Hmm, now I think, that instead of this, it is better to use pointer as
parent id for
nod-bds parents, to be sure they will not intersect. And add special
field to
qapi block relation info "parent-description" for such parents.
Also I'm afraid that this patch may break iotests..
16.08.2018 20:20, Vladimir Sementsov-Ogievskiy wrote:
> Make blk_root_get_parent_desc return different descriptions for
> different blk's in worst case.
>
> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
> ---
> block/block-backend.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/block/block-backend.c b/block/block-backend.c
> index fa120630be..e5707a0f8c 100644
> --- a/block/block-backend.c
> +++ b/block/block-backend.c
> @@ -141,7 +141,7 @@ static char *blk_root_get_parent_desc(BdrvChild *child)
> } else {
> /* TODO Callback into the BB owner for something more detailed */
> g_free(dev_id);
> - return g_strdup("a block device");
> + return g_strdup_printf("a block device %p", blk);
> }
> }
>
--
Best regards,
Vladimir