From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?Q?Bj=C3=B8rn_Mork?= Subject: Re: [PATCH] r8152: Add support for setting MAC to system's Auxiliary MAC address Date: Thu, 02 Jun 2016 09:27:28 +0200 Message-ID: <87lh2oyrrz.fsf@nemi.mork.no> References: <1464817844-27206-1-git-send-email-mario_limonciello@dell.com> <20160601230535.GA29279@kroah.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Greg KH , hayeswang@realtek.com, LKML , Netdev , Linux USB , pali.rohar@gmail.com, anthony.wong@canonical.com To: Mario Limonciello Return-path: In-Reply-To: <20160601230535.GA29279@kroah.com> (Greg KH's message of "Wed, 1 Jun 2016 16:05:35 -0700") Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Greg KH writes: > And finally, this seems odd overall given that a MAC address should b= e > associated with the specific network device, not the overall system. Definitely. I wonder if this isn't a perfect candidate for an x86 arch_get_platform_mac_address() implementation? Then you could just us= e the eth_platform_get_mac_address() helper in the driver and avoid any platform specific code there. Which will automagically make it work as expected on a Sparc too :) This will also make your job next year much easier, when the hardeware guys decided to replace the chip and you need to implement the exact same code in some other driver... See https://patchwork.ozlabs.org/patch/564100/ for a detailed discussion of this interface. Bj=C3=B8rn