Netdev Archive on lore.kernel.org
 help / color / Atom feed
From: Wei Wang <tracywwnj@gmail.com>
To: David Miller <davem@davemloft.net>, netdev@vger.kernel.org
Cc: Martin KaFai Lau <kafai@fb.com>,
	Eric Dumazet <edumazet@google.com>,
	Mahesh Bandewar <maheshb@google.com>,
	David Ahern <dsahern@gmail.com>, Wei Wang <weiwan@google.com>
Subject: [PATCH net-next 5/5] ipv6: convert major tx path to use RT6_LOOKUP_F_DST_NOREF
Date: Tue, 18 Jun 2019 11:25:43 -0700
Message-ID: <20190618182543.65477-6-tracywwnj@gmail.com> (raw)
In-Reply-To: <20190618182543.65477-1-tracywwnj@gmail.com>

From: Wei Wang <weiwan@google.com>

For tx path, in most cases, we still have to take refcnt on the dst
cause the caller is caching the dst somewhere. But it still is
beneficial to make use of RT6_LOOKUP_F_DST_NOREF flag while doing the
route lookup. It is cause this flag prevents manipulating refcnt on
net->ipv6.ip6_null_entry when doing fib6_rule_lookup() to traverse each
routing table. The null_entry is a shared object and constant updates on
it cause false sharing.

We converted the current major lookup function ip6_route_output_flags()
to make use of RT6_LOOKUP_F_DST_NOREF.

Together with the change in the rx path, we see noticable performance
boost:
I ran synflood tests between 2 hosts under the same switch. Both hosts
have 20G mlx NIC, and 8 tx/rx queues.
Sender sends pure SYN flood with random src IPs and ports using trafgen.
Receiver has a simple TCP listener on the target port.
Both hosts have multiple custom rules:
- For incoming packets, only local table is traversed.
- For outgoing packets, 3 tables are traversed to find the route.
The packet processing rate on the receiver is as follows:
- Before the fix: 3.78Mpps
- After the fix:  5.50Mpps

Signed-off-by: Wei Wang <weiwan@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Mahesh Bandewar <maheshb@google.com>
---
 drivers/net/vrf.c       | 11 ++++++-----
 include/net/ip6_route.h | 25 +++++++++++++++++++++++--
 include/net/l3mdev.h    | 11 +++++++----
 net/ipv6/route.c        | 10 ++++++----
 net/l3mdev/l3mdev.c     | 22 +++++++++++-----------
 5 files changed, 53 insertions(+), 26 deletions(-)

diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 11b9525dff27..1d1ac78b167e 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -1072,12 +1072,14 @@ static struct sk_buff *vrf_l3_rcv(struct net_device *vrf_dev,
 #if IS_ENABLED(CONFIG_IPV6)
 /* send to link-local or multicast address via interface enslaved to
  * VRF device. Force lookup to VRF table without changing flow struct
+ * No refcnt is taken on the dst.
  */
-static struct dst_entry *vrf_link_scope_lookup(const struct net_device *dev,
-					      struct flowi6 *fl6)
+static struct dst_entry *vrf_link_scope_lookup_noref(
+					    const struct net_device *dev,
+					    struct flowi6 *fl6)
 {
 	struct net *net = dev_net(dev);
-	int flags = RT6_LOOKUP_F_IFACE;
+	int flags = RT6_LOOKUP_F_IFACE | RT6_LOOKUP_F_DST_NOREF;
 	struct dst_entry *dst = NULL;
 	struct rt6_info *rt;
 
@@ -1087,7 +1089,6 @@ static struct dst_entry *vrf_link_scope_lookup(const struct net_device *dev,
 	 */
 	if (fl6->flowi6_oif == dev->ifindex) {
 		dst = &net->ipv6.ip6_null_entry->dst;
-		dst_hold(dst);
 		return dst;
 	}
 
@@ -1107,7 +1108,7 @@ static const struct l3mdev_ops vrf_l3mdev_ops = {
 	.l3mdev_l3_rcv		= vrf_l3_rcv,
 	.l3mdev_l3_out		= vrf_l3_out,
 #if IS_ENABLED(CONFIG_IPV6)
-	.l3mdev_link_scope_lookup = vrf_link_scope_lookup,
+	.l3mdev_link_scope_lookup_noref = vrf_link_scope_lookup_noref,
 #endif
 };
 
