Merge remote-tracking branch 'remotes/dgilbert/tags/pull-migration-20190814a' into staging

Migration pull 2019-08-15

Marcel's vmxnet3 live migraiton fix (that breaks vmxnet3 compatibility
but makes it work)

Error description improvements from Yury.

Multifd fixes from Ivan and Juan.

A load of small cleanups from Wei.

A small cleanup from Marc-André for a future patch.

# gpg: Signature made Wed 14 Aug 2019 19:00:39 BST
# gpg:                using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <dgilbert@redhat.com>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A  9FA9 0516 331E BC5B FDE7

* remotes/dgilbert/tags/pull-migration-20190814a: (33 commits)
  migration: add some multifd traces
  migration: Make global sem_sync semaphore by channel
  migration: Add traces for multifd terminate threads
  qemu-file: move qemu_{get,put}_counted_string() declarations
  migration/postcopy: use mis->bh instead of allocating a QEMUBH
  migration: rename migration_bitmap_sync_range to ramblock_sync_dirty_bitmap
  migration: update ram_counters for multifd sync packet
  migration: add speed limit for multifd migration
  migration: add qemu_file_update_transfer interface
  migration: always initialise ram_counters for a new migration
  migration: remove unused field bytes_xfer
  hmp: Remove migration capabilities from "info migrate"
  migration/postcopy: use QEMU_IS_ALIGNED to replace host_offset
  migration/postcopy: simplify calculation of run_start and fixup_start_addr
  migration/postcopy: make PostcopyDiscardState a static variable
  migration: extract ram_load_precopy
  migration: return -EINVAL directly when version_id mismatch
  migration: equation is more proper than and to check LOADVM_QUIT
  migration: just pass RAMBlock is enough
  migration: use migration_in_postcopy() to check POSTCOPY_ACTIVE
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

11 files changed, 342 insertions, 264 deletions

diff --git a/migration/migration.c b/migration/migration.c
index 8a607fe1e2..c49e9dc035 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -823,6 +823,25 @@ bool migration_is_setup_or_active(int state)
     }
 }
 
+static void populate_time_info(MigrationInfo *info, MigrationState *s)
+{
+    info->has_status = true;
+    info->has_setup_time = true;
+    info->setup_time = s->setup_time;
+    if (s->state == MIGRATION_STATUS_COMPLETED) {
+        info->has_total_time = true;
+        info->total_time = s->total_time;
+        info->has_downtime = true;
+        info->downtime = s->downtime;
+    } else {
+        info->has_total_time = true;
+        info->total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) -
+                           s->start_time;
+        info->has_expected_downtime = true;
+        info->expected_downtime = s->expected_downtime;
+    }
+}
+
 static void populate_ram_info(MigrationInfo *info, MigrationState *s)
 {
     info->has_ram = true;
@@ -908,16 +927,8 @@ static void fill_source_migration_info(MigrationInfo *info)
     case MIGRATION_STATUS_DEVICE:
     case MIGRATION_STATUS_POSTCOPY_PAUSED:
     case MIGRATION_STATUS_POSTCOPY_RECOVER:
-         /* TODO add some postcopy stats */
-        info->has_status = true;
-        info->has_total_time = true;
-        info->total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME)
-            - s->start_time;
-        info->has_expected_downtime = true;
-        info->expected_downtime = s->expected_downtime;
-        info->has_setup_time = true;
-        info->setup_time = s->setup_time;
-
+        /* TODO add some postcopy stats */
+        populate_time_info(info, s);
         populate_ram_info(info, s);
         populate_disk_info(info);
         break;
@@ -926,14 +937,7 @@ static void fill_source_migration_info(MigrationInfo *info)
         /* TODO: display COLO specific information (checkpoint info etc.) */
         break;
     case MIGRATION_STATUS_COMPLETED:
-        info->has_status = true;
-        info->has_total_time = true;
-        info->total_time = s->total_time;
-        info->has_downtime = true;
-        info->downtime = s->downtime;
-        info->has_setup_time = true;
-        info->setup_time = s->setup_time;
-
+        populate_time_info(info, s);
         populate_ram_info(info, s);
         break;
     case MIGRATION_STATUS_FAILED:
@@ -1695,7 +1699,6 @@ void migrate_init(MigrationState *s)
      * parameters/capabilities that the user set, and
      * locks.
      */
-    s->bytes_xfer = 0;
     s->cleanup_bh = 0;
     s->to_dst_file = NULL;
     s->state = MIGRATION_STATUS_NONE;
@@ -1908,6 +1911,11 @@ static bool migrate_prepare(MigrationState *s, bool blk, bool blk_inc,
     }
 
     migrate_init(s);
+    /*
+     * set ram_counters memory to zero for a
+     * new migration
+     */
+    memset(&ram_counters, 0, sizeof(ram_counters));
 
     return true;
 }
@@ -2963,6 +2971,7 @@ static MigThrError migration_detect_error(MigrationState *s)
 {
     int ret;
     int state = s->state;
+    Error *local_error = NULL;
 
     if (state == MIGRATION_STATUS_CANCELLING ||
         state == MIGRATION_STATUS_CANCELLED) {
@@ -2971,13 +2980,18 @@ static MigThrError migration_detect_error(MigrationState *s)
     }
 
     /* Try to detect any file errors */
-    ret = qemu_file_get_error(s->to_dst_file);
-
+    ret = qemu_file_get_error_obj(s->to_dst_file, &local_error);
     if (!ret) {
         /* Everything is fine */
+        assert(!local_error);
         return MIG_THR_ERR_NONE;
     }
 
+    if (local_error) {
+        migrate_set_error(s, local_error);
+        error_free(local_error);
+    }
+
     if (state == MIGRATION_STATUS_POSTCOPY_ACTIVE && ret == -EIO) {
         /*
          * For postcopy, we allow the network to be down for a
@@ -3025,6 +3039,17 @@ static void migration_calculate_complete(MigrationState *s)
     }
 }
 
+static void update_iteration_initial_status(MigrationState *s)
+{
+    /*
+     * Update these three fields at the same time to avoid mismatch info lead
+     * wrong speed calculation.
+     */
+    s->iteration_start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+    s->iteration_initial_bytes = migration_total_bytes(s);
+    s->iteration_initial_pages = ram_get_total_transferred_pages();
+}
+
 static void migration_update_counters(MigrationState *s,
                                       int64_t current_time)
 {
@@ -3060,9 +3085,7 @@ static void migration_update_counters(MigrationState *s,
 
     qemu_file_reset_rate_limit(s->to_dst_file);
 
-    s->iteration_start_time = current_time;
-    s->iteration_initial_bytes = current_bytes;
-    s->iteration_initial_pages = ram_get_total_transferred_pages();
+    update_iteration_initial_status(s);
 
     trace_migrate_transferred(transferred, time_spent,
                               bandwidth, s->threshold_size);
@@ -3093,8 +3116,7 @@ static MigIterateState migration_iteration_run(MigrationState *s)
 
     if (pending_size && pending_size >= s->threshold_size) {
         /* Still a significant amount to transfer */
-        if (migrate_postcopy() && !in_postcopy &&
-            pend_pre <= s->threshold_size &&
+        if (!in_postcopy && pend_pre <= s->threshold_size &&
             atomic_read(&s->start_postcopy)) {
             if (postcopy_start(s)) {
                 error_report("%s: postcopy failed to start", __func__);
@@ -3186,7 +3208,7 @@ static void *migration_thread(void *opaque)
     rcu_register_thread();
 
     object_ref(OBJECT(s));
-    s->iteration_start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+    update_iteration_initial_status(s);
 
     qemu_savevm_state_header(s->to_dst_file);
 
