1 |
;/***************************************************************************** |
;/***************************************************************************** |
2 |
; * |
; * |
3 |
; * XVID MPEG-4 VIDEO CODEC |
; * XVID MPEG-4 VIDEO CODEC |
4 |
; * 3dnow 8x8 block-based halfpel interpolation |
; * - 3dnow 8x8 block-based halfpel interpolation - |
5 |
; * |
; * |
6 |
; * Copyright(C) 2002 Peter Ross <pross@xvid.org> |
; * Copyright(C) 2001 Peter Ross <pross@xvid.org> |
7 |
; * Copyright(C) 2002 Michael Militzer <michael@xvid.org> |
; * 2002 Michael Militzer <isibaar@xvid.org> |
8 |
|
; * 2002 Pascal Massimino <skal@planet-d.net> |
9 |
; * |
; * |
10 |
; * 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 |
11 |
; * |
; * 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 |
|
12 |
; * the Free Software Foundation; either version 2 of the License, or |
; * the Free Software Foundation; either version 2 of the License, or |
13 |
; * (at your option) any later version. |
; * (at your option) any later version. |
14 |
; * |
; * |
21 |
; * along with this program; if not, write to the Free Software |
; * along with this program; if not, write to the Free Software |
22 |
; * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
; * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
23 |
; * |
; * |
|
; * 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: interpolate8x8_3dn.asm,v 1.3 2002-11-17 00:20:30 edgomez Exp $ |
|
|
; * |
|
24 |
; ****************************************************************************/ |
; ****************************************************************************/ |
25 |
|
|
26 |
bits 32 |
BITS 32 |
27 |
|
|
28 |
%macro cglobal 1 |
%macro cglobal 1 |
29 |
%ifdef PREFIX |
%ifdef PREFIX |
34 |
%endif |
%endif |
35 |
%endmacro |
%endmacro |
36 |
|
|
37 |
section .data |
;============================================================================= |
38 |
|
; Read Only data |
39 |
|
;============================================================================= |
40 |
|
|
41 |
align 16 |
%ifdef FORMAT_COFF |
42 |
|
SECTION .rodata |
43 |
|
%else |
44 |
|
SECTION .rodata align=16 |
45 |
|
%endif |
46 |
|
|
47 |
mmx_one |
ALIGN 16 |
48 |
|
mmx_one: |
49 |
times 8 db 1 |
times 8 db 1 |
50 |
|
|
51 |
section .text |
;============================================================================= |
52 |
|
; Code |
53 |
|
;============================================================================= |
54 |
|
|
55 |
|
SECTION .text |
56 |
|
|
57 |
cglobal interpolate8x8_halfpel_h_3dn |
cglobal interpolate8x8_halfpel_h_3dn |
58 |
cglobal interpolate8x8_halfpel_v_3dn |
cglobal interpolate8x8_halfpel_v_3dn |
59 |
cglobal interpolate8x8_halfpel_hv_3dn |
cglobal interpolate8x8_halfpel_hv_3dn |
60 |
|
|
61 |
|
;----------------------------------------------------------------------------- |
|
;=========================================================================== |
|
62 |
; |
; |
63 |
; void interpolate8x8_halfpel_h_3dn(uint8_t * const dst, |
; void interpolate8x8_halfpel_h_3dn(uint8_t * const dst, |
64 |
; const uint8_t * const src, |
; const uint8_t * const src, |
65 |
; const uint32_t stride, |
; const uint32_t stride, |
66 |
; const uint32_t rounding); |
; const uint32_t rounding); |
67 |
; |
; |
68 |
;=========================================================================== |
;----------------------------------------------------------------------------- |
69 |
|
|
70 |
%macro COPY_H_3DN_RND0 0 |
%macro COPY_H_3DN_RND0 0 |
71 |
movq mm0, [eax] |
movq mm0, [eax] |
97 |
movq [ecx+edx], mm1 |
movq [ecx+edx], mm1 |
98 |
%endmacro |
%endmacro |
99 |
|
|
100 |
align 16 |
ALIGN 16 |
101 |
interpolate8x8_halfpel_h_3dn: |
interpolate8x8_halfpel_h_3dn: |
102 |
|
|
103 |
mov eax, [esp+16] ; rounding |
mov eax, [esp+16] ; rounding |
130 |
ret |
ret |
131 |
|
|
132 |
|
|
133 |
;=========================================================================== |
;----------------------------------------------------------------------------- |
134 |
; |
; |
135 |
; void interpolate8x8_halfpel_v_3dn(uint8_t * const dst, |
; void interpolate8x8_halfpel_v_3dn(uint8_t * const dst, |
136 |
; const uint8_t * const src, |
; const uint8_t * const src, |
137 |
; const uint32_t stride, |
; const uint32_t stride, |
138 |
; const uint32_t rounding); |
; const uint32_t rounding); |
139 |
; |
; |
140 |
;=========================================================================== |
;----------------------------------------------------------------------------- |
141 |
|
|
142 |
%macro COPY_V_3DN_RND0 0 |
%macro COPY_V_3DN_RND0 0 |
143 |
movq mm0, [eax] |
movq mm0, [eax] |
168 |
movq [ecx+edx], mm1 |
movq [ecx+edx], mm1 |
169 |
%endmacro |
%endmacro |
170 |
|
|
171 |
align 16 |
ALIGN 16 |
172 |
interpolate8x8_halfpel_v_3dn: |
interpolate8x8_halfpel_v_3dn: |
173 |
|
|
174 |
mov eax, [esp+16] ; rounding |
mov eax, [esp+16] ; rounding |
206 |
ret |
ret |
207 |
|
|
208 |
|
|
209 |
;=========================================================================== |
;----------------------------------------------------------------------------- |
210 |
; |
; |
211 |
; void interpolate8x8_halfpel_hv_3dn(uint8_t * const dst, |
; void interpolate8x8_halfpel_hv_3dn(uint8_t * const dst, |
212 |
; const uint8_t * const src, |
; const uint8_t * const src, |
214 |
; const uint32_t rounding); |
; const uint32_t rounding); |
215 |
; |
; |
216 |
; |
; |
217 |
;=========================================================================== |
;----------------------------------------------------------------------------- |
218 |
|
|
219 |
; The trick is to correct the result of 'pavgusb' with some combination of the |
; The trick is to correct the result of 'pavgusb' with some combination of the |
220 |
; lsb's of the 4 input values i,j,k,l, and their intermediate 'pavgusb' (s and t). |
; lsb's of the 4 input values i,j,k,l, and their intermediate 'pavgusb' (s and t). |
306 |
movq [ecx], mm0 |
movq [ecx], mm0 |
307 |
%endmacro |
%endmacro |
308 |
|
|
309 |
align 16 |
ALIGN 16 |
310 |
interpolate8x8_halfpel_hv_3dn |
interpolate8x8_halfpel_hv_3dn |
311 |
mov eax, [esp+16] ; rounding |
mov eax, [esp+16] ; rounding |
312 |
mov ecx, [esp+ 4] ; Dst |
mov ecx, [esp+ 4] ; Dst |