Bug 13964

Summary:

Exception on startup

Product:

[Eclipse Project] JDT

Reporter:

DJ Houghton <dj.houghton>

Component:

Core

Assignee:

Philipe Mulet <philippe_mulet>

Status:

RESOLVED WONTFIX

QA Contact:

Severity:

normal

Priority:

Version:

2.0

Target Milestone:

2.0 M6

Hardware:

OS:

Windows XP

Whiteboard:

Attachments:

Description	Flags
console message and core dump.	none

Description DJ Houghton

2002-04-16 21:22:47 EDT

build 2002-04-16 on WinXP.

I tried to start the new build on top of my old workspace and got an exception 
to the console and Eclipse wouldn't start at all. Will attach exception and 
core dump.

Comment 1 DJ Houghton

2002-04-16 21:23:06 EDT

Created attachment 637 [details]
console message and core dump.

Comment 2 DJ Houghton

2002-04-16 21:31:47 EDT

Note this is reproducable every time.

Also note I:
- went back to 2002-04-12
- started up ok
- shutdown ok
- tried 2002-04-16
- got the same exceptions (without the "mru list" errors)

Comment 3 Jerome Lanneluc

2002-04-17 05:20:59 EDT

I could not reproduce with DJ's workspace and IBM JDK 1.3.1.
However, by looking at the provided trace this looks like a jit problem with 
this VM.

DJ, could you try another VM and see if you still have the problem?

Comment 4 Philipe Mulet

2002-04-17 06:02:35 EDT

Cannot reproduce it either. DJ, could you alternatively try to disable the JIT 
and see if the problem still occurs ?

Comment 5 DJ Houghton

2002-04-17 08:13:07 EDT

IBM JRE 1.3.1 (no options) -> same results
IBM JRE 1.3.1 -Dcompiler=none -> same results
IBM JRE 1.3.0 (no options) -> works OK

Note that I have been using the 1.3.1 JRE on that workspace (but with previous 
builds) for about 2 weeks.

Comment 6 Philipe Mulet

2002-04-17 09:31:37 EDT

You need to use "-Djava.compiler=none" to disable the jit.

Comment 7 DJ Houghton

2002-04-17 09:37:46 EDT

I was close! :-)

It worked OK when I started up 1.3.1 with -Djava.compiler=none.

Comment 8 Philipe Mulet

2002-04-17 13:59:51 EDT

Then it sounds like an IBM jit 1.3.1 bug.

Comment 9 Philipe Mulet

2002-04-23 05:35:57 EDT

Closing as an IBM 1.3.1 jit bug.