最近遇到了一个C语言中使用矢量运算的问题,栽了不少跟头,这里就做个总结,免得后面再犯类似错误.
该数据结构的定义如下所示:
typedef int v4si __attribute__ ((vector_size (16)))
基本的操作:
typedef int v4si __attribute__ ((vector_size (16))); v4si a, b, c; long l; a = b + 1; /* a = b + {1,1,1,1}; */ a = 2 * b; /* a = {2,2,2,2} * b; */ a = l + a; /* Error, cannot convert long to int. */
特殊的运算:
typedef int v4si __attribute__ ((vector_size (16))); v4si a = {1,2,3,4}; v4si b = {3,2,1,4}; v4si c; c = a > b; /* The result would be {0, 0,-1, 0} */ c = a == b; /* The result would be {0,-1, 0,-1} */
使用内敛函数的运算结果:
typedef int v4si __attribute__ ((vector_size (16))); v4si a = {1,2,3,4}; v4si b = {5,6,7,8}; v4si mask1 = {0,1,1,3}; v4si mask2 = {0,4,2,5}; v4si res; res = __builtin_shuffle (a, mask1); /* res is {1,2,2,4} */ res = __builtin_shuffle (a, b, mask2); /* res is {1,5,3,6} */
一个例子:
#includetypedef int v4si __attribute__ ((vector_size (16))); int main() { v4si a = {1,2,3,4}; v4si b = {5,6,7,8}; v4si mask1 = {0,1,1,3}; v4si mask2 = {0,4,2,5}; v4si res; v4si res1; res = __builtin_shuffle (a, mask1); /* res is {1,2,2,4} */ res1 = __builtin_shuffle (a, b, mask2); /* res is {1,5,3,6} */ for(int cnt =0; cnt < 4; cnt++) { printf("%d \n",res[cnt]); printf("%d \n",res1[cnt]); } return 0; } ~
运算结果:
1 1 2 5 2 3 4 6
其实,详细的针对这个概念的解释请参考后面参考文档,我只是负责知道这个东西怎么用了.
参考文档:
1 https://gcc.gnu.org/onlinedocs/gcc/Vector-Extensions.html
2 https://gcc.gnu.org/onlinedocs/gcc-4.0.0/gcc/Vector-Extensions.html