Horovod 0.26.1 for PyTorch 1.11.0

easybuild/easyconfigs/h/Horovod/Horovod-0.26.1-foss-2021b-CUDA-11.4.1-PyTorch-1.11.0.eb

@@ -0,0 +1,48 @@
+easyblock = 'PythonBundle'
+
+name = 'Horovod'
+version = '0.26.1'
+local_pt_version = '1.11.0'
+local_cuda_suffix = '-CUDA-%(cudaver)s'
+versionsuffix = local_cuda_suffix + '-PyTorch-%s' % local_pt_version
+
+homepage = 'https://github.com/uber/horovod'
+description = "Horovod is a distributed training framework for TensorFlow."
+
+toolchain = {'name': 'foss', 'version': '2021b'}
+
+builddependencies = [
+    ('CMake', '3.22.1'),
+    ('flatbuffers', '2.0.0'),
+]
+dependencies = [
+    ('Python', '3.9.6'),
+    ('PyYAML', '5.4.1'),
+    ('CUDA', '11.4.1', '', SYSTEM),
+    ('NCCL', '2.10.3', local_cuda_suffix),
+    ('PyTorch', local_pt_version, local_cuda_suffix),
+]
+
+use_pip = True
+sanity_pip_check = True
+
+preinstallopts = 'HOROVOD_WITH_MPI=1 HOROVOD_GPU_ALLREDUCE=NCCL HOROVOD_GPU_BROADCAST=NCCL '
+preinstallopts += 'HOROVOD_WITHOUT_TENSORFLOW=1 HOROVOD_WITH_PYTORCH=1 HOROVOD_WITHOUT_MXNET=1 '
+
+parallel = 1  # Bug in CMake causes a race condition on horovod_cuda_kernels_generated_cuda_kernels.cu.o.NVCC-depend
+
+exts_list = [
+    ('cloudpickle', '2.2.0', {
+        'checksums': ['3f4219469c55453cfe4737e564b67c2a149109dabf7f242478948b895f61106f'],
+    }),
+    ('horovod', version, {
+        'checksums': ['0079e4efb4ffe3f042e9e00470fbb457973903e748f75909de71e492d2474e94'],
+    }),
+]
+
+sanity_check_paths = {
+    'files': ['bin/horovodrun'],
+    'dirs': ['lib/python%(pyshortver)s/site-packages'],
+}
+
+moduleclass = 'tools'

easybuild/easyconfigs/p/Pillow/Pillow-8.2.0-GCCcore-11.2.0.eb

@@ -0,0 +1,36 @@
+easyblock = 'PythonPackage'
+
+name = 'Pillow'
+version = '8.2.0'
+
+homepage = 'https://pillow.readthedocs.org/'
+description = """Pillow is the 'friendly PIL fork' by Alex Clark and Contributors.
+ PIL is the Python Imaging Library by Fredrik Lundh and Contributors."""
+
+toolchain = {'name': 'GCCcore', 'version': '11.2.0'}
+
+sources = [SOURCE_TAR_GZ]
+patches = ['Pillow-%(version_major)s_CVE-2021-23437.patch']
+checksums = [
+    'a787ab10d7bb5494e5f76536ac460741788f1fbce851068d73a87ca7c35fc3e1',  # Pillow-8.2.0.tar.gz
+    '12bac0ff4f815d1e60ed129106c2990c8c6157891fd90ec8aaec88a5a642d521',  # Pillow-8_CVE-2021-23437.patch
+]
+
+builddependencies = [('binutils', '2.37')]
+
+dependencies = [
+    ('Python', '3.9.6'),
+    ('libjpeg-turbo', '2.0.6'),
+    ('libpng', '1.6.37'),
+    ('zlib', '1.2.11'),
+    ('LibTIFF', '4.3.0'),
+    ('freetype', '2.11.0')
+]
+
+download_dep_fail = True
+use_pip = True
+sanity_pip_check = True
+
+options = {'modulename': 'PIL'}
+
+moduleclass = 'vis'