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 82bced2fc1e3..d587d73a5120 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -84,8 +84,29 @@ struct dst_entry *ip6_route_input_lookup(struct net *net,
 					 struct flowi6 *fl6,
 					 const struct sk_buff *skb, int flags);
 
-struct dst_entry *ip6_route_output_flags(struct net *net, const struct sock *sk,
-					 struct flowi6 *fl6, int flags);
+struct dst_entry *ip6_route_output_flags_noref(struct net *net,
+					       const struct sock *sk,
+					       struct flowi6 *fl6, int flags);
+
+static inline struct dst_entry *ip6_route_output_flags(struct net *net,
+						       const struct sock *sk,
+						       struct flowi6 *fl6,
+						       int flags) {
+	struct dst_entry *dst;
+	struct rt6_info *rt6;
+
+	rcu_read_lock();
+	dst = ip6_route_output_flags_noref(net, sk, fl6, flags);
+	rt6 = (struct rt6_info *)dst;
+	/* For dst cached in uncached_list, refcnt is already taken. */
+	if (list_empty(&rt6->rt6i_uncached) && !dst_hold_safe(dst)) {
+		dst = &net->ipv6.ip6_null_entry->dst;
+		dst_hold(dst);
+	}
+	rcu_read_unlock();
+
+	return dst;
+}
 
 static inline struct dst_entry *ip6_route_output(struct net *net,
 						 const struct sock *sk,
diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h
index e942372b077b..d8c37317bb86 100644
--- a/include/net/l3mdev.h
+++ b/include/net/l3mdev.h
@@ -31,8 +31,9 @@ struct l3mdev_ops {
 					  u16 proto);
 
 	/* IPv6 ops */
-	struct dst_entry * (*l3mdev_link_scope_lookup)(const struct net_device *dev,
-						 struct flowi6 *fl6);
+	struct dst_entry * (*l3mdev_link_scope_lookup_noref)(
+					    const struct net_device *dev,
+					    struct flowi6 *fl6);
 };
 
 #ifdef CONFIG_NET_L3_MASTER_DEV
@@ -140,7 +141,8 @@ static inline bool netif_index_is_l3_master(struct net *net, int ifindex)
 	return rc;
 }
 
-struct dst_entry *l3mdev_link_scope_lookup(struct net *net, struct flowi6 *fl6);
+struct dst_entry *l3mdev_link_scope_lookup_noref(struct net *net,
+						 struct flowi6 *fl6);
 
 static inline
 struct sk_buff *l3mdev_l3_rcv(struct sk_buff *skb, u16 proto)
@@ -251,7 +253,8 @@ static inline bool netif_index_is_l3_master(struct net *net, int ifindex)
 }
 
 static inline
-struct dst_entry *l3mdev_link_scope_lookup(struct net *net, struct flowi6 *fl6)
+struct dst_entry *l3mdev_link_scope_lookup_noref(struct net *net,
+						 struct flowi6 *fl6)
 {
 	return NULL;
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index d2b287635aab..602d00794b30 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2415,8 +2415,9 @@ static struct rt6_info *ip6_pol_route_output(struct net *net,
 	return ip6_pol_route(net, table, fl6->flowi6_oif, fl6, skb, flags);
 }
 
-struct dst_entry *ip6_route_output_flags(struct net *net, const struct sock *sk,
-					 struct flowi6 *fl6, int flags)
+struct dst_entry *ip6_route_output_flags_noref(struct net *net,
+					       const struct sock *sk,
+					       struct flowi6 *fl6, int flags)
 {
 	bool any_src;
 
@@ -2424,13 +2425,14 @@ struct dst_entry *ip6_route_output_flags(struct net *net, const struct sock *sk,
 	    (IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL)) {
 		struct dst_entry *dst;
 
-		dst = l3mdev_link_scope_lookup(net, fl6);
+		dst = l3mdev_link_scope_lookup_noref(net, fl6);
 		if (dst)
 			return dst;
 	}
 
 	fl6->flowi6_iif = LOOPBACK_IFINDEX;
 
