[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20221215083045.2lfplx6fhuwpau2s@sgarzare-redhat>
Date: Thu, 15 Dec 2022 09:30:45 +0100
From: Stefano Garzarella <sgarzare@...hat.com>
To: Paolo Abeni <pabeni@...hat.com>
Cc: Bobby Eshleman <bobby.eshleman@...edance.com>,
Bobby Eshleman <bobbyeshleman@...il.com>,
Cong Wang <cong.wang@...edance.com>,
Jiang Wang <jiang.wang@...edance.com>,
Krasnov Arseniy <oxffffaa@...il.com>,
Stefan Hajnoczi <stefanha@...hat.com>,
"Michael S. Tsirkin" <mst@...hat.com>,
Jason Wang <jasowang@...hat.com>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>, kvm@...r.kernel.org,
virtualization@...ts.linux-foundation.org, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v7] virtio/vsock: replace virtio_vsock_pkt with
sk_buff
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
Powered by blists - more mailing lists