search

PaddlePaddle / PaddleNLP

👑 Easy-to-use and powerful NLP and LLM library with 🤗 Awesome model zoo, supporting wide-range of NLP tasks from research to industrial applications, including 🗂Text Classification, 🔍 Neural Search, ❓ Question Answering, ℹ️ Information Extraction, 📄 Document Intelligence, 💌 Sentiment Analysis etc.
https://paddlenlp.readthedocs.io
Apache License 2.0
12.1k stars 2.93k forks source link

[Bug]: 编译paddlenlp_ops报错,编译这个需要的环境能否提供下？ #9303

Closed yaphet266 closed 2 weeks ago

yaphet266 commented 2 weeks ago

软件环境

paddlenlp_ops报错,编译这个需要的环境能否提供下？
比如 gcc版本
cuda版本等

目前paddle版本如下：
paddle2onnx              1.2.11
paddlefsl                1.1.0
paddlenlp                3.0.0b2
paddlepaddle-gpu         3.0.0.dev20241019
paddleslim               2.6.0

gcc版本:gcc version 11.2.0 (GCC) 
make:GNU Make 4.2.1
cmake version 3.16.3

重复问题

错误描述

(py39_paddlenlp) root@90d36e064828:/workspace2/yyc/PaddleNLP/csrc# python setup_cuda.py install
/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/utils/cpp_extension/extension_utils.py:686: UserWarning: No ccache found. Please be aware that recompiling all source files may be required. You can download and install ccache from: https://github.com/ccache/ccache/blob/master/doc/INSTALL.md
  warnings.warn(warning_message)
Cloning into 'third_party/cutlass'...
fatal: unable to access 'https://github.com/NVIDIA/cutlass.git/': GnuTLS recv error (-110): The TLS connection was non-properly terminated.
Git clone https://github.com/NVIDIA/cutlass.git operation failed with the following error: Command '['git', 'clone', '-b', 'v3.5.0', '--single-branch', 'https://github.com/NVIDIA/cutlass.git', 'third_party/cutlass', '--depth=1']' returned non-zero exit status 128.
Please check your network connection or access rights to the repository.
If the problem persists, please refer to the README file for instructions on how to manually download and install the necessary components.
[2024-10-23 09:43:16,115] [    INFO] dist.py:970 - running install
/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/setuptools/_distutils/cmd.py:66: SetuptoolsDeprecationWarning: setup.py install is deprecated.
!!

        ********************************************************************************
        Please avoid running ``setup.py`` directly.
        Instead, use pypa/build, pypa/installer or other
        standards-based tools.

        See https://blog.ganssle.io/articles/2021/10/setup-py-deprecated.html for details.
        ********************************************************************************

!!
  self.initialize_options()
/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/setuptools/_distutils/cmd.py:66: EasyInstallDeprecationWarning: easy_install command is deprecated.
!!

        ********************************************************************************
        Please avoid running ``setup.py`` and ``easy_install``.
        Instead, use pypa/build, pypa/installer or other
        standards-based tools.

        See https://github.com/pypa/setuptools/issues/917 for details.
        ********************************************************************************

