[FFmpeg-cvslog] avcodec/sbrdsp_fixed: Fix integer overflow in sbr_hf_apply_noise()

Michael Niedermayer git at videolan.org
Sun Jul 9 00:46:11 EEST 2017


ffmpeg | branch: master | Michael Niedermayer <michael at niedermayer.cc> | Sun Jul  2 03:30:54 2017 +0200| [2061de8a3f73f14806e5f6ccaf9a635f740a54e6] | committer: Michael Niedermayer

avcodec/sbrdsp_fixed: Fix integer overflow in sbr_hf_apply_noise()

Fixes: runtime error: signed integer overflow: -2049425300 + -117591631 cannot be represented in type 'int'
Fixes: part of 2096/clusterfuzz-testcase-minimized-4901566068817920

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

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=2061de8a3f73f14806e5f6ccaf9a635f740a54e6
---

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

diff --git a/libavcodec/sbrdsp_fixed.c b/libavcodec/sbrdsp_fixed.c
index 218bb1acd8..fe48104021 100644
--- a/libavcodec/sbrdsp_fixed.c
+++ b/libavcodec/sbrdsp_fixed.c
@@ -264,8 +264,8 @@ static av_always_inline int sbr_hf_apply_noise(int (*Y)[2],
     int m;
 
     for (m = 0; m < m_max; m++) {
-        int y0 = Y[m][0];
-        int y1 = Y[m][1];
+        unsigned y0 = Y[m][0];
+        unsigned y1 = Y[m][1];
         noise = (noise + 1) & 0x1ff;
         if (s_m[m].mant) {
             int shift, round;



More information about the ffmpeg-cvslog mailing list