2016-10-26 08:10:19 +02:00
|
|
|
/*
|
|
|
|
* generic decoding-related code
|
|
|
|
*
|
|
|
|
* 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 <stdint.h>
|
|
|
|
#include <string.h>
|
|
|
|
|
|
|
|
#include "config.h"
|
|
|
|
|
|
|
|
#include "libavutil/avassert.h"
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
#include "libavutil/avstring.h"
|
2016-10-26 08:10:19 +02:00
|
|
|
#include "libavutil/common.h"
|
|
|
|
#include "libavutil/frame.h"
|
|
|
|
#include "libavutil/hwcontext.h"
|
|
|
|
#include "libavutil/imgutils.h"
|
|
|
|
|
|
|
|
#include "avcodec.h"
|
|
|
|
#include "bytestream.h"
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
#include "decode.h"
|
2016-10-26 08:10:19 +02:00
|
|
|
#include "internal.h"
|
|
|
|
#include "thread.h"
|
|
|
|
|
|
|
|
static int apply_param_change(AVCodecContext *avctx, AVPacket *avpkt)
|
|
|
|
{
|
|
|
|
int size = 0, ret;
|
|
|
|
const uint8_t *data;
|
|
|
|
uint32_t flags;
|
|
|
|
|
|
|
|
data = av_packet_get_side_data(avpkt, AV_PKT_DATA_PARAM_CHANGE, &size);
|
|
|
|
if (!data)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (!(avctx->codec->capabilities & AV_CODEC_CAP_PARAM_CHANGE)) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "This decoder does not support parameter "
|
|
|
|
"changes, but PARAM_CHANGE side data was sent to it.\n");
|
|
|
|
ret = AVERROR(EINVAL);
|
|
|
|
goto fail2;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (size < 4)
|
|
|
|
goto fail;
|
|
|
|
|
|
|
|
flags = bytestream_get_le32(&data);
|
|
|
|
size -= 4;
|
|
|
|
|
|
|
|
if (flags & AV_SIDE_DATA_PARAM_CHANGE_CHANNEL_COUNT) {
|
|
|
|
if (size < 4)
|
|
|
|
goto fail;
|
|
|
|
avctx->channels = bytestream_get_le32(&data);
|
|
|
|
size -= 4;
|
|
|
|
}
|
|
|
|
if (flags & AV_SIDE_DATA_PARAM_CHANGE_CHANNEL_LAYOUT) {
|
|
|
|
if (size < 8)
|
|
|
|
goto fail;
|
|
|
|
avctx->channel_layout = bytestream_get_le64(&data);
|
|
|
|
size -= 8;
|
|
|
|
}
|
|
|
|
if (flags & AV_SIDE_DATA_PARAM_CHANGE_SAMPLE_RATE) {
|
|
|
|
if (size < 4)
|
|
|
|
goto fail;
|
|
|
|
avctx->sample_rate = bytestream_get_le32(&data);
|
|
|
|
size -= 4;
|
|
|
|
}
|
|
|
|
if (flags & AV_SIDE_DATA_PARAM_CHANGE_DIMENSIONS) {
|
|
|
|
if (size < 8)
|
|
|
|
goto fail;
|
|
|
|
avctx->width = bytestream_get_le32(&data);
|
|
|
|
avctx->height = bytestream_get_le32(&data);
|
|
|
|
size -= 8;
|
|
|
|
ret = ff_set_dimensions(avctx, avctx->width, avctx->height);
|
|
|
|
if (ret < 0)
|
|
|
|
goto fail2;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
fail:
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "PARAM_CHANGE side data too small.\n");
|
|
|
|
ret = AVERROR_INVALIDDATA;
|
|
|
|
fail2:
|
|
|
|
if (ret < 0) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "Error applying parameter changes.\n");
|
|
|
|
if (avctx->err_recognition & AV_EF_EXPLODE)
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-10-26 13:41:12 +02:00
|
|
|
static int extract_packet_props(AVCodecInternal *avci, const AVPacket *pkt)
|
|
|
|
{
|
|
|
|
av_packet_unref(avci->last_pkt_props);
|
|
|
|
if (pkt)
|
|
|
|
return av_packet_copy_props(avci->last_pkt_props, pkt);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-10-26 08:10:19 +02:00
|
|
|
static int unrefcount_frame(AVCodecInternal *avci, AVFrame *frame)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
/* move the original frame to our backup */
|
|
|
|
av_frame_unref(avci->to_free);
|
|
|
|
av_frame_move_ref(avci->to_free, frame);
|
|
|
|
|
|
|
|
/* now copy everything except the AVBufferRefs back
|
|
|
|
* note that we make a COPY of the side data, so calling av_frame_free() on
|
|
|
|
* the caller's frame will work properly */
|
|
|
|
ret = av_frame_copy_props(frame, avci->to_free);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
memcpy(frame->data, avci->to_free->data, sizeof(frame->data));
|
|
|
|
memcpy(frame->linesize, avci->to_free->linesize, sizeof(frame->linesize));
|
|
|
|
if (avci->to_free->extended_data != avci->to_free->data) {
|
|
|
|
int planes = av_get_channel_layout_nb_channels(avci->to_free->channel_layout);
|
|
|
|
int size = planes * sizeof(*frame->extended_data);
|
|
|
|
|
|
|
|
if (!size) {
|
|
|
|
av_frame_unref(frame);
|
|
|
|
return AVERROR_BUG;
|
|
|
|
}
|
|
|
|
|
|
|
|
frame->extended_data = av_malloc(size);
|
|
|
|
if (!frame->extended_data) {
|
|
|
|
av_frame_unref(frame);
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
}
|
|
|
|
memcpy(frame->extended_data, avci->to_free->extended_data,
|
|
|
|
size);
|
|
|
|
} else
|
|
|
|
frame->extended_data = frame->data;
|
|
|
|
|
|
|
|
frame->format = avci->to_free->format;
|
|
|
|
frame->width = avci->to_free->width;
|
|
|
|
frame->height = avci->to_free->height;
|
|
|
|
frame->channel_layout = avci->to_free->channel_layout;
|
|
|
|
frame->nb_samples = avci->to_free->nb_samples;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
static int bsfs_init(AVCodecContext *avctx)
|
|
|
|
{
|
|
|
|
AVCodecInternal *avci = avctx->internal;
|
|
|
|
DecodeFilterContext *s = &avci->filter;
|
|
|
|
const char *bsfs_str;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (s->nb_bsfs)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
bsfs_str = avctx->codec->bsfs ? avctx->codec->bsfs : "null";
|
|
|
|
while (bsfs_str && *bsfs_str) {
|
|
|
|
AVBSFContext **tmp;
|
|
|
|
const AVBitStreamFilter *filter;
|
|
|
|
char *bsf;
|
|
|
|
|
|
|
|
bsf = av_get_token(&bsfs_str, ",");
|
|
|
|
if (!bsf) {
|
|
|
|
ret = AVERROR(ENOMEM);
|
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
|
|
|
|
filter = av_bsf_get_by_name(bsf);
|
|
|
|
if (!filter) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "A non-existing bitstream filter %s "
|
|
|
|
"requested by a decoder. This is a bug, please report it.\n",
|
|
|
|
bsf);
|
|
|
|
ret = AVERROR_BUG;
|
|
|
|
av_freep(&bsf);
|
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
av_freep(&bsf);
|
|
|
|
|
|
|
|
tmp = av_realloc_array(s->bsfs, s->nb_bsfs + 1, sizeof(*s->bsfs));
|
|
|
|
if (!tmp) {
|
|
|
|
ret = AVERROR(ENOMEM);
|
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
s->bsfs = tmp;
|
|
|
|
s->nb_bsfs++;
|
|
|
|
|
|
|
|
ret = av_bsf_alloc(filter, &s->bsfs[s->nb_bsfs - 1]);
|
|
|
|
if (ret < 0)
|
|
|
|
goto fail;
|
|
|
|
|
|
|
|
if (s->nb_bsfs == 1) {
|
|
|
|
/* We do not currently have an API for passing the input timebase into decoders,
|
|
|
|
* but no filters used here should actually need it.
|
|
|
|
* So we make up some plausible-looking number (the MPEG 90kHz timebase) */
|
|
|
|
s->bsfs[s->nb_bsfs - 1]->time_base_in = (AVRational){ 1, 90000 };
|
|
|
|
ret = avcodec_parameters_from_context(s->bsfs[s->nb_bsfs - 1]->par_in,
|
|
|
|
avctx);
|
|
|
|
} else {
|
|
|
|
s->bsfs[s->nb_bsfs - 1]->time_base_in = s->bsfs[s->nb_bsfs - 2]->time_base_out;
|
|
|
|
ret = avcodec_parameters_copy(s->bsfs[s->nb_bsfs - 1]->par_in,
|
|
|
|
s->bsfs[s->nb_bsfs - 2]->par_out);
|
|
|
|
}
|
|
|
|
if (ret < 0)
|
|
|
|
goto fail;
|
|
|
|
|
|
|
|
ret = av_bsf_init(s->bsfs[s->nb_bsfs - 1]);
|
|
|
|
if (ret < 0)
|
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
fail:
|
|
|
|
ff_decode_bsfs_uninit(avctx);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* try to get one output packet from the filter chain */
|
|
|
|
static int bsfs_poll(AVCodecContext *avctx, AVPacket *pkt)
|
|
|
|
{
|
|
|
|
DecodeFilterContext *s = &avctx->internal->filter;
|
|
|
|
int idx, ret;
|
|
|
|
|
|
|
|
/* start with the last filter in the chain */
|
|
|
|
idx = s->nb_bsfs - 1;
|
|
|
|
while (idx >= 0) {
|
|
|
|
/* request a packet from the currently selected filter */
|
|
|
|
ret = av_bsf_receive_packet(s->bsfs[idx], pkt);
|
|
|
|
if (ret == AVERROR(EAGAIN)) {
|
|
|
|
/* no packets available, try the next filter up the chain */
|
|
|
|
ret = 0;
|
|
|
|
idx--;
|
|
|
|
continue;
|
|
|
|
} else if (ret < 0 && ret != AVERROR_EOF) {
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* got a packet or EOF -- pass it to the caller or to the next filter
|
|
|
|
* down the chain */
|
|
|
|
if (idx == s->nb_bsfs - 1) {
|
|
|
|
return ret;
|
|
|
|
} else {
|
|
|
|
idx++;
|
|
|
|
ret = av_bsf_send_packet(s->bsfs[idx], ret < 0 ? NULL : pkt);
|
|
|
|
if (ret < 0) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR,
|
|
|
|
"Error pre-processing a packet before decoding\n");
|
|
|
|
av_packet_unref(pkt);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return AVERROR(EAGAIN);
|
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
int ff_decode_get_packet(AVCodecContext *avctx, AVPacket *pkt)
|
2016-10-26 08:10:19 +02:00
|
|
|
{
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
AVCodecInternal *avci = avctx->internal;
|
2016-10-26 08:10:19 +02:00
|
|
|
int ret;
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (avci->draining)
|
|
|
|
return AVERROR_EOF;
|
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
ret = bsfs_poll(avctx, pkt);
|
|
|
|
if (ret == AVERROR_EOF)
|
|
|
|
avci->draining = 1;
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
|
|
|
|
ret = extract_packet_props(avctx->internal, pkt);
|
|
|
|
if (ret < 0)
|
|
|
|
goto finish;
|
|
|
|
|
|
|
|
ret = apply_param_change(avctx, pkt);
|
|
|
|
if (ret < 0)
|
|
|
|
goto finish;
|
|
|
|
|
|
|
|
if (avctx->codec->receive_frame)
|
|
|
|
avci->compat_decode_consumed += pkt->size;
|
2016-10-26 08:10:19 +02:00
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
return 0;
|
|
|
|
finish:
|
|
|
|
av_packet_unref(pkt);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The core of the receive_frame_wrapper for the decoders implementing
|
|
|
|
* the simple API. Certain decoders might consume partial packets without
|
|
|
|
* returning any output, so this function needs to be called in a loop until it
|
|
|
|
* returns EAGAIN.
|
|
|
|
**/
|
|
|
|
static int decode_simple_internal(AVCodecContext *avctx, AVFrame *frame)
|
|
|
|
{
|
|
|
|
AVCodecInternal *avci = avctx->internal;
|
|
|
|
DecodeSimpleContext *ds = &avci->ds;
|
|
|
|
AVPacket *pkt = ds->in_pkt;
|
|
|
|
int got_frame;
|
|
|
|
int ret;
|
2016-10-26 08:10:19 +02:00
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (!pkt->data && !avci->draining) {
|
|
|
|
av_packet_unref(pkt);
|
|
|
|
ret = ff_decode_get_packet(avctx, pkt);
|
|
|
|
if (ret < 0 && ret != AVERROR_EOF)
|
|
|
|
return ret;
|
|
|
|
}
|
2016-10-26 08:10:19 +02:00
|
|
|
|
|
|
|
// Some codecs (at least wma lossless) will crash when feeding drain packets
|
|
|
|
// after EOF was signaled.
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (avci->draining_done)
|
|
|
|
return AVERROR_EOF;
|
|
|
|
|
|
|
|
if (!pkt->data &&
|
|
|
|
!(avctx->codec->capabilities & AV_CODEC_CAP_DELAY ||
|
|
|
|
avctx->active_thread_type & FF_THREAD_FRAME))
|
2016-10-26 08:10:19 +02:00
|
|
|
return AVERROR_EOF;
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
got_frame = 0;
|
|
|
|
|
|
|
|
if (HAVE_THREADS && avctx->active_thread_type & FF_THREAD_FRAME) {
|
|
|
|
ret = ff_thread_decode_frame(avctx, frame, &got_frame, pkt);
|
2016-10-26 08:10:19 +02:00
|
|
|
} else {
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
ret = avctx->codec->decode(avctx, frame, &got_frame, pkt);
|
|
|
|
|
|
|
|
if (!(avctx->codec->caps_internal & FF_CODEC_CAP_SETS_PKT_DTS))
|
|
|
|
frame->pkt_dts = pkt->dts;
|
|
|
|
/* get_buffer is supposed to set frame parameters */
|
|
|
|
if (!(avctx->codec->capabilities & AV_CODEC_CAP_DR1)) {
|
|
|
|
frame->sample_aspect_ratio = avctx->sample_aspect_ratio;
|
|
|
|
frame->width = avctx->width;
|
|
|
|
frame->height = avctx->height;
|
|
|
|
frame->format = avctx->codec->type == AVMEDIA_TYPE_VIDEO ?
|
|
|
|
avctx->pix_fmt : avctx->sample_fmt;
|
|
|
|
}
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
emms_c();
|
|
|
|
|
|
|
|
if (!got_frame)
|
|
|
|
av_frame_unref(frame);
|
|
|
|
|
|
|
|
if (ret >= 0 && avctx->codec->type == AVMEDIA_TYPE_VIDEO)
|
|
|
|
ret = pkt->size;
|
|
|
|
|
|
|
|
#if FF_API_AVCTX_TIMEBASE
|
|
|
|
if (avctx->framerate.num > 0 && avctx->framerate.den > 0)
|
|
|
|
avctx->time_base = av_inv_q(avctx->framerate);
|
|
|
|
#endif
|
2016-10-26 08:10:19 +02:00
|
|
|
|
|
|
|
if (avctx->internal->draining && !got_frame)
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
avci->draining_done = 1;
|
|
|
|
|
|
|
|
avci->compat_decode_consumed += ret;
|
2016-10-26 08:10:19 +02:00
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (ret >= pkt->size || ret < 0) {
|
|
|
|
av_packet_unref(pkt);
|
2016-10-26 08:10:19 +02:00
|
|
|
} else {
|
|
|
|
int consumed = ret;
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
pkt->data += consumed;
|
|
|
|
pkt->size -= consumed;
|
|
|
|
pkt->pts = AV_NOPTS_VALUE;
|
|
|
|
pkt->dts = AV_NOPTS_VALUE;
|
|
|
|
avci->last_pkt_props->pts = AV_NOPTS_VALUE;
|
|
|
|
avci->last_pkt_props->dts = AV_NOPTS_VALUE;
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (got_frame)
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
av_assert0(frame->buf[0]);
|
|
|
|
|
|
|
|
return ret < 0 ? ret : 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int decode_simple_receive_frame(AVCodecContext *avctx, AVFrame *frame)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
while (!frame->buf[0]) {
|
|
|
|
ret = decode_simple_internal(avctx, frame);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
|
|
|
}
|
2016-10-26 08:10:19 +02:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
static int decode_receive_frame_internal(AVCodecContext *avctx, AVFrame *frame)
|
2016-10-26 08:10:19 +02:00
|
|
|
{
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
AVCodecInternal *avci = avctx->internal;
|
2016-10-26 08:10:19 +02:00
|
|
|
int ret;
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
av_assert0(!frame->buf[0]);
|
|
|
|
|
|
|
|
if (avctx->codec->receive_frame)
|
|
|
|
ret = avctx->codec->receive_frame(avctx, frame);
|
|
|
|
else
|
|
|
|
ret = decode_simple_receive_frame(avctx, frame);
|
|
|
|
|
|
|
|
if (ret == AVERROR_EOF)
|
|
|
|
avci->draining_done = 1;
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int attribute_align_arg avcodec_send_packet(AVCodecContext *avctx, const AVPacket *avpkt)
|
|
|
|
{
|
|
|
|
AVCodecInternal *avci = avctx->internal;
|
|
|
|
int ret = 0;
|
|
|
|
|
2016-10-26 08:10:19 +02:00
|
|
|
if (!avcodec_is_open(avctx) || !av_codec_is_decoder(avctx->codec))
|
|
|
|
return AVERROR(EINVAL);
|
|
|
|
|
|
|
|
if (avctx->internal->draining)
|
|
|
|
return AVERROR_EOF;
|
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
ret = bsfs_init(avctx);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
av_packet_unref(avci->buffer_pkt);
|
|
|
|
if (avpkt && (avpkt->data || avpkt->side_data_elems)) {
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
ret = av_packet_ref(avci->buffer_pkt, avpkt);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
ret = av_bsf_send_packet(avci->filter.bsfs[0], avci->buffer_pkt);
|
|
|
|
if (ret < 0) {
|
|
|
|
av_packet_unref(avci->buffer_pkt);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (!avci->buffer_frame->buf[0]) {
|
|
|
|
ret = decode_receive_frame_internal(avctx, avci->buffer_frame);
|
|
|
|
if (ret < 0 && ret != AVERROR(EAGAIN) && ret != AVERROR_EOF)
|
|
|
|
return ret;
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
return 0;
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
int attribute_align_arg avcodec_receive_frame(AVCodecContext *avctx, AVFrame *frame)
|
|
|
|
{
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
AVCodecInternal *avci = avctx->internal;
|
2016-10-26 08:10:19 +02:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
av_frame_unref(frame);
|
|
|
|
|
|
|
|
if (!avcodec_is_open(avctx) || !av_codec_is_decoder(avctx->codec))
|
|
|
|
return AVERROR(EINVAL);
|
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
ret = bsfs_init(avctx);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (avci->buffer_frame->buf[0]) {
|
|
|
|
av_frame_move_ref(frame, avci->buffer_frame);
|
|
|
|
} else {
|
|
|
|
ret = decode_receive_frame_internal(avctx, frame);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
avctx->frame_number++;
|
2016-10-26 08:10:19 +02:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
static int compat_decode(AVCodecContext *avctx, AVFrame *frame,
|
|
|
|
int *got_frame, AVPacket *pkt)
|
2016-10-26 08:10:19 +02:00
|
|
|
{
|
|
|
|
AVCodecInternal *avci = avctx->internal;
|
|
|
|
int ret;
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
av_assert0(avci->compat_decode_consumed == 0);
|
2016-10-26 08:10:19 +02:00
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
*got_frame = 0;
|
|
|
|
avci->compat_decode = 1;
|
2016-10-26 08:10:19 +02:00
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (avci->compat_decode_partial_size > 0 &&
|
|
|
|
avci->compat_decode_partial_size != pkt->size) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR,
|
|
|
|
"Got unexpected packet size after a partial decode\n");
|
|
|
|
ret = AVERROR(EINVAL);
|
|
|
|
goto finish;
|
|
|
|
}
|
2016-10-26 13:41:12 +02:00
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (!avci->compat_decode_partial_size) {
|
|
|
|
ret = avcodec_send_packet(avctx, pkt);
|
|
|
|
if (ret == AVERROR_EOF)
|
|
|
|
ret = 0;
|
|
|
|
else if (ret == AVERROR(EAGAIN)) {
|
|
|
|
/* we fully drain all the output in each decode call, so this should not
|
|
|
|
* ever happen */
|
|
|
|
ret = AVERROR_BUG;
|
|
|
|
goto finish;
|
|
|
|
} else if (ret < 0)
|
|
|
|
goto finish;
|
|
|
|
}
|
2016-10-26 08:10:19 +02:00
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
while (ret >= 0) {
|
|
|
|
ret = avcodec_receive_frame(avctx, frame);
|
|
|
|
if (ret < 0) {
|
|
|
|
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
|
|
|
|
ret = 0;
|
|
|
|
goto finish;
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
if (frame != avci->compat_decode_frame) {
|
2016-10-26 08:10:19 +02:00
|
|
|
if (!avctx->refcounted_frames) {
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
ret = unrefcount_frame(avci, frame);
|
|
|
|
if (ret < 0)
|
|
|
|
goto finish;
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
*got_frame = 1;
|
|
|
|
frame = avci->compat_decode_frame;
|
|
|
|
} else {
|
|
|
|
if (!avci->compat_decode_warned) {
|
|
|
|
av_log(avctx, AV_LOG_WARNING, "The deprecated avcodec_decode_* "
|
|
|
|
"API cannot return all the frames for this decoder. "
|
|
|
|
"Some frames will be dropped. Update your code to the "
|
|
|
|
"new decoding API to fix this.\n");
|
|
|
|
avci->compat_decode_warned = 1;
|
|
|
|
}
|
|
|
|
}
|
2016-10-26 08:10:19 +02:00
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
if (avci->draining || (!avctx->codec->bsfs && avci->compat_decode_consumed < pkt->size))
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
finish:
|
2016-11-12 18:43:55 +02:00
|
|
|
if (ret == 0) {
|
|
|
|
/* if there are any bsfs then assume full packet is always consumed */
|
|
|
|
if (avctx->codec->bsfs)
|
|
|
|
ret = pkt->size;
|
|
|
|
else
|
|
|
|
ret = FFMIN(avci->compat_decode_consumed, pkt->size);
|
|
|
|
}
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
avci->compat_decode_consumed = 0;
|
|
|
|
avci->compat_decode_partial_size = (ret >= 0) ? pkt->size - ret : 0;
|
2016-10-26 08:10:19 +02:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
int attribute_align_arg avcodec_decode_video2(AVCodecContext *avctx, AVFrame *picture,
|
|
|
|
int *got_picture_ptr,
