@@ -11,6 +11,7 @@ version <next>:

 - Sun Rasterfile Encoder

 - remove libpostproc

 - ID3v2 attached pictures reading and writing

+- WMA Lossless decoder

 version 0.8:

@@ -768,6 +768,7 @@ following image formats are supported:

 @item Westwood Audio (SND1)  @tab     @tab  X

 @item Windows Media Audio 1  @tab  X  @tab  X

 @item Windows Media Audio 2  @tab  X  @tab  X

+@item Windows Media Audio Lossless @tab  @tab  X

 @item Windows Media Audio Pro @tab    @tab  X

 @item Windows Media Audio Voice @tab  @tab  X

 @end multitable

@@ -418,6 +418,7 @@ OBJS-$(CONFIG_VP6_DECODER)             += vp6.o vp56.o vp56data.o vp56dsp.o \

 OBJS-$(CONFIG_VP8_DECODER)             += vp8.o vp8dsp.o vp56rac.o

 OBJS-$(CONFIG_VQA_DECODER)             += vqavideo.o

 OBJS-$(CONFIG_WAVPACK_DECODER)         += wavpack.o

+OBJS-$(CONFIG_WMALOSSLESS_DECODER)     += wmalosslessdec.o wma.o

 OBJS-$(CONFIG_WMAPRO_DECODER)          += wmaprodec.o wma.o

 OBJS-$(CONFIG_WMAV1_DECODER)           += wmadec.o wma.o aactab.o

 OBJS-$(CONFIG_WMAV1_ENCODER)           += wmaenc.o wma.o aactab.o

@@ -289,6 +289,7 @@ void avcodec_register_all(void)

     REGISTER_DECODER (VMDAUDIO, vmdaudio);

     REGISTER_ENCDEC  (VORBIS, vorbis);

     REGISTER_DECODER (WAVPACK, wavpack);

+    REGISTER_DECODER (WMALOSSLESS, wmalossless);

     REGISTER_DECODER (WMAPRO, wmapro);

     REGISTER_ENCDEC  (WMAV1, wmav1);

     REGISTER_ENCDEC  (WMAV2, wmav2);

@@ -21,7 +21,7 @@

 #define AVCODEC_VERSION_H

 #define LIBAVCODEC_VERSION_MAJOR 54

-#define LIBAVCODEC_VERSION_MINOR  4

+#define LIBAVCODEC_VERSION_MINOR  5

 #define LIBAVCODEC_VERSION_MICRO  0

 #define LIBAVCODEC_VERSION_INT  AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \

new file mode 100644

@@ -0,0 +1,1248 @@

+/*

+ * Windows Media Audio Lossless decoder

+ * Copyright (c) 2007 Baptiste Coudurier, Benjamin Larsson, Ulion

+ * Copyright (c) 2008 - 2011 Sascha Sommer, Benjamin Larsson

+ * Copyright (c) 2011 Andreas Öman

+ * Copyright (c) 2011 - 2012 Mashiat Sarker Shakkhar

+ *

+ * 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

+ */

+

+#include "avcodec.h"

+#include "internal.h"

+#include "get_bits.h"

+#include "put_bits.h"

+#include "wma.h"

+

+/** current decoder limitations */

+#define WMALL_MAX_CHANNELS      8                       ///< max number of handled channels

+#define MAX_SUBFRAMES          32                       ///< max number of subframes per channel

+#define MAX_BANDS              29                       ///< max number of scale factor bands

+#define MAX_FRAMESIZE       32768                       ///< maximum compressed frame size

+

+#define WMALL_BLOCK_MIN_BITS    6                       ///< log2 of min block size

+#define WMALL_BLOCK_MAX_BITS   12                       ///< log2 of max block size

+#define WMALL_BLOCK_MAX_SIZE (1 << WMALL_BLOCK_MAX_BITS)    ///< maximum block size

+#define WMALL_BLOCK_SIZES    (WMALL_BLOCK_MAX_BITS - WMALL_BLOCK_MIN_BITS + 1) ///< possible block sizes

+

+

+/**

+ * @brief frame-specific decoder context for a single channel

+ */

+typedef struct {

+    int16_t     prev_block_len;                         ///< length of the previous block

+    uint8_t     transmit_coefs;

+    uint8_t     num_subframes;

+    uint16_t    subframe_len[MAX_SUBFRAMES];            ///< subframe length in samples

+    uint16_t    subframe_offsets[MAX_SUBFRAMES];        ///< subframe positions in the current frame

+    uint8_t     cur_subframe;                           ///< current subframe number

+    uint16_t    decoded_samples;                        ///< number of already processed samples

+    int         quant_step;                             ///< quantization step for the current subframe

+    int         transient_counter;                      ///< number of transient samples from the beginning of the transient zone

+} WmallChannelCtx;

