Re: [Qemu-devel] [PATCH 12/12] block/get_block_status: fix BDRV_BLOCK_ZERO for unallocated blocks

[Peter Lieven <pl@kamp.de>]

On 13.09.2013 12:25, Peter Lieven wrote:
> this patch does 2 things:
> a) only do additional call outs if BDRV_BLOCK_ZERO is not already set.
> b) use bdi.discard_zeroes to return the zero state of an unallocated block.
>     the callout to bdrv_has_zero_init() is only valid right after bdrv_create.
>
> Signed-off-by: Peter Lieven <pl@kamp.de>
> ---
>   block.c |    6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/block.c b/block.c
> index 282943b..6698c10 100644
> --- a/block.c
> +++ b/block.c
> @@ -3153,8 +3153,10 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
>           return ret;
>       }
>   
> -    if (!(ret & BDRV_BLOCK_DATA)) {
> -        if (bdrv_has_zero_init(bs)) {
> +    if (!(ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO)) {
> +        BlockDriverInfo bdi;
> +        if (bdrv_get_info(bs, &bdi) == 0 &&
> +            bdi.discard_zeroes) {
>               ret |= BDRV_BLOCK_ZERO;
>           } else if (bs->backing_hd) {
>               BlockDriverState *bs2 = bs->backing_hd;
Kevin, according to your comment that we need a separate function like
bdrv_has_discard_zeroes. It should also be ok to change the order and
first check for bs->backing_hd here or not?

Peter