Re: [SAM] New Version draft common multicast API

[Gustavo Carneiro <[e-mail]>]

On Wed, Jul 28, 2010 at 22:22, Matthias Waehlisch <[e-mail]>wrote:

> Hi all,
>
>  we submitted an updated version of the draft "A Common API for
> Transparent Hybrid Multicast":
>
>  * http://www.ietf.org/id/draft-waehlisch-sam-common-api-04.txt
>
>  This includes the following changes:
>
>   1.  Use cases added for illustration.
>
>   2.  Service calls added for inquiring on the multicast distribution
>       system.
>
>   3.  Namespace examples added.
>
>   4.  Clarifications and editorial improvements.
>
>
>  Comments are very welcome!
>

Hi,

I am proposing to use/implement a subset of this API inside a european
research project; let's see how it goes.

I have a question regarding the send() API.


4.3.2.  Receive


>    The receive call passes multicast data and the corresponding Group

   Name to the application.


>        int receive(int s, const uri group_name,

                   size_t msg_len, msg *msg_buf);


>    The s argument identifies the multicast socket.


>    The group_name argument identifies the multicast group for which data

   was received.


>    The msg_len argument holds the length of the received message.


>    The msg_buf argument points to the payload of the received multicast

   data.


>    On success the value 0 is returned, otherwise -1.



Is this a synchronous call?  What if there are multiple multicast sockets to
listen to, we need to create a separate thread for each?

I think we could use a select/poll/WaitForMultipleObjects equivalent for the
abstract multicast sockets.  I'd be happy to contribute text or ideas to
this end if you think is a good idea.

Best regards.

-- 
Gustavo J. A. M. Carneiro
INESC Porto, UTM, WiN, http://win.inescporto.pt/gjc
"The universe is always one step beyond logic." -- Frank Herbert