1 |
;/************************************************************************** |
;/**************************************************************************** |
2 |
; * |
; * |
3 |
; * XVID MPEG-4 VIDEO CODEC |
; * XVID MPEG-4 VIDEO CODEC |
4 |
; * sse2 cbp calc |
; * - SSE2 CBP computation - |
5 |
; * |
; * |
6 |
; * This file is part of XviD, a free MPEG-4 video encoder/decoder |
; * Copyright (C) 2002 Daniel Smith <danielsmith@astroboymail.com> |
7 |
|
; * 2002 Pascal Massimino <skal@planet-d.net> |
8 |
; * |
; * |
9 |
; * XviD is free software; you can redistribute it and/or modify it |
; * This program is free software ; you can redistribute it and/or modify |
10 |
; * under the terms of the GNU General Public License as published by |
; * it under the terms of the GNU General Public License as published by |
11 |
; * the Free Software Foundation; either version 2 of the License, or |
; * the Free Software Foundation; either version 2 of the License, or |
12 |
; * (at your option) any later version. |
; * (at your option) any later version. |
13 |
; * |
; * |
20 |
; * along with this program; if not, write to the Free Software |
; * along with this program; if not, write to the Free Software |
21 |
; * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
; * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
22 |
; * |
; * |
23 |
; * Under section 8 of the GNU General Public License, the copyright |
; * $Id: cbp_sse2.asm,v 1.14 2009-09-16 17:07:58 Isibaar Exp $ |
24 |
; * holders of XVID explicitly forbid distribution in the following |
; * |
25 |
; * countries: |
; ***************************************************************************/ |
|
; * |
|
|
; * - Japan |
|
|
; * - United States of America |
|
|
; * |
|
|
; * Linking XviD statically or dynamically with other modules is making a |
|
|
; * combined work based on XviD. Thus, the terms and conditions of the |
|
|
; * GNU General Public License cover the whole combination. |
|
|
; * |
|
|
; * As a special exception, the copyright holders of XviD give you |
|
|
; * permission to link XviD with independent modules that communicate with |
|
|
; * XviD solely through the VFW1.1 and DShow interfaces, regardless of the |
|
|
; * license terms of these independent modules, and to copy and distribute |
|
|
; * the resulting combined work under terms of your choice, provided that |
|
|
; * every copy of the combined work is accompanied by a complete copy of |
|
|
; * the source code of XviD (the version of XviD used to produce the |
|
|
; * combined work), being distributed under the terms of the GNU General |
|
|
; * Public License plus this exception. An independent module is a module |
|
|
; * which is not derived from or based on XviD. |
|
|
; * |
|
|
; * Note that people who make modified versions of XviD are not obligated |
|
|
; * to grant this special exception for their modified versions; it is |
|
|
; * their choice whether to do so. The GNU General Public License gives |
|
|
; * permission to release a modified version without this exception; this |
|
|
; * exception also makes it possible to release a modified version which |
|
|
; * carries forward this exception. |
|
|
; * |
|
|
; * $Id: cbp_sse2.asm,v 1.2 2002-11-17 00:57:58 edgomez Exp $ |
|
|
; * |
|
|
; *************************************************************************/ |
|
|
|
|
|
bits 32 |
|
|
|
|
|
section .data |
|
|
|
|
|
%macro cglobal 1 |
|
|
%ifdef PREFIX |
|
|
global _%1 |
|
|
%define %1 _%1 |
|
|
%else |
|
|
global %1 |
|
|
%endif |
|
|
%endmacro |
|
|
|
|
|
align 16 |
|
|
|
|
|
ignore_dc dw 0, -1, -1, -1, -1, -1, -1, -1 |
|
|
|
|
|
section .text |
|
|
|
|
|
cglobal calc_cbp_sse2 |
|
26 |
|
|
27 |
;=========================================================================== |
