Numpy 构建错误：gcc: build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.c

【问题标题】：Numpy building error: gcc: build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.cNumpy 构建错误：gcc: build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.c
【发布时间】：2022-07-08 08:03:21
【问题描述】：

我正在尝试通过pip 安装 Numpy v1.22.4，接下来通过编译源代码。
执行$ python setup.py build时遇到以下错误。

error: Command "gcc -pthread -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -fPIC -DNPY_INTERNAL_BUILD=1 -DHAVE_NPY_CONFIG_H=1 -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE=1 -D_LARGEFILE64_SOURCE=1 -DHAVE_CBLAS -DNO_ATLAS_INFO=-1 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/umath -Inumpy/core/include -Ibuild/src.linux-x86_64-3.10/numpy/core/include/numpy -Ibuild/src.linux-x86_64-3.10/numpy/distutils/include -Inumpy/core/src/common -Inumpy/core/src -Inumpy/core -Inumpy/core/src/npymath -Inumpy/core/src/multiarray -Inumpy/core/src/umath -Inumpy/core/src/npysort -Inumpy/core/src/_simd -I/var/www/project/my-product/.venv/include -I/usr/local/include/python3.10 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/npymath -c build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.c -o build/temp.linux-x86_64-cpython-310/build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.o -MMD -MF build/temp.linux-x86_64-cpython-310/build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.o.d -msse -msse2 -msse3" failed with exit status 1

以下是我尝试过的事情：

  • 在 CentOS 6 上安装 Python 3.10.4。
    *因为似乎 Python 3.10.3 无法在 CentOS 6 上运行。
  • 正在安装 gcc v9.5.0，以及 v11.3.0。
    *与默认 gcc (v4.4.7) 分开。
    *我在安装 gcc v11.3.0 时使用了contrib/download_prerequisites。
  • 从 gcc v11.3.0 更新 libstdc++。

*我稍后会描述更多细节。

您能告诉我如何解决这个错误吗？

我的开发环境资料

(.venv) [vagrant@localhost ~]$ cat /etc/redhat-release 
CentOS release 6.10 (Final)

(.venv) [vagrant@localhost ~]$ which python
/var/www/project/my-product/.venv/bin/python
(.venv) [vagrant@localhost ~]$ python -V
Python 3.10.4

(.venv) [vagrant@localhost ~]$ which pip
/var/www/project/my-product/.venv/bin/pip
(.venv) [vagrant@localhost ~]$ pip -V
pip 22.1.1 from /var/www/project/my-product/.venv/lib/python3.10/site-packages/pip (python 3.10)
(.venv) [vagrant@localhost ~]$ pip list
Package    Version
---------- -------
Cython     0.29.30
pip        22.1.1
setuptools 62.3.2
WARNING: There was an error checking the latest version of pip.
(.venv) [vagrant@localhost ~]$ python -m cython --version
Cython version 0.29.30

(.venv) [vagrant@localhost ~]$ which gcc
/usr/local/bin/gcc
(.venv) [vagrant@localhost ~]$ gcc --version
gcc (GCC) 11.3.0
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

(.venv) [vagrant@localhost ~]$ which g++
/usr/local/bin/g++
(.venv) [vagrant@localhost ~]$ g++ --version
g++ (GCC) 11.3.0
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

(.venv) [vagrant@localhost ~]$ strings /usr/lib64/libstdc++.so.6 | grep GLIBCXX
GLIBCXX_3.4
GLIBCXX_3.4.1
GLIBCXX_3.4.2
GLIBCXX_3.4.3
GLIBCXX_3.4.4
GLIBCXX_3.4.5
GLIBCXX_3.4.6
GLIBCXX_3.4.7
GLIBCXX_3.4.8
GLIBCXX_3.4.9
GLIBCXX_3.4.10
GLIBCXX_3.4.11
GLIBCXX_3.4.12
GLIBCXX_3.4.13
GLIBCXX_3.4.14
GLIBCXX_3.4.15
GLIBCXX_3.4.16
GLIBCXX_3.4.17
GLIBCXX_3.4.18
GLIBCXX_3.4.19
GLIBCXX_3.4.20
GLIBCXX_3.4.21
GLIBCXX_3.4.22
GLIBCXX_3.4.23
GLIBCXX_3.4.24
GLIBCXX_3.4.25
GLIBCXX_3.4.26
GLIBCXX_3.4.27
GLIBCXX_3.4.28
GLIBCXX_FORCE_NEW
GLIBCXX_DEBUG_MESSAGE_LENGTH

