On Mon, 05/25 16:43, Paolo Bonzini wrote:
>
>
> On 22/05/2015 05:40, Fam Zheng wrote:
> > +{
> > +BlockDriverState *p;
> > +int64_t ret;
> > +
> > +assert(bs != base);
> > +for (p = bs; p != base; p = p->backing_hd) {
> > +ret = bdrv_co_get_block_status(p, sector_num, nb_s
On 22/05/2015 05:40, Fam Zheng wrote:
> +{
> +BlockDriverState *p;
> +int64_t ret;
> +
> +assert(bs != base);
> +for (p = bs; p != base; p = p->backing_hd) {
> +ret = bdrv_co_get_block_status(p, sector_num, nb_sectors, pnum);
It's a bit ugly to have different parameters f
On 05/21/2015 09:40 PM, Fam Zheng wrote:
> Now this function follows the backing chain until seeing BDRV_BLOCK_ALLOCATED.
> Base is not included, and it can be NULL just like bdrv_is_allocated_above().
>
> Existing callers pass in bs->backing_hd to keep the old behavior.
>
> Signed-off-by: Fam Zh
Now this function follows the backing chain until seeing BDRV_BLOCK_ALLOCATED.
Base is not included, and it can be NULL just like bdrv_is_allocated_above().
Existing callers pass in bs->backing_hd to keep the old behavior.
Signed-off-by: Fam Zheng
---
block/io.c| 39