@@ -3251,8 +3273,7 @@ static void *migration_thread(void *opaque)
              * the local variables. This is important to avoid
              * breaking transferred_bytes and bandwidth calculation
              */
-            s->iteration_start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
-            s->iteration_initial_bytes = 0;
+            update_iteration_initial_status(s);
         }
 
         current_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
diff --git a/migration/migration.h b/migration/migration.h
index 1fdd7b21fd..5bc60709db 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -132,7 +132,6 @@ struct MigrationState
     DeviceState parent_obj;
 
     /*< public >*/
-    size_t bytes_xfer;
     QemuThread thread;
     QEMUBH *cleanup_bh;
     QEMUFile *to_dst_file;
diff --git a/migration/postcopy-ram.c b/migration/postcopy-ram.c
index 9faacacc9e..2cb1a69752 100644
--- a/migration/postcopy-ram.c
+++ b/migration/postcopy-ram.c
@@ -1377,22 +1377,16 @@ void postcopy_fault_thread_notify(MigrationIncomingState *mis)
  *   asking to discard individual ranges.
  *
  * @ms: The current migration state.
- * @offset: the bitmap offset of the named RAMBlock in the migration
- *   bitmap.
+ * @offset: the bitmap offset of the named RAMBlock in the migration bitmap.
  * @name: RAMBlock that discards will operate on.
- *
- * returns: a new PDS.
  */
-PostcopyDiscardState *postcopy_discard_send_init(MigrationState *ms,
-                                                 const char *name)
+static PostcopyDiscardState pds = {0};
+void postcopy_discard_send_init(MigrationState *ms, const char *name)
 {
-    PostcopyDiscardState *res = g_malloc0(sizeof(PostcopyDiscardState));
-
-    if (res) {
-        res->ramblock_name = name;
-    }
-
-    return res;
+    pds.ramblock_name = name;
+    pds.cur_entry = 0;
+    pds.nsentwords = 0;
+    pds.nsentcmds = 0;
 }
 
 /**
@@ -1401,30 +1395,29 @@ PostcopyDiscardState *postcopy_discard_send_init(MigrationState *ms,
  *   be sent later.
  *
  * @ms: Current migration state.
- * @pds: Structure initialised by postcopy_discard_send_init().
  * @start,@length: a range of pages in the migration bitmap in the
  *   RAM block passed to postcopy_discard_send_init() (length=1 is one page)
  */
-void postcopy_discard_send_range(MigrationState *ms, PostcopyDiscardState *pds,
-                                unsigned long start, unsigned long length)
+void postcopy_discard_send_range(MigrationState *ms, unsigned long start,
+                                 unsigned long length)
 {
     size_t tp_size = qemu_target_page_size();
     /* Convert to byte offsets within the RAM block */
-    pds->start_list[pds->cur_entry] = start  * tp_size;
-    pds->length_list[pds->cur_entry] = length * tp_size;
-    trace_postcopy_discard_send_range(pds->ramblock_name, start, length);
-    pds->cur_entry++;
-    pds->nsentwords++;
+    pds.start_list[pds.cur_entry] = start  * tp_size;
+    pds.length_list[pds.cur_entry] = length * tp_size;
+    trace_postcopy_discard_send_range(pds.ramblock_name, start, length);
+    pds.cur_entry++;
+    pds.nsentwords++;
 
-    if (pds->cur_entry == MAX_DISCARDS_PER_COMMAND) {
+    if (pds.cur_entry == MAX_DISCARDS_PER_COMMAND) {
         /* Full set, ship it! */
         qemu_savevm_send_postcopy_ram_discard(ms->to_dst_file,
-                                              pds->ramblock_name,
-                                              pds->cur_entry,
-                                              pds->start_list,
-                                              pds->length_list);
-        pds->nsentcmds++;
-        pds->cur_entry = 0;
+                                              pds.ramblock_name,
+                                              pds.cur_entry,
+                                              pds.start_list,
+                                              pds.length_list);
+        pds.nsentcmds++;
+        pds.cur_entry = 0;
     }
 }
 
@@ -1433,24 +1426,21 @@ void postcopy_discard_send_range(MigrationState *ms, PostcopyDiscardState *pds,
  * bitmap code. Sends any outstanding discard messages, frees the PDS
  *
  * @ms: Current migration state.
- * @pds: Structure initialised by postcopy_discard_send_init().
  */
-void postcopy_discard_send_finish(MigrationState *ms, PostcopyDiscardState *pds)
+void postcopy_discard_send_finish(MigrationState *ms)
 {
     /* Anything unsent? */
-    if (pds->cur_entry) {
+    if (pds.cur_entry) {
         qemu_savevm_send_postcopy_ram_discard(ms->to_dst_file,
-                                              pds->ramblock_name,
-                                              pds->cur_entry,
-                                              pds->start_list,
-                                              pds->length_list);
-        pds->nsentcmds++;
+                                              pds.ramblock_name,
+                                              pds.cur_entry,
+                                              pds.start_list,
+                                              pds.length_list);
+        pds.nsentcmds++;
     }
 
-    trace_postcopy_discard_send_finish(pds->ramblock_name, pds->nsentwords,
-                                       pds->nsentcmds);
-
-    g_free(pds);
+    trace_postcopy_discard_send_finish(pds.ramblock_name, pds.nsentwords,
+                                       pds.nsentcmds);
 }
 
 /*
diff --git a/migration/postcopy-ram.h b/migration/postcopy-ram.h
index 9d55536fd1..9c8bd2bae0 100644
--- a/migration/postcopy-ram.h
+++ b/migration/postcopy-ram.h
@@ -43,10 +43,8 @@ int postcopy_ram_prepare_discard(MigrationIncomingState *mis);
 
 /*
  * Called at the start of each RAMBlock by the bitmap code.
- * Returns a new PDS
  */
-PostcopyDiscardState *postcopy_discard_send_init(MigrationState *ms,
-                                                 const char *name);
+void postcopy_discard_send_init(MigrationState *ms, const char *name);
 
 /*
  * Called by the bitmap code for each chunk to discard.
@@ -55,15 +53,14 @@ PostcopyDiscardState *postcopy_discard_send_init(MigrationState *ms,
  * @start,@length: a range of pages in the migration bitmap in the
  *  RAM block passed to postcopy_discard_send_init() (length=1 is one page)
  */
-void postcopy_discard_send_range(MigrationState *ms, PostcopyDiscardState *pds,
-                                 unsigned long start, unsigned long length);
+void postcopy_discard_send_range(MigrationState *ms, unsigned long start,
+                                 unsigned long length);
 
 /*
  * Called at the end of each RAMBlock by the bitmap code.
- * Sends any outstanding discard messages, frees the PDS.
+ * Sends any outstanding discard messages.
  */
-void postcopy_discard_send_finish(MigrationState *ms,
-                                  PostcopyDiscardState *pds);
+void postcopy_discard_send_finish(MigrationState *ms);
 
