2 * simple math operations
3 * Copyright (c) 2001, 2002 Fabrice Bellard
4 * Copyright (c) 2006 Michael Niedermayer <michaelni@gmx.at> et al
6 * This file is part of FFmpeg.
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.
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.
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
22 #ifndef AVCODEC_MATHOPS_H
23 #define AVCODEC_MATHOPS_H
27 #include "libavutil/common.h"
30 extern const uint32_t ff_inverse[257];
31 extern const uint8_t ff_sqrt_tab[256];
34 # include "arm/mathops.h"
36 # include "avr32/mathops.h"
38 # include "bfin/mathops.h"
40 # include "mips/mathops.h"
42 # include "ppc/mathops.h"
44 # include "x86/mathops.h"
47 /* generic implementation */
50 # define MUL64(a,b) ((int64_t)(a) * (int64_t)(b))
54 # define MULL(a,b,s) (MUL64(a, b) >> (s))
58 static av_always_inline int MULH(int a, int b){
59 return MUL64(a, b) >> 32;
64 static av_always_inline unsigned UMULH(unsigned a, unsigned b){
65 return ((uint64_t)(a) * (uint64_t)(b))>>32;
70 # define MAC64(d, a, b) ((d) += MUL64(a, b))
74 # define MLS64(d, a, b) ((d) -= MUL64(a, b))
77 /* signed 16x16 -> 32 multiply add accumulate */
79 # define MAC16(rt, ra, rb) rt += (ra) * (rb)
82 /* signed 16x16 -> 32 multiply */
84 # define MUL16(ra, rb) ((ra) * (rb))
88 # define MLS16(rt, ra, rb) ((rt) -= (ra) * (rb))
93 #define mid_pred mid_pred
94 static inline av_const int mid_pred(int a, int b, int c)
97 int t= (a-b)&((a-b)>>31);
100 b-= (b-c)&((b-c)>>31);
101 b+= (a-b)&((a-b)>>31);
122 static inline av_const int sign_extend(int val, unsigned bits)
124 unsigned shift = 8 * sizeof(int) - bits;
125 union { unsigned u; int s; } v = { (unsigned) val << shift };
131 static inline av_const unsigned zero_extend(unsigned val, unsigned bits)
133 return (val << ((8 * sizeof(int)) - bits)) >> ((8 * sizeof(int)) - bits);
138 #define COPY3_IF_LT(x, y, a, b, c, d)\
147 #define MASK_ABS(mask, level) do { \
148 mask = level >> 31; \
149 level = (level ^ mask) - mask; \
154 # define NEG_SSR32(a,s) ((( int32_t)(a))>>(32-(s)))
158 # define NEG_USR32(a,s) (((uint32_t)(a))>>(32-(s)))
163 # define PACK_2U8(a,b) (((a) << 8) | (b))
166 # define PACK_4U8(a,b,c,d) (((a) << 24) | ((b) << 16) | ((c) << 8) | (d))
169 # define PACK_2U16(a,b) (((a) << 16) | (b))
173 # define PACK_2U8(a,b) (((b) << 8) | (a))
176 # define PACK_4U8(a,b,c,d) (((d) << 24) | ((c) << 16) | ((b) << 8) | (a))
179 # define PACK_2U16(a,b) (((b) << 16) | (a))
184 # define PACK_2S8(a,b) PACK_2U8((a)&255, (b)&255)
187 # define PACK_4S8(a,b,c,d) PACK_4U8((a)&255, (b)&255, (c)&255, (d)&255)
190 # define PACK_2S16(a,b) PACK_2U16((a)&0xffff, (b)&0xffff)
194 # define FASTDIV(a,b) ((uint32_t)((((uint64_t)a) * ff_inverse[b]) >> 32))
197 static inline av_const unsigned int ff_sqrt(unsigned int a)
201 if (a < 255) return (ff_sqrt_tab[a + 1] - 1) >> 4;
202 else if (a < (1 << 12)) b = ff_sqrt_tab[a >> 4] >> 2;
204 else if (a < (1 << 14)) b = ff_sqrt_tab[a >> 6] >> 1;
205 else if (a < (1 << 16)) b = ff_sqrt_tab[a >> 8] ;
208 int s = av_log2_16bit(a >> 16) >> 1;
209 unsigned int c = a >> (s + 2);
210 b = ff_sqrt_tab[c >> (s + 8)];
211 b = FASTDIV(c,b) + (b << s);
214 return b - (a < b * b);
217 #endif /* AVCODEC_MATHOPS_H */