| 3 |
; * XVID MPEG-4 VIDEO CODEC |
; * XVID MPEG-4 VIDEO CODEC |
| 4 |
; * mmx cbp calc |
; * mmx cbp calc |
| 5 |
; * |
; * |
| 6 |
; * This program is an implementation of a part of one or more MPEG-4 |
; * This file is part of XviD, a free MPEG-4 video encoder/decoder |
|
; * Video tools as specified in ISO/IEC 14496-2 standard. Those intending |
|
|
; * to use this software module in hardware or software products are |
|
|
; * advised that its use may infringe existing patents or copyrights, and |
|
|
; * any such use would be at such party's own risk. The original |
|
|
; * developer of this software module and his/her company, and subsequent |
|
|
; * editors and their companies, will have no liability for use of this |
|
|
; * software or modifications or derivatives thereof. |
|
| 7 |
; * |
; * |
| 8 |
; * This program is free software; you can redistribute it and/or modify |
; * XviD is free software; you can redistribute it and/or modify it |
| 9 |
; * it under the terms of the GNU General Public License as published by |
; * under the terms of the GNU General Public License as published by |
| 10 |
; * the Free Software Foundation; either version 2 of the License, or |
; * the Free Software Foundation; either version 2 of the License, or |
| 11 |
; * (at your option) any later version. |
; * (at your option) any later version. |
| 12 |
; * |
; * |
| 17 |
; * |
; * |
| 18 |
; * You should have received a copy of the GNU General Public License |
; * You should have received a copy of the GNU General Public License |
| 19 |
; * along with this program; if not, write to the Free Software |
; * along with this program; if not, write to the Free Software |
| 20 |
; * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
; * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
| 21 |
; * |
; * |
| 22 |
; *************************************************************************/ |
; * Under section 8 of the GNU General Public License, the copyright |
| 23 |
|
; * holders of XVID explicitly forbid distribution in the following |
| 24 |
;/************************************************************************** |
; * countries: |
| 25 |
|
; * |
| 26 |
|
; * - Japan |
| 27 |
|
; * - United States of America |
| 28 |
|
; * |
| 29 |
|
; * Linking XviD statically or dynamically with other modules is making a |
| 30 |
|
; * combined work based on XviD. Thus, the terms and conditions of the |
| 31 |
|
; * GNU General Public License cover the whole combination. |
| 32 |
|
; * |
| 33 |
|
; * As a special exception, the copyright holders of XviD give you |
| 34 |
|
; * permission to link XviD with independent modules that communicate with |
| 35 |
|
; * XviD solely through the VFW1.1 and DShow interfaces, regardless of the |
| 36 |
|
; * license terms of these independent modules, and to copy and distribute |
| 37 |
|
; * the resulting combined work under terms of your choice, provided that |
| 38 |
|
; * every copy of the combined work is accompanied by a complete copy of |
| 39 |
|
; * the source code of XviD (the version of XviD used to produce the |
| 40 |
|
; * combined work), being distributed under the terms of the GNU General |
| 41 |
|
; * Public License plus this exception. An independent module is a module |
| 42 |
|
; * which is not derived from or based on XviD. |
| 43 |
|
; * |
