Re: [Xen-devel] [PATCH linux-2.6.18] support suspend/resume in pvscsi drivers

[Juergen Gross <jgross@xxxxxxxx>]

On 01/30/2015 04:05 PM, Jan Beulich wrote:
> > > > On 30.01.15 at 15:54, <JGross@xxxxxxxx> wrote:
> > On 01/30/2015 03:46 PM, Jan Beulich wrote:
> > > > > > On 30.01.15 at 14:52, <"jgross@xxxxxxxx".non-mime.internet> wrote:
> > > > @@ -231,8 +242,23 @@ static int scsifront_cmd_done(struct vsc
> > > >         return more_to_do;
> > > >    }
> > > >
> > > > +void scsifront_finish_all(struct vscsifrnt_info *info)
> > > > +{
> > > > +       unsigned i;
> > > > +       struct vscsiif_response resp;
> > > >
> > > > +       scsifront_ring_drain(info);
> > > Shouldn't you at least issue some kind of warning when this returns
> > > non-zero?
> > If a warning should be issued, then this should be done after the
> > following loop in case of at least one request terminated there.
> >
> > I'm really not sure, whether a warning is required here. If you like,
> > I can add one.
> I'm not sure, I'm merely asking because I saw the function return
> value being ignored here.
I think it can be 0 only. We are handling resume, so the ring which is
being drained will no longer be filled by the backend.

Juergen


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel