Files
zoneminder/src/zm_ffmpeg.cpp
Aaron Kling e710127924 ffmpeg: Rework decode and encode flows
send_packet will fail with -EAGAIN if receive_frame hasn't consumed the
last packet. Similarly send_frame will fail if receive_packet has
pending frames.

This fixes problems with the raspberry pi hardware decoder. The change
likely affects many other codecs as well.
2024-02-19 17:17:24 -06:00

514 lines
17 KiB
C++

/*
* ZoneMinder FFMPEG implementation, $Date$, $Revision$
* Copyright (C) 2001-2008 Philip Coombes
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*/
#include "zm_ffmpeg.h"
#include "zm_logger.h"
#include "zm_rgb.h"
#include "zm_utils.h"
extern "C" {
#include <libavutil/pixdesc.h>
}
void log_libav_callback(void *ptr, int level, const char *fmt, va_list vargs) {
Logger *log = Logger::fetch();
int log_level = 0;
if (level == AV_LOG_QUIET) { // -8
log_level = Logger::NOLOG;
} else if (level == AV_LOG_PANIC) { //0
log_level = Logger::PANIC;
} else if (level == AV_LOG_FATAL) { // 8
log_level = Logger::FATAL;
} else if (level == AV_LOG_ERROR) { // 16
log_level = Logger::WARNING; // ffmpeg outputs a lot of errors that don't really affect anything.
} else if (level == AV_LOG_WARNING) { //24
log_level = Logger::INFO;
} else if (level == AV_LOG_INFO) { //32
log_level = Logger::DEBUG1;
} else if (level == AV_LOG_VERBOSE) { //40
log_level = Logger::DEBUG2;
} else if (level == AV_LOG_DEBUG) { //48
log_level = Logger::DEBUG3;
#ifdef AV_LOG_TRACE
} else if (level == AV_LOG_TRACE) {
log_level = Logger::DEBUG8;
#endif
#ifdef AV_LOG_MAX_OFFSET
} else if (level == AV_LOG_MAX_OFFSET) {
log_level = Logger::DEBUG9;
#endif
} else {
Error("Unknown log level %d", level);
}
if (log and (log->level() >= log_level) ) {
char logString[8192];
int length = vsnprintf(logString, sizeof(logString)-1, fmt, vargs);
if (length > 0) {
if (static_cast<size_t>(length) > sizeof(logString)-1) length = sizeof(logString)-1;
// ffmpeg logs have a carriage return, so replace it with terminator
logString[length-1] = 0;
log->logPrint(false, __FILE__, __LINE__, log_level, "%s", logString);
} else {
log->logPrint(false, __FILE__, __LINE__, AV_LOG_ERROR, "Can't encode log from av. fmt was %s", fmt);
}
}
}
static bool bInit = false;
void FFMPEGInit() {
if (!bInit) {
if (logDebugging() && config.log_ffmpeg) {
av_log_set_level(AV_LOG_DEBUG);
av_log_set_callback(log_libav_callback);
Info("Enabling ffmpeg logs, as LOG_DEBUG+LOG_FFMPEG are enabled in options");
} else {
Debug(1,"Not enabling ffmpeg logs, as LOG_FFMPEG and/or LOG_DEBUG is disabled in options, or this monitor is not part of your debug targets");
av_log_set_level(AV_LOG_QUIET);
}
#if !LIBAVFORMAT_VERSION_CHECK(58, 9, 58, 9, 0)
av_register_all();
#endif
avformat_network_init();
bInit = true;
}
}
void FFMPEGDeInit() {
avformat_network_deinit();
bInit = false;
}
enum _AVPIXELFORMAT GetFFMPEGPixelFormat(unsigned int p_colours, unsigned p_subpixelorder) {
enum _AVPIXELFORMAT pf;
Debug(8,"Colours: %d SubpixelOrder: %d",p_colours,p_subpixelorder);
switch (p_colours) {
case ZM_COLOUR_RGB24:
if(p_subpixelorder == ZM_SUBPIX_ORDER_BGR) {
/* BGR subpixel order */
pf = AV_PIX_FMT_BGR24;
} else {
/* Assume RGB subpixel order */
pf = AV_PIX_FMT_RGB24;
}
break;
case ZM_COLOUR_RGB32:
if (p_subpixelorder == ZM_SUBPIX_ORDER_ARGB) {
/* ARGB subpixel order */
pf = AV_PIX_FMT_ARGB;
} else if (p_subpixelorder == ZM_SUBPIX_ORDER_ABGR) {
/* ABGR subpixel order */
pf = AV_PIX_FMT_ABGR;
} else if (p_subpixelorder == ZM_SUBPIX_ORDER_BGRA) {
/* BGRA subpixel order */
pf = AV_PIX_FMT_BGRA;
} else {
/* Assume RGBA subpixel order */
pf = AV_PIX_FMT_RGBA;
}
break;
case ZM_COLOUR_GRAY8:
pf = AV_PIX_FMT_GRAY8;
break;
default:
Panic("Unexpected colours: %d", p_colours);
pf = AV_PIX_FMT_GRAY8; /* Just to shush gcc variable may be unused warning */
break;
}
return pf;
}
#if LIBAVUTIL_VERSION_CHECK(56, 0, 0, 17, 100)
int64_t av_rescale_delta(AVRational in_tb, int64_t in_ts, AVRational fs_tb, int duration, int64_t *last, AVRational out_tb){
int64_t a, b, this_thing;
av_assert0(in_ts != AV_NOPTS_VALUE);
av_assert0(duration >= 0);
if (*last == AV_NOPTS_VALUE || !duration || in_tb.num*(int64_t)out_tb.den <= out_tb.num*(int64_t)in_tb.den) {
simple_round:
*last = av_rescale_q(in_ts, in_tb, fs_tb) + duration;
return av_rescale_q(in_ts, in_tb, out_tb);
}
a = av_rescale_q_rnd(2*in_ts-1, in_tb, fs_tb, AV_ROUND_DOWN) >>1;
b = (av_rescale_q_rnd(2*in_ts+1, in_tb, fs_tb, AV_ROUND_UP )+1)>>1;
if (*last < 2*a - b || *last > 2*b - a)
goto simple_round;
this_thing = av_clip64(*last, a, b);
*last = this_thing + duration;
return av_rescale_q(this_thing, fs_tb, out_tb);
}
#endif
static void zm_log_fps(double d, const char *postfix) {
uint64_t v = lrintf(d * 100);
if (!v) {
Debug(1, "%1.4f %s", d, postfix);
} else if (v % 100) {
Debug(1, "%3.2f %s", d, postfix);
} else if (v % (100 * 1000)) {
Debug(1, "%1.0f %s", d, postfix);
} else {
Debug(1, "%1.0fk %s", d / 1000, postfix);
}
}
void zm_dump_codecpar(const AVCodecParameters *par) {
Debug(1, "Dumping codecpar codec_type %d %s codec_id %d %s codec_tag %" PRIu32
" width %d height %d bit_rate%" PRIu64 " bpcs %d bprs %d format%d %s"
" extradata:%d:%s profile %d level %d field order %d color_range %d"
" color_primaries %d color_trc %d color_space %d location %d video_delay %d",
static_cast<int>(par->codec_type),
av_get_media_type_string(par->codec_type),
static_cast<int>(par->codec_id),
avcodec_get_name(par->codec_id),
par->codec_tag,
par->width,
par->height,
par->bit_rate,
par->bits_per_coded_sample,
par->bits_per_raw_sample,
par->format,
(((AVPixelFormat)par->format == AV_PIX_FMT_NONE) ? "none" : av_get_pix_fmt_name((AVPixelFormat)par->format)),
par->extradata_size, ByteArrayToHexString(nonstd::span<const uint8>{
par->extradata,
static_cast<nonstd::span_lite::span<const unsigned char>::size_type>(par->extradata_size)
}).c_str(),
par->profile,
par->level,
static_cast<int>(par->field_order),
static_cast<int>(par->color_range),
static_cast<int>(par->color_primaries),
static_cast<int>(par->color_trc),
static_cast<int>(par->color_space),
static_cast<int>(par->chroma_location),
static_cast<int>(par->video_delay)
);
}
void zm_dump_codec(const AVCodecContext *codec) {
Debug(1, "Dumping codec_context codec_type %d %s codec_id %d %s width %d height %d timebase %d/%d format %s profile %d level %d "