(.venv) [vagrant@localhost ~]$ sudo yum list installed | grep -e "atlas" -e "lapack" -e "blas" -e "gfortran" -e "gmp" -e "mpfr" -e "mpc" -e "python-devel" -e "python34-devel" -e "Cython"
atlas.x86_64                        3.8.4-2.el6                      @base      
atlas-devel.x86_64                  3.8.4-2.el6                      @base      
blas.x86_64                         3.2.1-5.el6                      @base      
blas-devel.x86_64                   3.2.1-5.el6                      @base      
gcc-gfortran.x86_64                 4.4.7-23.el6                     @base      
gmp.x86_64                          4.3.1-13.el6                     @base      
gmp-devel.x86_64                    4.3.1-13.el6                     @base      
lapack.x86_64                       3.2.1-5.el6                      @base      
lapack-devel.x86_64                 3.2.1-5.el6                      @base      
libgfortran.x86_64                  4.4.7-23.el6                     @base      
libmpc.x86_64                       0.8-3.el6                        @epel      
libmpc-devel.x86_64                 0.8-3.el6                        @epel      
mpfr.x86_64                         2.4.1-6.el6                      @base      
mpfr-devel.x86_64                   2.4.1-6.el6                      @base      
openblas.x86_64                     0.3.3-2.el6                      @epel      
python-devel.x86_64                 2.6.6-68.el6_10                  @updates   
python34-devel.x86_64               3.4.10-4.el6                     @epel

我执行的命令

(.venv) [vagrant@localhost src]$ pwd
/home/vagrant/src
(.venv) [vagrant@localhost src]$ git clone https://github.com/numpy/numpy.git
(.venv) [vagrant@localhost src]$ cd numpy
(.venv) [vagrant@localhost numpy]$ git checkout refs/tags/v1.22.4
(.venv) [vagrant@localhost numpy]$ git submodule update --init
(.venv) [vagrant@localhost numpy]$ python setup.py build 2>&1 | tee build_log.txt

... *Omitting

INFO: compile options: '-DNPY_INTERNAL_BUILD=1 -DHAVE_NPY_CONFIG_H=1 -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE=1 -D_LARGEFILE64_SOURCE=1 -DHAVE_CBLAS -DNO_ATLAS_INFO=-1 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/umath -Inumpy/core/include -Ibuild/src.linux-x86_64-3.10/numpy/core/include/numpy -Ibuild/src.linux-x86_64-3.10/numpy/distutils/include -Inumpy/core/src/common -Inumpy/core/src -Inumpy/core -Inumpy/core/src/npymath -Inumpy/core/src/multiarray -Inumpy/core/src/umath -Inumpy/core/src/npysort -Inumpy/core/src/_simd -I/var/www/project/my-product/.venv/include -I/usr/local/include/python3.10 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/npymath -c'
extra options: '-O3 -msse -msse2 -msse3'
INFO: compiling C sources
INFO: C compiler: gcc -pthread -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -fPIC

INFO: compile options: '-DNPY_INTERNAL_BUILD=1 -DHAVE_NPY_CONFIG_H=1 -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE=1 -D_LARGEFILE64_SOURCE=1 -DHAVE_CBLAS -DNO_ATLAS_INFO=-1 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/umath -Inumpy/core/include -Ibuild/src.linux-x86_64-3.10/numpy/core/include/numpy -Ibuild/src.linux-x86_64-3.10/numpy/distutils/include -Inumpy/core/src/common -Inumpy/core/src -Inumpy/core -Inumpy/core/src/npymath -Inumpy/core/src/multiarray -Inumpy/core/src/umath -Inumpy/core/src/npysort -Inumpy/core/src/_simd -I/var/www/project/my-product/.venv/include -I/usr/local/include/python3.10 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/npymath -c'
extra options: '-msse -msse2 -msse3'
INFO: gcc: build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.c
/tmp/ccuh177O.s: Assembler messages:
/tmp/ccuh177O.s:36: Error: no such instruction: `vpbroadcastd %edx,%zmm1'
/tmp/ccuh177O.s:37: Error: bad register name `%zmm1'
/tmp/ccuh177O.s:63: Error: bad register name `%zmm2'
/tmp/ccuh177O.s:69: Error: no such instruction: `kxnorw %k1,%k1,%k1'

... *Omitting

/tmp/ccuh177O.s:18728: Error: no such instruction: `kmovw %ecx,%k1'
/tmp/ccuh177O.s:18741: Error: invalid character '{' in operand 2
/tmp/ccuh177O.s:18754: Error: invalid character '{' in operand 2
/tmp/ccuh177O.s:18777: Error: bad register name `%zmm1'
/tmp/ccuh177O.s:18792: Error: bad register name `%zmm0'
/tmp/ccuh177O.s:18801: Error: bad register name `%zmm1'
/tmp/ccuh177O.s:18810: Error: bad register name `%zmm0'
error: Command "gcc -pthread -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -fPIC -DNPY_INTERNAL_BUILD=1 -DHAVE_NPY_CONFIG_H=1 -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE=1 -D_LARGEFILE64_SOURCE=1 -DHAVE_CBLAS -DNO_ATLAS_INFO=-1 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/umath -Inumpy/core/include -Ibuild/src.linux-x86_64-3.10/numpy/core/include/numpy -Ibuild/src.linux-x86_64-3.10/numpy/distutils/include -Inumpy/core/src/common -Inumpy/core/src -Inumpy/core -Inumpy/core/src/npymath -Inumpy/core/src/multiarray -Inumpy/core/src/umath -Inumpy/core/src/npysort -Inumpy/core/src/_simd -I/var/www/project/my-product/.venv/include -I/usr/local/include/python3.10 -Ibuild/src.linux-x86_64-3.10/numpy/core/src/common -Ibuild/src.linux-x86_64-3.10/numpy/core/src/npymath -c build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.c -o build/temp.linux-x86_64-cpython-310/build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.o -MMD -MF build/temp.linux-x86_64-cpython-310/build/src.linux-x86_64-3.10/numpy/core/src/umath/loops.o.d -msse -msse2 -msse3" failed with exit status 1
INFO: 
########### EXT COMPILER OPTIMIZATION ###########
INFO: Platform      : 
  Architecture: x64
  Compiler    : gcc

