ffmpeg/libavcodec/encode.c

910 lines
29 KiB
C
Raw Normal View History

/*
* generic encoding-related code
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "libavutil/attributes.h"
#include "libavutil/avassert.h"
#include "libavutil/channel_layout.h"
#include "libavutil/emms.h"
#include "libavutil/frame.h"
#include "libavutil/imgutils.h"
#include "libavutil/internal.h"
#include "libavutil/mem.h"
#include "libavutil/pixdesc.h"
#include "libavutil/samplefmt.h"
#include "avcodec.h"
#include "avcodec_internal.h"
#include "codec_desc.h"
#include "codec_internal.h"
#include "encode.h"
#include "frame_thread_encoder.h"
#include "internal.h"
typedef struct EncodeContext {
AVCodecInternal avci;
/**
* This is set to AV_PKT_FLAG_KEY for encoders that encode intra-only
* formats (i.e. whose codec descriptor has AV_CODEC_PROP_INTRA_ONLY set).
* This is used to set said flag generically for said encoders.
*/
int intra_only_flag;
/**
* An audio frame with less than required samples has been submitted (and
* potentially padded with silence). Reject all subsequent frames.
*/
int last_audio_frame;
} EncodeContext;
static EncodeContext *encode_ctx(AVCodecInternal *avci)
{
return (EncodeContext*)avci;
}
int ff_alloc_packet(AVCodecContext *avctx, AVPacket *avpkt, int64_t size)
{
if (size < 0 || size > INT_MAX - AV_INPUT_BUFFER_PADDING_SIZE) {
av_log(avctx, AV_LOG_ERROR, "Invalid minimum required packet size %"PRId64" (max allowed is %d)\n",
size, INT_MAX - AV_INPUT_BUFFER_PADDING_SIZE);
return AVERROR(EINVAL);
}
av_assert0(!avpkt->data);
av_fast_padded_malloc(&avctx->internal->byte_buffer,
&avctx->internal->byte_buffer_size, size);
avpkt->data = avctx->internal->byte_buffer;
if (!avpkt->data) {
av_log(avctx, AV_LOG_ERROR, "Failed to allocate packet of size %"PRId64"\n", size);
return AVERROR(ENOMEM);
}
avpkt->size = size;
return 0;
}
int avcodec_default_get_encode_buffer(AVCodecContext *avctx, AVPacket *avpkt, int flags)
{
int ret;
if (avpkt->size < 0 || avpkt->size > INT_MAX - AV_INPUT_BUFFER_PADDING_SIZE)
return AVERROR(EINVAL);
if (avpkt->data || avpkt->buf) {
av_log(avctx, AV_LOG_ERROR, "avpkt->{data,buf} != NULL in avcodec_default_get_encode_buffer()\n");
return AVERROR(EINVAL);
}
ret = av_buffer_realloc(&avpkt->buf, avpkt->size + AV_INPUT_BUFFER_PADDING_SIZE);
if (ret < 0) {
av_log(avctx, AV_LOG_ERROR, "Failed to allocate packet of size %d\n", avpkt->size);
return ret;
}
avpkt->data = avpkt->buf->data;
return 0;
}
int ff_get_encode_buffer(AVCodecContext *avctx, AVPacket *avpkt, int64_t size, int flags)
{
int ret;
if (size < 0 || size > INT_MAX - AV_INPUT_BUFFER_PADDING_SIZE)
return AVERROR(EINVAL);
av_assert0(!avpkt->data && !avpkt->buf);
avpkt->size = size;
ret = avctx->get_encode_buffer(avctx, avpkt, flags);
if (ret < 0)
goto fail;
if (!avpkt->data || !avpkt->buf) {
av_log(avctx, AV_LOG_ERROR, "No buffer returned by get_encode_buffer()\n");
ret = AVERROR(EINVAL);
goto fail;
}
memset(avpkt->data + avpkt->size, 0, AV_INPUT_BUFFER_PADDING_SIZE);
ret = 0;
fail:
if (ret < 0) {
av_log(avctx, AV_LOG_ERROR, "get_encode_buffer() failed\n");
av_packet_unref(avpkt);
}
return ret;
}
static int encode_make_refcounted(AVCodecContext *avctx, AVPacket *avpkt)
{
uint8_t *data = avpkt->data;
int ret;
if (avpkt->buf)
return 0;
avpkt->data = NULL;
ret = ff_get_encode_buffer(avctx, avpkt, avpkt->size, 0);
if (ret < 0)
return ret;
memcpy(avpkt->data, data, avpkt->size);
return 0;
}
/**
* Pad last frame with silence.
