FFmpeg
vp9_superframe_bsf.c
Go to the documentation of this file.
1 /*
2  * Vp9 invisible (alt-ref) frame to superframe merge bitstream filter
3  * Copyright (c) 2016 Ronald S. Bultje <rsbultje@gmail.com>
4  *
5  * This file is part of FFmpeg.
6  *
7  * FFmpeg is free software; you can redistribute it and/or
8  * modify it under the terms of the GNU Lesser General Public
9  * License as published by the Free Software Foundation; either
10  * version 2.1 of the License, or (at your option) any later version.
11  *
12  * FFmpeg is distributed in the hope that it will be useful,
13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15  * Lesser General Public License for more details.
16  *
17  * You should have received a copy of the GNU Lesser General Public
18  * License along with FFmpeg; if not, write to the Free Software
19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20  */
21 
22 #include "libavutil/avassert.h"
23 #include "avcodec.h"
24 #include "bsf.h"
25 #include "get_bits.h"
26 
27 #define MAX_CACHE 8
28 typedef struct VP9BSFContext {
29  int n_cache;
32 
33 static void stats(AVPacket * const *in, int n_in,
34  unsigned *_max, unsigned *_sum)
35 {
36  int n;
37  unsigned max = 0, sum = 0;
38 
39  for (n = 0; n < n_in; n++) {
40  unsigned sz = in[n]->size;
41 
42  if (sz > max)
43  max = sz;
44  sum += sz;
45  }
46 
47  *_max = max;
48  *_sum = sum;
49 }
50 
51 static int merge_superframe(AVPacket * const *in, int n_in, AVPacket *out)
52 {
53  unsigned max, sum, mag, marker, n, sz;
54  uint8_t *ptr;
55  int res;
56 
57  stats(in, n_in, &max, &sum);
58  mag = av_log2(max) >> 3;
59  marker = 0xC0 + (mag << 3) + (n_in - 1);
60  sz = sum + 2 + (mag + 1) * n_in;
61  res = av_new_packet(out, sz);
62  if (res < 0)
63  return res;
64  ptr = out->data;
65  for (n = 0; n < n_in; n++) {
66  memcpy(ptr, in[n]->data, in[n]->size);
67  ptr += in[n]->size;
68  }
69 
70 #define wloop(mag, wr) \
71  do { \
72  for (n = 0; n < n_in; n++) { \
73  wr; \
74  ptr += mag + 1; \
75  } \
76  } while (0)
77 
78  // write superframe with marker 110[mag:2][nframes:3]
79  *ptr++ = marker;
80  switch (mag) {
81  case 0:
82  wloop(mag, *ptr = in[n]->size);
83  break;
84  case 1:
85  wloop(mag, AV_WL16(ptr, in[n]->size));
86  break;
87  case 2:
88  wloop(mag, AV_WL24(ptr, in[n]->size));
89  break;
90  case 3:
91  wloop(mag, AV_WL32(ptr, in[n]->size));
92  break;
93  }
94  *ptr++ = marker;
95  av_assert0(ptr == &out->data[out->size]);
96 
97  return 0;
98 }
99 
101 {
102  GetBitContext gb;
103  VP9BSFContext *s = ctx->priv_data;
104  int res, invisible, profile, marker, uses_superframe_syntax = 0, n;
105 
106  res = ff_bsf_get_packet_ref(ctx, pkt);
107  if (res < 0)
108  return res;
109 
110  marker = pkt->data[pkt->size - 1];
111  if ((marker & 0xe0) == 0xc0) {
112  int nbytes = 1 + ((marker >> 3) & 0x3);
113  int n_frames = 1 + (marker & 0x7), idx_sz = 2 + n_frames * nbytes;
114 
115  uses_superframe_syntax = pkt->size >= idx_sz && pkt->data[pkt->size - idx_sz] == marker;
116  }
117 
118  if ((res = init_get_bits8(&gb, pkt->data, pkt->size)) < 0)
119  goto done;
120 
121  get_bits(&gb, 2); // frame marker
122  profile = get_bits1(&gb);
123  profile |= get_bits1(&gb) << 1;
124  if (profile == 3) profile += get_bits1(&gb);
125 
126  if (get_bits1(&gb)) {
127  invisible = 0;
128  } else {
129  get_bits1(&gb); // keyframe
130  invisible = !get_bits1(&gb);
131  }
132 
133  if (uses_superframe_syntax && s->n_cache > 0) {
134  av_log(ctx, AV_LOG_ERROR,
135  "Mixing of superframe syntax and naked VP9 frames not supported\n");
136  res = AVERROR(ENOSYS);
137  goto done;
138  } else if ((!invisible || uses_superframe_syntax) && !s->n_cache) {
139  // passthrough
140  return 0;
141  } else if (s->n_cache + 1 >= MAX_CACHE) {
142  av_log(ctx, AV_LOG_ERROR,
143  "Too many invisible frames\n");
144  res = AVERROR_INVALIDDATA;
145  goto done;
146  }
147 
148  av_packet_move_ref(s->cache[s->n_cache++], pkt);
149 
150  if (invisible) {
151  return AVERROR(EAGAIN);
152  }
153  av_assert0(s->n_cache > 0);
154 
155  // build superframe
156  if ((res = merge_superframe(s->cache, s->n_cache, pkt)) < 0)
157  goto done;
158 
159  res = av_packet_copy_props(pkt, s->cache[s->n_cache - 1]);
160  if (res < 0)
161  goto done;
162 
163  for (n = 0; n < s->n_cache; n++)
164  av_packet_unref(s->cache[n]);
165  s->n_cache = 0;
166 
167 done:
168  if (res < 0)
169  av_packet_unref(pkt);
170  return res;
171 }
172 
174 {
175  VP9BSFContext *s = ctx->priv_data;
176  int n;
177 
178  // alloc cache packets
179  for (n = 0; n < MAX_CACHE; n++) {
180  s->cache[n] = av_packet_alloc();
181  if (!s->cache[n])
182  return AVERROR(ENOMEM);
183  }
184 
185  return 0;
186 }
187 
189 {
190  VP9BSFContext *s = ctx->priv_data;
191  int n;
192 
193  // unref cached data
194  for (n = 0; n < s->n_cache; n++)
195  av_packet_unref(s->cache[n]);
196  s->n_cache = 0;
197 }
198 
200 {
201  VP9BSFContext *s = ctx->priv_data;
202  int n;
203 
204  // free cached data
205  for (n = 0; n < MAX_CACHE; n++)
206  av_packet_free(&s->cache[n]);
207 }
208 
209 static const enum AVCodecID codec_ids[] = {
211 };
212 
214  .name = "vp9_superframe",
215  .priv_data_size = sizeof(VP9BSFContext),
219  .close = vp9_superframe_close,
220  .codec_ids = codec_ids,
221 };
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
Definition: error.h:59
AVPacket * cache[MAX_CACHE]
ptrdiff_t const GLvoid * data
Definition: opengl_enc.c:100
static void flush(AVCodecContext *avctx)
static unsigned int get_bits(GetBitContext *s, int n)
Read 1-25 bits.
