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

[Nikita Popov 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) {
----------------
nikic wrote:

Isn't this what the preceding test case checks?

Though we should also have test variants for monotonic store. This is currently only testing monotonic load.

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