Lines Matching refs:mm1
440 movq mm1,mask1
443 pand mm1,mm7
446 pcmpeqb mm1,mm6
464 pand mm5,mm1
465 movq mm7,mm1
559 movq mm1,mask1
563 pand mm1,mm7
567 pcmpeqb mm1,mm6
587 pand mm5,mm1
588 movq mm7,mm1
695 movq mm1,mask1
700 pand mm1,mm7
705 pcmpeqb mm1,mm6
726 pand mm5,mm1
727 movq mm7,mm1
840 movq mm1,mask1
847 pand mm1,mm7
854 pcmpeqb mm1,mm6
876 pand mm6,mm1
877 movq mm7,mm1
1237 movq mm1, mm0 ; 0 0 0 0 0 v2 v1 v0
1241 psrlq mm1, 8 ; 0 0 0 0 0 0 v2 v1
1243 por mm0, mm1 ; v2 v1 v0 v2 v1 v0 v2 v1
1272 movq mm1, mm0 ; 0 0 0 0 0 v2 v1 v0
1276 psrlq mm1, 8 ; 0 0 0 0 0 0 v2 v1
1278 por mm0, mm1 ; v2 v1 v0 v2 v1 v0 v2 v1
1362 movq mm1, mm0 ; X X X X v0 v1 v2 v3
1430 movq mm1, mm0 ; v0 v0 v1 v1 v2 v2 v3 v3
1432 punpckhwd mm1, mm1 ; v0 v0 v0 v0 v1 v1 v1 v1
1435 movq [edi+8], mm1 ; move to memory v1 and v0
1471 movq mm1, mm0 ; v0 v1 v2 v3 v4 v5 v6 v7
1473 //movq mm1, mm0 ; v0 v0 v1 v1 v2 v2 v3 v3
1474 punpckhbw mm1, mm1 ;v0 v0 v1 v1 v2 v2 v3 v3
1475 movq [edi+8], mm1 ; move to memory v0 v1 v2 and v3
1519 movq mm1, mm0 ; v1 v0 v1 v0 v3 v2 v3 v2
1521 punpckhdq mm1, mm1 ; v1 v0 v1 v0 v1 v0 v1 v0
1524 movq [edi + 16], mm1
1525 movq [edi + 24], mm1
1565 movq mm1, mm0 ; v1 v0 v1 v0 v3 v2 v3 v2
1567 punpckhdq mm1, mm1 ; v1 v0 v1 v0 v1 v0 v1 v0
1570 movq [edi + 8], mm1
1653 movq mm1, mm0 ; v3 v2 v1 v0 v7 v6 v5 v4
1655 punpckhdq mm1, mm1 ; v3 v2 v1 v0 v3 v2 v1 v0
1660 movq [edi+32], mm1
1661 movq [edi + 40], mm1
1662 movq [edi+ 48], mm1
1664 movq [edi + 56], mm1
1702 movq mm1, mm0 ; v3 v2 v1 v0 v7 v6 v5 v4
1704 punpckhdq mm1, mm1 ; v3 v2 v1 v0 v3 v2 v1 v0
1707 movq [edi+16], mm1
1708 movq [edi + 24], mm1
1747 movq mm1, mm0 ; v3 v2 v1 v0 v7 v6 v5 v4
1749 punpckhdq mm1, mm1 ; v3 v2 v1 v0 v3 v2 v1 v0
1752 movq [edi + 8], mm1
2014 movq mm1, [esi + ebx] // Load mm1 with Prior(x)
2016 pand mm3, mm1 // get lsb for each prev_row byte
2017 psrlq mm1, 1 // divide prev_row bytes by 2
2018 pand mm1, mm4 // clear invalid bit 7 of each byte
2019 paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte
2021 movq mm1, mm3 // now use mm1 for getting LBCarrys
2022 pand mm1, mm2 // get LBCarrys for each byte where both
2026 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2034 movq mm1, mm3 // now use mm1 for getting LBCarrys
2035 pand mm1, mm2 // get LBCarrys for each byte where both
2039 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2051 movq mm1, mm3 // now use mm1 for getting LBCarrys
2052 pand mm1, mm2 // get LBCarrys for each byte where both
2056 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2099 movq mm1, [esi + ebx]
2102 pand mm3, mm1 // get lsb for each prev_row byte
2103 psrlq mm1, 1 // divide prev_row bytes by 2
2104 pand mm1, mm4 // clear invalid bit 7 of each byte
2105 paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte
2107 movq mm1, mm3 // now use mm1 for getting LBCarrys
2108 pand mm1, mm2 // get LBCarrys for each byte where both
2112 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2120 movq mm1, mm3 // now use mm1 for getting LBCarrys
