[FFmpeg-cvslog] avcodec/diracdec: Use 64bit in intermediate of global motion vector field generation

Michael Niedermayer git at videolan.org
Thu Jun 27 20:47:41 EEST 2019


ffmpeg | branch: release/4.1 | Michael Niedermayer <michael at niedermayer.cc> | Sun Apr  7 16:44:53 2019 +0200| [fb3135ae97948c9dab2b68cfe8a59f35f1e7d287] | committer: Michael Niedermayer

avcodec/diracdec: Use 64bit in intermediate of global motion vector field generation

It seems the specification does not limit the value to 32bit

Fixes: signed integer overflow: -109611143 * 24 cannot be represented in type 'int'
Fixes: 13477/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DIRAC_fuzzer-5648337460527104

Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
(cherry picked from commit 837820f385af699f9bee5e2ba3169dda15e5894d)
Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=fb3135ae97948c9dab2b68cfe8a59f35f1e7d287
---

 libavcodec/diracdec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libavcodec/diracdec.c b/libavcodec/diracdec.c
index fd23139062..0216cf2650 100644
--- a/libavcodec/diracdec.c
+++ b/libavcodec/diracdec.c
@@ -1427,7 +1427,7 @@ static void global_mv(DiracContext *s, DiracBlock *block, int x, int y, int ref)
     int *b      = s->globalmc[ref].pan_tilt;
     int *c      = s->globalmc[ref].perspective;
 
-    int m       = (1<<ep) - (c[0]*x + c[1]*y);
+    int64_t m   = (1<<ep) - (c[0]*(int64_t)x + c[1]*(int64_t)y);
     int64_t mx  = m * (int64_t)((A[0][0] * (int64_t)x + A[0][1]*(int64_t)y) + (1<<ez) * b[0]);
     int64_t my  = m * (int64_t)((A[1][0] * (int64_t)x + A[1][1]*(int64_t)y) + (1<<ez) * b[1]);
 



More information about the ffmpeg-cvslog mailing list