Merge remote-tracking branch 'qemu-kvm/uq/master' into staging

[qemu.git] / block-migration.c

diff --git a/block-migration.c b/block-migration.c
index b56be7908c736f96b4fb8d20c2eba2442f7d55eb..0936c7d5ea347de19d37afe4ae7d9956381bc6d7 100644 (file)
--- a/block-migration.c
+++ b/block-migration.c
@@ -15,26 +15,28 @@
 #include "block_int.h"
 #include "hw/hw.h"
 #include "qemu-queue.h"
 #include "block_int.h"
 #include "hw/hw.h"
 #include "qemu-queue.h"

+#include "qemu-timer.h"
+#include "monitor.h"

 #include "block-migration.h"
 #include "block-migration.h"

+#include "migration.h"
+#include "blockdev.h"

 #include <assert.h>
 
 #define BLOCK_SIZE (BDRV_SECTORS_PER_DIRTY_CHUNK << BDRV_SECTOR_BITS)
 
 #define BLK_MIG_FLAG_DEVICE_BLOCK       0x01
 #define BLK_MIG_FLAG_EOS                0x02
 #include <assert.h>
 
 #define BLOCK_SIZE (BDRV_SECTORS_PER_DIRTY_CHUNK << BDRV_SECTOR_BITS)
 
 #define BLK_MIG_FLAG_DEVICE_BLOCK       0x01
 #define BLK_MIG_FLAG_EOS                0x02

+#define BLK_MIG_FLAG_PROGRESS           0x04

 
 #define MAX_IS_ALLOCATED_SEARCH 65536
 
 #define MAX_IS_ALLOCATED_SEARCH 65536

-#define MAX_BLOCKS_READ 10000
-#define BLOCKS_READ_CHANGE 100
-#define INITIAL_BLOCKS_READ 100

 
 //#define DEBUG_BLK_MIGRATION
 
 #ifdef DEBUG_BLK_MIGRATION
 
 //#define DEBUG_BLK_MIGRATION
 
 #ifdef DEBUG_BLK_MIGRATION

-#define dprintf(fmt, ...) \
+#define DPRINTF(fmt, ...) \

     do { printf("blk_migration: " fmt, ## __VA_ARGS__); } while (0)
 #else
     do { printf("blk_migration: " fmt, ## __VA_ARGS__); } while (0)
 #else

-#define dprintf(fmt, ...) \
+#define DPRINTF(fmt, ...) \

     do { } while (0)
 #endif
 
     do { } while (0)
 #endif
 


@@ -43,16 +45,19 @@ typedef struct BlkMigDevState {
     int bulk_completed;
     int shared_base;
     int64_t cur_sector;
     int bulk_completed;
     int shared_base;
     int64_t cur_sector;

+    int64_t cur_dirty;

     int64_t completed_sectors;
     int64_t total_sectors;
     int64_t dirty;
     QSIMPLEQ_ENTRY(BlkMigDevState) entry;
     int64_t completed_sectors;
     int64_t total_sectors;
     int64_t dirty;
     QSIMPLEQ_ENTRY(BlkMigDevState) entry;

+    unsigned long *aio_bitmap;

 } BlkMigDevState;
 
 typedef struct BlkMigBlock {
     uint8_t *buf;
     BlkMigDevState *bmds;
     int64_t sector;
 } BlkMigDevState;
 
 typedef struct BlkMigBlock {
     uint8_t *buf;
     BlkMigDevState *bmds;
     int64_t sector;

+    int nr_sectors;

     struct iovec iov;
     QEMUIOVector qiov;
     BlockDriverAIOCB *aiocb;
     struct iovec iov;
     QEMUIOVector qiov;
     BlockDriverAIOCB *aiocb;


@@ -69,7 +74,11 @@ typedef struct BlkMigState {
     int read_done;
     int transferred;
     int64_t total_sector_sum;
     int read_done;
     int transferred;
     int64_t total_sector_sum;

-    int64_t print_completion;
+    int prev_progress;
+    int bulk_completed;
+    long double total_time;
+    long double prev_time_offset;
+    int reads;