+

+/**

+ * @brief main decoder context

+ */

+typedef struct WmallDecodeCtx {

+    /* generic decoder variables */

+    AVCodecContext  *avctx;

+    AVFrame         frame;

+    uint8_t         frame_data[MAX_FRAMESIZE + FF_INPUT_BUFFER_PADDING_SIZE];  ///< compressed frame data

+    PutBitContext   pb;                             ///< context for filling the frame_data buffer

+

+    /* frame size dependent frame information (set during initialization) */

+    uint32_t        decode_flags;                   ///< used compression features

+    int             len_prefix;                     ///< frame is prefixed with its length

+    int             dynamic_range_compression;      ///< frame contains DRC data

+    uint8_t         bits_per_sample;                ///< integer audio sample size for the unscaled IMDCT output (used to scale to [-1.0, 1.0])

+    uint16_t        samples_per_frame;              ///< number of samples to output

+    uint16_t        log2_frame_size;

+    int8_t          num_channels;                   ///< number of channels in the stream (same as AVCodecContext.num_channels)

+    int8_t          lfe_channel;                    ///< lfe channel index

+    uint8_t         max_num_subframes;

+    uint8_t         subframe_len_bits;              ///< number of bits used for the subframe length

+    uint8_t         max_subframe_len_bit;           ///< flag indicating that the subframe is of maximum size when the first subframe length bit is 1

+    uint16_t        min_samples_per_subframe;

+

+    /* packet decode state */

+    GetBitContext   pgb;                            ///< bitstream reader context for the packet

+    int             next_packet_start;              ///< start offset of the next WMA packet in the demuxer packet

+    uint8_t         packet_offset;                  ///< offset to the frame in the packet

+    uint8_t         packet_sequence_number;         ///< current packet number

+    int             num_saved_bits;                 ///< saved number of bits

+    int             frame_offset;                   ///< frame offset in the bit reservoir

+    int             subframe_offset;                ///< subframe offset in the bit reservoir

+    uint8_t         packet_loss;                    ///< set in case of bitstream error

+    uint8_t         packet_done;                    ///< set when a packet is fully decoded

+

+    /* frame decode state */

+    uint32_t        frame_num;                      ///< current frame number (not used for decoding)

+    GetBitContext   gb;                             ///< bitstream reader context

+    int             buf_bit_size;                   ///< buffer size in bits

+    int16_t         *samples_16;                    ///< current samplebuffer pointer (16-bit)

+    int16_t         *samples_16_end;                ///< maximum samplebuffer pointer

+    int             *samples_32;                    ///< current samplebuffer pointer (24-bit)

+    int             *samples_32_end;                ///< maximum samplebuffer pointer

+    uint8_t         drc_gain;                       ///< gain for the DRC tool

+    int8_t          skip_frame;                     ///< skip output step

+    int8_t          parsed_all_subframes;           ///< all subframes decoded?

+

+    /* subframe/block decode state */

+    int16_t         subframe_len;                   ///< current subframe length

+    int8_t          channels_for_cur_subframe;      ///< number of channels that contain the subframe

+    int8_t          channel_indexes_for_cur_subframe[WMALL_MAX_CHANNELS];

+

+    WmallChannelCtx channel[WMALL_MAX_CHANNELS];    ///< per channel data

+

+    // WMA Lossless-specific

+

+    uint8_t do_arith_coding;

+    uint8_t do_ac_filter;

+    uint8_t do_inter_ch_decorr;

+    uint8_t do_mclms;

+    uint8_t do_lpc;

+

+    int8_t  acfilter_order;

+    int8_t  acfilter_scaling;

+    int64_t acfilter_coeffs[16];

+    int     acfilter_prevvalues[2][16];

+

+    int8_t  mclms_order;

+    int8_t  mclms_scaling;

+    int16_t mclms_coeffs[128];

+    int16_t mclms_coeffs_cur[4];

+    int16_t mclms_prevvalues[64];

+    int16_t mclms_updates[64];

+    int     mclms_recent;

+

+    int     movave_scaling;

+    int     quant_stepsize;

+

+    struct {

+        int order;

+        int scaling;

+        int coefsend;

+        int bitsend;

+        int16_t coefs[256];

+        int16_t lms_prevvalues[512];

+        int16_t lms_updates[512];

+        int recent;

+    } cdlms[2][9];

+

+    int cdlms_ttl[2];

+

+    int bV3RTM;

+

+    int is_channel_coded[2];

+    int update_speed[2];

+

+    int transient[2];

+    int transient_pos[2];

+    int seekable_tile;

+

+    int ave_sum[2];

+

+    int channel_residues[2][2048];

+

+    int lpc_coefs[2][40];

+    int lpc_order;

+    int lpc_scaling;

+    int lpc_intbits;

+

+    int channel_coeffs[2][2048];

+} WmallDecodeCtx;

+

+

+static av_cold int decode_init(AVCodecContext *avctx)

