[FFmpeg-devel] [PATCH 4/5] avcodec/hevc_mp4toannexb_bsf: check that nalu size doesnt overflow

[Andriy Gelman]

On Fri, 13. Dec 18:51, Michael Niedermayer wrote:
> On Thu, Dec 12, 2019 at 07:36:50PM -0500, Andriy Gelman wrote:
> > On Fri, 13. Dec 01:28, Michael Niedermayer wrote:
> > > Fixes: Out of array access
> > > Fixes: 19299/clusterfuzz-testcase-minimized-ffmpeg_BSF_HEVC_MP4TOANNEXB_fuzzer-5169193398042624
> > > 
> > > 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/hevc_mp4toannexb_bsf.c | 3 +--
> > >  1 file changed, 1 insertion(+), 2 deletions(-)
> > > 
> > > diff --git a/libavcodec/hevc_mp4toannexb_bsf.c b/libavcodec/hevc_mp4toannexb_bsf.c
> > > index d0f1b94f0e..baa93628ed 100644
> > > --- a/libavcodec/hevc_mp4toannexb_bsf.c
> > > +++ b/libavcodec/hevc_mp4toannexb_bsf.c
> > > @@ -152,8 +152,7 @@ static int hevc_mp4toannexb_filter(AVBSFContext *ctx, AVPacket *out)
> > >          extra_size    = add_extradata * ctx->par_out->extradata_size;
> > >          got_irap     |= is_irap;
> > >  
> > > -        if (SIZE_MAX - nalu_size < 4 ||
> > > -            SIZE_MAX - 4 - nalu_size < extra_size) {
> > > +        if (FFMIN(INT_MAX, SIZE_MAX) < 4ULL + nalu_size + extra_size) {
> > >              ret = AVERROR_INVALIDDATA;
> > >              goto fail;
> > >          }
> > > -- 
> > > 2.24.0
> > 
> > I already sent a patch for this:
> > http://ffmpeg.org/pipermail/ffmpeg-devel/2019-December/253972.html
> 
> yes but it depends on "distant" code behaving in specific ways. I prefer
> a single self contained check. Because its more robust to someone changing
> that distant code.
> So i would prefer to apply some "dumb" check. Do you agree ?

I agree it's more robust. I'll keep this in mind for future

Thanks,
-- 
Andriy