[reSIProcate] reSIP Process model

[rchakraborty at velankani.com]

Are FIFOs in reSIProcate stack strictly used to communicate between threads?

Quoting Byron Campen <bcampen at estacado.net>:

> 	This architecture appears to have been chosen so that, at some point
> in the future, it would be compatible with multithreading the
> transports. (I am theorizing here, someone who was involved in the
> early design stages will have to elaborate.)
>
> Best regards,
> Byron Campen
>
>> Thanks Byron and Scott,
>>
>> What I am not able to understand is about the Transaction FIFO and   
>>  Transport FIFO; they must be used to communicate between two  
>> threads.
>> Which are those threads. I was trying to compare the architecture    
>> document http://warsaw.sjc.purplecomm.com/wiki/index.php?   
>> title=Architecture_Overview with the source code. Does the    
>> architecture depict current implementation?
>>
>> Thanks,
>> Rishin
>>
>> Quoting Byron Campen <bcampen at estacado.net>:
>>
>>> 	The stack itself is a single thread. Each TransactionUser (TU) can
>>> also be a separate thread. DUM is an example of a TU, repro is   another
>>> example (see Proxy.[hc]xx).
>>>
>>> Best regards,
>>> Byron Campen
>>>
>>>> Hi,
>>>>
>>>>
>>>> I am going through the reSIProcate code. I have a doubt about    
>>>> the   stacks process model.
>>>>
>>>> The wiki pages say that the FIFOs in reSIP are used to    
>>>> communicate  between threads; however, I could not find any    
>>>> thread other than   SipThread within reSIP.
>>>>
>>>> Am I missing something here? Can someone help let me know about    
>>>> the  process or thread model of reSIP?
>>>>
>>>>
>>>> Thanks,
>>>>
>>>> Rishin
>>>>
>>>>
>>>> _______________________________________________
>>>> resiprocate-devel mailing list
>>>> resiprocate-devel at list.sipfoundry.org
>>>> https://list.sipfoundry.org/mailman/listinfo/resiprocate-devel
>>
>>
>>
>>