test/v4.asm

   1         bits 64
   2         v4fmaddps zmm0,zmm1+3,[rax]
   3         v4fnmaddps zmm2,zmm3,[rax]
   4         v4fmaddss zmm4,zmm5+3,[rax]
   5         v4fnmaddss zmm6,zmm7+3,[rax]
   6
   7         v4dpwssds zmm8,zmm9,[rax]
   8         v4dpwssd zmm10,zmm11+3,[rax]
   9         v4dpwssd zmm10+0,zmm11+3,[rax]
  10 ;       v4dpwssd zmm10+1,zmm11+3,[rax]
  11 ;       v4dpwssd zmm10,zmm11+4,[rax]
  12 ;       v4dpwssd zmm10,zmm11+7,[rax]