*
*/
+#include "qemu/osdep.h"
#include "trace.h"
#include "block/blockjob.h"
#include "block/block_int.h"
#include "sysemu/block-backend.h"
+#include "qapi/error.h"
#include "qapi/qmp/qerror.h"
#include "qemu/ratelimit.h"
#include "qemu/bitmap.h"
typedef struct MirrorBlockJob {
BlockJob common;
RateLimit limit;
- BlockDriverState *target;
+ BlockBackend *target;
BlockDriverState *base;
/* The name of the graph node to replace */
char *replaces;
/* Used to block operations on the drive-mirror-replace target */
Error *replace_blocker;
bool is_none_mode;
+ BlockMirrorBackingMode backing_mode;
BlockdevOnError on_source_error, on_target_error;
bool synced;
bool should_complete;
- int64_t sector_num;
int64_t granularity;
size_t buf_size;
int64_t bdev_length;
int ret;
bool unmap;
bool waiting_for_io;
+ int target_cluster_sectors;
+ int max_iov;
} MirrorBlockJob;
typedef struct MirrorOp {
{
s->synced = false;
if (read) {
- return block_job_error_action(&s->common, s->common.bs,
- s->on_source_error, true, error);
+ return block_job_error_action(&s->common, s->on_source_error,
+ true, error);
} else {
- return block_job_error_action(&s->common, s->target,
- s->on_target_error, false, error);
+ return block_job_error_action(&s->common, s->on_target_error,
+ false, error);
}
}
sectors_per_chunk = s->granularity >> BDRV_SECTOR_BITS;
chunk_num = op->sector_num / sectors_per_chunk;
- nb_chunks = op->nb_sectors / sectors_per_chunk;
+ nb_chunks = DIV_ROUND_UP(op->nb_sectors, sectors_per_chunk);
bitmap_clear(s->in_flight_bitmap, chunk_num, nb_chunks);
if (ret >= 0) {
if (s->cow_bitmap) {
g_free(op);
if (s->waiting_for_io) {
- qemu_coroutine_enter(s->common.co, NULL);
+ qemu_coroutine_enter(s->common.co);
}
}
mirror_iteration_done(op, ret);
return;
}
- bdrv_aio_writev(s->target, op->sector_num, &op->qiov, op->nb_sectors,
- mirror_write_complete, op);
+ blk_aio_pwritev(s->target, op->sector_num * BDRV_SECTOR_SIZE, &op->qiov,
+ 0, mirror_write_complete, op);
}
-static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
+static inline void mirror_clip_sectors(MirrorBlockJob *s,
+ int64_t sector_num,
+ int *nb_sectors)
{
- BlockDriverState *source = s->common.bs;
- int nb_sectors, sectors_per_chunk, nb_chunks;
- int64_t end, sector_num, next_chunk, next_sector, hbitmap_next_sector;
- uint64_t delay_ns = 0;
- MirrorOp *op;
- int pnum;
- int64_t ret;
-
- s->sector_num = hbitmap_iter_next(&s->hbi);
- if (s->sector_num < 0) {
- bdrv_dirty_iter_init(s->dirty_bitmap, &s->hbi);
- s->sector_num = hbitmap_iter_next(&s->hbi);
- trace_mirror_restart_iter(s, bdrv_get_dirty_count(s->dirty_bitmap));
- assert(s->sector_num >= 0);
- }
-
- hbitmap_next_sector = s->sector_num;
- sector_num = s->sector_num;
- sectors_per_chunk = s->granularity >> BDRV_SECTOR_BITS;
- end = s->bdev_length / BDRV_SECTOR_SIZE;
-
- /* Extend the QEMUIOVector to include all adjacent blocks that will
- * be copied in this operation.
- *
- * We have to do this if we have no backing file yet in the destination,
- * and the cluster size is very large. Then we need to do COW ourselves.
- * The first time a cluster is copied, copy it entirely. Note that,
- * because both the granularity and the cluster size are powers of two,
- * the number of sectors to copy cannot exceed one cluster.
- *
- * We also want to extend the QEMUIOVector to include more adjacent
- * dirty blocks if possible, to limit the number of I/O operations and
- * run efficiently even with a small granularity.
