* This work is licensed under the terms of the GNU GPL, version 2. See
* the COPYING file in the top-level directory.
*
+ * Contributions after 2012-01-13 are licensed under the terms of the
+ * GNU GPL, version 2 or (at your option) any later version.
*/
#include "qemu-common.h"
-#include "block_int.h"
+#include "block/block_int.h"
#include "hw/hw.h"
-#include "qemu-queue.h"
-#include "monitor.h"
-#include "block-migration.h"
+#include "qemu/queue.h"
+#include "qemu/timer.h"
+#include "migration/block.h"
+#include "migration/migration.h"
+#include "sysemu/blockdev.h"
#include <assert.h>
-#define BLOCK_SIZE (BDRV_SECTORS_PER_DIRTY_CHUNK << BDRV_SECTOR_BITS)
+#define BLOCK_SIZE (1 << 20)
+#define BDRV_SECTORS_PER_DIRTY_CHUNK (BLOCK_SIZE >> 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_BLOCKS_READ 10000
-#define BLOCKS_READ_CHANGE 100
-#define INITIAL_BLOCKS_READ 100
//#define DEBUG_BLK_MIGRATION
#endif
typedef struct BlkMigDevState {
+ /* Written during setup phase. Can be read without a lock. */
BlockDriverState *bs;
- int bulk_completed;
int shared_base;
- int64_t cur_sector;
- int64_t completed_sectors;
int64_t total_sectors;
- int64_t dirty;
QSIMPLEQ_ENTRY(BlkMigDevState) entry;
+
+ /* Only used by migration thread. Does not need a lock. */
+ int bulk_completed;
+ int64_t cur_sector;
+ int64_t cur_dirty;
+
+ /* Protected by block migration lock. */
+ unsigned long *aio_bitmap;
+ int64_t completed_sectors;
} BlkMigDevState;
typedef struct BlkMigBlock {
+ /* Only used by migration thread. */
uint8_t *buf;
BlkMigDevState *bmds;
int64_t sector;
+ int nr_sectors;
struct iovec iov;
QEMUIOVector qiov;
BlockDriverAIOCB *aiocb;
+
+ /* Protected by block migration lock. */
int ret;
QSIMPLEQ_ENTRY(BlkMigBlock) entry;
} BlkMigBlock;
typedef struct BlkMigState {
+ /* Written during setup phase. Can be read without a lock. */
int blk_enable;
int shared_base;
QSIMPLEQ_HEAD(bmds_list, BlkMigDevState) bmds_list;
+ int64_t total_sector_sum;
+
+ /* Protected by lock. */
QSIMPLEQ_HEAD(blk_list, BlkMigBlock) blk_list;
int submitted;
int read_done;
+
+ /* Only used by migration thread. Does not need a lock. */
int transferred;
- int64_t total_sector_sum;
int prev_progress;
+ int bulk_completed;
+
+ /* Lock must be taken _inside_ the iothread lock. */
+ QemuMutex lock;
} BlkMigState;
static BlkMigState block_mig_state;
+static void blk_mig_lock(void)
+{
+ qemu_mutex_lock(&block_mig_state.lock);
+}
+
+static void blk_mig_unlock(void)
+{
+ qemu_mutex_unlock(&block_mig_state.lock);
+}
+
+/* Must run outside of the iothread lock during the bulk phase,
+ * or the VM will stall.
