net: introduce and use route hint

PATCH In review Paolo Abeni 4 participants 13 messages Nov 20, 2019 – Nov 21, 2019

Subsystems (3)

Networking Network drivers IPv4/IPv6

lore

Share & tools

lore

Paolo Abeni

author maintainer

2019-11-20 12:48 UTC #0 lore raw

v4 cover net: introduce and use route hint

This series leverages the listification infrastructure to avoid unnecessary route lookup on ingress packets. In absence of custom rules, packets with equal daddr will usually land on the same dst.

When processing packet bursts (lists) we can easily reference the previous dst entry. When we hit the 'same destination' condition we can avoid the route lookup, coping the already available dst.

Detailed performance numbers are available in the individual commit messages.

v3 -> v4:
- move helpers to their own patches (Eric D.)
- enable hints for SUBTREE builds (David A.)
- re-enable hints for ipv4 forward (David A.)

v2 -> v3:
- use fib*_has_custom_rules() helpers (David A.)
- add ip*_extract_route_hint() helper (Edward C.)
- use prev skb as hint instead of copying data (Willem )

v1 -> v2:
- fix build issue with !CONFIG_IP*_MULTIPLE_TABLES
- fix potential race in ip6_list_rcv_finish()

Paolo Abeni (5):
ipv6: add fib6_has_custom_rules() helper
ipv6: keep track of routes using src
ipv6: introduce and uses route look hints for list input.
ipv4: move fib4_has_custom_rules() helper to public header
ipv4: use dst hint for ipv4 list receive

 include/net/ip6_fib.h    | 39 +++++++++++++++++++++++++++++++++++++
 include/net/ip_fib.h     | 10 ++++++++++
 include/net/netns/ipv6.h |  3 +++
 include/net/route.h      |  4 ++++
 net/ipv4/fib_frontend.c  | 10 ----------
 net/ipv4/ip_input.c      | 35 +++++++++++++++++++++++++++++----
 net/ipv4/route.c         | 42 ++++++++++++++++++++++++++++++++++++++++
 net/ipv6/ip6_fib.c       |  4 ++++
 net/ipv6/ip6_input.c     | 26 +++++++++++++++++++++++--
 net/ipv6/route.c         |  3 +++
 10 files changed, 160 insertions(+), 16 deletions(-)

Signature · 1 line

2.21.0

Paolo Abeni

author maintainer

2019-11-20 12:48 UTC #1 lore raw

PATCH v4 1/5 ipv6: add fib6_has_custom_rules() helper

It wraps the namespace field with the same name, to easily access it regardless of build options.

Signed-off-by Paolo Abeni <[email protected]>

---
 include/net/ip6_fib.h | 9 +++++++++
 1 file changed, 9 insertions(+)

include/net/ip6_fib.h +9 -0

index 5d1615463138..8ac3a59e5126 100644
@@ -502,6 +502,11 @@ static inline bool fib6_metric_locked(struct fib6_info *f6i, int metric)
 }
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
+static inline bool fib6_has_custom_rules(const struct net *net)
+{
+	return net->ipv6.fib6_has_custom_rules;
+}
+
 int fib6_rules_init(void);
 void fib6_rules_cleanup(void);
 bool fib6_rule_default(const struct fib_rule *rule);

@@ -527,6 +532,10 @@ static inline bool fib6_rules_early_flow_dissect(struct net *net,
 	return true;
 }
 #else
+static inline bool fib6_has_custom_rules(const struct net *net)
+{
+	return false;
+}
 static inline int               fib6_rules_init(void)
 {
 	return 0;

Signature · 1 line

2.21.0

David Ahern

Reviewed-by on v4 1/5

2019-11-21 20:07 UTC #2 lore raw

Re: [PATCH net-next v4 1/5] ipv6: add fib6_has_custom_rules() helper

Quoted 10 lines · Paolo Abeni

It wraps the namespace field with the same name, to easily access it regardless of build options.
Suggested-by David Ahern <[email protected]>
Suggested-by Eric Dumazet <[email protected]>
Signed-off-by Paolo Abeni <[email protected]>
---
 include/net/ip6_fib.h | 9 +++++++++
 1 file changed, 9 insertions(+)

Reviewed-by David Ahern <[email protected]>

Paolo Abeni

author maintainer

2019-11-20 12:48 UTC #3 lore raw

PATCH v4 2/5 ipv6: keep track of routes using src

In reply to Paolo Abeni

Use a per namespace counter, increment it on successful creation of any route using the source address, decrement it on deletion of such routes.

