TransformerEngine icon indicating copy to clipboard operation
TransformerEngine copied to clipboard

Impossible to compile the package

Open mrinaldi97 opened this issue 1 month ago • 0 comments

Hello, I am trying to install transformer-engine using pip install --no-build-isolation transformer_engine[pytorch] on two different systems but I always get compilation errors:

Personal system (Nvidia 3090, torch 2.6):

Building wheels for collected packages: transformer_engine_torch
  Building wheel for transformer_engine_torch (pyproject.toml) ... error
  error: subprocess-exited-with-error
  
  × Building wheel for transformer_engine_torch (pyproject.toml) did not run successfully.
  │ exit code: 1
  ╰─> [226 lines of output]
      /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/dist.py:289: UserWarning: Unknown distribution option: 'tests_require'
        warnings.warn(msg)
      running bdist_wheel
      Guessing wheel URL:  https://github.com/NVIDIA/TransformerEngine/releases/download/v2.9.0/transformer_engine_torch-2.9.0+cu12torch2.6cxx11abiFALSE-cp313-cp313-linux_x86_64.whl
      Precompiled wheel not found. Building from source...
      running build
      running build_ext
      building 'transformer_engine_torch' extension
      creating /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc
      creating /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions
      creating /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/multi_tensor
      Emitting ninja build file /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/build.ninja...
      Compiling objects...
      Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
      [1/28] c++ -MMD -MF /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/common.o.d -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O3 -Wall -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -ffat-lto-objects -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -fPIC -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvcc/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cublas/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cudnn/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_cccl/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/nvtx/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvrtc/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/torch/csrc/api/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/TH -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/THC -I/home/matteo/envs/torch-cuda/include -I/usr/include/python3.13 -c -c /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/common.cpp -o /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/common.o -O3 -fvisibility=hidden -g0 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=transformer_engine_torch -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++17
      FAILED: /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/common.o
      c++ -MMD -MF /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/common.o.d -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O3 -Wall -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -ffat-lto-objects -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -fPIC -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvcc/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cublas/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cudnn/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_cccl/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/nvtx/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvrtc/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/torch/csrc/api/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/TH -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/THC -I/home/matteo/envs/torch-cuda/include -I/usr/include/python3.13 -c -c /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/common.cpp -o /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/common.o -O3 -fvisibility=hidden -g0 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=transformer_engine_torch -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++17
      In file included from /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/ATen/cuda/CUDAContextLight.h:6,
                       from /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/ATen/cuda/CUDAContext.h:3,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/common.h:12,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/common.cpp:7:
      /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include/cuda_runtime_api.h:148:10: fatal error: crt/host_defines.h: File o directory non esistente
        148 | #include "crt/host_defines.h"
            |          ^~~~~~~~~~~~~~~~~~~~
      compilation terminated.
      [2/28] c++ -MMD -MF /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/apply_rope.o.d -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O3 -Wall -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -ffat-lto-objects -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -fPIC -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvcc/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cublas/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cudnn/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_cccl/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/nvtx/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvrtc/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/torch/csrc/api/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/TH -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/THC -I/home/matteo/envs/torch-cuda/include -I/usr/include/python3.13 -c -c /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/apply_rope.cpp -o /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/apply_rope.o -O3 -fvisibility=hidden -g0 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=transformer_engine_torch -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++17
      FAILED: /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/apply_rope.o
