utils.c

/*
 * utils for libavcodec
 * Copyright (c) 2001 Fabrice Bellard
 * Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
 *
 * This file is part of Libav.
 *
 * Libav 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.
 *
 * Libav 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 Libav; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * utils.
 */

#include "libavutil/avassert.h"
#include "libavutil/avstring.h"
#include "libavutil/crc.h"
#include "libavutil/mathematics.h"
#include "libavutil/pixdesc.h"
#include "libavutil/audioconvert.h"
#include "libavutil/imgutils.h"
#include "libavutil/samplefmt.h"
#include "libavutil/dict.h"
#include "avcodec.h"
#include "dsputil.h"
#include "libavutil/opt.h"
#include "imgconvert.h"
#include "thread.h"
#include "audioconvert.h"
#include "internal.h"
#include "bytestream.h"
#include <stdlib.h>
#include <stdarg.h>
#include <limits.h>
#include <float.h>

static int volatile entangled_thread_counter=0;
static int (*ff_lockmgr_cb)(void **mutex, enum AVLockOp op);
static void *codec_mutex;
static void *avformat_mutex;

void *av_fast_realloc(void *ptr, unsigned int *size, size_t min_size)
{
    if(min_size < *size)
        return ptr;

    min_size= FFMAX(17*min_size/16 + 32, min_size);

    ptr= av_realloc(ptr, min_size);
    if(!ptr) //we could set this to the unmodified min_size but this is safer if the user lost the ptr and uses NULL now
        min_size= 0;

    *size= min_size;

    return ptr;
}

void av_fast_malloc(void *ptr, unsigned int *size, size_t min_size)
{
    void **p = ptr;
    if (min_size < *size)
        return;
    min_size= FFMAX(17*min_size/16 + 32, min_size);
    av_free(*p);
    *p = av_malloc(min_size);
    if (!*p) min_size = 0;
    *size= min_size;
}

void av_fast_padded_malloc(void *ptr, unsigned int *size, size_t min_size)
{
    void **p = ptr;
    if (min_size > SIZE_MAX - FF_INPUT_BUFFER_PADDING_SIZE) {
        av_freep(p);
        *size = 0;
        return;
    }
    av_fast_malloc(p, size, min_size + FF_INPUT_BUFFER_PADDING_SIZE);
    if (*size)
        memset((uint8_t *)*p + min_size, 0, FF_INPUT_BUFFER_PADDING_SIZE);
}

/* encoder management */
static AVCodec *first_avcodec = NULL;

AVCodec *av_codec_next(AVCodec *c){
    if(c) return c->next;
    else  return first_avcodec;
}

static void avcodec_init(void)
{
    static int initialized = 0;

    if (initialized != 0)
        return;
    initialized = 1;

    ff_dsputil_static_init();
}

int av_codec_is_encoder(AVCodec *codec)
{
    return codec && (codec->encode || codec->encode2);
}

int av_codec_is_decoder(AVCodec *codec)
{
    return codec && codec->decode;
}

void avcodec_register(AVCodec *codec)
{
    AVCodec **p;
    avcodec_init();
    p = &first_avcodec;
    while (*p != NULL) p = &(*p)->next;
    *p = codec;
    codec->next = NULL;

    if (codec->init_static_data)
        codec->init_static_data(codec);
}

unsigned avcodec_get_edge_width(void)
{
    return EDGE_WIDTH;
}

void avcodec_set_dimensions(AVCodecContext *s, int width, int height){
    s->coded_width = width;
    s->coded_height= height;
    s->width  = width;
    s->height = height;
}

#define INTERNAL_BUFFER_SIZE (32+1)

void avcodec_align_dimensions2(AVCodecContext *s, int *width, int *height,
                               int linesize_align[AV_NUM_DATA_POINTERS])
{
    int i;
    int w_align= 1;
    int h_align= 1;

