> Stephen Crawley wrote:
> > Stephen Crawley wrote:
> > 
> > I understand this.  Unfortunately it does not work for us at the moment
> > because Kissme's native methods do not "obey the rules", and are hence
> > not GC safe.
> 
> How about adding the special code in *kissme*'s native methods then? 
> That is to say, assume that any JNI function is GC safe unless it 
> specifies, somehow, that it isn't. Then go through all the 
> Kissme-internal native calls that worry you (probably automatically, 
> using a script) and add the necessary specification.
> 
> This "specification" could be as simple as finding the start of a 
> function, adding a call to "kissme_disable_gc()" or whatever, and then 
> going to all exit points and adding "kissme_enable_gc()". Or just adding 
> "kissme_disable_gc_for_this_function()" and letting kissme itself figure 
> out what to do when the function returns.
> 
> That seems like a lot less effort than doing the same kind of thing, but 
> in reverse, for all *classpath*'s native methods.

This is a REALLY GOOD IDEA!  I wish I'd thought of it.

I'll start trying to implement this approach tonight.

-- Steve


-- Steve



_______________________________________________
Classpath mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/classpath

Reply via email to