""" This file must not depend on any other CuPy modules. """ import ctypes import importlib.metadata import json import os import os.path import platform import re import shutil import sys from typing import Any, Dict, List, Optional, Tuple import warnings # '' for uninitialized, None for non-existing _cuda_path = '' _nvcc_path = '' _rocm_path = '' _hipcc_path = '' _cub_path = '' """ Library Preloading ------------------ Wheel packages are built against specific versions of CUDA libraries (cuTENSOR/NCCL/cuDNN). To avoid loading wrong version, these shared libraries are manually preloaded. # TODO(kmaehashi): Support NCCL Example of `_preload_config` is as follows: { # installation source 'packaging': 'pip', # CUDA version string 'cuda': '11.0', 'cudnn': { # cuDNN version string 'version': '8.0.0', # names of the shared library 'filenames': ['libcudnn.so.X.Y.Z'] # or `cudnn64_X.dll` for Windows } } The configuration file is intended solely for internal purposes and not expected to be parsed by end-users. """ _preload_config = None _preload_libs = { 'cudnn': None, 'nccl': None, 'cutensor': None, } _debug = os.environ.get('CUPY_DEBUG_LIBRARY_LOAD', '0') == '1' def _log(msg: str) -> None: if _debug: sys.stderr.write(f'[CUPY_DEBUG_LIBRARY_LOAD] {msg}\n') sys.stderr.flush() def get_cuda_path(): # Returns the CUDA installation path or None if not found. global _cuda_path if _cuda_path == '': _cuda_path = _get_cuda_path() return _cuda_path def get_nvcc_path(): # Returns the path to the nvcc command or None if not found. global _nvcc_path if _nvcc_path == '': _nvcc_path = _get_nvcc_path() return _nvcc_path def get_rocm_path(): # Returns the ROCm installation path or None if not found. global _rocm_path if _rocm_path == '': _rocm_path = _get_rocm_path() return _rocm_path def get_hipcc_path(): # Returns the path to the hipcc command or None if not found. global _hipcc_path if _hipcc_path == '': _hipcc_path = _get_hipcc_path() return _hipcc_path def get_cub_path(): # Returns the CUB header path or None if not found. global _cub_path if _cub_path == '': _cub_path = _get_cub_path() return _cub_path def _get_cuda_path(): # Use environment variable cuda_path = os.environ.get('CUDA_PATH', '') # Nvidia default on Windows if os.path.exists(cuda_path): return cuda_path # Use nvcc path nvcc_path = shutil.which('nvcc') if nvcc_path is not None: return os.path.dirname(os.path.dirname(nvcc_path)) # Use typical path if os.path.exists('/usr/local/cuda'): return '/usr/local/cuda' return None def _get_nvcc_path(): # Honor the "NVCC" env var nvcc_path = os.environ.get('NVCC', None) if nvcc_path is not None: return nvcc_path # Lookup /bin cuda_path = get_cuda_path() if cuda_path is None: return None return shutil.which('nvcc', path=os.path.join(cuda_path, 'bin')) def _get_rocm_path(): # Use environment variable rocm_path = os.environ.get('ROCM_HOME', '') if os.path.exists(rocm_path): return rocm_path # Use hipcc path hipcc_path = shutil.which('hipcc') if hipcc_path is not None: return os.path.dirname(os.path.dirname(hipcc_path)) # Use typical path if os.path.exists('/opt/rocm'): return '/opt/rocm' return None def _get_hipcc_path(): # TODO(leofang): Introduce an env var HIPCC? # Lookup /bin rocm_path = get_rocm_path() if rocm_path is None: return None return shutil.which('hipcc', path=os.path.join(rocm_path, 'bin')) def _get_cub_path(): # runtime discovery of CUB headers from cupy_backends.cuda.api import runtime current_dir = os.path.dirname(os.path.abspath(__file__)) if not runtime.is_hip: if os.path.isdir( os.path.join(current_dir, '_core/include/cupy/_cccl/cub')): _cub_path = '' else: _cub_path = None else: # the bundled CUB does not work in ROCm rocm_path = get_rocm_path() if rocm_path is not None and os.path.isdir( os.path.join(rocm_path, 'include/hipcub')): # use hipCUB _cub_path = '' else: _cub_path = None return _cub_path def _setup_win32_dll_directory(): # Setup DLL directory to load CUDA Toolkit libs and