Received-SPF: pass (google.com: domain of linux-kernel+bounces-125647-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Date: Sat, 30 Mar 2024 15:00:27 +0530
From: Varadarajan Narayanan <quic_varada@quicinc.com>
To: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
CC: Stephen Boyd <sboyd@kernel.org>, <andersson@kernel.org>,
        <conor+dt@kernel.org>, <devicetree@vger.kernel.org>,
        <djakov@kernel.org>, <dmitry.baryshkov@linaro.org>,
        <konrad.dybcio@linaro.org>, <krzysztof.kozlowski+dt@linaro.org>,
        <linux-arm-msm@vger.kernel.org>, <linux-clk@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>, <linux-pm@vger.kernel.org>,
        <mturquette@baylibre.com>, <quic_anusha@quicinc.com>,
        <robh@kernel.org>
Subject: Re: [PATCH v5 4/5] clk: qcom: ipq9574: Use icc-clk for enabling NoC
 related clocks
Message-ID: <Zgfbs5SFN2cA0gSK@hu-varada-blr.qualcomm.com>
References: <20240328075936.223461-1-quic_varada@quicinc.com>
 <20240328075936.223461-5-quic_varada@quicinc.com>
 <95f4e99a60cc97770fc3cee850b62faf.sboyd@kernel.org>
 <ZgaeGZL7QXh75aSA@hu-varada-blr.qualcomm.com>
 <031d0a35-b192-4161-beef-97b89d5d1da6@linaro.org>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <031d0a35-b192-4161-beef-97b89d5d1da6@linaro.org>

On Fri, Mar 29, 2024 at 01:10:03PM +0100, Krzysztof Kozlowski wrote:
> On 29/03/2024 11:55, Varadarajan Narayanan wrote:
> >>> +
> >>> +enum {
> >>> +       ICC_ANOC_PCIE0,
> >>> +       ICC_SNOC_PCIE0,
> >>> +       ICC_ANOC_PCIE1,
> >>> +       ICC_SNOC_PCIE1,
> >>> +       ICC_ANOC_PCIE2,
> >>> +       ICC_SNOC_PCIE2,
> >>> +       ICC_ANOC_PCIE3,
> >>> +       ICC_SNOC_PCIE3,
> >>> +       ICC_SNOC_USB,
> >>> +       ICC_ANOC_USB_AXI,
> >>> +       ICC_NSSNOC_NSSCC,
> >>> +       ICC_NSSNOC_SNOC_0,
> >>> +       ICC_NSSNOC_SNOC_1,
> >>> +       ICC_NSSNOC_PCNOC_1,
> >>> +       ICC_NSSNOC_QOSGEN_REF,
> >>> +       ICC_NSSNOC_TIMEOUT_REF,
> >>> +       ICC_NSSNOC_XO_DCD,
> >>> +       ICC_NSSNOC_ATB,
> >>> +       ICC_MEM_NOC_NSSNOC,
> >>> +       ICC_NSSNOC_MEMNOC,
> >>> +       ICC_NSSNOC_MEM_NOC_1,
> >>> +};
> >>
> >> Are these supposed to be in a dt-binding header?
> >
> > Since these don't directly relate to the ids in the dt-bindings
> > not sure if they will be permitted there. Will move and post a
> > new version and get feedback.
>
> You can answer this by yourself by looking at your DTS. Do you use them
> as well in the DTS?

I can use them in the DTS. The icc-clk framework automatically
creates master and slave nodes as 'n' and 'n+1'. Hence I can have
something like this in the dt-bindings include file

	#define ICC_ANOC_PCIE0		0
	#define ICC_SNOC_PCIE0		1
		.
		.
		.
	#define ICC_NSSNOC_MEM_NOC_1	20

	#define MASTER(x)	((ICC_ ## x) * 2)
	#define SLAVE(x)	(MASTER(x) + 1)

> It's a pity we see here only parts of DTS, instead of full interconnect
> usage.

Unfortunately cannot include the pcie dts changes with this
patch, but you can refer to them at https://lore.kernel.org/linux-arm-msm/20230519090219.15925-5-quic_devipriy@quicinc.com/

The above macros will be used in the pcie node as follows

pcie0: pci@28000000 {
	compatible = "qcom,pcie-ipq9574";
	. . .
	interconnects = <&gcc MASTER(ANOC_PCIE0) &gcc SLAVE(ANOC_PCIE0)>,
			<&gcc MASTER(SNOC_PCIE0) &gcc SLAVE(SNOC_PCIE0)>;
	interconnect-names = "pcie-mem", "cpu-pcie";
	. . .
};

Hope this is acceptable.

Thanks
Varada