Re: compute-branch

[Lloyd Wood <L.Wood@surrey.ac.uk>]

On Fri, 26 Feb 1999, Huang Hai wrote:

> I have downloaded 1.16 of CtrMcastComp.tcl and re-built ns, but after that,
> ns cannot work.

since, as you've discovered, major changes to a file are rarely made
in isolation, I'd suggest downloading and building an entirely new ns
snapshot.

> _o3: unable to dispatch method maybeEnableTraceAll
[..]
> I think maybeEnableTraceAll is a new method but I cannot find it in
> "CVS Tree". (I have checked the lastest version of CtrMcast.tcl, CtrRPComp.tcl,
> ns-mcast.tcl, etc.) 

This method is in:

~ns/tcl/lib/ns-trace.tcl

I would build an ns snapshot, so you get all the updates at once. IMO,
snapshots are the only way to run ns...


> Please tell me which files should I also update. And I
> think there should be a dependence description for each file.

well, dependencies on stuff in ~ns/tcl/lib seems obvious - but Tcl
dependencies are rather less obvious/visible than C includes, or even
perl use statements, say.

That's arguably something of a language/namespace management failing
on OTcl's part, but it means you don't have to have a zillion
includes in your simulation script before it does anything. Having
visible dependencies in one but not in the other strikes me as
difficult without some sort of 'include everything' get-out, which
everyone would then start using within ns as easier to do.

Otherwise, dependence descriptions become documentation problems - and
documentation is always out of sync with code.

> I have browsed compute-branch of 1.16. It's very different with the old one.
> Since it doesn't find upstream node when $tmp==$target, I think my script
> can run smoothly with it. However, I'm not sure before I have tested it.

...on a snapshot.

Cheers,

L.


> LW> On Fri, 26 Feb 1999, Huang Hai wrote:
> 
> LW> Huang, take a look at the CVS information for this file:
> 
> LW> http://www-mash.cs.berkeley.edu/cgi-bin/cvsweb/ns-2/tcl/ctr-mcast/CtrMcastComp.tcl
> 
> LW> It appears that you're looking at version 1.1.3 or earlier of that
> LW> file (ideally the version # would be somewhere in the comments at the
> LW> top).
> 
> LW> You will want to look at Polly/Kannans's small changes in
> LW> 1.1.4 and further changes in 1.1.5/1.1.6, which modify this bit of
> LW> code extensively, and see how those changes affect your problem and
> LW> your proposed fix.

<L.Wood@surrey.ac.uk>PGP<http://www.ee.surrey.ac.uk/Personal/L.Wood/>