+ */
+
static void blk_send(QEMUFile *f, BlkMigBlock * blk)
{
int len;
BlkMigDevState *bmds;
uint64_t sum = 0;
+ blk_mig_lock();
QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
sum += bmds->completed_sectors;
}
+ blk_mig_unlock();
return sum << BDRV_SECTOR_BITS;
}
return sum << BDRV_SECTOR_BITS;
}
+
+/* Called with migration lock held. */
+
+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;
+ }
+}
+
+/* Called with migration lock held. */
+
+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 = g_malloc0(bitmap_size);
+}
+
+/* Never hold migration lock when yielding to the main loop! */
+
static void blk_mig_read_cb(void *opaque, int ret)
{
BlkMigBlock *blk = opaque;
+ blk_mig_lock();
blk->ret = ret;
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);
+ blk_mig_unlock();
}
-static int mig_save_device_bulk(Monitor *mon, QEMUFile *f,
- BlkMigDevState *bmds, int is_async)
+/* Called with no lock taken. */
+
+static int mig_save_device_bulk(QEMUFile *f, BlkMigDevState *bmds)
{
int64_t total_sectors = bmds->total_sectors;
int64_t cur_sector = bmds->cur_sector;
int nr_sectors;
if (bmds->shared_base) {
+ qemu_mutex_lock_iothread();
while (cur_sector < total_sectors &&
!bdrv_is_allocated(bs, cur_sector, MAX_IS_ALLOCATED_SEARCH,
&nr_sectors)) {
cur_sector += nr_sectors;
}
+ qemu_mutex_unlock_iothread();
}
if (cur_sector >= total_sectors) {
nr_sectors = total_sectors - cur_sector;
}
- blk = qemu_malloc(sizeof(BlkMigBlock));
- blk->buf = qemu_malloc(BLOCK_SIZE);
+ blk = g_malloc(sizeof(BlkMigBlock));
+ blk->buf = g_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);
+ blk_mig_lock();
+ block_mig_state.submitted++;
+ blk_mig_unlock();
- qemu_free(blk->buf);
- qemu_free(blk);
- }
+ qemu_mutex_lock_iothread();
+ blk->aiocb = bdrv_aio_readv(bs, cur_sector, &blk->qiov,
+ nr_sectors, blk_mig_read_cb, blk);
bdrv_reset_dirty(bs, cur_sector, nr_sectors);
- bmds->cur_sector = cur_sector + nr_sectors;
+ qemu_mutex_unlock_iothread();
+ bmds->cur_sector = cur_sector + nr_sectors;
return (bmds->cur_sector >= total_sectors);
-
-error:
- monitor_printf(mon, "Error reading sector %" PRId64 "\n", cur_sector);
- qemu_file_set_error(f);
- qemu_free(blk->buf);
- qemu_free(blk);
- return 0;
}
+/* Called with iothread lock taken. */
+
static void set_dirty_tracking(int enable)
{
BlkMigDevState *bmds;
QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
- bdrv_set_dirty_tracking(bmds->bs, enable);
+ bdrv_set_dirty_tracking(bmds->bs, enable ? BLOCK_SIZE : 0);
}
}
-static void init_blk_migration(Monitor *mon, QEMUFile *f)
+static void init_blk_migration_it(void *opaque, BlockDriverState *bs)
{
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 = g_malloc0(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) {
+ DPRINTF("Start migration for %s with shared base image\n",
+ bs->device_name);
+ } else {
+ DPRINTF("Start full migration for %s\n", bs->device_name);
+ }
+
+ QSIMPLEQ_INSERT_TAIL(&block_mig_state.bmds_list, bmds, entry);
+ }
+}
+
+static void init_blk_migration(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.prev_progress = -1;
+ block_mig_state.bulk_completed = 0;
- for (bs = bdrv_first; bs != NULL; bs = bs->next) {
- if (bs->type == BDRV_TYPE_HD) {
- sectors = bdrv_getlength(bs) >> BDRV_SECTOR_BITS;
- if (sectors == 0) {
- continue;
- }
-
- 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;
-
- 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);
- }
- }
+ bdrv_iterate(init_blk_migration_it, NULL);
}
-static int blk_mig_save_bulked_block(Monitor *mon, QEMUFile *f, int is_async)
+/* Called with no lock taken. */
+
+static int blk_mig_save_bulked_block(QEMUFile *f)
{
int64_t completed_sector_sum = 0;
BlkMigDevState *bmds;
QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
if (bmds->bulk_completed == 0) {
- if (mig_save_device_bulk(mon, f, bmds, is_async) == 1) {
+ if (mig_save_device_bulk(f, bmds) == 1) {
/* completed bulk section for this device */
bmds->bulk_completed = 1;
}
}
}
- progress = completed_sector_sum * 100 / block_mig_state.total_sector_sum;
+ 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);
+ DPRINTF("Completed %d %%\r", progress);
}
return ret;
}
-#define MAX_NUM_BLOCKS 4
-
-static void blk_mig_save_dirty_blocks(Monitor *mon, QEMUFile *f)
+static void blk_mig_reset_dirty_cursor(void)
{
BlkMigDevState *bmds;
- BlkMigBlock blk;
+
+ QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+ bmds->cur_dirty = 0;
+ }
+}
+
+/* Called with iothread lock taken. */
+
+static int mig_save_device_dirty(QEMUFile *f, BlkMigDevState *bmds,
+ int is_async)
+{
+ BlkMigBlock *blk;
+ int64_t total_sectors = bmds->total_sectors;
int64_t sector;
+ int nr_sectors;
+ int ret = -EIO;
- blk.buf = qemu_malloc(BLOCK_SIZE);
+ for (sector = bmds->cur_dirty; sector < bmds->total_sectors;) {
+ blk_mig_lock();
+ if (bmds_aio_inflight(bmds, sector)) {
+ blk_mig_unlock();
+ bdrv_drain_all();
+ } else {
+ blk_mig_unlock();
+ }
+ 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) {
- monitor_printf(mon, "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 = g_malloc(sizeof(BlkMigBlock));
+ blk->buf = g_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);
+
+ blk->aiocb = bdrv_aio_readv(bmds->bs, sector, &blk->qiov,
+ nr_sectors, blk_mig_read_cb, blk);
+
+ blk_mig_lock();
+ block_mig_state.submitted++;
+ bmds_set_aio_inflight(bmds, sector, nr_sectors, 1);
+ blk_mig_unlock();
+ } else {
+ ret = bdrv_read(bmds->bs, sector, blk->buf, nr_sectors);
+ if (ret < 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);
+ g_free(blk->buf);
+ g_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:
+ DPRINTF("Error reading sector %" PRId64 "\n", sector);
+ g_free(blk->buf);
+ g_free(blk);
+ return ret;
+}
+
+/* Called with iothread lock taken.
+ *
+ * return value:
+ * 0: too much data for max_downtime
+ * 1: few enough data for max_downtime
+*/
+static int blk_mig_save_dirty_block(QEMUFile *f, int is_async)
+{
+ BlkMigDevState *bmds;
+ int ret = 1;
+
+ QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+ ret = mig_save_device_dirty(f, bmds, is_async);
+ if (ret <= 0) {
+ break;
}
}
- qemu_free(blk.buf);
+ return ret;
}
-static void flush_blks(QEMUFile* f)
+/* Called with no locks taken. */
+
+static int flush_blks(QEMUFile *f)
{
BlkMigBlock *blk;
+ int ret = 0;
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);
+ blk_mig_lock();
while ((blk = QSIMPLEQ_FIRST(&block_mig_state.blk_list)) != NULL) {
if (qemu_file_rate_limit(f)) {
break;
}
if (blk->ret < 0) {
- qemu_file_set_error(f);
+ ret = blk->ret;
break;
}
- blk_send(f, blk);
QSIMPLEQ_REMOVE_HEAD(&block_mig_state.blk_list, entry);
- qemu_free(blk->buf);
- qemu_free(blk);
+ blk_mig_unlock();
+ blk_send(f, blk);
+ blk_mig_lock();
+
+ g_free(blk->buf);
+ g_free(blk);
block_mig_state.read_done--;
block_mig_state.transferred++;
assert(block_mig_state.read_done >= 0);
}
+ blk_mig_unlock();
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);
+ return ret;
}
-static int is_stage2_completed(void)
+/* Called with iothread lock taken. */
+
+static int64_t get_remaining_dirty(void)
{
BlkMigDevState *bmds;
-
- if (block_mig_state.submitted > 0) {
- return 0;
- }
+ int64_t dirty = 0;
QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
- if (bmds->bulk_completed == 0) {
- return 0;
- }
+ dirty += bdrv_get_dirty_count(bmds->bs);
}
- return 1;
