[Omp] A memory model question

[Michael Suess]

On Friday 14 July 2006 02:15, Georg Hager wrote:
> Hi,
>
> > Yes, this code is broken.
>
> A thorough examination of the failure of the double-checked
> locking idiom for C++ can be found in this article by
> Scott Meyers and Andrei Alexandrescu:
>
> http://www.aristeia.com/Papers/DDJ_Jul_Aug_2004_revised.pdf

Hi Georg,

thanks for pointing me into this direction, I did not know about the article 
and had assumed the failure of the double-checked locking pattern was fixable 
for OpenMP / C++. Should have done my homework better ;-).

Thanks again, Michael