+#if defined(USE_SSE) && defined(ARCH_X86_64)
+ if ( !luma && width > 7 )
+ {
+ composite_line_yuv_sse2_simple(dest, src, width, alpha_b, alpha_a, weight);
+ j = width - width % 8;
+ dest += j * 2;
+ src += j * 2;
+ alpha_a += j;
+ alpha_b += j;
+ }
+#endif
+
+ for ( ; j < width; j ++ )