[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200130102104.GA48466@bogus>
Date: Thu, 30 Jan 2020 10:21:04 +0000
From: Sudeep Holla <sudeep.holla@....com>
To: Viresh Kumar <viresh.kumar@...aro.org>
Cc: Etienne Carriere <etienne.carriere@...aro.org>,
linux-kernel@...r.kernel.org, Peng Fan <peng.fan@....com>,
Sudeep Holla <sudeep.holla@....com>
Subject: Re: [PATCH V5] firmware: arm_scmi: Make scmi core independent of the
transport type
On Thu, Jan 30, 2020 at 03:11:03PM +0530, Viresh Kumar wrote:
> On 30-01-20, 10:25, Etienne Carriere wrote:
> > I've made a first port (draft) for adding new transport channels, next
> > to existing mailbox channel, on top of your change.
> > You can find it here: https://github.com/etienne-lms/linux/pull/1.
> >
> > I don't have specific comments on your change but the one below.
> > I think SMT header should move out of mailbox.c, but that may be a bit
> > out of the scope of your change.
>
> If it is guaranteed that someone will end up using those routines
> apart from mailbox.c, then surely it can be done.
>
I thought about it and decided to take up when we add new transport
instead of doing now and having to redo again when we add new transport
mostly SMC/HVC.
> > I would prefer an optional mak_txdone callback:
> >
> > if (info->desc->ops->mark_txdone)
> > info->desc->ops->mark_txdone(cinfo, ret);
>
> So you are sure that mark_txdone won't be required in your case? I can
> make it optional then.
>
Yes this can be done. Might even help virtio and keeps Peter happy :)
--
Regards,
Sudeep
Powered by blists - more mailing lists