 /*
  * Place a page (from) at (host) efficiently
diff --git a/migration/qemu-file-channel.c b/migration/qemu-file-channel.c
index 8e639eb496..c382ea2d78 100644
--- a/migration/qemu-file-channel.c
+++ b/migration/qemu-file-channel.c
@@ -33,7 +33,8 @@
 static ssize_t channel_writev_buffer(void *opaque,
                                      struct iovec *iov,
                                      int iovcnt,
-                                     int64_t pos)
+                                     int64_t pos,
+                                     Error **errp)
 {
     QIOChannel *ioc = QIO_CHANNEL(opaque);
     ssize_t done = 0;
@@ -47,7 +48,7 @@ static ssize_t channel_writev_buffer(void *opaque,
 
     while (nlocal_iov > 0) {
         ssize_t len;
-        len = qio_channel_writev(ioc, local_iov, nlocal_iov, NULL);
+        len = qio_channel_writev(ioc, local_iov, nlocal_iov, errp);
         if (len == QIO_CHANNEL_ERR_BLOCK) {
             if (qemu_in_coroutine()) {
                 qio_channel_yield(ioc, G_IO_OUT);
@@ -57,7 +58,6 @@ static ssize_t channel_writev_buffer(void *opaque,
             continue;
         }
         if (len < 0) {
-            /* XXX handle Error objects */
             done = -EIO;
             goto cleanup;
         }
@@ -75,13 +75,14 @@ static ssize_t channel_writev_buffer(void *opaque,
 static ssize_t channel_get_buffer(void *opaque,
                                   uint8_t *buf,
                                   int64_t pos,
-                                  size_t size)
+                                  size_t size,
+                                  Error **errp)
 {
     QIOChannel *ioc = QIO_CHANNEL(opaque);
     ssize_t ret;
 
     do {
-        ret = qio_channel_read(ioc, (char *)buf, size, NULL);
+        ret = qio_channel_read(ioc, (char *)buf, size, errp);
         if (ret < 0) {
             if (ret == QIO_CHANNEL_ERR_BLOCK) {
                 if (qemu_in_coroutine()) {
@@ -90,7 +91,6 @@ static ssize_t channel_get_buffer(void *opaque,
                     qio_channel_wait(ioc, G_IO_IN);
                 }
             } else {
-                /* XXX handle Error * object */
                 return -EIO;
             }
         }
@@ -100,18 +100,20 @@ static ssize_t channel_get_buffer(void *opaque,
 }
 
 
-static int channel_close(void *opaque)
+static int channel_close(void *opaque, Error **errp)
 {
+    int ret;
     QIOChannel *ioc = QIO_CHANNEL(opaque);
-    qio_channel_close(ioc, NULL);
+    ret = qio_channel_close(ioc, errp);
     object_unref(OBJECT(ioc));
-    return 0;
+    return ret;
 }
 
 
 static int channel_shutdown(void *opaque,
                             bool rd,
-                            bool wr)
+                            bool wr,
+                            Error **errp)
 {
     QIOChannel *ioc = QIO_CHANNEL(opaque);
 
@@ -125,8 +127,7 @@ static int channel_shutdown(void *opaque,
         } else {
             mode = QIO_CHANNEL_SHUTDOWN_WRITE;
         }
-        if (qio_channel_shutdown(ioc, mode, NULL) < 0) {
-            /* XXX handler Error * object */
+        if (qio_channel_shutdown(ioc, mode, errp) < 0) {
             return -EIO;
         }
     }
@@ -135,11 +136,12 @@ static int channel_shutdown(void *opaque,
 
 
 static int channel_set_blocking(void *opaque,
-                                bool enabled)
+                                bool enabled,
+                                Error **errp)
 {
     QIOChannel *ioc = QIO_CHANNEL(opaque);
 
-    if (qio_channel_set_blocking(ioc, enabled, NULL) < 0) {
+    if (qio_channel_set_blocking(ioc, enabled, errp) < 0) {
         return -1;
     }
     return 0;
diff --git a/migration/qemu-file.c b/migration/qemu-file.c
index 0431585502..e33c46764f 100644
--- a/migration/qemu-file.c
+++ b/migration/qemu-file.c
@@ -28,6 +28,7 @@
 #include "migration.h"
 #include "qemu-file.h"
 #include "trace.h"
+#include "qapi/error.h"
 
 #define IO_BUF_SIZE 32768
 #define MAX_IOV_SIZE MIN(IOV_MAX, 64)
@@ -51,6 +52,7 @@ struct QEMUFile {
     unsigned int iovcnt;
 
     int last_error;
+    Error *last_error_obj;
 };
 
 /*
@@ -62,7 +64,7 @@ int qemu_file_shutdown(QEMUFile *f)
     if (!f->ops->shut_down) {
         return -ENOSYS;
     }
-    return f->ops->shut_down(f->opaque, true, true);
+    return f->ops->shut_down(f->opaque, true, true, NULL);
 }
 
 /*
@@ -107,24 +109,55 @@ void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks)
 }
 
 /*
- * Get last error for stream f
+ * Get last error for stream f with optional Error*
  *
  * Return negative error value if there has been an error on previous
  * operations, return 0 if no error happened.
+ * Optional, it returns Error* in errp, but it may be NULL even if return value
+ * is not 0.
  *
  */
-int qemu_file_get_error(QEMUFile *f)
+int qemu_file_get_error_obj(QEMUFile *f, Error **errp)
 {
+    if (errp) {
+        *errp = f->last_error_obj ? error_copy(f->last_error_obj) : NULL;
+    }
     return f->last_error;
 }
 
-void qemu_file_set_error(QEMUFile *f, int ret)
+/*
+ * Set the last error for stream f with optional Error*
+ */
+void qemu_file_set_error_obj(QEMUFile *f, int ret, Error *err)
 {
-    if (f->last_error == 0) {
+    if (f->last_error == 0 && ret) {
         f->last_error = ret;
+        error_propagate(&f->last_error_obj, err);
+    } else if (err) {
+        error_report_err(err);
     }
 }
 
+/*
+ * Get last error for stream f
+ *
+ * Return negative error value if there has been an error on previous
+ * operations, return 0 if no error happened.
+ *
+ */
+int qemu_file_get_error(QEMUFile *f)
+{
+    return qemu_file_get_error_obj(f, NULL);
+}
+
+/*
+ * Set the last error for stream f
+ */
+void qemu_file_set_error(QEMUFile *f, int ret)
+{
+    qemu_file_set_error_obj(f, ret, NULL);
+}
+
 bool qemu_file_is_writable(QEMUFile *f)
 {
     return f->ops->writev_buffer;
@@ -176,6 +209,7 @@ void qemu_fflush(QEMUFile *f)
 {
     ssize_t ret = 0;
     ssize_t expect = 0;
+    Error *local_error = NULL;
 
     if (!qemu_file_is_writable(f)) {
         return;
@@ -183,7 +217,8 @@ void qemu_fflush(QEMUFile *f)
 
     if (f->iovcnt > 0) {
         expect = iov_size(f->iov, f->iovcnt);
-        ret = f->ops->writev_buffer(f->opaque, f->iov, f->iovcnt, f->pos);
+        ret = f->ops->writev_buffer(f->opaque, f->iov, f->iovcnt, f->pos,
+                                    &local_error);
 
         qemu_iovec_release_ram(f);
     }
@@ -195,7 +230,7 @@ void qemu_fflush(QEMUFile *f)
      * data set we requested, so sanity check that.
      */
     if (ret != expect) {
-        qemu_file_set_error(f, ret < 0 ? ret : -EIO);
+        qemu_file_set_error_obj(f, ret < 0 ? ret : -EIO, local_error);
     }
     f->buf_index = 0;
     f->iovcnt = 0;
@@ -283,6 +318,7 @@ static ssize_t qemu_fill_buffer(QEMUFile *f)
 {
     int len;
     int pending;
+    Error *local_error = NULL;
 
     assert(!qemu_file_is_writable(f));
 
