[llvm] [InstCombine] Generalise optimisation of redundant floating point comparisons with `ConstantFPRange` (PR #159315)

[Yingwei Zheng via llvm-commits]

================
@@ -638,8 +637,7 @@ define i1 @issubnormal_or_finite_olt(half %x) {
 ; inf | nan | zero | subnormal
 define i1 @issubnormal_or_finite_uge(half %x) {
 ; CHECK-LABEL: @issubnormal_or_finite_uge(
-; CHECK-NEXT:    [[OR:%.*]] = call i1 @llvm.is.fpclass.f16(half [[X:%.*]], i32 759)
-; CHECK-NEXT:    ret i1 [[OR]]
+; CHECK-NEXT:    ret i1 true
----------------
dtcxzyw wrote:

Miscompilation: https://alive2.llvm.org/ce/z/jAaejK

https://github.com/llvm/llvm-project/pull/159315