vmdk: Don't update L2 table for zero write on zero cluster

If a cluster is already zeroed, we don't have to call vmdk_L2update(),
which is rather slow because it flushes the image file.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200430133007.170335-5-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block/vmdk.c

@@ -2013,7 +2013,7 @@ static int vmdk_pwritev(BlockDriverState *bs, uint64_t offset,
                     offset_in_cluster == 0 &&
                     n_bytes >= extent->cluster_sectors * BDRV_SECTOR_SIZE) {
                 n_bytes = extent->cluster_sectors * BDRV_SECTOR_SIZE;
-                if (!zero_dry_run) {
+                if (!zero_dry_run && ret != VMDK_ZEROED) {
                     /* update L2 tables */
                     if (vmdk_L2update(extent, &m_data, VMDK_GTE_ZEROED)
                             != VMDK_OK) {