easybuild/easyconfigs/p/PyTorch/PyTorch-1.11.0-foss-2021b-CUDA-11.4.1.eb

@@ -0,0 +1,126 @@
+name = 'PyTorch'
+version = '1.11.0'
+versionsuffix = '-CUDA-%(cudaver)s'
+
+homepage = 'https://pytorch.org/'
+description = """Tensors and Dynamic neural networks in Python with strong GPU acceleration.
+PyTorch is a deep learning framework that puts Python first."""
+
+toolchain = {'name': 'foss', 'version': '2021b'}
+
+sources = [{
+    'filename': '%(name)s-%(version)s.tar.gz',
+    'git_config': {
+        'url': 'https://github.com/pytorch',
+        'repo_name': 'pytorch',
+        'tag': 'v%(version)s',
+        'recursive': True,
+    },
+}]
+patches = [
+    'PyTorch-1.7.0_avoid-nan-in-test-torch.patch',
+    'PyTorch-1.7.0_disable-dev-shm-test.patch',
+    'PyTorch-1.8.1_dont-use-gpu-ccc-in-test.patch',
+    'PyTorch-1.9.0_limit-world-size-for-zero-redundancy-opt-test.patch',
+    'PyTorch-1.10.0_fix-test-dataloader-fixed-affinity.patch',
+    'PyTorch-1.10.0_skip_cmake_rpath.patch',
+    'PyTorch-1.11.0_increase-distributed-test-timeout.patch',
+    'PyTorch-1.11.0_skip_failing_ops_tests.patch',
+    'PyTorch-1.11.0_fix_skip_jit_cuda_fuser.patch',
+    'PyTorch-1.11.0_fix_sharded_imports.patch',
+    'PyTorch-1.11.0_increase_test_tolerances_TF32.patch',
+    'PyTorch-1.11.0_increase_c10d_gloo_timeout.patch',
+    'PyTorch-1.11.0_disable_failing_jit_cuda_fuser_tests.patch',
+]
+checksums = [
+    None,  # can't add proper SHA256 checksum, because source tarball is created locally after recursive 'git clone'
+    'b899aa94d9e60f11ee75a706563312ccefa9cf432756c470caa8e623991c8f18',  # PyTorch-1.7.0_avoid-nan-in-test-torch.patch
+    '622cb1eaeadc06e13128a862d9946bcc1f1edd3d02b259c56a9aecc4d5406b8a',  # PyTorch-1.7.0_disable-dev-shm-test.patch
+    '89ac7a8e9e7df2e64cf8404fe3a279f5e9b759fee41c9de3aaff9c22f385c2c6',  # PyTorch-1.8.1_dont-use-gpu-ccc-in-test.patch
+    # PyTorch-1.9.0_limit-world-size-for-zero-redundancy-opt-test.patch
+    'ff573660913ce055e24cfd194ce747ba5685091c631cfd443eae2a99d56b57ea',
+    # PyTorch-1.10.0_fix-test-dataloader-fixed-affinity.patch
+    '313dca681f45ce3bc7c4557fdcdcbe0b77216d2c708fa30a2ec0e22c44876707',
+    'ac05943bb205623f91ef140aa00869efc5fe844184bd666bebf5405808610448',  # PyTorch-1.10.0_skip_cmake_rpath.patch
+    # PyTorch-1.11.0_increase-distributed-test-timeout.patch
+    '087ad20163a1291773ae3457569b80523080eb3731e210946459b2333a919f3f',
+    '8eaca92d64fcadb0552d28e9c7ea5c4bc669d2fe33004e45a3519ce8d0d136a2',  # PyTorch-1.11.0_skip_failing_ops_tests.patch
+    '21fc678febcdfbb9dabd72235be23cd392044e9a954f6580d15b530e1f69dcc1',  # PyTorch-1.11.0_fix_skip_jit_cuda_fuser.patch
+    '9a04f4285b800dad8a00c3014af0a9713d40d5dd35d10931c7c0da4e89c558e9',  # PyTorch-1.11.0_fix_sharded_imports.patch
+    # PyTorch-1.11.0_increase_test_tolerances_TF32.patch
+    '26e179a4f6f57e49209092612ae5f5cd8c03fd2ca84566ba0244eabefc3736ba',
+    # PyTorch-1.11.0_increase_c10d_gloo_timeout.patch
+    '20cd4a8663f74ab326fdb032b926bf5c7e94d9750c515ab9050927ba00cf1953',
+    # PyTorch-1.11.0_disable_failing_jit_cuda_fuser_tests.patch
+    'e7bfe120a8b3fe2b40dac6839852a5fbab3cb3429fbe44a0fc3a1800adaaee51',
+]
+
+osdependencies = [OS_PKG_IBVERBS_DEV]
+
+builddependencies = [
+    ('CMake', '3.22.1'),
+    ('hypothesis', '6.14.6'),
+]
+
+dependencies = [
+    ('CUDA', '11.4.1', '', SYSTEM),
+    ('Ninja', '1.10.2'),  # Required for JIT compilation of C++ extensions
+    ('Python', '3.9.6'),
+    ('protobuf', '3.17.3'),
+    ('protobuf-python', '3.17.3'),
+    ('pybind11', '2.7.1'),
+    ('SciPy-bundle', '2021.10'),
+    ('typing-extensions', '4.3.0'),
+    ('PyYAML', '5.4.1'),
+    ('MPFR', '4.1.0'),
+    ('GMP', '6.2.1'),
+    ('numactl', '2.0.14'),
+    ('FFmpeg', '4.3.2'),
+    ('Pillow', '8.2.0'),
+    ('cuDNN', '8.2.2.26', '-CUDA-%(cudaver)s', SYSTEM),
+    ('magma', '2.6.2', '-CUDA-%(cudaver)s'),
+    ('NCCL', '2.10.3', '-CUDA-%(cudaver)s'),
+    ('expecttest', '0.1.3'),
+]
+
+# default CUDA compute capabilities to use (override via --cuda-compute-capabilities)
+cuda_compute_capabilities = ['3.5', '3.7', '5.2', '6.0', '6.1', '7.0', '7.2', '7.5', '8.0', '8.6']
+
+custom_opts = ["USE_CUPTI_SO=1"]
+
+excluded_tests = {
+    '': [
+        # Bad tests: https://github.com/pytorch/pytorch/issues/60260
+        'distributed/elastic/utils/distributed_test',
+        'distributed/elastic/multiprocessing/api_test',
+        # These tests fail on A10s at the very least, they time out forever no matter how long the timeout is.
+        # Possibly related to NCCL 2.8.3: https://docs.nvidia.com/deeplearning/nccl/release-notes/rel_2-8-3.html
+        # 'distributed/test_distributed_fork',
+        'distributed/test_distributed_spawn',
+        # Fails on A10s: https://github.com/pytorch/pytorch/issues/63079
+        'test_optim',
+        # Test from this suite timeout often. The process group backend is deprecated anyway
+        # 'distributed/rpc/test_process_group_agent',
+        # This test fails constently when run as part of the test suite, but succeeds when run interactively
+        'test_model_dump',
+        # These tests appear flaky, possibly related to number of GPUs that are used
+        'distributed/fsdp/test_fsdp_memory',
+        'distributed/fsdp/test_fsdp_overlap',
+    ]
+}
+
+runtest = 'cd test && PYTHONUNBUFFERED=1 %(python)s run_test.py --continue-through-error  --verbose %(excluded_tests)s'
+
+# several tests are known to be flaky, and fail in some contexts (like having multiple GPUs available),
+# so we allow up to 10 (out of ~90k) tests to fail before treating the installation to be faulty
+max_failed_tests = 10
+
+# The readelf sanity check command can be taken out once the TestRPATH test from 
+# https://github.com/pytorch/pytorch/pull/68912 is accepted, since it is then checked as part of the PyTorch test suite
+local_libcaffe2 = "$EBROOTPYTORCH/lib/python%%(pyshortver)s/site-packages/torch/lib/libcaffe2_nvrtc.%s" % SHLIB_EXT
+sanity_check_commands = [
+    "readelf -d %s | egrep 'RPATH|RUNPATH' | grep -v stubs" % local_libcaffe2,
+]
+tests = ['PyTorch-check-cpp-extension.py']
+
+moduleclass = 'devel'

easybuild/easyconfigs/t/typing-extensions/typing-extensions-4.3.0-GCCcore-11.2.0.eb

@@ -0,0 +1,22 @@
+easyblock = 'PythonPackage'
+
+name = 'typing-extensions'
+version = '4.3.0'
+
+homepage = 'https://github.com/python/typing/blob/master/typing_extensions/README.rst'
+description = 'Typing Extensions – Backported and Experimental Type Hints for Python'
+
+toolchain = {'name': 'GCCcore', 'version': '11.2.0'}
+
+sources = ['typing_extensions-%(version)s.tar.gz']
+checksums = ['e6d2677a32f47fc7eb2795db1dd15c1f34eff616bcaf2cfb5e997f854fa1c4a6']
+
+builddependencies = [('binutils', '2.37')]
+
+dependencies = [('Python', '3.9.6')]
+
+use_pip = True
+download_dep_fail = True
+sanity_pip_check = True
+
+moduleclass = 'devel'