| 44 |
|
; * Note that people who make modified versions of XviD are not obligated |
| 45 |
|
; * to grant this special exception for their modified versions; it is |
| 46 |
|
; * their choice whether to do so. The GNU General Public License gives |
| 47 |
|
; * permission to release a modified version without this exception; this |
| 48 |
|
; * exception also makes it possible to release a modified version which |
| 49 |
|
; * carries forward this exception. |
| 50 |
; * |
; * |
| 51 |
; * History: |
; * $Id: cbp_mmx.asm,v 1.7 2002-11-17 00:57:57 edgomez Exp $ |
|
; * |
|
|
; * 17.04.2002 sse2 stuff |
|
|
; * 22.03.2002 0.01 ; Min Chen <chenm001@163.com> |
|
|
; * ; use 386 cpu's 'BTS' to replace 'cbp |= 1 << (edx-1)' |
|
|
; * 24.11.2001 inital version; (c)2001 peter ross <pross@cs.rmit.edu.au> |
|
| 52 |
; * |
; * |
| 53 |
; *************************************************************************/ |
; *************************************************************************/ |
| 54 |
|
|
|
|
|
| 55 |
bits 32 |
bits 32 |
| 56 |
|
|
| 57 |
section .data |
section .data |
| 69 |
|
|
| 70 |
ignore_dc dw 0, -1, -1, -1, -1, -1, -1, -1 |
ignore_dc dw 0, -1, -1, -1, -1, -1, -1, -1 |
| 71 |
|
|
|
|
|
| 72 |
section .text |
section .text |
| 73 |
|
|
| 74 |
|
cglobal calc_cbp_mmx |
| 75 |
|
|
| 76 |
;=========================================================================== |
;=========================================================================== |
| 77 |
; |
; |
| 80 |
;=========================================================================== |
;=========================================================================== |
| 81 |
|
|
| 82 |
align 16 |
align 16 |
| 83 |
cglobal calc_cbp_mmx |
calc_cbp_mmx: |
|
calc_cbp_mmx |
|
| 84 |
push ebx |
push ebx |
|
push ecx |
|
|
push edx |
|
| 85 |
push esi |
push esi |
| 86 |
|
|
| 87 |
mov esi, [esp + 16 + 4] ; coeff |
mov esi, [esp + 8 + 4] ; coeff |
| 88 |
|
xor eax, eax ; cbp = 0 |
| 89 |
|
mov edx, (1 << 5) |
| 90 |
|
|
| 91 |
movq mm7, [ignore_dc] |
movq mm7, [ignore_dc] |
| 92 |
|
|
|
xor eax, eax ; cbp = 0 |
|
|
mov edx, 6 |
|
| 93 |
.loop |
.loop |
| 94 |
movq mm0, [esi] |
movq mm0, [esi] |
|
pand mm0, mm7 |
|
| 95 |
movq mm1, [esi+8] |
movq mm1, [esi+8] |
| 96 |
|
pand mm0, mm7 |
| 97 |
|
|
| 98 |
por mm0, [esi+16] |
por mm0, [esi+16] |
| 99 |
por mm1, [esi+24] |
por mm1, [esi+24] |
| 119 |
por mm0, mm1 |
por mm0, mm1 |
| 120 |
movq mm1, mm0 |
movq mm1, mm0 |
| 121 |
psrlq mm1, 32 |
psrlq mm1, 32 |
| 122 |
|
lea esi, [esi + 128] |
| 123 |
|
|
| 124 |
por mm0, mm1 |
por mm0, mm1 |
| 125 |
movd ebx, mm0 |
movd ebx, mm0 |
| 126 |
|
|
| 127 |
add esi, 128 |
test ebx, ebx |
| 128 |
|
jz .next |
| 129 |
or ebx, ebx |
or eax, edx ; cbp |= 1 << (5-i) |
| 130 |
jz .iterate |
|
| 131 |
|
.next |
| 132 |
; cbp |= 1 << (edx-1) |
shr edx,1 |
| 133 |
|
jnc .loop |
|
; Change by Chenm001 <chenm001@163.com> |
|
|
;mov ecx, edx |
|
|
;dec ecx |
|
|
;mov ebx, 1 |
|
|
;shl ebx, cl |
|
|
;or eax, ebx |
|
|
lea ebx,[edx-1] |
|
|
bts eax,ebx |
|
|
|
|
|
.iterate dec edx |
|
|
jnz .loop |
|
| 134 |
|
|
| 135 |
pop esi |
pop esi |
|
pop edx |
|
|
pop ecx |
|
| 136 |
pop ebx |
pop ebx |
| 137 |
|
|
| 138 |
ret |
ret |
|
|
|
|
|
|
|
|
|
|
;=========================================================================== |
|
|
; |
|
|
; uint32_t calc_cbp_sse2(const int16_t coeff[6][64]); |
