summary refs log tree commit diff stats
path: root/block
diff options
context:
space:
mode:
Diffstat (limited to 'block')
-rw-r--r--block/Makefile.objs2
-rw-r--r--block/backup.c246
-rw-r--r--block/block-backend.c50
-rw-r--r--block/commit.c13
-rw-r--r--block/crypto.c3
-rw-r--r--block/dirty-bitmap.c9
-rw-r--r--block/io.c28
-rw-r--r--block/linux-aio.c72
-rw-r--r--block/mirror.c4
-rw-r--r--block/parallels.c3
-rw-r--r--block/qcow.c3
-rw-r--r--block/qcow2-bitmap.c3
-rw-r--r--block/qcow2-cache.c1
-rw-r--r--block/qcow2-cluster.c10
-rw-r--r--block/qcow2-refcount.c40
-rw-r--r--block/qcow2-snapshot.c1
-rw-r--r--block/qcow2-threads.c268
-rw-r--r--block/qcow2.c337
-rw-r--r--block/qcow2.h30
-rw-r--r--block/qed.c3
-rw-r--r--block/quorum.c1
-rw-r--r--block/sheepdog.c3
-rw-r--r--block/trace-events4
-rw-r--r--block/vdi.c3
-rw-r--r--block/vhdx.c3
-rw-r--r--block/vmdk.c3
-rw-r--r--block/vpc.c3
27 files changed, 673 insertions, 473 deletions
diff --git a/block/Makefile.objs b/block/Makefile.objs
index 7a81892a52..ae11605c9f 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -6,7 +6,7 @@ block-obj-$(CONFIG_BOCHS) += bochs.o
 block-obj-$(CONFIG_VVFAT) += vvfat.o
 block-obj-$(CONFIG_DMG) += dmg.o
 
-block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o qcow2-bitmap.o
+block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o qcow2-bitmap.o qcow2-threads.o
 block-obj-$(CONFIG_QED) += qed.o qed-l2-cache.o qed-table.o qed-cluster.o
 block-obj-$(CONFIG_QED) += qed-check.o
 block-obj-y += vhdx.o vhdx-endian.o vhdx-log.o
diff --git a/block/backup.c b/block/backup.c
index 916817d8b1..715e1d3be8 100644
--- a/block/backup.c
+++ b/block/backup.c
@@ -112,7 +112,8 @@ static int coroutine_fn backup_cow_with_bounce_buffer(BackupBlockJob *job,
     int read_flags = is_write_notifier ? BDRV_REQ_NO_SERIALISING : 0;
     int write_flags = job->serialize_target_writes ? BDRV_REQ_SERIALISING : 0;
 
-    hbitmap_reset(job->copy_bitmap, start / job->cluster_size, 1);
+    assert(QEMU_IS_ALIGNED(start, job->cluster_size));
+    hbitmap_reset(job->copy_bitmap, start, job->cluster_size);
     nbytes = MIN(job->cluster_size, job->len - start);
     if (!*bounce_buffer) {
         *bounce_buffer = blk_blockalign(blk, job->cluster_size);
@@ -145,7 +146,7 @@ static int coroutine_fn backup_cow_with_bounce_buffer(BackupBlockJob *job,
 
     return nbytes;
 fail:
-    hbitmap_set(job->copy_bitmap, start / job->cluster_size, 1);
+    hbitmap_set(job->copy_bitmap, start, job->cluster_size);
     return ret;
 
 }
@@ -165,16 +166,15 @@ static int coroutine_fn backup_cow_with_offload(BackupBlockJob *job,
     int write_flags = job->serialize_target_writes ? BDRV_REQ_SERIALISING : 0;
 
     assert(QEMU_IS_ALIGNED(job->copy_range_size, job->cluster_size));
+    assert(QEMU_IS_ALIGNED(start, job->cluster_size));
     nbytes = MIN(job->copy_range_size, end - start);
     nr_clusters = DIV_ROUND_UP(nbytes, job->cluster_size);
-    hbitmap_reset(job->copy_bitmap, start / job->cluster_size,
-                  nr_clusters);
+    hbitmap_reset(job->copy_bitmap, start, job->cluster_size * nr_clusters);
     ret = blk_co_copy_range(blk, start, job->target, start, nbytes,
                             read_flags, write_flags);
     if (ret < 0) {
         trace_backup_do_cow_copy_range_fail(job, start, ret);
-        hbitmap_set(job->copy_bitmap, start / job->cluster_size,
-                    nr_clusters);
+        hbitmap_set(job->copy_bitmap, start, job->cluster_size * nr_clusters);
         return ret;
     }
 
@@ -202,7 +202,7 @@ static int coroutine_fn backup_do_cow(BackupBlockJob *job,
     cow_request_begin(&cow_request, job, start, end);
 
     while (start < end) {
-        if (!hbitmap_get(job->copy_bitmap, start / job->cluster_size)) {
+        if (!hbitmap_get(job->copy_bitmap, start)) {
             trace_backup_do_cow_skip(job, start);
             start += job->cluster_size;
             continue; /* already copied */
@@ -298,12 +298,16 @@ static void backup_clean(Job *job)
     assert(s->target);
     blk_unref(s->target);
     s->target = NULL;
+
+    if (s->copy_bitmap) {
+        hbitmap_free(s->copy_bitmap);
+        s->copy_bitmap = NULL;
+    }
 }
 
 void backup_do_checkpoint(BlockJob *job, Error **errp)
 {
     BackupBlockJob *backup_job = container_of(job, BackupBlockJob, common);
-    int64_t len;
 
     assert(block_job_driver(job) == &backup_job_driver);
 
@@ -313,8 +317,7 @@ void backup_do_checkpoint(BlockJob *job, Error **errp)
         return;
     }
 
-    len = DIV_ROUND_UP(backup_job->len, backup_job->cluster_size);
-    hbitmap_set(backup_job->copy_bitmap, 0, len);
+    hbitmap_set(backup_job->copy_bitmap, 0, backup_job->len);
 }
 
 static void backup_drain(BlockJob *job)
@@ -365,20 +368,44 @@ static bool coroutine_fn yield_and_check(BackupBlockJob *job)
     return false;
 }
 
-static int coroutine_fn backup_run_incremental(BackupBlockJob *job)
+static bool bdrv_is_unallocated_range(BlockDriverState *bs,
+                                      int64_t offset, int64_t bytes)
+{
+    int64_t end = offset + bytes;
+
+    while (offset < end && !bdrv_is_allocated(bs, offset, bytes, &bytes)) {
+        if (bytes == 0) {
+            return true;
+        }
+        offset += bytes;
+        bytes = end - offset;
+    }
+
+    return offset >= end;
+}
+
+static int coroutine_fn backup_loop(BackupBlockJob *job)
 {
     int ret;
     bool error_is_read;
-    int64_t cluster;
+    int64_t offset;
     HBitmapIter hbi;
+    BlockDriverState *bs = blk_bs(job->common.blk);
 
     hbitmap_iter_init(&hbi, job->copy_bitmap, 0);
-    while ((cluster = hbitmap_iter_next(&hbi)) != -1) {
+    while ((offset = hbitmap_iter_next(&hbi)) != -1) {
+        if (job->sync_mode == MIRROR_SYNC_MODE_TOP &&
+            bdrv_is_unallocated_range(bs, offset, job->cluster_size))
+        {
+            hbitmap_reset(job->copy_bitmap, offset, job->cluster_size);
+            continue;
+        }
+
         do {
             if (yield_and_check(job)) {
                 return 0;
             }
-            ret = backup_do_cow(job, cluster * job->cluster_size,
+            ret = backup_do_cow(job, offset,
                                 job->cluster_size, &error_is_read, false);
             if (ret < 0 && backup_error_action(job, error_is_read, -ret) ==
                            BLOCK_ERROR_ACTION_REPORT)
@@ -394,66 +421,43 @@ static int coroutine_fn backup_run_incremental(BackupBlockJob *job)
 /* init copy_bitmap from sync_bitmap */
 static void backup_incremental_init_copy_bitmap(BackupBlockJob *job)
 {
-    BdrvDirtyBitmapIter *dbi;
-    int64_t offset;
-    int64_t end = DIV_ROUND_UP(bdrv_dirty_bitmap_size(job->sync_bitmap),
-                               job->cluster_size);
+    uint64_t offset = 0;
+    uint64_t bytes = job->len;
 
-    dbi = bdrv_dirty_iter_new(job->sync_bitmap);
-    while ((offset = bdrv_dirty_iter_next(dbi)) != -1) {
-        int64_t cluster = offset / job->cluster_size;
-        int64_t next_cluster;
-
-        offset += bdrv_dirty_bitmap_granularity(job->sync_bitmap);
-        if (offset >= bdrv_dirty_bitmap_size(job->sync_bitmap)) {
-            hbitmap_set(job->copy_bitmap, cluster, end - cluster);
-            break;
-        }
+    while (bdrv_dirty_bitmap_next_dirty_area(job->sync_bitmap,
+                                             &offset, &bytes))
+    {
+        hbitmap_set(job->copy_bitmap, offset, bytes);
 
-        offset = bdrv_dirty_bitmap_next_zero(job->sync_bitmap, offset,
-                                             UINT64_MAX);
-        if (offset == -1) {
-            hbitmap_set(job->copy_bitmap, cluster, end - cluster);
+        offset += bytes;
+        if (offset >= job->len) {
             break;
         }
-
-        next_cluster = DIV_ROUND_UP(offset, job->cluster_size);
-        hbitmap_set(job->copy_bitmap, cluster, next_cluster - cluster);
-        if (next_cluster >= end) {
-            break;
-        }
-
-        bdrv_set_dirty_iter(dbi, next_cluster * job->cluster_size);
+        bytes = job->len - offset;
     }
 