- */
- nb_chunks = 0;
- nb_sectors = 0;
- next_sector = sector_num;
- next_chunk = sector_num / sectors_per_chunk;
+ *nb_sectors = MIN(*nb_sectors,
+ s->bdev_length / BDRV_SECTOR_SIZE - sector_num);
+}
- /* Wait for I/O to this cluster (from a previous iteration) to be done. */
- while (test_bit(next_chunk, s->in_flight_bitmap)) {
- trace_mirror_yield_in_flight(s, sector_num, s->in_flight);
- s->waiting_for_io = true;
- qemu_coroutine_yield();
- s->waiting_for_io = false;
+/* Round sector_num and/or nb_sectors to target cluster if COW is needed, and
+ * return the offset of the adjusted tail sector against original. */
+static int mirror_cow_align(MirrorBlockJob *s,
+ int64_t *sector_num,
+ int *nb_sectors)
+{
+ bool need_cow;
+ int ret = 0;
+ int chunk_sectors = s->granularity >> BDRV_SECTOR_BITS;
+ int64_t align_sector_num = *sector_num;
+ int align_nb_sectors = *nb_sectors;
+ int max_sectors = chunk_sectors * s->max_iov;
+
+ need_cow = !test_bit(*sector_num / chunk_sectors, s->cow_bitmap);
+ need_cow |= !test_bit((*sector_num + *nb_sectors - 1) / chunk_sectors,
+ s->cow_bitmap);
+ if (need_cow) {
+ bdrv_round_sectors_to_clusters(blk_bs(s->target), *sector_num,
+ *nb_sectors, &align_sector_num,
+ &align_nb_sectors);
+ }
+
+ if (align_nb_sectors > max_sectors) {
+ align_nb_sectors = max_sectors;
+ if (need_cow) {
+ align_nb_sectors = QEMU_ALIGN_DOWN(align_nb_sectors,
+ s->target_cluster_sectors);
+ }
}
+ /* Clipping may result in align_nb_sectors unaligned to chunk boundary, but
+ * that doesn't matter because it's already the end of source image. */
+ mirror_clip_sectors(s, align_sector_num, &align_nb_sectors);
- do {
- int added_sectors, added_chunks;
+ ret = align_sector_num + align_nb_sectors - (*sector_num + *nb_sectors);
+ *sector_num = align_sector_num;
+ *nb_sectors = align_nb_sectors;
+ assert(ret >= 0);
+ return ret;
+}
- if (!bdrv_get_dirty(source, s->dirty_bitmap, next_sector) ||
- test_bit(next_chunk, s->in_flight_bitmap)) {
- assert(nb_sectors > 0);
- break;
- }
+static inline void mirror_wait_for_io(MirrorBlockJob *s)
+{
+ assert(!s->waiting_for_io);
+ s->waiting_for_io = true;
+ qemu_coroutine_yield();
+ s->waiting_for_io = false;
+}
- added_sectors = sectors_per_chunk;
- if (s->cow_bitmap && !test_bit(next_chunk, s->cow_bitmap)) {
- bdrv_round_to_clusters(s->target,
- next_sector, added_sectors,
- &next_sector, &added_sectors);
+/* Submit async read while handling COW.
+ * Returns: The number of sectors copied after and including sector_num,
+ * excluding any sectors copied prior to sector_num due to alignment.
+ * This will be nb_sectors if no alignment is necessary, or
+ * (new_end - sector_num) if tail is rounded up or down due to
+ * alignment or buffer limit.
+ */
+static int mirror_do_read(MirrorBlockJob *s, int64_t sector_num,
+ int nb_sectors)
+{
+ BlockBackend *source = s->common.blk;
+ int sectors_per_chunk, nb_chunks;
+ int ret;
+ MirrorOp *op;
+ int max_sectors;
- /* On the first iteration, the rounding may make us copy
- * sectors before the first dirty one.
- */
- if (next_sector < sector_num) {
- assert(nb_sectors == 0);
- sector_num = next_sector;
- next_chunk = next_sector / sectors_per_chunk;
- }
- }
+ sectors_per_chunk = s->granularity >> BDRV_SECTOR_BITS;
+ max_sectors = sectors_per_chunk * s->max_iov;
- added_sectors = MIN(added_sectors, end - (sector_num + nb_sectors));
- added_chunks = (added_sectors + sectors_per_chunk - 1) / sectors_per_chunk;
+ /* We can only handle as much as buf_size at a time. */
+ nb_sectors = MIN(s->buf_size >> BDRV_SECTOR_BITS, nb_sectors);
+ nb_sectors = MIN(max_sectors, nb_sectors);
+ assert(nb_sectors);
+ ret = nb_sectors;
- /* When doing COW, it may happen that there is not enough space for
- * a full cluster. Wait if that is the case.
- */
- while (nb_chunks == 0 && s->buf_free_count < added_chunks) {
- trace_mirror_yield_buf_busy(s, nb_chunks, s->in_flight);
- s->waiting_for_io = true;
- qemu_coroutine_yield();
- s->waiting_for_io = false;
- }
- if (s->buf_free_count < nb_chunks + added_chunks) {
- trace_mirror_break_buf_busy(s, nb_chunks, s->in_flight);
- break;
- }
- if (IOV_MAX < nb_chunks + added_chunks) {
- trace_mirror_break_iov_max(s, nb_chunks, added_chunks);
- break;
- }
-
- /* We have enough free space to copy these sectors. */
- bitmap_set(s->in_flight_bitmap, next_chunk, added_chunks);
+ if (s->cow_bitmap) {
+ ret += mirror_cow_align(s, §or_num, &nb_sectors);
+ }
+ assert(nb_sectors << BDRV_SECTOR_BITS <= s->buf_size);
+ /* The sector range must meet granularity because:
+ * 1) Caller passes in aligned values;
+ * 2) mirror_cow_align is used only when target cluster is larger. */
+ assert(!(sector_num % sectors_per_chunk));
+ nb_chunks = DIV_ROUND_UP(nb_sectors, sectors_per_chunk);
- nb_sectors += added_sectors;
- nb_chunks += added_chunks;
- next_sector += added_sectors;
- next_chunk += added_chunks;
- if (!s->synced && s->common.speed) {
- delay_ns = ratelimit_calculate_delay(&s->limit, added_sectors);
- }
- } while (delay_ns == 0 && next_sector < end);
+ while (s->buf_free_count < nb_chunks) {
+ trace_mirror_yield_in_flight(s, sector_num, s->in_flight);
+ mirror_wait_for_io(s);
+ }
/* Allocate a MirrorOp that is used as an AIO callback. */
op = g_new(MirrorOp, 1);
* from s->buf_free.
*/
qemu_iovec_init(&op->qiov, nb_chunks);
- next_sector = sector_num;
while (nb_chunks-- > 0) {
MirrorBuffer *buf = QSIMPLEQ_FIRST(&s->buf_free);
- size_t remaining = (nb_sectors * BDRV_SECTOR_SIZE) - op->qiov.size;
+ size_t remaining = nb_sectors * BDRV_SECTOR_SIZE - op->qiov.size;
QSIMPLEQ_REMOVE_HEAD(&s->buf_free, next);
s->buf_free_count--;
qemu_iovec_add(&op->qiov, buf, MIN(s->granularity, remaining));
-
- /* Advance the HBitmapIter in parallel, so that we do not examine
- * the same sector twice.