@@ -294,14 +330,16 @@ static ssize_t qemu_fill_buffer(QEMUFile *f)
     f->buf_size = pending;
 
     len = f->ops->get_buffer(f->opaque, f->buf + pending, f->pos,
-                        IO_BUF_SIZE - pending);
+                             IO_BUF_SIZE - pending, &local_error);
     if (len > 0) {
         f->buf_size += len;
         f->pos += len;
     } else if (len == 0) {
-        qemu_file_set_error(f, -EIO);
+        qemu_file_set_error_obj(f, -EIO, local_error);
     } else if (len != -EAGAIN) {
-        qemu_file_set_error(f, len);
+        qemu_file_set_error_obj(f, len, local_error);
+    } else {
+        error_free(local_error);
     }
 
     return len;
@@ -327,7 +365,7 @@ int qemu_fclose(QEMUFile *f)
     ret = qemu_file_get_error(f);
 
     if (f->ops->close) {
-        int ret2 = f->ops->close(f->opaque);
+        int ret2 = f->ops->close(f->opaque, NULL);
         if (ret >= 0) {
             ret = ret2;
         }
@@ -338,6 +376,7 @@ int qemu_fclose(QEMUFile *f)
     if (f->last_error) {
         ret = f->last_error;
     }
+    error_free(f->last_error_obj);
     g_free(f);
     trace_qemu_file_fclose();
     return ret;
@@ -615,6 +654,11 @@ void qemu_file_reset_rate_limit(QEMUFile *f)
     f->bytes_xfer = 0;
 }
 
+void qemu_file_update_transfer(QEMUFile *f, int64_t len)
+{
+    f->bytes_xfer += len;
+}
+
 void qemu_put_be16(QEMUFile *f, unsigned int v)
 {
     qemu_put_byte(f, v >> 8);
@@ -783,6 +827,6 @@ void qemu_put_counted_string(QEMUFile *f, const char *str)
 void qemu_file_set_blocking(QEMUFile *f, bool block)
 {
     if (f->ops->set_blocking) {
-        f->ops->set_blocking(f->opaque, block);
+        f->ops->set_blocking(f->opaque, block, NULL);
     }
 }
diff --git a/migration/qemu-file.h b/migration/qemu-file.h
index 13baf896bd..b6303dbeef 100644
--- a/migration/qemu-file.h
+++ b/migration/qemu-file.h
@@ -32,7 +32,8 @@
  * bytes actually read should be returned.
  */
 typedef ssize_t (QEMUFileGetBufferFunc)(void *opaque, uint8_t *buf,
-                                        int64_t pos, size_t size);
+                                        int64_t pos, size_t size,
+                                        Error **errp);
 
 /* Close a file
  *
@@ -41,7 +42,7 @@ typedef ssize_t (QEMUFileGetBufferFunc)(void *opaque, uint8_t *buf,
  * The meaning of return value on success depends on the specific back-end being
  * used.
  */
-typedef int (QEMUFileCloseFunc)(void *opaque);
+typedef int (QEMUFileCloseFunc)(void *opaque, Error **errp);
 
 /* Called to return the OS file descriptor associated to the QEMUFile.
  */
@@ -49,14 +50,15 @@ typedef int (QEMUFileGetFD)(void *opaque);
 
 /* Called to change the blocking mode of the file
  */
-typedef int (QEMUFileSetBlocking)(void *opaque, bool enabled);
+typedef int (QEMUFileSetBlocking)(void *opaque, bool enabled, Error **errp);
 
 /*
  * This function writes an iovec to file. The handler must write all
  * of the data or return a negative errno value.
  */
 typedef ssize_t (QEMUFileWritevBufferFunc)(void *opaque, struct iovec *iov,
-                                           int iovcnt, int64_t pos);
+                                           int iovcnt, int64_t pos,
+                                           Error **errp);
 
 /*
  * This function provides hooks around different
@@ -97,7 +99,8 @@ typedef QEMUFile *(QEMURetPathFunc)(void *opaque);
  * Existing blocking reads/writes must be woken
  * Returns 0 on success, -err on error
  */
-typedef int (QEMUFileShutdownFunc)(void *opaque, bool rd, bool wr);
+typedef int (QEMUFileShutdownFunc)(void *opaque, bool rd, bool wr,
+                                   Error **errp);
 
 typedef struct QEMUFileOps {
     QEMUFileGetBufferFunc *get_buffer;
@@ -147,16 +150,17 @@ int qemu_peek_byte(QEMUFile *f, int offset);
 void qemu_file_skip(QEMUFile *f, int size);
 void qemu_update_position(QEMUFile *f, size_t size);
 void qemu_file_reset_rate_limit(QEMUFile *f);
+void qemu_file_update_transfer(QEMUFile *f, int64_t len);
 void qemu_file_set_rate_limit(QEMUFile *f, int64_t new_rate);
 int64_t qemu_file_get_rate_limit(QEMUFile *f);
+int qemu_file_get_error_obj(QEMUFile *f, Error **errp);
+void qemu_file_set_error_obj(QEMUFile *f, int ret, Error *err);
 void qemu_file_set_error(QEMUFile *f, int ret);
 int qemu_file_shutdown(QEMUFile *f);
 QEMUFile *qemu_file_get_return_path(QEMUFile *f);
 void qemu_fflush(QEMUFile *f);
 void qemu_file_set_blocking(QEMUFile *f, bool block);
 
-size_t qemu_get_counted_string(QEMUFile *f, char buf[256]);
-
 void ram_control_before_iterate(QEMUFile *f, uint64_t flags);
 void ram_control_after_iterate(QEMUFile *f, uint64_t flags);
 void ram_control_load_hook(QEMUFile *f, uint64_t flags, void *data);
@@ -175,6 +179,4 @@ size_t ram_control_save_page(QEMUFile *f, ram_addr_t block_offset,
                              ram_addr_t offset, size_t size,
                              uint64_t *bytes_sent);
 
-void qemu_put_counted_string(QEMUFile *f, const char *name);
-
 #endif
diff --git a/migration/ram.c b/migration/ram.c
index 889148dd84..35552c090b 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -661,6 +661,8 @@ typedef struct {
     uint64_t num_packets;
     /* pages sent through this channel */
     uint64_t num_pages;
+    /* syncs main thread and channels */
+    QemuSemaphore sem_sync;
 }  MultiFDSendParams;
 
 typedef struct {
@@ -896,8 +898,6 @@ struct {
     MultiFDSendParams *params;
     /* array of pages to sent */
     MultiFDPages_t *pages;
-    /* syncs main thread and channels */
-    QemuSemaphore sem_sync;
     /* global number of generated multifd packets */
     uint64_t packet_num;
     /* send channels ready */
@@ -922,7 +922,7 @@ struct {
  * false.
  */
 
-static int multifd_send_pages(void)
+static int multifd_send_pages(RAMState *rs)
 {
     int i;
     static int next_channel;
@@ -954,6 +954,7 @@ static int multifd_send_pages(void)
     multifd_send_state->pages = p->pages;
     p->pages = pages;
     transferred = ((uint64_t) pages->used) * TARGET_PAGE_SIZE + p->packet_len;
+    qemu_file_update_transfer(rs->f, transferred);
     ram_counters.multifd_bytes += transferred;
     ram_counters.transferred += transferred;;
     qemu_mutex_unlock(&p->mutex);
@@ -962,7 +963,7 @@ static int multifd_send_pages(void)
     return 1;
 }
 