     /* TODO job_progress_set_remaining() would make more sense */
     job_progress_update(&job->common.job,
-        job->len - hbitmap_count(job->copy_bitmap) * job->cluster_size);
-
-    bdrv_dirty_iter_free(dbi);
+        job->len - hbitmap_count(job->copy_bitmap));
 }
 
 static int coroutine_fn backup_run(Job *job, Error **errp)
 {
     BackupBlockJob *s = container_of(job, BackupBlockJob, common.job);
     BlockDriverState *bs = blk_bs(s->common.blk);
-    int64_t offset, nb_clusters;
     int ret = 0;
 
     QLIST_INIT(&s->inflight_reqs);
     qemu_co_rwlock_init(&s->flush_rwlock);
 
-    nb_clusters = DIV_ROUND_UP(s->len, s->cluster_size);
     job_progress_set_remaining(job, s->len);
 
-    s->copy_bitmap = hbitmap_alloc(nb_clusters, 0);
     if (s->sync_mode == MIRROR_SYNC_MODE_INCREMENTAL) {
         backup_incremental_init_copy_bitmap(s);
     } else {
-        hbitmap_set(s->copy_bitmap, 0, nb_clusters);
+        hbitmap_set(s->copy_bitmap, 0, s->len);
     }
 
-
     s->before_write.notify = backup_before_write_notify;
     bdrv_add_before_write_notifier(bs, &s->before_write);
 
@@ -465,68 +469,8 @@ static int coroutine_fn backup_run(Job *job, Error **errp)
              * notify callback service CoW requests. */
             job_yield(job);
         }
-    } else if (s->sync_mode == MIRROR_SYNC_MODE_INCREMENTAL) {
-        ret = backup_run_incremental(s);
     } else {
-        /* Both FULL and TOP SYNC_MODE's require copying.. */
-        for (offset = 0; offset < s->len;
-             offset += s->cluster_size) {
-            bool error_is_read;
-            int alloced = 0;
-
-            if (yield_and_check(s)) {
-                break;
-            }
-
-            if (s->sync_mode == MIRROR_SYNC_MODE_TOP) {
-                int i;
-                int64_t n;
-
-                /* Check to see if these blocks are already in the
-                 * backing file. */
-
-                for (i = 0; i < s->cluster_size;) {
-                    /* bdrv_is_allocated() only returns true/false based
-                     * on the first set of sectors it comes across that
-                     * are are all in the same state.
-                     * For that reason we must verify each sector in the
-                     * backup cluster length.  We end up copying more than
-                     * needed but at some point that is always the case. */
-                    alloced =
-                        bdrv_is_allocated(bs, offset + i,
-                                          s->cluster_size - i, &n);
-                    i += n;
-
-                    if (alloced || n == 0) {
-                        break;
-                    }
-                }
-
-                /* If the above loop never found any sectors that are in
-                 * the topmost image, skip this backup. */
-                if (alloced == 0) {
-                    continue;
-                }
-            }
-            /* FULL sync mode we copy the whole drive. */
-            if (alloced < 0) {
-                ret = alloced;
-            } else {
-                ret = backup_do_cow(s, offset, s->cluster_size,
-                                    &error_is_read, false);
-            }
-            if (ret < 0) {
-                /* Depending on error action, fail now or retry cluster */
-                BlockErrorAction action =
-                    backup_error_action(s, error_is_read, -ret);
-                if (action == BLOCK_ERROR_ACTION_REPORT) {
-                    break;
-                } else {
-                    offset -= s->cluster_size;
-                    continue;
-                }
-            }
-        }
+        ret = backup_loop(s);
     }
 
     notifier_with_return_remove(&s->before_write);
@@ -534,7 +478,6 @@ static int coroutine_fn backup_run(Job *job, Error **errp)
     /* wait until pending backup_do_cow() calls have completed */
     qemu_co_rwlock_wrlock(&s->flush_rwlock);
     qemu_co_rwlock_unlock(&s->flush_rwlock);
-    hbitmap_free(s->copy_bitmap);
 
     return ret;
 }
@@ -554,6 +497,42 @@ static const BlockJobDriver backup_job_driver = {
     .drain                  = backup_drain,
 };
 
+static int64_t backup_calculate_cluster_size(BlockDriverState *target,
+                                             Error **errp)
+{
+    int ret;
+    BlockDriverInfo bdi;
+
+    /*
+     * If there is no backing file on the target, we cannot rely on COW if our
+     * backup cluster size is smaller than the target cluster size. Even for
+     * targets with a backing file, try to avoid COW if possible.
+     */
+    ret = bdrv_get_info(target, &bdi);
+    if (ret == -ENOTSUP && !target->backing) {
+        /* Cluster size is not defined */
+        warn_report("The target block device doesn't provide "
+                    "information about the block size and it doesn't have a "
+                    "backing file. The default block size of %u bytes is "
+                    "used. If the actual block size of the target exceeds "
+                    "this default, the backup may be unusable",
+                    BACKUP_CLUSTER_SIZE_DEFAULT);
+        return BACKUP_CLUSTER_SIZE_DEFAULT;
+    } else if (ret < 0 && !target->backing) {
+        error_setg_errno(errp, -ret,
+            "Couldn't determine the cluster size of the target image, "
+            "which has no backing file");
+        error_append_hint(errp,
+            "Aborting, since this may create an unusable destination image\n");
+        return ret;
+    } else if (ret < 0 && target->backing) {
+        /* Not fatal; just trudge on ahead. */
+        return BACKUP_CLUSTER_SIZE_DEFAULT;
+    }
+
+    return MAX(BACKUP_CLUSTER_SIZE_DEFAULT, bdi.cluster_size);
+}
+
 BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
                   BlockDriverState *target, int64_t speed,
                   MirrorSyncMode sync_mode, BdrvDirtyBitmap *sync_bitmap,
@@ -565,9 +544,10 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
                   JobTxn *txn, Error **errp)
 {
     int64_t len;
-    BlockDriverInfo bdi;
     BackupBlockJob *job = NULL;
     int ret;
+    int64_t cluster_size;
+    HBitmap *copy_bitmap = NULL;
 
     assert(bs);
     assert(target);
@@ -629,6 +609,13 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
         goto error;
     }
 
+    cluster_size = backup_calculate_cluster_size(target, errp);
+    if (cluster_size < 0) {
+        goto error;
+    }
+
+    copy_bitmap = hbitmap_alloc(len, ctz32(cluster_size));
+
     /* job->len is fixed, so we can't allow resize */
     job = block_job_create(job_id, &backup_job_driver, txn, bs,
                            BLK_PERM_CONSISTENT_READ,
@@ -640,7 +627,8 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
     }
 
     /* The target must match the source in size, so no resize here either */
-    job->target = blk_new(BLK_PERM_WRITE,
+    job->target = blk_new(job->common.job.aio_context,
+                          BLK_PERM_WRITE,
                           BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE |
                           BLK_PERM_WRITE_UNCHANGED | BLK_PERM_GRAPH_MOD);
     ret = blk_insert_bs(job->target, target, errp);
@@ -657,33 +645,9 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
 
     /* Detect image-fleecing (and similar) schemes */
     job->serialize_target_writes = bdrv_chain_contains(target, bs);
-
-    /* If there is no backing file on the target, we cannot rely on COW if our
-     * backup cluster size is smaller than the target cluster size. Even for
-     * targets with a backing file, try to avoid COW if possible. */
-    ret = bdrv_get_info(target, &bdi);
-    if (ret == -ENOTSUP && !target->backing) {
-        /* Cluster size is not defined */
-        warn_report("The target block device doesn't provide "
-                    "information about the block size and it doesn't have a "
-                    "backing file. The default block size of %u bytes is "
-                    "used. If the actual block size of the target exceeds "
-                    "this default, the backup may be unusable",
-                    BACKUP_CLUSTER_SIZE_DEFAULT);
-        job->cluster_size = BACKUP_CLUSTER_SIZE_DEFAULT;
-    } else if (ret < 0 && !target->backing) {
-        error_setg_errno(errp, -ret,
-            "Couldn't determine the cluster size of the target image, "
-            "which has no backing file");
-        error_append_hint(errp,
-            "Aborting, since this may create an unusable destination image\n");
-        goto error;
-    } else if (ret < 0 && target->backing) {
-        /* Not fatal; just trudge on ahead. */
-        job->cluster_size = BACKUP_CLUSTER_SIZE_DEFAULT;
-    } else {
-        job->cluster_size = MAX(BACKUP_CLUSTER_SIZE_DEFAULT, bdi.cluster_size);
-    }
+    job->cluster_size = cluster_size;
+    job->copy_bitmap = copy_bitmap;
+    copy_bitmap = NULL;
     job->use_copy_range = true;
     job->copy_range_size = MIN_NON_ZERO(blk_get_max_transfer(job->common.blk),
                                         blk_get_max_transfer(job->target));
@@ -699,6 +663,10 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
     return &job->common;
 
  error:
+    if (copy_bitmap) {
+        assert(!job || !job->copy_bitmap);
+        hbitmap_free(copy_bitmap);
+    }
     if (sync_bitmap) {
         bdrv_reclaim_dirty_bitmap(bs, sync_bitmap, NULL);
     }
diff --git a/block/block-backend.c b/block/block-backend.c
index 4c0a8ef88d..f5d9407d20 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -42,6 +42,7 @@ struct BlockBackend {
     char *name;
     int refcnt;
     BdrvChild *root;
+    AioContext *ctx;
     DriveInfo *legacy_dinfo;    /* null unless created by drive_new() */
     QTAILQ_ENTRY(BlockBackend) link;         /* for block_backends */
     QTAILQ_ENTRY(BlockBackend) monitor_link; /* for monitor_block_backends */
@@ -322,12 +323,13 @@ static const BdrvChildRole child_root = {
  *
  * Return the new BlockBackend on success, null on failure.
  */
-BlockBackend *blk_new(uint64_t perm, uint64_t shared_perm)
+BlockBackend *blk_new(AioContext *ctx, uint64_t perm, uint64_t shared_perm)
 {
     BlockBackend *blk;
 
     blk = g_new0(BlockBackend, 1);
     blk->refcnt = 1;
+    blk->ctx = ctx;
     blk->perm = perm;
     blk->shared_perm = shared_perm;
     blk_set_enable_write_cache(blk, true);
@@ -347,6 +349,7 @@ BlockBackend *blk_new(uint64_t perm, uint64_t shared_perm)
 
