[PATCH 06/11] firmware: arm_scmi: add support for protocol modularization

[Cristian Marussi]

On Tue, Oct 20, 2020 at 10:47:28PM -0400, Thara Gopinath wrote:
> 
> 
> On 10/14/20 11:05 AM, Cristian Marussi wrote:
> > Modify protocol initialization callback adding a new parameter representing
> > a reference to the available xfer core operations and introduce a macro to
> > simply register with the core new protocols as loadable drivers.
> > Keep standard protocols as builtin.
> > 
> > Signed-off-by: Cristian Marussi <cristian.marussi at arm.com>
> > ---
> >   drivers/firmware/arm_scmi/base.c    | 56 ++++++++++--------
> >   drivers/firmware/arm_scmi/bus.c     | 14 ++++-
> >   drivers/firmware/arm_scmi/clock.c   | 56 +++++++++---------
> >   drivers/firmware/arm_scmi/common.h  | 42 +++++++++-----
> >   drivers/firmware/arm_scmi/driver.c  | 50 ++++++++++------
> >   drivers/firmware/arm_scmi/perf.c    | 88 +++++++++++++++--------------
> >   drivers/firmware/arm_scmi/power.c   | 46 ++++++++-------
> >   drivers/firmware/arm_scmi/reset.c   | 46 ++++++++-------
> >   drivers/firmware/arm_scmi/sensors.c | 52 +++++++++--------
> >   drivers/firmware/arm_scmi/system.c  | 16 ++++--
> >   include/linux/scmi_protocol.h       | 18 +++++-
> >   11 files changed, 288 insertions(+), 196 deletions(-)
> > 
> > diff --git a/drivers/firmware/arm_scmi/base.c b/drivers/firmware/arm_scmi/base.c
> > index f40821eeb103..8d7214fd2187 100644
> > --- a/drivers/firmware/arm_scmi/base.c
> > +++ b/drivers/firmware/arm_scmi/base.c
> > @@ -15,6 +15,8 @@
> >   #define SCMI_BASE_NUM_SOURCES		1
> >   #define SCMI_BASE_MAX_CMD_ERR_COUNT	1024
> > +static const struct scmi_xfer_ops *ops;
> 
> Minor nit. I would consider renaming ops to something more
> meaningful like xfer_ops (or anything that makes sense). ops by
> itself leads to confusion with ops in scmo_protocol and  in
> scmi_protocol_events.
> 
> Same suggestion for all other declarations of ops in this patch.
> 
I agree, but I'm in fact reworking deeply the interface for protocols
initialization (following some idea from Florian) in order to pass down
a protocol object including the xfer_ops instead of a handle+ops like
now, so this hopefully will solve also this concern.

Thanks

Cristian
> -- 
> Warm Regards
> Thara