On 26.02.2023 18:37, Rob Herring wrote: > On Fri, Feb 17, 2023 at 12:16:08PM +0100, Konrad Dybcio wrote: >> qcom,smmu-500 was introduced to prevent people from adding new >> compatibles for what seems to roughly be the same hardware. Use it for >> qcom,adreno-smmu-compatible targets as well. >> >> Signed-off-by: Konrad Dybcio <konrad.dybcio@xxxxxxxxxx> >> --- >> v1 -> v2: >> - Add this patch, omitted previously (big oops) >> >> .../devicetree/bindings/iommu/arm,smmu.yaml | 14 ++++++++++++-- >> 1 file changed, 12 insertions(+), 2 deletions(-) >> >> diff --git a/Documentation/devicetree/bindings/iommu/arm,smmu.yaml b/Documentation/devicetree/bindings/iommu/arm,smmu.yaml >> index 807cb511fe18..4d7f61700cae 100644 >> --- a/Documentation/devicetree/bindings/iommu/arm,smmu.yaml >> +++ b/Documentation/devicetree/bindings/iommu/arm,smmu.yaml >> @@ -75,9 +75,19 @@ properties: >> - qcom,sm8350-smmu-500 >> - qcom,sm8450-smmu-500 >> - const: arm,mmu-500 >> - >> - - description: Qcom Adreno GPUs implementing "arm,smmu-500" >> + - description: Qcom Adreno GPUs implementing "qcom,smmu-500" and "arm,smmu-500" >> + items: >> + - enum: >> + - qcom,sc7280-smmu-500 >> + - qcom,sm8150-smmu-500 >> + - qcom,sm8250-smmu-500 >> + - const: qcom,adreno-smmu >> + - const: qcom,smmu-500 >> + - const: arm,mmu-500 > > 4 compatibles seems excessive. Is adding one that helpful? Is > 'arm,mmu-500' useful on its own? Yes. per-soc compatible is there for per-soc quirks should there be any qcom,adreno-smmu enabled per-process pagetables qcom,smmu-500 matches the qcom smmu implementation arm,mmu-500 matches the smmu driver as a whole > >> + - description: Qcom Adreno GPUs implementing "arm,smmu-500" (legacy binding) > > Perhaps fix the existing typo: arm,mmu-500 Ack Konrad > >> + deprecated: true >> items: >> + # Do not add additional SoC to this list. Instead use previous list. >> - enum: >> - qcom,sc7280-smmu-500 >> - qcom,sm8150-smmu-500 >> -- >> 2.39.1 >>
