mirror of
https://github.com/cp2k/cp2k.git
synced 2026-07-21 06:25:15 -04:00
Drop local spack recipe for dbcsr (#5576)
This commit is contained in:
Matthias Krack
GitHub
parent
29578c7298
commit
e2d766a6da
1 changed files with 0 additions and 262 deletions
|
|
@ -1,262 +0,0 @@
|
|||
# Copyright Spack Project Developers. See COPYRIGHT file for details.
|
||||
#
|
||||
# SPDX-License-Identifier: (Apache-2.0 OR MIT)
|
||||
|
||||
import os
|
||||
|
||||
from spack_repo.builtin.build_systems.cmake import CMakePackage, generator
|
||||
from spack_repo.builtin.build_systems.cuda import CudaPackage
|
||||
from spack_repo.builtin.build_systems.rocm import ROCmPackage
|
||||
|
||||
from spack.package import *
|
||||
|
||||
|
||||
class Dbcsr(CMakePackage, CudaPackage, ROCmPackage):
|
||||
"""The Distributed Block Compressed Sparse Row (DBCSR) matrix library is a
|
||||
library designed to efficiently perform sparse matrix-matrix multiplication,
|
||||
among other operations. It is MPI and OpenMP parallel and can exploit Nvidia
|
||||
and AMD GPUs via CUDA and HIP."""
|
||||
|
||||
homepage = "https://github.com/cp2k/dbcsr"
|
||||
git = "https://github.com/cp2k/dbcsr.git"
|
||||
url = "https://github.com/cp2k/dbcsr/releases/download/v2.2.0/dbcsr-2.2.0.tar.gz"
|
||||
list_url = "https://github.com/cp2k/dbcsr/releases"
|
||||
|
||||
maintainers("dev-zero", "mtaillefumier", "RMeli", "hfp", "mkrack")
|
||||
|
||||
license("GPL-2.0-or-later")
|
||||
|
||||
version("develop", branch="develop")
|
||||
version("2.10.0", sha256="3d897220fbb4498215331efad6905eb7744881b4cf04eb5c5fb4db7c48a56ef9")
|
||||
version("2.9.1", sha256="fa5a4aeba0a07761511af2c26c779bd811b5ea0ef06a5d94535b6dd7b2e0ce59")
|
||||
version("2.9.0", sha256="a04cacd2203bd97a31ac993f9ab84237a48191140bba29efadbc27db544bbcd6")
|
||||
version("2.8.0", sha256="d55e4f052f28d1ed0faeaa07557241439243287a184d1fd27f875c8b9ca6bd96")
|
||||
|
||||
new_smm_versions = "@2.9.2-preview,2.9.2:,develop"
|
||||
|
||||
variant("tests", default=False, description="Build DBCSR unit tests")
|
||||
variant("tests", default=True, description="Build DBCSR unit tests", when="@2.1:2.2")
|
||||
variant("mpi", default=True, description="Compile with MPI")
|
||||
variant("openmp", default=True, description="Build with OpenMP support")
|
||||
variant("shared", default=True, description="Build shared library")
|
||||
variant(
|
||||
"smm",
|
||||
default="libxsmm",
|
||||
values=("libxsmm", "libxs", "blas"),
|
||||
description="Library for small matrix multiplications",
|
||||
)
|
||||
variant(
|
||||
"cuda_arch_35_k20x",
|
||||
default=False,
|
||||
description=(
|
||||
"CP2K (resp. DBCSR) has specific parameter sets for"
|
||||
" different GPU models. Enable this when building"
|
||||
" with cuda_arch=35 for a K20x instead of a K40"
|
||||
),
|
||||
)
|
||||
variant("examples", default=True, description="Build examples")
|
||||
|
||||
variant("opencl", default=False, description="Enable OpenCL backend")
|
||||
variant("mpi_f08", default=False, when="@2.6:", description="Use mpi F08 module")
|
||||
|
||||
depends_on("c", type="build")
|
||||
depends_on("cxx", type="build")
|
||||
depends_on("fortran", type="build")
|
||||
|
||||
depends_on("blas")
|
||||
depends_on("lapack")
|
||||
depends_on("mpi", when="+mpi")
|
||||
|
||||
conflicts(
|
||||
"smm=libxs", when="@:2.9.1", msg="smm=libxs is only supported by DBCSR 2.9.2 or newer"
|
||||
)
|
||||
|
||||
depends_on("libxsmm@1.11:", when="@:2.9.1 smm=libxsmm")
|
||||
|
||||
with when(f"{new_smm_versions} smm=libxs"):
|
||||
depends_on("libxs")
|
||||
|
||||
with when(f"{new_smm_versions} smm=libxsmm"):
|
||||
depends_on("libxs")
|
||||
depends_on("libxsmm")
|
||||
|
||||
depends_on("cmake@3.10:", type="build")
|
||||
depends_on("cmake@3.12:", type="build", when="@2.1:")
|
||||
depends_on("cmake@3.17:", type="build", when="@2.2:")
|
||||
depends_on("cmake@3.22:", type="build", when="@2.3:")
|
||||
|
||||
depends_on("py-fypp", type="build")
|
||||
depends_on("py-fypp@3.1:", type="build", when="@2.6:")
|
||||
depends_on("pkgconfig", type="build")
|
||||
depends_on("python@3.6:", type="build", when="+cuda")
|
||||
|
||||
depends_on("hipblas", when="+rocm")
|
||||
|
||||
# Several packages provide "opencl" (incl. ICD/loader), e.g., "cuda"
|
||||
depends_on("opencl", when="+opencl")
|
||||
depends_on("libxstream", when="+opencl")
|
||||
opencl_loader_header_version = "2022.10.24"
|
||||
depends_on(f"opencl-c-headers@{opencl_loader_header_version}:", when="+opencl")
|
||||
requires(f"%opencl=opencl-icd-loader@{opencl_loader_header_version}:", when="+opencl")
|
||||
|
||||
# All examples require MPI
|
||||
conflicts("+examples", when="~mpi", msg="Examples require MPI")
