Updated I’ve got no explanation for you, since I’m on Haswell, but I do have code to share that might help you or someone else with Broadwell or Skylake hardware isolate your problem. If you could please run it on your machine and share the results, we could gain an insight into what’s happening to … Read more
IACA Analysis Using IACA (the Intel Architecture Code Analyzer) reveals that macro-op fusion is indeed occurring, and that it is not the problem. It is Mysticial who is correct: The problem is that the store isn’t using Port 7 at all. IACA reports the following: Intel(R) Architecture Code Analyzer Version – 2.1 Analyzed File – … Read more
The compiler is allowed to fuse a separated add and multiply, even though this changes the final result (by making it more accurate). An FMA has only one rounding (it effectively keeps infinite precision for the internal temporary multiply result), while an ADD + MUL has two. The IEEE and C standards allow this when … Read more