This allows us to check easily if the routing decision in the current namespace depends on the packet source. Will be used by the next patch.

Signed-off-by Paolo Abeni <[email protected]>

---
 include/net/ip6_fib.h    | 30 ++++++++++++++++++++++++++++++
 include/net/netns/ipv6.h |  3 +++
 net/ipv6/ip6_fib.c       |  4 ++++
 net/ipv6/route.c         |  3 +++
 4 files changed, 40 insertions(+)

include/net/ip6_fib.h +30 -0

index 8ac3a59e5126..f1535f172935 100644
@@ -90,7 +90,32 @@ struct fib6_gc_args {
 
 #ifndef CONFIG_IPV6_SUBTREES
 #define FIB6_SUBTREE(fn)	NULL
+
+static inline bool fib6_routes_require_src(const struct net *net)
+{
+	return false;
+}
+
+static inline void fib6_routes_require_src_inc(struct net *net) {}
+static inline void fib6_routes_require_src_dec(struct net *net) {}
+
 #else
+
+static inline bool fib6_routes_require_src(const struct net *net)
+{
+	return net->ipv6.fib6_routes_require_src > 0;
+}
+
+static inline void fib6_routes_require_src_inc(struct net *net)
+{
+	net->ipv6.fib6_routes_require_src++;
+}
+
+static inline void fib6_routes_require_src_dec(struct net *net)
+{
+	net->ipv6.fib6_routes_require_src--;
+}
+
 #define FIB6_SUBTREE(fn)	(rcu_dereference_protected((fn)->subtree, 1))
 #endif

@@ -212,6 +237,11 @@ static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst)
 	return ((struct rt6_info *)dst)->rt6i_idev;
 }
 
+static inline bool fib6_requires_src(const struct fib6_info *rt)
+{
+	return rt->fib6_src.plen > 0;
+}
+
 static inline void fib6_clean_expires(struct fib6_info *f6i)
 {
 	f6i->fib6_flags &= ~RTF_EXPIRES;

include/net/netns/ipv6.h +3 -0

index 022a0fd1a5a4..5ec054473d81 100644
@@ -83,6 +83,9 @@ struct netns_ipv6 {
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	unsigned int		fib6_rules_require_fldissect;
 	bool			fib6_has_custom_rules;
+#ifdef CONFIG_IPV6_SUBTREES
+	unsigned int		fib6_routes_require_src;
+#endif
 	struct rt6_info         *ip6_prohibit_entry;
 	struct rt6_info         *ip6_blk_hole_entry;
 	struct fib6_table       *fib6_local_tbl;

net/ipv6/ip6_fib.c +4 -0

index f66bc2af4e9d..7bae6a91b487 100644
@@ -1461,6 +1461,8 @@ int fib6_add(struct fib6_node *root, struct fib6_info *rt,
 		}
 #endif
 		goto failure;
+	} else if (fib6_requires_src(rt)) {
+		fib6_routes_require_src_inc(info->nl_net);
 	}
 	return err;

@@ -1933,6 +1935,8 @@ int fib6_del(struct fib6_info *rt, struct nl_info *info)
 		struct fib6_info *cur = rcu_dereference_protected(*rtp,
 					lockdep_is_held(&table->tb6_lock));
 		if (rt == cur) {
+			if (fib6_requires_src(cur))
+				fib6_routes_require_src_dec(info->nl_net);
 			fib6_del_route(table, fn, rtp, info);
 			return 0;
 		}

net/ipv6/route.c +3 -0

index edcb52543518..c92b367e058d 100644
@@ -6199,6 +6199,9 @@ static int __net_init ip6_route_net_init(struct net *net)
 	dst_init_metrics(&net->ipv6.ip6_blk_hole_entry->dst,
 			 ip6_template_metrics, true);
 	INIT_LIST_HEAD(&net->ipv6.ip6_blk_hole_entry->rt6i_uncached);
+#ifdef CONFIG_IPV6_SUBTREES
+	net->ipv6.fib6_routes_require_src = 0;
+#endif
 #endif
 
 	net->ipv6.sysctl.flush_delay = 0;

Signature · 1 line

2.21.0

David Ahern

Reviewed-by on v4 2/5

2019-11-21 20:09 UTC #4 lore raw

Re: [PATCH net-next v4 2/5] ipv6: keep track of routes using src

Quoted 8 lines · Paolo Abeni

Use a per namespace counter, increment it on successful creation of any route using the source address, decrement it on deletion of such routes.
This allows us to check easily if the routing decision in the current namespace depends on the packet source. Will be used by the next patch.

Quoted 9 lines

Suggested-by David Ahern <[email protected]>
Signed-off-by Paolo Abeni <[email protected]>
---
 include/net/ip6_fib.h    | 30 ++++++++++++++++++++++++++++++
 include/net/netns/ipv6.h |  3 +++
 net/ipv6/ip6_fib.c       |  4 ++++
 net/ipv6/route.c         |  3 +++
 4 files changed, 40 insertions(+)

Reviewed-by David Ahern <[email protected]>

Paolo Abeni

author maintainer

2019-11-20 12:48 UTC #5 lore raw

PATCH v4 3/5 ipv6: introduce and uses route look hints for list input.

In reply to Paolo Abeni

When doing RX batch packet processing, we currently always repeat the route lookup for each ingress packet. When no custom rules are in place, and there aren't routes depending on source addresses, we know that packets with the same destination address will use the same dst.

This change tries to avoid per packet route lookup caching the destination address of the latest successful lookup, and reusing it for the next packet when the above conditions are in place. Ingress traffic for most servers should fit.

The measured performance delta under UDP flood vs a recvmmsg receiver is as follow:

vanilla patched delta
Kpps Kpps %
1431 1674 +17

In the worst-case scenario - each packet has a different destination address - the performance delta is within noise range.

v3 -> v4:
- support hints for SUBFLOW build, too (David A.)
- several style fixes (Eric)

v2 -> v3:
- add fib6_has_custom_rules() helpers (David A.)
- add ip6_extract_route_hint() helper (Edward C.)
- use hint directly in ip6_list_rcv_finish() (Willem)

v1 -> v2:
- fix build issue with !CONFIG_IPV6_MULTIPLE_TABLES
- fix potential race when fib6_has_custom_rules is set while processing a packet batch

Signed-off-by Paolo Abeni <[email protected]>

---
 net/ipv6/ip6_input.c | 26 ++++++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

net/ipv6/ip6_input.c +24 -2

index ef7f707d9ae3..7b089d0ac8cd 100644
@@ -86,11 +86,27 @@ static void ip6_sublist_rcv_finish(struct list_head *head)
 	}
 }
 
+static bool ip6_can_use_hint(const struct sk_buff *skb,
+			     const struct sk_buff *hint)
+{
+	return hint && !skb_dst(skb) &&
+	       ipv6_addr_equal(&ipv6_hdr(hint)->daddr, &ipv6_hdr(skb)->daddr);
+}
+
+static struct sk_buff *ip6_extract_route_hint(const struct net *net,
+					      struct sk_buff *skb)
+{
+	if (fib6_routes_require_src(net) || fib6_has_custom_rules(net))
+		return NULL;
+
+	return skb;
+}
+
 static void ip6_list_rcv_finish(struct net *net, struct sock *sk,
 				struct list_head *head)
 {
+	struct sk_buff *skb, *next, *hint = NULL;
 	struct dst_entry *curr_dst = NULL;
-	struct sk_buff *skb, *next;
 	struct list_head sublist;
 
 	INIT_LIST_HEAD(&sublist);

@@ -104,9 +120,15 @@ static void ip6_list_rcv_finish(struct net *net, struct sock *sk,
 		skb = l3mdev_ip6_rcv(skb);
 		if (!skb)
 			continue;
-		ip6_rcv_finish_core(net, sk, skb);
+
+		if (ip6_can_use_hint(skb, hint))
+			skb_dst_copy(skb, hint);
+		else
+			ip6_rcv_finish_core(net, sk, skb);
 		dst = skb_dst(skb);
 		if (curr_dst != dst) {
+			hint = ip6_extract_route_hint(net, skb);
+
 			/* dispatch old sublist */
 			if (!list_empty(&sublist))
 				ip6_sublist_rcv_finish(&sublist);

Signature · 1 line

2.21.0

David Ahern

Reviewed-by on v4 3/5

2019-11-21 20:11 UTC #6 lore raw

Re: [PATCH net-next v4 3/5] ipv6: introduce and uses route look hints for list input.

Quoted 22 lines · Paolo Abeni

When doing RX batch packet processing, we currently always repeat the route lookup for each ingress packet. When no custom rules are in place, and there aren't routes depending on source addresses, we know that packets with the same destination address will use the same dst.
This change tries to avoid per packet route lookup caching the destination address of the latest successful lookup, and reusing it for the next packet when the above conditions are in place. Ingress traffic for most servers should fit.
The measured performance delta under UDP flood vs a recvmmsg receiver is as follow:
vanilla patched delta
Kpps Kpps %
1431 1674 +17
In the worst-case scenario - each packet has a different destination address - the performance delta is within noise range.

...

Quoted 5 lines

Signed-off-by Paolo Abeni <[email protected]>

---
 net/ipv6/ip6_input.c | 26 ++++++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

Reviewed-by David Ahern <[email protected]>

Paolo Abeni

author maintainer

2019-11-20 12:51 UTC #7 lore raw

PATCH v4 4/5 ipv4: move fib4_has_custom_rules() helper to public header

In reply to Paolo Abeni

So that we can use it in the next patch.
Additionally constify the helper argument.

Signed-off-by Paolo Abeni <[email protected]>

---
 include/net/ip_fib.h    | 10 ++++++++++
 net/ipv4/fib_frontend.c | 10 ----------
 2 files changed, 10 insertions(+), 10 deletions(-)

include/net/ip_fib.h +10 -0

index 52b2406a5dfc..b9cba41c6d4f 100644
@@ -311,6 +311,11 @@ static inline int fib_lookup(struct net *net, const struct flowi4 *flp,
 	return err;
 }
 
+static inline bool fib4_has_custom_rules(const struct net *net)
+{
+	return false;
+}
+
 static inline bool fib4_rule_default(const struct fib_rule *rule)
 {
 	return true;

@@ -378,6 +383,11 @@ static inline int fib_lookup(struct net *net, struct flowi4 *flp,
 	return err;
 }
 
+static inline bool fib4_has_custom_rules(const struct net *net)
+{
+	return net->ipv4.fib_has_custom_rules;
+}
+
 bool fib4_rule_default(const struct fib_rule *rule);
 int fib4_rules_dump(struct net *net, struct notifier_block *nb,
 		    struct netlink_ext_ack *extack);

net/ipv4/fib_frontend.c +0 ~~-10~~

index 71c78d223dfd..577db1d50a24 100644
@@ -70,11 +70,6 @@ static int __net_init fib4_rules_init(struct net *net)
 	fib_free_table(main_table);
 	return -ENOMEM;
 }
-
-static bool fib4_has_custom_rules(struct net *net)
-{
-	return false;
-}
 #else
 
 struct fib_table *fib_new_table(struct net *net, u32 id)

@@ -131,11 +126,6 @@ struct fib_table *fib_get_table(struct net *net, u32 id)
 	}
 	return NULL;
 }
-
-static bool fib4_has_custom_rules(struct net *net)
-{
-	return net->ipv4.fib_has_custom_rules;
-}
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
 static void fib_replace_table(struct net *net, struct fib_table *old,

Signature · 1 line

2.21.0

David Ahern

Reviewed-by on v4 4/5

2019-11-21 20:12 UTC #8 lore raw

Re: [PATCH net-next v4 4/5] ipv4: move fib4_has_custom_rules() helper to public header

Quoted 10 lines · Paolo Abeni

So that we can use it in the next patch.
Additionally constify the helper argument.
Suggested-by David Ahern <[email protected]>
Signed-off-by Paolo Abeni <[email protected]>
---
 include/net/ip_fib.h    | 10 ++++++++++
 net/ipv4/fib_frontend.c | 10 ----------
 2 files changed, 10 insertions(+), 10 deletions(-)

Reviewed-by David Ahern <[email protected]>

Paolo Abeni

author maintainer

2019-11-20 12:48 UTC #9 lore raw

PATCH v4 5/5 ipv4: use dst hint for ipv4 list receive

In reply to Paolo Abeni

This is alike the previous change, with some additional ipv4 specific quirk. Even when using the route hint we still have to do perform additional per packet checks about source address validity: a new helper is added to wrap them.

Hints are explicitly disabled if the destination is a local broadcast, that keeps the code simple and local broadcast are a slower path anyway.

