[Mlir-commits] [mlir] 24a1869 - [MLIR][GPU] Update GPUToROCDL to account for ControlFlow dialect

[Krzysztof Drewniak]

Author: Krzysztof Drewniak
Date: 2022-02-08T16:34:34Z
New Revision: 24a1869d007b57026cd10561ea9e8bbe1b38d996

URL: https://github.com/llvm/llvm-project/commit/24a1869d007b57026cd10561ea9e8bbe1b38d996
DIFF: https://github.com/llvm/llvm-project/commit/24a1869d007b57026cd10561ea9e8bbe1b38d996.diff

LOG: [MLIR][GPU] Update GPUToROCDL to account for ControlFlow dialect

The conversion to the new ControlFlow dialect didn't change the
GPUToROCDL pass - this commit fixes this issue.

Reviewed By: rriddle

Differential Revision: https://reviews.llvm.org/D119188

Added: 
    

Modified: 
    mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp

Removed: 
    


################################################################################
diff  --git a/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp b/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
index e95e2854be0cf..38bf0efcdf084 100644
--- a/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
+++ b/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
@@ -11,6 +11,7 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "mlir/Conversion/ControlFlowToLLVM/ControlFlowToLLVM.h"
 #include "mlir/Conversion/GPUToROCDL/GPUToROCDLPass.h"
 
 #include "mlir/Conversion/ArithmeticToLLVM/ArithmeticToLLVM.h"
@@ -78,6 +79,7 @@ struct LowerGpuOpsToROCDLOpsPass
                                                             llvmPatterns);
     populateVectorToLLVMConversionPatterns(converter, llvmPatterns);
     populateVectorToROCDLConversionPatterns(converter, llvmPatterns);
+    cf::populateControlFlowToLLVMConversionPatterns(converter, llvmPatterns);
     populateStdToLLVMConversionPatterns(converter, llvmPatterns);
     populateMemRefToLLVMConversionPatterns(converter, llvmPatterns);
     populateGpuToROCDLConversionPatterns(converter, llvmPatterns, runtime);