    switch(s->pix_fmt){
    case PIX_FMT_YUV420P:
    case PIX_FMT_YUYV422:
    case PIX_FMT_UYVY422:
    case PIX_FMT_YUV422P:
    case PIX_FMT_YUV440P:
    case PIX_FMT_YUV444P:
    case PIX_FMT_GBRP:
    case PIX_FMT_GRAY8:
    case PIX_FMT_GRAY16BE:
    case PIX_FMT_GRAY16LE:
    case PIX_FMT_YUVJ420P:
    case PIX_FMT_YUVJ422P:
    case PIX_FMT_YUVJ440P:
    case PIX_FMT_YUVJ444P:
    case PIX_FMT_YUVA420P:
    case PIX_FMT_YUV420P9LE:
    case PIX_FMT_YUV420P9BE:
    case PIX_FMT_YUV420P10LE:
    case PIX_FMT_YUV420P10BE:
    case PIX_FMT_YUV422P9LE:
    case PIX_FMT_YUV422P9BE:
    case PIX_FMT_YUV422P10LE:
    case PIX_FMT_YUV422P10BE:
    case PIX_FMT_YUV444P9LE:
    case PIX_FMT_YUV444P9BE:
    case PIX_FMT_YUV444P10LE:
    case PIX_FMT_YUV444P10BE:
    case PIX_FMT_GBRP9LE:
    case PIX_FMT_GBRP9BE:
    case PIX_FMT_GBRP10LE:
    case PIX_FMT_GBRP10BE:
        w_align = 16; //FIXME assume 16 pixel per macroblock
        h_align = 16 * 2; // interlaced needs 2 macroblocks height
        break;
    case PIX_FMT_YUV411P:
    case PIX_FMT_UYYVYY411:
        w_align=32;
        h_align=8;
        break;
    case PIX_FMT_YUV410P:
        if(s->codec_id == AV_CODEC_ID_SVQ1){
            w_align=64;
            h_align=64;
        }
    case PIX_FMT_RGB555:
        if(s->codec_id == AV_CODEC_ID_RPZA){
            w_align=4;
            h_align=4;
        }
    case PIX_FMT_PAL8:
    case PIX_FMT_BGR8:
    case PIX_FMT_RGB8:
        if(s->codec_id == AV_CODEC_ID_SMC){
            w_align=4;
            h_align=4;
        }
        break;
    case PIX_FMT_BGR24:
        if((s->codec_id == AV_CODEC_ID_MSZH) || (s->codec_id == AV_CODEC_ID_ZLIB)){
            w_align=4;
            h_align=4;
        }
        break;
    default:
        w_align= 1;
        h_align= 1;
        break;
    }

    *width = FFALIGN(*width , w_align);
    *height= FFALIGN(*height, h_align);
    if (s->codec_id == AV_CODEC_ID_H264)
        *height+=2; // some of the optimized chroma MC reads one line too much

    for (i = 0; i < 4; i++)
        linesize_align[i] = STRIDE_ALIGN;
}

void avcodec_align_dimensions(AVCodecContext *s, int *width, int *height){
    int chroma_shift = av_pix_fmt_descriptors[s->pix_fmt].log2_chroma_w;
    int linesize_align[AV_NUM_DATA_POINTERS];
    int align;
    avcodec_align_dimensions2(s, width, height, linesize_align);
    align = FFMAX(linesize_align[0], linesize_align[3]);
    linesize_align[1] <<= chroma_shift;
    linesize_align[2] <<= chroma_shift;
    align = FFMAX3(align, linesize_align[1], linesize_align[2]);
    *width=FFALIGN(*width, align);
}

int avcodec_fill_audio_frame(AVFrame *frame, int nb_channels,
                             enum AVSampleFormat sample_fmt, const uint8_t *buf,
                             int buf_size, int align)
{
    int ch, planar, needed_size, ret = 0;

    needed_size = av_samples_get_buffer_size(NULL, nb_channels,
                                             frame->nb_samples, sample_fmt,
                                             align);
    if (buf_size < needed_size)
        return AVERROR(EINVAL);

    planar = av_sample_fmt_is_planar(sample_fmt);
    if (planar && nb_channels > AV_NUM_DATA_POINTERS) {
        if (!(frame->extended_data = av_mallocz(nb_channels *
                                                sizeof(*frame->extended_data))))
            return AVERROR(ENOMEM);
    } else {
        frame->extended_data = frame->data;
    }

    if ((ret = av_samples_fill_arrays(frame->extended_data, &frame->linesize[0],
                                      buf, nb_channels, frame->nb_samples,
                                      sample_fmt, align)) < 0) {
        if (frame->extended_data != frame->data)
            av_free(frame->extended_data);
        return ret;
    }
    if (frame->extended_data != frame->data) {
        for (ch = 0; ch < AV_NUM_DATA_POINTERS; ch++)
            frame->data[ch] = frame->extended_data[ch];
    }

    return ret;
}

static int audio_get_buffer(AVCodecContext *avctx, AVFrame *frame)
{
    AVCodecInternal *avci = avctx->internal;
    InternalBuffer *buf;
    int buf_size, ret;

    buf_size = av_samples_get_buffer_size(NULL, avctx->channels,
                                          frame->nb_samples, avctx->sample_fmt,
                                          0);
    if (buf_size < 0)
        return AVERROR(EINVAL);

