[llvm] [AA] Improve precision for monotonic atomic load/store operations (PR #158169)

[David Li via llvm-commits]

================
@@ -21,3 +20,17 @@ define i32 @test9() {
   store i32 1, ptr @x
   ret i32 %x
 }
+
+; DSE across monotonic load (blocked if the atomic load's address isn't NoAlias)
+define i32 @test9a(ptr %ptr) {
----------------
david-xl wrote:

Add a case where DSE or CSE is not blocked by a non-aliased monotonic load or store.

https://github.com/llvm/llvm-project/pull/158169