On Wed, Feb 3, 2010 at 3:07 PM, Kirk, Benjamin (JSC-EG311)
<[email protected]> wrote:
>> Surprised me too, actually.  I think it would be better to turn off
>> counting entirely when asked via configure (especially since it
>> requires an atomic unsigned int when threading is on), do the "one
>> total count" stuff in opt mode with counting on, then do the detailed
>> stuff in dbg/devel mode with counting on.
>
> My thought was that the global ++/-- reference count was a cheap, good thing
> to check for leaks at the grossest level.  Not sure how many threads you are
> on, or the actual overhead of the atomic operation...
>
> Then again, you're the guy who has advocated leaking memory on this list
> before, perhaps the reference counter is right? ;-)

The RC class is like The Tell-Tale Heart, you think you've killed it
but it just keeps reminding you of your sins ;-)

-- 
John

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
_______________________________________________
Libmesh-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-devel

Reply via email to