+{

+    WmallDecodeCtx *s  = avctx->priv_data;

+    uint8_t *edata_ptr = avctx->extradata;

+    unsigned int channel_mask;

+    int i, log2_max_num_subframes, num_possible_block_sizes;

+

+    s->avctx = avctx;

+    init_put_bits(&s->pb, s->frame_data, MAX_FRAMESIZE);

+

+    if (avctx->extradata_size >= 18) {

+        s->decode_flags    = AV_RL16(edata_ptr + 14);

+        channel_mask       = AV_RL32(edata_ptr +  2);

+        s->bits_per_sample = AV_RL16(edata_ptr);

+        if (s->bits_per_sample == 16)

+            avctx->sample_fmt = AV_SAMPLE_FMT_S16;

+        else if (s->bits_per_sample == 24) {

+            avctx->sample_fmt = AV_SAMPLE_FMT_S32;

+            av_log_missing_feature(avctx, "bit-depth higher than 16", 0);

+            return AVERROR_PATCHWELCOME;

+        } else {

+            av_log(avctx, AV_LOG_ERROR, "Unknown bit-depth: %d\n",

+                   s->bits_per_sample);

+            return AVERROR_INVALIDDATA;

+        }

+        /* dump the extradata */

+        for (i = 0; i < avctx->extradata_size; i++)

+            av_dlog(avctx, AV_LOG_DEBUG, "[%x] ", avctx->extradata[i]);

+        av_dlog(avctx, AV_LOG_DEBUG, "\n");

+

+    } else {

+        av_log_ask_for_sample(avctx, "Unsupported extradata size\n");

+        return AVERROR_INVALIDDATA;

+    }

+

+    /* generic init */

+    s->log2_frame_size = av_log2(avctx->block_align) + 4;

+

+    /* frame info */

+    s->skip_frame  = 1; /* skip first frame */

+    s->packet_loss = 1;

+    s->len_prefix  = s->decode_flags & 0x40;

+

+    /* get frame len */

+    s->samples_per_frame = 1 << ff_wma_get_frame_len_bits(avctx->sample_rate,

+                                                          3, s->decode_flags);

+

+    /* init previous block len */

+    for (i = 0; i < avctx->channels; i++)

+        s->channel[i].prev_block_len = s->samples_per_frame;

+

+    /* subframe info */

+    log2_max_num_subframes  = (s->decode_flags & 0x38) >> 3;

+    s->max_num_subframes    = 1 << log2_max_num_subframes;

+    s->max_subframe_len_bit = 0;

+    s->subframe_len_bits    = av_log2(log2_max_num_subframes) + 1;

+

+    num_possible_block_sizes     = log2_max_num_subframes + 1;

+    s->min_samples_per_subframe  = s->samples_per_frame / s->max_num_subframes;

+    s->dynamic_range_compression = s->decode_flags & 0x80;

+    s->bV3RTM                    = s->decode_flags & 0x100;

+

+    if (s->max_num_subframes > MAX_SUBFRAMES) {

+        av_log(avctx, AV_LOG_ERROR, "invalid number of subframes %i\n",

+               s->max_num_subframes);

+        return AVERROR_INVALIDDATA;

+    }

+

+    s->num_channels = avctx->channels;

+

+    /* extract lfe channel position */

+    s->lfe_channel = -1;

+

+    if (channel_mask & 8) {

+        unsigned int mask;

+        for (mask = 1; mask < 16; mask <<= 1)

+            if (channel_mask & mask)

+                ++s->lfe_channel;

+    }

+

+    if (s->num_channels < 0) {

+        av_log(avctx, AV_LOG_ERROR, "invalid number of channels %d\n",

+               s->num_channels);

+        return AVERROR_INVALIDDATA;

+    } else if (s->num_channels > WMALL_MAX_CHANNELS) {

+        av_log_ask_for_sample(avctx, "unsupported number of channels\n");

+        return AVERROR_PATCHWELCOME;

+    }

+

+    avcodec_get_frame_defaults(&s->frame);

+    avctx->coded_frame    = &s->frame;

+    avctx->channel_layout = channel_mask;

+    return 0;

+}

+

+/**

+ * @brief Decode the subframe length.

+ * @param s      context

+ * @param offset sample offset in the frame

+ * @return decoded subframe length on success, < 0 in case of an error

+ */

+static int decode_subframe_length(WmallDecodeCtx *s, int offset)

+{

+    int frame_len_ratio, subframe_len, len;

+

+    /* no need to read from the bitstream when only one length is possible */

+    if (offset == s->samples_per_frame - s->min_samples_per_subframe)

+        return s->min_samples_per_subframe;

+

+    len             = av_log2(s->max_num_subframes - 1) + 1;

+    frame_len_ratio = get_bits(&s->gb, len);

+    subframe_len    = s->min_samples_per_subframe * (frame_len_ratio + 1);

+

+    /* sanity check the length */

+    if (subframe_len < s->min_samples_per_subframe ||

+        subframe_len > s->samples_per_frame) {

+        av_log(s->avctx, AV_LOG_ERROR, "broken frame: subframe_len %i\n",

+               subframe_len);

+        return AVERROR_INVALIDDATA;

+    }

+    return subframe_len;

+}

+

+/**

+ * @brief Decode how the data in the frame is split into subframes.

+ *       Every WMA frame contains the encoded data for a fixed number of

+ *       samples per channel. The data for every channel might be split

+ *       into several subframes. This function will reconstruct the list of

+ *       subframes for every channel.

+ *

+ *       If the subframes are not evenly split, the algorithm estimates the

+ *       channels with the lowest number of total samples.

+ *       Afterwards, for each of these channels a bit is read from the

+ *       bitstream that indicates if the channel contains a subframe with the

+ *       next subframe size that is going to be read from the bitstream or not.

+ *       If a channel contains such a subframe, the subframe size gets added to

+ *       the channel's subframe list.

+ *       The algorithm repeats these steps until the frame is properly divided

+ *       between the individual channels.

+ *

+ * @param s context

+ * @return 0 on success, < 0 in case of an error

+ */

+static int decode_tilehdr(WmallDecodeCtx *s)

+{

+    uint16_t num_samples[WMALL_MAX_CHANNELS] = { 0 }; /* sum of samples for all currently known subframes of a channel */

+    uint8_t  contains_subframe[WMALL_MAX_CHANNELS];   /* flag indicating if a channel contains the current subframe */

+    int channels_for_cur_subframe = s->num_channels;  /* number of channels that contain the current subframe */

+    int fixed_channel_layout = 0;                     /* flag indicating that all channels use the same subfra2me offsets and sizes */

+    int min_channel_len = 0;                          /* smallest sum of samples (channels with this length will be processed first) */

+    int c, tile_aligned;

+

+    /* reset tiling information */

+    for (c = 0; c < s->num_channels; c++)

+        s->channel[c].num_subframes = 0;

+

+    tile_aligned = get_bits1(&s->gb);

+    if (s->max_num_subframes == 1 || tile_aligned)

+        fixed_channel_layout = 1;

+

+    /* loop until the frame data is split between the subframes */

+    do {

+        int subframe_len;

+

+        /* check which channels contain the subframe */

+        for (c = 0; c < s->num_channels; c++) {

+            if (num_samples[c] == min_channel_len) {

+                if (fixed_channel_layout || channels_for_cur_subframe == 1 ||

+                   (min_channel_len == s->samples_per_frame - s->min_samples_per_subframe)) {

+                    contains_subframe[c] = 1;

+                } else {

+                    contains_subframe[c] = get_bits1(&s->gb);

+                }

+            } else

+                contains_subframe[c] = 0;

+        }

+

+        /* get subframe length, subframe_len == 0 is not allowed */

+        if ((subframe_len = decode_subframe_length(s, min_channel_len)) <= 0)

+            return AVERROR_INVALIDDATA;

+        /* add subframes to the individual channels and find new min_channel_len */

+        min_channel_len += subframe_len;

+        for (c = 0; c < s->num_channels; c++) {

+            WmallChannelCtx *chan = &s->channel[c];

+

+            if (contains_subframe[c]) {

+                if (chan->num_subframes >= MAX_SUBFRAMES) {

+                    av_log(s->avctx, AV_LOG_ERROR,

+                           "broken frame: num subframes > 31\n");

+                    return AVERROR_INVALIDDATA;

+                }

+                chan->subframe_len[chan->num_subframes] = subframe_len;

+                num_samples[c] += subframe_len;

+                ++chan->num_subframes;

+                if (num_samples[c] > s->samples_per_frame) {

+                    av_log(s->avctx, AV_LOG_ERROR, "broken frame: "

+                           "channel len(%d) > samples_per_frame(%d)\n",

+                           num_samples[c], s->samples_per_frame);

+                    return AVERROR_INVALIDDATA;

+                }

+            } else if (num_samples[c] <= min_channel_len) {

+                if (num_samples[c] < min_channel_len) {

+                    channels_for_cur_subframe = 0;

+                    min_channel_len = num_samples[c];

+                }

+                ++channels_for_cur_subframe;

+            }

+        }

+    } while (min_channel_len < s->samples_per_frame);

+

+    for (c = 0; c < s->num_channels; c++) {

+        int i, offset = 0;

+        for (i = 0; i < s->channel[c].num_subframes; i++) {

+            s->channel[c].subframe_offsets[i] = offset;

+            offset += s->channel[c].subframe_len[i];

+        }

+    }

+

+    return 0;

+}

+

+static void decode_ac_filter(WmallDecodeCtx *s)

+{

+    int i;

+    s->acfilter_order   = get_bits(&s->gb, 4) + 1;

+    s->acfilter_scaling = get_bits(&s->gb, 4);

+

+    for (i = 0; i < s->acfilter_order; i++)

+        s->acfilter_coeffs[i] = get_bits(&s->gb, s->acfilter_scaling) + 1;

+}

+

+static void decode_mclms(WmallDecodeCtx *s)

+{

+    s->mclms_order   = (get_bits(&s->gb, 4) + 1) * 2;

+    s->mclms_scaling = get_bits(&s->gb, 4);

+    if (get_bits1(&s->gb)) {

+        int i, send_coef_bits;

+        int cbits = av_log2(s->mclms_scaling + 1);

+        assert(cbits == my_log2(s->mclms_scaling + 1));

+        if (1 << cbits < s->mclms_scaling + 1)

+            cbits++;

+

+        send_coef_bits = (cbits ? get_bits(&s->gb, cbits) : 0) + 2;

+

+        for (i = 0; i < s->mclms_order * s->num_channels * s->num_channels; i++)

+            s->mclms_coeffs[i] = get_bits(&s->gb, send_coef_bits);

+

+        for (i = 0; i < s->num_channels; i++) {

+            int c;

+            for (c = 0; c < i; c++)

+                s->mclms_coeffs_cur[i * s->num_channels + c] = get_bits(&s->gb, send_coef_bits);

+        }

+    }

+}

+

+static void decode_cdlms(WmallDecodeCtx *s)

+{

+    int c, i;

+    int cdlms_send_coef = get_bits1(&s->gb);

+

+    for (c = 0; c < s->num_channels; c++) {

+        s->cdlms_ttl[c] = get_bits(&s->gb, 3) + 1;

+        for (i = 0; i < s->cdlms_ttl[c]; i++)

+            s->cdlms[c][i].order = (get_bits(&s->gb, 7) + 1) * 8;

+

+        for (i = 0; i < s->cdlms_ttl[c]; i++)

+            s->cdlms[c][i].scaling = get_bits(&s->gb, 4);

+

+        if (cdlms_send_coef) {

+            for (i = 0; i < s->cdlms_ttl[c]; i++) {

+                int cbits, shift_l, shift_r, j;

+                cbits = av_log2(s->cdlms[c][i].order);

+                if ((1 << cbits) < s->cdlms[c][i].order)

+                    cbits++;

+                s->cdlms[c][i].coefsend = get_bits(&s->gb, cbits) + 1;

+

+                cbits = av_log2(s->cdlms[c][i].scaling + 1);

+                if ((1 << cbits) < s->cdlms[c][i].scaling + 1)

+                    cbits++;

+

+                s->cdlms[c][i].bitsend = get_bits(&s->gb, cbits) + 2;

+                shift_l = 32 - s->cdlms[c][i].bitsend;

+                shift_r = 32 - s->cdlms[c][i].scaling - 2;

+                for (j = 0; j < s->cdlms[c][i].coefsend; j++)

+                    s->cdlms[c][i].coefs[j] =

+                        (get_bits(&s->gb, s->cdlms[c][i].bitsend) << shift_l) >> shift_r;

+            }

+        }

+    }

+}

+

+static int decode_channel_residues(WmallDecodeCtx *s, int ch, int tile_size)

+{

+    int i = 0;

+    unsigned int ave_mean;

+    s->transient[ch] = get_bits1(&s->gb);

+    if (s->transient[ch]) {

+        s->transient_pos[ch] = get_bits(&s->gb, av_log2(tile_size));

+        if (s->transient_pos[ch])

+            s->transient[ch] = 0;

+        s->channel[ch].transient_counter =

+            FFMAX(s->channel[ch].transient_counter, s->samples_per_frame / 2);

+    } else if (s->channel[ch].transient_counter)

+        s->transient[ch] = 1;

+

+    if (s->seekable_tile) {

+        ave_mean = get_bits(&s->gb, s->bits_per_sample);

+        s->ave_sum[ch] = ave_mean << (s->movave_scaling + 1);

+    }

+

+    if (s->seekable_tile) {

+        if (s->do_inter_ch_decorr)

+            s->channel_residues[ch][0] = get_sbits(&s->gb, s->bits_per_sample + 1);

+        else

+            s->channel_residues[ch][0] = get_sbits(&s->gb, s->bits_per_sample);

+        i++;

+    }

+    for (; i < tile_size; i++) {

+        int quo = 0, rem, rem_bits, residue;

+        while(get_bits1(&s->gb)) {

+            quo++;

+            if (get_bits_left(&s->gb) <= 0)

+                return -1;

+        }

+        if (quo >= 32)

+            quo += get_bits_long(&s->gb, get_bits(&s->gb, 5) + 1);

+

+        ave_mean = (s->ave_sum[ch] + (1 << s->movave_scaling)) >> (s->movave_scaling + 1);

