netdev - Re: [PATCH] net: phy: return EPROBE_DEFER if PHY is not accessible

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 20 Mar 2023 10:12:10 +0000
From:   "Russell King (Oracle)" <linux@...linux.org.uk>
To:     "Buzarra, Arturo" <Arturo.Buzarra@...i.com>
Cc:     Heiner Kallweit <hkallweit1@...il.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Florian Fainelli <f.fainelli@...il.com>,
        Andrew Lunn <andrew@...n.ch>
Subject: Re: [PATCH] net: phy: return EPROBE_DEFER if PHY is not accessible

On Mon, Mar 20, 2023 at 09:45:38AM +0000, Buzarra, Arturo wrote:
> Hi,
> 
> I will try to answer all your questions:
> 
> - "We need more specifics here, what type of PHY device are you seeing this with?"
> - " So best start with some details about your use case, which MAC, which PHY, etc"
> I'm using a LAN8720A PHY (10/100 on RMII mode) with a ST MAC ( in particular is a stm32mp1 processor).

I can only find:

arch/arm/boot/dts/rk3066a-marsboard.dts:        lan8720a {
arch/arm/boot/dts/rk3188-radxarock.dts: lan8720a  {
arch/arm/boot/dts/imx6sx-softing-vining-2000.dts:                       /* LAN8720 PHY Reset */
arch/arm/boot/dts/imx6sx-softing-vining-2000.dts:                       /* LAN8720 PHY Reset */
arch/mips/boot/dts/ingenic/cu1000-neo.dts:      phy-handle = <&lan8720a>;
arch/mips/boot/dts/ingenic/cu1000-neo.dts:      lan8720a: ethernet-phy@0 {

using this PHY in the mainline kernel, none of them look like a stm32mp1
processor.

Please can you give details of how the MDIO bus is connected to this
PHY, in particular the hardware setup, and which driver is being used?

> We have two PHYs one is a Gigabit PHY (RGMII mode) and the another one is a 10/100 (RMII mode).
> In the boot process, I think that there is a race condition between configuring the Ethernet MACs and the two PHYs. At same point the RGMII Ethernet MAC is configured and starts the PHY probes.
> When the 10/100 PHY starts the probe, it tries to read the genphy_read_abilities() and always reads 0xFFFF ( I assume that this is the default electrical values for that lines before it are configured).
> At that point, the PHY initialization assumes like a valid value 0xFFFF and obviously it reports capabilities that the LAN8720A PHY does not have, like for example gigabit support.

So the questions that need to be asked are:

  What is causing 0xffff to be returned?

and two sub-questions on that which may help you answer it in a way
that is most relevant:

  Is it the PHY itself that is not responding?

  Is the MDIO driver not functional because the MAC or some resource it
  needs hasn't been setup, and thus is returning 0xffff?

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!