!!
  self.initialize_options()
[2024-10-23 09:43:16,346] [    INFO] dist.py:970 - running bdist_egg
[2024-10-23 09:43:16,371] [    INFO] dist.py:970 - running egg_info
[2024-10-23 09:43:16,377] [    INFO] egg_info.py:648 - writing paddlenlp_ops.egg-info/PKG-INFO
[2024-10-23 09:43:16,378] [    INFO] egg_info.py:282 - writing dependency_links to paddlenlp_ops.egg-info/dependency_links.txt
[2024-10-23 09:43:16,378] [    INFO] egg_info.py:282 - writing top-level names to paddlenlp_ops.egg-info/top_level.txt
[2024-10-23 09:43:16,386] [    INFO] sdist.py:202 - reading manifest file 'paddlenlp_ops.egg-info/SOURCES.txt'
[2024-10-23 09:43:16,389] [    INFO] util.py:324 - writing manifest file 'paddlenlp_ops.egg-info/SOURCES.txt'
[2024-10-23 09:43:16,389] [    INFO] bdist_egg.py:162 - installing library code to build/paddlenlp_ops/bdist.linux-x86_64/egg
[2024-10-23 09:43:16,389] [    INFO] dist.py:970 - running install_lib
[2024-10-23 09:43:16,390] [    INFO] dist.py:970 - running build_ext
Compiling user custom op, it will cost a few seconds.....
[2024-10-23 09:43:16,452] [    INFO] build_ext.py:530 - building 'paddlenlp_ops' extension
[2024-10-23 09:43:16,455] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/dequant_int8.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,457] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/encode_rotary_qk.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,459] [    INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/flash_attn_bwd.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/flash_attn_bwd.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17
[2024-10-23 09:43:16,475] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/fused_get_rope.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,486] [    INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/get_output.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_output.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17
/workspace2/yyc/PaddleNLP/csrc/gpu/dequant_int8.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
[2024-10-23 09:43:16,489] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
/workspace2/yyc/PaddleNLP/csrc/gpu/encode_rotary_qk.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
nvcc fatal   : Value 'c++17' is not defined for option 'std'
/workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
[2024-10-23 09:43:16,510] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,512] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/qkv_transpose_split.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,521] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/quant_int8.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,542] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,545] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,564] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/sample_kernels/top_p_sampling_reject.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
/workspace2/yyc/PaddleNLP/csrc/gpu/fused_get_rope.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,585] [    INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/save_with_output.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17
/workspace2/yyc/PaddleNLP/csrc/gpu/quant_int8.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
[2024-10-23 09:43:16,589] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,590] [    INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/save_with_output_msg.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output_msg.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17
/workspace2/yyc/PaddleNLP/csrc/gpu/qkv_transpose_split.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,599] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,601] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/step.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,630] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
/workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
[2024-10-23 09:43:16,632] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
/workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
[2024-10-23 09:43:16,635] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,654] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
/workspace2/yyc/PaddleNLP/csrc/gpu/sample_kernels/top_p_sampling_reject.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
[2024-10-23 09:43:16,665] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/tune_cublaslt_gemm.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
/workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
[2024-10-23 09:43:16,666] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/transpose_removing_padding.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,680] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/update_inputs.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,682] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/write_cache_kv.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
[2024-10-23 09:43:16,684] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/write_int8_cache_kv.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
/workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/step.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
[2024-10-23 09:43:16,711] [    INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/int8_gemm_with_cutlass/gemm_dequant.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_BFLOAT16_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ -U__CUDA_NO_BFLOAT162_OPERATORS__ -U__CUDA_NO_BFLOAT162_CONVERSIONS__ -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
nvcc fatal   : Value 'c++17' is not defined for option 'std'
/workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/write_cache_kv.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/transpose_removing_padding.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/update_inputs.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/gpu/tune_cublaslt_gemm.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/gpu/int8_gemm_with_cutlass/gemm_dequant.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o is compiled
nvcc fatal   : Value 'c++17' is not defined for option 'std'
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/gpu/write_int8_cache_kv.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o is compiled
nvcc fatal   : Value 'c++17' is not defined for option 'std'
/workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_output.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output_msg.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output.o is compiled
/workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/flash_attn_bwd.o is compiled
[2024-10-23 09:43:23,693] [    INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -shared /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/flash_attn_bwd.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_output.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output_msg.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o -L/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/cv2/../../lib64: -L/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/libs -L/usr/lib64 -L/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/base -Wl,--enable-new-dtags,-rpath,/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/libs -Wl,--enable-new-dtags,-rpath,/usr/lib64 -Wl,--enable-new-dtags,-rpath,/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/base -lcublasLt -o build/paddlenlp_ops/lib.linux-x86_64-cpython-39/paddlenlp_ops.so -l:libpaddle.so -lcudart
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o: No such file or directory
/root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o: No such file or directory
collect2: error: ld returned 1 exit status
error: command '/usr/local/bin/g++' failed with exit code 1

稳定复现步骤 & 代码

(py39_paddlenlp) root@90d36e064828:/workspace2/yyc/PaddleNLP/csrc# python setup_cuda.py install

ZHUI commented 2 weeks ago

你好，看起来是 git clone cutlass 出现了网络错误。

yaphet266 commented 2 weeks ago

@ZHUI 你好，这个cutlass项目是在哪里，我可以在本地下载好吗，下载好是放在哪个位置

yaphet266 commented 2 weeks ago

这次git clone cutlass可以了，但是还是报下面的错误 (py39_paddlenlp) root@90d36e064828:/workspace2/yyc/PaddleNLP/csrc# python setup_cuda.py install /root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/utils/cpp_extension/extension_utils.py:686: UserWarning: No ccache found. Please be aware that recompiling all source files may be required. You can download and install ccache from: https://github.com/ccache/ccache/blob/master/doc/INSTALL.md warnings.warn(warning_message) Cloning into 'third_party/cutlass'... remote: Enumerating objects: 5992, done. remote: Counting objects: 100% (5992/5992), done. remote: Compressing objects: 100% (1639/1639), done. remote: Total 5992 (delta 3485), reused 4943 (delta 3065), pack-reused 0 (from 0) Receiving objects: 100% (5992/5992), 27.23 MiB | 9.39 MiB/s, done. Resolving deltas: 100% (3485/3485), done. Note: switching to '7d49e6c7e2f8896c47f586706e67e1fb215529dc'.

