On Tue, Aug 23, 2011 at 02:38:00PM +0200, Kevin Wolf wrote:
> Am 16.08.2011 01:16, schrieb Devin Nakamura:
> > Signed-off-by: Devin Nakamura <devin...@gmail.com>
> > ---
> >  block/qed.c |    2 ++
> >  1 files changed, 2 insertions(+), 0 deletions(-)
> > 
> > diff --git a/block/qed.c b/block/qed.c
> > index 333f067..9a1e49c 100644
> > --- a/block/qed.c
> > +++ b/block/qed.c
> > @@ -263,6 +263,8 @@ static int qed_read_string(BlockDriverState *file, 
> > uint64_t offset, size_t n,
> >   */
> >  static uint64_t qed_alloc_clusters(BDRVQEDState *s, unsigned int n)
> >  {
> > +    s->file_size =  qed_start_of_cluster(s, s->file_size +
> > +                                         s->header.cluster_size - 1);
> >      uint64_t offset = s->file_size;
> >      s->file_size += n * s->header.cluster_size;
> >      return offset;
> 
> Stefan, Devin, have you come to a conclusion about this patch?

Yes, I suggested keeping the current contraint that file_size is
cluster-aligned.

Devin, does that work for you?

Stefan

Reply via email to