avcodec/aacdec_fixed: use 64bit to avoid overflow in rounding in apply_dependent_coup...
authorMichael Niedermayer <michael@niedermayer.cc>
Fri, 25 May 2018 20:06:48 +0000 (22:06 +0200)
committerMichael Niedermayer <michael@niedermayer.cc>
Mon, 9 Jul 2018 23:52:56 +0000 (01:52 +0200)
Fixes: signed integer overflow: -2141499320 + -14469590 cannot be represented in type 'int'
Fixes: 7351/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AAC_FIXED_fuzzer-6351214791884800

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 90475db97e2e5931d295df6ab86519fa2e14d259)
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
libavcodec/aacdec_fixed.c

index a37ed2a..e0a54ff 100644 (file)
@@ -384,7 +384,7 @@ static void apply_dependent_coupling_fixed(AACContext *ac,
                         for (k = offsets[i]; k < offsets[i + 1]; k++) {
                             tmp = (int)(((int64_t)src[group * 128 + k] * c + \
                                        (int64_t)0x1000000000) >> 37);
-                            dest[group * 128 + k] += (tmp + round) >> shift;
+                            dest[group * 128 + k] += (tmp + (int64_t)round) >> shift;
                         }
                     }
                 }