same as previous (improve in sum part)
At 2015-03-11 18:50:23,[email protected] wrote: ># HG changeset patch ># User Sumalatha Polureddy<[email protected]> ># Date 1426071015 -19800 ># Node ID 8dd95d229097e31d4fc0edc7873664cc24bdeb5f ># Parent bc9f8dbe8f655fe0b1a218bb6abb8fa88536c6dd >asm: avx2 code for sad[32x8] for 8bpp > >SSE3 >sad[ 32x8] 27.37x 218.86 5989.18 > >AVX2 >sad[ 32x8] 37.41x 158.53 5930.84 > >diff -r bc9f8dbe8f65 -r 8dd95d229097 source/common/x86/asm-primitives.cpp >--- a/source/common/x86/asm-primitives.cpp Wed Mar 11 15:18:06 2015 +0530 >+++ b/source/common/x86/asm-primitives.cpp Wed Mar 11 16:20:15 2015 +0530 >@@ -1444,6 +1444,7 @@ > > p.pu[LUMA_32x32].sad = x265_pixel_sad_32x32_avx2; > p.pu[LUMA_64x64].sad = x265_pixel_sad_64x64_avx2; >+ p.pu[LUMA_32x8].sad = x265_pixel_sad_32x8_avx2; > > p.pu[LUMA_8x4].sad_x3 = x265_pixel_sad_x3_8x4_avx2; > p.pu[LUMA_8x8].sad_x3 = x265_pixel_sad_x3_8x8_avx2; >diff -r bc9f8dbe8f65 -r 8dd95d229097 source/common/x86/sad-a.asm >--- a/source/common/x86/sad-a.asm Wed Mar 11 15:18:06 2015 +0530 >+++ b/source/common/x86/sad-a.asm Wed Mar 11 16:20:15 2015 +0530 >@@ -3925,6 +3925,34 @@ > RET > > INIT_YMM avx2 >+cglobal pixel_sad_32x8, 4,4,5 >+ xorps m0, m0 >+%assign x 0 >+%rep 4 >+ movu m1, [r0] ; row 0 of pix0 >+ movu m2, [r2] ; row 0 of pix1 >+ movu m3, [r0 + r1] ; row 1 of pix0 >+ movu m4, [r2 + r3] ; row 1 of pix1 >+ >+ psadbw m1, m2 >+ psadbw m3, m4 >+ paddd m0, m1 >+ paddd m0, m3 >+%assign x x+1 >+ %if x < 4 >+ lea r2, [r2 + 2 * r3] >+ lea r0, [r0 + 2 * r1] >+ %endif >+%endrep >+ vextracti128 xm1, m0, 1 >+ paddd xm0, xm1 >+ pshufd xm1, xm0, 2 >+ paddd xm0,xm1 >+ movd eax, xm0 >+ RET >+ >+ >+INIT_YMM avx2 > cglobal pixel_sad_64x64, 4,4,5 > xorps m0, m0 > >_______________________________________________ >x265-devel mailing list >[email protected] >https://mailman.videolan.org/listinfo/x265-devel
_______________________________________________ x265-devel mailing list [email protected] https://mailman.videolan.org/listinfo/x265-devel
