Re: Questions about GC, _Jv_MarkObj() and reflection data.

[Andrew Haley <aph at redhat dot com>]

Tom Tromey writes:
 > >>>>> "David" == David Daney <ddaney@avtrex.com> writes:
 > 
 > David> Currently in make_class_data (in class.c) the call to
 > David> get_dispatch_table (which creates the vtable and thus the GC
 > David> descriptor) is done after all the reflection data is generated.  My
 > David> plan is to move it before the field and method arrays are built.  Then
 > David> it should be fairly simple to extract the GC descriptor from the
 > David> dispatch table tree and use that to decide if field reflection data is
 > David> needed for reference fields.
 > 
 > David> Does anyone foresee a problem with doing it this way?
 > 
 > Not offhand.
 > 
 > Another approach which is more space efficient is to fall back to
 > conservative scanning for classes which can't be described using a
 > bitmap.  Off the top of my head I don't know exactly how you'd
 > implement this.

Just make _Jv_MarkObj() do it.  It's the easiest solution, for sure.

Andrew.