[FFmpeg-devel] [PATCH 2/2] avcodec/error_resilience: Fix integer overflow in filter181()

Michael Niedermayer michael at niedermayer.cc
Sun Apr 22 23:20:32 EEST 2018


Fixes: runtime error: signed integer overflow: 197710 * 10923 cannot be represented in type 'int'
Fixes: 7010/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MPEG4_fuzzer-5667127596941312

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

diff --git a/libavcodec/error_resilience.c b/libavcodec/error_resilience.c
index 25e54a535b..339042e206 100644
--- a/libavcodec/error_resilience.c
+++ b/libavcodec/error_resilience.c
@@ -107,7 +107,7 @@ static void filter181(int16_t *data, int width, int height, ptrdiff_t stride)
             dc = -prev_dc +
                  data[x     + y * stride] * 8 -
                  data[x + 1 + y * stride];
-            dc = (dc * 10923 + 32768) >> 16;
+            dc = (av_clip(dc, INT_MIN/10923, INT_MAX/10923 - 32768) * 10923 + 32768) >> 16;
             prev_dc = data[x + y * stride];
             data[x + y * stride] = dc;
         }
@@ -123,7 +123,7 @@ static void filter181(int16_t *data, int width, int height, ptrdiff_t stride)
             dc = -prev_dc +
                  data[x +  y      * stride] * 8 -
                  data[x + (y + 1) * stride];
-            dc = (dc * 10923 + 32768) >> 16;
+            dc = (av_clip(dc, INT_MIN/10923, INT_MAX/10923 - 32768) * 10923 + 32768) >> 16;
             prev_dc = data[x + y * stride];
             data[x + y * stride] = dc;
         }
-- 
2.17.0



More information about the ffmpeg-devel mailing list