[llvm] [clang] [Instcombine] use zext's nneg flag for icmp folding (PR #70845)

[Yingwei Zheng via llvm-commits]

================
@@ -0,0 +1,126 @@
+; RUN: opt < %s --O3 -S | FileCheck %s
+
+define signext i16 @vecreduce_smax_v2i16(i32 noundef %0, ptr noundef %1) #0 {
----------------
dtcxzyw wrote:

Could you please reduce the test case and move it into `InstCombine`?

https://github.com/llvm/llvm-project/pull/70845