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: <20200716172611.5349-1-bruno.thomsen@gmail.com>
 <20200716172611.5349-2-bruno.thomsen@gmail.com> <CAOMZO5B7cJmF4e78xbErzBHfMw4MuF09eXti4xAUFcKOGKXqNw@mail.gmail.com>
In-Reply-To: <CAOMZO5B7cJmF4e78xbErzBHfMw4MuF09eXti4xAUFcKOGKXqNw@mail.gmail.com>
From:   Bruno Thomsen <bruno.thomsen@gmail.com>
Date:   Thu, 16 Jul 2020 21:32:06 +0200
Message-ID: <CAH+2xPCa2GNgtJ_5hr3dcaXzA+EB4JLfCsqDJQQY4pbxuw-4JQ@mail.gmail.com>
Subject: Re: [PATCH v2 2/2] ARM: dts: imx7: add support for kamstrup flex concentrator
To:     Fabio Estevam <festevam@gmail.com>
Cc:     linux-kernel <linux-kernel@vger.kernel.org>,
        "moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE" 
        <linux-arm-kernel@lists.infradead.org>,
        "open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS" 
        <devicetree@vger.kernel.org>, Shawn Guo <shawnguo@kernel.org>,
        Pengutronix Kernel Team <kernel@pengutronix.de>,
        NXP Linux Team <linux-imx@nxp.com>,
        Bruno Thomsen <bth@kamstrup.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi Fabio,

Den tor. 16. jul. 2020 kl. 20.01 skrev Fabio Estevam <festevam@gmail.com>:
> On Thu, Jul 16, 2020 at 2:26 PM Bruno Thomsen <bruno.thomsen@gmail.com> wrote:
>
> > Limitations: Ethernet PHY type auto detection does not
> > work when using reset-{assert-us,deassert-us,gpios}
> > properties so it's using a fixed PHY type ID for now. Auto
> > detection worked when using the deprecated FEC properties
> > phy-reset-{gpios,duration,post-delay}.
>
> I think we need to understand this better. Why does it fail?

Yes, we need to locate the root cause.

> I gave a test on an imx6q-sabresd and the Ethernet PHY (AR8031) could
> be properly detected using reset-{assert-us,deassert-us,gpios}
> properties inside the mdio node.

Okay, thanks.

> Is this a Micrel KSZ8081 specific issue?

Maybe, I'm actually beginning to suspect that the issue might be related
to polling vs interrupt mode. As all partially related examples I have seen
uses interrupt mode and I have only configured poll mode. But the hardware
does have the interrupt signal connected to the imx7.

> Please report this issue to the Ethernet PHY folks.

I'm going to send an Ethernet PHY mail tomorrow with the details I know
at the moment. Just wanted to get an updated version of the device tree
out in case of any other issues not related to the PHY part.

/Bruno