1/*
2 *  sh4 dsputil
3 *
4 * Copyright (c) 2003 BERO <bero@geocities.co.jp>
5 *
6 * This file is part of FFmpeg.
7 *
8 * FFmpeg is free software; you can redistribute it and/or
9 * modify it under the terms of the GNU Lesser General Public
10 * License as published by the Free Software Foundation; either
11 * version 2.1 of the License, or (at your option) any later version.
12 *
13 * FFmpeg is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
16 * Lesser General Public License for more details.
17 *
18 * You should have received a copy of the GNU Lesser General Public
19 * License along with FFmpeg; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21 */
22
23#include "libavcodec/avcodec.h"
24#include "libavcodec/dsputil.h"
25#include "sh4.h"
26
27static void memzero_align8(void *dst,size_t size)
28{
29        int fpscr;
30        fp_single_enter(fpscr);
31        dst = (char *)dst + size;
32        size /= 32;
33        __asm__ volatile (
34        " fldi0 fr0\n"
35        " fldi0 fr1\n"
36        " fschg\n"  // double
37        "1: \n" \
38        " dt %1\n"
39        " fmov  dr0,@-%0\n"
40        " fmov  dr0,@-%0\n"
41        " fmov  dr0,@-%0\n"
42        " bf.s 1b\n"
43        " fmov  dr0,@-%0\n"
44        " fschg" //back to single
45        : "+r"(dst),"+r"(size) :: "memory" );
46        fp_single_leave(fpscr);
47}
48
49static void clear_blocks_sh4(DCTELEM *blocks)
50{
51        memzero_align8(blocks,sizeof(DCTELEM)*6*64);
52}
53
54void idct_sh4(DCTELEM *block);
55static void idct_put(uint8_t *dest, int line_size, DCTELEM *block)
56{
57        int i;
58        uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
59        idct_sh4(block);
60        for(i=0;i<8;i++) {
61                dest[0] = cm[block[0]];
62                dest[1] = cm[block[1]];
63                dest[2] = cm[block[2]];
64                dest[3] = cm[block[3]];
65                dest[4] = cm[block[4]];
66                dest[5] = cm[block[5]];
67                dest[6] = cm[block[6]];
68                dest[7] = cm[block[7]];
69                dest+=line_size;
70                block+=8;
71        }
72}
73static void idct_add(uint8_t *dest, int line_size, DCTELEM *block)
74{
75        int i;
76        uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
77        idct_sh4(block);
78        for(i=0;i<8;i++) {
79                dest[0] = cm[dest[0]+block[0]];
80                dest[1] = cm[dest[1]+block[1]];
81                dest[2] = cm[dest[2]+block[2]];
82                dest[3] = cm[dest[3]+block[3]];
83                dest[4] = cm[dest[4]+block[4]];
84                dest[5] = cm[dest[5]+block[5]];
85                dest[6] = cm[dest[6]+block[6]];
86                dest[7] = cm[dest[7]+block[7]];
87                dest+=line_size;
88                block+=8;
89        }
90}
91
92void dsputil_init_align(DSPContext* c, AVCodecContext *avctx);
93
94void dsputil_init_sh4(DSPContext* c, AVCodecContext *avctx)
95{
96        const int idct_algo= avctx->idct_algo;
97        dsputil_init_align(c,avctx);
98
99        c->clear_blocks = clear_blocks_sh4;
100        if(idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_SH4){
101                c->idct_put = idct_put;
102                c->idct_add = idct_add;
103               c->idct     = idct_sh4;
104                c->idct_permutation_type= FF_NO_IDCT_PERM;
105        }
106}
107