shared libraries # added during the build process. if sys.platform.startswith('win32'): # see _can_attempt_preload() config = get_preload_config() is_conda = (config is not None and (config['packaging'] == 'conda')) # Path to the CUDA Toolkit binaries cuda_path = get_cuda_path() if cuda_path is not None: if is_conda: cuda_bin_path = cuda_path else: cuda_bin_path = os.path.join(cuda_path, 'bin') else: cuda_bin_path = None if not is_conda: warnings.warn( 'CUDA path could not be detected.' ' Set CUDA_PATH environment variable if CuPy ' 'fails to load.') _log('CUDA_PATH: {}'.format(cuda_path)) # Path to shared libraries in wheel wheel_libdir = os.path.join( get_cupy_install_path(), 'cupy', '.data', 'lib') if os.path.isdir(wheel_libdir): _log('Wheel shared libraries: {}'.format(wheel_libdir)) else: _log('Not wheel distribution ({} not found)'.format( wheel_libdir)) wheel_libdir = None if (3, 8) <= sys.version_info: if cuda_bin_path is not None: _log('Adding DLL search path: {}'.format(cuda_bin_path)) os.add_dll_directory(cuda_bin_path) if wheel_libdir is not None: _log('Adding DLL search path: {}'.format(wheel_libdir)) os.add_dll_directory(wheel_libdir) else: # Users are responsible for adding `%CUDA_PATH%/bin` to PATH. if wheel_libdir is not None: _log('Adding to PATH: {}'.format(wheel_libdir)) path = os.environ.get('PATH', '') os.environ['PATH'] = wheel_libdir + os.pathsep + path def get_cupy_install_path(): # Path to the directory where the package is installed. return os.path.abspath( os.path.join(os.path.dirname(__file__), '..')) def get_cupy_cuda_lib_path(): """Returns the directory where CUDA external libraries are installed. This environment variable only affects wheel installations. Shared libraries are looked up from `$CUPY_CUDA_LIB_PATH/$CUDA_VER/$LIB_NAME/$LIB_VER/{lib,lib64,bin}`, e.g., `~/.cupy/cuda_lib/11.2/cudnn/8.1.1/lib64/libcudnn.so.8.1.1`. The default $CUPY_CUDA_LIB_PATH is `~/.cupy/cuda_lib`. """ cupy_cuda_lib_path = os.environ.get('CUPY_CUDA_LIB_PATH', None) if cupy_cuda_lib_path is None: return os.path.expanduser('~/.cupy/cuda_lib') return os.path.abspath(cupy_cuda_lib_path) def get_preload_config() -> Optional[Dict[str, Any]]: global _preload_config if _preload_config is None: _preload_config = _get_json_data('_wheel.json') return _preload_config def _get_json_data(name: str) -> Optional[Dict[str, Any]]: config_path = os.path.join( get_cupy_install_path(), 'cupy', '.data', name) if not os.path.exists(config_path): return None with open(config_path) as f: return json.load(f) def _can_attempt_preload(lib: str) -> bool: """Returns if the preload can be attempted.""" config = get_preload_config() if (config is None) or (config['packaging'] == 'conda'): # We don't do preload if CuPy is installed from Conda-Forge, as we # cannot guarantee the version pinned in _wheel.json, which is # encoded in config[lib]['filenames'], is always available on # Conda-Forge. See here for the configuration files used in # Conda-Forge distributions. # https://github.com/conda-forge/cupy-feedstock/blob/master/recipe/preload_config/ _log(f'Not preloading {lib} as this is not a pip wheel installation') return False if lib not in _preload_libs: raise AssertionError(f'Unknown preload library: {lib}') if lib not in config: _log(f'Preload {lib} not configured in wheel') return False if _preload_libs[lib] is not None: _log(f'Preload already attempted: {lib}') return False return True def _preload_library(lib): """Preload dependent shared libraries. The preload configuration file (cupy/.data/_wheel.json) will be added during the wheel build process. """ _log(f'Preloading triggered for library: {lib}') if not _can_attempt_preload(lib): return _preload_libs[lib] = {} config = get_preload_config() cuda_version = config['cuda'] _log('CuPy wheel package built for CUDA {}'.format(cuda_version)) cupy_cuda_lib_path = get_cupy_cuda_lib_path() _log('CuPy CUDA library directory: {}'.format(cupy_cuda_lib_path)) version = config[lib]['version'] filenames = config[lib]['filenames'] for filename in filenames: _log(f'Looking for {lib} version {version} ({filename})') # "lib": cuTENSOR (Linux/Windows) / NCCL (Linux) # "lib64": cuDNN (Linux) # "bin": cuDNN (Windows) libpath_cands = [ os.path.join( cupy_cuda_lib_path, config['cuda'], lib, version, x, filename) for x in ['lib', 'lib64', 'bin']] if lib == 'cutensor': libpath_cands = ( _get_cutensor_from_wheel(version, config['cuda']) + libpath_cands) for libpath in libpath_cands: if not os.path.exists(libpath): _log('Rejected candidate (not found): {}'.format(libpath)) continue try: _log(f'Trying to load {libpath}') # Keep reference to the preloaded module. _preload_libs[lib][libpath] = ctypes.CDLL(libpath) _log('Loaded') break except Exception as e: e_type = type(e).__name__ # NOQA msg = ( f'CuPy failed to preload library ({libpath}): ' f'{e_type} ({e})') _log(msg) warnings.warn(msg) else: _log('File {} could not be found'.format(filename)) # Lookup library with fully-qualified version (e.g., # `libcudnn.so.X.Y.Z`). _log(f'Trying to load {filename} from default search path') try: _preload_libs[lib][filename] = ctypes.CDLL(filename) _log('Loaded') except Exception as e: # Fallback to the standard shared library lookup which only # uses the major version (e.g., `libcudnn.so.X`). _log(f'Library {lib} could not be preloaded: {e}') def _parse_version(version: str) -> Tuple[int, int, int]: parts = re.split(r'[^\d]', version, maxsplit=3) major = int(parts[0]) minor = int(parts[1]) if len(parts) >= 2 else 0 patch = int(parts[2]) if len(parts) >= 3 else 0 return major, minor, patch def _get_cutensor_from_wheel(version: str, cuda: str) -> List[str]: """ Returns the list of shared library path candidates for cuTENSOR installed via Pip (cutensor-cuXX package). """ cuda_major_ver, _ = cuda.split('.') cutensor_pkg = f'cutensor-cu{cuda_major_ver}' try: cutensor_dist = importlib.metadata.distribution(cutensor_pkg) except importlib.metadata.PackageNotFoundError: _log(f'cuTENSOR wheel package not installed: {cutensor_pkg}') return [] actual = _parse_version(cutensor_dist.version) expected = _parse_version(version) is_compatible = ( actual[0] == expected[0] and actual[1] >= expected[1] and actual[2] >= expected[2] ) if not is_compatible: _log('cuTENSOR wheel incompatible: ' f'expected {version}, found {cutensor_dist.version}') return [] if sys.platform == 'linux': shared_lib = cutensor_dist.locate_file( f'cutensor/lib/libcutensor.so.{version.split(".")[0]}' ) else: shared_lib = cutensor_dist.locate_file('cutensor\\bin\\cutensor.dll') return [str(shared_lib)] def _preload_warning(lib, exc): config = get_preload_config() if config is None or lib not in config: return if config['packaging'] == 'pip': cuda = config['cuda'] if lib == 'cutensor': cuda_major = cuda.split('.')[0] version = config['cutensor']['version'] cmd = f'pip install "cutensor-cu{cuda_major}~={version}"' else: cmd = f'python -m cupyx.tools.install_library --library {lib} --cuda {cuda}' # NOQA elif config['packaging'] == 'conda': cmd = f'conda install -c conda-forge {lib}' else: raise AssertionError warnings.warn(f''' {lib} library could not be loaded. Reason: {type(exc).__name__} ({str(exc)}) You can install the library by: $ {cmd} ''') def _get_include_dir_from_conda_or_wheel(major: int, minor: int) -> List[str]: # FP16 headers from CUDA 12.2+ depends on headers from CUDA Runtime. # See https://github.com/cupy/cupy/issues/8466. if major < 12 or (major == 12 and minor < 2): return [] config = get_preload_config() if config is not None and config['packaging'] == 'conda': if sys.platform.startswith('linux'): arch = platform.machine() if arch == "aarch64": arch = "sbsa" assert arch, "platform.machine() returned an empty string" target_dir = f"{arch}-linux" return [ os.path.join(sys.prefix, "targets", target_dir, "include"), os.path.join(sys.prefix, "include"), ] elif sys.platform.startswith('win'): return [ os.path.join(sys.prefix, "Library", "include"), ] else: # No idea what this platform is. Do nothing? return [] # Look for headers in wheels pkg_name = f'nvidia-cuda-runtime-cu{major}' ver_str = f'{major}.