|
|
|
|
AVPacket *avpkt)
|
|
|
|
{
|
|
|
|
return compat_decode(avctx, picture, got_picture_ptr, avpkt);
|
|
|
|
}
|
|
|
|
|
2016-10-26 08:10:19 +02:00
|
|
|
int attribute_align_arg avcodec_decode_audio4(AVCodecContext *avctx,
|
|
|
|
AVFrame *frame,
|
|
|
|
int *got_frame_ptr,
|
|
|
|
AVPacket *avpkt)
|
|
|
|
{
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
return compat_decode(avctx, frame, got_frame_ptr, avpkt);
|
2016-10-26 08:10:19 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
int avcodec_decode_subtitle2(AVCodecContext *avctx, AVSubtitle *sub,
|
|
|
|
int *got_sub_ptr,
|
|
|
|
AVPacket *avpkt)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
2016-10-26 13:41:12 +02:00
|
|
|
ret = extract_packet_props(avctx->internal, avpkt);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
|
|
|
|
2016-10-26 08:10:19 +02:00
|
|
|
*got_sub_ptr = 0;
|
|
|
|
ret = avctx->codec->decode(avctx, sub, got_sub_ptr, avpkt);
|
|
|
|
if (*got_sub_ptr)
|
|
|
|
avctx->frame_number++;
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int is_hwaccel_pix_fmt(enum AVPixelFormat pix_fmt)
|
|
|
|
{
|
|
|
|
const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
|
|
|
|
return desc->flags & AV_PIX_FMT_FLAG_HWACCEL;
|
|
|
|
}
|
|
|
|
|
|
|
|
enum AVPixelFormat avcodec_default_get_format(struct AVCodecContext *s, const enum AVPixelFormat *fmt)
|
|
|
|
{
|
|
|
|
while (*fmt != AV_PIX_FMT_NONE && is_hwaccel_pix_fmt(*fmt))
|
|
|
|
++fmt;
|
|
|
|
return fmt[0];
|
|
|
|
}
|
|
|
|
|
|
|
|
static AVHWAccel *find_hwaccel(enum AVCodecID codec_id,
|
|
|
|
enum AVPixelFormat pix_fmt)
|
|
|
|
{
|
|
|
|
AVHWAccel *hwaccel = NULL;
|
|
|
|
|
|
|
|
while ((hwaccel = av_hwaccel_next(hwaccel)))
|
|
|
|
if (hwaccel->id == codec_id
|
|
|
|
&& hwaccel->pix_fmt == pix_fmt)
|
|
|
|
return hwaccel;
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int setup_hwaccel(AVCodecContext *avctx,
|
|
|
|
const enum AVPixelFormat fmt,
|
|
|
|
const char *name)
|
|
|
|
{
|
|
|
|
AVHWAccel *hwa = find_hwaccel(avctx->codec_id, fmt);
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
if (!hwa) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR,
|
|
|
|
"Could not find an AVHWAccel for the pixel format: %s",
|
|
|
|
name);
|
|
|
|
return AVERROR(ENOENT);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (hwa->priv_data_size) {
|
|
|
|
avctx->internal->hwaccel_priv_data = av_mallocz(hwa->priv_data_size);
|
|
|
|
if (!avctx->internal->hwaccel_priv_data)
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (hwa->init) {
|
|
|
|
ret = hwa->init(avctx);
|
|
|
|
if (ret < 0) {
|
|
|
|
av_freep(&avctx->internal->hwaccel_priv_data);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
avctx->hwaccel = hwa;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int ff_get_format(AVCodecContext *avctx, const enum AVPixelFormat *fmt)
|
|
|
|
{
|
|
|
|
const AVPixFmtDescriptor *desc;
|
|
|
|
enum AVPixelFormat *choices;
|
|
|
|
enum AVPixelFormat ret;
|
|
|
|
unsigned n = 0;
|
|
|
|
|
|
|
|
while (fmt[n] != AV_PIX_FMT_NONE)
|
|
|
|
++n;
|
|
|
|
|
|
|
|
av_assert0(n >= 1);
|
|
|
|
avctx->sw_pix_fmt = fmt[n - 1];
|
|
|
|
av_assert2(!is_hwaccel_pix_fmt(avctx->sw_pix_fmt));
|
|
|
|
|
|
|
|
choices = av_malloc_array(n + 1, sizeof(*choices));
|
|
|
|
if (!choices)
|
|
|
|
return AV_PIX_FMT_NONE;
|
|
|
|
|
|
|
|
memcpy(choices, fmt, (n + 1) * sizeof(*choices));
|
|
|
|
|
|
|
|
for (;;) {
|
|
|
|
if (avctx->hwaccel && avctx->hwaccel->uninit)
|
|
|
|
avctx->hwaccel->uninit(avctx);
|
|
|
|
av_freep(&avctx->internal->hwaccel_priv_data);
|
|
|
|
avctx->hwaccel = NULL;
|
|
|
|
|
|
|
|
av_buffer_unref(&avctx->hw_frames_ctx);
|
|
|
|
|
|
|
|
ret = avctx->get_format(avctx, choices);
|
|
|
|
|
|
|
|
desc = av_pix_fmt_desc_get(ret);
|
|
|
|
if (!desc) {
|
|
|
|
ret = AV_PIX_FMT_NONE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!(desc->flags & AV_PIX_FMT_FLAG_HWACCEL))
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (avctx->hw_frames_ctx) {
|
|
|
|
AVHWFramesContext *hw_frames_ctx = (AVHWFramesContext*)avctx->hw_frames_ctx->data;
|
|
|
|
if (hw_frames_ctx->format != ret) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "Format returned from get_buffer() "
|
|
|
|
"does not match the format of provided AVHWFramesContext\n");
|
|
|
|
ret = AV_PIX_FMT_NONE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!setup_hwaccel(avctx, ret, desc->name))
|
|
|
|
break;
|
|
|
|
|
|
|
|
/* Remove failed hwaccel from choices */
|
|
|
|
for (n = 0; choices[n] != ret; n++)
|
|
|
|
av_assert0(choices[n] != AV_PIX_FMT_NONE);
|
|
|
|
|
|
|
|
do
|
|
|
|
choices[n] = choices[n + 1];
|
|
|
|
while (choices[n++] != AV_PIX_FMT_NONE);
|
|
|
|
}
|
|
|
|
|
|
|
|
av_freep(&choices);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int update_frame_pool(AVCodecContext *avctx, AVFrame *frame)
