The 02/17/2021 10:34, Vladimir Oltean wrote:
Hi Vladimir,
>
> On Tue, Feb 16, 2021 at 10:41:59PM +0100, Horatiu Vultur wrote:
> > Add the member sw_backup to the structures switchdev_obj_ring_role_mrp
> > and switchdev_obj_in_role_mrp. In this way the SW can call the driver in
> > 2 ways, once when sw_backup is set to false, meaning that the driver
> > should implement this completely in HW. And if that is not supported the
> > SW will call again but with sw_backup set to true, meaning that the
> > HW should help or allow the SW to run the protocol.
> >
> > For example when role is MRM, if the HW can't detect when it stops
> > receiving MRP Test frames but it can trap these frames to CPU, then it
> > needs to return -EOPNOTSUPP when sw_backup is false and return 0 when
> > sw_backup is true.
> >
> > Signed-off-by: Horatiu Vultur <horatiu.vultur@xxxxxxxxxxxxx>
> > ---
> > include/net/switchdev.h | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/include/net/switchdev.h b/include/net/switchdev.h
> > index 465362d9d063..b7fc7d0f54e2 100644
> > --- a/include/net/switchdev.h
> > +++ b/include/net/switchdev.h
> > @@ -127,6 +127,7 @@ struct switchdev_obj_ring_role_mrp {
> > struct switchdev_obj obj;
> > u8 ring_role;
> > u32 ring_id;
> > + u8 sw_backup;
> > };
> >
> > #define SWITCHDEV_OBJ_RING_ROLE_MRP(OBJ) \
> > @@ -161,6 +162,7 @@ struct switchdev_obj_in_role_mrp {
> > u32 ring_id;
> > u16 in_id;
> > u8 in_role;
> > + u8 sw_backup;
>
> What was wrong with 'bool'?
Yeah, that would be a better match.
>
> > };
> >
> > #define SWITCHDEV_OBJ_IN_ROLE_MRP(OBJ) \
> > --
> > 2.27.0
> >
>
> If a driver implements full MRP offload for a ring/interconnect
> manager/automanager, should it return -EOPNOTSUPP when sw_backup=false?
In that case it should return 0.
So if the driver can:
- fully support MRP, when sw_backup = false, return 0. Then end of story.
- partially support MRP, when sw_backup = false, return -EOPNOTSUPP,
when sw_backup = true, return 0.
- no support at all, return -EOPNOTSUPP.
--
/Horatiu
