Re: [Xen-devel] [for-4.7 2/2] xen/arm: Document the behavior of XENMAPSPACE_dev_mmio

[Stefano Stabellini <sstabellini@xxxxxxxxxx>]

On Wed, 25 May 2016, Julien Grall wrote:
> Hi Jan,
> 
> On 25/05/16 14:14, Jan Beulich wrote:
> > > > > On 25.05.16 at 13:41, <julien.grall@xxxxxxx> wrote:
> > > --- a/xen/include/public/memory.h
> > > +++ b/xen/include/public/memory.h
> > > @@ -220,7 +220,10 @@ DEFINE_XEN_GUEST_HANDLE(xen_machphys_mapping_t);
> > >   #define XENMAPSPACE_gmfn_range   3 /* GMFN range, XENMEM_add_to_physmap
> > > only. */
> > >   #define XENMAPSPACE_gmfn_foreign 4 /* GMFN from another dom,
> > >                                       * XENMEM_add_to_physmap_batch only.
> > > */
> > > -#define XENMAPSPACE_dev_mmio     5 /* device mmio region */
> > > +#define XENMAPSPACE_dev_mmio     5 /* device mmio region
> > > +                                      On ARM, the region will be mapped
> > > +                                      in stage-2 using the memory
> > > attribute
> > > +                                      Device_nGnRE. */
> > 
> > Since this is unimplemented on x86, may I suggest to make this "ARM
> > only; the region will be ..."?
> 
> Sounds good.
> 
> > 
> > Also - is Device_nGnRE a term uniform between ARM32 and ARM64?
> 
> Actually it should be Device-nGnRE to match with the spec. This has been
> introduced on Armv8. Previously for Armv7, the name was "Device", although the
> behavior is exactly the same.
> 
> I could say:
> 
> "ARM only; the region will be mapped in Stage-2 using the memory attribute
> 'Device-nGnRE' (previously named 'Device' on ARMv7)".

Small nit: I would say "the region gets mapped" or "is mapped" rather
than "will be" because that makes me think that it is a future behavior
of following Xen releases rather than the current behavior.

Acked-by: Stefano Stabellini <sstabellini@xxxxxxxxxx>

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