atm: fix use-after-free in sigd_put_skb()
Subsystems (3)
PATCH v2 1/1 atm: fix use-after-free in sigd_put_skb()
sigd_put_skb() delivers a signalling message to the daemon socket named by the global @sigd pointer, ending in a call to sk_data_ready(). It reads @sigd with no synchronisation, so it can race with a close of the daemon socket: sigd_close() clears @sigd and the socket is then torn down and freed.
Holding a reference on the socket is not enough to make this safe. The daemon fd close runs __sock_release(), which frees the struct socket -- and the wait queue that sk->sk_wq points at -- via iput() once ->release() has returned. sk_data_ready() (sock_def_readable()) then dereferences a freed sk_wq:
KASAN: null-ptr-deref in range [0x0000000000000188-0x000000000000018f] RIP: 0010:sigd_put_skb (net/atm/signaling.c:65) sigd_enq2 (net/atm/signaling.c:228) sigd_enq (net/atm/signaling.c:237) svc_bind (net/atm/svc.c:135) __sys_bind __x64_sys_bind do_syscall_64
Fix it on both sides. sigd_close() now calls sock_orphan(), which under sk_callback_lock sets SOCK_DEAD and clears sk_wq before the socket is freed. sigd_put_skb() latches @sigd with READ_ONCE(), pins the socket with find_get_vcc(), and then takes sk_callback_lock; it delivers only while the socket is still the signalling daemon and not yet SOCK_DEAD, otherwise it drops the skb. sk_callback_lock is used rather than lock_sock() because sigd_put_skb() can be reached from vcc_sendmsg() -> sigd_send(), which already holds lock_sock() on the daemon socket.
The READ_ONCE(sigd) != vcc check also rejects a reused VCC: a caller preempted between latching @sigd and validating it in find_get_vcc() could otherwise match a freed-and-reallocated, live (non-SOCK_DEAD) VCC. This needs a concurrent (privileged) daemon teardown, so it is hardening.
Triggering the use-after-free requires CAP_NET_ADMIN and CAP_SYS_RAWIO to attach the daemon.
---
v2:
- Also bail when READ_ONCE(sigd) != vcc, rejecting a reused VCC that a preempted caller could match in find_get_vcc() (hardening, no stable).
net/atm/signaling.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-)
net/atm/signaling.c +25 -4
@@ -54,14 +54,32 @@ static struct atm_vcc *find_get_vcc(struct atm_vcc *vcc)static void sigd_put_skb (struct sk_buff *skb) { -if (!sigd) { +struct atm_vcc *vcc; +struct sock *sk; + + vcc =find_get_vcc (READ_ONCE (sigd)); +if (!vcc) {pr_debug ("atmsvc: no signaling daemon\n" );kfree_skb (skb);return ; } -atm_force_charge (sigd, skb->truesize); -skb_queue_tail (&sk_atm (sigd)->sk_receive_queue, skb); -sk_atm (sigd)->sk_data_ready (sk_atm (sigd)); + sk =sk_atm (vcc); + +/* Pairs with sock_orphan() in sigd_close(). */ +read_lock_bh (&sk->sk_callback_lock); +/* Bail if torn down, or if the slot was reused by another VCC. */ +if (sock_flag (sk,SOCK_DEAD ) ||READ_ONCE (sigd) != vcc) { +read_unlock_bh (&sk->sk_callback_lock); +sock_put (sk); +kfree_skb (skb); +return ; + } +atm_force_charge (vcc, skb->truesize); +skb_queue_tail (&sk->sk_receive_queue, skb); + sk->sk_data_ready (sk); +read_unlock_bh (&sk->sk_callback_lock); + +sock_put (sk); }static void modify_qos (struct atm_vcc *vcc,struct atmsvc_msg *msg)
@@ -257,6 +275,9 @@ static void sigd_close(struct atm_vcc *vcc)pr_err ("closing with requests pending\n" );skb_queue_purge (&sk_atm (vcc)->sk_receive_queue); +/* Make a concurrent sigd_put_skb() observe SOCK_DEAD and bail. */ +sock_orphan (sk_atm (vcc)); +read_lock (&vcc_sklist_lock);for (i =0 ; i <VCC_HTABLE_SIZE ; ++i) {struct hlist_head *head = &vcc_hash[i];
Signature · 1 line
2.43.0
Quoted 101 lines · Weiming Shi
sigd_put_skb()delivers a signalling message to the daemon socket named by the global @sigd pointer, ending in a call tosk_data_ready(). It reads @sigd with no synchronisation, so it can race with a close of the daemon socket:sigd_close()clears @sigd and the socket is then torn down and freed.Holding a reference on the socket is not enough to make this safe. The daemon fd close runs
__sock_release(), which frees the struct socket -- and the wait queue thatsk->sk_wqpoints at -- viaiput()once ->release() has returned.sk_data_ready()(sock_def_readable()) then dereferences a freedsk_wq:KASAN: null-ptr-deref in range [0x0000000000000188-0x000000000000018f] RIP: 0010:sigd_put_skb (net/atm/signaling.c:65) sigd_enq2 (net/atm/signaling.c:228) sigd_enq (net/atm/signaling.c:237) svc_bind (net/atm/svc.c:135) __sys_bind __x64_sys_bind do_syscall_64Fix it on both sides.
