[PATCH] D36335: Add ‘llvm.experimental.constrained.fma‘ Intrinsic

[Andy Kaylor via Phabricator via llvm-commits]

andrew.w.kaylor added inline comments.


================
Comment at: test/CodeGen/X86/fp-intrinsics.ll:245
 
+; Verify that fma(1.0) isn't simplified when the rounding mode is
+; unknown.
----------------
These values could be constant folded without rounding, so even though this test case works now it's testing something that we don't necessarily want to be true.  At some point, we're going to want to teach optimizations to recognize these intrinsics and fold cases like this.  That's why I was using 42.1 in the other tests.  It's just an arbitrary value that introduces rounding errors.


Repository:
  rL LLVM

https://reviews.llvm.org/D36335