| commit | e6f997e51d66d6321881648fc8050c02cbd4c605 | |
| author | Dan Carpenter <dan.carpenter@oracle.com> | |
| Fri, 15 Jun 2012 10:01:57 +0000 (15 13:01 +0300) | ||
| committer | Dan Carpenter <dan.carpenter@oracle.com> | |
| Fri, 15 Jun 2012 10:01:57 +0000 (15 13:01 +0300) | ||
| tree | 8ec3549a1a8d32f20d9d410733f2d26035d2742b | treesnapshot (tar.gz zip) |
| parent | 57a6294bef41645178338aba79539e7f0d836ba6 | commitdiff |
buf_size: if the member size is not known then store that in the DB
The thought here is that some drivers allocate a buffer in the
module_init() and it's always the same size. The problem is that we were
only storing the size if it was known, and not if it was unknown. So it
created a problem that sometimes we thought we knew the buffer size but
actually we didn't.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
The thought here is that some drivers allocate a buffer in the
module_init() and it's always the same size. The problem is that we were
only storing the size if it was known, and not if it was unknown. So it
created a problem that sometimes we thought we knew the buffer size but
actually we didn't.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
| smatch_buf_size.c | diffblobblamehistory |