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;
Date:   Thu, 12 Nov 2020 00:59:20 -0800
From:   Prashant Malani <pmalani@chromium.org>
To:     Pi-Hsun Shih <pihsun@chromium.org>
Cc:     Nicolas Boichat <drinkcat@chromium.org>,
        Andrzej Hajda <a.hajda@samsung.com>,
        Neil Armstrong <narmstrong@baylibre.com>,
        Laurent Pinchart <Laurent.pinchart@ideasonboard.com>,
        Jonas Karlman <jonas@kwiboo.se>,
        Jernej Skrabec <jernej.skrabec@siol.net>,
        David Airlie <airlied@linux.ie>,
        Daniel Vetter <daniel@ffwll.ch>, Xin Ji <xji@analogixsemi.com>,
        Sam Ravnborg <sam@ravnborg.org>,
        "open list:DRM DRIVERS" <dri-devel@lists.freedesktop.org>,
        open list <linux-kernel@vger.kernel.org>,
        heikki.krogerus@linux.intel.com
Subject: Re: [PATCH] drm/bridge: anx7625: Add anx7625 port switching.
Message-ID: <20201112085920.GB1367855@google.com>
References: <20201112064051.3716968-1-pihsun@chromium.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20201112064051.3716968-1-pihsun@chromium.org>
Precedence: bulk

Hi Pi-Hsun,

I haven't gone through the code, but did have a high-level comment
(kindly see inline)

On Thu, Nov 12, 2020 at 02:40:40PM +0800, Pi-Hsun Shih wrote:
> When output 2 lanes DP data, anx7625 can output to either TX1/RX1 or
> TX2/RX2. In typical usage, these two TX/RX pairs corresponds to two
> orientations of typec.
> 
> On some board one anx7625 is used as DPI to DP converter for two typec
> ports. In this case, the TX1/RX1 and TX2/RX2 are connected to two usb
> muxes, which mux the DP data with the rest of the USB3 data, and
> connects to the two typec ports.
> 
> This patch adds option for anx7625 to acts as a usb typec switch and
> switch output lanes based on the typec orientation, or acts as two usb
> typec mux and switch output lanes depending on whether the two ports
> currently has DP enabled.
> 
> Signed-off-by: Pi-Hsun Shih <pihsun@chromium.org>
> 
> ====================================================================
> 
> This is an attempt to use typec framework with how we're using anx7625
> on Chrome OS asurada board.
> 
> An example of the dts for the two ports case can be found at
> https://crrev.com/c/2507199/6

Do you plan on submitting DT schemas & bindings documentation for the switch(es)
that are intended to be used?

I would strongly recommend that for usb-c-connector since AFAIK they don't exist, and
I don't believe there is explicit support for them in the Type C connector class framework
(even .

IMO this would be needed to ensure an implementation here doesn't break
in the event of modifications to the connector class framework (or Type
C port drivers like cros-ec-typec) in the future. I think some patches
were floated for this for orientation switch [1] so those might provide
some hints about how to proceed.

I've CC-ed Heikki (Type C maintainer) in case he has additional comments regarding this.

> 
> Sending this as a RFC patch since I'm not sure about the best approach
> here. Should the logic of switching output lanes depends on ports be
> coupled inside anx7625 driver, or in another driver, or is there
> any existing way to accomplish this?

Might be good to add [RFC] as a tag instead of [PATCH] in case this
iteration is chiefly to solicit comments.

Best regards,

-Prashant

[1]:
https://lore.kernel.org/linux-usb/1604403610-16577-1-git-send-email-jun.li@nxp.com/
>