[Insight-developers] Valgrind is back

[Bradley Lowekamp]

Doh,

  this does not make very much since:

for(i=0; i<nWrds; i++)
  {
    delete [] wrds[i++];
  }

I care not to share how long it took me to figure out what is wrong  
with that code.

Brad


On Jul 14, 2009, at 4:25 PM, Bill Lorensen wrote:

> Still some issues...
>
> http://www.cdash.org/CDash/viewDynamicAnalysisFile.php?id=377269
>
>
> On Mon, Jul 13, 2009 at 11:39 AM, Julien
> Jomier<julien.jomier at kitware.com> wrote:
>> Brad,
>>
>> I have committed the fix. Thanks for looking into it.
>>
>> -Julien
>>
>> Bradley Lowekamp wrote:
>>>
>>> Hello,
>>>
>>> I have attached a patch which I believe fixes this leak. I am  
>>> still in the
>>> process of an Experimental build and test. The freeing of  
>>> variables did not
>>> match the allocation in MET_StringToWordArray. I check it's usage  
>>> in other
>>> files and it was performed correctly in the rest of the library.
>>>
>>> So why was this not detected before? This method is used with  
>>> "LIST" and
>>> "%" in files. It appears that this is the only test which utilizes  
>>> this
>>> option that metaio image  file can have. As I reported last week  
>>> this test
>>> also does not produce the correct results on all systems (This is an
>>> original streaming IO test, before I improved the ImageFileWriter,  
>>> and this
>>> was it's state before as well).  So again, I am just confused as  
>>> to why
>>> valgrind did not show this leak before. It is also clean to me  
>>> that the
>>>  "LIST" and "%" options should likely be tested to see if they are
>>> functional with streaming.
>>>
>>> Brad
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>>
>>> On Jul 13, 2009, at 8:25 AM, Lowekamp, Bradley (NIH/NLM/LHC) [C]  
>>> wrote:
>>>
>>>> I think I see where the leak is. I wonder why it was not detected
>>>>  before, it looks like it has been there for a while.
>>>>
>>>> Brad
>>>>
>>>>
>>>> On Jul 10, 2009, at 7:23 PM, Bill Lorensen wrote:
>>>>
>>>>> After a hiatus of a couple of weeks, valgrind is back and  
>>>>> reporting a
>>>>> Meta issue:
>>>>>
>>>>> http://www.cdash.org/CDash/viewDynamicAnalysisFile.php?id=371803
>>>>>
>>>>> Bill
>>>>> \
>>>>
>>>> _______________________________________________
>>>> Powered by www.kitware.com <http://www.kitware.com>
>>>>
>>>> Visit other Kitware open-source projects at
>>>> http://www.kitware.com/opensource/opensource.html
>>>>
>>>> Please keep messages on-topic and check the ITK FAQ at:
>>>> http://www.itk.org/Wiki/ITK_FAQ
>>>>
>>>> Follow this link to subscribe/unsubscribe:
>>>> http://www.itk.org/mailman/listinfo/insight-developers
>>>
>>> ========================================================
>>>
>>> Bradley Lowekamp
>>> Lockheed Martin Contractor for
>>>
>>> Office of High Performance Computing and Communications
>>>
>>> National Library of Medicine
>>> blowekamp at mail.nih.gov <mailto:blowekamp at mail.nih.gov>
>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Please keep messages on-topic and check the ITK FAQ at:
>>> http://www.itk.org/Wiki/ITK_FAQ
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.itk.org/mailman/listinfo/insight-developers
>>
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ITK FAQ at:
>> http://www.itk.org/Wiki/ITK_FAQ
>>
>> Follow this link to subscribe/unsubscribe:
>> http://www.itk.org/mailman/listinfo/insight-developers
>>