| commit | a1c6fd9f23cdecc8d3cbe791dd535d0e5717eb65 | |
| author | Dan Carpenter <dan.carpenter@oracle.com> | |
| Wed, 1 Jul 2015 10:52:15 +0000 (1 13:52 +0300) | ||
| committer | Dan Carpenter <dan.carpenter@oracle.com> | |
| Wed, 1 Jul 2015 10:52:15 +0000 (1 13:52 +0300) | ||
| tree | c21d7e79a98fdbcc60abfb36defa3e1427ac26e4 | treesnapshot (tar.gz zip) |
| parent | 67a2b631f1f3557e9262974fc9d25d63cb581672 | commitdiff |
user_data2: handle pointers to user data better
There are a couple related issues here. If you had a buffer containing
user data then we should record that *buf is USER_DATA. But even if it
is a struct pointer containing user data and we have tagged the members,
we should still record that *p is USER_DATA in case the buffer gets
casted to something else.
This information needs to be recorded in caller_info etc.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
There are a couple related issues here. If you had a buffer containing
user data then we should record that *buf is USER_DATA. But even if it
is a struct pointer containing user data and we have tagged the members,
we should still record that *p is USER_DATA in case the buffer gets
casted to something else.
This information needs to be recorded in caller_info etc.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
| check_user_data2.c | diffblobblamehistory |