harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mikhail Fursov" <mike.fur...@gmail.com>
Subject Re: [dlrvm] ClassCircularityError in recursive compilation (Was: Re: [drlvm] smoke test : gc PhantomReferenceQueueTest is really unstable)
Date Fri, 13 Oct 2006 05:40:36 GMT
On 10/13/06, Gregory Shimansky <gshimansky@gmail.com> wrote:
> On Thursday 12 October 2006 14:06 Mikhail Fursov wrote:
> > On 10/12/06, Pavel Pervov <pmcfirst@gmail.com> wrote:
> > > Yes, lazy (interpreter style) resolution will solve the problem - JIT
> > > will never ask to load classes while compiling.
> >
> > Here we have different vision. What do you think, is it possible to go
> into
> > the endless resursion with lazy resolution?
> AFAIK it is not a matter of endless recursion. It just the a matter of how
> VM
> class loader determines CCE condition. If a class loading requested for a
> class which is being loaded already somewhere up the stack, it is
> determined
> as a CCE. This method is very effective and works just fine, I think RI
> uses
> something similar. It doesn't require analyzing the whole classes tree and
> exception is thrown only on demand.
Gregory, I just tried to hint that it's possible to write a code in Java
when to resolve a class you need to execute one of its methods :)
Note, that I do not mention compilation or JIT here at all. It's more
general problem.

Mikhail Fursov

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message