From 7a60d91c690bf73c2c78e763efa29f294e217c3a Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 11 Apr 2025 13:52:32 -0700 Subject: net: Add ->exit_rtnl() hook to struct pernet_operations. struct pernet_operations provides two batching hooks; ->exit_batch() and ->exit_batch_rtnl(). The batching variant is beneficial if ->exit() meets any of the following conditions: 1) ->exit() repeatedly acquires a global lock for each netns 2) ->exit() has a time-consuming operation that can be factored out (e.g. synchronize_rcu(), smp_mb(), etc) 3) ->exit() does not need to repeat the same iterations for each netns (e.g. inet_twsk_purge()) Currently, none of the ->exit_batch_rtnl() functions satisfy any of the above conditions because RTNL is factored out and held by the caller and all of these functions iterate over the dying netns list. Also, we want to hold per-netns RTNL there but avoid spreading __rtnl_net_lock() across multiple locations. Let's add ->exit_rtnl() hook and run it under __rtnl_net_lock(). The following patches will convert all ->exit_batch_rtnl() users to ->exit_rtnl(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20250411205258.63164-4-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/net_namespace.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index bd57d8fb54f1..b071e6eed9d5 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -475,6 +475,8 @@ struct pernet_operations { void (*exit)(struct net *net); void (*exit_batch)(struct list_head *net_exit_list); /* Following method is called with RTNL held. */ + void (*exit_rtnl)(struct net *net, + struct list_head *dev_kill_list); void (*exit_batch_rtnl)(struct list_head *net_exit_list, struct list_head *dev_kill_list); unsigned int * const id; -- cgit v1.2.3 From a967e01e2ad201f6ddc778ed65a5dae1c68ee8a5 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 11 Apr 2025 13:52:35 -0700 Subject: ipv4: ip_tunnel: Convert ip_tunnel_delete_nets() callers to ->exit_rtnl(). ip_tunnel_delete_nets() iterates the dying netns list and performs the same operations for each. Let's export ip_tunnel_destroy() as ip_tunnel_delete_net() and call it from ->exit_rtnl(). Signed-off-by: Kuniyuki Iwashima Reviewed-by: David Ahern Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20250411205258.63164-7-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/ip_tunnels.h | 7 +++---- net/ipv4/ip_gre.c | 27 ++++++++++++--------------- net/ipv4/ip_tunnel.c | 25 +++++++------------------ net/ipv4/ip_vti.c | 9 ++++----- net/ipv4/ipip.c | 9 ++++----- 5 files changed, 30 insertions(+), 47 deletions(-) (limited to 'include') diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index a36a335cef9f..0c3d571a04a1 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -377,10 +377,9 @@ struct net *ip_tunnel_get_link_net(const struct net_device *dev); int ip_tunnel_get_iflink(const struct net_device *dev); int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id, struct rtnl_link_ops *ops, char *devname); - -void ip_tunnel_delete_nets(struct list_head *list_net, unsigned int id, - struct rtnl_link_ops *ops, - struct list_head *dev_to_kill); +void ip_tunnel_delete_net(struct net *net, unsigned int id, + struct rtnl_link_ops *ops, + struct list_head *dev_to_kill); void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, const struct iphdr *tnl_params, const u8 protocol); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 26d15f907551..f5b9004d6938 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -1066,16 +1066,15 @@ static int __net_init ipgre_init_net(struct net *net) return ip_tunnel_init_net(net, ipgre_net_id, &ipgre_link_ops, NULL); } -static void __net_exit ipgre_exit_batch_rtnl(struct list_head *list_net, - struct list_head *dev_to_kill) +static void __net_exit ipgre_exit_rtnl(struct net *net, + struct list_head *dev_to_kill) { - ip_tunnel_delete_nets(list_net, ipgre_net_id, &ipgre_link_ops, - dev_to_kill); + ip_tunnel_delete_net(net, ipgre_net_id, &ipgre_link_ops, dev_to_kill); } static struct pernet_operations ipgre_net_ops = { .init = ipgre_init_net, - .exit_batch_rtnl = ipgre_exit_batch_rtnl, + .exit_rtnl = ipgre_exit_rtnl, .id = &ipgre_net_id, .size = sizeof(struct ip_tunnel_net), }; @@ -1752,16 +1751,15 @@ static int __net_init ipgre_tap_init_net(struct net *net) return ip_tunnel_init_net(net, gre_tap_net_id, &ipgre_tap_ops, "gretap0"); } -static void __net_exit ipgre_tap_exit_batch_rtnl(struct list_head *list_net, - struct list_head *dev_to_kill) +static void __net_exit ipgre_tap_exit_rtnl(struct net *net, + struct list_head *dev_to_kill) { - ip_tunnel_delete_nets(list_net, gre_tap_net_id, &ipgre_tap_ops, - dev_to_kill); + ip_tunnel_delete_net(net, gre_tap_net_id, &ipgre_tap_ops, dev_to_kill); } static struct pernet_operations ipgre_tap_net_ops = { .init = ipgre_tap_init_net, - .exit_batch_rtnl = ipgre_tap_exit_batch_rtnl, + .exit_rtnl = ipgre_tap_exit_rtnl, .id = &gre_tap_net_id, .size = sizeof(struct ip_tunnel_net), }; @@ -1772,16 +1770,15 @@ static int __net_init erspan_init_net(struct net *net) &erspan_link_ops, "erspan0"); } -static void __net_exit erspan_exit_batch_rtnl(struct list_head *net_list, - struct list_head *dev_to_kill) +static void __net_exit erspan_exit_rtnl(struct