Re: [PATCH v4 04/17] KVM: arm64: Add a heap allocator for the pKVM hyp

[Vincent Donnefort <vdonnefort@google.com>]

On Tue, Aug 18, 2026 at 03:22:55PM +0100, Fuad Tabba wrote:
> Hi Vincent,
> 
> On Fri, 31 Jul 2026 at 15:36, 'Vincent Donnefort' via kernel-team
> <kernel-team@android.com> wrote:
> ...
> > +static int hyp_allocator_destroy_chunk(struct hyp_allocator *allocator,
> > +                                      struct chunk_hdr *prev,
> > +                                      struct chunk_hdr *chunk)
> > +{
> > +       struct chunk_hdr *next;
> > +
> > +       next = prev ? chunk_get_next(chunk) : NULL;
> > +       if (!hyp_allocator_can_destroy_chunk(allocator, prev, next, chunk))
> > +               return -EINVAL;
> > +
> > +       /* Last chunk in the allocator */
> > +       if (!prev) {
> > +               allocator->first_unmapped = chunk_unmapped(chunk);
> > +               allocator->head = allocator->tail = NULL;
> > +               return 0;
> > +       }
> 
> hyp_allocator_destroy_chunk() has only two callers, both in
> hyp_allocator_free(), and both pass a non-NULL prev. So this !prev
> branch cannot be reached, and nor can the matching "if (!prev) return
> true;" in hyp_allocator_can_destroy_chunk() or the false arm of "next
> = prev ? chunk_get_next(chunk) : NULL".
> 
> The empty-allocator reset it looks like it is handling already happens
> in hyp_allocator_reclaim_chunk(), in the addr == allocator->start
> case. Could these be dropped?

Ha yes, correct, that's leftover of the Android version. I can then remove that
allocator argument in hyp_allocator_can_destroy_chunk().

> 
> With that:
> 
> Reviewed-by: Fuad Tabba <fuad.tabba@linux.dev>
> 
> Cheers,
> /fuad
> 
> To unsubscribe from this group and stop receiving emails from it, send an email to kernel-team+unsubscribe@android.com.
>