From: Zheng Wang <zyytlz.wz@xxxxxxx>
[ Upstream commit f486893288f3e9b171b836f43853a6426515d800 ]
mptlan_probe() calls mpt_register_lan_device() which initializes the
&priv->post_buckets_task workqueue. A call to
mpt_lan_wake_post_buckets_task() will subsequently start the work.
During driver unload in mptlan_remove() the following race may occur:
CPU0 CPU1
|mpt_lan_post_receive_buckets_work()
mptlan_remove() |
free_netdev() |
kfree(dev); |
|
| dev->mtu
| //use
Fix this by finishing the work prior to cleaning up in mptlan_remove().
[mkp: we really should remove mptlan instead of attempting to fix it]
Signed-off-by: Zheng Wang <zyytlz.wz@xxxxxxx>
Link: https://lore.kernel.org/r/20230318081635.796479-1-zyytlz.wz@xxxxxxx
Signed-off-by: Martin K. Petersen <martin.petersen@xxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/message/fusion/mptlan.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/message/fusion/mptlan.c b/drivers/message/fusion/mptlan.c
index 3261cac762def..ec3ee356078db 100644
--- a/drivers/message/fusion/mptlan.c
+++ b/drivers/message/fusion/mptlan.c
@@ -1427,7 +1427,9 @@ mptlan_remove(struct pci_dev *pdev)
{
MPT_ADAPTER *ioc = pci_get_drvdata(pdev);
struct net_device *dev = ioc->netdev;
+ struct mpt_lan_priv *priv = netdev_priv(dev);
+ cancel_delayed_work_sync(&priv->post_buckets_task);
if(dev != NULL) {
unregister_netdev(dev);
free_netdev(dev);
--
2.39.2