    /* allocate InternalBuffer if needed */
    if (!avci->buffer) {
        avci->buffer = av_mallocz(sizeof(InternalBuffer));
        if (!avci->buffer)
            return AVERROR(ENOMEM);
    }
    buf = avci->buffer;

    /* if there is a previously-used internal buffer, check its size and
       channel count to see if we can reuse it */
    if (buf->extended_data) {
        /* if current buffer is too small, free it */
        if (buf->extended_data[0] && buf_size > buf->audio_data_size) {
            av_free(buf->extended_data[0]);
            if (buf->extended_data != buf->data)
                av_free(&buf->extended_data);
            buf->extended_data = NULL;
            buf->data[0] = NULL;
        }
        /* if number of channels has changed, reset and/or free extended data
           pointers but leave data buffer in buf->data[0] for reuse */
        if (buf->nb_channels != avctx->channels) {
            if (buf->extended_data != buf->data)
                av_free(buf->extended_data);
            buf->extended_data = NULL;
        }
    }

    /* if there is no previous buffer or the previous buffer cannot be used
       as-is, allocate a new buffer and/or rearrange the channel pointers */
    if (!buf->extended_data) {
        if (!buf->data[0]) {
            if (!(buf->data[0] = av_mallocz(buf_size)))
                return AVERROR(ENOMEM);
            buf->audio_data_size = buf_size;
        }
        if ((ret = avcodec_fill_audio_frame(frame, avctx->channels,
                                            avctx->sample_fmt, buf->data[0],
                                            buf->audio_data_size, 0)))
            return ret;

        if (frame->extended_data == frame->data)
            buf->extended_data = buf->data;
        else
            buf->extended_data = frame->extended_data;
        memcpy(buf->data, frame->data, sizeof(frame->data));
        buf->linesize[0] = frame->linesize[0];
        buf->nb_channels = avctx->channels;
    } else {
        /* copy InternalBuffer info to the AVFrame */
        frame->extended_data = buf->extended_data;
        frame->linesize[0]   = buf->linesize[0];
        memcpy(frame->data, buf->data, sizeof(frame->data));
    }

    frame->type          = FF_BUFFER_TYPE_INTERNAL;

    if (avctx->pkt) frame->pkt_pts = avctx->pkt->pts;
    else            frame->pkt_pts = AV_NOPTS_VALUE;
    frame->reordered_opaque = avctx->reordered_opaque;

    frame->sample_rate    = avctx->sample_rate;
    frame->format         = avctx->sample_fmt;
    frame->channel_layout = avctx->channel_layout;

    if (avctx->debug & FF_DEBUG_BUFFERS)
        av_log(avctx, AV_LOG_DEBUG, "default_get_buffer called on frame %p, "
               "internal audio buffer used\n", frame);

    return 0;
}

static int video_get_buffer(AVCodecContext *s, AVFrame *pic)
{
    int i;
    int w= s->width;
    int h= s->height;
    InternalBuffer *buf;
    AVCodecInternal *avci = s->internal;

    if(pic->data[0]!=NULL) {
        av_log(s, AV_LOG_ERROR, "pic->data[0]!=NULL in avcodec_default_get_buffer\n");
        return -1;
    }
    if(avci->buffer_count >= INTERNAL_BUFFER_SIZE) {
        av_log(s, AV_LOG_ERROR, "buffer_count overflow (missing release_buffer?)\n");
        return -1;
    }

    if(av_image_check_size(w, h, 0, s))
        return -1;

    if (!avci->buffer) {
        avci->buffer = av_mallocz((INTERNAL_BUFFER_SIZE+1) *
                                  sizeof(InternalBuffer));
    }

    buf = &avci->buffer[avci->buffer_count];

    if(buf->base[0] && (buf->width != w || buf->height != h || buf->pix_fmt != s->pix_fmt)){
        for (i = 0; i < AV_NUM_DATA_POINTERS; i++) {
            av_freep(&buf->base[i]);
            buf->data[i]= NULL;
        }
    }

    if (!buf->base[0]) {
        int h_chroma_shift, v_chroma_shift;
        int size[4] = {0};
        int tmpsize;
        int unaligned;
        AVPicture picture;
        int stride_align[AV_NUM_DATA_POINTERS];
        const int pixel_size = av_pix_fmt_descriptors[s->pix_fmt].comp[0].step_minus1+1;

