RE: [Gomp-discuss] Organization of libgomp source code

[Scott Robert Ladd]

>   Also are we not rushing a bit on thee implementation side ? I thought
> we should try to find a minimal subset ? Or shall we implement straight
> off against the threading lib ? ( POSIX ).

My idea was to implement the well-defined OpenMP functions, given that the
Standards provide "minimal" examples.

1) Some of the Standard function can be implemented and tested without
making any changes to the gcc tree. We can figure out how to determine the
number of processors, for example, in implementing omp_get_num_procs.

2) By implementing these standard functions, we learn things useful in
parsing the pragmas, such as determining the number of processors (again).

3) Given the simplicity of the standard functions, this gives us a chance to
iron out proceedures, cvs issues, and the like, before we get into modifyign
existing gcc code.

4) I figured we could "kick start" the process if I threw out some code. ;)

I am, of course, open to suggestions as to how we proceed.

..Scott