Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 340261

Summary:	[debug view] Label of cores does not get refreshed properly
Product:	[Tools] CDT	Reporter:	Marc Khouzam <marc.khouzam>
Component:	cdt-debug-dsf-gdb	Assignee:	Project Inbox <cdt-debug-dsf-gdb-inbox>
Status:	NEW ---	QA Contact:	Jonah Graham <jonah>
Severity:	normal
Priority:	P3	CC:	cdtdoug, pawel.1.piech
Version:	7.0.2
Target Milestone:	---
Hardware:	PC
OS:	Linux
Whiteboard:

Description Marc Khouzam

2011-03-16 21:12:52 EDT

I was running a process with 5 threads in non-stop mode.  I noticed that when some threads where stopping or starting, the label '[core: 0]' on the thread was not being updated properly.  Pressing the refresh button would cause the labels to change.

I'm not sure if the core label on the process node was being updated properly or not.  What I did see what that process said [cores 0,1] while all threads where showing [core 0].  When pressing refresh, some threads would update their label and show [core: 1].

Comment 1 Marc Khouzam

2011-05-18 14:20:11 EDT

This could be the same issue as bug 337899, where when a thread resumes, the process node is not updated.  Therefore if the process node was showing [cores 0,1] maybe it was old data, but when I pressed refresh, a thread had moved back to core 1, and it made it look like it was the threads that had it wrong, while it actually was a process node problem.

This is just a theory.

Comment 2 Marc Khouzam

2012-10-16 06:09:31 EDT

(In reply to comment #1)
> This could be the same issue as bug 337899, where when a thread resumes, the
> process node is not updated.  Therefore if the process node was showing
> [cores 0,1] maybe it was old data, but when I pressed refresh, a thread had
> moved back to core 1, and it made it look like it was the threads that had
> it wrong, while it actually was a process node problem.
> 
> This is just a theory.

This is not the case.

I saw this again today while also using the Multicore Visualizer.  The visualizer was showing a thread on core one but the debug view showed [core 0] for all threads, although the process node showed [core 0, 1].  After pressing refresh, one of the threads did show on [core 1] as the visualizer had indicated.  An important thing to note is that the thread affected was running at the time.  I believe the refresh may not be perfect in that case.