From f38a9eb1f77b296ff07e000823884a0f64d67b2a Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 21 Jul 2015 10:43:56 +0200 Subject: dst: Metadata destinations Introduces a new dst_metadata which enables to carry per packet metadata between forwarding and processing elements via the skb->dst pointer. The structure is set up to be a union. Thus, each separate type of metadata requires its own dst instance. If demand arises to carry multiple types of metadata concurrently, metadata dst entries can be made stackable. The metadata dst entry is refcnt'ed as expected for now but a non reference counted use is possible if the reference is forced before queueing the skb. In order to allow allocating dsts with variable length, the existing dst_alloc() is split into a dst_alloc() and dst_init() function. The existing dst_init() function to initialize the subsystem is being renamed to dst_subsys_init() to make it clear what is what. The check before ip_route_input() is changed to ignore metadata dsts and drop the dst inside the routing function thus allowing to interpret metadata in a later commit. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/dst.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/net/dst.h') diff --git a/include/net/dst.h b/include/net/dst.h index 2bc73f8a00a9..2578811cef51 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -57,6 +57,7 @@ struct dst_entry { #define DST_FAKE_RTABLE 0x0040 #define DST_XFRM_TUNNEL 0x0080 #define DST_XFRM_QUEUE 0x0100 +#define DST_METADATA 0x0200 unsigned short pending_confirm; @@ -356,6 +357,9 @@ static inline int dst_discard(struct sk_buff *skb) } void *dst_alloc(struct dst_ops *ops, struct net_device *dev, int initial_ref, int initial_obsolete, unsigned short flags); +void dst_init(struct dst_entry *dst, struct dst_ops *ops, + struct net_device *dev, int initial_ref, int initial_obsolete, + unsigned short flags); void __dst_free(struct dst_entry *dst); struct dst_entry *dst_destroy(struct dst_entry *dst); @@ -457,7 +461,7 @@ static inline struct dst_entry *dst_check(struct dst_entry *dst, u32 cookie) return dst; } -void dst_init(void); +void dst_subsys_init(void); /* Flags for xfrm_lookup flags argument. */ enum { -- cgit v1.2.3 From 61adedf3e3f1d3f032c5a6a299978d91eff6d555 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 20 Aug 2015 13:56:25 +0200 Subject: route: move lwtunnel state to dst_entry Currently, the lwtunnel state resides in per-protocol data. This is a problem if we encapsulate ipv6 traffic in an ipv4 tunnel (or vice versa). The xmit function of the tunnel does not know whether the packet has been routed to it by ipv4 or ipv6, yet it needs the lwtstate data. Moving the lwtstate data to dst_entry makes such inter-protocol tunneling possible. As a bonus, this brings a nice diffstat. Signed-off-by: Jiri Benc Acked-by: Roopa Prabhu Acked-by: Thomas Graf Signed-off-by: David S. Miller --- drivers/net/vrf.c | 1 - drivers/net/vxlan.c | 4 +-- include/net/dst.h | 3 +- include/net/dst_metadata.h | 15 +++------ include/net/ip6_fib.h | 1 - include/net/lwtunnel.h | 12 -------- include/net/route.h | 1 - net/core/dst.c | 3 ++ net/core/filter.c | 2 +- net/core/lwtunnel.c | 70 ++++++------------------------------------ net/ipv4/ip_gre.c | 2 +- net/ipv4/route.c | 20 +++++------- net/ipv6/ila.c | 14 +++------ net/ipv6/ip6_fib.c | 1 - net/ipv6/route.c | 20 ++++++------ net/mpls/mpls_iptunnel.c | 7 ++--- net/openvswitch/vport-netdev.c | 2 +- 17 files changed, 48 insertions(+), 130 deletions(-) (limited to 'include/net/dst.h') diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index