Re: [PATCH v4 2/3] target/ppc: change ppc_hash32_xlate to use mmu_idx

[David Gibson <david@gibson.dropbear.id.au>]

[-- Attachment #1: Type: text/plain, Size: 1685 bytes --]

On Mon, Jul 05, 2021 at 11:31:13AM -0300, Bruno Piazera Larsen wrote:
> 
> On 05/07/2021 01:24, David Gibson wrote:
> 
> > > Changed hash32 address translation to use the supplied mmu_idx, instead
> > > of using what was stored in the msr, for parity purposes (radix64
> > > already uses that).
> 
> > Well.. parity and conceptual correctness.  The translation is supposed
> > to use mmu_idx, not look at the CPU again to get the right context.
> > AFAIK there isn't a situation in hash32 where they'll get out of sync,
> > but nothing guarantees that.
> 
> 
> Fair point, I can change the description if I do end up with a new
> version, but
> 
> > I think the right approach is to duplicate the helper macros in
> > mmu-hash32.h for now.  We can unify them later with a more thorough
> > review (which would probably involve creating a new header for things
> > common to all BookS family MMUs).
> 
> This doesn't work directly. I'd need to put in an ifndef
> PPC_MMU_BOOK3S_V3_H, which also feels a bit dubious to me. I can go
> with whichever one you prefer

Ah... good point, because both headers are included in some places.

Ok, in that case let's jump ahead instead.  Let's create a new
mmu-books.h to cover all MMUs based on the "classic" powerpc model,
and put them in there.  hash32 and book3s-v3 can include that, BookE,
4xx etc. will not.

For now these will be the only things in there, but there will
probably be more we can add later on.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]