Messages in this thread | | | Date | Wed, 27 May 2009 16:48:34 -0400 | From | Josh Boyer <> | Subject | Re: [RFC] [PATCH] Device Tree on ARM platform |
| |
On Wed, May 27, 2009 at 10:27:01PM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote: >On 13:23 Wed 27 May , David Miller wrote: >> From: Russell King <rmk+lkml@arm.linux.org.uk> >> Date: Wed, 27 May 2009 20:29:10 +0100 >> >> > To program them into the hardware registers, which is not what we in >> > the ARM community say, but it's what the _network_ guys tell people >> > they should be doing. >> > >> > I've suggested in the past having a standard kernel parameter such >> > that you can specify a mac address on a per-device basis in a totally >> > platform independent way, but the network folk don't like that idea. >> >> As a "network guy" I can tell you that when the system firmware >> provides a system-wide or device specific MAC address in it's device >> tree, that is what you should use. >> >> And that's what I've been doing on sparc FOR 15 YEARS. >> >> Device trees are the only systematic generic mechanism for >> describing device layouts on the myriad of embedded boards >> out there which you will ever find. >> >> Two platforms, and now a third, have been using this scheme >> over a very proven period of time. It's not a hack, there is >> infrastructure to get the boot loaders to do the right thing, >> and only ARM seems to resist it. :-) >The SH, avr32, mips, x86 does not support it too IIRC
x86 has a variant. The OLPC uses OF.
josh
| |