55     movq      xmm5, qword ptr [eax + edi]  in TransposeWx8_SSSE3()
56     punpcklbw xmm4, xmm5  in TransposeWx8_SSSE3()
58     movdqa    xmm5, xmm4  in TransposeWx8_SSSE3()
60     palignr   xmm5, xmm5, 8  in TransposeWx8_SSSE3()
74     punpcklwd xmm5, xmm7  in TransposeWx8_SSSE3()
76     movdqa    xmm7, xmm5  in TransposeWx8_SSSE3()
91     punpckldq xmm1, xmm5  in TransposeWx8_SSSE3()
94     movdqa    xmm5, xmm1  in TransposeWx8_SSSE3()
96     palignr   xmm5, xmm5, 8  in TransposeWx8_SSSE3()
98     movq      qword ptr [edx + esi], xmm5  in TransposeWx8_SSSE3()
158     movdqu    xmm5, [eax + edi]  in TransposeUVWx8_SSE2()
161     punpcklbw xmm4, xmm5  in TransposeUVWx8_SSE2()
162     punpckhbw xmm7, xmm5  in TransposeUVWx8_SSE2()
163     movdqa    xmm5, xmm7  in TransposeUVWx8_SSE2()
167     movdqu    [esp], xmm5  // backup xmm5  in TransposeUVWx8_SSE2()
169     movdqa    xmm5, xmm6  // use xmm5 as temp register.  in TransposeUVWx8_SSE2()
171     punpckhbw xmm5, xmm7  in TransposeUVWx8_SSE2()
172     movdqa    xmm7, xmm5  in TransposeUVWx8_SSE2()
176     movdqa    xmm5, xmm0  in TransposeUVWx8_SSE2()
178     punpckhwd xmm5, xmm2  in TransposeUVWx8_SSE2()
179     movdqa    xmm2, xmm5  in TransposeUVWx8_SSE2()
180     movdqa    xmm5, xmm1  in TransposeUVWx8_SSE2()
182     punpckhwd xmm5, xmm3  in TransposeUVWx8_SSE2()
183     movdqa    xmm3, xmm5  in TransposeUVWx8_SSE2()
184     movdqa    xmm5, xmm4  in TransposeUVWx8_SSE2()
186     punpckhwd xmm5, xmm6  in TransposeUVWx8_SSE2()
187     movdqa    xmm6, xmm5  in TransposeUVWx8_SSE2()
188     movdqu    xmm5, [esp]  // restore xmm5  in TransposeUVWx8_SSE2()
190     movdqa    xmm6, xmm5  // use xmm6 as temp register.  in TransposeUVWx8_SSE2()
191     punpcklwd xmm5, xmm7  in TransposeUVWx8_SSE2()
218     punpckldq xmm1, xmm5  in TransposeUVWx8_SSE2()
221     punpckhdq xmm0, xmm5  in TransposeUVWx8_SSE2()