[llvm] [X86] Try Folding icmp of v8i32 -> fcmp of v8f32 on AVX (PR #82290)

[via llvm-commits]

================
@@ -41125,6 +41255,154 @@ static SDValue combineShuffle(SDNode *N, SelectionDAG &DAG,
   return SDValue();
 }
 
+// Simplify a decomposed (sext (setcc)). Assumes prior check that
+// bitwidth(sext)==bitwidth(setcc operands).
+static SDValue simplifySExtOfDecomposedSetCCImpl(
----------------
goldsteinn wrote:

So this PR will have some regressions in it, ill post the other PR as rebased ontop of this one.

https://github.com/llvm/llvm-project/pull/82290