This is the mail archive of the
java-patches@gcc.gnu.org
mailing list for the Java project.
Re: 3.0.1 PATCH: Support JNI_OnLoad on systems without weak definitions
Jeff Sturm writes:
> On 24 Jul 2001, Tom Tromey wrote:
> > I wonder if we really need all this JNI_OnLoad stuff here. It seems
> > to me that even if a JNI library is linked in to the final executable,
> > the Java code will still try to load the library dynamically, and the
> > code in natRuntime.cc will still look in the library for the
> > JNI_OnLoad symbol at runtime.
>
> It wouldn't do the right thing anyway if two such libraries each export
> JNI_OnLoad.
It would certainly depend on the details of the respective runtime linker's
symbol resolution order, and, while predictible for a given system, quite
hard to use portably. But whoever tries this gets what he deserves :-)
Rainer
-----------------------------------------------------------------------------
Rainer Orth, Faculty of Technology, Bielefeld University
Email: ro@TechFak.Uni-Bielefeld.DE