Re: [FFmpeg-devel] [PATCH 5/5] avformat/pcm: Use 64bit in bitrate computation

From: Michael Niedermayer <michael@niedermayer.cc>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH 5/5] avformat/pcm: Use 64bit in bitrate computation
Date: Thu, 4 Apr 2024 20:41:59 +0200
Message-ID: <20240404184159.GA6420@pb2> (raw)
In-Reply-To: <956d190b-ccde-74af-aebc-77667f9108ca@passwd.hu>

[-- Attachment #1.1: Type: text/plain, Size: 1862 bytes --]

On Thu, Apr 04, 2024 at 07:12:40PM +0200, Marton Balint wrote:
> 
> 
> On Thu, 4 Apr 2024, Michael Niedermayer wrote:
> 
> > Fixes: signed integer overflow: 65792 * 65312 cannot be represented in type 'int'
> > Fixes: 67819/clusterfuzz-testcase-minimized-ffmpeg_dem_WADY_fuzzer-5236100912185344
> > 
> > Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
> > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
> > ---
> > libavformat/pcm.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/libavformat/pcm.c b/libavformat/pcm.c
> > index 051e86dd464..a774dbc3726 100644
> > --- a/libavformat/pcm.c
> > +++ b/libavformat/pcm.c
> > @@ -41,7 +41,7 @@ int ff_pcm_default_packet_size(AVCodecParameters *par)
> >     /* Don't trust the codecpar bitrate if we can calculate it ourselves */
> >     if (bits_per_sample > 0 && par->sample_rate > 0 && par->ch_layout.nb_channels > 0)
> >         if ((int64_t)par->sample_rate * par->ch_layout.nb_channels < INT64_MAX / bits_per_sample)
> > -            bitrate = bits_per_sample * par->sample_rate * par->ch_layout.nb_channels;
> > +            bitrate = bits_per_sample * (int64_t)par->sample_rate * par->ch_layout.nb_channels;
> 
> LGTM, thanks.
> 
> I wonder why we usually cast the second operand and not the first to 64 bit,
> since cast has higher precedence than multiplication, it should not matter,
> should it?

If the reader isnt sure about the precedence, writing it as
a * (int64_t)b
has the advantage that precedence doesnt matter, so the code is easier to
understand and verify than

(int64_t)a * b

thx

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

In a rich man's house there is no place to spit but his face.
-- Diogenes of Sinope

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

[-- Attachment #2: Type: text/plain, Size: 251 bytes --]

_______________________________________________
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".