From: xiaohuizhang98 <[email protected]>
We detected a suspected bug with our code clone detection tool.
Similar to the handling of l2tp_tunnel_get in commit a622b40035d1
("l2ip: fix possible use-after-free"), we thought a patch might
be needed here as well.
Before taking a refcount on a rcu protected structure,
we need to make sure the refcount is not zero.
Signed-off-by: xiaohuizhang98 <[email protected]>
---
net/l2tp/l2tp_core.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index b759fbd09b65..c5de6d4e0818 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -273,8 +273,8 @@ struct l2tp_session *l2tp_session_get(const struct net *net, u32 session_id)
rcu_read_lock_bh();
hlist_for_each_entry_rcu(session, session_list, global_hlist)
- if (session->session_id == session_id) {
- l2tp_session_inc_refcount(session);
+ if (session->session_id == session_id &&
+ refcount_inc_not_zero(&session->ref_count)) {
rcu_read_unlock_bh();
return session;
--
2.17.1
Hello,
On Sun, 2022-06-05 at 19:54 +0800, Xiaohui Zhang wrote:
> From: xiaohuizhang98 <[email protected]>
>
> We detected a suspected bug with our code clone detection tool.
>
> Similar to the handling of l2tp_tunnel_get in commit a622b40035d1
> ("l2ip: fix possible use-after-free"), we thought a patch might
> be needed here as well.
>
> Before taking a refcount on a rcu protected structure,
> we need to make sure the refcount is not zero.
>
> Signed-off-by: xiaohuizhang98 <[email protected]>
It looks like series this did not make to patchwork, due to a typo in
the ML address, please fix it and resend: [email protected] ->
[email protected]
Thanks,
Paolo
On Tue, Jun 7, 2022 at 9:16 AM Xiaohui Zhang <[email protected]> wrote:
>
> We detected a suspected bug with our code clone detection tool.
>
> Similar to the handling of l2tp_tunnel_get in commit a622b40035d1
> ("l2ip: fix possible use-after-free"), we thought a patch might
> be needed here as well.
>
> Before taking a refcount on a rcu protected structure,
> we need to make sure the refcount is not zero.
>
> Signed-off-by: Xiaohui Zhang <[email protected]>
> ---
> net/l2tp/l2tp_core.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
> index b759fbd09b65..c5de6d4e0818 100644
> --- a/net/l2tp/l2tp_core.c
> +++ b/net/l2tp/l2tp_core.c
> @@ -273,8 +273,8 @@ struct l2tp_session *l2tp_session_get(const struct net *net, u32 session_id)
>
> rcu_read_lock_bh();
> hlist_for_each_entry_rcu(session, session_list, global_hlist)
> - if (session->session_id == session_id) {
> - l2tp_session_inc_refcount(session);
> + if (session->session_id == session_id &&
> + refcount_inc_not_zero(&session->ref_count)) {
> rcu_read_unlock_bh();
>
> return session;
> --
> 2.17.1
>
Please fix all bugs in a single patch.
net/l2tp/l2tp_core.c contains four suspect calls to l2tp_session_inc_refcount()