[Mlir-commits] [mlir] [mlir][gpu] Fix breaking constructor from GPUSubgroupSizeToROCDL (PR #137439)

[Stanley Winata]

https://github.com/raikonenfnu created https://github.com/llvm/llvm-project/pull/137439

We are adding GPUSubgroupSizeToROCDL to patterns function that do not have a valid constructor for it. This is causing said error below:

error: constructor inherited by 'GPUSubgroupSizeOpToROCDL' from base class 'ConvertOpToLLVMPattern<mlir::gpu::SubgroupSizeOp>' is implicitly deleted

>From 2e715c5a052089a9ea2a532c0876db8e8080db49 Mon Sep 17 00:00:00 2001
From: Stanley Winata <stanley.winata at amd.com>
Date: Fri, 25 Apr 2025 20:13:42 -0700
Subject: [PATCH] [mlir][gpu] Fix breaking constructor from
 GPUSubgroupSizeToROCDL

We are adding GPUSubgroupSizeToROCDL to patterns function that do not
have a valid constructor for it. This is causing said error below:

error: constructor inherited by 'GPUSubgroupSizeOpToROCDL'
from base class 'ConvertOpToLLVMPattern<mlir::gpu::SubgroupSizeOp>' is implicitly deleted

Signed-off-by: Stanley Winata <stanley.winata at amd.com>
---
 mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp b/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
index 0a6fb6451f700..dd16ec4b73e9f 100644
--- a/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
+++ b/mlir/lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp
@@ -455,9 +455,7 @@ void mlir::populateGpuToROCDLConversionPatterns(
   // TODO: Add alignment for workgroup memory
   patterns.add<GPUDynamicSharedMemoryOpLowering>(converter);
 
-  patterns
-      .add<GPUShuffleOpLowering, GPULaneIdOpToROCDL, GPUSubgroupSizeOpToROCDL>(
-          converter);
+  patterns.add<GPUShuffleOpLowering, GPULaneIdOpToROCDL>(converter);
   patterns.add<GPUSubgroupSizeOpToROCDL>(converter, chipset);
 
   populateMathToROCDLConversionPatterns(converter, patterns);