| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <200803052320.m25NKbr7003898@quickie.katalix.com>
Date: Wed, 5 Mar 2008 23:20:37 GMT
From: James Chapman <jchapman@...alix.com>
To: netdev@...r.kernel.org
Cc: jarkao2@...il.com
Subject: [PATCH 1/2][PPPOL2TP]: Make locking calls softirq-safe
Fix locking issues in the pppol2tp driver which can cause a kernel
crash on SMP boxes. There were two problems:-
1. The driver was violating read_lock() and write_lock() scheduling
rules because it wasn't using softirq-safe locks in softirq
contexts. So we now consistently use the _bh variants of the lock
functions.
2. The driver was calling sk_dst_get() in pppol2tp_xmit() which was
taking sk_dst_lock in softirq context. We now call __sk_dst_get().
Signed-off-by: James Chapman <jchapman@...alix.com>
Index: linux-2.6.24.3/drivers/net/pppol2tp.c
===================================================================
--- linux-2.6.24.3.orig/drivers/net/pppol2tp.c
+++ linux-2.6.24.3/drivers/net/pppol2tp.c
@@ -302,14 +302,14 @@ pppol2tp_session_find(struct pppol2tp_tu
struct pppol2tp_session *session;
struct hlist_node *walk;
- read_lock(&tunnel->hlist_lock);
+ read_lock_bh(&tunnel->hlist_lock);
hlist_for_each_entry(session, walk, session_list, hlist) {
if (session->tunnel_addr.s_session == session_id) {
- read_unlock(&tunnel->hlist_lock);
+ read_unlock_bh(&tunnel->hlist_lock);
return session;
}
}
- read_unlock(&tunnel->hlist_lock);
+ read_unlock_bh(&tunnel->hlist_lock);
return NULL;
}
@@ -320,14 +320,14 @@ static struct pppol2tp_tunnel *pppol2tp_
{
struct pppol2tp_tunnel *tunnel = NULL;
- read_lock(&pppol2tp_tunnel_list_lock);
+ read_lock_bh(&pppol2tp_tunnel_list_lock);
list_for_each_entry(tunnel, &pppol2tp_tunnel_list, list) {
if (tunnel->stats.tunnel_id == tunnel_id) {
- read_unlock(&pppol2tp_tunnel_list_lock);
+ read_unlock_bh(&pppol2tp_tunnel_list_lock);
return tunnel;
}
}
- read_unlock(&pppol2tp_tunnel_list_lock);
+ read_unlock_bh(&pppol2tp_tunnel_list_lock);
return NULL;
}
@@ -344,7 +344,7 @@ static void pppol2tp_recv_queue_skb(stru
struct sk_buff *skbp;
u16 ns = PPPOL2TP_SKB_CB(skb)->ns;
- spin_lock(&session->reorder_q.lock);
+ spin_lock_bh(&session->reorder_q.lock);
skb_queue_walk(&session->reorder_q, skbp) {
if (PPPOL2TP_SKB_CB(skbp)->ns > ns) {
__skb_insert(skb, skbp->prev, skbp, &session->reorder_q);
@@ -360,7 +360,7 @@ static void pppol2tp_recv_queue_skb(stru
__skb_queue_tail(&session->reorder_q, skb);
out:
- spin_unlock(&session->reorder_q.lock);
+ spin_unlock_bh(&session->reorder_q.lock);
}
/* Dequeue a single skb.
@@ -442,7 +442,7 @@ static void pppol2tp_recv_dequeue(struct
* expect to send up next, dequeue it and any other
* in-sequence packets behind it.
*/
- spin_lock(&session->reorder_q.lock);
+ spin_lock_bh(&session->reorder_q.lock);
skb_queue_walk_safe(&session->reorder_q, skb, tmp) {
if (time_after(jiffies, PPPOL2TP_SKB_CB(skb)->expires)) {
session->stats.rx_seq_discards++;
@@ -469,13 +469,13 @@ static void pppol2tp_recv_dequeue(struct
goto out;
}
}
- spin_unlock(&session->reorder_q.lock);
+ spin_unlock_bh(&session->reorder_q.lock);
pppol2tp_recv_dequeue_skb(session, skb);
- spin_lock(&session->reorder_q.lock);
+ spin_lock_bh(&session->reorder_q.lock);
}
out:
- spin_unlock(&session->reorder_q.lock);
+ spin_unlock_bh(&session->reorder_q.lock);
}
/* Internal receive frame. Do the real work of receiving an L2TP data frame
@@ -1058,7 +1058,7 @@ static int pppol2tp_xmit(struct ppp_chan
/* Get routing info from the tunnel socket */
dst_release(skb->dst);
- skb->dst = sk_dst_get(sk_tun);
+ skb->dst = dst_clone(__sk_dst_get(sk_tun));
skb_orphan(skb);
skb->sk = sk_tun;
@@ -1106,7 +1106,7 @@ static void pppol2tp_tunnel_closeall(str
PRINTK(tunnel->debug, PPPOL2TP_MSG_CONTROL, KERN_INFO,
"%s: closing all sessions...\n", tunnel->name);
- write_lock(&tunnel->hlist_lock);
+ write_lock_bh(&tunnel->hlist_lock);
for (hash = 0; hash < PPPOL2TP_HASH_SIZE; hash++) {
again:
hlist_for_each_safe(walk, tmp, &tunnel->session_hlist[hash]) {
@@ -1126,7 +1126,7 @@ again:
* disappear as we're jumping between locks.
