[x265] [PATCH] 16bpp: assembly code for intra_pred_dc4

chen chenm003 at 163.com
Wed Dec 4 11:43:15 CET 2013


Please don't put empty/dummy function in patch. 

At 2013-12-04 18:15:48,dnyaneshwar at multicorewareinc.com wrote:
># HG changeset patch
># User Dnyaneshwar G <dnyaneshwar at multicorewareinc.com>
># Date 1386151818 -19800
>#      Wed Dec 04 15:40:18 2013 +0530
># Node ID dfde78a1d1eaad7385823831dd5a121861e0d82d
># Parent  759a5eee46d38ffa0205fc1551291653086208c9
>16bpp: assembly code for intra_pred_dc4
>
>diff -r 759a5eee46d3 -r dfde78a1d1ea source/common/CMakeLists.txt
>--- a/source/common/CMakeLists.txt	Wed Dec 04 12:24:59 2013 +0530
>+++ b/source/common/CMakeLists.txt	Wed Dec 04 15:40:18 2013 +0530
>@@ -119,12 +119,12 @@
> if(ENABLE_PRIMITIVES_ASM)
>     set(C_SRCS asm-primitives.cpp pixel.h mc.h ipfilter8.h blockcopy8.h dct8.h)
>     set(A_SRCS pixel-a.asm const-a.asm cpu-a.asm ssd-a.asm mc-a.asm
>-               mc-a2.asm ipfilter8.asm pixel-util8.asm blockcopy8.asm intrapred8.asm
>+               mc-a2.asm ipfilter8.asm pixel-util8.asm blockcopy8.asm
>                pixeladd8.asm dct8.asm)
>     if(HIGH_BIT_DEPTH)
>-        set(A_SRCS ${A_SRCS} sad16-a.asm)
>+        set(A_SRCS ${A_SRCS} sad16-a.asm intrapred16.asm)
>     else()
>-        set(A_SRCS ${A_SRCS} sad-a.asm)
>+        set(A_SRCS ${A_SRCS} sad-a.asm intrapred8.asm)
>     endif()
> 
>     if (NOT X64)
>diff -r 759a5eee46d3 -r dfde78a1d1ea source/common/x86/asm-primitives.cpp
>--- a/source/common/x86/asm-primitives.cpp	Wed Dec 04 12:24:59 2013 +0530
>+++ b/source/common/x86/asm-primitives.cpp	Wed Dec 04 15:40:18 2013 +0530
>@@ -563,6 +563,7 @@
>     }
>     if (cpuMask & X265_CPU_SSE4)
>     {
>+        p.intra_pred_dc[BLOCK_4x4] = x265_intra_pred_dc4_sse4;
>     }
>     if (cpuMask & X265_CPU_XOP)
>     {
>diff -r 759a5eee46d3 -r dfde78a1d1ea source/common/x86/const-a.asm
>--- a/source/common/x86/const-a.asm	Wed Dec 04 12:24:59 2013 +0530
>+++ b/source/common/x86/const-a.asm	Wed Dec 04 15:40:18 2013 +0530
>@@ -36,6 +36,7 @@
> const pw_32,       times 16 dw 32
> const pw_512,      times 16 dw 512
> const pw_1024,     times 16 dw 1024
>+const pw_4096,     times 16 dw 4096
> const pw_00ff,     times 16 dw 0x00ff
> const pw_pixel_max,times 16 dw ((1 << BIT_DEPTH)-1)
> const pd_1,        times 8 dd 1
>diff -r 759a5eee46d3 -r dfde78a1d1ea source/common/x86/intrapred16.asm
>--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
>+++ b/source/common/x86/intrapred16.asm	Wed Dec 04 15:40:18 2013 +0530
>@@ -0,0 +1,117 @@
>+;*****************************************************************************
>+;* Copyright (C) 2013 x265 project
>+;*
>+;* Authors: Dnyaneshwar Gorade <dnyaneshwar at multicorewareinc.com>
>+;*
>+;* This program is free software; you can redistribute it and/or modify
>+;* it under the terms of the GNU General Public License as published by
>+;* the Free Software Foundation; either version 2 of the License, or
>+;* (at your option) any later version.
>+;*
>+;* This program is distributed in the hope that it will be useful,
>+;* but WITHOUT ANY WARRANTY; without even the implied warranty of
>+;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>+;* GNU General Public License for more details.
>+;*
>+;* You should have received a copy of the GNU General Public License
>+;* along with this program; if not, write to the Free Software
>+;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02111, USA.
>+;*
>+;* This program is also available under a commercial proprietary license.
>+;* For more information, contact us at licensing at multicorewareinc.com.
