Files
TheRock/rocm-systems/projects/rccl/bindings/nccl4py/setup.py
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

109 lines
3.2 KiB
Python

import os
import re
import subprocess
from pathlib import Path
from Cython.Build import cythonize
from setuptools import setup, Extension
# No CUDA_HOME requirement: after patching cynccl.pxd to drop the CUDA
# driver-types include block, no CUDA/HIP headers are needed at compile
# time — cudaStream_t and friends are opaque void* aliases. librccl.so
# is resolved at runtime via dlopen.
ext_modules = ["nccl.bindings.nccl"]
def calculate_modules(module: str):
module_parts = module.split(".")
# nccl.bindings.nccl -> nccl/bindings/nccl.pyx
lowpp_mod = module_parts.copy()
lowpp_pyx = os.path.join(*lowpp_mod[:-1], f"{lowpp_mod[-1]}.pyx")
lowpp_mod = ".".join(lowpp_mod)
lowpp_ext = Extension(
lowpp_mod,
sources=[lowpp_pyx],
language="c++",
extra_compile_args=["-std=c++14"],
libraries=["dl"],
)
# cy variant: nccl.bindings.nccl -> nccl/bindings/cynccl.pyx
cy_mod = module_parts.copy()
cy_mod[-1] = f"cy{cy_mod[-1]}"
cy_mod_pyx = os.path.join(*cy_mod[:-1], f"{cy_mod[-1]}.pyx")
cy_mod = ".".join(cy_mod)
cy_ext = Extension(
cy_mod,
sources=[cy_mod_pyx],
language="c++",
extra_compile_args=["-std=c++14"],
libraries=["dl"],
)
# internal variant: source is nccl_linux.pyx, but published module name is nccl.bindings._internal.nccl
inter_mod = module_parts.copy()
inter_mod.insert(-1, "_internal")
inter_mod_pyx = os.path.join(*inter_mod[:-1], f"{inter_mod[-1]}_linux.pyx")
inter_mod = ".".join(inter_mod)
inter_ext = Extension(
inter_mod,
sources=[inter_mod_pyx],
language="c++",
extra_compile_args=["-std=c++14"],
libraries=["dl"],
)
# internal variant: insert _internal and use utils.pyx
inter_utils_mod = module_parts.copy()
inter_utils_mod.insert(-1, "_internal")
inter_utils_mod[-1] = "utils"
inter_utils_mod_pyx = os.path.join(*inter_utils_mod[:-1], f"{inter_utils_mod[-1]}.pyx")
inter_utils_mod = ".".join(inter_utils_mod)
inter_utils_ext = Extension(
inter_utils_mod,
sources=[inter_utils_mod_pyx],
language="c++",
extra_compile_args=["-std=c++14"],
libraries=["dl"],
)
return lowpp_ext, cy_ext, inter_ext, inter_utils_ext
# Note: the extension attributes are overwritten in build_extension()
ext_modules = [e for ext in ext_modules for e in calculate_modules(ext)]
# Hand-written companion module exposing RCCL-only collectives
# (ncclAllReduceWithBias, ncclAllToAllv) under the same snake_case
# style as the autogenerated upstream wrappers.
ext_modules.append(
Extension(
"nccl.bindings.rocm_extensions",
sources=["nccl/bindings/rocm_extensions.pyx"],
language="c++",
extra_compile_args=["-std=c++14"],
libraries=["dl"],
)
)
compiler_directives = {
"embedsignature": True,
"show_performance_hints": True,
"freethreading_compatible": True,
}
setup(
ext_modules=cythonize(
ext_modules,
verbose=True,
language_level=3,
compiler_directives=compiler_directives,
),
zip_safe=False,
options={"build_ext": {"inplace": False}},
)