On Thu, 2 Jul 2026 11:16:22 +0200, Geert Uytterhoeven
[off-list ref] said:

Hi Bartosz,

On Thu, 2 Jul 2026 at 11:12, Bartosz Golaszewski [off-list ref] wrote:

quoted

On Tue, 30 Jun 2026 12:23:16 +0200, Vinod Koul [off-list ref] said:

quoted

On 29-06-26, 16:51, Geert Uytterhoeven wrote:

quoted

quoted

Russell King asked me to put the PHY logic for SCMI pm domains into the PHY
driver instead of the MAC driver where it was previously. Instead of cramming
both HLOS and firmware handling into the same driver, I figured it makes more
sense to have a dedicated, cleaner driver as the two share very little code (if
any).

I think you are mixing up DT bindings and driver implementation?

Should the bindings change if we have different driver and firmware
implementations? Isn't binding supposed to be agnostic of
implementations..?

I've thought about it some more and I believe this question is philosophical in
nature.

sa8775p and sa8255p are *the same* hardware. I can flash different firmware on
the same Lemans Ride board and it becomes one or the other. Yet they are not
described by the same DTS and the bindings differ as well. I don't see why we
wouldn't allow the same approach for the this PHY.

We treat it as different HW variant when it's managed by firmware - just like
we do with the rest of the SoC.

DT describes hardware, not software policy.

I'll defer to DT maintainers then for that particular case because it affects
more than just this platform. For instance: Qualcomm Nord[1] is already being
upstreamed with a similar split into common parts and then sources specific to
the SCMI and non-SCMI variants - even though it's the same SoC.

Bartosz

[1] https://lore.kernel.org/all/20260526051300.1669201-1-shengchao.guo@oss.qualcomm.com/ (local)