1/*
2 * rectangle filling function
3 * Copyright (c) 2003 Michael Niedermayer <michaelni@gmx.at>
4 *
5 * This file is part of FFmpeg.
6 *
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.
11 *
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.
16 *
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
20 */
21
22/**
23 * @file
24 * useful rectangle filling function
25 * @author Michael Niedermayer <michaelni@gmx.at>
26 */
27
28#ifndef AVCODEC_RECTANGLE_H
29#define AVCODEC_RECTANGLE_H
30
31#include <assert.h>
32#include "config.h"
33#include "libavutil/common.h"
34#include "dsputil.h"
35
36/**
37 * fill a rectangle.
38 * @param h height of the rectangle, should be a constant
39 * @param w width of the rectangle, should be a constant
40 * @param size the size of val (1, 2 or 4), should be a constant
41 */
42static av_always_inline void fill_rectangle(void *vp, int w, int h, int stride, uint32_t val, int size){
43    uint8_t *p= (uint8_t*)vp;
44    assert(size==1 || size==2 || size==4);
45    assert(w<=4);
46
47    w      *= size;
48    stride *= size;
49
50    assert((((long)vp)&(FFMIN(w, STRIDE_ALIGN)-1)) == 0);
51    assert((stride&(w-1))==0);
52    if(w==2){
53        const uint16_t v= size==4 ? val : val*0x0101;
54        *(uint16_t*)(p + 0*stride)= v;
55        if(h==1) return;
56        *(uint16_t*)(p + 1*stride)= v;
57        if(h==2) return;
58        *(uint16_t*)(p + 2*stride)= v;
59        *(uint16_t*)(p + 3*stride)= v;
60    }else if(w==4){
61        const uint32_t v= size==4 ? val : size==2 ? val*0x00010001 : val*0x01010101;
62        *(uint32_t*)(p + 0*stride)= v;
63        if(h==1) return;
64        *(uint32_t*)(p + 1*stride)= v;
65        if(h==2) return;
66        *(uint32_t*)(p + 2*stride)= v;
67        *(uint32_t*)(p + 3*stride)= v;
68    }else if(w==8){
69    //gcc can't optimize 64bit math on x86_32
70#if HAVE_FAST_64BIT
71        const uint64_t v=  size==2 ? val*0x0001000100010001ULL : val*0x0100000001ULL;
72        *(uint64_t*)(p + 0*stride)= v;
73        if(h==1) return;
74        *(uint64_t*)(p + 1*stride)= v;
75        if(h==2) return;
76        *(uint64_t*)(p + 2*stride)= v;
77        *(uint64_t*)(p + 3*stride)= v;
78    }else if(w==16){
79        const uint64_t v= val*0x0100000001ULL;
80        *(uint64_t*)(p + 0+0*stride)= v;
81        *(uint64_t*)(p + 8+0*stride)= v;
82        *(uint64_t*)(p + 0+1*stride)= v;
83        *(uint64_t*)(p + 8+1*stride)= v;
84        if(h==2) return;
85        *(uint64_t*)(p + 0+2*stride)= v;
86        *(uint64_t*)(p + 8+2*stride)= v;
87        *(uint64_t*)(p + 0+3*stride)= v;
88        *(uint64_t*)(p + 8+3*stride)= v;
89#else
90        const uint32_t v= size==2 ? val*0x00010001 : val;
91        *(uint32_t*)(p + 0+0*stride)= v;
92        *(uint32_t*)(p + 4+0*stride)= v;
93        if(h==1) return;
94        *(uint32_t*)(p + 0+1*stride)= v;
95        *(uint32_t*)(p + 4+1*stride)= v;
96        if(h==2) return;
97        *(uint32_t*)(p + 0+2*stride)= v;
98        *(uint32_t*)(p + 4+2*stride)= v;
99        *(uint32_t*)(p + 0+3*stride)= v;
100        *(uint32_t*)(p + 4+3*stride)= v;
101    }else if(w==16){
102        *(uint32_t*)(p + 0+0*stride)= val;
103        *(uint32_t*)(p + 4+0*stride)= val;
104        *(uint32_t*)(p + 8+0*stride)= val;
105        *(uint32_t*)(p +12+0*stride)= val;
106        *(uint32_t*)(p + 0+1*stride)= val;
107        *(uint32_t*)(p + 4+1*stride)= val;
108        *(uint32_t*)(p + 8+1*stride)= val;
109        *(uint32_t*)(p +12+1*stride)= val;
110        if(h==2) return;
111        *(uint32_t*)(p + 0+2*stride)= val;
112        *(uint32_t*)(p + 4+2*stride)= val;
113        *(uint32_t*)(p + 8+2*stride)= val;
114        *(uint32_t*)(p +12+2*stride)= val;
115        *(uint32_t*)(p + 0+3*stride)= val;
116        *(uint32_t*)(p + 4+3*stride)= val;
117        *(uint32_t*)(p + 8+3*stride)= val;
118        *(uint32_t*)(p +12+3*stride)= val;
119#endif
120    }else
121        assert(0);
122    assert(h==4);
123}
124
125#endif /* AVCODEC_RECTANGLE_H */
126