On Nov 7, 2012, at 5:31 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> If BLAS is a (direct) dependency, it should be listed in the file. Regardless
> of depth, pkgconfig --libs --static thepackage should show it.
>
> Multiple compilers only matters insofar as they have a different ABI.
> Different MPIs have different ABIs,
Right, but since we are using MPI we live in the world where each compiler
has a different ABI which pkg-config doesn't live in. That is why I say it is
not a great model for us.
> and that is my primary complaint about pkgconfig. They don't seem to think
> this is very important (or it's too complicated to get right across all
> systems so they don't try), but the natural thing is to keep different
> pkgconfig databases under different PKG_CONFIG_PATH.
>
>
> On Wed, Nov 7, 2012 at 5:25 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
>
> Ok, pkg-config seems really really limited anyways, it basically lists the
> libraries and include paths so would be useful if packages provided it but
> there is no place for information about compilers/mpi/blas/etc used to build
> the libraries so is just not very informative for us.
>
> Barry
>
> On Nov 7, 2012, at 4:46 PM, Matthew Knepley <knepley at gmail.com> wrote:
>
> > On Wed, Nov 7, 2012 at 5:24 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
> >>
> >> In BuildSystem I find
> >>
> >> if hasattr(self, 'usePkgConfig') and 'with-'+self.package+'-pkg-config'
> >> in self.framework.argDB:
> >> self.framework.argDB['with-'+self.package] = 1
> >> self.usePkgConfig()
> >>
> >> and
> >>
> >> if hasattr(self, 'usePkgConfig'):
> >> help.addArgument(self.PACKAGE,
> >> '-with-'+self.package+'-pkg-config=<dir>', nargs.ArgDir(None, None,
> >> 'Indicate the root directory of the '+self.name+' installation'))
> >>
> >> not signs of any implemented usePkgConfig()
> >>
> >> Am I missing something or are these just plans that never came to
> >> fruition?
> >
> > I think I implemented more on my machine for FEniCS SCons but decided
> > it was not the right thing. So yes, you are right.
> >
> > Matt
> >
> >> Thanks
> >>
> >> Barry
> >>
> >
> >
> >
> > --
> > What most experimenters take for granted before they begin their
> > experiments is infinitely more interesting than any results to which
> > their experiments lead.
> > -- Norbert Wiener
>
>