 } BlkMigState;
 
 static BlkMigState block_mig_state;
 } BlkMigState;
 
 static BlkMigState block_mig_state;


@@ -90,20 +99,111 @@ static void blk_send(QEMUFile *f, BlkMigBlock * blk)
     qemu_put_buffer(f, blk->buf, BLOCK_SIZE);
 }
 
     qemu_put_buffer(f, blk->buf, BLOCK_SIZE);
 }
 

+int blk_mig_active(void)
+{
+    return !QSIMPLEQ_EMPTY(&block_mig_state.bmds_list);
+}
+
+uint64_t blk_mig_bytes_transferred(void)
+{
+    BlkMigDevState *bmds;
+    uint64_t sum = 0;
+
+    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+        sum += bmds->completed_sectors;
+    }
+    return sum << BDRV_SECTOR_BITS;
+}
+
+uint64_t blk_mig_bytes_remaining(void)
+{
+    return blk_mig_bytes_total() - blk_mig_bytes_transferred();
+}
+
+uint64_t blk_mig_bytes_total(void)
+{
+    BlkMigDevState *bmds;
+    uint64_t sum = 0;
+
+    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+        sum += bmds->total_sectors;
+    }
+    return sum << BDRV_SECTOR_BITS;
+}
+
+static inline long double compute_read_bwidth(void)
+{
+    assert(block_mig_state.total_time != 0);
+    return (block_mig_state.reads / block_mig_state.total_time) * BLOCK_SIZE;
+}
+
+static int bmds_aio_inflight(BlkMigDevState *bmds, int64_t sector)
+{
+    int64_t chunk = sector / (int64_t)BDRV_SECTORS_PER_DIRTY_CHUNK;
+
+    if ((sector << BDRV_SECTOR_BITS) < bdrv_getlength(bmds->bs)) {
+        return !!(bmds->aio_bitmap[chunk / (sizeof(unsigned long) * 8)] &
+            (1UL << (chunk % (sizeof(unsigned long) * 8))));
+    } else {
+        return 0;
+    }
+}
+
+static void bmds_set_aio_inflight(BlkMigDevState *bmds, int64_t sector_num,
+                             int nb_sectors, int set)
+{
+    int64_t start, end;
+    unsigned long val, idx, bit;
+
+    start = sector_num / BDRV_SECTORS_PER_DIRTY_CHUNK;
+    end = (sector_num + nb_sectors - 1) / BDRV_SECTORS_PER_DIRTY_CHUNK;
+
+    for (; start <= end; start++) {
+        idx = start / (sizeof(unsigned long) * 8);
+        bit = start % (sizeof(unsigned long) * 8);
+        val = bmds->aio_bitmap[idx];
+        if (set) {
+            val |= 1UL << bit;
+        } else {
+            val &= ~(1UL << bit);
+        }
+        bmds->aio_bitmap[idx] = val;
+    }
+}
+
+static void alloc_aio_bitmap(BlkMigDevState *bmds)
+{
+    BlockDriverState *bs = bmds->bs;
+    int64_t bitmap_size;
+
+    bitmap_size = (bdrv_getlength(bs) >> BDRV_SECTOR_BITS) +
+            BDRV_SECTORS_PER_DIRTY_CHUNK * 8 - 1;
+    bitmap_size /= BDRV_SECTORS_PER_DIRTY_CHUNK * 8;
+
+    bmds->aio_bitmap = qemu_mallocz(bitmap_size);
+}
+

 static void blk_mig_read_cb(void *opaque, int ret)
 {
 static void blk_mig_read_cb(void *opaque, int ret)
 {

+    long double curr_time = qemu_get_clock_ns(rt_clock);

     BlkMigBlock *blk = opaque;
 
     blk->ret = ret;
 
     BlkMigBlock *blk = opaque;
 
     blk->ret = ret;
 

+    block_mig_state.reads++;
+    block_mig_state.total_time += (curr_time - block_mig_state.prev_time_offset);
+    block_mig_state.prev_time_offset = curr_time;
+

     QSIMPLEQ_INSERT_TAIL(&block_mig_state.blk_list, blk, entry);
     QSIMPLEQ_INSERT_TAIL(&block_mig_state.blk_list, blk, entry);