;============================================================================= |
28 |
; |
; Macros |
29 |
; uint32_t calc_cbp_sse2(const int16_t coeff[6][64]); |
;============================================================================= |
30 |
; |
|
31 |
; not enabled - slower than mmx? |
%include "nasm.inc" |
32 |
; |
|
33 |
;=========================================================================== |
%macro LOOP_SSE2 2 |
34 |
|
movdqa xmm0, [%2+(%1)*128] |
35 |
%macro LOOP_SSE2 1 |
pand xmm0, xmm3 |
36 |
movdqa xmm0, [edx+(%1)*128] |
movdqa xmm1, [%2+(%1)*128+16] |
37 |
pand xmm0, xmm7 |
|
38 |
movdqa xmm1, [edx+(%1)*128+16] |
por xmm0, [%2+(%1)*128+32] |
39 |
|
por xmm1, [%2+(%1)*128+48] |
40 |
por xmm0, [edx+(%1)*128+32] |
por xmm0, [%2+(%1)*128+64] |
41 |
por xmm1, [edx+(%1)*128+48] |
por xmm1, [%2+(%1)*128+80] |
42 |
por xmm0, [edx+(%1)*128+64] |
por xmm0, [%2+(%1)*128+96] |
43 |
por xmm1, [edx+(%1)*128+80] |
por xmm1, [%2+(%1)*128+112] |
|
por xmm0, [edx+(%1)*128+96] |
|
|
por xmm1, [edx+(%1)*128+112] |
|
44 |
|
|
45 |
por xmm0, xmm1 ; xmm0 = xmm1 = 128 bits worth of info |
por xmm0, xmm1 ; xmm0 = xmm1 = 128 bits worth of info |
46 |
psadbw xmm0, xmm6 ; contains 2 dwords with sums |
psadbw xmm0, xmm2 ; contains 2 dwords with sums |
47 |
movhlps xmm1, xmm0 ; move high dword from xmm0 to low xmm1 |
movhlps xmm1, xmm0 ; move high dword from xmm0 to low xmm1 |
48 |
por xmm0, xmm1 ; combine |
por xmm0, xmm1 ; combine |
49 |
movd ecx, xmm0 ; if ecx set, values were found |
movd ecx, xmm0 ; if ecx set, values were found |
50 |
test ecx, ecx |
test _ECX, _ECX |
51 |
%endmacro |
%endmacro |
52 |
|
|
53 |
align 16 |
;============================================================================= |
54 |
|
; Data (Read Only) |
55 |
|
;============================================================================= |
56 |
|
|
57 |
|
DATA |
58 |
|
|
59 |
|
ALIGN SECTION_ALIGN |
60 |
|
ignore_dc: |
61 |
|
dw 0, -1, -1, -1, -1, -1, -1, -1 |
62 |
|
|
63 |
|
;============================================================================= |
64 |
|
; Code |
65 |
|
;============================================================================= |
66 |
|
|
67 |
|
TEXT |
68 |
|
|
69 |
|
;----------------------------------------------------------------------------- |
70 |
|
; uint32_t calc_cbp_sse2(const int16_t coeff[6*64]); |
71 |
|
;----------------------------------------------------------------------------- |
72 |
|
|
73 |
|
ALIGN SECTION_ALIGN |
74 |
|
cglobal calc_cbp_sse2 |
75 |
calc_cbp_sse2: |
calc_cbp_sse2: |
76 |
mov edx, [esp+4] ; coeff[] |
mov _EDX, prm1 ; coeff[] |
77 |
xor eax, eax ; cbp = 0 |
xor _EAX, _EAX ; cbp = 0 |
78 |
|
|
79 |
movdqu xmm7, [ignore_dc] ; mask to ignore dc value |
movdqu xmm3, [ignore_dc] ; mask to ignore dc value |
80 |
pxor xmm6, xmm6 ; zero |
pxor xmm2, xmm2 ; zero |
81 |
|
|
82 |
LOOP_SSE2 0 |
LOOP_SSE2 0, _EDX |
|
test ecx, ecx |
|
83 |
jz .blk2 |
jz .blk2 |
84 |
or eax, (1<<5) |
or _EAX, (1<<5) |
85 |
.blk2 |
|
86 |
LOOP_SSE2 1 |
.blk2: |
87 |
test ecx, ecx |
LOOP_SSE2 1, _EDX |
88 |
jz .blk3 |
jz .blk3 |
89 |
or eax, (1<<4) |
or _EAX, (1<<4) |
90 |
.blk3 |
|
91 |
LOOP_SSE2 2 |
.blk3: |
92 |
test ecx, ecx |
LOOP_SSE2 2, _EDX |
93 |
jz .blk4 |
jz .blk4 |
94 |
or eax, (1<<3) |
or _EAX, (1<<3) |
95 |
.blk4 |
|
96 |
LOOP_SSE2 3 |
.blk4: |
97 |
test ecx, ecx |
LOOP_SSE2 3, _EDX |
98 |
jz .blk5 |
jz .blk5 |
99 |
or eax, (1<<2) |
or _EAX, (1<<2) |
100 |
.blk5 |
|
101 |
LOOP_SSE2 4 |
.blk5: |
102 |
test ecx, ecx |
LOOP_SSE2 4, _EDX |
103 |
jz .blk6 |
jz .blk6 |
104 |
or eax, (1<<1) |
or _EAX, (1<<1) |
105 |
.blk6 |
|
106 |
LOOP_SSE2 5 |
.blk6: |
107 |
test ecx, ecx |
LOOP_SSE2 5, _EDX |
108 |
jz .finished |
jz .finished |
109 |
or eax, (1<<0) |
or _EAX, (1<<0) |
110 |
.finished |
|
111 |
|
.finished: |
112 |
|
|
113 |
ret |
ret |
114 |
|
ENDFUNC |
115 |
|
|
116 |
|
NON_EXEC_STACK |