Final methods and the BC-ABI

Andrew Haley aph@redhat.com
Mon Nov 10 13:46:00 GMT 2003


Bryce McKinlay writes:
 > On Nov 10, 2003, at 5:56 PM, Per Bothner wrote:
 > 
 > > Bryce McKinlay wrote:
 > >
 > >> Currently, the compiler does not generate vtable entries at all for 
 > >> final methods under the "current" ABI. This poses a problem in that 
 > >> --indirect-dispatch code will not be able to inter-operate with 
 > >> "current ABI" code. So, I propose that the compiler be changed to 
 > >> generate vtable entries for all final methods.
 > >
 > > Is this the only (or primary) binary interoperability between the two 
 > > ABIs?  If the long-term migrataion path is towards the binary 
 > > compatible ABI as the default, then I don't see why interoperability 
 > > matters.
 > 
 > Right, the long-term goal is for the BC-ABI to be the default for 
 > systems with shared libraries. However while the BC-ABI is still in 
 > development, its very helpful to have interoperability between the two 
 > ABIs - this way we can build a binary with --indirect-dispatch and have 
 > it work against a libgcj built with the old ABI.

I don't want to rule out the possibility of building libgcj with
-Bsymbolic (so that all internal calls are direct where possible).
I'm not saying we *should* do this, but let's not rule out the
possibility before we've measured it.

Andrew.



More information about the Java mailing list