- */
- if (next_sector > hbitmap_next_sector
- && bdrv_get_dirty(source, s->dirty_bitmap, next_sector)) {
- hbitmap_next_sector = hbitmap_iter_next(&s->hbi);
- }
-
- next_sector += sectors_per_chunk;
}
- bdrv_reset_dirty_bitmap(s->dirty_bitmap, sector_num, nb_sectors);
-
/* Copy the dirty cluster. */
s->in_flight++;
s->sectors_in_flight += nb_sectors;
trace_mirror_one_iteration(s, sector_num, nb_sectors);
- ret = bdrv_get_block_status_above(source, NULL, sector_num,
- nb_sectors, &pnum);
- if (ret < 0 || pnum < nb_sectors ||
- (ret & BDRV_BLOCK_DATA && !(ret & BDRV_BLOCK_ZERO))) {
- bdrv_aio_readv(source, sector_num, &op->qiov, nb_sectors,
- mirror_read_complete, op);
- } else if (ret & BDRV_BLOCK_ZERO) {
- bdrv_aio_write_zeroes(s->target, sector_num, op->nb_sectors,
+ blk_aio_preadv(source, sector_num * BDRV_SECTOR_SIZE, &op->qiov, 0,
+ mirror_read_complete, op);
+ return ret;
+}
+
+static void mirror_do_zero_or_discard(MirrorBlockJob *s,
+ int64_t sector_num,
+ int nb_sectors,
+ bool is_discard)
+{
+ MirrorOp *op;
+
+ /* Allocate a MirrorOp that is used as an AIO callback. The qiov is zeroed
+ * so the freeing in mirror_iteration_done is nop. */
+ op = g_new0(MirrorOp, 1);
+ op->s = s;
+ op->sector_num = sector_num;
+ op->nb_sectors = nb_sectors;
+
+ s->in_flight++;
+ s->sectors_in_flight += nb_sectors;
+ if (is_discard) {
+ blk_aio_discard(s->target, sector_num, op->nb_sectors,
+ mirror_write_complete, op);
+ } else {
+ blk_aio_pwrite_zeroes(s->target, sector_num * BDRV_SECTOR_SIZE,
+ op->nb_sectors * BDRV_SECTOR_SIZE,
s->unmap ? BDRV_REQ_MAY_UNMAP : 0,
mirror_write_complete, op);
- } else {
- assert(!(ret & BDRV_BLOCK_DATA));
- bdrv_aio_discard(s->target, sector_num, op->nb_sectors,
- mirror_write_complete, op);
+ }
+}
+
+static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
+{
+ BlockDriverState *source = blk_bs(s->common.blk);
+ int64_t sector_num, first_chunk;
+ uint64_t delay_ns = 0;
+ /* At least the first dirty chunk is mirrored in one iteration. */
+ int nb_chunks = 1;
+ int64_t end = s->bdev_length / BDRV_SECTOR_SIZE;
+ int sectors_per_chunk = s->granularity >> BDRV_SECTOR_BITS;
+
+ sector_num = hbitmap_iter_next(&s->hbi);
+ if (sector_num < 0) {
+ bdrv_dirty_iter_init(s->dirty_bitmap, &s->hbi);
+ sector_num = hbitmap_iter_next(&s->hbi);
+ trace_mirror_restart_iter(s, bdrv_get_dirty_count(s->dirty_bitmap));
+ assert(sector_num >= 0);
+ }
+
+ first_chunk = sector_num / sectors_per_chunk;
+ while (test_bit(first_chunk, s->in_flight_bitmap)) {
+ trace_mirror_yield_in_flight(s, sector_num, s->in_flight);
+ mirror_wait_for_io(s);
+ }
+
+ block_job_pause_point(&s->common);
+
+ /* Find the number of consective dirty chunks following the first dirty
+ * one, and wait for in flight requests in them. */
+ while (nb_chunks * sectors_per_chunk < (s->buf_size >> BDRV_SECTOR_BITS)) {
+ int64_t hbitmap_next;
+ int64_t next_sector = sector_num + nb_chunks * sectors_per_chunk;
+ int64_t next_chunk = next_sector / sectors_per_chunk;
+ if (next_sector >= end ||
+ !bdrv_get_dirty(source, s->dirty_bitmap, next_sector)) {
+ break;
+ }
+ if (test_bit(next_chunk, s->in_flight_bitmap)) {
+ break;
+ }
+
+ hbitmap_next = hbitmap_iter_next(&s->hbi);
+ if (hbitmap_next > next_sector || hbitmap_next < 0) {
+ /* The bitmap iterator's cache is stale, refresh it */
+ bdrv_set_dirty_iter(&s->hbi, next_sector);
+ hbitmap_next = hbitmap_iter_next(&s->hbi);
+ }
+ assert(hbitmap_next == next_sector);
+ nb_chunks++;
+ }
+
+ /* Clear dirty bits before querying the block status, because
+ * calling bdrv_get_block_status_above could yield - if some blocks are
+ * marked dirty in this window, we need to know.
