This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: PowerPC64 long double
- From: Alan Modra <amodra at bigpond dot net dot au>
- To: Ulrich Drepper <drepper at redhat dot com>
- Cc: libc-alpha at sources dot redhat dot com,Steve Munroe <sjmunroe at us dot ibm dot com>, dgm69 at us dot ibm dot com
- Date: Tue, 22 Feb 2005 14:55:25 +1030
- Subject: Re: PowerPC64 long double
- References: <20050215041116.GM22790@bubble.modra.org> <421A8FB7.1090805@redhat.com>
On Mon, Feb 21, 2005 at 05:49:43PM -0800, Ulrich Drepper wrote:
> Alan Modra wrote:
> >Needed for compiling with -mlong-double-128 on powerpc64-linux
>
> Is this the only change needed?
Probably not. This particular change merely fixes an obvious
compile-time error. Steve can no doubt answer better as to how much
long double support is missing.
Steve/Dwayne, I'd like to enable -mlong-double-128 by default for
powerpc64-linux in gcc-4.0. Whether we do this depends on the state
of glibc long double support for powerpc64-linux. It would be
particularly nice if glibc could support both 64-bit and 128-bit
long double, for example by defining inline math functions that select
between the double or long double version depending on the size of
long double. I guess this opens up a can of worms regarding backward
compatibility, which would need some versioning tricks to solve. ie.
old code using 64-bit long doubles will reference fabsl@@some_version
and expect fabsl@@some_version to continue to be 64-bit.
--
Alan Modra
IBM OzLabs - Linux Technology Centre