*/
static int pad_last_frame(AVCodecContext *s, AVFrame *frame, const AVFrame *src, int out_samples)
{
int ret;
frame->format = src->format;
frame->nb_samples = out_samples;
ret = av_channel_layout_copy(&frame->ch_layout, &s->ch_layout);
if (ret < 0)
goto fail;
ret = av_frame_get_buffer(frame, 0);
if (ret < 0)
goto fail;
ret = av_frame_copy_props(frame, src);
if (ret < 0)
goto fail;
if ((ret = av_samples_copy(frame->extended_data, src->extended_data, 0, 0,
src->nb_samples, s->ch_layout.nb_channels,
s->sample_fmt)) < 0)
goto fail;
if ((ret = av_samples_set_silence(frame->extended_data, src->nb_samples,
frame->nb_samples - src->nb_samples,
s->ch_layout.nb_channels, s->sample_fmt)) < 0)
goto fail;
return 0;
fail:
av_frame_unref(frame);
encode_ctx(s->internal)->last_audio_frame = 0;
return ret;
}
int avcodec_encode_subtitle(AVCodecContext *avctx, uint8_t *buf, int buf_size,
const AVSubtitle *sub)
{
int ret;
if (sub->start_display_time) {
av_log(avctx, AV_LOG_ERROR, "start_display_time must be 0.\n");
return -1;
}
ret = ffcodec(avctx->codec)->cb.encode_sub(avctx, buf, buf_size, sub);
avctx->frame_num++;
return ret;
}
int ff_encode_get_frame(AVCodecContext *avctx, AVFrame *frame)
{
AVCodecInternal *avci = avctx->internal;
if (avci->draining)
return AVERROR_EOF;
if (!avci->buffer_frame->buf[0])
return AVERROR(EAGAIN);
av_frame_move_ref(frame, avci->buffer_frame);
#if FF_API_FRAME_KEY
FF_DISABLE_DEPRECATION_WARNINGS
if (frame->key_frame)
frame->flags |= AV_FRAME_FLAG_KEY;
FF_ENABLE_DEPRECATION_WARNINGS
#endif
#if FF_API_INTERLACED_FRAME
FF_DISABLE_DEPRECATION_WARNINGS
if (frame->interlaced_frame)
frame->flags |= AV_FRAME_FLAG_INTERLACED;
if (frame->top_field_first)
frame->flags |= AV_FRAME_FLAG_TOP_FIELD_FIRST;
FF_ENABLE_DEPRECATION_WARNINGS
#endif
return 0;
}
int ff_encode_reordered_opaque(AVCodecContext *avctx,
AVPacket *pkt, const AVFrame *frame)
{
if (avctx->flags & AV_CODEC_FLAG_COPY_OPAQUE) {
int ret = av_buffer_replace(&pkt->opaque_ref, frame->opaque_ref);
if (ret < 0)
return ret;
pkt->opaque = frame->opaque;
}
return 0;
}
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
int ff_encode_encode_cb(AVCodecContext *avctx, AVPacket *avpkt,
AVFrame *frame, int *got_packet)
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
{
const FFCodec *const codec = ffcodec(avctx->codec);
int ret;
ret = codec->cb.encode(avctx, avpkt, frame, got_packet);
emms_c();
av_assert0(ret <= 0);
if (!ret && *got_packet) {
if (avpkt->data) {
ret = encode_make_refcounted(avctx, avpkt);
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
if (ret < 0)
goto unref;
// Date returned by encoders must always be ref-counted
av_assert0(avpkt->buf);
}
// set the timestamps for the simple no-delay case
// encoders with delay have to set the timestamps themselves
if (!(avctx->codec->capabilities & AV_CODEC_CAP_DELAY) ||
(frame && (codec->caps_internal & FF_CODEC_CAP_EOF_FLUSH))) {
if (avpkt->pts == AV_NOPTS_VALUE)
avpkt->pts = frame->pts;
if (!avpkt->duration) {
if (frame->duration)
avpkt->duration = frame->duration;
else if (avctx->codec->type == AVMEDIA_TYPE_AUDIO) {
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
avpkt->duration = ff_samples_to_time_base(avctx,
frame->nb_samples);
}
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
}
ret = ff_encode_reordered_opaque(avctx, avpkt, frame);
if (ret < 0)
goto unref;
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
}
// dts equals pts unless there is reordering
// there can be no reordering if there is no encoder delay
if (!(avctx->codec_descriptor->props & AV_CODEC_PROP_REORDER) ||
!(avctx->codec->capabilities & AV_CODEC_CAP_DELAY) ||
(codec->caps_internal & FF_CODEC_CAP_EOF_FLUSH))
