[llvm] [GlobalISel] Improve bf16 converts with fast-math flags. (PR #200741)

[Jay Foad via llvm-commits]

================
@@ -148,6 +148,13 @@ class LegalizerHelper {
   LLVM_ABI void widenScalarSrc(MachineInstr &MI, LLT WideTy, unsigned OpIdx,
                                unsigned ExtOpcode);
 
+  /// Legalize a single operand \p OpIdx of the machine instruction \p MI as a
+  /// Use by extending the operand's type to \p WideTy using the G_FPEXT for the
+  /// extension instruction, and replacing the vreg of the operand in place.
+  /// Flags are copied from MI to the new extend.
+  LLVM_ABI void widenScalarSrcUsingFPExt(MachineInstr &MI, LLT WideTy,
----------------
jayfoad wrote:

> Do you think it will cause problems?

No, it just seemed redundant.

https://github.com/llvm/llvm-project/pull/200741