Re: [PULL 06/13] nbd: remove peppering of nbd_client_connected

[Eric Blake]

On Thu, May 12, 2022 at 05:16:04PM +0100, Peter Maydell wrote:
> On Tue, 26 Apr 2022 at 21:21, Eric Blake <eblake@redhat.com> wrote:
> >
> > From: Paolo Bonzini <pbonzini@redhat.com>
> >
> > It is unnecessary to check nbd_client_connected() because every time
> > s->state is moved out of NBD_CLIENT_CONNECTED the socket is shut down
> > and all coroutines are resumed.
> 
> Hi; Coverity points out (CID 1488362) that this part of this change
> has resulted in some dead code:
> 
> > @@ -512,7 +508,7 @@ static int coroutine_fn 
> > nbd_co_send_request(BlockDriverState *bs,
> >      if (qiov) {
> >          qio_channel_set_cork(s->ioc, true);
> >          rc = nbd_send_request(s->ioc, request);
> > -        if (nbd_client_connected(s) && rc >= 0) {
> > +        if (rc >= 0) {
> >              if (qio_channel_writev_all(s->ioc, qiov->iov, qiov->niov,
> >                                         NULL) < 0) {
> >                  rc = -EIO;
> 
> because the change means this code is now
> 
>         if (rc >= 0) {
>             if (qio_channel_writev_all(s->ioc, qiov->iov, qiov->niov,
>                                        NULL) < 0) {
>                 rc = -EIO;
>             }
>         } else if (rc >= 0) {
>             rc = -EIO;
>         }
> 
> and the "else if" clause is dead and can be deleted.

Thanks for reporting it.  I can prepare a patch.

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org