Comment # 14 on bug 1141105 from Martin Li��ka

(In reply to J��rg Schiling from comment #10)
> (In reply to Martin Li��ka from comment #9)
> 
> > That is still under investigation and since we don't have a reasonable
> > test-case, we can't prove that.
> 
> Well you have a test case. mkisofs can be used as a test case, but I guess
> that any three liner that calls error() is sufficient. You already have a
> 100% prove for my statements in the two bug reports with the same signature
> of problems.

I'm asking you here for a reasonable small test-case. To be honest, the build
system
of schily is complicated and I one can't report such a huge project as a
reproducer.
That's why I asked you for a reproducer.

...

> 
> You are wrong, these functions have different prototypes and mkisofs

>From the linker perspective, the resolution can be fulfilled by any of the
shared libraries. Yes, the order of shared libraries should be probably
followed, but it seems to me that providing a clashing functions by multiple
libraries is unhappy. Maybe your signature is the right from 1980s perspective,
but glibc's is much more standard from today's perspective.