Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[Eclipse Bug] JVM Crash #4854

Closed
amitjoy opened this issue Sep 20, 2021 · 8 comments · Fixed by #4862
Closed

[Eclipse Bug] JVM Crash #4854

amitjoy opened this issue Sep 20, 2021 · 8 comments · Fixed by #4862

Comments

@amitjoy
Copy link
Member

amitjoy commented Sep 20, 2021

Even though it is a SWT problem, I thought it might be a good idea to keep you informed about it so that we can track the resolution of this SWT bug.

Here is the link: https://bugs.eclipse.org/bugs/show_bug.cgi?id=576099

@bjhargrave
Copy link
Member

I guess we can close this since there is nothing for Bndtools to do?

@bjhargrave bjhargrave changed the title [Bug] JVM Crash [Eclipse Bug] JVM Crash Sep 22, 2021
@pkriens
Copy link
Member

pkriens commented Sep 23, 2021

Leave it a bit open. It happens in the Bndexplorer but not in the package explorer. It is consistent, maybe others have similar experiences.

@bjhargrave
Copy link
Member

I get it all the time and hate it. I did not realize it was only happening in the Bndtools Explorer view. I will go back to using the Package Explorer view and save myself from having to learn to not collapse things in the tree view...

@kriegfrj
Copy link
Contributor

For now, at a minimum, we should make a note in the release notes for 5.3.0 and 6.0.0?

amitjoy added a commit to amitjoy/bnd that referenced this issue Sep 27, 2021
This workaround ensures that the action is performed on the display thread properly. However, the bug still exists in the SWT/JFace which needs to be fixed. Other plugins using `getTreeViewer().refresh()` from `PackageExplorerPart` are prone to crash the JVM as well.

Fixes bndtools#4854
amitjoy added a commit to amitjoy/bnd that referenced this issue Sep 27, 2021
This workaround ensures that the action is performed on the display thread properly. However, the bug still exists in the SWT/JFace which needs to be fixed. Other plugins using `getTreeViewer().refresh()` from `PackageExplorerPart` are prone to crash the JVM as well.

Fixes bndtools#4854
Signed-off-by: Amit Kumar Mondal <admin@amitinside.com>
amitjoy added a commit to amitjoy/bnd that referenced this issue Sep 28, 2021
This workaround ensures that the action is performed on the display thread properly. However, the bug still exists in the SWT/JFace which needs to be fixed. Other plugins using `getTreeViewer().refresh()` from `PackageExplorerPart` are prone to crash the JVM as well.

Fixes bndtools#4854
Signed-off-by: Amit Kumar Mondal <admin@amitinside.com>
bjhargrave pushed a commit to bjhargrave/bnd that referenced this issue Sep 28, 2021
This workaround ensures that the action is performed on the display thread properly. However, the bug still exists in the SWT/JFace which needs to be fixed. Other plugins using `getTreeViewer().refresh()` from `PackageExplorerPart` are prone to crash the JVM as well.

Fixes bndtools#4854
Signed-off-by: Amit Kumar Mondal <admin@amitinside.com>
Signed-off-by: BJ Hargrave <bj@hargrave.dev>
@amitjoy
Copy link
Member Author

amitjoy commented Jan 17, 2022

I just recently noticed that this problem reappeared in the latest snapshot (in combination with Eclipse 2021-12)

@kriegfrj
Copy link
Contributor

I don't think I've ever seen this. Is it a particular platform implementation of SWT (Mac maybe) that has the issue?

@amitjoy
Copy link
Member Author

amitjoy commented Jan 17, 2022

@kriegfrj Yeah, it's a known issue in Eclipse SWT project which happens on Mac. Previously in Bndtools REL 6.1, we managed to find a workaround which prevented this from happening but it reappeared once again in 6.2.x-SNAPSHOT.

@chrisrueger
Copy link
Contributor

chrisrueger commented Jul 4, 2022

Just want to confirm I get it all the time now with bndtools 6.3.1.REL-202206071316-g30a66dc on Mac.

Strangly I got it since upgrading to Eclipse 2022-06 and downgraded again to 2022-03... But still getting it (also cleaned Eclipse .metadata folder).

Unfortunatelly I cannot tell exactly when I upgraded the bndtools lib to which version. But I have the feeling that my Eclipse just crashed at least 10 times since last week.

But maybe something helpful: I got that error when doing simple java changes, pressing save which triggered a rebuilt.
Eclipse just closed without any warning. Then Apple Crash Log appeared:

Crashed Thread:        94  Java: Worker-63: Building

Exception Type:        EXC_BAD_ACCESS (SIGABRT)
Exception Codes:       KERN_INVALID_ADDRESS at 0x0000000000000000
Exception Codes:       0x0000000000000001, 0x0000000000000000
Exception Note:        EXC_CORPSE_NOTIFY

Thread 94 Crashed:: Java: Worker-63: Building
0   libsystem_kernel.dylib        	       0x1a14d6d98 __pthread_kill + 8
1   libsystem_pthread.dylib       	       0x1a150bee0 pthread_kill + 288
2   libsystem_c.dylib             	       0x1a1446340 abort + 168
3   libjvm.dylib                  	       0x1107c972c os::abort(bool, void*, void const*) + 56
4   libjvm.dylib                  	       0x110a09fa8 VMError::report_and_die(int, char const*, char const*, char*, Thread*, unsigned char*, void*, void*, char const*, int, unsigned long) + 2684
5   libjvm.dylib                  	       0x110a09520 VMError::report_and_die(Thread*, unsigned int, unsigned char*, void*, void*, char const*, ...) + 72
6   libjvm.dylib                  	       0x110a0a0f0 VMError::report_and_die(Thread*, unsigned int, unsigned char*, void*, void*) + 36
7   libjvm.dylib                  	       0x1108e5320 JVM_handle_bsd_signal + 420
8   libsystem_platform.dylib      	       0x1a15234a4 _sigtramp + 56
9   ???                           	       0x150026f0c ???
10  ???                           	       0x159f94ba8 ???

# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

5 participants