[FFmpeg-devel] [PATCH 4/4] avcodec/wmalosslessdec: Fix 2 overflows in mclms

Michael Niedermayer michael at niedermayer.cc
Thu Nov 21 00:23:30 EET 2019


Fixes: signed integer overflow: 2038337026 + 109343477 cannot be represented in type 'int'
Fixes: 18886/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_WMALOSSLESS_fuzzer-5673660505653248

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

diff --git a/libavcodec/wmalosslessdec.c b/libavcodec/wmalosslessdec.c
index d4f18b9841..d454478687 100644
--- a/libavcodec/wmalosslessdec.c
+++ b/libavcodec/wmalosslessdec.c
@@ -628,7 +628,7 @@ static void mclms_update(WmallDecodeCtx *s, int icoef, int *pred)
     int range        = 1 << (s->bits_per_sample - 1);
 
     for (ich = 0; ich < num_channels; ich++) {
-        pred_error = s->channel_residues[ich][icoef] - pred[ich];
+        pred_error = s->channel_residues[ich][icoef] - (unsigned)pred[ich];
         if (pred_error > 0) {
             for (i = 0; i < order * num_channels; i++)
                 s->mclms_coeffs[i + ich * order * num_channels] +=
@@ -680,7 +680,7 @@ static void mclms_predict(WmallDecodeCtx *s, int icoef, int *pred)
                          s->mclms_coeffs_cur[i + num_channels * ich];
         pred[ich] += 1 << s->mclms_scaling - 1;
         pred[ich] >>= s->mclms_scaling;
-        s->channel_residues[ich][icoef] += pred[ich];
+        s->channel_residues[ich][icoef] += (unsigned)pred[ich];
     }
 }
 
-- 
2.23.0



More information about the ffmpeg-devel mailing list