
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_icmp(Packet *spkt, Packet *ppkt)
       * other field like TOS,TTL,IP Checksum. we only need to compare
       * the ip payload here.
       */
 -    if (colo_packet_compare_common(ppkt, spkt,
 -                                   network_header_length + ETH_HLEN,
 -                                   network_header_length + ETH_HLEN)) {
 +    if (ppkt->size != spkt->size) {
 +        trace_colo_compare_main("ICMP: payload size of packets are different");
 +        return -1;
 +    }
 +    if (colo_compare_packet_payload(ppkt, spkt, offset, offset,
 +                                    ppkt->size - offset)) {
          trace_colo_compare_icmp_miscompare("primary pkt size",
                                             ppkt->size);
          trace_colo_compare_icmp_miscompare("Secondary pkt size",
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_icmp(Packet *spkt, Packet *ppkt)
   */
  static int colo_packet_compare_other(Packet *spkt, Packet *ppkt)
  {
 +    uint16_t offset = ppkt->vnet_hdr_len;
 +
      trace_colo_compare_main("compare other");
      if (trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
          char pri_ip_src[20], pri_ip_dst[20], sec_ip_src[20], sec_ip_dst[20];
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_other(Packet *spkt, Packet *ppkt)
                                     sec_ip_src, sec_ip_dst);
      }
 -    return colo_packet_compare_common(ppkt, spkt, 0, 0);
 +    if (ppkt->size != spkt->size) {
 +        trace_colo_compare_main("Other: payload size of packets are different");
 +        return -1;
 +    }
 +    return colo_compare_packet_payload(ppkt, spkt, offset, offset,
 +                                       ppkt->size - offset);
  }
  static int colo_old_packet_check_one(Packet *pkt, int64_t *check_time)
 --
-.7.4
+.5.0

-New patch
+[Qemu-devel] [PULL V2 04/26] COLO: integrate colo compare with colo frame
+From: Zhang Chen <zhangckid@gmail.com>
+For COLO FT, both the PVM and SVM run at the same time,
+only sync the state while it needs.
+So here, let SVM runs while not doing checkpoint, change
+DEFAULT_MIGRATE_X_CHECKPOINT_DELAY to 200*100.
+Besides, we forgot to release colo_checkpoint_semd and
+colo_delay_timer, fix them here.
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/colo.c      | 42 ++++++++++++++++++++++++++++++++++++++++--
+ migration/migration.c |  6 ++----
+files changed, 42 insertions(+), 6 deletions(-)
+diff --git a/migration/colo.c b/migration/colo.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo.c
++++ b/migration/colo.c
+@@ -XXX,XX +XXX,XX @@
+ #include "qemu/error-report.h"
+ #include "migration/failover.h"
+ #include "replication.h"
++#include "net/colo-compare.h"
++#include "net/colo.h"
+ static bool vmstate_loading;
++static Notifier packets_compare_notifier;
+ #define COLO_BUFFER_BASE_SIZE (4 * 1024 * 1024)
+@@ -XXX,XX +XXX,XX @@ static int colo_do_checkpoint_transaction(MigrationState *s,
+         goto out;
+     }
++    colo_notify_compares_event(NULL, COLO_EVENT_CHECKPOINT, &local_err);
++    if (local_err) {
++        goto out;
++    }
++
+     /* Disable block migration */
+     migrate_set_block_enabled(false, &local_err);
+     qemu_savevm_state_header(fb);
+@@ -XXX,XX +XXX,XX @@ out:
+     return ret;
+ }
++static void colo_compare_notify_checkpoint(Notifier *notifier, void *data)
++{
++    colo_checkpoint_notify(data);
++}
++
+ static void colo_process_checkpoint(MigrationState *s)
+ {
+     QIOChannelBuffer *bioc;
+@@ -XXX,XX +XXX,XX @@ static void colo_process_checkpoint(MigrationState *s)
+         goto out;
+     }
++    packets_compare_notifier.notify = colo_compare_notify_checkpoint;
++    colo_compare_register_notifier(&packets_compare_notifier);
++
+     /*
+      * Wait for Secondary finish loading VM states and enter COLO
+      * restore.
+@@ -XXX,XX +XXX,XX @@ out:
+         qemu_fclose(fb);
+     }
+-    timer_del(s->colo_delay_timer);
+-
+     /* Hope this not to be too long to wait here */
+     qemu_sem_wait(&s->colo_exit_sem);
+     qemu_sem_destroy(&s->colo_exit_sem);
++
++    /*
++     * It is safe to unregister notifier after failover finished.
++     * Besides, colo_delay_timer and colo_checkpoint_sem can't be
++     * released befor unregister notifier, or there will be use-after-free
++     * error.
++     */
++    colo_compare_unregister_notifier(&packets_compare_notifier);
++    timer_del(s->colo_delay_timer);
++    timer_free(s->colo_delay_timer);
++    qemu_sem_destroy(&s->colo_checkpoint_sem);
++
+     /*
+      * Must be called after failover BH is completed,
+      * Or the failover BH may shutdown the wrong fd that
+@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
+     fb = qemu_fopen_channel_input(QIO_CHANNEL(bioc));
+     object_unref(OBJECT(bioc));
++    qemu_mutex_lock_iothread();
++    vm_start();
++    trace_colo_vm_state_change("stop", "run");
++    qemu_mutex_unlock_iothread();
++
+     colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_READY,
+                       &local_err);
+     if (local_err) {
+@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
+             goto out;
+         }
++        qemu_mutex_lock_iothread();
++        vm_stop_force_state(RUN_STATE_COLO);
++        trace_colo_vm_state_change("run", "stop");
++        qemu_mutex_unlock_iothread();
++
+         /* FIXME: This is unnecessary for periodic checkpoint mode */
+         colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_REPLY,
+                      &local_err);
+@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
+         }
+         vmstate_loading = false;
++        vm_start();
++        trace_colo_vm_state_change("stop", "run");
+         qemu_mutex_unlock_iothread();
+         if (failover_get_state() == FAILOVER_STATUS_RELAUNCH) {
+diff --git a/migration/migration.c b/migration/migration.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/migration.c
++++ b/migration/migration.c
+@@ -XXX,XX +XXX,XX @@
+ /* Migration XBZRLE default cache size */
+ #define DEFAULT_MIGRATE_XBZRLE_CACHE_SIZE (64 * 1024 * 1024)
+-/* The delay time (in ms) between two COLO checkpoints
+- * Note: Please change this default value to 10000 when we support hybrid mode.
+- */
+-#define DEFAULT_MIGRATE_X_CHECKPOINT_DELAY 200
++/* The delay time (in ms) between two COLO checkpoints */
++#define DEFAULT_MIGRATE_X_CHECKPOINT_DELAY (200 * 100)
+ #define DEFAULT_MIGRATE_MULTIFD_CHANNELS 2
+ #define DEFAULT_MIGRATE_MULTIFD_PAGE_COUNT 16
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 05/26] COLO: Add block replication into colo process
+From: Zhang Chen <zhangckid@gmail.com>
+Make sure master start block replication after slave's block
+replication started.
+Besides, we need to activate VM's blocks before goes into
+COLO state.
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/colo.c      | 43 +++++++++++++++++++++++++++++++++++++++++++
+ migration/migration.c | 10 ++++++++++
+files changed, 53 insertions(+)
+diff --git a/migration/colo.c b/migration/colo.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo.c
++++ b/migration/colo.c
+@@ -XXX,XX +XXX,XX @@
+ #include "replication.h"
+ #include "net/colo-compare.h"
+ #include "net/colo.h"
++#include "block/block.h"
+ static bool vmstate_loading;
+ static Notifier packets_compare_notifier;
+@@ -XXX,XX +XXX,XX @@ static void secondary_vm_do_failover(void)
+ {
+     int old_state;
+     MigrationIncomingState *mis = migration_incoming_get_current();
++    Error *local_err = NULL;
+     /* Can not do failover during the process of VM's loading VMstate, Or
+      * it will break the secondary VM.
+@@ -XXX,XX +XXX,XX @@ static void secondary_vm_do_failover(void)
+     migrate_set_state(&mis->state, MIGRATION_STATUS_COLO,
+                       MIGRATION_STATUS_COMPLETED);
++    replication_stop_all(true, &local_err);
++    if (local_err) {
++        error_report_err(local_err);
++    }
++
+     if (!autostart) {
+         error_report("\"-S\" qemu option will be ignored in secondary side");
+         /* recover runstate to normal migration finish state */
+@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
+ {
+     MigrationState *s = migrate_get_current();
+     int old_state;
++    Error *local_err = NULL;
+     migrate_set_state(&s->state, MIGRATION_STATUS_COLO,
+                       MIGRATION_STATUS_COMPLETED);
+@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
+                      FailoverStatus_str(old_state));
+         return;
+     }
++
++    replication_stop_all(true, &local_err);
++    if (local_err) {
++        error_report_err(local_err);
++        local_err = NULL;
++    }
++
+     /* Notify COLO thread that failover work is finished */
+     qemu_sem_post(&s->colo_exit_sem);
+ }
+@@ -XXX,XX +XXX,XX @@ static int colo_do_checkpoint_transaction(MigrationState *s,
+     qemu_savevm_state_header(fb);
+     qemu_savevm_state_setup(fb);
+     qemu_mutex_lock_iothread();
++    replication_do_checkpoint_all(&local_err);
++    if (local_err) {
++        qemu_mutex_unlock_iothread();
++        goto out;
++    }
+     qemu_savevm_state_complete_precopy(fb, false, false);
+     qemu_mutex_unlock_iothread();
+@@ -XXX,XX +XXX,XX @@ static void colo_process_checkpoint(MigrationState *s)
+     object_unref(OBJECT(bioc));
+     qemu_mutex_lock_iothread();
++    replication_start_all(REPLICATION_MODE_PRIMARY, &local_err);
++    if (local_err) {
++        qemu_mutex_unlock_iothread();
++        goto out;
++    }
++
+     vm_start();
+     qemu_mutex_unlock_iothread();
+     trace_colo_vm_state_change("stop", "run");
+@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
+     object_unref(OBJECT(bioc));
+     qemu_mutex_lock_iothread();
++    replication_start_all(REPLICATION_MODE_SECONDARY, &local_err);
++    if (local_err) {
++        qemu_mutex_unlock_iothread();
++        goto out;
++    }
+     vm_start();
+     trace_colo_vm_state_change("stop", "run");
+     qemu_mutex_unlock_iothread();
+@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
+             goto out;
+         }
++        replication_get_error_all(&local_err);
++        if (local_err) {
++            qemu_mutex_unlock_iothread();
++            goto out;
++        }
++        /* discard colo disk buffer */
++        replication_do_checkpoint_all(&local_err);
++        if (local_err) {
++            qemu_mutex_unlock_iothread();
++            goto out;
++        }
++
+         vmstate_loading = false;
+         vm_start();
+         trace_colo_vm_state_change("stop", "run");
+diff --git a/migration/migration.c b/migration/migration.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/migration.c
++++ b/migration/migration.c
+@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
+     MigrationIncomingState *mis = migration_incoming_get_current();
+     PostcopyState ps;
+     int ret;
++    Error *local_err = NULL;
+     assert(mis->from_src_file);
+     mis->migration_incoming_co = qemu_coroutine_self();
+@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
+     /* we get COLO info, and know if we are in COLO mode */
+     if (!ret && migration_incoming_enable_colo()) {
++        /* Make sure all file formats flush their mutable metadata */
++        bdrv_invalidate_cache_all(&local_err);
++        if (local_err) {
++            migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
++                    MIGRATION_STATUS_FAILED);
++            error_report_err(local_err);
++            exit(EXIT_FAILURE);
++        }
++
+         qemu_thread_create(&mis->colo_incoming_thread, "COLO incoming",
+              colo_process_incoming_thread, mis, QEMU_THREAD_JOINABLE);
+         mis->have_colo_incoming_thread = true;
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 06/26] COLO: Remove colo_state migration struct
+From: Zhang Chen <zhangckid@gmail.com>
 We need to know if migration is going into COLO state for
 incoming side before start normal migration.
 Instead by using the VMStateDescription to send colo_state
 from source side to destination side, we use MIG_CMD_ENABLE_COLO
 to indicate whether COLO is enabled or not.
 Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
 Signed-off-by: Zhang Chen <zhangckid@gmail.com>
 Signed-off-by: Zhang Chen <chen.zhang@intel.com>
 Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
  include/migration/colo.h |  5 ++--
  migration/Makefile.objs  |  2 +-
  migration/colo-comm.c    | 76 ------------------------------------------------
  migration/colo.c         | 13 ++++++++-
  migration/migration.c    | 23 ++++++++++++++-
  migration/savevm.c       | 17 +++++++++++
  migration/savevm.h       |  1 +
  migration/trace-events   |  1 +
  vl.c                     |  2 --
 files changed, 57 insertions(+), 83 deletions(-)
  delete mode 100644 migration/colo-comm.c
 diff --git a/include/migration/colo.h b/include/migration/colo.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/migration/colo.h
 +++ b/include/migration/colo.h
@@ -XXX,XX +XXX,XX @@ void migrate_start_colo_process(MigrationState *s);
  bool migration_in_colo_state(void);
  /* loadvm */
 -bool migration_incoming_enable_colo(void);
 -void migration_incoming_exit_colo(void);
 +void migration_incoming_enable_colo(void);
 +void migration_incoming_disable_colo(void);
 +bool migration_incoming_colo_enabled(void);
  void *colo_process_incoming_thread(void *opaque);
  bool migration_incoming_in_colo_state(void);
 diff --git a/migration/Makefile.objs b/migration/Makefile.objs
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/Makefile.objs
 +++ b/migration/Makefile.objs
@@ -XXX,XX +XXX,XX @@
  common-obj-y += migration.o socket.o fd.o exec.o
  common-obj-y += tls.o channel.o savevm.o
 -common-obj-y += colo-comm.o colo.o colo-failover.o
 +common-obj-y += colo.o colo-failover.o
  common-obj-y += vmstate.o vmstate-types.o page_cache.o
  common-obj-y += qemu-file.o global_state.o
  common-obj-y += qemu-file-channel.o
 diff --git a/migration/colo-comm.c b/migration/colo-comm.c
 deleted file mode 100644
 index XXXXXXX..XXXXXXX
 --- a/migration/colo-comm.c
 +++ /dev/null
@@ -XXX,XX +XXX,XX @@
 -/*
 - * COarse-grain LOck-stepping Virtual Machines for Non-stop Service (COLO)
 - * (a.k.a. Fault Tolerance or Continuous Replication)
 - *
 - * Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
 - * Copyright (c) 2016 FUJITSU LIMITED
 - * Copyright (c) 2016 Intel Corporation
 - *
 - * This work is licensed under the terms of the GNU GPL, version 2 or
 - * later. See the COPYING file in the top-level directory.
 - *
 - */
 -
 -#include "qemu/osdep.h"
 -#include "migration.h"
 -#include "migration/colo.h"
 -#include "migration/vmstate.h"
 -#include "trace.h"
 -
 -typedef struct {
 -     bool colo_requested;
 -} COLOInfo;
 -
 -static COLOInfo colo_info;
 -
 -COLOMode get_colo_mode(void)
 -{
 -    if (migration_in_colo_state()) {
 -        return COLO_MODE_PRIMARY;
 -    } else if (migration_incoming_in_colo_state()) {
 -        return COLO_MODE_SECONDARY;
 -    } else {
 -        return COLO_MODE_UNKNOWN;
 -    }
 -}
 -
 -static int colo_info_pre_save(void *opaque)
 -{
 -    COLOInfo *s = opaque;
 -
 -    s->colo_requested = migrate_colo_enabled();
 -
 -    return 0;
 -}
 -
 -static bool colo_info_need(void *opaque)
 -{
 -   return migrate_colo_enabled();
 -}
 -
 -static const VMStateDescription colo_state = {
 -    .name = "COLOState",
 -    .version_id = 1,
 -    .minimum_version_id = 1,
 -    .pre_save = colo_info_pre_save,
 -    .needed = colo_info_need,
 -    .fields = (VMStateField[]) {
 -        VMSTATE_BOOL(colo_requested, COLOInfo),
 -        VMSTATE_END_OF_LIST()
 -    },
 -};
 -
 -void colo_info_init(void)
 -{
 -    vmstate_register(NULL, 0, &colo_state, &colo_info);
 -}
 -
 -bool migration_incoming_enable_colo(void)
 -{
 -    return colo_info.colo_requested;
 -}
 -
 -void migration_incoming_exit_colo(void)
 -{
 -    colo_info.colo_requested = false;
 -}
 diff --git a/migration/colo.c b/migration/colo.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/colo.c
 +++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
      qemu_sem_post(&s->colo_exit_sem);
  }
 +COLOMode get_colo_mode(void)
 +{
 +    if (migration_in_colo_state()) {
 +        return COLO_MODE_PRIMARY;
 +    } else if (migration_incoming_in_colo_state()) {
 +        return COLO_MODE_SECONDARY;
 +    } else {
 +        return COLO_MODE_UNKNOWN;
 +    }
 +}
 +
  void colo_do_failover(MigrationState *s)
  {
      /* Make sure VM stopped while failover happened. */
@@ -XXX,XX +XXX,XX @@ out:
      if (mis->to_src_file) {
          qemu_fclose(mis->to_src_file);
      }
 -    migration_incoming_exit_colo();
 +    migration_incoming_disable_colo();
      rcu_unregister_thread();
      return NULL;
 diff --git a/migration/migration.c b/migration/migration.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/migration.c
 +++ b/migration/migration.c
@@ -XXX,XX +XXX,XX @@ int migrate_send_rp_req_pages(MigrationIncomingState *mis, const char *rbname,
      return migrate_send_rp_message(mis, msg_type, msglen, bufc);
  }
 +static bool migration_colo_enabled;
 +bool migration_incoming_colo_enabled(void)
 +{
 +    return migration_colo_enabled;
 +}
 +
 +void migration_incoming_disable_colo(void)
 +{
 +    migration_colo_enabled = false;
 +}
 +
 +void migration_incoming_enable_colo(void)
 +{
 +    migration_colo_enabled = true;
 +}
 +
  void qemu_start_incoming_migration(const char *uri, Error **errp)
  {
      const char *p;
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
      }
      /* we get COLO info, and know if we are in COLO mode */
 -    if (!ret && migration_incoming_enable_colo()) {
 +    if (!ret && migration_incoming_colo_enabled()) {
          /* Make sure all file formats flush their mutable metadata */
          bdrv_invalidate_cache_all(&local_err);
          if (local_err) {
@@ -XXX,XX +XXX,XX @@ static void *migration_thread(void *opaque)
          qemu_savevm_send_postcopy_advise(s->to_dst_file);
      }
 +    if (migrate_colo_enabled()) {
 +        /* Notify migration destination that we enable COLO */
 +        qemu_savevm_send_colo_enable(s->to_dst_file);
 +    }
 +
      qemu_savevm_state_setup(s->to_dst_file);
      s->setup_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) - setup_start;
 diff --git a/migration/savevm.c b/migration/savevm.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/savevm.c
 +++ b/migration/savevm.c
@@ -XXX,XX +XXX,XX @@
  #include "io/channel-file.h"
  #include "sysemu/replay.h"
  #include "qjson.h"
 +#include "migration/colo.h"
  #ifndef ETH_P_RARP
  #define ETH_P_RARP 0x8035
@@ -XXX,XX +XXX,XX @@ enum qemu_vm_cmd {
                                        were previously sent during
                                        precopy but are dirty. */
      MIG_CMD_PACKAGED,          /* Send a wrapped stream within this stream */
 +    MIG_CMD_ENABLE_COLO,       /* Enable COLO */
      MIG_CMD_POSTCOPY_RESUME,   /* resume postcopy on dest */
      MIG_CMD_RECV_BITMAP,       /* Request for recved bitmap on dst */
      MIG_CMD_MAX
@@ -XXX,XX +XXX,XX @@ static void qemu_savevm_command_send(QEMUFile *f,
      qemu_fflush(f);
  }
 +void qemu_savevm_send_colo_enable(QEMUFile *f)
 +{
 +    trace_savevm_send_colo_enable();
 +    qemu_savevm_command_send(f, MIG_CMD_ENABLE_COLO, 0, NULL);
 +}
 +
  void qemu_savevm_send_ping(QEMUFile *f, uint32_t value)
  {
      uint32_t buf;
@@ -XXX,XX +XXX,XX @@ static int loadvm_handle_recv_bitmap(MigrationIncomingState *mis,
      return 0;
  }
 +static int loadvm_process_enable_colo(MigrationIncomingState *mis)
 +{
 +    migration_incoming_enable_colo();
 +    return 0;
 +}
 +
  /*
   * Process an incoming 'QEMU_VM_COMMAND'
   * 0           just a normal return
@@ -XXX,XX +XXX,XX @@ static int loadvm_process_command(QEMUFile *f)
      case MIG_CMD_RECV_BITMAP:
          return loadvm_handle_recv_bitmap(mis, len);
 +
 +    case MIG_CMD_ENABLE_COLO:
 +        return loadvm_process_enable_colo(mis);
      }
      return 0;
 diff --git a/migration/savevm.h b/migration/savevm.h
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/savevm.h
 +++ b/migration/savevm.h
@@ -XXX,XX +XXX,XX @@ void qemu_savevm_send_postcopy_ram_discard(QEMUFile *f, const char *name,
                                             uint16_t len,
                                             uint64_t *start_list,
                                             uint64_t *length_list);
 +void qemu_savevm_send_colo_enable(QEMUFile *f);
  int qemu_loadvm_state(QEMUFile *f);
  void qemu_loadvm_state_cleanup(void);
 diff --git a/migration/trace-events b/migration/trace-events
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/trace-events
 +++ b/migration/trace-events
@@ -XXX,XX +XXX,XX @@ savevm_send_ping(uint32_t val) "0x%x"
  savevm_send_postcopy_listen(void) ""
  savevm_send_postcopy_run(void) ""
  savevm_send_postcopy_resume(void) ""
 +savevm_send_colo_enable(void) ""
  savevm_send_recv_bitmap(char *name) "%s"
  savevm_state_setup(void) ""
  savevm_state_resume_prepare(void) ""
 diff --git a/vl.c b/vl.c
 index XXXXXXX..XXXXXXX 100644
 --- a/vl.c
 +++ b/vl.c
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv, char **envp)
  #endif
      }
 -    colo_info_init();
 -
      if (net_init_clients(&err) < 0) {
          error_report_err(err);
          exit(1);
 --
 .5.0

-New patch
+[Qemu-devel] [PULL V2 07/26] COLO: Load dirty pages into SVM's RAM cache firstly
+From: Zhang Chen <zhangckid@gmail.com>
 We should not load PVM's state directly into SVM, because there maybe some
 errors happen when SVM is receving data, which will break SVM.
 We need to ensure receving all data before load the state into SVM. We use
 an extra memory to cache these data (PVM's ram). The ram cache in secondary side
 is initially the same as SVM/PVM's memory. And in the process of checkpoint,
 we cache the dirty pages of PVM into this ram cache firstly, so this ram cache
 always the same as PVM's memory at every checkpoint, then we flush this cached ram
 to SVM after we receive all PVM's state.
 Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
 Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
 Signed-off-by: Zhang Chen <zhangckid@gmail.com>
 Signed-off-by: Zhang Chen <chen.zhang@intel.com>
 Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
  include/exec/ram_addr.h |  1 +
  migration/migration.c   |  7 +++++
  migration/ram.c         | 83 +++++++++++++++++++++++++++++++++++++++++++++++--
  migration/ram.h         |  4 +++
  migration/savevm.c      |  2 +-
 files changed, 94 insertions(+), 3 deletions(-)
 diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
 index XXXXXXX..XXXXXXX 100644
 --- a/include/exec/ram_addr.h
 +++ b/include/exec/ram_addr.h
@@ -XXX,XX +XXX,XX @@ struct RAMBlock {
      struct rcu_head rcu;
      struct MemoryRegion *mr;
      uint8_t *host;
 +    uint8_t *colo_cache; /* For colo, VM's ram cache */
      ram_addr_t offset;
      ram_addr_t used_length;
      ram_addr_t max_length;
 diff --git a/migration/migration.c b/migration/migration.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/migration.c
 +++ b/migration/migration.c
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
              exit(EXIT_FAILURE);
          }
 +        if (colo_init_ram_cache() < 0) {
 +            error_report("Init ram cache failed");
 +            exit(EXIT_FAILURE);
 +        }
 +
          qemu_thread_create(&mis->colo_incoming_thread, "COLO incoming",
               colo_process_incoming_thread, mis, QEMU_THREAD_JOINABLE);
          mis->have_colo_incoming_thread = true;
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
          /* Wait checkpoint incoming thread exit before free resource */
          qemu_thread_join(&mis->colo_incoming_thread);
 +        /* We hold the global iothread lock, so it is safe here */
 +        colo_release_ram_cache();
      }
      if (ret < 0) {
 diff --git a/migration/ram.c b/migration/ram.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/ram.c
 +++ b/migration/ram.c
@@ -XXX,XX +XXX,XX @@ static inline void *host_from_ram_block_offset(RAMBlock *block,
      return block->host + offset;
  }
 +static inline void *colo_cache_from_block_offset(RAMBlock *block,
 +                                                 ram_addr_t offset)
 +{
 +    if (!offset_in_ramblock(block, offset)) {
 +        return NULL;
 +    }
 +    if (!block->colo_cache) {
 +        error_report("%s: colo_cache is NULL in block :%s",
 +                     __func__, block->idstr);
 +        return NULL;
 +    }
 +    return block->colo_cache + offset;
 +}
 +
  /**
   * ram_handle_compressed: handle the zero page case
   *
@@ -XXX,XX +XXX,XX @@ static void decompress_data_with_multi_threads(QEMUFile *f,
      qemu_mutex_unlock(&decomp_done_lock);
  }
 +/*
 + * colo cache: this is for secondary VM, we cache the whole
 + * memory of the secondary VM, it is need to hold the global lock
 + * to call this helper.
 + */
 +int colo_init_ram_cache(void)
 +{
 +    RAMBlock *block;
 +
 +    rcu_read_lock();
 +    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
 +        block->colo_cache = qemu_anon_ram_alloc(block->used_length,
 +                                                NULL,
 +                                                false);
 +        if (!block->colo_cache) {
 +            error_report("%s: Can't alloc memory for COLO cache of block %s,"
 +                         "size 0x" RAM_ADDR_FMT, __func__, block->idstr,
 +                         block->used_length);
 +            goto out_locked;
 +        }
 +        memcpy(block->colo_cache, block->host, block->used_length);
 +    }
 +    rcu_read_unlock();
 +    return 0;
 +
 +out_locked:
 +    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
 +        if (block->colo_cache) {
 +            qemu_anon_ram_free(block->colo_cache, block->used_length);
 +            block->colo_cache = NULL;
 +        }
 +    }
 +
 +    rcu_read_unlock();
 +    return -errno;
 +}
 +
 +/* It is need to hold the global lock to call this helper */
 +void colo_release_ram_cache(void)
 +{
 +    RAMBlock *block;
 +
 +    rcu_read_lock();
 +    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
 +        if (block->colo_cache) {
 +            qemu_anon_ram_free(block->colo_cache, block->used_length);
 +            block->colo_cache = NULL;
 +        }
 +    }
 +    rcu_read_unlock();
 +}
 +
  /**
   * ram_load_setup: Setup RAM for migration incoming side
   *
@@ -XXX,XX +XXX,XX @@ static int ram_load_setup(QEMUFile *f, void *opaque)
      xbzrle_load_setup();
      ramblock_recv_map_init();
 +
      return 0;
  }
@@ -XXX,XX +XXX,XX @@ static int ram_load_cleanup(void *opaque)
          g_free(rb->receivedmap);
          rb->receivedmap = NULL;
      }
 +
      return 0;
  }
@@ -XXX,XX +XXX,XX @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
                       RAM_SAVE_FLAG_COMPRESS_PAGE | RAM_SAVE_FLAG_XBZRLE)) {
              RAMBlock *block = ram_block_from_stream(f, flags);
 -            host = host_from_ram_block_offset(block, addr);
 +            /*
 +             * After going into COLO, we should load the Page into colo_cache.
 +             */
 +            if (migration_incoming_in_colo_state()) {
 +                host = colo_cache_from_block_offset(block, addr);
 +            } else {
 +                host = host_from_ram_block_offset(block, addr);
 +            }
              if (!host) {
                  error_report("Illegal RAM offset " RAM_ADDR_FMT, addr);
                  ret = -EINVAL;
                  break;
              }
 -            ramblock_recv_bitmap_set(block, host);
 +
 +            if (!migration_incoming_in_colo_state()) {
 +                ramblock_recv_bitmap_set(block, host);
 +            }
 +
              trace_ram_load_loop(block->idstr, (uint64_t)addr, flags, host);
          }
 diff --git a/migration/ram.h b/migration/ram.h
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/ram.h
 +++ b/migration/ram.h
@@ -XXX,XX +XXX,XX @@ int64_t ramblock_recv_bitmap_send(QEMUFile *file,
                                    const char *block_name);
  int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *rb);
 +/* ram cache */
 +int colo_init_ram_cache(void);
 +void colo_release_ram_cache(void);
 +
  #endif
 diff --git a/migration/savevm.c b/migration/savevm.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/savevm.c
 +++ b/migration/savevm.c
@@ -XXX,XX +XXX,XX @@ static int loadvm_handle_recv_bitmap(MigrationIncomingState *mis,
  static int loadvm_process_enable_colo(MigrationIncomingState *mis)
  {
      migration_incoming_enable_colo();
 -    return 0;
 +    return colo_init_ram_cache();
  }
  /*
 --
 .5.0

