From 85e482285bbbd508483cbe08de69c8fe00cdbbfe Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Wed, 27 Sep 2017 08:23:11 +0200 Subject: fib: notifier: Add VIF add and delete event types In order for an interface to forward packets according to the kernel multicast routing table, it must be configured with a VIF index according to the mroute user API. The VIF index is then used to refer to that interface in the mroute user API, for example, to set the iif and oifs of an MFC entry. In order to allow drivers to be aware and offload multicast routes, they have to be aware of the VIF add and delete notifications. Due to the fact that a specific VIF can be deleted and re-added pointing to another netdevice, and the MFC routes that point to it will forward the matching packets to the new netdevice, a driver willing to offload MFC cache entries must be aware of the VIF add and delete events in addition to MFC routes notifications. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/net/fib_notifier.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/net/fib_notifier.h b/include/net/fib_notifier.h index 669b9716dc7a..54cd6b839d2f 100644 --- a/include/net/fib_notifier.h +++ b/include/net/fib_notifier.h @@ -20,6 +20,8 @@ enum fib_event_type { FIB_EVENT_RULE_DEL, FIB_EVENT_NH_ADD, FIB_EVENT_NH_DEL, + FIB_EVENT_VIF_ADD, + FIB_EVENT_VIF_DEL, }; struct fib_notifier_ops { -- cgit v1.2.3 From 310ebbba3b7396b00bce08a33f1d2de2c74fa257 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Wed, 27 Sep 2017 08:23:12 +0200 Subject: ipmr: Add reference count to MFC entries Next commits will introduce MFC notifications through the atomic fib_notification chain, thus allowing modules to be aware of MFC entries. Due to the fact that modules may need to hold a reference to an MFC entry, add reference count to MFC entries to prevent them from being freed while these modules use them. The reference counting is done only on resolved MFC entries currently. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 21 +++++++++++++++++++++ net/ipv4/ipmr.c | 8 +++++--- 2 files changed, 26 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index d7f63339ef0b..10028f208efb 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -109,6 +109,7 @@ struct mfc_cache_cmp_arg { * @wrong_if: number of wrong source interface hits * @lastuse: time of last use of the group (traffic or update) * @ttls: OIF TTL threshold array + * @refcount: reference count for this entry * @list: global entry list * @rcu: used for entry destruction */ @@ -138,6 +139,7 @@ struct mfc_cache { unsigned long wrong_if; unsigned long lastuse; unsigned char ttls[MAXVIFS]; + refcount_t refcount; } res; } mfc_un; struct list_head list; @@ -148,4 +150,23 @@ struct rtmsg; int ipmr_get_route(struct net *net, struct sk_buff *skb, __be32 saddr, __be32 daddr, struct rtmsg *rtm, u32 portid); + +#ifdef CONFIG_IP_MROUTE +void ipmr_cache_free(struct mfc_cache *mfc_cache); +#else +static inline void ipmr_cache_free(struct mfc_cache *mfc_cache) +{ +} +#endif + +static inline void ipmr_cache_put(struct mfc_cache *c) +{ + if (refcount_dec_and_test(&c->mfc_un.res.refcount)) + ipmr_cache_free(c); +} +static inline void ipmr_cache_hold(struct mfc_cache *c) +{ + refcount_inc(&c->mfc_un.res.refcount); +} + #endif diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index c9b3e6e069ae..86dc5f98c5dd 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -652,10 +652,11 @@ static void ipmr_cache_free_rcu(struct rcu_head *head) kmem_cache_free(mrt_cachep, c); } -static inline void ipmr_cache_free(struct mfc_cache *c) +void ipmr_cache_free(struct mfc_cache *c) { call_rcu(&c->rcu, ipmr_cache_free_rcu); } +EXPORT_SYMBOL(ipmr_cache_free); /* Destroy an unresolved cache entry, killing queued skbs * and reporting error to netlink readers. @@ -949,6 +950,7 @@ static struct mfc_cache *ipmr_cache_alloc(void) if (c) { c->mfc_un.res.last_assert = jiffies - MFC_ASSERT_THRESH - 1; c->mfc_un.res.minvif = MAXVIFS; + refcount_set(&c->mfc_un.res.refcount, 1); } return c; } @@ -1162,7 +1164,7 @@ static int ipmr_mfc_delete(struct mr_table *mrt, struct mfcctl *mfc, int parent) rhltable_remove(&mrt->mfc_hash, &c->mnode, ipmr_rht_params); list_del_rcu(&c->list); mroute_netlink_event(mrt, c, RTM_DELROUTE); - ipmr_cache_free(c); + ipmr_cache_put(c); return 0; } @@ -1264,7 +1266,7 @@ static void mroute_clean_tables(struct mr_table *mrt, bool all) rhltable_remove(&mrt->mfc_hash, &c->mnode, ipmr_rht_params); list_del_rcu(&c->list); mroute_netlink_event(mrt, c, RTM_DELROUTE); - ipmr_cache_free(c); + ipmr_cache_put(c); } if (atomic_read(&mrt->cache_resolve_queue_len) != 0) { -- cgit v1.2.3 From 4d65b9487831170e699b2fc64a91b839d729bd78 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Wed, 27 Sep 2017 08:23:13 +0200 Subject: ipmr: Add FIB notification access functions Make the ipmr module register as a FIB notifier. To do that, implement both the ipmr_seq_read and ipmr_dump ops. The ipmr_seq_read op returns a sequence counter that is incremented on every notification related operation done by the ipmr. To implement that, add a sequence counter in the netns_ipv4 struct and increment it whenever a new MFC route or VIF are added or deleted. The sequence operations are protected by the RTNL lock. The ipmr_dump iterates the list of MFC routes and the list of VIF entries and sends notifications about them. The entries dump is done under RCU where the VIF dump uses the mrt_lock too, as the vif->dev field can change under RCU. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 15 ++++++ include/net/netns/ipv4.h | 3 ++ net/ipv4/ipmr.c | 137 ++++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 153 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index 10028f208efb..54c5cb82ddcb 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -5,6 +5,7 @@ #include #include #include +#include #include #ifdef CONFIG_IP_MROUTE @@ -58,6 +59,14 @@ struct vif_device { int link; /* Physical interface index */ }; +struct vif_entry_notifier_info { + struct fib_notifier_info info; + struct net_device *dev; + vifi_t vif_index; + unsigned short vif_flags; + u32 tb_id; +}; + #define VIFF_STATIC 0x8000 #define VIF_EXISTS(_mrt, _idx) ((_mrt)->vif_table[_idx].dev != NULL) @@ -146,6 +155,12 @@ struct mfc_cache { struct rcu_head rcu; }; +struct mfc_entry_notifier_info { + struct fib_notifier_info info; + struct mfc_cache *mfc; + u32 tb_id; +}; + struct rtmsg; int ipmr_get_route(struct net *net, struct sk_buff *skb, __be32 saddr, __be32 daddr, diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 8387f099115e..abc84d986da4 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -163,6 +163,9 @@ struct netns_ipv4 { struct fib_notifier_ops *notifier_ops; unsigned int fib_seq; /* protected by rtnl_mutex */ + struct fib_notifier_ops *ipmr_notifier_ops; + unsigned int ipmr_seq; /* protected by rtnl_mutex */ + atomic_t rt_genid; }; #endif diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 86dc5f98c5dd..49879c338357 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -264,6 +264,16 @@ static void __net_exit ipmr_rules_exit(struct net *net) fib_rules_unregister(net->ipv4.mr_rules_ops); rtnl_unlock(); } + +static int ipmr_rules_dump(struct net *net, struct notifier_block *nb) +{ + return fib_rules_dump(net, nb, RTNL_FAMILY_IPMR); +} + +static unsigned int ipmr_rules_seq_read(struct net *net) +{ + return fib_rules_seq_read(net, RTNL_FAMILY_IPMR); +} #else #define ipmr_for_each_table(mrt, net) \ for (mrt = net->ipv4.mrt; mrt; mrt = NULL) @@ -298,6 +308,16 @@ static void __net_exit ipmr_rules_exit(struct net *net) net->ipv4.mrt = NULL; rtnl_unlock(); } + +static int ipmr_rules_dump(struct net *net, struct notifier_block *nb) +{ + return 0; +} + +static unsigned int ipmr_rules_seq_read(struct net *net) +{ + return 0; +} #endif static inline int ipmr_hash_cmp(struct rhashtable_compare_arg *arg, @@ -587,6 +607,43 @@ static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) } #endif +static int call_ipmr_vif_entry_notifier(struct notifier_block *nb, + struct net *net, + enum fib_event_type event_type, + struct vif_device *vif, + vifi_t vif_index, u32 tb_id) +{ + struct vif_entry_notifier_info info = { + .info = { + .family = RTNL_FAMILY_IPMR, + .net = net, + }, + .dev = vif->dev, + .vif_index = vif_index, + .vif_flags = vif->flags, + .tb_id = tb_id, + }; + + return call_fib_notifier(nb, net, event_type, &info.info); +} + +static int call_ipmr_mfc_entry_notifier(struct notifier_block *nb, + struct net *net, + enum fib_event_type event_type, + struct mfc_cache *mfc, u32 tb_id) +{ + struct mfc_entry_notifier_info info = { + .info = { + .family = RTNL_FAMILY_IPMR, + .net = net, + }, + .mfc = mfc, + .tb_id = tb_id + }; + + return call_fib_notifier(nb, net, event_type, &info.info); +} + /** * vif_delete - Delete a VIF entry * @notify: Set to 1, if the caller is a notifier_call @@ -3050,14 +3107,87 @@ static const struct net_protocol pim_protocol = { }; #endif +static unsigned int ipmr_seq_read(struct net *net) +{ + ASSERT_RTNL(); + + return net->ipv4.ipmr_seq + ipmr_rules_seq_read(net); +} + +static int ipmr_dump(struct net *net, struct notifier_block *nb) +{ + struct mr_table *mrt; + int err; + + err = ipmr_rules_dump(net, nb); + if (err) + return err; + + ipmr_for_each_table(mrt, net) { + struct vif_device *v = &mrt->vif_table[0]; + struct mfc_cache *mfc; + int vifi; + + /* Notifiy on table VIF entries */ + read_lock(&mrt_lock); + for (vifi = 0; vifi < mrt->maxvif; vifi++, v++) { + if (!v->dev) + continue; + + call_ipmr_vif_entry_notifier(nb, net, FIB_EVENT_VIF_ADD, + v, vifi, mrt->id); + } + read_unlock(&mrt_lock); + + /* Notify on table MFC entries */ + list_for_each_entry_rcu(mfc, &mrt->mfc_cache_list, list) + call_ipmr_mfc_entry_notifier(nb, net, + FIB_EVENT_ENTRY_ADD, mfc, + mrt->id); + } + + return 0; +} + +static const struct fib_notifier_ops ipmr_notifier_ops_template = { + .family = RTNL_FAMILY_IPMR, + .fib_seq_read = ipmr_seq_read, + .fib_dump = ipmr_dump, + .owner = THIS_MODULE, +}; + +int __net_init ipmr_notifier_init(struct net *net) +{ + struct fib_notifier_ops *ops; + + net->ipv4.ipmr_seq = 0; + + ops = fib_notifier_ops_register(&ipmr_notifier_ops_template, net); + if (IS_ERR(ops)) + return PTR_ERR(ops); + net->ipv4.ipmr_notifier_ops = ops; + + return 0; +} + +static void __net_exit ipmr_notifier_exit(struct net *net) +{ + fib_notifier_ops_unregister(net->ipv4.ipmr_notifier_ops); + net->ipv4.ipmr_notifier_ops = NULL; +} + /* Setup for IP multicast routing */ static int __net_init ipmr_net_init(struct net *net) { int err; + err = ipmr_notifier_init(net); + if (err) + goto ipmr_notifier_fail; + err = ipmr_rules_init(net); if (err < 0) - goto fail; + goto ipmr_rules_fail; #ifdef CONFIG_PROC_FS err = -ENOMEM; @@ -3074,7 +3204,9 @@ proc_cache_fail: proc_vif_fail: ipmr_rules_exit(net); #endif -fail: +ipmr_rules_fail: + ipmr_notifier_exit(net); +ipmr_notifier_fail: return err; } @@ -3084,6 +3216,7 @@ static void __net_exit ipmr_net_exit(struct net *net) remove_proc_entry("ip_mr_cache", net->proc_net); remove_proc_entry("ip_mr_vif", net->proc_net); #endif + ipmr_notifier_exit(net); ipmr_rules_exit(net); } -- cgit v1.2.3 From c7c0bbeae9501a7e42f2fd306d6a6399aca688b6 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Wed, 27 Sep 2017 08:23:15 +0200 Subject: net: ipmr: Add MFC offload indication Allow drivers, registered to the fib notification chain indicate whether a multicast MFC route is offloaded or not, similarly to unicast routes. The indication of whether a route is offloaded is done using the mfc_flags field on an mfc_cache struct, and the information is sent to the userspace via the RTNetlink interface only. Currently, MFC routes are either offloaded or not, thus there is no need to add per-VIF offload indication. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 2 ++ net/ipv4/ipmr.c | 3 +++ 2 files changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index 54c5cb82ddcb..5566580811ce 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -90,9 +90,11 @@ struct mr_table { /* mfc_flags: * MFC_STATIC - the entry was added statically (not by a routing daemon) + * MFC_OFFLOAD - the entry was offloaded to the hardware */ enum { MFC_STATIC = BIT(0), + MFC_OFFLOAD = BIT(1), }; struct mfc_cache_cmp_arg { diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index ba71bc402336..2a795d2c0502 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -2268,6 +2268,9 @@ static int __ipmr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, nla_put_u32(skb, RTA_IIF, mrt->vif_table[c->mfc_parent].dev->ifindex) < 0) return -EMSGSIZE; + if (c->mfc_flags & MFC_OFFLOAD) + rtm->rtm_flags |= RTNH_F_OFFLOAD; + if (!(mp_attr = nla_nest_start(skb, RTA_MULTIPATH))) return -EMSGSIZE; -- cgit v1.2.3 From 478e4c2f0067d57d7c17059caafab026ca32084a Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Wed, 27 Sep 2017 08:23:16 +0200 Subject: net: mroute: Check if rule is a default rule When the ipmr starts, it adds one default FIB rule that matches all packets and sends them to the DEFAULT (multicast) FIB table. A more complex rule can be added by user to specify that for a specific interface, a packet should be look up at either an arbitrary table or according to the l3mdev of the interface. For drivers willing to offload the ipmr logic into a hardware but don't want to offload all the FIB rules functionality, provide a function that can indicate whether the FIB rule is the default multicast rule, thus only one routing table is needed. This way, a driver can register to the FIB notification chain, get notifications about FIB rules added and trigger some kind of an internal abort mechanism when a non default rule is added by the user. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 7 +++++++ net/ipv4/ipmr.c | 12 ++++++++++++ 2 files changed, 19 insertions(+) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index 5566580811ce..b072a84fbe1c 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -5,6 +5,7 @@ #include #include #include +#include #include #include @@ -19,6 +20,7 @@ int ip_mroute_getsockopt(struct sock *, int, char __user *, int __user *); int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg); int ipmr_compat_ioctl(struct sock *sk, unsigned int cmd, void __user *arg); int ip_mr_init(void); +bool ipmr_rule_default(const struct fib_rule *rule); #else static inline int ip_mroute_setsockopt(struct sock *sock, int optname, char __user *optval, unsigned int optlen) @@ -46,6 +48,11 @@ static inline int ip_mroute_opt(int opt) { return 0; } + +static inline bool ipmr_rule_default(const struct fib_rule *rule) +{ + return true; +} #endif struct vif_device { diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 2a795d2c0502..292a8e80bdfa 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -274,6 +274,12 @@ static unsigned int ipmr_rules_seq_read(struct net *net) { return fib_rules_seq_read(net, RTNL_FAMILY_IPMR); } + +bool ipmr_rule_default(const struct fib_rule *rule) +{ + return fib_rule_matchall(rule) && rule->table == RT_TABLE_DEFAULT; +} +EXPORT_SYMBOL(ipmr_rule_default); #else #define ipmr_for_each_table(mrt, net) \ for (mrt = net->ipv4.mrt; mrt; mrt = NULL) @@ -318,6 +324,12 @@ static unsigned int ipmr_rules_seq_read(struct net *net) { return 0; } + +bool ipmr_rule_default(const struct fib_rule *rule) +{ + return true; +} +EXPORT_SYMBOL(ipmr_rule_default); #endif static inline int ipmr_hash_cmp(struct rhashtable_compare_arg *arg, -- cgit v1.2.3