Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH] arm64: dts: meson: vim3l: add NNA node
To:     Martin Blumenstingl <martin.blumenstingl@googlemail.com>,
        xieqinick@gmail.com
Cc:     robh+dt@kernel.org, khilman@baylibre.com, jbrunet@baylibre.com,
        devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
        linux-amlogic@lists.infradead.org, linux-kernel@vger.kernel.org,
        nick@khadas.com, artem@khadas.com
References: <20210605030915.411391-1-xieqinick@gmail.com>
 <CAFBinCA+Rm3X43fT4oePS0vXphPPEZMweoM_7c0MR1ZwSbsoaQ@mail.gmail.com>
From:   Neil Armstrong <narmstrong@baylibre.com>
Organization: Baylibre
Message-ID: <f42f343d-0d32-bbd8-9066-17b72b69f99a@baylibre.com>
Date:   Mon, 7 Jun 2021 09:03:49 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.8.1
MIME-Version: 1.0
In-Reply-To: <CAFBinCA+Rm3X43fT4oePS0vXphPPEZMweoM_7c0MR1ZwSbsoaQ@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi,

On 05/06/2021 11:49, Martin Blumenstingl wrote:
> Hi Nick,
> 
> On Sat, Jun 5, 2021 at 5:09 AM <xieqinick@gmail.com> wrote:
>>
>> From: Nick Xie <nick@khadas.com>
>>
>> Add Neural Network Accelerator device tree node
>> for Khadas VIM3L.
> we neither have a driver for this upstream nor any binding
> documentation in Documentation/devicetree/bindings/
> are you planning to add this?
> 
> Especially the missing binding documentation is problematic as it
> causes some warnings when validating the .dts files against the
> bindings from Documentation/devicetree/bindings/

Actually the NNA core is a verisilicon VIVANTE core, but I have no idea what is the difference
with the VIVANTE GPU cores ?

the userspace libraries & SDK shipped by Amlogic exposes OpenCL and (AFAIK) NNA over OpenCL, so I think
the HW description should be similar to the current Etnaviv bindings.

Neil

> 
> 
> Best regards,
> Martin
>