Re: [PATCH v4 6/7] arm64: use activity monitors for frequency invariance

Hi Valentin, Lukasz,

On Tuesday 25 Feb 2020 at 09:59:20 (+0000), Lukasz Luba wrote:
[..]

On 2/24/20 6:40 PM, Valentin Schneider wrote:

quoted

Ionela Voinescu writes:

quoted

+static inline int

That should be bool, seeing what it returns.

Will do!

[..]

quoted

quoted

+#ifndef arch_cpu_freq_counters
+static __always_inline
+bool arch_cpu_freq_counters(struct cpumask *cpus)
+{
+	return false;
+}
+#endif

Apologies for commenting on this only now, I had missed it in my earlier
round of review.

I would've liked to keep this contained within arm64 stuff until we agreed
on a more generic counter-driven FIE interface, but seems like we can't evade
it due to the arch_topology situation.

Would it make sense to relocate this stub to arch_topology.h instead, at
least for the time being? That way the only non-arm64 changes are condensed
in arch_topology (even if it doesn't change much in terms of header files,
since topology.h imports arch_topology.h)

Or make it as a 'weak' and place it just above the arch_set_freq_scale()
in arch_topology.c, not touching headers?

Yes, you guys are right, this works better nicely confined to
arch_topology.c/h. As Lukasz suggested, I'll make
arch_cpu_freq_counters (while here, it probably works better renamed to
arch_freq_counters_available) a weak function in arch_topology.c with
its strong definition in arm64/kernel/topology.c.

The diff is large(ish) so I'll push v5 directly with this change.

Thank you both for the review,
Ionela.