wjeaton
Results
1
comments of
wjeaton
The problem is caused by the gcc -ffast-math optimizations. This flag turns on -ffinite-math-only, which acts as a promise to the compiler that code will never use floating point infinities....