Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Xingjing Deng <micro6947-at-gmail.com@ffmpeg.org>
To: Kacper Michajlow <kasper93@gmail.com>
Cc: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH] libavcodec/utils: Ensure allocated buffer is zero-initialized
Date: Sun, 29 Jun 2025 10:25:18 +0800
Message-ID: <CAK+ZN9qCy0Aiy_+Fq0OpJjhggau74hi8S6NhCf2zA6=W0jUXxQ@mail.gmail.com> (raw)
In-Reply-To: <CABPLASTbJ-NDGRN1_WFpUfzVyyFutLX6AOj2vYX3H152n9i+dQ@mail.gmail.com>

 Thank you very much — I think you've convinced me. This was my first time
asking a question in the FFmpeg community, and my limited understanding of
the code led to a false positive. I apologize for the delayed response, and
I truly appreciate your time and help.

Kacper Michajlow <kasper93@gmail.com> 于2025年6月27日周五 02:16写道:

> n Thu, 26 Jun 2025 at 16:07, Zhao Zhili
> <quinkblack-at-foxmail.com@ffmpeg.org> wrote:
> >
> >
> >
> > > On Jun 26, 2025, at 12:07, xjdeng <micro6947-at-gmail.com@ffmpeg.org>
> wrote:
> > >
> > > In `av_fast_padded_mallocz`, the allocated buffer's
> zero-initialization is not guaranteed.
> > > This is because it calls `av_fast_malloc`, which in turn calls
> `fast_malloc` with `zero_realloc=0`.
> > > Consequently, the memory returned by the underlying `av_malloc` (used
> within `fast_malloc`)
> > > is not guaranteed to be zero-initialized.
> > >
> > > Furthermore, if `*size` is adjusted to be greater than `min_size +
> AV_INPUT_BUFFER_PADDING_SIZE`,
> > > the subsequent `memset` operation will not cover the entire allocated
> buffer,
> > > leaving a portion of it uninitialized.
> > >
> > > To ensure the entire allocated buffer is properly zero-initialized, we
> should use `FFMAX`
> > > to adjust the `memset` range.
> >
> > I think memset size is enough. size >= min_size +
> AV_INPUT_BUFFER_PADDING_SIZE.
> >
> > >
> > > Signed-off-by: xjdeng <micro6947@gmail.com>
> > >
> > > ---
> > > libavcodec/utils.c | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/libavcodec/utils.c b/libavcodec/utils.c
> > > index f2686b6863..e2afce71ef 100644
> > > --- a/libavcodec/utils.c
> > > +++ b/libavcodec/utils.c
> > > @@ -72,8 +72,8 @@ void av_fast_padded_mallocz(void *ptr, unsigned int
> *size, size_t min_size)
> > >         return;
> > >     }
> > >     av_fast_malloc(p, size, min_size + AV_INPUT_BUFFER_PADDING_SIZE);
> > > -    if (*p)
> > > -        memset(*p, 0, min_size + AV_INPUT_BUFFER_PADDING_SIZE);
> > > +    if (*p)
> > > +        memset(*p, 0, FFMAX(*size, min_size +
> AV_INPUT_BUFFER_PADDING_SIZE));
> > > }
>
> I think the current code is intentional. In av_fast_malloc() and
> siblings the `*size` is the real allocation size, but your working
> area that you should be using is `
> min_size` and hence why this area (+padding) is zeroed. You shouldn't
> be using anything beyond that, as it is allocated space, but "not
> active" right now.
>
> These functions are meant to be `fast` by not reallocating when there
> is enough space for user payload, and part of the `fast` part is also
> not zeroing the memory area that you as a user don't request in
> `min_size` parameter.
>
> - Kacper
>
_______________________________________________
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:[~2025-06-29  2:25 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-26  4:07 xjdeng
2025-06-26 14:07 ` Zhao Zhili
2025-06-26 18:16   ` Kacper Michajlow
2025-06-29  2:25     ` Xingjing Deng [this message]

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='CAK+ZN9qCy0Aiy_+Fq0OpJjhggau74hi8S6NhCf2zA6=W0jUXxQ@mail.gmail.com' \
    --to=micro6947-at-gmail.com@ffmpeg.org \
    --cc=ffmpeg-devel@ffmpeg.org \
    --cc=kasper93@gmail.com \
    /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