[Mlir-commits] [mlir] [mlir][dataflow] Delete visitRegionSuccessors not run code (NFC) (PR #174790)

[Matthias Springer]

================
@@ -307,29 +307,20 @@ void AbstractSparseForwardDataFlowAnalysis::visitRegionSuccessors(
            "expected the same number of successor inputs as operands");
 
     unsigned firstIndex = 0;
-    if (inputs.size() != lattices.size()) {
-      if (!point->isBlockStart()) {
-        if (!inputs.empty())
-          firstIndex = cast<OpResult>(inputs.front()).getResultNumber();
-        visitNonControlFlowArgumentsImpl(
----------------
matthias-springer wrote:

I don't understand why it's safe to remove the case for OpResults. Imagine a hypothetical op that's like `scf.for`, but has an additional op result, which is the number of iterations.

```
%num_iter, %loop_carried = scf.fancy_for_loop ... iter_args(%arg0 = %0) {
  ...
}
```

In the above example, `%num_iter` is not a successor input, so `inputs.size() != lattices.size()`. (`inputs.size()` is 1, `lattices.size()` is 2.)

https://github.com/llvm/llvm-project/pull/174790