Re: [PATCH 1/2] dt-bindings: perf: arm,smmu-v3-pmcg: Allow power-domains

From: Robin Murphy

Date: Wed Nov 18 2026 - 08:07:20 EST


On 2026-11-18 10:19 am, Amita Mishra wrote:
On some SoCs, SMMUv3 Performance Monitor Counter Groups (PMCGs) reside
in switchable power domains rather than an always-on domain.

Document the optional power-domains property so that runtime power
management can manage the PMCG power state.

Isn't power-domain-consumer one of the common schemas that's automatically applicable to any node anyway? I would think that unless we have a specific expectation and description of what each power-domains entry should represent - which we cannot reasonably do in a binding that is for the architectural PMCG interface rather than any specific implementation of such - we'd be better off just leaving it implicit.

Signed-off-by: Amita Mishra <amitamishra@xxxxxxxxxx>
---
Documentation/devicetree/bindings/perf/arm,smmu-v3-pmcg.yaml | 3 +++
1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/perf/arm,smmu-v3-pmcg.yaml b/Documentation/devicetree/bindings/perf/arm,smmu-v3-pmcg.yaml
index a4b53a6a1ebf..aea6623bb98c 100644
--- a/Documentation/devicetree/bindings/perf/arm,smmu-v3-pmcg.yaml
+++ b/Documentation/devicetree/bindings/perf/arm,smmu-v3-pmcg.yaml
@@ -36,6 +36,9 @@ properties:
msi-parent: true
+ power-domains:
+ maxItems: 1

Even for existing implementations it would be quite possible for a PMCG to be functionally dependent on multiple power domains (yes some of them should usually be hierarchical, but still...) - what justifies 1 being the only "correct" number?

Thanks,
Robin.

+
required:
- compatible
- reg