Re: [RFC PATCH V1 04/12] xen/arm: Introduce arch specific bits for IOREQ/DM features

[Stefano Stabellini <sstabellini@xxxxxxxxxx>]

On Wed, 5 Aug 2020, Jan Beulich wrote:
> On 05.08.2020 01:22, Stefano Stabellini wrote:
> > On Mon, 3 Aug 2020, Oleksandr Tyshchenko wrote:
> >> --- a/xen/include/asm-arm/p2m.h
> >> +++ b/xen/include/asm-arm/p2m.h
> >> @@ -385,10 +385,11 @@ static inline int set_foreign_p2m_entry(struct 
> >> domain *d, unsigned long gfn,
> >>                                          mfn_t mfn)
> >>  {
> >>      /*
> >> -     * NOTE: If this is implemented then proper reference counting of
> >> -     *       foreign entries will need to be implemented.
> >> +     * XXX: handle properly reference. It looks like the page may not 
> >> always
> >> +     * belong to d.
> > 
> > Just as a reference, and without taking away anything from the comment,
> > I think that QEMU is doing its own internal reference counting for these
> > mappings.
> 
> Which of course in no way replaces the need to do proper ref counting
> in Xen. (Just FAOD, as I'm not sure why you've said what you've said.)

Given the state of the series, which is a RFC, I only meant to say that
the lack of refcounting shouldn't prevent things from working when using
QEMU. In the sense that if somebody wants to give it a try for an early
demo, they should be able to see it running without crashes.

Of course, refcounting needs to be implemented.