+    bmds_set_aio_inflight(blk->bmds, blk->sector, blk->nr_sectors, 0);

 
     block_mig_state.submitted--;
     block_mig_state.read_done++;
     assert(block_mig_state.submitted >= 0);
 }
 
 
     block_mig_state.submitted--;
     block_mig_state.read_done++;
     assert(block_mig_state.submitted >= 0);
 }
 

-static int mig_save_device_bulk(QEMUFile *f, BlkMigDevState *bmds, int is_async)
+static int mig_save_device_bulk(Monitor *mon, QEMUFile *f,
+                                BlkMigDevState *bmds)

 {
     int64_t total_sectors = bmds->total_sectors;
     int64_t cur_sector = bmds->cur_sector;
 {
     int64_t total_sectors = bmds->total_sectors;
     int64_t cur_sector = bmds->cur_sector;


@@ -139,27 +239,22 @@ static int mig_save_device_bulk(QEMUFile *f, BlkMigDevState *bmds, int is_async)
     blk->buf = qemu_malloc(BLOCK_SIZE);
     blk->bmds = bmds;
     blk->sector = cur_sector;
     blk->buf = qemu_malloc(BLOCK_SIZE);
     blk->bmds = bmds;
     blk->sector = cur_sector;

+    blk->nr_sectors = nr_sectors;

 
 

-    if (is_async) {
-        blk->iov.iov_base = blk->buf;
-        blk->iov.iov_len = nr_sectors * BDRV_SECTOR_SIZE;
-        qemu_iovec_init_external(&blk->qiov, &blk->iov, 1);
+    blk->iov.iov_base = blk->buf;
+    blk->iov.iov_len = nr_sectors * BDRV_SECTOR_SIZE;
+    qemu_iovec_init_external(&blk->qiov, &blk->iov, 1);

 
 

-        blk->aiocb = bdrv_aio_readv(bs, cur_sector, &blk->qiov,
-                                    nr_sectors, blk_mig_read_cb, blk);
-        if (!blk->aiocb) {
-            goto error;
-        }
-        block_mig_state.submitted++;
-    } else {
-        if (bdrv_read(bs, cur_sector, blk->buf, nr_sectors) < 0) {
-            goto error;
-        }
-        blk_send(f, blk);
+    if (block_mig_state.submitted == 0) {
+        block_mig_state.prev_time_offset = qemu_get_clock_ns(rt_clock);
+    }

 
 

-        qemu_free(blk->buf);
-        qemu_free(blk);
+    blk->aiocb = bdrv_aio_readv(bs, cur_sector, &blk->qiov,
+                                nr_sectors, blk_mig_read_cb, blk);
+    if (!blk->aiocb) {
+        goto error;

     }
     }

+    block_mig_state.submitted++;

 
     bdrv_reset_dirty(bs, cur_sector, nr_sectors);
     bmds->cur_sector = cur_sector + nr_sectors;
 
     bdrv_reset_dirty(bs, cur_sector, nr_sectors);
     bmds->cur_sector = cur_sector + nr_sectors;


@@ -167,7 +262,7 @@ static int mig_save_device_bulk(QEMUFile *f, BlkMigDevState *bmds, int is_async)
     return (bmds->cur_sector >= total_sectors);
 
 error:
     return (bmds->cur_sector >= total_sectors);
 
 error:

-    printf("Error reading sector %" PRId64 "\n", cur_sector);
+    monitor_printf(mon, "Error reading sector %" PRId64 "\n", cur_sector);

     qemu_file_set_error(f);
     qemu_free(blk->buf);
     qemu_free(blk);
     qemu_file_set_error(f);
     qemu_free(blk->buf);
     qemu_free(blk);


@@ -183,49 +278,67 @@ static void set_dirty_tracking(int enable)
     }
 }
 
     }
 }
 

-static void init_blk_migration(QEMUFile *f)
+static void init_blk_migration_it(void *opaque, BlockDriverState *bs)

 {
 {

+    Monitor *mon = opaque;

     BlkMigDevState *bmds;
     BlkMigDevState *bmds;

-    BlockDriverState *bs;
+    int64_t sectors;
+
+    if (!bdrv_is_read_only(bs)) {
+        sectors = bdrv_getlength(bs) >> BDRV_SECTOR_BITS;
+        if (sectors <= 0) {
+            return;
+        }
+
+        bmds = qemu_mallocz(sizeof(BlkMigDevState));
+        bmds->bs = bs;
+        bmds->bulk_completed = 0;
+        bmds->total_sectors = sectors;
+        bmds->completed_sectors = 0;
+        bmds->shared_base = block_mig_state.shared_base;
+        alloc_aio_bitmap(bmds);
+        drive_get_ref(drive_get_by_blockdev(bs));
+        bdrv_set_in_use(bs, 1);
+
+        block_mig_state.total_sector_sum += sectors;
+
+        if (bmds->shared_base) {
+            monitor_printf(mon, "Start migration for %s with shared base "
+                                "image\n",
+                           bs->device_name);
+        } else {
+            monitor_printf(mon, "Start full migration for %s\n",
+                           bs->device_name);
+        }

 
 

+        QSIMPLEQ_INSERT_TAIL(&block_mig_state.bmds_list, bmds, entry);
+    }
+}
+
+static void init_blk_migration(Monitor *mon, QEMUFile *f)
+{

     block_mig_state.submitted = 0;
     block_mig_state.read_done = 0;
     block_mig_state.transferred = 0;
     block_mig_state.total_sector_sum = 0;
     block_mig_state.submitted = 0;
     block_mig_state.read_done = 0;
     block_mig_state.transferred = 0;
     block_mig_state.total_sector_sum = 0;

-    block_mig_state.print_completion = 0;
-
-    for (bs = bdrv_first; bs != NULL; bs = bs->next) {
-        if (bs->type == BDRV_TYPE_HD) {
-            bmds = qemu_mallocz(sizeof(BlkMigDevState));
-            bmds->bs = bs;
-            bmds->bulk_completed = 0;
-            bmds->total_sectors = bdrv_getlength(bs) >> BDRV_SECTOR_BITS;
-            bmds->completed_sectors = 0;
-            bmds->shared_base = block_mig_state.shared_base;
-
-            block_mig_state.total_sector_sum += bmds->total_sectors;
-
-            if (bmds->shared_base) {
-                printf("Start migration for %s with shared base image\n",
-                       bs->device_name);
-            } else {
-                printf("Start full migration for %s\n", bs->device_name);
-            }
+    block_mig_state.prev_progress = -1;
+    block_mig_state.bulk_completed = 0;
+    block_mig_state.total_time = 0;
+    block_mig_state.reads = 0;

 
 

-            QSIMPLEQ_INSERT_TAIL(&block_mig_state.bmds_list, bmds, entry);
-        }
-    }
+    bdrv_iterate(init_blk_migration_it, mon);

 }
 
 }
 

-static int blk_mig_save_bulked_block(QEMUFile *f, int is_async)
+static int blk_mig_save_bulked_block(Monitor *mon, QEMUFile *f)

 {
     int64_t completed_sector_sum = 0;
     BlkMigDevState *bmds;
 {
     int64_t completed_sector_sum = 0;
     BlkMigDevState *bmds;

+    int progress;

     int ret = 0;
 
     QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
         if (bmds->bulk_completed == 0) {
     int ret = 0;
 
     QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
         if (bmds->bulk_completed == 0) {

-            if (mig_save_device_bulk(f, bmds, is_async) == 1) {
+            if (mig_save_device_bulk(mon, f, bmds) == 1) {

                 /* completed bulk section for this device */
                 bmds->bulk_completed = 1;
             }
                 /* completed bulk section for this device */
                 bmds->bulk_completed = 1;
             }


@@ -237,56 +350,121 @@ static int blk_mig_save_bulked_block(QEMUFile *f, int is_async)
         }
     }
 
         }
     }
 

-    if (completed_sector_sum >= block_mig_state.print_completion) {
-        printf("Completed %" PRId64 " %%\r",
-               completed_sector_sum * 100 / block_mig_state.total_sector_sum);
-        fflush(stdout);
-        block_mig_state.print_completion +=
-            (BDRV_SECTORS_PER_DIRTY_CHUNK * 10000);
+    if (block_mig_state.total_sector_sum != 0) {
+        progress = completed_sector_sum * 100 /
+                   block_mig_state.total_sector_sum;
+    } else {
+        progress = 100;
+    }
+    if (progress != block_mig_state.prev_progress) {
+        block_mig_state.prev_progress = progress;
+        qemu_put_be64(f, (progress << BDRV_SECTOR_BITS)
+                         | BLK_MIG_FLAG_PROGRESS);
+        monitor_printf(mon, "Completed %d %%\r", progress);
+        monitor_flush(mon);

     }
 
     return ret;
 }
 
     }
 
     return ret;
 }
 

-#define MAX_NUM_BLOCKS 4
-
-static void blk_mig_save_dirty_blocks(QEMUFile *f)
+static void blk_mig_reset_dirty_cursor(void)

 {
     BlkMigDevState *bmds;
 {
     BlkMigDevState *bmds;

-    BlkMigBlock blk;
+
+    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+        bmds->cur_dirty = 0;
+    }
+}
+
+static int mig_save_device_dirty(Monitor *mon, QEMUFile *f,
+                                 BlkMigDevState *bmds, int is_async)
+{
+    BlkMigBlock *blk;
+    int64_t total_sectors = bmds->total_sectors;

     int64_t sector;
     int64_t sector;

+    int nr_sectors;

 
 

-    blk.buf = qemu_malloc(BLOCK_SIZE);
+    for (sector = bmds->cur_dirty; sector < bmds->total_sectors;) {
+        if (bmds_aio_inflight(bmds, sector)) {
+            qemu_aio_flush();
+        }
+        if (bdrv_get_dirty(bmds->bs, sector)) {

 
 

-    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
-        for (sector = 0; sector < bmds->cur_sector;) {
-            if (bdrv_get_dirty(bmds->bs, sector)) {
-                if (bdrv_read(bmds->bs, sector, blk.buf,
-                              BDRV_SECTORS_PER_DIRTY_CHUNK) < 0) {
-                    printf("Error reading sector %" PRId64 "\n", sector);
-                    qemu_file_set_error(f);
-                    qemu_free(blk.buf);
-                    return;
+            if (total_sectors - sector < BDRV_SECTORS_PER_DIRTY_CHUNK) {
+                nr_sectors = total_sectors - sector;
+            } else {
+                nr_sectors = BDRV_SECTORS_PER_DIRTY_CHUNK;
+            }
+            blk = qemu_malloc(sizeof(BlkMigBlock));
+            blk->buf = qemu_malloc(BLOCK_SIZE);
+            blk->bmds = bmds;
+            blk->sector = sector;
+            blk->nr_sectors = nr_sectors;
+
+            if (is_async) {
+                blk->iov.iov_base = blk->buf;
+                blk->iov.iov_len = nr_sectors * BDRV_SECTOR_SIZE;
+                qemu_iovec_init_external(&blk->qiov, &blk->iov, 1);
+
+                if (block_mig_state.submitted == 0) {
+                    block_mig_state.prev_time_offset = qemu_get_clock_ns(rt_clock);
+                }
+
+                blk->aiocb = bdrv_aio_readv(bmds->bs, sector, &blk->qiov,
+                                            nr_sectors, blk_mig_read_cb, blk);
+                if (!blk->aiocb) {
+                    goto error;
+                }
+                block_mig_state.submitted++;
+                bmds_set_aio_inflight(bmds, sector, nr_sectors, 1);
+            } else {
+                if (bdrv_read(bmds->bs, sector, blk->buf,
+                              nr_sectors) < 0) {
+                    goto error;

                 }
                 }

-                blk.bmds = bmds;
-                blk.sector = sector;
-                blk_send(f, &blk);
+                blk_send(f, blk);

 
 

-                bdrv_reset_dirty(bmds->bs, sector,
-                                 BDRV_SECTORS_PER_DIRTY_CHUNK);
+                qemu_free(blk->buf);
+                qemu_free(blk);

             }
             }

-            sector += BDRV_SECTORS_PER_DIRTY_CHUNK;
+
+            bdrv_reset_dirty(bmds->bs, sector, nr_sectors);
+            break;
+        }
+        sector += BDRV_SECTORS_PER_DIRTY_CHUNK;
+        bmds->cur_dirty = sector;
+    }
+
+    return (bmds->cur_dirty >= bmds->total_sectors);
+
+error:
+    monitor_printf(mon, "Error reading sector %" PRId64 "\n", sector);
+    qemu_file_set_error(f);
+    qemu_free(blk->buf);
+    qemu_free(blk);
+    return 0;
+}
+
+static int blk_mig_save_dirty_block(Monitor *mon, QEMUFile *f, int is_async)
+{
+    BlkMigDevState *bmds;
+    int ret = 0;
+
+    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+        if (mig_save_device_dirty(mon, f, bmds, is_async) == 0) {
+            ret = 1;
+            break;

         }
     }
 
         }
     }
 

-    qemu_free(blk.buf);
+    return ret;

 }
 
 static void flush_blks(QEMUFile* f)
 {
     BlkMigBlock *blk;
 
 }
 
 static void flush_blks(QEMUFile* f)
 {
     BlkMigBlock *blk;
 

-    dprintf("%s Enter submitted %d read_done %d transferred %d\n",
+    DPRINTF("%s Enter submitted %d read_done %d transferred %d\n",

             __FUNCTION__, block_mig_state.submitted, block_mig_state.read_done,
             block_mig_state.transferred);
 
             __FUNCTION__, block_mig_state.submitted, block_mig_state.read_done,
             block_mig_state.transferred);
 


@@ -309,35 +487,61 @@ static void flush_blks(QEMUFile* f)
         assert(block_mig_state.read_done >= 0);
     }
 
         assert(block_mig_state.read_done >= 0);
     }
 

-    dprintf("%s Exit submitted %d read_done %d transferred %d\n", __FUNCTION__,
+    DPRINTF("%s Exit submitted %d read_done %d transferred %d\n", __FUNCTION__,

             block_mig_state.submitted, block_mig_state.read_done,
             block_mig_state.transferred);
 }
 
             block_mig_state.submitted, block_mig_state.read_done,
             block_mig_state.transferred);
 }
 

-static int is_stage2_completed(void)
+static int64_t get_remaining_dirty(void)

 {
     BlkMigDevState *bmds;
 {
     BlkMigDevState *bmds;

+    int64_t dirty = 0;

 
 

-    if (block_mig_state.submitted > 0) {
-        return 0;
+    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+        dirty += bdrv_get_dirty_count(bmds->bs);

     }
 
     }
 

-    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
-        if (bmds->bulk_completed == 0) {
-            return 0;
+    return dirty * BLOCK_SIZE;
+}
+
+static int is_stage2_completed(void)
+{
+    int64_t remaining_dirty;
+    long double bwidth;
+
+    if (block_mig_state.bulk_completed == 1) {
+
+        remaining_dirty = get_remaining_dirty();
+        if (remaining_dirty == 0) {
+            return 1;
+        }
+
+        bwidth = compute_read_bwidth();
+
+        if ((remaining_dirty / bwidth) <=
+            migrate_max_downtime()) {
+            /* finish stage2 because we think that we can finish remaing work
+               below max_downtime */
+
+            return 1;

         }
     }
 
         }
     }
 

-    return 1;
+    return 0;

 }
 
 }
 

-static void blk_mig_cleanup(void)
+static void blk_mig_cleanup(Monitor *mon)

 {
     BlkMigDevState *bmds;
     BlkMigBlock *blk;
 
 {
     BlkMigDevState *bmds;
     BlkMigBlock *blk;
 

+    set_dirty_tracking(0);
+

     while ((bmds = QSIMPLEQ_FIRST(&block_mig_state.bmds_list)) != NULL) {
         QSIMPLEQ_REMOVE_HEAD(&block_mig_state.bmds_list, entry);
     while ((bmds = QSIMPLEQ_FIRST(&block_mig_state.bmds_list)) != NULL) {
         QSIMPLEQ_REMOVE_HEAD(&block_mig_state.bmds_list, entry);

