2 # This file is part of the GROMACS molecular simulation package.
4 # Copyright (c) 2014,2015,2016,2017,2018 by the GROMACS development team.
5 # Copyright (c) 2019,2020, by the GROMACS development team, led by
6 # Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
7 # and including many others, as listed in the AUTHORS file in the
8 # top-level source directory and at http://www.gromacs.org.
10 # GROMACS is free software; you can redistribute it and/or
11 # modify it under the terms of the GNU Lesser General Public License
12 # as published by the Free Software Foundation; either version 2.1
13 # of the License, or (at your option) any later version.
15 # GROMACS is distributed in the hope that it will be useful,
16 # but WITHOUT ANY WARRANTY; without even the implied warranty of
17 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18 # Lesser General Public License for more details.
20 # You should have received a copy of the GNU Lesser General Public
21 # License along with GROMACS; if not, see
22 # http://www.gnu.org/licenses, or write to the Free Software Foundation,
23 # Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
25 # If you want to redistribute modifications to GROMACS, please
26 # consider that scientific software is very special. Version
27 # control is crucial - bugs must be traceable. We will be happy to
28 # consider code for inclusion in the official distribution, but
29 # derived work must not be called official GROMACS. Details are found
30 # in the README & COPYING files - if they are missing, get the
31 # official version at http://www.gromacs.org.
33 # To help us fund GROMACS development, we humbly ask that you cite
34 # the research papers on the package. Check out http://www.gromacs.org.
36 gmx_add_libgromacs_sources(
37 calculate_spline_moduli.cpp
40 long_range_correction.cpp
44 pme_load_balancing.cpp
51 # Files that implement stubs
53 pme_pp_comm_gpu_impl.cpp
54 pme_coordinate_receiver_gpu_impl.cpp
55 pme_force_sender_gpu_impl.cpp
58 gmx_add_libgromacs_sources(
59 # CUDA-specific sources
64 pme_gpu_program_impl.cu
65 pme_pp_comm_gpu_impl.cu
66 pme_force_sender_gpu_impl.cu
67 pme_coordinate_receiver_gpu_impl.cu
68 # GPU-specific sources
73 gmx_compile_cpp_as_cuda(
78 elseif (GMX_USE_OPENCL)
79 gmx_add_libgromacs_sources(
80 # OpenCL-specific sources
82 pme_gpu_program_impl_ocl.cpp
83 # GPU-specific sources
89 gmx_add_libgromacs_sources(
90 # Files that implement stubs
91 pme_gpu_program_impl.cpp
96 add_subdirectory(tests)
100 set(PME_OCL_KERNEL_SOURCES
101 "${CMAKE_CURRENT_SOURCE_DIR}/pme_gpu_utils.clh"
102 "${CMAKE_CURRENT_SOURCE_DIR}/pme_solve.clh"
103 "${CMAKE_CURRENT_SOURCE_DIR}/pme_gather.clh"
104 "${CMAKE_CURRENT_SOURCE_DIR}/pme_spread.clh")
107 set(OCL_COMPILER "${CLANG_TIDY_EXE}")
108 set(CLANG_TIDY_ARGS "-quiet;-checks=*,-readability-implicit-bool-conversion,-llvm-header-guard,-hicpp-signed-bitwise,-clang-analyzer-deadcode.DeadStores,-google-readability-todo,-clang-diagnostic-padded,-fcomment-block-commands=internal;--;${CMAKE_C_COMPILER}")
110 set(OCL_COMPILER "${CMAKE_C_COMPILER}")
113 # TODO: test all warp sizes on all vendor targets?
114 foreach(VENDOR AMD NVIDIA INTEL)
115 foreach(WARPSIZE 16 32 64)
116 math(EXPR SPREAD_WG_SIZE "8*${WARPSIZE}")
117 math(EXPR SOLVE_WG_SIZE "8*${WARPSIZE}")
118 math(EXPR GATHER_WG_SIZE "4*${WARPSIZE}")
119 set(OBJ_FILE pme_ocl_kernel_warpSize${WARPSIZE}_${VENDOR}.o)
120 add_custom_command(OUTPUT ${OBJ_FILE} COMMAND ${OCL_COMPILER}
121 ${CMAKE_CURRENT_SOURCE_DIR}/pme_program.cl ${CLANG_TIDY_ARGS}
122 -Xclang -finclude-default-header -D_${VENDOR}_SOURCE_
123 -Dwarp_size=${WARPSIZE}
126 -Dc_pmeMaxUnitcellShift=2
127 -Dc_skipNeutralAtoms=false
128 -Dc_virialAndEnergyCount=7
129 -Dc_spreadWorkGroupSize=${SPREAD_WG_SIZE}
130 -Dc_solveMaxWorkGroupSize=${SOLVE_WG_SIZE}
131 -Dc_gatherWorkGroupSize=${GATHER_WG_SIZE}
132 -DDIM=3 -DXX=0 -DYY=1 -DZZ=2
133 -DwrapX=true -DwrapY=true
134 -c -I ${CMAKE_SOURCE_DIR}/src -std=cl1.2
135 -Weverything -Wno-conversion -Wno-missing-variable-declarations -Wno-used-but-marked-unused
136 -Wno-cast-align -Wno-incompatible-pointer-types
137 # to avoid "warning: unknown command tag name" for \internal
138 -Wno-documentation-unknown-command
139 # to avoid pme_gpu_types.h:100:52: warning: padding struct 'struct PmeGpuConstParams' with 4 bytes to align 'd_virialAndEnergy'
143 list(APPEND PME_OCL_KERNELS ${OBJ_FILE})
147 add_custom_target(ocl_pme_kernels DEPENDS ${PME_OCL_KERNELS} )
148 gmx_set_custom_target_output(ocl_pme_kernels ${PME_OCL_KERNELS})