|
||||
|
||||
# We only support specific gpu archs for which we have parameter files
|
||||
# for optimal kernels. Note that we don't override the parent class arch
|
||||
# properties, since the parent class defines constraints for different archs
|
||||
# Instead just mark all unsupported cuda archs as conflicting.
|
||||
dbcsr_cuda_archs = ("35", "37", "60", "70", "80", "90")
|
||||
cuda_msg = f"dbcsr only supports cuda_arch {dbcsr_cuda_archs}"
|
||||
|
||||
for arch in CudaPackage.cuda_arch_values:
|
||||
if arch not in dbcsr_cuda_archs:
|
||||
conflicts("+cuda", when=f"cuda_arch={arch}", msg=cuda_msg)
|
||||
|
||||
conflicts("+cuda", when="cuda_arch=none", msg=cuda_msg)
|
||||
|
||||
dbcsr_amdgpu_targets = (
|
||||
"gfx906",
|
||||
"gfx908",
|
||||
"gfx90a",
|
||||
"gfx90a:xnack-",
|
||||
"gfx90a:xnack+",
|
||||
"gfx942",
|
||||
"gfx950",
|
||||
)
|
||||
amd_msg = f"""DBCSR supports these AMD gpu targets: {", ".join(dbcsr_amdgpu_targets)}.
|
||||
Set amdgpu_target explicitly to one of the supported targets"""
|
||||
|
||||
for arch in ROCmPackage.amdgpu_targets:
|
||||
if arch not in dbcsr_amdgpu_targets:
|
||||
conflicts("+rocm", when=f"amdgpu_target={arch}", msg=amd_msg)
|
||||
|
||||
# GPU runtimes are usually mutually exclusive
|
||||
accel_msg = "CUDA and ROCm are mutually exlusive"
|
||||
conflicts("+cuda", when="+rocm", msg=accel_msg)
|
||||
conflicts("+cuda", when="+opencl", msg=accel_msg)
|
||||
conflicts("+rocm", when="+opencl", msg=accel_msg)
|
||||
|
||||
# Require OpenMP threading by making other options conflict
|
||||
conflicts("^intel-oneapi-mkl threads=none", when="+openmp")
|
||||
conflicts("^intel-oneapi-mkl threads=tbb", when="+openmp")
|
||||
conflicts("^openblas threads=pthreads", when="+openmp")
|
||||
conflicts("^openblas threads=none", when="+openmp")
|
||||
|
||||
# OpenCL follows the legacy LIBXSMM SMM path up to DBCSR 2.9.1.
|
||||
requires("smm=libxsmm", when="+opencl @:2.9.1")