-[Qemu-devel] [PULL 4/6] net: Allow netdevs to be used with 'hostfwd_add' and 'hostfwd_remove'
+[Qemu-devel] [PULL V2 08/26] ram/COLO: Record the dirty pages that SVM received
-From: Thomas Huth <thuth@redhat.com>
+From: Zhang Chen <zhangckid@gmail.com>
-It does not make much sense to limit these commands to the legacy 'vlan'
+We record the address of the dirty pages that received,
-concept only, they should work with the modern netdevs, too. So now
+it will help flushing pages that cached into SVM.
 it is possible to use this command with one, two or three parameters.
-With one parameter, the command installs a hostfwd rule on the default
+Here, it is a trick, we record dirty pages by re-using migration
-"user" network:
+dirty bitmap. In the later patch, we will start the dirty log
-    hostfwd_add tcp:...
+for SVM, just like migration, in this way, we can record both
 the dirty pages caused by PVM and SVM, we only flush those dirty
 pages from RAM cache while do checkpoint.
-With two parameters, the command installs a hostfwd rule on a netdev
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
-(that's the new way of using this command):
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
-    hostfwd_add netdev_id tcp:...
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
 With three parameters, the command installs a rule on a 'vlan' (aka hub):
     hostfwd_add hub_id name tcp:...
 Same applies to the hostfwd_remove command now.
 Signed-off-by: Thomas Huth <thuth@redhat.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
- hmp-commands.hx |  4 ++--
+ migration/ram.c | 43 ++++++++++++++++++++++++++++++++++++++++---
- net/slirp.c     | 33 +++++++++++++++++++++++----------
+file changed, 40 insertions(+), 3 deletions(-)
 files changed, 25 insertions(+), 12 deletions(-)
-diff --git a/hmp-commands.hx b/hmp-commands.hx
+diff --git a/migration/ram.c b/migration/ram.c
 index XXXXXXX..XXXXXXX 100644
---- a/hmp-commands.hx
+--- a/migration/ram.c
-+++ b/hmp-commands.hx
++++ b/migration/ram.c
-@@ -XXX,XX +XXX,XX @@ ETEXI
+@@ -XXX,XX +XXX,XX @@ static inline void *colo_cache_from_block_offset(RAMBlock *block,
-     {
+                      __func__, block->idstr);
-         .name       = "hostfwd_add",
+         return NULL;
-         .args_type  = "arg1:s,arg2:s?,arg3:s?",
+     }
--        .params     = "[vlan_id name] [tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport",
++
-+        .params     = "[hub_id name]|[netdev_id] [tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport",
++    /*
-         .help       = "redirect TCP or UDP connections from host to guest (requires -net user)",
++    * During colo checkpoint, we need bitmap of these migrated pages.
-         .cmd        = hmp_hostfwd_add,
++    * It help us to decide which pages in ram cache should be flushed
-     },
++    * into VM's RAM later.
-@@ -XXX,XX +XXX,XX @@ ETEXI
++    */
-     {
++    if (!test_and_set_bit(offset >> TARGET_PAGE_BITS, block->bmap)) {
-         .name       = "hostfwd_remove",
++        ram_state->migration_dirty_pages++;
-         .args_type  = "arg1:s,arg2:s?,arg3:s?",
++    }
--        .params     = "[vlan_id name] [tcp|udp]:[hostaddr]:hostport",
+     return block->colo_cache + offset;
 +        .params     = "[hub_id name]|[netdev_id] [tcp|udp]:[hostaddr]:hostport",
          .help       = "remove host-to-guest TCP or UDP redirection",
          .cmd        = hmp_hostfwd_remove,
      },
 diff --git a/net/slirp.c b/net/slirp.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/slirp.c
 +++ b/net/slirp.c
@@ -XXX,XX +XXX,XX @@ error:
      return -1;
  }
--static SlirpState *slirp_lookup(Monitor *mon, const char *vlan,
+@@ -XXX,XX +XXX,XX @@ int colo_init_ram_cache(void)
--                                const char *stack)
+     RAMBlock *block;
-+static SlirpState *slirp_lookup(Monitor *mon, const char *hub_id,
-+                                const char *name)
+     rcu_read_lock();
 -    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
 +    RAMBLOCK_FOREACH_MIGRATABLE(block) {
          block->colo_cache = qemu_anon_ram_alloc(block->used_length,
                                                  NULL,
                                                  false);
@@ -XXX,XX +XXX,XX @@ int colo_init_ram_cache(void)
          memcpy(block->colo_cache, block->host, block->used_length);
      }
      rcu_read_unlock();
 +    /*
 +    * Record the dirty pages that sent by PVM, we use this dirty bitmap together
 +    * with to decide which page in cache should be flushed into SVM's RAM. Here
 +    * we use the same name 'ram_bitmap' as for migration.
 +    */
 +    if (ram_bytes_total()) {
 +        RAMBlock *block;
 +
 +        RAMBLOCK_FOREACH_MIGRATABLE(block) {
 +            unsigned long pages = block->max_length >> TARGET_PAGE_BITS;
 +
 +            block->bmap = bitmap_new(pages);
 +            bitmap_set(block->bmap, 0, pages);
 +        }
 +    }
 +    ram_state = g_new0(RAMState, 1);
 +    ram_state->migration_dirty_pages = 0;
 +
      return 0;
  out_locked:
 -    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
 +
 +    RAMBLOCK_FOREACH_MIGRATABLE(block) {
          if (block->colo_cache) {
              qemu_anon_ram_free(block->colo_cache, block->used_length);
              block->colo_cache = NULL;
@@ -XXX,XX +XXX,XX @@ void colo_release_ram_cache(void)
  {
--
+     RAMBlock *block;
--    if (vlan) {
-+    if (name) {
++    RAMBLOCK_FOREACH_MIGRATABLE(block) {
-         NetClientState *nc;
++        g_free(block->bmap);
--        nc = net_hub_find_client_by_name(strtol(vlan, NULL, 0), stack);
++        block->bmap = NULL;
--        if (!nc) {
++    }
--            monitor_printf(mon, "unrecognized (vlan-id, stackname) pair\n");
++
--            return NULL;
+     rcu_read_lock();
-+        if (hub_id) {
+-    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
-+            nc = net_hub_find_client_by_name(strtol(hub_id, NULL, 0), name);
++
-+            if (!nc) {
++    RAMBLOCK_FOREACH_MIGRATABLE(block) {
-+                monitor_printf(mon, "unrecognized (vlan-id, stackname) pair\n");
+         if (block->colo_cache) {
-+                return NULL;
+             qemu_anon_ram_free(block->colo_cache, block->used_length);
-+            }
+             block->colo_cache = NULL;
 +        } else {
 +            nc = qemu_find_netdev(name);
 +            if (!nc) {
 +                monitor_printf(mon, "unrecognized netdev id '%s'\n", name);
 +                return NULL;
 +            }
          }
-         if (strcmp(nc->model, "user")) {
+     }
-             monitor_printf(mon, "invalid device specified\n");
++
-@@ -XXX,XX +XXX,XX @@ void hmp_hostfwd_remove(Monitor *mon, const QDict *qdict)
+     rcu_read_unlock();
-     const char *arg2 = qdict_get_try_str(qdict, "arg2");
++    g_free(ram_state);
-     const char *arg3 = qdict_get_try_str(qdict, "arg3");
++    ram_state = NULL;
+ }
--    if (arg2) {
-+    if (arg3) {
+ /**
          s = slirp_lookup(mon, arg1, arg2);
          src_str = arg3;
 +    } else if (arg2) {
 +        s = slirp_lookup(mon, NULL, arg1);
 +        src_str = arg2;
      } else {
          s = slirp_lookup(mon, NULL, NULL);
          src_str = arg1;
@@ -XXX,XX +XXX,XX @@ void hmp_hostfwd_add(Monitor *mon, const QDict *qdict)
      const char *arg2 = qdict_get_try_str(qdict, "arg2");
      const char *arg3 = qdict_get_try_str(qdict, "arg3");
 -    if (arg2) {
 +    if (arg3) {
          s = slirp_lookup(mon, arg1, arg2);
          redir_str = arg3;
 +    } else if (arg2) {
 +        s = slirp_lookup(mon, NULL, arg1);
 +        redir_str = arg2;
      } else {
          s = slirp_lookup(mon, NULL, NULL);
          redir_str = arg1;
 --
-.7.4
+.5.0

-[Qemu-devel] [PULL 3/6] net: Allow hubports to connect to other netdevs
+[Qemu-devel] [PULL V2 09/26] COLO: Flush memory data from ram cache
-From: Thomas Huth <thuth@redhat.com>
+From: Zhang Chen <zhangckid@gmail.com>
-QEMU can emulate hubs to connect NICs and netdevs. This is currently
+During the time of VM's running, PVM may dirty some pages, we will transfer
-primarily used for the mis-named 'vlan' feature of the networking
+PVM's dirty pages to SVM and store them into SVM's RAM cache at next checkpoint
-subsystem. Now the 'vlan' feature has been marked as deprecated, since
+time. So, the content of SVM's RAM cache will always be same with PVM's memory
-its name is rather confusing and the users often rather mis-configure
+after checkpoint.
 their network when trying to use it. But while the 'vlan' parameter
 should be removed at one point in time, the basic idea of emulating
 a hub in QEMU is still good: It's useful for bundling up the output of
 multiple NICs into one single l2tp netdev for example.
-Now to be able to use the hubport feature without 'vlan's, there is one
+Instead of flushing all content of PVM's RAM cache into SVM's MEMORY,
-missing piece: The possibility to connect a hubport to a netdev, too.
+we do this in a more efficient way:
-This patch adds this possibility by introducing a new "netdev=..."
+Only flush any page that dirtied by PVM since last checkpoint.
-parameter to the hubports.
+In this way, we can ensure SVM's memory same with PVM's.
-To bundle up the output of multiple NICs into one socket netdev, you can
+Besides, we must ensure flush RAM cache before load device state.
 now run QEMU with these parameters for example:
-qemu-system-ppc64 ... -netdev socket,id=s1,connect=:11122 \
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
-    -netdev hubport,hubid=1,id=h1,netdev=s1 \
+Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
-    -netdev hubport,hubid=1,id=h2 -device e1000,netdev=h2 \
+Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
     -netdev hubport,hubid=1,id=h3 -device virtio-net-pci,netdev=h3
 For using the socket netdev, you have got to start another QEMU as the
 receiving side first, for example with network dumping enabled:
 qemu-system-x86_64 -M isapc -netdev socket,id=s0,listen=:11122 \
     -device ne2k_isa,netdev=s0 \
     -object filter-dump,id=f1,netdev=s0,file=/tmp/dump.dat
 After the ppc64 guest tried to boot from both NICs, you can see in the
 dump file (using Wireshark, for example), that the output of both NICs
 (the e1000 and the virtio-net-pci) has been successfully transfered
 via the socket netdev in this case.
 Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
 Signed-off-by: Thomas Huth <thuth@redhat.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
- net/hub.c       | 27 +++++++++++++++++++++------
+ migration/ram.c        | 37 +++++++++++++++++++++++++++++++++++++
- net/hub.h       |  3 ++-
+ migration/trace-events |  2 ++
- net/net.c       |  2 +-
+files changed, 39 insertions(+)
  qapi/net.json   |  4 +++-
  qemu-options.hx |  8 +++++---
 files changed, 32 insertions(+), 12 deletions(-)
-diff --git a/net/hub.c b/net/hub.c
+diff --git a/migration/ram.c b/migration/ram.c
 index XXXXXXX..XXXXXXX 100644
---- a/net/hub.c
+--- a/migration/ram.c
-+++ b/net/hub.c
++++ b/migration/ram.c
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ static bool postcopy_is_running(void)
-  */
+     return ps >= POSTCOPY_INCOMING_LISTENING && ps < POSTCOPY_INCOMING_END;
  #include "qemu/osdep.h"
 +#include "qapi/error.h"
  #include "monitor/monitor.h"
  #include "net/net.h"
  #include "clients.h"
@@ -XXX,XX +XXX,XX @@ static NetClientInfo net_hub_port_info = {
      .cleanup = net_hub_port_cleanup,
  };
 -static NetHubPort *net_hub_port_new(NetHub *hub, const char *name)
 +static NetHubPort *net_hub_port_new(NetHub *hub, const char *name,
 +                                    NetClientState *hubpeer)
  {
      NetClientState *nc;
      NetHubPort *port;
@@ -XXX,XX +XXX,XX @@ static NetHubPort *net_hub_port_new(NetHub *hub, const char *name)
          name = default_name;
      }
 -    nc = qemu_new_net_client(&net_hub_port_info, NULL, "hub", name);
 +    nc = qemu_new_net_client(&net_hub_port_info, hubpeer, "hub", name);
      port = DO_UPCAST(NetHubPort, nc, nc);
      port->id = id;
      port->hub = hub;
@@ -XXX,XX +XXX,XX @@ static NetHubPort *net_hub_port_new(NetHub *hub, const char *name)
  /**
   * Create a port on a given hub
 + * @hub_id: Number of the hub
   * @name: Net client name or NULL for default name.
 + * @hubpeer: Peer to use (if "netdev=id" has been specified)
   *
   * If there is no existing hub with the given id then a new hub is created.
   */
 -NetClientState *net_hub_add_port(int hub_id, const char *name)
 +NetClientState *net_hub_add_port(int hub_id, const char *name,
 +                                 NetClientState *hubpeer)
  {
      NetHub *hub;
      NetHubPort *port;
@@ -XXX,XX +XXX,XX @@ NetClientState *net_hub_add_port(int hub_id, const char *name)
          hub = net_hub_new(hub_id);
      }
 -    port = net_hub_port_new(hub, name);
 +    port = net_hub_port_new(hub, name, hubpeer);
      return &port->nc;
  }
-@@ -XXX,XX +XXX,XX @@ NetClientState *net_hub_port_find(int hub_id)
++/*
-         }
++ * Flush content of RAM cache into SVM's memory.
-     }
++ * Only flush the pages that be dirtied by PVM or SVM or both.
++ */
--    nc = net_hub_add_port(hub_id, NULL);
++static void colo_flush_ram_cache(void)
-+    nc = net_hub_add_port(hub_id, NULL, NULL);
++{
-     return nc;
++    RAMBlock *block = NULL;
- }
++    void *dst_host;
++    void *src_host;
-@@ -XXX,XX +XXX,XX @@ int net_init_hubport(const Netdev *netdev, const char *name,
++    unsigned long offset = 0;
-                      NetClientState *peer, Error **errp)
++
- {
++    trace_colo_flush_ram_cache_begin(ram_state->migration_dirty_pages);
-     const NetdevHubPortOptions *hubport;
++    rcu_read_lock();
-+    NetClientState *hubpeer = NULL;
++    block = QLIST_FIRST_RCU(&ram_list.blocks);
++
-     assert(netdev->type == NET_CLIENT_DRIVER_HUBPORT);
++    while (block) {
-     assert(!peer);
++        offset = migration_bitmap_find_dirty(ram_state, block, offset);
-     hubport = &netdev->u.hubport;
++
++        if (offset << TARGET_PAGE_BITS >= block->used_length) {
--    net_hub_add_port(hubport->hubid, name);
++            offset = 0;
-+    if (hubport->has_netdev) {
++            block = QLIST_NEXT_RCU(block, next);
-+        hubpeer = qemu_find_netdev(hubport->netdev);
++        } else {
-+        if (!hubpeer) {
++            migration_bitmap_clear_dirty(ram_state, block, offset);
-+            error_setg(errp, "netdev '%s' not found", hubport->netdev);
++            dst_host = block->host + (offset << TARGET_PAGE_BITS);
-+            return -1;
++            src_host = block->colo_cache + (offset << TARGET_PAGE_BITS);
 +            memcpy(dst_host, src_host, TARGET_PAGE_SIZE);
 +        }
 +    }
 +
-+    net_hub_add_port(hubport->hubid, name, hubpeer);
++    rcu_read_unlock();
 +    trace_colo_flush_ram_cache_end();
 +}
 +
-     return 0;
+ static int ram_load(QEMUFile *f, void *opaque, int version_id)
  {
      int flags = 0, ret = 0, invalid_flags = 0;
@@ -XXX,XX +XXX,XX @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
      ret |= wait_for_decompress_done();
      rcu_read_unlock();
      trace_ram_load_complete(ret, seq_iter);
 +
 +    if (!ret  && migration_incoming_in_colo_state()) {
 +        colo_flush_ram_cache();
 +    }
      return ret;
  }
-diff --git a/net/hub.h b/net/hub.h
+diff --git a/migration/trace-events b/migration/trace-events
 index XXXXXXX..XXXXXXX 100644
---- a/net/hub.h
+--- a/migration/trace-events
-+++ b/net/hub.h
++++ b/migration/trace-events
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ ram_dirty_bitmap_sync_start(void) ""
+ ram_dirty_bitmap_sync_wait(void) ""
- #include "qemu-common.h"
+ ram_dirty_bitmap_sync_complete(void) ""
+ ram_state_resume_prepare(uint64_t v) "%" PRId64
--NetClientState *net_hub_add_port(int hub_id, const char *name);
++colo_flush_ram_cache_begin(uint64_t dirty_pages) "dirty_pages %" PRIu64
-+NetClientState *net_hub_add_port(int hub_id, const char *name,
++colo_flush_ram_cache_end(void) ""
-+                                 NetClientState *hubpeer);
- NetClientState *net_hub_find_client_by_name(int hub_id, const char *name);
+ # migration/migration.c
- void net_hub_info(Monitor *mon);
+ await_return_path_close_on_source_close(void) ""
  void net_hub_check_clients(void);
 diff --git a/net/net.c b/net/net.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/net.c
 +++ b/net/net.c
@@ -XXX,XX +XXX,XX @@ static int net_client_init1(const void *object, bool is_netdev, Error **errp)
          /* Do not add to a vlan if it's a nic with a netdev= parameter. */
          if (netdev->type != NET_CLIENT_DRIVER_NIC ||
              !opts->u.nic.has_netdev) {
 -            peer = net_hub_add_port(net->has_vlan ? net->vlan : 0, NULL);
 +            peer = net_hub_add_port(net->has_vlan ? net->vlan : 0, NULL, NULL);
          }
          if (net->has_vlan && !vlan_warned) {
 diff --git a/qapi/net.json b/qapi/net.json
 index XXXXXXX..XXXXXXX 100644
 --- a/qapi/net.json
 +++ b/qapi/net.json
@@ -XXX,XX +XXX,XX @@
  # Connect two or more net clients through a software hub.
  #
  # @hubid: hub identifier number
 +# @netdev: used to connect hub to a netdev instead of a device (since 2.12)
  #
  # Since: 1.2
  ##
  { 'struct': 'NetdevHubPortOptions',
    'data': {
 -    'hubid':     'int32' } }
 +    'hubid':     'int32',
 +    '*netdev':    'str' } }
  ##
  # @NetdevNetmapOptions:
 diff --git a/qemu-options.hx b/qemu-options.hx
 index XXXXXXX..XXXXXXX 100644
 --- a/qemu-options.hx
 +++ b/qemu-options.hx
@@ -XXX,XX +XXX,XX @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
  #endif
      "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
      "                configure a vhost-user network, backed by a chardev 'dev'\n"
 -    "-netdev hubport,id=str,hubid=n\n"
 +    "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
      "                configure a hub port on QEMU VLAN 'n'\n", QEMU_ARCH_ALL)
  DEF("net", HAS_ARG, QEMU_OPTION_net,
      "-net nic[,vlan=n][,netdev=nd][,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
@@ -XXX,XX +XXX,XX @@ vde_switch -F -sock /tmp/myswitch
  qemu-system-i386 linux.img -net nic -net vde,sock=/tmp/myswitch
  @end example
 -@item -netdev hubport,id=@var{id},hubid=@var{hubid}
 +@item -netdev hubport,id=@var{id},hubid=@var{hubid}[,netdev=@var{nd}]
  Create a hub port on QEMU "vlan" @var{hubid}.
  The hubport netdev lets you connect a NIC to a QEMU "vlan" instead of a single
  netdev.  @code{-net} and @code{-device} with parameter @option{vlan} create the
 -required hub automatically.
 +required hub automatically. Alternatively, you can also connect the hubport
 +to another netdev with ID @var{nd} by using the @option{netdev=@var{nd}}
 +option.
  @item -netdev vhost-user,chardev=@var{id}[,vhostforce=on|off][,queues=n]
 --
-.7.4
+.5.0

-New patch
+[Qemu-devel] [PULL V2 10/26] qmp event: Add COLO_EXIT event to notify users while exited COLO
+From: zhanghailiang <zhang.zhanghailiang@huawei.com>
+If some errors happen during VM's COLO FT stage, it's important to
+notify the users of this event. Together with 'x-colo-lost-heartbeat',
+Users can intervene in COLO's failover work immediately.
+If users don't want to get involved in COLO's failover verdict,
+it is still necessary to notify users that we exited COLO mode.
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/colo.c    | 31 +++++++++++++++++++++++++++++++
+ qapi/migration.json | 38 ++++++++++++++++++++++++++++++++++++++
+files changed, 69 insertions(+)
+diff --git a/migration/colo.c b/migration/colo.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo.c
++++ b/migration/colo.c
+@@ -XXX,XX +XXX,XX @@
+ #include "net/colo-compare.h"
+ #include "net/colo.h"
+ #include "block/block.h"
++#include "qapi/qapi-events-migration.h"
+ static bool vmstate_loading;
+ static Notifier packets_compare_notifier;
+@@ -XXX,XX +XXX,XX @@ out:
+         qemu_fclose(fb);
+     }
++    /*
++     * There are only two reasons we can get here, some error happened
++     * or the user triggered failover.
++     */
++    switch (failover_get_state()) {
++    case FAILOVER_STATUS_NONE:
++        qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
++                                  COLO_EXIT_REASON_ERROR);
++        break;
++    case FAILOVER_STATUS_REQUIRE:
++        qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
++                                  COLO_EXIT_REASON_REQUEST);
++        break;
++    default:
++        abort();
++    }
++
+     /* Hope this not to be too long to wait here */
+     qemu_sem_wait(&s->colo_exit_sem);
+     qemu_sem_destroy(&s->colo_exit_sem);
+@@ -XXX,XX +XXX,XX @@ out:
+         error_report_err(local_err);
+     }
++    switch (failover_get_state()) {
++    case FAILOVER_STATUS_NONE:
++        qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
++                                  COLO_EXIT_REASON_ERROR);
++        break;
++    case FAILOVER_STATUS_REQUIRE:
++        qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
++                                  COLO_EXIT_REASON_REQUEST);
++        break;
++    default:
++        abort();
++    }
++
+     if (fb) {
+         qemu_fclose(fb);
+     }
+diff --git a/qapi/migration.json b/qapi/migration.json
+index XXXXXXX..XXXXXXX 100644
+--- a/qapi/migration.json
++++ b/qapi/migration.json
+@@ -XXX,XX +XXX,XX @@
+   'data': [ 'none', 'require', 'active', 'completed', 'relaunch' ] }
+ ##
++# @COLO_EXIT:
++#
++# Emitted when VM finishes COLO mode due to some errors happening or
++# at the request of users.
++#
++# @mode: report COLO mode when COLO exited.
++#
++# @reason: describes the reason for the COLO exit.
++#
++# Since: 3.1
++#
++# Example:
++#
++# <- { "timestamp": {"seconds": 2032141960, "microseconds": 417172},
++#      "event": "COLO_EXIT", "data": {"mode": "primary", "reason": "request" } }
++#
++##
++{ 'event': 'COLO_EXIT',
++  'data': {'mode': 'COLOMode', 'reason': 'COLOExitReason' } }
++
++##
++# @COLOExitReason:
++#
++# The reason for a COLO exit
++#
++# @none: no failover has ever happened. This can't occur in the
++# COLO_EXIT event, only in the result of query-colo-status.
++#
++# @request: COLO exit is due to an external request
++#
++# @error: COLO exit is due to an internal error
++#
++# Since: 3.1
++##
++{ 'enum': 'COLOExitReason',
++  'data': [ 'none', 'request', 'error' ] }
++
++##
+ # @x-colo-lost-heartbeat:
+ #
+ # Tell qemu that heartbeat is lost, request it to do takeover procedures.
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 11/26] qapi/migration.json: Rename COLO unknown mode to none mode.
+From: Zhang Chen <chen.zhang@intel.com>
+Suggested by Markus Armbruster rename COLO unknown mode to none mode.
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Reviewed-by: Eric Blake <eblake@redhat.com>
+Reviewed-by: Markus Armbruster <armbru@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/colo-failover.c |  2 +-
+ migration/colo.c          |  2 +-
+ qapi/migration.json       | 10 +++++-----
+files changed, 7 insertions(+), 7 deletions(-)
+diff --git a/migration/colo-failover.c b/migration/colo-failover.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo-failover.c
++++ b/migration/colo-failover.c
+@@ -XXX,XX +XXX,XX @@ FailoverStatus failover_get_state(void)
+ void qmp_x_colo_lost_heartbeat(Error **errp)
+ {
+-    if (get_colo_mode() == COLO_MODE_UNKNOWN) {
++    if (get_colo_mode() == COLO_MODE_NONE) {
+         error_setg(errp, QERR_FEATURE_DISABLED, "colo");
+         return;
+     }
+diff --git a/migration/colo.c b/migration/colo.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo.c
++++ b/migration/colo.c
+@@ -XXX,XX +XXX,XX @@ COLOMode get_colo_mode(void)
+     } else if (migration_incoming_in_colo_state()) {
+         return COLO_MODE_SECONDARY;
+     } else {
+-        return COLO_MODE_UNKNOWN;
++        return COLO_MODE_NONE;
+     }
+ }
+diff --git a/qapi/migration.json b/qapi/migration.json
+index XXXXXXX..XXXXXXX 100644
+--- a/qapi/migration.json
++++ b/qapi/migration.json
+@@ -XXX,XX +XXX,XX @@
+ ##
+ # @COLOMode:
+ #
+-# The colo mode
++# The COLO current mode.
+ #
+-# @unknown: unknown mode
++# @none: COLO is disabled.
+ #
+-# @primary: master side
++# @primary: COLO node in primary side.
+ #
+-# @secondary: slave side
++# @secondary: COLO node in slave side.
+ #
+ # Since: 2.8
+ ##
+ { 'enum': 'COLOMode',
+-  'data': [ 'unknown', 'primary', 'secondary'] }
++  'data': [ 'none', 'primary', 'secondary'] }
+ ##
+ # @FailoverStatus:
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 12/26] qapi: Add new command to query colo status
+From: Zhang Chen <zhangckid@gmail.com>
+Libvirt or other high level software can use this command query colo status.
+You can test this command like that:
+{'execute':'query-colo-status'}
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/colo.c    | 21 +++++++++++++++++++++
+ qapi/migration.json | 32 ++++++++++++++++++++++++++++++++
+files changed, 53 insertions(+)
+diff --git a/migration/colo.c b/migration/colo.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo.c
++++ b/migration/colo.c
+@@ -XXX,XX +XXX,XX @@
+ #include "net/colo.h"
+ #include "block/block.h"
+ #include "qapi/qapi-events-migration.h"
++#include "qapi/qmp/qerror.h"
+ static bool vmstate_loading;
+ static Notifier packets_compare_notifier;
+@@ -XXX,XX +XXX,XX @@ void qmp_xen_colo_do_checkpoint(Error **errp)
+ #endif
+ }
++COLOStatus *qmp_query_colo_status(Error **errp)
++{
++    COLOStatus *s = g_new0(COLOStatus, 1);
++
++    s->mode = get_colo_mode();
++
++    switch (failover_get_state()) {
++    case FAILOVER_STATUS_NONE:
++        s->reason = COLO_EXIT_REASON_NONE;
++        break;
++    case FAILOVER_STATUS_REQUIRE:
++        s->reason = COLO_EXIT_REASON_REQUEST;
++        break;
++    default:
++        s->reason = COLO_EXIT_REASON_ERROR;
++    }
++
++    return s;
++}
++
+ static void colo_send_message(QEMUFile *f, COLOMessage msg,
+                               Error **errp)
+ {
+diff --git a/qapi/migration.json b/qapi/migration.json
+index XXXXXXX..XXXXXXX 100644
+--- a/qapi/migration.json
++++ b/qapi/migration.json
+@@ -XXX,XX +XXX,XX @@
+ { 'command': 'xen-colo-do-checkpoint' }
+ ##
++# @COLOStatus:
++#
++# The result format for 'query-colo-status'.
++#
++# @mode: COLO running mode. If COLO is running, this field will return
++#        'primary' or 'secondary'.
++#
++# @reason: describes the reason for the COLO exit.
++#
++# Since: 3.0
++##
++{ 'struct': 'COLOStatus',
++  'data': { 'mode': 'COLOMode', 'reason': 'COLOExitReason' } }
++
++##
++# @query-colo-status:
++#
++# Query COLO status while the vm is running.
++#
++# Returns: A @COLOStatus object showing the status.
++#
++# Example:
++#
++# -> { "execute": "query-colo-status" }
++# <- { "return": { "mode": "primary", "active": true, "reason": "request" } }
++#
++# Since: 3.0
++##
++{ 'command': 'query-colo-status',
++  'returns': 'COLOStatus' }
++
++##
+ # @migrate-recover:
+ #
+ # Provide a recovery migration stream URI.
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 13/26] savevm: split the process of different stages for loadvm/savevm
+From: Zhang Chen <zhangckid@gmail.com>
 There are several stages during loadvm/savevm process. In different stage,
 migration incoming processes different types of sections.
 We want to control these stages more accuracy, it will benefit COLO
 performance, we don't have to save type of QEMU_VM_SECTION_START
 sections everytime while do checkpoint, besides, we want to separate
 the process of saving/loading memory and devices state.
 So we add three new helper functions: qemu_load_device_state() and
 qemu_savevm_live_state() to achieve different process during migration.
 Besides, we make qemu_loadvm_state_main() and qemu_save_device_state()
 public, and simplify the codes of qemu_save_device_state() by calling the
 wrapper qemu_savevm_state_header().
 Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
 Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
 Signed-off-by: Zhang Chen <zhangckid@gmail.com>
 Signed-off-by: Zhang Chen <chen.zhang@intel.com>
 Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
  migration/colo.c   | 41 ++++++++++++++++++++++++++++++++---------
  migration/savevm.c | 36 +++++++++++++++++++++++++++++-------
  migration/savevm.h |  4 ++++
 files changed, 65 insertions(+), 16 deletions(-)
 diff --git a/migration/colo.c b/migration/colo.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/colo.c
 +++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@
  #include "block/block.h"
  #include "qapi/qapi-events-migration.h"
  #include "qapi/qmp/qerror.h"
 +#include "sysemu/cpus.h"
  static bool vmstate_loading;
  static Notifier packets_compare_notifier;
@@ -XXX,XX +XXX,XX @@ static int colo_do_checkpoint_transaction(MigrationState *s,
      /* Disable block migration */
      migrate_set_block_enabled(false, &local_err);
 -    qemu_savevm_state_header(fb);
 -    qemu_savevm_state_setup(fb);
      qemu_mutex_lock_iothread();
      replication_do_checkpoint_all(&local_err);
      if (local_err) {
          qemu_mutex_unlock_iothread();
          goto out;
      }
 -    qemu_savevm_state_complete_precopy(fb, false, false);
 -    qemu_mutex_unlock_iothread();
 -
 -    qemu_fflush(fb);
      colo_send_message(s->to_dst_file, COLO_MESSAGE_VMSTATE_SEND, &local_err);
      if (local_err) {
 +        qemu_mutex_unlock_iothread();
 +        goto out;
 +    }
 +    /* Note: device state is saved into buffer */
 +    ret = qemu_save_device_state(fb);
 +
 +    qemu_mutex_unlock_iothread();
 +    if (ret < 0) {
          goto out;
      }
      /*
 +     * Only save VM's live state, which not including device state.
 +     * TODO: We may need a timeout mechanism to prevent COLO process
 +     * to be blocked here.
 +     */
 +    qemu_savevm_live_state(s->to_dst_file);
 +
 +    qemu_fflush(fb);
 +
 +    /*
       * We need the size of the VMstate data in Secondary side,
       * With which we can decide how much data should be read.
       */
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
      uint64_t total_size;
      uint64_t value;
      Error *local_err = NULL;
 +    int ret;
      rcu_register_thread();
      qemu_sem_init(&mis->colo_incoming_sem, 0);
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
              goto out;
          }
 +        qemu_mutex_lock_iothread();
 +        cpu_synchronize_all_pre_loadvm();
 +        ret = qemu_loadvm_state_main(mis->from_src_file, mis);
 +        qemu_mutex_unlock_iothread();
 +
 +        if (ret < 0) {
 +            error_report("Load VM's live state (ram) error");
 +            goto out;
 +        }
 +
          value = colo_receive_message_value(mis->from_src_file,
                                   COLO_MESSAGE_VMSTATE_SIZE, &local_err);
          if (local_err) {
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
          }
          qemu_mutex_lock_iothread();
 -        qemu_system_reset(SHUTDOWN_CAUSE_NONE);
          vmstate_loading = true;
 -        if (qemu_loadvm_state(fb) < 0) {
 -            error_report("COLO: loadvm failed");
 +        ret = qemu_load_device_state(fb);
 +        if (ret < 0) {
 +            error_report("COLO: load device state failed");
              qemu_mutex_unlock_iothread();
              goto out;
          }
 diff --git a/migration/savevm.c b/migration/savevm.c
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/savevm.c
 +++ b/migration/savevm.c
@@ -XXX,XX +XXX,XX @@ done:
      return ret;
  }
 -static int qemu_save_device_state(QEMUFile *f)
 +void qemu_savevm_live_state(QEMUFile *f)
  {
 -    SaveStateEntry *se;
 +    /* save QEMU_VM_SECTION_END section */
 +    qemu_savevm_state_complete_precopy(f, true, false);
 +    qemu_put_byte(f, QEMU_VM_EOF);
 +}
 -    qemu_put_be32(f, QEMU_VM_FILE_MAGIC);
 -    qemu_put_be32(f, QEMU_VM_FILE_VERSION);
 +int qemu_save_device_state(QEMUFile *f)
 +{
 +    SaveStateEntry *se;
 +    if (!migration_in_colo_state()) {
 +        qemu_put_be32(f, QEMU_VM_FILE_MAGIC);
 +        qemu_put_be32(f, QEMU_VM_FILE_VERSION);
 +    }
      cpu_synchronize_all_states();
      QTAILQ_FOREACH(se, &savevm_state.handlers, entry) {
@@ -XXX,XX +XXX,XX @@ enum LoadVMExitCodes {
      LOADVM_QUIT     =  1,
  };
 -static int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis);
 -
  /* ------ incoming postcopy messages ------ */
  /* 'advise' arrives before any transfers just to tell us that a postcopy
   * *might* happen - it might be skipped if precopy transferred everything
@@ -XXX,XX +XXX,XX @@ static bool postcopy_pause_incoming(MigrationIncomingState *mis)
      return true;
  }
 -static int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis)
 +int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis)
  {
      uint8_t section_type;
      int ret = 0;
@@ -XXX,XX +XXX,XX @@ int qemu_loadvm_state(QEMUFile *f)
      return ret;
  }
 +int qemu_load_device_state(QEMUFile *f)
 +{
 +    MigrationIncomingState *mis = migration_incoming_get_current();
 +    int ret;
 +
 +    /* Load QEMU_VM_SECTION_FULL section */
 +    ret = qemu_loadvm_state_main(f, mis);
 +    if (ret < 0) {
 +        error_report("Failed to load device state: %d", ret);
 +        return ret;
 +    }
 +
 +    cpu_synchronize_all_post_init();
 +    return 0;
 +}
 +
  int save_snapshot(const char *name, Error **errp)
  {
      BlockDriverState *bs, *bs1;
 diff --git a/migration/savevm.h b/migration/savevm.h
 index XXXXXXX..XXXXXXX 100644
 --- a/migration/savevm.h
 +++ b/migration/savevm.h
@@ -XXX,XX +XXX,XX @@ void qemu_savevm_send_postcopy_ram_discard(QEMUFile *f, const char *name,
                                             uint64_t *start_list,
                                             uint64_t *length_list);
  void qemu_savevm_send_colo_enable(QEMUFile *f);
 +void qemu_savevm_live_state(QEMUFile *f);
 +int qemu_save_device_state(QEMUFile *f);
  int qemu_loadvm_state(QEMUFile *f);
  void qemu_loadvm_state_cleanup(void);
 +int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis);
 +int qemu_load_device_state(QEMUFile *f);
  #endif
 --
 .5.0

