[v5] block: Deal with filters

[Qemu-devel] [PATCH v5 07/42] block: *filtered_cow_child() for *has_zero_init()

Posted by Max Reitz 6 years, 8 months ago

bdrv_has_zero_init() and the related bdrv_unallocated_blocks_are_zero()
should use bdrv_filtered_cow_child() if they want to check whether the
given BDS has a COW backing file.

Signed-off-by: Max Reitz <mreitz@redhat.com>
---
 block.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/block.c b/block.c
index be18130944..64d6190984 100644
--- a/block.c
+++ b/block.c
@@ -4933,7 +4933,7 @@ int bdrv_has_zero_init(BlockDriverState *bs)
 
     /* If BS is a copy on write image, it is initialized to
        the contents of the base image, which may not be zeroes.  */
-    if (bs->backing) {
+    if (bdrv_filtered_cow_child(bs)) {
         return 0;
     }
     if (bs->drv->bdrv_has_zero_init) {
@@ -4951,7 +4951,7 @@ bool bdrv_unallocated_blocks_are_zero(BlockDriverState *bs)
 {
     BlockDriverInfo bdi;
 
-    if (bs->backing) {
+    if (bdrv_filtered_cow_child(bs)) {
         return false;
     }
 
-- 
2.21.0

Re: [Qemu-devel] [PATCH v5 07/42] block: *filtered_cow_child() for *has_zero_init()

Posted by Vladimir Sementsov-Ogievskiy 6 years, 8 months ago

13.06.2019 1:09, Max Reitz wrote:
> bdrv_has_zero_init() and the related bdrv_unallocated_blocks_are_zero()
> should use bdrv_filtered_cow_child() if they want to check whether the
> given BDS has a COW backing file.
> 
> Signed-off-by: Max Reitz <mreitz@redhat.com>
> ---
>   block.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/block.c b/block.c
> index be18130944..64d6190984 100644
> --- a/block.c
> +++ b/block.c
> @@ -4933,7 +4933,7 @@ int bdrv_has_zero_init(BlockDriverState *bs)
>   
>       /* If BS is a copy on write image, it is initialized to
>          the contents of the base image, which may not be zeroes.  */
> -    if (bs->backing) {
> +    if (bdrv_filtered_cow_child(bs)) {
>           return 0;
>       }

Hmm, if you are fixing bdrv_has_zero_init around filters, I'd prefere to fix the whole
function, converting the following here too:
     if (bs->file && bs->drv->is_filter) {
         return bdrv_has_zero_init(bs->file->bs);
     }


But it's not a real problem:

Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>

>       if (bs->drv->bdrv_has_zero_init) {
> @@ -4951,7 +4951,7 @@ bool bdrv_unallocated_blocks_are_zero(BlockDriverState *bs)
>   {
>       BlockDriverInfo bdi;
>   
> -    if (bs->backing) {
> +    if (bdrv_filtered_cow_child(bs)) {
>           return false;
>       }
>   
> 


-- 
Best regards,
Vladimir