forked from alexeedm/pytorch-fortran
-
Notifications
You must be signed in to change notification settings - Fork 0
/
make_all.sh
executable file
·75 lines (65 loc) · 2.99 KB
/
make_all.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
# Copyright (c) 2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# MIT License
#
# 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.
#!/bin/bash
set -ex
# This wrapper uses the NVHPC compiler to build
# training and inference examples
#
# Need to kick off 2 separate cmake builds with different compilers:
# 1. with GCC, build pytorch C++ wrapper lib that exposes things to Fortran
# 2. with NVHPC, build fortran bindings that just bind(c) to built lib from (1)
# TODO: set path for nvhpc compilers (subshell?)
# Note for installing in Bridges-2 (for CPU support only and intel compilers): Sungduk
# [1] load intelmpi, eg, module load intelmpi/20.4-intel20.4
# [2] activate cmake env for a newer cmake version, eg, conda activate cmake
# (pytorch, torchvision, cmake, ... are installed in env cmake, Pillow needed some code mod)
# [3] install libtorch (https://pytorch.org/cppdocs/installing.html) -> CMAKE_PREFIX_PATH
NVPATH=/opt/nvidia/hpc_sdk/Linux_x86_64/21.9/compilers/bin/:$PATH
CMAKE_PREFIX_PATH=/ocean/projects/atm200007p/shared/libtorch
CONFIG=Debug
OPENACC=0
# List CUDA compute capabilities
TORCH_CUDA_ARCH_LIST=7.0
INST=${1:-$(pwd -P)/install}
mkdir -p build_proxy build_fortproxy build_example
# c++ wrappers
(
cd build_proxy
cmake -DCMAKE_INSTALL_PREFIX=$INST -DCMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH -DCMAKE_CXX_COMPILER=mpiicc ../src/proxy_lib
cmake --build . --config $CONFIG --parallel
make install
)
# fortran bindings
(
export PATH=$NVPATH:$PATH
cd build_fortproxy
cmake -DOPENACC=$OPENACC -DCMAKE_Fortran_COMPILER=mpiifort -DCMAKE_CXX_COMPILER=mpiicc -DCMAKE_INSTALL_PREFIX=$INST -DCMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH/lib ../src/f90_bindings/
cmake --build . --config $CONFIG --parallel
make install
)
# fortran examples
(
export PATH=$NVPATH:$PATH
cd build_example
cmake -DOPENACC=$OPENACC -DCMAKE_Fortran_COMPILER=mpiifort -DCMAKE_CXX_COMPILER=mpiicc -DCMAKE_INSTALL_PREFIX=$INST ../examples/
cmake --build . --config $CONFIG --parallel
make install
)