home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Turnbull China Bikeride
/
Turnbull China Bikeride - Disc 2.iso
/
BARNET
/
ARMLINUX
/
MAIL
/
9709
/
000050_alan@snowcrash.cymru.net _Mon Sep 22 15:27:29 1997.msg
< prev
next >
Wrap
Internet Message Format
|
1997-11-30
|
2KB
Return-Path: <alan@snowcrash.cymru.net>
Received: from snowcrash.cymru.net (snowcrash.cymru.net [163.164.160.3])
by odie.barnet.ac.uk (8.8.6/8.8.6) with ESMTP id PAA03734
for <willy@odie.barnet.ac.uk>; Mon, 22 Sep 1997 15:27:25 +0100
Received: (from alan@localhost) by snowcrash.cymru.net (8.8.5-q-beta3/8.7.1) id PAA09911; Mon, 22 Sep 1997 15:25:53 +0100
From: Alan Cox <alan@cymru.net>
Message-Id: <199709221425.PAA09911@snowcrash.cymru.net>
Subject: Re: Latest kernel in development?
To: willy@odie.barnet.ac.uk (Matthew Wilcox)
Date: Mon, 22 Sep 1997 15:25:51 +0100 (BST)
Cc: alan@cymru.net, linux-arm@vger.rutgers.edu
In-Reply-To: <199709221417.PAA03707@odie.barnet.ac.uk> from "Matthew Wilcox" at Sep 22, 97 03:17:31 pm
Content-Type: text
Status: RO
> ARM have an officially allocated ELF number.
> They have defined the E(xecutable) stuff, but not the L(inkable) stuff.
> They do intend to do so at some undefined stage in the future.
> They would be 'unhappy' if there were a competing standard. Even if it
> uses a different magic number.
> In particular, they wouldn't be happy for it to be released.
Tough shit. Sun aren't terribly happy about Sparc Linux either.
> We could use the ARM number and come up with our own definitions for the
> Linking standard and hope that ours becomes accepted by everyone. This
> would upset ARM a lot and probably cause problems if we ever changed over
> to an ARM-sanctioned one that was different to our proposed one.
Better to use the ARM number and spec for executables and a random number
for linking. gnu ld supports multiple link formats at once so it can then
link old and new format.
> Or we could wait for ARM to get their act together.
Every minute you wait another vendor doesnt bother using Linux and/or ARM.
We've made a decision not to use ARM almost solely on the lack of ELF tools.
Im sure NatSemi and Intel are happy with ARM right now.
Alan