+ return dirty << BDRV_SECTOR_BITS;
}
-static void blk_mig_cleanup(Monitor *mon)
+/* Called with iothread lock taken. */
+
+static void blk_mig_cleanup(void)
{
BlkMigDevState *bmds;
BlkMigBlock *blk;
+ bdrv_drain_all();
+
+ set_dirty_tracking(0);
+
+ blk_mig_lock();
while ((bmds = QSIMPLEQ_FIRST(&block_mig_state.bmds_list)) != NULL) {
QSIMPLEQ_REMOVE_HEAD(&block_mig_state.bmds_list, entry);
- qemu_free(bmds);
+ bdrv_set_in_use(bmds->bs, 0);
+ drive_put_ref(drive_get_by_blockdev(bmds->bs));
+ g_free(bmds->aio_bitmap);
+ g_free(bmds);
}
while ((blk = QSIMPLEQ_FIRST(&block_mig_state.blk_list)) != NULL) {
QSIMPLEQ_REMOVE_HEAD(&block_mig_state.blk_list, entry);
- qemu_free(blk->buf);
- qemu_free(blk);
+ g_free(blk->buf);
+ g_free(blk);
}
+ blk_mig_unlock();
+}
- set_dirty_tracking(0);
-
- monitor_printf(mon, "\n");
+static void block_migration_cancel(void *opaque)
+{
+ blk_mig_cleanup();
}
-static int block_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
+static int block_save_setup(QEMUFile *f, void *opaque)
{
- DPRINTF("Enter save live stage %d submitted %d transferred %d\n",
- stage, block_mig_state.submitted, block_mig_state.transferred);
+ int ret;
- if (stage < 0) {
- blk_mig_cleanup(mon);
- return 0;
- }
+ DPRINTF("Enter save live setup submitted %d transferred %d\n",
+ block_mig_state.submitted, block_mig_state.transferred);
- if (block_mig_state.blk_enable != 1) {
- /* no need to migrate storage */
- qemu_put_be64(f, BLK_MIG_FLAG_EOS);
- return 1;
- }
+ qemu_mutex_lock_iothread();
+ init_blk_migration(f);
- if (stage == 1) {
- init_blk_migration(mon, f);
+ /* start track dirty blocks */
+ set_dirty_tracking(1);
+ qemu_mutex_unlock_iothread();
- /* start track dirty blocks */
- set_dirty_tracking(1);
- }
+ ret = flush_blks(f);
+ blk_mig_reset_dirty_cursor();
+ qemu_put_be64(f, BLK_MIG_FLAG_EOS);
- flush_blks(f);
+ return ret;
+}
- if (qemu_file_has_error(f)) {
- blk_mig_cleanup(mon);
- return 0;
+static int block_save_iterate(QEMUFile *f, void *opaque)
+{
+ int ret;
+ int64_t last_ftell = qemu_ftell(f);
+
+ DPRINTF("Enter save live iterate submitted %d transferred %d\n",
+ block_mig_state.submitted, block_mig_state.transferred);
+
+ ret = flush_blks(f);
+ if (ret) {
+ return ret;
}
+ blk_mig_reset_dirty_cursor();
+
/* control the rate of transfer */
+ blk_mig_lock();
while ((block_mig_state.submitted +
block_mig_state.read_done) * BLOCK_SIZE <
qemu_file_get_rate_limit(f)) {
- if (blk_mig_save_bulked_block(mon, f, 1) == 0) {
- /* no more bulk blocks for now */
+ blk_mig_unlock();
+ if (block_mig_state.bulk_completed == 0) {
+ /* first finish the bulk phase */
+ if (blk_mig_save_bulked_block(f) == 0) {
+ /* finished saving bulk on all devices */
+ block_mig_state.bulk_completed = 1;
+ }
+ ret = 0;
+ } else {
+ /* Always called with iothread lock taken for
+ * simplicity, block_save_complete also calls it.
+ */
+ qemu_mutex_lock_iothread();
+ ret = blk_mig_save_dirty_block(f, 1);
+ qemu_mutex_unlock_iothread();
+ }
+ if (ret < 0) {
+ return ret;
+ }
+ blk_mig_lock();
+ if (ret != 0) {
+ /* no more dirty blocks */
break;
}
}
+ blk_mig_unlock();
- flush_blks(f);
-
- if (qemu_file_has_error(f)) {
- blk_mig_cleanup(mon);
- return 0;
+ ret = flush_blks(f);
+ if (ret) {
+ return ret;
}
- if (stage == 3) {
- while (blk_mig_save_bulked_block(mon, f, 0) != 0) {
- /* empty */
- }
+ qemu_put_be64(f, BLK_MIG_FLAG_EOS);
+ return qemu_ftell(f) - last_ftell;
+}
- blk_mig_save_dirty_blocks(mon, f);
- blk_mig_cleanup(mon);
+/* Called with iothread lock taken. */
- /* report completion */
- qemu_put_be64(f, (100 << BDRV_SECTOR_BITS) | BLK_MIG_FLAG_PROGRESS);
+static int block_save_complete(QEMUFile *f, void *opaque)