dbeffe789185..b3d9c5546c79 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -295,7 +295,6 @@ static struct rtable *vrf_rtable_create(struct net_device *dev) rth->rt_uses_gateway = 0; INIT_LIST_HEAD(&rth->rt_uncached); rth->rt_uncached_list = NULL; - rth->rt_lwtstate = NULL; } return rth; diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index ebeb3def06c5..93613ffd8d7e 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -1909,7 +1909,7 @@ static void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, u32 flags = vxlan->flags; /* FIXME: Support IPv6 */ - info = skb_tunnel_info(skb, AF_INET); + info = skb_tunnel_info(skb); if (rdst) { dst_port = rdst->remote_port ? rdst->remote_port : vxlan->cfg.dst_port; @@ -2105,7 +2105,7 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev) struct vxlan_fdb *f; /* FIXME: Support IPv6 */ - info = skb_tunnel_info(skb, AF_INET); + info = skb_tunnel_info(skb); skb_reset_mac_header(skb); eth = eth_hdr(skb); diff --git a/include/net/dst.h b/include/net/dst.h index 2578811cef51..0a9a723f6c19 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -44,6 +44,7 @@ struct dst_entry { #else void *__pad1; #endif + struct lwtunnel_state *lwtstate; int (*input)(struct sk_buff *); int (*output)(struct sock *sk, struct sk_buff *skb); @@ -89,7 +90,7 @@ struct dst_entry { * (L1_CACHE_SIZE would be too much) */ #ifdef CONFIG_64BIT - long __pad_to_align_refcnt[2]; + long __pad_to_align_refcnt[1]; #endif /* * __refcnt wants to be on a different cache line from diff --git a/include/net/dst_metadata.h b/include/net/dst_metadata.h index 075f523ff23f..2cb52d562272 100644 --- a/include/net/dst_metadata.h +++ b/include/net/dst_metadata.h @@ -23,22 +23,17 @@ static inline struct metadata_dst *skb_metadata_dst(struct sk_buff *skb) return NULL; } -static inline struct ip_tunnel_info *skb_tunnel_info(struct sk_buff *skb, - int family) +static inline struct ip_tunnel_info *skb_tunnel_info(struct sk_buff *skb) { struct metadata_dst *md_dst = skb_metadata_dst(skb); - struct rtable *rt; + struct dst_entry *dst; if (md_dst) return &md_dst->u.tun_info; - switch (family) { - case AF_INET: - rt = (struct rtable *)skb_dst(skb); - if (rt && rt->rt_lwtstate) - return lwt_tun_info(rt->rt_lwtstate); - break; - } + dst = skb_dst(skb); + if (dst && dst->lwtstate) + return lwt_tun_info(dst->lwtstate); return NULL; } diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 276328e3daa6..063d30474cf6 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -133,7 +133,6 @@ struct rt6_info { /* more non-fragment space at head required */ unsigned short rt6i_nfheader_len; u8 rt6i_protocol; - struct lwtunnel_state *rt6i_lwtstate; }; static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst) diff --git a/include/net/lwtunnel.h b/include/net/lwtunnel.h index cfee53916ba5..843489884448 100644 --- a/include/net/lwtunnel.h +++ b/include/net/lwtunnel.h @@ -87,9 +87,7 @@ int lwtunnel_get_encap_size(struct lwtunnel_state *lwtstate); struct lwtunnel_state *lwtunnel_state_alloc(int hdr_len); int lwtunnel_cmp_encap(struct lwtunnel_state *a, struct lwtunnel_state *b); int lwtunnel_output(struct sock *sk, struct sk_buff *skb); -int lwtunnel_output6(struct sock *sk, struct sk_buff *skb); int lwtunnel_input(struct sk_buff *skb); -int lwtunnel_input6(struct sk_buff *skb); #else @@ -164,21 +162,11 @@ static inline int lwtunnel_output(struct sock *sk, struct sk_buff *skb) return -EOPNOTSUPP; } -static inline int lwtunnel_output6(struct sock *sk, struct sk_buff *skb) -{ - return -EOPNOTSUPP; -} - static inline int