diff --git a/paddle/platform/device_context.cc b/paddle/platform/device_context.cc
index 7afcdfce9371e29aad968a1729931173fb2309b5..ae4f0bf896dce013d301aa0bf9f732f0fd9cc6bf 100644
--- a/paddle/platform/device_context.cc
+++ b/paddle/platform/device_context.cc
@@ -122,10 +122,6 @@ Place CUDADeviceContext::GetPlace() const { return place_; }
 
 void CUDADeviceContext::Wait() const {
   PADDLE_ENFORCE(cudaStreamSynchronize(stream_));
-}
-
-void CUDADeviceContext::Finish() const {
-  Wait();
   PADDLE_ENFORCE(cudaGetLastError());
 }
 
diff --git a/paddle/platform/device_context.h b/paddle/platform/device_context.h
index 526d089e35da9c9f89a3852095ad3a4c82d4d85d..ef5f19214d9ccb23b9c946bee28cb764122bd7cd 100644
--- a/paddle/platform/device_context.h
+++ b/paddle/platform/device_context.h
@@ -46,8 +46,6 @@ class DeviceContext {
   DeviceType* GetEigenDevice() const;
 
   virtual void Wait() const {}
-
-  virtual void Finish() const {}
 };
 
 class CPUDeviceContext : public DeviceContext {
@@ -79,9 +77,6 @@ class CUDADeviceContext : public DeviceContext {
   /*! \brief  Wait for all operations completion in the stream. */
   void Wait() const override;
 
-  /*! \brief  Check potential errors for the cuda kernel calls. */
-  void Finish() const override;
-
   /*! \brief  Return place in the device context. */
   Place GetPlace() const override;