|
|
|
|
{
|
|
|
|
FramePool *pool = avctx->internal->pool;
|
|
|
|
int i, ret;
|
|
|
|
|
|
|
|
switch (avctx->codec_type) {
|
|
|
|
case AVMEDIA_TYPE_VIDEO: {
|
|
|
|
uint8_t *data[4];
|
|
|
|
int linesize[4];
|
|
|
|
int size[4] = { 0 };
|
|
|
|
int w = frame->width;
|
|
|
|
int h = frame->height;
|
|
|
|
int tmpsize, unaligned;
|
|
|
|
|
|
|
|
if (pool->format == frame->format &&
|
|
|
|
pool->width == frame->width && pool->height == frame->height)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
avcodec_align_dimensions2(avctx, &w, &h, pool->stride_align);
|
|
|
|
|
|
|
|
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(linesize, avctx->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 |= linesize[i] % pool->stride_align[i];
|
|
|
|
} while (unaligned);
|
|
|
|
|
|
|
|
tmpsize = av_image_fill_pointers(data, avctx->pix_fmt, h,
|
|
|
|
NULL, linesize);
|
|
|
|
if (tmpsize < 0)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
for (i = 0; i < 3 && data[i + 1]; i++)
|
|
|
|
size[i] = data[i + 1] - data[i];
|
|
|
|
size[i] = tmpsize - (data[i] - data[0]);
|
|
|
|
|
|
|
|
for (i = 0; i < 4; i++) {
|
|
|
|
av_buffer_pool_uninit(&pool->pools[i]);
|
|
|
|
pool->linesize[i] = linesize[i];
|
|
|
|
if (size[i]) {
|
|
|
|
pool->pools[i] = av_buffer_pool_init(size[i] + 16, NULL);
|
|
|
|
if (!pool->pools[i]) {
|
|
|
|
ret = AVERROR(ENOMEM);
|
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
pool->format = frame->format;
|
|
|
|
pool->width = frame->width;
|
|
|
|
pool->height = frame->height;
|
|
|
|
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
case AVMEDIA_TYPE_AUDIO: {
|
|
|
|
int ch = av_get_channel_layout_nb_channels(frame->channel_layout);
|
|
|
|
int planar = av_sample_fmt_is_planar(frame->format);
|
|
|
|
int planes = planar ? ch : 1;
|
|
|
|
|
|
|
|
if (pool->format == frame->format && pool->planes == planes &&
|
|
|
|
pool->channels == ch && frame->nb_samples == pool->samples)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
av_buffer_pool_uninit(&pool->pools[0]);
|
|
|
|
ret = av_samples_get_buffer_size(&pool->linesize[0], ch,
|
|
|
|
frame->nb_samples, frame->format, 0);
|
|
|
|
if (ret < 0)
|
|
|
|
goto fail;
|
|
|
|
|
|
|
|
pool->pools[0] = av_buffer_pool_init(pool->linesize[0], NULL);
|
|
|
|
if (!pool->pools[0]) {
|
|
|
|
ret = AVERROR(ENOMEM);
|
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
|
|
|
|
pool->format = frame->format;
|
|
|
|
pool->planes = planes;
|
|
|
|
pool->channels = ch;
|
|
|
|
pool->samples = frame->nb_samples;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
default: av_assert0(0);
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
fail:
|
|
|
|
for (i = 0; i < 4; i++)
|
|
|
|
av_buffer_pool_uninit(&pool->pools[i]);
|
|
|
|
pool->format = -1;
|
|
|
|
pool->planes = pool->channels = pool->samples = 0;
|
|
|
|
pool->width = pool->height = 0;
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int audio_get_buffer(AVCodecContext *avctx, AVFrame *frame)
|
|
|
|
{
|
|
|
|
FramePool *pool = avctx->internal->pool;
|
|
|
|
int planes = pool->planes;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
frame->linesize[0] = pool->linesize[0];
|
|
|
|
|
|
|
|
if (planes > AV_NUM_DATA_POINTERS) {
|
|
|
|
frame->extended_data = av_mallocz(planes * sizeof(*frame->extended_data));
|
|
|
|
frame->nb_extended_buf = planes - AV_NUM_DATA_POINTERS;
|
|
|
|
frame->extended_buf = av_mallocz(frame->nb_extended_buf *
|
|
|
|
sizeof(*frame->extended_buf));
|
|
|
|
if (!frame->extended_data || !frame->extended_buf) {
|
|
|
|
av_freep(&frame->extended_data);
|
|
|
|
av_freep(&frame->extended_buf);
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
}
|
|
|
|
} else
|
|
|
|
frame->extended_data = frame->data;
|
|
|
|
|
|
|
|
for (i = 0; i < FFMIN(planes, AV_NUM_DATA_POINTERS); i++) {
|
|
|
|
frame->buf[i] = av_buffer_pool_get(pool->pools[0]);
|
|
|
|
if (!frame->buf[i])
|
|
|
|
goto fail;
|
|
|
|
frame->extended_data[i] = frame->data[i] = frame->buf[i]->data;
|
|
|
|
}
|
|
|
|
for (i = 0; i < frame->nb_extended_buf; i++) {
|
|
|
|
frame->extended_buf[i] = av_buffer_pool_get(pool->pools[0]);
|
|
|
|
if (!frame->extended_buf[i])
|
|
|
|
goto fail;
|
|
|
|
frame->extended_data[i + AV_NUM_DATA_POINTERS] = frame->extended_buf[i]->data;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (avctx->debug & FF_DEBUG_BUFFERS)
|
|
|
|
av_log(avctx, AV_LOG_DEBUG, "default_get_buffer called on frame %p", frame);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
fail:
|
|
|
|
av_frame_unref(frame);
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int video_get_buffer(AVCodecContext *s, AVFrame *pic)
|
|
|
|
{
|
|
|
|
FramePool *pool = s->internal->pool;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
