#include <stdio.h>

int main() {
  char text[] = "go fast!";
  char subMe[] = { 32, 32, 0, 32, 32, 32, 32, 0 };
  printf("%s \n", text);

  __asm {
    movq mm0, text;//move 64 bits into mm0 register
    psubsb mm0, subMe;//substract content of "text" and "subMe" in parallel
    movq text, mm0
      emms;// emms = Empty MMX Technology State (reactivate FPU registers)
  }

  printf("%s \n", text);
  return 0;
}