net *net, + struct list_head *dev_to_kill) { - ip_tunnel_delete_nets(net_list, erspan_net_id, &erspan_link_ops, - dev_to_kill); + ip_tunnel_delete_net(net, erspan_net_id, &erspan_link_ops, dev_to_kill); } static struct pernet_operations erspan_net_ops = { .init = erspan_init_net, - .exit_batch_rtnl = erspan_exit_batch_rtnl, + .exit_rtnl = erspan_exit_rtnl, .id = &erspan_net_id, .size = sizeof(struct ip_tunnel_net), }; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 1024f961ec9a..3913ec89ad20 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -1174,13 +1174,16 @@ int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id, } EXPORT_SYMBOL_GPL(ip_tunnel_init_net); -static void ip_tunnel_destroy(struct net *net, struct ip_tunnel_net *itn, - struct list_head *head, - struct rtnl_link_ops *ops) +void ip_tunnel_delete_net(struct net *net, unsigned int id, + struct rtnl_link_ops *ops, + struct list_head *head) { + struct ip_tunnel_net *itn = net_generic(net, id); struct net_device *dev, *aux; int h; + ASSERT_RTNL_NET(net); + for_each_netdev_safe(net, dev, aux) if (dev->rtnl_link_ops == ops) unregister_netdevice_queue(dev, head); @@ -1198,21 +1201,7 @@ static void ip_tunnel_destroy(struct net *net, struct ip_tunnel_net *itn, unregister_netdevice_queue(t->dev, head); } } - -void ip_tunnel_delete_nets(struct list_head *net_list, unsigned int id, - struct rtnl_link_ops *ops, - struct list_head *dev_to_kill) -{ - struct ip_tunnel_net *itn; - struct net *net; - - ASSERT_RTNL(); - list_for_each_entry(net, net_list, exit_list) { - itn = net_generic(net, id); - ip_tunnel_destroy(net, itn, dev_to_kill, ops); - } -} -EXPORT_SYMBOL_GPL(ip_tunnel_delete_nets); +EXPORT_SYMBOL_GPL(ip_tunnel_delete_net); int ip_tunnel_newlink(struct net *net, struct net_device *dev, struct nlattr *tb[], struct ip_tunnel_parm_kern *p, diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 159b4473290e..686e4f3d83aa 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -523,16 +523,15 @@ static int __net_init vti_init_net(struct net *net) return 0; } -static void __net_exit vti_exit_batch_rtnl(struct list_head *list_net, - struct list_head *dev_to_kill) +static void __net_exit vti_exit_rtnl(struct net *net, + struct list_head *dev_to_kill) { - ip_tunnel_delete_nets(list_net, vti_net_id, &vti_link_ops, - dev_to_kill); + ip_tunnel_delete_net(net, vti_net_id, &vti_link_ops, dev_to_kill); } static struct pernet_operations vti_net_ops = { .init = vti_init_net, - .exit_batch_rtnl = vti_exit_batch_rtnl, + .exit_rtnl = vti_exit_rtnl, .id = &vti_net_id, .size = sizeof(struct ip_tunnel_net), }; diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index bab0bf90c908..3e03af073a1c 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -604,16 +604,15 @@ static int __net_init ipip_init_net(struct net *net) return ip_tunnel_init_net(net, ipip_net_id, &ipip_link_ops, "tunl0"); } -static void __net_exit ipip_exit_batch_rtnl(struct list_head *list_net, - struct list_head *dev_to_kill) +static void __net_exit ipip_exit_rtnl(struct net *net, + struct list_head *dev_to_kill) { - ip_tunnel_delete_nets(list_net, ipip_net_id, &ipip_link_ops, - dev_to_kill); + ip_tunnel_delete_net(net, ipip_net_id, &ipip_link_ops, dev_to_kill); } static struct pernet_operations ipip_net_ops = { .init = ipip_init_net, - .exit_batch_rtnl = ipip_exit_batch_rtnl, + .exit_rtnl = ipip_exit_rtnl, .id = &ipip_net_id, .size = sizeof(struct ip_tunnel_net), }; -- cgit v1.2.3 From c57a9c503543cd8829eeaaf88362199e0491c0d7 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 11 Apr 2025 13:52:43 -0700 Subject: net: Remove ->exit_batch_rtnl(). There are no ->exit_batch_rtnl() users remaining. Let's remove the hook. Signed-off-by: Kuniyuki Iwashima Reviewed-by: Sabrina Dubroca Link: https://patch.msgid.link/20250411205258.63164-15-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/net_namespace.h | 2 -- net/core/net_namespace.c | 8 +------- 2 files changed, 1 insertion(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index b071e6eed9d5..025a7574b275 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -477,8 +477,6 @@ struct pernet_operations { /* Following method is called with RTNL held. */ void (*exit_rtnl)(struct net *net, struct list_head *dev_kill_list); - void (*exit_batch_rtnl)(struct list_head *net_exit_list, - struct list_head *dev_kill_list); unsigned int * const id; const size_t size; }; diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index afaa3d1bda8d..0a2b24af4028 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -185,12 +185,6 @@ static void ops_exit_rtnl_list(const struct list_head *ops_list, __rtnl_net_unlock(net); } - ops = saved_ops; - list_for_each_entry_continue_reverse(ops, ops_list, list) { - if (ops->exit_batch_rtnl) - ops->exit_batch_rtnl(net_exit_list, &dev_kill_list); - } - unregister_netdevice_many(&dev_kill_list); rtnl_unlock(); @@ -263,7 +257,7 @@ static void ops_undo_list(const struct list_head *ops_list, static void ops_undo_single(struct pernet_operations *ops, struct list_head *net_exit_list) { - bool hold_rtnl = ops->exit_rtnl || ops->exit_batch_rtnl; + bool hold_rtnl = !!ops->exit_rtnl; LIST_HEAD(ops_list); list_add(&ops->list, &ops_list); -- cgit v1.2.3