On Mon, Mar 23, 2020 at 12:26:17PM -0400, Peter Xu wrote:
> On Mon, Mar 23, 2020 at 08:42:16AM -0700, Sean Christopherson wrote:
> > > > Regarding the HVA, it's a bit confusing saying that it's guaranteed to be
> > > > valid, and then contradicting that in the second clause. Maybe something
> > > > like this to explain the GPA->HVA is guaranteed to be valid, but the
> > > > HVA->HPA is not.
> > > >
> > > > /*
> > > > * before use. Note, KVM internal memory slots are guaranteed to remain valid
> > > > * and unchanged until the VM is destroyed, i.e. the GPA->HVA translation will
> > > > * not change. However, the HVA is a user address, i.e. its accessibility is
> > > > * not guaranteed, and must be accessed via __copy_{to,from}_user().
> > > > */
> > >
> > > Sure I can switch to this, though note that I still think the GPA->HVA
> > > is not guaranteed logically because the userspace can unmap any HVA it
> > > wants..
> >
> > You're conflating the GPA->HVA translation with the validity of the HVA,
> > i.e. the HVA->HPA and/or HVA->VMA translation/association. GPA->HVA is
> > guaranteed because userspace doesn't have access to the memslot which
> > defines that transation.
>
> Yes I completely agree if you mean the pure mapping of GPA->HVA.
>
> I think it's a matter of how to define the "valid" when you say
> "guaranteed to remain valid", because I don't think the mapping is
> still valid from the most strict sense if e.g. the backing HVA does
> not exist any more for that GPA->HVA mapping, then the memslot won't
> be anything useful.
Yes. That's why my proposed comment is worded to state that the _memslot_
will remain valid. It deliberately avoids mentioning "valid HVA".
