Commit 54764bb6 authored by Eric Dumazet's avatar Eric Dumazet Committed by David S. Miller
Browse files

ipv4: Restore old dst_free() behavior. 



commit 404e0a8b (net: ipv4: fix RCU races on dst refcounts) tried
to solve a race but added a problem at device/fib dismantle time :

We really want to call dst_free() as soon as possible, even if sockets
still have dst in their cache.
dst_release() calls in free_fib_info_rcu() are not welcomed.

Root of the problem was that now we also cache output routes (in
nh_rth_output), we must use call_rcu() instead of call_rcu_bh() in
rt_free(), because output route lookups are done in process context.

Based on feedback and initial patch from David Miller (adding another
call_rcu_bh() call in fib, but it appears it was not the right fix)

I left the inet_sk_rx_dst_set() helper and added __rcu attributes
to nh_rth_output and nh_rth_input to better document what is going on in
this code.

Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 5a0d513b

include/net/dst.h

+6 −1
Original line number Diff line number Diff line
@@ -61,7 +61,6 @@ struct dst_entry { 
#define DST_NOPEER		0x0040

#define DST_FAKE_RTABLE		0x0080

#define DST_XFRM_TUNNEL		0x0100

#define DST_RCU_FREE		0x0200



	unsigned short		pending_confirm;
@@ -383,6 +382,12 @@ static inline void dst_free(struct dst_entry *dst) 
	__dst_free(dst);

}



static inline void dst_rcu_free(struct rcu_head *head)

{

	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);

	dst_free(dst);

}



static inline void dst_confirm(struct dst_entry *dst)

{

	dst->pending_confirm = 1;

include/net/inet_sock.h

+3 −7
Original line number Diff line number Diff line
@@ -253,13 +253,9 @@ static inline void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb 
{

	struct dst_entry *dst = skb_dst(skb);



	if (atomic_inc_not_zero(&dst->__refcnt)) {

		if (!(dst->flags & DST_RCU_FREE))

			dst->flags |= DST_RCU_FREE;



	dst_hold(dst);

	sk->sk_rx_dst = dst;

	inet_sk(sk)->rx_dst_ifindex = skb->skb_iif;

}

}



#endif	/* _INET_SOCK_H */

include/net/ip_fib.h

+2 −2
Original line number Diff line number Diff line
@@ -81,8 +81,8 @@ struct fib_nh { 
	__be32			nh_gw;

	__be32			nh_saddr;

	int			nh_saddr_genid;

	struct rtable		*nh_rth_output;

	struct rtable		*nh_rth_input;

	struct rtable __rcu	*nh_rth_output;

	struct rtable __rcu	*nh_rth_input;

	struct fnhe_hash_bucket	*nh_exceptions;

};

net/core/dst.c

+5 −21
Original line number Diff line number Diff line
@@ -258,15 +258,6 @@ again: 
}

EXPORT_SYMBOL(dst_destroy);



static void dst_rcu_destroy(struct rcu_head *head)

{

	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);



	dst = dst_destroy(dst);

	if (dst)

		__dst_free(dst);

}



void dst_release(struct dst_entry *dst)

{

	if (dst) {
@@ -274,17 +265,13 @@ void dst_release(struct dst_entry *dst) 


		newrefcnt = atomic_dec_return(&dst->__refcnt);

		WARN_ON(newrefcnt < 0);

		if (unlikely(dst->flags & (DST_NOCACHE | DST_RCU_FREE)) && !newrefcnt) {

			if (dst->flags & DST_RCU_FREE) {

				call_rcu_bh(&dst->rcu_head, dst_rcu_destroy);

			} else {

		if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) {

			dst = dst_destroy(dst);

			if (dst)

				__dst_free(dst);

		}

	}

}

}

EXPORT_SYMBOL(dst_release);



u32 *dst_cow_metrics_generic(struct dst_entry *dst, unsigned long old)
@@ -333,14 +320,11 @@ EXPORT_SYMBOL(__dst_destroy_metrics_generic); 
 */

void skb_dst_set_noref(struct sk_buff *skb, struct dst_entry *dst)

{

	bool hold;



	WARN_ON(!rcu_read_lock_held() && !rcu_read_lock_bh_held());

	/* If dst not in cache, we must take a reference, because

	 * dst_release() will destroy dst as soon as its refcount becomes zero

	 */

	hold = (dst->flags & (DST_NOCACHE | DST_RCU_FREE)) == DST_NOCACHE;

	if (unlikely(hold)) {

	if (unlikely(dst->flags & DST_NOCACHE)) {

		dst_hold(dst);

		skb_dst_set(skb, dst);

	} else {

net/decnet/dn_route.c

+0 −6
Original line number Diff line number Diff line
@@ -184,12 +184,6 @@ static __inline__ unsigned int dn_hash(__le16 src, __le16 dst) 
	return dn_rt_hash_mask & (unsigned int)tmp;

}



static inline void dst_rcu_free(struct rcu_head *head)

{

	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);

	dst_free(dst);

}



static inline void dnrt_free(struct dn_route *rt)

{

	call_rcu_bh(&rt->dst.rcu_head, dst_rcu_free);

CRA Git | Maintained and supported by SUSTech CRA and CCSE