On 11.03.2023 12:24, Dmitry Baryshkov wrote: > On Sat, 11 Mar 2023 at 03:21, Konrad Dybcio <konrad.dybcio@xxxxxxxxxx> wrote: >> >> The Qualcomm RPM message RAM is a part of the RPM low-power processor's >> actual RAM that's exposed to the Application Processors as a MMIO-mapped >> region. It contains a couple of things, most notably the RPM master stats, >> detailing per-master (read: per-subsystem) sleep/wake statistics and an >> interrupt controller that's crucial to achieving full SoC clock and/or >> power collapse states (the MPM). This series tries to prepare the MSG >> RAM node to be a host for such devices. >> >> Depends on https://lore.kernel.org/linux-arm-msm/20230308011705.291337-1-konrad.dybcio@xxxxxxxxxx/ >> >> to solve a warning in the example. >> >> The DTS commits don't add -cells or -ranges, as without children they're >> not of very much use. They'll be introduced along with subnodes. >> >> Signed-off-by: Konrad Dybcio <konrad.dybcio@xxxxxxxxxx> >> --- >> Konrad Dybcio (15): >> dt-bindings: mfd: Split out Qcom RPM MSG RAM >> arm64: dts: qcom: msm8916: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: msm8953: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: msm8976: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: msm8994: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: msm8996: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: msm8998: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: qcs404: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: sdm630: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: sm6115: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: sm6125: Add simple-mfd to rpm_msg_ram >> arm64: dts: qcom: sm6375: Add simple-mfd to rpm_msg_ram >> ARM: dts: qcom-apq8084: Add simple-mfd to rpm_msg_ram >> ARM: dts: qcom-msm8226: Add simple-mfd to rpm_msg_ram >> ARM: dts: qcom-msm8974: Add simple-mfd to rpm_msg_ram > > These commits do not demonstrate the need for simple-mfd (other than > qcom,mpm being mentioned in bindings). I'd suggest limiting the RFC to > a single platform populated with rpm_msg_ram child devices. This way > we can better understand the needs and the requirements for these > patches. Right.. I was debating whether to just send the bindings as a RFC, but then I'd have to make simple-mfd optional just to retire it soon.. Konrad > >
