[PATCH net-next] vsock/vmci: use sk_acceptq_is_full() helper

[PATCH net-next] vsock/vmci: use sk_acceptq_is_full() helper

[Raf Dickson]

Replace the open-coded backlog check with sk_acceptq_is_full().
The helper uses > instead of >=, which is the correct comparison
per commit 64a146513f8f ("[NET]: Revert incorrect accept queue
backlog changes."), and adds READ_ONCE() for proper memory ordering.

Suggested-by: Stefano Garzarella <sgarzare@redhat.com>

Signed-off-by: Raf Dickson <rafdog35@gmail.com>
---
 net/vmw_vsock/vmci_transport.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 91516488a7..56503bee31 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -1010,7 +1010,7 @@ static int vmci_transport_recv_listen(struct sock *sk,
 	 * reset.  Otherwise we create and initialize a child socket and reply
 	 * with a connection negotiation.
 	 */
-	if (sk->sk_ack_backlog >= sk->sk_max_ack_backlog) {
+	if (sk_acceptq_is_full(sk)) {
 		vmci_transport_reply_reset(pkt);
 		return -ECONNREFUSED;
 	}
-- 
2.54.0

Re: [PATCH net-next] vsock/vmci: use sk_acceptq_is_full() helper

[Luigi Leonardi]

On Thu, Jun 11, 2026 at 02:38:30AM +0000, Raf Dickson wrote:
>Replace the open-coded backlog check with sk_acceptq_is_full().
>The helper uses > instead of >=, which is the correct comparison
>per commit 64a146513f8f ("[NET]: Revert incorrect accept queue
>backlog changes."), and adds READ_ONCE() for proper memory ordering.
>
>Suggested-by: Stefano Garzarella <sgarzare@redhat.com>
>

this blank line should be dropped
>Signed-off-by: Raf Dickson <rafdog35@gmail.com>
>---
> net/vmw_vsock/vmci_transport.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
>diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
>index 91516488a7..56503bee31 100644
>--- a/net/vmw_vsock/vmci_transport.c
>+++ b/net/vmw_vsock/vmci_transport.c
>@@ -1010,7 +1010,7 @@ static int vmci_transport_recv_listen(struct sock *sk,
> 	 * reset.  Otherwise we create and initialize a child socket and reply
> 	 * with a connection negotiation.
> 	 */
>-	if (sk->sk_ack_backlog >= sk->sk_max_ack_backlog) {
>+	if (sk_acceptq_is_full(sk)) {
> 		vmci_transport_reply_reset(pkt);
> 		return -ECONNREFUSED;
> 	}
>-- 
>2.54.0
>

Thanks for the patch!

note: according to patchwork [1] you forgot to CC some maintainers,
please be more careful next time :)

Reviewed-by: Luigi Leonardi <leonardi@redhat.com>

[1] https://patchwork.kernel.org/project/netdevbpf/patch/20260611023830.106259-1-rafdog35@gmail.com/

Re: [PATCH net-next] vsock/vmci: use sk_acceptq_is_full() helper

[Raf Dickson]

On Thu, Jun 11, 2026 at 07:58AM, Luigi Leonardi wrote:
> this blank line should be dropped
>
> note: according to patchwork [1] you forgot to CC some maintainers,
> please be more careful next time :)
>
> Reviewed-by: Luigi Leonardi <leonardi@redhat.com>

Thanks for the review! Fixed the blank line and will add the missing
maintainers (horms, edumazet, kuba) in v2. Will send after the 24h
window. Also tried get_maintainer.pl but it kept refusing to recognize
the sparse checkout as a kernel tree -- lesson learned to check
patchwork next time.

Raf

Re: [PATCH net-next] vsock/vmci: use sk_acceptq_is_full() helper

[Stefano Garzarella]

On Thu, Jun 11, 2026 at 02:38:30AM +0000, Raf Dickson wrote:
>Replace the open-coded backlog check with sk_acceptq_is_full().
>The helper uses > instead of >=, which is the correct comparison
>per commit 64a146513f8f ("[NET]: Revert incorrect accept queue
>backlog changes."), and adds READ_ONCE() for proper memory ordering.
>
>Suggested-by: Stefano Garzarella <sgarzare@redhat.com>
>
>Signed-off-by: Raf Dickson <rafdog35@gmail.com>
>---
> net/vmw_vsock/vmci_transport.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>

Thanks for this, what about fixing also hyperv_transport ?

Stefano

>
>diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
>index 91516488a7..56503bee31 100644
>--- a/net/vmw_vsock/vmci_transport.c
>+++ b/net/vmw_vsock/vmci_transport.c
>@@ -1010,7 +1010,7 @@ static int vmci_transport_recv_listen(struct sock *sk,
> 	 * reset.  Otherwise we create and initialize a child socket and reply
> 	 * with a connection negotiation.
> 	 */
>-	if (sk->sk_ack_backlog >= sk->sk_max_ack_backlog) {
>+	if (sk_acceptq_is_full(sk)) {
> 		vmci_transport_reply_reset(pkt);
> 		return -ECONNREFUSED;
> 	}
>-- 
>2.54.0
>

Re: [PATCH net-next] vsock/vmci: use sk_acceptq_is_full() helper

[Raf Dickson]

On Thu, Jun 11, 2026 at 10:37:30AM +0200, Stefano Garzarella wrote:
> Thanks for this, what about fixing also hyperv_transport ?

nice spotting, hyperv_transport.c:326 has the same open-coded check.
Will include it in v2 of this patch.

Raf

Re: [PATCH net-next] vsock/vmci: use sk_acceptq_is_full() helper

[Jakub Kicinski]

On Thu, 11 Jun 2026 09:58:22 +0200 Luigi Leonardi wrote:
> note: according to patchwork [1] you forgot to CC some maintainers,
> please be more careful next time :)

Please focus your reviews on the code not patchwork checks.