UDP flood performances vs recvmmsg() receiver:

vanilla patched delta
Kpps Kpps %
1683 1871 +11

In the worst case scenario - each packet has a different destination address - the performance delta is within noise range.

v3 -> v4:
- re-enable hints for forward

v2 -> v3:
- really fix build (sic) and hint usage check
- use fib4_has_custom_rules() helpers (David A.)
- add ip_extract_route_hint() helper (Edward C.)
- use prev skb as hint instead of copying data (Willem)

v1 -> v2:
- fix build issue with !CONFIG_IP_MULTIPLE_TABLES

Signed-off-by Paolo Abeni <[email protected]>

---
 include/net/route.h |  4 ++++
 net/ipv4/ip_input.c | 35 +++++++++++++++++++++++++++++++----
 net/ipv4/route.c    | 42 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 77 insertions(+), 4 deletions(-)

include/net/route.h +4 -0

index 6c516840380d..a9c60fc68e36 100644
@@ -185,6 +185,10 @@ int ip_route_input_rcu(struct sk_buff *skb, __be32 dst, __be32 src,
 		       u8 tos, struct net_device *devin,
 		       struct fib_result *res);
 
+int ip_route_use_hint(struct sk_buff *skb, __be32 dst, __be32 src,
+		      u8 tos, struct net_device *devin,
+		      const struct sk_buff *hint);
+
 static inline int ip_route_input(struct sk_buff *skb, __be32 dst, __be32 src,
 				 u8 tos, struct net_device *devin)
 {

net/ipv4/ip_input.c +31 -4

index 24a95126e698..aa438c6758a7 100644
@@ -302,16 +302,31 @@ static inline bool ip_rcv_options(struct sk_buff *skb, struct net_device *dev)
 	return true;
 }
 
+static bool ip_can_use_hint(const struct sk_buff *skb, const struct iphdr *iph,
+			    const struct sk_buff *hint)
+{
+	return hint && !skb_dst(skb) && ip_hdr(hint)->daddr == iph->daddr &&
+	       ip_hdr(hint)->tos == iph->tos;
+}
+
 INDIRECT_CALLABLE_DECLARE(int udp_v4_early_demux(struct sk_buff *));
 INDIRECT_CALLABLE_DECLARE(int tcp_v4_early_demux(struct sk_buff *));
 static int ip_rcv_finish_core(struct net *net, struct sock *sk,
-			      struct sk_buff *skb, struct net_device *dev)
+			      struct sk_buff *skb, struct net_device *dev,
+			      const struct sk_buff *hint)
 {
 	const struct iphdr *iph = ip_hdr(skb);
 	int (*edemux)(struct sk_buff *skb);
 	struct rtable *rt;
 	int err;
 
+	if (ip_can_use_hint(skb, iph, hint)) {
+		err = ip_route_use_hint(skb, iph->daddr, iph->saddr, iph->tos,
+					dev, hint);
+		if (unlikely(err))
+			goto drop_error;
+	}
+
 	if (net->ipv4.sysctl_ip_early_demux &&
 	    !skb_dst(skb) &&
 	    !skb->sk &&

@@ -408,7 +423,7 @@ static int ip_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
 	if (!skb)
 		return NET_RX_SUCCESS;
 
-	ret = ip_rcv_finish_core(net, sk, skb, dev);
+	ret = ip_rcv_finish_core(net, sk, skb, dev, NULL);
 	if (ret != NET_RX_DROP)
 		ret = dst_input(skb);
 	return ret;

@@ -535,11 +550,20 @@ static void ip_sublist_rcv_finish(struct list_head *head)
 	}
 }
 
+static struct sk_buff *ip_extract_route_hint(const struct net *net,
+					     struct sk_buff *skb, int rt_type)
+{
+	if (fib4_has_custom_rules(net) || rt_type == RTN_BROADCAST)
+		return NULL;
+
+	return skb;
+}
+
 static void ip_list_rcv_finish(struct net *net, struct sock *sk,
 			       struct list_head *head)
 {
+	struct sk_buff *skb, *next, *hint = NULL;
 	struct dst_entry *curr_dst = NULL;
-	struct sk_buff *skb, *next;
 	struct list_head sublist;
 
 	INIT_LIST_HEAD(&sublist);

@@ -554,11 +578,14 @@ static void ip_list_rcv_finish(struct net *net, struct sock *sk,
 		skb = l3mdev_ip_rcv(skb);
 		if (!skb)
 			continue;
-		if (ip_rcv_finish_core(net, sk, skb, dev) == NET_RX_DROP)
+		if (ip_rcv_finish_core(net, sk, skb, dev, hint) == NET_RX_DROP)
 			continue;
 
 		dst = skb_dst(skb);
 		if (curr_dst != dst) {
+			hint = ip_extract_route_hint(net, skb,
+					       ((struct rtable *)dst)->rt_type);
+
 			/* dispatch old sublist */
 			if (!list_empty(&sublist))
 				ip_sublist_rcv_finish(&sublist);

