On Tue, 30 Mar 2010 10:26:59 +0200 Wolfgang Denk wd@denx.de wrote:
Personally, I'd prefer using Heikos approach until Arm Linux has moved to device trees. I know it's a deviation from how it's supposed to work, but it also solves a real problem without introducing kludges elsewhere.
If we do not even raise issues with the current Linux code with the Linux developers they will not even be aware that there are problems. In the end, things will never change.
I believe they are aware of this especially since many developers work on both projects anyway. If I remember the discussion on ARM device trees a year ago or so correct, this was one of the issues brought up in support of the device trees (or it should have, anyway).
I think most people (myself included) would just "solve" the problem by carrying a private patch to setup the MAC address in U-boot anyway.
Interesting. Why would you do this? Why would you not rather fix the Linux driver instead? [This is what I would do.]
Basically two reasons: First, it's a simpler fix in U-boot (a oneliner for Kirkwood), and secondly because (as far as I understand, correct me if I'm wrong), it lacks any well-defined protocol to transfer this knowledge to the kernel driver.
I know mostly how it looks on the OpenRD board, where the MAC address is stored in the U-boot environment. Easy to access in U-boot, but a lot trickier from Linux. Sure, you could transfer it via a command-line parameter or something, but personally I think this is uglier than setting it up in U-boot anyway.
// Simon