[x265] [PATCH] asm: avx2 code for add_ps[8x8] for 8bpp -- 22x

[chen]

[i420]  add_ps[  8x8]  6.68x    250.21          1670.68
 
[i420]  add_ps[  8x8]  6.56x    246.99          1620.94
 
same speed,
 
wrong debug code like below:
p.pu[LUMA_8x8].sad_x3



At 2015-03-05 17:31:34,sumalatha at multicorewareinc.com wrote:
># HG changeset patch
># User Sumalatha Polureddy<sumalatha at multicorewareinc.com>
># Date 1425547841 -19800
># Node ID b12a6a607b55ac1a347ca9c0026d9eabefe5ed80
># Parent  ea9bdb10353fcb06cea1045ba0186c22c448df63
>asm: avx2 code for add_ps[8x8] for 8bpp -- 22x
>
>add_ps[  8x8]  22.34x   320.31          7154.80
>
>diff -r ea9bdb10353f -r b12a6a607b55 source/common/x86/asm-primitives.cpp
>--- a/source/common/x86/asm-primitives.cpp	Wed Mar 04 13:20:55 2015 +0530
>+++ b/source/common/x86/asm-primitives.cpp	Thu Mar 05 15:00:41 2015 +0530
>@@ -1376,6 +1376,7 @@
>         ASSIGN_SSE_SS(avx);
>         LUMA_VAR(avx);
> 
>+        p.pu[LUMA_8x8].sad_x3 = x265_pixel_sad_x3_8x8_avx;
>         p.pu[LUMA_12x16].sad_x3 = x265_pixel_sad_x3_12x16_avx;
>         p.pu[LUMA_16x4].sad_x3  = x265_pixel_sad_x3_16x4_avx;
>         HEVC_SAD_X3(avx);
>@@ -1432,6 +1433,7 @@
>     }
>     if (cpuMask & X265_CPU_AVX2)
>     {
>+        p.cu[BLOCK_8x8].add_ps = x265_pixel_add_ps_8x8_avx2;
>         p.cu[BLOCK_16x16].add_ps = x265_pixel_add_ps_16x16_avx2;
>         p.cu[BLOCK_32x32].add_ps = x265_pixel_add_ps_32x32_avx2;
>         p.cu[BLOCK_64x64].add_ps = x265_pixel_add_ps_64x64_avx2;
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20150306/175003b6/attachment.html>