netdev - Re: [PATCH vhost 21/22] virtio_net: update tx timeout record

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1697101953.6236846-1-xuanzhuo@linux.alibaba.com>
Date: Thu, 12 Oct 2023 17:12:33 +0800
From: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
To: "Michael S. Tsirkin" <mst@...hat.com>
Cc: virtualization@...ts.linux-foundation.org,
 "David S. Miller" <davem@...emloft.net>,
 Eric Dumazet <edumazet@...gle.com>,
 Jakub Kicinski <kuba@...nel.org>,
 Paolo Abeni <pabeni@...hat.com>,
 Jason Wang <jasowang@...hat.com>,
 Alexei Starovoitov <ast@...nel.org>,
 Daniel Borkmann <daniel@...earbox.net>,
 Jesper Dangaard Brouer <hawk@...nel.org>,
 John Fastabend <john.fastabend@...il.com>,
 netdev@...r.kernel.org,
 bpf@...r.kernel.org
Subject: Re: [PATCH vhost 21/22] virtio_net: update tx timeout record

On Thu, 12 Oct 2023 05:10:55 -0400, "Michael S. Tsirkin" <mst@...hat.com> wrote:
> On Wed, Oct 11, 2023 at 05:27:27PM +0800, Xuan Zhuo wrote:
> > If send queue sent some packets, we update the tx timeout
> > record to prevent the tx timeout.
> >
> > Signed-off-by: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
> > ---
> >  drivers/net/virtio/xsk.c | 10 ++++++++++
> >  1 file changed, 10 insertions(+)
> >
> > diff --git a/drivers/net/virtio/xsk.c b/drivers/net/virtio/xsk.c
> > index 7abd46bb0e3d..e605f860edb6 100644
> > --- a/drivers/net/virtio/xsk.c
> > +++ b/drivers/net/virtio/xsk.c
> > @@ -274,6 +274,16 @@ bool virtnet_xsk_xmit(struct virtnet_sq *sq, struct xsk_buff_pool *pool,
> >
> >  	virtnet_xsk_check_queue(sq);
> >
> > +	if (stats.packets) {
> > +		struct netdev_queue *txq;
> > +		struct virtnet_info *vi;
> > +
> > +		vi = sq->vq->vdev->priv;
> > +
> > +		txq = netdev_get_tx_queue(vi->dev, sq - vi->sq);
> > +		txq_trans_cond_update(txq);
> > +	}
> > +
> >  	u64_stats_update_begin(&sq->stats.syncp);
> >  	sq->stats.packets += stats.packets;
> >  	sq->stats.bytes += stats.bytes;
>
> I don't get what this is doing. Is there some kind of race here you
> are trying to address? And what introduced the race?

Because the xsk xmit shares the send queue with the kernel xmit,
then when I do benchmark, the xsk will always use the send queue,
so the kernel may have no chance to do xmit, the tx watchdog
thinks that the send queue is hang and prints tx timeout log.

So I call the txq_trans_cond_update() to tell the tx watchdog
that the send queue is working.

Thanks.

>
> > --
> > 2.32.0.3.g01195cf9f
>
>