>+;*****************************************************************************/
>+
>+%include "x86inc.asm"
>+%include "x86util.asm"
>+
>+SECTION_RODATA 32
>+
>+SECTION .text
>+
>+cextern pw_8
>+cextern pw_4096
>+cextern multiH
>+cextern multiH2
>+cextern multiH3
>+cextern multi_2Row
>+
>+;-----------------------------------------------------------------------------
>+; void intra_pred_dc(pixel* above, pixel* left, pixel* dst, intptr_t dstStride, int filter)
>+;-----------------------------------------------------------------------------
>+INIT_XMM sse4
>+cglobal intra_pred_dc4, 5,6,3
>+    pxor        m0,             m0
>+    movh        m1,             [r0]           ; sumAbove
>+    movh        m2,             [r1]           ; sumLeft
>+
>+    paddw       m1,             m2
>+    pshufd      m2,             m1, 1
>+    paddw       m1,             m2
>+    phaddw      m1,             m1             ; m1 = sum
>+
>+    test        r4d,            r4d
>+
>+    pmulhrsw    m1,             [pw_4096]      ; m1 = (sum + 4) / 8
>+    movd        r4d,            m1             ; r4d = dc_val
>+    movzx       r4d,            r4w
>+    pshuflw     m1,             m1, 0          ; m1 = word [dc_val ...]
>+
>+    ; store DC 4x4
>+    movh        [r2],           m1
>+    movh        [r2 + r3 * 2],  m1
>+    lea         r5,             [r2 + r3 * 4]
>+    movh        [r5],           m1
>+    movh        [r5 + r3 * 2],  m1
>+
>+    ; do DC filter
>+    jz          .end
>+    lea         r5d,            [r4d * 2 + 2]  ; r5d = DC * 2 + 2
>+    add         r4d,            r5d            ; r4d = DC * 3 + 2
>+    movd        m1,             r4d
>+    pshuflw     m1,             m1, 0          ; m1 = pixDCx3
>+
>+    ; filter top
>+    movu        m2,             [r0]
>+    paddw       m2,             m1
>+    psraw       m2,             2
>+    movh        [r2],           m2             ; overwrite top-left pixel, we will update it later
>+
>+    ; filter top-left
>+    movzx       r0d, word       [r0]
>+    add         r5d,            r0d
>+    movzx       r0d, word       [r1]
>+    add         r0d,            r5d
>+    shr         r0d,            2
>+    mov         [r2],           r0w
>+
>+    ; filter left
>+    lea         r2,             [r2 + r3 * 2]
>+    movu        m2,             [r1 + 2]
>+    paddw       m2,             m1
>+    psraw       m2,             2
>+    movd        r0d,            m2
>+    mov         [r2],           r0w
>+    shr         r0d,            16
>+    mov         [r2 + 2 * r3],  r0w
>+    pextrw      [r2 + r3 * 4],  m2, 2
>+
>+.end:
>+
>+    RET
>+
>+;-----------------------------------------------------------------------------
>+; void intraPredAng(pixel* dst, intptr_t dstStride, pixel *refLeft, pixel *refAbove, int dirMode, int bFilter)
>+;-----------------------------------------------------------------------------
>+INIT_XMM ssse3
>+cglobal intra_pred_ang4_2, 3,3,4
>+
>+    RET
>+
>+
>+;-----------------------------------------------------------------------------
>+; void intraPredAng(pixel* dst, intptr_t dstStride, pixel *refLeft, pixel *refAbove, int dirMode, int bFilter)
>+;-----------------------------------------------------------------------------
>+INIT_XMM ssse3
>+cglobal intra_pred_ang4_3, 3,3,4
>+
>+    RET
>_______________________________________________
>x265-devel mailing list
>x265-devel at videolan.org
>https://mailman.videolan.org/listinfo/x265-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20131204/d447d42b/attachment-0001.html>


More information about the x265-devel mailing list