[llvm-branch-commits] [flang] [flang][OpenMP] Update `do concurrent` mapping pass to use `fir.do_concurrent` op (PR #138489)

[Tom Eccles via llvm-branch-commits]

================
@@ -548,19 +357,16 @@ class DoConcurrentConversionPass
       return;
     }
 
-    llvm::DenseSet<fir::DoLoopOp> concurrentLoopsToSkip;
+    llvm::DenseSet<fir::DoConcurrentOp> concurrentLoopsToSkip;
----------------
tblah wrote:

The correctness of this depends upon the traversal order used by `applyFullConversion`. I'm not certain that is guaranteed. Also modifying this (conceptually) global list is harder to follow. Now that nesting is much simpler, could we instead now skip any `fir::DoConcurrentOp` which has an ancestor which is a `fir::DoConcurrentOp`?

https://github.com/llvm/llvm-project/pull/138489