From: hwren <hwrenx@126.com> To: "FFmpeg development discussions and patches" <ffmpeg-devel@ffmpeg.org> Subject: Re: [FFmpeg-devel] [PATCH v3 3/5] avcodec/avs2_parser: parse more info Date: Tue, 21 Jun 2022 16:57:06 +0800 (CST) Message-ID: <41cafcda.58f8.181857acb92.Coremail.hwrenx@126.com> (raw) In-Reply-To: <tencent_9C35ED800A345D3E7957439CEAE09C5A5209@qq.com> 在 2022-06-21 16:28:37,"\"zhilizhao(赵志立)\"" <quinkblack@foxmail.com> 写道: > > >> On Jun 21, 2022, at 4:25 PM, zhilizhao(赵志立) <quinkblack@foxmail.com> wrote: >> >>> On Jun 21, 2022, at 10:45 AM, hwren <hwrenx@126.com> wrote: >>> >>> At 2022-06-13 11:36:34, "Zhao Zhili" <quinkblack@foxmail.com> wrote: >>>> Including video resolution, framerate and picture type, etc. >>>> >>>> Signed-off-by: Zhao Zhili <zhilizhao@tencent.com> >>>> --- >>>> libavcodec/Makefile | 2 +- >>>> libavcodec/avs2.c | 42 ++++++++++++++++ >>>> libavcodec/avs2.h | 10 ++++ >>>> libavcodec/avs2_parser.c | 105 +++++++++++++++++++++++++++++++++++++++ >>>> 4 files changed, 158 insertions(+), 1 deletion(-) >>>> create mode 100644 libavcodec/avs2.c >>>> >>>> diff --git a/libavcodec/Makefile b/libavcodec/Makefile >>>> index 3b8f7b5e01..6cc6f8437c 100644 >>>> --- a/libavcodec/Makefile >>>> +++ b/libavcodec/Makefile >>>> @@ -1114,7 +1114,7 @@ OBJS-$(CONFIG_AC3_PARSER) += aac_ac3_parser.o ac3tab.o \ >>>> OBJS-$(CONFIG_ADX_PARSER) += adx_parser.o adx.o >>>> OBJS-$(CONFIG_AMR_PARSER) += amr_parser.o >>>> OBJS-$(CONFIG_AV1_PARSER) += av1_parser.o >>>> -OBJS-$(CONFIG_AVS2_PARSER) += avs2_parser.o >>>> +OBJS-$(CONFIG_AVS2_PARSER) += avs2.o avs2_parser.o >>>> OBJS-$(CONFIG_AVS3_PARSER) += avs3_parser.o >>>> OBJS-$(CONFIG_BMP_PARSER) += bmp_parser.o >>>> OBJS-$(CONFIG_CAVSVIDEO_PARSER) += cavs_parser.o >>>> diff --git a/libavcodec/avs2.c b/libavcodec/avs2.c >>>> new file mode 100644 >>>> index 0000000000..ead8687d0a >>>> --- /dev/null >>>> +++ b/libavcodec/avs2.c >>>> @@ -0,0 +1,42 @@ >>>> +/* >>>> + * AVS2 related definitions >>>> + * >>>> + * Copyright (C) 2022 Zhao Zhili, <zhilizhao@tencent.com> >>>> + * >>>> + * 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 "avs2.h" >>>> + >>>> +const AVRational ff_avs2_frame_rate_tab[16] = { >>>> + { 0 , 0 }, // forbid >>>> + { 24000, 1001}, >>>> + { 24 , 1 }, >>>> + { 25 , 1 }, >>>> + { 30000, 1001}, >>>> + { 30 , 1 }, >>>> + { 50 , 1 }, >>>> + { 60000, 1001}, >>>> + { 60 , 1 }, >>>> + { 100 , 1 }, >>>> + { 120 , 1 }, >>>> + { 200 , 1 }, >>>> + { 240 , 1 }, >>>> + { 300 , 1 }, >>>> + { 0 , 0 }, // reserved >>>> + { 0 , 0 } // reserved >>>> +}; >>> >>> That table seems to appear only in avs2 decoder. If so, they are supposed to be declared in libdavs2.c. >>> If not, please point out the usage of this separate source file. >> >> It’s used in this patch, parse_avs2_seq_header(). LGTM if it is used in multiple files and need to be optimized to reduce binary size. >> >>> >>> Persionally, the frame_rate table can be used in both encoder and decoder. So it is what should be >>> defined in the avs2.h (with related enc/dec changes). >> >> It’s declared in avs2.h, so it can be used by enc/dec. >> It’s defined in .c to reduce binary size. >> >>> >>>> diff --git a/libavcodec/avs2.h b/libavcodec/avs2.h >>>> index f342ba52a0..544cf502d7 100644 >>>> --- a/libavcodec/avs2.h >>>> +++ b/libavcodec/avs2.h >>>> @@ -23,6 +23,8 @@ >>>> #ifndef AVCODEC_AVS2_H >>>> #define AVCODEC_AVS2_H >>>> >>>> +#include "libavutil/rational.h" >>>> + >>>> #define AVS2_SLICE_MAX_START_CODE 0x000001AF >>>> >>>> enum { >>>> @@ -38,4 +40,12 @@ enum { >>>> #define AVS2_ISPIC(x) ((x) == AVS2_INTRA_PIC_START_CODE || (x) == AVS2_INTER_PIC_START_CODE) >>>> #define