Define -msoft-float on ppc when --without-fp specified

Daniel Jacobowitz drow@mvista.com
Fri Dec 7 05:23:00 GMT 2001


On Thu, Dec 06, 2001 at 03:23:20PM -0800, Geoff Keating wrote:
> > Date: Wed, 5 Dec 2001 20:44:58 -0500
> > From: Daniel Jacobowitz <drow@mvista.com>
> 
> > On Wed, Dec 05, 2001 at 05:22:50PM -0800, Geoff Keating wrote:
> > > 
> > > This patch makes the user's life easier by automatically passing the
> > > appropriate flags to GCC to disable use of the FPU when --without-fp
> > > is specified.
> > > 
> > > OK to commit?
> > 
> > While you're at it, you might also want to:
> >   - shrink the size of a jmpbuf while you're already changing the ABI
> 
> I thought about doing this, but it seemed like it would be better to
> allow the space in the jmpbuf.  That way, code that doesn't pass
> floating-point arguments is compatible with both ABIs.

I only mention it by analogy with AltiVec, which needs a larger
jumpbuf.  Which reminds me - does Red Hat have plans to submit any
patches which would cover AltiVec setjmp/longjmp, and if not and I
repost the patch I sent several times this summer, would it at least
get a little more reaction?

Some of the configury would probably have to change, given the decision
to use new triplets.

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer



More information about the Libc-alpha mailing list