[llvm] [NVPTX] Consistently check fast-math flags when lowering fsqrt (PR #143776)

[Alex MacLean via llvm-commits]

================
@@ -1134,7 +1143,8 @@ SDValue NVPTXTargetLowering::getSqrtEstimate(SDValue Operand, SelectionDAG &DAG,
                                              bool &UseOneConst,
                                              bool Reciprocal) const {
   if (!(Enabled == ReciprocalEstimate::Enabled ||
-        (Enabled == ReciprocalEstimate::Unspecified && !usePrecSqrtF32())))
+        (Enabled == ReciprocalEstimate::Unspecified &&
+         !usePrecSqrtF32(DAG.getMachineFunction()))))
----------------
AlexMaclean wrote:

Shoot, sorry I missed this one!

https://github.com/llvm/llvm-project/pull/143776