On Mon, Feb 09, 2026 at 03:18:43PM +0000, Peter Maydell wrote:
> On Tue, 23 Dec 2025 at 01:22, Mark Brown <[email protected]> wrote:

> A late reply, but I just noticed that the cover letter says:

> > Userspace access to ZA and (if configured) ZT0 is always available, they
> > will be zeroed when the guest runs if disabled in SVCR and the value
> > read will be zero if the guest stops with them disabled. This mirrors
> > the behaviour of the architecture, enabling access causes ZA and ZT0 to
> > be zeroed, while allowing access to SVCR, ZA and ZT0 to be performed in
> > any order.

> but the doc patch itself says:

> > +Access to the ZA and ZT0 registers is only available if SVCR.ZA is set
> > +to 1.

> Which one is the intention here ?

I guess the second if we're going with the approach of making everything
depend on the state of SVCR, it's less pressing to do so than with
SVCR.SM but it'd be consistent.  I'll update, thanks for spotting this.

Attachment: signature.asc
Description: PGP signature

Reply via email to