void SSE_MatrixF_x_MatrixF(const F32 *matA, const F32 *matB, F32 *result)
{
   asm
   (
      "movss      (%%edx),%%xmm0\n"
      "movups     (%%ecx),%%xmm1\n"
      "shufps     $0,%%xmm0,%%xmm0\n"      
      "movss      4(%%edx),%%xmm2\n"
      "mulps      %%xmm1,%%xmm0\n"
      "shufps     $0,%%xmm2,%%xmm2\n"
      "movups     0x10(%%ecx),%%xmm3\n"
      "movss      8(%%edx),%%xmm7\n"
      "mulps      %%xmm3,%%xmm2\n"
      "shufps     $0,%%xmm7,%%xmm7\n"
      "addps      %%xmm2,%%xmm0\n"
      "movups     0x20(%%ecx),%%xmm4\n"
      "movss      0x0c(%%edx),%%xmm2\n"
      "mulps      %%xmm4,%%xmm7\n"
      "shufps     $0,%%xmm2,%%xmm2\n"
      "addps      %%xmm7,%%xmm0\n"
      "movups     0x30(%%ecx),%%xmm5\n"
      "movss      0x10(%%edx),%%xmm6\n"
      "mulps      %%xmm5,%%xmm2\n"
      "movss      0x14(%%edx),%%xmm7\n"
      "shufps     $0,%%xmm6,%%xmm6\n"
      "addps      %%xmm2,%%xmm0\n"
      "shufps     $0,%%xmm7,%%xmm7\n"
      "movlps     %%xmm0,(%%eax)\n"
      "movhps     %%xmm0,8(%%eax)\n"
      "mulps      %%xmm3,%%xmm7\n"
      "movss      0x18(%%edx),%%xmm0\n"
      "mulps      %%xmm1,%%xmm6\n"
      "shufps     $0,%%xmm0,%%xmm0\n"
      "addps      %%xmm7,%%xmm6\n"
      "mulps      %%xmm4,%%xmm0\n"
      "movss      0x24(%%edx),%%xmm2\n"
      "addps      %%xmm0,%%xmm6\n"
      "movss      0x1c(%%edx),%%xmm0\n"
      "movss      0x20(%%edx),%%xmm7\n"
      "shufps     $0,%%xmm0,%%xmm0\n"
      "shufps     $0,%%xmm7,%%xmm7\n"
      "mulps      %%xmm5,%%xmm0\n"
      "mulps      %%xmm1,%%xmm7\n"
      "addps      %%xmm0,%%xmm6\n"
      "shufps     $0,%%xmm2,%%xmm2\n"
      "movlps     %%xmm6,0x10(%%eax)\n"
      "movhps     %%xmm6,0x18(%%eax)\n"
      "mulps      %%xmm3,%%xmm2\n"
      "movss      0x28(%%edx),%%xmm6\n"
      "addps      %%xmm2,%%xmm7\n"
      "shufps     $0,%%xmm6,%%xmm6\n"
      "movss      0x2c(%%edx),%%xmm2\n"
      "mulps      %%xmm4,%%xmm6\n"
      "shufps     $0,%%xmm2,%%xmm2\n"
      "addps      %%xmm6,%%xmm7\n"
      "mulps      %%xmm5,%%xmm2\n"
      "movss      0x34(%%edx),%%xmm0\n"
      "addps      %%xmm2,%%xmm7\n"
      "shufps     $0,%%xmm0,%%xmm0\n"
      "movlps     %%xmm7,0x20(%%eax)\n"
      "movss      0x30(%%edx),%%xmm2\n"
      "movhps     %%xmm7,0x28(%%eax)\n"
      "mulps      %%xmm3,%%xmm0\n"
      "shufps     $0,%%xmm2,%%xmm2\n"
      "movss      0x38(%%edx),%%xmm6\n"
      "mulps      %%xmm1,%%xmm2\n"
      "shufps     $0,%%xmm6,%%xmm6\n"
      "addps      %%xmm0,%%xmm2\n"
      "mulps      %%xmm4,%%xmm6\n"
      "movss      0x3c(%%edx),%%xmm7\n"
      "shufps     $0,%%xmm7,%%xmm7\n"
      "addps      %%xmm6,%%xmm2\n"
      "mulps      %%xmm5,%%xmm7\n"
      "addps      %%xmm7,%%xmm2\n"
      "movups     %%xmm2,0x30(%%eax)\n"
      
      :
      : "d" ( matA ),
        "c" ( matB ),
        "a" ( result )
   );
}