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]
Message-ID: <Y4eZg56XBWwR+pkr@x1>
Date:   Wed, 30 Nov 2022 12:57:23 -0500
From:   Brian Masney <bmasney@...hat.com>
To:     irusskikh@...vell.com, davem@...emloft.net, edumazet@...gle.com,
        kuba@...nel.org, pabeni@...hat.com
Cc:     netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
        cth451@...il.com
Subject: Re: [PATCH] net: atlantic: fix check for invalid ethernet addresses

On Wed, Nov 30, 2022 at 12:42:59PM -0500, Brian Masney wrote:
> The Qualcomm sa8540p automotive development board (QDrive3) has an
> Aquantia NIC wired over PCIe. The ethernet MAC address assigned to
> all of the boards in our lab is 00:17:b6:00:00:00. The existing
> check in aq_nic_is_valid_ether_addr() only checks for leading zeros
> in the MAC address. Let's update the check to also check for trailing
> zeros in the MAC address so that a random MAC address is assigned
> in this case.
> 
> Signed-off-by: Brian Masney <bmasney@...hat.com>

I have a question about the original commit that introduced this check:
553217c24426 ("ethernet: aquantia: Try MAC address from device tree").
The commit message talks about getting the MAC address from device tree,
however I don't see any compatible lines in this driver, nor a
of_match_table. As far as I can tell, this driver is only setup to be
accessed over PCIe.

The random MAC address is not ideal for our lab since we'd like to have
stable addresses. I'd like to have the bootloader be able to inject a
MAC address that's generated based on the board's serial number. I
assume that it would go in the chosen node in device tree. One of the
issues is that there are multiple NICs on this board, so I'm not sure
how that would go in the chosen node and identify this particular NIC.
Does anyone know of a place in the kernel where this is already done?

Brian

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