        avcodec_get_chroma_sub_sample(s->pix_fmt, &h_chroma_shift, &v_chroma_shift);

        avcodec_align_dimensions2(s, &w, &h, stride_align);

        if(!(s->flags&CODEC_FLAG_EMU_EDGE)){
            w+= EDGE_WIDTH*2;
            h+= EDGE_WIDTH*2;
        }

        do {
            // NOTE: do not align linesizes individually, this breaks e.g. assumptions
            // that linesize[0] == 2*linesize[1] in the MPEG-encoder for 4:2:2
            av_image_fill_linesizes(picture.linesize, s->pix_fmt, w);
            // increase alignment of w for next try (rhs gives the lowest bit set in w)
            w += w & ~(w-1);

            unaligned = 0;
            for (i=0; i<4; i++){
                unaligned |= picture.linesize[i] % stride_align[i];
            }
        } while (unaligned);

        tmpsize = av_image_fill_pointers(picture.data, s->pix_fmt, h, NULL, picture.linesize);
        if (tmpsize < 0)
            return -1;

        for (i=0; i<3 && picture.data[i+1]; i++)
            size[i] = picture.data[i+1] - picture.data[i];
        size[i] = tmpsize - (picture.data[i] - picture.data[0]);

        memset(buf->base, 0, sizeof(buf->base));
        memset(buf->data, 0, sizeof(buf->data));

        for(i=0; i<4 && size[i]; i++){
            const int h_shift= i==0 ? 0 : h_chroma_shift;
            const int v_shift= i==0 ? 0 : v_chroma_shift;

            buf->linesize[i]= picture.linesize[i];

            buf->base[i]= av_malloc(size[i]+16); //FIXME 16
            if(buf->base[i]==NULL) return -1;
            memset(buf->base[i], 128, size[i]);

            // no edge if EDGE EMU or not planar YUV
            if((s->flags&CODEC_FLAG_EMU_EDGE) || !size[2])
                buf->data[i] = buf->base[i];
            else
                buf->data[i] = buf->base[i] + FFALIGN((buf->linesize[i]*EDGE_WIDTH>>v_shift) + (pixel_size*EDGE_WIDTH>>h_shift), stride_align[i]);
        }
        for (; i < AV_NUM_DATA_POINTERS; i++) {
            buf->base[i] = buf->data[i] = NULL;
            buf->linesize[i] = 0;
        }
        if(size[1] && !size[2])
            ff_set_systematic_pal2((uint32_t*)buf->data[1], s->pix_fmt);
        buf->width  = s->width;
        buf->height = s->height;
        buf->pix_fmt= s->pix_fmt;
    }
    pic->type= FF_BUFFER_TYPE_INTERNAL;

    for (i = 0; i < AV_NUM_DATA_POINTERS; i++) {
        pic->base[i]= buf->base[i];
        pic->data[i]= buf->data[i];
        pic->linesize[i]= buf->linesize[i];
    }
    pic->extended_data = pic->data;
    avci->buffer_count++;
    pic->width  = buf->width;
    pic->height = buf->height;
    pic->format = buf->pix_fmt;
    pic->sample_aspect_ratio = s->sample_aspect_ratio;

    if(s->pkt) pic->pkt_pts= s->pkt->pts;
    else       pic->pkt_pts= AV_NOPTS_VALUE;
    pic->reordered_opaque= s->reordered_opaque;

    if(s->debug&FF_DEBUG_BUFFERS)
        av_log(s, AV_LOG_DEBUG, "default_get_buffer called on pic %p, %d "
               "buffers used\n", pic, avci->buffer_count);

    return 0;
}

int avcodec_default_get_buffer(AVCodecContext *avctx, AVFrame *frame)
{
    switch (avctx->codec_type) {
    case AVMEDIA_TYPE_VIDEO:
        return video_get_buffer(avctx, frame);
    case AVMEDIA_TYPE_AUDIO:
        return audio_get_buffer(avctx, frame);
    default:
        return -1;
    }
}

void avcodec_default_release_buffer(AVCodecContext *s, AVFrame *pic){
    int i;
    InternalBuffer *buf, *last;
    AVCodecInternal *avci = s->internal;

    assert(s->codec_type == AVMEDIA_TYPE_VIDEO);

    assert(pic->type==FF_BUFFER_TYPE_INTERNAL);
    assert(avci->buffer_count);

    if (avci->buffer) {
        buf = NULL; /* avoids warning */
        for (i = 0; i < avci->buffer_count; i++) { //just 3-5 checks so is not worth to optimize
            buf = &avci->buffer[i];
            if (buf->data[0] == pic->data[0])
                break;
        }
        assert(i < avci->buffer_count);
        avci->buffer_count--;
        last = &avci->buffer[avci->buffer_count];

        if (buf != last)
            FFSWAP(InternalBuffer, *buf, *last);
    }

    for (i = 0; i < AV_NUM_DATA_POINTERS; i++) {
        pic->data[i]=NULL;
//        pic->base[i]=NULL;
    }
//printf("R%X\n", pic->opaque);

    if(s->debug&FF_DEBUG_BUFFERS)
        av_log(s, AV_LOG_DEBUG, "default_release_buffer called on pic %p, %d "
               "buffers used\n", pic, avci->buffer_count);
}

int avcodec_default_reget_buffer(AVCodecContext *s, AVFrame *pic){
    AVFrame temp_pic;
    int i;

    assert(s->codec_type == AVMEDIA_TYPE_VIDEO);

    /* If no picture return a new buffer */
    if(pic->data[0] == NULL) {
        /* We will copy from buffer, so must be readable */
        pic->buffer_hints |= FF_BUFFER_HINTS_READABLE;
        return s->get_buffer(s, pic);
    }

    assert(s->pix_fmt == pic->format);

    /* If internal buffer type return the same buffer */
    if(pic->type == FF_BUFFER_TYPE_INTERNAL) {
        if(s->pkt) pic->pkt_pts= s->pkt->pts;
        else       pic->pkt_pts= AV_NOPTS_VALUE;
        pic->reordered_opaque= s->reordered_opaque;
        return 0;
    }

    /*
     * Not internal type and reget_buffer not overridden, emulate cr buffer
     */
    temp_pic = *pic;
    for(i = 0; i < AV_NUM_DATA_POINTERS; i++)
        pic->data[i] = pic->base[i] = NULL;
    pic->opaque = NULL;
    /* Allocate new frame */
    if (s->get_buffer(s, pic))
        return -1;
    /* Copy image data from old buffer to new buffer */
    av_picture_copy((AVPicture*)pic, (AVPicture*)&temp_pic, s->pix_fmt, s->width,
             s->height);
    s->release_buffer(s, &temp_pic); // Release old frame
    return 0;
}

int avcodec_default_execute(AVCodecContext *c, int (*func)(AVCodecContext *c2, void *arg2),void *arg, int *ret, int count, int size){
    int i;

    for(i=0; i<count; i++){
        int r= func(c, (char*)arg + i*size);
        if(ret) ret[i]= r;
    }
    return 0;
}

int avcodec_default_execute2(AVCodecContext *c, int (*func)(AVCodecContext *c2, void *arg2, int jobnr, int threadnr),void *arg, int *ret, int count){
    int i;

    for(i=0; i<count; i++){
        int r= func(c, arg, i, 0);
        if(ret) ret[i]= r;
    }
    return 0;
}

enum PixelFormat avcodec_default_get_format(struct AVCodecContext *s, const enum PixelFormat *fmt){
    while (*fmt != PIX_FMT_NONE && ff_is_hwaccel_pix_fmt(*fmt))
        ++fmt;
    return fmt[0];
}

void avcodec_get_frame_defaults(AVFrame *pic){
    memset(pic, 0, sizeof(AVFrame));

    pic->pts= AV_NOPTS_VALUE;
    pic->key_frame= 1;
    pic->sample_aspect_ratio = (AVRational){0, 1};
    pic->format = -1;           /* unknown */
}

AVFrame *avcodec_alloc_frame(void){
    AVFrame *pic= av_malloc(sizeof(AVFrame));

    if(pic==NULL) return NULL;

    avcodec_get_frame_defaults(pic);

    return pic;
}

int attribute_align_arg avcodec_open2(AVCodecContext *avctx, AVCodec *codec, AVDictionary **options)
{
    int ret = 0;
    AVDictionary *tmp = NULL;

    if (avcodec_is_open(avctx))
        return 0;

    if ((!codec && !avctx->codec)) {
        av_log(avctx, AV_LOG_ERROR, "No codec provided to avcodec_open2().\n");
        return AVERROR(EINVAL);
    }
    if ((codec && avctx->codec && codec != avctx->codec)) {
        av_log(avctx, AV_LOG_ERROR, "This AVCodecContext was allocated for %s, "
               "but %s passed to avcodec_open2().\n", avctx->codec->name, codec->name);
        return AVERROR(EINVAL);
    }
    if (!codec)
        codec = avctx->codec;

    if (avctx->extradata_size < 0 || avctx->extradata_size >= FF_MAX_EXTRADATA_SIZE)
        return AVERROR(EINVAL);

    if (options)
        av_dict_copy(&tmp, *options, 0);

