The IEEE standard for floating point numbers has certain mathematical requirements that makes it a bit more "correct" but possibly also a bit slower. What Linus

The IEEE standard for floating point numbers has certain mathematical requirements that makes it a bit more "correct" but possibly also a bit slower. What Linus is saying [1] is that you don't need that in most applications [2] and definitely not when you specify -ffast-math [1] https://gcc.gnu.org/ml/gcc/2001-07/msg01877.html https://gcc.gnu.org/ml/gcc/2001-07/msg01877.html [2] and many agree with him, Java for example moved away from IEEE around the same time this was written.