[okl4-developer] Asynchronous message

[Jean-Christophe Dubois]

Hi Nelson,

I am not sure the question was correctly understood. 

In IDL syntax if you precede an operation definition with the 
keyword "oneway", a client that calls the operation will not block while the 
object processes the call. This makes things really asynchronous ...

I think the point Charles is making is that the "oneway" keyword seems to be 
ignored by the OKL4 IDL compiler and there is no difference in the generated 
code between a normal operation and a "oneway" operation. In both case 
the "client" will wait for the "server" to process the operation to completion 
even if the return value is "void" (which is required for "oneway" 
operation).

So the question is to know if the "oneway" IDL keyword receives today a 
different treatement in OKL4 and/or if this will be the case in a future 
release? Or maybe this is not considered at all for any future OKL4 
release ...

Regards

JC

Le Monday 04 August 2008 01:53:16 Nelson Tam, vous avez écrit :
> Hi Charles,
>
> There are certainly ways to set up threads in OKL4 2.1.1 such that IPC
> calls will only succeed in one direction, and calls to the other
> direction will either fail or result in a nop (this applies to both
> synchronous and asynchronous IPC).  In (near) future versions of OKL4,
> it will be possible to prevent IPC calls from being made in the first
> place, on a directional basis.
>
> I'm afraid I'm not sure what amount of support IDL4 provides for
> ensuring one-way IPC only.  But as always, writing the interface code
> manually will give you finest control over how the interface is
> implemented.
>
> Hope this helps,
> Nelson.
>
> On 01/08/2008, at 7:43 PM, Charles.VANWITTENBERGHE at fr.thalesgroup.com
>
> wrote:
> > This time I have a simple question : Is the keyword "Oneway", for
> > asynchronous messages, currently supported by IDL4 & OKL4 v.2.1.1-
> > patch 9 ?
> >
> > It seems to me that there is no differences between communication
> > interfaces
> > defined with or without oneway. And I as far as I can see, there
> > aren't any
> > differences in the generated code.
>
> --
> (nt)
>
> Nelson Tam
>
>
> _______________________________________________
> Developer mailing list
> Developer at okl4.org
> https://lists.okl4.org/mailman/listinfo/developer