+        bdrv_set_in_use(bmds->bs, 0);
+        drive_put_ref(drive_get_by_blockdev(bmds->bs));
+        qemu_free(bmds->aio_bitmap);

         qemu_free(bmds);
     }
 
         qemu_free(bmds);
     }
 


@@ -347,18 +551,16 @@ static void blk_mig_cleanup(void)
         qemu_free(blk);
     }
 
         qemu_free(blk);
     }
 

-    set_dirty_tracking(0);
-
-    printf("\n");
+    monitor_printf(mon, "\n");

 }
 
 static int block_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
 {
 }
 
 static int block_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
 {

-    dprintf("Enter save live stage %d submitted %d transferred %d\n",
+    DPRINTF("Enter save live stage %d submitted %d transferred %d\n",

             stage, block_mig_state.submitted, block_mig_state.transferred);
 
     if (stage < 0) {
             stage, block_mig_state.submitted, block_mig_state.transferred);
 
     if (stage < 0) {

-        blk_mig_cleanup();
+        blk_mig_cleanup(mon);

         return 0;
     }
 
         return 0;
     }
 


@@ -369,7 +571,7 @@ static int block_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
     }
 
     if (stage == 1) {
     }
 
     if (stage == 1) {

-        init_blk_migration(f);
+        init_blk_migration(mon, f);

 
         /* start track dirty blocks */
         set_dirty_tracking(1);
 
         /* start track dirty blocks */
         set_dirty_tracking(1);


@@ -378,40 +580,55 @@ static int block_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
     flush_blks(f);
 
     if (qemu_file_has_error(f)) {
     flush_blks(f);
 
     if (qemu_file_has_error(f)) {

-        blk_mig_cleanup();
+        blk_mig_cleanup(mon);

         return 0;
     }
 
         return 0;
     }
 

-    /* control the rate of transfer */
-    while ((block_mig_state.submitted +
-            block_mig_state.read_done) * BLOCK_SIZE <
-           qemu_file_get_rate_limit(f)) {
-        if (blk_mig_save_bulked_block(f, 1) == 0) {
-            /* no more bulk blocks for now */
-            break;
+    blk_mig_reset_dirty_cursor();
+
+    if (stage == 2) {
+        /* control the rate of transfer */
+        while ((block_mig_state.submitted +
+                block_mig_state.read_done) * BLOCK_SIZE <
+               qemu_file_get_rate_limit(f)) {
+            if (block_mig_state.bulk_completed == 0) {
+                /* first finish the bulk phase */
+                if (blk_mig_save_bulked_block(mon, f) == 0) {
+                    /* finished saving bulk on all devices */
+                    block_mig_state.bulk_completed = 1;
+                }
+            } else {
+                if (blk_mig_save_dirty_block(mon, f, 1) == 0) {
+                    /* no more dirty blocks */
+                    break;
+                }
+            }

         }
         }

-    }

 
 

-    flush_blks(f);
+        flush_blks(f);

 
 

-    if (qemu_file_has_error(f)) {
-        blk_mig_cleanup();
-        return 0;
+        if (qemu_file_has_error(f)) {
+            blk_mig_cleanup(mon);
+            return 0;
+        }

     }
 
     if (stage == 3) {
     }
 
     if (stage == 3) {

-        while (blk_mig_save_bulked_block(f, 0) != 0) {
-            /* empty */
-        }
+        /* we know for sure that save bulk is completed and
+           all async read completed */
+        assert(block_mig_state.submitted == 0);

 
 

-        blk_mig_save_dirty_blocks(f);
-        blk_mig_cleanup();
+        while (blk_mig_save_dirty_block(mon, f, 0) != 0);
+        blk_mig_cleanup(mon);
+
+        /* report completion */
+        qemu_put_be64(f, (100 << BDRV_SECTOR_BITS) | BLK_MIG_FLAG_PROGRESS);

 
         if (qemu_file_has_error(f)) {
             return 0;
         }
 
 
         if (qemu_file_has_error(f)) {
             return 0;
         }
 

-        printf("Block migration completed\n");
+        monitor_printf(mon, "Block migration completed\n");

