- 版本、环境信息:
1)PaddlePaddle版本:develop
2)CPU:默认
3)GPU:
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 435.21 Driver Version: 435.21 CUDA Version: 10.1 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GT 710 Off | 00000000:01:00.0 N/A | N/A |
| 35% 46C P0 N/A / N/A | 418MiB / 974MiB | N/A Default |
+-------------------------------+----------------------+----------------------+
4)系统环境:Ubuntu 18.04 + conda
(smu) vimos@x86_64-conda_cos6-linux-gnu ➜ ~ conda env export
name: smu
channels:
- conda-forge
- defaults
dependencies:
- _libgcc_mutex=0.1=conda_forge
- _openmp_mutex=4.5=0_gnu
- binutils_impl_linux-64=2.33.1=he1b5a44_7
- binutils_linux-64=2.33.1=h9595d00_16
- bzip2=1.0.8=h516909a_2
- ca-certificates=2019.11.28=hecc5488_0
- certifi=2019.11.28=py38_0
- cmake=3.16.2=h28c56e5_0
- cudatoolkit=10.2.89=hfd86e86_0
- cudnn=7.6.5=cuda10.2_0
- expat=2.2.5=he1b5a44_1004
- gcc_impl_linux-64=7.3.0=hd420e75_4
- gcc_linux-64=7.3.0=h553295d_16
- gxx_impl_linux-64=7.3.0=hdf63c60_4
- gxx_linux-64=7.3.0=h553295d_16
- krb5=1.16.4=h2fd8d38_0
- ld_impl_linux-64=2.33.1=h53a641e_7
- libblas=3.8.0=14_openblas
- libcblas=3.8.0=14_openblas
- libcurl=7.65.3=hda55be3_0
- libedit=3.1.20170329=hf8c457e_1001
- libffi=3.2.1=he1b5a44_1006
- libgcc-ng=9.2.0=h24d8f2e_2
- libgfortran-ng=7.3.0=hdf63c60_4
- libgomp=9.2.0=h24d8f2e_2
- liblapack=3.8.0=14_openblas
- libopenblas=0.3.7=h5ec1e0e_6
- libprotobuf=3.11.2=h8b12597_0
- libssh2=1.8.2=h22169c7_2
- libstdcxx-ng=9.2.0=hdf63c60_2
- libuv=1.34.0=h516909a_0
- make=4.2.1=h14c3975_2004
- nccl=2.5.6.1=hc6a2c23_0
- ncurses=6.1=hf484d3e_1002
- numpy=1.17.3=py38h95a1406_0
- openssl=1.1.1d=h516909a_0
- patchelf=0.10=he1b5a44_0
- pip=19.3.1=py38_0
- protobuf=3.11.2=py38he1b5a44_0
- python=3.8.1=h357f687_1
- readline=8.0=hf8c457e_0
- rhash=1.3.6=h14c3975_1001
- setuptools=45.0.0=py38_1
- six=1.13.0=py38_0
- sqlite=3.30.1=hcee41ef_0
- tk=8.6.10=hed695b0_0
- wheel=0.33.6=py38_0
- xz=5.2.4=h14c3975_1001
- zlib=1.2.11=h516909a_1006
prefix: /home/vimos/anaconda3/envs/smu
- 安装方式信息:
本地编译:请提供cmake命令,编译命令
#=============================================================================
# Target rules for targets named nccl_wrapper
# Build rule for target.
nccl_wrapper: cmake_check_build_system
$(MAKE) -f CMakeFiles/Makefile2 nccl_wrapper
.PHONY : nccl_wrapper
# fast build rule for target.
nccl_wrapper/fast:
$(MAKE) -f paddle/fluid/framework/fleet/CMakeFiles/nccl_wrapper.dir/build.make paddle/fluid/framework/fleet/CMakeFiles/nccl_wrapper.dir/build
.PHONY : nccl_wrapper/fast
3)docker编译:请提供docker镜像,编译命令
特殊环境请注明:如离线安装等
- 复现信息:如为报错,请给出复现环境、复现步骤
- 问题描述:请详细描述您的问题,同步贴出报错信息、日志/代码关键片段
[ 19%] Building NVCC (Device) object paddle/fluid/operators/math/CMakeFiles/sequence_padding.dir/sequence_padding_generated_sequence_padding.cu.o
[ 19%] Linking CXX static library libnccl_context.a
[ 19%] Built target nccl_context
[ 19%] Building NVCC (Device) object paddle/fluid/operators/math/CMakeFiles/sequence_scale.dir/sequence_scale_generated_sequence_scale.cu.o
/home/vimos/anaconda3/envs/smu/lib/gcc/x86_64-conda_cos6-linux-gnu/7.3.0/include/mm_malloc.h(34): error: allowing all exceptions is incompatible with previous function "posix_memalign"
/usr/include/stdlib.h(580): here
1 error detected in the compilation of "/tmp/tmpxft_0000267c_00000000-13_sequence_padding.compute_75.cpp1.ii".
CMake Error at sequence_padding_generated_sequence_padding.cu.o.Debug.cmake:276 (message):
Error generating file
/home/vimos/git/Paddle/cmake-build-debug-smu/paddle/fluid/operators/math/CMakeFiles/sequence_padding.dir//./sequence_padding_generated_sequence_padding.cu.o
make[3]: *** [paddle/fluid/operators/math/CMakeFiles/sequence_padding.dir/build.make:65: paddle/fluid/operators/math/CMakeFiles/sequence_padding.dir/sequence_padding_generated_sequence_padding.cu.o] Error 1
make[2]: *** [CMakeFiles/Makefile2:49574: paddle/fluid/operators/math/CMakeFiles/sequence_padding.dir/all] Error 2
make[2]: *** Waiting for unfinished jobs....
/home/vimos/anaconda3/envs/smu/lib/gcc/x86_64-conda_cos6-linux-gnu/7.3.0/include/mm_malloc.h(34): error: allowing all exceptions is incompatible with previous function "posix_memalign"
/usr/include/stdlib.h(580): here
1 error detected in the compilation of "/tmp/tmpxft_0000269a_00000000-13_sequence_scale.compute_75.cpp1.ii".
CMake Error at sequence_scale_generated_sequence_scale.cu.o.Debug.cmake:276 (message):
Error generating file
/home/vimos/git/Paddle/cmake-build-debug-smu/paddle/fluid/operators/math/CMakeFiles/sequence_scale.dir//./sequence_scale_generated_sequence_scale.cu.o
make[3]: *** [paddle/fluid/operators/math/CMakeFiles/sequence_scale.dir/build.make:65: paddle/fluid/operators/math/CMakeFiles/sequence_scale.dir/sequence_scale_generated_sequence_scale.cu.o] Error 1
make[2]: *** [CMakeFiles/Makefile2:51019: paddle/fluid/operators/math/CMakeFiles/sequence_scale.dir/all] Error 2
make[1]: *** [CMakeFiles/Makefile2:72335: python/CMakeFiles/paddle_python.dir/rule] Error 2
make: *** [Makefile:8224: paddle_python] Error 2
1)PaddlePaddle版本:develop
2)CPU:默认
3)GPU:
4)系统环境:Ubuntu 18.04 + conda
本地编译:请提供cmake命令,编译命令
3)docker编译:请提供docker镜像,编译命令
特殊环境请注明:如离线安装等