| Summary: | StringIndexOutOfBounds exception viewing monitor information in the Debug view | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Eclipse Project] JDT | Reporter: | Michael Rennie <Michael_Rennie> | ||||
| Component: | Debug | Assignee: | Michael Rennie <Michael_Rennie> | ||||
| Status: | VERIFIED FIXED | QA Contact: | |||||
| Severity: | normal | ||||||
| Priority: | P3 | CC: | daniel_megert | ||||
| Version: | 3.6 | ||||||
| Target Milestone: | 3.8 M3 | ||||||
| Hardware: | PC | ||||||
| OS: | Windows 7 | ||||||
| Whiteboard: | |||||||
| Attachments: |
|
||||||
|
Description
Michael Rennie
The problem is two-fold: 1. I had manually suspended the debug target, meaning object information is not available 2. this causes the value of '<unknown>' to be passed around as native object names, which is what is causing the SIOOBE because we think it is a generic signature. Created attachment 203441 [details]
fix
Pushed changes to master. This fix is not good. I now get "null null" displayed as value in the 'Variables' view.
Test case:
1. paste:
public static void main(String[] args) {
Object a= null;
System.out.println(a);
}
2. set breakpoint on second line in body
3. Debug
pushed updated fix + regression tests to master. Please verify Curtis. >JDT / Platform Debug would also like to contribute to a re-build. We are tagged
>and ready to go.
Did that not work? I still see the bug in I20111011-1046. Works in N20111011-2000.
(In reply to comment #6) > >JDT / Platform Debug would also like to contribute to a re-build. We are tagged > >and ready to go. > Did that not work? I still see the bug in I20111011-1046. Works in > N20111011-2000. Chatting with Kim, apparently my email did not make it to her prior to the build starting, so the fix is *not* in the I-build. verified in I20111027-1800 As a note, a good test case to verify the fix is to follow the steps from bug 345298, and once the deadlock occurs - in your target workspace - suspend the VM in your host. |