Re: [PATCH 2/4] Revert "arm64: dts: renesas: Add compatible properties to AR8031 Ethernet PHYs"

From: Geert Uytterhoeven <geert@linux-m68k.org>
To: Adam Ford <aford173@gmail.com>
Cc: linux-renesas-soc@vger.kernel.org, aford@beaconembedded.com,
	Magnus Damm <magnus.damm@gmail.com>,
	Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	Andrew Lunn <andrew@lunn.ch>,
	Heiner Kallweit <hkallweit1@gmail.com>,
	Russell King <linux@armlinux.org.uk>,
	netdev <netdev@vger.kernel.org>
Subject: Re: [PATCH 2/4] Revert "arm64: dts: renesas: Add compatible properties to AR8031 Ethernet PHYs"
Date: Fri, 6 Jan 2023 15:27:56 +0100	[thread overview]
Message-ID: <CAMuHMdXQfAJUVsYeN37T_KvXUoEaSqYJ+UWtUehLv-9R9goVzA@mail.gmail.com> (raw)
In-Reply-To: <20230104141245.8407-2-aford173@gmail.com>

Hi Adam,

CC Ethernet phy

On Wed, Jan 4, 2023 at 3:12 PM Adam Ford <aford173@gmail.com> wrote:
> This reverts commit 18a2427146bf8a3da8fc7825051d6aadb9c2d8fb.
>
> Due to the part shortage, the AR8031 PHY was replaced with a
> Micrel KSZ9131.  Hard-coding the ID of the PHY makes this new
> PHY non-operational.  Since previous hardware had shipped,
> it's not as simple as just replacing the ID number as it would
> break the older hardware.  Since the generic mode can correctly
> identify both versions of hardware, it seems safer to revert
> this patch.
>
> Signed-off-by: Adam Ford <aford173@gmail.com>

Thanks for your patch!

> --- a/arch/arm64/boot/dts/renesas/beacon-renesom-som.dtsi
> +++ b/arch/arm64/boot/dts/renesas/beacon-renesom-som.dtsi
> @@ -59,8 +59,6 @@ &avb {
>         status = "okay";
>
>         phy0: ethernet-phy@0 {
> -               compatible = "ethernet-phy-id004d.d074",
> -                            "ethernet-phy-ieee802.3-c22";
>                 reg = <0>;
>                 interrupt-parent = <&gpio2>;
>                 interrupts = <11 IRQ_TYPE_LEVEL_LOW>;

The next line:

                reset-gpios = <&gpio2 10 GPIO_ACTIVE_LOW>;

Unfortunately, removing the compatible value will cause regressions
for kexec/kdump and for Ethernet driver unbind, as the PHY reset will
be asserted before starting the new kernel, or on driver unbind.
Due to a deficiency in the Ethernet PHY subsystem, the PHY will be
probed while the reset is still asserted, and thus fail probing[1].

Is there a (new) proper way to handle this?
Perhaps the issue has been fixed in the PHY subsystem meanwhile?

Thanks!

[1] https://lore.kernel.org/all/cover.1631174218.git.geert+renesas@glider.be

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds