Re: [PATCH v4 09/12] block/block-copy: merge the two backup scan steps into block_copy_prescan()

[Eric Blake <eblake@redhat.com>]

On Fri, Oct 30, 2026 at 02:06:09AM +0100, Denis V. Lunev wrote:
> From: Denis V. Lunev <den@openvz.org>
> 
> block_copy_reset_unallocated() and block_copy_calculate_zero_bitmap()
> are both one step of the same scan and differ only in what they do with
> the answer. Fold them into block_copy_prescan(), which drops unallocated
> clusters from copy_bitmap while skip_unallocated is set, so
> backup_scan_source() loses its per-mode branch.
> 
> sync=top now walks the dirty areas the way full and bitmap do, so it no
> longer queries what copy-before-write has already copied during the
> scan. A failed query still fails sync=top, as nothing else can tell
> which clusters to skip, and still leaves the area to be copied as data
> otherwise.
> 
> Suggested-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
> Signed-off-by: Denis V. Lunev <den@openvz.org>
> CC: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
> CC: John Snow <jsnow@redhat.com>
> CC: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>
> ---
>  block/backup.c             |  7 +----
>  block/block-copy.c         | 63 ++++++++++++++------------------------
>  include/block/block-copy.h |  9 ++----
>  3 files changed, 27 insertions(+), 52 deletions(-)

Reviewed-by: Eric Blake <eblake@redhat.com>

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.
Virtualization:  qemu.org | libguestfs.org