netem: fix skb_orphan_partial()

I should have known that lowering skb->truesize was dangerous :/

In case packets are not leaving the host via a standard Ethernet device,
but looped back to local sockets, bad things can happen, as reported
by Michael Madsen ( https://bugzilla.kernel.org/show_bug.cgi?id=195713 )

So instead of tweaking skb->truesize, lets change skb->destructor
and keep a reference on the owner socket via its sk_refcnt.

Fixes: f2f872f927 ("netem: Introduce skb_orphan_partial() helper")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Michael Madsen <mkm@nabto.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Eric Dumazet 2017-05-11 15:24:41 -07:00 committed by David S. Miller
parent 4e3c60ed2f
commit f6ba8d33cf
1 changed files with 8 additions and 12 deletions

View File

@ -1803,28 +1803,24 @@ EXPORT_SYMBOL(skb_set_owner_w);
* delay queue. We want to allow the owner socket to send more
* packets, as if they were already TX completed by a typical driver.
* But we also want to keep skb->sk set because some packet schedulers
* rely on it (sch_fq for example). So we set skb->truesize to a small
* amount (1) and decrease sk_wmem_alloc accordingly.
* rely on it (sch_fq for example).
*/
void skb_orphan_partial(struct sk_buff *skb)
{
/* If this skb is a TCP pure ACK or already went here,
* we have nothing to do. 2 is already a very small truesize.
*/
if (skb->truesize <= 2)
if (skb_is_tcp_pure_ack(skb))
return;
/* TCP stack sets skb->ooo_okay based on sk_wmem_alloc,
* so we do not completely orphan skb, but transfert all
* accounted bytes but one, to avoid unexpected reorders.
*/
if (skb->destructor == sock_wfree
#ifdef CONFIG_INET
|| skb->destructor == tcp_wfree
#endif
) {
atomic_sub(skb->truesize - 1, &skb->sk->sk_wmem_alloc);
skb->truesize = 1;
struct sock *sk = skb->sk;
if (atomic_inc_not_zero(&sk->sk_refcnt)) {
atomic_sub(skb->truesize, &sk->sk_wmem_alloc);
skb->destructor = sock_efree;
}
} else {
skb_orphan(skb);
}