+ */
+ bdrv_reset_dirty_bitmap(s->dirty_bitmap, sector_num,
+ nb_chunks * sectors_per_chunk);
+ bitmap_set(s->in_flight_bitmap, sector_num / sectors_per_chunk, nb_chunks);
+ while (nb_chunks > 0 && sector_num < end) {
+ int ret;
+ int io_sectors;
+ BlockDriverState *file;
+ enum MirrorMethod {
+ MIRROR_METHOD_COPY,
+ MIRROR_METHOD_ZERO,
+ MIRROR_METHOD_DISCARD
+ } mirror_method = MIRROR_METHOD_COPY;
+
+ assert(!(sector_num % sectors_per_chunk));
+ ret = bdrv_get_block_status_above(source, NULL, sector_num,
+ nb_chunks * sectors_per_chunk,
+ &io_sectors, &file);
+ if (ret < 0) {
+ io_sectors = nb_chunks * sectors_per_chunk;
+ }
+
+ io_sectors -= io_sectors % sectors_per_chunk;
+ if (io_sectors < sectors_per_chunk) {
+ io_sectors = sectors_per_chunk;
+ } else if (ret >= 0 && !(ret & BDRV_BLOCK_DATA)) {
+ int64_t target_sector_num;
+ int target_nb_sectors;
+ bdrv_round_sectors_to_clusters(blk_bs(s->target), sector_num,
+ io_sectors, &target_sector_num,
+ &target_nb_sectors);
+ if (target_sector_num == sector_num &&
+ target_nb_sectors == io_sectors) {
+ mirror_method = ret & BDRV_BLOCK_ZERO ?
+ MIRROR_METHOD_ZERO :
+ MIRROR_METHOD_DISCARD;
+ }
+ }
+
+ mirror_clip_sectors(s, sector_num, &io_sectors);
+ switch (mirror_method) {
+ case MIRROR_METHOD_COPY:
+ io_sectors = mirror_do_read(s, sector_num, io_sectors);
+ break;
+ case MIRROR_METHOD_ZERO:
+ mirror_do_zero_or_discard(s, sector_num, io_sectors, false);
+ break;
+ case MIRROR_METHOD_DISCARD:
+ mirror_do_zero_or_discard(s, sector_num, io_sectors, true);
+ break;
+ default:
+ abort();
+ }
+ assert(io_sectors);
+ sector_num += io_sectors;
+ nb_chunks -= DIV_ROUND_UP(io_sectors, sectors_per_chunk);
+ delay_ns += ratelimit_calculate_delay(&s->limit, io_sectors);
}
return delay_ns;
}
static void mirror_drain(MirrorBlockJob *s)
{
while (s->in_flight > 0) {
- s->waiting_for_io = true;
- qemu_coroutine_yield();
- s->waiting_for_io = false;
+ mirror_wait_for_io(s);
}
}
MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
MirrorExitData *data = opaque;
AioContext *replace_aio_context = NULL;
- BlockDriverState *src = s->common.bs;
+ BlockDriverState *src = blk_bs(s->common.blk);
+ BlockDriverState *target_bs = blk_bs(s->target);
/* Make sure that the source BDS doesn't go away before we called
* block_job_completed(). */
}
if (s->should_complete && data->ret == 0) {
- BlockDriverState *to_replace = s->common.bs;
+ BlockDriverState *to_replace = src;
if (s->to_replace) {
to_replace = s->to_replace;
}
- if (bdrv_get_flags(s->target) != bdrv_get_flags(to_replace)) {
- bdrv_reopen(s->target, bdrv_get_flags(to_replace), NULL);
+
+ if (bdrv_get_flags(target_bs) != bdrv_get_flags(to_replace)) {
+ bdrv_reopen(target_bs, bdrv_get_flags(to_replace), NULL);
}
- bdrv_replace_in_backing_chain(to_replace, s->target);
+
+ /* The mirror job has no requests in flight any more, but we need to
+ * drain potential other users of the BDS before changing the graph. */
+ bdrv_drained_begin(target_bs);
+ bdrv_replace_in_backing_chain(to_replace, target_bs);
+ bdrv_drained_end(target_bs);
+
+ /* We just changed the BDS the job BB refers to */
+ blk_remove_bs(job->blk);
+ blk_insert_bs(job->blk, src);
}
if (s->to_replace) {
bdrv_op_unblock_all(s->to_replace, s->replace_blocker);
aio_context_release(replace_aio_context);
}
g_free(s->replaces);
- bdrv_op_unblock_all(s->target, s->common.blocker);
- bdrv_unref(s->target);
+ bdrv_op_unblock_all(target_bs, s->common.blocker);
+ blk_unref(s->target);
block_job_completed(&s->common, data->ret);
g_free(data);
bdrv_drained_end(src);
+ if (qemu_get_aio_context() == bdrv_get_aio_context(src)) {
+ aio_enable_external(iohandler_get_aio_context());
+ }
bdrv_unref(src);
}
{
MirrorBlockJob *s = opaque;
MirrorExitData *data;
- BlockDriverState *bs = s->common.bs;
+ BlockDriverState *bs = blk_bs(s->common.blk);
+ BlockDriverState *target_bs = blk_bs(s->target);
int64_t sector_num, end, length;
uint64_t last_pause_ns;
BlockDriverInfo bdi;
checking for a NULL string */
int ret = 0;
int n;
+ int target_cluster_size = BDRV_SECTOR_SIZE;
if (block_job_is_cancelled(&s->common)) {
goto immediate_exit;
* the destination do COW. Instead, we copy sectors around the
* dirty data if needed. We need a bitmap to do that.
