This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH v2 0/6] Enable float128 on powerpc64le
On Thu, 22 Jun 2017 14:03:39 +0000
Joseph Myers <joseph@codesourcery.com> wrote:
> The following is about future developments after the float128 support is
> ready and checked in, not about any issues with the patches themselves or
> requirements for them to be included.
>
> Once the support is included, I intend to enable it in glibc for the other
> platforms where GCC supports float128 as ABI-distinct from long double
> (x86_64, x86, ia64). (I intend eventually to enable support for *fN /
> *fNx aliases for existing functions where the types are not ABI-distinct,
> but I don't expect that to be for 2.26.)
>
> Do you have any plans for additional float128 toolchain features, such as
> tgmath.h support, *f128 GCC built-in functions beyond the minimal
> __builtin_* set present for all _FloatN/_FloatNx types in GCC 7, or
> libstdc++ support?
My current intention, after the enabling of float128 for powerpc64le is
in, is to start working on transitiong long double from IBM Extended
Precision to IEEE 754 128-bits FP. This will be only for powerpc64le and
not for glibc 2.26.