Re: [Qemu-devel] [PATCH RESEND 16/50] block: Prepare remaining BB functions for NULL BDS

[Max Reitz]

On 2015-01-27 at 16:36, Eric Blake wrote:
> On 01/27/2015 12:45 PM, Max Reitz wrote:
> > There are several BlockBackend functions which, in theory, cannot fail.
> > This patch makes them cope with the BlockDriverState pointer being NULL
> > by making them fall back to some default action like ignoring the value
> > in setters and returning the default in getters.
> >
> > Signed-off-by: Max Reitz <address@hidden>
> > ---
> >   block/block-backend.c | 70 ++++++++++++++++++++++++++++++++++++++++-----------
> >   1 file changed, 55 insertions(+), 15 deletions(-)
> >
> > @@ -926,22 +944,32 @@ void *blk_blockalign(BlockBackend *blk, size_t size)
> >   
> >   bool blk_op_is_blocked(BlockBackend *blk, BlockOpType op, Error **errp)
> >   {
> > +    if (!blk->bs) {
> > +        return false;
> > +    }
> > +
> >       return bdrv_op_is_blocked(blk->bs, op, errp);
> >   }
> I don't know if this is correct - maybe a no-medium drive should block
> actions like commit by default?

Somehow I don't think they should be blocked. For instance, commit will 
fail the instant you actually try to commit something (that is, read 
some data or access the backing hierarchy).

Generally, I think we don't need to block operations on an empty 
BlockBackend due to the fact that any operation will fail anyway.

Max

> If my design question gets answered satisfactorily that you went with
> the right default, then
>
> Reviewed-by: Eric Blake <address@hidden>