[CIG-SHORT] Trouble Updating PyLith

Brad Aagaard baagaard at usgs.gov
Sat Sep 11 08:52:46 PDT 2010


Jeff-

Some of the arguments for configuring petsc-dev have changed recently 
(they don't shy away from making changes).

You need to configure PETSc (petsc-dev) with

--with-shared-libraries=1 --with-dynamic-loading=1

rather than with

--with-shared=1 --with-dynamic=1

Update PETSc, reconfigure, and rebuild, and then update PyLith, 
reconfigure, and rebuild.

Brad


On 9/10/10 11:07 AM, Jeffrey Thompson wrote:
> Hi,
>
> I'm trying to update PyLith and I'm running into trouble successfully
> making it.  I get the following error:
>
> /usr/bin/ld:
> /home/jeffremt/tools/petsc-dev/linux_gcc-4.1.1_64_opt/lib/libpetsc.a(filev.o):
> relocation R_X86_64_32S against `a local symbol' can not be used when
> making a shared object; recompile with -fPIC
> /home/jeffremt/tools/petsc-dev/linux_gcc-4.1.1_64_opt/lib/libpetsc.a:
> could not read symbols: Bad value
> collect2: ld returned 1 exit status
> make[3]: *** [libpylith.la <http://libpylith.la>] Error 1
> make[3]: Leaving directory `/home/jeffremt/tools/cig/pylith/libsrc'
> make[2]: *** [all-recursive] Error 1
> make[2]: Leaving directory `/home/jeffremt/tools/cig/pylith/libsrc'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/home/jeffremt/tools/cig/pylith'
> make: *** [all] Error 2
>
> I'm confused about the suggestion to "recompile with -fPIC."  Is this
> saying I should recompile PyLith or PETSc using this option?  I've
> attached my PETSc and PyLith configuration files, as well as the
> config.log and configure.log from PyLith to this email in case these are
> helpful.
>
> Thanks,
> Jeff Thompson
> California Institute of Technology
>
> PS, I've got PETSc version 3.1.0.4, Spatialdata version 0.6.0, and the
> most recent (as of this email) pull of PyLith (revision 17183).  All the
> other needed programs are at the recommended version or more recent.
>
>
>
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short



More information about the CIG-SHORT mailing list