[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20210107165659.512fe71c@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
Date: Thu, 7 Jan 2021 16:56:59 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: wangyunjian <wangyunjian@...wei.com>
Cc: <netdev@...r.kernel.org>, <mst@...hat.com>,
<willemdebruijn.kernel@...il.com>, <jasowang@...hat.com>,
<virtualization@...ts.linux-foundation.org>,
<jerry.lilijun@...wei.com>, <xudingke@...wei.com>,
<chenchanghu@...wei.com>, <brian.huangbin@...wei.com>
Subject: Re: [PATCH v6] vhost_net: avoid tx queue stuck when sendmsg fails
On Thu, 7 Jan 2021 17:49:45 +0800 wangyunjian wrote:
> From: Yunjian Wang <wangyunjian@...wei.com>
>
> Currently the driver doesn't drop a packet which can't be sent by tun
> (e.g bad packet). In this case, the driver will always process the
> same packet lead to the tx queue stuck.
>
> To fix this issue:
> 1. in the case of persistent failure (e.g bad packet), the driver
> can skip this descriptor by ignoring the error.
> 2. in the case of transient failure (e.g -ENOBUFS, -EAGAIN and -ENOMEM),
> the driver schedules the worker to try again.
>
> Signed-off-by: Yunjian Wang <wangyunjian@...wei.com>
Hi! AFAIU this needs to go to net-next. You need to wait until net is
merged into net-next and then repost. It doesn't apply at the moment
and we get in excess of 100 patches a day right now so keeping track
of dependencies on the maintainer side does not work. Also build bot
can't test it.
net should get merged into net-next some time tomorrow or on Saturday.
Powered by blists - more mailing lists