[PATCH 1/1] qede: sync udp_tunnel ports outside qede_lock in the recovery path
From: Denis V. Lunev <hidden>
Date: 2026-07-26 10:43:16
Subsystem:
networking drivers, qlogic ql4xxx ethernet driver, the rest · Maintainers:
Andrew Lunn, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds
A TX timeout on a qede NIC that has VXLAN/GENEVE tunnel ports
configured wedges the rtnetlink control plane of the whole machine:
NETDEV WATCHDOG: ens6f1 (qede): transmit queue 2 timed out 10226 ms
[qede_tx_timeout:586(ens6f1)]TX timeout on queue 2!
[qede_recovery_handler:2665(ens6f0)]Starting a recovery process
The recovery path deadlocks on the driver's own mutex:
qede_sp_task
rtnl_lock()
mutex_lock(&edev->qede_lock) <- taken
qede_recovery_handler
qede_load
udp_tunnel_nic_reset_ntf
__udp_tunnel_nic_device_sync
info->sync_table == qede_udp_tunnel_sync
mutex_lock(&edev->qede_lock) <- same task: deadlock
The mutex is not recursive, so the kworker blocks on itself with
rtnl_lock held, and neither lock is ever released. Every task that
calls rtnl_lock() afterwards (ip, ovs-vswitchd, lldpad, IPv6
addrconf, sshd) blocks forever while the node still answers ping.
In a vmcore from an affected production node rtnl_mutex.owner
decodes to the very kworker blocked at the innermost mutex_lock()
above.
Re-sync the tunnel ports from qede_sp_task() after the internal lock
is dropped, still under rtnl_lock as the udp_tunnel API requires.
This mirrors qede_open(), which calls udp_tunnel_nic_reset_ntf()
under rtnl without the internal lock.
qede_recovery_handler() now returns whether it has successfully
reloaded an open device, and the caller re-syncs the ports only in
that case. This keeps the old gating exactly: a device that was down
or a failed recovery returns false, as those paths never reached the
udp_tunnel_nic_reset_ntf() call before either.
This was the only user of the qede_lock()/qede_unlock() helpers, so
remove them.
Fixes: 8cd160a29415 ("qede: convert to new udp_tunnel_nic infra")
Signed-off-by: Denis V. Lunev <redacted>
CC: Andrew Lunn <andrew+netdev@lunn.ch>
CC: "David S. Miller" <davem@davemloft.net>
CC: Eric Dumazet <edumazet@google.com>
CC: Jakub Kicinski <kuba@kernel.org>
CC: Paolo Abeni <pabeni@redhat.com>
---
drivers/net/ethernet/qlogic/qede/qede_main.c | 44 ++++++++++----------
1 file changed, 22 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c
index cb0ae0650905..7ed17faced54 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_main.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_main.c@@ -107,7 +107,7 @@ static void qede_remove(struct pci_dev *pdev); static void qede_shutdown(struct pci_dev *pdev); static void qede_link_update(void *dev, struct qed_link_output *link); static void qede_schedule_recovery_handler(void *dev); -static void qede_recovery_handler(struct qede_dev *edev); +static bool qede_recovery_handler(struct qede_dev *edev); static void qede_schedule_hw_err_handler(void *dev, enum qed_hw_err_type err_type); static void qede_get_eth_tlv_data(void *edev, void *data);
@@ -1043,21 +1043,6 @@ void __qede_unlock(struct qede_dev *edev) mutex_unlock(&edev->qede_lock); } -/* This version of the lock should be used when acquiring the RTNL lock is also - * needed in addition to the internal qede lock. - */ -static void qede_lock(struct qede_dev *edev) -{ - rtnl_lock(); - __qede_lock(edev); -} - -static void qede_unlock(struct qede_dev *edev) -{ - __qede_unlock(edev); - rtnl_unlock(); -} - static void qede_periodic_task(struct work_struct *work) { struct qede_dev *edev = container_of(work, struct qede_dev,
@@ -1094,6 +1079,8 @@ static void qede_sp_task(struct work_struct *work) */ if (test_and_clear_bit(QEDE_SP_RECOVERY, &edev->sp_flags)) { + bool reloaded; + cancel_delayed_work_sync(&edev->periodic_task); #ifdef CONFIG_QED_SRIOV /* SRIOV must be disabled outside the lock to avoid a deadlock.
@@ -1102,9 +1089,17 @@ static void qede_sp_task(struct work_struct *work) if (pci_num_vf(edev->pdev)) qede_sriov_configure(edev->pdev, 0); #endif - qede_lock(edev); - qede_recovery_handler(edev); - qede_unlock(edev); + rtnl_lock(); + __qede_lock(edev); + reloaded = qede_recovery_handler(edev); + __qede_unlock(edev); + + /* The udp_tunnel core synchronously calls back into + * qede_udp_tunnel_sync(), which takes the qede lock. + */ + if (reloaded) + udp_tunnel_nic_reset_ntf(edev->ndev); + rtnl_unlock(); } __qede_lock(edev);
@@ -2645,9 +2640,13 @@ static void qede_recovery_failed(struct qede_dev *edev) edev->ops->common->set_power_state(edev->cdev, PCI_D3hot); } -static void qede_recovery_handler(struct qede_dev *edev) +/* Returns true if an open device was successfully reloaded and its + * udp_tunnel ports need to be re-synced by the caller. + */ +static bool qede_recovery_handler(struct qede_dev *edev) { u32 curr_state = edev->state; + bool reloaded = false; int rc; DP_NOTICE(edev, "Starting a recovery process\n");
@@ -2677,17 +2676,18 @@ static void qede_recovery_handler(struct qede_dev *edev) goto err; qede_config_rx_mode(edev->ndev); - udp_tunnel_nic_reset_ntf(edev->ndev); + reloaded = true; } edev->state = curr_state; DP_NOTICE(edev, "Recovery handling is done\n"); - return; + return reloaded; err: qede_recovery_failed(edev); + return false; } static void qede_atomic_hw_err_handler(struct qede_dev *edev)
--
2.53.0