This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: OpenMP and its testing
- From: Biagio Lucini <lucini at phys dot ethz dot ch>
- To: gcc at gcc dot gnu dot org
- Cc: Nick Maclaren <nmm1 at cus dot cam dot ac dot uk>
- Date: Wed, 10 Nov 2004 15:30:27 +0100
- Subject: Re: OpenMP and its testing
- References: <E1CRrNc-00076A-Go@libra.cus.cam.ac.uk>
On Wednesday 10 November 2004 13.15, Nick Maclaren wrote:
> Frankly, I think that this is asking for trouble. We are a parallel
> shop, and my experience is that most C OpenMP implementations are
> very sick - my experience is that only half of the major vendors'
> implementations work at all. The reasons should be well-known, but
> please ask me if anyone wants to know more.
>
I would be surely interested in knowing the details. I have experience with
the Intel compiler (whose OpenMP implementation I consider broken or at least
unreliable) and with the Portland group compiler (which for me gives a good
implementation). You seem to know also why that happens. Could you please
provide some more information? This might help our effort.
Thanks
Biagio
--
=========================================================
Biagio Lucini
Institut Fuer Theoretische Physik
ETH Hoenggerberg
CH-8093 Zuerich - Switzerland
Tel. +41 (0)1 6332562
=========================================================