RE: [abc] FW: Aprove als Benchmark fuer Compileroptimierung

Ok, I will just check in the bytecode then. Weaving is no problem, yes. Obviously they asked us that *if* we find any way to speed up Aprove that we would tell them how to do so, because they have those term rewriting tool competitions every year, where you have to (dis)prove termination on as many systems as possible within a given amount of time. Hence, speeding up their tool would also give a big benefit for them.
 
Eric
 

--
Eric Bodden
Sable Research Group
McGill University, Montréal, Canada
________________________________
From: Majordomo list server on behalf of Ganesh Sittampalam
Sent: Wed 25.01.2006 08:29
To: abc@comlab.ox.ac.uk
Subject: Re: [abc] FW: Aprove als Benchmark fuer Compileroptimierung
On Wed, 25 Jan 2006, Eric Bodden wrote:
> Good news: I just got the permission to use Aprove in source format for
> our purposes. However, we have to make sure that NEVER any of the source
> would be accessible to people not being member of our team, especially
> not to the outer world, i.e. the internet etc.
>
> I will try to check in the code by the end of the week.
I suggest you don't do this until we're sure we want to use it in source
form. At the present moment, we don't, because we can't compile Java 5
anyway. Putting it into our repository means we will forever have to worry
about restricting access to the repository, so we may want to keep it
elsewhere, even once we are ready for it.
For now we can keep working with the bytecode, which presumably we also
effectively have permission to weave into now.
Cheers,
Ganesh