[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <48c03b5a-6557-9eee-8b85-24d72cfbfcd1@ti.com>
Date: Wed, 29 Mar 2023 15:19:33 -0500
From: Hari Nagalla <hnagalla@...com>
To: Nishanth Menon <nm@...com>
CC: <vigneshr@...com>, <kristo@...nel.org>, <robh+dt@...nel.org>,
<krzysztof.kozlowski+dt@...aro.org>,
<linux-arm-kernel@...ts.infradead.org>,
<devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/3] arm64: dts: ti: k3-j784s4-main: Add MAIN domain R5F
cluster nodes
On 3/29/23 07:52, Nishanth Menon wrote:
>> MAIN R5FSS0 Core0: j784s4-main-r5f0_0-fw (both in LockStep and Split modes)
>> MAIN R5FSS0 Core1: j784s4-main-r5f0_1-fw (needed only in Split mode)
>> MAIN R5FSS1 Core0: j784s4-main-r5f1_0-fw (both in LockStep and Split modes)
>> MAIN R5FSS1 Core1: j784s4-main-r5f1_1-fw (needed only in Split mode)
>> MAIN R5FSS2 Core0: j784s4-main-r5f2_0-fw (both in LockStep and Split modes)
>> MAIN R5FSS2 Core1: j784s4-main-r5f2_1-fw (needed only in Split mode)
> Why are the patches split up into main and mcu - if you are adding r5f
> cores, do them as a single patch.
>
Thought would be cleaner with separated patches for resolving potential
merge conflicts. But, can combine into one for v2.
>> +
>> + main_r5fss0: r5fss@...0000 {
>> + compatible = "ti,j721s2-r5fss";
>> + ti,cluster-mode = <1>;
>> + #address-cells = <1>;
>> + #size-cells = <1>;
>> + ranges = <0x5c00000 0x00 0x5c00000 0x20000>,
>> + <0x5d00000 0x00 0x5d00000 0x20000>;
>> + power-domains = <&k3_pds 336 TI_SCI_PD_EXCLUSIVE>;
>> +
>> + main_r5fss0_core0: r5f@...0000 {
>> + compatible = "ti,j721s2-r5f";
>> + reg = <0x5c00000 0x00010000>,
>> + <0x5c10000 0x00010000>;
>> + reg-names = "atcm", "btcm";
>> + ti,sci = <&sms>;
>> + ti,sci-dev-id = <339>;
>> + ti,sci-proc-ids = <0x06 0xff>;
>> + resets = <&k3_reset 339 1>;
>> + firmware-name = "j784s4-main-r5f0_0-fw";
>> + ti,atcm-enable = <1>;
>> + ti,btcm-enable = <1>;
>> + ti,loczrama = <1>;
>> + status = "disabled";
> Why are these disabled by default?
Well, the idea is to let the board specific device tree enable needed
remote core nodes in *-evm/sk.dts and disable by default in SoC device
tree files by default.
Powered by blists - more mailing lists