-New patch
+[Qemu-devel] [PULL V2 14/26] COLO: flush host dirty ram from cache
+From: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Don't need to flush all VM's ram from cache, only
+flush the dirty pages since last checkpoint
+Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/ram.c | 9 +++++++++
+file changed, 9 insertions(+)
+diff --git a/migration/ram.c b/migration/ram.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/ram.c
++++ b/migration/ram.c
+@@ -XXX,XX +XXX,XX @@ int colo_init_ram_cache(void)
+     }
+     ram_state = g_new0(RAMState, 1);
+     ram_state->migration_dirty_pages = 0;
++    memory_global_dirty_log_start();
+     return 0;
+@@ -XXX,XX +XXX,XX @@ void colo_release_ram_cache(void)
+ {
+     RAMBlock *block;
++    memory_global_dirty_log_stop();
+     RAMBLOCK_FOREACH_MIGRATABLE(block) {
+         g_free(block->bmap);
+         block->bmap = NULL;
+@@ -XXX,XX +XXX,XX @@ static void colo_flush_ram_cache(void)
+     void *src_host;
+     unsigned long offset = 0;
++    memory_global_dirty_log_sync();
++    rcu_read_lock();
++    RAMBLOCK_FOREACH_MIGRATABLE(block) {
++        migration_bitmap_sync_range(ram_state, block, 0, block->used_length);
++    }
++    rcu_read_unlock();
++
+     trace_colo_flush_ram_cache_begin(ram_state->migration_dirty_pages);
+     rcu_read_lock();
+     block = QLIST_FIRST_RCU(&ram_list.blocks);
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 15/26] filter: Add handle_event method for NetFilterClass
+From: Zhang Chen <zhangckid@gmail.com>
+Filter needs to process the event of checkpoint/failover or
+other event passed by COLO frame.
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ include/net/filter.h |  5 +++++
+ net/filter.c         | 17 +++++++++++++++++
+ net/net.c            | 19 +++++++++++++++++++
+files changed, 41 insertions(+)
+diff --git a/include/net/filter.h b/include/net/filter.h
+index XXXXXXX..XXXXXXX 100644
+--- a/include/net/filter.h
++++ b/include/net/filter.h
+@@ -XXX,XX +XXX,XX @@ typedef ssize_t (FilterReceiveIOV)(NetFilterState *nc,
+ typedef void (FilterStatusChanged) (NetFilterState *nf, Error **errp);
++typedef void (FilterHandleEvent) (NetFilterState *nf, int event, Error **errp);
++
+ typedef struct NetFilterClass {
+     ObjectClass parent_class;
+@@ -XXX,XX +XXX,XX @@ typedef struct NetFilterClass {
+     FilterSetup *setup;
+     FilterCleanup *cleanup;
+     FilterStatusChanged *status_changed;
++    FilterHandleEvent *handle_event;
+     /* mandatory */
+     FilterReceiveIOV *receive_iov;
+ } NetFilterClass;
+@@ -XXX,XX +XXX,XX @@ ssize_t qemu_netfilter_pass_to_next(NetClientState *sender,
+                                     int iovcnt,
+                                     void *opaque);
++void colo_notify_filters_event(int event, Error **errp);
++
+ #endif /* QEMU_NET_FILTER_H */
+diff --git a/net/filter.c b/net/filter.c
+index XXXXXXX..XXXXXXX 100644
+--- a/net/filter.c
++++ b/net/filter.c
+@@ -XXX,XX +XXX,XX @@
+ #include "net/vhost_net.h"
+ #include "qom/object_interfaces.h"
+ #include "qemu/iov.h"
++#include "net/colo.h"
++#include "migration/colo.h"
+ static inline bool qemu_can_skip_netfilter(NetFilterState *nf)
+ {
+@@ -XXX,XX +XXX,XX @@ static void netfilter_finalize(Object *obj)
+     g_free(nf->netdev_id);
+ }
++static void default_handle_event(NetFilterState *nf, int event, Error **errp)
++{
++    switch (event) {
++    case COLO_EVENT_CHECKPOINT:
++        break;
++    case COLO_EVENT_FAILOVER:
++        object_property_set_str(OBJECT(nf), "off", "status", errp);
++        break;
++    default:
++        break;
++    }
++}
++
+ static void netfilter_class_init(ObjectClass *oc, void *data)
+ {
+     UserCreatableClass *ucc = USER_CREATABLE_CLASS(oc);
++    NetFilterClass *nfc = NETFILTER_CLASS(oc);
+     ucc->complete = netfilter_complete;
++    nfc->handle_event = default_handle_event;
+ }
+ static const TypeInfo netfilter_info = {
+diff --git a/net/net.c b/net/net.c
+index XXXXXXX..XXXXXXX 100644
+--- a/net/net.c
++++ b/net/net.c
+@@ -XXX,XX +XXX,XX @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
+     }
+ }
++void colo_notify_filters_event(int event, Error **errp)
++{
++    NetClientState *nc;
++    NetFilterState *nf;
++    NetFilterClass *nfc = NULL;
++    Error *local_err = NULL;
++
++    QTAILQ_FOREACH(nc, &net_clients, next) {
++        QTAILQ_FOREACH(nf, &nc->filters, next) {
++            nfc = NETFILTER_GET_CLASS(OBJECT(nf));
++            nfc->handle_event(nf, event, &local_err);
++            if (local_err) {
++                error_propagate(errp, local_err);
++                return;
++            }
++        }
++    }
++}
++
+ void qmp_set_link(const char *name, bool up, Error **errp)
+ {
+     NetClientState *ncs[MAX_QUEUE_NUM];
+--
+.5.0

-[Qemu-devel] [PULL 2/6] colo: compare the packet based on the tcp sequence number
+[Qemu-devel] [PULL V2 16/26] filter-rewriter: handle checkpoint and failover event
-From: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>
+From: Zhang Chen <zhangckid@gmail.com>
-Packet size some time different or when network is busy.
+After one round of checkpoint, the states between PVM and SVM
-Based on same payload size, but TCP protocol can not
+become consistent, so it is unnecessary to adjust the sequence
-guarantee send the same one packet in the same way,
+of net packets for old connections, besides, while failover
 happens, filter-rewriter will into failover mode that needn't
 handle the new TCP connection.
-like that:
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
 We send this payload:
 ------------------------------
 | header |1|2|3|4|5|6|7|8|9|0|
 ------------------------------
 primary:
 ppkt1:
 ----------------
 | header |1|2|3|
 ----------------
 ppkt2:
 ------------------------
 | header |4|5|6|7|8|9|0|
 ------------------------
 secondary:
 spkt1:
 ------------------------------
 | header |1|2|3|4|5|6|7|8|9|0|
 ------------------------------
 In the original method, ppkt1 and ppkt2 are different in size and
 spkt1, so they can't compare and trigger the checkpoint.
 I have tested FTP get 200M and 1G file many times, I found that
 the performance was less than 1% of the native.
 Now I reconstructed the comparison of TCP packets based on the
 TCP sequence number. first of all, ppkt1 and spkt1 have the same
 starting sequence number, so they can compare, even though their
 length is different. And then ppkt1 with a smaller payload length
 is used as the comparison length, if the payload is same, send
 out the ppkt1 and record the offset(the length of ppkt1 payload)
 in spkt1. The next comparison, ppkt2 and spkt1 can be compared
 from the recorded position of spkt1.
 like that:
 ----------------
 | header |1|2|3| ppkt1
 ---------|-----|
          |     |
 ---------v-----v--------------
 | header |1|2|3|4|5|6|7|8|9|0| spkt1
 ---------------|\------------|
                | \offset     |
       ---------v-------------v
       | header |4|5|6|7|8|9|0| ppkt2
       ------------------------
 In this way, the performance can reach native 20% in my multiple
 tests.
 Cc: Zhang Chen <zhangckid@gmail.com>
 Cc: Li Zhijian <lizhijian@cn.fujitsu.com>
 Cc: Jason Wang <jasowang@redhat.com>
 Signed-off-by: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>
 Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
 Signed-off-by: Zhang Chen <zhangckid@gmail.com>
-Reviewed-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
 Tested-by: Zhang Chen <zhangckid@gmail.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
- net/colo-compare.c | 343 +++++++++++++++++++++++++++++++++++------------------
+ net/colo-compare.c    | 12 +++++------
- net/colo.c         |   9 ++
+ net/colo.c            |  8 ++++++++
- net/colo.h         |  15 +++
+ net/colo.h            |  2 ++
- net/trace-events   |   2 +-
+ net/filter-rewriter.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++
-files changed, 250 insertions(+), 119 deletions(-)
+files changed, 73 insertions(+), 6 deletions(-)
 diff --git a/net/colo-compare.c b/net/colo-compare.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/colo-compare.c
 +++ b/net/colo-compare.c
-@@ -XXX,XX +XXX,XX @@
+@@ -XXX,XX +XXX,XX @@ enum {
- #define COMPARE_READ_LEN_MAX NET_BUFSIZE
+     SECONDARY_IN,
- #define MAX_QUEUE_SIZE 1024
+ };
-+#define COLO_COMPARE_FREE_PRIMARY     0x01
++static void colo_compare_inconsistency_notify(void)
 +#define COLO_COMPARE_FREE_SECONDARY   0x02
 +
  /* TODO: Should be configurable */
  #define REGULAR_PACKET_CHECK_MS 3000
@@ -XXX,XX +XXX,XX @@ static gint seq_sorter(Packet *a, Packet *b, gpointer data)
      return ntohl(atcp->th_seq) - ntohl(btcp->th_seq);
  }
 +static void fill_pkt_tcp_info(void *data, uint32_t *max_ack)
 +{
-+    Packet *pkt = data;
++    notifier_list_notify(&colo_compare_notifiers,
-+    struct tcphdr *tcphd;
++                migrate_get_current());
 +
 +    tcphd = (struct tcphdr *)pkt->transport_header;
 +
 +    pkt->tcp_seq = ntohl(tcphd->th_seq);
 +    pkt->tcp_ack = ntohl(tcphd->th_ack);
 +    *max_ack = *max_ack > pkt->tcp_ack ? *max_ack : pkt->tcp_ack;
 +    pkt->header_size = pkt->transport_header - (uint8_t *)pkt->data
 +                       + (tcphd->th_off << 2) - pkt->vnet_hdr_len;
 +    pkt->payload_size = pkt->size - pkt->header_size;
 +    pkt->seq_end = pkt->tcp_seq + pkt->payload_size;
 +    pkt->flags = tcphd->th_flags;
 +}
 +
  /*
   * Return 1 on success, if return 0 means the
   * packet will be dropped
   */
 -static int colo_insert_packet(GQueue *queue, Packet *pkt)
 +static int colo_insert_packet(GQueue *queue, Packet *pkt, uint32_t *max_ack)
  {
      if (g_queue_get_length(queue) <= MAX_QUEUE_SIZE) {
          if (pkt->ip->ip_p == IPPROTO_TCP) {
 +            fill_pkt_tcp_info(pkt, max_ack);
              g_queue_insert_sorted(queue,
                                    pkt,
                                    (GCompareDataFunc)seq_sorter,
@@ -XXX,XX +XXX,XX @@ static int packet_enqueue(CompareState *s, int mode, Connection **con)
      }
      if (mode == PRIMARY_IN) {
 -        if (!colo_insert_packet(&conn->primary_list, pkt)) {
 +        if (!colo_insert_packet(&conn->primary_list, pkt, &conn->pack)) {
              error_report("colo compare primary queue size too big,"
                           "drop packet");
          }
      } else {
 -        if (!colo_insert_packet(&conn->secondary_list, pkt)) {
 +        if (!colo_insert_packet(&conn->secondary_list, pkt, &conn->sack)) {
              error_report("colo compare secondary queue size too big,"
                           "drop packet");
          }
@@ -XXX,XX +XXX,XX @@ static int packet_enqueue(CompareState *s, int mode, Connection **con)
      return 0;
  }
 +static inline bool after(uint32_t seq1, uint32_t seq2)
 +{
 +        return (int32_t)(seq1 - seq2) > 0;
 +}
 +
 +static void colo_release_primary_pkt(CompareState *s, Packet *pkt)
 +{
 +    int ret;
 +    ret = compare_chr_send(s,
 +                           pkt->data,
 +                           pkt->size,
 +                           pkt->vnet_hdr_len);
 +    if (ret < 0) {
 +        error_report("colo send primary packet failed");
 +    }
 +    trace_colo_compare_main("packet same and release packet");
 +    packet_destroy(pkt, NULL);
 +}
 +
  /*
   * The IP packets sent by primary and secondary
   * will be compared in here
@@ -XXX,XX +XXX,XX @@ static int colo_compare_packet_payload(Packet *ppkt,
  }
  /*
 - * Called from the compare thread on the primary
 - * for compare tcp packet
 - * compare_tcp copied from Dr. David Alan Gilbert's branch
 - */
 -static int colo_packet_compare_tcp(Packet *spkt, Packet *ppkt)
 + * return true means that the payload is consist and
 + * need to make the next comparison, false means do
 + * the checkpoint
 +*/
 +static bool colo_mark_tcp_pkt(Packet *ppkt, Packet *spkt,
 +                              int8_t *mark, uint32_t max_ack)
  {
 -    struct tcphdr *ptcp, *stcp;
 -    int res;
 +    *mark = 0;
 +
 +    if (ppkt->tcp_seq == spkt->tcp_seq && ppkt->seq_end == spkt->seq_end) {
 +        if (colo_compare_packet_payload(ppkt, spkt,
 +                                        ppkt->header_size, spkt->header_size,
 +                                        ppkt->payload_size)) {
 +            *mark = COLO_COMPARE_FREE_SECONDARY | COLO_COMPARE_FREE_PRIMARY;
 +            return true;
 +        }
 +    }
 +    if (ppkt->tcp_seq == spkt->tcp_seq && ppkt->seq_end == spkt->seq_end) {
 +        if (colo_compare_packet_payload(ppkt, spkt,
 +                                        ppkt->header_size, spkt->header_size,
 +                                        ppkt->payload_size)) {
 +            *mark = COLO_COMPARE_FREE_SECONDARY | COLO_COMPARE_FREE_PRIMARY;
 +            return true;
 +        }
 +    }
 +
 +    /* one part of secondary packet payload still need to be compared */
 +    if (!after(ppkt->seq_end, spkt->seq_end)) {
 +        if (colo_compare_packet_payload(ppkt, spkt,
 +                                        ppkt->header_size + ppkt->offset,
 +                                        spkt->header_size + spkt->offset,
 +                                        ppkt->payload_size - ppkt->offset)) {
 +            if (!after(ppkt->tcp_ack, max_ack)) {
 +                *mark = COLO_COMPARE_FREE_PRIMARY;
 +                spkt->offset += ppkt->payload_size - ppkt->offset;
 +                return true;
 +            } else {
 +                /* secondary guest hasn't ack the data, don't send
 +                 * out this packet
 +                 */
 +                return false;
 +            }
 +        }
 +    } else {
 +        /* primary packet is longer than secondary packet, compare
 +         * the same part and mark the primary packet offset
 +         */
 +        if (colo_compare_packet_payload(ppkt, spkt,
 +                                        ppkt->header_size + ppkt->offset,
 +                                        spkt->header_size + spkt->offset,
 +                                        spkt->payload_size - spkt->offset)) {
 +            *mark = COLO_COMPARE_FREE_SECONDARY;
 +            ppkt->offset += spkt->payload_size - spkt->offset;
 +            return true;
 +        }
 +    }
 -    trace_colo_compare_main("compare tcp");
 +    return false;
 +}
 -    ptcp = (struct tcphdr *)ppkt->transport_header;
 -    stcp = (struct tcphdr *)spkt->transport_header;
 +static void colo_compare_tcp(CompareState *s, Connection *conn)
 +{
 +    Packet *ppkt = NULL, *spkt = NULL;
 +    int8_t mark;
      /*
 -     * The 'identification' field in the IP header is *very* random
 -     * it almost never matches.  Fudge this by ignoring differences in
 -     * unfragmented packets; they'll normally sort themselves out if different
 -     * anyway, and it should recover at the TCP level.
 -     * An alternative would be to get both the primary and secondary to rewrite
 -     * somehow; but that would need some sync traffic to sync the state
 -     */
 -    if (ntohs(ppkt->ip->ip_off) & IP_DF) {
 -        spkt->ip->ip_id = ppkt->ip->ip_id;
 -        /* and the sum will be different if the IDs were different */
 -        spkt->ip->ip_sum = ppkt->ip->ip_sum;
 +     * If ppkt and spkt have the same payload, but ppkt's ACK
 +     * is greater than spkt's ACK, in this case we can not
 +     * send the ppkt because it will cause the secondary guest
 +     * to miss sending some data in the next. Therefore, we
 +     * record the maximum ACK in the current queue at both
 +     * primary side and secondary side. Only when the ack is
 +     * less than the smaller of the two maximum ack, then we
 +     * can ensure that the packet's payload is acknowledged by
 +     * primary and secondary.
 +    */
 +    uint32_t min_ack = conn->pack > conn->sack ? conn->sack : conn->pack;
 +
 +pri:
 +    if (g_queue_is_empty(&conn->primary_list)) {
 +        return;
      }
 +    ppkt = g_queue_pop_head(&conn->primary_list);
 +sec:
 +    if (g_queue_is_empty(&conn->secondary_list)) {
 +        g_queue_push_head(&conn->primary_list, ppkt);
 +        return;
 +    }
 +    spkt = g_queue_pop_head(&conn->secondary_list);
 -    /*
 -     * Check tcp header length for tcp option field.
 -     * th_off > 5 means this tcp packet have options field.
 -     * The tcp options maybe always different.
 -     * for example:
 -     * From RFC 7323.
 -     * TCP Timestamps option (TSopt):
 -     * Kind: 8
 -     *
 -     * Length: 10 bytes
 -     *
 -     *    +-------+-------+---------------------+---------------------+
 -     *    |Kind=8 |  10   |   TS Value (TSval)  |TS Echo Reply (TSecr)|
 -     *    +-------+-------+---------------------+---------------------+
 -     *       1       1              4                     4
 -     *
 -     * In this case the primary guest's timestamp always different with
 -     * the secondary guest's timestamp. COLO just focus on payload,
 -     * so we just need skip this field.
 -     */
 +    if (ppkt->tcp_seq == ppkt->seq_end) {
 +        colo_release_primary_pkt(s, ppkt);
 +        ppkt = NULL;
 +    }
 -    ptrdiff_t ptcp_offset, stcp_offset;
 +    if (ppkt && conn->compare_seq && !after(ppkt->seq_end, conn->compare_seq)) {
 +        trace_colo_compare_main("pri: this packet has compared");
 +        colo_release_primary_pkt(s, ppkt);
 +        ppkt = NULL;
 +    }
 -    ptcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
 -                  + (ptcp->th_off << 2) - ppkt->vnet_hdr_len;
 -    stcp_offset = spkt->transport_header - (uint8_t *)spkt->data
 -                  + (stcp->th_off << 2) - spkt->vnet_hdr_len;
 -    if (ppkt->size - ptcp_offset == spkt->size - stcp_offset) {
 -        res = colo_compare_packet_payload(ppkt, spkt,
 -                                          ptcp_offset, stcp_offset,
 -                                          ppkt->size - ptcp_offset);
 +    if (spkt->tcp_seq == spkt->seq_end) {
 +        packet_destroy(spkt, NULL);
 +        if (!ppkt) {
 +            goto pri;
 +        } else {
 +            goto sec;
 +        }
      } else {
 -        trace_colo_compare_main("TCP: payload size of packets are different");
 -        res = -1;
 +        if (conn->compare_seq && !after(spkt->seq_end, conn->compare_seq)) {
 +            trace_colo_compare_main("sec: this packet has compared");
 +            packet_destroy(spkt, NULL);
 +            if (!ppkt) {
 +                goto pri;
 +            } else {
 +                goto sec;
 +            }
 +        }
 +        if (!ppkt) {
 +            g_queue_push_head(&conn->secondary_list, spkt);
 +            goto pri;
 +        }
      }
 -    if (res != 0 &&
 -        trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
 -        char pri_ip_src[20], pri_ip_dst[20], sec_ip_src[20], sec_ip_dst[20];
 -
 -        strcpy(pri_ip_src, inet_ntoa(ppkt->ip->ip_src));
 -        strcpy(pri_ip_dst, inet_ntoa(ppkt->ip->ip_dst));
 -        strcpy(sec_ip_src, inet_ntoa(spkt->ip->ip_src));
 -        strcpy(sec_ip_dst, inet_ntoa(spkt->ip->ip_dst));
 -
 -        trace_colo_compare_ip_info(ppkt->size, pri_ip_src,
 -                                   pri_ip_dst, spkt->size,
 -                                   sec_ip_src, sec_ip_dst);
 -
 -        trace_colo_compare_tcp_info("pri tcp packet",
 -                                    ntohl(ptcp->th_seq),
 -                                    ntohl(ptcp->th_ack),
 -                                    res, ptcp->th_flags,
 -                                    ppkt->size);
 -
 -        trace_colo_compare_tcp_info("sec tcp packet",
 -                                    ntohl(stcp->th_seq),
 -                                    ntohl(stcp->th_ack),
 -                                    res, stcp->th_flags,
 -                                    spkt->size);
 +    if (colo_mark_tcp_pkt(ppkt, spkt, &mark, min_ack)) {
 +        trace_colo_compare_tcp_info("pri",
 +                                    ppkt->tcp_seq, ppkt->tcp_ack,
 +                                    ppkt->header_size, ppkt->payload_size,
 +                                    ppkt->offset, ppkt->flags);
 +
 +        trace_colo_compare_tcp_info("sec",
 +                                    spkt->tcp_seq, spkt->tcp_ack,
 +                                    spkt->header_size, spkt->payload_size,
 +                                    spkt->offset, spkt->flags);
 +
 +        if (mark == COLO_COMPARE_FREE_PRIMARY) {
 +            conn->compare_seq = ppkt->seq_end;
 +            colo_release_primary_pkt(s, ppkt);
 +            g_queue_push_head(&conn->secondary_list, spkt);
 +            goto pri;
 +        }
 +        if (mark == COLO_COMPARE_FREE_SECONDARY) {
 +            conn->compare_seq = spkt->seq_end;
 +            packet_destroy(spkt, NULL);
 +            goto sec;
 +        }
 +        if (mark == (COLO_COMPARE_FREE_PRIMARY | COLO_COMPARE_FREE_SECONDARY)) {
 +            conn->compare_seq = ppkt->seq_end;
 +            colo_release_primary_pkt(s, ppkt);
 +            packet_destroy(spkt, NULL);
 +            goto pri;
 +        }
 +    } else {
 +        g_queue_push_head(&conn->primary_list, ppkt);
 +        g_queue_push_head(&conn->secondary_list, spkt);
          qemu_hexdump((char *)ppkt->data, stderr,
                       "colo-compare ppkt", ppkt->size);
          qemu_hexdump((char *)spkt->data, stderr,
                       "colo-compare spkt", spkt->size);
 -    }
 -    return res;
 +        /*
 +         * colo_compare_inconsistent_notify();
 +         * TODO: notice to checkpoint();
 +         */
 +    }
  }
 +
  /*
   * Called from the compare thread on the primary
   * for compare udp packet
@@ -XXX,XX +XXX,XX @@ static void colo_old_packet_check(void *opaque)
                          (GCompareFunc)colo_old_packet_check_one_conn);
  }
 -/*
 - * Called from the compare thread on the primary
 - * for compare packet with secondary list of the
 - * specified connection when a new packet was
 - * queued to it.
 - */
 -static void colo_compare_connection(void *opaque, void *user_data)
 +static void colo_compare_packet(CompareState *s, Connection *conn,
 +                                int (*HandlePacket)(Packet *spkt,
 +                                Packet *ppkt))
  {
 -    CompareState *s = user_data;
 -    Connection *conn = opaque;
      Packet *pkt = NULL;
      GList *result = NULL;
 -    int ret;
      while (!g_queue_is_empty(&conn->primary_list) &&
             !g_queue_is_empty(&conn->secondary_list)) {
          pkt = g_queue_pop_head(&conn->primary_list);
 -        switch (conn->ip_proto) {
 -        case IPPROTO_TCP:
 -            result = g_queue_find_custom(&conn->secondary_list,
 -                     pkt, (GCompareFunc)colo_packet_compare_tcp);
 -            break;
 -        case IPPROTO_UDP:
 -            result = g_queue_find_custom(&conn->secondary_list,
 -                     pkt, (GCompareFunc)colo_packet_compare_udp);
 -            break;
 -        case IPPROTO_ICMP:
 -            result = g_queue_find_custom(&conn->secondary_list,
 -                     pkt, (GCompareFunc)colo_packet_compare_icmp);
 -            break;
 -        default:
 -            result = g_queue_find_custom(&conn->secondary_list,
 -                     pkt, (GCompareFunc)colo_packet_compare_other);
 -            break;
 -        }
 +        result = g_queue_find_custom(&conn->secondary_list,
 +                 pkt, (GCompareFunc)HandlePacket);
          if (result) {
 -            ret = compare_chr_send(s,
 -                                   pkt->data,
 -                                   pkt->size,
 -                                   pkt->vnet_hdr_len);
 -            if (ret < 0) {
 -                error_report("colo_send_primary_packet failed");
 -            }
 -            trace_colo_compare_main("packet same and release packet");
 +            colo_release_primary_pkt(s, pkt);
              g_queue_remove(&conn->secondary_list, result->data);
 -            packet_destroy(pkt, NULL);
          } else {
              /*
               * If one packet arrive late, the secondary_list or
@@ -XXX,XX +XXX,XX @@ static void colo_compare_connection(void *opaque, void *user_data)
      }
  }
 +/*
 + * Called from the compare thread on the primary
 + * for compare packet with secondary list of the
 + * specified connection when a new packet was
 + * queued to it.
 + */
 +static void colo_compare_connection(void *opaque, void *user_data)
 +{
 +    CompareState *s = user_data;
 +    Connection *conn = opaque;
 +
 +    switch (conn->ip_proto) {
 +    case IPPROTO_TCP:
 +        colo_compare_tcp(s, conn);
 +        break;
 +    case IPPROTO_UDP:
 +        colo_compare_packet(s, conn, colo_packet_compare_udp);
 +        break;
 +    case IPPROTO_ICMP:
 +        colo_compare_packet(s, conn, colo_packet_compare_icmp);
 +        break;
 +    default:
 +        colo_compare_packet(s, conn, colo_packet_compare_other);
 +        break;
 +    }
 +}
 +
  static int compare_chr_send(CompareState *s,
                              const uint8_t *buf,
                              uint32_t size,
+@@ -XXX,XX +XXX,XX @@ static bool colo_mark_tcp_pkt(Packet *ppkt, Packet *spkt,
+     return false;
+ }
+-static void colo_compare_inconsistency_notify(void)
+-{
+-    notifier_list_notify(&colo_compare_notifiers,
+-                migrate_get_current());
+-}
+-
+ static void colo_compare_tcp(CompareState *s, Connection *conn)
+ {
+     Packet *ppkt = NULL, *spkt = NULL;
 diff --git a/net/colo.c b/net/colo.c
 index XXXXXXX..XXXXXXX 100644
 --- a/net/colo.c
 +++ b/net/colo.c
-@@ -XXX,XX +XXX,XX @@ Connection *connection_new(ConnectionKey *key)
+@@ -XXX,XX +XXX,XX @@ Connection *connection_get(GHashTable *connection_track_table,
-     conn->processing = false;
-     conn->offset = 0;
+     return conn;
      conn->syn_flag = 0;
 +    conn->pack = 0;
 +    conn->sack = 0;
      g_queue_init(&conn->primary_list);
      g_queue_init(&conn->secondary_list);
@@ -XXX,XX +XXX,XX @@ Packet *packet_new(const void *data, int size, int vnet_hdr_len)
      pkt->size = size;
      pkt->creation_ms = qemu_clock_get_ms(QEMU_CLOCK_HOST);
      pkt->vnet_hdr_len = vnet_hdr_len;
 +    pkt->tcp_seq = 0;
 +    pkt->tcp_ack = 0;
 +    pkt->seq_end = 0;
 +    pkt->header_size = 0;
 +    pkt->payload_size = 0;
 +    pkt->offset = 0;
 +    pkt->flags = 0;
      return pkt;
  }
++
++bool connection_has_tracked(GHashTable *connection_track_table,
++                            ConnectionKey *key)
++{
++    Connection *conn = g_hash_table_lookup(connection_track_table, key);
++
++    return conn ? true : false;
++}
 diff --git a/net/colo.h b/net/colo.h
 index XXXXXXX..XXXXXXX 100644
 --- a/net/colo.h
 +++ b/net/colo.h
-@@ -XXX,XX +XXX,XX @@ typedef struct Packet {
+@@ -XXX,XX +XXX,XX @@ void connection_destroy(void *opaque);
-     int64_t creation_ms;
+ Connection *connection_get(GHashTable *connection_track_table,
-     /* Get vnet_hdr_len from filter */
+                            ConnectionKey *key,
-     uint32_t vnet_hdr_len;
+                            GQueue *conn_list);
-+    uint32_t tcp_seq; /* sequence number */
++bool connection_has_tracked(GHashTable *connection_track_table,
-+    uint32_t tcp_ack; /* acknowledgement number */
++                            ConnectionKey *key);
-+    /* the sequence number of the last byte of the packet */
+ void connection_hashtable_reset(GHashTable *connection_track_table);
-+    uint32_t seq_end;
+ Packet *packet_new(const void *data, int size, int vnet_hdr_len);
-+    uint8_t header_size;  /* the header length */
+ void packet_destroy(void *opaque, void *user_data);
-+    uint16_t payload_size; /* the payload length */
+diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c
 +    /* record the payload offset(the length that has been compared) */
 +    uint16_t offset;
 +    uint8_t flags; /* Flags(aka Control bits) */
  } Packet;
  typedef struct ConnectionKey {
@@ -XXX,XX +XXX,XX @@ typedef struct Connection {
      /* flag to enqueue unprocessed_connections */
      bool processing;
      uint8_t ip_proto;
 +    /* record the sequence number that has been compared */
 +    uint32_t compare_seq;
 +    /* the maximum of acknowledgement number in primary_list queue */
 +    uint32_t pack;
 +    /* the maximum of acknowledgement number in secondary_list queue */
 +    uint32_t sack;
      /* offset = secondary_seq - primary_seq */
      tcp_seq  offset;
      /*
 diff --git a/net/trace-events b/net/trace-events
 index XXXXXXX..XXXXXXX 100644
---- a/net/trace-events
+--- a/net/filter-rewriter.c
-+++ b/net/trace-events
++++ b/net/filter-rewriter.c
-@@ -XXX,XX +XXX,XX @@ colo_compare_icmp_miscompare(const char *sta, int size) ": %s = %d"
+@@ -XXX,XX +XXX,XX @@
- colo_compare_ip_info(int psize, const char *sta, const char *stb, int ssize, const char *stc, const char *std) "ppkt size = %d, ip_src = %s, ip_dst = %s, spkt size = %d, ip_src = %s, ip_dst = %s"
+ #include "qemu/main-loop.h"
- colo_old_packet_check_found(int64_t old_time) "%" PRId64
+ #include "qemu/iov.h"
- colo_compare_miscompare(void) ""
+ #include "net/checksum.h"
--colo_compare_tcp_info(const char *pkt, uint32_t seq, uint32_t ack, int res, uint32_t flag, int size) "side: %s seq/ack= %u/%u res= %d flags= 0x%x pkt_size: %d\n"
++#include "net/colo.h"
-+colo_compare_tcp_info(const char *pkt, uint32_t seq, uint32_t ack, int hdlen, int pdlen, int offset, int flags) "%s: seq/ack= %u/%u hdlen= %d pdlen= %d offset= %d flags=%d\n"
++#include "migration/colo.h"
- # net/filter-rewriter.c
+ #define FILTER_COLO_REWRITER(obj) \
- colo_filter_rewriter_debug(void) ""
+     OBJECT_CHECK(RewriterState, (obj), TYPE_FILTER_REWRITER)
  #define TYPE_FILTER_REWRITER "filter-rewriter"
 +#define FAILOVER_MODE_ON  true
 +#define FAILOVER_MODE_OFF false
  typedef struct RewriterState {
      NetFilterState parent_obj;
@@ -XXX,XX +XXX,XX @@ typedef struct RewriterState {
      /* hashtable to save connection */
      GHashTable *connection_track_table;
      bool vnet_hdr;
 +    bool failover_mode;
  } RewriterState;
 +static void filter_rewriter_failover_mode(RewriterState *s)
 +{
 +    s->failover_mode = FAILOVER_MODE_ON;
 +}
 +
  static void filter_rewriter_flush(NetFilterState *nf)
  {
      RewriterState *s = FILTER_COLO_REWRITER(nf);
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
               */
              reverse_connection_key(&key);
          }
 +
 +        /* After failover we needn't change new TCP packet */
 +        if (s->failover_mode &&
 +            !connection_has_tracked(s->connection_track_table, &key)) {
 +            goto out;
 +        }
 +
          conn = connection_get(s->connection_track_table,
                                &key,
                                NULL);
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
          }
      }
 +out:
      packet_destroy(pkt, NULL);
      pkt = NULL;
      return 0;
  }
 +static void reset_seq_offset(gpointer key, gpointer value, gpointer user_data)
 +{
 +    Connection *conn = (Connection *)value;
 +
 +    conn->offset = 0;
 +}
 +
 +static gboolean offset_is_nonzero(gpointer key,
 +                                  gpointer value,
 +                                  gpointer user_data)
 +{
 +    Connection *conn = (Connection *)value;
 +
 +    return conn->offset ? true : false;
 +}
 +
 +static void colo_rewriter_handle_event(NetFilterState *nf, int event,
 +                                       Error **errp)
 +{
 +    RewriterState *rs = FILTER_COLO_REWRITER(nf);
 +
 +    switch (event) {
 +    case COLO_EVENT_CHECKPOINT:
 +        g_hash_table_foreach(rs->connection_track_table,
 +                            reset_seq_offset, NULL);
 +        break;
 +    case COLO_EVENT_FAILOVER:
 +        if (!g_hash_table_find(rs->connection_track_table,
 +                              offset_is_nonzero, NULL)) {
 +            filter_rewriter_failover_mode(rs);
 +        }
 +        break;
 +    default:
 +        break;
 +    }
 +}
 +
  static void colo_rewriter_cleanup(NetFilterState *nf)
  {
      RewriterState *s = FILTER_COLO_REWRITER(nf);
@@ -XXX,XX +XXX,XX @@ static void filter_rewriter_init(Object *obj)
      RewriterState *s = FILTER_COLO_REWRITER(obj);
      s->vnet_hdr = false;
 +    s->failover_mode = FAILOVER_MODE_OFF;
      object_property_add_bool(obj, "vnet_hdr_support",
                               filter_rewriter_get_vnet_hdr,
                               filter_rewriter_set_vnet_hdr, NULL);
@@ -XXX,XX +XXX,XX @@ static void colo_rewriter_class_init(ObjectClass *oc, void *data)
      nfc->setup = colo_rewriter_setup;
      nfc->cleanup = colo_rewriter_cleanup;
      nfc->receive_iov = colo_rewriter_receive_iov;
 +    nfc->handle_event = colo_rewriter_handle_event;
  }
  static const TypeInfo colo_rewriter_info = {
 --
-.7.4
+.5.0

-New patch
+[Qemu-devel] [PULL V2 17/26] COLO: notify net filters about checkpoint/failover event
+From: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Notify all net filters about the checkpoint and failover event.
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/colo.c | 15 +++++++++++++++
+file changed, 15 insertions(+)
+diff --git a/migration/colo.c b/migration/colo.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo.c
++++ b/migration/colo.c
+@@ -XXX,XX +XXX,XX @@
+ #include "qapi/qapi-events-migration.h"
+ #include "qapi/qmp/qerror.h"
+ #include "sysemu/cpus.h"
++#include "net/filter.h"
+ static bool vmstate_loading;
+ static Notifier packets_compare_notifier;
+@@ -XXX,XX +XXX,XX @@ static void secondary_vm_do_failover(void)
+         error_report_err(local_err);
+     }
++    /* Notify all filters of all NIC to do checkpoint */
++    colo_notify_filters_event(COLO_EVENT_FAILOVER, &local_err);
++    if (local_err) {
++        error_report_err(local_err);
++    }
++
+     if (!autostart) {
+         error_report("\"-S\" qemu option will be ignored in secondary side");
+         /* recover runstate to normal migration finish state */
+@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
+             goto out;
+         }
++        /* Notify all filters of all NIC to do checkpoint */
++        colo_notify_filters_event(COLO_EVENT_CHECKPOINT, &local_err);
++
++        if (local_err) {
++            qemu_mutex_unlock_iothread();
++            goto out;
++        }
++
+         vmstate_loading = false;
+         vm_start();
+         trace_colo_vm_state_change("stop", "run");
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 18/26] COLO: quick failover process by kick COLO thread
+From: zhanghailiang <zhang.zhanghailiang@huawei.com>
+COLO thread may sleep at qemu_sem_wait(&s->colo_checkpoint_sem),
+while failover works begin, It's better to wakeup it to quick
+the process.
+Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
+Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ migration/colo.c | 8 ++++++++
+file changed, 8 insertions(+)
+diff --git a/migration/colo.c b/migration/colo.c
+index XXXXXXX..XXXXXXX 100644
+--- a/migration/colo.c
++++ b/migration/colo.c
+@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
+     migrate_set_state(&s->state, MIGRATION_STATUS_COLO,
+                       MIGRATION_STATUS_COMPLETED);
++    /*
++     * kick COLO thread which might wait at
++     * qemu_sem_wait(&s->colo_checkpoint_sem).
++     */
++    colo_checkpoint_notify(migrate_get_current());
+     /*
+      * Wake up COLO thread which may blocked in recv() or send(),
+@@ -XXX,XX +XXX,XX @@ static void colo_process_checkpoint(MigrationState *s)
+         qemu_sem_wait(&s->colo_checkpoint_sem);
++        if (s->state != MIGRATION_STATUS_COLO) {
++            goto out;
++        }
+         ret = colo_do_checkpoint_transaction(s, bioc, fb);
+         if (ret < 0) {
+             goto out;
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 19/26] docs: Add COLO status diagram to COLO-FT.txt
+From: Zhang Chen <chen.zhang@intel.com>
+This diagram make user better understand COLO.
+Suggested by Markus Armbruster.
+Signed-off-by: Zhang Chen <zhangckid@gmail.com>
+Signed-off-by: Zhang Chen <chen.zhang@intel.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ docs/COLO-FT.txt | 34 ++++++++++++++++++++++++++++++++++
+file changed, 34 insertions(+)
+diff --git a/docs/COLO-FT.txt b/docs/COLO-FT.txt
+index XXXXXXX..XXXXXXX 100644
+--- a/docs/COLO-FT.txt
++++ b/docs/COLO-FT.txt
+@@ -XXX,XX +XXX,XX @@ Note:
+ HeartBeat has not been implemented yet, so you need to trigger failover process
+ by using 'x-colo-lost-heartbeat' command.
++== COLO operation status ==
++
+++-----------------+
++|                 |
++|    Start COLO   |
++|                 |
+++--------+--------+
++         |
++         |  Main qmp command:
++         |  migrate-set-capabilities with x-colo
++         |  migrate
++         |
++         v
+++--------+--------+
++|                 |
++|  COLO running   |
++|                 |
+++--------+--------+
++         |
++         |  Main qmp command:
++         |  x-colo-lost-heartbeat
++         |  or
++         |  some error happened
++         v
+++--------+--------+
++|                 |  send qmp event:
++|  COLO failover  |  COLO_EXIT
++|                 |
+++-----------------+
++
++COLO use the qmp command to switch and report operation status.
++The diagram just shows the main qmp command, you can get the detail
++in test procedure.
++
+ == Test procedure ==
+. Startup qemu
+ Primary:
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 20/26] clean up callback when del virtqueue
+From: liujunjie <liujunjie23@huawei.com>
+Before, we did not clear callback like handle_output when delete
+the virtqueue which may result be segmentfault.
+The scene is as follows:
+. Start a vm with multiqueue vhost-net,
+. then we write VIRTIO_PCI_GUEST_FEATURES in PCI configuration to
+triger multiqueue disable in this vm which will delete the virtqueue.
+In this step, the tx_bh is deleted but the callback virtio_net_handle_tx_bh
+still exist.
+. Finally, we write VIRTIO_PCI_QUEUE_NOTIFY in PCI configuration to
+notify the deleted virtqueue. In this way, virtio_net_handle_tx_bh
+will be called and qemu will be crashed.
+Although the way described above is uncommon, we had better reinforce it.
+CC: qemu-stable@nongnu.org
+Signed-off-by: liujunjie <liujunjie23@huawei.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ hw/virtio/virtio.c | 2 ++
+file changed, 2 insertions(+)
+diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/virtio/virtio.c
++++ b/hw/virtio/virtio.c
+@@ -XXX,XX +XXX,XX @@ void virtio_del_queue(VirtIODevice *vdev, int n)
+     vdev->vq[n].vring.num = 0;
+     vdev->vq[n].vring.num_default = 0;
++    vdev->vq[n].handle_output = NULL;
++    vdev->vq[n].handle_aio_output = NULL;
+ }
+ static void virtio_set_isr(VirtIODevice *vdev, int value)
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 21/26] ne2000: fix possible out of bound access in ne2000_receive
+In ne2000_receive(), we try to assign size_ to size which converts
+from size_t to integer. This will cause troubles when size_ is greater
+INT_MAX, this will lead a negative value in size and it can then pass
+the check of size < MIN_BUF_SIZE which may lead out of bound access of
+for both buf and buf1.
+Fixing by converting the type of size to size_t.
+CC: qemu-stable@nongnu.org
+Reported-by: Daniel Shapira <daniel@twistlock.com>
+Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ hw/net/ne2000.c | 4 ++--
+file changed, 2 insertions(+), 2 deletions(-)
+diff --git a/hw/net/ne2000.c b/hw/net/ne2000.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/net/ne2000.c
++++ b/hw/net/ne2000.c
+@@ -XXX,XX +XXX,XX @@ static int ne2000_buffer_full(NE2000State *s)
+ ssize_t ne2000_receive(NetClientState *nc, const uint8_t *buf, size_t size_)
+ {
+     NE2000State *s = qemu_get_nic_opaque(nc);
+-    int size = size_;
++    size_t size = size_;
+     uint8_t *p;
+     unsigned int total_len, next, avail, len, index, mcast_idx;
+     uint8_t buf1[60];
+@@ -XXX,XX +XXX,XX @@ ssize_t ne2000_receive(NetClientState *nc, const uint8_t *buf, size_t size_)
+         { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
+ #if defined(DEBUG_NE2000)
+-    printf("NE2000: received len=%d\n", size);
++    printf("NE2000: received len=%zu\n", size);
+ #endif
+     if (s->cmd & E8390_STOP || ne2000_buffer_full(s))
+--
+.5.0

-[Qemu-devel] [PULL 6/6] MAINTAINERS: update Dmitry Fleytman email
+[Qemu-devel] [PULL V2 22/26] rtl8139: fix possible out of bound access
-From: Philippe Mathieu-Daudé <f4bug@amsat.org>
+In rtl8139_do_receive(), we try to assign size_ to size which converts
 from size_t to integer. This will cause troubles when size_ is greater
 INT_MAX, this will lead a negative value in size and it can then pass
 the check of size < MIN_BUF_SIZE which may lead out of bound access of
 for both buf and buf1.
-gently asked by his automatic reply :)
+Fixing by converting the type of size to size_t.
-Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
+CC: qemu-stable@nongnu.org
 Reported-by: Daniel Shapira <daniel@twistlock.com>
 Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
- MAINTAINERS | 8 ++++----
+ hw/net/rtl8139.c | 8 ++++----
 file changed, 4 insertions(+), 4 deletions(-)
-diff --git a/MAINTAINERS b/MAINTAINERS
+diff --git a/hw/net/rtl8139.c b/hw/net/rtl8139.c
 index XXXXXXX..XXXXXXX 100644
---- a/MAINTAINERS
+--- a/hw/net/rtl8139.c
-+++ b/MAINTAINERS
++++ b/hw/net/rtl8139.c
-@@ -XXX,XX +XXX,XX @@ F: hw/scsi/mfi.h
+@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
- F: tests/megasas-test.c
+     RTL8139State *s = qemu_get_nic_opaque(nc);
+     PCIDevice *d = PCI_DEVICE(s);
- Network packet abstractions
+     /* size is the length of the buffer passed to the driver */
--M: Dmitry Fleytman <dmitry@daynix.com>
+-    int size = size_;
-+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
++    size_t size = size_;
- S: Maintained
+     const uint8_t *dot1q_buf = NULL;
- F: include/net/eth.h
- F: net/eth.c
+     uint32_t packet_header = 0;
-@@ -XXX,XX +XXX,XX @@ F: hw/net/net_rx_pkt*
+@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
- F: hw/net/net_tx_pkt*
+     static const uint8_t broadcast_macaddr[6] =
+         { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
- Vmware
--M: Dmitry Fleytman <dmitry@daynix.com>
+-    DPRINTF(">>> received len=%d\n", size);
-+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
++    DPRINTF(">>> received len=%zu\n", size);
- S: Maintained
- F: hw/net/vmxnet*
+     /* test if board clock is stopped */
- F: hw/scsi/vmw_pvscsi*
+     if (!s->clock_enabled)
-@@ -XXX,XX +XXX,XX @@ F: hw/mem/nvdimm.c
+@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
- F: include/hw/mem/nvdimm.h
+         if (size+4 > rx_space)
- e1000x
+         {
--M: Dmitry Fleytman <dmitry@daynix.com>
+-            DPRINTF("C+ Rx mode : descriptor %d size %d received %d + 4\n",
-+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
++            DPRINTF("C+ Rx mode : descriptor %d size %d received %zu + 4\n",
- S: Maintained
+                 descriptor, rx_space, size);
- F: hw/net/e1000x*
+             s->IntrStatus |= RxOverflow;
- e1000e
+@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
--M: Dmitry Fleytman <dmitry@daynix.com>
+         if (avail != 0 && RX_ALIGN(size + 8) >= avail)
-+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
+         {
- S: Maintained
+             DPRINTF("rx overflow: rx buffer length %d head 0x%04x "
- F: hw/net/e1000e*
+-                "read 0x%04x === available 0x%04x need 0x%04x\n",
++                "read 0x%04x === available 0x%04x need 0x%04zx\n",
                  s->RxBufferSize, s->RxBufAddr, s->RxBufPtr, avail, size + 8);
              s->IntrStatus |= RxOverflow;
 --
-.7.4
+.5.0

-New patch
+[Qemu-devel] [PULL V2 23/26] pcnet: fix possible buffer overflow
+In pcnet_receive(), we try to assign size_ to size which converts from
+size_t to integer. This will cause troubles when size_ is greater
+INT_MAX, this will lead a negative value in size and it can then pass
+the check of size < MIN_BUF_SIZE which may lead out of bound access
+for both buf and buf1.
+Fixing by converting the type of size to size_t.
+CC: qemu-stable@nongnu.org
+Reported-by: Daniel Shapira <daniel@twistlock.com>
+Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ hw/net/pcnet.c | 4 ++--
+file changed, 2 insertions(+), 2 deletions(-)
+diff --git a/hw/net/pcnet.c b/hw/net/pcnet.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/net/pcnet.c
++++ b/hw/net/pcnet.c
+@@ -XXX,XX +XXX,XX @@ ssize_t pcnet_receive(NetClientState *nc, const uint8_t *buf, size_t size_)
+     uint8_t buf1[60];
+     int remaining;
+     int crc_err = 0;
+-    int size = size_;
++    size_t size = size_;
+     if (CSR_DRX(s) || CSR_STOP(s) || CSR_SPND(s) || !size ||
+         (CSR_LOOP(s) && !s->looptest)) {
+         return -1;
+     }
+ #ifdef PCNET_DEBUG
+-    printf("pcnet_receive size=%d\n", size);
++    printf("pcnet_receive size=%zu\n", size);
+ #endif
+     /* if too small buffer, then expand it */
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 24/26] net: ignore packet size greater than INT_MAX
+There should not be a reason for passing a packet size greater than
+INT_MAX. It's usually a hint of bug somewhere, so ignore packet size
+greater than INT_MAX in qemu_deliver_packet_iov()
+CC: qemu-stable@nongnu.org
+Reported-by: Daniel Shapira <daniel@twistlock.com>
+Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ net/net.c | 7 ++++++-
+file changed, 6 insertions(+), 1 deletion(-)
+diff --git a/net/net.c b/net/net.c
+index XXXXXXX..XXXXXXX 100644
+--- a/net/net.c
++++ b/net/net.c
+@@ -XXX,XX +XXX,XX @@ ssize_t qemu_deliver_packet_iov(NetClientState *sender,
+                                 void *opaque)
+ {
+     NetClientState *nc = opaque;
++    size_t size = iov_size(iov, iovcnt);
+     int ret;
++    if (size > INT_MAX) {
++        return size;
++    }
++
+     if (nc->link_down) {
+-        return iov_size(iov, iovcnt);
++        return size;
+     }
+     if (nc->receive_disabled) {
+--
+.5.0

-New patch
+[Qemu-devel] [PULL V2 25/26] e1000: indicate dropped packets in HW counters
+The e1000 emulation silently discards RX packets if there's
+insufficient space in the ring buffer. This leads to errors
+on higher-level protocols in the guest, with no indication
+about the error cause.
+This patch increments the "Missed Packets Count" (MPC) and
+"Receive No Buffers Count" (RNBC) HW counters in this case.
+As the emulation has no FIFO for buffering packets that can't
+immediately be pushed to the guest, these two registers are
+practically equivalent (see 10.2.7.4, 10.2.7.33 in
+https://www.intel.com/content/www/us/en/embedded/products/networking/82574l-gbe-controller-datasheet.html).
+On a Linux guest, the register content  will be reflected in
+the "rx_missed_errors" and "rx_no_buffer_count" stats from
+"ethtool -S", and in the "missed" stat from "ip -s -s link show",
+giving at least some hint about the error cause inside the guest.
+If the cause is known, problems like this can often be avoided
+easily, by increasing the number of RX descriptors in the guest
+e1000 driver (e.g under Linux, "e1000.RxDescriptors=1024").
+The patch also adds a qemu trace message for this condition.
+Signed-off-by: Martin Wilck <mwilck@suse.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+---
+ hw/net/e1000.c      | 16 +++++++++++++---
+ hw/net/trace-events |  3 +++
+files changed, 16 insertions(+), 3 deletions(-)
+diff --git a/hw/net/e1000.c b/hw/net/e1000.c
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/net/e1000.c
++++ b/hw/net/e1000.c
+@@ -XXX,XX +XXX,XX @@
+ #include "qemu/range.h"
+ #include "e1000x_common.h"
++#include "trace.h"
+ static const uint8_t bcast[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
+@@ -XXX,XX +XXX,XX @@ static uint64_t rx_desc_base(E1000State *s)
+     return (bah << 32) + bal;
+ }
++static void
++e1000_receiver_overrun(E1000State *s, size_t size)
++{
++    trace_e1000_receiver_overrun(size, s->mac_reg[RDH], s->mac_reg[RDT]);
++    e1000x_inc_reg_if_not_full(s->mac_reg, RNBC);
++    e1000x_inc_reg_if_not_full(s->mac_reg, MPC);
++    set_ics(s, 0, E1000_ICS_RXO);
++}
++
+ static ssize_t
+ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
+ {
+@@ -XXX,XX +XXX,XX @@ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
+     desc_offset = 0;
+     total_size = size + e1000x_fcs_len(s->mac_reg);
+     if (!e1000_has_rxbufs(s, total_size)) {
+-            set_ics(s, 0, E1000_ICS_RXO);
+-            return -1;
++        e1000_receiver_overrun(s, total_size);
++        return -1;
+     }
+     do {
+         desc_size = total_size - desc_offset;
+@@ -XXX,XX +XXX,XX @@ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
+             rdh_start >= s->mac_reg[RDLEN] / sizeof(desc)) {
+             DBGOUT(RXERR, "RDH wraparound @%x, RDT %x, RDLEN %x\n",
+                    rdh_start, s->mac_reg[RDT], s->mac_reg[RDLEN]);
+-            set_ics(s, 0, E1000_ICS_RXO);
++            e1000_receiver_overrun(s, total_size);
+             return -1;
+         }
+     } while (desc_offset < total_size);
+diff --git a/hw/net/trace-events b/hw/net/trace-events
+index XXXXXXX..XXXXXXX 100644
+--- a/hw/net/trace-events
++++ b/hw/net/trace-events
+@@ -XXX,XX +XXX,XX @@ net_rx_pkt_rss_ip6_ex(void) "Calculating IPv6/EX RSS  hash"
+ net_rx_pkt_rss_hash(size_t rss_length, uint32_t rss_hash) "RSS hash for %zu bytes: 0x%X"
+ net_rx_pkt_rss_add_chunk(void* ptr, size_t size, size_t input_offset) "Add RSS chunk %p, %zu bytes, RSS input offset %zu bytes"
++# hw/net/e1000.c
++e1000_receiver_overrun(size_t s, uint32_t rdh, uint32_t rdt) "Receiver overrun: dropped packet of %zu bytes, RDH=%u, RDT=%u"
++
+ # hw/net/e1000x_common.c
+ e1000x_rx_can_recv_disabled(bool link_up, bool rx_enabled, bool pci_master) "link_up: %d, rx_enabled %d, pci_master %d"
+ e1000x_vlan_is_vlan_pkt(bool is_vlan_pkt, uint16_t eth_proto, uint16_t vet) "Is VLAN packet: %d, ETH proto: 0x%X, VET: 0x%X"
+--
+.5.0

-[Qemu-devel] [PULL 5/6] qemu-doc: Get rid of "vlan=X" example in the documentation
+[Qemu-devel] [PULL V2 26/26] qemu-options: Fix bad "macaddr" property in the documentation
 From: Thomas Huth <thuth@redhat.com>
-The vlan concept is marked as deprecated, so we should not use
+When using the "-device" option, the property is called "mac".
-this for examples in the documentation anymore.
+"macaddr" is only used for the legacy "-net nic" option.
+Reported-by: Harald Hoyer <harald@redhat.com>
+Reviewed-by: Markus Armbruster <armbru@redhat.com>
 Signed-off-by: Thomas Huth <thuth@redhat.com>
 Signed-off-by: Jason Wang <jasowang@redhat.com>
 ---
- qemu-options.hx | 4 ++--
+ qemu-options.hx | 2 +-
-file changed, 2 insertions(+), 2 deletions(-)
+file changed, 1 insertion(+), 1 deletion(-)
 diff --git a/qemu-options.hx b/qemu-options.hx
 index XXXXXXX..XXXXXXX 100644
 --- a/qemu-options.hx
 +++ b/qemu-options.hx
-@@ -XXX,XX +XXX,XX @@ qemu-system-i386 linux.img -net nic -net tap
+@@ -XXX,XX +XXX,XX @@ qemu-system-i386 linux.img \
- #launch a QEMU instance with two NICs, each one connected
+                  -netdev socket,id=n2,mcast=230.0.0.1:1234
- #to a TAP device
+ # launch yet another QEMU instance on same "bus"
  qemu-system-i386 linux.img \
--                 -net nic,vlan=0 -net tap,vlan=0,ifname=tap0 \
+-                 -device e1000,netdev=n3,macaddr=52:54:00:12:34:58 \
--                 -net nic,vlan=1 -net tap,vlan=1,ifname=tap1
++                 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \
-+        -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \
+                  -netdev socket,id=n3,mcast=230.0.0.1:1234
 +        -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1
  @end example
- @example
 --
-.7.4
+.5.0

The following changes since commit e607bbee553cfe73072870cef458cfa4e78133e2:

Merge remote-tracking branch 'remotes/edgar/tags/edgar/xilinx-next-2018-01-26.for-upstream' into staging (2018-01-26 14:24:25 +0000)

are available in the git repository at:

https://github.com/jasowang/qemu.git tags/net-pull-request

for you to fetch changes up to bf4835a4d5338bb7424827715df22570a8adc67c:

MAINTAINERS: update Dmitry Fleytman email (2018-01-29 16:05:38 +0800)

----------------------------------------------------------------

----------------------------------------------------------------
Mao Zhongyi (2):
      colo: modified the payload compare function
      colo: compare the packet based on the tcp sequence number

Philippe Mathieu-Daudé (1):
      MAINTAINERS: update Dmitry Fleytman email

Thomas Huth (3):
      net: Allow hubports to connect to other netdevs
      net: Allow netdevs to be used with 'hostfwd_add' and 'hostfwd_remove'
      qemu-doc: Get rid of "vlan=X" example in the documentation

From: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>

Modified the function colo_packet_compare_common to prepare for the
tcp packet comparison in the next patch.

Cc: Zhang Chen <zhangckid@gmail.com>
Cc: Li Zhijian <lizhijian@cn.fujitsu.com>
Cc: Jason Wang <jasowang@redhat.com>

Signed-off-by: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Reviewed-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/colo-compare.c | 88 +++++++++++++++++++++++++++---------------------------
 1 file changed, 44 insertions(+), 44 deletions(-)

diff --git a/net/colo-compare.c b/net/colo-compare.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -XXX,XX +XXX,XX @@ static int packet_enqueue(CompareState *s, int mode, Connection **con)
  * return:    0  means packet same
  *            > 0 || < 0 means packet different
  */
-static int colo_packet_compare_common(Packet *ppkt,
-                                      Packet *spkt,
-                                      int poffset,
-                                      int soffset)
+static int colo_compare_packet_payload(Packet *ppkt,
+                                       Packet *spkt,
+                                       uint16_t poffset,
+                                       uint16_t soffset,
+                                       uint16_t len)
+
 {
     if (trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
         char pri_ip_src[20], pri_ip_dst[20], sec_ip_src[20], sec_ip_dst[20];
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_common(Packet *ppkt,
                                    sec_ip_src, sec_ip_dst);
     }
 
-    poffset = ppkt->vnet_hdr_len + poffset;
-    soffset = ppkt->vnet_hdr_len + soffset;
-
-    if (ppkt->size - poffset == spkt->size - soffset) {
-        return memcmp(ppkt->data + poffset,
-                      spkt->data + soffset,
-                      spkt->size - soffset);
-    } else {
-        trace_colo_compare_main("Net packet size are not the same");
-        return -1;
-    }
+    return memcmp(ppkt->data + poffset, spkt->data + soffset, len);
 }
 
 /*
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_tcp(Packet *spkt, Packet *ppkt)
      * the secondary guest's timestamp. COLO just focus on payload,
      * so we just need skip this field.
      */
-    if (ptcp->th_off > 5) {
-        ptrdiff_t ptcp_offset, stcp_offset;
 
-        ptcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
-                      + (ptcp->th_off * 4) - ppkt->vnet_hdr_len;
-        stcp_offset = spkt->transport_header - (uint8_t *)spkt->data
-                      + (stcp->th_off * 4) - spkt->vnet_hdr_len;
+    ptrdiff_t ptcp_offset, stcp_offset;
 
-        /*
-         * When network is busy, some tcp options(like sack) will unpredictable
-         * occur in primary side or secondary side. it will make packet size
-         * not same, but the two packet's payload is identical. colo just
-         * care about packet payload, so we skip the option field.
-         */
-        res = colo_packet_compare_common(ppkt, spkt, ptcp_offset, stcp_offset);
-    } else if (ptcp->th_sum == stcp->th_sum) {
-        res = colo_packet_compare_common(ppkt, spkt, ETH_HLEN, ETH_HLEN);
+    ptcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
+                  + (ptcp->th_off << 2) - ppkt->vnet_hdr_len;
+    stcp_offset = spkt->transport_header - (uint8_t *)spkt->data
+                  + (stcp->th_off << 2) - spkt->vnet_hdr_len;
+    if (ppkt->size - ptcp_offset == spkt->size - stcp_offset) {
+        res = colo_compare_packet_payload(ppkt, spkt,
+                                          ptcp_offset, stcp_offset,
+                                          ppkt->size - ptcp_offset);
     } else {
+        trace_colo_compare_main("TCP: payload size of packets are different");
         res = -1;
     }
 
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_tcp(Packet *spkt, Packet *ppkt)
  */
 static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
 {
-    int ret;
-    int network_header_length = ppkt->ip->ip_hl * 4;
+    uint16_t network_header_length = ppkt->ip->ip_hl << 2;
+    uint16_t offset = network_header_length + ETH_HLEN + ppkt->vnet_hdr_len;
 
     trace_colo_compare_main("compare udp");
 
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
      * other field like TOS,TTL,IP Checksum. we only need to compare
      * the ip payload here.
      */
-    ret = colo_packet_compare_common(ppkt, spkt,
-                                     network_header_length + ETH_HLEN,
-                                     network_header_length + ETH_HLEN);
-
-    if (ret) {
+    if (ppkt->size != spkt->size) {
+        trace_colo_compare_main("UDP: payload size of packets are different");
+        return -1;
+    }
+    if (colo_compare_packet_payload(ppkt, spkt, offset, offset,
+                                    ppkt->size - offset)) {
         trace_colo_compare_udp_miscompare("primary pkt size", ppkt->size);
         trace_colo_compare_udp_miscompare("Secondary pkt size", spkt->size);
         if (trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
             qemu_hexdump((char *)spkt->data, stderr, "colo-compare sec pkt",
                          spkt->size);
         }
+        return -1;
+    } else {
+        return 0;
     }
-
-    return ret;
 }
 
 /*
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
  */
 static int colo_packet_compare_icmp(Packet *spkt, Packet *ppkt)
 {
-    int network_header_length = ppkt->ip->ip_hl * 4;
+    uint16_t network_header_length = ppkt->ip->ip_hl << 2;
+    uint16_t offset = network_header_length + ETH_HLEN + ppkt->vnet_hdr_len;
 
     trace_colo_compare_main("compare icmp");
 
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_icmp(Packet *spkt, Packet *ppkt)
      * other field like TOS,TTL,IP Checksum. we only need to compare
      * the ip payload here.
      */
-    if (colo_packet_compare_common(ppkt, spkt,
-                                   network_header_length + ETH_HLEN,
-                                   network_header_length + ETH_HLEN)) {
+    if (ppkt->size != spkt->size) {
+        trace_colo_compare_main("ICMP: payload size of packets are different");
+        return -1;
+    }
+    if (colo_compare_packet_payload(ppkt, spkt, offset, offset,
+                                    ppkt->size - offset)) {
         trace_colo_compare_icmp_miscompare("primary pkt size",
                                            ppkt->size);
         trace_colo_compare_icmp_miscompare("Secondary pkt size",
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_icmp(Packet *spkt, Packet *ppkt)
  */
 static int colo_packet_compare_other(Packet *spkt, Packet *ppkt)
 {
+    uint16_t offset = ppkt->vnet_hdr_len;
+
     trace_colo_compare_main("compare other");
     if (trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
         char pri_ip_src[20], pri_ip_dst[20], sec_ip_src[20], sec_ip_dst[20];
@@ -XXX,XX +XXX,XX @@ static int colo_packet_compare_other(Packet *spkt, Packet *ppkt)
                                    sec_ip_src, sec_ip_dst);
     }
 
-    return colo_packet_compare_common(ppkt, spkt, 0, 0);
+    if (ppkt->size != spkt->size) {
+        trace_colo_compare_main("Other: payload size of packets are different");
+        return -1;
+    }
+    return colo_compare_packet_payload(ppkt, spkt, offset, offset,
+                                       ppkt->size - offset);
 }
 
 static int colo_old_packet_check_one(Packet *pkt, int64_t *check_time)
-- 
2.7.4

From: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>

Packet size some time different or when network is busy.
Based on same payload size, but TCP protocol can not
guarantee send the same one packet in the same way,

like that:
We send this payload:
------------------------------
| header |1|2|3|4|5|6|7|8|9|0|
------------------------------

primary:
ppkt1:
----------------
| header |1|2|3|
----------------
ppkt2:
------------------------
| header |4|5|6|7|8|9|0|
------------------------

secondary:
spkt1:
------------------------------
| header |1|2|3|4|5|6|7|8|9|0|
------------------------------

In the original method, ppkt1 and ppkt2 are different in size and
spkt1, so they can't compare and trigger the checkpoint.

I have tested FTP get 200M and 1G file many times, I found that
the performance was less than 1% of the native.