CPU baseline  : 
  Requested   : 'min'
  Enabled     : SSE SSE2 SSE3
  Flags       : -msse -msse2 -msse3
  Extra checks: none

CPU dispatch  : 
  Requested   : 'max -xop -fma4'
  Enabled     : SSSE3 SSE41 POPCNT SSE42 AVX F16C FMA3
  Generated   : 
              : 
  SSE41       : SSE SSE2 SSE3 SSSE3
  Flags       : -msse -msse2 -msse3 -mssse3 -msse4.1
  Extra checks: none
  Detect      : SSE SSE2 SSE3 SSSE3 SSE41
              : build/src.linux-x86_64-3.10/numpy/core/src/umath/loops_arithmetic.dispatch.c
              : build/src.linux-x86_64-3.10/numpy/core/src/umath/loops_unary_fp.dispatch.c
INFO: CCompilerOpt.cache_flush[825] : write cache to path -> /home/vagrant/src/numpy/build/temp.linux-x86_64-cpython-310/ccompiler_opt_cache_ext.py
INFO: 
########### CLIB COMPILER OPTIMIZATION ###########
INFO: Platform      : 
  Architecture: x64
  Compiler    : gcc

CPU baseline  : 
  Requested   : 'min'
  Enabled     : SSE SSE2 SSE3
  Flags       : -msse -msse2 -msse3
  Extra checks: none

CPU dispatch  : 
  Requested   : 'max -xop -fma4'
  Enabled     : SSSE3 SSE41 POPCNT SSE42 AVX F16C FMA3
  Generated   : none
INFO: CCompilerOpt.cache_flush[825] : write cache to path -> /home/vagrant/src/numpy/build/temp.linux-x86_64-cpython-310/ccompiler_opt_cache_clib.py

【问题讨论】：

  • 当您只请求 SSE（>15 岁）时，您似乎遇到了一些与 AVX-512 相关的问题。如果您在最新的 Intel 处理器（例如 IceLake/CannonLake 或 Skylake SP）上运行，那么启用 AVX-512 怎么样？顺便说一句，请检查 GCC 的默认版本是不是 4.4.7（很旧）。另外，您的确切目标平台是什么？
  • 感谢您的帮助。我的开发环境的处理器是 Intel(R) Core(TM) i7-8700 CPU @ 3.20GHz。 （我通过/proc/cpuinfo查看）默认gcc版本是4.4.7...如何确认目标平台...？
  • 好的，所以您的处理器不支持 AVX-512（但它支持 AVX-2，因此最好使用 -mavx2 编译标志添加它）。这意味着某些东西是在 AVX-512 的支持下构建的，然后 GCC 汇编器会抱怨，因为它是意外的（编译器生成了 AVX-512 汇编指令）。假设问题不是来自 Numpy（现在 1.22.4 应该很稳定），可能是由于使用 AVX-512 编译的依赖库（例如 glibc、libm 等），一些坏的旧文件编译为AVX-512 仍在缓存中，等等。很难说更多。
  • 好的，非常感谢！我会特别研究与 AVX512 相关的内容。

标签： python c++ c linux numpy


【解决方案1】：

看来I need at least gcc 11.2，对于-O3，然后我通过Binutils 2.38包升级了GNU汇编器。 我可以通过升级解决我的问题。

【讨论】：

    猜你喜欢
    • 2011-08-12
    • 2021-11-30
    • 2015-06-01
    • 1970-01-01
    • 2022-06-30
    • 1970-01-01
    • 2014-03-29
    • 1970-01-01
    相关资源
    最近更新 更多
    热门标签
    Java Python linux javascript Mysql C# Docker 算法 前端 SpringBoot Redis Vue spring 设计模式 .net core .net kubernetes c++ 数据库 数据结构 大数据 js 机器学习 微服务 Android Go 程序员 面试 JVM ASP.net core 云原生 人工智能 后端 PHP git CSS golang k8s Nginx Django mybatis 深度学习 多线程 React 架构 devops 爬虫 云计算 Spring Boot LeetCode