Re: OpenOffice and gcj at runtime

[Bryce McKinlay <mckinlay at redhat dot com>]

Tom Tromey wrote:

> So when we get to the call that hangs, we see that the method index is
> "1".  But this points to some other method in the vtable -- since in
> this case the index is really an interface table index, not a vtable
> index.
>
> I agree that this fix is just a band-aid for 4.0.
>
> Retro-fitting the declaring class into the jmethodID is somewhat ugly,
> since it means more memory use (another pointer per method) just for
> this one obscure case.
>  
Yeah. Unfortunately I don't see a way to make JNI interface calls 
efficient without having the pointer, but perhaps such calls are rare 
enough that we can ignore them (the small number of complaints about 
this bug suggests this might be true). But, note that there are some 
other cases where it would make things easier if we had this pointer.

When we have fully converted to the BC-ABI, the extra memory will be 
less of a concern because the jmethodIDs will be allocated at runtime, 
and only methods from initialized/linked classes will occupy memory. 
Since they are currently taking up space in binaries, size is much more 
of a concern.

In the short term, perhaps adding another hidden method modifier/flag 
would be the best fix.

Bryce