"gop_size %d has_b_frames %d max_b_frames %d me_cmp %d me_range %d qmin %d qmax %d bit_rate %" PRId64 " extradata:%d:%s",
codec->codec_type,
av_get_media_type_string(codec->codec_type),
codec->codec_id,
avcodec_get_name(codec->codec_id),
codec->width,
codec->height,
codec->time_base.num,
codec->time_base.den,
(codec->pix_fmt == AV_PIX_FMT_NONE ? "none" : av_get_pix_fmt_name(codec->pix_fmt)),
codec->profile,
codec->level,
codec->gop_size,
codec->has_b_frames,
codec->max_b_frames,
codec->me_cmp,
codec->me_range,
codec->qmin,
codec->qmax,
codec->bit_rate,
codec->extradata_size,
ByteArrayToHexString(nonstd::span<const uint8>{
codec->extradata,
static_cast<nonstd::span_lite::span<const unsigned char>::size_type>(codec->extradata_size)
}).c_str()
);
}
/* "user interface" functions */
void zm_dump_stream_format(AVFormatContext *ic, int i, int index, int is_output) {
Debug(1, "Dumping stream index i(%d) index(%d)", i, index);
int flags = (is_output ? ic->oformat->flags : ic->iformat->flags);
AVStream *st = ic->streams[i];
AVDictionaryEntry *lang = av_dict_get(st->metadata, "language", nullptr, 0);
AVCodecParameters *codec = st->codecpar;
Debug(1, " Stream #%d:%d", index, i);
/* the pid is an important information, so we display it */
/* XXX: add a generic system */
if (flags & AVFMT_SHOW_IDS)
Debug(1, "ids [0x%x]", st->id);
if (lang)
Debug(1, "language (%s)", lang->value);
Debug(1, "frame_size:%d stream timebase: %d/%d",
codec->frame_size,
st->time_base.num, st->time_base.den
);
Debug(1, "codec: %s %s",
avcodec_get_name(st->codecpar->codec_id),
av_get_media_type_string(st->codecpar->codec_type)
);
if (st->sample_aspect_ratio.num && // default
av_cmp_q(st->sample_aspect_ratio, codec->sample_aspect_ratio)
) {
AVRational display_aspect_ratio;
av_reduce(&display_aspect_ratio.num,
&display_aspect_ratio.den,
codec->width * (int64_t)st->sample_aspect_ratio.num,
codec->height * (int64_t)st->sample_aspect_ratio.den,
1024 * 1024);
Debug(1, ", SAR %d:%d DAR %d:%d",
st->sample_aspect_ratio.num, st->sample_aspect_ratio.den,
display_aspect_ratio.num, display_aspect_ratio.den);
} else {
Debug(1, ", SAR %d:%d ",
st->sample_aspect_ratio.num, st->sample_aspect_ratio.den);
}
if (codec->codec_type == AVMEDIA_TYPE_VIDEO) {
int fps = st->avg_frame_rate.den && st->avg_frame_rate.num;
int tbn = st->time_base.den && st->time_base.num;
if (fps)
zm_log_fps(av_q2d(st->avg_frame_rate), "fps");
if (tbn)
zm_log_fps(1 / av_q2d(st->time_base), "stream tb numerator");
} else if (codec->codec_type == AVMEDIA_TYPE_AUDIO) {
#if LIBAVUTIL_VERSION_CHECK(57, 28, 100, 28, 0)
Debug(1, "profile %d channels %d sample_rate %d",
codec->profile, codec->ch_layout.nb_channels, codec->sample_rate);
#else
Debug(1, "profile %d channels %d sample_rate %d",
codec->profile, codec->channels, codec->sample_rate);
#endif
} else {
Debug(1, "Unknown codec type %d", codec->codec_type);
}
if (st->disposition & AV_DISPOSITION_DEFAULT)
Debug(1, " (default)");
if (st->disposition & AV_DISPOSITION_DUB)
Debug(1, " (dub)");
if (st->disposition & AV_DISPOSITION_ORIGINAL)
Debug(1, " (original)");