    /* If there is a user-supplied mutex locking routine, call it. */
    if (ff_lockmgr_cb) {
        if ((*ff_lockmgr_cb)(&codec_mutex, AV_LOCK_OBTAIN))
            return -1;
    }

    entangled_thread_counter++;
    if(entangled_thread_counter != 1){
        av_log(avctx, AV_LOG_ERROR, "insufficient thread locking around avcodec_open/close()\n");
        ret = -1;
        goto end;
    }

    avctx->internal = av_mallocz(sizeof(AVCodecInternal));
    if (!avctx->internal) {
        ret = AVERROR(ENOMEM);
        goto end;
    }

    if (codec->priv_data_size > 0) {
      if(!avctx->priv_data){
        avctx->priv_data = av_mallocz(codec->priv_data_size);
        if (!avctx->priv_data) {
            ret = AVERROR(ENOMEM);
            goto end;
        }
        if (codec->priv_class) {
            *(AVClass**)avctx->priv_data= codec->priv_class;
            av_opt_set_defaults(avctx->priv_data);
        }
      }
      if (codec->priv_class && (ret = av_opt_set_dict(avctx->priv_data, &tmp)) < 0)
          goto free_and_end;
    } else {
        avctx->priv_data = NULL;
    }
    if ((ret = av_opt_set_dict(avctx, &tmp)) < 0)
        goto free_and_end;

    if(avctx->coded_width && avctx->coded_height)
        avcodec_set_dimensions(avctx, avctx->coded_width, avctx->coded_height);
    else if(avctx->width && avctx->height)
        avcodec_set_dimensions(avctx, avctx->width, avctx->height);

    if ((avctx->coded_width || avctx->coded_height || avctx->width || avctx->height)
        && (  av_image_check_size(avctx->coded_width, avctx->coded_height, 0, avctx) < 0
           || av_image_check_size(avctx->width,       avctx->height,       0, avctx) < 0)) {
        av_log(avctx, AV_LOG_WARNING, "ignoring invalid width/height values\n");
        avcodec_set_dimensions(avctx, 0, 0);
    }

    /* if the decoder init function was already called previously,
       free the already allocated subtitle_header before overwriting it */
    if (av_codec_is_decoder(codec))
        av_freep(&avctx->subtitle_header);

#define SANE_NB_CHANNELS 128U
    if (avctx->channels > SANE_NB_CHANNELS) {
        ret = AVERROR(EINVAL);
        goto free_and_end;
    }

    avctx->codec = codec;
    if ((avctx->codec_type == AVMEDIA_TYPE_UNKNOWN || avctx->codec_type == codec->type) &&
        avctx->codec_id == AV_CODEC_ID_NONE) {
        avctx->codec_type = codec->type;
        avctx->codec_id   = codec->id;
    }
    if (avctx->codec_id != codec->id || (avctx->codec_type != codec->type
                           && avctx->codec_type != AVMEDIA_TYPE_ATTACHMENT)) {
        av_log(avctx, AV_LOG_ERROR, "codec type or id mismatches\n");
        ret = AVERROR(EINVAL);
        goto free_and_end;
    }
    avctx->frame_number = 0;

    if (avctx->codec_type == AVMEDIA_TYPE_AUDIO &&
        (!avctx->time_base.num || !avctx->time_base.den)) {
        avctx->time_base.num = 1;
        avctx->time_base.den = avctx->sample_rate;
    }

    if (HAVE_THREADS && !avctx->thread_opaque) {
        ret = ff_thread_init(avctx);
        if (ret < 0) {
            goto free_and_end;
        }
    }
    if (!HAVE_THREADS && !(codec->capabilities & CODEC_CAP_AUTO_THREADS))
        avctx->thread_count = 1;