|
|
; |
|
|
; not enabled - slower than mmx? |
|
|
; |
|
|
;=========================================================================== |
|
|
|
|
|
align 16 |
|
|
cglobal calc_cbp_sse2 |
|
|
calc_cbp_sse2 |
|
|
push esi |
|
|
|
|
|
mov esi, [esp + 4 + 4] ; coeff |
|
|
movdqa xmm7, [ignore_dc] ; mask to ignore dc value |
|
|
|
|
|
xor eax, eax ; cbp = 0 |
|
|
pxor xmm6, xmm6 ; zeroes to help psadbw |
|
|
|
|
|
.first movdqa xmm0, [esi] |
|
|
pand xmm0, xmm7 |
|
|
movdqa xmm1, [esi+16] |
|
|
|
|
|
por xmm0, [esi+32] |
|
|
por xmm1, [esi+48] |
|
|
por xmm0, [esi+64] |
|
|
por xmm1, [esi+80] |
|
|
por xmm0, [esi+96] |
|
|
por xmm1, [esi+112] |
|
|
|
|
|
por xmm0, xmm1 ; xmm0 = xmm1 = 128 bits worth of info |
|
|
psadbw xmm0, xmm6 ; contains 2 dwords with sums |
|
|
movhlps xmm1, xmm0 ; move high dword from xmm0 to low xmm1 |
|
|
por xmm0, xmm1 ; combine |
|
|
movd ecx, xmm0 ; if ecx set, values were found |
|
|
|
|
|
add esi, 128 |
|
|
|
|
|
or ecx, ecx |
|
|
jz .second |
|
|
|
|
|
bts eax, 5 |
|
|
|
|
|
.second movdqa xmm0, [esi] |
|
|
pand xmm0, xmm7 |
|
|
movdqa xmm1, [esi+16] |
|
|
|
|
|
por xmm0, [esi+32] |
|
|
por xmm1, [esi+48] |
|
|
por xmm0, [esi+64] |
|
|
por xmm1, [esi+80] |
|
|
por xmm0, [esi+96] |
|
|
por xmm1, [esi+112] |
|
|
|
|
|
por xmm0, xmm1 |
|
|
psadbw xmm0, xmm6 |
|
|
movhlps xmm1, xmm0 |
|
|
por xmm0, xmm1 |
|
|
movd ecx, xmm0 |
|
|
|
|
|
add esi, 128 |
|
|
|
|
|
or ecx, ecx |
|
|
jz .third |
|
|
|
|
|
bts eax, 4 |
|
|
|
|
|
.third movdqa xmm0, [esi] |
|
|
pand xmm0, xmm7 |
|
|
movdqa xmm1, [esi+16] |
|
|
|
|
|
por xmm0, [esi+32] |
|
|
por xmm1, [esi+48] |
|
|
por xmm0, [esi+64] |
|
|
por xmm1, [esi+80] |
|
|
por xmm0, [esi+96] |
|
|
por xmm1, [esi+112] |
|
|
|
|
|
por xmm0, xmm1 |
|
|
psadbw xmm0, xmm6 |
|
|
movhlps xmm1, xmm0 |
|
|
por xmm0, xmm1 |
|
|
movd ecx, xmm0 |
|
|
|
|
|
add esi, 128 |
|
|
|
|
|
or ecx, ecx |
|
|
jz .fourth |
|
|
|
|
|
bts eax, 3 |
|
|
|
|
|
.fourth movdqa xmm0, [esi] |
|
|
pand xmm0, xmm7 |
|
|
movdqa xmm1, [esi+16] |
|
|
|
|
|
por xmm0, [esi+32] |
|
|
por xmm1, [esi+48] |
|
|
por xmm0, [esi+64] |
|
|
por xmm1, [esi+80] |
|
|
por xmm0, [esi+96] |
|
|
por xmm1, [esi+112] |
|
|
|
|
|
por xmm0, xmm1 |
|
|
psadbw xmm0, xmm6 |
|
|
movhlps xmm1, xmm0 |
|
|
por xmm0, xmm1 |
|
|
movd ecx, xmm0 |
|
|
|
|
|
add esi, 128 |
|
|
|
|
|
or ecx, ecx |
|
|
jz .fifth |
|
|
|
|
|
bts eax, 2 |
|
|
|
|
|
.fifth movdqa xmm0, [esi] |
|
|
pand xmm0, xmm7 |
|
|
movdqa xmm1, [esi+16] |
|
|
|
|
|
por xmm0, [esi+32] |
|
|
por xmm1, [esi+48] |
|
|
por xmm0, [esi+64] |
|
|
por xmm1, [esi+80] |
|
|
por xmm0, [esi+96] |
|
|
por xmm1, [esi+112] |
|
|
|
|
|
por xmm0, xmm1 |
|
|
psadbw xmm0, xmm6 |
|
|
movhlps xmm1, xmm0 |
|
|
por xmm0, xmm1 |
|
|
movd ecx, xmm0 |
|
|
|
|
|
add esi, 128 |
|
|
|
|
|
or ecx, ecx |
|
|
jz .sixth |
|
|
|
|
|
bts eax, 1 |
|
|
|
|
|
.sixth movdqa xmm0, [esi] |
|
|
pand xmm0, xmm7 |
|
|
movdqa xmm1, [esi+16] |
|
|
|
|
|
por xmm0, [esi+32] |
|
|
por xmm1, [esi+48] |
|
|
por xmm0, [esi+64] |
|
|
por xmm1, [esi+80] |
|
|
por xmm0, [esi+96] |
|
|
por xmm1, [esi+112] |
|
|
|
|
|
por xmm0, xmm1 |
|
|
psadbw xmm0, xmm6 |
|
|
movhlps xmm1, xmm0 |
|
|
por xmm0, xmm1 |
|
|
movd ecx, xmm0 |
|
|
|
|
|
or ecx, ecx |
|
|
jz .end |
|
|
|
|
|
bts eax, 0 |
|
|
|
|
|
.end pop esi |
|
|
|
|
|
ret |
|
Parent Directory
| 
Revision Log
| 
Patch