      c++ -MMD -MF /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/apply_rope.o.d -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O3 -Wall -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -ffat-lto-objects -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -fPIC -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvcc/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cublas/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cudnn/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_cccl/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/nvtx/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvrtc/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/torch/csrc/api/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/TH -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/THC -I/home/matteo/envs/torch-cuda/include -I/usr/include/python3.13 -c -c /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/apply_rope.cpp -o /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/apply_rope.o -O3 -fvisibility=hidden -g0 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=transformer_engine_torch -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++17
      In file included from /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/ATen/cuda/CUDAContextLight.h:6,
                       from /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/ATen/cuda/CUDAContext.h:3,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/../common.h:12,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/../extensions.h:12,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/apply_rope.cpp:7:
      /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include/cuda_runtime_api.h:148:10: fatal error: crt/host_defines.h: File o directory non esistente
        148 | #include "crt/host_defines.h"
            |          ^~~~~~~~~~~~~~~~~~~~
      compilation terminated.
      [3/28] c++ -MMD -MF /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/activation.o.d -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O3 -Wall -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -ffat-lto-objects -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -fPIC -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvcc/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cublas/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cudnn/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_cccl/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/nvtx/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvrtc/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/torch/csrc/api/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/TH -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/THC -I/home/matteo/envs/torch-cuda/include -I/usr/include/python3.13 -c -c /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/activation.cpp -o /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/activation.o -O3 -fvisibility=hidden -g0 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=transformer_engine_torch -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++17
      FAILED: /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/activation.o
      c++ -MMD -MF /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/activation.o.d -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O3 -Wall -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -ffat-lto-objects -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -march=x86-64 -mtune=generic -O3 -pipe -fno-plt -fexceptions -Wp,-D_FORTIFY_SOURCE=3 -Wformat -Werror=format-security -fstack-clash-protection -fcf-protection -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -g -ffile-prefix-map=/build/python/src=/usr/src/debug/python -flto=auto -fPIC -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvcc/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cublas/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cudnn/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_cccl/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/nvtx/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_nvrtc/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/common_headers/common/include -I/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/torch/csrc/api/include -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/TH -I/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/THC -I/home/matteo/envs/torch-cuda/include -I/usr/include/python3.13 -c -c /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/activation.cpp -o /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build/temp.linux-x86_64-cpython-313/csrc/extensions/activation.o -O3 -fvisibility=hidden -g0 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=transformer_engine_torch -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++17
      In file included from /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/ATen/cuda/CUDAContextLight.h:6,
                       from /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/include/ATen/cuda/CUDAContext.h:3,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/../common.h:12,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/../extensions.h:12,
                       from /tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/csrc/extensions/activation.cpp:6:
      /home/matteo/envs/torch-cuda/lib/python3.13/site-packages/nvidia/cuda_runtime/include/cuda_runtime_api.h:148:10: fatal error: crt/host_defines.h: File o directory non esistente
        148 | #include "crt/host_defines.h"
            |          ^~~~~~~~~~~~~~~~~~~~
      compilation terminated.
      ninja: build stopped: subcommand failed.
      Traceback (most recent call last):
        File "<string>", line 118, in run
        File "/usr/lib/python3.13/urllib/request.py", line 214, in urlretrieve
          with contextlib.closing(urlopen(url, data)) as fp:
                                  ~~~~~~~^^^^^^^^^^^
        File "/usr/lib/python3.13/urllib/request.py", line 189, in urlopen
          return opener.open(url, data, timeout)
                 ~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^
        File "/usr/lib/python3.13/urllib/request.py", line 495, in open
          response = meth(req, response)
        File "/usr/lib/python3.13/urllib/request.py", line 604, in http_response
          response = self.parent.error(
              'http', request, response, code, msg, hdrs)
        File "/usr/lib/python3.13/urllib/request.py", line 533, in error
          return self._call_chain(*args)
                 ~~~~~~~~~~~~~~~~^^^^^^^
        File "/usr/lib/python3.13/urllib/request.py", line 466, in _call_chain
          result = func(*args)
        File "/usr/lib/python3.13/urllib/request.py", line 613, in http_error_default
          raise HTTPError(req.full_url, code, msg, hdrs, fp)
      urllib.error.HTTPError: HTTP Error 404: Not Found
      
      During handling of the above exception, another exception occurred:
      
      Traceback (most recent call last):
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/utils/cpp_extension.py", line 2209, in _run_ninja_build
          subprocess.run(
          ~~~~~~~~~~~~~~^
              command,
              ^^^^^^^^
          ...<3 lines>...
              check=True,
              ^^^^^^^^^^^
              env=env)
              ^^^^^^^^
        File "/usr/lib/python3.13/subprocess.py", line 577, in run
          raise CalledProcessError(retcode, process.args,
                                   output=stdout, stderr=stderr)
      subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.
      
      The above exception was the direct cause of the following exception:
      
      Traceback (most recent call last):
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 389, in <module>
          main()
          ~~~~^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 373, in main
          json_out["return_val"] = hook(**hook_input["kwargs"])
                                   ~~~~^^^^^^^^^^^^^^^^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 280, in build_wheel
          return _build_backend().build_wheel(
                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
              wheel_directory, config_settings, metadata_directory
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
          )
          ^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/build_meta.py", line 435, in build_wheel
          return _build(['bdist_wheel', '--dist-info-dir', str(metadata_directory)])
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/build_meta.py", line 423, in _build
          return self._build_with_temp_dir(
                 ~~~~~~~~~~~~~~~~~~~~~~~~~^
              cmd,
              ^^^^
          ...<3 lines>...
              self._arbitrary_args(config_settings),
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
          )
          ^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/build_meta.py", line 404, in _build_with_temp_dir
          self.run_setup()
          ~~~~~~~~~~~~~~^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/build_meta.py", line 512, in run_setup
          super().run_setup(setup_script=setup_script)
          ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/build_meta.py", line 317, in run_setup
          exec(code, locals())
          ~~~~^^^^^^^^^^^^^^^^
        File "<string>", line 159, in <module>
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/__init__.py", line 117, in setup
          return distutils.core.setup(**attrs)
                 ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/core.py", line 186, in setup
          return run_commands(dist)
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/core.py", line 202, in run_commands
          dist.run_commands()
          ~~~~~~~~~~~~~~~~~^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/dist.py", line 1002, in run_commands
          self.run_command(cmd)
          ~~~~~~~~~~~~~~~~^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/dist.py", line 1104, in run_command
          super().run_command(command)
          ~~~~~~~~~~~~~~~~~~~^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/dist.py", line 1021, in run_command
          cmd_obj.run()
          ~~~~~~~~~~~^^
        File "<string>", line 135, in run
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/command/bdist_wheel.py", line 370, in run
          self.run_command("build")
          ~~~~~~~~~~~~~~~~^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/cmd.py", line 357, in run_command
          self.distribution.run_command(command)
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/dist.py", line 1104, in run_command
          super().run_command(command)
          ~~~~~~~~~~~~~~~~~~~^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/dist.py", line 1021, in run_command
          cmd_obj.run()
          ~~~~~~~~~~~^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/command/build.py", line 135, in run
          self.run_command(cmd_name)
          ~~~~~~~~~~~~~~~~^^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/cmd.py", line 357, in run_command
          self.distribution.run_command(command)
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/dist.py", line 1104, in run_command
          super().run_command(command)
          ~~~~~~~~~~~~~~~~~~~^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/dist.py", line 1021, in run_command
          cmd_obj.run()
          ~~~~~~~~~~~^^
        File "/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build_tools/build_ext.py", line 131, in run
          super().run()
          ~~~~~~~~~~~^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/command/build_ext.py", line 99, in run
          _build_ext.run(self)
          ~~~~~~~~~~~~~~^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/command/build_ext.py", line 368, in run
          self.build_extensions()
          ~~~~~~~~~~~~~~~~~~~~~^^
        File "/tmp/pip-install-wt_vvn5q/transformer-engine-torch_c18f5f484dfb4d589bf55a29582498d4/build_tools/build_ext.py", line 208, in build_extensions
          super().build_extensions()
          ~~~~~~~~~~~~~~~~~~~~~~~~^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/utils/cpp_extension.py", line 900, in build_extensions
          build_ext.build_extensions(self)
          ~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/command/build_ext.py", line 484, in build_extensions
          self._build_extensions_serial()
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/command/build_ext.py", line 510, in _build_extensions_serial
          self.build_extension(ext)
          ~~~~~~~~~~~~~~~~~~~~^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/command/build_ext.py", line 264, in build_extension
          _build_ext.build_extension(self, ext)
          ~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/Cython/Distutils/build_ext.py", line 136, in build_extension
          super().build_extension(ext)
          ~~~~~~~~~~~~~~~~~~~~~~~^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/setuptools/_distutils/command/build_ext.py", line 565, in build_extension
          objects = self.compiler.compile(
              sources,
          ...<5 lines>...
              depends=ext.depends,
          )
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/utils/cpp_extension.py", line 713, in unix_wrap_ninja_compile
          _write_ninja_file_and_compile_objects(
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
              sources=sources,
              ^^^^^^^^^^^^^^^^
          ...<7 lines>...
              verbose=True,
              ^^^^^^^^^^^^^
              with_cuda=with_cuda)
              ^^^^^^^^^^^^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/utils/cpp_extension.py", line 1869, in _write_ninja_file_and_compile_objects
          _run_ninja_build(
          ~~~~~~~~~~~~~~~~^
              build_directory,
              ^^^^^^^^^^^^^^^^
          ...<2 lines>...
              # that failed to build but there isn't a good way to get it here.
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^



