Re: [PATCH 1/2] KVM: arm64: Infer the PA offset from IPA in stage-2 map walker

[Oliver Upton <oliver.upton@linux.dev>]

On Fri, Apr 21, 2023 at 10:28:58AM +0100, Marc Zyngier wrote:

[...]

> So my conclusion is that after these two patches, data->phys should
> never be updated, right?

Yep, that's the intention.

> Then I'd suggest an additional patch to constify a couple of things
> and make sure we don't accidentally update them. Something like the
> patch below (compile-tested only).

I'm always a fan of more idiot proofing, especially when I am said idiot
:)

> From a2eb08ce793c1cf01c79df13a619815e9d7c1d41 Mon Sep 17 00:00:00 2001
> From: Marc Zyngier <maz@kernel.org>
> Date: Fri, 21 Apr 2023 10:18:34 +0100
> Subject: [PATCH] KVM: arm64: Constify start/end/phys fields of the pgtable
>  walker data
> 
> As we are revamping the way the pgtable walker evaluates some of the
> data, make it clear that we rely on somew of the fields to be constant
> across the lifetime of a walk.
> 
> For this, flag the start, end and pjys fields of the walk data as

typo: phys

> 'const', which will generate an error if we were to accidentally
> update these fields again.
> 
> Signed-off-by: Marc Zyngier <maz@kernel.org>

Looks good.

Reviewed-by: Oliver Upton <oliver.upton@linux.dev>

Let's see if I've confused b4 into thinking I'm reviewing my own patches
:-P

> ---
>  arch/arm64/kvm/hyp/pgtable.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> index 356a3fd5220c..5282cb9ca4cf 100644
> --- a/arch/arm64/kvm/hyp/pgtable.c
> +++ b/arch/arm64/kvm/hyp/pgtable.c
> @@ -58,9 +58,9 @@
>  struct kvm_pgtable_walk_data {
>  	struct kvm_pgtable_walker	*walker;
>  
> -	u64				start;
> +	const u64			start;
>  	u64				addr;
> -	u64				end;
> +	const u64			end;
>  };
>  
>  static bool kvm_phys_is_valid(u64 phys)
> @@ -352,7 +352,7 @@ int kvm_pgtable_get_leaf(struct kvm_pgtable *pgt, u64 addr,
>  }
>  
>  struct hyp_map_data {
> -	u64				phys;
> +	const u64			phys;
>  	kvm_pte_t			attr;
>  };
>  
> @@ -578,7 +578,7 @@ void kvm_pgtable_hyp_destroy(struct kvm_pgtable *pgt)
>  }
>  
>  struct stage2_map_data {
> -	u64				phys;
> +	const u64			phys;
>  	kvm_pte_t			attr;
>  	u8				owner_id;
>  
> -- 
> 2.34.1
> 
> -- 
> Without deviation from the norm, progress is not possible.

-- 
Thanks,
Oliver