Re: [Xen-devel] [PATCH v4] x86/altp2m: Add a subop for obtaining the mem access of a page

[Wei Liu <wei.liu2@xxxxxxxxxx>]

On Wed, Sep 19, 2018 at 08:48:54AM -0600, Jan Beulich wrote:
> >>> On 19.09.18 at 16:44, <wei.liu2@xxxxxxxxxx> wrote:
> > On Mon, Sep 03, 2018 at 06:47:34PM +0300, Adrian Pop wrote:
> >> --- a/xen/include/public/hvm/hvm_op.h
> >> +++ b/xen/include/public/hvm/hvm_op.h
> >> @@ -234,6 +234,7 @@ struct xen_hvm_altp2m_view {
> >>  typedef struct xen_hvm_altp2m_view xen_hvm_altp2m_view_t;
> >>  DEFINE_XEN_GUEST_HANDLE(xen_hvm_altp2m_view_t);
> >>  
> >> +#if __XEN_INTERFACE_VERSION__ < 0x00040a00
> >>  struct xen_hvm_altp2m_set_mem_access {
> >>      /* view */
> >>      uint16_t view;
> >> @@ -245,6 +246,19 @@ struct xen_hvm_altp2m_set_mem_access {
> >>  };
> >>  typedef struct xen_hvm_altp2m_set_mem_access 
> >> xen_hvm_altp2m_set_mem_access_t;
> >>  DEFINE_XEN_GUEST_HANDLE(xen_hvm_altp2m_set_mem_access_t);
> >> +#endif /* __XEN_INTERFACE_VERSION__ < 0x00040a00 */
> > 
> > #else here please, otherwise you're adding this new interface to a set
> > of old interfaces.
> 
> I don't think we do so elsewhere, so I'd prefer if we didn't do so
> here either. By unconditionally exposing the new interface we
> allow people to morph their code in steps, rather than all in one
> go at the point they bump their __XEN_INTERFACE_VERSION__
> setting.
> 

Okay, I'm not too fussed. Your argument sounds reasonable.

Wei.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel