Re: [PATCH v4 10/10] KVM: arm64: np-guest CMOs with PMD_SIZE fixmap

[Marc Zyngier <maz@kernel.org>]

On Fri, 16 May 2025 19:03:14 +0100,
Vincent Donnefort <vdonnefort@google.com> wrote:
> 
> > > +}
> > > +
> > >  static void clean_dcache_guest_page(void *va, size_t size)
> > >  {
> > >  	WARN_ON(!PAGE_ALIGNED(size));
> > >  
> > >  	while (size) {
> > > -		__clean_dcache_guest_page(hyp_fixmap_map(__hyp_pa(va)),
> > > -					  PAGE_SIZE);
> > > -		hyp_fixmap_unmap();
> > > -		va += PAGE_SIZE;
> > > -		size -= PAGE_SIZE;
> > > +		size_t fixmap_size = size == PMD_SIZE ? size : PAGE_SIZE;
> > > +		void *addr = __fixmap_guest_page(va, &fixmap_size);
> > > +
> > > +		__clean_dcache_guest_page(addr, fixmap_size);
> > > +		__fixunmap_guest_page(fixmap_size);
> > > +
> > > +		size -= fixmap_size;
> > > +		va += fixmap_size;
> > 
> > Can this ever be called with a *multiple* of PMD_SIZE? In this case
> > you'd still end-up doing PAGE_SIZEd-bite CMOs until there is only
> > PMD_SIZE left, ruining the optimisation.
> > 
> > I think this needs fixing.
> 
> So this can be only called with size either equal to PAGE_SIZE or PMD_SIZE. I
> wasn't sure if it was worth to make it more generic than it needs.
> 
> But like for the first patch, I can make it more future-proof by handling size >
> PMD_SIZE.

Yup. These things are hard enough to debug that we should try and make
it fool-proof, even if that's not immediately used (and fixing that is
pretty easy).

Thanks,

	M.

-- 
Without deviation from the norm, progress is not possible.