On Mon, 4 Aug 2025 at 13:44, Peter Maydell <peter.mayd...@linaro.org> wrote:
>
> On Wed, 16 Jul 2025 at 10:56, Luc Michel <luc.mic...@amd.com> wrote:
> >
> > Add support for the ARM Cortex-A78AE CPU.
>
>
>
> > static const ARMCPUInfo aarch64_cpus[] = {
> > { .name = "cortex-a35", .initfn = aarch64_a35_initfn },
> > { .name = "cortex-a55", .initfn = aarch64_a55_initfn },
> > { .name = "cortex-a72", .initfn = aarch64_a72_initfn },
> > { .name = "cortex-a76", .initfn = aarch64_a76_initfn },
> > + { .name = "cortex-a78ae", .initfn = aarch64_a78ae_initfn },
>
> Do we really need to specifically call this the Cortex-A78AE ?
> Is there anything we would model differently between this and
> the Cortex-A78 ?
Looking a bit more closely at the TRMs, although they're quite
similar they're not identical -- for example the 78AE has
FEAT_FlagM (as advertised in ID_AA64ISAR0_EL1.TS) and the 78
does not. So we should keep this "cortex-a78ae" name, with
a comment something like "This is not quite the same as the
plain Cortex-A78; we don't currently model the latter".
thanks
-- PMM
