-
Bug
-
Resolution: Duplicate
-
P1
-
None
-
5.0u8
-
x86
-
windows, windows_xp
POGO.com report a memory regression in 5u8. Run a few simple applets (or the applet supplied from pogo) in the same browser, and then navigate the browser to a page without any applet. Perform GC.
If you hook up JProfiler, you will still see instances of AppletViewer and IExplorerEmbedded frame there, with a JNI Gloal Ref referenceing it, preventing it from GC.
JProfiler screenshot:
http://javaweb.sfbay.sun.com/~ngthomas/pogo/applet-jni-ref.PNG
Pogo.com simple applet source:
import java.applet.Applet;
public class JNIGlobalReferenceLeakDemoApplet extends Applet {
private int[] mBigLeak;
public void init() {
mBigLeak = new int[1000000];
for (int i = 0; i < mBigLeak.length; i++) {
mBigLeak[i] = i;
}
}
}
With 5u7, the problem cannot be reproduced.
Also, if I replace awt.dll and rt.jar from 5u7 to 5u8, I cannot reproduce the problem too. So I am assigning the bug to awt so they can take a look.
If you hook up JProfiler, you will still see instances of AppletViewer and IExplorerEmbedded frame there, with a JNI Gloal Ref referenceing it, preventing it from GC.
JProfiler screenshot:
http://javaweb.sfbay.sun.com/~ngthomas/pogo/applet-jni-ref.PNG
Pogo.com simple applet source:
import java.applet.Applet;
public class JNIGlobalReferenceLeakDemoApplet extends Applet {
private int[] mBigLeak;
public void init() {
mBigLeak = new int[1000000];
for (int i = 0; i < mBigLeak.length; i++) {
mBigLeak[i] = i;
}
}
}
With 5u7, the problem cannot be reproduced.
Also, if I replace awt.dll and rt.jar from 5u7 to 5u8, I cannot reproduce the problem too. So I am assigning the bug to awt so they can take a look.
- duplicates
-
JDK-6471044 Memory leak in native cursor code
- Closed
-
JDK-6351698 Regression: 4506928 testcase is passing with 142_10-b03 but failing with 142_11-b01
- Resolved
- relates to
-
JDK-6462383 Unbounded memory leak in Windows XP JRE for Applets and applications that open JFrame's
- Closed
-
JDK-6474128 REGRESSION: Frame Memory Leak
- Closed