On Wed, Dec 14, 2022 at 11:58:47AM +0100, Paolo Abeni wrote:
On Tue, 2022-12-13 at 19:28 +0000, Bobby Eshleman wrote:
diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
index 5703775af129..2a5994b029b2 100644
--- a/drivers/vhost/vsock.c
+++ b/drivers/vhost/vsock.c
@@ -51,8 +51,7 @@ struct vhost_vsock {
        struct hlist_node hash;

        struct vhost_work send_pkt_work;
-       spinlock_t send_pkt_list_lock;
-       struct list_head send_pkt_list; /* host->guest pending packets */
+       struct sk_buff_head send_pkt_queue; /* host->guest pending packets */

        atomic_t queued_replies;

@@ -108,40 +107,33 @@ vhost_transport_do_send_pkt(struct vhost_vsock *vsock,
        vhost_disable_notify(&vsock->dev, vq);

        do {
-               struct virtio_vsock_pkt *pkt;
+               struct virtio_vsock_hdr *hdr;
+               size_t iov_len, payload_len;
                struct iov_iter iov_iter;
+               u32 flags_to_restore = 0;
+               struct sk_buff *skb;
                unsigned out, in;
                size_t nbytes;
-               size_t iov_len, payload_len;
                int head;
-               u32 flags_to_restore = 0;

-               spin_lock_bh(&vsock->send_pkt_list_lock);
-               if (list_empty(&vsock->send_pkt_list)) {
-                       spin_unlock_bh(&vsock->send_pkt_list_lock);
+               spin_lock(&vsock->send_pkt_queue.lock);
+               skb = __skb_dequeue(&vsock->send_pkt_queue);
+               spin_unlock(&vsock->send_pkt_queue.lock);

Here you use a plain spin_lock(), but every other lock has the _bh()
variant. A few lines above this functions acquires a mutex, so this is
process context (and not BH context): I guess you should use _bh()
here, too.

Maybe we can use directly the new virtio_vsock_skb_dequeue().
IIRC we added it exactly to use the same kind of lock everywhere.

Thanks,
Stefano

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

Reply via email to