| Message ID | 20160607120416.GA2175@mwanda |
|---|---|
| State | Accepted, archived |
| Delegated to: | David Miller |
| Headers | show |
> In the current code "ent_per_page" could be more than "conn_num" making > "conn_num" negative after the subtraction. In the next iteration through the > loop then the negative is treated as a very high positive meaning we don't put a > limit on "ent_num". It could lead to memory corruption. > > Fixes: dbb799c39717 ('qed: Initialize hardware for new protocols') > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> I don't think we can actually hit an issue here, since the number of iterations was also calculated based on conn_num, so this can only happen on the last iteration. Regardless, yours is the correct way to go. Thanks, Yuval. Acked-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
From: Dan Carpenter <dan.carpenter@oracle.com> Date: Tue, 7 Jun 2016 15:04:16 +0300 > In the current code "ent_per_page" could be more than "conn_num" making > "conn_num" negative after the subtraction. In the next iteration > through the loop then the negative is treated as a very high positive > meaning we don't put a limit on "ent_num". It could lead to memory > corruption. > > Fixes: dbb799c39717 ('qed: Initialize hardware for new protocols') > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Applied, thanks.
diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c index d85b7ba..1c35f37 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c @@ -850,7 +850,7 @@ static int qed_cxt_src_t2_alloc(struct qed_hwfn *p_hwfn) val = 0; entries[j].next = cpu_to_be64(val); - conn_num -= ent_per_page; + conn_num -= ent_num; } return 0;
In the current code "ent_per_page" could be more than "conn_num" making "conn_num" negative after the subtraction. In the next iteration through the loop then the negative is treated as a very high positive meaning we don't put a limit on "ent_num". It could lead to memory corruption. Fixes: dbb799c39717 ('qed: Initialize hardware for new protocols') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>