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 
24 #include "bsf.h"
25 #include "bsf_internal.h"
26 #include "get_bits.h"
27 
28 #define MAX_CACHE 8
29 typedef struct VP9BSFContext {
30  int n_cache;
33 
34 static void stats(AVPacket * const *in, int n_in,
35  unsigned *_max, unsigned *_sum)
36 {
37  int n;
38  unsigned max = 0, sum = 0;
39 
40  for (n = 0; n < n_in; n++) {
41  unsigned sz = in[n]->size;
42 
43  if (sz > max)
44  max = sz;
45  sum += sz;
46  }
47 
48  *_max = max;
49  *_sum = sum;
50 }
51 
52 static int merge_superframe(AVPacket * const *in, int n_in, AVPacket *out)
53 {
54  unsigned max, sum, mag, marker, n, sz;
55  uint8_t *ptr;
56  int res;
57 
58  stats(in, n_in, &max, &sum);
59  mag = av_log2(max) >> 3;
60  marker = 0xC0 + (mag << 3) + (n_in - 1);
61  sz = sum + 2 + (mag + 1) * n_in;
62  res = av_new_packet(out, sz);
63  if (res < 0)
64  return res;
65  ptr = out->data;
66  for (n = 0; n < n_in; n++) {
67  memcpy(ptr, in[n]->data, in[n]->size);
68  ptr += in[n]->size;
69  }
70 
71 #define wloop(mag, wr) \
72  do { \
73  for (n = 0; n < n_in; n++) { \
74  wr; \
75  ptr += mag + 1; \
76  } \
77  } while (0)
78 
79  // write superframe with marker 110[mag:2][nframes:3]
80  *ptr++ = marker;
81  switch (mag) {
82  case 0:
83  wloop(mag, *ptr = in[n]->size);
84  break;
85  case 1:
86  wloop(mag, AV_WL16(ptr, in[n]->size));
87  break;
88  case 2:
89  wloop(mag, AV_WL24(ptr, in[n]->size));
90  break;
91  case 3:
92  wloop(mag, AV_WL32(ptr, in[n]->size));
93  break;
94  }
95  *ptr++ = marker;
96  av_assert0(ptr == &out->data[out->size]);
97 
98  return 0;
99 }
100 
102 {
103  GetBitContext gb;
104  VP9BSFContext *s = ctx->priv_data;
105  int res, invisible, profile, marker, uses_superframe_syntax = 0, n;
106 
107  res = ff_bsf_get_packet_ref(ctx, pkt);
108  if (res < 0)
109  return res;
110 
111  marker = pkt->data[pkt->size - 1];
112  if ((marker & 0xe0) == 0xc0) {
113  int nbytes = 1 + ((marker >> 3) & 0x3);
114  int n_frames = 1 + (marker & 0x7), idx_sz = 2 + n_frames * nbytes;
115 
116  uses_superframe_syntax = pkt->size >= idx_sz && pkt->data[pkt->size - idx_sz] == marker;
117  }
118 
119  if ((res = init_get_bits8(&gb, pkt->data, pkt->size)) < 0)
120  goto done;
121 
122  get_bits(&gb, 2); // frame marker
123  profile = get_bits1(&gb);
124  profile |= get_bits1(&gb) << 1;
125  if (profile == 3) profile += get_bits1(&gb);
126 
127  if (get_bits1(&gb)) {
128  invisible = 0;
129  } else {
130  get_bits1(&gb); // keyframe
131  invisible = !get_bits1(&gb);
132  }
133 
134  if (uses_superframe_syntax && s->n_cache > 0) {
135  av_log(ctx, AV_LOG_ERROR,
136  "Mixing of superframe syntax and naked VP9 frames not supported\n");
137  res = AVERROR(ENOSYS);
138  goto done;
139  } else if ((!invisible || uses_superframe_syntax) && !s->n_cache) {
140  // passthrough
141  return 0;
142  } else if (s->n_cache + 1 >= MAX_CACHE) {
143  av_log(ctx, AV_LOG_ERROR,
144  "Too many invisible frames\n");
145  res = AVERROR_INVALIDDATA;
146  goto done;
147  }
148 
149  av_packet_move_ref(s->cache[s->n_cache++], pkt);
150 
151  if (invisible) {
152  return AVERROR(EAGAIN);
153  }
154  av_assert0(s->n_cache > 0);
155 
156  // build superframe
157  if ((res = merge_superframe(s->cache, s->n_cache, pkt)) < 0)
158  goto done;
159 
160  res = av_packet_copy_props(pkt, s->cache[s->n_cache - 1]);
161  if (res < 0)
162  goto done;
163 
164  for (n = 0; n < s->n_cache; n++)
165  av_packet_unref(s->cache[n]);
166  s->n_cache = 0;
167 
168 done:
169  if (res < 0)
170  av_packet_unref(pkt);
171  return res;
172 }
173 
175 {
176  VP9BSFContext *s = ctx->priv_data;
177  int n;
178 
179  // alloc cache packets
180  for (n = 0; n < MAX_CACHE; n++) {
181  s->cache[n] = av_packet_alloc();
182  if (!s->cache[n])
183  return AVERROR(ENOMEM);
184  }
185 
186  return 0;
187 }
188 
190 {
191  VP9BSFContext *s = ctx->priv_data;
192  int n;
193 
194  // unref cached data
195  for (n = 0; n < s->n_cache; n++)
196  av_packet_unref(s->cache[n]);
197  s->n_cache = 0;
198 }
199 
201 {
202  VP9BSFContext *s = ctx->priv_data;
203  int n;
204 
205  // free cached data
206  for (n = 0; n < MAX_CACHE; n++)
207  av_packet_free(&s->cache[n]);
208 }
209 
210 static const enum AVCodecID codec_ids[] = {
212 };
213 
215  .name = "vp9_superframe",
216  .priv_data_size = sizeof(VP9BSFContext),
220  .close = vp9_superframe_close,
221  .codec_ids = codec_ids,
222 };
#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: bsf.h:49
int size
Definition: packet.h:364
int av_log2(unsigned v)
Definition: intmath.c:26
static AVPacket pkt
void * priv_data
Opaque filter-specific private data.
Definition: bsf.h:70
#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:64
uint8_t
const char * name
Definition: bsf.h:99
uint8_t * data
Definition: packet.h:363
void av_packet_move_ref(AVPacket *dst, AVPacket *src)
Move every field in src to dst and reset src.
Definition: avpacket.c:664
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:88
AVCodecID
Identify the syntax and semantics of the bitstream.
Definition: codec_id.h:46
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:194
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:572
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
static int merge_superframe(AVPacket *const *in, int n_in, AVPacket *out)
#define wloop(mag, wr)
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:606
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:45
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:53
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: packet.h:340
int ff_bsf_get_packet_ref(AVBSFContext *ctx, AVPacket *pkt)
Called by bitstream filters to get packet for filtering.
Definition: bsf.c:252
#define AV_WL32(p, v)
Definition: intreadwrite.h:426