...ero 9 stall ;delayed branch slot 每次循环都有数据相关/控制相关引起的停顿 每次循环需要6 clocks Unroll loop (循环展开) 4 times code to make faster?
基于12个网页-相关网页
A standard optimization is loop unrolling; the compiler will unroll the loops to reduce the number of branches.
标准的优化是循环展开;编译器将展开循环,以减少分支的数量。
I used one unsupported list comprehension in my code, which I had to unroll into a regular for loop.
在代码中,我使用了一个不支持的列表,将其放入一个普通的for循环。
The article discuss the advantage and meaning integrating software pipelining and loop unroll in IA - 64 compiler, moreover do some deep research by experiment result.
论述了在IA-64编译器中结合使用软件流水和循环展开的优点和意义,并结合实验进行了深度探讨。
应用推荐