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  PHC 
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Date:   Wed, 8 Apr 2020 17:49:31 +0300
From:   "Matwey V. Kornilov" <>
To:, Michal Simek <>,, Anton Gerasimov <>,,,
        Linux ARM Kernel List <>
Subject: zynq-zturn: macb: Could not attach PHY


I am running Linux 5.6 on MYIR Z-turn board and see the following issue:

[   38.485960] macb e000b000.ethernet eth0: Cadence GEM rev 0x00020118
at 0xe000b000 irq 28 (a6:78:9c:86:65:d3)
[   42.990885] macb e000b000.ethernet eth0: Could not attach PHY (-19)

Setting 0x3 for ethernet-phy address in zynq-zturn.dts fixes this issue:

[   23.445231] macb e000b000.ethernet eth0: Cadence GEM rev 0x00020118
at 0xe000b000 irq 28 (5a:77:b7:82:c4:b3)
[   27.843706] macb e000b000.ethernet eth0: PHY
[e000b000.ethernet-ffffffff:03] driver [Micrel KSZ9031 Gigabit PHY]
[   27.843726] macb e000b000.ethernet eth0: configuring for
phy/rgmii-id link mod

However, I've found that there are at least two Z-turn board
revisions. The first one has v4 schematics as described in
The second one has v5 schematics. I've found v5 schematics PDF file at
DVD disk supplied with my board. I am not sure whether I am allowed to
attach it here.
My board seems to be v5 schematics. The only described difference
between board revisions is that v4 has Atheros AR8035 PHY at 0b000
address, and v5 has Micrel KSZ9031 PHY at 0b011 address.

What should be preferred fix for this issue? I have not found a way to
specify a list of PHY addresses to probe in DTS file.
u-boot has similar issue with this board [1]. While it can be
workarounded by PHY auto scan for u-boot, in Linux
drivers/of/of_mdio.c, I see that auto scan is generally not


With best regards,
Matwey V. Kornilov

Powered by blists - more mailing lists