    if (av_codec_is_encoder(avctx->codec)) {
        int i;
        if (avctx->codec->sample_fmts) {
            for (i = 0; avctx->codec->sample_fmts[i] != AV_SAMPLE_FMT_NONE; i++)
                if (avctx->sample_fmt == avctx->codec->sample_fmts[i])
                    break;
            if (avctx->codec->sample_fmts[i] == AV_SAMPLE_FMT_NONE) {
                av_log(avctx, AV_LOG_ERROR, "Specified sample_fmt is not supported.\n");
                ret = AVERROR(EINVAL);
                goto free_and_end;
            }
        }
        if (avctx->codec->pix_fmts) {
            for (i = 0; avctx->codec->pix_fmts[i] != PIX_FMT_NONE; i++)
                if (avctx->pix_fmt == avctx->codec->pix_fmts[i])
                    break;
            if (avctx->codec->pix_fmts[i] == PIX_FMT_NONE) {
                av_log(avctx, AV_LOG_ERROR, "Specified pix_fmt is not supported\n");
                ret = AVERROR(EINVAL);
                goto free_and_end;
            }
        }
        if (avctx->codec->supported_samplerates) {
            for (i = 0; avctx->codec->supported_samplerates[i] != 0; i++)
                if (avctx->sample_rate == avctx->codec->supported_samplerates[i])
                    break;
            if (avctx->codec->supported_samplerates[i] == 0) {
                av_log(avctx, AV_LOG_ERROR, "Specified sample_rate is not supported\n");
                ret = AVERROR(EINVAL);
                goto free_and_end;
            }
        }
        if (avctx->codec->channel_layouts) {
            if (!avctx->channel_layout) {
                av_log(avctx, AV_LOG_WARNING, "channel_layout not specified\n");
            } else {
                for (i = 0; avctx->codec->channel_layouts[i] != 0; i++)
                    if (avctx->channel_layout == avctx->codec->channel_layouts[i])
                        break;
                if (avctx->codec->channel_layouts[i] == 0) {
                    av_log(avctx, AV_LOG_ERROR, "Specified channel_layout is not supported\n");
                    ret = AVERROR(EINVAL);
                    goto free_and_end;
                }
            }
        }
        if (avctx->channel_layout && avctx->channels) {
            if (av_get_channel_layout_nb_channels(avctx->channel_layout) != avctx->channels) {
                av_log(avctx, AV_LOG_ERROR, "channel layout does not match number of channels\n");
                ret = AVERROR(EINVAL);
                goto free_and_end;
            }
        } else if (avctx->channel_layout) {
            avctx->channels = av_get_channel_layout_nb_channels(avctx->channel_layout);
        }
    }

    if(avctx->codec->init && !(avctx->active_thread_type&FF_THREAD_FRAME)){
        ret = avctx->codec->init(avctx);
        if (ret < 0) {
            goto free_and_end;
        }
    }

    if (av_codec_is_decoder(avctx->codec)) {
        /* validate channel layout from the decoder */
        if (avctx->channel_layout &&
            av_get_channel_layout_nb_channels(avctx->channel_layout) != avctx->channels) {
            av_log(avctx, AV_LOG_WARNING, "channel layout does not match number of channels\n");
            avctx->channel_layout = 0;
        }
    }
end:
    entangled_thread_counter--;

    /* Release any user-supplied mutex. */
    if (ff_lockmgr_cb) {
        (*ff_lockmgr_cb)(&codec_mutex, AV_LOCK_RELEASE);
    }
    if (options) {
        av_dict_free(options);
        *options = tmp;
    }

    return ret;
free_and_end:
    av_dict_free(&tmp);
    av_freep(&avctx->priv_data);
    av_freep(&avctx->internal);
    avctx->codec= NULL;
    goto end;
}

int ff_alloc_packet(AVPacket *avpkt, int size)
{
    if (size > INT_MAX - FF_INPUT_BUFFER_PADDING_SIZE)
        return AVERROR(EINVAL);

    if (avpkt->data) {
        void *destruct = avpkt->destruct;

        if (avpkt->size < size)
            return AVERROR(EINVAL);

        av_init_packet(avpkt);
        avpkt->destruct = destruct;
        avpkt->size = size;
        return 0;
    } else {
        return av_new_packet(avpkt, size);
    }
}

