Submitted by Anonymous on Mon, 07/28/2003 - 21:26. General Questions
Hi there,
I am using the latest version of OpenMotif (2.2.2-3_ICS.i386.rpm) on a RedHat Linux 7.2 (Enigma Kernel 2.4.7-10 on an i686) system. Just recently I am trying to use a Java window with my application using JNI to call down to JVM and displaying a JavaHelp window. I am getting a problem where if I closed the window, the whole application crash. I am using Java version
java version "1.4.1_03"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.1_03-b02)
Java HotSpot(TM) Client VM (build 1.4.1_03-b02, mixed mode)
And the error msg is shown below
An unexpected exception has been detected in native code outside the VM.
Unexpected Signal 11 occurred at PC=0x40F3F630
Function=(null)+0x40F3F630
Library=/lib/i686/libc.so.6
NOTE We are unable to locate the function name symbol for the error
just occurred. Please refer to release documentation for possible
reason and solutions.
Current Java thread
at sun.awt.motif.MWindowPeer.pHide(Native Method)
at sun.awt.motif.MComponentPeer.setVisible(MComponentPeer.java741)
at sun.awt.motif.MComponentPeer.hide(MComponentPeer.java838)
at java.awt.Component.hide(Component.java1168)
- locked <0x4593e818> (a java.awt.Component$AWTTreeLock)
at java.awt.Window.hide(Window.java507)
at java.awt.Component.show(Component.java1137)
at java.awt.Component.setVisible(Component.java1089)
at javax.help.DefaultHelpBroker$2.windowClosing(DefaultHelpBroker.java855)
at java.awt.Window.processWindowEvent(Window.java1107)
at javax.swing.JFrame.processWindowEvent(JFrame.java266)
at java.awt.Window.processEvent(Window.java1066)
at java.awt.Component.dispatchEventImpl(Component.java3639)
at java.awt.Container.dispatchEventImpl(Container.java1623)
at java.awt.Window.dispatchEventImpl(Window.java1590)
at java.awt.Component.dispatchEvent(Component.java3480)
at java.awt.EventQueue.dispatchEvent(EventQueue.java450)
at java.awt.EventDispatchThread.pumpOneEventForHierarchy(EventDispatchThread.java197)
at java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java150)
at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java144)
at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java136)
at java.awt.EventDispatchThread.run(EventDispatchThread.java99)
I was unable to find out the exact cause of this problem and is wondering if anyone has come accross this before.
If I have an application that do not use any GUI or link against the -ljawt libraries on Linux, I don`t get this problem. I am suspecting that perhaps this is related to something with OpenMotif - but is not sure exactly.
I came across an article from the motifdeveloper web site
(http//www.motifdeveloper.com/tips/tip22.html) stating the following
..."In summary, OpenMotif 2.2 is not product quality, and is a hurried attempt to incorporate some legacy and proprietary Motif 1.x widgets into the Motif 2.1 toolkit, without upgrading to the new model. This causes a large number of problems, sufficiently serious as to break the toolkit in a number of ways. Since the proposed new components simply replicate functionality already present in the standard Motif 2.1.30 toolkit, and since the entire set of additions is completely thread-unsafe, there is regrettably no gain but pain. "
And reviews at http//www.motifdeveloper.com/tips/Motif22Review.pdf
My questions are
1) Is this review true?
2) Is OpenMotif going to respond to this? If some of these points are true. Is there any plan for OpenMotif to fix them? if so, when will this be addressed?
3) Would OpenMotif 2.3 be better?
Thanks,
John
Hi,
Some of the OpenMotif team took a look at your problem, and they are 90% convinced this is a Java problem not a Motif one. Did you rebuild the Java subsystem (and especially AWT that relies on Motif) on your system or did you just install an RPM?
OM 2.1.30 is not compatible with OM 2.2.2 which is not compatible with lesstif (the standard Motif on RH 7.2). If AWT was originally built with OM 2.1.30 and it tries to link with the OM 2.2.2 shared library, you will see a core dump like you reported. (This could also happen if Java was built against lesstif and you tried to link it into one of the Motif shared libraries). That would be my suggested first path of investigation.
In regards to your questions
1) Is this review (Fountain`s) true?
In some areas. However, most of the comments were on architectural issues that would prevent the new widgets from playing well with the new OM 2.x functionality (traits, multi-threading, etc.) For most end users, I think that having builtin tooltips and new widgets (trees, etc.) is a better situation than not having them.
I think the community as a whole would have benefited more if the author had decided to help (i.e., write some code!) than take the easy way out and just turn up his nose and complain.
2) Is OpenMotif going to respond to this? If some of these points are true. Is there any plan for OpenMotif to fix them? if so, when will this be addressed?
Yup. Stay tuned for news. (Actually, many are already fixed in CVS. This review is quite old.)
3) Would OpenMotif 2.3 be better?
Yup. Isn`t every new release better than the last? ;-)
Mark