On 29/02/2024 23:24, Chris Goldsworthy wrote: > On Thu, Feb 29, 2024 at 10:09:34PM +0200, Georgi Djakov wrote: >> Hi Krzysztof, >> >> On 29.02.24 19:53, Krzysztof Kozlowski wrote: >>> On 26/02/2024 18:22, Georgi Djakov wrote: >>>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) >>>> +%YAML 1.2 >>>> +--- >>>> +$id: http://devicetree.org/schemas/iommu/qcom,tbu.yaml# >>>> +$schema: http://devicetree.org/meta-schemas/core.yaml# >>>> + >>>> +title: Qualcomm TBU (Translation Buffer Unit) >>>> + >>>> +maintainers: >>>> + - Georgi Djakov <quic_c_gdjako@xxxxxxxxxxx> >>>> + >>>> +description: >>>> + The Qualcomm SMMU500 implementation consists of TCU and TBU. The TBU contains >>>> + a Translation Lookaside Buffer (TLB) that caches page tables. TBUs provides >>>> + debug features to trace and trigger debug transactions. There are multiple TBU >>>> + instances with each client core. >>>> + >>>> +properties: >>>> + compatible: >>>> + const: qcom,qsmmuv500-tbu >>> >>> Why we don't have SoC specific compatibles? If that's for SDM845, then >>> it should be qcom,sdm845-tbu or qcom,sdm845-qsmmuv500-tbu >>> >> >> Because they should be all compatible (as registers). Adding a SoC compatible >> might get overly-specific, but i can also see the benefits in that, so ok will >> do it! >> > > Hi Krzysztof, > > JFYI that the TBUs are used on our mobile SoCs going up until the SoC > we commercialized in early 2022, Snapdragon 8 Gen 1. Including SDM845 > there are three more premium tier SoCs using TBUs plus all of their > value-tier derivatives. There will also be prior generation premium > tier SoCs along with their derivatives that use the TBU as well. Does > it make sense to have a target-specific compatible string given this? This does not explain me anything. Why an exemption from usual bindings rules should apply here? https://elixir.bootlin.com/linux/v6.8-rc6/source/Documentation/devicetree/bindings/writing-bindings.rst Best regards, Krzysztof
