[openib-general] Re: Re: [CMA] re-connect procedure?

[Michael S. Tsirkin]

Quoting r. Sean Hefty <mshefty at ichips.intel.com>:
> Subject: Re: Re: [CMA] re-connect procedure?
> 
> Sean Hefty wrote:
> >I think that a better fix would be to transition to the correct state 
> >after disconnect occurred.  I.e. back to CMA_ROUTE_RESOLVED.  This kind 
> >of gets back to what is the correct state flow.  Should a user be able 
> >to walk backwards through the state diagram, or should the state return 
> >to IDLE?
> 
> Giving this more thought, I think that the behavior of least surprise would 
> be for the rdma_cm_id to return to IDLE after being disconnected, rather 
> than returning to ROUTE_RESOLVED.
> 
> - Sean

As was discussed here previously, there's still the need to retry after REJ
indicating MTU problem (tavor work-around).
In that case we'll be going back to ROUTE_RESOLVED, right?

-- 
Michael S. Tsirkin
Staff Engineer, Mellanox Technologies