netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1)
@ 2025-10-14 14:06 Eric Dumazet
  2025-10-15  2:06 ` Kuniyuki Iwashima
                   ` (2 more replies)
  0 siblings, 3 replies; 5+ messages in thread
From: Eric Dumazet @ 2025-10-14 14:06 UTC (permalink / raw)
  To: David S . Miller, Jakub Kicinski, Paolo Abeni
  Cc: Simon Horman, netdev, eric.dumazet, Eric Dumazet, Xuanqiang Luo,
	Kuniyuki Iwashima

sk->sk_refcnt has been converted to refcount_t in 2017.

__sock_put(sk) being refcount_dec(&sk->sk_refcnt), it will complain
loudly if the current refcnt is 1 (or less) in a non racy way.

We can remove four WARN_ON() in favor of the generic refcount_dec()
check.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Xuanqiang Luo <luoxuanqiang@kylinos.cn>
Cc: Kuniyuki Iwashima <kuniyu@google.com>
---
 include/net/sock.h       | 12 ++++--------
 net/netlink/af_netlink.c |  4 +---
 net/tipc/socket.c        |  4 +---
 3 files changed, 6 insertions(+), 14 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 60bcb13f045c..596302bffedf 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -828,11 +828,9 @@ static inline bool sk_del_node_init(struct sock *sk)
 {
 	bool rc = __sk_del_node_init(sk);
 
-	if (rc) {
-		/* paranoid for a while -acme */
-		WARN_ON(refcount_read(&sk->sk_refcnt) == 1);
+	if (rc)
 		__sock_put(sk);
-	}
+
 	return rc;
 }
 #define sk_del_node_init_rcu(sk)	sk_del_node_init(sk)
@@ -850,11 +848,9 @@ static inline bool sk_nulls_del_node_init_rcu(struct sock *sk)
 {
 	bool rc = __sk_nulls_del_node_init_rcu(sk);
 
-	if (rc) {
-		/* paranoid for a while -acme */
-		WARN_ON(refcount_read(&sk->sk_refcnt) == 1);
+	if (rc)
 		__sock_put(sk);
-	}
+
 	return rc;
 }
 
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 2b46c0cd752a..687a84c48882 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -596,10 +596,8 @@ static void netlink_remove(struct sock *sk)
 
 	table = &nl_table[sk->sk_protocol];
 	if (!rhashtable_remove_fast(&table->hash, &nlk_sk(sk)->node,
-				    netlink_rhashtable_params)) {
-		WARN_ON(refcount_read(&sk->sk_refcnt) == 1);
+				    netlink_rhashtable_params))
 		__sock_put(sk);
-	}
 
 	netlink_table_grab();
 	if (nlk_sk(sk)->subscriptions) {
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 1574a83384f8..bc614a1f019c 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -3031,10 +3031,8 @@ static void tipc_sk_remove(struct tipc_sock *tsk)
 	struct sock *sk = &tsk->sk;
 	struct tipc_net *tn = net_generic(sock_net(sk), tipc_net_id);
 
-	if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params)) {
-		WARN_ON(refcount_read(&sk->sk_refcnt) == 1);
+	if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params))
 		__sock_put(sk);
-	}
 }
 
 static const struct rhashtable_params tsk_rht_params = {
-- 
2.51.0.788.g6d19910ace-goog


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1)
  2025-10-14 14:06 [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1) Eric Dumazet
@ 2025-10-15  2:06 ` Kuniyuki Iwashima
  2025-10-15  6:03 ` luoxuanqiang
  2025-10-16  0:40 ` patchwork-bot+netdevbpf
  2 siblings, 0 replies; 5+ messages in thread
From: Kuniyuki Iwashima @ 2025-10-15  2:06 UTC (permalink / raw)
  To: Eric Dumazet
  Cc: David S . Miller, Jakub Kicinski, Paolo Abeni, Simon Horman,
	netdev, eric.dumazet, Xuanqiang Luo

On Tue, Oct 14, 2025 at 7:06 AM Eric Dumazet <edumazet@google.com> wrote:
>
> sk->sk_refcnt has been converted to refcount_t in 2017.
>
> __sock_put(sk) being refcount_dec(&sk->sk_refcnt), it will complain
> loudly if the current refcnt is 1 (or less) in a non racy way.
>
> We can remove four WARN_ON() in favor of the generic refcount_dec()
> check.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>

Thanks for catching this!

