Bug 438724 - [flex] REVEAL delta does not always work reliably
Summary: [flex] REVEAL delta does not always work reliably
Status: VERIFIED FIXED
Alias: None
Product: Platform
Classification: Eclipse Project
Component: Debug (show other bugs)
Version: 4.4   Edit
Hardware: PC Windows 7
: P3 normal (vote)
Target Milestone: 4.5 M3   Edit
Assignee: Anton Leherbauer CLA
QA Contact:
URL:
Whiteboard:
Keywords: contributed
Depends on:
Blocks:
 
Reported: 2014-07-02 08:30 EDT by Anton Leherbauer CLA
Modified: 2014-10-30 10:37 EDT (History)
3 users (show)

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Anton Leherbauer CLA 2014-07-02 08:30:51 EDT 
When a thread or stack frame is revealed in the Debug view (on hitting a breakpoint) it may get shifted up or down due to subsequent content updates of items higher up in the tree.
Note that this is similar to bug 324100, which was fixed by delaying the handling of reveal until all content updates have been completed.
I am not sure if it is reproducible with the Java debugger, but I will provide a JUnit test which simulates the issue together with a proposed fix.
Comment 1 Anton Leherbauer CLA 2014-07-02 09:35:28 EDT 
Proposed fix and test case for review: https://git.eclipse.org/r/29316
Comment 2 Anton Leherbauer CLA 2014-08-05 05:55:29 EDT 
I have pushed the patch again. Last time it was on the wrong branch: https://git.eclipse.org/r/#/c/31022/
Comment 3 Dani Megert CLA 2014-08-05 07:11:09 EDT 
(In reply to Anton Leherbauer from comment #2)
> I have pushed the patch again. Last time it was on the wrong branch:
> https://git.eclipse.org/r/#/c/31022/

Hi Anton. We usually fix a bug in the current stream (master) and only when the fix proofs to be OK, we backport it.

Is this a regression compared to 4.3?
Comment 4 Anton Leherbauer CLA 2014-08-05 07:13:07 EDT 
(In reply to Dani Megert from comment #3)
> Hi Anton. We usually fix a bug in the current stream (master) and only when
> the fix proofs to be OK, we backport it.
> 
> Is this a regression compared to 4.3?

No, I don't think it is a regression.
I'll post another patch against master, then.
Comment 5 Anton Leherbauer CLA 2014-08-05 07:36:44 EDT 
New patch against master: https://git.eclipse.org/r/31029
Comment 6 Sarika Sinha CLA 2014-10-06 07:42:34 EDT 
Anton, 
Do you have steps to reproduce in some other debugger which we can test with the fix ?
Comment 7 Anton Leherbauer CLA 2014-10-06 08:08:26 EDT 
(In reply to Sarika Sinha from comment #6)
> Anton, 
> Do you have steps to reproduce in some other debugger which we can test with
> the fix ?

So far I can only reproduce with the TCF debugger + Simics backend.
Comment 8 Sarika Sinha CLA 2014-10-21 02:31:31 EDT 
Patch looks fine with the tests.

No Regression in JDT.
Comment 9 Dani Megert CLA 2014-10-21 06:41:46 EDT 
(In reply to Sarika Sinha from comment #8)
> Patch looks fine with the tests.
> 
> No Regression in JDT.

(In reply to Anton Leherbauer from comment #5)
> New patch against master: https://git.eclipse.org/r/31029

Please see my comments there.
Comment 10 Dani Megert CLA 2014-10-23 07:29:08 EDT 
(In reply to Anton Leherbauer from comment #5)
> New patch against master: https://git.eclipse.org/r/31029

Thanks Anton. Submitted with http://git.eclipse.org/c/platform/eclipse.platform.debug.git/commit/?id=e598699286ab39f12bbcc0d7b1fc2475f11d40f5
Comment 11 Sarika Sinha CLA 2014-10-29 02:41:44 EDT 
Anton,
 Can you verify that the bug has been resolved ?
Comment 12 Anton Leherbauer CLA 2014-10-29 10:33:50 EDT 
(In reply to Sarika Sinha from comment #11)
> Anton,
>  Can you verify that the bug has been resolved ?

I'll try to verify it by the end of this week.
Comment 13 Anton Leherbauer CLA 2014-10-30 10:37:38 EDT 
Verified with I20141028-2000. Thanks!