cuvidCreateDecoder memory leak

Accelerated Computing CUDA CUDA Programming and Performance
1

I need restart movie in the my program. But cuvidDecoder has memory leak in GPU memory.

// CUDA_VERSION 4000
// NVIDIA GeForce GTX 560
// DisplayDriver: 6.14.12.7533
// OS: WindowsXP Professional

// test

	cuCtxPushCurrent(m_Context);
	do{
		oResult = cuvidCreateDecoder(&oDecoder_, &oVideoDecodeCreateInfo_);
		if(oDecoder_)
			cuvidDestroyDecoder(oDecoder_);
		oDecoder_ = NULL;

	}while(1);
	cuCtxPopCurrent(NULL);

oVideoDecodeCreateInfo_ initialization:

m_Context          = rContext;
m_VideoCreateFlags = cudaVideoCodec_MPEG2;
switch (eCreateFlags) {
    case cudaVideoCreate_Default:    printf("Default (VP)\n"); break;
    case cudaVideoCreate_PreferCUDA: printf("Use CUDA decoder\n"); break;
    case cudaVideoCreate_PreferDXVA: printf("Use DXVA decoder\n"); break;
    case cudaVideoCreate_PreferCUVID: printf("Use CUVID decoder\n"); break;
    default: printf("Unknown value\n"); break;
}
assert(cudaVideoCodec_MPEG1 == rVideoFormat.codec || 
	   cudaVideoCodec_MPEG2 == rVideoFormat.codec || 
	   cudaVideoCodec_MPEG4 == rVideoFormat.codec ||
	   cudaVideoCodec_VC1   == rVideoFormat.codec || 
	   cudaVideoCodec_H264  == rVideoFormat.codec ||
	   cudaVideoCodec_JPEG  == rVideoFormat.codec ||
	   cudaVideoCodec_YUV420== rVideoFormat.codec ||
	   cudaVideoCodec_YV12  == rVideoFormat.codec ||
	   cudaVideoCodec_NV12  == rVideoFormat.codec ||
       cudaVideoCodec_YUYV  == rVideoFormat.codec ||
	   cudaVideoCodec_UYVY  == rVideoFormat.codec );

assert(cudaVideoChromaFormat_Monochrome == rVideoFormat.chroma_format ||
	   cudaVideoChromaFormat_420        == rVideoFormat.chroma_format ||
	   cudaVideoChromaFormat_422        == rVideoFormat.chroma_format ||
	   cudaVideoChromaFormat_444        == rVideoFormat.chroma_format);

// Fill the decoder-create-info struct from the given video-format struct.
memset(&oVideoDecodeCreateInfo_, 0, sizeof(CUVIDDECODECREATEINFO));
        // Create video decoder
oVideoDecodeCreateInfo_.CodecType           = rVideoFormat.codec;
oVideoDecodeCreateInfo_.ulWidth             = rVideoFormat.coded_width;
oVideoDecodeCreateInfo_.ulHeight            = rVideoFormat.coded_height;
oVideoDecodeCreateInfo_.ulNumDecodeSurfaces = FrameQueue::cnMaximumSize;
        // Limit decode memory to 24MB (16M pixels at 4:2:0 = 24M bytes)
while (oVideoDecodeCreateInfo_.ulNumDecodeSurfaces * rVideoFormat.coded_width * rVideoFormat.coded_height > 16*1024*1024)
{
    oVideoDecodeCreateInfo_.ulNumDecodeSurfaces--;
}
oVideoDecodeCreateInfo_.ChromaFormat        = rVideoFormat.chroma_format;
oVideoDecodeCreateInfo_.OutputFormat        = cudaVideoSurfaceFormat_NV12;
oVideoDecodeCreateInfo_.DeinterlaceMode     = cudaVideoDeinterlaceMode_Adaptive;

        // No scaling
oVideoDecodeCreateInfo_.ulTargetWidth       = oVideoDecodeCreateInfo_.ulWidth;
oVideoDecodeCreateInfo_.ulTargetHeight      = oVideoDecodeCreateInfo_.ulHeight;
oVideoDecodeCreateInfo_.ulNumOutputSurfaces = MAX_FRAME_COUNT;  // We won't simultaneously map more than 8 surfaces
oVideoDecodeCreateInfo_.ulCreationFlags     = m_VideoCreateFlags;
oVideoDecodeCreateInfo_.vidLock             = m_VidCtxLock;
        // create the decoder