新建 hello.cpp 文件:

#include <omp.h>
#include <stdio.h>
int main() {
#pragma omp parallel
printf("Hello from thread %d, nthreads %d\n", omp_get_thread_num(), omp_get_num_threads());
return 0;
}

编译会遇到如下错误:

hello.cpp:1:10: fatal error: 'omp.h' file not found
#include <omp.h>
^~~~~~
1 error generated.

因为默认的 g++ 编译器不支持 openmp,我们可以设置 LLVM/Clang 编译器来编译 openmp。

执行以下命令:

brew install llvm # 安装 LLVM 编译器
brew install libomp # 安装 OpenMP 库
echo 'export PATH="/usr/local/opt/llvm/bin:$PATH"' >> ~/.bash_profile # 将 llvm 的可执行文件添加到 PATH 目录

然后执行

clang -fopenmp hello.cpp -o hello
./hello

我的CPP文件中用到了STL 中的 vector,然后就遇到了新的错误:

Undefined symbols for architecture x86_64:
"std::__1::__vector_base_common<true>::__throw_length_error() const", referenced from:
std::__1::vector<int, std::__1::allocator<int> >::assign(unsigned long, int const&) in seq-3efcea.o
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
void std::__1::vector<int, std::__1::allocator<int> >::__push_back_slow_path<int>(int&&) in seq-3efcea.o
"std::logic_error::logic_error(char const*)", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
"std::length_error::~length_error()", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
"std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::__init(char const*, unsigned long)", referenced from:
_main in seq-3efcea.o
"std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::insert(unsigned long, char const*)", referenced from:
_main in seq-3efcea.o
"std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::~basic_string()", referenced from:
_main in seq-3efcea.o
"std::terminate()", referenced from:
___clang_call_terminate in seq-3efcea.o
"typeinfo for std::length_error", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
"vtable for std::length_error", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
NOTE: a missing vtable usually means the first non-inline virtual member function has no definition.
"operator delete(void*)", referenced from:
std::__1::__vector_base<int, std::__1::allocator<int> >::~__vector_base() in seq-3efcea.o
std::__1::__deque_base<int, std::__1::allocator<int> >::~__deque_base() in seq-3efcea.o
std::__1::__deque_base<int, std::__1::allocator<int> >::clear() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*> >::~__split_buffer() in seq-3efcea.o
std::__1::__vector_base<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::~__vector_base() in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::~__split_buffer() in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
...
"operator new(unsigned long)", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
"___cxa_allocate_exception", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
"___cxa_begin_catch", referenced from:
___clang_call_terminate in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*> >::shrink_to_fit() in seq-3efcea.o
"___cxa_end_catch", referenced from:
std::__1::__split_buffer<int*, std::__1::allocator<int*> >::shrink_to_fit() in seq-3efcea.o
"___cxa_free_exception", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
"___cxa_throw", referenced from:
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::__split_buffer<int, std::__1::allocator<int>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int>&) in seq-3efcea.o
std::__1::deque<int, std::__1::allocator<int> >::__add_back_capacity() in seq-3efcea.o
std::__1::__split_buffer<int*, std::__1::allocator<int*>&>::__split_buffer(unsigned long, unsigned long, std::__1::allocator<int*>&) in seq-3efcea.o
"___gxx_personality_v0", referenced from:
_main in seq-3efcea.o
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::vector(unsigned long, std::__1::vector<int, std::__1::allocator<int> > const&) in seq-3efcea.o
std::__1::vector<std::__1::vector<int, std::__1::allocator<int> >, std::__1::allocator<std::__1::vector<int, std::__1::allocator<int> > > >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::vector(std::__1::vector<int, std::__1::allocator<int> > const&) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::allocate(unsigned long) in seq-3efcea.o
std::__1::vector<int, std::__1::allocator<int> >::vector(unsigned long, int const&) in seq-3efcea.o
void std::__1::vector<int, std::__1::allocator<int> >::__push_back_slow_path<int>(int&&) in seq-3efcea.o
...
ld: symbol(s) not found for architecture x86_64
clang-6.0: error: linker command failed with exit code 1 (use -v to see invocation)

是因为我们编译的是 C++文件,clang 没有链接 STL 库所以出错了,我们可以显式地链接标准库:

clang -fopenmp hello.cpp -o hello -lstdc++
#或者 clang -fopenmp hello.cpp -o hello -lc++
./hello

也可以直接用 clang++:

clang++ -fopenmp hello.cpp -o hello
./hello

Mac 使用 OpenMP/Clang的更多相关文章

  1. 尝试在Mac/iOS上使用tcmalloc库

    概述        TCMalloc 是 Google 开发的内存分配器,在不少项目中都有使用,例如在 Golang 中就使用了类似的算法进行内存分配.它具有现代化内存分配器的基本特征:对抗内存碎片. ...