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
avpkt->dts = avpkt->pts;
} else {
unref:
av_packet_unref(avpkt);
}
if (frame)
av_frame_unref(frame);
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
return ret;
}
static int encode_simple_internal(AVCodecContext *avctx, AVPacket *avpkt)
{
AVCodecInternal *avci = avctx->internal;
AVFrame *frame = avci->in_frame;
const FFCodec *const codec = ffcodec(avctx->codec);
int got_packet;
int ret;
if (avci->draining_done)
return AVERROR_EOF;
if (!frame->buf[0] && !avci->draining) {
av_frame_unref(frame);
ret = ff_encode_get_frame(avctx, frame);
if (ret < 0 && ret != AVERROR_EOF)
return ret;
}
if (!frame->buf[0]) {
if (!(avctx->codec->capabilities & AV_CODEC_CAP_DELAY ||
avci->frame_thread_encoder))
return AVERROR_EOF;
// Flushing is signaled with a NULL frame
frame = NULL;
}
got_packet = 0;
av_assert0(codec->cb_type == FF_CODEC_CB_TYPE_ENCODE);
if (CONFIG_FRAME_THREAD_ENCODER && avci->frame_thread_encoder)
/* This will unref frame. */
ret = ff_thread_video_encode_frame(avctx, avpkt, frame, &got_packet);
else {
avcodec/encode, frame_thread_encoder: Unify calling encode callback The encode-callback (the callback used by the FF_CODEC_CB_TYPE_ENCODE encoders) is currently called in two places: encode_simple_internal() and by the worker threads of frame-threaded encoders. After the call, some packet properties are set based upon the corresponding AVFrame properties and the packet is made refcounted if it isn't already. So there is some code duplication. There was also non-duplicated code in encode_simple_internal() which is executed even when using frame-threading. This included an emms_c() (which is needed for frame-threading, too, if it is needed for the single-threaded case, because there are allocations (via av_packet_make_refcounted()) immediately after returning from the encode-callback). Furthermore, some further properties are only set in encode_simple_internal(): For audio, pts and duration are derived from the corresponding fields of the frame if the encoder does not have the AV_CODEC_CAP_DELAY set. Yet this is wrong for frame-threaded encoders, because frame-threading always introduces delay regardless of whether the underlying codec has said cap. This only worked because there are no frame-threaded audio encoders. This commit fixes the code duplication and the above issue by factoring this code out and reusing it in both places. It would work in case of audio codecs with frame-threading, because now the values are derived from the correct AVFrame. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-08-23 23:26:57 +02:00
ret = ff_encode_encode_cb(avctx, avpkt, frame, &got_packet);
}
if (avci->draining && !got_packet)
avci->draining_done = 1;
return ret;
}
static int encode_simple_receive_packet(AVCodecContext *avctx, AVPacket *avpkt)
{
int ret;
while (!avpkt->data && !avpkt->side_data) {
ret = encode_simple_internal(avctx, avpkt);
if (ret < 0)
return ret;
}
return 0;
}
static int encode_receive_packet_internal(AVCodecContext *avctx, AVPacket *avpkt)
{
AVCodecInternal *avci = avctx->internal;
int ret;
if (avci->draining_done)
return AVERROR_EOF;
av_assert0(!avpkt->data && !avpkt->side_data);
if (avctx->codec->type == AVMEDIA_TYPE_VIDEO) {
if ((avctx->flags & AV_CODEC_FLAG_PASS1) && avctx->stats_out)
avctx->stats_out[0] = '\0';
if (av_image_check_size2(avctx->width, avctx->height, avctx->max_pixels, AV_PIX_FMT_NONE, 0, avctx))
return AVERROR(EINVAL);
}
if (ffcodec(avctx->codec)->cb_type == FF_CODEC_CB_TYPE_RECEIVE_PACKET) {
ret = ffcodec(avctx->codec)->cb.receive_packet(avctx, avpkt);
if (ret < 0)
av_packet_unref(avpkt);
else
// Encoders must always return ref-counted buffers.
