[Mlir-commits] [mlir] [Affine] Avoid repeated hash lookups (NFC) (PR #111226)

[llvmlistbot at llvm.org]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir

Author: Kazu Hirata (kazutakahirata)

<details>
<summary>Changes</summary>



---
Full diff: https://github.com/llvm/llvm-project/pull/111226.diff


1 Files Affected:

- (modified) mlir/lib/Dialect/Affine/Transforms/PipelineDataTransfer.cpp (+1-2) 


``````````diff
diff --git a/mlir/lib/Dialect/Affine/Transforms/PipelineDataTransfer.cpp b/mlir/lib/Dialect/Affine/Transforms/PipelineDataTransfer.cpp
index deb530b4cf1c95..4be99aa1973805 100644
--- a/mlir/lib/Dialect/Affine/Transforms/PipelineDataTransfer.cpp
+++ b/mlir/lib/Dialect/Affine/Transforms/PipelineDataTransfer.cpp
@@ -350,8 +350,7 @@ void PipelineDataTransfer::runOnAffineForOp(AffineForOp forOp) {
   }
   // Everything else (including compute ops and dma finish) are shifted by one.
   for (auto &op : forOp.getBody()->without_terminator())
-    if (!instShiftMap.contains(&op))
-      instShiftMap[&op] = 1;
+    instShiftMap.try_emplace(&op, 1);
 
   // Get shifts stored in map.
   SmallVector<uint64_t, 8> shifts(forOp.getBody()->getOperations().size());

``````````

</details>


https://github.com/llvm/llvm-project/pull/111226