Re: [FFmpeg-devel] [PATCH 4/4] avcodec/bonk: Check for undefined overflow in predictor_calc_error()

From: Michael Niedermayer <michael@niedermayer.cc>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: Re: [FFmpeg-devel] [PATCH 4/4] avcodec/bonk: Check for undefined overflow in predictor_calc_error()
Date: Sun, 6 Nov 2022 13:30:36 +0100
Message-ID: <20221106123036.GV1814017@pb2> (raw)
In-Reply-To: <CAPYw7P4Kt4iMDTWQDfwhHH_wPOzSBSe0Rie7JO3Hr-vYBH-PAg@mail.gmail.com>

[-- Attachment #1.1: Type: text/plain, Size: 2766 bytes --]

On Sun, Nov 06, 2022 at 09:50:47AM +0100, Paul B Mahol wrote:
> On 11/5/22, Michael Niedermayer <michael@niedermayer.cc> wrote:
> > Fixes: signed integer overflow: 22 * -2107998208 cannot be represented in
> > type 'int'
> > Fixes:
> > 51363/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_BONK_fuzzer-5660734784143360
> >
> > Found-by: continuous fuzzing process
> > https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
> > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
> > ---
> >  libavcodec/bonk.c | 11 +++++++++--
> >  1 file changed, 9 insertions(+), 2 deletions(-)
> >
> > diff --git a/libavcodec/bonk.c b/libavcodec/bonk.c
> > index 1695229dbd..40963aa7c6 100644
> > --- a/libavcodec/bonk.c
> > +++ b/libavcodec/bonk.c
> > @@ -278,10 +278,13 @@ static int predictor_calc_error(int *k, int *state,
> > int order, int error)
> >          *state_ptr = &(state[order-2]);
> >
> >      for (i = order-2; i >= 0; i--, k_ptr--, state_ptr--) {
> > -        int k_value = *k_ptr, state_value = *state_ptr;
> > +        int64_t k_value = *k_ptr, state_value = *state_ptr;
> >
> >          x -= shift_down(k_value * state_value, LATTICE_SHIFT);
> > -        state_ptr[1] = state_value + shift_down(k_value * x,
> > LATTICE_SHIFT);
> > +        k_value *= x;
> > +        if ((int32_t)k_value != k_value)
> > +            return AVERROR_INVALIDDATA;
> > +        state_ptr[1] = state_value + shift_down(k_value, LATTICE_SHIFT);
> >      }
> >
> >      // don't drift too far, to avoid overflows
> > @@ -366,6 +369,8 @@ static int bonk_decode(AVCodecContext *avctx, AVFrame
> > *frame,
> >              int64_t t64;
> >              for (int j = 0; j < s->down_sampling - 1; j++) {
> >                  sample[0] = predictor_calc_error(s->k, state, s->n_taps,
> > 0);
> > +                if (sample[0] == AVERROR_INVALIDDATA)
> > +                    return sample[0];
> >                  sample++;
> >              }
> >
> > @@ -374,6 +379,8 @@ static int bonk_decode(AVCodecContext *avctx, AVFrame
> > *frame,
> >                  return AVERROR_INVALIDDATA;
> >
> >              sample[0] = predictor_calc_error(s->k, state, s->n_taps, t64);
> > +            if (sample[0] == AVERROR_INVALIDDATA)
> > +                return sample[0];
> >              sample++;
> >          }
> >
> > --
> > 2.17.1
> >
> 
> 
> NAK,
> 
> slowing things down by using int64_t

ill post some other solution, it seems slower indeed, i was hoping
gcc would produce smarter code

thx

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

Elect your leaders based on what they did after the last election, not
based on what they say before an election.

[-- 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".