sigd_close()now callssock_orphan(), which undersk_callback_locksetsSOCK_DEADand clearssk_wqbefore the socket is freed.sigd_put_skb()latches @sigd withREAD_ONCE(), pins the socket withfind_get_vcc(), and then takessk_callback_lock; it delivers only while the socket is still the signalling daemon and not yetSOCK_DEAD, otherwise it drops the skb.sk_callback_lockis used rather thanlock_sock()becausesigd_put_skb()can be reached fromvcc_sendmsg()->sigd_send(), which already holdslock_sock()on the daemon socket.The
READ_ONCE(sigd)!= vcc check also rejects a reused VCC: a caller preempted between latching @sigd and validating it infind_get_vcc()could otherwise match a freed-and-reallocated, live (non-SOCK_DEAD) VCC. This needs a concurrent (privileged) daemon teardown, so it is hardening.Triggering the use-after-free requires
CAP_NET_ADMINandCAP_SYS_RAWIOto attach the daemon.Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")Reported-by Xiang Mei <[email protected]>Assisted-by Claude:claude-opus-4-8Signed-off-by Weiming Shi <[email protected]>---
v2:
- Also bail whenREAD_ONCE(sigd)!= vcc, rejecting a reused VCC that a preempted caller could match infind_get_vcc()(hardening, no stable).net/atm/signaling.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-)net/atm/signaling.c +25
-4@@ -54,14 +54,32 @@ static struct atm_vcc *find_get_vcc(struct atm_vcc *vcc)static void sigd_put_skb (struct sk_buff *skb) { -if (!sigd) { +struct atm_vcc *vcc; +struct sock *sk; + + vcc =find_get_vcc (READ_ONCE (sigd)); +if (!vcc) {pr_debug ("atmsvc: no signaling daemon\n" );kfree_skb (skb);return ; } -atm_force_charge (sigd, skb->truesize); -skb_queue_tail (&sk_atm (sigd)->sk_receive_queue, skb); -sk_atm (sigd)->sk_data_ready (sk_atm (sigd)); + sk =sk_atm (vcc); + +/* Pairs with sock_orphan() in sigd_close(). */ +read_lock_bh (&sk->sk_callback_lock); +/* Bail if torn down, or if the slot was reused by another VCC. */ +if (sock_flag (sk,SOCK_DEAD ) ||READ_ONCE (sigd) != vcc) { +read_unlock_bh (&sk->sk_callback_lock); +sock_put (sk); +kfree_skb (skb); +return ; + } +atm_force_charge (vcc, skb->truesize); +skb_queue_tail (&sk->sk_receive_queue, skb); + sk->sk_data_ready (sk); +read_unlock_bh (&sk->sk_callback_lock); + +sock_put (sk); }static void modify_qos (struct atm_vcc *vcc,struct atmsvc_msg *msg)@@ -257,6 +275,9 @@ static void sigd_close(struct atm_vcc *vcc)pr_err ("closing with requests pending\n" );skb_queue_purge (&sk_atm (vcc)->sk_receive_queue); +/* Make a concurrent sigd_put_skb() observe SOCK_DEAD and bail. */ +sock_orphan (sk_atm (vcc)); +read_lock (&vcc_sklist_lock);for (i =0 ; i <VCC_HTABLE_SIZE ; ++i) {struct hlist_head *head = &vcc_hash[i];
Jakub Kicinski
maintainerQuoted 5 lines · On Wed, 10 Jun 2026 00:21:08 +0800 Weiming Shi wrote:
sigd_put_skb()delivers a signalling message to the daemon socket named by the global @sigd pointer, ending in a call tosk_data_ready(). It reads @sigd with no synchronisation, so it can race with a close of the daemon socket:sigd_close()clears @sigd and the socket is then torn down and freed.
Hm, we intend to only retain the portions of the ATM stack which are still used in PPPoATM and ADSL. I don't believe the signaling stuff is used there. I will post a patch to delete this code.
Signature · 1 line
pw-bot: nap
Quoted 10 lines · Jakub Kicinski
Quoted 5 lines · On Wed, 10 Jun 2026 00:21:08 +0800 Weiming Shi wrote:
sigd_put_skb()delivers a signalling message to the daemon socket named by the global @sigd pointer, ending in a call tosk_data_ready(). It reads @sigd with no synchronisation, so it can race with a close of the daemon socket:sigd_close()clears @sigd and the socket is then torn down and freed.Hm, we intend to only retain the portions of the ATM stack which are still used in PPPoATM and ADSL. I don't believe the signaling stuff is used there. I will post a patch to delete this code.
Thanks, that makes sense to me.