[Click] On initialize()

[Adam Greenhalgh]

Ed et al.,

I've up dated Ross's patch for linux to take in your recent changes, do
you have a time line for further integration of click into linux 2.6.x ?

If I've missed something please yell.

http://cellini.cs.ucl.ac.uk/~adam/linux-click/linux-kernel-2.6.10_click.patch

stuff I removed from Ross's patch is :

http://cellini.cs.ucl.ac.uk/~adam/linux-click/linux-kernel-2.6.10_click.patch.removed

Adam 

On Tue, 2005-03-22 at 08:35 -0800, Eddie Kohler wrote:
> Hi Chema,
> 
> We generally implement tasks like this usign a Timer that's set in initialize(), 
> or a Task.  I strongly suggest you go this way.  But the configure_phase() 
> mechanism can be used to determine the configure+initialize order; elements are 
> config/initialized in increasing configure_phase order.
> 
> E
> 
> 
> José María González wrote:
> > Hi, 
> > 
> > I'm wondering whether there is any mechanism in click to permit elements 
> > carrying out some work between initialize() and the start of the packet 
> > processing. I'm writing an element whose very first task is to beacon
> > its existence. I can't send the beacon during initialize(), as the next
> > element (a Queue) is no initialized yet (click seems to initialize()
> > elements following the packet stream direction), so it has to be after
> > that. 
> > 
> > As an alternative, is there any mechanism to alter the order in which 
> > elements are initialized? 
> > 
> > Thanks in advance. 
> > -Chema
> > 
> > _______________________________________________
> > click mailing list
> > click at amsterdam.lcs.mit.edu
> > https://amsterdam.lcs.mit.edu/mailman/listinfo/click
> 
> 
> _______________________________________________
> click mailing list
> click at amsterdam.lcs.mit.edu
> https://amsterdam.lcs.mit.edu/mailman/listinfo/click