为什么uint32_t与uint64_t的速度不同？

Question

Trying to understand something about how g++/cpu processes integers at runtime. 试图了解有关g ++ / cpu如何在运行时处理整数的信息。

I'm measuring how long the following function takes to run: 我正在测量以下功能需要运行多长时间：

template<class T>
void speedTest() {
    for(T d=0;d<4294967295u;d++)int number;
}

This simple method will run a dumb loop the max value of uint32_t many times 这个简单的方法将多次运行一个uint32_t最大值的哑循环

and when I call: 当我打电话时：

speedTest<uint32_t>();

the software takes an average of 8.15 seconds but when I call: 该软件平均需要8.15秒，但在我致电时：

speedTest<uint64_t>();

the software takes an average of 10.35 seconds. 该软件平均需要10.35秒。

Why is this happening? 为什么会这样呢？

Answer 1

Some possible reasons: 一些可能的原因：

Larger data types require more memory bandwidth in general 较大的数据类型通常需要更多的内存带宽
Even if that loop counter is kept inside a register, the CPU is probably taking more time to do calculations with large values, especially, if it needs multiple registers (eg if your CPU has just 32bit wide registers) 即使该循环计数器保留在寄存器中，CPU可能仍会花费更多时间来进行大数值的计算，尤其是当它需要多个寄存器时（例如，如果您的CPU只有32位宽的寄存器）
The compiler would need to emit extra machine instructions to emulate any type not directly supported by the CPU 编译器将需要发出额外的机器指令来模拟CPU不直接支持的任何类型
It also depends on optimization. 它还取决于优化。 Such a loop without side effects could be optimized out completely, regardless of int number; 不管int number; ，都可以完全优化掉这种无副作用的循环int number; (could just be for(T d=0;d<4294967295u;d++); ) （可能只是for(T d=0;d<4294967295u;d++); ）

You could continue your investigation/exercise by providing some assembly. 您可以通过提供一些程序集来继续进行调查/练习。