net/ipv4/route.c +42 -0

52 lines not on this page · raw message

Signature · 1 line

2.21.0

David Ahern

Reviewed-by on v4 5/5

2019-11-21 21:16 UTC #10 lore raw

Re: [PATCH net-next v4 5/5] ipv4: use dst hint for ipv4 list receive

Quoted 30 lines · Paolo Abeni

This is alike the previous change, with some additional ipv4 specific quirk. Even when using the route hint we still have to do perform additional per packet checks about source address validity: a new helper is added to wrap them.
Hints are explicitly disabled if the destination is a local broadcast, that keeps the code simple and local broadcast are a slower path anyway.
UDP flood performances vs recvmmsg() receiver:
vanilla patched delta
Kpps Kpps %
1683 1871 +11
In the worst case scenario - each packet has a different destination address - the performance delta is within noise range.
v3 -> v4:
- re-enable hints for forward
v2 -> v3:
- really fix build (sic) and hint usage check
- use fib4_has_custom_rules() helpers (David A.)
- add ip_extract_route_hint() helper (Edward C.)
- use prev skb as hint instead of copying data (Willem)
v1 -> v2:
- fix build issue with !CONFIG_IP_MULTIPLE_TABLES

Quoted 7 lines

Signed-off-by Paolo Abeni <[email protected]>

---
 include/net/route.h |  4 ++++
 net/ipv4/ip_input.c | 35 +++++++++++++++++++++++++++++++----
 net/ipv4/route.c    | 42 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 77 insertions(+), 4 deletions(-)

Reviewed-by David Ahern <[email protected]>

Edward Cree

Acked-by on the v4 series

2019-11-20 16:55 UTC #11 lore raw

In reply to Paolo Abeni

Quoted 10 lines · Paolo Abeni

This series leverages the listification infrastructure to avoid unnecessary route lookup on ingress packets. In absence of custom rules, packets with equal daddr will usually land on the same dst.
When processing packet bursts (lists) we can easily reference the previous dst entry. When we hit the 'same destination' condition we can avoid the route lookup, coping the already available dst.
Detailed performance numbers are available in the individual commit messages.

I wonder if you could use static keys for the fib*_has_custom_rules() and if that would gain you any extra speed?
Other than that,

Acked-by Edward Cree <[email protected]>

for the series.

Quoted 34 lines

v3 -> v4:
- move helpers to their own patches (Eric D.)
- enable hints for SUBTREE builds (David A.)
- re-enable hints for ipv4 forward (David A.)
v2 -> v3:
- use fib*_has_custom_rules() helpers (David A.)
- add ip*_extract_route_hint() helper (Edward C.)
- use prev skb as hint instead of copying data (Willem )
v1 -> v2:
- fix build issue with !CONFIG_IP*_MULTIPLE_TABLES
- fix potential race in ip6_list_rcv_finish()
Paolo Abeni (5):
ipv6: add fib6_has_custom_rules() helper
ipv6: keep track of routes using src
ipv6: introduce and uses route look hints for list input.
ipv4: move fib4_has_custom_rules() helper to public header
ipv4: use dst hint for ipv4 list receive
 include/net/ip6_fib.h    | 39 +++++++++++++++++++++++++++++++++++++
 include/net/ip_fib.h     | 10 ++++++++++
 include/net/netns/ipv6.h |  3 +++
 include/net/route.h      |  4 ++++
 net/ipv4/fib_frontend.c  | 10 ----------
 net/ipv4/ip_input.c      | 35 +++++++++++++++++++++++++++++----
 net/ipv4/route.c         | 42 ++++++++++++++++++++++++++++++++++++++++
 net/ipv6/ip6_fib.c       |  4 ++++
 net/ipv6/ip6_input.c     | 26 +++++++++++++++++++++++--
 net/ipv6/route.c         |  3 +++
 10 files changed, 160 insertions(+), 16 deletions(-)

David Miller

maintainer

2019-11-21 22:46 UTC #12 lore raw

In reply to Paolo Abeni

From: Paolo Abeni <[email protected]>
Date: Wed, 20 Nov 2019 13:47:32 +0100

Quoted 10 lines

This series leverages the listification infrastructure to avoid unnecessary route lookup on ingress packets. In absence of custom rules, packets with equal daddr will usually land on the same dst.
When processing packet bursts (lists) we can easily reference the previous dst entry. When we hit the 'same destination' condition we can avoid the route lookup, coping the already available dst.
Detailed performance numbers are available in the individual commit messages.

...

Series applied, thanks.