netdev - Re: net: phy: micrel KSZ9031 ifdown ifup issue

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 for Android: free password hash cracker in your pocket

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

Date:   Sat, 14 Sep 2019 16:54:43 +0200
From:   Andrew Lunn <andrew@...n.ch>
To:     Paul Thomas <pthomas8589@...il.com>
Cc:     Florian Fainelli <f.fainelli@...il.com>,
        Heiner Kallweit <hkallweit1@...il.com>,
        "David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org
Subject: Re: net: phy: micrel KSZ9031 ifdown ifup issue

On Fri, Sep 13, 2019 at 10:42:38AM -0400, Paul Thomas wrote:
> Hello,
> 
> I think I'm seeing an issue with the PHY hardware or PHY driver. What
> happens is sometimes (but not always) when I do 'ip link set eth0
> down' followed by 'ip link set eth0 up' I don't ever see an
> auto-negotiation again. LEDs don't come on, ethtool reports 'Link
> detected: no'. Even physically unplugging and plugging the network
> cable doesn't bring it back. I have to do a reboot to get the
> networking back.
> 
> When the networking is started I don't see any issue forcing
> negotiations by unplugging and plugging the cable. I get standard
> messages like this all day long:
> [   21.031793] 003: macb ff0b0000.ethernet eth0: link down
> [   26.142835] 003: macb ff0b0000.ethernet eth0: link up (1000/Full)
> 
> One thing that makes me think this is the PHY is that we have another
> Ethernet port using the DP83867 PHY and I can always do ifdown/ifup
> with it.
> 
> This is using a 5.2.10 kernel on arm64 zynqmp platform with the macb driver.
> 
> Is this something anyone else has seen? I know there is some Errata
> with this part, but I'm hoping there is something to fix or work
> around this. Any thoughts on where to look or add debugging would
> appreciated.

Hi Paul

Are you using interrupts, or polling? If interrupts, try polling?
Seems unlikely, but you could be missing an interrupt.

There is a fix from Antoine Tenart which suggests asym pause can be an
issue? What pause setup are you using? But this is a known issue,
which 5.2 should have the fix for.

    Andrew