Re: [Qemu-devel] [PATCH v8 01/15] block: Add "file" output parameter to block status query functions

[Fam Zheng <famz@redhat.com>]

On Tue, 01/26 11:27, Fam Zheng wrote:
> On Mon, 01/25 14:04, Kevin Wolf wrote:
> > Am 25.01.2016 um 03:44 hat Fam Zheng geschrieben:
> > > diff --git a/block/vvfat.c b/block/vvfat.c
> > > index 2ea5a4a..b8d29e1 100644
> > > --- a/block/vvfat.c
> > > +++ b/block/vvfat.c
> > > @@ -2884,7 +2884,7 @@ static coroutine_fn int vvfat_co_write(BlockDriverState *bs, int64_t sector_num,
> > >  }
> > >  
> > >  static int64_t coroutine_fn vvfat_co_get_block_status(BlockDriverState *bs,
> > > -	int64_t sector_num, int nb_sectors, int* n)
> > > +	int64_t sector_num, int nb_sectors, int *n, BlockDriverState **file)
> > >  {
> > >      BDRVVVFATState* s = bs->opaque;
> > >      *n = s->sector_count - sector_num;
> > 
> > This still returns NULL at the end of the series. Shouldn't it return bs
> > like other protocol drivers do?
> 
> Yes, we need another patch for vvfat.

No, I now remember why vvfat didn't need a patch: it never sets the
BDRV_BLOCK_OFFSET_VALID bit.

Fam