lwtunnel_input(struct sk_buff *skb) { return -EOPNOTSUPP; } -static inline int lwtunnel_input6(struct sk_buff *skb) -{ - return -EOPNOTSUPP; -} - #endif #endif /* __NET_LWTUNNEL_H */ diff --git a/include/net/route.h b/include/net/route.h index 6dda2c1bf8c6..395d79bb556c 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -66,7 +66,6 @@ struct rtable { struct list_head rt_uncached; struct uncached_list *rt_uncached_list; - struct lwtunnel_state *rt_lwtstate; }; static inline bool rt_is_input_route(const struct rtable *rt) diff --git a/net/core/dst.c b/net/core/dst.c index f8694d1b8702..50dcdbb0ee46 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -184,6 +185,7 @@ void dst_init(struct dst_entry *dst, struct dst_ops *ops, #ifdef CONFIG_IP_ROUTE_CLASSID dst->tclassid = 0; #endif + dst->lwtstate = NULL; atomic_set(&dst->__refcnt, initial_ref); dst->__use = 0; dst->lastuse = jiffies; @@ -264,6 +266,7 @@ again: kfree(dst); else kmem_cache_free(dst->ops->kmem_cachep, dst); + lwtstate_put(dst->lwtstate); dst = child; if (dst) { diff --git a/net/core/filter.c b/net/core/filter.c index 379568562ffb..b4adc961413f 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1489,7 +1489,7 @@ static u64 bpf_skb_get_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5) { struct sk_buff *skb = (struct sk_buff *) (long) r1; struct bpf_tunnel_key *to = (struct bpf_tunnel_key *) (long) r2; - struct ip_tunnel_info *info = skb_tunnel_info(skb, AF_INET); + struct ip_tunnel_info *info = skb_tunnel_info(skb); if (unlikely(size != sizeof(struct bpf_tunnel_key) || flags || !info)) return -EINVAL; diff --git a/net/core/lwtunnel.c b/net/core/lwtunnel.c index 3331585174d9..e924c2e08554 100644 --- a/net/core/lwtunnel.c +++ b/net/core/lwtunnel.c @@ -179,14 +179,16 @@ int lwtunnel_cmp_encap(struct lwtunnel_state *a, struct lwtunnel_state *b) } EXPORT_SYMBOL(lwtunnel_cmp_encap); -int __lwtunnel_output(struct sock *sk, struct sk_buff *skb, - struct lwtunnel_state *lwtstate) +int lwtunnel_output(struct sock *sk, struct sk_buff *skb) { + struct dst_entry *dst = skb_dst(skb); const struct lwtunnel_encap_ops *ops; + struct lwtunnel_state *lwtstate; int ret = -EINVAL; - if (!lwtstate) + if (!dst) goto drop; + lwtstate = dst->lwtstate; if (lwtstate->type == LWTUNNEL_ENCAP_NONE || lwtstate->type > LWTUNNEL_ENCAP_MAX) @@ -209,47 +211,18 @@ drop: return ret; } - -int lwtunnel_output6(struct sock *sk, struct sk_buff *skb) -{ - struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); - struct lwtunnel_state *lwtstate = NULL; - - if (rt) { - lwtstate = rt->rt6i_lwtstate; - skb->dev = rt->dst.dev; - } - - skb->protocol = htons(ETH_P_IPV6); - - return __lwtunnel_output(sk, skb, lwtstate); -} -EXPORT_SYMBOL(lwtunnel_output6); - -int lwtunnel_output(struct sock *sk, struct sk_buff *skb) -{ - struct rtable *rt = (struct rtable *)skb_dst(skb); - struct lwtunnel_state *lwtstate = NULL; - - if (rt) { - lwtstate = rt->rt_lwtstate; - skb->dev = rt->dst.dev; - } - - skb->protocol = htons(ETH_P_IP); - - return __lwtunnel_output(sk, skb, lwtstate); -} EXPORT_SYMBOL(lwtunnel_output); -int __lwtunnel_input(struct sk_buff *skb, - struct lwtunnel_state *lwtstate) +int lwtunnel_input(struct sk_buff *skb) { + struct dst_entry *dst = skb_dst(skb); const struct lwtunnel_encap_ops *ops; + struct lwtunnel_state *lwtstate; int ret = -EINVAL; - if (!lwtstate) + if (!dst) goto drop; + lwtstate = dst->lwtstate; if (lwtstate->type == LWTUNNEL_ENCAP_NONE || lwtstate->type > LWTUNNEL_ENCAP_MAX) @@ -272,27 +245,4 @@ drop: return ret; } - -int lwtunnel_input6(struct sk_buff *skb) -{ - struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); - struct lwtunnel_state *lwtstate = NULL; - - if (rt) - lwtstate = rt->rt6i_lwtstate; - - return __lwtunnel_input(skb, lwtstate); -} -EXPORT_SYMBOL(lwtunnel_input6); - -int lwtunnel_input(struct sk_buff *skb) -{ - struct rtable *rt = (struct rtable *)skb_dst(skb); - struct lwtunnel_state *lwtstate = NULL; - - if (rt) - lwtstate = rt->rt_lwtstate; - - return __lwtunnel_input(skb, lwtstate); -} EXPORT_SYMBOL(lwtunnel_input); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 5193618b2600..1bf328182697 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -521,7 +521,7 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev) __be16 df, flags; int err; - tun_info = skb_tunnel_info(skb, AF_INET); + tun_info = skb_tunnel_info(skb); if (unlikely(!tun_info || tun_info->mode != IP_TUNNEL_INFO_TX)) goto err_free_skb; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 2403e85107f0..f3087aaa6dd8 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1359,7 +1359,6 @@ static void ipv4_dst_destroy(struct dst_entry *dst) list_del(&rt->rt_uncached); spin_unlock_bh(&ul->lock); } - lwtstate_put(rt->rt_lwtstate); } void rt_flush_dev(struct net_device *dev) @@ -1408,7 +1407,7 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr, #ifdef CONFIG_IP_ROUTE_CLASSID rt->dst.tclassid = nh->nh_tclassid; #endif - rt->rt_lwtstate = lwtstate_get(nh->nh_lwtstate); + rt->dst.lwtstate = lwtstate_get(nh->nh_lwtstate); if (unlikely(fnhe)) cached = rt_bind_exception(rt, fnhe, daddr); else if (!(rt->dst.flags & DST_NOCACHE)) @@ -1494,7 +1493,6 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->rt_gateway = 0; rth->rt_uses_gateway = 0; INIT_LIST_HEAD(&rth->rt_uncached); - rth->rt_lwtstate = NULL; if (our) { rth->dst.input= ip_local_deliver; rth->rt_flags |= RTCF_LOCAL; @@ -1624,19 +1622,18 @@ static int __mkroute_input(struct sk_buff *skb, rth->rt_gateway = 0; rth->rt_uses_gateway = 0; INIT_LIST_HEAD(&rth->rt_uncached); - rth->rt_lwtstate = NULL; RT_CACHE_STAT_INC(in_slow_tot); rth->dst.input = ip_forward; rth->dst.output = ip_output; rt_set_nexthop(rth, daddr, res, fnhe, res->fi, res->type, itag); - if (lwtunnel_output_redirect(rth->rt_lwtstate)) { - rth->rt_lwtstate->orig_output = rth->dst.output; + if (lwtunnel_output_redirect(rth->dst.lwtstate)) { + rth->dst.lwtstate->orig_output = rth->dst.output; rth->dst.output = lwtunnel_output; } - if (lwtunnel_input_redirect(rth->rt_lwtstate)) { - rth->rt_lwtstate->orig_input = rth->dst.input; + if (lwtunnel_input_redirect(rth->dst.lwtstate)) { + rth->dst.lwtstate->orig_input = rth->dst.input; rth->dst.input = lwtunnel_input; } skb_dst_set(skb, &rth->dst); @@ -1695,7 +1692,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, by fib_lookup. */ - tun_info = skb_tunnel_info(skb, AF_INET); + tun_info = skb_tunnel_info(skb); if (tun_info && tun_info->mode == IP_TUNNEL_INFO_RX) fl4.flowi4_tun_key.tun_id = tun_info->key.tun_id; else @@ -1815,7 +1812,6 @@ local_input: rth->rt_gateway = 0; rth->rt_uses_gateway = 0; INIT_LIST_HEAD(&rth->rt_uncached); - rth->rt_lwtstate = NULL; RT_CACHE_STAT_INC(in_slow_tot); if (res.type == RTN_UNREACHABLE) { @@ -2006,7 +2002,6 @@ add: rth->rt_gateway = 0; rth->rt_uses_gateway = 0; INIT_LIST_HEAD(&rth->rt_uncached); - rth->rt_lwtstate = NULL; RT_CACHE_STAT_INC(out_slow_tot); if (flags & RTCF_LOCAL) @@ -2029,7 +2024,7 @@ add: } rt_set_nexthop(rth, fl4->daddr, res, fnhe, fi, type, 0); - if (lwtunnel_output_redirect(rth->rt_lwtstate)) + if (lwtunnel_output_redirect(rth->dst.lwtstate)) rth->dst.output = lwtunnel_output; return rth; @@ -2293,7 +2288,6 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or rt->rt_uses_gateway = ort->rt_uses_gateway; INIT_LIST_HEAD(&rt->rt_uncached); - rt->rt_lwtstate = NULL; dst_free(new); } diff --git a/net/ipv6/ila.c b/net/ipv6/ila.c index 2540ab4b76d1..f011c3d5ca40 100644 --- a/net/ipv6/ila.c +++ b/net/ipv6/ila.c @@ -89,16 +89,13 @@ static void update_ipv6_locator(struct sk_buff *skb, struct ila_params *p) static int ila_output(struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); - struct rt6_info *rt6 = NULL; if (skb->protocol != htons(ETH_P_IPV6)) goto drop; - rt6 = (struct rt6_info *)dst; + update_ipv6_locator(skb, ila_params_lwtunnel(dst->lwtstate)); - update_ipv6_locator(skb, ila_params_lwtunnel(rt6->rt6i_lwtstate)); - - return rt6->rt6i_lwtstate->orig_output(sk, skb); + return dst->lwtstate->orig_output(sk, skb); drop: kfree_skb(skb); @@ -108,16 +105,13 @@ drop: static int ila_input(struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); - struct rt6_info *rt6 = NULL; if (skb->protocol != htons(ETH_P_IPV6)) goto drop; - rt6 = (struct rt6_info *)dst; - - update_ipv6_locator(skb, ila_params_lwtunnel(rt6->rt6i_lwtstate)); + update_ipv6_locator(skb, ila_params_lwtunnel(dst->lwtstate)); - return rt6->rt6i_lwtstate->orig_input(skb); + return dst->lwtstate->orig_input(skb); drop: kfree_skb(skb); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 5693b5eb8482..865e777ae20c 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -178,7 +178,6 @@ static void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) static void rt6_release(struct rt6_info *rt) { if (atomic_dec_and_test(&rt->rt6i_ref)) { - lwtstate_put(rt->rt6i_lwtstate); rt6_free_pcpu(rt); dst_free(&rt->dst); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index c3733049715e..e6bbcdee7707 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1784,14 +1784,14 @@ int ip6_route_add(struct fib6_config *cfg) cfg->fc_encap, &lwtstate); if (err) goto out; - rt->rt6i_lwtstate = lwtstate_get(lwtstate); - if (lwtunnel_output_redirect(rt->rt6i_lwtstate)) { - rt->rt6i_lwtstate->orig_output = rt->dst.output; - rt->dst.output = lwtunnel_output6; + rt->dst.lwtstate = lwtstate_get(lwtstate); + if (lwtunnel_output_redirect(rt->dst.lwtstate)) { + rt->dst.lwtstate->orig_output = rt->dst.output; + rt->dst.output = lwtunnel_output; } - if (lwtunnel_input_redirect(rt->rt6i_lwtstate)) { - rt->rt6i_lwtstate->orig_input = rt->dst.input; - rt->dst.input = lwtunnel_input6; + if (lwtunnel_input_redirect(rt->dst.lwtstate)) { + rt->dst.lwtstate->orig_input = rt->dst.input; + rt->dst.input = lwtunnel_input; } } @@ -2174,7 +2174,7 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct rt6_info *ort) #endif rt->rt6i_prefsrc = ort->rt6i_prefsrc; rt->rt6i_table = ort->rt6i_table; - rt->rt6i_lwtstate = lwtstate_get(ort->rt6i_lwtstate); + rt->dst.lwtstate = lwtstate_get(ort->dst.lwtstate); } #ifdef CONFIG_IPV6_ROUTE_INFO @@ -2838,7 +2838,7 @@ static inline size_t rt6_nlmsg_size(struct rt6_info *rt) + nla_total_size(sizeof(struct rta_cacheinfo)) + nla_total_size(TCP_CA_NAME_MAX) /* RTAX_CC_ALGO */ + nla_total_size(1) /* RTA_PREF */ - + lwtunnel_get_encap_size(rt->rt6i_lwtstate); + + lwtunnel_get_encap_size(rt->dst.lwtstate); } static int rt6_fill_node(struct net *net, @@ -2991,7 +2991,7 @@ static int rt6_fill_node(struct net *net, if (nla_put_u8(skb, RTA_PREF, IPV6_EXTRACT_PREF(rt->rt6i_flags))) goto nla_put_failure; - lwtunnel_fill_encap(skb, rt->rt6i_lwtstate); + lwtunnel_fill_encap(skb, rt->dst.lwtstate); nlmsg_end(skb, nlh); return 0; diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index 