要返回std :: move(x)或不? [英] To return std::move (x) or not?

查看:152
本文介绍了要返回std :: move(x)或不?的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!

问题描述

  std :: vector< double> foo()
{
std :: vector< double> t;
...

return t;
}

  std :: vector< double> foo()
{
std :: vector< double> t;
...

return std :: move(t);
}

当量?



更准确地说, return x 总是等于 return std :: move(x)

解决方案

它们不是等效的,你应该总是使用 return t; p>

更长的版本是,当且仅当return语句有资格进行返回值优化时,returnee绑定到rvalue引用(或者俗称为 move 是隐式的)。



通过拼写出来 return std :: move(t); ,但是,你实际上禁止返回值优化!


Are

std::vector<double> foo ()
{
    std::vector<double> t;
    ...

    return t;
}

and

std::vector<double> foo ()
{
    std::vector<double> t;
    ...

    return std::move (t);
}

equivalent ?

More precisely, is return x always equivalent to return std::move (x) ?

解决方案

They're not equivalent, and you should always use return t;.

The longer version is that if and only if a return statement is eligible for return value optimization, then the returnee binds to rvalue reference (or colloquially, "the move is implicit").

By spelling out return std::move(t);, however, you actually inhibit return value optimization!

这篇关于要返回std :: move(x)或不?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!

查看全文
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