Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220805103605.23096-1-yuji2.ishikawa@toshiba.co.jp>
In-Reply-To: <20220805103605.23096-1-yuji2.ishikawa@toshiba.co.jp>
From:   Oded Gabbay <oded.gabbay@gmail.com>
Date:   Mon, 8 Aug 2022 23:27:34 +0300
Message-ID: <CAFCwf10PL+kO5YcaBTmA1N34tSF18NJ0GLJoJ0L2VG5GuV-6XQ@mail.gmail.com>
Subject: Re: FYI: misc: visconti: Toshiba Visconti DSP accelerator driver sample
To:     Yuji Ishikawa <yuji2.ishikawa@toshiba.co.jp>
Cc:     Nobuhiro Iwamatsu <nobuhiro1.iwamatsu@toshiba.co.jp>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        "list@263.net:IOMMU DRIVERS <iommu@lists.linux-foundation.org>, Joerg
        Roedel <joro@8bytes.org>," <linux-arm-kernel@lists.infradead.org>,
        "Linux-Kernel@Vger. Kernel. Org" <linux-kernel@vger.kernel.org>,
        Linux Media Mailing List <linux-media@vger.kernel.org>,
        Maling list - DRI developers 
        <dri-devel@lists.freedesktop.org>,
        "moderated list:DMA BUFFER SHARING FRAMEWORK" 
        <linaro-mm-sig@lists.linaro.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Aug 5, 2022 at 1:41 PM Yuji Ishikawa
<yuji2.ishikawa@toshiba.co.jp> wrote:
>
> Hello Odded
>
> This is a sample (wip) driver for a DSP found on Toshiba Visconti SoC.
> The DSP typically accepts some images, apply an algorithm on them and yields resulting one.
> Therefore (image-in, image-out), they say this driver should be classified to media driver category.
> However, it can handle various data format (wider than v4l2 officially supports)
> if userland provide firmware (=algorithm) for its own.
>
> Yes, this rough implementation is the first step only our staff could go.
> I'm not for sure whether we could carry out experiments earlier if we chose to use existing frameworks,
> instead of writing our own routine for handling DMA-BUFs and interrupts.
> I hope this post will help your case-study.
>
> Regards
>         Yuji Ishikawa
>
Thanks Yuji.

I've gone over the IOCTLs and they seem fairly straightforward. I
think that we will be able to accommodate at least part of them in the
common code of an accel subsystem.
Oded