1 |
;/***************************************************************************** |
;/***************************************************************************** |
2 |
; * |
; * |
3 |
; * XVID MPEG-4 VIDEO CODEC |
; * XVID MPEG-4 VIDEO CODEC |
4 |
; * xmm 8x8 block-based halfpel interpolation |
; * - mmx 8x8 block-based halfpel interpolation - |
5 |
; * |
; * |
6 |
; * Copyright(C) 2002 Michael Militzer <michael@xvid.org> |
; * Copyright(C) 2002 Michael Militzer <isibaar@xvid.org> |
7 |
; * Copyright(C) 2002 -Skal- |
; * 2002 Pascal Massimino <skal@planet-d.net> |
8 |
; * |
; * |
9 |
; * This file is part of XviD, a free MPEG-4 video encoder/decoder |
; * This program is free software ; you can redistribute it and/or modify |
10 |
; * |
; * it under the terms of the GNU General Public License as published by |
|
; * XviD is free software; you can redistribute it and/or modify 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 |
; * |
; * |
|
; * 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. |
|
|
; * |
|
|
; * $Id$ |
|
|
; * |
|
23 |
; ****************************************************************************/ |
; ****************************************************************************/ |
24 |
|
|
25 |
bits 32 |
BITS 32 |
26 |
|
|
27 |
%macro cglobal 1 |
%macro cglobal 1 |
28 |
%ifdef PREFIX |
%ifdef PREFIX |
33 |
%endif |
%endif |
34 |
%endmacro |
%endmacro |
35 |
|
|
36 |
section .data |
;============================================================================= |
37 |
|
; Read only data |
38 |
|
;============================================================================= |
39 |
|
|
40 |
align 16 |
%ifdef FORMAT_COFF |
41 |
|
SECTION .rodata |
42 |
|
%else |
43 |
|
SECTION .rodata align=16 |
44 |
|
%endif |
45 |
|
|
46 |
mmx_one |
ALIGN 16 |
47 |
|
mmx_one: |
48 |
times 8 db 1 |
times 8 db 1 |
49 |
|
|
50 |
section .text |
SECTION .text |
51 |
|
|
52 |
cglobal interpolate8x8_halfpel_h_xmm |
cglobal interpolate8x8_halfpel_h_xmm |
53 |
cglobal interpolate8x8_halfpel_v_xmm |
cglobal interpolate8x8_halfpel_v_xmm |
92 |
movq [ecx+edx], mm1 |
movq [ecx+edx], mm1 |
93 |
%endmacro |
%endmacro |
94 |
|
|
95 |
align 16 |
ALIGN 16 |
96 |
interpolate8x8_halfpel_h_xmm: |
interpolate8x8_halfpel_h_xmm: |
97 |
|
|
98 |
mov eax, [esp+16]; rounding |
mov eax, [esp+16]; rounding |
162 |
movq [ecx+edx], mm1 |
movq [ecx+edx], mm1 |
163 |
%endmacro |
%endmacro |
164 |
|
|
165 |
align 16 |
ALIGN 16 |
166 |
interpolate8x8_halfpel_v_xmm: |
interpolate8x8_halfpel_v_xmm: |
167 |
|
|
168 |
mov eax, [esp+16]; rounding |
mov eax, [esp+16]; rounding |
172 |
mov edx, [esp+12] ; stride |
mov edx, [esp+12] ; stride |
173 |
|
|
174 |
; we process 2 line at a time |
; we process 2 line at a time |
|
|
|
175 |
jnz near .rounding1 |
jnz near .rounding1 |
176 |
|
|
177 |
COPY_V_SSE_RND0 |
COPY_V_SSE_RND0 |
298 |
movq [ecx], mm0 |
movq [ecx], mm0 |
299 |
%endmacro |
%endmacro |
300 |
|
|
301 |
align 16 |
ALIGN 16 |
302 |
interpolate8x8_halfpel_hv_xmm: |
interpolate8x8_halfpel_hv_xmm: |
303 |
mov eax, [esp+16] ; rounding |
mov eax, [esp+16] ; rounding |
304 |
mov ecx, [esp+ 4] ; Dst |
mov ecx, [esp+ 4] ; Dst |