netdev - [PATCH v1 net-next 03/14] net: Add ->exit_rtnl() hook to struct pernet

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20250410022004.8668-4-kuniyu@amazon.com>
Date: Wed, 9 Apr 2025 19:19:24 -0700
From: Kuniyuki Iwashima <kuniyu@...zon.com>
To: "David S. Miller" <davem@...emloft.net>, Eric Dumazet
	<edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>, Paolo Abeni
	<pabeni@...hat.com>
CC: Simon Horman <horms@...nel.org>, Kuniyuki Iwashima <kuniyu@...zon.com>,
	Kuniyuki Iwashima <kuni1840@...il.com>, <netdev@...r.kernel.org>
Subject: [PATCH v1 net-next 03/14] 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 <kuniyu@...zon.com>
---
 include/net/net_namespace.h |  2 ++
 net/core/net_namespace.c    | 53 +++++++++++++++++++++++++++----------
 2 files changed, 41 insertions(+), 14 deletions(-)

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;
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 5e0ee117cf25..7ea818342721 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -163,16 +163,51 @@ static void ops_pre_exit_list(const struct pernet_operations *ops,
 	}
 }
 
+static void ops_exit_rtnl_list(const struct list_head *ops_list,
+			       const struct pernet_operations *ops,
+			       struct list_head *net_exit_list)
+{
+	const struct pernet_operations *saved_ops = ops;
+	LIST_HEAD(dev_kill_list);
+	struct net *net;
+
+	rtnl_lock();
+
+	list_for_each_entry(net, net_exit_list, exit_list) {
+		__rtnl_net_lock(net);
+
+		ops = saved_ops;
+		list_for_each_entry_continue_reverse(ops, ops_list, list) {
+			if (ops->exit_rtnl)
+				ops->exit_rtnl(net, &dev_kill_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();
+}
+
 static void ops_exit_list(const struct pernet_operations *ops,
 			  struct list_head *net_exit_list)
 {
-	struct net *net;
 	if (ops->exit) {
+		struct net *net;
+
 		list_for_each_entry(net, net_exit_list, exit_list) {
 			ops->exit(net);
 			cond_resched();
 		}
 	}
+
 	if (ops->exit_batch)
 		ops->exit_batch(net_exit_list);
 }
@@ -213,18 +248,8 @@ static void ops_undo_list(const struct list_head *ops_list,
 	else
 		synchronize_rcu();
 
-	if (hold_rtnl) {
-		LIST_HEAD(dev_kill_list);
-
-		ops = saved_ops;
-		rtnl_lock();
-		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();
-	}
+	if (hold_rtnl)
+		ops_exit_rtnl_list(ops_list, saved_ops, net_exit_list);
 
 	ops = saved_ops;
 	list_for_each_entry_continue_reverse(ops, ops_list, list)
@@ -238,7 +263,7 @@ 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_batch_rtnl;
+	bool hold_rtnl = ops->exit_rtnl || ops->exit_batch_rtnl;
 	LIST_HEAD(ops_list);
 
 	list_add(&ops->list, &ops_list);
-- 
2.49.0