avcodec/ilbcdec: fix integer overflow in energy
authorMichael Niedermayer <michael@niedermayer.cc>
Sun, 9 Dec 2018 01:26:18 +0000 (02:26 +0100)
committerMichael Niedermayer <michael@niedermayer.cc>
Sun, 23 Dec 2018 19:24:47 +0000 (20:24 +0100)
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@niedermayer.cc>
libavcodec/ilbcdec.c

index 8a6dbe0..50012c0 100644 (file)
@@ -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;
     }