Now I reconstructed the comparison of TCP packets based on the
TCP sequence number. first of all, ppkt1 and spkt1 have the same
starting sequence number, so they can compare, even though their
length is different. And then ppkt1 with a smaller payload length
is used as the comparison length, if the payload is same, send
out the ppkt1 and record the offset(the length of ppkt1 payload)
in spkt1. The next comparison, ppkt2 and spkt1 can be compared
from the recorded position of spkt1.

like that:
----------------
| header |1|2|3| ppkt1
---------|-----|
         |     |
---------v-----v--------------
| header |1|2|3|4|5|6|7|8|9|0| spkt1
---------------|\------------|
               | \offset     |
      ---------v-------------v
      | header |4|5|6|7|8|9|0| ppkt2
      ------------------------

In this way, the performance can reach native 20% in my multiple
tests.

Cc: Zhang Chen <zhangckid@gmail.com>
Cc: Li Zhijian <lizhijian@cn.fujitsu.com>
Cc: Jason Wang <jasowang@redhat.com>

Signed-off-by: Mao Zhongyi <maozy.fnst@cn.fujitsu.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Reviewed-by: Zhang Chen <zhangckid@gmail.com>
Tested-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/colo-compare.c | 343 +++++++++++++++++++++++++++++++++++------------------
 net/colo.c         |   9 ++
 net/colo.h         |  15 +++
 net/trace-events   |   2 +-
 4 files changed, 250 insertions(+), 119 deletions(-)

diff --git a/net/colo-compare.c b/net/colo-compare.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -XXX,XX +XXX,XX @@
 #define COMPARE_READ_LEN_MAX NET_BUFSIZE
 #define MAX_QUEUE_SIZE 1024
 
+#define COLO_COMPARE_FREE_PRIMARY     0x01
+#define COLO_COMPARE_FREE_SECONDARY   0x02
+
 /* TODO: Should be configurable */
 #define REGULAR_PACKET_CHECK_MS 3000
 
@@ -XXX,XX +XXX,XX @@ static gint seq_sorter(Packet *a, Packet *b, gpointer data)
     return ntohl(atcp->th_seq) - ntohl(btcp->th_seq);
 }
 
+static void fill_pkt_tcp_info(void *data, uint32_t *max_ack)
+{
+    Packet *pkt = data;
+    struct tcphdr *tcphd;
+
+    tcphd = (struct tcphdr *)pkt->transport_header;
+
+    pkt->tcp_seq = ntohl(tcphd->th_seq);
+    pkt->tcp_ack = ntohl(tcphd->th_ack);
+    *max_ack = *max_ack > pkt->tcp_ack ? *max_ack : pkt->tcp_ack;
+    pkt->header_size = pkt->transport_header - (uint8_t *)pkt->data
+                       + (tcphd->th_off << 2) - pkt->vnet_hdr_len;
+    pkt->payload_size = pkt->size - pkt->header_size;
+    pkt->seq_end = pkt->tcp_seq + pkt->payload_size;
+    pkt->flags = tcphd->th_flags;
+}
+
 /*
  * Return 1 on success, if return 0 means the
  * packet will be dropped
  */
-static int colo_insert_packet(GQueue *queue, Packet *pkt)
+static int colo_insert_packet(GQueue *queue, Packet *pkt, uint32_t *max_ack)
 {
     if (g_queue_get_length(queue) <= MAX_QUEUE_SIZE) {
         if (pkt->ip->ip_p == IPPROTO_TCP) {
+            fill_pkt_tcp_info(pkt, max_ack);
             g_queue_insert_sorted(queue,
                                   pkt,
                                   (GCompareDataFunc)seq_sorter,
@@ -XXX,XX +XXX,XX @@ static int packet_enqueue(CompareState *s, int mode, Connection **con)
     }
 
     if (mode == PRIMARY_IN) {
-        if (!colo_insert_packet(&conn->primary_list, pkt)) {
+        if (!colo_insert_packet(&conn->primary_list, pkt, &conn->pack)) {
             error_report("colo compare primary queue size too big,"
                          "drop packet");
         }
     } else {
-        if (!colo_insert_packet(&conn->secondary_list, pkt)) {
+        if (!colo_insert_packet(&conn->secondary_list, pkt, &conn->sack)) {
             error_report("colo compare secondary queue size too big,"
                          "drop packet");
         }
@@ -XXX,XX +XXX,XX @@ static int packet_enqueue(CompareState *s, int mode, Connection **con)
     return 0;
 }
 
+static inline bool after(uint32_t seq1, uint32_t seq2)
+{
+        return (int32_t)(seq1 - seq2) > 0;
+}
+
+static void colo_release_primary_pkt(CompareState *s, Packet *pkt)
+{
+    int ret;
+    ret = compare_chr_send(s,
+                           pkt->data,
+                           pkt->size,
+                           pkt->vnet_hdr_len);
+    if (ret < 0) {
+        error_report("colo send primary packet failed");
+    }
+    trace_colo_compare_main("packet same and release packet");
+    packet_destroy(pkt, NULL);
+}
+
 /*
  * The IP packets sent by primary and secondary
  * will be compared in here
@@ -XXX,XX +XXX,XX @@ static int colo_compare_packet_payload(Packet *ppkt,
 }
 
 /*
- * Called from the compare thread on the primary
- * for compare tcp packet
- * compare_tcp copied from Dr. David Alan Gilbert's branch
- */
-static int colo_packet_compare_tcp(Packet *spkt, Packet *ppkt)
+ * return true means that the payload is consist and
+ * need to make the next comparison, false means do
+ * the checkpoint
+*/
+static bool colo_mark_tcp_pkt(Packet *ppkt, Packet *spkt,
+                              int8_t *mark, uint32_t max_ack)
 {
-    struct tcphdr *ptcp, *stcp;
-    int res;
+    *mark = 0;
+
+    if (ppkt->tcp_seq == spkt->tcp_seq && ppkt->seq_end == spkt->seq_end) {
+        if (colo_compare_packet_payload(ppkt, spkt,
+                                        ppkt->header_size, spkt->header_size,
+                                        ppkt->payload_size)) {
+            *mark = COLO_COMPARE_FREE_SECONDARY | COLO_COMPARE_FREE_PRIMARY;
+            return true;
+        }
+    }
+    if (ppkt->tcp_seq == spkt->tcp_seq && ppkt->seq_end == spkt->seq_end) {
+        if (colo_compare_packet_payload(ppkt, spkt,
+                                        ppkt->header_size, spkt->header_size,
+                                        ppkt->payload_size)) {
+            *mark = COLO_COMPARE_FREE_SECONDARY | COLO_COMPARE_FREE_PRIMARY;
+            return true;
+        }
+    }
+
+    /* one part of secondary packet payload still need to be compared */
+    if (!after(ppkt->seq_end, spkt->seq_end)) {
+        if (colo_compare_packet_payload(ppkt, spkt,
+                                        ppkt->header_size + ppkt->offset,
+                                        spkt->header_size + spkt->offset,
+                                        ppkt->payload_size - ppkt->offset)) {
+            if (!after(ppkt->tcp_ack, max_ack)) {
+                *mark = COLO_COMPARE_FREE_PRIMARY;
+                spkt->offset += ppkt->payload_size - ppkt->offset;
+                return true;
+            } else {
+                /* secondary guest hasn't ack the data, don't send
+                 * out this packet
+                 */
+                return false;
+            }
+        }
+    } else {
+        /* primary packet is longer than secondary packet, compare
+         * the same part and mark the primary packet offset
+         */
+        if (colo_compare_packet_payload(ppkt, spkt,
+                                        ppkt->header_size + ppkt->offset,
+                                        spkt->header_size + spkt->offset,
+                                        spkt->payload_size - spkt->offset)) {
+            *mark = COLO_COMPARE_FREE_SECONDARY;
+            ppkt->offset += spkt->payload_size - spkt->offset;
+            return true;
+        }
+    }
 
-    trace_colo_compare_main("compare tcp");
+    return false;
+}
 
-    ptcp = (struct tcphdr *)ppkt->transport_header;
-    stcp = (struct tcphdr *)spkt->transport_header;
+static void colo_compare_tcp(CompareState *s, Connection *conn)
+{
+    Packet *ppkt = NULL, *spkt = NULL;
+    int8_t mark;
 
     /*
-     * The 'identification' field in the IP header is *very* random
-     * it almost never matches.  Fudge this by ignoring differences in
-     * unfragmented packets; they'll normally sort themselves out if different
-     * anyway, and it should recover at the TCP level.
-     * An alternative would be to get both the primary and secondary to rewrite
-     * somehow; but that would need some sync traffic to sync the state
-     */
-    if (ntohs(ppkt->ip->ip_off) & IP_DF) {
-        spkt->ip->ip_id = ppkt->ip->ip_id;
-        /* and the sum will be different if the IDs were different */
-        spkt->ip->ip_sum = ppkt->ip->ip_sum;
+     * If ppkt and spkt have the same payload, but ppkt's ACK
+     * is greater than spkt's ACK, in this case we can not
+     * send the ppkt because it will cause the secondary guest
+     * to miss sending some data in the next. Therefore, we
+     * record the maximum ACK in the current queue at both
+     * primary side and secondary side. Only when the ack is
+     * less than the smaller of the two maximum ack, then we
+     * can ensure that the packet's payload is acknowledged by
+     * primary and secondary.
+    */
+    uint32_t min_ack = conn->pack > conn->sack ? conn->sack : conn->pack;
+
+pri:
+    if (g_queue_is_empty(&conn->primary_list)) {
+        return;
     }
+    ppkt = g_queue_pop_head(&conn->primary_list);
+sec:
+    if (g_queue_is_empty(&conn->secondary_list)) {
+        g_queue_push_head(&conn->primary_list, ppkt);
+        return;
+    }
+    spkt = g_queue_pop_head(&conn->secondary_list);
 
-    /*
-     * Check tcp header length for tcp option field.
-     * th_off > 5 means this tcp packet have options field.
-     * The tcp options maybe always different.
-     * for example:
-     * From RFC 7323.
-     * TCP Timestamps option (TSopt):
-     * Kind: 8
-     *
-     * Length: 10 bytes
-     *
-     *    +-------+-------+---------------------+---------------------+
-     *    |Kind=8 |  10   |   TS Value (TSval)  |TS Echo Reply (TSecr)|
-     *    +-------+-------+---------------------+---------------------+
-     *       1       1              4                     4
-     *
-     * In this case the primary guest's timestamp always different with
-     * the secondary guest's timestamp. COLO just focus on payload,
-     * so we just need skip this field.
-     */
+    if (ppkt->tcp_seq == ppkt->seq_end) {
+        colo_release_primary_pkt(s, ppkt);
+        ppkt = NULL;
+    }
 
-    ptrdiff_t ptcp_offset, stcp_offset;
+    if (ppkt && conn->compare_seq && !after(ppkt->seq_end, conn->compare_seq)) {
+        trace_colo_compare_main("pri: this packet has compared");
+        colo_release_primary_pkt(s, ppkt);
+        ppkt = NULL;
+    }
 
-    ptcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
-                  + (ptcp->th_off << 2) - ppkt->vnet_hdr_len;
-    stcp_offset = spkt->transport_header - (uint8_t *)spkt->data
-                  + (stcp->th_off << 2) - spkt->vnet_hdr_len;
-    if (ppkt->size - ptcp_offset == spkt->size - stcp_offset) {
-        res = colo_compare_packet_payload(ppkt, spkt,
-                                          ptcp_offset, stcp_offset,
-                                          ppkt->size - ptcp_offset);
+    if (spkt->tcp_seq == spkt->seq_end) {
+        packet_destroy(spkt, NULL);
+        if (!ppkt) {
+            goto pri;
+        } else {
+            goto sec;
+        }
     } else {
-        trace_colo_compare_main("TCP: payload size of packets are different");
-        res = -1;
+        if (conn->compare_seq && !after(spkt->seq_end, conn->compare_seq)) {
+            trace_colo_compare_main("sec: this packet has compared");
+            packet_destroy(spkt, NULL);
+            if (!ppkt) {
+                goto pri;
+            } else {
+                goto sec;
+            }
+        }
+        if (!ppkt) {
+            g_queue_push_head(&conn->secondary_list, spkt);
+            goto pri;
+        }
     }
 
-    if (res != 0 &&
-        trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
-        char pri_ip_src[20], pri_ip_dst[20], sec_ip_src[20], sec_ip_dst[20];
-
-        strcpy(pri_ip_src, inet_ntoa(ppkt->ip->ip_src));
-        strcpy(pri_ip_dst, inet_ntoa(ppkt->ip->ip_dst));
-        strcpy(sec_ip_src, inet_ntoa(spkt->ip->ip_src));
-        strcpy(sec_ip_dst, inet_ntoa(spkt->ip->ip_dst));
-
-        trace_colo_compare_ip_info(ppkt->size, pri_ip_src,
-                                   pri_ip_dst, spkt->size,
-                                   sec_ip_src, sec_ip_dst);
-
-        trace_colo_compare_tcp_info("pri tcp packet",
-                                    ntohl(ptcp->th_seq),
-                                    ntohl(ptcp->th_ack),
-                                    res, ptcp->th_flags,
-                                    ppkt->size);
-
-        trace_colo_compare_tcp_info("sec tcp packet",
-                                    ntohl(stcp->th_seq),
-                                    ntohl(stcp->th_ack),
-                                    res, stcp->th_flags,
-                                    spkt->size);
+    if (colo_mark_tcp_pkt(ppkt, spkt, &mark, min_ack)) {
+        trace_colo_compare_tcp_info("pri",
+                                    ppkt->tcp_seq, ppkt->tcp_ack,
+                                    ppkt->header_size, ppkt->payload_size,
+                                    ppkt->offset, ppkt->flags);
+
+        trace_colo_compare_tcp_info("sec",
+                                    spkt->tcp_seq, spkt->tcp_ack,
+                                    spkt->header_size, spkt->payload_size,
+                                    spkt->offset, spkt->flags);
+
+        if (mark == COLO_COMPARE_FREE_PRIMARY) {
+            conn->compare_seq = ppkt->seq_end;
+            colo_release_primary_pkt(s, ppkt);
+            g_queue_push_head(&conn->secondary_list, spkt);
+            goto pri;
+        }
+        if (mark == COLO_COMPARE_FREE_SECONDARY) {
+            conn->compare_seq = spkt->seq_end;
+            packet_destroy(spkt, NULL);
+            goto sec;
+        }
+        if (mark == (COLO_COMPARE_FREE_PRIMARY | COLO_COMPARE_FREE_SECONDARY)) {
+            conn->compare_seq = ppkt->seq_end;
+            colo_release_primary_pkt(s, ppkt);
+            packet_destroy(spkt, NULL);
+            goto pri;
+        }
+    } else {
+        g_queue_push_head(&conn->primary_list, ppkt);
+        g_queue_push_head(&conn->secondary_list, spkt);
 
         qemu_hexdump((char *)ppkt->data, stderr,
                      "colo-compare ppkt", ppkt->size);
         qemu_hexdump((char *)spkt->data, stderr,
                      "colo-compare spkt", spkt->size);
-    }
 
-    return res;
+        /*
+         * colo_compare_inconsistent_notify();
+         * TODO: notice to checkpoint();
+         */
+    }
 }
 
+
 /*
  * Called from the compare thread on the primary
  * for compare udp packet
@@ -XXX,XX +XXX,XX @@ static void colo_old_packet_check(void *opaque)
                         (GCompareFunc)colo_old_packet_check_one_conn);
 }
 
-/*
- * Called from the compare thread on the primary
- * for compare packet with secondary list of the
- * specified connection when a new packet was
- * queued to it.
- */
-static void colo_compare_connection(void *opaque, void *user_data)
+static void colo_compare_packet(CompareState *s, Connection *conn,
+                                int (*HandlePacket)(Packet *spkt,
+                                Packet *ppkt))
 {
-    CompareState *s = user_data;
-    Connection *conn = opaque;
     Packet *pkt = NULL;
     GList *result = NULL;
-    int ret;
 
     while (!g_queue_is_empty(&conn->primary_list) &&
            !g_queue_is_empty(&conn->secondary_list)) {
         pkt = g_queue_pop_head(&conn->primary_list);
-        switch (conn->ip_proto) {
-        case IPPROTO_TCP:
-            result = g_queue_find_custom(&conn->secondary_list,
-                     pkt, (GCompareFunc)colo_packet_compare_tcp);
-            break;
-        case IPPROTO_UDP:
-            result = g_queue_find_custom(&conn->secondary_list,
-                     pkt, (GCompareFunc)colo_packet_compare_udp);
-            break;
-        case IPPROTO_ICMP:
-            result = g_queue_find_custom(&conn->secondary_list,
-                     pkt, (GCompareFunc)colo_packet_compare_icmp);
-            break;
-        default:
-            result = g_queue_find_custom(&conn->secondary_list,
-                     pkt, (GCompareFunc)colo_packet_compare_other);
-            break;
-        }
+        result = g_queue_find_custom(&conn->secondary_list,
+                 pkt, (GCompareFunc)HandlePacket);
 
         if (result) {
-            ret = compare_chr_send(s,
-                                   pkt->data,
-                                   pkt->size,
-                                   pkt->vnet_hdr_len);
-            if (ret < 0) {
-                error_report("colo_send_primary_packet failed");
-            }
-            trace_colo_compare_main("packet same and release packet");
+            colo_release_primary_pkt(s, pkt);
             g_queue_remove(&conn->secondary_list, result->data);
-            packet_destroy(pkt, NULL);
         } else {
             /*
              * If one packet arrive late, the secondary_list or
@@ -XXX,XX +XXX,XX @@ static void colo_compare_connection(void *opaque, void *user_data)
     }
 }
 
+/*
+ * Called from the compare thread on the primary
+ * for compare packet with secondary list of the
+ * specified connection when a new packet was
+ * queued to it.
+ */
+static void colo_compare_connection(void *opaque, void *user_data)
+{
+    CompareState *s = user_data;
+    Connection *conn = opaque;
+
+    switch (conn->ip_proto) {
+    case IPPROTO_TCP:
+        colo_compare_tcp(s, conn);
+        break;
+    case IPPROTO_UDP:
+        colo_compare_packet(s, conn, colo_packet_compare_udp);
+        break;
+    case IPPROTO_ICMP:
+        colo_compare_packet(s, conn, colo_packet_compare_icmp);
+        break;
+    default:
+        colo_compare_packet(s, conn, colo_packet_compare_other);
+        break;
+    }
+}
+
 static int compare_chr_send(CompareState *s,
                             const uint8_t *buf,
                             uint32_t size,
diff --git a/net/colo.c b/net/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo.c
+++ b/net/colo.c
@@ -XXX,XX +XXX,XX @@ Connection *connection_new(ConnectionKey *key)
     conn->processing = false;
     conn->offset = 0;
     conn->syn_flag = 0;
+    conn->pack = 0;
+    conn->sack = 0;
     g_queue_init(&conn->primary_list);
     g_queue_init(&conn->secondary_list);
 
@@ -XXX,XX +XXX,XX @@ Packet *packet_new(const void *data, int size, int vnet_hdr_len)
     pkt->size = size;
     pkt->creation_ms = qemu_clock_get_ms(QEMU_CLOCK_HOST);
     pkt->vnet_hdr_len = vnet_hdr_len;
+    pkt->tcp_seq = 0;
+    pkt->tcp_ack = 0;
+    pkt->seq_end = 0;
+    pkt->header_size = 0;
+    pkt->payload_size = 0;
+    pkt->offset = 0;
+    pkt->flags = 0;
 
     return pkt;
 }
diff --git a/net/colo.h b/net/colo.h
index XXXXXXX..XXXXXXX 100644
--- a/net/colo.h
+++ b/net/colo.h
@@ -XXX,XX +XXX,XX @@ typedef struct Packet {
     int64_t creation_ms;
     /* Get vnet_hdr_len from filter */
     uint32_t vnet_hdr_len;
+    uint32_t tcp_seq; /* sequence number */
+    uint32_t tcp_ack; /* acknowledgement number */
+    /* the sequence number of the last byte of the packet */
+    uint32_t seq_end;
+    uint8_t header_size;  /* the header length */
+    uint16_t payload_size; /* the payload length */
+    /* record the payload offset(the length that has been compared) */
+    uint16_t offset;
+    uint8_t flags; /* Flags(aka Control bits) */
 } Packet;
 
 typedef struct ConnectionKey {
@@ -XXX,XX +XXX,XX @@ typedef struct Connection {
     /* flag to enqueue unprocessed_connections */
     bool processing;
     uint8_t ip_proto;
+    /* record the sequence number that has been compared */
+    uint32_t compare_seq;
+    /* the maximum of acknowledgement number in primary_list queue */
+    uint32_t pack;
+    /* the maximum of acknowledgement number in secondary_list queue */
+    uint32_t sack;
     /* offset = secondary_seq - primary_seq */
     tcp_seq  offset;
     /*
diff --git a/net/trace-events b/net/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/net/trace-events
+++ b/net/trace-events
@@ -XXX,XX +XXX,XX @@ colo_compare_icmp_miscompare(const char *sta, int size) ": %s = %d"
 colo_compare_ip_info(int psize, const char *sta, const char *stb, int ssize, const char *stc, const char *std) "ppkt size = %d, ip_src = %s, ip_dst = %s, spkt size = %d, ip_src = %s, ip_dst = %s"
 colo_old_packet_check_found(int64_t old_time) "%" PRId64
 colo_compare_miscompare(void) ""
-colo_compare_tcp_info(const char *pkt, uint32_t seq, uint32_t ack, int res, uint32_t flag, int size) "side: %s seq/ack= %u/%u res= %d flags= 0x%x pkt_size: %d\n"
+colo_compare_tcp_info(const char *pkt, uint32_t seq, uint32_t ack, int hdlen, int pdlen, int offset, int flags) "%s: seq/ack= %u/%u hdlen= %d pdlen= %d offset= %d flags=%d\n"
 
 # net/filter-rewriter.c
 colo_filter_rewriter_debug(void) ""
-- 
2.7.4

From: Thomas Huth <thuth@redhat.com>

QEMU can emulate hubs to connect NICs and netdevs. This is currently
primarily used for the mis-named 'vlan' feature of the networking
subsystem. Now the 'vlan' feature has been marked as deprecated, since
its name is rather confusing and the users often rather mis-configure
their network when trying to use it. But while the 'vlan' parameter
should be removed at one point in time, the basic idea of emulating
a hub in QEMU is still good: It's useful for bundling up the output of
multiple NICs into one single l2tp netdev for example.

Now to be able to use the hubport feature without 'vlan's, there is one
missing piece: The possibility to connect a hubport to a netdev, too.
This patch adds this possibility by introducing a new "netdev=..."
parameter to the hubports.

To bundle up the output of multiple NICs into one socket netdev, you can
now run QEMU with these parameters for example:

qemu-system-ppc64 ... -netdev socket,id=s1,connect=:11122 \
    -netdev hubport,hubid=1,id=h1,netdev=s1 \
    -netdev hubport,hubid=1,id=h2 -device e1000,netdev=h2 \
    -netdev hubport,hubid=1,id=h3 -device virtio-net-pci,netdev=h3

For using the socket netdev, you have got to start another QEMU as the
receiving side first, for example with network dumping enabled:

qemu-system-x86_64 -M isapc -netdev socket,id=s0,listen=:11122 \
    -device ne2k_isa,netdev=s0 \
    -object filter-dump,id=f1,netdev=s0,file=/tmp/dump.dat

After the ppc64 guest tried to boot from both NICs, you can see in the
dump file (using Wireshark, for example), that the output of both NICs
(the e1000 and the virtio-net-pci) has been successfully transfered
via the socket netdev in this case.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/hub.c       | 27 +++++++++++++++++++++------
 net/hub.h       |  3 ++-
 net/net.c       |  2 +-
 qapi/net.json   |  4 +++-
 qemu-options.hx |  8 +++++---
 5 files changed, 32 insertions(+), 12 deletions(-)

diff --git a/net/hub.c b/net/hub.c
index XXXXXXX..XXXXXXX 100644
--- a/net/hub.c
+++ b/net/hub.c
@@ -XXX,XX +XXX,XX @@
  */
 
 #include "qemu/osdep.h"
+#include "qapi/error.h"
 #include "monitor/monitor.h"
 #include "net/net.h"
 #include "clients.h"
@@ -XXX,XX +XXX,XX @@ static NetClientInfo net_hub_port_info = {
     .cleanup = net_hub_port_cleanup,
 };
 
-static NetHubPort *net_hub_port_new(NetHub *hub, const char *name)
+static NetHubPort *net_hub_port_new(NetHub *hub, const char *name,
+                                    NetClientState *hubpeer)
 {
     NetClientState *nc;
     NetHubPort *port;
@@ -XXX,XX +XXX,XX @@ static NetHubPort *net_hub_port_new(NetHub *hub, const char *name)
         name = default_name;
     }
 
-    nc = qemu_new_net_client(&net_hub_port_info, NULL, "hub", name);
+    nc = qemu_new_net_client(&net_hub_port_info, hubpeer, "hub", name);
     port = DO_UPCAST(NetHubPort, nc, nc);
     port->id = id;
     port->hub = hub;
@@ -XXX,XX +XXX,XX @@ static NetHubPort *net_hub_port_new(NetHub *hub, const char *name)
 
 /**
  * Create a port on a given hub
+ * @hub_id: Number of the hub
  * @name: Net client name or NULL for default name.
+ * @hubpeer: Peer to use (if "netdev=id" has been specified)
  *
  * If there is no existing hub with the given id then a new hub is created.
  */
-NetClientState *net_hub_add_port(int hub_id, const char *name)
+NetClientState *net_hub_add_port(int hub_id, const char *name,
+                                 NetClientState *hubpeer)
 {
     NetHub *hub;
     NetHubPort *port;
@@ -XXX,XX +XXX,XX @@ NetClientState *net_hub_add_port(int hub_id, const char *name)
         hub = net_hub_new(hub_id);
     }
 
-    port = net_hub_port_new(hub, name);
+    port = net_hub_port_new(hub, name, hubpeer);
     return &port->nc;
 }
 
@@ -XXX,XX +XXX,XX @@ NetClientState *net_hub_port_find(int hub_id)
         }
     }
 
-    nc = net_hub_add_port(hub_id, NULL);
+    nc = net_hub_add_port(hub_id, NULL, NULL);
     return nc;
 }
 
@@ -XXX,XX +XXX,XX @@ int net_init_hubport(const Netdev *netdev, const char *name,
                      NetClientState *peer, Error **errp)
 {
     const NetdevHubPortOptions *hubport;
+    NetClientState *hubpeer = NULL;
 
     assert(netdev->type == NET_CLIENT_DRIVER_HUBPORT);
     assert(!peer);
     hubport = &netdev->u.hubport;
 
-    net_hub_add_port(hubport->hubid, name);
+    if (hubport->has_netdev) {
+        hubpeer = qemu_find_netdev(hubport->netdev);
+        if (!hubpeer) {
+            error_setg(errp, "netdev '%s' not found", hubport->netdev);
+            return -1;
+        }
+    }
+
+    net_hub_add_port(hubport->hubid, name, hubpeer);
+
     return 0;
 }
 
diff --git a/net/hub.h b/net/hub.h
index XXXXXXX..XXXXXXX 100644
--- a/net/hub.h
+++ b/net/hub.h
@@ -XXX,XX +XXX,XX @@
 
 #include "qemu-common.h"
 
-NetClientState *net_hub_add_port(int hub_id, const char *name);
+NetClientState *net_hub_add_port(int hub_id, const char *name,
+                                 NetClientState *hubpeer);
 NetClientState *net_hub_find_client_by_name(int hub_id, const char *name);
 void net_hub_info(Monitor *mon);
 void net_hub_check_clients(void);
diff --git a/net/net.c b/net/net.c
index XXXXXXX..XXXXXXX 100644
--- a/net/net.c
+++ b/net/net.c
@@ -XXX,XX +XXX,XX @@ static int net_client_init1(const void *object, bool is_netdev, Error **errp)
         /* Do not add to a vlan if it's a nic with a netdev= parameter. */
         if (netdev->type != NET_CLIENT_DRIVER_NIC ||
             !opts->u.nic.has_netdev) {
-            peer = net_hub_add_port(net->has_vlan ? net->vlan : 0, NULL);
+            peer = net_hub_add_port(net->has_vlan ? net->vlan : 0, NULL, NULL);
         }
 
         if (net->has_vlan && !vlan_warned) {
diff --git a/qapi/net.json b/qapi/net.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/net.json
+++ b/qapi/net.json
@@ -XXX,XX +XXX,XX @@
 # Connect two or more net clients through a software hub.
 #
 # @hubid: hub identifier number
+# @netdev: used to connect hub to a netdev instead of a device (since 2.12)
 #
 # Since: 1.2
 ##
 { 'struct': 'NetdevHubPortOptions',
   'data': {
-    'hubid':     'int32' } }
+    'hubid':     'int32',
+    '*netdev':    'str' } }
 
 ##
 # @NetdevNetmapOptions:
diff --git a/qemu-options.hx b/qemu-options.hx
index XXXXXXX..XXXXXXX 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -XXX,XX +XXX,XX @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
 #endif
     "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
     "                configure a vhost-user network, backed by a chardev 'dev'\n"
-    "-netdev hubport,id=str,hubid=n\n"
+    "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
     "                configure a hub port on QEMU VLAN 'n'\n", QEMU_ARCH_ALL)
 DEF("net", HAS_ARG, QEMU_OPTION_net,
     "-net nic[,vlan=n][,netdev=nd][,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
@@ -XXX,XX +XXX,XX @@ vde_switch -F -sock /tmp/myswitch
 qemu-system-i386 linux.img -net nic -net vde,sock=/tmp/myswitch
 @end example
 
-@item -netdev hubport,id=@var{id},hubid=@var{hubid}
+@item -netdev hubport,id=@var{id},hubid=@var{hubid}[,netdev=@var{nd}]
 
 Create a hub port on QEMU "vlan" @var{hubid}.
 
 The hubport netdev lets you connect a NIC to a QEMU "vlan" instead of a single
 netdev.  @code{-net} and @code{-device} with parameter @option{vlan} create the
-required hub automatically.
+required hub automatically. Alternatively, you can also connect the hubport
+to another netdev with ID @var{nd} by using the @option{netdev=@var{nd}}
+option.
 
 @item -netdev vhost-user,chardev=@var{id}[,vhostforce=on|off][,queues=n]
 
-- 
2.7.4

From: Thomas Huth <thuth@redhat.com>

It does not make much sense to limit these commands to the legacy 'vlan'
concept only, they should work with the modern netdevs, too. So now
it is possible to use this command with one, two or three parameters.

With one parameter, the command installs a hostfwd rule on the default
"user" network:
    hostfwd_add tcp:...

With two parameters, the command installs a hostfwd rule on a netdev
(that's the new way of using this command):
    hostfwd_add netdev_id tcp:...

With three parameters, the command installs a rule on a 'vlan' (aka hub):
    hostfwd_add hub_id name tcp:...

Same applies to the hostfwd_remove command now.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hmp-commands.hx |  4 ++--
 net/slirp.c     | 33 +++++++++++++++++++++++----------
 2 files changed, 25 insertions(+), 12 deletions(-)

diff --git a/hmp-commands.hx b/hmp-commands.hx
index XXXXXXX..XXXXXXX 100644
--- a/hmp-commands.hx
+++ b/hmp-commands.hx
@@ -XXX,XX +XXX,XX @@ ETEXI
     {
         .name       = "hostfwd_add",
         .args_type  = "arg1:s,arg2:s?,arg3:s?",
-        .params     = "[vlan_id name] [tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport",
+        .params     = "[hub_id name]|[netdev_id] [tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport",
         .help       = "redirect TCP or UDP connections from host to guest (requires -net user)",
         .cmd        = hmp_hostfwd_add,
     },
@@ -XXX,XX +XXX,XX @@ ETEXI
     {
         .name       = "hostfwd_remove",
         .args_type  = "arg1:s,arg2:s?,arg3:s?",
-        .params     = "[vlan_id name] [tcp|udp]:[hostaddr]:hostport",
+        .params     = "[hub_id name]|[netdev_id] [tcp|udp]:[hostaddr]:hostport",
         .help       = "remove host-to-guest TCP or UDP redirection",
         .cmd        = hmp_hostfwd_remove,
     },
diff --git a/net/slirp.c b/net/slirp.c
index XXXXXXX..XXXXXXX 100644
--- a/net/slirp.c
+++ b/net/slirp.c
@@ -XXX,XX +XXX,XX @@ error:
     return -1;
 }
 
-static SlirpState *slirp_lookup(Monitor *mon, const char *vlan,
-                                const char *stack)
+static SlirpState *slirp_lookup(Monitor *mon, const char *hub_id,
+                                const char *name)
 {
-
-    if (vlan) {
+    if (name) {
         NetClientState *nc;
-        nc = net_hub_find_client_by_name(strtol(vlan, NULL, 0), stack);
-        if (!nc) {
-            monitor_printf(mon, "unrecognized (vlan-id, stackname) pair\n");
-            return NULL;
+        if (hub_id) {
+            nc = net_hub_find_client_by_name(strtol(hub_id, NULL, 0), name);
+            if (!nc) {
+                monitor_printf(mon, "unrecognized (vlan-id, stackname) pair\n");
+                return NULL;
+            }
+        } else {
+            nc = qemu_find_netdev(name);
+            if (!nc) {
+                monitor_printf(mon, "unrecognized netdev id '%s'\n", name);
+                return NULL;
+            }
         }
         if (strcmp(nc->model, "user")) {
             monitor_printf(mon, "invalid device specified\n");
@@ -XXX,XX +XXX,XX @@ void hmp_hostfwd_remove(Monitor *mon, const QDict *qdict)
     const char *arg2 = qdict_get_try_str(qdict, "arg2");
     const char *arg3 = qdict_get_try_str(qdict, "arg3");
 
-    if (arg2) {
+    if (arg3) {
         s = slirp_lookup(mon, arg1, arg2);
         src_str = arg3;
+    } else if (arg2) {
+        s = slirp_lookup(mon, NULL, arg1);
+        src_str = arg2;
     } else {
         s = slirp_lookup(mon, NULL, NULL);
         src_str = arg1;
@@ -XXX,XX +XXX,XX @@ void hmp_hostfwd_add(Monitor *mon, const QDict *qdict)
     const char *arg2 = qdict_get_try_str(qdict, "arg2");
     const char *arg3 = qdict_get_try_str(qdict, "arg3");
 
-    if (arg2) {
+    if (arg3) {
         s = slirp_lookup(mon, arg1, arg2);
         redir_str = arg3;
+    } else if (arg2) {
+        s = slirp_lookup(mon, NULL, arg1);
+        redir_str = arg2;
     } else {
         s = slirp_lookup(mon, NULL, NULL);
         redir_str = arg1;
-- 
2.7.4

From: Philippe Mathieu-Daudé <f4bug@amsat.org>

gently asked by his automatic reply :)

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 MAINTAINERS | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index XXXXXXX..XXXXXXX 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -XXX,XX +XXX,XX @@ F: hw/scsi/mfi.h
 F: tests/megasas-test.c
 
 Network packet abstractions
-M: Dmitry Fleytman <dmitry@daynix.com>
+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
 S: Maintained
 F: include/net/eth.h
 F: net/eth.c
@@ -XXX,XX +XXX,XX @@ F: hw/net/net_rx_pkt*
 F: hw/net/net_tx_pkt*
 
 Vmware
-M: Dmitry Fleytman <dmitry@daynix.com>
+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
 S: Maintained
 F: hw/net/vmxnet*
 F: hw/scsi/vmw_pvscsi*
@@ -XXX,XX +XXX,XX @@ F: hw/mem/nvdimm.c
 F: include/hw/mem/nvdimm.h
 
 e1000x
-M: Dmitry Fleytman <dmitry@daynix.com>
+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
 S: Maintained
 F: hw/net/e1000x*
 
 e1000e
-M: Dmitry Fleytman <dmitry@daynix.com>
+M: Dmitry Fleytman <dmitry.fleytman@gmail.com>
 S: Maintained
 F: hw/net/e1000e*
 
-- 
2.7.4

The following changes since commit 77f7c747193662edfadeeb3118d63eed0eac51a6:

Merge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2018-10-17' into staging (2018-10-18 13:40:19 +0100)

are available in the git repository at:

https://github.com/jasowang/qemu.git tags/net-pull-request

for you to fetch changes up to 37a4442a76d010f5d957e3ee09dfb23364281b37:

qemu-options: Fix bad "macaddr" property in the documentation (2018-10-19 11:15:04 +0800)

----------------------------------------------------------------

----------------------------------------------------------------
Jason Wang (5):
      ne2000: fix possible out of bound access in ne2000_receive
      rtl8139: fix possible out of bound access
      pcnet: fix possible buffer overflow
      net: ignore packet size greater than INT_MAX
      e1000: indicate dropped packets in HW counters

Thomas Huth (1):
      qemu-options: Fix bad "macaddr" property in the documentation

Zhang Chen (15):
      filter-rewriter: Add TCP state machine and fix memory leak in connection_track_table
      colo-compare: implement the process of checkpoint
      colo-compare: use notifier to notify packets comparing result
      COLO: integrate colo compare with colo frame
      COLO: Add block replication into colo process
      COLO: Remove colo_state migration struct
      COLO: Load dirty pages into SVM's RAM cache firstly
      ram/COLO: Record the dirty pages that SVM received
      COLO: Flush memory data from ram cache
      qapi/migration.json: Rename COLO unknown mode to none mode.
      qapi: Add new command to query colo status
      savevm: split the process of different stages for loadvm/savevm
      filter: Add handle_event method for NetFilterClass
      filter-rewriter: handle checkpoint and failover event
      docs: Add COLO status diagram to COLO-FT.txt

liujunjie (1):
      clean up callback when del virtqueue

zhanghailiang (4):
      qmp event: Add COLO_EXIT event to notify users while exited COLO
      COLO: flush host dirty ram from cache
      COLO: notify net filters about checkpoint/failover event
      COLO: quick failover process by kick COLO thread

From: Zhang Chen <zhangckid@gmail.com>

We add almost full TCP state machine in filter-rewriter, except
TCPS_LISTEN and some simplify in VM active close FIN states.
The reason for this simplify job is because guest kernel will track
the TCP status and wait 2MSL time too, if client resend the FIN packet,
guest will resend the last ACK, so we needn't wait 2MSL time in filter-rewriter.

After a net connection is closed, we didn't clear its related resources
in connection_track_table, which will lead to memory leak.

Let's track the state of net connection, if it is closed, its related
resources will be cleared up.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/colo.c            |   2 +-
 net/colo.h            |   9 ++---
 net/filter-rewriter.c | 109 +++++++++++++++++++++++++++++++++++++++++++++-----
 3 files changed, 104 insertions(+), 16 deletions(-)

diff --git a/net/colo.c b/net/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo.c
+++ b/net/colo.c
@@ -XXX,XX +XXX,XX @@ Connection *connection_new(ConnectionKey *key)
     conn->ip_proto = key->ip_proto;
     conn->processing = false;
     conn->offset = 0;
-    conn->syn_flag = 0;
+    conn->tcp_state = TCPS_CLOSED;
     conn->pack = 0;
     conn->sack = 0;
     g_queue_init(&conn->primary_list);
diff --git a/net/colo.h b/net/colo.h
index XXXXXXX..XXXXXXX 100644
--- a/net/colo.h
+++ b/net/colo.h
@@ -XXX,XX +XXX,XX @@
 #include "slirp/slirp.h"
 #include "qemu/jhash.h"
 #include "qemu/timer.h"
+#include "slirp/tcp.h"
 
 #define HASHTABLE_MAX_SIZE 16384
 
@@ -XXX,XX +XXX,XX @@ typedef struct Connection {
     uint32_t sack;
     /* offset = secondary_seq - primary_seq */
     tcp_seq  offset;
-    /*
-     * we use this flag update offset func
-     * run once in independent tcp connection
-     */
-    int syn_flag;
+
+    int tcp_state; /* TCP FSM state */
+    tcp_seq fin_ack_seq; /* the seq of 'fin=1,ack=1' */
 } Connection;
 
 uint32_t connection_key_hash(const void *opaque);
diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c
index XXXXXXX..XXXXXXX 100644
--- a/net/filter-rewriter.c
+++ b/net/filter-rewriter.c
@@ -XXX,XX +XXX,XX @@ static int is_tcp_packet(Packet *pkt)
 }
 
 /* handle tcp packet from primary guest */
-static int handle_primary_tcp_pkt(NetFilterState *nf,
+static int handle_primary_tcp_pkt(RewriterState *rf,
                                   Connection *conn,
-                                  Packet *pkt)
+                                  Packet *pkt, ConnectionKey *key)
 {
     struct tcphdr *tcp_pkt;
 
@@ -XXX,XX +XXX,XX @@ static int handle_primary_tcp_pkt(NetFilterState *nf,
         trace_colo_filter_rewriter_conn_offset(conn->offset);
     }
 
+    if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == (TH_ACK | TH_SYN)) &&
+        conn->tcp_state == TCPS_SYN_SENT) {
+        conn->tcp_state = TCPS_ESTABLISHED;
+    }
+
     if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_SYN)) {
         /*
          * we use this flag update offset func
          * run once in independent tcp connection
          */
-        conn->syn_flag = 1;
+        conn->tcp_state = TCPS_SYN_RECEIVED;
     }
 
     if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_ACK)) {
-        if (conn->syn_flag) {
+        if (conn->tcp_state == TCPS_SYN_RECEIVED) {
             /*
              * offset = secondary_seq - primary seq
              * ack packet sent by guest from primary node,
              * so we use th_ack - 1 get primary_seq
              */
             conn->offset -= (ntohl(tcp_pkt->th_ack) - 1);
-            conn->syn_flag = 0;
+            conn->tcp_state = TCPS_ESTABLISHED;
         }
         if (conn->offset) {
             /* handle packets to the secondary from the primary */
@@ -XXX,XX +XXX,XX @@ static int handle_primary_tcp_pkt(NetFilterState *nf,
             net_checksum_calculate((uint8_t *)pkt->data + pkt->vnet_hdr_len,
                                    pkt->size - pkt->vnet_hdr_len);
         }
+
+        /*
+         * Passive close step 3
+         */
+        if ((conn->tcp_state == TCPS_LAST_ACK) &&
+            (ntohl(tcp_pkt->th_ack) == (conn->fin_ack_seq + 1))) {
+            conn->tcp_state = TCPS_CLOSED;
+            g_hash_table_remove(rf->connection_track_table, key);
+        }
+    }
+
+    if ((tcp_pkt->th_flags & TH_FIN) == TH_FIN) {
+        /*
+         * Passive close.
+         * Step 1:
+         * The *server* side of this connect is VM, *client* tries to close
+         * the connection. We will into CLOSE_WAIT status.
+         *
+         * Step 2:
+         * In this step we will into LAST_ACK status.
+         *
+         * We got 'fin=1, ack=1' packet from server side, we need to
+         * record the seq of 'fin=1, ack=1' packet.
+         *
+         * Step 3:
+         * We got 'ack=1' packets from client side, it acks 'fin=1, ack=1'
+         * packet from server side. From this point, we can ensure that there
+         * will be no packets in the connection, except that, some errors
+         * happen between the path of 'filter object' and vNIC, if this rare
+         * case really happen, we can still create a new connection,
+         * So it is safe to remove the connection from connection_track_table.
+         *
+         */
+        if (conn->tcp_state == TCPS_ESTABLISHED) {
+            conn->tcp_state = TCPS_CLOSE_WAIT;
+        }
+
+        /*
+         * Active close step 2.
+         */
+        if (conn->tcp_state == TCPS_FIN_WAIT_1) {
+            conn->tcp_state = TCPS_TIME_WAIT;
+            /*
+             * For simplify implementation, we needn't wait 2MSL time
+             * in filter rewriter. Because guest kernel will track the
+             * TCP status and wait 2MSL time, if client resend the FIN
+             * packet, guest will apply the last ACK too.
+             */
+            conn->tcp_state = TCPS_CLOSED;
+            g_hash_table_remove(rf->connection_track_table, key);
+        }
     }
 
     return 0;
 }
 
 /* handle tcp packet from secondary guest */
-static int handle_secondary_tcp_pkt(NetFilterState *nf,
+static int handle_secondary_tcp_pkt(RewriterState *rf,
                                     Connection *conn,
-                                    Packet *pkt)
+                                    Packet *pkt, ConnectionKey *key)
 {
     struct tcphdr *tcp_pkt;
 
@@ -XXX,XX +XXX,XX @@ static int handle_secondary_tcp_pkt(NetFilterState *nf,
         trace_colo_filter_rewriter_conn_offset(conn->offset);
     }
 
-    if (((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == (TH_ACK | TH_SYN))) {
+    if (conn->tcp_state == TCPS_SYN_RECEIVED &&
+        ((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == (TH_ACK | TH_SYN))) {
         /*
          * save offset = secondary_seq and then
          * in handle_primary_tcp_pkt make offset
@@ -XXX,XX +XXX,XX @@ static int handle_secondary_tcp_pkt(NetFilterState *nf,
         conn->offset = ntohl(tcp_pkt->th_seq);
     }
 
+    /* VM active connect */
+    if (conn->tcp_state == TCPS_CLOSED &&
+        ((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_SYN)) {
+        conn->tcp_state = TCPS_SYN_SENT;
+    }
+
     if ((tcp_pkt->th_flags & (TH_ACK | TH_SYN)) == TH_ACK) {
         /* Only need to adjust seq while offset is Non-zero */
         if (conn->offset) {
@@ -XXX,XX +XXX,XX @@ static int handle_secondary_tcp_pkt(NetFilterState *nf,
         }
     }
 
+    /*
+     * Passive close step 2:
+     */
+    if (conn->tcp_state == TCPS_CLOSE_WAIT &&
+        (tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == (TH_ACK | TH_FIN)) {
+        conn->fin_ack_seq = ntohl(tcp_pkt->th_seq);
+        conn->tcp_state = TCPS_LAST_ACK;
+    }
+
+    /*
+     * Active close
+     *
+     * Step 1:
+     * The *server* side of this connect is VM, *server* tries to close
+     * the connection.
+     *
+     * Step 2:
+     * We will into CLOSE_WAIT status.
+     * We simplify the TCPS_FIN_WAIT_2, TCPS_TIME_WAIT and
+     * CLOSING status.
+     */
+    if (conn->tcp_state == TCPS_ESTABLISHED &&
+        (tcp_pkt->th_flags & (TH_ACK | TH_FIN)) == TH_FIN) {
+        conn->tcp_state = TCPS_FIN_WAIT_1;
+    }
+
     return 0;
 }
 
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
 
         if (sender == nf->netdev) {
             /* NET_FILTER_DIRECTION_TX */
-            if (!handle_primary_tcp_pkt(nf, conn, pkt)) {
+            if (!handle_primary_tcp_pkt(s, conn, pkt, &key)) {
                 qemu_net_queue_send(s->incoming_queue, sender, 0,
                 (const uint8_t *)pkt->data, pkt->size, NULL);
                 packet_destroy(pkt, NULL);
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
             }
         } else {
             /* NET_FILTER_DIRECTION_RX */
-            if (!handle_secondary_tcp_pkt(nf, conn, pkt)) {
+            if (!handle_secondary_tcp_pkt(s, conn, pkt, &key)) {
                 qemu_net_queue_send(s->incoming_queue, sender, 0,
                 (const uint8_t *)pkt->data, pkt->size, NULL);
                 packet_destroy(pkt, NULL);
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

While do checkpoint, we need to flush all the unhandled packets,
By using the filter notifier mechanism, we can easily to notify
every compare object to do this process, which runs inside
of compare threads as a coroutine.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 include/migration/colo.h |  6 ++++
 net/colo-compare.c       | 78 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/colo-compare.h       | 22 ++++++++++++++
 3 files changed, 106 insertions(+)
 create mode 100644 net/colo-compare.h

diff --git a/include/migration/colo.h b/include/migration/colo.h
index XXXXXXX..XXXXXXX 100644
--- a/include/migration/colo.h
+++ b/include/migration/colo.h
@@ -XXX,XX +XXX,XX @@
 #include "qemu-common.h"
 #include "qapi/qapi-types-migration.h"
 
+enum colo_event {
+    COLO_EVENT_NONE,
+    COLO_EVENT_CHECKPOINT,
+    COLO_EVENT_FAILOVER,
+};
+
 void colo_info_init(void);
 
 void migrate_start_colo_process(MigrationState *s);
diff --git a/net/colo-compare.c b/net/colo-compare.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/sockets.h"
 #include "colo.h"
 #include "sysemu/iothread.h"
+#include "net/colo-compare.h"
+#include "migration/colo.h"
 
 #define TYPE_COLO_COMPARE "colo-compare"
 #define COLO_COMPARE(obj) \
     OBJECT_CHECK(CompareState, (obj), TYPE_COLO_COMPARE)
 
+static QTAILQ_HEAD(, CompareState) net_compares =
+       QTAILQ_HEAD_INITIALIZER(net_compares);
+
 #define COMPARE_READ_LEN_MAX NET_BUFSIZE
 #define MAX_QUEUE_SIZE 1024
 
@@ -XXX,XX +XXX,XX @@
 /* TODO: Should be configurable */
 #define REGULAR_PACKET_CHECK_MS 3000
 
+static QemuMutex event_mtx;
+static QemuCond event_complete_cond;
+static int event_unhandled_count;
+
 /*
  *  + CompareState ++
  *  |               |
@@ -XXX,XX +XXX,XX @@ typedef struct CompareState {
     IOThread *iothread;
     GMainContext *worker_context;
     QEMUTimer *packet_check_timer;
+
+    QEMUBH *event_bh;
+    enum colo_event event;
+
+    QTAILQ_ENTRY(CompareState) next;
 } CompareState;
 
 typedef struct CompareClass {
@@ -XXX,XX +XXX,XX @@ static void check_old_packet_regular(void *opaque)
                 REGULAR_PACKET_CHECK_MS);
 }
 
+/* Public API, Used for COLO frame to notify compare event */
+void colo_notify_compares_event(void *opaque, int event, Error **errp)
+{
+    CompareState *s;
+
+    qemu_mutex_lock(&event_mtx);
+    QTAILQ_FOREACH(s, &net_compares, next) {
+        s->event = event;
+        qemu_bh_schedule(s->event_bh);
+        event_unhandled_count++;
+    }
+    /* Wait all compare threads to finish handling this event */
+    while (event_unhandled_count > 0) {
+        qemu_cond_wait(&event_complete_cond, &event_mtx);
+    }
+
+    qemu_mutex_unlock(&event_mtx);
+}
+
 static void colo_compare_timer_init(CompareState *s)
 {
     AioContext *ctx = iothread_get_aio_context(s->iothread);
@@ -XXX,XX +XXX,XX @@ static void colo_compare_timer_del(CompareState *s)
     }
  }
 
+static void colo_flush_packets(void *opaque, void *user_data);
+
+static void colo_compare_handle_event(void *opaque)
+{
+    CompareState *s = opaque;
+
+    switch (s->event) {
+    case COLO_EVENT_CHECKPOINT:
+        g_queue_foreach(&s->conn_list, colo_flush_packets, s);
+        break;
+    case COLO_EVENT_FAILOVER:
+        break;
+    default:
+        break;
+    }
+
+    assert(event_unhandled_count > 0);
+
+    qemu_mutex_lock(&event_mtx);
+    event_unhandled_count--;
+    qemu_cond_broadcast(&event_complete_cond);
+    qemu_mutex_unlock(&event_mtx);
+}
+
 static void colo_compare_iothread(CompareState *s)
 {
     object_ref(OBJECT(s->iothread));
@@ -XXX,XX +XXX,XX @@ static void colo_compare_iothread(CompareState *s)
                              s, s->worker_context, true);
 
     colo_compare_timer_init(s);
+    s->event_bh = qemu_bh_new(colo_compare_handle_event, s);
 }
 
 static char *compare_get_pri_indev(Object *obj, Error **errp)
@@ -XXX,XX +XXX,XX @@ static void colo_compare_complete(UserCreatable *uc, Error **errp)
     net_socket_rs_init(&s->pri_rs, compare_pri_rs_finalize, s->vnet_hdr);
     net_socket_rs_init(&s->sec_rs, compare_sec_rs_finalize, s->vnet_hdr);
 
+    QTAILQ_INSERT_TAIL(&net_compares, s, next);
+
     g_queue_init(&s->conn_list);
 
+    qemu_mutex_init(&event_mtx);
+    qemu_cond_init(&event_complete_cond);
+
     s->connection_track_table = g_hash_table_new_full(connection_key_hash,
                                                       connection_key_equal,
                                                       g_free,
@@ -XXX,XX +XXX,XX @@ static void colo_compare_init(Object *obj)
 static void colo_compare_finalize(Object *obj)
 {
     CompareState *s = COLO_COMPARE(obj);
+    CompareState *tmp = NULL;
 
     qemu_chr_fe_deinit(&s->chr_pri_in, false);
     qemu_chr_fe_deinit(&s->chr_sec_in, false);
@@ -XXX,XX +XXX,XX @@ static void colo_compare_finalize(Object *obj)
     if (s->iothread) {
         colo_compare_timer_del(s);
     }
+
+    qemu_bh_delete(s->event_bh);
+
+    QTAILQ_FOREACH(tmp, &net_compares, next) {
+        if (tmp == s) {
+            QTAILQ_REMOVE(&net_compares, s, next);
+            break;
+        }
+    }
+
     /* Release all unhandled packets after compare thead exited */
     g_queue_foreach(&s->conn_list, colo_flush_packets, s);
 
@@ -XXX,XX +XXX,XX @@ static void colo_compare_finalize(Object *obj)
     if (s->iothread) {
         object_unref(OBJECT(s->iothread));
     }
+
+    qemu_mutex_destroy(&event_mtx);
+    qemu_cond_destroy(&event_complete_cond);
+
     g_free(s->pri_indev);
     g_free(s->sec_indev);
     g_free(s->outdev);
diff --git a/net/colo-compare.h b/net/colo-compare.h
new file mode 100644
index XXXXXXX..XXXXXXX
--- /dev/null
+++ b/net/colo-compare.h
@@ -XXX,XX +XXX,XX @@
+/*
+ * COarse-grain LOck-stepping Virtual Machines for Non-stop Service (COLO)
+ * (a.k.a. Fault Tolerance or Continuous Replication)
+ *
+ * Copyright (c) 2017 HUAWEI TECHNOLOGIES CO., LTD.
+ * Copyright (c) 2017 FUJITSU LIMITED
+ * Copyright (c) 2017 Intel Corporation
+ *
+ * Authors:
+ *    zhanghailiang <zhang.zhanghailiang@huawei.com>
+ *    Zhang Chen <zhangckid@gmail.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or
+ * later.  See the COPYING file in the top-level directory.
+ */
+
+#ifndef QEMU_COLO_COMPARE_H
+#define QEMU_COLO_COMPARE_H
+
+void colo_notify_compares_event(void *opaque, int event, Error **errp);
+
+#endif /* QEMU_COLO_COMPARE_H */
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

It's a good idea to use notifier to notify COLO frame of
inconsistent packets comparing.

Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/colo-compare.c | 37 ++++++++++++++++++++++++++-----------
 net/colo-compare.h |  2 ++
 2 files changed, 28 insertions(+), 11 deletions(-)

diff --git a/net/colo-compare.c b/net/colo-compare.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -XXX,XX +XXX,XX @@
 #include "sysemu/iothread.h"
 #include "net/colo-compare.h"
 #include "migration/colo.h"
+#include "migration/migration.h"
 
 #define TYPE_COLO_COMPARE "colo-compare"
 #define COLO_COMPARE(obj) \
@@ -XXX,XX +XXX,XX @@
 static QTAILQ_HEAD(, CompareState) net_compares =
        QTAILQ_HEAD_INITIALIZER(net_compares);
 
+static NotifierList colo_compare_notifiers =
+    NOTIFIER_LIST_INITIALIZER(colo_compare_notifiers);
+
 #define COMPARE_READ_LEN_MAX NET_BUFSIZE
 #define MAX_QUEUE_SIZE 1024
 
@@ -XXX,XX +XXX,XX @@ static bool colo_mark_tcp_pkt(Packet *ppkt, Packet *spkt,
     return false;
 }
 
+static void colo_compare_inconsistency_notify(void)
+{
+    notifier_list_notify(&colo_compare_notifiers,
+                migrate_get_current());
+}
+
 static void colo_compare_tcp(CompareState *s, Connection *conn)
 {
     Packet *ppkt = NULL, *spkt = NULL;
@@ -XXX,XX +XXX,XX @@ sec:
         qemu_hexdump((char *)spkt->data, stderr,
                      "colo-compare spkt", spkt->size);
 
-        /*
-         * colo_compare_inconsistent_notify();
-         * TODO: notice to checkpoint();
-         */
+        colo_compare_inconsistency_notify();
     }
 }
 
@@ -XXX,XX +XXX,XX @@ static int colo_old_packet_check_one(Packet *pkt, int64_t *check_time)
     }
 }
 
+void colo_compare_register_notifier(Notifier *notify)
+{
+    notifier_list_add(&colo_compare_notifiers, notify);
+}
+
+void colo_compare_unregister_notifier(Notifier *notify)
+{
+    notifier_remove(notify);
+}
+
 static int colo_old_packet_check_one_conn(Connection *conn,
-                                          void *user_data)
+                                           void *user_data)
 {
     GList *result = NULL;
     int64_t check_time = REGULAR_PACKET_CHECK_MS;
@@ -XXX,XX +XXX,XX @@ static int colo_old_packet_check_one_conn(Connection *conn,
 
     if (result) {
         /* Do checkpoint will flush old packet */
-        /*
-         * TODO: Notify colo frame to do checkpoint.
-         * colo_compare_inconsistent_notify();
-         */
+        colo_compare_inconsistency_notify();
         return 0;
     }
 
@@ -XXX,XX +XXX,XX @@ static void colo_compare_packet(CompareState *s, Connection *conn,
             /*
              * If one packet arrive late, the secondary_list or
              * primary_list will be empty, so we can't compare it
-             * until next comparison.
+             * until next comparison. If the packets in the list are
+             * timeout, it will trigger a checkpoint request.
              */
             trace_colo_compare_main("packet different");
             g_queue_push_head(&conn->primary_list, pkt);
-            /* TODO: colo_notify_checkpoint();*/
+            colo_compare_inconsistency_notify();
             break;
         }
     }
diff --git a/net/colo-compare.h b/net/colo-compare.h
index XXXXXXX..XXXXXXX 100644
--- a/net/colo-compare.h
+++ b/net/colo-compare.h
@@ -XXX,XX +XXX,XX @@
 #define QEMU_COLO_COMPARE_H
 
 void colo_notify_compares_event(void *opaque, int event, Error **errp);
+void colo_compare_register_notifier(Notifier *notify);
+void colo_compare_unregister_notifier(Notifier *notify);
 
 #endif /* QEMU_COLO_COMPARE_H */
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

For COLO FT, both the PVM and SVM run at the same time,
only sync the state while it needs.

So here, let SVM runs while not doing checkpoint, change
DEFAULT_MIGRATE_X_CHECKPOINT_DELAY to 200*100.

Besides, we forgot to release colo_checkpoint_semd and
colo_delay_timer, fix them here.

diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/error-report.h"
 #include "migration/failover.h"
 #include "replication.h"
+#include "net/colo-compare.h"
+#include "net/colo.h"
 
 static bool vmstate_loading;
+static Notifier packets_compare_notifier;
 
 #define COLO_BUFFER_BASE_SIZE (4 * 1024 * 1024)
 
@@ -XXX,XX +XXX,XX @@ static int colo_do_checkpoint_transaction(MigrationState *s,
         goto out;
     }
 
+    colo_notify_compares_event(NULL, COLO_EVENT_CHECKPOINT, &local_err);
+    if (local_err) {
+        goto out;
+    }
+
     /* Disable block migration */
     migrate_set_block_enabled(false, &local_err);
     qemu_savevm_state_header(fb);
@@ -XXX,XX +XXX,XX @@ out:
     return ret;
 }
 
+static void colo_compare_notify_checkpoint(Notifier *notifier, void *data)
+{
+    colo_checkpoint_notify(data);
+}
+
 static void colo_process_checkpoint(MigrationState *s)
 {
     QIOChannelBuffer *bioc;
@@ -XXX,XX +XXX,XX @@ static void colo_process_checkpoint(MigrationState *s)
         goto out;
     }
 
+    packets_compare_notifier.notify = colo_compare_notify_checkpoint;
+    colo_compare_register_notifier(&packets_compare_notifier);
+
     /*
      * Wait for Secondary finish loading VM states and enter COLO
      * restore.
@@ -XXX,XX +XXX,XX @@ out:
         qemu_fclose(fb);
     }
 
-    timer_del(s->colo_delay_timer);
-
     /* Hope this not to be too long to wait here */
     qemu_sem_wait(&s->colo_exit_sem);
     qemu_sem_destroy(&s->colo_exit_sem);
+
+    /*
+     * It is safe to unregister notifier after failover finished.
+     * Besides, colo_delay_timer and colo_checkpoint_sem can't be
+     * released befor unregister notifier, or there will be use-after-free
+     * error.
+     */
+    colo_compare_unregister_notifier(&packets_compare_notifier);
+    timer_del(s->colo_delay_timer);
+    timer_free(s->colo_delay_timer);
+    qemu_sem_destroy(&s->colo_checkpoint_sem);
+
     /*
      * Must be called after failover BH is completed,
      * Or the failover BH may shutdown the wrong fd that
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
     fb = qemu_fopen_channel_input(QIO_CHANNEL(bioc));
     object_unref(OBJECT(bioc));
 
+    qemu_mutex_lock_iothread();
+    vm_start();
+    trace_colo_vm_state_change("stop", "run");
+    qemu_mutex_unlock_iothread();
+
     colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_READY,
                       &local_err);
     if (local_err) {
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
             goto out;
         }
 
+        qemu_mutex_lock_iothread();
+        vm_stop_force_state(RUN_STATE_COLO);
+        trace_colo_vm_state_change("run", "stop");
+        qemu_mutex_unlock_iothread();
+
         /* FIXME: This is unnecessary for periodic checkpoint mode */
         colo_send_message(mis->to_src_file, COLO_MESSAGE_CHECKPOINT_REPLY,
                      &local_err);
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
         }
 
         vmstate_loading = false;
+        vm_start();
+        trace_colo_vm_state_change("stop", "run");
         qemu_mutex_unlock_iothread();
 
         if (failover_get_state() == FAILOVER_STATUS_RELAUNCH) {
diff --git a/migration/migration.c b/migration/migration.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -XXX,XX +XXX,XX @@
 /* Migration XBZRLE default cache size */
 #define DEFAULT_MIGRATE_XBZRLE_CACHE_SIZE (64 * 1024 * 1024)
 
-/* The delay time (in ms) between two COLO checkpoints
- * Note: Please change this default value to 10000 when we support hybrid mode.
- */
-#define DEFAULT_MIGRATE_X_CHECKPOINT_DELAY 200
+/* The delay time (in ms) between two COLO checkpoints */
+#define DEFAULT_MIGRATE_X_CHECKPOINT_DELAY (200 * 100)
 #define DEFAULT_MIGRATE_MULTIFD_CHANNELS 2
 #define DEFAULT_MIGRATE_MULTIFD_PAGE_COUNT 16
 
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

Make sure master start block replication after slave's block
replication started.

Besides, we need to activate VM's blocks before goes into
COLO state.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/colo.c      | 43 +++++++++++++++++++++++++++++++++++++++++++
 migration/migration.c | 10 ++++++++++
 2 files changed, 53 insertions(+)

diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@
 #include "replication.h"
 #include "net/colo-compare.h"
 #include "net/colo.h"
+#include "block/block.h"
 
 static bool vmstate_loading;
 static Notifier packets_compare_notifier;
@@ -XXX,XX +XXX,XX @@ static void secondary_vm_do_failover(void)
 {
     int old_state;
     MigrationIncomingState *mis = migration_incoming_get_current();
+    Error *local_err = NULL;
 
     /* Can not do failover during the process of VM's loading VMstate, Or
      * it will break the secondary VM.
@@ -XXX,XX +XXX,XX @@ static void secondary_vm_do_failover(void)
     migrate_set_state(&mis->state, MIGRATION_STATUS_COLO,
                       MIGRATION_STATUS_COMPLETED);
 
+    replication_stop_all(true, &local_err);
+    if (local_err) {
+        error_report_err(local_err);
+    }
+
     if (!autostart) {
         error_report("\"-S\" qemu option will be ignored in secondary side");
         /* recover runstate to normal migration finish state */
@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
 {
     MigrationState *s = migrate_get_current();
     int old_state;
+    Error *local_err = NULL;
 
     migrate_set_state(&s->state, MIGRATION_STATUS_COLO,
                       MIGRATION_STATUS_COMPLETED);
@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
                      FailoverStatus_str(old_state));
         return;
     }
+
+    replication_stop_all(true, &local_err);
+    if (local_err) {
+        error_report_err(local_err);
+        local_err = NULL;
+    }
+
     /* Notify COLO thread that failover work is finished */
     qemu_sem_post(&s->colo_exit_sem);
 }
@@ -XXX,XX +XXX,XX @@ static int colo_do_checkpoint_transaction(MigrationState *s,
     qemu_savevm_state_header(fb);
     qemu_savevm_state_setup(fb);
     qemu_mutex_lock_iothread();
+    replication_do_checkpoint_all(&local_err);
+    if (local_err) {
+        qemu_mutex_unlock_iothread();
+        goto out;
+    }
     qemu_savevm_state_complete_precopy(fb, false, false);
     qemu_mutex_unlock_iothread();
 
@@ -XXX,XX +XXX,XX @@ static void colo_process_checkpoint(MigrationState *s)
     object_unref(OBJECT(bioc));
 
     qemu_mutex_lock_iothread();
+    replication_start_all(REPLICATION_MODE_PRIMARY, &local_err);
+    if (local_err) {
+        qemu_mutex_unlock_iothread();
+        goto out;
+    }
+
     vm_start();
     qemu_mutex_unlock_iothread();
     trace_colo_vm_state_change("stop", "run");
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
     object_unref(OBJECT(bioc));
 
     qemu_mutex_lock_iothread();
+    replication_start_all(REPLICATION_MODE_SECONDARY, &local_err);
+    if (local_err) {
+        qemu_mutex_unlock_iothread();
+        goto out;
+    }
     vm_start();
     trace_colo_vm_state_change("stop", "run");
     qemu_mutex_unlock_iothread();
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
             goto out;
         }
 
+        replication_get_error_all(&local_err);
+        if (local_err) {
+            qemu_mutex_unlock_iothread();
+            goto out;
+        }
+        /* discard colo disk buffer */
+        replication_do_checkpoint_all(&local_err);
+        if (local_err) {
+            qemu_mutex_unlock_iothread();
+            goto out;
+        }
+
         vmstate_loading = false;
         vm_start();
         trace_colo_vm_state_change("stop", "run");
diff --git a/migration/migration.c b/migration/migration.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
     MigrationIncomingState *mis = migration_incoming_get_current();
     PostcopyState ps;
     int ret;
+    Error *local_err = NULL;
 
     assert(mis->from_src_file);
     mis->migration_incoming_co = qemu_coroutine_self();
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
 
     /* we get COLO info, and know if we are in COLO mode */
     if (!ret && migration_incoming_enable_colo()) {
+        /* Make sure all file formats flush their mutable metadata */
+        bdrv_invalidate_cache_all(&local_err);
+        if (local_err) {
+            migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
+                    MIGRATION_STATUS_FAILED);
+            error_report_err(local_err);
+            exit(EXIT_FAILURE);
+        }
+
         qemu_thread_create(&mis->colo_incoming_thread, "COLO incoming",
              colo_process_incoming_thread, mis, QEMU_THREAD_JOINABLE);
         mis->have_colo_incoming_thread = true;
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

We need to know if migration is going into COLO state for
incoming side before start normal migration.

Instead by using the VMStateDescription to send colo_state
from source side to destination side, we use MIG_CMD_ENABLE_COLO
to indicate whether COLO is enabled or not.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 include/migration/colo.h |  5 ++--
 migration/Makefile.objs  |  2 +-
 migration/colo-comm.c    | 76 ------------------------------------------------
 migration/colo.c         | 13 ++++++++-
 migration/migration.c    | 23 ++++++++++++++-
 migration/savevm.c       | 17 +++++++++++
 migration/savevm.h       |  1 +
 migration/trace-events   |  1 +
 vl.c                     |  2 --
 9 files changed, 57 insertions(+), 83 deletions(-)
 delete mode 100644 migration/colo-comm.c

diff --git a/include/migration/colo.h b/include/migration/colo.h
index XXXXXXX..XXXXXXX 100644
--- a/include/migration/colo.h
+++ b/include/migration/colo.h
@@ -XXX,XX +XXX,XX @@ void migrate_start_colo_process(MigrationState *s);
 bool migration_in_colo_state(void);
 
 /* loadvm */
-bool migration_incoming_enable_colo(void);
-void migration_incoming_exit_colo(void);
+void migration_incoming_enable_colo(void);
+void migration_incoming_disable_colo(void);
+bool migration_incoming_colo_enabled(void);
 void *colo_process_incoming_thread(void *opaque);
 bool migration_incoming_in_colo_state(void);
 
diff --git a/migration/Makefile.objs b/migration/Makefile.objs
index XXXXXXX..XXXXXXX 100644
--- a/migration/Makefile.objs
+++ b/migration/Makefile.objs
@@ -XXX,XX +XXX,XX @@
 common-obj-y += migration.o socket.o fd.o exec.o
 common-obj-y += tls.o channel.o savevm.o
-common-obj-y += colo-comm.o colo.o colo-failover.o
+common-obj-y += colo.o colo-failover.o
 common-obj-y += vmstate.o vmstate-types.o page_cache.o
 common-obj-y += qemu-file.o global_state.o
 common-obj-y += qemu-file-channel.o
diff --git a/migration/colo-comm.c b/migration/colo-comm.c
deleted file mode 100644
index XXXXXXX..XXXXXXX
--- a/migration/colo-comm.c
+++ /dev/null
@@ -XXX,XX +XXX,XX @@
-/*
- * COarse-grain LOck-stepping Virtual Machines for Non-stop Service (COLO)
- * (a.k.a. Fault Tolerance or Continuous Replication)
- *
- * Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
- * Copyright (c) 2016 FUJITSU LIMITED
- * Copyright (c) 2016 Intel Corporation
- *
- * This work is licensed under the terms of the GNU GPL, version 2 or
- * later. See the COPYING file in the top-level directory.
- *
- */
-
-#include "qemu/osdep.h"
-#include "migration.h"
-#include "migration/colo.h"
-#include "migration/vmstate.h"
-#include "trace.h"
-
-typedef struct {
-     bool colo_requested;
-} COLOInfo;
-
-static COLOInfo colo_info;
-
-COLOMode get_colo_mode(void)
-{
-    if (migration_in_colo_state()) {
-        return COLO_MODE_PRIMARY;
-    } else if (migration_incoming_in_colo_state()) {
-        return COLO_MODE_SECONDARY;
-    } else {
-        return COLO_MODE_UNKNOWN;
-    }
-}
-
-static int colo_info_pre_save(void *opaque)
-{
-    COLOInfo *s = opaque;
-
-    s->colo_requested = migrate_colo_enabled();
-
-    return 0;
-}
-
-static bool colo_info_need(void *opaque)
-{
-   return migrate_colo_enabled();
-}
-
-static const VMStateDescription colo_state = {
-    .name = "COLOState",
-    .version_id = 1,
-    .minimum_version_id = 1,
-    .pre_save = colo_info_pre_save,
-    .needed = colo_info_need,
-    .fields = (VMStateField[]) {
-        VMSTATE_BOOL(colo_requested, COLOInfo),
-        VMSTATE_END_OF_LIST()
-    },
-};
-
-void colo_info_init(void)
-{
-    vmstate_register(NULL, 0, &colo_state, &colo_info);
-}
-
-bool migration_incoming_enable_colo(void)
-{
-    return colo_info.colo_requested;
-}
-
-void migration_incoming_exit_colo(void)
-{
-    colo_info.colo_requested = false;
-}
diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
     qemu_sem_post(&s->colo_exit_sem);
 }
 
+COLOMode get_colo_mode(void)
+{
+    if (migration_in_colo_state()) {
+        return COLO_MODE_PRIMARY;
+    } else if (migration_incoming_in_colo_state()) {
+        return COLO_MODE_SECONDARY;
+    } else {
+        return COLO_MODE_UNKNOWN;
+    }
+}
+
 void colo_do_failover(MigrationState *s)
 {
     /* Make sure VM stopped while failover happened. */
@@ -XXX,XX +XXX,XX @@ out:
     if (mis->to_src_file) {
         qemu_fclose(mis->to_src_file);
     }
-    migration_incoming_exit_colo();
+    migration_incoming_disable_colo();
 
     rcu_unregister_thread();
     return NULL;
diff --git a/migration/migration.c b/migration/migration.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -XXX,XX +XXX,XX @@ int migrate_send_rp_req_pages(MigrationIncomingState *mis, const char *rbname,
     return migrate_send_rp_message(mis, msg_type, msglen, bufc);
 }
 
+static bool migration_colo_enabled;
+bool migration_incoming_colo_enabled(void)
+{
+    return migration_colo_enabled;
+}
+
+void migration_incoming_disable_colo(void)
+{
+    migration_colo_enabled = false;
+}
+
+void migration_incoming_enable_colo(void)
+{
+    migration_colo_enabled = true;
+}
+
 void qemu_start_incoming_migration(const char *uri, Error **errp)
 {
     const char *p;
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
     }
 
     /* we get COLO info, and know if we are in COLO mode */
-    if (!ret && migration_incoming_enable_colo()) {
+    if (!ret && migration_incoming_colo_enabled()) {
         /* Make sure all file formats flush their mutable metadata */
         bdrv_invalidate_cache_all(&local_err);
         if (local_err) {
@@ -XXX,XX +XXX,XX @@ static void *migration_thread(void *opaque)
         qemu_savevm_send_postcopy_advise(s->to_dst_file);
     }
 
+    if (migrate_colo_enabled()) {
+        /* Notify migration destination that we enable COLO */
+        qemu_savevm_send_colo_enable(s->to_dst_file);
+    }
+
     qemu_savevm_state_setup(s->to_dst_file);
 
     s->setup_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) - setup_start;
diff --git a/migration/savevm.c b/migration/savevm.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -XXX,XX +XXX,XX @@
 #include "io/channel-file.h"
 #include "sysemu/replay.h"
 #include "qjson.h"
+#include "migration/colo.h"
 
 #ifndef ETH_P_RARP
 #define ETH_P_RARP 0x8035
@@ -XXX,XX +XXX,XX @@ enum qemu_vm_cmd {
                                       were previously sent during
                                       precopy but are dirty. */
     MIG_CMD_PACKAGED,          /* Send a wrapped stream within this stream */
+    MIG_CMD_ENABLE_COLO,       /* Enable COLO */
     MIG_CMD_POSTCOPY_RESUME,   /* resume postcopy on dest */
     MIG_CMD_RECV_BITMAP,       /* Request for recved bitmap on dst */
     MIG_CMD_MAX
@@ -XXX,XX +XXX,XX @@ static void qemu_savevm_command_send(QEMUFile *f,
     qemu_fflush(f);
 }
 
+void qemu_savevm_send_colo_enable(QEMUFile *f)
+{
+    trace_savevm_send_colo_enable();
+    qemu_savevm_command_send(f, MIG_CMD_ENABLE_COLO, 0, NULL);
+}
+
 void qemu_savevm_send_ping(QEMUFile *f, uint32_t value)
 {
     uint32_t buf;
@@ -XXX,XX +XXX,XX @@ static int loadvm_handle_recv_bitmap(MigrationIncomingState *mis,
     return 0;
 }
 
+static int loadvm_process_enable_colo(MigrationIncomingState *mis)
+{
+    migration_incoming_enable_colo();
+    return 0;
+}
+
 /*
  * Process an incoming 'QEMU_VM_COMMAND'
  * 0           just a normal return
@@ -XXX,XX +XXX,XX @@ static int loadvm_process_command(QEMUFile *f)
 
     case MIG_CMD_RECV_BITMAP:
         return loadvm_handle_recv_bitmap(mis, len);
+
+    case MIG_CMD_ENABLE_COLO:
+        return loadvm_process_enable_colo(mis);
     }
 
     return 0;
diff --git a/migration/savevm.h b/migration/savevm.h
index XXXXXXX..XXXXXXX 100644
--- a/migration/savevm.h
+++ b/migration/savevm.h
@@ -XXX,XX +XXX,XX @@ void qemu_savevm_send_postcopy_ram_discard(QEMUFile *f, const char *name,
                                            uint16_t len,
                                            uint64_t *start_list,
                                            uint64_t *length_list);
+void qemu_savevm_send_colo_enable(QEMUFile *f);
 
 int qemu_loadvm_state(QEMUFile *f);
 void qemu_loadvm_state_cleanup(void);
diff --git a/migration/trace-events b/migration/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/migration/trace-events
+++ b/migration/trace-events
@@ -XXX,XX +XXX,XX @@ savevm_send_ping(uint32_t val) "0x%x"
 savevm_send_postcopy_listen(void) ""
 savevm_send_postcopy_run(void) ""
 savevm_send_postcopy_resume(void) ""
+savevm_send_colo_enable(void) ""
 savevm_send_recv_bitmap(char *name) "%s"
 savevm_state_setup(void) ""
 savevm_state_resume_prepare(void) ""
diff --git a/vl.c b/vl.c
index XXXXXXX..XXXXXXX 100644
--- a/vl.c
+++ b/vl.c
@@ -XXX,XX +XXX,XX @@ int main(int argc, char **argv, char **envp)
 #endif
     }
 
-    colo_info_init();
-
     if (net_init_clients(&err) < 0) {
         error_report_err(err);
         exit(1);
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

We should not load PVM's state directly into SVM, because there maybe some
errors happen when SVM is receving data, which will break SVM.

We need to ensure receving all data before load the state into SVM. We use
an extra memory to cache these data (PVM's ram). The ram cache in secondary side
is initially the same as SVM/PVM's memory. And in the process of checkpoint,
we cache the dirty pages of PVM into this ram cache firstly, so this ram cache
always the same as PVM's memory at every checkpoint, then we flush this cached ram
to SVM after we receive all PVM's state.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 include/exec/ram_addr.h |  1 +
 migration/migration.c   |  7 +++++
 migration/ram.c         | 83 +++++++++++++++++++++++++++++++++++++++++++++++--
 migration/ram.h         |  4 +++
 migration/savevm.c      |  2 +-
 5 files changed, 94 insertions(+), 3 deletions(-)

diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
index XXXXXXX..XXXXXXX 100644
--- a/include/exec/ram_addr.h
+++ b/include/exec/ram_addr.h
@@ -XXX,XX +XXX,XX @@ struct RAMBlock {
     struct rcu_head rcu;
     struct MemoryRegion *mr;
     uint8_t *host;
+    uint8_t *colo_cache; /* For colo, VM's ram cache */
     ram_addr_t offset;
     ram_addr_t used_length;
     ram_addr_t max_length;
diff --git a/migration/migration.c b/migration/migration.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
             exit(EXIT_FAILURE);
         }
 
+        if (colo_init_ram_cache() < 0) {
+            error_report("Init ram cache failed");
+            exit(EXIT_FAILURE);
+        }
+
         qemu_thread_create(&mis->colo_incoming_thread, "COLO incoming",
              colo_process_incoming_thread, mis, QEMU_THREAD_JOINABLE);
         mis->have_colo_incoming_thread = true;
@@ -XXX,XX +XXX,XX @@ static void process_incoming_migration_co(void *opaque)
 
         /* Wait checkpoint incoming thread exit before free resource */
         qemu_thread_join(&mis->colo_incoming_thread);
+        /* We hold the global iothread lock, so it is safe here */
+        colo_release_ram_cache();
     }
 
     if (ret < 0) {
diff --git a/migration/ram.c b/migration/ram.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -XXX,XX +XXX,XX @@ static inline void *host_from_ram_block_offset(RAMBlock *block,
     return block->host + offset;
 }
 
+static inline void *colo_cache_from_block_offset(RAMBlock *block,
+                                                 ram_addr_t offset)
+{
+    if (!offset_in_ramblock(block, offset)) {
+        return NULL;
+    }
+    if (!block->colo_cache) {
+        error_report("%s: colo_cache is NULL in block :%s",
+                     __func__, block->idstr);
+        return NULL;
+    }
+    return block->colo_cache + offset;
+}
+
 /**
  * ram_handle_compressed: handle the zero page case
  *
@@ -XXX,XX +XXX,XX @@ static void decompress_data_with_multi_threads(QEMUFile *f,
     qemu_mutex_unlock(&decomp_done_lock);
 }
 
+/*
+ * colo cache: this is for secondary VM, we cache the whole
+ * memory of the secondary VM, it is need to hold the global lock
+ * to call this helper.
+ */
+int colo_init_ram_cache(void)
+{
+    RAMBlock *block;
+
+    rcu_read_lock();
+    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+        block->colo_cache = qemu_anon_ram_alloc(block->used_length,
+                                                NULL,
+                                                false);
+        if (!block->colo_cache) {
+            error_report("%s: Can't alloc memory for COLO cache of block %s,"
+                         "size 0x" RAM_ADDR_FMT, __func__, block->idstr,
+                         block->used_length);
+            goto out_locked;
+        }
+        memcpy(block->colo_cache, block->host, block->used_length);
+    }
+    rcu_read_unlock();
+    return 0;
+
+out_locked:
+    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+        if (block->colo_cache) {
+            qemu_anon_ram_free(block->colo_cache, block->used_length);
+            block->colo_cache = NULL;
+        }
+    }
+
+    rcu_read_unlock();
+    return -errno;
+}
+
+/* It is need to hold the global lock to call this helper */
+void colo_release_ram_cache(void)
+{
+    RAMBlock *block;
+
+    rcu_read_lock();
+    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+        if (block->colo_cache) {
+            qemu_anon_ram_free(block->colo_cache, block->used_length);
+            block->colo_cache = NULL;
+        }
+    }
+    rcu_read_unlock();
+}
+
 /**
  * ram_load_setup: Setup RAM for migration incoming side
  *
@@ -XXX,XX +XXX,XX @@ static int ram_load_setup(QEMUFile *f, void *opaque)
 
     xbzrle_load_setup();
     ramblock_recv_map_init();
+
     return 0;
 }
 
@@ -XXX,XX +XXX,XX @@ static int ram_load_cleanup(void *opaque)
         g_free(rb->receivedmap);
         rb->receivedmap = NULL;
     }
+
     return 0;
 }
 
@@ -XXX,XX +XXX,XX @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
                      RAM_SAVE_FLAG_COMPRESS_PAGE | RAM_SAVE_FLAG_XBZRLE)) {
             RAMBlock *block = ram_block_from_stream(f, flags);
 
-            host = host_from_ram_block_offset(block, addr);
+            /*
+             * After going into COLO, we should load the Page into colo_cache.
+             */
+            if (migration_incoming_in_colo_state()) {
+                host = colo_cache_from_block_offset(block, addr);
+            } else {
+                host = host_from_ram_block_offset(block, addr);
+            }
             if (!host) {
                 error_report("Illegal RAM offset " RAM_ADDR_FMT, addr);
                 ret = -EINVAL;
                 break;
             }
-            ramblock_recv_bitmap_set(block, host);
+
+            if (!migration_incoming_in_colo_state()) {
+                ramblock_recv_bitmap_set(block, host);
+            }
+
             trace_ram_load_loop(block->idstr, (uint64_t)addr, flags, host);
         }
 
diff --git a/migration/ram.h b/migration/ram.h
index XXXXXXX..XXXXXXX 100644
--- a/migration/ram.h
+++ b/migration/ram.h
@@ -XXX,XX +XXX,XX @@ int64_t ramblock_recv_bitmap_send(QEMUFile *file,
                                   const char *block_name);
 int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *rb);
 
+/* ram cache */
+int colo_init_ram_cache(void);
+void colo_release_ram_cache(void);
+
 #endif
diff --git a/migration/savevm.c b/migration/savevm.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -XXX,XX +XXX,XX @@ static int loadvm_handle_recv_bitmap(MigrationIncomingState *mis,
 static int loadvm_process_enable_colo(MigrationIncomingState *mis)
 {
     migration_incoming_enable_colo();
-    return 0;
+    return colo_init_ram_cache();
 }
 
 /*
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

We record the address of the dirty pages that received,
it will help flushing pages that cached into SVM.

Here, it is a trick, we record dirty pages by re-using migration
dirty bitmap. In the later patch, we will start the dirty log
for SVM, just like migration, in this way, we can record both
the dirty pages caused by PVM and SVM, we only flush those dirty
pages from RAM cache while do checkpoint.

diff --git a/migration/ram.c b/migration/ram.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -XXX,XX +XXX,XX @@ static inline void *colo_cache_from_block_offset(RAMBlock *block,
                      __func__, block->idstr);
         return NULL;
     }
+
+    /*
+    * During colo checkpoint, we need bitmap of these migrated pages.
+    * It help us to decide which pages in ram cache should be flushed
+    * into VM's RAM later.
+    */
+    if (!test_and_set_bit(offset >> TARGET_PAGE_BITS, block->bmap)) {
+        ram_state->migration_dirty_pages++;
+    }
     return block->colo_cache + offset;
 }
 
@@ -XXX,XX +XXX,XX @@ int colo_init_ram_cache(void)
     RAMBlock *block;
 
     rcu_read_lock();
-    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+    RAMBLOCK_FOREACH_MIGRATABLE(block) {
         block->colo_cache = qemu_anon_ram_alloc(block->used_length,
                                                 NULL,
                                                 false);
@@ -XXX,XX +XXX,XX @@ int colo_init_ram_cache(void)
         memcpy(block->colo_cache, block->host, block->used_length);
     }
     rcu_read_unlock();
+    /*
+    * Record the dirty pages that sent by PVM, we use this dirty bitmap together
+    * with to decide which page in cache should be flushed into SVM's RAM. Here
+    * we use the same name 'ram_bitmap' as for migration.
+    */
+    if (ram_bytes_total()) {
+        RAMBlock *block;
+
+        RAMBLOCK_FOREACH_MIGRATABLE(block) {
+            unsigned long pages = block->max_length >> TARGET_PAGE_BITS;
+
+            block->bmap = bitmap_new(pages);
+            bitmap_set(block->bmap, 0, pages);
+        }
+    }
+    ram_state = g_new0(RAMState, 1);
+    ram_state->migration_dirty_pages = 0;
+
     return 0;
 
 out_locked:
-    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+
+    RAMBLOCK_FOREACH_MIGRATABLE(block) {
         if (block->colo_cache) {
             qemu_anon_ram_free(block->colo_cache, block->used_length);
             block->colo_cache = NULL;
@@ -XXX,XX +XXX,XX @@ void colo_release_ram_cache(void)
 {
     RAMBlock *block;
 
+    RAMBLOCK_FOREACH_MIGRATABLE(block) {
+        g_free(block->bmap);
+        block->bmap = NULL;
+    }
+
     rcu_read_lock();
-    QLIST_FOREACH_RCU(block, &ram_list.blocks, next) {
+
+    RAMBLOCK_FOREACH_MIGRATABLE(block) {
         if (block->colo_cache) {
             qemu_anon_ram_free(block->colo_cache, block->used_length);
             block->colo_cache = NULL;
         }
     }
+
     rcu_read_unlock();
+    g_free(ram_state);
+    ram_state = NULL;
 }
 
 /**
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

During the time of VM's running, PVM may dirty some pages, we will transfer
PVM's dirty pages to SVM and store them into SVM's RAM cache at next checkpoint
time. So, the content of SVM's RAM cache will always be same with PVM's memory
after checkpoint.

Instead of flushing all content of PVM's RAM cache into SVM's MEMORY,
we do this in a more efficient way:
Only flush any page that dirtied by PVM since last checkpoint.
In this way, we can ensure SVM's memory same with PVM's.

Besides, we must ensure flush RAM cache before load device state.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/ram.c        | 37 +++++++++++++++++++++++++++++++++++++
 migration/trace-events |  2 ++
 2 files changed, 39 insertions(+)

diff --git a/migration/ram.c b/migration/ram.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -XXX,XX +XXX,XX @@ static bool postcopy_is_running(void)
     return ps >= POSTCOPY_INCOMING_LISTENING && ps < POSTCOPY_INCOMING_END;
 }
 
+/*
+ * Flush content of RAM cache into SVM's memory.
+ * Only flush the pages that be dirtied by PVM or SVM or both.
+ */
+static void colo_flush_ram_cache(void)
+{
+    RAMBlock *block = NULL;
+    void *dst_host;
+    void *src_host;
+    unsigned long offset = 0;
+
+    trace_colo_flush_ram_cache_begin(ram_state->migration_dirty_pages);
+    rcu_read_lock();
+    block = QLIST_FIRST_RCU(&ram_list.blocks);
+
+    while (block) {
+        offset = migration_bitmap_find_dirty(ram_state, block, offset);
+
+        if (offset << TARGET_PAGE_BITS >= block->used_length) {
+            offset = 0;
+            block = QLIST_NEXT_RCU(block, next);
+        } else {
+            migration_bitmap_clear_dirty(ram_state, block, offset);
+            dst_host = block->host + (offset << TARGET_PAGE_BITS);
+            src_host = block->colo_cache + (offset << TARGET_PAGE_BITS);
+            memcpy(dst_host, src_host, TARGET_PAGE_SIZE);
+        }
+    }
+
+    rcu_read_unlock();
+    trace_colo_flush_ram_cache_end();
+}
+
 static int ram_load(QEMUFile *f, void *opaque, int version_id)
 {
     int flags = 0, ret = 0, invalid_flags = 0;
@@ -XXX,XX +XXX,XX @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
     ret |= wait_for_decompress_done();
     rcu_read_unlock();
     trace_ram_load_complete(ret, seq_iter);
+
+    if (!ret  && migration_incoming_in_colo_state()) {
+        colo_flush_ram_cache();
+    }
     return ret;
 }
 
diff --git a/migration/trace-events b/migration/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/migration/trace-events
+++ b/migration/trace-events
@@ -XXX,XX +XXX,XX @@ ram_dirty_bitmap_sync_start(void) ""
 ram_dirty_bitmap_sync_wait(void) ""
 ram_dirty_bitmap_sync_complete(void) ""
 ram_state_resume_prepare(uint64_t v) "%" PRId64
+colo_flush_ram_cache_begin(uint64_t dirty_pages) "dirty_pages %" PRIu64
+colo_flush_ram_cache_end(void) ""
 
 # migration/migration.c
 await_return_path_close_on_source_close(void) ""
-- 
2.5.0

From: zhanghailiang <zhang.zhanghailiang@huawei.com>

If some errors happen during VM's COLO FT stage, it's important to
notify the users of this event. Together with 'x-colo-lost-heartbeat',
Users can intervene in COLO's failover work immediately.
If users don't want to get involved in COLO's failover verdict,
it is still necessary to notify users that we exited COLO mode.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/colo.c    | 31 +++++++++++++++++++++++++++++++
 qapi/migration.json | 38 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 69 insertions(+)

diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@
 #include "net/colo-compare.h"
 #include "net/colo.h"
 #include "block/block.h"
+#include "qapi/qapi-events-migration.h"
 
 static bool vmstate_loading;
 static Notifier packets_compare_notifier;
@@ -XXX,XX +XXX,XX @@ out:
         qemu_fclose(fb);
     }
 
+    /*
+     * There are only two reasons we can get here, some error happened
+     * or the user triggered failover.
+     */
+    switch (failover_get_state()) {
+    case FAILOVER_STATUS_NONE:
+        qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
+                                  COLO_EXIT_REASON_ERROR);
+        break;
+    case FAILOVER_STATUS_REQUIRE:
+        qapi_event_send_colo_exit(COLO_MODE_PRIMARY,
+                                  COLO_EXIT_REASON_REQUEST);
+        break;
+    default:
+        abort();
+    }
+
     /* Hope this not to be too long to wait here */
     qemu_sem_wait(&s->colo_exit_sem);
     qemu_sem_destroy(&s->colo_exit_sem);
@@ -XXX,XX +XXX,XX @@ out:
         error_report_err(local_err);
     }
 
+    switch (failover_get_state()) {
+    case FAILOVER_STATUS_NONE:
+        qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
+                                  COLO_EXIT_REASON_ERROR);
+        break;
+    case FAILOVER_STATUS_REQUIRE:
+        qapi_event_send_colo_exit(COLO_MODE_SECONDARY,
+                                  COLO_EXIT_REASON_REQUEST);
+        break;
+    default:
+        abort();
+    }
+
     if (fb) {
         qemu_fclose(fb);
     }
diff --git a/qapi/migration.json b/qapi/migration.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -XXX,XX +XXX,XX @@
   'data': [ 'none', 'require', 'active', 'completed', 'relaunch' ] }
 
 ##
+# @COLO_EXIT:
+#
+# Emitted when VM finishes COLO mode due to some errors happening or
+# at the request of users.
+#
+# @mode: report COLO mode when COLO exited.
+#
+# @reason: describes the reason for the COLO exit.
+#
+# Since: 3.1
+#
+# Example:
+#
+# <- { "timestamp": {"seconds": 2032141960, "microseconds": 417172},
+#      "event": "COLO_EXIT", "data": {"mode": "primary", "reason": "request" } }
+#
+##
+{ 'event': 'COLO_EXIT',
+  'data': {'mode': 'COLOMode', 'reason': 'COLOExitReason' } }
+
+##
+# @COLOExitReason:
+#
+# The reason for a COLO exit
+#
+# @none: no failover has ever happened. This can't occur in the
+# COLO_EXIT event, only in the result of query-colo-status.
+#
+# @request: COLO exit is due to an external request
+#
+# @error: COLO exit is due to an internal error
+#
+# Since: 3.1
+##
+{ 'enum': 'COLOExitReason',
+  'data': [ 'none', 'request', 'error' ] }
+
+##
 # @x-colo-lost-heartbeat:
 #
 # Tell qemu that heartbeat is lost, request it to do takeover procedures.
-- 
2.5.0

From: Zhang Chen <chen.zhang@intel.com>

Suggested by Markus Armbruster rename COLO unknown mode to none mode.

Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/colo-failover.c |  2 +-
 migration/colo.c          |  2 +-
 qapi/migration.json       | 10 +++++-----
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/migration/colo-failover.c b/migration/colo-failover.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo-failover.c
+++ b/migration/colo-failover.c
@@ -XXX,XX +XXX,XX @@ FailoverStatus failover_get_state(void)
 
 void qmp_x_colo_lost_heartbeat(Error **errp)
 {
-    if (get_colo_mode() == COLO_MODE_UNKNOWN) {
+    if (get_colo_mode() == COLO_MODE_NONE) {
         error_setg(errp, QERR_FEATURE_DISABLED, "colo");
         return;
     }
diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@ COLOMode get_colo_mode(void)
     } else if (migration_incoming_in_colo_state()) {
         return COLO_MODE_SECONDARY;
     } else {
-        return COLO_MODE_UNKNOWN;
+        return COLO_MODE_NONE;
     }
 }
 
diff --git a/qapi/migration.json b/qapi/migration.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -XXX,XX +XXX,XX @@
 ##
 # @COLOMode:
 #
-# The colo mode
+# The COLO current mode.
 #
-# @unknown: unknown mode
+# @none: COLO is disabled.
 #
-# @primary: master side
+# @primary: COLO node in primary side.
 #
-# @secondary: slave side
+# @secondary: COLO node in slave side.
 #
 # Since: 2.8
 ##
 { 'enum': 'COLOMode',
-  'data': [ 'unknown', 'primary', 'secondary'] }
+  'data': [ 'none', 'primary', 'secondary'] }
 
 ##
 # @FailoverStatus:
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

Libvirt or other high level software can use this command query colo status.
You can test this command like that:
{'execute':'query-colo-status'}

Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/colo.c    | 21 +++++++++++++++++++++
 qapi/migration.json | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 53 insertions(+)

diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@
 #include "net/colo.h"
 #include "block/block.h"
 #include "qapi/qapi-events-migration.h"
+#include "qapi/qmp/qerror.h"
 
 static bool vmstate_loading;
 static Notifier packets_compare_notifier;
@@ -XXX,XX +XXX,XX @@ void qmp_xen_colo_do_checkpoint(Error **errp)
 #endif
 }
 
+COLOStatus *qmp_query_colo_status(Error **errp)
+{
+    COLOStatus *s = g_new0(COLOStatus, 1);
+
+    s->mode = get_colo_mode();
+
+    switch (failover_get_state()) {
+    case FAILOVER_STATUS_NONE:
+        s->reason = COLO_EXIT_REASON_NONE;
+        break;
+    case FAILOVER_STATUS_REQUIRE:
+        s->reason = COLO_EXIT_REASON_REQUEST;
+        break;
+    default:
+        s->reason = COLO_EXIT_REASON_ERROR;
+    }
+
+    return s;
+}
+
 static void colo_send_message(QEMUFile *f, COLOMessage msg,
                               Error **errp)
 {
diff --git a/qapi/migration.json b/qapi/migration.json
index XXXXXXX..XXXXXXX 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -XXX,XX +XXX,XX @@
 { 'command': 'xen-colo-do-checkpoint' }
 
 ##
+# @COLOStatus:
+#
+# The result format for 'query-colo-status'.
+#
+# @mode: COLO running mode. If COLO is running, this field will return
+#        'primary' or 'secondary'.
+#
+# @reason: describes the reason for the COLO exit.
+#
+# Since: 3.0
+##
+{ 'struct': 'COLOStatus',
+  'data': { 'mode': 'COLOMode', 'reason': 'COLOExitReason' } }
+
+##
+# @query-colo-status:
+#
+# Query COLO status while the vm is running.
+#
+# Returns: A @COLOStatus object showing the status.
+#
+# Example:
+#
+# -> { "execute": "query-colo-status" }
+# <- { "return": { "mode": "primary", "active": true, "reason": "request" } }
+#
+# Since: 3.0
+##
+{ 'command': 'query-colo-status',
+  'returns': 'COLOStatus' }
+
+##
 # @migrate-recover:
 #
 # Provide a recovery migration stream URI.
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

There are several stages during loadvm/savevm process. In different stage,
migration incoming processes different types of sections.
We want to control these stages more accuracy, it will benefit COLO
performance, we don't have to save type of QEMU_VM_SECTION_START
sections everytime while do checkpoint, besides, we want to separate
the process of saving/loading memory and devices state.