// Side-data only packets have no data and can be not ref-counted.
av_assert0(!avpkt->data || avpkt->buf);
} else
ret = encode_simple_receive_packet(avctx, avpkt);
if (ret >= 0)
avpkt->flags |= encode_ctx(avci)->intra_only_flag;
if (ret == AVERROR_EOF)
avci->draining_done = 1;
return ret;
}
#if CONFIG_LCMS2
static int encode_generate_icc_profile(AVCodecContext *avctx, AVFrame *frame)
{
enum AVColorTransferCharacteristic trc = frame->color_trc;
enum AVColorPrimaries prim = frame->color_primaries;
const FFCodec *const codec = ffcodec(avctx->codec);
AVCodecInternal *avci = avctx->internal;
cmsHPROFILE profile;
int ret;
/* don't generate ICC profiles if disabled or unsupported */
if (!(avctx->flags2 & AV_CODEC_FLAG2_ICC_PROFILES))
return 0;
if (!(codec->caps_internal & FF_CODEC_CAP_ICC_PROFILES))
return 0;
if (trc == AVCOL_TRC_UNSPECIFIED)
trc = avctx->color_trc;
if (prim == AVCOL_PRI_UNSPECIFIED)
prim = avctx->color_primaries;
if (trc == AVCOL_TRC_UNSPECIFIED || prim == AVCOL_PRI_UNSPECIFIED)
return 0; /* can't generate ICC profile with missing csp tags */
if (av_frame_get_side_data(frame, AV_FRAME_DATA_ICC_PROFILE))
return 0; /* don't overwrite existing ICC profile */
if (!avci->icc.avctx) {
ret = ff_icc_context_init(&avci->icc, avctx);
if (ret < 0)
return ret;
}
ret = ff_icc_profile_generate(&avci->icc, prim, trc, &profile);
if (ret < 0)
return ret;
ret = ff_icc_profile_attach(&avci->icc, profile, frame);
cmsCloseProfile(profile);
return ret;
}
#else /* !CONFIG_LCMS2 */
static int encode_generate_icc_profile(av_unused AVCodecContext *c, av_unused AVFrame *f)
{
return 0;
}
#endif
static int encode_send_frame_internal(AVCodecContext *avctx, const AVFrame *src)
{
AVCodecInternal *avci = avctx->internal;
EncodeContext *ec = encode_ctx(avci);
AVFrame *dst = avci->buffer_frame;
int ret;
if (avctx->codec->type == AVMEDIA_TYPE_AUDIO) {
/* extract audio service type metadata */
AVFrameSideData *sd = av_frame_get_side_data(src, AV_FRAME_DATA_AUDIO_SERVICE_TYPE);
if (sd && sd->size >= sizeof(enum AVAudioServiceType))
avctx->audio_service_type = *(enum AVAudioServiceType*)sd->data;
/* check for valid frame size */
if (!(avctx->codec->capabilities & AV_CODEC_CAP_VARIABLE_FRAME_SIZE)) {
/* if we already got an undersized frame, that must have been the last */
if (ec->last_audio_frame) {
av_log(avctx, AV_LOG_ERROR, "frame_size (%d) was not respected for a non-last frame\n", avctx->frame_size);
return AVERROR(EINVAL);
}
if (src->nb_samples > avctx->frame_size) {
av_log(avctx, AV_LOG_ERROR, "nb_samples (%d) > frame_size (%d)\n", src->nb_samples, avctx->frame_size);
return AVERROR(EINVAL);
}
if (src->nb_samples < avctx->frame_size) {
ec->last_audio_frame = 1;
if (!(avctx->codec->capabilities & AV_CODEC_CAP_SMALL_LAST_FRAME)) {
int pad_samples = avci->pad_samples ? avci->pad_samples : avctx->frame_size;
int out_samples = (src->nb_samples + pad_samples - 1) / pad_samples * pad_samples;