-static int multifd_queue_page(RAMBlock *block, ram_addr_t offset)
+static int multifd_queue_page(RAMState *rs, RAMBlock *block, ram_addr_t offset)
 {
     MultiFDPages_t *pages = multifd_send_state->pages;
 
@@ -981,12 +982,12 @@ static int multifd_queue_page(RAMBlock *block, ram_addr_t offset)
         }
     }
 
-    if (multifd_send_pages() < 0) {
+    if (multifd_send_pages(rs) < 0) {
         return -1;
     }
 
     if (pages->block != block) {
-        return  multifd_queue_page(block, offset);
+        return  multifd_queue_page(rs, block, offset);
     }
 
     return 1;
@@ -996,6 +997,8 @@ static void multifd_send_terminate_threads(Error *err)
 {
     int i;
 
+    trace_multifd_send_terminate_threads(err != NULL);
+
     if (err) {
         MigrationState *s = migrate_get_current();
         migrate_set_error(s, err);
@@ -1036,6 +1039,7 @@ void multifd_save_cleanup(void)
         p->c = NULL;
         qemu_mutex_destroy(&p->mutex);
         qemu_sem_destroy(&p->sem);
+        qemu_sem_destroy(&p->sem_sync);
         g_free(p->name);
         p->name = NULL;
         multifd_pages_clear(p->pages);
@@ -1045,7 +1049,6 @@ void multifd_save_cleanup(void)
         p->packet = NULL;
     }
     qemu_sem_destroy(&multifd_send_state->channels_ready);
-    qemu_sem_destroy(&multifd_send_state->sem_sync);
     g_free(multifd_send_state->params);
     multifd_send_state->params = NULL;
     multifd_pages_clear(multifd_send_state->pages);
@@ -1054,7 +1057,7 @@ void multifd_save_cleanup(void)
     multifd_send_state = NULL;
 }
 
-static void multifd_send_sync_main(void)
+static void multifd_send_sync_main(RAMState *rs)
 {
     int i;
 
@@ -1062,7 +1065,7 @@ static void multifd_send_sync_main(void)
         return;
     }
     if (multifd_send_state->pages->used) {
-        if (multifd_send_pages() < 0) {
+        if (multifd_send_pages(rs) < 0) {
             error_report("%s: multifd_send_pages fail", __func__);
             return;
         }
@@ -1083,6 +1086,9 @@ static void multifd_send_sync_main(void)
         p->packet_num = multifd_send_state->packet_num++;
         p->flags |= MULTIFD_FLAG_SYNC;
         p->pending_job++;
+        qemu_file_update_transfer(rs->f, p->packet_len);
+        ram_counters.multifd_bytes += p->packet_len;
+        ram_counters.transferred += p->packet_len;
         qemu_mutex_unlock(&p->mutex);
         qemu_sem_post(&p->sem);
     }
@@ -1090,7 +1096,7 @@ static void multifd_send_sync_main(void)
         MultiFDSendParams *p = &multifd_send_state->params[i];
 
         trace_multifd_send_sync_main_wait(p->id);
-        qemu_sem_wait(&multifd_send_state->sem_sync);
+        qemu_sem_wait(&p->sem_sync);
     }
     trace_multifd_send_sync_main(multifd_send_state->packet_num);
 }
@@ -1150,7 +1156,7 @@ static void *multifd_send_thread(void *opaque)
             qemu_mutex_unlock(&p->mutex);
 
             if (flags & MULTIFD_FLAG_SYNC) {
-                qemu_sem_post(&multifd_send_state->sem_sync);
+                qemu_sem_post(&p->sem_sync);
             }
             qemu_sem_post(&multifd_send_state->channels_ready);
         } else if (p->quit) {
@@ -1164,6 +1170,7 @@ static void *multifd_send_thread(void *opaque)
 
 out:
     if (local_err) {
+        trace_multifd_send_error(p->id);
         multifd_send_terminate_threads(local_err);
     }
 
@@ -1173,7 +1180,7 @@ out:
      */
     if (ret != 0) {
         if (flags & MULTIFD_FLAG_SYNC) {
-            qemu_sem_post(&multifd_send_state->sem_sync);
+            qemu_sem_post(&p->sem_sync);
         }
         qemu_sem_post(&multifd_send_state->channels_ready);
     }
@@ -1194,6 +1201,7 @@ static void multifd_new_send_channel_async(QIOTask *task, gpointer opaque)
     QIOChannel *sioc = QIO_CHANNEL(qio_task_get_source(task));
     Error *local_err = NULL;
 
+    trace_multifd_new_send_channel_async(p->id);
     if (qio_task_propagate_error(task, &local_err)) {
         migrate_set_error(migrate_get_current(), local_err);
         multifd_save_cleanup();
@@ -1219,7 +1227,6 @@ int multifd_save_setup(void)
     multifd_send_state = g_malloc0(sizeof(*multifd_send_state));
     multifd_send_state->params = g_new0(MultiFDSendParams, thread_count);
     multifd_send_state->pages = multifd_pages_init(page_count);
-    qemu_sem_init(&multifd_send_state->sem_sync, 0);
     qemu_sem_init(&multifd_send_state->channels_ready, 0);
 
     for (i = 0; i < thread_count; i++) {
@@ -1227,6 +1234,7 @@ int multifd_save_setup(void)
 
         qemu_mutex_init(&p->mutex);
         qemu_sem_init(&p->sem, 0);
+        qemu_sem_init(&p->sem_sync, 0);
         p->quit = false;
         p->pending_job = 0;
         p->id = i;
@@ -1254,6 +1262,8 @@ static void multifd_recv_terminate_threads(Error *err)
 {
     int i;
 
+    trace_multifd_recv_terminate_threads(err != NULL);
+
     if (err) {
         MigrationState *s = migrate_get_current();
         migrate_set_error(s, err);
@@ -1478,6 +1488,7 @@ bool multifd_recv_new_channel(QIOChannel *ioc, Error **errp)
                                 atomic_read(&multifd_recv_state->count));
         return false;
     }
+    trace_multifd_recv_new_channel(id);
 
     p = &multifd_recv_state->params[id];
     if (p->c != NULL) {
@@ -1748,11 +1759,10 @@ static inline bool migration_bitmap_clear_dirty(RAMState *rs,
 }
 
 /* Called with RCU critical section */
-static void migration_bitmap_sync_range(RAMState *rs, RAMBlock *rb,
-                                        ram_addr_t length)
+static void ramblock_sync_dirty_bitmap(RAMState *rs, RAMBlock *rb)
 {
     rs->migration_dirty_pages +=
-        cpu_physical_memory_sync_dirty_bitmap(rb, 0, length,
+        cpu_physical_memory_sync_dirty_bitmap(rb, 0, rb->used_length,
                                               &rs->num_dirty_pages_period);
 }
 