+	flags |= RT6_LOOKUP_F_DST_NOREF;
 	any_src = ipv6_addr_any(&fl6->saddr);
 	if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr) ||
 	    (fl6->flowi6_oif && any_src))
@@ -2443,7 +2445,7 @@ struct dst_entry *ip6_route_output_flags(struct net *net, const struct sock *sk,
 
 	return fib6_rule_lookup(net, fl6, NULL, flags, ip6_pol_route_output);
 }
-EXPORT_SYMBOL_GPL(ip6_route_output_flags);
+EXPORT_SYMBOL_GPL(ip6_route_output_flags_noref);
 
 struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_orig)
 {
diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c
index cfc9fcb97465..06133426549b 100644
--- a/net/l3mdev/l3mdev.c
+++ b/net/l3mdev/l3mdev.c
@@ -114,35 +114,35 @@ u32 l3mdev_fib_table_by_index(struct net *net, int ifindex)
 EXPORT_SYMBOL_GPL(l3mdev_fib_table_by_index);
 
 /**
- *	l3mdev_link_scope_lookup - IPv6 route lookup based on flow for link
- *			     local and multicast addresses
+ *	l3mdev_link_scope_lookup_noref - IPv6 route lookup based on flow
+ *			     for link local and multicast addresses
  *	@net: network namespace for device index lookup
  *	@fl6: IPv6 flow struct for lookup
+ *	This function does not hold refcnt on the returned dst.
+ *	Caller must hold rcu_read_lock().
  */
 
-struct dst_entry *l3mdev_link_scope_lookup(struct net *net,
-					   struct flowi6 *fl6)
+struct dst_entry *l3mdev_link_scope_lookup_noref(struct net *net,
+						 struct flowi6 *fl6)
 {
 	struct dst_entry *dst = NULL;
 	struct net_device *dev;
 
+	WARN_ON_ONCE(!rcu_read_lock_held());
 	if (fl6->flowi6_oif) {
-		rcu_read_lock();
-
 		dev = dev_get_by_index_rcu(net, fl6->flowi6_oif);
 		if (dev && netif_is_l3_slave(dev))
 			dev = netdev_master_upper_dev_get_rcu(dev);
 
 		if (dev && netif_is_l3_master(dev) &&
-		    dev->l3mdev_ops->l3mdev_link_scope_lookup)
-			dst = dev->l3mdev_ops->l3mdev_link_scope_lookup(dev, fl6);
-
-		rcu_read_unlock();
+		    dev->l3mdev_ops->l3mdev_link_scope_lookup_noref)
+			dst = dev->l3mdev_ops->
+				l3mdev_link_scope_lookup_noref(dev, fl6);
 	}
 
 	return dst;
 }
-EXPORT_SYMBOL_GPL(l3mdev_link_scope_lookup);
+EXPORT_SYMBOL_GPL(l3mdev_link_scope_lookup_noref);
 
 /**
  *	l3mdev_fib_rule_match - Determine if flowi references an
-- 
2.22.0.410.gd8fdbe21b5-goog


  parent reply index

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-18 18:25 [PATCH net-next 0/5] ipv6: avoid taking refcnt on dst during route lookup Wei Wang
2019-06-18 18:25 ` [PATCH net-next 1/5] ipv6: introduce RT6_LOOKUP_F_DST_NOREF flag in ip6_pol_route() Wei Wang
2019-06-18 18:25 ` [PATCH net-next 2/5] ipv6: initialize rt6->rt6i_uncached in all pre-allocated dst entries Wei Wang
2019-06-18 18:25 ` [PATCH net-next 3/5] ipv6: honor RT6_LOOKUP_F_DST_NOREF in rule lookup logic Wei Wang
2019-06-19 16:07   ` David Miller
2019-06-19 16:51     ` Wei Wang
2019-06-18 18:25 ` [PATCH net-next 4/5] ipv6: convert rx data path to not take refcnt on dst Wei Wang
2019-06-18 18:25 ` Wei Wang [this message]
2019-06-20 22:47   ` [PATCH net-next 5/5] ipv6: convert major tx path to use RT6_LOOKUP_F_DST_NOREF kbuild test robot

Reply instructions:

You may reply publically to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190618182543.65477-6-tracywwnj@gmail.com \
    --to=tracywwnj@gmail.com \
    --cc=davem@davemloft.net \
    --cc=dsahern@gmail.com \
    --cc=edumazet@google.com \
    --cc=kafai@fb.com \
    --cc=maheshb@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=weiwan@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Netdev Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/netdev/0 netdev/git/0.git
	git clone --mirror https://lore.kernel.org/netdev/1 netdev/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 netdev netdev/ https://lore.kernel.org/netdev \
		netdev@vger.kernel.org netdev@archiver.kernel.org
	public-inbox-index netdev


Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.netdev


AGPL code for this site: git clone https://public-inbox.org/ public-inbox