if (out_samples != src->nb_samples) {
ret = pad_last_frame(avctx, dst, src, out_samples);
if (ret < 0)
return ret;
goto finish;
}
}
}
}
}
ret = av_frame_ref(dst, src);
if (ret < 0)
return ret;
finish:
if (avctx->codec->type == AVMEDIA_TYPE_VIDEO) {
ret = encode_generate_icc_profile(avctx, dst);
if (ret < 0)
return ret;
}
// unset frame duration unless AV_CODEC_FLAG_FRAME_DURATION is set,
// since otherwise we cannot be sure that whatever value it has is in the
// right timebase, so we would produce an incorrect value, which is worse
// than none at all
if (!(avctx->flags & AV_CODEC_FLAG_FRAME_DURATION))
dst->duration = 0;
return 0;
}
int attribute_align_arg avcodec_send_frame(AVCodecContext *avctx, const AVFrame *frame)
{
AVCodecInternal *avci = avctx->internal;
int ret;
if (!avcodec_is_open(avctx) || !av_codec_is_encoder(avctx->codec))
return AVERROR(EINVAL);
if (avci->draining)
return AVERROR_EOF;
if (avci->buffer_frame->buf[0])
return AVERROR(EAGAIN);
if (!frame) {
avci->draining = 1;
} else {
ret = encode_send_frame_internal(avctx, frame);
if (ret < 0)
return ret;
}
if (!avci->buffer_pkt->data && !avci->buffer_pkt->side_data) {
ret = encode_receive_packet_internal(avctx, avci->buffer_pkt);
if (ret < 0 && ret != AVERROR(EAGAIN) && ret != AVERROR_EOF)
return ret;
}
avctx->frame_num++;
return 0;
}
int attribute_align_arg avcodec_receive_packet(AVCodecContext *avctx, AVPacket *avpkt)
{
AVCodecInternal *avci = avctx->internal;
int ret;
av_packet_unref(avpkt);
if (!avcodec_is_open(avctx) || !av_codec_is_encoder(avctx->codec))
return AVERROR(EINVAL);
if (avci->buffer_pkt->data || avci->buffer_pkt->side_data) {
av_packet_move_ref(avpkt, avci->buffer_pkt);
} else {
ret = encode_receive_packet_internal(avctx, avpkt);
if (ret < 0)
return ret;
}
return 0;
}
static int encode_preinit_video(AVCodecContext *avctx)
{
const AVCodec *c = avctx->codec;
2022-01-20 16:53:18 +01:00
const AVPixFmtDescriptor *pixdesc = av_pix_fmt_desc_get(avctx->pix_fmt);
int i;
if (!av_get_pix_fmt_name(avctx->pix_fmt)) {
av_log(avctx, AV_LOG_ERROR, "Invalid video pixel format: %d\n",
avctx->pix_fmt);
return AVERROR(EINVAL);
}
if (c->pix_fmts) {
for (i = 0; c->pix_fmts[i] != AV_PIX_FMT_NONE; i++)
if (avctx->pix_fmt == c->pix_fmts[i])
break;
if (c->pix_fmts[i] == AV_PIX_FMT_NONE) {
av_log(avctx, AV_LOG_ERROR,
"Specified pixel format %s is not supported by the %s encoder.\n",
av_get_pix_fmt_name(avctx->pix_fmt), c->name);
av_log(avctx, AV_LOG_ERROR, "Supported pixel formats:\n");
for (int p = 0; c->pix_fmts[p] != AV_PIX_FMT_NONE; p++) {
av_log(avctx, AV_LOG_ERROR, " %s\n",
av_get_pix_fmt_name(c->pix_fmts[p]));
}
return AVERROR(EINVAL);
}
if (c->pix_fmts[i] == AV_PIX_FMT_YUVJ420P ||
c->pix_fmts[i] == AV_PIX_FMT_YUVJ411P ||
c->pix_fmts[i] == AV_PIX_FMT_YUVJ422P ||
c->pix_fmts[i] == AV_PIX_FMT_YUVJ440P ||
c->pix_fmts[i] == AV_PIX_FMT_YUVJ444P)
avctx->color_range = AVCOL_RANGE_JPEG;
}
2022-01-20 16:53:18 +01:00
if ( avctx->bits_per_raw_sample < 0
