*/
#include "qemu-common.h"
-#include "block_int.h"
+#include "block/block_int.h"
#include "hw/hw.h"
-#include "qemu-queue.h"
-#include "qemu-timer.h"
-#include "block-migration.h"
-#include "migration.h"
-#include "blockdev.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
int64_t cur_dirty;
int64_t completed_sectors;
int64_t total_sectors;
- int64_t dirty;
QSIMPLEQ_ENTRY(BlkMigDevState) entry;
unsigned long *aio_bitmap;
} BlkMigDevState;
int64_t total_sector_sum;
int prev_progress;
int bulk_completed;
- long double total_time;
- long double prev_time_offset;
- int reads;
} BlkMigState;
static BlkMigState block_mig_state;
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;
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;
- 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);
bmds_set_aio_inflight(blk->bmds, blk->sector, blk->nr_sectors, 0);
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(bs, cur_sector, &blk->qiov,
nr_sectors, blk_mig_read_cb, blk);
block_mig_state.submitted++;
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);
}
}
block_mig_state.total_sector_sum = 0;
block_mig_state.prev_progress = -1;
block_mig_state.bulk_completed = 0;
- block_mig_state.total_time = 0;
- block_mig_state.reads = 0;
bdrv_iterate(init_blk_migration_it, NULL);
}
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);
block_mig_state.submitted++;
error:
DPRINTF("Error reading sector %" PRId64 "\n", sector);
- qemu_file_set_error(f, ret);
g_free(blk->buf);
g_free(blk);
- return 0;
+ return ret;
}
+/* 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 = 0;
+ int ret = 1;
QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
- if (mig_save_device_dirty(f, bmds, is_async) == 0) {
- ret = 1;
+ ret = mig_save_device_dirty(f, bmds, is_async);
+ if (ret <= 0) {
break;
}
}
return ret;
}
-static void flush_blks(QEMUFile* f)
+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,
break;
}
if (blk->ret < 0) {
- qemu_file_set_error(f, blk->ret);
+ ret = blk->ret;
break;
}
blk_send(f, blk);
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 int64_t get_remaining_dirty(void)
dirty += bdrv_get_dirty_count(bmds->bs);
}
- 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 remaining work
- below max_downtime */
-
- return 1;
- }
- }
-
- return 0;
+ return dirty << BDRV_SECTOR_BITS;
}
static void blk_mig_cleanup(void)
BlkMigDevState *bmds;
BlkMigBlock *blk;
+ bdrv_drain_all();
+
set_dirty_tracking(0);
while ((bmds = QSIMPLEQ_FIRST(&block_mig_state.bmds_list)) != NULL) {
}
}
-static int block_save_live(QEMUFile *f, int stage, void *opaque)
+static void block_migration_cancel(void *opaque)
+{
+ blk_mig_cleanup();
+}
+
+static int block_save_setup(QEMUFile *f, void *opaque)
{
int ret;
- DPRINTF("Enter save live stage %d submitted %d transferred %d\n",
- stage, block_mig_state.submitted, block_mig_state.transferred);
+ DPRINTF("Enter save live setup submitted %d transferred %d\n",
+ block_mig_state.submitted, block_mig_state.transferred);
- if (stage < 0) {
- blk_mig_cleanup();
- return 0;
- }
+ init_blk_migration(f);
- if (block_mig_state.blk_enable != 1) {
- /* no need to migrate storage */
- qemu_put_be64(f, BLK_MIG_FLAG_EOS);
- return 1;
- }
+ /* 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);
- if (stage == 1) {
- init_blk_migration(f);
+ return ret;
+}
- /* start track dirty blocks */
- set_dirty_tracking(1);
- }
+static int block_save_iterate(QEMUFile *f, void *opaque)
+{
+ int ret;
+ int64_t last_ftell = qemu_ftell(f);
- flush_blks(f);
+ DPRINTF("Enter save live iterate submitted %d transferred %d\n",
+ block_mig_state.submitted, block_mig_state.transferred);
- ret = qemu_file_get_error(f);
+ ret = flush_blks(f);
if (ret) {
- blk_mig_cleanup();
return ret;
}
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(f) == 0) {
- /* finished saving bulk on all devices */
- block_mig_state.bulk_completed = 1;
- }
- } else {
- if (blk_mig_save_dirty_block(f, 1) == 0) {
- /* no more dirty blocks */
- break;
- }
+ /* 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(f) == 0) {
+ /* finished saving bulk on all devices */
+ block_mig_state.bulk_completed = 1;
+ }
+ } else {
+ ret = blk_mig_save_dirty_block(f, 1);
+ if (ret < 0) {
+ return ret;
+ }
+ if (ret != 0) {
+ /* no more dirty blocks */
+ break;
}
}
+ }
- flush_blks(f);
-
- ret = qemu_file_get_error(f);
- if (ret) {
- blk_mig_cleanup();
- return ret;
- }
+ ret = flush_blks(f);
+ if (ret) {
+ return ret;
}
- if (stage == 3) {
- /* we know for sure that save bulk is completed and
- all async read completed */
- assert(block_mig_state.submitted == 0);
+ qemu_put_be64(f, BLK_MIG_FLAG_EOS);
- while (blk_mig_save_dirty_block(f, 0) != 0);
- blk_mig_cleanup();
+ return qemu_ftell(f) - last_ftell;
+}
- /* 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;
- ret = qemu_file_get_error(f);
- if (ret) {
+ DPRINTF("Enter save live complete 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();
+
+ /* we know for sure that save bulk is completed and
+ all async read completed */
+ assert(block_mig_state.submitted == 0);
+
+ do {
+ ret = blk_mig_save_dirty_block(f, 0);
+ if (ret < 0) {
return ret;
}
+ } while (ret == 0);
- DPRINTF("Block migration completed\n");
- }
+ /* 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 = 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;
+ }
+
+ DPRINTF("Enter save live pending %" PRIu64 "\n", pending);
+ return pending;
}
static int block_load(QEMUFile *f, void *opaque, int version_id)
(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;
}
ret = qemu_file_get_error(f);
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);
- register_savevm_live(NULL, "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