     }
 
     qemu_put_be64(f, BLK_MIG_FLAG_EOS);
     }
 
     qemu_put_be64(f, BLK_MIG_FLAG_EOS);


@@ -421,11 +638,14 @@ static int block_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
 
 static int block_load(QEMUFile *f, void *opaque, int version_id)
 {
 
 static int block_load(QEMUFile *f, void *opaque, int version_id)
 {

+    static int banner_printed;

     int len, flags;
     char device_name[256];
     int64_t addr;
     int len, flags;
     char device_name[256];
     int64_t addr;

-    BlockDriverState *bs;
+    BlockDriverState *bs, *bs_prev = NULL;

     uint8_t *buf;
     uint8_t *buf;

+    int64_t total_sectors = 0;
+    int nr_sectors;

 
     do {
         addr = qemu_get_be64(f);
 
     do {
         addr = qemu_get_be64(f);


@@ -434,6 +654,7 @@ static int block_load(QEMUFile *f, void *opaque, int version_id)
         addr >>= BDRV_SECTOR_BITS;
 
         if (flags & BLK_MIG_FLAG_DEVICE_BLOCK) {
         addr >>= BDRV_SECTOR_BITS;
 
         if (flags & BLK_MIG_FLAG_DEVICE_BLOCK) {

+            int ret;

             /* get device name */
             len = qemu_get_byte(f);
             qemu_get_buffer(f, (uint8_t *)device_name, len);
             /* get device name */
             len = qemu_get_byte(f);
             qemu_get_buffer(f, (uint8_t *)device_name, len);


@@ -446,12 +667,39 @@ static int block_load(QEMUFile *f, void *opaque, int version_id)
                 return -EINVAL;
             }
 
                 return -EINVAL;
             }
 

+            if (bs != bs_prev) {
+                bs_prev = bs;
+                total_sectors = bdrv_getlength(bs) >> BDRV_SECTOR_BITS;
+                if (total_sectors <= 0) {
+                    error_report("Error getting length of block device %s",
+                                 device_name);
+                    return -EINVAL;
+                }
+            }
+
+            if (total_sectors - addr < BDRV_SECTORS_PER_DIRTY_CHUNK) {
+                nr_sectors = total_sectors - addr;
+            } else {
+                nr_sectors = BDRV_SECTORS_PER_DIRTY_CHUNK;
+            }
+

             buf = qemu_malloc(BLOCK_SIZE);
 
             qemu_get_buffer(f, buf, BLOCK_SIZE);
             buf = qemu_malloc(BLOCK_SIZE);
 
             qemu_get_buffer(f, buf, BLOCK_SIZE);

-            bdrv_write(bs, addr, buf, BDRV_SECTORS_PER_DIRTY_CHUNK);
+            ret = bdrv_write(bs, addr, buf, nr_sectors);

 
             qemu_free(buf);
 
             qemu_free(buf);

+            if (ret < 0) {
+                return ret;
+            }
+        } else if (flags & BLK_MIG_FLAG_PROGRESS) {
+            if (!banner_printed) {
+                printf("Receiving block device images\n");
+                banner_printed = 1;
+            }
+            printf("Completed %d %%%c", (int)addr,
+                   (addr == 100) ? '\n' : '\r');
+            fflush(stdout);

         } else if (!(flags & BLK_MIG_FLAG_EOS)) {
             fprintf(stderr, "Unknown flags\n");
             return -EINVAL;
         } else if (!(flags & BLK_MIG_FLAG_EOS)) {
             fprintf(stderr, "Unknown flags\n");
             return -EINVAL;


@@ -478,6 +726,6 @@ void blk_mig_init(void)
     QSIMPLEQ_INIT(&block_mig_state.bmds_list);
     QSIMPLEQ_INIT(&block_mig_state.blk_list);
 
     QSIMPLEQ_INIT(&block_mig_state.bmds_list);
     QSIMPLEQ_INIT(&block_mig_state.blk_list);
 

-    register_savevm_live("block", 0, 1, block_set_params, block_save_live,
-                         NULL, block_load, &block_mig_state);
+    register_savevm_live(NULL, "block", 0, 1, block_set_params,
+                         block_save_live, NULL, block_load, &block_mig_state);

 }
 }