Re: [Qemu-devel] [PATCH v2] vhost: Allow adjoining regions

[Alex Williamson <alex.williamson@redhat.com>]

On Thu, 5 Apr 2018 20:46:55 +0100
"Dr. David Alan Gilbert" <dgilbert@redhat.com> wrote:

> * Alex Williamson (alex.williamson@redhat.com) wrote:
> > On Fri, 23 Mar 2018 15:39:39 +0000
> > "Dr. David Alan Gilbert (git)" <dgilbert@redhat.com> wrote:
> >   
> > > From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> > > 
> > > My rework of section adding combines overlapping or adjoining regions,
> > > but checks they're actually the same underlying RAM block.
> > > Fix the case where two blocks adjoin but don't overlap; that new region
> > > should get added (but not combined), but my previous patch was disallowing it.
> > > 
> > > Fixes: c1ece84e7c9
> > > 
> > > Reported-by: Alex Williamson <alex.williamson@redhat.com>
> > > Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> > > ---  
> > 
> > 
> > Ping?  This error report is a regression and should be considered for
> > 2.12.
> > 
> > Tested-by: Alex Williamson <alex.williamson@redhat.com>  
> 
> It's a change in vhost - so mst is taking it ?

That's my assumption, sorry, I should have set to/cc explicitly rather
than just replying.  Thanks,

Alex

> > >  hw/virtio/vhost.c | 13 +++++++++----
> > >  1 file changed, 9 insertions(+), 4 deletions(-)
> > > 
> > > diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
> > > index 250f886acb..2a720cca3d 100644
> > > --- a/hw/virtio/vhost.c
> > > +++ b/hw/virtio/vhost.c
> > > @@ -595,10 +595,15 @@ static void vhost_region_add_section(struct vhost_dev *dev,
> > >                                          prev_sec->offset_within_address_space,
> > >                                          prev_sec->offset_within_region);
> > >              } else {
> > > -                error_report("%s: Overlapping but not coherent sections "
> > > -                             "at %"PRIx64,
> > > -                             __func__, mrs_gpa);
> > > -                return;
> > > +                /* adjoining regions are fine, but overlapping ones with
> > > +                 * different blocks/offsets shouldn't happen
> > > +                 */
> > > +                if (mrs_gpa != prev_gpa_end + 1) {
> > > +                    error_report("%s: Overlapping but not coherent sections "
> > > +                                 "at %"PRIx64,
> > > +                                 __func__, mrs_gpa);
> > > +                    return;
> > > +                }
> > >              }
> > >          }
> > >      }  
> >   
> --
> Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK