[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1jcyrxi2l0.fsf@starbuckisacylon.baylibre.com>
Date: Thu, 14 Mar 2024 10:26:26 +0100
From: Jerome Brunet <jbrunet@...libre.com>
To: Xianwei Zhao <xianwei.zhao@...ogic.com>
Cc: Jerome Brunet <jbrunet@...libre.com>, Rob Herring <robh+dt@...nel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>, Conor Dooley
<conor+dt@...nel.org>, Neil Armstrong <neil.armstrong@...aro.org>, Martin
Blumenstingl <martin.blumenstingl@...glemail.com>, Kevin Hilman
<khilman@...libre.com>, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-amlogic@...ts.infradead.org
Subject: Re: [PATCH 3/4] arm64: dts: add support for A4 based Amlogic BA400
On Thu 14 Mar 2024 at 16:08, Xianwei Zhao <xianwei.zhao@...ogic.com> wrote:
>>> +
>>> +#include "amlogic-a4.dtsi"
>> Could you describe how the a4 and a5 differs from each other ?
>> The description given in the commit description is the same.
>> Beside the a53 vs a55, I'm not seeing much of a difference.
>> Admittedly, there is not much yet but I wonder if a4 and a5 should have
>> a common dtsi.
>>
> They are mostly the same, A5 include HiFi-DSP and NPU, but A4 is not. And
> some peripheral modules are different, such as SPI and Ehernet phy.
>
> I would like to wait for the follow-on chips to come out before considering
> a merger with common dtsi file.
>
No, Please do it now. There is no reason for the community to review the
same thing twice if the SoCs are "mostly the same".
>>> +
>>> +/ {
>>> + model = "Amlogic A113L2 ba400 Development Board";
>>> + compatible = "amlogic,ba400","amlogic,a4";
>>> + interrupt-parent = <&gic>;
>>> + #address-cells = <2>;
>>> + #size-cells = <2>;
>>> +
>>> + aliases {
>>> + serial0 = &uart_b;
>>> + };
>>> +
>>> + memory@0 {
>>> + device_type = "memory";
>>> + reg = <0x0 0x0 0x0 0x40000000>;
>>> + };
>>> +
>>> + reserved-memory {
>>> + #address-cells = <2>;
>>> + #size-cells = <2>;
>>> + ranges;
>>> +
>>> + /* 52 MiB reserved for ARM Trusted Firmware */
>> That's a lot of memory to blindly reserve.
>> Any chance we can stop doing that and have u-boot amend reserved memory
>> zone based on the actual needs of the device ?
> Yes. U-boot will change size of reserved memory base on actual usage.
Then u-boot should add (not change) the memory if necessary.
Please drop this.
>>
>>> + secmon_reserved:linux,secmon {
>>> + compatible = "shared-dma-pool";
>>> + no-map;
>>> + alignment = <0x0 0x400000>;
>>> + reg = <0x0 0x05000000 0x0 0x3400000>;
>>> + };
>>> + };
>>> +};
>>> +
Powered by blists - more mailing lists