On Fri, Mar 22, 2024 at 04:55:04PM +0000, David Woodhouse wrote:
> On Fri, 2024-03-22 at 16:37 +0000, Marc Zyngier wrote:
> >
> > I agree that nothing really breaks, but I also hold the view that
> > broken firmware implementations should be given the finger, specially
> > given that you have done this work *ahead* of the spec. I would really
> > like this to fail immediately on these and not even try to suspend.
> >
> > With that in mind, if doesn't really matter whether HIBERNATE_OFF is
> > mandatory or not. We really should check for it and pretend it doesn't
> > exist if the correct flag isn't set.
>
> Ack.
>
> I'll rename that variable to 'psci_system_off2_hibernate_supported' then.
>
> static void __init psci_init_system_off2(void)
> {
> int ret;
>
> ret = psci_features(PSCI_FN_NATIVE(1_3, SYSTEM_OFF2));
> if (ret < 0)
> return;
>
> if (ret & (1 << PSCI_1_3_HIBERNATE_TYPE_OFF))
> psci_system_off2_hibernate_supported = true;
>
Ah OK, you have already agreed to do this, please ignore my response then.
--
Regards,
Sudeep
