Hi Ben,
   Will be happy to do so. At this point I¹m developing a library of
linear solvers, with various options for test systems and other things. So
would need to pare this down significantly to give a simple test. I¹ll
keep that in mind.
Richard
--
Richard Barrett
PO Box 5800, MS-0621
Sandia National Laboratories
Albuquerque, NM 87185


Phone: 505-845-7655
Pager: 505-951-8087

www.sandia.gov/~rfbarre









On 1/4/16, 6:50 PM, "Ben Albrecht" <[email protected]> wrote:

>Hello Richard,
>
>After some manual testing with the RANSAC application from the Chapel By
>Example guide, it appears this bug has been fixed in the master branch of
>Chapel. If you have a small test-case highlighting the error, it would be
>valuable to add into our test suite.
>
>Best,
>Ben
>
>On 1/4/16, 11:14 AM, "Barrett, Richard F" <[email protected]> wrote:
>
>>Thanks, Brad, this works.
>>Richard
>>-- 
>>Richard Barrett
>>PO Box 5800, MS-0621
>>Sandia National Laboratories
>>Albuquerque, NM 87185
>>
>>Phone: 505-845-7655
>>Pager: 505-951-8087
>>
>>www.sandia.gov/~rfbarre
>>
>>
>>On 1/4/16, 12:11 PM, "Brad Chamberlain" <[email protected]> wrote:
>>
>>>
>>>Hi Richard --
>>>
>>>This is definitely a bug on our part and one that current work is
>>>striving 
>>>to address.
>>>
>>>If your program only uses a few arrays that live for the duration of the
>>>program (i.e., it'd be OK to leak them), a _potential_ workaround might
>>>be 
>>>to compile with -snoRefCount=true.  This is a big hammer that turns off
>>>reference counting for all arrays as a short-term workaround for some
>>>reference counting-related issues.  It wasn't designed to help with this
>>>case, but seems like it might.
>>>
>>>-Brad
>>>
>>>
>>>On Mon, 4 Jan 2016, Barrett, Richard F wrote:
>>>
>>>> Folks,
>>>>
>>>> I've been wrestling with an odd problem,
>>>>
>>>> error: halt reached - array reference count is negative!
>>>>
>>>> then found what appears to be the issue:
>>>> http://www.primordand.com/chapel_by_ex_practical.html Is there new
>>>> information regarding this issue?
>>>>
>>>> Btw, this on my Mac laptop, OS 10.9.5, Chapel v1.12.0.
>>>>
>>>> Thanks,
>>>> Richard
>>>> --
>>>> Richard Barrett
>>>> PO Box 5800, MS-0621
>>>> Sandia National Laboratories
>>>> Albuquerque, NM 87185
>>>>
>>>> Phone: 505-845-7655
>>>> Pager: 505-951-8087
>>>>
>>>> www.sandia.gov/~rfbarre
>>>>
>>
>>
>>-------------------------------------------------------------------------
>>-
>>----
>>_______________________________________________
>>Chapel-bugs mailing list
>>[email protected]
>>https://lists.sourceforge.net/lists/listinfo/chapel-bugs
>


------------------------------------------------------------------------------
_______________________________________________
Chapel-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/chapel-bugs

Reply via email to