2121 pand mm1, mm2 // get LBCarrys for each byte where both
2125 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2158 movq mm1, [esi + ebx]
2161 pand mm3, mm1 // get lsb for each prev_row byte
2162 psrlq mm1, 1 // divide prev_row bytes by 2
2163 pand mm1, mm4 // clear invalid bit 7 of each byte
2165 paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte
2167 movq mm1, mm3 // now use mm1 for getting LBCarrys
2168 pand mm1, mm2 // get LBCarrys for each byte where both
2172 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2179 movq mm1, mm3 // now use mm1 for getting LBCarrys
2180 pand mm1, mm2 // get LBCarrys for each byte where both
2184 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2194 movq mm1, mm3 // now use mm1 for getting LBCarrys
2195 pand mm1, mm2 // get LBCarrys for each byte where both
2199 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2210 movq mm1, mm3 // now use mm1 for getting LBCarrys
2211 pand mm1, mm2 // get LBCarrys for each byte where both
2215 paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte
2275 movq mm1, [esi + ebx]
2277 pand mm3, mm1 // get lsb for each prev_row byte
2278 psrlq mm1, 1 // divide prev_row bytes by 2
2282 pand mm1, mm4 // clear invalid bit 7 of each byte
2285 paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte
2308 movq mm1, [esi + ebx]
2309 pand mm3, mm1 // get lsb for each prev_row byte
2311 psrlq mm1, 1 // divide prev_row bytes by 2
2315 pand mm1, mm4 // clear invalid bit 7 of each byte
2318 paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte
2488 movq mm1, [edi+ebx-8]
2490 psrlq mm1, ShiftRem // shift last 3 bytes to 1st 3 bytes
2492 punpcklbw mm1, mm0 // Unpack High bytes of a
2500 movq mm5, mm1
2533 pandn mm0, mm1
2538 pxor mm1, mm1
2543 packuswb mm7, mm1
2550 movq mm1, mm7 // Now mm1 will be used as Raw(x-bpp)
2553 punpcklbw mm1, mm0 // Unpack High bytes of a
2557 movq mm5, mm1
2592 pandn mm0, mm1
2600 pxor mm1, mm1
2603 packuswb mm7, mm1
2613 movq mm1, mm7
2615 psllq mm1, ShiftBpp // Shift bytes
2616 // Now mm1 will be used as Raw(x-bpp)
2619 punpckhbw mm1, mm0 // Unpack High bytes of a
2622 movq mm5, mm1
2653 pandn mm0, mm1
2662 pxor mm1, mm1
2663 packuswb mm1, mm7
2666 pand mm1, ActiveMaskEnd
2667 paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x)
2671 movq [edi + ebx - 8], mm1 // write back updated value
2672 // mm1 will be used as Raw(x-bpp) next loop
2693 movq mm1, [edi+ebx-8]
2697 psrlq mm1, ShiftRem
2700 punpcklbw mm1, mm0 // Unpack Low bytes of a
2709 movq mm5, mm1
2741 pandn mm0, mm1
2746 pxor mm1, mm1
2751 packuswb mm7, mm1
2759 movq mm1, [edi+ebx-8]
2762 psrlq mm1, ShiftRem
2766 por mm1, mm5
2769 punpckhbw mm1, mm0 // Unpack High bytes of a
2773 movq mm5, mm1
2805 pandn mm0, mm1
2810 pxor mm1, mm1
2813 pxor mm1, mm1
2818 packuswb mm1, mm7
2819 paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x)
2821 movq [edi + ebx - 8], mm1 // write back updated value
2822 // mm1 will be used as Raw(x-bpp) next loop
2837 movq mm1, [edi+ebx-8] // Only time should need to read
2842 punpckhbw mm1, mm0 // Unpack Low bytes of a
