"The entries in an IFD must be sorted in ascending order by Tag. Note that this 
is
 not the order in which the fields are described in this document."

This way various dimensions, sample and bit sizes cannot be changed at
arbitrary times which reduces the potential for bugs.
The tag reading code also on various places assumes that numerically previous
tags have already been parsed, so this needs to be enforced one way or another.

If this commit causes problems with real world files which are not easy to fix
then some other form of checks are needed to ensure the various dependencies
in the tag reading are not violated.

Fixes: out of array access
Fixes: 
24825/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_TIFF_fuzzer-6326925027704832

Found-by: continuous fuzzing process 
https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <mich...@niedermayer.cc>
---
 libavcodec/tiff.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/libavcodec/tiff.c b/libavcodec/tiff.c
index 8a5f6e030d..86f8487086 100644
--- a/libavcodec/tiff.c
+++ b/libavcodec/tiff.c
@@ -79,6 +79,7 @@ typedef struct TiffContext {
     int fill_order;
     uint32_t res[4];
     int is_thumbnail;
+    unsigned last_tag;
 
     int is_bayer;
     uint8_t pattern[4];
@@ -1252,6 +1253,12 @@ static int tiff_decode_tag(TiffContext *s, AVFrame 
*frame)
     if (ret < 0) {
         goto end;
     }
+    if (tag <= s->last_tag)
+        return AVERROR_INVALIDDATA;
+
+    // We ignore TIFF_STRIP_SIZE as it is sometimes in the logic but wrong 
order around TIFF_STRIP_OFFS
+    if (tag != TIFF_STRIP_SIZE)
+        s->last_tag = tag;
 
     off = bytestream2_tell(&s->gb);
     if (count == 1) {
@@ -1807,6 +1814,7 @@ again:
     s->is_tiled    = 0;
     s->is_jpeg     = 0;
     s->cur_page    = 0;
+    s->last_tag    = 0;
 
     for (i = 0; i < 65536; i++)
         s->dng_lut[i] = i;
-- 
2.17.1

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Reply via email to