Medium
Four floating-point identities
Compiled with Clang or GCC at -O2 and no fast-math flags of any kind, which of these functions compile to a bare ret that hands x back unchanged?
main.cpp
double addZero(double x) { return x + 0.0; }
double subZero(double x) { return x - 0.0; }
double mulOne(double x) { return x * 1.0; }
double mulZero(double x) { return x * 0.0; }42 of 44 in Compilers