if (pic->data[0]) {
|
|
|
|
av_log(s, AV_LOG_ERROR, "pic->data[0]!=NULL in avcodec_default_get_buffer\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
memset(pic->data, 0, sizeof(pic->data));
|
|
|
|
pic->extended_data = pic->data;
|
|
|
|
|
|
|
|
for (i = 0; i < 4 && pool->pools[i]; i++) {
|
|
|
|
pic->linesize[i] = pool->linesize[i];
|
|
|
|
|
|
|
|
pic->buf[i] = av_buffer_pool_get(pool->pools[i]);
|
|
|
|
if (!pic->buf[i])
|
|
|
|
goto fail;
|
|
|
|
|
|
|
|
pic->data[i] = pic->buf[i]->data;
|
|
|
|
}
|
|
|
|
for (; i < AV_NUM_DATA_POINTERS; i++) {
|
|
|
|
pic->data[i] = NULL;
|
|
|
|
pic->linesize[i] = 0;
|
|
|
|
}
|
|
|
|
if (pic->data[1] && !pic->data[2])
|
|
|
|
avpriv_set_systematic_pal2((uint32_t *)pic->data[1], s->pix_fmt);
|
|
|
|
|
|
|
|
if (s->debug & FF_DEBUG_BUFFERS)
|
|
|
|
av_log(s, AV_LOG_DEBUG, "default_get_buffer called on pic %p\n", pic);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
fail:
|
|
|
|
av_frame_unref(pic);
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
}
|
|
|
|
|
|
|
|
int avcodec_default_get_buffer2(AVCodecContext *avctx, AVFrame *frame, int flags)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (avctx->hw_frames_ctx)
|
|
|
|
return av_hwframe_get_buffer(avctx->hw_frames_ctx, frame, 0);
|
|
|
|
|
|
|
|
if ((ret = update_frame_pool(avctx, frame)) < 0)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
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;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
int ff_decode_frame_props(AVCodecContext *avctx, AVFrame *frame)
|
|
|
|
{
|
2016-10-26 13:41:12 +02:00
|
|
|
AVPacket *pkt = avctx->internal->last_pkt_props;
|
2016-10-26 08:10:19 +02:00
|
|
|
int i;
|
|
|
|
struct {
|
|
|
|
enum AVPacketSideDataType packet;
|
|
|
|
enum AVFrameSideDataType frame;
|
|
|
|
} sd[] = {
|
|
|
|
{ AV_PKT_DATA_REPLAYGAIN , AV_FRAME_DATA_REPLAYGAIN },
|
|
|
|
{ AV_PKT_DATA_DISPLAYMATRIX, AV_FRAME_DATA_DISPLAYMATRIX },
|
2016-11-02 17:28:54 +02:00
|
|
|
{ AV_PKT_DATA_SPHERICAL, AV_FRAME_DATA_SPHERICAL },
|
2016-10-26 08:10:19 +02:00
|
|
|
{ AV_PKT_DATA_STEREO3D, AV_FRAME_DATA_STEREO3D },
|
|
|
|
{ AV_PKT_DATA_AUDIO_SERVICE_TYPE, AV_FRAME_DATA_AUDIO_SERVICE_TYPE },
|
|
|
|
};
|
|
|
|
|
|
|
|
frame->color_primaries = avctx->color_primaries;
|
|
|
|
frame->color_trc = avctx->color_trc;
|
|
|
|
frame->colorspace = avctx->colorspace;
|
|
|
|
frame->color_range = avctx->color_range;
|
|
|
|
frame->chroma_location = avctx->chroma_sample_location;
|
|
|
|
|
|
|
|
frame->reordered_opaque = avctx->reordered_opaque;
|
|
|
|
|
|
|
|
#if FF_API_PKT_PTS
|
|
|
|
FF_DISABLE_DEPRECATION_WARNINGS
|
|
|
|
frame->pkt_pts = pkt->pts;
|
|
|
|
FF_ENABLE_DEPRECATION_WARNINGS
|
|
|
|
#endif
|
|
|
|
frame->pts = pkt->pts;
|
|
|
|
|
|
|
|
for (i = 0; i < FF_ARRAY_ELEMS(sd); i++) {
|
|
|
|
int size;
|
|
|
|
uint8_t *packet_sd = av_packet_get_side_data(pkt, sd[i].packet, &size);
|
|
|
|
if (packet_sd) {
|
|
|
|
AVFrameSideData *frame_sd = av_frame_new_side_data(frame,
|
|
|
|
sd[i].frame,
|
|
|
|
size);
|
|
|
|
if (!frame_sd)
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
|
|
|
|
memcpy(frame_sd->data, packet_sd, size);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int ff_get_buffer(AVCodecContext *avctx, AVFrame *frame, int flags)
|
|
|
|
{
|
|
|
|
const AVHWAccel *hwaccel = avctx->hwaccel;
|
|
|
|
int override_dimensions = 1;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
switch (avctx->codec_type) {
|
|
|
|
case AVMEDIA_TYPE_VIDEO:
|
|
|
|
if (frame->width <= 0 || frame->height <= 0) {
|
|
|
|
frame->width = FFMAX(avctx->width, avctx->coded_width);
|
|
|
|
frame->height = FFMAX(avctx->height, avctx->coded_height);
|
|
|
|
override_dimensions = 0;
|
|
|
|
}
|
|
|
|
if (frame->format < 0)
|
|
|
|
frame->format = avctx->pix_fmt;
|
|
|
|
if (!frame->sample_aspect_ratio.num)
|
|
|
|
frame->sample_aspect_ratio = avctx->sample_aspect_ratio;
|
|
|
|
|
|
|
|
if (av_image_check_sar(frame->width, frame->height,
|
|
|
|
frame->sample_aspect_ratio) < 0) {
|
|
|
|
av_log(avctx, AV_LOG_WARNING, "ignoring invalid SAR: %u/%u\n",
|
|
|
|
frame->sample_aspect_ratio.num,
|
|
|
|
frame->sample_aspect_ratio.den);
|
|
|
|
frame->sample_aspect_ratio = (AVRational){ 0, 1 };
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((ret = av_image_check_size(avctx->width, avctx->height, 0, avctx)) < 0)
|
|
|
|
return ret;
|
|
|
|
break;
|
|
|
|
case AVMEDIA_TYPE_AUDIO:
|
|
|
|
if (!frame->sample_rate)
|
|
|
|
frame->sample_rate = avctx->sample_rate;
|
|
|
|
if (frame->format < 0)
|
|
|
|
frame->format = avctx->sample_fmt;
|
|
|
|
if (!frame->channel_layout) {
|
|
|
|
if (avctx->channel_layout) {
|
|
|
|
if (av_get_channel_layout_nb_channels(avctx->channel_layout) !=
|
|
|
|
