2 * MMI optimized DSP utils
3 * Copyright (c) 2000, 2001 Fabrice Bellard.
5 * This file is part of FFmpeg.
7 * FFmpeg is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
12 * FFmpeg is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 * Lesser General Public License for more details.
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with FFmpeg; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21 * MMI optimization by Leon van Stuivenberg
22 * clear_blocks_mmi() by BroadQ
25 #include "../dsputil.h"
28 void ff_mmi_idct_put(uint8_t *dest, int line_size, DCTELEM *block);
29 void ff_mmi_idct_add(uint8_t *dest, int line_size, DCTELEM *block);
30 void ff_mmi_idct(DCTELEM *block);
32 static void clear_blocks_mmi(DCTELEM * blocks)
36 "addiu $9, %0, 768 \n"
48 : "+r" (blocks) :: "$8", "$9", "memory" );
52 static void get_pixels_mmi(DCTELEM *block, const uint8_t *pixels, int line_size)
62 "pextlb $8, $0, $8 \n\t"
66 "pextlb $9, $0, $9 \n\t"
70 "pextlb $10, $0, $10 \n\t"
74 "pextlb $8, $0, $8 \n\t"
78 "pextlb $9, $0, $9 \n\t"
82 "pextlb $10, $0, $10 \n\t"
84 "pextlb $8, $0, $8 \n\t"
86 "pextlb $9, $0, $9 \n\t"
89 : "+r" (pixels) : "r" (block), "r" (line_size) : "$8", "$9", "$10", "memory" );
93 static void put_pixels8_mmi(uint8_t *block, const uint8_t *pixels, int line_size, int h)
100 "addiu %2, %2, -1 \n\t"
102 "add %1, %1, %3 \n\t"
104 "add %0, %0, %3 \n\t"
107 : "+r" (block), "+r" (pixels), "+r" (h) : "r" (line_size)
112 static void put_pixels16_mmi(uint8_t *block, const uint8_t *pixels, int line_size, int h)
119 "add $11, %1, %3 \n\t"
121 "add $10, %0, %3 \n\t"
123 "ldl $9, 15(%1) \n\t"
124 "ldr $12, 0($11) \n\t"
125 "add %1, $11, %3 \n\t"
126 "ldl $12, 7($11) \n\t"
127 "pcpyld $8, $9, $8 \n\t"
129 "ldr $13, 8($11) \n\t"
130 "addiu %2, %2, -2 \n\t"
131 "ldl $13, 15($11) \n\t"
132 "add %0, $10, %3 \n\t"
133 "pcpyld $12, $13, $12 \n\t"
134 "sq $12, 0($10) \n\t"
137 : "+r" (block), "+r" (pixels), "+r" (h) : "r" (line_size)
138 : "$8", "$9", "$10", "$11", "$12", "$13", "memory" );
142 void dsputil_init_mmi(DSPContext* c, AVCodecContext *avctx)
144 const int idct_algo= avctx->idct_algo;
146 c->clear_blocks = clear_blocks_mmi;
148 c->put_pixels_tab[1][0] = put_pixels8_mmi;
149 c->put_no_rnd_pixels_tab[1][0] = put_pixels8_mmi;
151 c->put_pixels_tab[0][0] = put_pixels16_mmi;
152 c->put_no_rnd_pixels_tab[0][0] = put_pixels16_mmi;
154 c->get_pixels = get_pixels_mmi;
156 if(idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_PS2){
157 c->idct_put= ff_mmi_idct_put;
158 c->idct_add= ff_mmi_idct_add;
159 c->idct = ff_mmi_idct;
160 c->idct_permutation_type= FF_LIBMPEG2_IDCT_PERM;