[llvm] [OpenMPOpt] Fix incorrect end-of-kernel barrier removal (PR #65670)

[Johannes Doerfert via llvm-commits]

================
@@ -2676,17 +2676,19 @@ struct AAExecutionDomainFunction : public AAExecutionDomain {
       if (!ED.EncounteredAssumes.empty() && !A.isModulePass())
         return;
 
-      // We can remove this barrier, if it is one, or all aligned barriers
-      // reaching the kernel end. In the latter case we can transitively work
-      // our way back until we find a barrier that guards a side-effect if we
-      // are dealing with the kernel end here.
+      // We can remove this barrier, if it is one, or aligned barriers reaching
+      // the kernel end (if CB is nullptr). Aligned barriers reaching the kernel
+      // end may have other successors besides the kernel end (especially if
+      // they're in loops) with non-local side-effects, so those barriers can
+      // only be removed if they also only reach the kernel end. If those
----------------
jdoerfert wrote:

> only be removed if they also only reach the kernel end.

That is not true, nor does it reflect what happens below. The kernel end is not an aligned barrier which is why your test case starts to work (the explicit aligned barrier does not only reach aligned barriers but also the kernel end and is consequently not removed).

https://github.com/llvm/llvm-project/pull/65670