view rational.c @ 728:1fa3820b1a84 libavutil

ARM asm for AV_RN*() ARMv6 and later support unaligned loads and stores for single word/halfword but not double/multiple. GCC is ignorant of this and will always use bytewise accesses for unaligned data. Casting to an int32_t pointer is dangerous since a load/store double or multiple instruction might be used (this happens with some code in FFmpeg). Implementing the AV_[RW]* macros with inline asm using only supported instructions gives fast and safe unaligned accesses. ARM RVCT does the right thing with generic code. This gives an overall speedup of up to 10%.

/*
 * rational numbers
 * Copyright (c) 2003 Michael Niedermayer <[e-mail]>
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file libavutil/rational.c
 * rational numbers
 * @author Michael Niedermayer <[e-mail]>
 */

#include <assert.h>
//#include <math.h>
#include <limits.h>

#include "common.h"
#include "mathematics.h"
#include "rational.h"

int av_reduce(int *dst_num, int *dst_den, int64_t num, int64_t den, int64_t max){
    AVRational a0={0,1}, a1={1,0};
    int sign= (num<0) ^ (den<0);
    int64_t gcd= av_gcd(FFABS(num), FFABS(den));

    if(gcd){
        num = FFABS(num)/gcd;
        den = FFABS(den)/gcd;
    }
    if(num<=max && den<=max){
        a1= (AVRational){num, den};
        den=0;
    }

    while(den){
        uint64_t x      = num / den;
        int64_t next_den= num - den*x;
        int64_t a2n= x*a1.num + a0.num;
        int64_t a2d= x*a1.den + a0.den;

        if(a2n > max || a2d > max){
            if(a1.num) x= (max - a0.num) / a1.num;
            if(a1.den) x= FFMIN(x, (max - a0.den) / a1.den);

            if (den*(2*x*a1.den + a0.den) > num*a1.den)
                a1 = (AVRational){x*a1.num + a0.num, x*a1.den + a0.den};
            break;
        }

        a0= a1;
        a1= (AVRational){a2n, a2d};
        num= den;
        den= next_den;
    }
    assert(av_gcd(a1.num, a1.den) <= 1U);

    *dst_num = sign ? -a1.num : a1.num;
    *dst_den = a1.den;

    return den==0;
}

AVRational av_mul_q(AVRational b, AVRational c){
    av_reduce(&b.num, &b.den, b.num * (int64_t)c.num, b.den * (int64_t)c.den, INT_MAX);
    return b;
}

AVRational av_div_q(AVRational b, AVRational c){
    return av_mul_q(b, (AVRational){c.den, c.num});
}

AVRational av_add_q(AVRational b, AVRational c){
    av_reduce(&b.num, &b.den, b.num * (int64_t)c.den + c.num * (int64_t)b.den, b.den * (int64_t)c.den, INT_MAX);
    return b;
}

AVRational av_sub_q(AVRational b, AVRational c){
    return av_add_q(b, (AVRational){-c.num, c.den});
}

AVRational av_d2q(double d, int max){
    AVRational a;
#define LOG2  0.69314718055994530941723212145817656807550013436025
    int exponent= FFMAX( (int)(log(fabs(d) + 1e-20)/LOG2), 0);
    int64_t den= 1LL << (61 - exponent);
    av_reduce(&a.num, &a.den, (int64_t)(d * den + 0.5), den, max);

    return a;
}

int av_nearer_q(AVRational q, AVRational q1, AVRational q2)
{
    /* n/d is q, a/b is the median between q1 and q2 */
    int64_t a = q1.num * (int64_t)q2.den + q2.num * (int64_t)q1.den;
    int64_t b = 2 * (int64_t)q1.den * q2.den;

    /* rnd_up(a*d/b) > n => a*d/b > n */
    int64_t x_up = av_rescale_rnd(a, q.den, b, AV_ROUND_UP);

    /* rnd_down(a*d/b) < n => a*d/b < n */
    int64_t x_down = av_rescale_rnd(a, q.den, b, AV_ROUND_DOWN);

    return ((x_up > q.num) - (x_down < q.num)) * av_cmp_q(q2, q1);
}

int av_find_nearest_q_idx(AVRational q, const AVRational* q_list)
{
    int i, nearest_q_idx = 0;
    for(i=0; q_list[i].den; i++)
        if (av_nearer_q(q, q_list[i], q_list[nearest_q_idx]) > 0)
            nearest_q_idx = i;

    return nearest_q_idx;
}