Re: [PATCH v5 1/2] dt-bindings: clock: qcom: Add LPASS VA CSR HeartBeat pulse clock
Konrad Dybcio <[email protected]>
| Newsgroups | org.kernel.vger.linux-devicetree,org.kernel.vger.linux-arm-msm,org.kernel.vger.linux-clk,org.kernel.vger.linux-kernel |
|---|---|
| Message-ID | <[email protected]> |
On 8/10/26 6:01 PM, Bjorn Andersson wrote: > On Thu, Aug 06, 2026 at 08:21:36PM +0530, Sarath Ganapathiraju via B4 Relay wrote: >> From: Sarath Ganapathiraju <[email protected]> >> >> Add a binding for the Qualcomm LPASS VA CSR rate generator node that >> exposes the lpass_heartbeat_pulse clock on hawi. [...] >> +examples: >> + - | >> + soc { >> + #address-cells = <2>; >> + #size-cells = <2>; >> + >> + clock-controller@7eed000 { >> + compatible = "qcom,hawi-lpass-va-csr"; >> + reg = <0x0 0x07eed000 0x0 0x14>; > > The two 0x0 cells are just there in the normal dtsi because we need > address- and size-cells of 2. But you don't need that in your example. > > If you change this to <0x07eed000 0x14> you can drop the > address/size-cells and the whole soc {} level from the example. > > > That said, is this really a 20 byte large IP-block? Or is this part of > some larger function? Will we have overlapping/conflicting mappings of > this block? I raised a similar concern, but it seems my concern was misunderstood https://lore.kernel.org/linux-arm-msm/[email protected]/ I think we should still cover the full register space Konrad