Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
Message-ID: <dc3dd045-27d4-b684-e812-dc95faf0df50@quicinc.com>
Date:   Fri, 29 Sep 2023 16:42:15 +0530
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.5.1
Subject: Re: [PATCH 2/2] arm64: dts: qcom: qcm6490: Add qcm6490 dts file
Content-Language: en-US
To:     Konrad Dybcio <konrad.dybcio@linaro.org>,
        Andy Gross <agross@kernel.org>,
        Bjorn Andersson <andersson@kernel.org>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Conor Dooley <conor+dt@kernel.org>
CC:     <linux-arm-msm@vger.kernel.org>, <devicetree@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>,
        Luca Weiss <luca.weiss@fairphone.com>
References: <20230928133312.11371-1-quic_kbajaj@quicinc.com>
 <20230928133312.11371-3-quic_kbajaj@quicinc.com>
 <661bd908-a056-4d46-97a3-d3b12b14c050@linaro.org>
From:   Komal Bajaj <quic_kbajaj@quicinc.com>
In-Reply-To: <661bd908-a056-4d46-97a3-d3b12b14c050@linaro.org>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 9/28/2023 7:16 PM, Konrad Dybcio wrote:
> On 28.09.2023 15:33, Komal Bajaj wrote:
>> Add qcm6490 devicetree file for QCM6490 SoC and QCM6490 IDP
>> platform. QCM6490 is derived from SC7280 meant for various
>> form factor including IoT.
>>
>> Supported features are, as of now:
>> * Debug UART
>> * eMMC
>> * USB
>>
>> Signed-off-by: Komal Bajaj <quic_kbajaj@quicinc.com>
>> ---
> [...]
>
>> +/ {
>> +	model = "Qualcomm Technologies, Inc. QCM6490 IDP platform";
> Isomething Development Platform platform?

Will drop suffix platform.

>
>> +	compatible = "qcom,qcm6490-idp", "qcom,qcm6490";
>> +
>> +	aliases {
>> +		serial0 = &uart5;
>> +	};
>> +
>> +	chosen {
>> +		stdout-path = "serial0:115200n8";
>> +	};
>> +
>> +};
> Stray newline above

Noted.

>
>> +
>> +&apps_rsc {
>> +	regulators-0 {
>> +		compatible = "qcom,pm7325-rpmh-regulators";
>> +		qcom,pmic-id = "b";
>> +
>> +		vreg_s1b_1p8: smps1 {
>> +			regulator-min-microvolt = <1856000>;
>> +			regulator-max-microvolt = <2040000>;
> Hm, you didn't specify regulator-initial-mode on any vregs

Okay, will specify it in next patchset.

>
> [...]
>
>> --- /dev/null
>> +++ b/arch/arm64/boot/dts/qcom/qcm6490.dtsi
>> @@ -0,0 +1,137 @@
>> +// SPDX-License-Identifier: BSD-3-Clause
>> +/*
>> + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
>> + */
>> +
>> +#include "sc7280.dtsi"
>> +
>> +/*
>> + * Delete sc7280 lpass related nodes as qcm6490 intends to move away from
>> + * bypass configuration
>> + */
>> +/delete-node/ &lpass_ag_noc;
>> +/delete-node/ &lpass_aon;
>> +/delete-node/ &lpass_audiocc;
>> +/delete-node/ &lpass_core;
>> +/delete-node/ &lpass_cpu;
>> +/delete-node/ &lpass_hm;
>> +/delete-node/ &lpass_rx_macro;
>> +/delete-node/ &lpass_tx_macro;
>> +/delete-node/ &lpass_va_macro;
>> +/delete-node/ &lpass_tlmm;
>> +/delete-node/ &lpasscc;
>> +/delete-node/ &swr0;
>> +/delete-node/ &swr1;
> That's very unnecessary, most of these nodes are in use even
> when routing audio through ADSP.
>
> Ones that are not, are set to status = "reserved" and some
> will need more work to function based on the configuration.
>
> There was once a series from somebody at qc to introduce ADSP
> audio on 7280, but it was full of hacks and NAKed

Seeking clarification which nodes will be used and which should be disabled.
For now, will remove it.

>
>
>> +
>> +/*
>> + * Delete unused sc7280 memory nodes and define the memory regions
>> + * required by qcm6490
>> + */
> That's specific to your board.

No, these are not board specific, it's soc specific.

>
>> +/delete-node/ &rmtfs_mem;
>> +/delete-node/ &wlan_ce_mem;
>> +
>> +/{
>> +	reserved-memory {
>> +		secdata_apss_mem: secdata-apss@808ff000 {
>> +			reg = <0x0 0x808ff000 0x0 0x1000>;
>> +			no-map;
>> +		};
> This is identical to the entry in sc7280.dtsi.

Yeah, right. Will remove it.

>
>> +
>> +		cdsp_secure_heap_mem: cdsp-secure-heap@81800000 {
>> +			reg = <0x0 0x81800000 0x0 0x1e00000>;
>> +			no-map;
>> +		};
>> +
>> +		camera_mem: camera@84300000 {
>> +			reg = <0x0 0x84300000 0x0 0x500000>;
>> +			no-map;
>> +		};
>> +
>> +		adsp_mem: adsp@86100000 {
>> +			reg = <0x0 0x86100000 0x0 0x2800000>;
>> +			no-map;
>> +		};
>> +
>> +		cdsp_mem: cdsp@88900000 {
>> +			reg = <0x0 0x88900000 0x0 0x1e00000>;
>> +			no-map;
>> +		};
>> +
>> +		cvp_mem: cvp@8ac00000 {
>> +			reg = <0x0 0x8ac00000 0x0 0x500000>;
>> +			no-map;
>> +		};
>> +
>> +		ipa_gsi_mem: ipa-gsi@8b110000 {
>> +			reg = <0x0 0x8b110000 0x0 0xa000>;
>> +			no-map;
>> +		};
>> +
>> +		gpu_microcode_mem: gpu-microcode@8b11a000 {
>> +			reg = <0x0 0x8b11a000 0x0 0x2000>;
>> +			no-map;
>> +		};
>> +
>> +		mpss_mem: mpss@8b800000 {
>> +			reg = <0x0 0x8b800000 0x0 0xf600000>;
>> +			no-map;
>> +		};
>> +
>> +		wpss_mem: wpss@9ae00000 {
>> +			reg = <0x0 0x9ae00000 0x0 0x1900000>;
>> +			no-map;
>> +		};
> This entry is in both chrome-common and fairphone (meaning all boards
> use it), perhaps this one could be moved to 7280.dtsi

Missed to update region for wpss_mem. It's different what's used in 
chrome-common and fairphone.

>
>> +
>> +		tz_stat_mem: tz-stat@c0000000 {
>> +			reg = <0x0 0xc0000000 0x0 0x100000>;
>> +			no-map;
>> +		};
>> +
>> +		tags_mem: tags@c0100000 {
>> +			reg = <0x0 0xc0100000 0x0 0x1200000>;
>> +			no-map;
>> +		};
>> +
>> +		qtee_mem: qtee@c1300000 {
>> +			reg = <0x0 0xc1300000 0x0 0x500000>;
>> +			no-map;
>> +		};
>> +
>> +		trusted_apps_mem: trusted_apps@c1800000 {
>> +			reg = <0x0 0xc1800000 0x0 0x3900000>;
>> +			no-map;
>> +		};
>> +	};
>> +};
>> +
>> +&hyp_mem {
>> +	reg = <0x0 0x80000000 0x0 0x600000>;
> This is identical to the entry in sc7280.dtsi.

Will remove it.

>
>> +};
>> +
>> +&mpss_mem {
>> +	reg = <0x0 0x8b800000 0x0 0xf600000>;
> You're defining it here and overwriting it with an identical
> value.
>
> Looks like CrOS folks don't boot up the modem on non-LTE SKUs.
> Weird. Normally it would host some more software..
>
>> +};
>> +
>> +&remoteproc_mpss {
>> +	memory-region = <&mpss_mem>;
> This is identical to the entry in sc7280.dtsi.

Will remove it.

>
>> +};
>> +
>> +&video_mem {
>> +	reg = <0x0 0x8a700000 0x0 0x500000>;
>> +};
>> +
>> +&wifi {
>> +	memory-region = <&wlan_fw_mem>;
> No CE region?

Yes, no CE region is required here.

>
>> +};
>> +
>> +&wlan_fw_mem {
>> +	reg = <0x0 0x80c00000 0x0 0xc00000>;
> This is identical to the entry in sc7280.dtsi.

Will remove it.

>
>
> The memory map generally looks quite different to both chrome
> and fairphone.. are you sure it's all correct?

Yes, it's correct, it's different than chrome and fairphone.

Thanks
Komal

>
> Konrad