[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAA8EJprrcH3T8_aA7bZhZXKiWMXjUTZkvLkQzteHUG4_7e4i8w@mail.gmail.com>
Date: Tue, 5 Dec 2023 16:22:45 +0200
From: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>,
davem@...emloft.net, edumazet@...gle.com, pabeni@...hat.com,
mhi@...ts.linux.dev, linux-arm-msm@...r.kernel.org,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
loic.poulain@...aro.org
Subject: Re: [PATCH v2 0/2] Add MHI Endpoint network driver
On Mon, 4 Dec 2023 at 18:12, Jakub Kicinski <kuba@...nel.org> wrote:
>
> On Mon, 4 Dec 2023 14:12:12 +0200 Dmitry Baryshkov wrote:
> > Ok, here you are talking about the control path. I can then assume
> > that you consider it to be fine to use netdev for the EP data path, if
> > the control path is kept separate and those two can not be mixed. Does
> > that sound correct?
>
> If datapath == traffic which is intended to leave the card via
> the external port, then yes.
Then I think I understand what causes the confusion.
The MHI netdev is used to deliver network traffic to the modem CPU,
but it is not the controlpath.
For the control path we have non-IP MHI channels (QMI, IPCR, etc).
This can be the traffic targeting e.g. SSH or HTTP server running on
the EP side of the link.
I probably fail to see the difference between this scenario and the
proper virtio netdev which also allows us to send the same IPv4/v6
traffic to the CPU on the EP side.
--
With best wishes
Dmitry
Powered by blists - more mailing lists