/*
Copyright (c) 2017 - 2022 Advanced Micro Devices, Inc. All rights reserved.

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/

#include "kernels.h"

struct FullyConnectedLayerLocalData {
    NeuralNetworkCommonHandle * handle;
    miopenConvolutionDescriptor_t convdesc;
    miopenTensorDescriptor_t input_desc;
    miopenTensorDescriptor_t output_desc;
    miopenTensorDescriptor_t weight_desc;
    miopenTensorDescriptor_t bias_desc;
    miopenDataType_t data_type;          // data_type for the kernel
    void *input_mem;
    void *output_mem;
    void *weight_mem;
    void *bias_mem;
    miopenConvFwdAlgorithm_t algo;
    size_t workspace_size;
    float alpha;
    float beta;
    void *workspace;

};

static vx_status VX_CALLBACK validateFullyConnectedLayer(vx_node node, const vx_reference parameters[], vx_uint32 num, vx_meta_format metas[])
{
    // check scalar type
    vx_enum type, out_type;
    ERROR_CHECK_STATUS(vxQueryScalar((vx_scalar)parameters[3], VX_SCALAR_TYPE, &type, sizeof(type)));
    if(type != VX_TYPE_ENUM) return ERRMSG(VX_ERROR_INVALID_TYPE, "validate: FC: #3 type=%d (must be ENUM)\n", type);
    ERROR_CHECK_STATUS(vxQueryScalar((vx_scalar)parameters[4], VX_SCALAR_TYPE, &type, sizeof(type)));
    if(type != VX_TYPE_ENUM) return ERRMSG(VX_ERROR_INVALID_TYPE, "validate: FC: #4 type=%d (must be ENUM)\n", type);

    // check tensor dimensions
    vx_size num_dims;
    vx_size input_dims[4] = { 1, 1, 1, 1 }, weights_dims[4] = { 1, 1, 0, 0 }, output_dims[4] = { 1, 1, 1, 1 };
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_NUMBER_OF_DIMS, &num_dims, sizeof(num_dims)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_DATA_TYPE, &type, sizeof(type)));
    if(num_dims != 4) return ERRMSG(VX_ERROR_INVALID_DIMENSION, "validate: FC: #0 num_dims=%ld (must be 4)\n", num_dims);
    if((type != VX_TYPE_FLOAT32) && (type != VX_TYPE_FLOAT16)) return ERRMSG(VX_ERROR_INVALID_TYPE, "validate: FC: #0 type=%d (must be float)\n", type);
    out_type = type;        // has to be same as input
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_DIMS, input_dims, sizeof(input_dims)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_NUMBER_OF_DIMS, &num_dims, sizeof(num_dims)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_DATA_TYPE, &type, sizeof(type)));
    if(num_dims != 2 && num_dims != 4) return ERRMSG(VX_ERROR_INVALID_DIMENSION, "validate: FC: #1 num_dims=%ld (must be 2 or 4)\n", num_dims);
    if((type != VX_TYPE_FLOAT32) && (type != VX_TYPE_FLOAT16)) return ERRMSG(VX_ERROR_INVALID_TYPE, "validate: FC: #1 type=%d (must be float)\n", type);
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_DIMS, &weights_dims[4 - num_dims], num_dims * sizeof(vx_size)));
    if(parameters[2]) {
        ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_NUMBER_OF_DIMS, &num_dims, sizeof(num_dims)));
        ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_DATA_TYPE, &type, sizeof(type)));
        if(num_dims != 1 && num_dims != 2) return ERRMSG(VX_ERROR_INVALID_DIMENSION, "validate: FC: #2 num_dims=%ld (must be 1 or 2)\n", num_dims);
        if ((type != VX_TYPE_FLOAT32) && (type != VX_TYPE_FLOAT16)) return ERRMSG(VX_ERROR_INVALID_TYPE, "validate: FC: #2 type=%d (must be float)\n", type);
        vx_size bias_dims[2] = { 0, 1 };
        ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_DIMS, bias_dims, num_dims * sizeof(vx_size)));
        if(bias_dims[0] != weights_dims[3] || bias_dims[1] != 1) return ERRMSG(VX_ERROR_INVALID_DIMENSION, "validate: FC: bias[%ldx%ld] weights[%ldx%ldx%ldx%ld]\n", bias_dims[1], bias_dims[0], weights_dims[3], weights_dims[2], weights_dims[1], weights_dims[0]);
    }
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_NUMBER_OF_DIMS, &num_dims, sizeof(num_dims)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_DATA_TYPE, &out_type, sizeof(out_type)));
    if(num_dims != 2 && num_dims != 4) return ERRMSG(VX_ERROR_INVALID_DIMENSION, "validate: FC: #5 num_dims=%ld (must be 2 or 4)\n", num_dims);
    if((out_type != VX_TYPE_FLOAT32) && (out_type != VX_TYPE_FLOAT16)) return ERRMSG(VX_ERROR_INVALID_TYPE, "validate: FC: #5 type=%d (must be float)\n", type);
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_DIMS, &output_dims[4-num_dims], num_dims * sizeof(vx_size)));
    if(output_dims[3] != input_dims[3] || input_dims[2]*input_dims[1]*input_dims[0] != weights_dims[2]*weights_dims[1]*weights_dims[0] || output_dims[2] != weights_dims[3])
        return ERRMSG(VX_ERROR_INVALID_DIMENSION, "validate: FC: input[%ldx%ldx%ldx%ld] weights[%ldx%ldx%ldx%ld] output[%ldx%ldx%ldx%ld]\n",
            input_dims[3], input_dims[2], input_dims[1], input_dims[0],
            weights_dims[3], weights_dims[2], weights_dims[1], weights_dims[0],
            output_dims[3], output_dims[2], output_dims[1], output_dims[0]);

