diff --git a/lib/Conversion/TritonToTritonGPU/TritonToTritonGPU.cpp b/lib/Conversion/TritonToTritonGPU/TritonToTritonGPU.cpp
index 18ebd035d..7be905fab 100644
--- a/lib/Conversion/TritonToTritonGPU/TritonToTritonGPU.cpp
+++ b/lib/Conversion/TritonToTritonGPU/TritonToTritonGPU.cpp
@@ -11,6 +11,21 @@ using namespace mlir::triton;
 
 namespace {
 
+template<class Op>
+class ArithBinaryPattern : public OpConversionPattern<Op> {
+public:
+  using OpConversionPattern<Op>::OpConversionPattern;
+
+  LogicalResult matchAndRewrite(Op op, typename Op::Adaptor adaptor,
+                                ConversionPatternRewriter &rewriter) const override {
+    Type retType = this->getTypeConverter()->convertType(op.getType());
+    rewriter.replaceOpWithNewOp<Op>(
+      op.getOperation(), retType, op.getLhs(), op.getRhs()
+    );
+    return success();
+  }
+};
+
 class ConvertArithmeticOp: public ConversionPattern {
 public:
     ConvertArithmeticOp(TritonGPUTypeConverter &typeConverter, MLIRContext *context)
@@ -44,7 +59,38 @@ void populateArithmeticPatternsAndLegality(
       }
   );
   // Rewrite rule
-  patterns.add<ConvertArithmeticOp>(typeConverter, context);
+  // patterns.add<ConvertArithmeticOp>(typeConverter, context);
+  patterns.add<ArithBinaryPattern<arith::AddIOp>,
+               ArithBinaryPattern<arith::SubIOp>,
+               ArithBinaryPattern<arith::MulIOp>,
+               ArithBinaryPattern<arith::DivUIOp>,
+               ArithBinaryPattern<arith::DivSIOp>,
+               ArithBinaryPattern<arith::CeilDivUIOp>,
+               ArithBinaryPattern<arith::CeilDivSIOp>,
+               ArithBinaryPattern<arith::FloorDivSIOp>,
+               ArithBinaryPattern<arith::RemUIOp>,
+               ArithBinaryPattern<arith::RemSIOp>,
+               ArithBinaryPattern<arith::AndIOp>,
+               ArithBinaryPattern<arith::OrIOp>,
+               ArithBinaryPattern<arith::XOrIOp>,
+               ArithBinaryPattern<arith::ShLIOp>,
+               ArithBinaryPattern<arith::ShRUIOp>,
+               ArithBinaryPattern<arith::ShRSIOp>, // NegFOp
+               // Floating point
+               ArithBinaryPattern<arith::AddFOp>,
+               ArithBinaryPattern<arith::SubFOp>,
+               // MaxMin
+               ArithBinaryPattern<arith::MaxFOp>,
+               ArithBinaryPattern<arith::MaxSIOp>,
+               ArithBinaryPattern<arith::MaxUIOp>,
+               ArithBinaryPattern<arith::MinFOp>,
+               ArithBinaryPattern<arith::MinSIOp>,
+               ArithBinaryPattern<arith::MinUIOp>,
+               // Floating point
+               ArithBinaryPattern<arith::MulFOp>,
+               ArithBinaryPattern<arith::DivFOp>,
+               ArithBinaryPattern<arith::RemFOp>
+              >(typeConverter, context);
 }
 
 //