Re: [PATCH v8 20/28] KVM: arm64: Add clock support for the pKVM hyp

[Marc Zyngier <maz@kernel.org>]

On Thu, 20 Nov 2025 11:36:16 +0000,
Vincent Donnefort <vdonnefort@google.com> wrote:
> 
> [...]
> 
> > > +/* Using host provided data. Do not use for anything else than debugging. */
> > > +u64 trace_clock(void)
> > > +{
> > > +	struct clock_data *clock = &trace_clock_data;
> > > +	u64 bank = smp_load_acquire(&clock->cur);
> > > +	u64 cyc, ns;
> > > +
> > > +	cyc = __arch_counter_get_cntpct() - clock->data[bank].epoch_cyc;
> > 
> > I can only imagine that you don't care about the broken systems that
> > do not have a monotonic counter, and that will happily have their
> > counter swing back and forth?
> 
> I haven't used the _stable() variant for the affected devices... Hum this will
> not be trivial from the hypervisor. I'll see what I can do.

In all honestly, simply disabling tracing when you don't have a stable
counter should be enough. I don't think there is much to be gained
with supporting that on sub-standard HW (the bar is low enough).

> > Also, you may want to document why you are using the physical counter
> > instead of the virtual one. I guess that you don't want CNTVOFF_EL2 to
> > apply when HCR_EL2.E2H==0, but given that you never allow anything
> > else for pKVM, this is slightly odd.
> 
> You mean I might as well use __arch_counter_get_cntvct() as CNTVOFF_EL2 will
> always be ignored in the pKVM case?

That, plus the fact that CNTVOFF_EL2 is only evaluated at EL2 when
E2H=0 (and that case is going to disappear over time).

	M.

-- 
Jazz isn't dead. It just smells funny.