https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125623
John David Anglin <danglin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #64635|0 |1
is obsolete| |
--- Comment #6 from John David Anglin <danglin at gcc dot gnu.org> ---
Created attachment 64638
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=64638&action=edit
Patch
Testing. If the sizeof long double is the same as the sizeof __float128
(i.e. 16), I think we can use the libquadmath routines. libgfortran already
depends on libquadmath.
We still need to use builtins in write_float.def to avoid warnings.