+{
+ int ret;
- if (qemu_file_has_error(f)) {
- return 0;
- }
+ DPRINTF("Enter save live complete submitted %d transferred %d\n",
+ block_mig_state.submitted, block_mig_state.transferred);
- monitor_printf(mon, "Block migration completed\n");
+ ret = flush_blks(f);
+ if (ret) {
+ return ret;
}
+ blk_mig_reset_dirty_cursor();
+
+ /* we know for sure that save bulk is completed and
+ all async read completed */
+ blk_mig_lock();
+ assert(block_mig_state.submitted == 0);
+ blk_mig_unlock();
+
+ do {
+ ret = blk_mig_save_dirty_block(f, 0);
+ if (ret < 0) {
+ return ret;
+ }
+ } while (ret == 0);
+
+ /* report completion */
+ qemu_put_be64(f, (100 << BDRV_SECTOR_BITS) | BLK_MIG_FLAG_PROGRESS);
+
+ DPRINTF("Block migration completed\n");
+
qemu_put_be64(f, BLK_MIG_FLAG_EOS);
- return ((stage == 2) && is_stage2_completed());
+ blk_mig_cleanup();
+ return 0;
+}
+
+static uint64_t block_save_pending(QEMUFile *f, void *opaque, uint64_t max_size)
+{
+ /* Estimate pending number of bytes to send */
+ uint64_t pending;
+
+ qemu_mutex_lock_iothread();
+ blk_mig_lock();
+ pending = get_remaining_dirty() +
+ block_mig_state.submitted * BLOCK_SIZE +
+ block_mig_state.read_done * BLOCK_SIZE;
+
+ /* Report at least one block pending during bulk phase */
+ if (pending == 0 && !block_mig_state.bulk_completed) {
+ pending = BLOCK_SIZE;
+ }
+ blk_mig_unlock();
+ qemu_mutex_unlock_iothread();
+
+ DPRINTF("Enter save live pending %" PRIu64 "\n", pending);
+ return pending;
}
static int block_load(QEMUFile *f, void *opaque, int version_id)
int len, flags;
char device_name[256];
int64_t addr;
- BlockDriverState *bs;
+ BlockDriverState *bs, *bs_prev = NULL;
uint8_t *buf;
+ int64_t total_sectors = 0;
+ int nr_sectors;
+ int ret;
do {
addr = qemu_get_be64(f);
return -EINVAL;
}
- buf = qemu_malloc(BLOCK_SIZE);
+ 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 = g_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);
+ g_free(buf);
+ if (ret < 0) {
+ return ret;
+ }
} else if (flags & BLK_MIG_FLAG_PROGRESS) {
if (!banner_printed) {
printf("Receiving block device images\n");
(addr == 100) ? '\n' : '\r');
fflush(stdout);
} else if (!(flags & BLK_MIG_FLAG_EOS)) {
- fprintf(stderr, "Unknown flags\n");
+ fprintf(stderr, "Unknown block migration flags: %#x\n", flags);
return -EINVAL;
}
- if (qemu_file_has_error(f)) {
- return -EIO;
+ ret = qemu_file_get_error(f);
+ if (ret != 0) {
+ return ret;
}
} while (!(flags & BLK_MIG_FLAG_EOS));
return 0;
}
-static void block_set_params(int blk_enable, int shared_base, void *opaque)
+static void block_set_params(const MigrationParams *params, void *opaque)
{
- block_mig_state.blk_enable = blk_enable;
- block_mig_state.shared_base = shared_base;
+ block_mig_state.blk_enable = params->blk;
+ block_mig_state.shared_base = params->shared;
/* shared base means that blk_enable = 1 */
- block_mig_state.blk_enable |= shared_base;
+ block_mig_state.blk_enable |= params->shared;
}
+static bool block_is_active(void *opaque)
+{
+ return block_mig_state.blk_enable == 1;
+}
+
+SaveVMHandlers savevm_block_handlers = {
+ .set_params = block_set_params,
+ .save_live_setup = block_save_setup,
+ .save_live_iterate = block_save_iterate,
+ .save_live_complete = block_save_complete,
+ .save_live_pending = block_save_pending,
+ .load_state = block_load,
+ .cancel = block_migration_cancel,
+ .is_active = block_is_active,
+};
+
void blk_mig_init(void)
{
QSIMPLEQ_INIT(&block_mig_state.bmds_list);
QSIMPLEQ_INIT(&block_mig_state.blk_list);
+ qemu_mutex_init(&block_mig_state.lock);
- 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, &savevm_block_handlers,
+ &block_mig_state);
}
projects / qemu.git / blobdiff