[<prev] [next>] [day] [month] [year] [list]
Message-ID: <CACGkMEs0V7Hy2mkQymCyVBYAaM7tM=Wj7d+tfxTOg8zJdr4YDA@mail.gmail.com>
Date: Thu, 28 Oct 2021 10:28:45 +0800
From: Jason Wang <jasowang@...hat.com>
To: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
Cc: Jakub Kicinski <kuba@...nel.org>,
virtualization <virtualization@...ts.linux-foundation.org>,
netdev <netdev@...r.kernel.org>,
"Michael S. Tsirkin" <mst@...hat.com>,
"David S. Miller" <davem@...emloft.net>
Subject: Re: [PATCH 3/3] virtio-net: enable virtio indirect cache
On Thu, Oct 28, 2021 at 9:59 AM Xuan Zhuo <xuanzhuo@...ux.alibaba.com> wrote:
>
> On Wed, 27 Oct 2021 08:55:28 -0700, Jakub Kicinski <kuba@...nel.org> wrote:
> > On Wed, 27 Oct 2021 14:19:13 +0800 Xuan Zhuo wrote:
> > > +static bool virtio_desc_cache = true;
> > > module_param(csum, bool, 0444);
> > > module_param(gso, bool, 0444);
> > > module_param(napi_tx, bool, 0644);
> > > +module_param(virtio_desc_cache, bool, 0644);
> >
> > Can this be an ethtool priv flag? module params are discouraged because
> > they can't be controlled per-netdev.
>
>
> The current design can only be set when the device is initialized. So using
> ethtool to modify it will not work.
Anyhow you can add things like synchronization to make it work. But I
think what we want is to make it work unconditionally, so having a
module parameter seems useless. If you want to use it for
benchmarking?
Thanks
>
> Thanks.
>
Powered by blists - more mailing lists