 /*
  * Creates a new BlockBackend, opens a new BlockDriverState, and connects both.
+ * The new BlockBackend is in the main AioContext.
  *
  * Just as with bdrv_open(), after having called this function the reference to
  * @options belongs to the block layer (even on failure).
@@ -382,17 +385,16 @@ BlockBackend *blk_new_open(const char *filename, const char *reference,
         perm |= BLK_PERM_RESIZE;
     }
 
-    blk = blk_new(perm, BLK_PERM_ALL);
+    blk = blk_new(qemu_get_aio_context(), perm, BLK_PERM_ALL);
     bs = bdrv_open(filename, reference, options, flags, errp);
     if (!bs) {
         blk_unref(blk);
         return NULL;
     }
 
-    blk->root = bdrv_root_attach_child(bs, "root", &child_root,
+    blk->root = bdrv_root_attach_child(bs, "root", &child_root, blk->ctx,
                                        perm, BLK_PERM_ALL, blk, errp);
     if (!blk->root) {
-        bdrv_unref(bs);
         blk_unref(blk);
         return NULL;
     }
@@ -800,12 +802,12 @@ void blk_remove_bs(BlockBackend *blk)
 int blk_insert_bs(BlockBackend *blk, BlockDriverState *bs, Error **errp)
 {
     ThrottleGroupMember *tgm = &blk->public.throttle_group_member;
-    blk->root = bdrv_root_attach_child(bs, "root", &child_root,
+    bdrv_ref(bs);
+    blk->root = bdrv_root_attach_child(bs, "root", &child_root, blk->ctx,
                                        blk->perm, blk->shared_perm, blk, errp);
     if (blk->root == NULL) {
         return -EPERM;
     }
-    bdrv_ref(bs);
 
     notifier_list_notify(&blk->insert_bs_notifiers, blk);
     if (tgm->throttle_state) {
@@ -1857,7 +1859,14 @@ void blk_op_unblock_all(BlockBackend *blk, Error *reason)
 
 AioContext *blk_get_aio_context(BlockBackend *blk)
 {
-    return bdrv_get_aio_context(blk_bs(blk));
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        AioContext *ctx = bdrv_get_aio_context(blk_bs(blk));
+        assert(ctx == blk->ctx);
+    }
+
+    return blk->ctx;
 }
 
 static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb)
@@ -1866,30 +1875,37 @@ static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb)
     return blk_get_aio_context(blk_acb->blk);
 }
 
-static void blk_do_set_aio_context(BlockBackend *blk, AioContext *new_context,
-                                   bool update_root_node)
+static int blk_do_set_aio_context(BlockBackend *blk, AioContext *new_context,
+                                  bool update_root_node, Error **errp)
 {
     BlockDriverState *bs = blk_bs(blk);
     ThrottleGroupMember *tgm = &blk->public.throttle_group_member;
+    int ret;
 
     if (bs) {
+        if (update_root_node) {
+            ret = bdrv_child_try_set_aio_context(bs, new_context, blk->root,
+                                                 errp);
+            if (ret < 0) {
+                return ret;
+            }
+        }
         if (tgm->throttle_state) {
             bdrv_drained_begin(bs);
             throttle_group_detach_aio_context(tgm);
             throttle_group_attach_aio_context(tgm, new_context);
             bdrv_drained_end(bs);
         }
-        if (update_root_node) {
-            GSList *ignore = g_slist_prepend(NULL, blk->root);
-            bdrv_set_aio_context_ignore(bs, new_context, &ignore);
-            g_slist_free(ignore);
-        }
     }
+
+    blk->ctx = new_context;
+    return 0;
 }
 
-void blk_set_aio_context(BlockBackend *blk, AioContext *new_context)
+int blk_set_aio_context(BlockBackend *blk, AioContext *new_context,
+                        Error **errp)
 {
-    blk_do_set_aio_context(blk, new_context, true);
+    return blk_do_set_aio_context(blk, new_context, true, errp);
 }
 
 static bool blk_root_can_set_aio_ctx(BdrvChild *child, AioContext *ctx,
@@ -1916,7 +1932,7 @@ static void blk_root_set_aio_ctx(BdrvChild *child, AioContext *ctx,
                                  GSList **ignore)
 {
     BlockBackend *blk = child->opaque;
-    blk_do_set_aio_context(blk, ctx, false);
+    blk_do_set_aio_context(blk, ctx, false, &error_abort);
 }
 
 void blk_add_aio_context_notifier(BlockBackend *blk,
diff --git a/block/commit.c b/block/commit.c
index 14e5bb394c..c815def89a 100644
--- a/block/commit.c
+++ b/block/commit.c
@@ -301,7 +301,6 @@ void commit_start(const char *job_id, BlockDriverState *bs,
         commit_top_bs->implicit = true;
     }
     commit_top_bs->total_sectors = top->total_sectors;
-    bdrv_set_aio_context(commit_top_bs, bdrv_get_aio_context(top));
 
     bdrv_append(commit_top_bs, top, &local_err);
     if (local_err) {
@@ -338,7 +337,8 @@ void commit_start(const char *job_id, BlockDriverState *bs,
         goto fail;
     }
 
-    s->base = blk_new(BLK_PERM_CONSISTENT_READ
+    s->base = blk_new(s->common.job.aio_context,
+                      BLK_PERM_CONSISTENT_READ
                       | BLK_PERM_WRITE
                       | BLK_PERM_RESIZE,
                       BLK_PERM_CONSISTENT_READ
@@ -351,7 +351,7 @@ void commit_start(const char *job_id, BlockDriverState *bs,
     s->base_bs = base;
 
     /* Required permissions are already taken with block_job_add_bdrv() */
-    s->top = blk_new(0, BLK_PERM_ALL);
+    s->top = blk_new(s->common.job.aio_context, 0, BLK_PERM_ALL);
     ret = blk_insert_bs(s->top, top, errp);
     if (ret < 0) {
         goto fail;
@@ -395,6 +395,7 @@ int bdrv_commit(BlockDriverState *bs)
     BlockDriverState *backing_file_bs = NULL;
     BlockDriverState *commit_top_bs = NULL;
     BlockDriver *drv = bs->drv;
+    AioContext *ctx;
     int64_t offset, length, backing_length;
     int ro;
     int64_t n;
@@ -422,8 +423,9 @@ int bdrv_commit(BlockDriverState *bs)
         }
     }
 
-    src = blk_new(BLK_PERM_CONSISTENT_READ, BLK_PERM_ALL);
-    backing = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    ctx = bdrv_get_aio_context(bs);
+    src = blk_new(ctx, BLK_PERM_CONSISTENT_READ, BLK_PERM_ALL);
+    backing = blk_new(ctx, BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
 
     ret = blk_insert_bs(src, bs, &local_err);
     if (ret < 0) {
@@ -440,7 +442,6 @@ int bdrv_commit(BlockDriverState *bs)
         error_report_err(local_err);
         goto ro_cleanup;
     }
-    bdrv_set_aio_context(commit_top_bs, bdrv_get_aio_context(backing_file_bs));
 
     bdrv_set_backing_hd(commit_top_bs, backing_file_bs, &error_abort);
     bdrv_set_backing_hd(bs, commit_top_bs, &error_abort);
diff --git a/block/crypto.c b/block/crypto.c
index 3af46b805f..7351fd479d 100644
--- a/block/crypto.c
+++ b/block/crypto.c
@@ -257,7 +257,8 @@ static int block_crypto_co_create_generic(BlockDriverState *bs,
     QCryptoBlock *crypto = NULL;
     struct BlockCryptoCreateData data;
 
-    blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
 
     ret = blk_insert_bs(blk, bs, errp);
     if (ret < 0) {
diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c
index 59e6ebb861..49646a30e6 100644
--- a/block/dirty-bitmap.c
+++ b/block/dirty-bitmap.c
@@ -816,10 +816,10 @@ void bdrv_merge_dirty_bitmap(BdrvDirtyBitmap *dest, const BdrvDirtyBitmap *src,
 {
     bool ret;
 
-    /* only bitmaps from one bds are supported */
-    assert(dest->mutex == src->mutex);
-
     qemu_mutex_lock(dest->mutex);
+    if (src->mutex != dest->mutex) {
+        qemu_mutex_lock(src->mutex);
+    }
 
     if (bdrv_dirty_bitmap_check(dest, BDRV_BITMAP_DEFAULT, errp)) {
         goto out;
@@ -845,4 +845,7 @@ void bdrv_merge_dirty_bitmap(BdrvDirtyBitmap *dest, const BdrvDirtyBitmap *src,
 
 out:
     qemu_mutex_unlock(dest->mutex);
+    if (src->mutex != dest->mutex) {
+        qemu_mutex_unlock(src->mutex);
+    }
 }
diff --git a/block/io.c b/block/io.c
index 3134a60a48..9ba1bada36 100644
--- a/block/io.c
+++ b/block/io.c
@@ -422,11 +422,12 @@ static void bdrv_do_drained_end(BlockDriverState *bs, bool recursive,
         return;
     }
     assert(bs->quiesce_counter > 0);
-    old_quiesce_counter = atomic_fetch_dec(&bs->quiesce_counter);
 
     /* Re-enable things in child-to-parent order */
     bdrv_drain_invoke(bs, false);
     bdrv_parent_drained_end(bs, parent, ignore_bds_parents);
+
+    old_quiesce_counter = atomic_fetch_dec(&bs->quiesce_counter);
     if (old_quiesce_counter == 1) {
         aio_enable_external(bdrv_get_aio_context(bs));
     }
@@ -2092,6 +2093,12 @@ static int coroutine_fn bdrv_co_block_status(BlockDriverState *bs,
      */
     assert(*pnum && QEMU_IS_ALIGNED(*pnum, align) &&
            align > offset - aligned_offset);
+    if (ret & BDRV_BLOCK_RECURSE) {
+        assert(ret & BDRV_BLOCK_DATA);
+        assert(ret & BDRV_BLOCK_OFFSET_VALID);
+        assert(!(ret & BDRV_BLOCK_ZERO));
+    }
+
     *pnum -= offset - aligned_offset;
     if (*pnum > bytes) {
         *pnum = bytes;
@@ -2122,7 +2129,8 @@ static int coroutine_fn bdrv_co_block_status(BlockDriverState *bs,
         }
     }
 