|| (avctx->bits_per_raw_sample > 8 && pixdesc->comp[0].depth <= 8)) {
av_log(avctx, AV_LOG_WARNING, "Specified bit depth %d not possible with the specified pixel formats depth %d\n",
avctx->bits_per_raw_sample, pixdesc->comp[0].depth);
avctx->bits_per_raw_sample = pixdesc->comp[0].depth;
}
if (avctx->width <= 0 || avctx->height <= 0) {
av_log(avctx, AV_LOG_ERROR, "dimensions not set\n");
return AVERROR(EINVAL);
}
#if FF_API_TICKS_PER_FRAME
FF_DISABLE_DEPRECATION_WARNINGS
if (avctx->ticks_per_frame && avctx->time_base.num &&
avctx->ticks_per_frame > INT_MAX / avctx->time_base.num) {
av_log(avctx, AV_LOG_ERROR,
"ticks_per_frame %d too large for the timebase %d/%d.",
avctx->ticks_per_frame,
avctx->time_base.num,
avctx->time_base.den);
return AVERROR(EINVAL);
}
FF_ENABLE_DEPRECATION_WARNINGS
#endif
if (avctx->hw_frames_ctx) {
AVHWFramesContext *frames_ctx = (AVHWFramesContext*)avctx->hw_frames_ctx->data;
if (frames_ctx->format != avctx->pix_fmt) {
av_log(avctx, AV_LOG_ERROR,
"Mismatching AVCodecContext.pix_fmt and AVHWFramesContext.format\n");
return AVERROR(EINVAL);
}
if (avctx->sw_pix_fmt != AV_PIX_FMT_NONE &&
avctx->sw_pix_fmt != frames_ctx->sw_format) {
av_log(avctx, AV_LOG_ERROR,
"Mismatching AVCodecContext.sw_pix_fmt (%s) "
"and AVHWFramesContext.sw_format (%s)\n",
av_get_pix_fmt_name(avctx->sw_pix_fmt),
av_get_pix_fmt_name(frames_ctx->sw_format));
return AVERROR(EINVAL);
}
avctx->sw_pix_fmt = frames_ctx->sw_format;
}
return 0;
}
static int encode_preinit_audio(AVCodecContext *avctx)
{
const AVCodec *c = avctx->codec;
int i;
if (!av_get_sample_fmt_name(avctx->sample_fmt)) {
av_log(avctx, AV_LOG_ERROR, "Invalid audio sample format: %d\n",
avctx->sample_fmt);
return AVERROR(EINVAL);
}
if (avctx->sample_rate <= 0) {
av_log(avctx, AV_LOG_ERROR, "Invalid audio sample rate: %d\n",
avctx->sample_rate);
return AVERROR(EINVAL);
}
if (c->sample_fmts) {
for (i = 0; c->sample_fmts[i] != AV_SAMPLE_FMT_NONE; i++) {
if (avctx->sample_fmt == c->sample_fmts[i])
break;
if (avctx->ch_layout.nb_channels == 1 &&
av_get_planar_sample_fmt(avctx->sample_fmt) ==
av_get_planar_sample_fmt(c->sample_fmts[i])) {
avctx->sample_fmt = c->sample_fmts[i];
break;
}
}
if (c->sample_fmts[i] == AV_SAMPLE_FMT_NONE) {
av_log(avctx, AV_LOG_ERROR,
"Specified sample format %s is not supported by the %s encoder\n",
av_get_sample_fmt_name(avctx->sample_fmt), c->name);
av_log(avctx, AV_LOG_ERROR, "Supported sample formats:\n");
for (int p = 0; c->sample_fmts[p] != AV_SAMPLE_FMT_NONE; p++) {
av_log(avctx, AV_LOG_ERROR, " %s\n",
av_get_sample_fmt_name(c->sample_fmts[p]));
}
return AVERROR(EINVAL);
}
}
if (c->supported_samplerates) {
for (i = 0; c->supported_samplerates[i] != 0; i++)
if (avctx->sample_rate == c->supported_samplerates[i])
break;
if (c->supported_samplerates[i] == 0) {
av_log(avctx, AV_LOG_ERROR,
"Specified sample rate %d is not supported by the %s encoder\n",
avctx->sample_rate, c->name);