*/
sock_hold(sk);
- write_unlock(&tunnel->hlist_lock);
+ write_unlock_bh(&tunnel->hlist_lock);
lock_sock(sk);
if (sk->sk_state & (PPPOX_CONNECTED | PPPOX_BOUND)) {
@@ -1148,11 +1148,11 @@ again:
* list so we are guaranteed to make forward
* progress.
*/
- write_lock(&tunnel->hlist_lock);
+ write_lock_bh(&tunnel->hlist_lock);
goto again;
}
}
- write_unlock(&tunnel->hlist_lock);
+ write_unlock_bh(&tunnel->hlist_lock);
}
/* Really kill the tunnel.
@@ -1161,9 +1161,9 @@ again:
static void pppol2tp_tunnel_free(struct pppol2tp_tunnel *tunnel)
{
/* Remove from socket list */
- write_lock(&pppol2tp_tunnel_list_lock);
+ write_lock_bh(&pppol2tp_tunnel_list_lock);
list_del_init(&tunnel->list);
- write_unlock(&pppol2tp_tunnel_list_lock);
+ write_unlock_bh(&pppol2tp_tunnel_list_lock);
atomic_dec(&pppol2tp_tunnel_count);
kfree(tunnel);
@@ -1239,9 +1239,9 @@ static void pppol2tp_session_destruct(st
/* Delete the session socket from the
* hash
*/
- write_lock(&tunnel->hlist_lock);
+ write_lock_bh(&tunnel->hlist_lock);
hlist_del_init(&session->hlist);
- write_unlock(&tunnel->hlist_lock);
+ write_unlock_bh(&tunnel->hlist_lock);
atomic_dec(&pppol2tp_session_count);
}
@@ -1386,9 +1386,9 @@ static struct sock *pppol2tp_prepare_tun
/* Add tunnel to our list */
INIT_LIST_HEAD(&tunnel->list);
- write_lock(&pppol2tp_tunnel_list_lock);
+ write_lock_bh(&pppol2tp_tunnel_list_lock);
list_add(&tunnel->list, &pppol2tp_tunnel_list);
- write_unlock(&pppol2tp_tunnel_list_lock);
+ write_unlock_bh(&pppol2tp_tunnel_list_lock);
atomic_inc(&pppol2tp_tunnel_count);
/* Bump the reference count. The tunnel context is deleted
@@ -1593,11 +1593,11 @@ static int pppol2tp_connect(struct socke
sk->sk_user_data = session;
/* Add session to the tunnel's hash list */
- write_lock(&tunnel->hlist_lock);
+ write_lock_bh(&tunnel->hlist_lock);
hlist_add_head(&session->hlist,
pppol2tp_session_id_hash(tunnel,
session->tunnel_addr.s_session));
- write_unlock(&tunnel->hlist_lock);
+ write_unlock_bh(&tunnel->hlist_lock);
atomic_inc(&pppol2tp_session_count);
@@ -2199,7 +2199,7 @@ static struct pppol2tp_session *next_ses
int next = 0;
int i;
- read_lock(&tunnel->hlist_lock);
+ read_lock_bh(&tunnel->hlist_lock);
for (i = 0; i < PPPOL2TP_HASH_SIZE; i++) {
hlist_for_each_entry(session, walk, &tunnel->session_hlist[i], hlist) {
if (curr == NULL) {
@@ -2217,7 +2217,7 @@ static struct pppol2tp_session *next_ses
}
}
out:
- read_unlock(&tunnel->hlist_lock);
+ read_unlock_bh(&tunnel->hlist_lock);
if (!found)
session = NULL;
@@ -2228,13 +2228,13 @@ static struct pppol2tp_tunnel *next_tunn
{
struct pppol2tp_tunnel *tunnel = NULL;
- read_lock(&pppol2tp_tunnel_list_lock);
+ read_lock_bh(&pppol2tp_tunnel_list_lock);
if (list_is_last(&curr->list, &pppol2tp_tunnel_list)) {
goto out;
}
tunnel = list_entry(curr->list.next, struct pppol2tp_tunnel, list);
out:
- read_unlock(&pppol2tp_tunnel_list_lock);
+ read_unlock_bh(&pppol2tp_tunnel_list_lock);
return tunnel;
}
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists