Make cl_nbparam into a struct
[gromacs.git] / src / gromacs / gpu_utils / gputraits_ocl.h
bloba3eb510c95cf2135c9b82537c5bda106bdbe0a7a
1 /*
2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2018,2019,2020, by the GROMACS development team, led by
5 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
6 * and including many others, as listed in the AUTHORS file in the
7 * top-level source directory and at http://www.gromacs.org.
9 * GROMACS is free software; you can redistribute it and/or
10 * modify it under the terms of the GNU Lesser General Public License
11 * as published by the Free Software Foundation; either version 2.1
12 * of the License, or (at your option) any later version.
14 * GROMACS is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 * Lesser General Public License for more details.
19 * You should have received a copy of the GNU Lesser General Public
20 * License along with GROMACS; if not, see
21 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
22 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
24 * If you want to redistribute modifications to GROMACS, please
25 * consider that scientific software is very special. Version
26 * control is crucial - bugs must be traceable. We will be happy to
27 * consider code for inclusion in the official distribution, but
28 * derived work must not be called official GROMACS. Details are found
29 * in the README & COPYING files - if they are missing, get the
30 * official version at http://www.gromacs.org.
32 * To help us fund GROMACS development, we humbly ask that you cite
33 * the research papers on the package. Check out http://www.gromacs.org.
35 #ifndef GMX_GPU_UTILS_GPUTRAITS_OCL_H
36 #define GMX_GPU_UTILS_GPUTRAITS_OCL_H
38 /*! \libinternal \file
39 * \brief Declares the OpenCL type traits.
41 * \author Aleksei Iupinov <a.yupinov@gmail.com>
42 * \author Artem Zhmurov <zhmurov@gmail.com>
44 * \inlibraryapi
45 * \ingroup module_gpu_utils
48 #include "gromacs/gpu_utils/gmxopencl.h"
49 #include "gromacs/hardware/gpu_hw_info.h"
51 using DeviceTexture = void*;
53 //! OpenCL device vendors
54 enum class DeviceVendor : int
56 Unknown = 0, //!< No data
57 Nvidia = 1, //!< NVIDIA
58 Amd = 2, //!< Advanced Micro Devices
59 Intel = 3, //!< Intel
60 Count = 4
63 /*! \internal
64 * \brief OpenCL device information.
66 * The OpenCL device information is queried and set at detection and contains
67 * both information about the device/hardware returned by the runtime as well
68 * as additional data like support status.
70 struct DeviceInformation
72 cl_platform_id oclPlatformId; //!< OpenCL Platform ID.
73 cl_device_id oclDeviceId; //!< OpenCL Device ID.
74 char device_name[256]; //!< Device name.
75 char device_version[256]; //!< Device version.
76 char vendorName[256]; //!< Device vendor name.
77 int compute_units; //!< Number of compute units.
78 int adress_bits; //!< Number of address bits the device is capable of.
79 DeviceStatus stat; //!< Device status.
80 DeviceVendor deviceVendor; //!< Device vendor.
81 size_t maxWorkItemSizes[3]; //!< Workgroup size limits (CL_DEVICE_MAX_WORK_ITEM_SIZES).
82 size_t maxWorkGroupSize; //!< Workgroup total size limit (CL_DEVICE_MAX_WORK_GROUP_SIZE).
85 //! \brief Single GPU call timing event
86 using CommandEvent = cl_event;
88 /*! \internal \brief
89 * GPU kernels scheduling description. This is same in OpenCL/CUDA.
90 * Provides reasonable defaults, one typically only needs to set the GPU stream
91 * and non-1 work sizes.
93 struct KernelLaunchConfig
95 //! Work groups (CUDA blocks) counts
96 size_t gridSize[3] = { 1, 1, 1 };
97 //! Per work group (CUDA block) thread counts
98 size_t blockSize[3] = { 1, 1, 1 };
99 //! Shared memory size in bytes
100 size_t sharedMemorySize = 0;
103 /*! \brief Sets whether device code can use arrays that are embedded in structs.
104 * Note that OpenCL 2.x might be able to do this, but we use 1.2.
106 #define c_canEmbedBuffers false
108 #endif