[vtk-developers] Addition flag for valgrind

[David E DeMarle]

But valgrind is getting slow. I may have to remove that flag from karego-at
valgrind.

--track-origins slows the tests down significantly - as evidenced by the
new >500s timeout test failures on the valgrind results (
http://open.cdash.org/viewDynamicAnalysis.php?buildid=2836814). There is
some free time available overnight on that machine, so I'm checking how
much extra time the failing tests require. I had kind of wanted that time
though to turn on more tests...

What is our preference:
1) keep the flag, increase the timeout, and continue to test what we do
2) disable just these slow tests on that machine and possibly turn on other
tests (MPI, Qt, Tcl, Python)
3) remove the flag and turn on more tests

I'm leaning toward 1. When we get and setup a newer test machine it should
be less of an issue and we can have out cake and eat it too.


David E DeMarle
Kitware, Inc.
R&D Engineer
21 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-881-4909


On Tue, Mar 5, 2013 at 11:46 AM, Bill Lorensen <bill.lorensen at gmail.com>wrote:

> These kids are getting quick...
>
>
> On Tue, Mar 5, 2013 at 11:25 AM, David E DeMarle <dave.demarle at kitware.com
> > wrote:
>
>> Kyle beat you to the suggestion. I added it to karego-at this morning and
>> we should see it in tonight's dashboard.
>>
>>
>> David E DeMarle
>> Kitware, Inc.
>> R&D Engineer
>> 21 Corporate Drive
>> Clifton Park, NY 12065-8662
>> Phone: 518-881-4909
>>
>>
>> On Tue, Mar 5, 2013 at 11:22 AM, Bill Lorensen <bill.lorensen at gmail.com>wrote:
>>
>>> Foks,
>>>
>>> I suggest we add the flag:
>>> --track-origins=yes
>>> to valgrind.
>>> "Controls whether Memcheck tracks the origin of uninitialised values. By
>>> default, it does not, which means that although it can tell you that an
>>> uninitialised value is being used in a dangerous way, it cannot tell you
>>> where the uninitialised value came from. This often makes it difficult to
>>> track down the root problem. "
>>> http://valgrind.org/docs/manual/mc-manual.html
>>>
>>> For example, it helped me find this error:
>>> http://review.source.kitware.com/#/t/2301/
>>>
>>> which would have been hard to find without it.
>>>
>>> Bill
>>>
>>> --
>>> Unpaid intern in BillsBasement at noware dot com
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.vtk.org/mailman/listinfo/vtk-developers
>>>
>>>
>>>
>>
>
>
> --
> Unpaid intern in BillsBasement at noware dot com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtk-developers/attachments/20130306/30d2d0ef/attachment.html>