    // output tensor configuration
    ERROR_CHECK_STATUS(vxSetMetaFormatAttribute(metas[5], VX_TENSOR_DATA_TYPE, &out_type, sizeof(out_type)));
    ERROR_CHECK_STATUS(vxSetMetaFormatAttribute(metas[5], VX_TENSOR_NUMBER_OF_DIMS, &num_dims, sizeof(num_dims)));
    ERROR_CHECK_STATUS(vxSetMetaFormatAttribute(metas[5], VX_TENSOR_DIMS, &output_dims[4-num_dims], num_dims * sizeof(vx_size)));
    return VX_SUCCESS;
}

static vx_status VX_CALLBACK processFullyConnectedLayer(vx_node node, const vx_reference * parameters, vx_uint32 num)
{
PROFILER_START(VX_NN, Fully_Connected_Layer)
    FullyConnectedLayerLocalData * data = NULL;
    ERROR_CHECK_STATUS(vxQueryNode(node, VX_NODE_LOCAL_DATA_PTR, &data, sizeof(data)));
    miopenHandle_t miopen_handle = data->handle->miopen_handle;

#if ENABLE_OPENCL
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_BUFFER_OPENCL, &data->input_mem, sizeof(data->input_mem)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_BUFFER_OPENCL, &data->output_mem, sizeof(data->output_mem)));
#elif ENABLE_HIP
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_BUFFER_HIP, &data->input_mem, sizeof(data->input_mem)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_BUFFER_HIP, &data->output_mem, sizeof(data->output_mem)));
#endif

    //ConvolutionForward.
    ERROR_CHECK_MIOPEN_STATUS(miopenConvolutionForward(data->handle->miopen_handle, &data->alpha, data->input_desc, data->input_mem,
                                                       data->weight_desc, data->weight_mem, data->convdesc, data->algo, &data->beta, data->output_desc, data->output_mem, data->workspace, data->workspace_size));

    //Convolution Forward Bias.
	if(parameters[2]) {
	    ERROR_CHECK_MIOPEN_STATUS(miopenConvolutionForwardBias(data->handle->miopen_handle, &data->alpha, data->bias_desc, data->bias_mem,
                                                           &data->beta, data->output_desc, data->output_mem));
	}
    
    /*DUMP LAYER BUFFER*/
    #if ENABLE_DEBUG_DUMP_NN_LAYER_BUFFERS
        //dump the output layer
        nn_layer_test_dumpBuffer("conv_%04d.bin", (vx_tensor)parameters[5]);
    #endif

PROFILER_STOP(VX_NN, Fully_Connected_Layer)
    return VX_SUCCESS;
}

static vx_status VX_CALLBACK initializeFullyConnectedLayer(vx_node node, const vx_reference *parameters, vx_uint32 num)
{
    FullyConnectedLayerLocalData * data = new FullyConnectedLayerLocalData;
    memset(data, 0, sizeof(*data));
    ERROR_CHECK_STATUS(createGraphHandle(node, &data->handle));

    //input,weight,bias,output descriptors.
    miopenConvolutionMode_t mode = miopenConvolution;
    vx_size num_dims;
    vx_enum out_type;
    vx_size input_dims[4], weights_dims[4] = { 1, 1, 0, 0 }, output_dims[4], bias_dims[2] = { 0, 1 };
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_DIMS, input_dims, sizeof(input_dims)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_NUMBER_OF_DIMS, &num_dims, sizeof(vx_size)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_DIMS, &weights_dims[4 - num_dims], num_dims * sizeof(vx_size)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_DIMS, output_dims, sizeof(output_dims)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_DATA_TYPE, &out_type, sizeof(out_type)));

    if(parameters[2]) {
        ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_NUMBER_OF_DIMS, &num_dims, sizeof(vx_size)));
        ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_DIMS, bias_dims, num_dims * sizeof(vx_size)));
    }
    data->data_type = (out_type == VX_TYPE_FLOAT32)? miopenFloat:miopenHalf;

    // adjust weights to match input
    input_dims[2] = weights_dims[2];
    input_dims[1] = weights_dims[1];
    input_dims[0] = weights_dims[0];

    ERROR_CHECK_MIOPEN_STATUS(miopenCreateTensorDescriptor(&data->input_desc));
    ERROR_CHECK_MIOPEN_STATUS(miopenCreateTensorDescriptor(&data->weight_desc));
    ERROR_CHECK_MIOPEN_STATUS(miopenCreateTensorDescriptor(&data->output_desc));
    ERROR_CHECK_MIOPEN_STATUS(miopenCreateTensorDescriptor(&data->bias_desc));
    ERROR_CHECK_MIOPEN_STATUS(miopenSet4dTensorDescriptor(data->input_desc, data->data_type, input_dims[3], input_dims[2], input_dims[1], input_dims[0]));
    ERROR_CHECK_MIOPEN_STATUS(miopenSet4dTensorDescriptor(data->output_desc, data->data_type, output_dims[3], output_dims[2], output_dims[1], output_dims[0]));
    ERROR_CHECK_MIOPEN_STATUS(miopenSet4dTensorDescriptor(data->weight_desc, data->data_type, weights_dims[3], weights_dims[2], weights_dims[1], weights_dims[0]));
    if(parameters[2]) {
        ERROR_CHECK_MIOPEN_STATUS(miopenSet4dTensorDescriptor(data->bias_desc, data->data_type,1, bias_dims[0], 1, 1));
    }

    //fully connected to convolution conversion.
    ERROR_CHECK_MIOPEN_STATUS(miopenCreateConvolutionDescriptor(&data->convdesc));
    ERROR_CHECK_MIOPEN_STATUS(miopenInitConvolutionDescriptor(data->convdesc, mode, 0, 0, 1, 1, 1, 1));

#if ENABLE_OPENCL
    //Memory Declaration.
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_BUFFER_OPENCL, &data->input_mem, sizeof(data->input_mem)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_BUFFER_OPENCL, &data->output_mem, sizeof(data->output_mem)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_BUFFER_OPENCL, &data->weight_mem, sizeof(data->weight_mem)));
    if(parameters[2]) {
        ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_BUFFER_OPENCL, &data->bias_mem, sizeof(data->bias_mem)));
    }
#elif ENABLE_HIP
    //Memory Declaration.
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[0], VX_TENSOR_BUFFER_HIP, &data->input_mem, sizeof(data->input_mem)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[5], VX_TENSOR_BUFFER_HIP, &data->output_mem, sizeof(data->output_mem)));
    ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[1], VX_TENSOR_BUFFER_HIP, &data->weight_mem, sizeof(data->weight_mem)));
    if(parameters[2]) {
        ERROR_CHECK_STATUS(vxQueryTensor((vx_tensor)parameters[2], VX_TENSOR_BUFFER_HIP, &data->bias_mem, sizeof(data->bias_mem)));
    }
#endif
    data->alpha = 1;
    data->beta = 0;