{minor}' _log(f'Looking for {pkg_name}=={ver_str}.*') try: dist = importlib.metadata.distribution(pkg_name) except importlib.metadata.PackageNotFoundError: _log('The package could not be found') return [] if dist.version == ver_str or dist.version.startswith(f'{ver_str}.'): include_dir = dist.locate_file('nvidia/cuda_runtime/include') if not include_dir.exists(): _log('The include directory could not be found') return [] return [str(include_dir)] else: _log(f'Found incompatible version ({dist.version})') return [] def _detect_duplicate_installation(): # List of all CuPy packages, including out-dated ones. known = { 'cupy', 'cupy-cuda80', 'cupy-cuda90', 'cupy-cuda91', 'cupy-cuda92', 'cupy-cuda100', 'cupy-cuda101', 'cupy-cuda102', 'cupy-cuda110', 'cupy-cuda111', 'cupy-cuda112', 'cupy-cuda113', 'cupy-cuda114', 'cupy-cuda115', 'cupy-cuda116', 'cupy-cuda117', 'cupy-cuda118', 'cupy-cuda11x', 'cupy-cuda12x', 'cupy-rocm-4-0', 'cupy-rocm-4-1', 'cupy-rocm-4-2', 'cupy-rocm-4-3', 'cupy-rocm-5-0', } # use metadata.get to be resilient to namespace packages # that may be leftover in the user's path??? # something else might be triggering "Name" not existing # But without a safe ".get" a KeyError might be raised # not allowing us to get through the setup # https://github.com/cupy/cupy/issues/8440 installed_names = {d.metadata.get("Name", None) for d in importlib.metadata.distributions()} cupy_installed = known & installed_names if 1 < len(cupy_installed): cupy_packages_list = ', '.join(sorted(cupy_installed)) warnings.warn(f''' -------------------------------------------------------------------------------- CuPy may not function correctly because multiple CuPy packages are installed in your environment: {cupy_packages_list} Follow these steps to resolve this issue: 1. For all packages listed above, run the following command to remove all existing CuPy installations: $ pip uninstall If you previously installed CuPy via conda, also run the following: $ conda uninstall cupy 2. Install the appropriate CuPy package. Refer to the Installation Guide for detailed instructions. https://docs.cupy.dev/en/stable/install.html -------------------------------------------------------------------------------- ''') def _diagnose_import_error() -> str: # TODO(kmaehashi): provide better diagnostics. msg = '''\ Failed to import CuPy. If you installed CuPy via wheels (cupy-cudaXXX or cupy-rocm-X-X), make sure that the package matches with the version of CUDA or ROCm installed. On Linux, you may need to set LD_LIBRARY_PATH environment variable depending on how you installed CUDA/ROCm. On Windows, try setting CUDA_PATH environment variable. Check the Installation Guide for details: https://docs.cupy.dev/en/latest/install.html''' # NOQA if sys.platform == 'win32': try: msg += _diagnose_win32_dll_load() except Exception as e: msg += ( '\n\nThe cause could not be identified: ' f'{type(e).__name__}: {e}' ) return msg def _diagnose_win32_dll_load() -> str: depends = _get_json_data('_depends.json') if depends is None: return '' from ctypes import wintypes kernel32 = ctypes.CDLL('kernel32') kernel32.GetModuleFileNameW.argtypes = [ wintypes.HANDLE, wintypes.LPWSTR, wintypes.DWORD] kernel32.GetModuleFileNameW.restype = wintypes.DWORD # Show dependents in similar form of ldd on Linux. lines = [ '', '', f'CUDA Path: {get_cuda_path()}', 'DLL dependencies:' ] filepath = ctypes.create_unicode_buffer(2**15) for name in depends['depends']: try: dll = ctypes.CDLL(name) kernel32.GetModuleFileNameW(dll._handle, filepath, len(filepath)) lines.append(f' {name} -> {filepath.value}') except FileNotFoundError: lines.append(f' {name} -> not found') except Exception as e: lines.append(f' {name} -> error ({type(e).__name__}: {e})') return '\n'.join(lines)