> "solving" the fp32 incorrect-rounding problem by using fp64 is naive (not guaranteed to always work, although it will with high probability)
The key thing is there are only 2*32 float32s so you can check all of them. It sounds to me like the author did this, and realized they needed some tweaks for correct answers with log.
The key thing is there are only 2*32 float32s so you can check all of them. It sounds to me like the author did this, and realized they needed some tweaks for correct answers with log.