| Summary: | Detaching from a process needs to stop a running program | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Tools] CDT | Reporter: | Marc Khouzam <marc.khouzam> | ||||
| Component: | cdt-debug-dsf-gdb | Assignee: | Marc Khouzam <marc.khouzam> | ||||
| Status: | RESOLVED FIXED | QA Contact: | Marc Khouzam <marc.khouzam> | ||||
| Severity: | normal | ||||||
| Priority: | P3 | CC: | cdtdoug, ling.5.wang, overholt, pawel.1.piech | ||||
| Version: | 7.0 | Flags: | marc.khouzam:
review?
(ling.5.wang) |
||||
| Target Milestone: | 8.0 | ||||||
| Hardware: | PC | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Attachments: |
|
||||||
|
Description
Marc Khouzam
Thanks for catching this and knowing where the bug was, Mark :) Created attachment 192430 [details]
Fix
This fix interrupts the target before doing a detach from the process when in all-stop mode.
This will not work properly for multi-process because we don't restart the target after we have detached from the one process. But all-stop does not work properly with multi-process yet anyway. I will open a separate bug about it.
Committed to HEAD.
Fixed. Ling, can you review? It is an extension of your solution of interrupting the target when doing a terminate. I have opened Bug 341735 to deal with the multi-process issue. *** cdt cvs genie on behalf of mkhouzam *** Bug 340601 - Detaching from a process needs to stop a running program [*] GDBProcesses_7_2.java 1.10 http://dev.eclipse.org/viewcvs/index.cgi/org.eclipse.cdt/dsf-gdb/org.eclipse.cdt.dsf.gdb/src/org/eclipse/cdt/dsf/gdb/service/GDBProcesses_7_2.java?root=Tools_Project&r1=1.9&r2=1.10 [*] GDBProcesses_7_0.java 1.46 http://dev.eclipse.org/viewcvs/index.cgi/org.eclipse.cdt/dsf-gdb/org.eclipse.cdt.dsf.gdb/src/org/eclipse/cdt/dsf/gdb/service/GDBProcesses_7_0.java?root=Tools_Project&r1=1.45&r2=1.46 [*] MIProcesses.java 1.14 http://dev.eclipse.org/viewcvs/index.cgi/org.eclipse.cdt/dsf-gdb/org.eclipse.cdt.dsf.gdb/src/org/eclipse/cdt/dsf/mi/service/MIProcesses.java?root=Tools_Project&r1=1.13&r2=1.14 (In reply to comment #3) > Fixed. > > Ling, can you review? It is an extension of your solution of interrupting the > target when doing a terminate. Marc, I'm very sorry this fell through cracks as I've been busy with other stuff. One issue I see in your fix is it does not take into account possible failure of the interrupt. Maybe we could use backend.interruptAndWait() with proper timeout, and report failure through the request monitor so that user can also get a feedback that Disconnect fails. |