meton-robean

### XLOOP 几个例子 

### 带有循环指示变量： ``` for(int i = 0; i < 8; ++i) { a[i] = b[i]+c[i]; tmp += 3*i+1; //j++; printf("a: %d, tmp: %d j: %d \n", a[4], tmp, j );...

``` for(int i = 0; i < 128; ++i) { A[i] += 2; } ``` ``` 80001048: 7101 addi sp,sp,-512 8000104a: 878a mv a5,sp 8000104c: 0414 addi a3,sp,512 8000104e: 4398...

``` for(int i = 0; i < 128; ++i ){ B[i] = a + A[i]; a = C[i] + c; } ``` ``` 10194: 4394 lw a3,0(a5) 10196: 9f35 addw...

``` for(int i=0; i

``` for(int i=0; i

### dynamic programing (这个例子来自 xloop 的[dynprog](https://github.com/cornell-brg/xloops-bmarks/tree/master/dynprog)) ``` __attribute__((noinline)) void dynprog_scalar_longlong( int length, long long *c, long long *W, long long *out ) { int out_l = 0; int i, j,...

### warshall computing 例子来自xloop的benchmark ``` __attribute__((noinline)) void warshall_scalar( int n, float *path, float *path_in ) { int i, j, k; // initially copy the input path to path memcpy( path,...

为什么在平均情况下快速排序比堆排序要优秀？ - 陈运文的回答 - 知乎 https://www.zhihu.com/question/23873747/answer/281183029

大O标记法展示的是算法时间随着数据规模变化的大趋势，实际运行时间还要看内存带宽限制，cache局部性原理，指令效率，数据分布情况，以及一些参数项对运行时间的影响