So we add three new helper functions: qemu_load_device_state() and
qemu_savevm_live_state() to achieve different process during migration.

Besides, we make qemu_loadvm_state_main() and qemu_save_device_state()
public, and simplify the codes of qemu_save_device_state() by calling the
wrapper qemu_savevm_state_header().

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/colo.c   | 41 ++++++++++++++++++++++++++++++++---------
 migration/savevm.c | 36 +++++++++++++++++++++++++++++-------
 migration/savevm.h |  4 ++++
 3 files changed, 65 insertions(+), 16 deletions(-)

diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@
 #include "block/block.h"
 #include "qapi/qapi-events-migration.h"
 #include "qapi/qmp/qerror.h"
+#include "sysemu/cpus.h"
 
 static bool vmstate_loading;
 static Notifier packets_compare_notifier;
@@ -XXX,XX +XXX,XX @@ static int colo_do_checkpoint_transaction(MigrationState *s,
 
     /* Disable block migration */
     migrate_set_block_enabled(false, &local_err);
-    qemu_savevm_state_header(fb);
-    qemu_savevm_state_setup(fb);
     qemu_mutex_lock_iothread();
     replication_do_checkpoint_all(&local_err);
     if (local_err) {
         qemu_mutex_unlock_iothread();
         goto out;
     }
-    qemu_savevm_state_complete_precopy(fb, false, false);
-    qemu_mutex_unlock_iothread();
-
-    qemu_fflush(fb);
 
     colo_send_message(s->to_dst_file, COLO_MESSAGE_VMSTATE_SEND, &local_err);
     if (local_err) {
+        qemu_mutex_unlock_iothread();
+        goto out;
+    }
+    /* Note: device state is saved into buffer */
+    ret = qemu_save_device_state(fb);
+
+    qemu_mutex_unlock_iothread();
+    if (ret < 0) {
         goto out;
     }
     /*
+     * Only save VM's live state, which not including device state.
+     * TODO: We may need a timeout mechanism to prevent COLO process
+     * to be blocked here.
+     */
+    qemu_savevm_live_state(s->to_dst_file);
+
+    qemu_fflush(fb);
+
+    /*
      * We need the size of the VMstate data in Secondary side,
      * With which we can decide how much data should be read.
      */
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
     uint64_t total_size;
     uint64_t value;
     Error *local_err = NULL;
+    int ret;
 
     rcu_register_thread();
     qemu_sem_init(&mis->colo_incoming_sem, 0);
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
             goto out;
         }
 
+        qemu_mutex_lock_iothread();
+        cpu_synchronize_all_pre_loadvm();
+        ret = qemu_loadvm_state_main(mis->from_src_file, mis);
+        qemu_mutex_unlock_iothread();
+
+        if (ret < 0) {
+            error_report("Load VM's live state (ram) error");
+            goto out;
+        }
+
         value = colo_receive_message_value(mis->from_src_file,
                                  COLO_MESSAGE_VMSTATE_SIZE, &local_err);
         if (local_err) {
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
         }
 
         qemu_mutex_lock_iothread();
-        qemu_system_reset(SHUTDOWN_CAUSE_NONE);
         vmstate_loading = true;
-        if (qemu_loadvm_state(fb) < 0) {
-            error_report("COLO: loadvm failed");
+        ret = qemu_load_device_state(fb);
+        if (ret < 0) {
+            error_report("COLO: load device state failed");
             qemu_mutex_unlock_iothread();
             goto out;
         }
diff --git a/migration/savevm.c b/migration/savevm.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -XXX,XX +XXX,XX @@ done:
     return ret;
 }
 
-static int qemu_save_device_state(QEMUFile *f)
+void qemu_savevm_live_state(QEMUFile *f)
 {
-    SaveStateEntry *se;
+    /* save QEMU_VM_SECTION_END section */
+    qemu_savevm_state_complete_precopy(f, true, false);
+    qemu_put_byte(f, QEMU_VM_EOF);
+}
 
-    qemu_put_be32(f, QEMU_VM_FILE_MAGIC);
-    qemu_put_be32(f, QEMU_VM_FILE_VERSION);
+int qemu_save_device_state(QEMUFile *f)
+{
+    SaveStateEntry *se;
 
+    if (!migration_in_colo_state()) {
+        qemu_put_be32(f, QEMU_VM_FILE_MAGIC);
+        qemu_put_be32(f, QEMU_VM_FILE_VERSION);
+    }
     cpu_synchronize_all_states();
 
     QTAILQ_FOREACH(se, &savevm_state.handlers, entry) {
@@ -XXX,XX +XXX,XX @@ enum LoadVMExitCodes {
     LOADVM_QUIT     =  1,
 };
 
-static int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis);
-
 /* ------ incoming postcopy messages ------ */
 /* 'advise' arrives before any transfers just to tell us that a postcopy
  * *might* happen - it might be skipped if precopy transferred everything
@@ -XXX,XX +XXX,XX @@ static bool postcopy_pause_incoming(MigrationIncomingState *mis)
     return true;
 }
 
-static int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis)
+int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis)
 {
     uint8_t section_type;
     int ret = 0;
@@ -XXX,XX +XXX,XX @@ int qemu_loadvm_state(QEMUFile *f)
     return ret;
 }
 
+int qemu_load_device_state(QEMUFile *f)
+{
+    MigrationIncomingState *mis = migration_incoming_get_current();
+    int ret;
+
+    /* Load QEMU_VM_SECTION_FULL section */
+    ret = qemu_loadvm_state_main(f, mis);
+    if (ret < 0) {
+        error_report("Failed to load device state: %d", ret);
+        return ret;
+    }
+
+    cpu_synchronize_all_post_init();
+    return 0;
+}
+
 int save_snapshot(const char *name, Error **errp)
 {
     BlockDriverState *bs, *bs1;
diff --git a/migration/savevm.h b/migration/savevm.h
index XXXXXXX..XXXXXXX 100644
--- a/migration/savevm.h
+++ b/migration/savevm.h
@@ -XXX,XX +XXX,XX @@ void qemu_savevm_send_postcopy_ram_discard(QEMUFile *f, const char *name,
                                            uint64_t *start_list,
                                            uint64_t *length_list);
 void qemu_savevm_send_colo_enable(QEMUFile *f);
+void qemu_savevm_live_state(QEMUFile *f);
+int qemu_save_device_state(QEMUFile *f);
 
 int qemu_loadvm_state(QEMUFile *f);
 void qemu_loadvm_state_cleanup(void);
+int qemu_loadvm_state_main(QEMUFile *f, MigrationIncomingState *mis);
+int qemu_load_device_state(QEMUFile *f);
 
 #endif
-- 
2.5.0

From: zhanghailiang <zhang.zhanghailiang@huawei.com>

Don't need to flush all VM's ram from cache, only
flush the dirty pages since last checkpoint

Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/ram.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/migration/ram.c b/migration/ram.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -XXX,XX +XXX,XX @@ int colo_init_ram_cache(void)
     }
     ram_state = g_new0(RAMState, 1);
     ram_state->migration_dirty_pages = 0;
+    memory_global_dirty_log_start();
 
     return 0;
 
@@ -XXX,XX +XXX,XX @@ void colo_release_ram_cache(void)
 {
     RAMBlock *block;
 
+    memory_global_dirty_log_stop();
     RAMBLOCK_FOREACH_MIGRATABLE(block) {
         g_free(block->bmap);
         block->bmap = NULL;
@@ -XXX,XX +XXX,XX @@ static void colo_flush_ram_cache(void)
     void *src_host;
     unsigned long offset = 0;
 
+    memory_global_dirty_log_sync();
+    rcu_read_lock();
+    RAMBLOCK_FOREACH_MIGRATABLE(block) {
+        migration_bitmap_sync_range(ram_state, block, 0, block->used_length);
+    }
+    rcu_read_unlock();
+
     trace_colo_flush_ram_cache_begin(ram_state->migration_dirty_pages);
     rcu_read_lock();
     block = QLIST_FIRST_RCU(&ram_list.blocks);
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

Filter needs to process the event of checkpoint/failover or
other event passed by COLO frame.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 include/net/filter.h |  5 +++++
 net/filter.c         | 17 +++++++++++++++++
 net/net.c            | 19 +++++++++++++++++++
 3 files changed, 41 insertions(+)

diff --git a/include/net/filter.h b/include/net/filter.h
index XXXXXXX..XXXXXXX 100644
--- a/include/net/filter.h
+++ b/include/net/filter.h
@@ -XXX,XX +XXX,XX @@ typedef ssize_t (FilterReceiveIOV)(NetFilterState *nc,
 
 typedef void (FilterStatusChanged) (NetFilterState *nf, Error **errp);
 
+typedef void (FilterHandleEvent) (NetFilterState *nf, int event, Error **errp);
+
 typedef struct NetFilterClass {
     ObjectClass parent_class;
 
@@ -XXX,XX +XXX,XX @@ typedef struct NetFilterClass {
     FilterSetup *setup;
     FilterCleanup *cleanup;
     FilterStatusChanged *status_changed;
+    FilterHandleEvent *handle_event;
     /* mandatory */
     FilterReceiveIOV *receive_iov;
 } NetFilterClass;
@@ -XXX,XX +XXX,XX @@ ssize_t qemu_netfilter_pass_to_next(NetClientState *sender,
                                     int iovcnt,
                                     void *opaque);
 
+void colo_notify_filters_event(int event, Error **errp);
+
 #endif /* QEMU_NET_FILTER_H */
diff --git a/net/filter.c b/net/filter.c
index XXXXXXX..XXXXXXX 100644
--- a/net/filter.c
+++ b/net/filter.c
@@ -XXX,XX +XXX,XX @@
 #include "net/vhost_net.h"
 #include "qom/object_interfaces.h"
 #include "qemu/iov.h"
+#include "net/colo.h"
+#include "migration/colo.h"
 
 static inline bool qemu_can_skip_netfilter(NetFilterState *nf)
 {
@@ -XXX,XX +XXX,XX @@ static void netfilter_finalize(Object *obj)
     g_free(nf->netdev_id);
 }
 
+static void default_handle_event(NetFilterState *nf, int event, Error **errp)
+{
+    switch (event) {
+    case COLO_EVENT_CHECKPOINT:
+        break;
+    case COLO_EVENT_FAILOVER:
+        object_property_set_str(OBJECT(nf), "off", "status", errp);
+        break;
+    default:
+        break;
+    }
+}
+
 static void netfilter_class_init(ObjectClass *oc, void *data)
 {
     UserCreatableClass *ucc = USER_CREATABLE_CLASS(oc);
+    NetFilterClass *nfc = NETFILTER_CLASS(oc);
 
     ucc->complete = netfilter_complete;
+    nfc->handle_event = default_handle_event;
 }
 
 static const TypeInfo netfilter_info = {
diff --git a/net/net.c b/net/net.c
index XXXXXXX..XXXXXXX 100644
--- a/net/net.c
+++ b/net/net.c
@@ -XXX,XX +XXX,XX @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
     }
 }
 
+void colo_notify_filters_event(int event, Error **errp)
+{
+    NetClientState *nc;
+    NetFilterState *nf;
+    NetFilterClass *nfc = NULL;
+    Error *local_err = NULL;
+
+    QTAILQ_FOREACH(nc, &net_clients, next) {
+        QTAILQ_FOREACH(nf, &nc->filters, next) {
+            nfc = NETFILTER_GET_CLASS(OBJECT(nf));
+            nfc->handle_event(nf, event, &local_err);
+            if (local_err) {
+                error_propagate(errp, local_err);
+                return;
+            }
+        }
+    }
+}
+
 void qmp_set_link(const char *name, bool up, Error **errp)
 {
     NetClientState *ncs[MAX_QUEUE_NUM];
-- 
2.5.0

From: Zhang Chen <zhangckid@gmail.com>

After one round of checkpoint, the states between PVM and SVM
become consistent, so it is unnecessary to adjust the sequence
of net packets for old connections, besides, while failover
happens, filter-rewriter will into failover mode that needn't
handle the new TCP connection.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/colo-compare.c    | 12 +++++------
 net/colo.c            |  8 ++++++++
 net/colo.h            |  2 ++
 net/filter-rewriter.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 73 insertions(+), 6 deletions(-)

diff --git a/net/colo-compare.c b/net/colo-compare.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -XXX,XX +XXX,XX @@ enum {
     SECONDARY_IN,
 };
 
+static void colo_compare_inconsistency_notify(void)
+{
+    notifier_list_notify(&colo_compare_notifiers,
+                migrate_get_current());
+}
+
 static int compare_chr_send(CompareState *s,
                             const uint8_t *buf,
                             uint32_t size,
@@ -XXX,XX +XXX,XX @@ static bool colo_mark_tcp_pkt(Packet *ppkt, Packet *spkt,
     return false;
 }
 
-static void colo_compare_inconsistency_notify(void)
-{
-    notifier_list_notify(&colo_compare_notifiers,
-                migrate_get_current());
-}
-
 static void colo_compare_tcp(CompareState *s, Connection *conn)
 {
     Packet *ppkt = NULL, *spkt = NULL;
diff --git a/net/colo.c b/net/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/net/colo.c
+++ b/net/colo.c
@@ -XXX,XX +XXX,XX @@ Connection *connection_get(GHashTable *connection_track_table,
 
     return conn;
 }
+
+bool connection_has_tracked(GHashTable *connection_track_table,
+                            ConnectionKey *key)
+{
+    Connection *conn = g_hash_table_lookup(connection_track_table, key);
+
+    return conn ? true : false;
+}
diff --git a/net/colo.h b/net/colo.h
index XXXXXXX..XXXXXXX 100644
--- a/net/colo.h
+++ b/net/colo.h
@@ -XXX,XX +XXX,XX @@ void connection_destroy(void *opaque);
 Connection *connection_get(GHashTable *connection_track_table,
                            ConnectionKey *key,
                            GQueue *conn_list);
+bool connection_has_tracked(GHashTable *connection_track_table,
+                            ConnectionKey *key);
 void connection_hashtable_reset(GHashTable *connection_track_table);
 Packet *packet_new(const void *data, int size, int vnet_hdr_len);
 void packet_destroy(void *opaque, void *user_data);
diff --git a/net/filter-rewriter.c b/net/filter-rewriter.c
index XXXXXXX..XXXXXXX 100644
--- a/net/filter-rewriter.c
+++ b/net/filter-rewriter.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/main-loop.h"
 #include "qemu/iov.h"
 #include "net/checksum.h"
+#include "net/colo.h"
+#include "migration/colo.h"
 
 #define FILTER_COLO_REWRITER(obj) \
     OBJECT_CHECK(RewriterState, (obj), TYPE_FILTER_REWRITER)
 
 #define TYPE_FILTER_REWRITER "filter-rewriter"
+#define FAILOVER_MODE_ON  true
+#define FAILOVER_MODE_OFF false
 
 typedef struct RewriterState {
     NetFilterState parent_obj;
@@ -XXX,XX +XXX,XX @@ typedef struct RewriterState {
     /* hashtable to save connection */
     GHashTable *connection_track_table;
     bool vnet_hdr;
+    bool failover_mode;
 } RewriterState;
 
+static void filter_rewriter_failover_mode(RewriterState *s)
+{
+    s->failover_mode = FAILOVER_MODE_ON;
+}
+
 static void filter_rewriter_flush(NetFilterState *nf)
 {
     RewriterState *s = FILTER_COLO_REWRITER(nf);
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
              */
             reverse_connection_key(&key);
         }
+
+        /* After failover we needn't change new TCP packet */
+        if (s->failover_mode &&
+            !connection_has_tracked(s->connection_track_table, &key)) {
+            goto out;
+        }
+
         conn = connection_get(s->connection_track_table,
                               &key,
                               NULL);
@@ -XXX,XX +XXX,XX @@ static ssize_t colo_rewriter_receive_iov(NetFilterState *nf,
         }
     }
 
+out:
     packet_destroy(pkt, NULL);
     pkt = NULL;
     return 0;
 }
 
+static void reset_seq_offset(gpointer key, gpointer value, gpointer user_data)
+{
+    Connection *conn = (Connection *)value;
+
+    conn->offset = 0;
+}
+
+static gboolean offset_is_nonzero(gpointer key,
+                                  gpointer value,
+                                  gpointer user_data)
+{
+    Connection *conn = (Connection *)value;
+
+    return conn->offset ? true : false;
+}
+
+static void colo_rewriter_handle_event(NetFilterState *nf, int event,
+                                       Error **errp)
+{
+    RewriterState *rs = FILTER_COLO_REWRITER(nf);
+
+    switch (event) {
+    case COLO_EVENT_CHECKPOINT:
+        g_hash_table_foreach(rs->connection_track_table,
+                            reset_seq_offset, NULL);
+        break;
+    case COLO_EVENT_FAILOVER:
+        if (!g_hash_table_find(rs->connection_track_table,
+                              offset_is_nonzero, NULL)) {
+            filter_rewriter_failover_mode(rs);
+        }
+        break;
+    default:
+        break;
+    }
+}
+
 static void colo_rewriter_cleanup(NetFilterState *nf)
 {
     RewriterState *s = FILTER_COLO_REWRITER(nf);
@@ -XXX,XX +XXX,XX @@ static void filter_rewriter_init(Object *obj)
     RewriterState *s = FILTER_COLO_REWRITER(obj);
 
     s->vnet_hdr = false;
+    s->failover_mode = FAILOVER_MODE_OFF;
     object_property_add_bool(obj, "vnet_hdr_support",
                              filter_rewriter_get_vnet_hdr,
                              filter_rewriter_set_vnet_hdr, NULL);
@@ -XXX,XX +XXX,XX @@ static void colo_rewriter_class_init(ObjectClass *oc, void *data)
     nfc->setup = colo_rewriter_setup;
     nfc->cleanup = colo_rewriter_cleanup;
     nfc->receive_iov = colo_rewriter_receive_iov;
+    nfc->handle_event = colo_rewriter_handle_event;
 }
 
 static const TypeInfo colo_rewriter_info = {
-- 
2.5.0

From: zhanghailiang <zhang.zhanghailiang@huawei.com>

Notify all net filters about the checkpoint and failover event.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 migration/colo.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@
 #include "qapi/qapi-events-migration.h"
 #include "qapi/qmp/qerror.h"
 #include "sysemu/cpus.h"
+#include "net/filter.h"
 
 static bool vmstate_loading;
 static Notifier packets_compare_notifier;
@@ -XXX,XX +XXX,XX @@ static void secondary_vm_do_failover(void)
         error_report_err(local_err);
     }
 
+    /* Notify all filters of all NIC to do checkpoint */
+    colo_notify_filters_event(COLO_EVENT_FAILOVER, &local_err);
+    if (local_err) {
+        error_report_err(local_err);
+    }
+
     if (!autostart) {
         error_report("\"-S\" qemu option will be ignored in secondary side");
         /* recover runstate to normal migration finish state */
@@ -XXX,XX +XXX,XX @@ void *colo_process_incoming_thread(void *opaque)
             goto out;
         }
 
+        /* Notify all filters of all NIC to do checkpoint */
+        colo_notify_filters_event(COLO_EVENT_CHECKPOINT, &local_err);
+
+        if (local_err) {
+            qemu_mutex_unlock_iothread();
+            goto out;
+        }
+
         vmstate_loading = false;
         vm_start();
         trace_colo_vm_state_change("stop", "run");
-- 
2.5.0

From: zhanghailiang <zhang.zhanghailiang@huawei.com>

COLO thread may sleep at qemu_sem_wait(&s->colo_checkpoint_sem),
while failover works begin, It's better to wakeup it to quick
the process.

diff --git a/migration/colo.c b/migration/colo.c
index XXXXXXX..XXXXXXX 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -XXX,XX +XXX,XX @@ static void primary_vm_do_failover(void)
 
     migrate_set_state(&s->state, MIGRATION_STATUS_COLO,
                       MIGRATION_STATUS_COMPLETED);
+    /*
+     * kick COLO thread which might wait at
+     * qemu_sem_wait(&s->colo_checkpoint_sem).
+     */
+    colo_checkpoint_notify(migrate_get_current());
 
     /*
      * Wake up COLO thread which may blocked in recv() or send(),
@@ -XXX,XX +XXX,XX @@ static void colo_process_checkpoint(MigrationState *s)
 
         qemu_sem_wait(&s->colo_checkpoint_sem);
 
+        if (s->state != MIGRATION_STATUS_COLO) {
+            goto out;
+        }
         ret = colo_do_checkpoint_transaction(s, bioc, fb);
         if (ret < 0) {
             goto out;
-- 
2.5.0

From: Zhang Chen <chen.zhang@intel.com>

This diagram make user better understand COLO.
Suggested by Markus Armbruster.

Signed-off-by: Zhang Chen <zhangckid@gmail.com>
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 docs/COLO-FT.txt | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/docs/COLO-FT.txt b/docs/COLO-FT.txt
index XXXXXXX..XXXXXXX 100644
--- a/docs/COLO-FT.txt
+++ b/docs/COLO-FT.txt
@@ -XXX,XX +XXX,XX @@ Note:
 HeartBeat has not been implemented yet, so you need to trigger failover process
 by using 'x-colo-lost-heartbeat' command.
 
+== COLO operation status ==
+
++-----------------+
+|                 |
+|    Start COLO   |
+|                 |
++--------+--------+
+         |
+         |  Main qmp command:
+         |  migrate-set-capabilities with x-colo
+         |  migrate
+         |
+         v
++--------+--------+
+|                 |
+|  COLO running   |
+|                 |
++--------+--------+
+         |
+         |  Main qmp command:
+         |  x-colo-lost-heartbeat
+         |  or
+         |  some error happened
+         v
++--------+--------+
+|                 |  send qmp event:
+|  COLO failover  |  COLO_EXIT
+|                 |
++-----------------+
+
+COLO use the qmp command to switch and report operation status.
+The diagram just shows the main qmp command, you can get the detail
+in test procedure.
+
 == Test procedure ==
 1. Startup qemu
 Primary:
-- 
2.5.0

From: liujunjie <liujunjie23@huawei.com>

Before, we did not clear callback like handle_output when delete
the virtqueue which may result be segmentfault.
The scene is as follows:
1. Start a vm with multiqueue vhost-net,
2. then we write VIRTIO_PCI_GUEST_FEATURES in PCI configuration to
triger multiqueue disable in this vm which will delete the virtqueue.
In this step, the tx_bh is deleted but the callback virtio_net_handle_tx_bh
still exist.
3. Finally, we write VIRTIO_PCI_QUEUE_NOTIFY in PCI configuration to
notify the deleted virtqueue. In this way, virtio_net_handle_tx_bh
will be called and qemu will be crashed.

Although the way described above is uncommon, we had better reinforce it.

CC: qemu-stable@nongnu.org
Signed-off-by: liujunjie <liujunjie23@huawei.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/virtio/virtio.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -XXX,XX +XXX,XX @@ void virtio_del_queue(VirtIODevice *vdev, int n)
 
     vdev->vq[n].vring.num = 0;
     vdev->vq[n].vring.num_default = 0;
+    vdev->vq[n].handle_output = NULL;
+    vdev->vq[n].handle_aio_output = NULL;
 }
 
 static void virtio_set_isr(VirtIODevice *vdev, int value)
-- 
2.5.0

In ne2000_receive(), we try to assign size_ to size which converts
from size_t to integer. This will cause troubles when size_ is greater
INT_MAX, this will lead a negative value in size and it can then pass
the check of size < MIN_BUF_SIZE which may lead out of bound access of
for both buf and buf1.

Fixing by converting the type of size to size_t.

CC: qemu-stable@nongnu.org
Reported-by: Daniel Shapira <daniel@twistlock.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/ne2000.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/net/ne2000.c b/hw/net/ne2000.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/net/ne2000.c
+++ b/hw/net/ne2000.c
@@ -XXX,XX +XXX,XX @@ static int ne2000_buffer_full(NE2000State *s)
 ssize_t ne2000_receive(NetClientState *nc, const uint8_t *buf, size_t size_)
 {
     NE2000State *s = qemu_get_nic_opaque(nc);
-    int size = size_;
+    size_t size = size_;
     uint8_t *p;
     unsigned int total_len, next, avail, len, index, mcast_idx;
     uint8_t buf1[60];
@@ -XXX,XX +XXX,XX @@ ssize_t ne2000_receive(NetClientState *nc, const uint8_t *buf, size_t size_)
         { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
 
 #if defined(DEBUG_NE2000)
-    printf("NE2000: received len=%d\n", size);
+    printf("NE2000: received len=%zu\n", size);
 #endif
 
     if (s->cmd & E8390_STOP || ne2000_buffer_full(s))
-- 
2.5.0

In rtl8139_do_receive(), we try to assign size_ to size which converts
from size_t to integer. This will cause troubles when size_ is greater
INT_MAX, this will lead a negative value in size and it can then pass
the check of size < MIN_BUF_SIZE which may lead out of bound access of
for both buf and buf1.

Fixing by converting the type of size to size_t.

CC: qemu-stable@nongnu.org
Reported-by: Daniel Shapira <daniel@twistlock.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/rtl8139.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/hw/net/rtl8139.c b/hw/net/rtl8139.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/net/rtl8139.c
+++ b/hw/net/rtl8139.c
@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
     RTL8139State *s = qemu_get_nic_opaque(nc);
     PCIDevice *d = PCI_DEVICE(s);
     /* size is the length of the buffer passed to the driver */
-    int size = size_;
+    size_t size = size_;
     const uint8_t *dot1q_buf = NULL;
 
     uint32_t packet_header = 0;
@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
     static const uint8_t broadcast_macaddr[6] =
         { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
 
-    DPRINTF(">>> received len=%d\n", size);
+    DPRINTF(">>> received len=%zu\n", size);
 
     /* test if board clock is stopped */
     if (!s->clock_enabled)
@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
 
         if (size+4 > rx_space)
         {
-            DPRINTF("C+ Rx mode : descriptor %d size %d received %d + 4\n",
+            DPRINTF("C+ Rx mode : descriptor %d size %d received %zu + 4\n",
                 descriptor, rx_space, size);
 
             s->IntrStatus |= RxOverflow;
@@ -XXX,XX +XXX,XX @@ static ssize_t rtl8139_do_receive(NetClientState *nc, const uint8_t *buf, size_t
         if (avail != 0 && RX_ALIGN(size + 8) >= avail)
         {
             DPRINTF("rx overflow: rx buffer length %d head 0x%04x "
-                "read 0x%04x === available 0x%04x need 0x%04x\n",
+                "read 0x%04x === available 0x%04x need 0x%04zx\n",
                 s->RxBufferSize, s->RxBufAddr, s->RxBufPtr, avail, size + 8);
 
             s->IntrStatus |= RxOverflow;
-- 
2.5.0

In pcnet_receive(), we try to assign size_ to size which converts from
size_t to integer. This will cause troubles when size_ is greater
INT_MAX, this will lead a negative value in size and it can then pass
the check of size < MIN_BUF_SIZE which may lead out of bound access
for both buf and buf1.

Fixing by converting the type of size to size_t.

CC: qemu-stable@nongnu.org
Reported-by: Daniel Shapira <daniel@twistlock.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/pcnet.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/net/pcnet.c b/hw/net/pcnet.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/net/pcnet.c
+++ b/hw/net/pcnet.c
@@ -XXX,XX +XXX,XX @@ ssize_t pcnet_receive(NetClientState *nc, const uint8_t *buf, size_t size_)
     uint8_t buf1[60];
     int remaining;
     int crc_err = 0;
-    int size = size_;
+    size_t size = size_;
 
     if (CSR_DRX(s) || CSR_STOP(s) || CSR_SPND(s) || !size ||
         (CSR_LOOP(s) && !s->looptest)) {
         return -1;
     }
 #ifdef PCNET_DEBUG
-    printf("pcnet_receive size=%d\n", size);
+    printf("pcnet_receive size=%zu\n", size);
 #endif
 
     /* if too small buffer, then expand it */
-- 
2.5.0

The e1000 emulation silently discards RX packets if there's
insufficient space in the ring buffer. This leads to errors
on higher-level protocols in the guest, with no indication
about the error cause.

This patch increments the "Missed Packets Count" (MPC) and
"Receive No Buffers Count" (RNBC) HW counters in this case.
As the emulation has no FIFO for buffering packets that can't
immediately be pushed to the guest, these two registers are
practically equivalent (see 10.2.7.4, 10.2.7.33 in
https://www.intel.com/content/www/us/en/embedded/products/networking/82574l-gbe-controller-datasheet.html).

On a Linux guest, the register content  will be reflected in
the "rx_missed_errors" and "rx_no_buffer_count" stats from
"ethtool -S", and in the "missed" stat from "ip -s -s link show",
giving at least some hint about the error cause inside the guest.

If the cause is known, problems like this can often be avoided
easily, by increasing the number of RX descriptors in the guest
e1000 driver (e.g under Linux, "e1000.RxDescriptors=1024").

The patch also adds a qemu trace message for this condition.

Signed-off-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/net/e1000.c      | 16 +++++++++++++---
 hw/net/trace-events |  3 +++
 2 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/hw/net/e1000.c b/hw/net/e1000.c
index XXXXXXX..XXXXXXX 100644
--- a/hw/net/e1000.c
+++ b/hw/net/e1000.c
@@ -XXX,XX +XXX,XX @@
 #include "qemu/range.h"
 
 #include "e1000x_common.h"
+#include "trace.h"
 
 static const uint8_t bcast[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
 
@@ -XXX,XX +XXX,XX @@ static uint64_t rx_desc_base(E1000State *s)
     return (bah << 32) + bal;
 }
 
+static void
+e1000_receiver_overrun(E1000State *s, size_t size)
+{
+    trace_e1000_receiver_overrun(size, s->mac_reg[RDH], s->mac_reg[RDT]);
+    e1000x_inc_reg_if_not_full(s->mac_reg, RNBC);
+    e1000x_inc_reg_if_not_full(s->mac_reg, MPC);
+    set_ics(s, 0, E1000_ICS_RXO);
+}
+
 static ssize_t
 e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
 {
@@ -XXX,XX +XXX,XX @@ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
     desc_offset = 0;
     total_size = size + e1000x_fcs_len(s->mac_reg);
     if (!e1000_has_rxbufs(s, total_size)) {
-            set_ics(s, 0, E1000_ICS_RXO);
-            return -1;
+        e1000_receiver_overrun(s, total_size);
+        return -1;
     }
     do {
         desc_size = total_size - desc_offset;
@@ -XXX,XX +XXX,XX @@ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
             rdh_start >= s->mac_reg[RDLEN] / sizeof(desc)) {
             DBGOUT(RXERR, "RDH wraparound @%x, RDT %x, RDLEN %x\n",
                    rdh_start, s->mac_reg[RDT], s->mac_reg[RDLEN]);
-            set_ics(s, 0, E1000_ICS_RXO);
+            e1000_receiver_overrun(s, total_size);
             return -1;
         }
     } while (desc_offset < total_size);
diff --git a/hw/net/trace-events b/hw/net/trace-events
index XXXXXXX..XXXXXXX 100644
--- a/hw/net/trace-events
+++ b/hw/net/trace-events
@@ -XXX,XX +XXX,XX @@ net_rx_pkt_rss_ip6_ex(void) "Calculating IPv6/EX RSS  hash"
 net_rx_pkt_rss_hash(size_t rss_length, uint32_t rss_hash) "RSS hash for %zu bytes: 0x%X"
 net_rx_pkt_rss_add_chunk(void* ptr, size_t size, size_t input_offset) "Add RSS chunk %p, %zu bytes, RSS input offset %zu bytes"
 
+# hw/net/e1000.c
+e1000_receiver_overrun(size_t s, uint32_t rdh, uint32_t rdt) "Receiver overrun: dropped packet of %zu bytes, RDH=%u, RDT=%u"
+
 # hw/net/e1000x_common.c
 e1000x_rx_can_recv_disabled(bool link_up, bool rx_enabled, bool pci_master) "link_up: %d, rx_enabled %d, pci_master %d"
 e1000x_vlan_is_vlan_pkt(bool is_vlan_pkt, uint16_t eth_proto, uint16_t vet) "Is VLAN packet: %d, ETH proto: 0x%X, VET: 0x%X"
-- 
2.5.0