One thing I notice about the C program is that it uses the double-precision exp(), fabs(), and sqrt(), rather than the "q"-suffixed variants from quadmath.h, a

One thing I notice about the C program is that it uses the double-precision exp(), fabs(), and sqrt(), rather than the "q"-suffixed variants from quadmath.h, a feature that appears to be specific to GCC. However, changing the program had no effect for me with GCC 9.2. I can't get the C++ program to compile. Edit: the post has been rewritten since I saw it.