Re: More ARM binutils fuckage

["Michael K. Edwards" <medwards dot linux at gmail dot com>]

On 12/6/06, Daniel Jacobowitz <dan@debian.org> wrote:
> Why is this suddenly crossposted?

Sorry, that's my fault.  I didn't really mean to add two linux-arm-*
lists, and I cut linux-kernel (where Russell originally posted) after
the first follow-up.  It's copied to crossgcc because there have been
a lot of inquiries (and patches posted) there relating to EABI and
soft-float, and some of us who have posted there have been getting
private inquiries too.

> On Wed, Dec 06, 2006 at 04:34:44AM +0100, Lennert Buytenhek wrote:
> > You pass your old-ABI compiler the option -mabi=aapcs-linux, which works
> > fine with my gcc 4.1 old-ABI toolchain and is exactly what mainline 2.6
> > does.
>
> I don't recommend doing this.  The two compilers (...gnu-gcc
> -mabi=aapcs-linux and ...gnueabi-gcc) do not have exactly the same
> configuration; I don't know for sure what might be different between
> them, but I do know we only expect EABI compliance from the EABI
> compilers.  -mabi=aapcs-linux versus -mabi=aapcs was mainly for
> interoperation between arm-none-eabi-gcc and
> arm-none-linux-gnueabi-gcc.

It appears to me that anything that gets functionality from libgcc
would be miscompiled by an old-ABI compiler and -mabi=aapcs-linux.
But Lennert is only talking about using this toolchain for
cross-compiling kernels; doesn't -mabi=aapcs-linux wind up using
substantially the same code generation parameters as t-linux-eabi?

Cheers,
- Michael

--
For unsubscribe information see http://sourceware.org/lists.html#faq