Re: [PATCH v2 1/2] target/ppc: fix address translation bug for radix mmus

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

[-- Attachment #1: Type: text/plain, Size: 1395 bytes --]

On Tue, Jun 15, 2021 at 09:25:06AM -0300, Bruno Piazera Larsen wrote:
> 
> On 15/06/2021 00:20, Richard Henderson wrote:
> > On 6/14/21 6:41 PM, David Gibson wrote:
> > > I think move these to mmu-book3s-v3.h, since they're correct for both
> > > the radix and hash sides of the modern book3s mmu.
> > 
> > They're also correct for all non-booke mmus, e.g. hash32 and 6xx, which
> > is why I recommended internal.h (or some new mmu-internal.h).
> > 
> > While neither hash32 nor 6xx have HV, and thus there is no second tlb
> > bug, it would still be More Correct to use mmu_idx instead of direct
> > references to msr_pr et al.
> yeah, I agree that I should change the documentation. Before I send a new
> version with everything corrected, I wanna make sure if I should leave them
> in internal and use it for hash32 and 6xx MMUs, or do I put them in
> mmu-book3s-v3.h, since only radix64 has that the bug?

Put them in mmu-book3s-v3.h for the time being, we can move it later
if it makes sense.  I want to move to a mode model-first approach to
the softmmu code, so I think of these values as logically
per-MMU-style, even if they are the same in practice for many of them.

-- 
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 --]