av_log(avctx, AV_LOG_ERROR, "Supported sample rates:\n");
for (int p = 0; c->supported_samplerates[p]; p++)
av_log(avctx, AV_LOG_ERROR, " %d\n", c->supported_samplerates[p]);
return AVERROR(EINVAL);
}
}
if (c->ch_layouts) {
for (i = 0; c->ch_layouts[i].nb_channels; i++) {
if (!av_channel_layout_compare(&avctx->ch_layout, &c->ch_layouts[i]))
break;
}
if (!c->ch_layouts[i].nb_channels) {
char buf[512];
int ret = av_channel_layout_describe(&avctx->ch_layout, buf, sizeof(buf));
av_log(avctx, AV_LOG_ERROR,
"Specified channel layout '%s' is not supported by the %s encoder\n",
ret > 0 ? buf : "?", c->name);
av_log(avctx, AV_LOG_ERROR, "Supported channel layouts:\n");
for (int p = 0; c->ch_layouts[p].nb_channels; p++) {
ret = av_channel_layout_describe(&c->ch_layouts[p], buf, sizeof(buf));
av_log(avctx, AV_LOG_ERROR, " %s\n", ret > 0 ? buf : "?");
}
return AVERROR(EINVAL);
}
}
if (!avctx->bits_per_raw_sample)
avctx->bits_per_raw_sample = av_get_exact_bits_per_sample(avctx->codec_id);
if (!avctx->bits_per_raw_sample)
avctx->bits_per_raw_sample = 8 * av_get_bytes_per_sample(avctx->sample_fmt);
return 0;
}
int ff_encode_preinit(AVCodecContext *avctx)
{
AVCodecInternal *avci = avctx->internal;
EncodeContext *ec = encode_ctx(avci);
int ret = 0;
if (avctx->time_base.num <= 0 || avctx->time_base.den <= 0) {
av_log(avctx, AV_LOG_ERROR, "The encoder timebase is not set.\n");
return AVERROR(EINVAL);
}
if (avctx->flags & AV_CODEC_FLAG_COPY_OPAQUE &&
!(avctx->codec->capabilities & AV_CODEC_CAP_ENCODER_REORDERED_OPAQUE)) {
av_log(avctx, AV_LOG_ERROR, "The copy_opaque flag is set, but the "
"encoder does not support it.\n");
return AVERROR(EINVAL);
}
switch (avctx->codec_type) {
case AVMEDIA_TYPE_VIDEO: ret = encode_preinit_video(avctx); break;
case AVMEDIA_TYPE_AUDIO: ret = encode_preinit_audio(avctx); break;
}
if (ret < 0)
return ret;
if ( (avctx->codec_type == AVMEDIA_TYPE_VIDEO || avctx->codec_type == AVMEDIA_TYPE_AUDIO)
&& avctx->bit_rate>0 && avctx->bit_rate<1000) {
av_log(avctx, AV_LOG_WARNING, "Bitrate %"PRId64" is extremely low, maybe you mean %"PRId64"k\n", avctx->bit_rate, avctx->bit_rate);
}
if (!avctx->rc_initial_buffer_occupancy)
avctx->rc_initial_buffer_occupancy = avctx->rc_buffer_size * 3LL / 4;
if (avctx->codec_descriptor->props & AV_CODEC_PROP_INTRA_ONLY)
ec->intra_only_flag = AV_PKT_FLAG_KEY;
if (ffcodec(avctx->codec)->cb_type == FF_CODEC_CB_TYPE_ENCODE) {
avci->in_frame = av_frame_alloc();
if (!avci->in_frame)
return AVERROR(ENOMEM);
}
if ((avctx->flags & AV_CODEC_FLAG_RECON_FRAME)) {
if (!(avctx->codec->capabilities & AV_CODEC_CAP_ENCODER_RECON_FRAME)) {
av_log(avctx, AV_LOG_ERROR, "Reconstructed frame output requested "
"from an encoder not supporting it\n");
return AVERROR(ENOSYS);
}
avci->recon_frame = av_frame_alloc();
if (!avci->recon_frame)
return AVERROR(ENOMEM);
}
for (int i = 0; ff_sd_global_map[i].packet < AV_PKT_DATA_NB; i++) {
const enum AVPacketSideDataType type_packet = ff_sd_global_map[i].packet;
