[asterisk-dev] [Code Review] 3496: testsuite: add valgrind support

[Scott Griepentrog]

> On June 5, 2014, 3:57 p.m., Corey Farrell wrote:
> > /asterisk/trunk/runtests.py, lines 98-99
> > <https://reviewboard.asterisk.org/r/3496/diff/1/?file=58127#file58127line98>
> >
> >     I think a message could be good even when the test was already failed.  Leaks that happen due to error's are just as important as leaks on success.

My intention here was to indicate that the test itself was marked failure *only* because of the valgrind results -- but I see your point and have adjusted the message to be present in both cases with differentiation.


> On June 5, 2014, 3:57 p.m., Corey Farrell wrote:
> > /asterisk/trunk/tests/rest_api/tests.yaml, line 16
> > <https://reviewboard.asterisk.org/r/3496/diff/1/?file=58128#file58128line16>
> >
> >     Unrelated change?

Yes, created the diff in a directory I was doing ARI testing with valgrind in.

BTW, thanks for the review!


> On June 5, 2014, 3:57 p.m., Corey Farrell wrote:
> > /asterisk/trunk/lib/python/asterisk/valgrind.py, line 143
> > <https://reviewboard.asterisk.org/r/3496/diff/1/?file=58125#file58125line143>
> >
> >     As someone mentioned in your first review for valgrind support, XSLT would be better.  I had started writing an XSLT file to convert the results to test output before, I'll have to take another look at it.  If you want to open a JIRA ticket for this I can upload what I have so far.

I'm skeptical of being able to reproduce the fault stack compression as I've done it by using XSLT instead -- but I'll be happy to be wrong, and happier still to incorporate your improvement at any time.  There isn't even a JIRA issue for this work (totally back burner), you can just send it to me. 


- Scott


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3496/#review12061
-----------------------------------------------------------


On June 6, 2014, 9:06 a.m., Scott Griepentrog wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/3496/
> -----------------------------------------------------------
> 
> (Updated June 6, 2014, 9:06 a.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Repository: testsuite
> 
> 
> Description
> -------
> 
> This patch adds support for running Asterisk under Valgrind (said "Val-Grinned") to check for memory allocation and reference problems.
> 
> To use, add -V to ./runtests.py and get a cup of coffee.  Any errors will be displayed in a stack trace format.
> 
> Tests can also activate valgrind in the yaml configuration.
> 
> By default, only confirmed leaks will be shown - thoses where all copies of the pointer to the allocation have been released.  Additional leak checking can be enabled.
> 
> 
> Diffs
> -----
> 
>   /asterisk/trunk/runtests.py 5100 
>   /asterisk/trunk/lib/python/asterisk/valgrind.py PRE-CREATION 
>   /asterisk/trunk/lib/python/asterisk/test_config.py 5100 
>   /asterisk/trunk/lib/python/asterisk/test_case.py 5100 
>   /asterisk/trunk/lib/python/asterisk/asterisk.py 5100 
>   /asterisk/trunk/configs/valgrind.supp PRE-CREATION 
> 
> Diff: https://reviewboard.asterisk.org/r/3496/diff/
> 
> 
> Testing
> -------
> 
> Used to locate reference problems on Userevent issue, and incorporated several improvements over the prior attempt.
> 
> 
> Thanks,
> 
> Scott Griepentrog
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140606/68f32494/attachment.html>