Re: [PATCH v11 2/2] mailbox: introduce ARM SMC based mailbox

[Sudeep Holla]

On Wed, Jan 29, 2020 at 04:01:07PM +0100, Etienne Carriere wrote:
> Hello Peng,
>
> On Mon, 27 Jan 2020 at 13:58, Peng Fan <peng.fan@xxxxxxx> wrote:
> >
> > > Subject: Re: [PATCH v11 2/2] mailbox: introduce ARM SMC based mailbox
> > >
> > > Hello Peng and all,
> > >
> > >
> > > > From: Peng Fan <peng.fan@xxxxxxx>
> > > >
> > > > This mailbox driver implements a mailbox which signals transmitted
> > > > data via an ARM smc (secure monitor call) instruction. The mailbox
> > > > receiver is implemented in firmware and can synchronously return data
> > > > when it returns execution to the non-secure world again.
> > > > An asynchronous receive path is not implemented.
> > > > This allows the usage of a mailbox to trigger firmware actions on SoCs
> > > > which either don't have a separate management processor or on which
> > > > such a core is not available. A user of this mailbox could be the SCP
> > > > interface.
> > > >
> > > > Modified from Andre Przywara's v2 patch
> > > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore
> > > > .kernel.org%2Fpatchwork%2Fpatch%2F812999%2F&amp;data=02%7C01%7
> > > Cpeng.fa
> > > >
> > > n%40nxp.com%7C735cc6cd00404082bf8c08d79f67b93a%7C686ea1d3bc2b4
> > > c6fa92cd
> > > >
> > > 99c5c301635%7C0%7C0%7C637153140140878278&amp;sdata=m0lcAEIr0ZP
> > > tyPHorSW
> > > > NYgjfI5p0genJLlhqHMIHBg0%3D&amp;reserved=0
> > > >
> > > > Reviewed-by: Florian Fainelli <f.fainelli@xxxxxxxxx>
> > > > Reviewed-by: Andre Przywara <andre.przywara@xxxxxxx>
> > > > Signed-off-by: Peng Fan <peng.fan@xxxxxxx>
> > >
> > > I've successfully tested your change on my board. It is a stm32mp1 with TZ
> > > secure hardening and I run an OP-TEE firmware (possibly a TF-A
> > > sp_min) with a SCMI server for clock and reset. Upstream in progress.
> > > The platform uses 2 instances of your SMC based mailbox device driver
> > > (2 mailboxes). Works nice with your change.
> > >
> > > You can add my T-b tag: Tested-by: Etienne Carriere
> > > <etienne.carriere@xxxxxxxxxx>
> >
> > Thanks, but this patch has been dropped.
> >
> > Per Sudeep, we all use smc transport, not smc mailbox ,
> > I'll post patch in a few days based on the transport split patch.
>
> Ok, i am syncing.
>
> > >
> > > FYI, I'll (hopefully soon) post a change proposal in U-Boot ML for an equvalent
> > > 'SMC based mailbox' driver and SCMI agent protocol/device drivers for clock
> > > and reset controllers.
> >
> > Great to know you did scmi agent code in U-Boot. Do you have some public repo
> > for access?
>
> I've created a P-R on my github repo to share until I submit to u-boot:
>  https://github.com/etienne-lms/u-boot/pull/3
>
> I guess I will change my u-boot proposal and get a SMC SCMI transport
> outside of the mailbox framework.
>

Unless U-boot has mailbox framework or you are importing it, it's better
to keep U-boot implementation simple as SMC transport which I think you
already do. I had a look at the implementation[1], it shouldn't change
much other than if you prefer not to use "mailbox" terminology. I don't
understand the reason for even using the mailbox term there in the first
place.

--
Regards,
Sudeep

[1] https://github.com/etienne-lms/u-boot/pull/3/commits/34812c9175436f6a082f77347c5384393757c233