+        if (ave_mean <= 1)

+            residue = quo;

+        else {

+            rem_bits = av_ceil_log2(ave_mean);

+            rem      = rem_bits ? get_bits(&s->gb, rem_bits) : 0;

+            residue  = (quo << rem_bits) + rem;

+        }

+

+        s->ave_sum[ch] = residue + s->ave_sum[ch] -

+                         (s->ave_sum[ch] >> s->movave_scaling);

+

+        if (residue & 1)

+            residue = -(residue >> 1) - 1;

+        else

+            residue = residue >> 1;

+        s->channel_residues[ch][i] = residue;

+    }

+

+    return 0;

+

+}

+

+static void decode_lpc(WmallDecodeCtx *s)

+{

+    int ch, i, cbits;

+    s->lpc_order   = get_bits(&s->gb, 5) + 1;

+    s->lpc_scaling = get_bits(&s->gb, 4);

+    s->lpc_intbits = get_bits(&s->gb, 3) + 1;

+    cbits = s->lpc_scaling + s->lpc_intbits;

+    for (ch = 0; ch < s->num_channels; ch++)

+        for (i = 0; i < s->lpc_order; i++)

+            s->lpc_coefs[ch][i] = get_sbits(&s->gb, cbits);

+}

+

+static void clear_codec_buffers(WmallDecodeCtx *s)

+{

+    int ich, ilms;

+

+    memset(s->acfilter_coeffs,     0, sizeof(s->acfilter_coeffs));

+    memset(s->acfilter_prevvalues, 0, sizeof(s->acfilter_prevvalues));

+    memset(s->lpc_coefs,           0, sizeof(s->lpc_coefs));

+

+    memset(s->mclms_coeffs,     0, sizeof(s->mclms_coeffs));

+    memset(s->mclms_coeffs_cur, 0, sizeof(s->mclms_coeffs_cur));

+    memset(s->mclms_prevvalues, 0, sizeof(s->mclms_prevvalues));

+    memset(s->mclms_updates,    0, sizeof(s->mclms_updates));

+

+    for (ich = 0; ich < s->num_channels; ich++) {

+        for (ilms = 0; ilms < s->cdlms_ttl[ich]; ilms++) {

+            memset(s->cdlms[ich][ilms].coefs, 0,

+                   sizeof(s->cdlms[ich][ilms].coefs));

+            memset(s->cdlms[ich][ilms].lms_prevvalues, 0,

+                   sizeof(s->cdlms[ich][ilms].lms_prevvalues));

+            memset(s->cdlms[ich][ilms].lms_updates, 0,

+                   sizeof(s->cdlms[ich][ilms].lms_updates));

+        }

+        s->ave_sum[ich] = 0;

+    }

+}

+

+/**

+ * @brief Reset filter parameters and transient area at new seekable tile.

+ */

+static void reset_codec(WmallDecodeCtx *s)

+{

+    int ich, ilms;

+    s->mclms_recent = s->mclms_order * s->num_channels;

+    for (ich = 0; ich < s->num_channels; ich++) {

+        for (ilms = 0; ilms < s->cdlms_ttl[ich]; ilms++)

+            s->cdlms[ich][ilms].recent = s->cdlms[ich][ilms].order;

+        /* first sample of a seekable subframe is considered as the starting of

+            a transient area which is samples_per_frame samples long */

+        s->channel[ich].transient_counter = s->samples_per_frame;

+        s->transient[ich]     = 1;

+        s->transient_pos[ich] = 0;

+    }

+}

+

+static void mclms_update(WmallDecodeCtx *s, int icoef, int *pred)

+{

+    int i, j, ich, pred_error;

+    int order        = s->mclms_order;

+    int num_channels = s->num_channels;

+    int range        = 1 << (s->bits_per_sample - 1);

+

+    for (ich = 0; ich < num_channels; ich++) {

+        pred_error = s->channel_residues[ich][icoef] - pred[ich];

+        if (pred_error > 0) {

+            for (i = 0; i < order * num_channels; i++)

+                s->mclms_coeffs[i + ich * order * num_channels] +=

+                    s->mclms_updates[s->mclms_recent + i];

+            for (j = 0; j < ich; j++) {

+                if (s->channel_residues[j][icoef] > 0)

+                    s->mclms_coeffs_cur[ich * num_channels + j] += 1;

+                else if (s->channel_residues[j][icoef] < 0)

+                    s->mclms_coeffs_cur[ich * num_channels + j] -= 1;

+            }

+        } else if (pred_error < 0) {

+            for (i = 0; i < order * num_channels; i++)

+                s->mclms_coeffs[i + ich * order * num_channels] -=

+                    s->mclms_updates[s->mclms_recent + i];

+            for (j = 0; j < ich; j++) {

+                if (s->channel_residues[j][icoef] > 0)

+                    s->mclms_coeffs_cur[ich * num_channels + j] -= 1;

+                else if (s->channel_residues[j][icoef] < 0)

+                    s->mclms_coeffs_cur[ich * num_channels + j] += 1;

+            }

+        }

+    }

+

+    for (ich = num_channels - 1; ich >= 0; ich--) {

+        s->mclms_recent--;

+        s->mclms_prevvalues[s->mclms_recent] = s->channel_residues[ich][icoef];

+        if (s->channel_residues[ich][icoef] > range - 1)

+            s->mclms_prevvalues[s->mclms_recent] = range - 1;

+        else if (s->channel_residues[ich][icoef] < -range)

+            s->mclms_prevvalues[s->mclms_recent] = -range;

+

+        s->mclms_updates[s->mclms_recent] = 0;

+        if (s->channel_residues[ich][icoef] > 0)

+            s->mclms_updates[s->mclms_recent] = 1;

+        else if (s->channel_residues[ich][icoef] < 0)

+            s->mclms_updates[s->mclms_recent] = -1;

+    }

+

+    if (s->mclms_recent == 0) {

+        memcpy(&s->mclms_prevvalues[order * num_channels],

+               s->mclms_prevvalues,

+               2 * order * num_channels);

+        memcpy(&s->mclms_updates[order * num_channels],

+               s->mclms_updates,

+               2 * order * num_channels);

+        s->mclms_recent = num_channels * order;

+    }

+}

+

+static void mclms_predict(WmallDecodeCtx *s, int icoef, int *pred)

+{

+    int ich, i;

+    int order        = s->mclms_order;

+    int num_channels = s->num_channels;

+

+    for (ich = 0; ich < num_channels; ich++) {

+        if (!s->is_channel_coded[ich])

+            continue;

+        pred[ich] = 0;

+        for (i = 0; i < order * num_channels; i++)

+            pred[ich] += s->mclms_prevvalues[i + s->mclms_recent] *

+                         s->mclms_coeffs[i + order * num_channels * ich];

+        for (i = 0; i < ich; i++)

+            pred[ich] += s->channel_residues[i][icoef] *

+                         s->mclms_coeffs_cur[i + num_channels * ich];

+        pred[ich] += 1 << s->mclms_scaling - 1;

+        pred[ich] >>= s->mclms_scaling;

+        s->channel_residues[ich][icoef] += pred[ich];

+    }

+}

+

+static void revert_mclms(WmallDecodeCtx *s, int tile_size)

+{

+    int icoef, pred[WMALL_MAX_CHANNELS] = { 0 };

+    for (icoef = 0; icoef < tile_size; icoef++) {

+        mclms_predict(s, icoef, pred);

+        mclms_update(s, icoef, pred);

+    }

+}

+

+static int lms_predict(WmallDecodeCtx *s, int ich, int ilms)

+{

+    int pred = 0, icoef;

+    int recent = s->cdlms[ich][ilms].recent;

+

+    for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)

+        pred += s->cdlms[ich][ilms].coefs[icoef] *

+                s->cdlms[ich][ilms].lms_prevvalues[icoef + recent];

+

+    return pred;

+}

+

+static void lms_update(WmallDecodeCtx *s, int ich, int ilms,

+                       int input, int residue)

+{

+    int icoef;

+    int recent = s->cdlms[ich][ilms].recent;

+    int range  = 1 << s->bits_per_sample - 1;

+

+    if (residue < 0) {

+        for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)

+            s->cdlms[ich][ilms].coefs[icoef] -=

+                s->cdlms[ich][ilms].lms_updates[icoef + recent];

+    } else if (residue > 0) {

+        for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)

+            s->cdlms[ich][ilms].coefs[icoef] +=

+                s->cdlms[ich][ilms].lms_updates[icoef + recent];

+    }

+

+    if (recent)

+        recent--;

+    else {

+        memcpy(&s->cdlms[ich][ilms].lms_prevvalues[s->cdlms[ich][ilms].order],

+               s->cdlms[ich][ilms].lms_prevvalues,

+               2 * s->cdlms[ich][ilms].order);

+        memcpy(&s->cdlms[ich][ilms].lms_updates[s->cdlms[ich][ilms].order],

+               s->cdlms[ich][ilms].lms_updates,

+               2 * s->cdlms[ich][ilms].order);

+        recent = s->cdlms[ich][ilms].order - 1;

+    }

+

+    s->cdlms[ich][ilms].lms_prevvalues[recent] = av_clip(input, -range, range - 1);

+    if (!input)

+        s->cdlms[ich][ilms].lms_updates[recent] = 0;

+    else if (input < 0)

+        s->cdlms[ich][ilms].lms_updates[recent] = -s->update_speed[ich];

+    else

+        s->cdlms[ich][ilms].lms_updates[recent] = s->update_speed[ich];