/**
 * Pad last frame with silence.
 */
static int pad_last_frame(AVCodecContext *s, AVFrame **dst, const AVFrame *src)
{
    AVFrame *frame = NULL;
    uint8_t *buf   = NULL;
    int ret;

    if (!(frame = avcodec_alloc_frame()))
        return AVERROR(ENOMEM);
    *frame = *src;

    if ((ret = av_samples_get_buffer_size(&frame->linesize[0], s->channels,
                                          s->frame_size, s->sample_fmt, 0)) < 0)
        goto fail;

    if (!(buf = av_malloc(ret))) {
        ret = AVERROR(ENOMEM);
        goto fail;
    }

    frame->nb_samples = s->frame_size;
    if ((ret = avcodec_fill_audio_frame(frame, s->channels, s->sample_fmt,
                                        buf, ret, 0)) < 0)
        goto fail;
    if ((ret = av_samples_copy(frame->extended_data, src->extended_data, 0, 0,
                               src->nb_samples, s->channels, s->sample_fmt)) < 0)
        goto fail;
    if ((ret = av_samples_set_silence(frame->extended_data, src->nb_samples,
                                      frame->nb_samples - src->nb_samples,
                                      s->channels, s->sample_fmt)) < 0)
        goto fail;

    *dst = frame;

    return 0;

fail:
    if (frame->extended_data != frame->data)
        av_freep(&frame->extended_data);
    av_freep(&buf);
    av_freep(&frame);
    return ret;
}

int attribute_align_arg avcodec_encode_audio2(AVCodecContext *avctx,
                                              AVPacket *avpkt,
                                              const AVFrame *frame,
                                              int *got_packet_ptr)
{
    AVFrame tmp;
    AVFrame *padded_frame = NULL;
    int ret;
    int user_packet = !!avpkt->data;

    *got_packet_ptr = 0;

    if (!(avctx->codec->capabilities & CODEC_CAP_DELAY) && !frame) {
        av_free_packet(avpkt);
        av_init_packet(avpkt);
        return 0;
    }

    /* ensure that extended_data is properly set */
    if (frame && !frame->extended_data) {
        if (av_sample_fmt_is_planar(avctx->sample_fmt) &&
            avctx->channels > AV_NUM_DATA_POINTERS) {
            av_log(avctx, AV_LOG_ERROR, "Encoding to a planar sample format, "
                   "with more than %d channels, but extended_data is not set.\n",
                   AV_NUM_DATA_POINTERS);
            return AVERROR(EINVAL);
        }
        av_log(avctx, AV_LOG_WARNING, "extended_data is not set.\n");

        tmp = *frame;
        tmp.extended_data = tmp.data;
        frame = &tmp;
    }

    /* check for valid frame size */
    if (frame) {
        if (avctx->codec->capabilities & CODEC_CAP_SMALL_LAST_FRAME) {
            if (frame->nb_samples > avctx->frame_size)
                return AVERROR(EINVAL);
        } else if (!(avctx->codec->capabilities & CODEC_CAP_VARIABLE_FRAME_SIZE)) {
            if (frame->nb_samples < avctx->frame_size &&
                !avctx->internal->last_audio_frame) {
                ret = pad_last_frame(avctx, &padded_frame, frame);
                if (ret < 0)
                    return ret;

                frame = padded_frame;
                avctx->internal->last_audio_frame = 1;
            }

            if (frame->nb_samples != avctx->frame_size)
                return AVERROR(EINVAL);
        }
    }

    ret = avctx->codec->encode2(avctx, avpkt, frame, got_packet_ptr);
    if (!ret) {
        if (*got_packet_ptr) {
            if (!(avctx->codec->capabilities & CODEC_CAP_DELAY)) {
                if (avpkt->pts == AV_NOPTS_VALUE)
                    avpkt->pts = frame->pts;
                if (!avpkt->duration)
                    avpkt->duration = ff_samples_to_time_base(avctx,
                                                              frame->nb_samples);
            }
            avpkt->dts = avpkt->pts;
        } else {
            avpkt->size = 0;
        }

        if (!user_packet && avpkt->size) {
            uint8_t *new_data = av_realloc(avpkt->data, avpkt->size);
            if (new_data)
                avpkt->data = new_data;
        }

        avctx->frame_number++;
    }

    if (ret < 0 || !*got_packet_ptr) {
        av_free_packet(avpkt);
        av_init_packet(avpkt);
        return ret;
    }

    /* NOTE: if we add any audio encoders which output non-keyframe packets,
             this needs to be moved to the encoders, but for now we can do it
             here to simplify things */
    avpkt->flags |= AV_PKT_FLAG_KEY;

    if (padded_frame) {
        av_freep(&padded_frame->data[0]);
        if (padded_frame->extended_data != padded_frame->data)
            av_freep(&padded_frame->extended_data);
        av_freep(&padded_frame);