@@ -1841,7 +1851,7 @@ static void migration_bitmap_sync(RAMState *rs)
     qemu_mutex_lock(&rs->bitmap_mutex);
     rcu_read_lock();
     RAMBLOCK_FOREACH_NOT_IGNORED(block) {
-        migration_bitmap_sync_range(rs, block, block->used_length);
+        ramblock_sync_dirty_bitmap(rs, block);
     }
     ram_counters.remaining = ram_bytes_remaining();
     rcu_read_unlock();
@@ -2079,7 +2089,7 @@ static int ram_save_page(RAMState *rs, PageSearchStatus *pss, bool last_stage)
 static int ram_save_multifd_page(RAMState *rs, RAMBlock *block,
                                  ram_addr_t offset)
 {
-    if (multifd_queue_page(block, offset) < 0) {
+    if (multifd_queue_page(rs, block, offset) < 0) {
         return -1;
     }
     ram_counters.normal++;
@@ -2851,12 +2861,9 @@ void ram_postcopy_migrated_memory_release(MigrationState *ms)
  *       with the dirtymap; so a '1' means it's either dirty or unsent.
  *
  * @ms: current migration state
- * @pds: state for postcopy
  * @block: RAMBlock to discard
  */
-static int postcopy_send_discard_bm_ram(MigrationState *ms,
-                                        PostcopyDiscardState *pds,
-                                        RAMBlock *block)
+static int postcopy_send_discard_bm_ram(MigrationState *ms, RAMBlock *block)
 {
     unsigned long end = block->used_length >> TARGET_PAGE_BITS;
     unsigned long current;
@@ -2864,23 +2871,21 @@ static int postcopy_send_discard_bm_ram(MigrationState *ms,
 
     for (current = 0; current < end; ) {
         unsigned long one = find_next_bit(unsentmap, end, current);
+        unsigned long zero, discard_length;
+
+        if (one >= end) {
+            break;
+        }
 
-        if (one <= end) {
-            unsigned long zero = find_next_zero_bit(unsentmap, end, one + 1);
-            unsigned long discard_length;
+        zero = find_next_zero_bit(unsentmap, end, one + 1);
 
-            if (zero >= end) {
-                discard_length = end - one;
-            } else {
-                discard_length = zero - one;
-            }
-            if (discard_length) {
-                postcopy_discard_send_range(ms, pds, one, discard_length);
-            }
-            current = one + discard_length;
+        if (zero >= end) {
+            discard_length = end - one;
         } else {
-            current = one;
+            discard_length = zero - one;
         }
+        postcopy_discard_send_range(ms, one, discard_length);
+        current = one + discard_length;
     }
 
     return 0;
@@ -2905,16 +2910,15 @@ static int postcopy_each_ram_send_discard(MigrationState *ms)
     int ret;
 
     RAMBLOCK_FOREACH_NOT_IGNORED(block) {
-        PostcopyDiscardState *pds =
-            postcopy_discard_send_init(ms, block->idstr);
+        postcopy_discard_send_init(ms, block->idstr);
 
         /*
          * Postcopy sends chunks of bitmap over the wire, but it
          * just needs indexes at this point, avoids it having
          * target page specific code.
          */
-        ret = postcopy_send_discard_bm_ram(ms, pds, block);
-        postcopy_discard_send_finish(ms, pds);
+        ret = postcopy_send_discard_bm_ram(ms, block);
+        postcopy_discard_send_finish(ms);
         if (ret) {
             return ret;
         }
@@ -2937,11 +2941,9 @@ static int postcopy_each_ram_send_discard(MigrationState *ms)
  * @unsent_pass: if true we need to canonicalize partially unsent host pages
  *               otherwise we need to canonicalize partially dirty host pages
  * @block: block that contains the page we want to canonicalize
- * @pds: state for postcopy
  */
 static void postcopy_chunk_hostpages_pass(MigrationState *ms, bool unsent_pass,
-                                          RAMBlock *block,
-                                          PostcopyDiscardState *pds)
+                                          RAMBlock *block)
 {
     RAMState *rs = ram_state;
     unsigned long *bitmap = block->bmap;
@@ -2964,54 +2966,30 @@ static void postcopy_chunk_hostpages_pass(MigrationState *ms, bool unsent_pass,
     }
 
     while (run_start < pages) {
-        bool do_fixup = false;
-        unsigned long fixup_start_addr;
-        unsigned long host_offset;
 
         /*
          * If the start of this run of pages is in the middle of a host
          * page, then we need to fixup this host page.
          */
-        host_offset = run_start % host_ratio;
-        if (host_offset) {
-            do_fixup = true;
-            run_start -= host_offset;
-            fixup_start_addr = run_start;
-            /* For the next pass */
-            run_start = run_start + host_ratio;
-        } else {
+        if (QEMU_IS_ALIGNED(run_start, host_ratio)) {
             /* Find the end of this run */
-            unsigned long run_end;
             if (unsent_pass) {
-                run_end = find_next_bit(unsentmap, pages, run_start + 1);
+                run_start = find_next_bit(unsentmap, pages, run_start + 1);
             } else {
-                run_end = find_next_zero_bit(bitmap, pages, run_start + 1);
+                run_start = find_next_zero_bit(bitmap, pages, run_start + 1);
             }
             /*
              * If the end isn't at the start of a host page, then the
              * run doesn't finish at the end of a host page
              * and we need to discard.
              */
-            host_offset = run_end % host_ratio;
-            if (host_offset) {
-                do_fixup = true;
-                fixup_start_addr = run_end - host_offset;
-                /*
-                 * This host page has gone, the next loop iteration starts
-                 * from after the fixup
-                 */
-                run_start = fixup_start_addr + host_ratio;
-            } else {
-                /*
-                 * No discards on this iteration, next loop starts from
-                 * next sent/dirty page
-                 */
-                run_start = run_end + 1;
-            }
         }
 
-        if (do_fixup) {
+        if (!QEMU_IS_ALIGNED(run_start, host_ratio)) {
             unsigned long page;
+            unsigned long fixup_start_addr = QEMU_ALIGN_DOWN(run_start,
+                                                             host_ratio);
+            run_start = QEMU_ALIGN_UP(run_start, host_ratio);
 
             /* Tell the destination to discard this page */
             if (unsent_pass || !test_bit(fixup_start_addr, unsentmap)) {
@@ -3022,8 +3000,7 @@ static void postcopy_chunk_hostpages_pass(MigrationState *ms, bool unsent_pass,
                  *     (any partially sent pages were already discarded
                  *     by the previous unsent_pass)
                  */
-                postcopy_discard_send_range(ms, pds, fixup_start_addr,
-                                            host_ratio);
+                postcopy_discard_send_range(ms, fixup_start_addr, host_ratio);
             }
 
             /* Clean up the bitmap */
@@ -3066,18 +3043,17 @@ static void postcopy_chunk_hostpages_pass(MigrationState *ms, bool unsent_pass,
  */
 static int postcopy_chunk_hostpages(MigrationState *ms, RAMBlock *block)
 {
-    PostcopyDiscardState *pds =
-        postcopy_discard_send_init(ms, block->idstr);
+    postcopy_discard_send_init(ms, block->idstr);
 
     /* First pass: Discard all partially sent host pages */
-    postcopy_chunk_hostpages_pass(ms, true, block, pds);
+    postcopy_chunk_hostpages_pass(ms, true, block);
     /*
      * Second pass: Ensure that all partially dirty host pages are made
      * fully dirty.
      */
-    postcopy_chunk_hostpages_pass(ms, false, block, pds);
+    postcopy_chunk_hostpages_pass(ms, false, block);
 
-    postcopy_discard_send_finish(ms, pds);
+    postcopy_discard_send_finish(ms);
     return 0;
 }
 