              error_prefix='Error compiling objects for extension')
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        File "/home/matteo/envs/torch-cuda/lib/python3.13/site-packages/torch/utils/cpp_extension.py", line 2225, in _run_ninja_build
          raise RuntimeError(message) from e
      RuntimeError: Error compiling objects for extension
      [end of output]
  
  note: This error originates from a subprocess, and is likely not a problem with pip.
  ERROR: Failed building wheel for transformer_engine_torch
Failed to build transformer_engine_torch

[notice] A new release of pip is available: 25.2 -> 25.3
[notice] To update, run: pip install --upgrade pip
error: failed-wheel-build-for-install

× Failed to build installable wheels for some pyproject.toml based projects
╰─> transformer_engine_torch

University's server (4 Nvidia A40): trying with CUDNN_PATH=/home/matteo/.local/lib/python3.10/site-packages/nvidia/cudnn/include CPLUS_INCLUDE_PATH=/home/matteo/.local/lib/python3.10/site-packages/nvidia/cudnn/include but in both cases:

        File "/home/matteo/miniconda3/lib/python3.12/site-packages/setuptools/command/build_ext.py", line 96, in run
          _build_ext.run(self)
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/setuptools/_distutils/command/build_ext.py", line 368, in run
          self.build_extensions()
        File "/tmp/pip-install-v0d6dw0z/transformer-engine-torch_c11d10be70a74e29a480b47d50cf42b5/build_tools/build_ext.py", line 208, in build_extensions
          super().build_extensions()
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/torch/utils/cpp_extension.py", line 900, in build_extensions
          build_ext.build_extensions(self)
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/setuptools/_distutils/command/build_ext.py", line 484, in build_extensions
          self._build_extensions_serial()
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/setuptools/_distutils/command/build_ext.py", line 510, in _build_extensions_serial
          self.build_extension(ext)
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/setuptools/command/build_ext.py", line 261, in build_extension
          _build_ext.build_extension(self, ext)
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/setuptools/_distutils/command/build_ext.py", line 565, in build_extension
          objects = self.compiler.compile(
                    ^^^^^^^^^^^^^^^^^^^^^^
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/torch/utils/cpp_extension.py", line 713, in unix_wrap_ninja_compile
          _write_ninja_file_and_compile_objects(
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/torch/utils/cpp_extension.py", line 1869, in _write_ninja_file_and_compile_objects
          _run_ninja_build(
        File "/home/matteo/miniconda3/lib/python3.12/site-packages/torch/utils/cpp_extension.py", line 2225, in _run_ninja_build
          raise RuntimeError(message) from e
      RuntimeError: Error compiling objects for extension
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
  ERROR: Failed building wheel for transformer_engine_torch
Failed to build transformer_engine_torch
ERROR: Could not build wheels for transformer_engine_torch, which is required to install pyproject.toml-based projects

It looks it doesn't find cudnn.h even though I am pointing it to the right location

      /home/matteo/miniconda3/lib/python3.12/site-packages/torch/include/ATen/cudnn/cudnn-wrapper.h:3:10: fatal error: cudnn.h: No such file o
r directory
          3 | #include <cudnn.h>
            |          ^~~~~~~~~
      compilation terminated.

Is it a bug or am I doing something wrong? Thanks.

mrinaldi97 avatar Nov 14 '25 12:11 mrinaldi97