This is the mail archive of the java@gcc.gnu.org mailing list for the Java 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] | |
In GCC 4.1, some parts of libgcj, including org.xml, are compiled with the new "binary compatibility" ABI. It isn't safe to call BC-ABI code from code built with the old, "C++ ABI". Building your code with the BC-ABI ("-findirect-dispatch") should solve the problem.
The list of packages currently built with the BC-ABI in libgcj 4.1 includes:
gnu.xml javax.imageio javax.xml gnu.java.beans gnu.java.awt.peer.gtk org.xml org.w3c
Currently, code that refers directly to these classes should be built with -findirect-dispatch.
Bryce
S. Park wrote:
I'm experiencing some strange behavior with a simple class compiled with gcj 4.1 on FreeBSD 4.11 and 5.4 IA32. It compiles and runs as is, but if I subclass anything in the package org.xml.sax (e.g. by uncommenting the 'extends' clause below), the executable seg faults. What's strange is that I can instantiate objects of types in that same package within the body of the main method. The problem only appears when subtyping. The test class looks like this:
import java.util.HashMap; import org.xml.sax.helpers.DefaultHandler; import org.xml.sax.InputSource;
public class Foo .. extends DefaultHandler { public static void main(String argv[]) { InputSource is = new InputSource(); String enc = is.getEncoding();
DefaultHandler dh = new DefaultHandler();
try { dh.startDocument(); } catch (Exception e) { System.err.println(e.getMessage()); }
System.out.println(enc);
} }
Has anyone else run into this?
Sooyoung
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |