[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [cdt-dev] NPE exception when debugging LibreOffice using 8.1.0 DSF for long period of time
|
> -----Original Message-----
> From: cdt-dev-bounces@xxxxxxxxxxx
> [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Jeff Johnston
> Sent: Tuesday, January 08, 2013 2:49 PM
> To: CDT General developers list.
> Subject: [cdt-dev] NPE exception when debugging LibreOffice
> using 8.1.0 DSF for long period of time
>
> A bug has been reported to Fedora eclipse-cdt regarding debugging
> LibreOffice using
>
> CDT 8.1.0.
> gdb 7.4.50
> eclipse 4.2.1
>
> The following occurs after about 2 hours of debugging:
>
> --------------------------------------------------------------
> -------------------------------
>
> Description of problem:
>
> when using the Eclipse CDT with GDB to debug an overly large
> C++ application (LibreOffice) occasionally the following exception
> is printed:
>
> Exception in thread "MI TX Thread" java.lang.NullPointerException
> at
> org.eclipse.cdt.dsf.mi.service.command.commands.MICommand$MISt
andardParameterAdjustable.getAdjustedValue(MICommand.java:331)
> at
> org.eclipse.cdt.dsf.mi.service.command.commands.MICommand.para
> metersToString(MICommand.java:239)
> at
> org.eclipse.cdt.dsf.mi.service.command.commands.MICommand.cons
> tructCommand(MICommand.java:172)
> at
> org.eclipse.cdt.dsf.mi.service.command.AbstractMIControl$TxThr
> ead.run(AbstractMIControl.java:650)
>
> when this happens Eclipse CDT somehow gets confused and disables
> all of the debugging UI; only thing that still works then
> is the "Terminate" button to detach from the inferior.
>
> in case it matters, the inferior has some custom GDB python
> helpers to print various data types that are evidently used
> to display stuff in Eclipse CDT as well.
>
> --------------------------------------------------------------
> -----------------------------------
>
> Has anyone on the debugger team seen this occur before? Any ideas on
> how to get around the problem or if there is an issue in DSF
> that needs
> addressing or has already been fixed?
Nothing jumps out at me.
However, that code is called from a _lot_ of places, so there
may be a weird situation that passes a null by mistake.
It probably won't help much but you could ask the developer
to get the tail end (50 lines) of the gdb console longs
right after the problem happens:
http://wiki.eclipse.org/CDT/User/FAQ#I.27ve_been_asked_for_.27gdb_traces.27.2C_where_can_I_find_them.3F
Note also that GDB 7.4.50 is not an official version, although
I don't expect it would explain the problem.
Marc
>
> Unfortunately, the developer has admitted he does not have a simple
> reproducible scenario and that it occurs after a long period of time.
>
> -- Jeff J.
> _______________________________________________
> cdt-dev mailing list
> cdt-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>
