You should be able to see it in Sun libraries (used by almost all of the open C libraries that I know of, including glibc and FreeBSD).
I usually prefer BSD code for math code (more readable IMO). See here for the long dual format 80 bit (Intel). For this function, various implementations / architectures may be in different directories.
One thing to understand is that long double is not standardized (more precisely, only since 2008 IEEE754, which has not yet been implemented in most common processors). This means that each processor requires a different implementation for many things (IA32, AMD64, PPC, Alpha, Sparc are all different in this aspect ...).
source share