An OpenMP imprementation for gcc - Summary
This group is not part of the GNU Project.
Number crunching is an area that is attracting increasing interest coming not only from traditional sectors (e.g. the Accademic world), but also from industry. Many numerical computations exploit multiple CPU systems. Among those, SMP (Symmetric Multi Processor) machines are a cheap and effective way of achieving high computational speed.
However in order to exploit the full potential of SMP machines the code needs to be written specifically for a multi-CPU system. This might be a time consumming task that can put off people from parallel programming.OpenMP offers a simple way of exploiting parallelism without interfering too much with the design of the code: the directive-based parallelism method it offers makes it relatively easy to convert an existing serial code to an efficient parallel code.
While being an Open Standard, there are not many open (in the sense of the Free Software) OpenMP implementations (one noticeable exceptions being OdinMP, see http://odinmp.imit.kth.se).The aim of this project is to produce an implementations tightly integrated with the gcc compiler collection that will possibly be officially shipped in the future with gcc itself.
Registration Date: Mon 16 Dec 2002 11:56:49 AM UTC
License: GNU General Public License v2 or later
Development Status: 1 - Planning
- Browse open items
Submit a new item
- Browse open items
Submit a new item
Powered by Savane 3.14-8aba.
Corresponding source code