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1)
  2025-10-14 14:06 [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1) Eric Dumazet
  2025-10-15  2:06 ` Kuniyuki Iwashima
@ 2025-10-15  6:03 ` luoxuanqiang
  2025-10-15  6:44   ` Eric Dumazet
  2025-10-16  0:40 ` patchwork-bot+netdevbpf
  2 siblings, 1 reply; 5+ messages in thread
From: luoxuanqiang @ 2025-10-15  6:03 UTC (permalink / raw)
  To: Eric Dumazet, David S . Miller, Jakub Kicinski, Paolo Abeni
  Cc: Simon Horman, netdev, eric.dumazet, Kuniyuki Iwashima


在 2025/10/14 22:06, Eric Dumazet 写道:
> sk->sk_refcnt has been converted to refcount_t in 2017.
>
> __sock_put(sk) being refcount_dec(&sk->sk_refcnt), it will complain
> loudly if the current refcnt is 1 (or less) in a non racy way.
>
> We can remove four WARN_ON() in favor of the generic refcount_dec()
> check.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Reviewed-by: Xuanqiang Luo<luoxuanqiang@kylinos.cn>

Dear Eric,

Following your line of thought, I found there's also a point in btrfs that
needs modification.

Would you like to modify it together? Though it has nothing to do with
socket, or shall I modify it separately later?

diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 89ae0c7a610a..485bef0ba419 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -138,7 +138,6 @@ static const unsigned int btrfs_blocked_trans_types[TRANS_STATE_MAX] = {
  void btrfs_put_transaction(struct btrfs_transaction *transaction)
  {
-       WARN_ON(refcount_read(&transaction->use_count) == 0);
         if (refcount_dec_and_test(&transaction->use_count)) {
                 BUG_ON(!list_empty(&transaction->list));
                 WARN_ON(!xa_empty(&transaction->delayed_refs.head_refs));

Thanks!


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1)
  2025-10-15  6:03 ` luoxuanqiang
@ 2025-10-15  6:44   ` Eric Dumazet
  0 siblings, 0 replies; 5+ messages in thread
From: Eric Dumazet @ 2025-10-15  6:44 UTC (permalink / raw)
  To: luoxuanqiang
  Cc: David S . Miller, Jakub Kicinski, Paolo Abeni, Simon Horman,
	netdev, eric.dumazet, Kuniyuki Iwashima

On Tue, Oct 14, 2025 at 11:04 PM luoxuanqiang <xuanqiang.luo@linux.dev> wrote:
>
>
> 在 2025/10/14 22:06, Eric Dumazet 写道:
> > sk->sk_refcnt has been converted to refcount_t in 2017.
> >
> > __sock_put(sk) being refcount_dec(&sk->sk_refcnt), it will complain
> > loudly if the current refcnt is 1 (or less) in a non racy way.
> >
> > We can remove four WARN_ON() in favor of the generic refcount_dec()
> > check.
> >
> > Signed-off-by: Eric Dumazet <edumazet@google.com>
>
> Reviewed-by: Xuanqiang Luo<luoxuanqiang@kylinos.cn>
>
> Dear Eric,
>
> Following your line of thought, I found there's also a point in btrfs that
> needs modification.
>
> Would you like to modify it together? Though it has nothing to do with
> socket, or shall I modify it separately later?

I removed stuff around sk_refcnt only.

If you want, please send a patch to the appropriate lists and maintainers.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1)
  2025-10-14 14:06 [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1) Eric Dumazet
  2025-10-15  2:06 ` Kuniyuki Iwashima
  2025-10-15  6:03 ` luoxuanqiang
@ 2025-10-16  0:40 ` patchwork-bot+netdevbpf
  2 siblings, 0 replies; 5+ messages in thread
From: patchwork-bot+netdevbpf @ 2025-10-16  0:40 UTC (permalink / raw)
  To: Eric Dumazet
  Cc: davem, kuba, pabeni, horms, netdev, eric.dumazet, luoxuanqiang,
	kuniyu

Hello:

This patch was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Tue, 14 Oct 2025 14:06:05 +0000 you wrote:
> sk->sk_refcnt has been converted to refcount_t in 2017.
> 
> __sock_put(sk) being refcount_dec(&sk->sk_refcnt), it will complain
> loudly if the current refcnt is 1 (or less) in a non racy way.
> 
> We can remove four WARN_ON() in favor of the generic refcount_dec()
> check.
> 
> [...]

Here is the summary with links:
  - [net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1)
    https://git.kernel.org/netdev/net-next/c/e5b670e5439b

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html



^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2025-10-16  0:40 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-10-14 14:06 [PATCH net-next] net: remove obsolete WARN_ON(refcount_read(&sk->sk_refcnt) == 1) Eric Dumazet
2025-10-15  2:06 ` Kuniyuki Iwashima
2025-10-15  6:03 ` luoxuanqiang
2025-10-15  6:44   ` Eric Dumazet
2025-10-16  0:40 ` patchwork-bot+netdevbpf

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).