[ofa-general] IB verbs client/server programming...

[Dotan Barak]

Peter W. Morreale wrote:
> On Tue, 2008-07-15 at 17:07 +0300, Dotan Barak wrote:
>   
>> On Tue, Jul 15, 2008 at 4:07 PM, Peter W. Morreale <pmorreale at novell.com> wrote:
>>     
>>> On Tue, 2008-07-15 at 09:39 +0300, Dotan Barak wrote:
>>>
>>>       
>>>>> In other words, what interfaces do I have to look at to implement the
>>>>> "blocking" portion equivalent of a recv(2)?
>>>>>           
>>>> Do you want to block until the data will be received?
>>>>         
>>> Yes.  Right now, all the examples I see involve a tight polling loop
>>> waiting for, and acknowledging the events.
>>>
>>>       
>> You need to poll for completion on the receive queue
>> (IF you use opcode that will consume WR in the recieve queue at the receiver)
>>
>>     
>
> I'm sorry, I don't understand this.  What do you mean by "If you use
> opcode that will consume the work request in the receive queue..."  
>
> What opcodes are you referring to?  
>
>   
The opcodes: Send, Send with immediate, RDMA Write with immediate 
consumes WR from the
receive QP of the receiver, the other opcodes don't
> Are you saying that the receiver does not have to acknowledge the
> receipt of the data transfer?  (That would be ideal for the data
> transfers.)  If so, how do I set that up?   
>   
This is what RDMA is all about: the ability to write/read data directly 
to/from the remote process
memory.

It is out of the scope of Infiniband (but it is in the scope of the 
application) on how to inform the other
side that the operations was executed and its status.

> The examples I have seen have the receiver ack'ing events.   Can these
> be deferred (or ignored even?)
I don't really understand what you mean by saying this sentence, can you 
please try to explain
it to me (and write the code lines that does this operation).

Maybe you are referring to the code that read completions using events...
>   
>
> Thanks,
> -PWM
>   

Dotan