Le jeu 05/06/2003 � 18:41, Richard Eckart a �crit :
> > The major issue that need to be resolved before even starting is
> > defining the protocol that will be used between the GUI and the core.
> > I did not have any time to work on this specification yet.  I'm not
> > sure I even thought out all the issues there are with the separation.
> 
> Another problem is that to my knowledge gtk-gnutella relies very much on
> it's singlethreadedness. I sure a lot of ugly race conditions will turn
> up unless the gui/core protocol is very carefully designed and
> implemented or you have to make all calls from gui to core blocking. 

Maybe it would be 1 role more for the frontage: managing locks. This
protects the single threaded design of the core.

The server which uses the frontage receives multithreaded requests. Each
request make the server calls to the frontage. A timout of the server
detects possibly dead locks of the frontage and trace them for
debugging. The timout of the TCP connection detects a client failure.
This last timout does not forbid other requests to be processed.

I don't really knows but object oriented encapsulation seems to make
multithreading management easier. Doesn't it ?

J�r�me



-------------------------------------------------------
This SF.net email is sponsored by:  Etnus, makers of TotalView, The best
thread debugger on the planet. Designed with thread debugging features
you've never dreamed of, try TotalView 6 free at www.etnus.com.
_______________________________________________
Gtk-gnutella-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/gtk-gnutella-devel

Reply via email to