GCC为下面的shuffle()生成此代码:
movaps xmm0,XMMWORD PTR [rip+0x125]
pshufb xmm4,xmm0
Run Code Online (Sandbox Code Playgroud)
理想情况下,这应该是:
pshufb xmm4,XMMWORD PTR [rip+0x125]
Run Code Online (Sandbox Code Playgroud)
生成此单指令的扩展ASM语法是什么?
非常感谢,亚当
PS:注释掉的内在函数为此示例生成最佳代码.这通常不起作用(GCC很可能在存在全局寄存器变量时生成不必要的寄存器副本).
#include <stdint.h>
typedef int8_t xmm_t __attribute__ ((vector_size (16)));
const xmm_t xmm_shuf={128, 0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14, 15};
register xmm_t xmm __asm__("xmm4");
#define NTL ".intel_syntax noprefix\n"
#define ATT ".att_syntax\n"
void shuffle() {
//xmm=__builtin_ia32_pshufb128(xmm, xmm_shuf);
__asm__(NTL"pshufb %0, %1\n"ATT : "=x" (xmm) : "x" (xmm_shuf));
}
int main() {
}
Run Code Online (Sandbox Code Playgroud)
$ gcc -Os -std = gnu99 -msse4.1 -flax-vector-conversions pshufb_128bit_constant.c …