avctx->channels) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "Inconsistent channel "
|
|
|
|
"configuration.\n");
|
|
|
|
return AVERROR(EINVAL);
|
|
|
|
}
|
|
|
|
|
|
|
|
frame->channel_layout = avctx->channel_layout;
|
|
|
|
} else {
|
|
|
|
if (avctx->channels > FF_SANE_NB_CHANNELS) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "Too many channels: %d.\n",
|
|
|
|
avctx->channels);
|
|
|
|
return AVERROR(ENOSYS);
|
|
|
|
}
|
|
|
|
|
|
|
|
frame->channel_layout = av_get_default_channel_layout(avctx->channels);
|
|
|
|
if (!frame->channel_layout)
|
|
|
|
frame->channel_layout = (1ULL << avctx->channels) - 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
default: return AVERROR(EINVAL);
|
|
|
|
}
|
|
|
|
|
|
|
|
ret = ff_decode_frame_props(avctx, frame);
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
if (hwaccel) {
|
|
|
|
if (hwaccel->alloc_frame) {
|
|
|
|
ret = hwaccel->alloc_frame(avctx, frame);
|
|
|
|
goto end;
|
|
|
|
}
|
|
|
|
} else
|
|
|
|
avctx->sw_pix_fmt = avctx->pix_fmt;
|
|
|
|
|
|
|
|
ret = avctx->get_buffer2(avctx, frame, flags);
|
|
|
|
|
|
|
|
end:
|
|
|
|
if (avctx->codec_type == AVMEDIA_TYPE_VIDEO && !override_dimensions) {
|
|
|
|
frame->width = avctx->width;
|
|
|
|
frame->height = avctx->height;
|
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int ff_reget_buffer(AVCodecContext *avctx, AVFrame *frame)
|
|
|
|
{
|
|
|
|
AVFrame *tmp;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
av_assert0(avctx->codec_type == AVMEDIA_TYPE_VIDEO);
|
|
|
|
|
|
|
|
if (!frame->data[0])
|
|
|
|
return ff_get_buffer(avctx, frame, AV_GET_BUFFER_FLAG_REF);
|
|
|
|
|
|
|
|
if (av_frame_is_writable(frame))
|
|
|
|
return ff_decode_frame_props(avctx, frame);
|
|
|
|
|
|
|
|
tmp = av_frame_alloc();
|
|
|
|
if (!tmp)
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
|
|
|
|
av_frame_move_ref(tmp, frame);
|
|
|
|
|
|
|
|
ret = ff_get_buffer(avctx, frame, AV_GET_BUFFER_FLAG_REF);
|
|
|
|
if (ret < 0) {
|
|
|
|
av_frame_free(&tmp);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
av_frame_copy(frame, tmp);
|
|
|
|
av_frame_free(&tmp);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void avcodec_flush_buffers(AVCodecContext *avctx)
|
|
|
|
{
|
|
|
|
avctx->internal->draining = 0;
|
|
|
|
avctx->internal->draining_done = 0;
|
|
|
|
av_frame_unref(avctx->internal->buffer_frame);
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
av_frame_unref(avctx->internal->compat_decode_frame);
|
2016-10-26 08:10:19 +02:00
|
|
|
av_packet_unref(avctx->internal->buffer_pkt);
|
|
|
|
avctx->internal->buffer_pkt_valid = 0;
|
|
|
|
|
decode: restructure the core decoding code
Currently, the new decoding API is pretty much just a wrapper around the
old deprecated one. This is problematic, since it interferes with making
full use of the flexibility added by the new API. The old API should
also be removed at some future point.
Reorganize the code so that the new send_packet/receive_frame functions
call the actual decoding directly and change the old deprecated
avcodec_decode_* functions into wrappers around the new API.
The new internal API for decoders is now changing as well. Before this
commit, it mirrors the public API, so the decoders need to implement
send_packet() and receive_frame() callbacks. This turns out to require
awkward constructs in both the decoders and the generic code. After this
commit, the decoders only implement the receive_frame() callback and
call a new internal function, ff_decode_get_packet() to obtain input
data, in the same manner to how the bitstream filters now work.
avcodec will now always make a reference to the input packet, which means
that non-refcounted input packets will be copied. Keeping the previous
behaviour, where this copy could sometimes be avoided, would make the
code significantly more complex and fragile for only dubious gains,
since packets are typically small and everyone who cares about
performance should use refcounted packets anyway.
2016-10-26 13:59:15 +02:00
|
|
|
av_packet_unref(avctx->internal->ds.in_pkt);
|
|
|
|
|
2016-10-26 08:10:19 +02:00
|
|
|
if (HAVE_THREADS && avctx->active_thread_type & FF_THREAD_FRAME)
|
|
|
|
ff_thread_flush(avctx);
|
|
|
|
else if (avctx->codec->flush)
|
|
|
|
avctx->codec->flush(avctx);
|
|
|
|
|
2016-11-12 18:43:55 +02:00
|
|
|
ff_decode_bsfs_uninit(avctx);
|
|
|
|
|
2016-10-26 08:10:19 +02:00
|
|
|
if (!avctx->refcounted_frames)
|
|
|
|
av_frame_unref(avctx->internal->to_free);
|
|
|
|
}
|
2016-11-12 18:43:55 +02:00
|
|
|
|
|
|
|
void ff_decode_bsfs_uninit(AVCodecContext *avctx)
|
|
|
|
{
|
|
|
|
DecodeFilterContext *s = &avctx->internal->filter;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < s->nb_bsfs; i++)
|
|
|
|
av_bsf_free(&s->bsfs[i]);
|
|
|
|
av_freep(&s->bsfs);
|
|
|
|
s->nb_bsfs = 0;
|
|
|
|
}
|