Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
To: ffmpeg-devel@ffmpeg.org
Subject: Re: [FFmpeg-devel] [PATCH 4/5] lavc/bitstream: make skip_remaining() public
Date: Mon, 20 Jun 2022 13:39:13 +0200
Message-ID: <DB6PR0101MB22149C551B3C2AF9D37B0BCE8FB09@DB6PR0101MB2214.eurprd01.prod.exchangelabs.com> (raw)
In-Reply-To: <20220617133206.23643-4-anton@khirnov.net>

Anton Khirnov:
> Also rename it to bitstream_skip_cache(), which is more descriptive and
> follows the naming conventions of tis API.
> ---
>  libavcodec/bitstream.h          |  1 +
>  libavcodec/bitstream_template.h | 19 ++++++++++++-------
>  2 files changed, 13 insertions(+), 7 deletions(-)
> 
> diff --git a/libavcodec/bitstream.h b/libavcodec/bitstream.h
> index 3fa63695d3..364245453b 100644
> --- a/libavcodec/bitstream.h
> +++ b/libavcodec/bitstream.h
> @@ -116,6 +116,7 @@ static inline void bitstream_unget(BitstreamContext *bc, uint64_t value,
>  #define bitstream_peek          bitstream_peek_be
>  #define bitstream_peek_signed   bitstream_peek_signed_be
>  #define bitstream_skip          bitstream_skip_be
> +#define bitstream_skip_cache    bitstream_skip_cache_be
>  #define bitstream_seek          bitstream_seek_be
>  #define bitstream_align         bitstream_align_be
>  #define bitstream_read_xbits    bitstream_read_xbits_be
> diff --git a/libavcodec/bitstream_template.h b/libavcodec/bitstream_template.h
> index 14a420139c..717473ca40 100644
> --- a/libavcodec/bitstream_template.h
> +++ b/libavcodec/bitstream_template.h
> @@ -225,7 +225,12 @@ static inline int BS_FUNC(bitstream_peek_signed)(BitstreamContext *bc, unsigned
>      return sign_extend(BS_FUNC(bitstream_peek)(bc, n), n);
>  }
>  
> -static inline void BS_FUNC(skip_remaining)(BitstreamContext *bc, unsigned int n)
> +/**
> + * Skip n bits from the cache. This may only be called if at least n bits are
> + * guaranteed to be in the cache, e.g. right after bitstream_peek(n).
> + * Otherwise use bitstream_skip().
> + */
> +static inline void BS_FUNC(bitstream_skip_cache)(BitstreamContext *bc, unsigned int n)
>  {
>  #ifdef BITSTREAM_LE
>      bc->bits >>= n;
> @@ -241,7 +246,7 @@ static inline void BS_FUNC(skip_remaining)(BitstreamContext *bc, unsigned int n)
>  static inline void BS_FUNC(bitstream_skip)(BitstreamContext *bc, unsigned int n)
>  {
>      if (n < bc->bits_left)
> -        BS_FUNC(skip_remaining)(bc, n);
> +        BS_FUNC(bitstream_skip_cache)(bc, n);
>      else {
>          n -= bc->bits_left;
>          bc->bits      = 0;
> @@ -255,7 +260,7 @@ static inline void BS_FUNC(bitstream_skip)(BitstreamContext *bc, unsigned int n)
>          }
>          BS_FUNC(refill_64)(bc);
>          if (n)
> -            BS_FUNC(skip_remaining)(bc, n);
> +            BS_FUNC(bitstream_skip_cache)(bc, n);
>      }
>  }
>  
> @@ -291,7 +296,7 @@ static inline int BS_FUNC(bitstream_read_xbits)(BitstreamContext *bc, unsigned i
>  {
>      int32_t cache = BS_FUNC(bitstream_peek)(bc, 32);
>      int sign = ~cache >> 31;
> -    BS_FUNC(skip_remaining)(bc, n);
> +    BS_FUNC(bitstream_skip_cache)(bc, n);
>  
>      return ((((uint32_t)(sign ^ cache)) >> (32 - n)) ^ sign) - sign;
>  }
> @@ -374,14 +379,14 @@ static inline int BS_FUNC(bitstream_read_vlc)(BitstreamContext *bc, VLC_TYPE (*t
>      int n        = table[idx][1];
>  
>      if (max_depth > 1 && n < 0) {
> -        BS_FUNC(skip_remaining)(bc, bits);
> +        BS_FUNC(bitstream_skip_cache)(bc, bits);
>          code = BS_FUNC(set_idx)(bc, code, &n, &nb_bits, table);
>          if (max_depth > 2 && n < 0) {
> -            BS_FUNC(skip_remaining)(bc, nb_bits);
> +            BS_FUNC(bitstream_skip_cache)(bc, nb_bits);
>              code = BS_FUNC(set_idx)(bc, code, &n, &nb_bits, table);
>          }
>      }
> -    BS_FUNC(skip_remaining)(bc, n);
> +    BS_FUNC(bitstream_skip_cache)(bc, n);
>  
>      return code;
>  }

How do you intend to avoid the issue described here:
https://ffmpeg.org/pipermail/ffmpeg-devel/2022-June/297850.html?
Bitstream_peek() does not guarantee that there are bits bits in the
cache even though your documentation says otherwise.

- Andreas
_______________________________________________
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".

  reply	other threads:[~2022-06-20 11:39 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-17 13:32 [FFmpeg-devel] [PATCH 1/5] get_bits: move check_marker() to mpegvideodec.h Anton Khirnov
2022-06-17 13:32 ` [FFmpeg-devel] [PATCH 2/5] lavc: add standalone cached bitstream reader Anton Khirnov
2022-06-17 15:00   ` Paul B Mahol
2022-06-17 13:32 ` [FFmpeg-devel] [PATCH 3/5] lavc/bitstream: templatize for BE/LE Anton Khirnov
2022-06-17 13:32 ` [FFmpeg-devel] [PATCH 4/5] lavc/bitstream: make skip_remaining() public Anton Khirnov
2022-06-20 11:39   ` Andreas Rheinhardt [this message]
2022-06-17 13:32 ` [FFmpeg-devel] [PATCH 5/5] lavc/get_bits: add a compat wrapper for the cached bitstream reader Anton Khirnov
2022-06-17 15:00   ` Paul B Mahol
2022-06-17 15:00     ` James Almer
2022-06-17 15:26       ` Paul B Mahol
2022-06-18 14:31         ` Jean-Baptiste Kempf
2022-06-18 15:40           ` Paul B Mahol
2022-06-23 12:01           ` Anton Khirnov
2022-06-23 11:57         ` Anton Khirnov

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=DB6PR0101MB22149C551B3C2AF9D37B0BCE8FB09@DB6PR0101MB2214.eurprd01.prod.exchangelabs.com \
    --to=andreas.rheinhardt@outlook.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