Hi Michel,

On Wed, May 23, 2018 at 11:20 AM, M P [off-list ref] wrote:

On Wed, 23 May 2018 at 10:12, Geert Uytterhoeven [off-list ref]
wrote:

quoted

On Tue, May 22, 2018 at 12:01 PM, Michel Pollet
[off-list ref] wrote:

quoted

+       #address-cells = <1>;
+       #size-cells = <1>;
+
+       cpus {
+               #address-cells = <1>;
+               #size-cells = <0>;
+               clocks = <&clock RZN1_DIV_CA7>;

quoted

I think the clocks property should be moved to the individual CPU nodes.

Ah, I had a look around, and I found some instances that are in the cpu
sub-node, and others that are not -- it seems that having it in the cpu
sub-node would implies it's core specific... here if that clock is changed
both cores would change speed...

Assumed the driver code knows to look in the parent node, which I doubt
the cpufreq code does.

Either way, it's not used by the kernel in any way at the moment -- I had
hoped cpufreq or something would claim it, but it's not the case.

I guess you have to add your main SoC compatible value to the whitelist
in drivers/cpufreq/cpufreq-dt-platdev.c first.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds