int av_samples_get_buffer_size(int *linesize, int nb_channels, int nb_samples,
                               enum AVSampleFormat sample_fmt, int align)
{
    int line_size;
    int sample_size = av_get_bytes_per_sample(sample_fmt);
    int planar      = av_sample_fmt_is_planar(sample_fmt);

    /* validate parameter ranges */
    if (!sample_size || nb_samples <= 0 || nb_channels <= 0)
        return AVERROR(EINVAL);

    /* check for integer overflow */
    if (nb_channels > INT_MAX / align ||
        (int64_t)nb_channels * nb_samples > (INT_MAX - (align * nb_channels)) / sample_size)
        return AVERROR(EINVAL);

    line_size = planar ? FFALIGN(nb_samples * sample_size,               align) :
                         FFALIGN(nb_samples * sample_size * nb_channels, align);
    if (linesize)
        *linesize = line_size;

    return planar ? line_size * nb_channels : line_size;
}

static int dxv_init(AVCodecContext *avctx)
{
    DXVContext *ctx = avctx->priv_data;
    int ret = av_image_check_size(avctx->width, avctx->height, 0, avctx);

    if (ret < 0) {
        av_log(avctx, AV_LOG_ERROR, "Invalid image size %dx%d.\n",
               avctx->width, avctx->height);
        return ret;
    }

    /* Codec requires 16x16 alignment. */
    avctx->coded_width  = FFALIGN(avctx->width,  16);
    avctx->coded_height = FFALIGN(avctx->height, 16);

    ff_texturedsp_init(&ctx->texdsp);
    avctx->pix_fmt = AV_PIX_FMT_RGBA;

    ctx->slice_count = av_clip(avctx->thread_count, 1,
                               avctx->coded_height / TEXTURE_BLOCK_H);

    return 0;
}

/**
 * Backward synthesis filter, find the LPC coefficients from past speech data.
 */
static void backward_filter(RA288Context *ractx,
                            float *hist, float *rec, const float *window,
                            float *lpc, const float *tab,
                            int order, int n, int non_rec, int move_size)
{
    float temp[MAX_BACKWARD_FILTER_ORDER+1];

    do_hybrid_window(ractx, order, n, non_rec, temp, hist, rec, window);

    if (!compute_lpc_coefs(temp, order, lpc, 0, 1, 1))
        ractx->fdsp->vector_fmul(lpc, lpc, tab, FFALIGN(order, 16));

    memmove(hist, hist + n, move_size*sizeof(*hist));
}

static int ffmal_copy_frame(AVCodecContext *avctx,  AVFrame *frame,
                            MMAL_BUFFER_HEADER_T *buffer)
{
    MMALDecodeContext *ctx = avctx->priv_data;
    int ret = 0;

    if (avctx->pix_fmt == AV_PIX_FMT_MMAL) {
        if (!ctx->pool_out)
            return AVERROR_UNKNOWN; // format change code failed with OOM previously

        if ((ret = ff_decode_frame_props(avctx, frame)) < 0)
            goto done;

        if ((ret = ffmmal_set_ref(frame, ctx->pool_out, buffer)) < 0)
            goto done;
    } else {
        int w = FFALIGN(avctx->width, 32);
        int h = FFALIGN(avctx->height, 16);
        uint8_t *src[4];
        int linesize[4];

        if ((ret = ff_get_buffer(avctx, frame, 0)) < 0)
            goto done;

        av_image_fill_arrays(src, linesize,
                             buffer->data + buffer->type->video.offset[0],
                             avctx->pix_fmt, w, h, 1);
        av_image_copy(frame->data, frame->linesize, src, linesize,
                      avctx->pix_fmt, avctx->width, avctx->height);
    }

    frame->pkt_pts = buffer->pts == MMAL_TIME_UNKNOWN ? AV_NOPTS_VALUE : buffer->pts;
    frame->pkt_dts = AV_NOPTS_VALUE;

done:
    return ret;
}

static int msrle_decode_frame(AVCodecContext *avctx,
                              void *data, int *data_size,
                              AVPacket *avpkt)
{
    const uint8_t *buf = avpkt->data;
    int buf_size = avpkt->size;
    MsrleContext *s = avctx->priv_data;
    int istride = FFALIGN(avctx->width*avctx->bits_per_coded_sample, 32) / 8;

    s->buf = buf;
    s->size = buf_size;

    s->frame.reference = 1;
    s->frame.buffer_hints = FF_BUFFER_HINTS_VALID | FF_BUFFER_HINTS_PRESERVE | FF_BUFFER_HINTS_REUSABLE;
    if (avctx->reget_buffer(avctx, &s->frame)) {
        av_log(avctx, AV_LOG_ERROR, "reget_buffer() failed\n");
        return -1;
    }

    if (s->avctx->palctrl) {
        /* make the palette available */
        memcpy(s->frame.data[1], s->avctx->palctrl->palette, AVPALETTE_SIZE);
        if (s->avctx->palctrl->palette_changed) {
            s->frame.palette_has_changed = 1;
            s->avctx->palctrl->palette_changed = 0;
        }
    }

    /* FIXME how to correctly detect RLE ??? */
    if (avctx->height * istride == avpkt->size) { /* assume uncompressed */
        int linesize = avctx->width * avctx->bits_per_coded_sample / 8;
        uint8_t *ptr = s->frame.data[0];
        uint8_t *buf = avpkt->data + (avctx->height-1)*istride;
        int i, j;

        for (i = 0; i < avctx->height; i++) {
            if (avctx->bits_per_coded_sample == 4) {
                for (j = 0; j < avctx->width - 1; j += 2) {
                    ptr[j+0] = buf[j>>1] >> 4;
                    ptr[j+1] = buf[j>>1] & 0xF;
                }
                if (avctx->width & 1)
                    ptr[j+0] = buf[j>>1] >> 4;
            } else {
                memcpy(ptr, buf, linesize);
            }
            buf -= istride;
            ptr += s->frame.linesize[0];
        }

static int w64_read_header(AVFormatContext *s, AVFormatParameters *ap)
{
    int64_t size;
    ByteIOContext *pb  = s->pb;
    WAVContext    *wav = s->priv_data;
    AVStream *st;
    uint8_t guid[16];

    get_buffer(pb, guid, 16);
    if (memcmp(guid, guid_riff, 16))
        return -1;

    if (get_le64(pb) < 16 + 8 + 16 + 8 + 16 + 8) /* riff + wave + fmt + sizes */
        return -1;

    get_buffer(pb, guid, 16);
    if (memcmp(guid, guid_wave, 16)) {
        av_log(s, AV_LOG_ERROR, "could not find wave guid\n");
        return -1;
    }

    size = find_guid(pb, guid_fmt);
    if (size < 0) {
        av_log(s, AV_LOG_ERROR, "could not find fmt guid\n");
        return -1;
    }

    st = av_new_stream(s, 0);
    if (!st)
        return AVERROR(ENOMEM);

    /* subtract chunk header size - normal wav file doesn't count it */
    ff_get_wav_header(pb, st->codec, size - 24);
    url_fskip(pb, FFALIGN(size, INT64_C(8)) - size);

    st->need_parsing = AVSTREAM_PARSE_FULL;

    av_set_pts_info(st, 64, 1, st->codec->sample_rate);

    size = find_guid(pb, guid_data);
    if (size < 0) {
        av_log(s, AV_LOG_ERROR, "could not find data guid\n");
        return -1;
    }
    wav->data_end = url_ftell(pb) + size - 24;
    wav->w64      = 1;

    return 0;
}

// Filter data through filter
static af_data_t* play(struct af_instance_s* af, af_data_t* data)
{
  af_resample_t *s = af->setup;
  int ret;
  int8_t *in = (int8_t*)data->audio;
  int8_t *out;
  int chans   = data->nch;
  int in_len  = data->len;
  int out_len = in_len * af->mul + 10;

  if(AF_OK != RESIZE_LOCAL_BUFFER(af,data))
      return NULL;

  av_fast_malloc(&s->tmp[0], &s->tmp_alloc, FFALIGN(out_len,32));
  if(s->tmp[0] == NULL) return NULL;

  out= (int8_t*)af->data->audio;

  out_len= FFMIN(out_len, af->data->len);

  av_fast_malloc(&s->in[0], &s->in_alloc, FFALIGN(in_len,32));
  if(s->in[0] == NULL) return NULL;

  memcpy(s->in[0], in, in_len);

  ret = swr_convert(s->swrctx, &s->tmp[0], out_len/chans/2, &s->in[0], in_len/chans/2);
  if (ret < 0) return NULL;
  out_len= ret*chans*2;

  memcpy(out, s->tmp[0], out_len);

  data->audio = af->data->audio;
  data->len   = out_len;
  data->rate  = af->data->rate;
  return data;
}

static int codec_reinit(AVCodecContext *avctx, int width, int height,
                        int quality)
{
    NuvContext *c = avctx->priv_data;
    int ret;

    width  = FFALIGN(width,  2);
    height = FFALIGN(height, 2);
    if (quality >= 0)
        get_quant_quality(c, quality);
    if (width != c->width || height != c->height) {
        // also reserve space for a possible additional header
        int buf_size = height * width * 3 / 2
                     + FFMAX(AV_LZO_OUTPUT_PADDING, FF_INPUT_BUFFER_PADDING_SIZE)
                     + RTJPEG_HEADER_SIZE;
        if (buf_size > INT_MAX/8)
            return -1;
        if ((ret = av_image_check_size(height, width, 0, avctx)) < 0)
            return ret;
        avctx->width  = c->width  = width;
        avctx->height = c->height = height;
        av_fast_malloc(&c->decomp_buf, &c->decomp_size,
                       buf_size);
        if (!c->decomp_buf) {
            av_log(avctx, AV_LOG_ERROR,
                   "Can't allocate decompression buffer.\n");
            return AVERROR(ENOMEM);
        }
        ff_rtjpeg_decode_init(&c->rtj, c->width, c->height, c->lq, c->cq);
        av_frame_unref(c->pic);
        return 1;
    } else if (quality != c->quality)
        ff_rtjpeg_decode_init(&c->rtj, c->width, c->height, c->lq, c->cq);

    return 0;
}

static void put_meta(AVFormatContext *s, const char *key, uint32_t id)
{
    AVDictionaryEntry *tag;
    AVIOContext *pb = s->pb;

    if (tag = av_dict_get(s->metadata, key, NULL, 0)) {
        int size = strlen(tag->value);

        avio_wl32(pb, id);
        avio_wb32(pb, FFALIGN(size, 2));
        avio_write(pb, tag->value, size);
        if (size & 1)
            avio_w8(pb, 0);
    }
}

static void allocate_xvimage(struct vo *vo, int foo)
{
    struct xvctx *ctx = vo->priv;
    struct vo_x11_state *x11 = vo->x11;
    // align it for faster OSD rendering (draw_bmp.c swscale usage)
    int aligned_w = FFALIGN(ctx->image_width, 32);
#ifdef HAVE_SHM
    if (x11->display_is_local && XShmQueryExtension(x11->display)) {
        ctx->Shmem_Flag = 1;
        x11->ShmCompletionEvent = XShmGetEventBase(x11->display)
                                + ShmCompletion;
    } else {
        ctx->Shmem_Flag = 0;
        MP_INFO(vo, "Shared memory not supported\nReverting to normal Xv.\n");
    }
    if (ctx->Shmem_Flag) {
        ctx->xvimage[foo] =
            (XvImage *) XvShmCreateImage(x11->display, ctx->xv_port,
                                         ctx->xv_format, NULL,
                                         aligned_w, ctx->image_height,
                                         &ctx->Shminfo[foo]);

        ctx->Shminfo[foo].shmid = shmget(IPC_PRIVATE,
                                         ctx->xvimage[foo]->data_size,
                                         IPC_CREAT | 0777);
        ctx->Shminfo[foo].shmaddr = (char *) shmat(ctx->Shminfo[foo].shmid, 0,
                                                   0);
        ctx->Shminfo[foo].readOnly = False;

        ctx->xvimage[foo]->data = ctx->Shminfo[foo].shmaddr;
        XShmAttach(x11->display, &ctx->Shminfo[foo]);
        XSync(x11->display, False);
        shmctl(ctx->Shminfo[foo].shmid, IPC_RMID, 0);
    } else
#endif
    {
        ctx->xvimage[foo] =
            (XvImage *) XvCreateImage(x11->display, ctx->xv_port,
                                      ctx->xv_format, NULL, aligned_w,
                                      ctx->image_height);
        ctx->xvimage[foo]->data = av_malloc(ctx->xvimage[foo]->data_size);
        XSync(x11->display, False);
    }
    struct mp_image img = get_xv_buffer(vo, foo);
    img.w = aligned_w;
    mp_image_clear(&img, 0, 0, img.w, img.h);
    return;
}

static int msrle_decode_frame(AVCodecContext *avctx,
                              void *data, int *got_frame,
                              AVPacket *avpkt)
{
    const uint8_t *buf = avpkt->data;
    int buf_size = avpkt->size;
    MsrleContext *s = avctx->priv_data;
    int istride = FFALIGN(avctx->width*avctx->bits_per_coded_sample, 32) / 8;
    int ret;

    s->buf = buf;
    s->size = buf_size;

    if ((ret = ff_reget_buffer(avctx, s->frame)) < 0)
        return ret;

    if (avctx->bits_per_coded_sample > 1 && avctx->bits_per_coded_sample <= 8) {
        const uint8_t *pal = av_packet_get_side_data(avpkt, AV_PKT_DATA_PALETTE, NULL);

        if (pal) {
            s->frame->palette_has_changed = 1;
            memcpy(s->pal, pal, AVPALETTE_SIZE);
        }
        /* make the palette available */
        memcpy(s->frame->data[1], s->pal, AVPALETTE_SIZE);
    }

    /* FIXME how to correctly detect RLE ??? */
    if (avctx->height * istride == avpkt->size) { /* assume uncompressed */
        int linesize = (avctx->width * avctx->bits_per_coded_sample + 7) / 8;
        uint8_t *ptr = s->frame->data[0];
        uint8_t *buf = avpkt->data + (avctx->height-1)*istride;
        int i, j;

        for (i = 0; i < avctx->height; i++) {
            if (avctx->bits_per_coded_sample == 4) {
                for (j = 0; j < avctx->width - 1; j += 2) {
                    ptr[j+0] = buf[j>>1] >> 4;
                    ptr[j+1] = buf[j>>1] & 0xF;
                }
                if (avctx->width & 1)
                    ptr[j+0] = buf[j>>1] >> 4;
            } else {
                memcpy(ptr, buf, linesize);
            }
            buf -= istride;
            ptr += s->frame->linesize[0];
        }

int ff_image_copy_plane_uc_from_x86(uint8_t       *dst, ptrdiff_t dst_linesize,
                                    const uint8_t *src, ptrdiff_t src_linesize,
                                    ptrdiff_t bytewidth, int height)
{
    int cpu_flags = av_get_cpu_flags();
    ptrdiff_t bw_aligned = FFALIGN(bytewidth, 64);

    if (EXTERNAL_SSE4(cpu_flags) &&
        bw_aligned <= dst_linesize && bw_aligned <= src_linesize)
        ff_image_copy_plane_uc_from_sse4(dst, dst_linesize, src, src_linesize,
                                         bw_aligned, height);
    else
        return AVERROR(ENOSYS);

    return 0;
}

/** Find chunk with w64 GUID by skipping over other chunks
 * @return the size of the found chunk
 */
static int64_t find_guid(ByteIOContext *pb, const uint8_t guid1[16])
{
    uint8_t guid[16];
    int64_t size;

    while (!url_feof(pb)) {
        get_buffer(pb, guid, 16);
        size = get_le64(pb);
        if (size <= 24)
            return -1;
        if (!memcmp(guid, guid1, 16))
            return size;
        url_fskip(pb, FFALIGN(size, INT64_C(8)) - 24);
    }
    return -1;
}

/** Find chunk with w64 GUID by skipping over other chunks
 * @return the size of the found chunk
 */
static int64_t find_guid(AVIOContext *pb, const uint8_t guid1[16])
{
    uint8_t guid[16];
    int64_t size;

    while (!url_feof(pb)) {
        avio_read(pb, guid, 16);
        size = avio_rl64(pb);
        if (size <= 24)
            return -1;
        if (!memcmp(guid, guid1, 16))
            return size;
        avio_skip(pb, FFALIGN(size, INT64_C(8)) - 24);
    }
    return -1;
}

static int get_video_buffer(AVFrame *frame, int align)
{
    const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(frame->format);
    int ret, i;

    if (!desc)
        return AVERROR(EINVAL);

    if ((ret = av_image_check_size(frame->width, frame->height, 0, NULL)) < 0)
        return ret;

    if (!frame->linesize[0]) {
        ret = av_image_fill_linesizes(frame->linesize, frame->format,
                                      frame->width);
        if (ret < 0)
            return ret;

        for (i = 0; i < 4 && frame->linesize[i]; i++)
            frame->linesize[i] = FFALIGN(frame->linesize[i], align);
    }

    for (i = 0; i < 4 && frame->linesize[i]; i++) {
        int h = frame->height;
        if (i == 1 || i == 2)
            h = -((-h) >> desc->log2_chroma_h);

        frame->buf[i] = av_buffer_alloc(frame->linesize[i] * h);
        if (!frame->buf[i])
            goto fail;

        frame->data[i] = frame->buf[i]->data;
    }
    if (desc->flags & AV_PIX_FMT_FLAG_PAL || desc->flags & AV_PIX_FMT_FLAG_PSEUDOPAL) {
        av_buffer_unref(&frame->buf[1]);
        frame->buf[1] = av_buffer_alloc(1024);
        if (!frame->buf[1])
            goto fail;
        frame->data[1] = frame->buf[1]->data;
    }

    frame->extended_data = frame->data;

    return 0;
fail:
    av_frame_unref(frame);
    return AVERROR(ENOMEM);
}

void ff_mediacodec_sw_buffer_copy_yuv420_semi_planar(AVCodecContext *avctx,
                                                     MediaCodecDecContext *s,
                                                     uint8_t *data,
                                                     size_t size,
                                                     FFAMediaCodecBufferInfo *info,
                                                     AVFrame *frame)
{
    int i;
    uint8_t *src = NULL;

    for (i = 0; i < 2; i++) {
        int height;

        src = data + info->offset;
        if (i == 0) {
            height = avctx->height;

            src += s->crop_top * s->stride;
            src += s->crop_left;
        } else if (i == 1) {
            height = avctx->height / 2;

            src += s->slice_height * s->stride;
            src += s->crop_top * s->stride;
            src += s->crop_left;
        }

        if (frame->linesize[i] == s->stride) {
            memcpy(frame->data[i], src, height * s->stride);
        } else {
            int j, width;
            uint8_t *dst = frame->data[i];

            if (i == 0) {
                width = avctx->width;
            } else if (i == 1) {
                width = FFMIN(frame->linesize[i], FFALIGN(avctx->width, 2));
            }

            for (j = 0; j < height; j++) {
                memcpy(dst, src, width);
                src += s->stride;
                dst += frame->linesize[i];
            }
        }
    }
}

static int cuda_frames_init(AVHWFramesContext *ctx)
{
    CUDAFramesContext *priv = ctx->internal->priv;
    int aligned_width = FFALIGN(ctx->width, CUDA_FRAME_ALIGNMENT);
    int i;

    for (i = 0; i < FF_ARRAY_ELEMS(supported_formats); i++) {
        if (ctx->sw_format == supported_formats[i])
            break;
    }
    if (i == FF_ARRAY_ELEMS(supported_formats)) {
        av_log(ctx, AV_LOG_ERROR, "Pixel format '%s' is not supported\n",
               av_get_pix_fmt_name(ctx->sw_format));
        return AVERROR(ENOSYS);
    }

    av_pix_fmt_get_chroma_sub_sample(ctx->sw_format, &priv->shift_width, &priv->shift_height);

    if (!ctx->pool) {
        int size;

        switch (ctx->sw_format) {
        case AV_PIX_FMT_NV12:
        case AV_PIX_FMT_YUV420P:
            size = aligned_width * ctx->height * 3 / 2;
            break;
        case AV_PIX_FMT_YUV444P:
        case AV_PIX_FMT_P010:
        case AV_PIX_FMT_P016:
            size = aligned_width * ctx->height * 3;
            break;
        case AV_PIX_FMT_YUV444P16:
            size = aligned_width * ctx->height * 6;
            break;
        default:
            av_log(ctx, AV_LOG_ERROR, "BUG: Pixel format missing from size calculation.");
            return AVERROR_BUG;
        }

        ctx->internal->pool_internal = av_buffer_pool_init2(size, ctx, cuda_pool_alloc, NULL);
        if (!ctx->internal->pool_internal)
            return AVERROR(ENOMEM);
    }

    return 0;
}

static void draw_alpha(int x0,int y0, int w,int h, unsigned char* src, unsigned char *srca, int stride){
    uint32_t bespitch = FFALIGN(mga_vid_config.src_width, 32);
    x0+=mga_vid_config.src_width*(vo_panscan_x>>1)/(vo_dwidth+vo_panscan_x);
    switch(mga_vid_config.format){
    case MGA_VID_FORMAT_YV12:
    case MGA_VID_FORMAT_IYUV:
    case MGA_VID_FORMAT_I420:
        vo_draw_alpha_yv12(w,h,src,srca,stride,vid_data+bespitch*y0+x0,bespitch);
        break;
    case MGA_VID_FORMAT_YUY2:
        vo_draw_alpha_yuy2(w,h,src,srca,stride,vid_data+2*(bespitch*y0+x0),2*bespitch);
        break;
    case MGA_VID_FORMAT_UYVY:
        vo_draw_alpha_yuy2(w,h,src,srca,stride,vid_data+2*(bespitch*y0+x0)+1,2*bespitch);
        break;
    }
}

static int y41p_decode_frame(AVCodecContext *avctx, void *data,
                             int *got_frame, AVPacket *avpkt)
{
    AVFrame *pic = data;
    uint8_t *src = avpkt->data;
    uint8_t *y, *u, *v;
    int i, j, ret;

    if (avpkt->size < 3LL * avctx->height * FFALIGN(avctx->width, 8) / 2) {
        av_log(avctx, AV_LOG_ERROR, "Insufficient input data.\n");
        return AVERROR(EINVAL);
    }

    if ((ret = ff_get_buffer(avctx, pic, 0)) < 0)
        return ret;

    pic->key_frame = 1;
    pic->pict_type = AV_PICTURE_TYPE_I;

    for (i = avctx->height - 1; i >= 0 ; i--) {
        y = &pic->data[0][i * pic->linesize[0]];
        u = &pic->data[1][i * pic->linesize[1]];
        v = &pic->data[2][i * pic->linesize[2]];
        for (j = 0; j < avctx->width; j += 8) {
            *(u++) = *src++;
            *(y++) = *src++;
            *(v++) = *src++;
            *(y++) = *src++;

            *(u++) = *src++;
            *(y++) = *src++;
            *(v++) = *src++;
            *(y++) = *src++;

            *(y++) = *src++;
            *(y++) = *src++;
            *(y++) = *src++;
            *(y++) = *src++;
        }
    }

    *got_frame = 1;

    return avpkt->size;
}

/**
 * Read samples from the input FIFOs, mix, and write to the output link.
 */
static int output_frame(AVFilterLink *outlink, int nb_samples)
{
    AVFilterContext *ctx = outlink->src;
    MixContext      *s = ctx->priv;
    AVFrame *out_buf, *in_buf;
    int i;

    calculate_scales(s, nb_samples);

    out_buf = ff_get_audio_buffer(outlink, nb_samples);
    if (!out_buf)
        return AVERROR(ENOMEM);

    in_buf = ff_get_audio_buffer(outlink, nb_samples);
    if (!in_buf) {
        av_frame_free(&out_buf);
        return AVERROR(ENOMEM);
    }

    for (i = 0; i < s->nb_inputs; i++) {
        if (s->input_state[i] == INPUT_ON) {
            int planes, plane_size, p;

            av_audio_fifo_read(s->fifos[i], (void **)in_buf->extended_data,
                               nb_samples);

            planes     = s->planar ? s->nb_channels : 1;
            plane_size = nb_samples * (s->planar ? 1 : s->nb_channels);
            plane_size = FFALIGN(plane_size, 16);

            for (p = 0; p < planes; p++) {
                s->fdsp->vector_fmac_scalar((float *)out_buf->extended_data[p],
                                           (float *) in_buf->extended_data[p],
                                           s->input_scale[i], plane_size);
            }
        }
    }
    av_frame_free(&in_buf);

    out_buf->pts = s->next_pts;
    if (s->next_pts != AV_NOPTS_VALUE)
        s->next_pts += nb_samples;

    return ff_filter_frame(outlink, out_buf);
}