[lkml]   [2019]   [Nov]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[PATCH AUTOSEL 4.19 135/209] xsk: proper AF_XDP socket teardown ordering
From: Björn Töpel <>

[ Upstream commit 541d7fdd7694560404c502f64298a90ffe017e6b ]

The AF_XDP socket struct can exist in three different, implicit
states: setup, bound and released. Setup is prior the socket has been
bound to a device. Bound is when the socket is active for receive and
send. Released is when the process/userspace side of the socket is
released, but the sock object is still lingering, e.g. when there is a
reference to the socket in an XSKMAP after process termination.

The Rx fast-path code uses the "dev" member of struct xdp_sock to
check whether a socket is bound or relased, and the Tx code uses the
struct xdp_umem "xsk_list" member in conjunction with "dev" to
determine the state of a socket.

However, the transition from bound to released did not tear the socket
down in correct order.

On the Rx side "dev" was cleared after synchronize_net() making the
synchronization useless. On the Tx side, the internal queues were
destroyed prior removing them from the "xsk_list".

This commit corrects the cleanup order, and by doing so
xdp_del_sk_umem() can be simplified and one synchronize_net() can be

Fixes: 965a99098443 ("xsk: add support for bind for Rx")
Fixes: ac98d8aab61b ("xsk: wire upp Tx zero-copy functions")
Reported-by: Jesper Dangaard Brouer <>
Signed-off-by: Björn Töpel <>
Acked-by: Song Liu <>
Signed-off-by: Daniel Borkmann <>
Signed-off-by: Sasha Levin <>
net/xdp/xdp_umem.c | 11 +++--------
net/xdp/xsk.c | 13 ++++++++-----
2 files changed, 11 insertions(+), 13 deletions(-)

diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
index 8cab91c482ff5..d9117ab035f7c 100644
--- a/net/xdp/xdp_umem.c
+++ b/net/xdp/xdp_umem.c
@@ -32,14 +32,9 @@ void xdp_del_sk_umem(struct xdp_umem *umem, struct xdp_sock *xs)
unsigned long flags;

- if (xs->dev) {
- spin_lock_irqsave(&umem->xsk_list_lock, flags);
- list_del_rcu(&xs->list);
- spin_unlock_irqrestore(&umem->xsk_list_lock, flags);
- if (umem->zc)
- synchronize_net();
- }
+ spin_lock_irqsave(&umem->xsk_list_lock, flags);
+ list_del_rcu(&xs->list);
+ spin_unlock_irqrestore(&umem->xsk_list_lock, flags);

int xdp_umem_query(struct net_device *dev, u16 queue_id)
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 661504042d304..ff15207036dc5 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -343,12 +343,18 @@ static int xsk_release(struct socket *sock)

if (xs->dev) {
+ struct net_device *dev = xs->dev;
/* Wait for driver to stop using the xdp socket. */
- synchronize_net();
- dev_put(xs->dev);
+ xdp_del_sk_umem(xs->umem, xs);
xs->dev = NULL;
+ synchronize_net();
+ dev_put(dev);

+ xskq_destroy(xs->rx);
+ xskq_destroy(xs->tx);
sock->sk = NULL;

@@ -707,9 +713,6 @@ static void xsk_destruct(struct sock *sk)
if (!sock_flag(sk, SOCK_DEAD))

- xskq_destroy(xs->rx);
- xskq_destroy(xs->tx);
- xdp_del_sk_umem(xs->umem, xs);

 \ /
  Last update: 2019-11-13 03:22    [W:0.383 / U:2.584 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site