> -----Original Message-----
> From: Willem de Bruijn [mailto:willemdebruijn.ker...@gmail.com]
> Sent: Monday, December 28, 2020 12:15 AM
> To: Michael S. Tsirkin <m...@redhat.com>
> Cc: wangyunjian <wangyunj...@huawei.com>; Network Development
> <netdev@vger.kernel.org>; Jason Wang <jasow...@redhat.com>; Willem de
> Bruijn <willemdebruijn.ker...@gmail.com>;
> virtualizat...@lists.linux-foundation.org; Lilijun (Jerry)
> <jerry.lili...@huawei.com>; chenchanghu <chenchan...@huawei.com>;
> xudingke <xudin...@huawei.com>; huangbin (J)
> <brian.huang...@huawei.com>
> Subject: Re: [PATCH net v5 1/2] vhost_net: fix ubuf refcount incorrectly when
> sendmsg fails
>
> On Sun, Dec 27, 2020 at 6:26 AM Michael S. Tsirkin <m...@redhat.com> wrote:
> >
> > On Fri, Dec 25, 2020 at 03:24:25PM +0800, wangyunjian wrote:
> > > From: Yunjian Wang <wangyunj...@huawei.com>
> > >
> > > Currently the vhost_zerocopy_callback() maybe be called to decrease
> > > the refcount when sendmsg fails in tun. The error handling in vhost
> > > handle_tx_zerocopy() will try to decrease the same refcount again.
> > > This is wrong. To fix this issue, we only call vhost_net_ubuf_put()
> > > when vq->heads[nvq->desc].len == VHOST_DMA_IN_PROGRESS.
> > >
> > > Fixes: 0690899b4d45 ("tun: experimental zero copy tx support")
> >
> > Are you sure about this tag? the patch in question only affects tun,
> > while the fix only affects vhost.
>
> That was my suggestion. But you're right. Perhaps better:
>
> Fixes: bab632d69ee4 ("vhost: vhost TX zero-copy support")
OK, thanks, I will fix it in next version.
>
> That introduces the actual block that releases the buffer on error:
>
> "
> err = sock->ops->sendmsg(NULL, sock, &msg, len);
> if (unlikely(err < 0)) {
> + if (zcopy) {
> + if (ubufs)
> + vhost_ubuf_put(ubufs);
> + vq->upend_idx =
> ((unsigned)vq->upend_idx - 1) %
> + UIO_MAXIOV;
> + }
> "
