问题描述
在我正在开发的一个项目中，我正试图优化一个大文件中的四重嵌套for循环，我认为这将从编译器展开到-funroll-all-loops中受益。但是，当我将此标志添加到编译器时，它会将其他循环展开到文件的其余部分，并使整个程序运行速度更慢。有没有一种方法（可能通过#pragma）将编译器标志仅应用于文件中的某些函数而不是整个文件？
In a project I'm working on, there's a quadruple-nested for loop in a large file I'm trying to optimize I think would benefit from a compiler unroll with -funroll-all-loops. However, when I add this flag to the compiler, it unrolls the other loops the rest of the file and makes the overall program run more slowly. Is there a way (possibly via a #pragma) to apply compiler flags only to certain functions in the file instead of the entire file?
预先感谢。
推荐答案
GCC optimize 可用于为单个函数设置优化选项：
The GCC function attribute optimize can be used to set an optimization option for a single function:
void foo(int bar) __attribute__((optimize ("unroll-all-loops")))
{
}
这篇关于GCC - 仅在特定功能上启用编译器标志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！