[clang] [llvm] [mlir] [IR] Make semantics of strictfp consistent v2 (PR #211769)

[Nathan Corbyn via cfe-commits]

================
@@ -2800,7 +2800,8 @@ fn -> other_fn -> other_fn ; fn is norecurse
     optimizations that require assumptions about the floating-point rounding
     mode or that might alter the state of floating-point status flags that
     might otherwise be set or cleared by calling this function. LLVM will
-    not introduce any new floating-point instructions that may trap.
+    not introduce any new floating-point instructions that may trap. All
+    function definitions that contain strictfp calls must be marked strictfp.
----------------
cofibrant wrote:

I wasn't actually aware of such intrinsics, but there are two lines in the LangRef that will maybe offer some explanation:
> If any FP operation in a function is constrained then they all must be constrained. This is required for correct LLVM IR.

> All function calls done in a function that uses constrained floating point intrinsics must have the `strictfp` attribute either on the calling instruction or on the declaration or definition of the function being called.

(Otherwise the callee might raise an FP exception which is incorrectly optimised).

I'll look at #205061 when I find some time

https://github.com/llvm/llvm-project/pull/211769