86     movdqa    xmm1, xmm0  in TransposeWx8_SSSE3()
91     movdqa    xmm3, xmm2  in TransposeWx8_SSSE3()
97     movdqa    xmm5, xmm4  in TransposeWx8_SSSE3()
103     movdqa    xmm7, xmm6  in TransposeWx8_SSSE3()
108     movdqa    xmm2, xmm0  in TransposeWx8_SSSE3()
109     movdqa    xmm3, xmm1  in TransposeWx8_SSSE3()
114     movdqa    xmm6, xmm4  in TransposeWx8_SSSE3()
115     movdqa    xmm7, xmm5  in TransposeWx8_SSSE3()
122     movdqa    xmm4, xmm0  in TransposeWx8_SSSE3()
127     movdqa    xmm6, xmm2  in TransposeWx8_SSSE3()
133     movdqa    xmm5, xmm1  in TransposeWx8_SSSE3()
140     movdqa    xmm7, xmm3  in TransposeWx8_SSSE3()
181     movdqa    xmm0, [eax]  in TransposeUVWx8_SSE2()
182     movdqa    xmm1, [eax + edi]  in TransposeUVWx8_SSE2()
184     movdqa    xmm7, xmm0  // use xmm7 as temp register.  in TransposeUVWx8_SSE2()
187     movdqa    xmm1, xmm7  in TransposeUVWx8_SSE2()
188     movdqa    xmm2, [eax]  in TransposeUVWx8_SSE2()
189     movdqa    xmm3, [eax + edi]  in TransposeUVWx8_SSE2()
191     movdqa    xmm7, xmm2  in TransposeUVWx8_SSE2()
194     movdqa    xmm3, xmm7  in TransposeUVWx8_SSE2()
195     movdqa    xmm4, [eax]  in TransposeUVWx8_SSE2()
196     movdqa    xmm5, [eax + edi]  in TransposeUVWx8_SSE2()
198     movdqa    xmm7, xmm4  in TransposeUVWx8_SSE2()
201     movdqa    xmm5, xmm7  in TransposeUVWx8_SSE2()
202     movdqa    xmm6, [eax]  in TransposeUVWx8_SSE2()
203     movdqa    xmm7, [eax + edi]  in TransposeUVWx8_SSE2()
205     movdqa    [esp], xmm5  // backup xmm5  in TransposeUVWx8_SSE2()
207     movdqa    xmm5, xmm6   // use xmm5 as temp register.  in TransposeUVWx8_SSE2()
210     movdqa    xmm7, xmm5  in TransposeUVWx8_SSE2()
214     movdqa    xmm5, xmm0  in TransposeUVWx8_SSE2()
217     movdqa    xmm2, xmm5  in TransposeUVWx8_SSE2()
218     movdqa    xmm5, xmm1  in TransposeUVWx8_SSE2()
221     movdqa    xmm3, xmm5  in TransposeUVWx8_SSE2()
222     movdqa    xmm5, xmm4  in TransposeUVWx8_SSE2()
225     movdqa    xmm6, xmm5  in TransposeUVWx8_SSE2()
226     movdqa    xmm5, [esp]  // restore xmm5  in TransposeUVWx8_SSE2()
227     movdqa    [esp], xmm6  // backup xmm6  in TransposeUVWx8_SSE2()
228     movdqa    xmm6, xmm5    // use xmm6 as temp register.  in TransposeUVWx8_SSE2()
231     movdqa    xmm7, xmm6  in TransposeUVWx8_SSE2()
234     movdqa    xmm6, xmm0  in TransposeUVWx8_SSE2()
237     movdqa    xmm4, xmm6  in TransposeUVWx8_SSE2()
238     movdqa    xmm6, [esp]  // restore xmm6  in TransposeUVWx8_SSE2()
245     movdqa    xmm0, xmm2   // use xmm0 as the temp register.  in TransposeUVWx8_SSE2()
254     movdqa    xmm0, xmm1   // use xmm0 as the temp register.  in TransposeUVWx8_SSE2()
263     movdqa    xmm0, xmm3   // use xmm0 as the temp register.  in TransposeUVWx8_SSE2()