[llvm] [llubi] Initial support for floating-point numbers (PR #188453)

[Yingwei Zheng via llvm-commits]

================
@@ -87,6 +87,26 @@ cl::opt<ubi::UndefValueBehavior> UndefBehavior(
                clEnumVal(ubi::UndefValueBehavior::Zero,
                          "All uses of an uninitialized byte yield zero.")));
 
+cl::opt<ubi::NaNPropagationBehavior> NaNPropagationBehavior(
+    "", cl::desc("Choose NaN propagation behavior:"),
----------------
dtcxzyw wrote:

missing cl::init and tests.

https://github.com/llvm/llvm-project/pull/188453