Re: Loading native libraries

[Archie Cobbs]

David P Grove wrote:
> > > Before I go and implement this, does this sound correct to everyone?
> > 
> > I think we desperately need a VMClassLoader.getCallerClassLoader()
> > method. That way we can simply modify the various load/loadLibrary
> > methods to call that. It's also much more efficient to implement and can
> > be interacts better with method inlining.
> 
> In Jikes RVM we use an internal widget called VM_StackBrowser.  Think of 
> it as an interator on the thread stack.  It supports the following primary 
> functionality:
> (1) get the VMClass or java.lang.Classloader for the current stackframe
> (2) move up one stackframe.

That's a neat idea.. presumably it keeps some internal "pointer" to
the current (or next) stack frame. So does this pointer become invalid
if the method using the VM_StackBrowser returns? I.e., are you restricted
to only using a VM_StackBrowser instance within the same method that
created it? Guess so, because otherwise you'd have to copy all the pointers.

This would be acceptable I think, because we could make this VM_StackBrowser
class package private to java.lang and so ensure proper usage.

(In JC it's easy to see how to implement this using a struct _jc_stack_crawl.)

-Archie

__________________________________________________________________________
Archie Cobbs      *        CTO, Awarix        *      http://www.awarix.com