const enum AVFrameSideDataType type_frame = ff_sd_global_map[i].frame;
const AVFrameSideData *sd_frame;
AVPacketSideData *sd_packet;
sd_frame = av_frame_side_data_get(avctx->decoded_side_data,
avctx->nb_decoded_side_data,
type_frame);
if (!sd_frame ||
av_packet_side_data_get(avctx->coded_side_data, avctx->nb_coded_side_data,
type_packet))
continue;
sd_packet = av_packet_side_data_new(&avctx->coded_side_data, &avctx->nb_coded_side_data,
type_packet, sd_frame->size, 0);
if (!sd_packet)
return AVERROR(ENOMEM);
memcpy(sd_packet->data, sd_frame->data, sd_frame->size);
}
if (CONFIG_FRAME_THREAD_ENCODER) {
ret = ff_frame_thread_encoder_init(avctx);
if (ret < 0)
return ret;
}
return 0;
}
int ff_encode_alloc_frame(AVCodecContext *avctx, AVFrame *frame)
{
int ret;
switch (avctx->codec->type) {
case AVMEDIA_TYPE_VIDEO:
frame->format = avctx->pix_fmt;
if (frame->width <= 0 || frame->height <= 0) {
frame->width = FFMAX(avctx->width, avctx->coded_width);
frame->height = FFMAX(avctx->height, avctx->coded_height);
}
break;
case AVMEDIA_TYPE_AUDIO:
frame->sample_rate = avctx->sample_rate;
frame->format = avctx->sample_fmt;
if (!frame->ch_layout.nb_channels) {
ret = av_channel_layout_copy(&frame->ch_layout, &avctx->ch_layout);
if (ret < 0)
return ret;
}
break;
}
ret = avcodec_default_get_buffer2(avctx, frame, 0);
if (ret < 0) {
av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n");
av_frame_unref(frame);
return ret;
}
return 0;
}
int ff_encode_receive_frame(AVCodecContext *avctx, AVFrame *frame)
{
AVCodecInternal *avci = avctx->internal;
if (!avci->recon_frame)
return AVERROR(EINVAL);
if (!avci->recon_frame->buf[0])
return avci->draining_done ? AVERROR_EOF : AVERROR(EAGAIN);
av_frame_move_ref(frame, avci->recon_frame);
return 0;
}
void ff_encode_flush_buffers(AVCodecContext *avctx)
{
AVCodecInternal *avci = avctx->internal;
2023-06-20 12:53:48 +02:00
if (avci->in_frame)
av_frame_unref(avci->in_frame);
if (avci->recon_frame)
av_frame_unref(avci->recon_frame);
}
AVCodecInternal *ff_encode_internal_alloc(void)
{
return av_mallocz(sizeof(EncodeContext));
}
AVCPBProperties *ff_encode_add_cpb_side_data(AVCodecContext *avctx)
{
AVPacketSideData *tmp;
AVCPBProperties *props;
size_t size;
int i;
for (i = 0; i < avctx->nb_coded_side_data; i++)
if (avctx->coded_side_data[i].type == AV_PKT_DATA_CPB_PROPERTIES)
return (AVCPBProperties *)avctx->coded_side_data[i].data;
props = av_cpb_properties_alloc(&size);
if (!props)
return NULL;
tmp = av_realloc_array(avctx->coded_side_data, avctx->nb_coded_side_data + 1, sizeof(*tmp));
if (!tmp) {
av_freep(&props);
return NULL;
}
avctx->coded_side_data = tmp;
avctx->nb_coded_side_data++;
avctx->coded_side_data[avctx->nb_coded_side_data - 1].type = AV_PKT_DATA_CPB_PROPERTIES;
avctx->coded_side_data[avctx->nb_coded_side_data - 1].data = (uint8_t*)props;
avctx->coded_side_data[avctx->nb_coded_side_data - 1].size = size;
return props;
}