You are in 'detached HEAD' state. You can look around, make experimental changes and commit them, and you can discard any commits you make in this state without impacting any branches by switching back to a branch.

If you want to create a new branch to retain commits you create, you may do so (now or later) by using -c with the switch command. Example:

git switch -c

Or undo this operation with:

git switch -

Turn off this advice by setting config variable advice.detachedHead to false

[2024-10-23 11:07:28,262] [ INFO] dist.py:970 - running install /root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/setuptools/_distutils/cmd.py:66: SetuptoolsDeprecationWarning: setup.py install is deprecated. !!

    ********************************************************************************
    Please avoid running ``setup.py`` directly.
    Instead, use pypa/build, pypa/installer or other
    standards-based tools.

    See https://blog.ganssle.io/articles/2021/10/setup-py-deprecated.html for details.
    ********************************************************************************

!! self.initialize_options() /root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/setuptools/_distutils/cmd.py:66: EasyInstallDeprecationWarning: easy_install command is deprecated. !!

    ********************************************************************************
    Please avoid running ``setup.py`` and ``easy_install``.
    Instead, use pypa/build, pypa/installer or other
    standards-based tools.

    See https://github.com/pypa/setuptools/issues/917 for details.
    ********************************************************************************

!! self.initialize_options() [2024-10-23 11:07:28,506] [ INFO] dist.py:970 - running bdist_egg [2024-10-23 11:07:28,531] [ INFO] dist.py:970 - running egg_info [2024-10-23 11:07:28,538] [ INFO] egg_info.py:648 - writing paddlenlp_ops.egg-info/PKG-INFO [2024-10-23 11:07:28,539] [ INFO] egg_info.py:282 - writing dependency_links to paddlenlp_ops.egg-info/dependency_links.txt [2024-10-23 11:07:28,539] [ INFO] egg_info.py:282 - writing top-level names to paddlenlp_ops.egg-info/top_level.txt [2024-10-23 11:07:28,548] [ INFO] sdist.py:202 - reading manifest file 'paddlenlp_ops.egg-info/SOURCES.txt' [2024-10-23 11:07:28,551] [ INFO] util.py:324 - writing manifest file 'paddlenlp_ops.egg-info/SOURCES.txt' [2024-10-23 11:07:28,551] [ INFO] bdist_egg.py:162 - installing library code to build/paddlenlp_ops/bdist.linux-x86_64/egg [2024-10-23 11:07:28,551] [ INFO] dist.py:970 - running install_lib [2024-10-23 11:07:28,551] [ INFO] dist.py:970 - running build_ext Compiling user custom op, it will cost a few seconds..... [2024-10-23 11:07:28,619] [ INFO] build_ext.py:530 - building 'paddlenlp_ops' extension [2024-10-23 11:07:28,623] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/dequant_int8.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 [2024-10-23 11:07:28,624] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/encode_rotary_qk.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 [2024-10-23 11:07:28,626] [ INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/flash_attn_bwd.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/flash_attn_bwd.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17 [2024-10-23 11:07:28,645] [ INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/get_output.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_output.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' [2024-10-23 11:07:28,647] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/fused_get_rope.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/dequant_int8.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,651] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/encode_rotary_qk.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,653] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 [2024-10-23 11:07:28,654] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/qkv_transpose_split.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/fused_get_rope.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,664] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/quant_int8.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 [2024-10-23 11:07:28,686] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' nvcc fatal : Value 'c++17' is not defined for option 'std' [2024-10-23 11:07:28,701] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/get_padding_offset_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,702] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/sample_kernels/top_p_sampling_reject.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 [2024-10-23 11:07:28,714] [ INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/save_with_output.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/quant_int8.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,717] [ INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -fPIC -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/save_with_output_msg.cc -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output_msg.o -O3 -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -DPADDLE_WITH_CUDA -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' [2024-10-23 11:07:28,728] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 [2024-10-23 11:07:28,741] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/qkv_transpose_split.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,743] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/step.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' nvcc fatal : Value 'c++17' is not defined for option 'std' [2024-10-23 11:07:28,746] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 [2024-10-23 11:07:28,758] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 /workspace2/yyc/PaddleNLP/csrc/gpu/sample_kernels/top_p_sampling_reject.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/step.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,769] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,773] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores_v2.