Re: valgrinding with python stuff active

2012-09-10 Thread Stephan Bergmann
On 09/09/2012 12:45 AM, Julien Nabet wrote: So in http://wiki.documentfoundation.org/BugReport#How_to_get_valgrind_log_.28on_Linux.29 page, should I add this: export G_SLICE=always-malloc before this: valgrind --tool=memcheck --num-callers=50 --trace-children=yes ./soffice.bin 2>&1 | tee /tmp/

Re: valgrinding with python stuff active

2012-09-10 Thread Michael Meeks
On Sun, 2012-09-09 at 00:45 +0200, Julien Nabet wrote: > So in > http://wiki.documentfoundation.org/BugReport#How_to_get_valgrind_log_.28on_Linux.29 > > page, should I add this: > > export G_SLICE=always-malloc > > before this: > > valgrind --tool=memcheck --num-callers=50 --trace-children=ye

Re: valgrinding with python stuff active

2012-09-08 Thread Julien Nabet
On 08/09/2012 17:52, Norbert Thiebaud wrote: On Fri, Sep 7, 2012 at 5:49 PM, julien2412 wrote: on this page http://wiki.documentfoundation.org/Development/How_to_debug but what about if custom memory allocator is bugged? (BTW why do we use this? Historical reason?) Wouldn't Valgrind be helpful

Re: valgrinding with python stuff active

2012-09-08 Thread Norbert Thiebaud
On Fri, Sep 7, 2012 at 5:49 PM, julien2412 wrote: > on this page http://wiki.documentfoundation.org/Development/How_to_debug > but what about if custom memory allocator is bugged? (BTW why do we use > this? Historical reason?) Wouldn't Valgrind be helpful to detect it? > > I'm a bit lost here :-(

Re: valgrinding with python stuff active / no more false positives with valgrind 3.7?

2012-09-08 Thread Julien Nabet
On 08/09/2012 11:00, Andrzej J. R. Hunt wrote: Hi everyone, On 09/08/2012 12:49 AM, julien2412 wrote: ... so I get no false positive with Python and/or Java? Since upgrading my system valgrind no longer gives false positives. I'm pretty certain I didn't change any build options when changing

Re: valgrinding with python stuff active / no more false positives with valgrind 3.7?

2012-09-08 Thread Andrzej J. R. Hunt
Hi everyone, On 09/08/2012 12:49 AM, julien2412 wrote: Sorry, I didn't get the point about G_SLICE or the Stephan's commit. For a beginner like me, what must I change/add to this line: valgrind --tool=memcheck --num-callers=50 --trace-children=yes ./soffice.bin 2>&1 | tee /tmp/valgrind.log so I

Re: valgrinding with python stuff active

2012-09-07 Thread julien2412
Sorry, I didn't get the point about G_SLICE or the Stephan's commit. For a beginner like me, what must I change/add to this line: valgrind --tool=memcheck --num-callers=50 --trace-children=yes ./soffice.bin 2>&1 | tee /tmp/valgrind.log so I get no false positive with Python and/or Java? I read th

Re: valgrinding with python stuff active

2012-09-07 Thread Michael Meeks
On Fri, 2012-09-07 at 15:44 +0200, Stephan Bergmann wrote: > On 09/07/2012 03:29 PM, Michael Meeks wrote: > > Now if only we could do something similar for Java :-) > > FYI, works reasonably well for me with > > > "Force JVM into interpreted mode when running under Valgrind, to avoid > f

Re: valgrinding with python stuff active

2012-09-07 Thread Stephan Bergmann
On 09/07/2012 03:29 PM, Michael Meeks wrote: Now if only we could do something similar for Java :-) FYI, works reasonably well for me with "Force JVM into interpreted mode when running under Valgrind, to avoid false error reports" and a valgrind --suppressions= file containing just

Re: valgrinding with python stuff active

2012-09-07 Thread Michael Meeks
On Fri, 2012-09-07 at 13:30 +0100, Caolán McNamara wrote: > FWIW, I patched the *internal* to be valgrind friendly when G_SLICE is > set. So if you have to debug something where python is alive you could > try building against the internal python and it should "just work" with > your usual export