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;
MIME-Version: 1.0
References: <1596855231-5782-1-git-send-email-yongqiang.niu@mediatek.com> <1596855231-5782-4-git-send-email-yongqiang.niu@mediatek.com>
In-Reply-To: <1596855231-5782-4-git-send-email-yongqiang.niu@mediatek.com>
From:   Chun-Kuang Hu <chunkuang.hu@kernel.org>
Date:   Tue, 11 Aug 2020 07:35:10 +0800
Message-ID: <CAAOTY____7Yd1Gh-22oiOxsvftBjOHQ0T9CmKtGw6XLjw11gcQ@mail.gmail.com>
Subject: Re: [RESEND v7, PATCH 3/7] mtk-mmsys: add mmsys private data
To:     Yongqiang Niu <yongqiang.niu@mediatek.com>
Cc:     CK Hu <ck.hu@mediatek.com>, Philipp Zabel <p.zabel@pengutronix.de>,
        Rob Herring <robh+dt@kernel.org>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        Mark Rutland <mark.rutland@arm.com>,
        devicetree@vger.kernel.org, David Airlie <airlied@linux.ie>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        DRI Development <dri-devel@lists.freedesktop.org>,
        "moderated list:ARM/Mediatek SoC support" 
        <linux-mediatek@lists.infradead.org>,
        Daniel Vetter <daniel@ffwll.ch>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi, Yongqiang:

Yongqiang Niu <yongqiang.niu@mediatek.com> =E6=96=BC 2020=E5=B9=B48=E6=9C=
=888=E6=97=A5 =E9=80=B1=E5=85=AD =E4=B8=8A=E5=8D=8810:56=E5=AF=AB=E9=81=93=
=EF=BC=9A
>
> the reason why split out display connection function:
> 1. there will more and more Mediatek Soc upstream, and the display path
> connection function mtk_mmsys_ddp_mout_en, mtk_mmsys_ddp_sel_in and
> mtk_mmsys_ddp_sout_sel will complicated more and more
> 2. many of the connection are only used in some SoC, and useless for
> other SoC and not readable,
> 3. if we add a new SoC connection, we need check is this affect other
> Soc
>
> the reason why not apply the previous series method:
> this version is more readable and clear
> if go on use v6, except mt2701/mt8173/mt2712, we need add two more
> private data array for mt6779 and mt6797, and the connect function
> will add more if/else usecase
>
> move current display connection function into mt2701-mmsys.c
> keep mt2701/mt8173/mt2712/mt6779/mt6797 with original version
> connection function
> the corresponded SoC upstream member will coding these and test it
> on the SoC if it is need.
>

For this patch, I prefer [1]'s implementation. In [1], for each SoC,
the only difference is an array for routing register setting. In this
patch, the difference are three function which is more complicated
than [1]. If you like, you could pick [1] to upstream.

[1] https://chromium-review.googlesource.com/c/chromiumos/third_party/kerne=
l/+/2345186

Regards,
Chun-Kuang.

> Signed-off-by: Yongqiang Niu <yongqiang.niu@mediatek.com>
> ---
>  drivers/soc/mediatek/Makefile             |   1 +
>  drivers/soc/mediatek/mmsys/Makefile       |   2 +
>  drivers/soc/mediatek/mmsys/mt2701-mmsys.c | 250 ++++++++++++++++++++++++=
+++
>  drivers/soc/mediatek/mtk-mmsys.c          | 271 +++++-------------------=
------
>  include/linux/soc/mediatek/mtk-mmsys.h    |  15 ++
>  5 files changed, 314 insertions(+), 225 deletions(-)
>  create mode 100644 drivers/soc/mediatek/mmsys/Makefile
>  create mode 100644 drivers/soc/mediatek/mmsys/mt2701-mmsys.c
>