News for hackers. Stuff that matters.
Re:GCC optimized code gives strange floating point results by Kenji on Sunday February 26, 2017 @04:44PM (#13740093)
This is well-known by everyone who works with float determinism. You have to set the FPU flag with assembler to force the desired width (among other things). See: fldcw, LDMXCSR, etc.

Take into account rounding modes and denormals too, and make sure to disable float optimisations that break determinism in your compiler.

There are no comments on this comment yet.