49 lines
2.1 KiB
Python
49 lines
2.1 KiB
Python
# Copyright 2013-2019 Lawrence Livermore National Security, LLC and other
|
|
# Spack Project Developers. See the top-level COPYRIGHT file for details.
|
|
#
|
|
# SPDX-License-Identifier: (Apache-2.0 OR MIT)
|
|
|
|
from spack import *
|
|
|
|
|
|
class Aluminum(CMakePackage):
|
|
"""Aluminum provides a generic interface to high-performance
|
|
communication libraries, with a focus on allreduce
|
|
algorithms. Blocking and non-blocking algorithms and GPU-aware
|
|
algorithms are supported. Aluminum also contains custom
|
|
implementations of select algorithms to optimize for certain
|
|
situations."""
|
|
|
|
homepage = "https://github.com/LLNL/Aluminum"
|
|
url = "https://github.com/LLNL/Aluminum/archive/v0.1.tar.gz"
|
|
git = "https://github.com/LLNL/Aluminum.git"
|
|
|
|
version('master', branch='master')
|
|
version('0.3.3', sha256='26e7f263f53c6c6ee0fe216e981a558dfdd7ec997d0dd2a24285a609a6c68f3b')
|
|
version('0.3.2', sha256='09b6d1bcc02ac54ba269b1123eee7be20f0104b93596956c014b794ba96b037f')
|
|
version('0.2.1-1', sha256='066b750e9d1134871709a3e2414b96b166e0e24773efc7d512df2f1d96ee8eef')
|
|
version('0.2.1', sha256='3d5d15853cccc718f60df68205e56a2831de65be4d96e7f7e8497097e7905f89')
|
|
version('0.2', sha256='fc8f06c6d8faab17a2aedd408d3fe924043bf857da1094d5553f35c4d2af893b')
|
|
version('0.1', sha256='3880b736866e439dd94e6a61eeeb5bb2abccebbac82b82d52033bc6c94950bdb')
|
|
|
|
variant('gpu', default=False, description='Builds with support for GPUs via CUDA and cuDNN')
|
|
variant('nccl', default=False, description='Builds with support for NCCL communication lib')
|
|
variant('mpi_cuda', default=False, description='Builds with support for MPI-CUDA enabled library')
|
|
|
|
depends_on('cmake@3.9.0:', type='build')
|
|
depends_on('cuda', when='+gpu')
|
|
depends_on('mpi')
|
|
depends_on('nccl', when='+nccl')
|
|
depends_on('hwloc')
|
|
|
|
generator = 'Ninja'
|
|
depends_on('ninja', type='build')
|
|
|
|
def cmake_args(self):
|
|
spec = self.spec
|
|
args = [
|
|
'-DALUMINUM_ENABLE_CUDA:BOOL=%s' % ('+gpu' in spec),
|
|
'-DALUMINUM_ENABLE_MPI_CUDA:BOOL=%s' % ('+mpi_cuda' in spec),
|
|
'-DALUMINUM_ENABLE_NCCL:BOOL=%s' % ('+nccl' in spec)]
|
|
return args
|