[Mlir-commits] [mlir] [Conversion] Avoid repeated hash lookups (NFC) (PR #111637)

[llvmlistbot at llvm.org]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir-gpu

Author: Kazu Hirata (kazutakahirata)

<details>
<summary>Changes</summary>



---
Full diff: https://github.com/llvm/llvm-project/pull/111637.diff


1 Files Affected:

- (modified) mlir/lib/Conversion/SCFToGPU/SCFToGPU.cpp (+1-2) 


``````````diff
diff --git a/mlir/lib/Conversion/SCFToGPU/SCFToGPU.cpp b/mlir/lib/Conversion/SCFToGPU/SCFToGPU.cpp
index 2bcd082fb3e82f..dece254c325fcd 100644
--- a/mlir/lib/Conversion/SCFToGPU/SCFToGPU.cpp
+++ b/mlir/lib/Conversion/SCFToGPU/SCFToGPU.cpp
@@ -509,12 +509,11 @@ static LogicalResult processParallelLoop(
                   ensureLaunchIndependent(cloningMap.lookupOrDefault(step))});
           // todo(herhut,ravishankarm): Update the behavior of setMappingAttr
           // when this condition is relaxed.
-          if (bounds.contains(processor)) {
+          if (!bounds.try_emplace(processor, launchBound).second) {
             return rewriter.notifyMatchFailure(
                 parallelOp, "cannot redefine the bound for processor " +
                                 Twine(static_cast<int64_t>(processor)));
           }
-          bounds[processor] = launchBound;
         }
         if (!boundIsPrecise) {
           // We are using an approximation, create a surrounding conditional.

``````````

</details>


https://github.com/llvm/llvm-project/pull/111637