llvm
diff --git a/‎mlir/include/mlir/Conversion/Passes.td
Copy file name to clipboardExpand all lines: mlir/include/mlir/Conversion/Passes.td
+2-1Lines changed: 2 additions & 1 deletion b/‎mlir/include/mlir/Conversion/Passes.td
Copy file name to clipboardExpand all lines: mlir/include/mlir/Conversion/Passes.td
+2-1Lines changed: 2 additions & 1 deletion
diff --git a/‎mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
Copy file name to clipboardExpand all lines: mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
+2Lines changed: 2 additions & 0 deletions b/‎mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
Copy file name to clipboardExpand all lines: mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
+2Lines changed: 2 additions & 0 deletions
@@ -601,6 +601,7 @@ def ConvertGpuOpsToROCDLOps : Pass<"convert-gpu-to-rocdl", "gpu::GPUModuleOp"> {
   let constructor = "mlir::createLowerGpuOpsToROCDLOpsPass()";
   let dependentDialects = [
     "ROCDL::ROCDLDialect",
+    "amdgpu::AMDGPUDialect",
     "cf::ControlFlowDialect",
     "memref::MemRefDialect",
   ];
@@ -1415,7 +1416,7 @@ def ConvertVectorToLLVMPass : Pass<"convert-vector-to-llvm"> {
            "bool", /*default=*/"false",
            "Use the preferred alignment of a vector type in load/store "
            "operations instead of the alignment of the element type of the "
-           "memref. This flag is intended for use with hardware which requires" 
+           "memref. This flag is intended for use with hardware which requires"
            "vector alignment, or in application contexts where it is known all "
            "vector access are naturally aligned. ">,
     Option<"amx", "enable-amx",
 
@@ -27,6 +27,7 @@
 #include "mlir/Conversion/LLVMCommon/TypeConverter.h"
 #include "mlir/Conversion/MathToLLVM/MathToLLVM.h"
 #include "mlir/Conversion/MathToROCDL/MathToROCDL.h"
+#include "mlir/Dialect/AMDGPU/IR/AMDGPUDialect.h"
 #include "mlir/Dialect/ControlFlow/IR/ControlFlow.h"
 #include "mlir/Dialect/Func/IR/FuncOps.h"
 #include "mlir/Dialect/GPU/IR/GPUDialect.h"
@@ -318,6 +319,7 @@ struct LowerGpuOpsToROCDLOpsPass final
     {
       RewritePatternSet patterns(ctx);
       populateGpuRewritePatterns(patterns);
+      populateGpuPromoteShuffleToAMDGPUPatterns(patterns);
       (void)applyPatternsGreedily(m, std::move(patterns));
     }