Re: [Xen-devel] [PATCH 3/3] memory: restrict XENMEM_remove_from_physmap to translated guests

[Julien Grall <julien.grall@xxxxxxx>]

Hi,

On 08/04/2019 15:29, Jan Beulich wrote:
> > > > On 08.04.19 at 13:47, <julien.grall@xxxxxxx> wrote:
> > On 4/2/19 5:10 PM, Jan Beulich wrote:
> > > > > > On 02.04.19 at 12:26, <julien.grall@xxxxxxx> wrote:
> > > > On 05/03/2019 13:28, Jan Beulich wrote:
> > > > > The commit re-introducing it (14eb3b41d0 ["xen: reinstate previously
> > > > > unused XENMEM_remove_from_physmap hypercall"]) as well as the one having
> > > > > originally introduced it (d818f3cb7c ["hvm: Use main memory for video
> > > > > memory"]) and the one then purging it again (78c3097e4f ["Remove unused
> > > > > XENMEM_remove_from_physmap"]) make clear that this operation is intended
> > > > > for use on HVM (i.e. translated) guests only. Restrict it at least as
> > > > > much, because for PV guests documentation (in the public header) does
> > > > > not even match the implementation: It talks about GPFN as input, but
> > > > > get_page_from_gfn() assumes a GMFN in the non-translated case (and hands
> > > > > back the value passed in).
> > > > >
> > > > > Also lift the check in XENMEM_add_to_physmap{,_batch} handling up
> > > > > directly into top level hypercall handling, and clarify things in the
> > > > > public header accordingly.
> > > > >
> > > > > Take the liberty and also replace a pointless use of "current" with a
> > > > > more efficient use of an existing local variable (or function parameter
> > > > > to be precise).
> > > > >
> > > > > Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> > > > > ---
> > > > > TBD: It could be further restricted, disallowing its use by a HVM guest
> > > > >         on itself.
> > > > By HVM guest, do you refer to any auto-translated guest?
> > > Yes - sorry for using an x86 term.
> > >
> > > > The interface XENME_remove_from_physmap is used by Arm to remove foreign
> > > > mappings from its p2m. There are potentially other space with similar case
> > > > (e.g grant-table...).
> > > Oh, I see - this option goes away then.
> > >
> > > > > TBD: Is using P2M_ALLOC here really appropriate? It means e.g.
> > > > >         pointlessly populating a PoD slot just to unpopulate it again right
> > > > >         away, with the page then free floating, i.e. no longer available
> > > > >         for use to replace another PoD slot, and (afaict) no longer
> > > > >         accessible by the guest in any way.
> > > > > TBD: Is using guest_physmap_remove_page() here really appropriate? It
> > > > >         means that e.g. MMIO pages wouldn't be removed. Going through
> > > > >         guest_remove_page() (while skipping the de-allocation step) would
> > > > >         seem more appropriate to me, which would address the P2M_ALLOC
> > > > >         aspect above as well.
> > > > How is that an issue? Does XENMEM_add_to_physmap allows you to map MMIO
> > > > pages?
> > > Well, there's XENMAPSPACE_dev_mmio which xatp handles. But
> > > perhaps the MMIO example is more confusing than helpful. The
> > > question really just is whether guest_remove_page() shouldn't
> > > be used here instead of guest_physmap_remove_page()
> > de-allocation step aside, I am not really convinced you can reuse
> > guest_remove_page() here. On x86, the function will not work on certain
> > p2m types. Is it what we really want?
> Hmm, I'm confused. Afaics the only two types it refuses a request
> for are p2m_invalid and p2m_mmio_dm. These represent cases
> where there's no p2m entry anyway, i.e. nothing to remove. Am
> I perhaps overlooking something you see?
>
> Or are you referring to the mfn_invalid() check (which isn't x86-
> specific)? This ought to be covered by the p2m_is_paging() and
> p2m_mmio_direct special cases a few lines up from there. Other
> cases with invalid MFNs would indeed represent an error condition
> imo.
I am referring to get_page(...) which would not work for foreign pages.

> In the end it's actually quite the opposite that I'm thinking: For
> the caller it shouldn't, for example, matter whether the
> requested page was paged out. We wouldn't even call
> guest_physmap_remove_page() in that case, while
> guest_remove_page() would take care of it.
But those pages should never be removed via physmap, I am correct?

Anyway, guest_physmap_remove_page() would need some rework to do the correct 
things for physmap. I will wait an see your patch to comment.
Cheers,

--
Julien Grall

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