    //Workspace Size.
    ERROR_CHECK_MIOPEN_STATUS(miopenConvolutionForwardGetWorkSpaceSize(data->handle->miopen_handle, data->weight_desc, data->input_desc, data->convdesc, data->output_desc, &data->workspace_size ));
    if (data->workspace_size > 0) {
        vx_context   vxContext = vxGetContext((vx_reference)node);
#if ENABLE_OPENCL
        cl_context context;
        ERROR_CHECK_STATUS(vxQueryContext(vxContext, VX_CONTEXT_ATTRIBUTE_AMD_OPENCL_CONTEXT, &context, sizeof(context)));
        data->workspace_size = (data->workspace_size + 3) & ~3;
        data->workspace = clCreateBuffer(context, CL_MEM_READ_WRITE, data->workspace_size, NULL, NULL);
        if (!data->workspace) {
            return VX_FAILURE;
        }
        cl_float pattern = 0;
        cl_int err = 0;
        err = clEnqueueFillBuffer(data->handle->cmdq, (cl_mem)data->workspace, &pattern, sizeof(cl_float), 0, data->workspace_size, 0, NULL, NULL);
        if(err) return VX_FAILURE;
#elif ENABLE_HIP
        int hip_device = -1;
        ERROR_CHECK_STATUS(vxQueryContext(vxContext, VX_CONTEXT_ATTRIBUTE_AMD_HIP_DEVICE, &hip_device, sizeof(hip_device)));
        if (hip_device < 0) {
            return VX_FAILURE;
        }
        hipError_t errcode_ret = hipSuccess;
        data->workspace_size = (data->workspace_size + 3) & ~3;
        errcode_ret = hipMalloc(&data->workspace, data->workspace_size);
        if (errcode_ret != hipSuccess) {
            return VX_FAILURE;
        }

        errcode_ret = hipMemset(data->workspace, 0, data->workspace_size);
        if (errcode_ret != hipSuccess) {
            return VX_FAILURE;
        }
#endif
    }
    //Finding best Convolution Algorithm.
    miopenConvAlgoPerf_t perf;
    int algo_count;
    ERROR_CHECK_MIOPEN_STATUS(miopenFindConvolutionForwardAlgorithm(data->handle->miopen_handle, data->input_desc, data->input_mem, data->weight_desc, data->weight_mem, data->convdesc, data->output_desc, data->output_mem, 1, &algo_count, &perf, data->workspace, data->workspace_size, data->handle->exhaustiveSearch));
    data->algo = perf.fwd_algo;

#if ENABLE_DEBUG_PRINT_DIMS
    std::cout << "fullyconnected input " << input_dims[3] << " " << input_dims[2] << " " << input_dims[1] << " " << input_dims[0] << " ";
    std::cout << "weights " << weights_dims[3] << weights_dims[2] << weights_dims[1] << weights_dims[0] << " ";
    std::cout << "bias " << bias_dims[0] << " ";
    std::cout << "output " << output_dims[3] << " " << output_dims[2] << " " << output_dims[1] << " " << output_dims[0] << std::endl;
#endif

    //add to node attribute.
    ERROR_CHECK_STATUS(vxSetNodeAttribute(node, VX_NODE_LOCAL_DATA_PTR, &data, sizeof(data)));
    return VX_SUCCESS;
}

static vx_status VX_CALLBACK uninitializeFullyConnectedLayer(vx_node node, const vx_reference *parameters, vx_uint32 num)
{
    FullyConnectedLayerLocalData * data = NULL;
    ERROR_CHECK_STATUS(vxQueryNode(node, VX_NODE_LOCAL_DATA_PTR, &data, sizeof(data)));
#if ENABLE_OPENCL
    if(data && data->workspace && clReleaseMemObject((cl_mem)data->workspace) != 0 ) return VX_FAILURE;
#elif ENABLE_HIP
    if (data->workspace) {
        hipError_t errcode_ret = hipFree(data->workspace);
        if (errcode_ret != hipSuccess) {
            return VX_FAILURE;
        }
    }
#endif
    ERROR_CHECK_MIOPEN_STATUS(miopenDestroyConvolutionDescriptor(data->convdesc));
    ERROR_CHECK_MIOPEN_STATUS(miopenDestroyTensorDescriptor(data->input_desc));
    ERROR_CHECK_MIOPEN_STATUS(miopenDestroyTensorDescriptor(data->output_desc));
    ERROR_CHECK_MIOPEN_STATUS(miopenDestroyTensorDescriptor(data->weight_desc));
    ERROR_CHECK_MIOPEN_STATUS(miopenDestroyTensorDescriptor(data->bias_desc));
    if (data) {
        ERROR_CHECK_STATUS(releaseGraphHandle(node, data->handle));
        delete data;
    }
    return VX_SUCCESS;
}