@@ -3482,7 +3458,7 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
     ram_control_before_iterate(f, RAM_CONTROL_SETUP);
     ram_control_after_iterate(f, RAM_CONTROL_SETUP);
 
-    multifd_send_sync_main();
+    multifd_send_sync_main(*rsp);
     qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
     qemu_fflush(f);
 
@@ -3570,7 +3546,7 @@ static int ram_save_iterate(QEMUFile *f, void *opaque)
     ram_control_after_iterate(f, RAM_CONTROL_ROUND);
 
 out:
-    multifd_send_sync_main();
+    multifd_send_sync_main(rs);
     qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
     qemu_fflush(f);
     ram_counters.transferred += 8;
@@ -3629,7 +3605,7 @@ static int ram_save_complete(QEMUFile *f, void *opaque)
 
     rcu_read_unlock();
 
-    multifd_send_sync_main();
+    multifd_send_sync_main(rs);
     qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
     qemu_fflush(f);
 
@@ -4296,7 +4272,7 @@ static void colo_flush_ram_cache(void)
     memory_global_dirty_log_sync();
     rcu_read_lock();
     RAMBLOCK_FOREACH_NOT_IGNORED(block) {
-        migration_bitmap_sync_range(ram_state, block, block->used_length);
+        ramblock_sync_dirty_bitmap(ram_state, block);
     }
     rcu_read_unlock();
 
@@ -4322,40 +4298,26 @@ static void colo_flush_ram_cache(void)
     trace_colo_flush_ram_cache_end();
 }
 
-static int ram_load(QEMUFile *f, void *opaque, int version_id)
+/**
+ * ram_load_precopy: load pages in precopy case
+ *
+ * Returns 0 for success or -errno in case of error
+ *
+ * Called in precopy mode by ram_load().
+ * rcu_read_lock is taken prior to this being called.
+ *
+ * @f: QEMUFile where to send the data
+ */
+static int ram_load_precopy(QEMUFile *f)
 {
-    int flags = 0, ret = 0, invalid_flags = 0;
-    static uint64_t seq_iter;
-    int len = 0;
-    /*
-     * If system is running in postcopy mode, page inserts to host memory must
-     * be atomic
-     */
-    bool postcopy_running = postcopy_is_running();
+    int flags = 0, ret = 0, invalid_flags = 0, len = 0;
     /* ADVISE is earlier, it shows the source has the postcopy capability on */
     bool postcopy_advised = postcopy_is_advised();
-
-    seq_iter++;
-
-    if (version_id != 4) {
-        ret = -EINVAL;
-    }
-
     if (!migrate_use_compression()) {
         invalid_flags |= RAM_SAVE_FLAG_COMPRESS_PAGE;
     }
-    /* This RCU critical section can be very long running.
-     * When RCU reclaims in the code start to become numerous,
-     * it will be necessary to reduce the granularity of this
-     * critical section.
-     */
-    rcu_read_lock();
-
-    if (postcopy_running) {
-        ret = ram_load_postcopy(f);
-    }
 
-    while (!postcopy_running && !ret && !(flags & RAM_SAVE_FLAG_EOS)) {
+    while (!ret && !(flags & RAM_SAVE_FLAG_EOS)) {
         ram_addr_t addr, total_ram_bytes;
         void *host = NULL;
         uint8_t ch;
@@ -4506,6 +4468,39 @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
         }
     }
 
+    return ret;
+}
+
+static int ram_load(QEMUFile *f, void *opaque, int version_id)
+{
+    int ret = 0;
+    static uint64_t seq_iter;
+    /*
+     * If system is running in postcopy mode, page inserts to host memory must
+     * be atomic
+     */
+    bool postcopy_running = postcopy_is_running();
+
+    seq_iter++;
+
+    if (version_id != 4) {
+        return -EINVAL;
+    }
+
+    /*
+     * This RCU critical section can be very long running.
+     * When RCU reclaims in the code start to become numerous,
+     * it will be necessary to reduce the granularity of this
+     * critical section.
+     */
+    rcu_read_lock();
+
+    if (postcopy_running) {
+        ret = ram_load_postcopy(f);
+    } else {
+        ret = ram_load_precopy(f);
+    }
+
     ret |= wait_for_decompress_done();
     rcu_read_unlock();
     trace_ram_load_complete(ret, seq_iter);
diff --git a/migration/rdma.c b/migration/rdma.c
index 3036221ee8..0e73e759ca 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -3140,7 +3140,7 @@ static size_t qemu_rdma_save_page(QEMUFile *f, void *opaque,
 
     CHECK_ERROR_STATE();
 
-    if (migrate_get_current()->state == MIGRATION_STATUS_POSTCOPY_ACTIVE) {
+    if (migration_in_postcopy()) {
         rcu_read_unlock();
         return RAM_SAVE_CONTROL_NOT_SUPP;
     }
@@ -3775,7 +3775,7 @@ static int qemu_rdma_registration_start(QEMUFile *f, void *opaque,
 
     CHECK_ERROR_STATE();
 
-    if (migrate_get_current()->state == MIGRATION_STATUS_POSTCOPY_ACTIVE) {
+    if (migration_in_postcopy()) {
         rcu_read_unlock();
         return 0;
     }
@@ -3810,7 +3810,7 @@ static int qemu_rdma_registration_stop(QEMUFile *f, void *opaque,
 
     CHECK_ERROR_STATE();
 
-    if (migrate_get_current()->state == MIGRATION_STATUS_POSTCOPY_ACTIVE) {
+    if (migration_in_postcopy()) {
         rcu_read_unlock();
         return 0;
     }
diff --git a/migration/savevm.c b/migration/savevm.c
index 79ed44d475..6369a4ff7a 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -124,7 +124,7 @@ static struct mig_cmd_args {
 /* savevm/loadvm support */
 
 static ssize_t block_writev_buffer(void *opaque, struct iovec *iov, int iovcnt,
-                                   int64_t pos)
+                                   int64_t pos, Error **errp)
 {
     int ret;
     QEMUIOVector qiov;
@@ -139,12 +139,12 @@ static ssize_t block_writev_buffer(void *opaque, struct iovec *iov, int iovcnt,
 }
 
 static ssize_t block_get_buffer(void *opaque, uint8_t *buf, int64_t pos,
-                                size_t size)
+                                size_t size, Error **errp)
 {
     return bdrv_load_vmstate(opaque, buf, pos, size);
 }
 
-static int bdrv_fclose(void *opaque)
+static int bdrv_fclose(void *opaque, Error **errp)
 {
     return bdrv_flush(opaque);
 }
@@ -1246,29 +1246,16 @@ void qemu_savevm_state_complete_postcopy(QEMUFile *f)
     qemu_fflush(f);
 }
 
-int qemu_savevm_state_complete_precopy(QEMUFile *f, bool iterable_only,
-                                       bool inactivate_disks)
+static
+int qemu_savevm_state_complete_precopy_iterable(QEMUFile *f, bool in_postcopy)
 {
-    QJSON *vmdesc;
-    int vmdesc_len;
     SaveStateEntry *se;
     int ret;
-    bool in_postcopy = migration_in_postcopy();
-    Error *local_err = NULL;
-
-    if (precopy_notify(PRECOPY_NOTIFY_COMPLETE, &local_err)) {
-        error_report_err(local_err);
-    }
-
-    trace_savevm_state_complete_precopy();
-
-    cpu_synchronize_all_states();
 