Definition: get_bits.h:379
static void vp9_superframe_flush(AVBSFContext *ctx)
static av_cold int init(AVCodecContext *avctx)
Definition: avrndec.c:35
static int vp9_superframe_filter(AVBSFContext *ctx, AVPacket *pkt)
The bitstream filter state.
Definition: avcodec.h:5774
int size
Definition: avcodec.h:1481
int av_log2(unsigned v)
Definition: intmath.c:26
static AVPacket pkt
void * priv_data
Opaque filter-specific private data.
Definition: avcodec.h:5795
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:37
#define AV_WL24(p, d)
Definition: intreadwrite.h:464
void av_packet_free(AVPacket **pkt)
Free the packet, if the packet is reference counted, it will be unreferenced first.
Definition: avpacket.c:62
uint8_t
const char * name
Definition: avcodec.h:5824
uint8_t * data
Definition: avcodec.h:1480
void av_packet_move_ref(AVPacket *dst, AVPacket *src)
Move every field in src to dst and reset src.
Definition: avpacket.c:655
bitstream reader API header.
#define max(a, b)
Definition: cuda_runtime.h:33
ptrdiff_t size
Definition: opengl_enc.c:100
#define av_log(a,...)
int av_new_packet(AVPacket *pkt, int size)
Allocate the payload of a packet and initialize its fields with default values.
Definition: avpacket.c:86
AVCodecID
Identify the syntax and semantics of the bitstream.
Definition: avcodec.h:215
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:176
filter_frame For filters that do not use the this method is called when a frame is pushed to the filter s input It can be called at any time except in a reentrant way If the input frame is enough to produce then the filter should push the output frames on the output link immediately As an exception to the previous rule if the input frame is enough to produce several output frames then the filter needs output only at least one per link The additional frames can be left buffered in the filter
#define MAX_CACHE
simple assert() macros that are a bit more flexible than ISO C assert().
int av_packet_copy_props(AVPacket *dst, const AVPacket *src)
Copy only "properties" fields from src to dst.
Definition: avpacket.c:565
AVFormatContext * ctx
Definition: movenc.c:48
const AVBitStreamFilter ff_vp9_superframe_bsf
static void stats(AVPacket *const *in, int n_in, unsigned *_max, unsigned *_sum)
#define s(width, name)
Definition: cbs_vp9.c:257
int n
Definition: avisynth_c.h:760
static int merge_superframe(AVPacket *const *in, int n_in, AVPacket *out)
#define wloop(mag, wr)
Libavcodec external API header.
static int init_get_bits8(GetBitContext *s, const uint8_t *buffer, int byte_size)
Initialize GetBitContext.
Definition: get_bits.h:677
void av_packet_unref(AVPacket *pkt)
Wipe the packet.
Definition: avpacket.c:599
static unsigned int get_bits1(GetBitContext *s)
Definition: get_bits.h:498
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(const int16_t *) pi >> 8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(const int32_t *) pi >> 24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(const float *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(const float *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(const float *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(const double *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(const double *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(const double *) pi *(1U<< 31))))#define SET_CONV_FUNC_GROUP(ofmt, ifmt) static void set_generic_function(AudioConvert *ac){}void ff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enum AVSampleFormat out_fmt, enum AVSampleFormat in_fmt, int channels, int sample_rate, int apply_map){AudioConvert *ac;int in_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) return NULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt) > 2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);return NULL;}return ac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}else if(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;else ac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);return ac;}int ff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){int use_generic=1;int len=in->nb_samples;int p;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%d samples - audio_convert: %s to %s (dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));return ff_convert_dither(ac-> in
mfxU16 profile
Definition: qsvenc.c:44
static void vp9_superframe_close(AVBSFContext *ctx)
#define AV_WL16(p, v)
Definition: intreadwrite.h:412
static int vp9_superframe_init(AVBSFContext *ctx)
AVPacket * av_packet_alloc(void)
Allocate an AVPacket and set its fields to default values.
Definition: avpacket.c:51
static enum AVCodecID codec_ids[]
FILE * out
Definition: movenc.c:54
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later.That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another.Frame references ownership and permissions
This structure stores compressed data.
Definition: avcodec.h:1457
int ff_bsf_get_packet_ref(AVBSFContext *ctx, AVPacket *pkt)
Called by bitstream filters to get packet for filtering.
Definition: bsf.c:239
#define AV_WL32(p, v)
Definition: intreadwrite.h:426