This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH v3 1/2] float128: Add _Float128 make bits to libm.
On Mon, 8 May 2017, Gabriel F. T. Gomes wrote:
> On Mon, 08 May 2017 16:05:15 -0300
> "Tulio Magno Quites Machado Filho" <tuliom@linux.vnet.ibm.com> wrote:
>
> > "Gabriel F. T. Gomes" <gftg@linux.vnet.ibm.com> writes:
> >
> > > diff --git a/sysdeps/ieee754/float128/Versions
> > > b/sysdeps/ieee754/float128/Versions new file mode 100644
> > > [...]
> > > + __iscanonicalf128;
> >
> > Isn't it possible to reuse the implementation from bits/iscanonical.h and
> > avoid exporting this symbol?
> > I understand this would require to install iscanonicalf128.h with the new
> > implementation.
>
> You are right. I can provide __iscanonicalf128 as a macro through
> iscanonicalf128.h, then the line above (in the Versions file) is not
> necessary.
There should be no iscanonicalf128.h (or __iscanonicalf128 function).
Rather, bits/iscanonical.h defines the iscanonical macro in an appropriate
way for the applicable glibc configuration. That means that
sysdeps/ieee754/ldbl-128ibm/bits/iscanonical.h will need to define
__iscanonicalf128 as a macro (and math.h will need to define __MATH_TG
appropriately for the case where _Float128 is supported and ABI-distinct
from the other types).
--
Joseph S. Myers
joseph@codesourcery.com