+

+    s->cdlms[ich][ilms].lms_updates[recent + (s->cdlms[ich][ilms].order >> 4)] >>= 2;

+    s->cdlms[ich][ilms].lms_updates[recent + (s->cdlms[ich][ilms].order >> 3)] >>= 1;

+    s->cdlms[ich][ilms].recent = recent;

+}

+

+static void use_high_update_speed(WmallDecodeCtx *s, int ich)

+{

+    int ilms, recent, icoef;

+    for (ilms = s->cdlms_ttl[ich] - 1; ilms >= 0; ilms--) {

+        recent = s->cdlms[ich][ilms].recent;

+        if (s->update_speed[ich] == 16)

+            continue;

+        if (s->bV3RTM) {

+            for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)

+                s->cdlms[ich][ilms].lms_updates[icoef + recent] *= 2;

+        } else {

+            for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)

+                s->cdlms[ich][ilms].lms_updates[icoef] *= 2;

+        }

+    }

+    s->update_speed[ich] = 16;

+}

+

+static void use_normal_update_speed(WmallDecodeCtx *s, int ich)

+{

+    int ilms, recent, icoef;

+    for (ilms = s->cdlms_ttl[ich] - 1; ilms >= 0; ilms--) {

+        recent = s->cdlms[ich][ilms].recent;

+        if (s->update_speed[ich] == 8)

+            continue;

+        if (s->bV3RTM)

+            for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)

+                s->cdlms[ich][ilms].lms_updates[icoef + recent] /= 2;

+        else

+            for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)

+                s->cdlms[ich][ilms].lms_updates[icoef] /= 2;

+    }

+    s->update_speed[ich] = 8;

+}

+

+static void revert_cdlms(WmallDecodeCtx *s, int ch,

+                         int coef_begin, int coef_end)

+{

+    int icoef, pred, ilms, num_lms, residue, input;

+

+    num_lms = s->cdlms_ttl[ch];

+    for (ilms = num_lms - 1; ilms >= 0; ilms--) {

+        for (icoef = coef_begin; icoef < coef_end; icoef++) {

+            pred = 1 << (s->cdlms[ch][ilms].scaling - 1);

+            residue = s->channel_residues[ch][icoef];

+            pred += lms_predict(s, ch, ilms);

+            input = residue + (pred >> s->cdlms[ch][ilms].scaling);

+            lms_update(s, ch, ilms, input, residue);

+            s->channel_residues[ch][icoef] = input;

+        }

+    }

+}

+

+static void revert_inter_ch_decorr(WmallDecodeCtx *s, int tile_size)

+{

+    if (s->num_channels != 2)

+        return;

+    else if (s->is_channel_coded[0] && s->is_channel_coded[1]) {

+        int icoef;

+        for (icoef = 0; icoef < tile_size; icoef++) {

+            s->channel_residues[0][icoef] -= s->channel_residues[1][icoef] >> 1;

+            s->channel_residues[1][icoef] += s->channel_residues[0][icoef];

+        }

+    }

+}

+

+static void revert_acfilter(WmallDecodeCtx *s, int tile_size)

+{

+    int ich, pred, i, j;

+    int64_t *filter_coeffs = s->acfilter_coeffs;

+    int scaling            = s->acfilter_scaling;

+    int order              = s->acfilter_order;

+

+    for (ich = 0; ich < s->num_channels; ich++) {

+        int *prevvalues = s->acfilter_prevvalues[ich];

+        for (i = 0; i < order; i++) {

+            pred = 0;

+            for (j = 0; j < order; j++) {

+                if (i <= j)

+                    pred += filter_coeffs[j] * prevvalues[j - i];

+                else

+                    pred += s->channel_residues[ich][i - j - 1] * filter_coeffs[j];

+            }

+            pred >>= scaling;

+            s->channel_residues[ich][i] += pred;

+        }

+        for (i = order; i < tile_size; i++) {

+            pred = 0;

+            for (j = 0; j < order; j++)

+                pred += s->channel_residues[ich][i - j - 1] * filter_coeffs[j];

+            pred >>= scaling;

+            s->channel_residues[ich][i] += pred;

+        }

+        for (j = 0; j < order; j++)

+            prevvalues[j] = s->channel_residues[ich][tile_size - j - 1];

+    }

+}

+

+static int decode_subframe(WmallDecodeCtx *s)

+{

+    int offset        = s->samples_per_frame;

+    int subframe_len  = s->samples_per_frame;

+    int total_samples = s->samples_per_frame * s->num_channels;

+    int i, j, rawpcm_tile, padding_zeroes;

+

+    s->subframe_offset = get_bits_count(&s->gb);

+

+    /* reset channel context and find the next block offset and size

+        == the next block of the channel with the smallest number of

+        decoded samples */

+    for (i = 0; i < s->num_channels; i++) {

+        if (offset > s->channel[i].decoded_samples) {

+            offset = s->channel[i].decoded_samples;