This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: implement __builtin_copysign
Or, rather, an algorithm which achieves the Annex G treatment of NaNs and
infinities but also takes more care to avoid undue overflow (an attachment
to bug 18902 illustrates division with scaling to avoid overflow in
calculation of either numerator or denominator, but without dealing with
NaNs and infinities).
I don't remember the exact details, but parts of the C99 complex-type
specification were mathematically bogus, and very much so. I think it
was the handling of values where one or both values were infinities or NaNs.
There were such scary things as (-Inf, NaN) and so on...
We do not want them to be implemented neither in GCC nor in glibc,
right? :-/
Paolo