1. ffmpeg
  2. Commit 5b43aa7a80f5ddaa46fec9866061a7c0e0d7f72e
Browse code

avcodec/vc1_block: Fix integer overflow in AC rescaling in vc1_decode_i_block_adv()

Fixes: signed integer overflow: 50176 * 262144 cannot be represented in type 'int'
Fixes: 18629/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_VC1IMAGE_fuzzer-5182370286403584

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 0e010e489b70c044a67c47083cf8eb03209ee89f)
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Michael Niedermayer authored on 2019/11/09 02:31:02
Showing 1 changed files
libavcodec/vc1_block.c
History View file @ 5b43aa7
... ... 
@@ -867,7 +867,7 @@ static int vc1_decode_i_block_adv(VC1Context *v, int16_t block[64], int n,
867 867 
                 if (q1 < 1)
868 868 
                     return AVERROR_INVALIDDATA;
869 869 
                 for (k = 1; k < 8; k++)
870 
-                    ac_val2[k] = (ac_val2[k] * q2 * ff_vc1_dqscale[q1 - 1] + 0x20000) >> 18;
870 
+                    ac_val2[k] = (int)(ac_val2[k] * q2 * (unsigned)ff_vc1_dqscale[q1 - 1] + 0x20000) >> 18;
871 871 
             }
872 872 
             for (k = 1; k < 8; k++) {
873 873 
                 block[k << sh] = ac_val2[k] * scale;