This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: patch for merging graphite branch (before tuplification)
- From: "Joseph S. Myers" <joseph at codesourcery dot com>
- To: Sebastian Pop <sebpop at gmail dot com>
- Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>, Richard Guenther <richard dot guenther at gmail dot com>, Mark Mitchell <mark at codesourcery dot com>, Jakub Jelinek <jakub at redhat dot com>, David Edelsohn <edelsohn at gmail dot com>, "Harle, Christophe" <christophe dot harle at amd dot com>, Tobias Grosser <grosser at fim dot uni-passau dot de>, Konrad Trifunovic <konrad dot trifunovic at gmail dot com>, Albert Cohen <Albert dot Cohen at inria dot fr>
- Date: Sun, 3 Aug 2008 23:06:48 +0000 (UTC)
- Subject: Re: patch for merging graphite branch (before tuplification)
- References: <cb9d34b20807251914jb7fb76q4452be18461d7464@mail.gmail.com> <cb9d34b20807251919x59b6bca2vf171fb2304c602cc@mail.gmail.com> <Pine.LNX.4.64.0807261046270.16205@digraph.polyomino.org.uk> <cb9d34b20808031527x107314e9h50f83ef5ae76308b@mail.gmail.com>
On Sun, 3 Aug 2008, Sebastian Pop wrote:
> On Sat, Jul 26, 2008 at 6:11 AM, Joseph S. Myers
> <joseph@codesourcery.com> wrote:
> > Two new .exp files added in the patch should be GPL v3 or later with the
> > current GPLv3 or later notice used in GCC, not GPL v2 or later.
> >
> > The new libraries required, and the associated configure options, need to
> > be documented in install.texi.
> >
> > I see you have
> >
> > +#ifdef HAVE_cloog
> > + graphite_transform_loops ();
> > +#endif
> >
> > which would make the code generated depend on the libraries detected at
> > configure time. This is a bad idea. You should do one of two things:
> >
> > * Always require the library and give an error at configure time if not
> > detected.
> >
> > * Give an error if any command-line option that would require the library
> > is used but the library was not linked in. In this case the testcases
> > need to be appropriately conditioned, if they aren't already.
> >
>
> This second option is implemented in the attached patch that I
> committed to the graphite branch.
Where is the error in this patch? By an error, I mean the compiler prints
a message to stderr and exits with nonzero exit status if any unsupported
option (-fgraphite or one of the options implying it) is passed; not that
it puts something in a dump file.
--
Joseph S. Myers
joseph@codesourcery.com