[ofa-general] Re: [ewg] APM Example

[Dotan Barak]

Roland Dreier wrote:
>  > > You don't know the time that the transition occurred, except that it
>  > > is between when you called modify QP and when it returned.  But an
>  > > asynchronous event doesn't really help, does it?
>
>  > It does help. APM is not only defined for network fault tolerance, it can
>  > also be used for load-balancing. With this event, one can know when
>  > the path is loaded and it is safe to call modify_qp.
>
> I guess I don't really understand how you're using this event.  What
> advantage is there in getting an async event at some unknown time
> (maybe before the modify QP operation returns, maybe after)?  What
> does it let you do that you can't do with the verbs architecture as
> defined strictly by the verbs?
>   

Roland is right, this event wasn't defined in the IB spec.

If you wish to know when it is safe to call the modify_qp verb you can 
call query_qp and check the
path_mig_state: it the state is ARMED, it means that it is safe to use 
the alternate path.


Dotan