Comment # 4 on bug 1204790 from Ludwig Nussel

(In reply to Richard Biener from comment #3)
> (In reply to Ludwig Nussel from comment #2)
> > omg, yes I think we should get rid of that hack.
> > 
> > glibc-devel-32bit already requires glibc-32bit so I think the first
> > substitute is line is not needed at all.
> > 
> > So can we introduce eg %with_biarch or something like that and have the
> > basic logic in the gcc spec file itself?
> 
> Sure, but then the glibc-32bit "hack" is/was used from other packages as
> well.

Didn't encounter any others in ring0 at least.

> Somebody needs to document the canonical way to query enabled multilibs from
> the project config (and make that materialize in all supported code streams!)

So far I only found gcc that cares about multilib. Most other packages just
don't know about the baselibs mechanism. Do you know of any other package that
needs to behave differently when there is biarch support?

Anyway glibc in SLE12 also has correct requires according
SUSE:SLE-12-SP4:Update/glibc/baselibs.conf

So the hack isn't needed there either.