launch_parameters.h

Go to the documentation of this file.

/*
 * Copyright (c) 2014-2015, NVIDIA CORPORATION
 * Copyright (c) 2015, Nuno Subtil <subtil@gmail.com>
 * Copyright (c) 2015, Roche Molecular Systems Inc.
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *    * Redistributions of source code must retain the above copyright
 *      notice, this list of conditions and the following disclaimer.
 *    * Redistributions in binary form must reproduce the above copyright
 *      notice, this list of conditions and the following disclaimer in the
 *      documentation and/or other materials provided with the distribution.
 *    * Neither the name of the copyright holders nor the names of its
 *      contributors may be used to endorse or promote products derived from
 *      this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
 * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#pragma once

namespace lift {

// given a __global__ entry point and a number of elements to process,
// compute CUDA launch parameters that cover all elements in the input array and maximize occupancy
// http://devblogs.nvidia.com/parallelforall/cuda-pro-tip-occupancy-api-simplifies-launch-configuration/
// note that this is not necessarily the most efficient launch configuration, just "not too bad"
template <class T>
int2 launch_parameters(T kernel, size_t elements, int dynamic_smem_size = 0)
{
    int block_size;     // block size returned by launch configurator
    int min_grid_size;  // minimum grid required for full occupancy
    int grid_size;      // required grid size based on input size (elements)

    cudaError_t err;

    // figure out the largest potential block size for the function
    err = cudaOccupancyMaxPotentialBlockSize(&min_grid_size, &block_size,
                                             kernel, dynamic_smem_size, 0);

    if (err != cudaSuccess)
    {
        fprintf(stderr, "ERROR: cudaOccupancyMaxPotentialBlockSize failed (%d)\n", err);
        abort();
    }

    // round up the grid size according to the number of input elements
    grid_size = (elements + block_size - 1) / block_size;

    return make_int2(grid_size, block_size);
}

} // namespace lift