Re: [RFC PATCH v8 11/19] virtio/vsock: dequeue callback for SOCK_SEQPACKET

[Arseny Krasnov]

On 21.04.2021 11:56, Stefano Garzarella wrote:
> On Tue, Apr 13, 2021 at 03:44:40PM +0300, Arseny Krasnov wrote:
>> This adds transport callback and it's logic for SEQPACKET dequeue.
>> Callback fetches RW packets from rx queue of socket until whole record
>> is copied(if user's buffer is full, user is not woken up). This is done
>> to not stall sender, because if we wake up user and it leaves syscall,
>> nobody will send credit update for rest of record, and sender will wait
>> for next enter of read syscall at receiver's side. So if user buffer is
>> full, we just send credit update and drop data.
>>
>> Signed-off-by: Arseny Krasnov <arseny.krasnov@xxxxxxxxxxxxx>
>> ---
>> v7 -> v8:
>> - Things like SEQ_BEGIN, SEQ_END, 'msg_len' and 'msg_id' now removed.
>>   This callback fetches and copies RW packets to user's buffer, until
>>   last packet of message found(this packet is marked in 'flags' field
>>   of header).
>>
>> include/linux/virtio_vsock.h            |  5 ++
>> net/vmw_vsock/virtio_transport_common.c | 73 +++++++++++++++++++++++++
>> 2 files changed, 78 insertions(+)
>>
>> diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
>> index dc636b727179..02acf6e9ae04 100644
>> --- a/include/linux/virtio_vsock.h
>> +++ b/include/linux/virtio_vsock.h
>> @@ -80,6 +80,11 @@ virtio_transport_dgram_dequeue(struct vsock_sock *vsk,
>>                              struct msghdr *msg,
>>                              size_t len, int flags);
>>
>> +ssize_t
>> +virtio_transport_seqpacket_dequeue(struct vsock_sock *vsk,
>> +                                 struct msghdr *msg,
>> +                                 int flags,
>> +                                 bool *msg_ready);
>> s64 virtio_transport_stream_has_data(struct vsock_sock *vsk);
>> s64 virtio_transport_stream_has_space(struct vsock_sock *vsk);
>>
>> diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
>> index 833104b71a1c..8492b8bd5df5 100644
>> --- a/net/vmw_vsock/virtio_transport_common.c
>> +++ b/net/vmw_vsock/virtio_transport_common.c
>> @@ -393,6 +393,67 @@ virtio_transport_stream_do_dequeue(struct vsock_sock *vsk,
>>       return err;
>> }
>>
>> +static int virtio_transport_seqpacket_do_dequeue(struct vsock_sock *vsk,
>> +                                               struct msghdr *msg,
>> +                                               int flags,
>> +                                               bool *msg_ready)
>> +{
>> +      struct virtio_vsock_sock *vvs = vsk->trans;
>> +      struct virtio_vsock_pkt *pkt;
>> +      int err = 0;
>> +      size_t user_buf_len = msg->msg_iter.count;
>> +
>> +      *msg_ready = false;
>> +      spin_lock_bh(&vvs->rx_lock);
>> +
>> +      while (!*msg_ready && !list_empty(&vvs->rx_queue) && err >= 0) {
>> +              pkt = list_first_entry(&vvs->rx_queue, struct virtio_vsock_pkt, list);
>> +
>> +              if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RW) {
> Is this check still necessary, should they all be RW?
Ack
>
>> +                      size_t bytes_to_copy;
>> +                      size_t pkt_len;
>> +
>> +                      pkt_len = (size_t)le32_to_cpu(pkt->hdr.len);
>> +                      bytes_to_copy = min(user_buf_len, pkt_len);
>> +
> If bytes_to_copy == 0, we can avoid the next steps (release the lock try 
> to copy 0 bytes, reacquire the lock)
Ack
>
>> +                      /* sk_lock is held by caller so no one else can dequeue.
>> +                       * Unlock rx_lock since memcpy_to_msg() may sleep.
>> +                       */
>> +                      spin_unlock_bh(&vvs->rx_lock);
>> +
>> +                      if (memcpy_to_msg(msg, pkt->buf, bytes_to_copy)) {
>> +                              err = -EINVAL;
> Here we should reacquire the lock or prevent it from being released out
> of cycle.
Ack
>
>> +                              break;
>> +                      }
>> +
>> +                      spin_lock_bh(&vvs->rx_lock);
>> +
> As mentioned before, I think we could move this part into the core and 
> here always return the real dimension.
Ack
>
>> +                      /* If user sets 'MSG_TRUNC' we return real 
>> length
>> +                       * of message.
>> +                       */
>> +                      if (flags & MSG_TRUNC)
>> +                              err += pkt_len;
>> +                      else
>> +                              err += bytes_to_copy;
>> +
>> +                      user_buf_len -= bytes_to_copy;
>> +
>> +                      if (pkt->hdr.flags & VIRTIO_VSOCK_SEQ_EOR)
>                                      ^
> We should use le32_to_cpu() to read the flags.
>
Ack
>> +                              *msg_ready = true;
>> +              }
>> +
>> +              virtio_transport_dec_rx_pkt(vvs, pkt);
>> +              list_del(&pkt->list);
>> +              virtio_transport_free_pkt(pkt);
>> +      }
>> +
>> +      spin_unlock_bh(&vvs->rx_lock);
>> +
>> +      virtio_transport_send_credit_update(vsk);
>> +
>> +      return err;
>> +}
>> +
>> ssize_t
>> virtio_transport_stream_dequeue(struct vsock_sock *vsk,
>>                               struct msghdr *msg,
>> @@ -405,6 +466,18 @@ virtio_transport_stream_dequeue(struct vsock_sock *vsk,
>> }
>> EXPORT_SYMBOL_GPL(virtio_transport_stream_dequeue);
>>
>> +ssize_t
>> +virtio_transport_seqpacket_dequeue(struct vsock_sock *vsk,
>> +                                 struct msghdr *msg,
>> +                                 int flags, bool *msg_ready)
>> +{
>> +      if (flags & MSG_PEEK)
>> +              return -EOPNOTSUPP;
>> +
>> +      return virtio_transport_seqpacket_do_dequeue(vsk, msg, flags,
>> msg_ready);
>> +}
>> +EXPORT_SYMBOL_GPL(virtio_transport_seqpacket_dequeue);
>> +
>> int
>> virtio_transport_dgram_dequeue(struct vsock_sock *vsk,
>>                              struct msghdr *msg,
>> --
>> 2.25.1
>>
>