vx_status publishFullyConnectedLayer(vx_context context)
{
    // add kernel to the context with callbacks
    vx_kernel kernel = vxAddUserKernel(context, "org.khronos.nn_extension.fully_connected_layer", VX_KERNEL_FULLY_CONNECTED_LAYER, processFullyConnectedLayer, 6, validateFullyConnectedLayer, initializeFullyConnectedLayer, uninitializeFullyConnectedLayer);
    ERROR_CHECK_OBJECT(kernel);

    // enable OpenCL buffer access since the kernel_f callback uses OpenCL buffers instead of host accessible buffers
    vx_bool enableBufferAccess = vx_true_e;
    ERROR_CHECK_STATUS(vxSetKernelAttribute(kernel, VX_KERNEL_ATTRIBUTE_AMD_GPU_BUFFER_ACCESS_ENABLE, &enableBufferAccess, sizeof(enableBufferAccess)));

    // set kernel parameters
    ERROR_CHECK_STATUS(vxAddParameterToKernel(kernel, 0, VX_INPUT, VX_TYPE_TENSOR, VX_PARAMETER_STATE_REQUIRED));
    ERROR_CHECK_STATUS(vxAddParameterToKernel(kernel, 1, VX_INPUT, VX_TYPE_TENSOR, VX_PARAMETER_STATE_REQUIRED));
    ERROR_CHECK_STATUS(vxAddParameterToKernel(kernel, 2, VX_INPUT, VX_TYPE_TENSOR, VX_PARAMETER_STATE_OPTIONAL));
    ERROR_CHECK_STATUS(vxAddParameterToKernel(kernel, 3, VX_INPUT, VX_TYPE_SCALAR, VX_PARAMETER_STATE_REQUIRED));
    ERROR_CHECK_STATUS(vxAddParameterToKernel(kernel, 4, VX_INPUT, VX_TYPE_SCALAR, VX_PARAMETER_STATE_REQUIRED));
    ERROR_CHECK_STATUS(vxAddParameterToKernel(kernel, 5, VX_OUTPUT, VX_TYPE_TENSOR, VX_PARAMETER_STATE_REQUIRED));

    // finalize and release kernel object
    ERROR_CHECK_STATUS(vxFinalizeKernel(kernel));
    ERROR_CHECK_STATUS(vxReleaseKernel(&kernel));

    return VX_SUCCESS;
}

VX_API_ENTRY vx_node VX_API_CALL vxFullyConnectedLayer(vx_graph graph, vx_tensor inputs, vx_tensor weights,
                                                       vx_tensor biases, vx_enum overflow_policy, vx_enum rounding_policy, vx_tensor outputs)
{
    vx_node node = NULL;
    vx_context context = vxGetContext((vx_reference)graph);
    if(vxGetStatus((vx_reference)context) == VX_SUCCESS) {
        vx_scalar overflow = vxCreateScalarWithSize(context, VX_TYPE_ENUM, &overflow_policy, sizeof(overflow_policy));
        vx_scalar rounding = vxCreateScalarWithSize(context, VX_TYPE_ENUM, &rounding_policy, sizeof(rounding_policy));
        if(vxGetStatus((vx_reference)overflow) == VX_SUCCESS && vxGetStatus((vx_reference)rounding) == VX_SUCCESS) {
            vx_reference params[] = {
                (vx_reference)inputs,
                (vx_reference)weights,
                (vx_reference)biases,
                (vx_reference)overflow,
                (vx_reference)rounding,
                (vx_reference)outputs
            };
            node = createNode(graph, VX_KERNEL_FULLY_CONNECTED_LAYER, params, sizeof(params)/sizeof(params[0]));
            vxReleaseScalar(&overflow);
            vxReleaseScalar(&rounding);
        }
    }
    return node;
}