avcodec/truemotion2: Fix 2 integer overflows in tm2_update_block()
authorMichael Niedermayer <michael@niedermayer.cc>
Fri, 19 Apr 2019 23:05:44 +0000 (01:05 +0200)
committerMichael Niedermayer <michael@niedermayer.cc>
Thu, 27 Jun 2019 15:50:47 +0000 (17:50 +0200)
Fixes: signed integer overflow: -2147483648 + -1 cannot be represented in type 'int'
Fixes: 14107/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_TRUEMOTION2_fuzzer-5694078680825856

Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
(cherry picked from commit f4a1b8d409639b2394589efe20ad55410cce391c)
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
libavcodec/truemotion2.c

index 4bfe6e1..60354fd 100644 (file)
@@ -678,8 +678,8 @@ static inline void tm2_update_block(TM2Context *ctx, AVFrame *pic, int bx, int b
     /* update chroma */
     for (j = 0; j < 2; j++) {
         for (i = 0; i < 2; i++) {
-            U[i] = Uo[i] + GET_TOK(ctx, TM2_UPD);
-            V[i] = Vo[i] + GET_TOK(ctx, TM2_UPD);
+            U[i] = Uo[i] + (unsigned)GET_TOK(ctx, TM2_UPD);
+            V[i] = Vo[i] + (unsigned)GET_TOK(ctx, TM2_UPD);
         }
         U  += Ustride;
         V  += Vstride;