*/
- bdrv_get_backing_filename(s->target, backing_filename,
+ bdrv_get_backing_filename(target_bs, backing_filename,
sizeof(backing_filename));
- if (backing_filename[0] && !s->target->backing) {
- ret = bdrv_get_info(s->target, &bdi);
- if (ret < 0) {
- goto immediate_exit;
- }
- if (s->granularity < bdi.cluster_size) {
- s->buf_size = MAX(s->buf_size, bdi.cluster_size);
- s->cow_bitmap = bitmap_new(length);
- }
+ if (!bdrv_get_info(target_bs, &bdi) && bdi.cluster_size) {
+ target_cluster_size = bdi.cluster_size;
+ }
+ if (backing_filename[0] && !target_bs->backing
+ && s->granularity < target_cluster_size) {
+ s->buf_size = MAX(s->buf_size, target_cluster_size);
+ s->cow_bitmap = bitmap_new(length);
}
+ s->target_cluster_sectors = target_cluster_size >> BDRV_SECTOR_BITS;
+ s->max_iov = MIN(bs->bl.max_iov, target_bs->bl.max_iov);
end = s->bdev_length / BDRV_SECTOR_SIZE;
s->buf = qemu_try_blockalign(bs, s->buf_size);
if (!s->is_none_mode) {
/* First part, loop on the sectors and initialize the dirty bitmap. */
BlockDriverState *base = s->base;
- bool mark_all_dirty = s->base == NULL && !bdrv_has_zero_init(s->target);
+ bool mark_all_dirty = s->base == NULL && !bdrv_has_zero_init(target_bs);
for (sector_num = 0; sector_num < end; ) {
/* Just to make sure we are not exceeding int limit. */
if (now - last_pause_ns > SLICE_TIME) {
last_pause_ns = now;
block_job_sleep_ns(&s->common, QEMU_CLOCK_REALTIME, 0);
+ } else {
+ block_job_pause_point(&s->common);
}
if (block_job_is_cancelled(&s->common)) {
goto immediate_exit;
}
+ block_job_pause_point(&s->common);
+
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
/* s->common.offset contains the number of bytes already processed so
* far, cnt is the number of dirty sectors remaining and
if (s->in_flight == MAX_IN_FLIGHT || s->buf_free_count == 0 ||
(cnt == 0 && s->in_flight > 0)) {
trace_mirror_yield(s, s->in_flight, s->buf_free_count, cnt);
- s->waiting_for_io = true;
- qemu_coroutine_yield();
- s->waiting_for_io = false;
+ mirror_wait_for_io(s);
continue;
} else if (cnt != 0) {
delay_ns = mirror_iteration(s);
should_complete = false;
if (s->in_flight == 0 && cnt == 0) {
trace_mirror_before_flush(s);
- ret = bdrv_flush(s->target);
+ ret = blk_flush(s->target);
if (ret < 0) {
if (mirror_error_action(s, false, -ret) ==
BLOCK_ERROR_ACTION_REPORT) {
* mirror_populate runs.
*/
trace_mirror_before_drain(s, cnt);
- bdrv_drain(bs);
+ bdrv_co_drain(bs);
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
}
g_free(s->cow_bitmap);
g_free(s->in_flight_bitmap);
bdrv_release_dirty_bitmap(bs, s->dirty_bitmap);
- if (s->target->blk) {
- blk_iostatus_disable(s->target->blk);
- }
data = g_malloc(sizeof(*data));
data->ret = ret;
/* Before we switch to target in mirror_exit, make sure data doesn't
* change. */
- bdrv_drained_begin(s->common.bs);
+ bdrv_drained_begin(bs);
+ if (qemu_get_aio_context() == bdrv_get_aio_context(bs)) {
+ /* FIXME: virtio host notifiers run on iohandler_ctx, therefore the
+ * above bdrv_drained_end isn't enough to quiesce it. This is ugly, we
+ * need a block layer API change to achieve this. */
+ aio_disable_external(iohandler_get_aio_context());
+ }
block_job_defer_to_main_loop(&s->common, mirror_exit, data);
}
ratelimit_set_speed(&s->limit, speed / BDRV_SECTOR_SIZE, SLICE_TIME);
}
-static void mirror_iostatus_reset(BlockJob *job)
-{
- MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
-
- if (s->target->blk) {
- blk_iostatus_reset(s->target->blk);
- }
-}
-
static void mirror_complete(BlockJob *job, Error **errp)
{
MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
- Error *local_err = NULL;
- int ret;
+ BlockDriverState *src, *target;
+
+ src = blk_bs(job->blk);
+ target = blk_bs(s->target);
- ret = bdrv_open_backing_file(s->target, NULL, &local_err);
- if (ret < 0) {
- error_propagate(errp, local_err);
- return;
- }
if (!s->synced) {
- error_setg(errp, QERR_BLOCK_JOB_NOT_READY, job->id);
+ error_setg(errp, "The active block job '%s' cannot be completed",
+ job->id);
return;
}
- /* check the target bs is not blocked and block all operations on it */
+ if (s->backing_mode == MIRROR_OPEN_BACKING_CHAIN) {
+ int ret;
+
+ assert(!target->backing);
+ ret = bdrv_open_backing_file(target, NULL, "backing", errp);
+ if (ret < 0) {
+ return;
+ }
+ }
+
+ /* block all operations on to_replace bs */
if (s->replaces) {
AioContext *replace_aio_context;
aio_context_release(replace_aio_context);
}
+ if (s->backing_mode == MIRROR_SOURCE_BACKING_CHAIN) {
+ BlockDriverState *backing = s->is_none_mode ? src : s->base;
+ if (backing_bs(target) != backing) {
+ bdrv_set_backing_hd(target, backing);
+ }
+ }
+
s->should_complete = true;
block_job_enter(&s->common);
}
+/* There is no matching mirror_resume() because mirror_run() will begin
+ * iterating again when the job is resumed.
+ */
+static void coroutine_fn mirror_pause(BlockJob *job)
+{
+ MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
+
+ mirror_drain(s);
+}
+
+static void mirror_attached_aio_context(BlockJob *job, AioContext *new_context)
+{
+ MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
+
+ blk_set_aio_context(s->target, new_context);
+}
+
static const BlockJobDriver mirror_job_driver = {
- .instance_size = sizeof(MirrorBlockJob),
- .job_type = BLOCK_JOB_TYPE_MIRROR,
- .set_speed = mirror_set_speed,
- .iostatus_reset= mirror_iostatus_reset,
- .complete = mirror_complete,
+ .instance_size = sizeof(MirrorBlockJob),
+ .job_type = BLOCK_JOB_TYPE_MIRROR,
+ .set_speed = mirror_set_speed,
+ .complete = mirror_complete,
+ .pause = mirror_pause,
+ .attached_aio_context = mirror_attached_aio_context,
};
static const BlockJobDriver commit_active_job_driver = {
- .instance_size = sizeof(MirrorBlockJob),
- .job_type = BLOCK_JOB_TYPE_COMMIT,
- .set_speed = mirror_set_speed,
- .iostatus_reset
- = mirror_iostatus_reset,
- .complete = mirror_complete,
+ .instance_size = sizeof(MirrorBlockJob),
