diff --git a/migration/colo.c b/migration/colo.c
index 93c5a452fb..44942c4e23 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -26,6 +26,7 @@
 #include "qemu/main-loop.h"
 #include "qemu/rcu.h"
 #include "migration/failover.h"
+#include "migration/ram.h"
 #ifdef CONFIG_REPLICATION
 #include "replication.h"
 #endif
@@ -845,6 +846,8 @@ void *colo_process_incoming_thread(void *opaque)
      */
     qemu_file_set_blocking(mis->from_src_file, true);
 
+    colo_incoming_start_dirty_log();
+
     bioc = qio_channel_buffer_new(COLO_BUFFER_BASE_SIZE);
     fb = qemu_fopen_channel_input(QIO_CHANNEL(bioc));
     object_unref(OBJECT(bioc));
diff --git a/migration/ram.c b/migration/ram.c
index 02cfd76d19..31f4a9dc09 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -2994,7 +2994,6 @@ int colo_init_ram_cache(void)
                 }
                 return -errno;
             }
-            memcpy(block->colo_cache, block->host, block->used_length);
         }
     }
 
@@ -3008,19 +3007,36 @@ int colo_init_ram_cache(void)
 
         RAMBLOCK_FOREACH_NOT_IGNORED(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;
-    qemu_mutex_init(&ram_state->bitmap_mutex);
-    memory_global_dirty_log_start();
 
+    ram_state_init(&ram_state);
     return 0;
 }
 
+/* TODO: duplicated with ram_init_bitmaps */
+void colo_incoming_start_dirty_log(void)
+{
+    RAMBlock *block = NULL;
+    /* For memory_global_dirty_log_start below. */
+    qemu_mutex_lock_iothread();
+    qemu_mutex_lock_ramlist();
+
+    memory_global_dirty_log_sync();
+    WITH_RCU_READ_LOCK_GUARD() {
+        RAMBLOCK_FOREACH_NOT_IGNORED(block) {
+            ramblock_sync_dirty_bitmap(ram_state, block);
+            /* Discard this dirty bitmap record */
+            bitmap_zero(block->bmap, block->max_length >> TARGET_PAGE_BITS);
+        }
+        memory_global_dirty_log_start();
+    }
+    ram_state->migration_dirty_pages = 0;
+    qemu_mutex_unlock_ramlist();
+    qemu_mutex_unlock_iothread();
+}
+
 /* It is need to hold the global lock to call this helper */
 void colo_release_ram_cache(void)
 {
@@ -3040,9 +3056,7 @@ void colo_release_ram_cache(void)
             }
         }
     }
-    qemu_mutex_destroy(&ram_state->bitmap_mutex);
-    g_free(ram_state);
-    ram_state = NULL;
+    ram_state_cleanup(&ram_state);
 }
 
 /**
@@ -3356,7 +3370,7 @@ static int ram_load_precopy(QEMUFile *f)
 
     while (!ret && !(flags & RAM_SAVE_FLAG_EOS)) {
         ram_addr_t addr, total_ram_bytes;
-        void *host = NULL;
+        void *host = NULL, *host_bak = NULL;
         uint8_t ch;
 
         /*
@@ -3387,20 +3401,35 @@ static int ram_load_precopy(QEMUFile *f)
                      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.
+             * After going into COLO stage, we should not load the page
+             * into SVM's memory directly, we put them into colo_cache firstly.
+             * NOTE: We need to keep a copy of SVM's ram in colo_cache.
+             * Previously, we copied all these memory in preparing stage of COLO
+             * while we need to stop VM, which is a time-consuming process.
+             * Here we optimize it by a trick, back-up every page while in
+             * migration process while COLO is enabled, though it affects the
+             * speed of the migration, but it obviously reduce the downtime of
+             * back-up all SVM'S memory in COLO preparing stage.
              */
-            if (migration_incoming_in_colo_state()) {
-                host = colo_cache_from_block_offset(block, addr);
-            } else {
-                host = host_from_ram_block_offset(block, addr);
+            if (migration_incoming_colo_enabled()) {
+                if (migration_incoming_in_colo_state()) {
+                    /* In COLO stage, put all pages into cache temporarily */
+                    host = colo_cache_from_block_offset(block, addr);
+                } else {
+                   /*
+                    * In migration stage but before COLO stage,
+                    * Put all pages into both cache and SVM's memory.
+                    */
+                    host_bak = colo_cache_from_block_offset(block, addr);
+                }
             }
             if (!host) {
                 error_report("Illegal RAM offset " RAM_ADDR_FMT, addr);
                 ret = -EINVAL;
                 break;
             }
-
             if (!migration_incoming_in_colo_state()) {
                 ramblock_recv_bitmap_set(block, host);
             }
@@ -3514,6 +3543,9 @@ static int ram_load_precopy(QEMUFile *f)
         if (!ret) {
             ret = qemu_file_get_error(f);
         }
+        if (!ret && host_bak) {
+            memcpy(host_bak, host, TARGET_PAGE_SIZE);
+        }
     }
 
     ret |= wait_for_decompress_done();
diff --git a/migration/ram.h b/migration/ram.h
index a553d40751..5ceaff7cb4 100644
--- a/migration/ram.h
+++ b/migration/ram.h
@@ -66,5 +66,6 @@ int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *rb);
 /* ram cache */
 int colo_init_ram_cache(void);
 void colo_release_ram_cache(void);
+void colo_incoming_start_dirty_log(void);
 
 #endif