270 lines
11 KiB
Python
270 lines
11 KiB
Python
# Copyright 2013-2024 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.package import *
|
|
|
|
|
|
class DlaFuture(CMakePackage, CudaPackage, ROCmPackage):
|
|
"""DLA-Future library: Distributed Linear Algebra with Future"""
|
|
|
|
homepage = "https://github.com/eth-cscs/DLA-Future"
|
|
url = "https://github.com/eth-cscs/DLA-Future/archive/v0.0.0.tar.gz"
|
|
git = "https://github.com/eth-cscs/DLA-Future.git"
|
|
maintainers = ["rasolca", "albestro", "msimberg", "aurianer"]
|
|
|
|
license("BSD-3-Clause")
|
|
|
|
version("0.6.0", sha256="85dfcee36ff28fa44da3134408c40ebd611bccff8a295982a7c78eaf982524d9")
|
|
version("0.5.0", sha256="f964ee2a96bb58b3f0ee4563ae65fcd136e409a7c0e66beda33f926fc9515a8e")
|
|
version("0.4.1", sha256="ba95f26475ad68da1f3a24d091dc1b925525e269e4c83c1eaf1d37d29b526666")
|
|
version("0.4.0", sha256="34fd0da0d1a72b6981bed0bba029ba0947e0d0d99beb3e0aad0a478095c9527d")
|
|
version("0.3.1", sha256="350a7fd216790182aa52639a3d574990a9d57843e02b92d87b854912f4812bfe")
|
|
version("0.3.0", sha256="9887ac0b466ca03d704a8738bc89e68550ed33509578c576390e98e76b64911b")
|
|
version("0.2.1", sha256="4c2669d58f041304bd618a9d69d9879a42e6366612c2fc932df3894d0326b7fe")
|
|
version("0.2.0", sha256="da73cbd1b88287c86d84b1045a05406b742be924e65c52588bbff200abd81a10")
|
|
version("0.1.0", sha256="f7ffcde22edabb3dc24a624e2888f98829ee526da384cd752b2b271c731ca9b1")
|
|
version("master", branch="master")
|
|
|
|
depends_on("c", type="build")
|
|
depends_on("cxx", type="build")
|
|
|
|
variant("shared", default=True, description="Build shared libraries.")
|
|
|
|
variant(
|
|
"hdf5",
|
|
default=False,
|
|
when="@0.2.0:",
|
|
description="HDF5 support for dealing with matrices on disk.",
|
|
)
|
|
|
|
variant("doc", default=False, description="Build documentation.")
|
|
|
|
variant("miniapps", default=False, description="Build miniapps.")
|
|
|
|
variant(
|
|
"scalapack",
|
|
default=False,
|
|
when="@0.2.0:",
|
|
description="Build C API compatible with ScaLAPACK",
|
|
)
|
|
|
|
variant("mpi_gpu_aware", default=False, when="@0.5.0:", description="Use GPU-aware MPI.")
|
|
conflicts("+mpi_gpu_aware", when="~cuda ~rocm", msg="GPU-aware MPI requires +cuda or +rocm")
|
|
|
|
variant(
|
|
"mpi_gpu_force_contiguous",
|
|
default=True,
|
|
when="@0.5.0: +mpi_gpu_aware",
|
|
description="Force GPU communication buffers to be contiguous before communicating.",
|
|
)
|
|
|
|
generator("ninja")
|
|
|
|
depends_on("cmake@3.22:", type="build")
|
|
depends_on("pkgconfig", type="build")
|
|
depends_on("doxygen", type="build", when="+doc")
|
|
depends_on("mpi")
|
|
|
|
depends_on("blas")
|
|
depends_on("lapack")
|
|
depends_on("scalapack", when="+scalapack")
|
|
depends_on("blaspp@2022.05.00:")
|
|
depends_on("lapackpp@2022.05.00:")
|
|
depends_on("intel-oneapi-mkl +cluster", when="^[virtuals=scalapack] intel-oneapi-mkl")
|
|
|
|
conflicts("intel-oneapi-mkl", when="@:0.3")
|
|
|
|
depends_on("umpire~examples")
|
|
depends_on("umpire~cuda", when="~cuda")
|
|
depends_on("umpire~rocm", when="~rocm")
|
|
depends_on("umpire+cuda~shared", when="+cuda")
|
|
depends_on("umpire+rocm~shared", when="+rocm")
|
|
depends_on("umpire@4.1.0:")
|
|
|
|
depends_on("pika@0.15.1:", when="@0.1")
|
|
depends_on("pika@0.16:", when="@0.2.0")
|
|
depends_on("pika@0.17:", when="@0.2.1")
|
|
depends_on("pika@0.18:", when="@0.3")
|
|
depends_on("pika@0.19.1:", when="@0.4.0:")
|
|
depends_on("pika-algorithms@0.1:", when="@:0.2")
|
|
depends_on("pika +mpi")
|
|
depends_on("pika +cuda", when="+cuda")
|
|
depends_on("pika +rocm", when="+rocm")
|
|
|
|
conflicts("^pika cxxstd=20", when="+cuda")
|
|
|
|
depends_on("whip +cuda", when="+cuda")
|
|
depends_on("whip +rocm", when="+rocm")
|
|
|
|
depends_on("rocblas", when="+rocm")
|
|
depends_on("rocsolver", when="+rocm")
|
|
|
|
depends_on("rocprim", when="@:0.3 +rocm")
|
|
depends_on("rocthrust", when="@:0.3 +rocm")
