Re: [PATCH net 1/2] vsock/virtio: collapse receive queue under memory pressure
From: Bobby Eshleman <hidden>
Date: 2026-07-02 20:09:25
Also in:
kvm, lkml, stable, virtualization
On Thu, Jul 02, 2026 at 10:56:04AM +0200, Stefano Garzarella wrote:
On Wed, Jul 01, 2026 at 09:34:35AM -0700, Bobby Eshleman wrote:quoted
On Fri, Jun 26, 2026 at 03:48:22PM +0200, Stefano Garzarella wrote:[...]quoted
quoted
+out: + if (new_skb) + __skb_queue_tail(&new_queue, new_skb); + + skb_queue_splice(&new_queue, &vvs->rx_queue);I think the new skbs will also need skb_set_owner_sk_safe(skb, sk) when adding to rx_queue?IIRC we added it in the rx path, mainily for loopback to pass the ownership from the tx socket to the rx socket, but here we are already in the rx path, so the skb will never leave this socket.
Ah that's right, I stand corrected. There is no sender to leak in this case.
Maybe it's necessary for the eBPF path?
Looking through sockmap, I don't think it depends on skb->sk being non-null either (it reassigns owner to the redirect socket anyway using skb_set_owner_r()). Sorry for the false alarm. LGTM. Reviewed-by: Bobby Eshleman <redacted>