Re: Patch: RFC: fix non-BC / BC link problem

[Bryce McKinlay <mckinlay at redhat dot com>]

Andrew Haley wrote:

> > Index: libjava/link.cc
> > ===================================================================
> > RCS file: /cvs/gcc/gcc/libjava/link.cc,v
> > retrieving revision 1.16
> > diff -u -r1.16 link.cc
> > --- libjava/link.cc 15 Jun 2005 19:11:36 -0000 1.16
> > +++ libjava/link.cc 16 Aug 2005 19:05:59 -0000
> > @@ -1659,11 +1659,11 @@
> >    if (codesource == NULL)
> >      codesource = "<no code source>";
> >  
> > -  // We use a somewhat bogus test for the ABI here.
> >    char *abi;
> >    if (_Jv_IsInterpretedClass (klass))
> >      abi = "bytecode";
> > -  else if (klass->state == JV_STATE_PRELOADING)
> > +  // There isn't really a better test for the ABI type at this point.
> > +  else if (klass->otable_syms || klass->atable_syms || klass->itable_syms)
>
> It is possible for a class to have no syms at all?  What will happen?
>  

Its unfortunate that we don't have a proper way to check the ABI type.

klass->next_or_version would work (test for FLAG_BINARYCOMPAT_ABI), but 
we overwrite the version field during linking to use it for the class 
chain - probably this is not a good idea as the version field can come 
in useful later, for things like this, as well as debugging.

Perhaps this current hack could be encapsulated in a 
_Jv_IsBinaryCompatABI() function in order to make it easier to improve 
it later?

Bryce