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/rebuild_padding_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,785] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/transpose_removing_padding.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o is compiled [2024-10-23 11:07:28,809] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/tune_cublaslt_gemm.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/set_value_by_flags.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 nvcc fatal : Value 'c++17' is not defined for option 'std' [2024-10-23 11:07:28,838] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/update_inputs.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 /workspace2/yyc/PaddleNLP/csrc/gpu/stop_generation_multi_ends_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 [2024-10-23 11:07:28,842] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/write_cache_kv.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 nvcc fatal : Value 'c++17' is not defined for option 'std' nvcc fatal : Value 'c++17' is not defined for option 'std' [2024-10-23 11:07:28,843] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/write_int8_cache_kv.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 nvcc fatal : Value 'c++17' is not defined for option 'std' [2024-10-23 11:07:28,858] [ INFO] spawn.py:60 - /usr/bin/nvcc -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include -I/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/include/third_party -I/usr/include -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -I/root/miniconda3/envs/py39_paddlenlp/include/python3.9 -c /workspace2/yyc/PaddleNLP/csrc/gpu/int8_gemm_with_cutlass/gemm_dequant.cu -o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -ccbin cc -Xcompiler -fPIC --expt-relaxed-constexpr -DNVCC -gencode arch=compute_86,code=sm_86 -O3 -UCUDA_NO_HALF_OPERATORS -UCUDA_NO_HALF_CONVERSIONS -UCUDA_NO_BFLOAT16_OPERATORS -UCUDA_NO_BFLOAT16_CONVERSIONS -UCUDA_NO_BFLOAT162_OPERATORS -UCUDA_NO_BFLOAT162_CONVERSIONS -Igpu -Igpu/cutlass_kernels -Igpu/fp8_gemm_with_cutlass -Igpu/cutlass_kernels/fp8_gemm_fused/autogen -Ithird_party/cutlass/include -Ithird_party/nlohmann_json/single_include -Igpu/sample_kernels -w -DPADDLE_WITH_CUSTOM_KERNEL -D_GLIBCXX_USE_CXX11_ABI=1 -std=c++17 /workspace2/yyc/PaddleNLP/csrc/gpu/token_penalty_multi_scores_v2.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/tune_cublaslt_gemm.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/transpose_removing_padding.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/gpu/update_inputs.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 /workspace2/yyc/PaddleNLP/csrc/gpu/write_cache_kv.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/gpu/write_int8_cache_kv.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o is compiled nvcc fatal : Value 'c++17' is not defined for option 'std' /workspace2/yyc/PaddleNLP/csrc/gpu/int8_gemm_with_cutlass/gemm_dequant.cu compile failed, command '/usr/bin/nvcc' failed with exit code 1 /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output_msg.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_output.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/flash_attn_bwd.o is compiled /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output.o is compiled [2024-10-23 11:07:36,252] [ INFO] spawn.py:60 - g++ -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -I/root/miniconda3/envs/py39_paddlenlp/include -fPIC -O2 -isystem /root/miniconda3/envs/py39_paddlenlp/include -pthread -B /root/miniconda3/envs/py39_paddlenlp/compiler_compat -shared /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/flash_attn_bwd.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_output.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/save_with_output_msg.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o -L/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/cv2/../../lib64: -L/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/libs -L/usr/lib64 -L/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/base -Wl,--enable-new-dtags,-rpath,/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/libs -Wl,--enable-new-dtags,-rpath,/usr/lib64 -Wl,--enable-new-dtags,-rpath,/root/miniconda3/envs/py39_paddlenlp/lib/python3.9/site-packages/paddle/base -lcublasLt -o build/paddlenlp_ops/lib.linux-x86_64-cpython-39/paddlenlp_ops.so -l:libpaddle.so -lcudart /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/dequant_int8.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/encode_rotary_qk.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/fused_get_rope.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/get_padding_offset_v2.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/qkv_transpose_split.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/quant_int8.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/rebuild_padding_v2.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/top_p_sampling_reject.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/set_value_by_flags_v2.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/step.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/stop_generation_multi_ends_v2.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/token_penalty_multi_scores_v2.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/transpose_removing_padding.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/tune_cublaslt_gemm.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/update_inputs.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_cache_kv.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/write_int8_cache_kv.cu.o: No such file or directory /root/miniconda3/envs/py39_paddlenlp/compiler_compat/ld: cannot find /workspace2/yyc/PaddleNLP/csrc/build/paddlenlp_ops/lib.linux-x86_64-cpython-39/gemm_dequant.cu.o: No such file or directory collect2: error: ld returned 1 exit status error: command '/usr/local/bin/g++' failed with exit code 1

DrownFish19 commented 2 weeks ago

建议先把csrc目录清理感情，重新执行编译安装。可能是历史编译文件的问题

yaphet266 commented 2 weeks ago

@DrownFish19 问题已经解决了，可能是CUDA问题，我找了个干净的CUDA 11.8镜像，一把就编译过了，这个paddle-ops你们编译好python包，后续使用会方便些