|
|
|
|
# nvcc 11.2 and older is unable to detect fmt::formatter specializations.
|
|
# DLA-Future 0.3.1 includes a workaround to avoid including fmt in device
|
|
# code:
|
|
# https://github.com/pika-org/pika/issues/870
|
|
# https://github.com/eth-cscs/DLA-Future/pull/1045
|
|
conflicts("^fmt@10:", when="@:0.3.0 +cuda ^cuda@:11.2")
|
|
|
|
# Compilation problem triggered by the bundled fmt in Umpire together with
|
|
# fmt 10, which only happens with GCC 9 and nvcc 11.2 and older:
|
|
# https://github.com/eth-cscs/DLA-Future/issues/1044
|
|
conflicts("^fmt@10:", when="@:0.3.0 %gcc@9 +cuda ^cuda@:11.2 ^umpire@2022.10:")
|
|
|
|
# Pedantic warnings, triggered by GCC 9 and 10, are always errors until 0.3.1:
|
|
# https://github.com/eth-cscs/DLA-Future/pull/1043
|
|
conflicts("%gcc@9:10", when="@:0.3.0")
|
|
|
|
depends_on("hdf5 +cxx+mpi+threadsafe+shared", when="+hdf5")
|
|
|
|
conflicts("+cuda", when="+rocm")
|
|
|
|
with when("+rocm"):
|
|
for arch in ROCmPackage.amdgpu_targets:
|
|
depends_on(f"pika amdgpu_target={arch}", when=f"amdgpu_target={arch}")
|
|
depends_on(f"rocsolver amdgpu_target={arch}", when=f"amdgpu_target={arch}")
|
|
depends_on(f"rocblas amdgpu_target={arch}", when=f"amdgpu_target={arch}")
|
|
depends_on(f"whip amdgpu_target={arch}", when=f"amdgpu_target={arch}")
|
|
depends_on(f"umpire amdgpu_target={arch}", when=f"amdgpu_target={arch}")
|
|
|
|
with when("@:0.3 +rocm"):
|
|
for arch in ROCmPackage.amdgpu_targets:
|
|
depends_on(f"rocprim amdgpu_target={arch}", when=f"amdgpu_target={arch}")
|
|
depends_on(f"rocthrust amdgpu_target={arch}", when=f"amdgpu_target={arch}")
|
|
|
|
with when("+cuda"):
|
|
for arch in CudaPackage.cuda_arch_values:
|
|
depends_on(f"pika cuda_arch={arch}", when=f"cuda_arch={arch}")
|
|
depends_on(f"umpire cuda_arch={arch}", when=f"cuda_arch={arch}")
|
|
|
|
patch(
|
|
"https://github.com/eth-cscs/DLA-Future/commit/efc9c176a7a8c512b3f37d079dec8c25ac1b7389.patch?full_index=1",
|
|
sha256="f40e4a734650f56c39379717a682d00d6400a7a102d90821542652824a8f64cd",
|
|
when="@:0.3 %gcc@13:",
|
|
)
|
|
# https://github.com/spack/spack/issues/41511
|
|
patch("hip_complex_operator_overloads.patch", when="+rocm")
|
|
|
|
def cmake_args(self):
|
|
spec = self.spec
|
|
args = []
|
|
|
|
args.append(self.define_from_variant("BUILD_SHARED_LIBS", "shared"))
|
|
|
|
# BLAS/LAPACK
|
|
if spec.version <= Version("0.4") and spec["lapack"].name in INTEL_MATH_LIBRARIES:
|
|
mkl_provider = spec["lapack"].name
|
|
|
|
vmap = {
|
|
"intel-oneapi-mkl": {
|
|
"threading": {
|
|
"none": "sequential",
|
|
"openmp": "gnu_thread",
|
|
"tbb": "tbb_thread",
|
|
},
|
|
"mpi": {"intel-mpi": "intelmpi", "mpich": "mpich", "openmpi": "openmpi"},
|
|
},
|
|
"intel-mkl": {
|
|
"threading": {"none": "seq", "openmp": "omp", "tbb": "tbb"},
|
|
"mpi": {"intel-mpi": "mpich", "mpich": "mpich", "openmpi": "ompi"},
|
|
},
|
|
}
|
|
|
|
if mkl_provider not in vmap.keys():
|
|
raise RuntimeError(
|
|
f"dla-future does not support {mkl_provider} as lapack provider"
|
|
)
|
|
mkl_mapper = vmap[mkl_provider]
|
|
|
|
mkl_threads = mkl_mapper["threading"][spec[mkl_provider].variants["threads"].value]
|
|
if mkl_provider == "intel-oneapi-mkl":
|
|
args += [
|
|
self.define("DLAF_WITH_MKL", True),
|
|
self.define("MKL_INTERFACE", "lp64"),
|
|
self.define("MKL_THREADING", mkl_threads),
|
|
]
|
|
elif mkl_provider == "intel-mkl":
|
|
args += [
|
|
(
|
|
self.define("DLAF_WITH_MKL", True)
|
|
if spec.version <= Version("0.3")
|
|
else self.define("DLAF_WITH_MKL_LEGACY", True)
|
|
),
|
|
self.define("MKL_LAPACK_TARGET", f"mkl::mkl_intel_32bit_{mkl_threads}_dyn"),
|
|
]
|
|
|
|
if "+scalapack" in spec:
|
|
try:
|
|
mpi_provider = spec["mpi"].name
|
|
if mpi_provider in ["mpich", "cray-mpich", "mvapich", "mvapich2"]:
|
|
mkl_mpi = mkl_mapper["mpi"]["mpich"]
|
|
else:
|
|
mkl_mpi = mkl_mapper["mpi"][mpi_provider]
|
|
except KeyError:
|
|
raise RuntimeError(
|
|
f"dla-future does not support {spec['mpi'].name} as mpi provider with "
|
|
f"the selected scalapack provider {mkl_provider}"
|
|
)
|
|
|
|
if mkl_provider == "intel-oneapi-mkl":
|
|
args.append(self.define("MKL_MPI", mkl_mpi))
|
|
elif mkl_provider == "intel-mkl":
|
|
args.append(
|
|
self.define(
|
|
"MKL_SCALAPACK_TARGET",
|
|
f"mkl::scalapack_{mkl_mpi}_intel_32bit_{mkl_threads}_dyn",
|
|
)
|
|
)
|
|
else:
|
|
args.append(self.define("DLAF_WITH_MKL", spec["lapack"].name in INTEL_MATH_LIBRARIES))
|
|
args.append(
|
|
self.define(
|
|
"LAPACK_LIBRARY",
|
|
" ".join([spec[dep].libs.ld_flags for dep in ["blas", "lapack"]]),
|
|
)
|
|
)
|
|
if "+scalapack" in spec:
|
|
args.append(self.define("SCALAPACK_LIBRARY", spec["scalapack"].libs.ld_flags))
|
|
|
|
args.append(self.define_from_variant("DLAF_WITH_SCALAPACK", "scalapack"))
|
|
|
|
args.append(self.define_from_variant("DLAF_WITH_MPI_GPU_AWARE", "mpi_gpu_aware"))
|
|
args.append(
|
|
self.define_from_variant(
|
|
"DLAF_WITH_MPI_GPU_FORCE_CONTIGUOUS", "mpi_gpu_force_contiguous"
|
|
)
|
|
)
|
|
|
|
# CUDA/HIP
|
|
args.append(self.define_from_variant("DLAF_WITH_CUDA", "cuda"))
|
|
args.append(self.define_from_variant("DLAF_WITH_HIP", "rocm"))
|
|
if "+rocm" in spec:
|
|
archs = spec.variants["amdgpu_target"].value
|
|
if "none" not in archs:
|
|
arch_str = ";".join(archs)
|
|
args.append(self.define("CMAKE_HIP_ARCHITECTURES", arch_str))
|
|
if "+cuda" in spec:
|
|
archs = spec.variants["cuda_arch"].value
|
|
if "none" not in archs:
|
|
arch_str = ";".join(archs)
|
|
args.append(self.define("CMAKE_CUDA_ARCHITECTURES", arch_str))
|
|
|
|
# HDF5 support
|
|
args.append(self.define_from_variant("DLAF_WITH_HDF5", "hdf5"))
|
|
|
|
# DOC
|
|
args.append(self.define_from_variant("DLAF_BUILD_DOC", "doc"))
|
|
|
|
# TEST
|
|
args.append(self.define("DLAF_BUILD_TESTING", self.run_tests))
|
|
|
|
# MINIAPPS
|
|
args.append(self.define_from_variant("DLAF_BUILD_MINIAPPS", "miniapps"))
|
|
|
|
return args
|