[PATCH v2] ax25: Fix refcount imbalance on inbound connections

The first version of this patch was posted only to the linux-hams
mailing list. It has been difficult to get the patch reviewed, but the
patch has now been tested successfully by three people (that includes
me) who have all verified that it prevents the crashes that were
previously plaguing inbound ax.25 connections.

Related discussions:

- https://marc.info/?l=linux-hams&m=171629285223248&w=2
- https://marc.info/?l=linux-hams&m=171270115728031&w=2

8------------------------------------------------------8<

When releasing a socket in ax25_release(), we call netdev_put() to
decrease the refcount on the associated ax.25 device. However, the
execution path for accepting an incoming connection never calls
netdev_hold(). This imbalance leads to refcount errors, and ultimately
to kernel crashes.

A typical call trace for the above situation looks like this:

    Call Trace:
    <TASK>
    ? show_regs+0x64/0x70
    ? __warn+0x83/0x120
    ? refcount_warn_saturate+0xb2/0x100
    ? report_bug+0x158/0x190
    ? prb_read_valid+0x20/0x30
    ? handle_bug+0x3e/0x70
    ? exc_invalid_op+0x1c/0x70
    ? asm_exc_invalid_op+0x1f/0x30
    ? refcount_warn_saturate+0xb2/0x100
    ? refcount_warn_saturate+0xb2/0x100
    ax25_release+0x2ad/0x360
    __sock_release+0x35/0xa0
    sock_close+0x19/0x20
    [...]

On reboot (or any attempt to remove the interface), the kernel gets
stuck in an infinite loop:

    unregister_netdevice: waiting for ax0 to become free. Usage count = 0

This patch corrects these issues by ensuring that we call netdev_hold()
and ax25_dev_hold() for new connections in ax25_accept(), balancing the
calls to netdev_put() and ax25_dev_put() in ax25_release.

Fixes: 7d8a3a477b
Signed-off-by: Lars Kellogg-Stedman <redacted>
---
 net/ax25/af_ax25.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c
index 8077cf2ee44..ff921272d40 100644
--- a/net/ax25/af_ax25.c
+++ b/net/ax25/af_ax25.c

@@ -1381,6 +1381,8 @@ static int ax25_accept(struct socket *sock, struct socket *newsock,
 	DEFINE_WAIT(wait);
 	struct sock *sk;
 	int err = 0;
+	ax25_cb *ax25;
+	ax25_dev *ax25_dev;
 
 	if (sock->state != SS_UNCONNECTED)
 		return -EINVAL;

@@ -1434,6 +1436,10 @@ static int ax25_accept(struct socket *sock, struct socket *newsock,
 	kfree_skb(skb);
 	sk_acceptq_removed(sk);
 	newsock->state = SS_CONNECTED;
+	ax25 = sk_to_ax25(newsk);
+	ax25_dev = ax25->ax25_dev;
+	netdev_hold(ax25_dev->dev, &ax25->dev_tracker, GFP_ATOMIC);
+	ax25_dev_hold(ax25_dev);
 
 out:
 	release_sock(sk);

-- 
2.45.1

-- 
Lars Kellogg-Stedman <lars@oddbit.com> | larsks @ {irc,twitter,github}
http://blog.oddbit.com/                | N1LKS