AVS2_ISUNIT(x) ((x) == AVS2_SEQ_START_CODE || AVS2_ISPIC(x)) >>>> >>> >>> Same as above. Mentioned definitions now are only used in avs2_parser.c. If so, they are supposed to be >>> kept in avs2_parser.c. >> >> It’s not the same as ff_avs2_frame_rate_tab. Firstly, define enum >> and macros doesn’t generate binary. Secondly, they are be used >> by avs2 parser/decoder/encoder and so on, the same as avs3.h. > >I mean they ‘can’ be used by avs2 parser/decoder/encoder. LGTM > >> >> They are only used by avs2 parser now, but they are handy when debug >> decoder, for example, and future proof. >> >>> >>> >>>> +enum AVS2Profile { >>>> + AVS2_PROFILE_MAIN_PIC = 0x12, >>>> + AVS2_PROFILE_MAIN = 0x20, >>>> + AVS2_PROFILE_MAIN10 = 0x22, >>>> +}; >>> >>> Not mentioned elsewhere. If they are to work in the encoder/decoder, please package them with related changes. >> >> AVS2_PROFILE_MAIN10 is used in this patch. Got it. No more questions. >> >>> >>>> + >>>> +extern const AVRational ff_avs2_frame_rate_tab[16]; >>>> + >>>> #endif >>>> diff --git a/libavcodec/avs2_parser.c b/libavcodec/avs2_parser.c >>>> index 71cf442903..0350517493 100644 >>>> --- a/libavcodec/avs2_parser.c >>>> +++ b/libavcodec/avs2_parser.c >>>> @@ -19,7 +19,9 @@ >>>> * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA >>>> */ >>>> >>>> +#include "libavutil/avutil.h" >>>> #include "avs2.h" >>>> +#include "get_bits.h" >>>> #include "parser.h" >>>> >>>> static int avs2_find_frame_end(ParseContext *pc, const uint8_t *buf, int buf_size) >>>> @@ -58,6 +60,107 @@ static int avs2_find_frame_end(ParseContext *pc, const uint8_t *buf, int buf_siz >>>> return END_NOT_FOUND; >>>> } >>>> >>>> +static void parse_avs2_seq_header(AVCodecParserContext *s, const uint8_t *buf, >>>> + int buf_size, AVCodecContext *avctx) >>>> +{ >>>> + GetBitContext gb; >>>> + int profile, level; >>>> + int width, height; >>>> + int chroma, sample_precision, encoding_precision = 1; >>>> + // sample_precision and encoding_precision is 3 bits >>>> + static const uint8_t precision[8] = { 0, 8, 10 }; >>>> + unsigned aspect_ratio; >>>> + unsigned frame_rate_code; >>>> + int low_delay; >>>> + // update buf_size_min if parse more deeper >>>> + const int buf_size_min = 15; >>>> + >>>> + if (buf_size < buf_size_min) >>>> + return; >>>> + >>>> + init_get_bits8(&gb, buf, buf_size_min); >>>> + >>>> + s->key_frame = 1; >>>> + s->pict_type = AV_PICTURE_TYPE_I; >>>> + >>>> + profile = get_bits(&gb, 8); >>>> + level = get_bits(&gb, 8); >>>> + >>>> + // progressive_sequence u(1) >>>> + // field_coded_sequence u(1) >>>> + skip_bits(&gb, 2); >>>> + >>>> + width = get_bits(&gb, 14); >>>> + height = get_bits(&gb, 14); >>>> + >>>> + chroma = get_bits(&gb, 2); >>>> + sample_precision = get_bits(&gb, 3); >>>> + if (profile == AVS2_PROFILE_MAIN10) >>>> + encoding_precision = get_bits(&gb, 3); >>>> + >>>> + aspect_ratio = get_bits(&gb, 4); >>>> + frame_rate_code = get_bits(&gb, 4); >>>> + >>>> + // bit_rate_lower u(18) >>>> + // marker_bit f(1) >>>> + // bit_rate_upper u(12) >>>> + skip_bits(&gb, 31); >>>> + >>>> + low_delay = get_bits(&gb, 1); >>>> + >>>> + s->width = width; >>>> + s->height = height; >>>> + s->coded_width = FFALIGN(width, 8); >>>> + s->coded_height = FFALIGN(height, 8); >>>> + avctx->framerate.num = avctx->time_base.den = >>>> + ff_avs2_frame_rate_tab[frame_rate_code].num; >>>> + avctx->framerate.den = avctx->time_base.num = >>>> + ff_avs2_frame_rate_tab[frame_rate_code].den; >>>> + avctx->has_b_frames = FFMAX(avctx->has_b_frames, !low_delay); >>>> + >>>> + av_log(avctx, AV_LOG_DEBUG, >>>> + "AVS2 parse seq HDR: profile %x, level %x, " >>>> + "width %d, height %d, " >>>> + "chroma %d, sample_precision %d bits, encoding_precision %d bits, " >>>> + "aspect_ratio 0x%x, framerate %d/%d, low_delay %d\n", >>>> + profile, level, >>>> + width, height, >>>> + chroma, precision[sample_precision], precision[encoding_precision], >>>> + aspect_ratio, avctx->framerate.num, avctx->framerate.den, low_delay); >>>> +} >>>> + >>>> +static void parse_avs2_units(AVCodecParserContext *s, const uint8_t *buf, >>>> + int buf_size, AVCodecContext *avctx) >>>> +{ >>>> + if (buf_size < 5) >>>> + return; >>>> + >>>> + if (!(buf[0] == 0x0 && buf[1] == 0x0 && buf[2] == 0x1)) >>>> + return; >>>> + >>>> + switch (buf[3]) { >>>> + case AVS2_SEQ_START_CODE: >>>> + parse_avs2_seq_header(s, buf + 4, buf_size - 4, avctx); >>>> + return; >>>> + case AVS2_INTRA_PIC_START_CODE: >>>> + s->key_frame = 1; >>>> + s->pict_type = AV_PICTURE_TYPE_I; >>>> + return; >>>> + case AVS2_INTER_PIC_START_CODE: >>>> + s->key_frame = 0; >>>> + if (buf_size > 9) { >>>> + int pic_code_type = buf[8] & 0x3; >>>> + if (pic_code_type == 1) >>>> + s->pict_type = AV_PICTURE_TYPE_P; >>>> + else if (pic_code_type == 3) >>>> + s->pict_type = AV_PICTURE_TYPE_S; >>>> + else >>>> + s->pict_type = AV_PICTURE_TYPE_B; >>>> + } >>>> + return; >>>> + } >>>> +} >>>> + >>>> static int avs2_parse(AVCodecParserContext *s, AVCodecContext *avctx, >>>> const uint8_t **poutbuf, int *poutbuf_size, >>>> const uint8_t *buf, int buf_size) >>>> @@ -76,6 +179,8 @@ static int avs2_parse(AVCodecParserContext *s, AVCodecContext *avctx, >>>> } >>>> } >>>> >>>> + parse_avs2_units(s, buf, buf_size, avctx); >>>> + >>>> *poutbuf = buf; >>>> *poutbuf_size = buf_size; >>>> >>>> -- >>>> 2.35.3 >>> >>> >>> Best Regards, >>> Huiwen Ren >>> >>>> >>>> _______________________________________________ >>>> ffmpeg-devel mailing list >>>> ffmpeg-devel@ffmpeg.org >>>> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel >>>> >>>> To unsubscribe, visit link above, or email >>>> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >>> >>> >>> _______________________________________________ >>> ffmpeg-devel mailing list >>> ffmpeg-devel@ffmpeg.org >>> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel >>> >>> To unsubscribe, visit link above, or email >>> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
next prev parent reply other threads:[~2022-06-21 8:57 UTC|newest] Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top [not found] <20220613033636.84271-1-quinkblack@foxmail.com> 2022-06-13 3:36 ` [FFmpeg-devel] [PATCH v3 1/5] avcodec/avs2: add AVS2 related definitions Zhao Zhili 2022-06-13 3:36 ` [FFmpeg-devel] [PATCH v3 2/5] avcodec/avs2_parser: split data into frames Zhao Zhili 2022-06-13 3:36 ` [FFmpeg-devel] [PATCH v3 3/5] avcodec/avs2_parser: parse more info Zhao Zhili 2022-06-21 2:45 ` hwren 2022-06-21 8:25 ` "zhilizhao(赵志立)" 2022-06-21 8:28 ` "zhilizhao(赵志立)" 2022-06-21 8:57 ` hwren [this message] 2022-06-13 3:36 ` [FFmpeg-devel] [PATCH v3 4/5] configure: select avs2 parser for libdavs2 decoder Zhao Zhili 2022-06-13 3:36 ` [FFmpeg-devel] [PATCH v3 5/5] avcodec/libdavs2: use frame rate code table Zhao Zhili
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=41cafcda.58f8.181857acb92.Coremail.hwrenx@126.com \ --to=hwrenx@126.com \ --cc=ffmpeg-devel@ffmpeg.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel This inbox may be cloned and mirrored by anyone: git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \ ffmpegdev@gitmailbox.com public-inbox-index ffmpegdev Example config snippet for mirrors. AGPL code for this site: git clone https://public-inbox.org/public-inbox.git