[FFmpeg-devel] [PATCH 1/2] avcodec/aacdec_fixed: Fix integer overflow in predict()

Michael Niedermayer michael at niedermayer.cc
Fri Oct 27 03:23:20 EEST 2017


Fixes: runtime error: signed integer overflow: -2110708110 + -82837504 cannot be represented in type 'int'
Fixes: 3547/clusterfuzz-testcase-minimized-6009386439802880

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/aacdec_fixed.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/libavcodec/aacdec_fixed.c b/libavcodec/aacdec_fixed.c
index f6a533010f..5e9249ceea 100644
--- a/libavcodec/aacdec_fixed.c
+++ b/libavcodec/aacdec_fixed.c
@@ -307,9 +307,9 @@ static av_always_inline void predict(PredictorState *ps, int *coef,
 
         if (shift < 31) {
             if (shift > 0) {
-                *coef += (pv.mant + (1 << (shift - 1))) >> shift;
+                *coef += (unsigned)((pv.mant + (1 << (shift - 1))) >> shift);
             } else
-                *coef += pv.mant << -shift;
+                *coef += (unsigned)(pv.mant << -shift);
         }
     }
 
-- 
2.14.2



More information about the ffmpeg-devel mailing list