net: Do not clear the sock TX queue in sk_set_socket()

[ Upstream commit 41b14fb8724d5a4b382a63cb4a1a61880347ccb8 ]

Clearing the sock TX queue in sk_set_socket() might cause unexpected
out-of-order transmit when called from sock_orphan(), as outstanding
packets can pick a different TX queue and bypass the ones already queued.

This is undesired in general. More specifically, it breaks the in-order
scheduling property guarantee for device-offloaded TLS sockets.

Remove the call to sk_tx_queue_clear() in sk_set_socket(), and add it
explicitly only where needed.

Fixes: e022f0b4a0 ("net: Introduce sk_tx_queue_mapping")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Tariq Toukan 2020-06-22 23:26:04 +03:00 committed by Sasha Levin
parent c91b5162af
commit 31394a198c
2 changed files with 2 additions and 1 deletions

View file

@ -1678,7 +1678,6 @@ static inline int sk_tx_queue_get(const struct sock *sk)
static inline void sk_set_socket(struct sock *sk, struct socket *sock)
{
sk_tx_queue_clear(sk);
sk->sk_socket = sock;
}

View file

@ -1435,6 +1435,7 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority,
sock_update_classid(sk);
sock_update_netprioidx(sk);
sk_tx_queue_clear(sk);
}
return sk;
@ -1601,6 +1602,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
*/
sk_refcnt_debug_inc(newsk);
sk_set_socket(newsk, NULL);
sk_tx_queue_clear(newsk);
newsk->sk_wq = NULL;
sk_update_clone(sk, newsk);