[FFmpeg-cvslog] avcodec/ilbcdec: fix integer overflow in energy

Michael Niedermayer git at videolan.org
Mon Jan 21 13:20:25 EET 2019


ffmpeg | branch: release/4.1 | Michael Niedermayer <michael at niedermayer.cc> | Sun Dec  9 02:26:18 2018 +0100| [5550946ff4bdb4f6092e6aabd8d11888695ffbd0] | committer: Michael Niedermayer

avcodec/ilbcdec: fix integer overflow in energy

webrtc uses a int32_t like the existing code in ilbcdec

Fixes: signed integer overflow: 2080245063 + 257939661 cannot be represented in type 'int'
Fixes: 11037/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ILBC_fuzzer-5682976612941824

Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
(cherry picked from commit fbf409cd91aca2b4738c6b5bc963ae6041f26701)
Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=5550946ff4bdb4f6092e6aabd8d11888695ffbd0
---

 libavcodec/ilbcdec.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/libavcodec/ilbcdec.c b/libavcodec/ilbcdec.c
index 8a6dbe0b75..50012c0231 100644
--- a/libavcodec/ilbcdec.c
+++ b/libavcodec/ilbcdec.c
@@ -1303,7 +1303,8 @@ static int xcorr_coeff(int16_t *target, int16_t *regressor,
         pos += step;
 
         /* Do a +/- to get the next energy */
-        energy += step * ((*rp_end * *rp_end - *rp_beg * *rp_beg) >> shifts);
+        energy += (unsigned)step * ((*rp_end * *rp_end - *rp_beg * *rp_beg) >> shifts);
+
         rp_beg += step;
         rp_end += step;
     }



More information about the ffmpeg-cvslog mailing list