276f8c992218..3da5ca3ba563 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -48,7 +48,6 @@ int mpls_output(struct sock *sk, struct sk_buff *skb) struct dst_entry *dst = skb_dst(skb); struct rtable *rt = NULL; struct rt6_info *rt6 = NULL; - struct lwtunnel_state *lwtstate = NULL; int err = 0; bool bos; int i; @@ -58,11 +57,9 @@ int mpls_output(struct sock *sk, struct sk_buff *skb) if (skb->protocol == htons(ETH_P_IP)) { ttl = ip_hdr(skb)->ttl; rt = (struct rtable *)dst; - lwtstate = rt->rt_lwtstate; } else if (skb->protocol == htons(ETH_P_IPV6)) { ttl = ipv6_hdr(skb)->hop_limit; rt6 = (struct rt6_info *)dst; - lwtstate = rt6->rt6i_lwtstate; } else { goto drop; } @@ -72,12 +69,12 @@ int mpls_output(struct sock *sk, struct sk_buff *skb) /* Find the output device */ out_dev = dst->dev; if (!mpls_output_possible(out_dev) || - !lwtstate || skb_warn_if_lro(skb)) + !dst->lwtstate || skb_warn_if_lro(skb)) goto drop; skb_forward_csum(skb); - tun_encap_info = mpls_lwtunnel_encap(lwtstate); + tun_encap_info = mpls_lwtunnel_encap(dst->lwtstate); /* Verify the destination can hold the packet */ new_header_size = mpls_encap_size(tun_encap_info); diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index 4b70aaa4a746..a75011505039 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -57,7 +57,7 @@ static void netdev_port_receive(struct vport *vport, struct sk_buff *skb) skb_push(skb, ETH_HLEN); ovs_skb_postpush_rcsum(skb, skb->data, ETH_HLEN); - ovs_vport_receive(vport, skb, skb_tunnel_info(skb, AF_INET)); + ovs_vport_receive(vport, skb, skb_tunnel_info(skb)); return; error: -- cgit v1.2.3 From 751a587ac9f9a8bf314590fbac32d9e418060c5a Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Fri, 21 Aug 2015 12:41:14 +0200 Subject: route: fix breakage after moving lwtunnel state __recnt and related fields need to be in its own cacheline for performance reasons. Commit 61adedf3e3f1 ("route: move lwtunnel state to dst_entry") broke that on 32bit archs, causing BUILD_BUG_ON in dst_hold to be triggered. This patch fixes the breakage by moving the lwtunnel state to the end of dst_entry on 32bit archs. Unfortunately, this makes it share the cacheline with __refcnt and may affect performance, thus further patches may be needed. Reported-by: kbuild test robot Fixes: 61adedf3e3f1 ("route: move lwtunnel state to dst_entry") Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- include/net/dst.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'include/net/dst.h') diff --git a/include/net/dst.h b/include/net/dst.h index 0a9a723f6c19..ef8f1d43a203 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -44,7 +44,6 @@ struct dst_entry { #else void *__pad1; #endif - struct lwtunnel_state *lwtstate; int (*input)(struct sk_buff *); int (*output)(struct sock *sk, struct sk_buff *skb); @@ -85,11 +84,12 @@ struct dst_entry { __u32 __pad2; #endif +#ifdef CONFIG_64BIT + struct lwtunnel_state *lwtstate; /* * Align __refcnt to a 64 bytes alignment * (L1_CACHE_SIZE would be too much) */ -#ifdef CONFIG_64BIT long __pad_to_align_refcnt[1]; #endif /* @@ -99,6 +99,9 @@ struct dst_entry { atomic_t __refcnt; /* client references */ int __use; unsigned long lastuse; +#ifndef CONFIG_64BIT + struct lwtunnel_state *lwtstate; +#endif union { struct dst_entry *next; struct rtable __rcu *rt_next; -- cgit v1.2.3 From e79e259588a414589a016edc428ee8dd308f81ad Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Wed, 26 Aug 2015 11:31:47 -0700 Subject: dst: Add __skb_dst_copy() variation This variation on skb_dst_copy() doesn't require two skbs. Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Acked-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/dst.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'include/net/dst.h') diff --git a/include/net/dst.h b/include/net/dst.h index ef8f1d43a203..4c4801645371 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -289,13 +289,18 @@ static inline void skb_dst_drop(struct sk_buff *skb) } } -static inline void skb_dst_copy(struct sk_buff *nskb, const struct sk_buff *oskb) +static inline void __skb_dst_copy(struct sk_buff *nskb, unsigned long refdst) { - nskb->_skb_refdst = oskb->_skb_refdst; + nskb->_skb_refdst = refdst; if (!(nskb->_skb_refdst & SKB_DST_NOREF)) dst_clone(skb_dst(nskb)); } +static inline void skb_dst_copy(struct sk_buff *nskb, const struct sk_buff *oskb) +{ + __skb_dst_copy(nskb, oskb->_skb_refdst); +} + /** * skb_dst_force - makes sure skb dst is refcounted * @skb: buffer -- cgit v1.2.3 From c3a8d9474684d391b0afc3970d9b249add15ec07 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Mon, 31 Aug 2015 15:58:47 +0200 Subject: tcp: use dctcp if enabled on the route to the initiator Currently, the following case doesn't use DCTCP, even if it should: A responder has f.e. Cubic as system wide default, but for a specific route to the initiating host, DCTCP is being set in RTAX_CC_ALGO. The initiating host then uses DCTCP as congestion control, but since the initiator sets ECT(0), tcp_ecn_create_request() doesn't set ecn_ok, and we have to fall back to Reno after 3WHS completes. We were thinking on how to solve this in a minimal, non-intrusive way without bloating tcp_ecn_create_request() needlessly: lets cache the CA ecn option flag in RTAX_FEATURES. In other words, when ECT(0) is set on the SYN packet, set ecn_ok=1 iff route RTAX_FEATURES contains the unexposed (internal-only) DST_FEATURE_ECN_CA. This allows to only do a single metric feature lookup inside tcp_ecn_create_request(). Joint work with Florian Westphal. Signed-off-by: Daniel Borkmann Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/net/dst.h | 6 ++++++ include/net/tcp.h | 2 +- net/core/rtnetlink.c | 6 ++++++ net/ipv4/fib_semantics.c | 6 +++++- net/ipv4/tcp_cong.c | 9 ++++++--- net/ipv4/tcp_input.c | 7 +++++-- net/ipv6/route.c | 9 +++++++-- 7 files changed, 36 insertions(+), 9 deletions(-) (limited to 'include/net/dst.h') diff --git a/include/net/dst.h b/include/net/dst.h index 4c4801645371..9261d928303d 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -207,6 +207,12 @@ static inline void dst_metric_set(struct dst_entry *dst, int metric, u32 val) p[metric-1] = val; } +/* Kernel-internal feature bits that are unallocated in user space. */ +#define DST_FEATURE_ECN_CA (1 << 31) + +#define DST_FEATURE_MASK (DST_FEATURE_ECN_CA) +#define DST_FEATURE_ECN_MASK (DST_FEATURE_ECN_CA | RTAX_FEATURE_ECN) + static inline u32 dst_feature(const struct dst_entry *dst, u32 feature) { diff --git a/include/net/tcp.h b/include/net/tcp.h index 4a7b03947a38..0cab28cd43a9 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -888,7 +888,7 @@ void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 acked); extern struct tcp_congestion_ops tcp_reno; struct tcp_congestion_ops *tcp_ca_find_key(u32 key); -u32 tcp_ca_get_key_by_name(const char *name); +u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca); #ifdef CONFIG_INET char *tcp_ca_get_name_by_key(u32 key, char *buffer); #else diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 788ceed39463..a466821d1441 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -678,6 +678,12 @@ int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics) continue; if (nla_put_string(skb, i + 1, name)) goto nla_put_failure; + } else if (i == RTAX_FEATURES - 1) { + u32 user_features = metrics[i] & RTAX_FEATURE_MASK; + + BUILD_BUG_ON(RTAX_FEATURE_MASK & DST_FEATURE_MASK); + if (nla_put_u32(skb, i + 1, user_features)) + goto nla_put_failure; } else { if (nla_put_u32(skb, i + 1, metrics[i])) goto nla_put_failure; diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 115a08e70d43..992a9597daf8 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -879,6 +879,7 @@ static bool fib_valid_prefsrc(struct fib_config *cfg, __be32 fib_prefsrc) static int fib_convert_metrics(struct fib_info *fi, const struct fib_config *cfg) { + bool ecn_ca = false; struct nlattr *nla; int remaining; @@ -898,7 +899,7 @@ fib_convert_metrics(struct fib_info *fi, const struct fib_config *cfg) char tmp[TCP_CA_NAME_MAX]; nla_strlcpy(tmp, nla, sizeof(tmp)); - val = tcp_ca_get_key_by_name(tmp); + val = tcp_ca_get_key_by_name(tmp, &ecn_ca); if (val == TCP_CA_UNSPEC) return -EINVAL; } else { @@ -913,6 +914,9 @@ fib_convert_metrics(struct fib_info *fi, const struct fib_config *cfg) fi->fib_metrics[type - 1] = val; } + if (ecn_ca) + fi->fib_metrics[RTAX_FEATURES - 1] |= DST_FEATURE_ECN_CA; + return 0; } diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index a2ed23c595cf..93c4dc3ab23f 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -114,16 +114,19 @@ void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca) } EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control); -u32 tcp_ca_get_key_by_name(const char *name) +u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca) { const struct tcp_congestion_ops *ca; - u32 key; + u32 key = TCP_CA_UNSPEC; might_sleep(); rcu_read_lock(); ca = __tcp_ca_find_autoload(name); - key = ca ? ca->key : TCP_CA_UNSPEC; + if (ca) { + key = ca->key; + *ecn_ca = ca->flags & TCP_CONG_NEEDS_ECN; + } rcu_read_unlock(); return key; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index dc08e2352665..a8f515bb19c4 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6003,14 +6003,17 @@ static void tcp_ecn_create_request(struct request_sock *req, const struct net *net = sock_net(listen_sk); bool th_ecn = th->ece && th->cwr; bool ect, ecn_ok; + u32 ecn_ok_dst; if (!th_ecn) return; ect = !INET_ECN_is_not_ect(TCP_SKB_CB(skb)->ip_dsfield); - ecn_ok = net->ipv4.sysctl_tcp_ecn || dst_feature(dst, RTAX_FEATURE_ECN); + ecn_ok_dst = dst_feature(dst, DST_FEATURE_ECN_MASK); + ecn_ok = net->ipv4.sysctl_tcp_ecn || ecn_ok_dst; - if ((!ect && ecn_ok) || tcp_ca_needs_ecn(listen_sk)) + if ((!ect && ecn_ok) || tcp_ca_needs_ecn(listen_sk) || + (ecn_ok_dst & DST_FEATURE_ECN_CA)) inet_rsk(req)->ecn_ok = 1; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8771530df45e..f45cac6f8356 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1698,6 +1698,7 @@ out: static int ip6_convert_metrics(struct mx6_config *mxc, const struct fib6_config *cfg) { + bool ecn_ca = false; struct nlattr *nla; int remaining; u32 *mp; @@ -1722,7 +1723,7 @@ static int ip6_convert_metrics(struct mx6_config *mxc, char tmp[TCP_CA_NAME_MAX]; nla_strlcpy(tmp, nla, sizeof(tmp)); - val = tcp_ca_get_key_by_name(tmp); + val = tcp_ca_get_key_by_name(tmp, &ecn_ca); if (val == TCP_CA_UNSPEC) goto err; } else { @@ -1735,8 +1736,12 @@ static int ip6_convert_metrics(struct mx6_config *mxc, __set_bit(type - 1, mxc->mx_valid); } - mxc->mx = mp; + if (ecn_ca) { + __set_bit(RTAX_FEATURES - 1, mxc->mx_valid); + mp[RTAX_FEATURES - 1] |= DST_FEATURE_ECN_CA; + } + mxc->mx = mp; return 0; err: kfree(mp); -- cgit v1.2.3