  2. R语言︱XGBoost极端梯度上升以及forecastxgb(预测)+xgboost(回归)双案例解读

    XGBoost不仅仅可以用来做分类还可以做时间序列方面的预测,而且已经有人做的很好,可以见最后的案例. 应用一:XGBoost用来做预测 ------------------------------- ...

  3. xgboost: 速度快效果好的boosting模型

    转自:http://cos.name/2015/03/xgboost/ 本文作者:何通,SupStat Inc(总部在纽约,中国分部为北京数博思达信息科技有限公司)数据科学家,加拿大Simon Fra ...

  4. libc++

    今天测试最新的微信iOS SDK, 仅仅是建了一个空的工程,把sdk加进去运行,就报了以下错误: Undefined symbols for architecture x86_64: "op ...

  5. atomic, spinlock and mutex性能比较

    我非常好奇于不同同步原理的性能,于是对atomic, spinlock和mutex做了如下实验来比较: 1. 无同步的情况 #include <future> #include <i ...

  6. [BS] 小知识点总结-02

    1.  dispatch_GCD 可让某操作延迟x秒执行 //模拟网速慢,延迟3s返回数据(就会导致右侧数据和左侧标签不对应) dispatch_after(dispatch_time(DISPATC ...

  7. 李洪强iOS开发之【零基础学习iOS开发】【02-C语言】02-第一个C语言程序

    前言 前面已经唠叨了这么多理论知识,从这讲开始,就要通过接触代码来学习C语言的语法.学习任何一门语言,首先要掌握的肯定是语法.学习C语言语法的目的:就是能够利用C语言编写程序,然后运行程序跟硬件(计算 ...

  8. 学习C++之前要先学习C语言吗?

    C++ 读作“C加加”,是“C Plus Plus”的简称.顾名思义,C++ 是在C语言的基础上增加新特性,玩出了新花样,所以叫“C Plus Plus”,就像 iPhone 7S 和 iPhone ...

  9. 关于mac安装rails报错clang: error: unknown argument

    文章都是从我的个人博客上转载过来的,大家可以点击我的个人博客. www.iwangzheng.com mac上安装rails的时候报错, 安装rails的在终端执行了一句命令: $sudo gem i ...

随机推荐

  1. [转]Python in Visual Studio Code

    本文转自:https://code.visualstudio.com/docs/languages/python Working with Python in Visual Studio Code, ...

  2. Java多线程知识整理

    多线程 1. 多线程基础 多线程状态转换图 普通方法介绍 yeild yeild,线程让步.是当前线程执行完后所有线程又统一回到同一起跑线.让自己或者其他线程运行,并不是单纯的让给其他线程. join ...

  3. Yii2设计模式——注册树模式

    应用举例 在Yii.php中: <?php class ServiceLocator extends Component { //保存实例化的对象,每个对象都是单例,且有唯一string类型的I ...

  4. AJAX的创建

    20:29:50 创建的步骤: 1.创建异步请求的核心对象 2.设置请求方式和地址 3.设置结果产生的回调函数 4.进行结果的逻辑处理 5.获取结果并处理 6.发送请求 <!DOCTYPE ht ...

  5. video.js 一个页面同时播放多个视频

    $(data).each(function(i, item) { // innerHTML += '<li type-id="'+item.id+'">'+ // '& ...

  6. Windows Azure NotificationHub+Firebase Cloud Message 实现消息推动(付源码)

    前期项目一直用的是Windows azure NotificationHub+Google Cloud Message 实现消息推送, 但是GCM google已经不再推荐使用,慢慢就不再维护了, 现 ...

  7. 逐个使用C++11新特性

    C++11 auto & decltype auto:根据变量初始值来推导变量类型,并用右值初始化变量. decltype:从表达式推导出类型,并将变量定义为该类型,但不用表达式的值初始化该变 ...

  8. 你不可不知的Java引用类型之——Reference源码解析

    定义 Reference是所有引用类型的父类,定义了引用的公共行为和操作. reference指代引用对象本身,referent指代reference引用的对象,下文介绍会以reference,ref ...

  9. Numpy库的学习(三)

    今天我们继续学习一下Numpy库的学习 废话不多说 ,开始讲 比如我们现在想创建一个0-14这样一个15位的数组 可以直接写,但是很麻烦,Numpy中就给我们了一个方便创建的方法 numpy中有一个a ...

  10. PMM Client 安装异常报错

    1.PMM架构 如下图所示 2.Client主要组件 PMM Client是安装在你要监视的MySQL或MongoDB主机上的一组代理组件.组件收集关于一般系统和数据库性能的各种数据,并将该数据发送到 ...