[openib-general] Re: [PATCH] CMA: allow/require bind beforeconnect

[Caitlin Bestler]

Tom Tucker wrote:
> On Tue, 2006-03-28 at 09:50 -0800, Sean Hefty wrote:
>>>> The app _usually_ doesn't care.  See NFS discussion for a client
>>>> app that does care.  Also, providers DO care.  Because of this
>>>> issue, the chelsio iwarp provider right now has to allocate its own
>>>> ephemeral ports at connect time.  This logic should be moved into
>>>> the IWCM or maybe the CMA and an explicit bind() operation be
>>>> exported by the iwarp providers to allow the IWCM or CMA to track
>>>> all port allocations.
>> 
>> Somehow I didn't receive Steve's response.  Anyway, I'm referring to
>> apps that don't specify which port to use as the one's that don't
>> care, versus those that do want to know their port.
>> 
>> In reality, the port space for RDMA connections over IB is distinct
>> from the TCP port space.  No attempt is made to coordinate between
>> the two, and I am not convinced that those two port spaces should be
>> one and the same.  This is where I believe there's disconnect.
> 
> Whether it's a separate port space or not is as simple as
> defining a different hash_info structure.
> 

Tracking the port allocations in one or two data structures isn't
the real issue. The real issue is whether there is a way to describe
all connections created through the CMA in a simple manner that can
be understood by developers used to IP semantics. For example, how
do I compare the list of CMA approved connections vs. the netfilter
rules for establishing connections? Am I going to expect netfilter
code to undestand that a "connection" can have "Not applicable" as
the local port number?