| Message ID | 20190829194632.3918-2-michael@niedermayer.cc |
|---|---|
| State | Accepted |
| Commit | 12b909ba319d32ed100d9b26021aa9b6976424d7 |
| Headers | show |
On Thu, Aug 29, 2019 at 09:46:32PM +0200, Michael Niedermayer wrote: > Fixes: index 2304 out of bounds for type 'float [2304]' > Fixes: 16332/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_QDM2_fuzzer-5679142481166336 > > Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc> > --- > libavcodec/qdm2.c | 5 +++++ > 1 file changed, 5 insertions(+) will apply [...]
Are you sure this is not band aid but actually real fix? On 8/29/19, Michael Niedermayer <michael@niedermayer.cc> wrote: > Fixes: index 2304 out of bounds for type 'float [2304]' > Fixes: > 16332/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_QDM2_fuzzer-5679142481166336 > > Found-by: continuous fuzzing process > https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc> > --- > libavcodec/qdm2.c | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/libavcodec/qdm2.c b/libavcodec/qdm2.c > index eaffb36dcc..ac8ae8cbbb 100644 > --- a/libavcodec/qdm2.c > +++ b/libavcodec/qdm2.c > @@ -1727,6 +1727,11 @@ static av_cold int qdm2_decode_init(AVCodecContext > *avctx) > s->sub_sampling = s->fft_order - 7; > s->frequency_range = 255 / (1 << (2 - s->sub_sampling)); > > + if (s->frame_size * 4 >> s->sub_sampling > MPA_FRAME_SIZE) { > + avpriv_request_sample(avctx, "large frames"); > + return AVERROR_PATCHWELCOME; > + } > + > switch ((s->sub_sampling * 2 + s->channels - 1)) { > case 0: tmp = 40; break; > case 1: tmp = 48; break; > -- > 2.23.0 > > _______________________________________________ > 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".
On Fri, Sep 06, 2019 at 11:08:06PM +0200, Paul B Mahol wrote: > Are you sure this is not band aid but actually real fix? Sorry, I did not see your reply yesterday before applying the change. Without a real (non fuzzer) sample and no reference specification or reference implementation that is open source. Asking for a sample seems the obvious solution to me. Having a sample for this case would be quite usefull ... I do not know if this corner case ever occurs in a real sample, but if it does occur then yes this is not a real fix for it, it just avoids the anomaly found by the fuzzer. If this never occurs in any encoder output then it would be a real fix. (minus asking for a sample) Thanks > > On 8/29/19, Michael Niedermayer <michael@niedermayer.cc> wrote: > > Fixes: index 2304 out of bounds for type 'float [2304]' > > Fixes: > > 16332/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_QDM2_fuzzer-5679142481166336 > > > > Found-by: continuous fuzzing process > > https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg > > Signed-off-by: Michael Niedermayer <michael@niedermayer.cc> > > --- > > libavcodec/qdm2.c | 5 +++++ > > 1 file changed, 5 insertions(+) > > > > diff --git a/libavcodec/qdm2.c b/libavcodec/qdm2.c > > index eaffb36dcc..ac8ae8cbbb 100644 > > --- a/libavcodec/qdm2.c > > +++ b/libavcodec/qdm2.c > > @@ -1727,6 +1727,11 @@ static av_cold int qdm2_decode_init(AVCodecContext > > *avctx) > > s->sub_sampling = s->fft_order - 7; > > s->frequency_range = 255 / (1 << (2 - s->sub_sampling)); > > > > + if (s->frame_size * 4 >> s->sub_sampling > MPA_FRAME_SIZE) { > > + avpriv_request_sample(avctx, "large frames"); > > + return AVERROR_PATCHWELCOME; > > + } > > + > > switch ((s->sub_sampling * 2 + s->channels - 1)) { > > case 0: tmp = 40; break; > > case 1: tmp = 48; break; > > -- > > 2.23.0 > > > > _______________________________________________ > > 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". > _______________________________________________ > 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".
diff --git a/libavcodec/qdm2.c b/libavcodec/qdm2.c index eaffb36dcc..ac8ae8cbbb 100644 --- a/libavcodec/qdm2.c +++ b/libavcodec/qdm2.c @@ -1727,6 +1727,11 @@ static av_cold int qdm2_decode_init(AVCodecContext *avctx) s->sub_sampling = s->fft_order - 7; s->frequency_range = 255 / (1 << (2 - s->sub_sampling)); + if (s->frame_size * 4 >> s->sub_sampling > MPA_FRAME_SIZE) { + avpriv_request_sample(avctx, "large frames"); + return AVERROR_PATCHWELCOME; + } + switch ((s->sub_sampling * 2 + s->channels - 1)) { case 0: tmp = 40; break; case 1: tmp = 48; break;
Fixes: index 2304 out of bounds for type 'float [2304]' Fixes: 16332/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_QDM2_fuzzer-5679142481166336 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael@niedermayer.cc> --- libavcodec/qdm2.c | 5 +++++ 1 file changed, 5 insertions(+)