Re: [PATCH 6/6] target/arm: Implement FEAT_LPA2

[Richard Henderson]

On 1/8/22 01:39, Peter Maydell wrote:
> (1) The handling of the BaseADDR field for TLB range
> invalidates needs updating (there's a TODO to this effect in
> tlbi_aa64_range_get_base()).
>
> Side note: in that function, we shift the field by TARGET_PAGE_BITS,
> but the docs say that the shift should depend on the configured
> translation granule. Is that a bug?

Yes.

> (2) There are some new long-form fault status codes with FEAT_LPA2,
> corresponding to various fault types that can now occur at level -1.
> arm_fi_to_lfsc() needs updating to handle fi->level being -1.
> (You could do this bit as a preceding patch; it doesn't need to
> be squashed into this one.)

Yep, thanks.


r~