This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH, RFC] Enable IBM long double for PPC32 Linux
- From: "Ulrich Weigand" <uweigand at de dot ibm dot com>
- To: gerald at pfeifer dot com (Gerald Pfeifer)
- Cc: mark at codesourcery dot com (Mark Mitchell), krebbel1 at de dot ibm dot com (Andreas Krebbel), jakub at redhat dot com (Jakub Jelinek), gcc at gcc dot gnu dot org
- Date: Mon, 6 Feb 2006 14:12:14 +0100 (CET)
- Subject: Re: [PATCH, RFC] Enable IBM long double for PPC32 Linux
Gerald Pfeiffer wrote:
> Personally, and explicitly not speaking for my employer, I fully agree
> with Andrew Pinski that this kind of change is not appropriate for GCC
> 4.1 at this point in the release cycle.
>
> It is clearly against our development model and negatively impacts our
> schedule and the stabilization work done by many in the last months and
> was not even raised on the steering committee (which I would consider a
> requirement in such a case).
I'm not haüppy about the timing of this either. However, given the
circumstances, I'd considered it best for s390 to provide long double
support with 4.1 -- the patch I've approved doesn't change the default,
and should be low risk to code not built with -mlong-double-128.
The alternative -- either having a glibc not buildable with any released
GCC, or else having s390 unable to participate in the glibc long double
support together with all other affected platforms, would in my opinion be
worse than accepting this risk for 4.1.
I had been under the impression that the release manager was willing to
leave this choice to the affected platform maintainers, and the steering
committee was not involved. If I was mistaken in this, I'm sorry ...
Bye,
Ulrich
--
Dr. Ulrich Weigand
Linux on zSeries Development
Ulrich.Weigand@de.ibm.com