wireguard: device: use exit_rtnl callback instead of manual rtnl_lock in pre_exit
PATCH 1/1 wireguard: device: use exit_rtnl callback instead of manual rtnl_lock in pre_exit
From: Shardul Bankar <[email protected]>
[ Upstream commit 60a25ef8dacb3566b1a8c4de00572a498e2a3bf9 ]
wg_netns_pre_exit() manually acquires rtnl_lock() inside the pernet .pre_exit callback. This causes a hung task when another thread holds rtnl_mutex - the cleanup_net workqueue (or the setup_net failure rollback path) blocks indefinitely in wg_netns_pre_exit() waiting to acquire the lock.
Convert to .exit_rtnl, introduced in commit 7a60d91c690b ("net:
Add ->exit_rtnl() hook to struct pernet_operations."), where the framework already holds RTNL and batches all callbacks under a single rtnl_lock()/rtnl_unlock() pair, eliminating the contention window.
The rcu_assign_pointer(wg->creating_net, NULL) is safe to move from .pre_exit to .exit_rtnl (which runs after synchronize_rcu()) because all RCU readers of creating_net either use maybe_get_net()
- which returns NULL for a dying namespace with zero refcount - or access net->user_ns which remains valid throughout the entire ops_undo_list sequence.
--- drivers/net/wireguard/device.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-)
drivers/net/wireguard/device.c +3 -5
@@ -406,12 +406,11 @@ static struct rtnl_link_ops link_ops __read_mostly = { .newlink = wg_newlink, }; -static void wg_netns_pre_exit (struct net *net) +static void __net_exit wg_netns_exit_rtnl(struct net *net, struct list_head *dev_kill_list) {struct wg_device *wg;struct wg_peer *peer; -rtnl_lock ();list_for_each_entry (wg, &device_list, device_list) {if (rcu_access_pointer (wg->creating_net) == net) {pr_debug ("%s: Creating namespace exiting\n" , wg->dev->name);
@@ -424,11 +423,10 @@ static void wg_netns_pre_exit(struct net *net)mutex_unlock (&wg->device_update_lock); } } -rtnl_unlock (); } -static struct pernet_operations pernet_ops = { - .pre_exit = wg_netns_pre_exit +static struct pernet_operations pernet_ops__read_mostly = { + .exit_rtnl = wg_netns_exit_rtnl };int __init wg_device_init (void )
Signature · 1 line
2.53.0
Hi all,
Am Donnerstag, 11. Juni 2026, 16:46:15 Mitteleuropäische Sommerzeit schrieb Heiko Stuebner:
Quoted 23 lines
From: Shardul Bankar <[email protected]>
[ Upstream commit
60a25ef8dacb3566b1a8c4de00572a498e2a3bf9]
wg_netns_pre_exit()manually acquiresrtnl_lock()inside the pernet .pre_exit callback. This causes a hung task when another thread holdsrtnl_mutex- thecleanup_networkqueue (or thesetup_netfailure rollback path) blocks indefinitely inwg_netns_pre_exit()waiting to acquire the lock.Convert to .exit_rtnl, introduced in commit
7a60d91c690b("net:
Add ->exit_rtnl() hook to structpernet_operations."), where the framework already holds RTNL and batches all callbacks under a single rtnl_lock()/rtnl_unlock() pair, eliminating the contention window.The
rcu_assign_pointer(wg->creating_net, NULL)is safe to move from .pre_exit to .exit_rtnl (which runs aftersynchronize_rcu()) because all RCU readers ofcreating_neteither usemaybe_get_net()
- which returns NULL for a dying namespace with zero refcount - or accessnet->user_nswhich remains valid throughout the entireops_undo_listsequence.
Quoted 10 lines
Reported-by [email protected]Signed-off-by Shardul Bankar <[email protected]>[ Jason: added __net_exit and __read_mostly annotations that were missing. ]Fixes 900575aa33a3 ("wireguard: device: avoid circular netns references")Signed-off-by Jason A. Donenfeld <[email protected]>Signed-off-by Jakub Kicinski <[email protected]>Signed-off-by Heiko Stuebner <[email protected]>
sorry about the mail-noise, but I messed up sending this :-( Will re-try once I fixed my mistake.
Heiko