|
||||
|
||||
# Starting with the 2.9.2 interface, LIBXS is an explicit SMM backend.
|
||||
requires("smm=libxs", when=f"+opencl {new_smm_versions}")
|
||||
|
||||
with when("+mpi"):
|
||||
# When using mpich 4.1 or higher, mpi_f08 has to be used, otherwise:
|
||||
# Error: Type mismatch in argument 'baseptr' at (1); passed TYPE(c_ptr)
|
||||
# to INTEGER(8)
|
||||
conflicts("^mpich@4.1:", when="@:2.5")
|
||||
conflicts("~mpi_f08", when="^mpich@4.1:")
|
||||
depends_on("mpich+fortran", when="^[virtuals=mpi] mpich")
|
||||
|
||||
generator("ninja")
|
||||
depends_on("ninja@1.10:", type="build")
|
||||
|
||||
@when("+rocm")
|
||||
def patch(self):
|
||||
for directory, subdirectory, files in os.walk(os.getcwd()):
|
||||
for i in files:
|
||||
file_path = os.path.join(directory, i)
|
||||
filter_file("USE ISO_C_BINDING", "USE,INTRINSIC :: ISO_C_BINDING", file_path)
|
||||
filter_file("USE ISO_FORTRAN_ENV", "USE,INTRINSIC :: ISO_FORTRAN_ENV", file_path)
|
||||
filter_file("USE omp_lib", "USE,INTRINSIC :: omp_lib", file_path)
|
||||
filter_file("USE OMP_LIB", "USE,INTRINSIC :: OMP_LIB", file_path)
|
||||
filter_file("USE iso_c_binding", "USE,INTRINSIC :: iso_c_binding", file_path)
|
||||
filter_file("USE iso_fortran_env", "USE,INTRINSIC :: iso_fortran_env", file_path)
|
||||
|
||||
def cmake_args(self):
|
||||
spec = self.spec
|
||||
|
||||
if "+cuda" in spec and len(spec.variants["cuda_arch"].value) > 1:
|
||||
raise InstallError("dbcsr supports only one cuda_arch at a time")
|
||||
|
||||
if "+rocm" in spec and len(spec.variants["amdgpu_target"].value) > 1:
|
||||
raise InstallError("DBCSR supports only one amdgpu_arch at a time")
|
||||
|
||||
smm = spec.variants["smm"].value
|
||||
has_libxs_smm = (
|
||||
spec.satisfies("@2.9.2-preview")
|
||||
or spec.satisfies("@2.9.2:")
|
||||
or spec.satisfies("@develop")
|
||||
)
|
||||
|
||||
args = [
|
||||
self.define_from_variant("USE_MPI", "mpi"),
|
||||
self.define_from_variant("USE_OPENMP", "openmp"),
|
||||
# C API needs MPI
|
||||
self.define_from_variant("WITH_C_API", "mpi"),
|
||||
self.define_from_variant("BUILD_SHARED_LIBS", "shared"),
|
||||
self.define_from_variant("WITH_EXAMPLES", "examples"),
|
||||
self.define_from_variant("BUILD_TESTING", "tests"),
|
||||
]
|
||||
|
||||
if has_libxs_smm:
|
||||
args += [
|
||||
self.define("USE_LIBXS", smm in ("libxs", "libxsmm")),
|
||||
self.define("USE_LIBXSMM", smm == "libxsmm"),
|
||||
]
|
||||
else:
|
||||
args.append("-DUSE_SMM=%s" % ("libxsmm" if smm == "libxsmm" else "blas"))
|
||||
|
||||
lapack, blas = spec["lapack"], spec["blas"]
|
||||
if blas.name != "intel-oneapi-mkl":
|
||||
args += [
|
||||
"-DBLAS_FOUND=true",
|
||||
"-DBLAS_LIBRARIES=%s" % (blas.libs.joined(";")),
|
||||
"-DLAPACK_FOUND=true",
|
||||
"-DLAPACK_LIBRARIES=%s" % (lapack.libs.joined(";")),
|
||||
]
|
||||
|
||||
if self.spec.satisfies("+cuda"):
|
||||
cuda_arch = self.spec.variants["cuda_arch"].value[0]
|
||||
|
||||
gpu_map = {
|
||||
"35": "K40",
|
||||
"37": "K80",
|
||||
"60": "P100",
|
||||
"70": "V100",
|
||||
"80": "A100",
|
||||
"90": "H100",
|
||||
}
|
||||
|
||||
gpuver = gpu_map[cuda_arch]
|
||||
if cuda_arch == "35" and self.spec.satisfies("+cuda_arch_35_k20x"):
|
||||
gpuver = "K20X"
|
||||
|
||||
args += ["-DWITH_GPU=%s" % gpuver, "-DUSE_ACCEL=cuda"]
|
||||
|
||||
if self.spec.satisfies("+rocm"):
|
||||
amd_arch = self.spec.variants["amdgpu_target"].value[0]
|
||||
gpuver = {
|
||||
"gfx906": "Mi50",
|
||||
"gfx908": "Mi100",
|
||||
"gfx90a": "Mi250",
|
||||
"gfx90a:xnack-": "Mi250",
|
||||
"gfx90a:xnack+": "Mi250",
|
||||
"gfx942": "Mi300",
|
||||
"gfx950": "Mi350",
|
||||
}[amd_arch]
|
||||
|
||||
args += [f"-DWITH_GPU={gpuver}", "-DUSE_ACCEL=hip"]
|
||||
|
||||
if self.spec.satisfies("+opencl"):
|
||||
args += ["-DUSE_ACCEL=opencl"]
|
||||
|
||||
if self.spec.satisfies("+mpi_f08"):
|
||||
args += ["-DUSE_MPI_F08=ON"]
|
||||
|
||||
return args
|
||||
|
||||
def check(self):
|
||||
"""Override CMakePackage's check() to enforce seralized test runs
|
||||
since they are already parallelized"""
|
||||
with working_dir(self.build_directory):
|
||||
self._if_ninja_target_execute("test", parallel=False)
|
||||
Loading…
Add table
Add a link
Reference in a new issue