[PATCH v2 03/18] KVM: arm64: Add __hyp_allocator_map for the pKVM hyp

[Fuad Tabba]

On Wed, 15 Jul 2026 at 09:58, Vincent Donnefort <vdonnefort at google.com> wrote:
>
> On Tue, Jul 14, 2026 at 06:53:55PM +0100, Fuad Tabba wrote:
> > On Mon, 6 Jul 2026 at 18:54, Vincent Donnefort <vdonnefort at google.com> wrote:
> > >
> > > In preparation for the pKVM heap allocator, introduce __hyp_allocator_map()
> > > to map a single physical page to a given virtual address in the hypervisor
> > > stage-1 page-table.
> > >
> > > Signed-off-by: Vincent Donnefort <vdonnefort at google.com>
> > >
> > > diff --git a/arch/arm64/kvm/hyp/include/nvhe/mm.h b/arch/arm64/kvm/hyp/include/nvhe/mm.h
> > > index 725bb0fb941d..98a7774b541c 100644
> > > --- a/arch/arm64/kvm/hyp/include/nvhe/mm.h
> > > +++ b/arch/arm64/kvm/hyp/include/nvhe/mm.h
> > > @@ -32,5 +32,6 @@ int __pkvm_create_private_mapping(phys_addr_t phys, size_t size,
> > >  int pkvm_create_stack(phys_addr_t phys, unsigned long *haddr);
> > >  int pkvm_alloc_private_va_range(size_t size, unsigned long *haddr);
> > >  phys_addr_t __pkvm_private_range_pa(void *va);
> > > +int __hyp_allocator_map(void *va, phys_addr_t phys);
> > >
> > >  #endif /* __KVM_HYP_MM_H */
> > > diff --git a/arch/arm64/kvm/hyp/nvhe/mm.c b/arch/arm64/kvm/hyp/nvhe/mm.c
> > > index ec59da1322a7..c1c01f81ac5f 100644
> > > --- a/arch/arm64/kvm/hyp/nvhe/mm.c
> > > +++ b/arch/arm64/kvm/hyp/nvhe/mm.c
> > > @@ -160,6 +160,11 @@ void pkvm_remove_mappings(void *from, void *to)
> > >         hyp_spin_unlock(&pkvm_pgd_lock);
> > >  }
> > >
> > > +int __hyp_allocator_map(void *va, phys_addr_t phys)
> > > +{
> > > +       return __pkvm_create_mappings((unsigned long)va, PAGE_SIZE, phys, PAGE_HYP);
> > > +}
> > > +
> >
> > nit: __hyp_allocator_map() is named for its caller rather than what it does
> > (map a single hyp page). Its siblings here (pkvm_create_mappings,
> > pkvm_remove_mappings) are named for the operation. Consider a name like
> > pkvm_map_page() so it reads as a generic primitive.
>
> IIRC (it was a few years ago) this was a suggestion from Quentin who didn't want
> to expose too much of mm and suggested to make it clear that was a function
> private to the allocator.
>
> But looking at what we have now. perhaps
>
> int __pkvm_map_private_pages(phys_addr_t phys, void *addr, size_t size);
>
> And this could verify that addr is actually aprt of the private range
>
> WDYS?

I think this sounds better. That was a nit, so up to you.

Thanks,
/fuad

>
> >
> > That said, naming is hard:
> >
> > Reviewed-by: Fuad Tabba <fuad.tabba at linux.dev>
> >
> > Cheers,
> > /fuad
> >
> > >  int hyp_back_vmemmap(phys_addr_t back)
> > >  {
> > >         unsigned long i, start, size, end = 0;
> > > --
> > > 2.55.0.rc2.803.g1fd1e6609c-goog
> > >