-    if (want_zero && local_file && local_file != bs &&
+    if (want_zero && ret & BDRV_BLOCK_RECURSE &&
+        local_file && local_file != bs &&
         (ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO) &&
         (ret & BDRV_BLOCK_OFFSET_VALID)) {
         int64_t file_pnum;
@@ -2624,7 +2632,7 @@ int bdrv_flush(BlockDriverState *bs)
 typedef struct DiscardCo {
     BdrvChild *child;
     int64_t offset;
-    int bytes;
+    int64_t bytes;
     int ret;
 } DiscardCo;
 static void coroutine_fn bdrv_pdiscard_co_entry(void *opaque)
@@ -2635,14 +2643,15 @@ static void coroutine_fn bdrv_pdiscard_co_entry(void *opaque)
     aio_wait_kick();
 }
 
-int coroutine_fn bdrv_co_pdiscard(BdrvChild *child, int64_t offset, int bytes)
+int coroutine_fn bdrv_co_pdiscard(BdrvChild *child, int64_t offset,
+                                  int64_t bytes)
 {
     BdrvTrackedRequest req;
     int max_pdiscard, ret;
     int head, tail, align;
     BlockDriverState *bs = child->bs;
 
-    if (!bs || !bs->drv) {
+    if (!bs || !bs->drv || !bdrv_is_inserted(bs)) {
         return -ENOMEDIUM;
     }
 
@@ -2650,9 +2659,8 @@ int coroutine_fn bdrv_co_pdiscard(BdrvChild *child, int64_t offset, int bytes)
         return -EPERM;
     }
 
-    ret = bdrv_check_byte_request(bs, offset, bytes);
-    if (ret < 0) {
-        return ret;
+    if (offset < 0 || bytes < 0 || bytes > INT64_MAX - offset) {
+        return -EIO;
     }
 
     /* Do nothing if disabled.  */
@@ -2687,7 +2695,7 @@ int coroutine_fn bdrv_co_pdiscard(BdrvChild *child, int64_t offset, int bytes)
     assert(max_pdiscard >= bs->bl.request_alignment);
 
     while (bytes > 0) {
-        int num = bytes;
+        int64_t num = bytes;
 
         if (head) {
             /* Make small requests to get to alignment boundaries. */
@@ -2749,7 +2757,7 @@ out:
     return ret;
 }
 
-int bdrv_pdiscard(BdrvChild *child, int64_t offset, int bytes)
+int bdrv_pdiscard(BdrvChild *child, int64_t offset, int64_t bytes)
 {
     Coroutine *co;
     DiscardCo rwco = {
diff --git a/block/linux-aio.c b/block/linux-aio.c
index d4b61fb251..27100c2fd1 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -30,7 +30,6 @@
 #define MAX_EVENTS 128
 
 struct qemu_laiocb {
-    BlockAIOCB common;
     Coroutine *co;
     LinuxAioState *ctx;
     struct iocb iocb;
@@ -72,7 +71,7 @@ static inline ssize_t io_event_ret(struct io_event *ev)
 }
 
 /*
- * Completes an AIO request (calls the callback and frees the ACB).
+ * Completes an AIO request.
  */
 static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
 {
@@ -94,18 +93,15 @@ static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
     }
 
     laiocb->ret = ret;
-    if (laiocb->co) {
-        /* If the coroutine is already entered it must be in ioq_submit() and
-         * will notice laio->ret has been filled in when it eventually runs
-         * later.  Coroutines cannot be entered recursively so avoid doing
-         * that!
-         */
-        if (!qemu_coroutine_entered(laiocb->co)) {
-            aio_co_wake(laiocb->co);
-        }
-    } else {
-        laiocb->common.cb(laiocb->common.opaque, ret);
-        qemu_aio_unref(laiocb);
+
+    /*
+     * If the coroutine is already entered it must be in ioq_submit() and
+     * will notice laio->ret has been filled in when it eventually runs
+     * later.  Coroutines cannot be entered recursively so avoid doing
+     * that!
+     */
+    if (!qemu_coroutine_entered(laiocb->co)) {
+        aio_co_wake(laiocb->co);
     }
 }
 
@@ -273,30 +269,6 @@ static bool qemu_laio_poll_cb(void *opaque)
     return true;
 }
 
-static void laio_cancel(BlockAIOCB *blockacb)
-{
-    struct qemu_laiocb *laiocb = (struct qemu_laiocb *)blockacb;
-    struct io_event event;
-    int ret;
-
-    if (laiocb->ret != -EINPROGRESS) {
-        return;
-    }
-    ret = io_cancel(laiocb->ctx->ctx, &laiocb->iocb, &event);
-    laiocb->ret = -ECANCELED;
-    if (ret != 0) {
-        /* iocb is not cancelled, cb will be called by the event loop later */
-        return;
-    }
-
-    laiocb->common.cb(laiocb->common.opaque, laiocb->ret);
-}
-
-static const AIOCBInfo laio_aiocb_info = {
-    .aiocb_size         = sizeof(struct qemu_laiocb),
-    .cancel_async       = laio_cancel,
-};
-
 static void ioq_init(LaioQueue *io_q)
 {
     QSIMPLEQ_INIT(&io_q->pending);
@@ -431,30 +403,6 @@ int coroutine_fn laio_co_submit(BlockDriverState *bs, LinuxAioState *s, int fd,
     return laiocb.ret;
 }
 
-BlockAIOCB *laio_submit(BlockDriverState *bs, LinuxAioState *s, int fd,
-        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
-        BlockCompletionFunc *cb, void *opaque, int type)
-{
-    struct qemu_laiocb *laiocb;
-    off_t offset = sector_num * BDRV_SECTOR_SIZE;
-    int ret;
-
-    laiocb = qemu_aio_get(&laio_aiocb_info, bs, cb, opaque);
-    laiocb->nbytes = nb_sectors * BDRV_SECTOR_SIZE;
-    laiocb->ctx = s;
-    laiocb->ret = -EINPROGRESS;
-    laiocb->is_read = (type == QEMU_AIO_READ);
-    laiocb->qiov = qiov;
-
-    ret = laio_do_submit(fd, laiocb, offset, type);
-    if (ret < 0) {
-        qemu_aio_unref(laiocb);
-        return NULL;
-    }
-
-    return &laiocb->common;
-}
-
 void laio_detach_aio_context(LinuxAioState *s, AioContext *old_context)
 {
     aio_set_event_notifier(old_context, &s->e, false, NULL, NULL);
diff --git a/block/mirror.c b/block/mirror.c
index ec4bd9f404..f8bdb5b21b 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -1543,7 +1543,6 @@ static void mirror_start_job(const char *job_id, BlockDriverState *bs,
                                           BDRV_REQ_NO_FALLBACK;
     bs_opaque = g_new0(MirrorBDSOpaque, 1);
     mirror_top_bs->opaque = bs_opaque;
-    bdrv_set_aio_context(mirror_top_bs, bdrv_get_aio_context(bs));
 
     /* bdrv_append takes ownership of the mirror_top_bs reference, need to keep
      * it alive until block_job_create() succeeds even if bs has no parent. */
@@ -1584,7 +1583,8 @@ static void mirror_start_job(const char *job_id, BlockDriverState *bs,
      * We can allow anything except resize there.*/
     target_is_backing = bdrv_chain_contains(bs, target);
     target_graph_mod = (backing_mode != MIRROR_LEAVE_BACKING_CHAIN);
-    s->target = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE |
+    s->target = blk_new(s->common.job.aio_context,
+                        BLK_PERM_WRITE | BLK_PERM_RESIZE |
                         (target_graph_mod ? BLK_PERM_GRAPH_MOD : 0),
                         BLK_PERM_WRITE_UNCHANGED |
                         (target_is_backing ? BLK_PERM_CONSISTENT_READ |
diff --git a/block/parallels.c b/block/parallels.c
index 2747400577..00fae125d1 100644
--- a/block/parallels.c
+++ b/block/parallels.c
@@ -554,7 +554,8 @@ static int coroutine_fn parallels_co_create(BlockdevCreateOptions* opts,
         return -EIO;
     }
 
-    blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
     ret = blk_insert_bs(blk, bs, errp);
     if (ret < 0) {
         goto out;
diff --git a/block/qcow.c b/block/qcow.c
index 1bb8fd05e2..6dee5bb792 100644
--- a/block/qcow.c
+++ b/block/qcow.c
@@ -844,7 +844,8 @@ static int coroutine_fn qcow_co_create(BlockdevCreateOptions *opts,
         return -EIO;
     }
 
-    qcow_blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    qcow_blk = blk_new(bdrv_get_aio_context(bs),
+                       BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
     ret = blk_insert_bs(qcow_blk, bs, errp);
     if (ret < 0) {
         goto exit;
diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c
index 8a75366c92..b2487101ed 100644
--- a/block/qcow2-bitmap.c
+++ b/block/qcow2-bitmap.c
@@ -29,7 +29,6 @@
 #include "qapi/error.h"
 #include "qemu/cutils.h"
 
-#include "block/block_int.h"
 #include "qcow2.h"
 
 /* NOTICE: BME here means Bitmaps Extension and used as a namespace for
@@ -754,7 +753,7 @@ static int bitmap_list_store(BlockDriverState *bs, Qcow2BitmapList *bm_list,
         dir_offset = *offset;
     }
 
-    dir = g_try_malloc(dir_size);
+    dir = g_try_malloc0(dir_size);
     if (dir == NULL) {
         return -ENOMEM;
     }
diff --git a/block/qcow2-cache.c b/block/qcow2-cache.c
index df02e7b20a..b33bcbc984 100644
--- a/block/qcow2-cache.c
+++ b/block/qcow2-cache.c
@@ -23,7 +23,6 @@
  */
 
 #include "qemu/osdep.h"
-#include "block/block_int.h"
 #include "qemu-common.h"
 #include "qcow2.h"
 #include "trace.h"
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index b36f4aa84a..cf892f37a8 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -27,7 +27,6 @@
 
 #include "qapi/error.h"
 #include "qemu-common.h"
-#include "block/block_int.h"
 #include "qcow2.h"
 #include "qemu/bswap.h"
 #include "trace.h"
@@ -472,13 +471,12 @@ static bool coroutine_fn do_perform_cow_encrypt(BlockDriverState *bs,
 {
     if (bytes && bs->encrypted) {
         BDRVQcow2State *s = bs->opaque;
-        int64_t offset = (s->crypt_physical_offset ?
-                          (cluster_offset + offset_in_cluster) :
-                          (src_cluster_offset + offset_in_cluster));
         assert((offset_in_cluster & ~BDRV_SECTOR_MASK) == 0);
         assert((bytes & ~BDRV_SECTOR_MASK) == 0);
         assert(s->crypto);
-        if (qcrypto_block_encrypt(s->crypto, offset, buffer, bytes, NULL) < 0) {
+        if (qcow2_co_encrypt(bs, cluster_offset,
+                             src_cluster_offset + offset_in_cluster,
+                             buffer, bytes) < 0) {
             return false;
         }
     }
@@ -833,7 +831,7 @@ static int perform_cow(BlockDriverState *bs, QCowL2Meta *m)
     assert(start->offset + start->nb_bytes <= end->offset);
     assert(!m->data_qiov || m->data_qiov->size == data_bytes);
 
-    if (start->nb_bytes == 0 && end->nb_bytes == 0) {
+    if ((start->nb_bytes == 0 && end->nb_bytes == 0) || m->skip_cow) {
         return 0;
     }
 
diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index 0b09d6838b..22cadd79d5 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -25,11 +25,11 @@
 #include "qemu/osdep.h"
 #include "qapi/error.h"
 #include "qemu-common.h"
-#include "block/block_int.h"
 #include "qcow2.h"
 #include "qemu/range.h"
 #include "qemu/bswap.h"
 #include "qemu/cutils.h"
+#include "trace.h"
 
 static int64_t alloc_clusters_noref(BlockDriverState *bs, uint64_t size,
                                     uint64_t max);
@@ -738,7 +738,11 @@ void qcow2_process_discards(BlockDriverState *bs, int ret)
 
         /* Discard is optional, ignore the return value */
         if (ret >= 0) {
-            bdrv_pdiscard(bs->file, d->offset, d->bytes);
+            int r2 = bdrv_pdiscard(bs->file, d->offset, d->bytes);
+            if (r2 < 0) {
+                trace_qcow2_process_discards_failed_region(d->offset, d->bytes,
+                                                           r2);
+            }
         }
 
         g_free(d);
@@ -3445,3 +3449,35 @@ int64_t qcow2_get_last_cluster(BlockDriverState *bs, int64_t size)
                             "There are no references in the refcount table.");
     return -EIO;
 }
+
+int qcow2_detect_metadata_preallocation(BlockDriverState *bs)
+{
+    BDRVQcow2State *s = bs->opaque;
+    int64_t i, end_cluster, cluster_count = 0, threshold;
+    int64_t file_length, real_allocation, real_clusters;
+
+    file_length = bdrv_getlength(bs->file->bs);
+    if (file_length < 0) {
+        return file_length;
+    }
+
+    real_allocation = bdrv_get_allocated_file_size(bs->file->bs);
+    if (real_allocation < 0) {
+        return real_allocation;
+    }
+
+    real_clusters = real_allocation / s->cluster_size;
+    threshold = MAX(real_clusters * 10 / 9, real_clusters + 2);
+
+    end_cluster = size_to_clusters(s, file_length);
+    for (i = 0; i < end_cluster && cluster_count < threshold; i++) {
+        uint64_t refcount;
+        int ret = qcow2_get_refcount(bs, i, &refcount);
+        if (ret < 0) {
+            return ret;
+        }
+        cluster_count += !!refcount;
+    }
+
+    return cluster_count >= threshold;
+}
diff --git a/block/qcow2-snapshot.c b/block/qcow2-snapshot.c
index a6ffae89a6..d0e7fa9311 100644
--- a/block/qcow2-snapshot.c
+++ b/block/qcow2-snapshot.c
@@ -24,7 +24,6 @@
 
 #include "qemu/osdep.h"
 #include "qapi/error.h"
-#include "block/block_int.h"
 #include "qcow2.h"
 #include "qemu/bswap.h"
 #include "qemu/error-report.h"
diff --git a/block/qcow2-threads.c b/block/qcow2-threads.c
new file mode 100644
index 0000000000..3b1e63fe41
--- /dev/null
+++ b/block/qcow2-threads.c
@@ -0,0 +1,268 @@
+/*
+ * Threaded data processing for Qcow2: compression, encryption
+ *
+ * Copyright (c) 2004-2006 Fabrice Bellard
+ * Copyright (c) 2018 Virtuozzo International GmbH. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "qemu/osdep.h"
+
+#define ZLIB_CONST
+#include <zlib.h>
+
+#include "qcow2.h"
+#include "block/thread-pool.h"
+#include "crypto.h"
+
+static int coroutine_fn
+qcow2_co_process(BlockDriverState *bs, ThreadPoolFunc *func, void *arg)
+{
+    int ret;
+    BDRVQcow2State *s = bs->opaque;
+    ThreadPool *pool = aio_get_thread_pool(bdrv_get_aio_context(bs));
+
+    qemu_co_mutex_lock(&s->lock);
+    while (s->nb_threads >= QCOW2_MAX_THREADS) {
+        qemu_co_queue_wait(&s->thread_task_queue, &s->lock);
+    }
+    s->nb_threads++;
+    qemu_co_mutex_unlock(&s->lock);
+
+    ret = thread_pool_submit_co(pool, func, arg);
+
+    qemu_co_mutex_lock(&s->lock);
+    s->nb_threads--;
+    qemu_co_queue_next(&s->thread_task_queue);
+    qemu_co_mutex_unlock(&s->lock);
+
+    return ret;
+}
+
+
+/*
+ * Compression
+ */
+
+typedef ssize_t (*Qcow2CompressFunc)(void *dest, size_t dest_size,
+                                     const void *src, size_t src_size);
+typedef struct Qcow2CompressData {
+    void *dest;
+    size_t dest_size;
+    const void *src;
+    size_t src_size;
+    ssize_t ret;
+
+    Qcow2CompressFunc func;
+} Qcow2CompressData;
+
+/*
+ * qcow2_compress()
+ *
+ * @dest - destination buffer, @dest_size bytes
+ * @src - source buffer, @src_size bytes
+ *
+ * Returns: compressed size on success
+ *          -ENOMEM destination buffer is not enough to store compressed data
+ *          -EIO    on any other error
+ */
+static ssize_t qcow2_compress(void *dest, size_t dest_size,
+                              const void *src, size_t src_size)
+{
+    ssize_t ret;
+    z_stream strm;
+
+    /* best compression, small window, no zlib header */
+    memset(&strm, 0, sizeof(strm));
+    ret = deflateInit2(&strm, Z_DEFAULT_COMPRESSION, Z_DEFLATED,
+                       -12, 9, Z_DEFAULT_STRATEGY);
+    if (ret != Z_OK) {
+        return -EIO;
+    }
+
+    /*
+     * strm.next_in is not const in old zlib versions, such as those used on
+     * OpenBSD/NetBSD, so cast the const away
+     */
+    strm.avail_in = src_size;
+    strm.next_in = (void *) src;
+    strm.avail_out = dest_size;
+    strm.next_out = dest;
+
+    ret = deflate(&strm, Z_FINISH);
+    if (ret == Z_STREAM_END) {
+        ret = dest_size - strm.avail_out;
+    } else {
+        ret = (ret == Z_OK ? -ENOMEM : -EIO);
+    }
+
+    deflateEnd(&strm);
+
+    return ret;
+}
+
+/*
+ * qcow2_decompress()
+ *
+ * Decompress some data (not more than @src_size bytes) to produce exactly
+ * @dest_size bytes.
+ *
+ * @dest - destination buffer, @dest_size bytes
+ * @src - source buffer, @src_size bytes
+ *
+ * Returns: 0 on success
+ *          -1 on fail
+ */
+static ssize_t qcow2_decompress(void *dest, size_t dest_size,
+                                const void *src, size_t src_size)
+{
+    int ret = 0;
+    z_stream strm;
+
+    memset(&strm, 0, sizeof(strm));
+    strm.avail_in = src_size;
+    strm.next_in = (void *) src;
+    strm.avail_out = dest_size;
+    strm.next_out = dest;
+
+    ret = inflateInit2(&strm, -12);
+    if (ret != Z_OK) {
+        return -1;
+    }
+
+    ret = inflate(&strm, Z_FINISH);
+    if ((ret != Z_STREAM_END && ret != Z_BUF_ERROR) || strm.avail_out != 0) {
+        /*
+         * We approve Z_BUF_ERROR because we need @dest buffer to be filled, but
+         * @src buffer may be processed partly (because in qcow2 we know size of
+         * compressed data with precision of one sector)
+         */
+        ret = -1;
+    }
+
+    inflateEnd(&strm);
+
+    return ret;
+}
+
+static int qcow2_compress_pool_func(void *opaque)
+{
+    Qcow2CompressData *data = opaque;
+
+    data->ret = data->func(data->dest, data->dest_size,
+                           data->src, data->src_size);
+
+    return 0;
+}
+
+static ssize_t coroutine_fn
+qcow2_co_do_compress(BlockDriverState *bs, void *dest, size_t dest_size,
+                     const void *src, size_t src_size, Qcow2CompressFunc func)
+{
+    Qcow2CompressData arg = {
+        .dest = dest,
+        .dest_size = dest_size,
+        .src = src,
+        .src_size = src_size,
+        .func = func,
+    };
+
+    qcow2_co_process(bs, qcow2_compress_pool_func, &arg);
+
+    return arg.ret;
+}
+
+ssize_t coroutine_fn
+qcow2_co_compress(BlockDriverState *bs, void *dest, size_t dest_size,
+                  const void *src, size_t src_size)
+{
+    return qcow2_co_do_compress(bs, dest, dest_size, src, src_size,
+                                qcow2_compress);
+}
+
+ssize_t coroutine_fn
+qcow2_co_decompress(BlockDriverState *bs, void *dest, size_t dest_size,
+                    const void *src, size_t src_size)
+{
+    return qcow2_co_do_compress(bs, dest, dest_size, src, src_size,
+                                qcow2_decompress);
+}
+
+
+/*
+ * Cryptography
+ */
+
+/*
+ * Qcow2EncDecFunc: common prototype of qcrypto_block_encrypt() and
+ * qcrypto_block_decrypt() functions.
+ */
+typedef int (*Qcow2EncDecFunc)(QCryptoBlock *block, uint64_t offset,
+                               uint8_t *buf, size_t len, Error **errp);
+
+typedef struct Qcow2EncDecData {
+    QCryptoBlock *block;
+    uint64_t offset;
+    uint8_t *buf;
+    size_t len;
+
+    Qcow2EncDecFunc func;
+} Qcow2EncDecData;
+
+static int qcow2_encdec_pool_func(void *opaque)
+{
+    Qcow2EncDecData *data = opaque;
+
+    return data->func(data->block, data->offset, data->buf, data->len, NULL);
+}
+
+static int coroutine_fn
+qcow2_co_encdec(BlockDriverState *bs, uint64_t file_cluster_offset,
+                  uint64_t offset, void *buf, size_t len, Qcow2EncDecFunc func)
+{
+    BDRVQcow2State *s = bs->opaque;
+    Qcow2EncDecData arg = {
+        .block = s->crypto,
+        .offset = s->crypt_physical_offset ?
+                      file_cluster_offset + offset_into_cluster(s, offset) :
+                      offset,
+        .buf = buf,
+        .len = len,
+        .func = func,
+    };
+
+    return qcow2_co_process(bs, qcow2_encdec_pool_func, &arg);
+}
+
+int coroutine_fn
+qcow2_co_encrypt(BlockDriverState *bs, uint64_t file_cluster_offset,
+                 uint64_t offset, void *buf, size_t len)
+{
+    return qcow2_co_encdec(bs, file_cluster_offset, offset, buf, len,
+                             qcrypto_block_encrypt);
+}
+
+int coroutine_fn
+qcow2_co_decrypt(BlockDriverState *bs, uint64_t file_cluster_offset,
+                 uint64_t offset, void *buf, size_t len)
+{
+    return qcow2_co_encdec(bs, file_cluster_offset, offset, buf, len,
+                             qcrypto_block_decrypt);
+}
diff --git a/block/qcow2.c b/block/qcow2.c
index d39882785d..9396d490d5 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -24,10 +24,6 @@
 
 #include "qemu/osdep.h"
 
-#define ZLIB_CONST
-#include <zlib.h>
-
-#include "block/block_int.h"
 #include "block/qdict.h"
 #include "sysemu/block-backend.h"
 #include "qemu/module.h"
@@ -44,7 +40,6 @@
 #include "qapi/qobject-input-visitor.h"
 #include "qapi/qapi-visit-block-core.h"
 #include "crypto.h"
-#include "block/thread-pool.h"
 
 /*
   Differences with QCOW:
@@ -302,7 +297,7 @@ static int qcow2_read_extensions(BlockDriverState *bs, uint64_t start_offset,
             }
             s->crypto = qcrypto_block_open(s->crypto_opts, "encrypt.",
                                            qcow2_crypto_hdr_read_func,
-                                           bs, cflags, 1, errp);
+                                           bs, cflags, QCOW2_MAX_THREADS, errp);
             if (!s->crypto) {
                 return -EINVAL;
             }
@@ -1543,7 +1538,8 @@ static int coroutine_fn qcow2_do_open(BlockDriverState *bs, QDict *options,
                 cflags |= QCRYPTO_BLOCK_OPEN_NO_IO;
             }
             s->crypto = qcrypto_block_open(s->crypto_opts, "encrypt.",
-                                           NULL, NULL, cflags, 1, errp);
+                                           NULL, NULL, cflags,
+                                           QCOW2_MAX_THREADS, errp);
             if (!s->crypto) {
                 ret = -EINVAL;
                 goto fail;
@@ -1698,7 +1694,7 @@ static int coroutine_fn qcow2_do_open(BlockDriverState *bs, QDict *options,
     }
 #endif
 
-    qemu_co_queue_init(&s->compress_wait_queue);
+    qemu_co_queue_init(&s->thread_task_queue);
 
     return ret;
 
@@ -1899,6 +1895,12 @@ static int coroutine_fn qcow2_co_block_status(BlockDriverState *bs,
     unsigned int bytes;
     int status = 0;
 
+    if (!s->metadata_preallocation_checked) {
+        ret = qcow2_detect_metadata_preallocation(bs);
+        s->metadata_preallocation = (ret == 1);
+        s->metadata_preallocation_checked = true;
+    }
+
     bytes = MIN(INT_MAX, count);
     qemu_co_mutex_lock(&s->lock);
     ret = qcow2_get_cluster_offset(bs, offset, &bytes, &cluster_offset);
@@ -1921,6 +1923,11 @@ static int coroutine_fn qcow2_co_block_status(BlockDriverState *bs,
     } else if (ret != QCOW2_CLUSTER_UNALLOCATED) {
         status |= BDRV_BLOCK_DATA;
     }
+    if (s->metadata_preallocation && (status & BDRV_BLOCK_DATA) &&
+        (status & BDRV_BLOCK_OFFSET_VALID))
+    {
+        status |= BDRV_BLOCK_RECURSE;
+    }
     return status;
 }
 
@@ -1974,8 +1981,6 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
 
     qemu_iovec_init(&hd_qiov, qiov->niov);
 
-    qemu_co_mutex_lock(&s->lock);
-
     while (bytes != 0) {
 
         /* prepare next request */
@@ -1985,7 +1990,9 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
                             QCOW_MAX_CRYPT_CLUSTERS * s->cluster_size);
         }
 
+        qemu_co_mutex_lock(&s->lock);
         ret = qcow2_get_cluster_offset(bs, offset, &cur_bytes, &cluster_offset);
+        qemu_co_mutex_unlock(&s->lock);
         if (ret < 0) {
             goto fail;
         }
@@ -2000,10 +2007,8 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
 
             if (bs->backing) {
                 BLKDBG_EVENT(bs->file, BLKDBG_READ_BACKING_AIO);
-                qemu_co_mutex_unlock(&s->lock);
                 ret = bdrv_co_preadv(bs->backing, offset, cur_bytes,
                                      &hd_qiov, 0);
-                qemu_co_mutex_lock(&s->lock);
                 if (ret < 0) {
                     goto fail;
                 }
@@ -2019,11 +2024,9 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
             break;
 
         case QCOW2_CLUSTER_COMPRESSED:
-            qemu_co_mutex_unlock(&s->lock);
             ret = qcow2_co_preadv_compressed(bs, cluster_offset,
                                              offset, cur_bytes,
                                              &hd_qiov);
-            qemu_co_mutex_lock(&s->lock);
             if (ret < 0) {
                 goto fail;
             }
@@ -2060,11 +2063,9 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
             }
 
             BLKDBG_EVENT(bs->file, BLKDBG_READ_AIO);
-            qemu_co_mutex_unlock(&s->lock);
             ret = bdrv_co_preadv(s->data_file,
                                  cluster_offset + offset_in_cluster,
                                  cur_bytes, &hd_qiov, 0);
-            qemu_co_mutex_lock(&s->lock);
             if (ret < 0) {
                 goto fail;
             }
@@ -2072,13 +2073,8 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
                 assert(s->crypto);
                 assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
                 assert((cur_bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
-                if (qcrypto_block_decrypt(s->crypto,
-                                          (s->crypt_physical_offset ?
-                                           cluster_offset + offset_in_cluster :
-                                           offset),
-                                          cluster_data,
-                                          cur_bytes,
-                                          NULL) < 0) {
+                if (qcow2_co_decrypt(bs, cluster_offset, offset,
+                                     cluster_data, cur_bytes) < 0) {
                     ret = -EIO;
                     goto fail;
                 }
@@ -2099,8 +2095,6 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
     ret = 0;
 
 fail:
-    qemu_co_mutex_unlock(&s->lock);
-
     qemu_iovec_destroy(&hd_qiov);
     qemu_vfree(cluster_data);
 
@@ -2120,6 +2114,11 @@ static bool merge_cow(uint64_t offset, unsigned bytes,
             continue;
         }
 
+        /* If COW regions are handled already, skip this too */
+        if (m->skip_cow) {
+            continue;
+        }
+
         /* The data (middle) region must be immediately after the
          * start region */
         if (l2meta_cow_start(m) + m->cow_start.nb_bytes != offset) {
@@ -2145,6 +2144,80 @@ static bool merge_cow(uint64_t offset, unsigned bytes,
     return false;
 }
 
