At 2015-09-30 13:53:18,[email protected] wrote: ># HG changeset patch ># User Ramya Sriraman [email protected]> ># Date 1443592336 -19800 ># Wed Sep 30 11:22:16 2015 +0530 ># Node ID 29b61906162c657da241aecee9012e3f2da34b6d ># Parent 5f1451e5842252b31442e8b6519138d8033bbb2b >asm: Add sse_ss for [16x16],[32x32] & [64x64] for 8bpp avx2 > diff -r 5f1451e58422 -r 29b61906162c source/common/x86/ssd-a.asm >--- a/source/common/x86/ssd-a.asm Mon Sep 28 16:43:47 2015 +0530 >+++ b/source/common/x86/ssd-a.asm Wed Sep 30 11:22:16 2015 +0530 >@@ -1100,8 +1100,195 @@ > SSD_SS_32xN > SSD_SS_48 > SSD_SS_64xN >+ >+INIT_YMM avx2 >+cglobal pixel_ssd_ss_16x16, 4,4,5 >+ add r1d, r1d >+ add r3d, r3d >+ pxor m4, m4 >+ >+ movu m0, [r0] >+ movu m1, [r0+r1] >+ movu m2, [r2] >+ movu m3, [r2+r3] >+ psubw m0, m2 in avx2, vpsubw can work on unaligned address >+ psubw m1, m3 >+ lea r0, [r0+2*r1] >+ lea r2, [r2+2*r3] >+ pmaddwd m0, m0 >+ pmaddwd m1, m1 >+ paddd m0 , m1 >+ paddd m4, m0
_______________________________________________ x265-devel mailing list [email protected] https://mailman.videolan.org/listinfo/x265-devel
