3 |
; * XVID MPEG-4 VIDEO CODEC |
; * XVID MPEG-4 VIDEO CODEC |
4 |
; * mmx 8bit<->16bit transfers |
; * mmx 8bit<->16bit transfers |
5 |
; * |
; * |
6 |
; * This file is part of XviD, a free MPEG-4 video encoder/decoder |
; * This program is an implementation of a part of one or more MPEG-4 |
7 |
|
; * Video tools as specified in ISO/IEC 14496-2 standard. Those intending |
8 |
|
; * to use this software module in hardware or software products are |
9 |
|
; * advised that its use may infringe existing patents or copyrights, and |
10 |
|
; * any such use would be at such party's own risk. The original |
11 |
|
; * developer of this software module and his/her company, and subsequent |
12 |
|
; * editors and their companies, will have no liability for use of this |
13 |
|
; * software or modifications or derivatives thereof. |
14 |
; * |
; * |
15 |
; * XviD is free software; you can redistribute it and/or modify it |
; * This program is free software; you can redistribute it and/or modify |
16 |
; * 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 |
17 |
; * the Free Software Foundation; either version 2 of the License, or |
; * the Free Software Foundation; either version 2 of the License, or |
18 |
; * (at your option) any later version. |
; * (at your option) any later version. |
19 |
; * |
; * |
24 |
; * |
; * |
25 |
; * You should have received a copy of the GNU General Public License |
; * You should have received a copy of the GNU General Public License |
26 |
; * along with this program; if not, write to the Free Software |
; * along with this program; if not, write to the Free Software |
27 |
; * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
; * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
|
; * |
|
|
; * Under section 8 of the GNU General Public License, the copyright |
|
|
; * holders of XVID explicitly forbid distribution in the following |
|
|
; * 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: mem_transfer_mmx.asm,v 1.8 2002-11-17 00:51:11 edgomez Exp $ |
|
28 |
; * |
; * |
29 |
; *************************************************************************/ |
; *************************************************************************/ |
30 |
|
|
35 |
; * 04.06.2002 speed enhancement (unroll+overlap). -Skal- |
; * 04.06.2002 speed enhancement (unroll+overlap). -Skal- |
36 |
; * + added transfer_8to16sub2_mmx/xmm |
; * + added transfer_8to16sub2_mmx/xmm |
37 |
; * 07.01.2002 merge functions from compensate_mmx; rename functions |
; * 07.01.2002 merge functions from compensate_mmx; rename functions |
38 |
; * 07.11.2001 initial version; (c)2001 peter ross <pross@xvid.org> |
; * 07.11.2001 initial version; (c)2001 peter ross <pross@cs.rmit.edu.au> |
39 |
; * |
; * |
40 |
; *************************************************************************/ |
; *************************************************************************/ |
41 |
|
|
57 |
cglobal transfer_8to16copy_mmx |
cglobal transfer_8to16copy_mmx |
58 |
cglobal transfer_16to8copy_mmx |
cglobal transfer_16to8copy_mmx |
59 |
cglobal transfer_8to16sub_mmx |
cglobal transfer_8to16sub_mmx |
60 |
|
cglobal transfer_8to16subro_mmx |
61 |
cglobal transfer_8to16sub2_mmx |
cglobal transfer_8to16sub2_mmx |
62 |
cglobal transfer_8to16sub2_xmm |
cglobal transfer_8to16sub2_xmm |
63 |
cglobal transfer_16to8add_mmx |
cglobal transfer_16to8add_mmx |
152 |
; * 02.12.2001 loop unrolled, code runs 10% faster now (Isibaar) |
; * 02.12.2001 loop unrolled, code runs 10% faster now (Isibaar) |
153 |
; * 30.11.2001 16 pixels are processed per iteration (Isibaar) |
; * 30.11.2001 16 pixels are processed per iteration (Isibaar) |
154 |
; * 30.11.2001 .text missing |
; * 30.11.2001 .text missing |
155 |
; * 06.11.2001 inital version; (c)2001 peter ross <pross@xvid.org> |
; * 06.11.2001 inital version; (c)2001 peter ross <pross@cs.rmit.edu.au> |
156 |
; * |
; * |
157 |
; *************************************************************************/ |
; *************************************************************************/ |
158 |
|
|
159 |
%macro COPY_8_TO_16_SUB 1 |
; when second argument == 1, reference (ebx) block is to current (eax) |
160 |
|
%macro COPY_8_TO_16_SUB 2 |
161 |
movq mm0, [eax] ; cur |
movq mm0, [eax] ; cur |
162 |
movq mm2, [eax+edx] |
movq mm2, [eax+edx] |
163 |
movq mm1, mm0 |
movq mm1, mm0 |
171 |
movq mm5, [ebx+edx] ; ref |
movq mm5, [ebx+edx] ; ref |
172 |
|
|
173 |
movq mm6, mm4 |
movq mm6, mm4 |
174 |
|
%if %2 == 1 |
175 |
movq [eax], mm4 |
movq [eax], mm4 |
176 |
movq [eax+edx], mm5 |
movq [eax+edx], mm5 |
177 |
|
%endif |
178 |
punpcklbw mm4, mm7 |
punpcklbw mm4, mm7 |
179 |
punpckhbw mm6, mm7 |
punpckhbw mm6, mm7 |
180 |
psubsw mm0, mm4 |
psubsw mm0, mm4 |
202 |
mov edx, [esp+4+16] ; Stride |
mov edx, [esp+4+16] ; Stride |
203 |
pxor mm7, mm7 |
pxor mm7, mm7 |
204 |
|
|
205 |
COPY_8_TO_16_SUB 0 |
COPY_8_TO_16_SUB 0, 1 |
206 |
COPY_8_TO_16_SUB 1 |
COPY_8_TO_16_SUB 1, 1 |
207 |
COPY_8_TO_16_SUB 2 |
COPY_8_TO_16_SUB 2, 1 |
208 |
COPY_8_TO_16_SUB 3 |
COPY_8_TO_16_SUB 3, 1 |
209 |
|
|
210 |
pop ebx |
pop ebx |
211 |
ret |
ret |
212 |
|
|
213 |
|
|
214 |
|
align 16 |
215 |
|
transfer_8to16subro_mmx: |
216 |
|
mov ecx, [esp + 4] ; Dst |
217 |
|
mov eax, [esp + 8] ; Cur |
218 |
|
push ebx |
219 |
|
mov ebx, [esp+4+12] ; Ref |
220 |
|
mov edx, [esp+4+16] ; Stride |
221 |
|
pxor mm7, mm7 |
222 |
|
|
223 |
|
COPY_8_TO_16_SUB 0, 0 |
224 |
|
COPY_8_TO_16_SUB 1, 0 |
225 |
|
COPY_8_TO_16_SUB 2, 0 |
226 |
|
COPY_8_TO_16_SUB 3, 0 |
227 |
|
|
228 |
|
pop ebx |
229 |
|
ret |
230 |
|
|
231 |
|
|
232 |
;=========================================================================== |
;=========================================================================== |
233 |
; |
; |
234 |
; void transfer_8to16sub2_mmx(int16_t * const dct, |
; void transfer_8to16sub2_mmx(int16_t * const dct, |