+ .job_type = BLOCK_JOB_TYPE_COMMIT,
+ .set_speed = mirror_set_speed,
+ .complete = mirror_complete,
+ .pause = mirror_pause,
+ .attached_aio_context = mirror_attached_aio_context,
};
-static void mirror_start_job(BlockDriverState *bs, BlockDriverState *target,
- const char *replaces,
+static void mirror_start_job(const char *job_id, BlockDriverState *bs,
+ BlockDriverState *target, const char *replaces,
int64_t speed, uint32_t granularity,
int64_t buf_size,
+ BlockMirrorBackingMode backing_mode,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap,
assert ((granularity & (granularity - 1)) == 0);
- if ((on_source_error == BLOCKDEV_ON_ERROR_STOP ||
- on_source_error == BLOCKDEV_ON_ERROR_ENOSPC) &&
- (!bs->blk || !blk_iostatus_is_enabled(bs->blk))) {
- error_setg(errp, QERR_INVALID_PARAMETER, "on-source-error");
- return;
- }
-
if (buf_size < 0) {
error_setg(errp, "Invalid parameter 'buf-size'");
return;
buf_size = DEFAULT_MIRROR_BUF_SIZE;
}
- s = block_job_create(driver, bs, speed, cb, opaque, errp);
+ s = block_job_create(job_id, driver, bs, speed, cb, opaque, errp);
if (!s) {
return;
}
+ s->target = blk_new();
+ blk_insert_bs(s->target, target);
+
s->replaces = g_strdup(replaces);
s->on_source_error = on_source_error;
s->on_target_error = on_target_error;
- s->target = target;
s->is_none_mode = is_none_mode;
+ s->backing_mode = backing_mode;
s->base = base;
s->granularity = granularity;
s->buf_size = ROUND_UP(buf_size, granularity);
s->dirty_bitmap = bdrv_create_dirty_bitmap(bs, granularity, NULL, errp);
if (!s->dirty_bitmap) {
g_free(s->replaces);
+ blk_unref(s->target);
block_job_unref(&s->common);
return;
}
- bdrv_op_block_all(s->target, s->common.blocker);
+ bdrv_op_block_all(target, s->common.blocker);
- bdrv_set_enable_write_cache(s->target, true);
- if (s->target->blk) {
- blk_set_on_error(s->target->blk, on_target_error, on_target_error);
- blk_iostatus_enable(s->target->blk);
- }
- s->common.co = qemu_coroutine_create(mirror_run);
+ s->common.co = qemu_coroutine_create(mirror_run, s);
trace_mirror_start(bs, s, s->common.co, opaque);
- qemu_coroutine_enter(s->common.co, s);
+ qemu_coroutine_enter(s->common.co);
}
-void mirror_start(BlockDriverState *bs, BlockDriverState *target,
- const char *replaces,
+void mirror_start(const char *job_id, BlockDriverState *bs,
+ BlockDriverState *target, const char *replaces,
int64_t speed, uint32_t granularity, int64_t buf_size,
- MirrorSyncMode mode, BlockdevOnError on_source_error,
+ MirrorSyncMode mode, BlockMirrorBackingMode backing_mode,
+ BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap,
BlockCompletionFunc *cb,
}
is_none_mode = mode == MIRROR_SYNC_MODE_NONE;
base = mode == MIRROR_SYNC_MODE_TOP ? backing_bs(bs) : NULL;
- mirror_start_job(bs, target, replaces,
- speed, granularity, buf_size,
+ mirror_start_job(job_id, bs, target, replaces,
+ speed, granularity, buf_size, backing_mode,
on_source_error, on_target_error, unmap, cb, opaque, errp,
&mirror_job_driver, is_none_mode, base);
}
-void commit_active_start(BlockDriverState *bs, BlockDriverState *base,
- int64_t speed,
+void commit_active_start(const char *job_id, BlockDriverState *bs,
+ BlockDriverState *base, int64_t speed,
BlockdevOnError on_error,
BlockCompletionFunc *cb,
void *opaque, Error **errp)
}
}
- bdrv_ref(base);
- mirror_start_job(bs, base, NULL, speed, 0, 0,
+ mirror_start_job(job_id, bs, base, NULL, speed, 0, 0,
+ MIRROR_LEAVE_BACKING_CHAIN,
on_error, on_error, false, cb, opaque, &local_err,
&commit_active_job_driver, false, base);
if (local_err) {