On 09/09/2016 07:31 AM, Denis V. Lunev wrote: > bdrv_is_allocated_above() returns true in the case even for completel
s/completel/completely/
> zeroed areas as BDRV_BLOCK_ALLOCATED flag is set in both cases.
>
> The patch stops using bdrv_is_allocated_above() wrapper and switches to
> bdrv_get_block_status_above() to distinguish zeroed areas and areas with
> data to avoid extra IO operations if possible.
>
> Signed-off-by: Denis V. Lunev <d...@openvz.org>
> CC: Stefan Hajnoczi <stefa...@redhat.com>
> CC: Fam Zheng <f...@redhat.com>
> CC: Kevin Wolf <kw...@redhat.com>
> CC: Max Reitz <mre...@redhat.com>
> CC: Jeff Cody <jc...@redhat.com>
> ---
> block/mirror.c | 18 ++++++++++++------
> 1 file changed, 12 insertions(+), 6 deletions(-)
>
> @@ -597,13 +602,14 @@ static int coroutine_fn
> mirror_dirty_init(MirrorBlockJob *s)
> return 0;
> }
>
> - ret = bdrv_is_allocated_above(bs, base, sector_num, nb_sectors, &n);
> - if (ret < 0) {
> - return ret;
> + status = bdrv_get_block_status_above(bs, base, sector_num,
> + nb_sectors, &n, &file);
Eventually, we should probably fix bdrv_get_block_status_above() to be
byte-based, but that's not a problem with this patch.
Looks okay to me, but I haven't thought closely enough about potential
corner cases to feel comfortable with giving R-b yet...
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
