This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: patch to emit debug into emitted classes


On Apr 7, 2005 11:27 AM, Bryce McKinlay <mckinlay@redhat.com> wrote:
> 
> >For the verifier, I guess we can resurrect the
> >"-fnew-verifier" flag on the 4.0 branch (and, dare
> >I say, make it the default).
> >
> 
> This one I'm not really so worried about - anyone who's doing serious
> bytecode->native compilation ought to be using the BC-ABI anyway, which
> will always use the new verifier. I think porting the patch over would
> be pretty harmless, though, so we can do that if aph approves.

The "Binary Compatibility ABI" seems like a misnomer
for what has been implemented then. I mean, as a lay
developer if I just want to create a native executable
of my Java program (potentially depending on
classes external to my program and the JRE) and
am not worried about binary compatibility, I should
just be able to say:

  gcj -o foo foo.jar bar.jar --main=foo.Bar

I do not care if "bar.jar" is later changed by
upstream - I just want to have a self-contained
executable "foo" representing my Java application
in a native format.

In this case, without the new verifier I'll
stumble far more often than I would
otherwise.

Ranjit.

-- 
Ranjit Mathew      Email: rmathew AT gmail DOT com

Bangalore, INDIA.    Web: http://ranjitmathew.hostingzero.com/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]