[FFmpeg-devel] [PATCH 1/2] avcodec/speexdec: Check frames_per_packet more completely

[Michael Niedermayer]

On Fri, Oct 15, 2021 at 11:17:04PM +0200, Michael Niedermayer wrote:
> On Fri, Oct 15, 2021 at 07:30:20PM +0200, Paul B Mahol wrote:
> > On Fri, Oct 15, 2021 at 7:26 PM Michael Niedermayer <michael at niedermayer.cc>
> > wrote:
> > 
> > > On Fri, Oct 15, 2021 at 08:52:39AM +0200, Paul B Mahol wrote:
> > > > On Fri, Oct 15, 2021 at 12:20 AM James Almer <jamrial at gmail.com> wrote:
> > > >
> > > > > On 10/14/2021 7:13 PM, Michael Niedermayer wrote:
> > > > > > Fixes: signed integer overflow: 2105344 * 539033345 cannot be
> > > > > represented in type 'int'
> > > > > > Fixes: out of array write
> > > > > > Fixes:
> > > > >
> > > 39956/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_SPEEX_fuzzer-4766419250708480
> > > > > >
> > > > > > Found-by: continuous fuzzing process
> > > > > https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
> > > > > > Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
> > > > > > ---
> > > > > >   libavcodec/speexdec.c | 2 +-
> > > > > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > > > >
> > > > > > diff --git a/libavcodec/speexdec.c b/libavcodec/speexdec.c
> > > > > > index 4c50f54f27b..05f084de2e5 100644
> > > > > > --- a/libavcodec/speexdec.c
> > > > > > +++ b/libavcodec/speexdec.c
> > > > > > @@ -1428,7 +1428,7 @@ static int parse_speex_extradata(AVCodecContext
> > > > > *avctx,
> > > > > >           return AVERROR_INVALIDDATA;
> > > > > >       s->vbr = bytestream_get_le32(&buf);
> > > > > >       s->frames_per_packet = bytestream_get_le32(&buf);
> > > > > > -    if (s->frames_per_packet <= 0)
> > > > > > +    if (s->frames_per_packet <= 0 || s->frames_per_packet >=
> > > INT32_MAX
> > > > > / s->frame_size)
> > > > >
> > > > > s->frames_per_packet * s->frame_size is then multiplied by channel
> > > count
> > > > > (which can be at most 2) for the vector_fmul_scalar() call, so might as
> > > > > well check for that here too.
> > > > >
> > > > > Alternatively, all floatdsp functions could be adapted to take size_t
> > > > > buffer lengths.
> > > > >
> > > >
> > > > Limit it to 32/64.
> > >
> > > you mean like this: (64 channels) ?
> > >
> > 
> > I mean limit max frames per packet to 64.
> 
> that would give something like this: unless we can bound frame_size tighter
> 
> diff --git a/libavcodec/speexdec.c b/libavcodec/speexdec.c
> index 4c50f54f27b..d240e8a20bf 100644
> --- a/libavcodec/speexdec.c
> +++ b/libavcodec/speexdec.c

will apply

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

Old school: Use the lowest level language in which you can solve the problem
            conveniently.
New school: Use the highest level language in which the latest supercomputer
            can solve the problem without the user falling asleep waiting.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20211114/9d4e5a37/attachment.sig>