2849 movq mm5, mm1
2881 pandn mm0, mm1
2886 pxor mm1, mm1
2891 packuswb mm7, mm1
2898 movq mm1, mm7 // Now mm1 will be used as Raw(x-bpp)
2901 punpcklbw mm1, mm0 // Unpack Low bytes of a
2905 movq mm5, mm1
2937 pandn mm0, mm1
2942 pxor mm1, mm1
2945 pxor mm1, mm1
2950 packuswb mm1, mm7
2951 paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x)
2953 movq [edi + ebx - 8], mm1 // write back updated value
2954 // mm1 will be used as Raw(x-bpp) next loop
2968 movq mm1, [edi+ebx-8] // Only time should need to read
2973 punpcklbw mm1, mm0 // Unpack Low bytes of a
2980 movq mm5, mm1
3012 pandn mm0, mm1
3017 pxor mm1, mm1
3022 packuswb mm7, mm1
3029 movq mm1, [edi+ebx-8] // read a=Raw(x-bpp) bytes
3033 punpckhbw mm1, mm0 // Unpack High bytes of a
3037 movq mm5, mm1
3069 pandn mm0, mm1
3074 pxor mm1, mm1
3077 pxor mm1, mm1
3082 packuswb mm1, mm7
3083 paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x)
3085 movq [edi + ebx - 8], mm1 // write back updated value
3086 // mm1 will be used as Raw(x-bpp) next loop
3315 movq mm1, [edi+ebx-8]
3317 psrlq mm1, ShiftRem // Shift data for adding 1st bpp bytes
3321 paddb mm0, mm1
3323 movq mm1, mm0 // mov updated Raws to mm1
3324 psllq mm1, ShiftBpp // shift data to position correctly
3325 pand mm1, mm7 // mask to use only 2nd active group
3326 paddb mm0, mm1
3328 movq mm1, mm0 // mov updated Raws to mm1
3329 psllq mm1, ShiftBpp // shift data to position correctly
3330 pand mm1, mm6 // mask to use only 3rd active group
3332 paddb mm0, mm1
3336 movq mm1, mm0
3388 movq mm1, [edi+ebx-8]
3390 psrlq mm1, ShiftRem // Shift data for adding 1st bpp bytes
3393 paddb mm0, mm1
3395 movq mm1, mm0 // mov updated Raws to mm1
3396 psllq mm1, ShiftBpp // shift data to position correctly
3400 paddb mm0, mm1
3403 movq mm1, mm0 // Prep for doing 1st add at top of loop
3427 movq mm1, [edi+ebx-8]
3430 psrlq mm1, ShiftRem // Shift data for adding 1st bpp bytes
3434 paddb mm0, mm1
3436 movq mm1, mm0 // mov updated Raws to mm1
3437 psllq mm1, ShiftBpp // shift data to position correctly
3438 pand mm1, mm7 // mask to use only 2nd active group
3439 paddb mm0, mm1
3441 movq mm1, mm0 // mov updated Raws to mm1
3442 psllq mm1, ShiftBpp // shift data to position correctly
3443 pand mm1, mm6 // mask to use only 3rd active group
3444 paddb mm0, mm1
3446 movq mm1, mm0 // mov updated Raws to mm1
3447 psllq mm1, ShiftBpp // shift data to position correctly
3448 pand mm1, mm5 // mask to use only 4th active group
3450 paddb mm0, mm1
3453 movq mm1, mm0 // Prep for doing 1st add at top of loop
3472 movq mm1, [edi+ebx+8] // Load Sub(x) for 2nd 8 bytes
3479 paddb mm1, mm0
3481 movq [edi+ebx+8], mm1 // Write Raw(x) for 2nd 8 bytes
3482 paddb mm2, mm1
3510 movq mm7, mm0 // Move calculated Raw(x) data to mm1 to
3527 movq mm1, [esi+ebx]
3529 paddb mm0, mm1
3594 movq mm1, [esi+ebx]
3597 paddb mm0, mm1
3609 movq mm1, [esi+ebx+32]
3613 paddb mm0, mm1
3645 // Loop using MMX registers mm0 & mm1 to update 8 bytes simultaneously
3647 movq mm1, [esi+ebx]
3650 paddb mm0, mm1