diff --git a/mlir/include/mlir/InitAllPasses.h b/mlir/include/mlir/InitAllPasses.h
--- a/mlir/include/mlir/InitAllPasses.h
+++ b/mlir/include/mlir/InitAllPasses.h
@@ -91,10 +91,6 @@
 
   // CUDA
   createConvertGpuLaunchFuncToCudaCallsPass();
-#if MLIR_CUDA_CONVERSIONS_ENABLED
-  createConvertGPUKernelToCubinPass(
-      [](const std::string &, Location, StringRef) { return nullptr; });
-#endif
   createLowerGpuOpsToNVVMOpsPass();
 
   // Linalg
diff --git a/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp b/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp
--- a/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp
+++ b/mlir/lib/Conversion/GPUToCUDA/ConvertKernelFuncToCubin.cpp
@@ -49,8 +49,7 @@
 class GpuKernelToCubinPass
     : public OperationPass<GpuKernelToCubinPass, gpu::GPUModuleOp> {
 public:
-  GpuKernelToCubinPass(
-      CubinGenerator cubinGenerator = compilePtxToCubinForTesting)
+  GpuKernelToCubinPass(CubinGenerator cubinGenerator)
       : cubinGenerator(cubinGenerator) {}
 
   void runOnOperation() override {
@@ -76,9 +75,6 @@
   }
 
 private:
-  static OwnedCubin compilePtxToCubinForTesting(const std::string &ptx,
-                                                Location, StringRef);
-
   std::string translateModuleToPtx(llvm::Module &module,
                                    llvm::TargetMachine &target_machine);
 
@@ -112,13 +108,6 @@
   return ptx;
 }
 
-OwnedCubin
-GpuKernelToCubinPass::compilePtxToCubinForTesting(const std::string &ptx,
-                                                  Location, StringRef) {
-  const char data[] = "CUBIN";
-  return std::make_unique<std::vector<char>>(data, data + sizeof(data) - 1);
-}
-
 OwnedCubin GpuKernelToCubinPass::convertModuleToCubin(llvm::Module &llvmModule,
                                                       Location loc,
                                                       StringRef name) {
@@ -158,7 +147,3 @@
 mlir::createConvertGPUKernelToCubinPass(CubinGenerator cubinGenerator) {
   return std::make_unique<GpuKernelToCubinPass>(cubinGenerator);
 }
-
-static PassRegistration<GpuKernelToCubinPass>
-    pass("test-kernel-to-cubin",
-         "Convert all kernel functions to CUDA cubin blobs");
diff --git a/mlir/test/lib/Transforms/CMakeLists.txt b/mlir/test/lib/Transforms/CMakeLists.txt
--- a/mlir/test/lib/Transforms/CMakeLists.txt
+++ b/mlir/test/lib/Transforms/CMakeLists.txt
@@ -3,6 +3,7 @@
   TestAllReduceLowering.cpp
   TestCallGraph.cpp
   TestConstantFold.cpp
+  TestConvertGPUKernelToCubin.cpp
   TestLoopFusion.cpp
   TestGpuMemoryPromotion.cpp
   TestGpuParallelLoopMapping.cpp
diff --git a/mlir/test/lib/Transforms/TestConvertGPUKernelToCubin.cpp b/mlir/test/lib/Transforms/TestConvertGPUKernelToCubin.cpp
new file mode 100644
--- /dev/null
+++ b/mlir/test/lib/Transforms/TestConvertGPUKernelToCubin.cpp
@@ -0,0 +1,32 @@
+//===- TestConvertGPUKernelToCubin.cpp - Test gpu kernel cubin lowering ---===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Conversion/GPUToCUDA/GPUToCUDAPass.h"
+#include "mlir/Pass/Pass.h"
+#include "mlir/Pass/PassManager.h"
+
+using namespace mlir;
+
+namespace {
+static OwnedCubin compilePtxToCubinForTesting(const std::string &, Location,
+                                              StringRef) {
+  const char data[] = "CUBIN";
+  return std::make_unique<std::vector<char>>(data, data + sizeof(data) - 1);
+}
+} // end anonymous namespace
+
+namespace mlir {
+void registerTestConvertGPUKernelToCubinPass() {
+  PassPipelineRegistration<>("test-kernel-to-cubin",
+                             "Convert all kernel functions to CUDA cubin blobs",
+                             [](OpPassManager &pm) {
+                               pm.addPass(createConvertGPUKernelToCubinPass(
+                                   compilePtxToCubinForTesting));
+                             });
+}
+} // namespace mlir
diff --git a/mlir/tools/mlir-opt/mlir-opt.cpp b/mlir/tools/mlir-opt/mlir-opt.cpp
--- a/mlir/tools/mlir-opt/mlir-opt.cpp
+++ b/mlir/tools/mlir-opt/mlir-opt.cpp
@@ -41,6 +41,7 @@
 void registerTestAllReduceLoweringPass();
 void registerTestCallGraphPass();
 void registerTestConstantFold();
+void registerTestConvertGPUKernelToCubinPass();
 void registerTestFunc();
 void registerTestGpuMemoryPromotionPass();
 void registerTestLinalgTransforms();
@@ -96,6 +97,9 @@
   registerTestAllReduceLoweringPass();
   registerTestCallGraphPass();
   registerTestConstantFold();
+#if MLIR_CUDA_CONVERSIONS_ENABLED
+  registerTestConvertGPUKernelToCubinPass();
+#endif
   registerTestFunc();
   registerTestGpuMemoryPromotionPass();
   registerTestLinalgTransforms();