OpenMPI, Emacs and libotf problems

Mátyás Selmeci matyas at cs.wisc.edu
Wed Jul 6 20:24:08 UTC 2011


Adam Williamson wrote on Wed, Jul 06, 2011 at 12:54:48PM -0700:
> On Wed, 2011-07-06 at 22:46 +0300, Jussi Lehtola wrote:
> > On Wed, 06 Jul 2011 12:39:14 -0700
> > Adam Williamson <awilliam at redhat.com> wrote:
> > > On Wed, 2011-07-06 at 20:03 +1000, Amit Saha wrote:
> > > > So, this library is missing. However, this should have been
> > > > installed as its a dependency, right?
> > > > 
> > > > It can be seen that there are two providers listed for libotf.so.0.
> > > > Since 'openmpi' was already installed, so it didn't bother
> > > > installing 'libotf'. I could simulate the scenario on a Fedora 15
> > > > installation:
> > > 
> > > The problem is that openmpi includes a libotf.so.0, but it's not in a
> > > path that the system will usually look in for shared libraries, it's
> > > in (libdir)/openmpi/lib/ . openmpi probably shouldn't have a private
> > > copy of libotf in the first place (assuming that's what it is, and
> > > not just a naming coincidence), but if it's going to have one, it
> > > should have a line in the spec to prevent RPM auto-provides from
> > > giving the openmpi package a Provides for libotf.so.0.
> > 
> > It's certainly not the same libotf, since OpenMPI does not have
> > *anything* to do with truetype fonts.
> > 
> > Even though the library is installed in a non-system directory,
> > applications that link against libotf will get an automatically
> > generated Requires: against it anyways.
> 
> Well, packages get an auto-generated Requires: for libotf.so.0. Anything
> that claims to provide libotf.so.0 will satisfy this. The most correct
> solution is simply for openmpi to stop claiming to provide libotf.so.0
> because, for practical purposes, it does not provide it: even if the
> library in question were the same one, openmpi's copy is not in a
> location that other packages will know how to use, so in practical
> terms, it does not provide the library.

If I run into this kind of a situation (a false auto-generated
Provides), is it possible to disable just that one entry, or do I have
to disable AutoProv entirely and keep track of the libs myself?

Thanks,
-matyas



More information about the devel mailing list