Re: [PATCH net] net: fix NULL pointer dereference in l3mdev_l3_rcv

From
David Ahern <dsahern@kernel.org>
Date
2025-03-18 15:02:38
Also in
lkml
Reply to
Simon Horman, 2025-03-18
On 3/18/25 3:38 PM, Simon Horman wrote:
On Thu, Mar 13, 2025 at 09:27:13AM +0800, Wang Liang wrote:
quoted
When delete l3s ipvlan:

    ip link del link eth0 ipvlan1 type ipvlan mode l3s

This may cause a null pointer dereference:

    Call trace:
     ip_rcv_finish+0x48/0xd0
     ip_rcv+0x5c/0x100
     __netif_receive_skb_one_core+0x64/0xb0
     __netif_receive_skb+0x20/0x80
     process_backlog+0xb4/0x204
     napi_poll+0xe8/0x294
     net_rx_action+0xd8/0x22c
     __do_softirq+0x12c/0x354

This is because l3mdev_l3_rcv() visit dev->l3mdev_ops after
ipvlan_l3s_unregister() assign the dev->l3mdev_ops to NULL. The process
like this:

    (CPU1)                     | (CPU2)
    l3mdev_l3_rcv()            |
      check dev->priv_flags:   |
        master = skb->dev;     |
                               |
                               | ipvlan_l3s_unregister()
                               |   set dev->priv_flags
                               |   dev->l3mdev_ops = NULL;
                               |
      visit master->l3mdev_ops |

Add lock for dev->priv_flags and dev->l3mdev_ops is too expensive. Resolve
this issue by add check for master->l3mdev_ops.
Hi Wang Liang,

It seems to me that checking master->l3mdev_ops like this is racy.
vrf device leaves the l3mdev ops set; that is probably the better way to go.
Thread (7 messages) 7 messages, 3 authors, last reply 2025-03-21
Keyboard shortcuts
hback out one level
jnext message in thread / next series-diff position
kprevious message in thread / previous series-diff position
ldrill in
Escclose help / fold thread tree
?toggle this help