+static bool is_unallocated(BlockDriverState *bs, int64_t offset, int64_t bytes)
+{
+    int64_t nr;
+    return !bytes ||
+        (!bdrv_is_allocated_above(bs, NULL, offset, bytes, &nr) && nr == bytes);
+}
+
+static bool is_zero_cow(BlockDriverState *bs, QCowL2Meta *m)
+{
+    /*
+     * This check is designed for optimization shortcut so it must be
+     * efficient.
+     * Instead of is_zero(), use is_unallocated() as it is faster (but not
+     * as accurate and can result in false negatives).
+     */
+    return is_unallocated(bs, m->offset + m->cow_start.offset,
+                          m->cow_start.nb_bytes) &&
+           is_unallocated(bs, m->offset + m->cow_end.offset,
+                          m->cow_end.nb_bytes);
+}
+
+static int handle_alloc_space(BlockDriverState *bs, QCowL2Meta *l2meta)
+{
+    BDRVQcow2State *s = bs->opaque;
+    QCowL2Meta *m;
+
+    if (!(s->data_file->bs->supported_zero_flags & BDRV_REQ_NO_FALLBACK)) {
+        return 0;
+    }
+
+    if (bs->encrypted) {
+        return 0;
+    }
+
+    for (m = l2meta; m != NULL; m = m->next) {
+        int ret;
+
+        if (!m->cow_start.nb_bytes && !m->cow_end.nb_bytes) {
+            continue;
+        }
+
+        if (!is_zero_cow(bs, m)) {
+            continue;
+        }
+
+        /*
+         * instead of writing zero COW buffers,
+         * efficiently zero out the whole clusters
+         */
+
+        ret = qcow2_pre_write_overlap_check(bs, 0, m->alloc_offset,
+                                            m->nb_clusters * s->cluster_size,
+                                            true);
+        if (ret < 0) {
+            return ret;
+        }
+
+        BLKDBG_EVENT(bs->file, BLKDBG_CLUSTER_ALLOC_SPACE);
+        ret = bdrv_co_pwrite_zeroes(s->data_file, m->alloc_offset,
+                                    m->nb_clusters * s->cluster_size,
+                                    BDRV_REQ_NO_FALLBACK);
+        if (ret < 0) {
+            if (ret != -ENOTSUP && ret != -EAGAIN) {
+                return ret;
+            }
+            continue;
+        }
+
+        trace_qcow2_skip_cow(qemu_coroutine_self(), m->offset, m->nb_clusters);
+        m->skip_cow = true;
+    }
+    return 0;
+}
+
 static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
                                          uint64_t bytes, QEMUIOVector *qiov,
                                          int flags)
@@ -2181,11 +2254,20 @@ static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
         ret = qcow2_alloc_cluster_offset(bs, offset, &cur_bytes,
                                          &cluster_offset, &l2meta);
         if (ret < 0) {
-            goto fail;
+            goto out_locked;
         }
 
         assert((cluster_offset & 511) == 0);
 
+        ret = qcow2_pre_write_overlap_check(bs, 0,
+                                            cluster_offset + offset_in_cluster,
+                                            cur_bytes, true);
+        if (ret < 0) {
+            goto out_locked;
+        }
+
+        qemu_co_mutex_unlock(&s->lock);
+
         qemu_iovec_reset(&hd_qiov);
         qemu_iovec_concat(&hd_qiov, qiov, bytes_done, cur_bytes);
 
@@ -2197,7 +2279,7 @@ static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
                                                    * s->cluster_size);
                 if (cluster_data == NULL) {
                     ret = -ENOMEM;
-                    goto fail;
+                    goto out_unlocked;
                 }
             }
 
@@ -2205,24 +2287,20 @@ static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
                    QCOW_MAX_CRYPT_CLUSTERS * s->cluster_size);
             qemu_iovec_to_buf(&hd_qiov, 0, cluster_data, hd_qiov.size);
 
-            if (qcrypto_block_encrypt(s->crypto,
-                                      (s->crypt_physical_offset ?
-                                       cluster_offset + offset_in_cluster :
-                                       offset),
-                                      cluster_data,
-                                      cur_bytes, NULL) < 0) {
+            if (qcow2_co_encrypt(bs, cluster_offset, offset,
+                                 cluster_data, cur_bytes) < 0) {
                 ret = -EIO;
-                goto fail;
+                goto out_unlocked;
             }
 
             qemu_iovec_reset(&hd_qiov);
             qemu_iovec_add(&hd_qiov, cluster_data, cur_bytes);
         }
 
-        ret = qcow2_pre_write_overlap_check(bs, 0,
-                cluster_offset + offset_in_cluster, cur_bytes, true);
+        /* Try to efficiently initialize the physical space with zeroes */
+        ret = handle_alloc_space(bs, l2meta);
         if (ret < 0) {
-            goto fail;
+            goto out_unlocked;
         }
 
         /* If we need to do COW, check if it's possible to merge the
@@ -2230,22 +2308,22 @@ static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
          * If it's not possible (or not necessary) then write the
          * guest data now. */
         if (!merge_cow(offset, cur_bytes, &hd_qiov, l2meta)) {
-            qemu_co_mutex_unlock(&s->lock);
             BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
             trace_qcow2_writev_data(qemu_coroutine_self(),
                                     cluster_offset + offset_in_cluster);
             ret = bdrv_co_pwritev(s->data_file,
                                   cluster_offset + offset_in_cluster,
                                   cur_bytes, &hd_qiov, 0);
-            qemu_co_mutex_lock(&s->lock);
             if (ret < 0) {
-                goto fail;
+                goto out_unlocked;
             }
         }
 
+        qemu_co_mutex_lock(&s->lock);
+
         ret = qcow2_handle_l2meta(bs, &l2meta, true);
         if (ret) {
-            goto fail;
+            goto out_locked;
         }
 
         bytes -= cur_bytes;
@@ -2254,8 +2332,12 @@ static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
         trace_qcow2_writev_done_part(qemu_coroutine_self(), cur_bytes);
     }
     ret = 0;
+    goto out_locked;
 
-fail:
+out_unlocked:
+    qemu_co_mutex_lock(&s->lock);
+
+out_locked:
     qcow2_handle_l2meta(bs, &l2meta, false);
 
     qemu_co_mutex_unlock(&s->lock);
@@ -3092,7 +3174,8 @@ qcow2_co_create(BlockdevCreateOptions *create_options, Error **errp)
     }
 
     /* Create BlockBackend to write to the image */
-    blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
     ret = blk_insert_bs(blk, bs, errp);
     if (ret < 0) {
         goto out;
@@ -3921,171 +4004,6 @@ fail:
     return ret;
 }
 
-/*
- * qcow2_compress()
- *
- * @dest - destination buffer, @dest_size bytes
- * @src - source buffer, @src_size bytes
- *
- * Returns: compressed size on success
- *          -ENOMEM destination buffer is not enough to store compressed data
- *          -EIO    on any other error
- */
-static ssize_t qcow2_compress(void *dest, size_t dest_size,
-                              const void *src, size_t src_size)
-{
-    ssize_t ret;
-    z_stream strm;
-
-    /* best compression, small window, no zlib header */
-    memset(&strm, 0, sizeof(strm));
-    ret = deflateInit2(&strm, Z_DEFAULT_COMPRESSION, Z_DEFLATED,
-                       -12, 9, Z_DEFAULT_STRATEGY);
-    if (ret != Z_OK) {
-        return -EIO;
-    }
-
-    /* strm.next_in is not const in old zlib versions, such as those used on
-     * OpenBSD/NetBSD, so cast the const away */
-    strm.avail_in = src_size;
-    strm.next_in = (void *) src;
-    strm.avail_out = dest_size;
-    strm.next_out = dest;
-
-    ret = deflate(&strm, Z_FINISH);
-    if (ret == Z_STREAM_END) {
-        ret = dest_size - strm.avail_out;
-    } else {
-        ret = (ret == Z_OK ? -ENOMEM : -EIO);
-    }
-
-    deflateEnd(&strm);
-
-    return ret;
-}
-
-/*
- * qcow2_decompress()
- *
- * Decompress some data (not more than @src_size bytes) to produce exactly
- * @dest_size bytes.
- *
- * @dest - destination buffer, @dest_size bytes
- * @src - source buffer, @src_size bytes
- *
- * Returns: 0 on success
- *          -1 on fail
- */
-static ssize_t qcow2_decompress(void *dest, size_t dest_size,
-                                const void *src, size_t src_size)
-{
-    int ret = 0;
-    z_stream strm;
-
-    memset(&strm, 0, sizeof(strm));
-    strm.avail_in = src_size;
-    strm.next_in = (void *) src;
-    strm.avail_out = dest_size;
-    strm.next_out = dest;
-
-    ret = inflateInit2(&strm, -12);
-    if (ret != Z_OK) {
-        return -1;
-    }
-
-    ret = inflate(&strm, Z_FINISH);
-    if ((ret != Z_STREAM_END && ret != Z_BUF_ERROR) || strm.avail_out != 0) {
-        /* We approve Z_BUF_ERROR because we need @dest buffer to be filled, but
-         * @src buffer may be processed partly (because in qcow2 we know size of
-         * compressed data with precision of one sector) */
-        ret = -1;
-    }
-
-    inflateEnd(&strm);
-
-    return ret;
-}
-
-#define MAX_COMPRESS_THREADS 4
-
-typedef ssize_t (*Qcow2CompressFunc)(void *dest, size_t dest_size,
-                                     const void *src, size_t src_size);
-typedef struct Qcow2CompressData {
-    void *dest;
-    size_t dest_size;
-    const void *src;
-    size_t src_size;
-    ssize_t ret;
-
-    Qcow2CompressFunc func;
-} Qcow2CompressData;
-
-static int qcow2_compress_pool_func(void *opaque)
-{
-    Qcow2CompressData *data = opaque;
-
-    data->ret = data->func(data->dest, data->dest_size,
-                           data->src, data->src_size);
-
-    return 0;
-}
-
-static void qcow2_compress_complete(void *opaque, int ret)
-{
-    qemu_coroutine_enter(opaque);
-}
-
-static ssize_t coroutine_fn
-qcow2_co_do_compress(BlockDriverState *bs, void *dest, size_t dest_size,
-                     const void *src, size_t src_size, Qcow2CompressFunc func)
-{
-    BDRVQcow2State *s = bs->opaque;
-    BlockAIOCB *acb;
-    ThreadPool *pool = aio_get_thread_pool(bdrv_get_aio_context(bs));
-    Qcow2CompressData arg = {
-        .dest = dest,
-        .dest_size = dest_size,
-        .src = src,
-        .src_size = src_size,
-        .func = func,
-    };
-
-    while (s->nb_compress_threads >= MAX_COMPRESS_THREADS) {
-        qemu_co_queue_wait(&s->compress_wait_queue, NULL);
-    }
-
-    s->nb_compress_threads++;
-    acb = thread_pool_submit_aio(pool, qcow2_compress_pool_func, &arg,
-                                 qcow2_compress_complete,
-                                 qemu_coroutine_self());
-
-    if (!acb) {
-        s->nb_compress_threads--;
-        return -EINVAL;
-    }
-    qemu_coroutine_yield();
-    s->nb_compress_threads--;
-    qemu_co_queue_next(&s->compress_wait_queue);
-
-    return arg.ret;
-}
-
-static ssize_t coroutine_fn
-qcow2_co_compress(BlockDriverState *bs, void *dest, size_t dest_size,
-                  const void *src, size_t src_size)
-{
-    return qcow2_co_do_compress(bs, dest, dest_size, src, src_size,
-                                qcow2_compress);
-}
-
-static ssize_t coroutine_fn
-qcow2_co_decompress(BlockDriverState *bs, void *dest, size_t dest_size,
-                    const void *src, size_t src_size)
-{
-    return qcow2_co_do_compress(bs, dest, dest_size, src, src_size,
-                                qcow2_decompress);
-}
-
 /* XXX: put compressed sectors first, then all the cluster aligned
    tables to avoid losing bytes in alignment */
 static coroutine_fn int
@@ -5089,7 +5007,8 @@ static int qcow2_amend_options(BlockDriverState *bs, QemuOpts *opts,
     }
 
     if (new_size) {
-        BlockBackend *blk = blk_new(BLK_PERM_RESIZE, BLK_PERM_ALL);
+        BlockBackend *blk = blk_new(bdrv_get_aio_context(bs),
+                                    BLK_PERM_RESIZE, BLK_PERM_ALL);
         ret = blk_insert_bs(blk, bs, errp);
         if (ret < 0) {
             blk_unref(blk);
diff --git a/block/qcow2.h b/block/qcow2.h
index 8d92ef1fee..fc1b0d3c1e 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -28,6 +28,7 @@
 #include "crypto/block.h"
 #include "qemu/coroutine.h"
 #include "qemu/units.h"
+#include "block/block_int.h"
 
 //#define DEBUG_ALLOC
 //#define DEBUG_ALLOC2
@@ -267,6 +268,8 @@ typedef struct Qcow2BitmapHeaderExt {
     uint64_t bitmap_directory_offset;
 } QEMU_PACKED Qcow2BitmapHeaderExt;
 
+#define QCOW2_MAX_THREADS 4
+
 typedef struct BDRVQcow2State {
     int cluster_bits;
     int cluster_size;
@@ -349,10 +352,13 @@ typedef struct BDRVQcow2State {
     char *image_backing_format;
     char *image_data_file;
 
-    CoQueue compress_wait_queue;
-    int nb_compress_threads;
+    CoQueue thread_task_queue;
+    int nb_threads;
 
     BdrvChild *data_file;
+
+    bool metadata_preallocation_checked;
+    bool metadata_preallocation;
 } BDRVQcow2State;
 
 typedef struct Qcow2COWRegion {
@@ -402,6 +408,12 @@ typedef struct QCowL2Meta
      */
     Qcow2COWRegion cow_end;
 
+    /*
+     * Indicates that COW regions are already handled and do not require
+     * any more processing.
+     */
+    bool skip_cow;
+
     /**
      * The I/O vector with the data from the actual guest write request.
      * If non-NULL, this is meant to be merged together with the data
@@ -646,6 +658,7 @@ int qcow2_change_refcount_order(BlockDriverState *bs, int refcount_order,
                                 void *cb_opaque, Error **errp);
 int qcow2_shrink_reftable(BlockDriverState *bs);
 int64_t qcow2_get_last_cluster(BlockDriverState *bs, int64_t size);
+int qcow2_detect_metadata_preallocation(BlockDriverState *bs);
 
 /* qcow2-cluster.c functions */
 int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
@@ -737,4 +750,17 @@ void qcow2_remove_persistent_dirty_bitmap(BlockDriverState *bs,
                                           const char *name,
                                           Error **errp);
 
+ssize_t coroutine_fn
+qcow2_co_compress(BlockDriverState *bs, void *dest, size_t dest_size,
+                  const void *src, size_t src_size);
+ssize_t coroutine_fn
+qcow2_co_decompress(BlockDriverState *bs, void *dest, size_t dest_size,
+                    const void *src, size_t src_size);
+int coroutine_fn
+qcow2_co_encrypt(BlockDriverState *bs, uint64_t file_cluster_offset,
+                 uint64_t offset, void *buf, size_t len);
+int coroutine_fn
+qcow2_co_decrypt(BlockDriverState *bs, uint64_t file_cluster_offset,
+                 uint64_t offset, void *buf, size_t len);
+
 #endif
diff --git a/block/qed.c b/block/qed.c
index dcdcd62b4a..bb4f5c9863 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -649,7 +649,8 @@ static int coroutine_fn bdrv_qed_co_create(BlockdevCreateOptions *opts,
         return -EIO;
     }
 
-    blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
     ret = blk_insert_bs(blk, bs, errp);
     if (ret < 0) {
         goto out;
diff --git a/block/quorum.c b/block/quorum.c
index 352f729136..133ee18204 100644
--- a/block/quorum.c
+++ b/block/quorum.c
@@ -1019,7 +1019,6 @@ static void quorum_add_child(BlockDriverState *bs, BlockDriverState *child_bs,
     child = bdrv_attach_child(bs, child_bs, indexstr, &child_format, errp);
     if (child == NULL) {
         s->next_child_index--;
-        bdrv_unref(child_bs);
         goto out;
     }
     s->children = g_renew(BdrvChild *, s->children, s->num_children + 1);
diff --git a/block/sheepdog.c b/block/sheepdog.c
index cbdfe9ab6e..f76d6ddbbc 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -1800,7 +1800,8 @@ static int sd_prealloc(BlockDriverState *bs, int64_t old_size, int64_t new_size,
     void *buf = NULL;
     int ret;
 
-    blk = blk_new(BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE | BLK_PERM_RESIZE,
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE | BLK_PERM_RESIZE,
                   BLK_PERM_ALL);
 
     ret = blk_insert_bs(blk, bs, errp);
diff --git a/block/trace-events b/block/trace-events
index 79ccd8d824..eab51497fc 100644
--- a/block/trace-events
+++ b/block/trace-events
@@ -68,6 +68,7 @@ qcow2_writev_done_part(void *co, int cur_bytes) "co %p cur_bytes %d"
 qcow2_writev_data(void *co, uint64_t offset) "co %p offset 0x%" PRIx64
 qcow2_pwrite_zeroes_start_req(void *co, int64_t offset, int count) "co %p offset 0x%" PRIx64 " count %d"
 qcow2_pwrite_zeroes(void *co, int64_t offset, int count) "co %p offset 0x%" PRIx64 " count %d"
+qcow2_skip_cow(void *co, uint64_t offset, int nb_clusters) "co %p offset 0x%" PRIx64 " nb_clusters %d"
 
 # qcow2-cluster.c
 qcow2_alloc_clusters_offset(void *co, uint64_t offset, int bytes) "co %p offset 0x%" PRIx64 " bytes %d"
@@ -91,6 +92,9 @@ qcow2_cache_get_done(void *co, int c, int i) "co %p is_l2_cache %d index %d"
 qcow2_cache_flush(void *co, int c) "co %p is_l2_cache %d"
 qcow2_cache_entry_flush(void *co, int c, int i) "co %p is_l2_cache %d index %d"
 
+# qcow2-refcount.c
+qcow2_process_discards_failed_region(uint64_t offset, uint64_t bytes, int ret) "offset 0x%" PRIx64 " bytes 0x%" PRIx64 " ret %d"
+
 # qed-l2-cache.c
 qed_alloc_l2_cache_entry(void *l2_cache, void *entry) "l2_cache %p entry %p"
 qed_unref_l2_cache_entry(void *entry, int ref) "entry %p ref %d"
diff --git a/block/vdi.c b/block/vdi.c
index d7ef6628e7..b9845a4cbd 100644
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -803,7 +803,8 @@ static int coroutine_fn vdi_co_do_create(BlockdevCreateOptions *create_options,
         goto exit;
     }
 
-    blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    blk = blk_new(bdrv_get_aio_context(bs_file),
+                  BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
     ret = blk_insert_bs(blk, bs_file, errp);
     if (ret < 0) {
         goto exit;
diff --git a/block/vhdx.c b/block/vhdx.c
index a143a57657..d6070b6fa8 100644
--- a/block/vhdx.c
+++ b/block/vhdx.c
@@ -1900,7 +1900,8 @@ static int coroutine_fn vhdx_co_create(BlockdevCreateOptions *opts,
         return -EIO;
     }
 
-    blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
     ret = blk_insert_bs(blk, bs, errp);
     if (ret < 0) {
         goto delete_and_exit;
diff --git a/block/vmdk.c b/block/vmdk.c
index de8cb859f8..51067c774f 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -2356,7 +2356,8 @@ static BlockBackend *vmdk_co_create_cb(int64_t size, int idx,
     if (!bs) {
         return NULL;
     }
-    blk = blk_new(BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE | BLK_PERM_RESIZE,
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE | BLK_PERM_RESIZE,
                   BLK_PERM_ALL);
     if (blk_insert_bs(blk, bs, errp)) {
         bdrv_unref(bs);
diff --git a/block/vpc.c b/block/vpc.c
index 0c279b87c8..d4776ee8a5 100644
--- a/block/vpc.c
+++ b/block/vpc.c
@@ -1011,7 +1011,8 @@ static int coroutine_fn vpc_co_create(BlockdevCreateOptions *opts,
         return -EIO;
     }
 
-    blk = blk_new(BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
+    blk = blk_new(bdrv_get_aio_context(bs),
+                  BLK_PERM_WRITE | BLK_PERM_RESIZE, BLK_PERM_ALL);
     ret = blk_insert_bs(blk, bs, errp);
     if (ret < 0) {
         goto out;