This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: [Gomp] Status of the gomp branch?
- From: Tobias Schlüter <tobias dot schlueter at physik dot uni-muenchen dot de>
- To: Toon Moene <toon at moene dot indiv dot nluug dot nl>,Biagio Lucini <lucini at phys dot ethz dot ch>, gcc at gcc dot gnu dot org,Diego Novillo <dnovillo at redhat dot com>
- Date: Wed, 19 Jan 2005 22:15:31 +0100
- Subject: Re: [Gomp] Status of the gomp branch?
- Reply-to: 41EEC934 dot 4010803 at moene dot indiv dot nluug dot nl
Toon Moene wrote:
> Jacob is also still waiting for my analysis of where to put OpenMP
> directive parsing and processing in the gfortran frontend. Unfortunately, I
> haven't found enough time to even identify the files to look into.
>From scanner.c:
/* Skip a comment. When we come here the parse pointer is positioned
immediately after the comment character. If we ever implement
compiler directives withing comments, here is where we parse the
directive. */
static void
skip_comment_line (void)
{
If the directive can actually be parsed there, it would somehow have to be
linked to the instruction stream. Possible ways of solving this seems to be
adding the information to new_st, or maybe a new global veriable which is
dealt with in accept_statement or add_statement.
I hope this can simplify your analysis,
- Tobi