Re: math/fftw-long / math/fftw-quad broken on Linux as it pulls in math/libquadmath, which is superfluous and broken on Linux

[nia <nia%NetBSD.org@localhost>]

On Tue, Jun 22, 2021 at 03:09:36PM +0200, Dr. Thomas Orgis wrote:
> Am Tue, 22 Jun 2021 12:10:48 +0000
> schrieb nia <nia%NetBSD.org@localhost>: 
> 
> > Yes - libquamath from gcc 11 with gcc10 and gcc7 from NetBSD.
> 
> Not sure I like that. Doesn't matter in practice as this lib doesn't
> seem to change much, but I tend to follow the bundle GCC delivers and
> build this stuff together as toolchain.

We're already doing it for e.g. libatomic. It's pretty necessary
because the alternative is shipping 10 different versions in pkgsrc.
This is all over pkgsrc and not going away.

> I will patch out libquadmath usage locally anyway now. Preferrable
> would be a builtin check. A compile check for -lquadmath working?

It is extremely likely that you can get libquadmath to package with
the correct library installation prefix by playing with the configure
options. 

> > > The OpenMP and MPI options result in extra library files being built.
> > > And extra dependencies — that's why you don't want them enabled by
> > > default, and hence as separate packages. Speaking of which: The OpenMP
> > > one might miss the libomp dependency for the Darwin toolchain.
> 
> What to do about that? Rather split out MPI and OpenMP as separate
> packages, too? We don't ship the options at all right now. Would be
> stupid to let them slip through now just to remove them for split
> packages in the next release.

The FreeBSD package for example has OpenMP disabled by default.
I think we're unlikely to run into trouble.