if (st->disposition & AV_DISPOSITION_COMMENT)
Debug(1, " (comment)");
if (st->disposition & AV_DISPOSITION_LYRICS)
Debug(1, " (lyrics)");
if (st->disposition & AV_DISPOSITION_KARAOKE)
Debug(1, " (karaoke)");
if (st->disposition & AV_DISPOSITION_FORCED)
Debug(1, " (forced)");
if (st->disposition & AV_DISPOSITION_HEARING_IMPAIRED)
Debug(1, " (hearing impaired)");
if (st->disposition & AV_DISPOSITION_VISUAL_IMPAIRED)
Debug(1, " (visual impaired)");
if (st->disposition & AV_DISPOSITION_CLEAN_EFFECTS)
Debug(1, " (clean effects)");
//dump_metadata(NULL, st->metadata, " ");
//dump_sidedata(NULL, st, " ");
}
int check_sample_fmt(const AVCodec *codec, enum AVSampleFormat sample_fmt) {
const enum AVSampleFormat *p = codec->sample_fmts;
while (*p != AV_SAMPLE_FMT_NONE) {
if (*p == sample_fmt)
return 1;
else Debug(2, "Not %s", av_get_sample_fmt_name( *p ) );
p++;
}
return 0;
}
enum AVPixelFormat fix_deprecated_pix_fmt(enum AVPixelFormat fmt) {
// Fix deprecated formats
switch ( fmt ) {
case AV_PIX_FMT_YUVJ422P :
return AV_PIX_FMT_YUV422P;
case AV_PIX_FMT_YUVJ444P :
return AV_PIX_FMT_YUV444P;
case AV_PIX_FMT_YUVJ440P :
return AV_PIX_FMT_YUV440P;
case AV_PIX_FMT_NONE :
case AV_PIX_FMT_YUVJ420P :
return AV_PIX_FMT_YUV420P;
default:
return fmt;
}
}
bool is_video_stream(const AVStream * stream) {
if (stream->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
return true;
}
Debug(2, "Not a video type %d != %d", stream->codecpar->codec_type, AVMEDIA_TYPE_VIDEO);
return false;
}
bool is_video_context(const AVCodecContext *codec_context) {
return codec_context->codec_type == AVMEDIA_TYPE_VIDEO;
}
bool is_audio_stream(const AVStream *stream) {
return stream->codecpar->codec_type == AVMEDIA_TYPE_AUDIO;
}
bool is_audio_context(const AVCodecContext *codec_context) {
return codec_context->codec_type == AVMEDIA_TYPE_AUDIO;
}
int zm_receive_packet(AVCodecContext *context, AVPacket &packet) {
int ret = avcodec_receive_packet(context, &packet);
if ((ret < 0) and (AVERROR_EOF != ret)) {
Error("Error encoding (%d) (%s)", ret, av_err2str(ret));
}
return ret; // 1 or 0
} // end int zm_receive_packet(AVCodecContext *context, AVPacket &packet)
int zm_send_packet_receive_frame(AVCodecContext *context, AVFrame *frame, AVPacket &packet) {
int pkt_ret, frm_ret;
pkt_ret = avcodec_send_packet(context, &packet);
frm_ret = avcodec_receive_frame(context, frame);
if (pkt_ret == 0 && frm_ret == 0) {
// In this api the packet is always consumed, so return packet.bytes
return packet.size;
} else if (pkt_ret != 0 && pkt_ret != AVERROR(EAGAIN)) {
Error("Could not send packet (error %d = %s)", pkt_ret,
av_make_error_string(pkt_ret).c_str());
return pkt_ret;
} else if (frm_ret != 0 && frm_ret != AVERROR(EAGAIN)) {
Error("Could not receive frame (error %d = %s)", frm_ret,
av_make_error_string(frm_ret).c_str());
return frm_ret;
}
return 0;
} // end int zm_send_packet_receive_frame(AVCodecContext *context, AVFrame *frame, AVPacket &packet)
/* Returns < 0 on error, 0 if codec not ready, 1 on success
*/
int zm_send_frame_receive_packet(AVCodecContext *ctx, AVFrame *frame, AVPacket &packet) {
int frm_ret, pkt_ret;
frm_ret = avcodec_send_frame(ctx, frame);
pkt_ret = avcodec_receive_packet(ctx, &packet);
if (frm_ret != 0 && frame) {
Error("Could not send frame (error '%s')",
av_make_error_string(frm_ret).c_str());
return frm_ret;
} else if (pkt_ret != 0) {
if (pkt_ret == AVERROR(EAGAIN)) {
// The codec may need more samples than it has, perfectly valid
Debug(2, "Codec not ready to give us a packet");
return 0;
} else if (frame) {
// May get EOF if frame is NULL because it signals flushing
Error("Could not receive packet (error %d = '%s')", pkt_ret,
av_make_error_string(pkt_ret).c_str());
}
zm_av_packet_unref(&packet);
return pkt_ret;
}
return 1;
} // end int zm_send_frame_receive_packet
void zm_free_codec(AVCodecContext **ctx) {
if (*ctx) {
avcodec_close(*ctx);
// We allocate and copy in newer ffmpeg, so need to free it
avcodec_free_context(ctx);
*ctx = nullptr;
}
}
void zm_packet_copy_rescale_ts(const AVPacket *ipkt, AVPacket *opkt, const AVRational src_tb, const AVRational dst_tb) {
opkt->pts = ipkt->pts;
opkt->dts = ipkt->dts;
opkt->duration = ipkt->duration;
av_packet_rescale_ts(opkt, src_tb, dst_tb);
}
int zm_resample_audio(SwrContext *resample_ctx, AVFrame *in_frame, AVFrame *out_frame) {
if (in_frame) {
// Resample the in_frame into the audioSampleBuffer until we process the whole
// decoded data. Note: pts does not survive resampling or converting
Debug(2, "Converting %d to %d samples using swresample",
in_frame->nb_samples, out_frame->nb_samples);
} else {
Debug(2, "Sending NULL frame to flush resampler");
}
int ret = swr_convert_frame(resample_ctx, out_frame, in_frame);
if (ret < 0) {
Error("Could not resample frame (error '%s')",
av_make_error_string(ret).c_str());
return 0;
}
Debug(3, "swr_get_delay %" PRIi64, swr_get_delay(resample_ctx, out_frame->sample_rate));
zm_dump_frame(out_frame, "Out frame after resample");
return 1;
}
int zm_resample_get_delay(SwrContext *resample_ctx, int time_base) {
return swr_get_delay(resample_ctx, time_base);
}
int zm_add_samples_to_fifo(AVAudioFifo *fifo, AVFrame *frame) {
int ret = av_audio_fifo_realloc(fifo, av_audio_fifo_size(fifo) + frame->nb_samples);
if (ret < 0) {
Error("Could not reallocate FIFO to %d samples",
av_audio_fifo_size(fifo) + frame->nb_samples);
return 0;
}
/** Store the new samples in the FIFO buffer. */
ret = av_audio_fifo_write(fifo, (void **)frame->data, frame->nb_samples);
if (ret < frame->nb_samples) {
Error("Could not write data to FIFO. %d written, expecting %d. Reason %s",
ret, frame->nb_samples, av_make_error_string(ret).c_str());
return 0;
}
return 1;
}
int zm_get_samples_from_fifo(AVAudioFifo *fifo, AVFrame *frame) {
// AAC requires 1024 samples per encode. Our input tends to be something else, so need to buffer them.
if (frame->nb_samples > av_audio_fifo_size(fifo)) {
Debug(1, "Not enough samples in fifo for AAC codec frame_size %d > fifo size %d",
frame->nb_samples, av_audio_fifo_size(fifo));
return 0;
}
if (av_audio_fifo_read(fifo, (void **)frame->data, frame->nb_samples) < frame->nb_samples) {
Error("Could not read data from FIFO");
return 0;
}
//out_frame->nb_samples = frame_size;
zm_dump_frame(frame, "Out frame after fifo read");
return 1;
}