     QTAILQ_FOREACH(se, &savevm_state.handlers, entry) {
         if (!se->ops ||
             (in_postcopy && se->ops->has_postcopy &&
              se->ops->has_postcopy(se->opaque)) ||
-            (in_postcopy && !iterable_only) ||
             !se->ops->save_live_complete_precopy) {
             continue;
         }
@@ -1291,9 +1278,18 @@ int qemu_savevm_state_complete_precopy(QEMUFile *f, bool iterable_only,
         }
     }
 
-    if (iterable_only) {
-        return 0;
-    }
+    return 0;
+}
+
+static
+int qemu_savevm_state_complete_precopy_non_iterable(QEMUFile *f,
+                                                    bool in_postcopy,
+                                                    bool inactivate_disks)
+{
+    QJSON *vmdesc;
+    int vmdesc_len;
+    SaveStateEntry *se;
+    int ret;
 
     vmdesc = qjson_new();
     json_prop_int(vmdesc, "page_size", qemu_target_page_size());
@@ -1353,6 +1349,42 @@ int qemu_savevm_state_complete_precopy(QEMUFile *f, bool iterable_only,
     }
     qjson_destroy(vmdesc);
 
+    return 0;
+}
+
+int qemu_savevm_state_complete_precopy(QEMUFile *f, bool iterable_only,
+                                       bool inactivate_disks)
+{
+    int ret;
+    Error *local_err = NULL;
+    bool in_postcopy = migration_in_postcopy();
+
+    if (precopy_notify(PRECOPY_NOTIFY_COMPLETE, &local_err)) {
+        error_report_err(local_err);
+    }
+
+    trace_savevm_state_complete_precopy();
+
+    cpu_synchronize_all_states();
+
+    if (!in_postcopy || iterable_only) {
+        ret = qemu_savevm_state_complete_precopy_iterable(f, in_postcopy);
+        if (ret) {
+            return ret;
+        }
+    }
+
+    if (iterable_only) {
+        goto flush;
+    }
+
+    ret = qemu_savevm_state_complete_precopy_non_iterable(f, in_postcopy,
+                                                          inactivate_disks);
+    if (ret) {
+        return ret;
+    }
+
+flush:
     qemu_fflush(f);
     return 0;
 }
@@ -1424,6 +1456,7 @@ static int qemu_savevm_state(QEMUFile *f, Error **errp)
     }
 
     migrate_init(ms);
+    memset(&ram_counters, 0, sizeof(ram_counters));
     ms->to_dst_file = f;
 
     qemu_mutex_unlock_iothread();
@@ -1616,8 +1649,6 @@ static int loadvm_postcopy_handle_advise(MigrationIncomingState *mis,
         return -1;
     }
 
-    postcopy_state_set(POSTCOPY_INCOMING_ADVISE);
-
     return 0;
 }
 
@@ -1835,16 +1866,10 @@ static int loadvm_postcopy_handle_listen(MigrationIncomingState *mis)
     return 0;
 }
 
-
-typedef struct {
-    QEMUBH *bh;
-} HandleRunBhData;
-
 static void loadvm_postcopy_handle_run_bh(void *opaque)
 {
     Error *local_err = NULL;
-    HandleRunBhData *data = opaque;
-    MigrationIncomingState *mis = migration_incoming_get_current();
+    MigrationIncomingState *mis = opaque;
 
     /* TODO we should move all of this lot into postcopy_ram.c or a shared code
      * in migration.c
@@ -1876,15 +1901,13 @@ static void loadvm_postcopy_handle_run_bh(void *opaque)
         runstate_set(RUN_STATE_PAUSED);
     }
 
-    qemu_bh_delete(data->bh);
-    g_free(data);
+    qemu_bh_delete(mis->bh);
 }
 
 /* After all discards we can start running and asking for pages */
 static int loadvm_postcopy_handle_run(MigrationIncomingState *mis)
 {
     PostcopyState ps = postcopy_state_set(POSTCOPY_INCOMING_RUNNING);
-    HandleRunBhData *data;
 
     trace_loadvm_postcopy_handle_run();
     if (ps != POSTCOPY_INCOMING_LISTENING) {
@@ -1892,9 +1915,8 @@ static int loadvm_postcopy_handle_run(MigrationIncomingState *mis)
         return -1;
     }
 
-    data = g_new(HandleRunBhData, 1);
-    data->bh = qemu_bh_new(loadvm_postcopy_handle_run_bh, data);
-    qemu_bh_schedule(data->bh);
+    mis->bh = qemu_bh_new(loadvm_postcopy_handle_run_bh, mis);
+    qemu_bh_schedule(mis->bh);
 
     /* We need to finish reading the stream from the package
      * and also stop reading anything more from the stream that loaded the
@@ -2407,7 +2429,7 @@ retry:
         case QEMU_VM_COMMAND:
             ret = loadvm_process_command(f);
             trace_qemu_loadvm_state_section_command(ret);
-            if ((ret < 0) || (ret & LOADVM_QUIT)) {
+            if ((ret < 0) || (ret == LOADVM_QUIT)) {
                 goto out;
             }
             break;
diff --git a/migration/trace-events b/migration/trace-events
index d8e54c367a..00ffcd5930 100644
--- a/migration/trace-events
+++ b/migration/trace-events
@@ -81,16 +81,22 @@ migration_bitmap_sync_start(void) ""
 migration_bitmap_sync_end(uint64_t dirty_pages) "dirty_pages %" PRIu64
 migration_bitmap_clear_dirty(char *str, uint64_t start, uint64_t size, unsigned long page) "rb %s start 0x%"PRIx64" size 0x%"PRIx64" page 0x%lx"
 migration_throttle(void) ""
+multifd_new_send_channel_async(uint8_t id) "channel %d"
 multifd_recv(uint8_t id, uint64_t packet_num, uint32_t used, uint32_t flags, uint32_t next_packet_size) "channel %d packet_num %" PRIu64 " pages %d flags 0x%x next packet size %d"
+multifd_recv_new_channel(uint8_t id) "channel %d"
 multifd_recv_sync_main(long packet_num) "packet num %ld"
 multifd_recv_sync_main_signal(uint8_t id) "channel %d"
 multifd_recv_sync_main_wait(uint8_t id) "channel %d"
+multifd_recv_terminate_threads(bool error) "error %d"
 multifd_recv_thread_end(uint8_t id, uint64_t packets, uint64_t pages) "channel %d packets %" PRIu64 " pages %" PRIu64
 multifd_recv_thread_start(uint8_t id) "%d"
+multifd_save_setup_wait(uint8_t id) "%d"
 multifd_send(uint8_t id, uint64_t packet_num, uint32_t used, uint32_t flags, uint32_t next_packet_size) "channel %d packet_num %" PRIu64 " pages %d flags 0x%x next packet size %d"
+multifd_send_error(uint8_t id) "channel %d"
 multifd_send_sync_main(long packet_num) "packet num %ld"
 multifd_send_sync_main_signal(uint8_t id) "channel %d"
 multifd_send_sync_main_wait(uint8_t id) "channel %d"
+multifd_send_terminate_threads(bool error) "error %d"
 multifd_send_thread_end(uint8_t id, uint64_t packets, uint64_t pages) "channel %d packets %" PRIu64 " pages %"  PRIu64
 multifd_send_thread_start(uint8_t id) "%d"
 ram_discard_range(const char *rbname, uint64_t start, size_t len) "%s: start: %" PRIx64 " %zx"