<div>Thanks, Brad and Matt. </div><div><br></div><div>MPICH didn't build correctly either, by simply selecting it. There was a problem that was fixed by changing a variable in the Makefile after configuration:</div><div>
<br></div>line 305: mpich_sharedarch=gcc changeed to mpich_sharedarch=osx-gcc<div><br></div><div>Now, the build is failing during the spatiadata/spatialdb tests:</div><div><br></div><div><div>spatialdata::spatialdb::TestSimpleGridDB::testDataIndex : OK</div>
<div>spatialdata::spatialdb::TestSimpleGridDBLine1D::testQueryNearesttestspatial(91490) malloc: *** error for object 0xd000000000000000: pointer being freed was not allocated</div><div>*** set a breakpoint in malloc_error_break to debug</div>
<div>/bin/sh: line 1: 91490 Abort trap: 6 ${dir}$tst</div><div>FAIL: testspatial</div><div>==================================</div><div>1 of 1 test failed</div><div>Please report to <a href="mailto:baagaard@usgs.gov">baagaard@usgs.gov</a></div>
<div>==================================</div><div>make[8]: *** [check-TESTS] Error 1</div><div>make[7]: *** [check-am] Error 2</div><div>make[6]: *** [check-recursive] Error 1</div><div>make[5]: *** [check-recursive] Error 1</div>
<div>make[4]: *** [check-recursive] Error 1</div><div>make[3]: *** [check-recursive] Error 1</div><div>make[2]: *** [spatialdata] Error 2</div><div>make[1]: *** [installed_spatialdata] Error 2</div><div>make: *** [installed_pylith] Error 2</div>
</div><div><br></div><div>I'm not sure what this might be related to, or what log files might contain relevant info. Please let me know if there's more info I can give.</div><div><br>Thanks,</div><div>Richard</div>
<div><br><br><div class="gmail_quote">On Tue, Jan 29, 2013 at 2:23 PM, Matthew Knepley <span dir="ltr"><<a href="mailto:knepley@mcs.anl.gov" target="_blank">knepley@mcs.anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div>On Tue, Jan 29, 2013 at 2:00 PM, Brad Aagaard <span dir="ltr"><<a href="mailto:baagaard@usgs.gov" target="_blank">baagaard@usgs.gov</a>></span> wrote:<br>
</div><div class="gmail_extra"><div class="gmail_quote"><div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Richard,<br>
<br>
We have not encountered this problem you are describing with regards to<br>
openmpi and darwin. I am still using 10.6 on my Mac; I am not sure if we<br>
have tested building from source under 10.8.<br>
<br>
We didn't setup the pylith-installer configure to have openmpi options<br>
because we didn't anticipate the user needing to customize the openmpi<br>
options. For clusters and other computers where custom MPI settings are<br>
used, we expect a sysadmin to install MPI in a system wide location.<br>
<br>
There are two easy workarounds to this problem:<br>
(1) Install mpi yourself in a custom location that is in the path and<br>
then don't install it using the pylith installer. You can customize the<br>
mpi configure arguments to make them work on your machine.<br>
(2) Manually adjust the openmpi configure options used by the pylith<br>
installer by editing the install_openmpi target in the Makefile.<br></blockquote><div><br></div></div><div>(3) Have it install MPICH instead.</div><div><br></div><div> Thanks,</div><div><br></div><div>
Matt</div><div><div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Regards,<br>
Brad<br>
<br>
<br>
On 1/29/13 7:34 AM, Richard Styron wrote:<br>
> Hi all,<br>
><br>
> I am trying to build Pylith from source on a mac (OS X 10.8) following<br>
> the install instructions for Darwin/CUDA:<br>
><br>
> mkdir -p $HOME/build/pylith<br>
> $HOME/src/pylith/pylith-installer-1.8.0-0/configure \<br>
> --enable-gcc=fortran \<br>
> --enable-python \<br>
> --enable-mpi=openmpi \<br>
> --with-fetch=curl \<br>
> --with-make-threads=2 \<br>
> --prefix=$HOME/pylith \<br>
> --with-cuda (<-- also tried without this. CUDA is installed and<br>
> functional on my machine.)<br>
><br>
><br>
><br>
><br>
> However, OpenMPI fails when during the build:<br>
><br>
> libtool: link: g++ -fopenmp -DVT_OMP -g -O2 -DMPICH_IGNORE_CXX_SEEK<br>
> -fopenmp -o vtfilter vtfilter-vt_filter.o vtfilter-vt_filthandler.o<br>
> vtfilter-vt_otfhandler.o vtfilter-vt_tracefilter.o ../../util/util.o<br>
> -Wl,-bind_at_load<br>
> -L/Users/itchy/src/pylith/openmpi-build/ompi/contrib/vt/vt/extlib/otf/otflib/.libs -L/Users/itchy/pylith/lib -L/Users/itchy/src/pylith/openmpi-build/ompi/contrib/vt/vt/extlib/otf/otflib /Users/itchy/src/pylith/openmpi-build/ompi/contrib/vt/vt/extlib/otf/otflib/.libs/libotf.a -lz -lutil<br>
> Undefined symbols for architecture x86_64:<br>
> "___builtin_expect", referenced from:<br>
> _main.omp_fn.0 in vtfilter-vt_tracefilter.o<br>
> _main.omp_fn.1 in vtfilter-vt_tracefilter.o<br>
> ld: symbol(s) not found for architecture x86_64<br>
> collect2: ld returned 1 exit status<br>
> make[11]: *** [vtfilter] Error 1<br>
> make[10]: *** [all-recursive] Error 1<br>
> make[9]: *** [all-recursive] Error 1<br>
> make[8]: *** [all] Error 2<br>
> make[7]: *** [all-recursive] Error 1<br>
> make[6]: *** [all-recursive] Error 1<br>
> make[5]: *** [all-recursive] Error 1<br>
> make[4]: *** [openmpi] Error 2<br>
> make[3]: *** [installed_mpi] Error 2<br>
> make[2]: *** [installed_hdf5] Error 2<br>
> make[1]: *** [installed_netcdf] Error 2<br>
> make: *** [installed_pylith] Error 2<br>
><br>
> The error appears to be related to building VampirTrace, which may not<br>
> be critical (?), and following this<br>
> <a href="http://stackoverflow.com/questions/7785464/open-mpi-compilation-error" target="_blank">http://stackoverflow.com/questions/7785464/open-mpi-compilation-error</a> I'd like<br>
> to disable VampirTrace during the build, but I don't know how to pass<br>
> the 'disable-vt' flag during the pylith configuration process. I tried<br>
> building openmpi from the terminal using the configuration generated by<br>
> autoconf during the Pylith install ($ ../openmpi-1.4.5/configure<br>
> --prefix=/Users/itchy/pylith CPPFLAGS=-I/Users/itchy/pylith/include<br>
> LDFLAGS=-L/Users/itchy/pylith/lib CFLAGS=-g -O2 CXXFLAGS=-g -O2<br>
> -DMPICH_IGNORE_CXX_SEEK FCFLAGS= CC=gcc CXX=g++ FC=gfortran) plus<br>
> '--disable-vt' but that doesn't work because many of the flags there<br>
> aren't recognized as options from the command line. I also tried adding<br>
> '--mpi_options="--disable-vt" ' and '--openmpi_options="--disable-vt" '<br>
> during pylith configuration (similar to what can be done with petsc) but<br>
> those don't fly.<br>
><br>
> The config.log files for both pylith and openmpi are attached.<br>
><br>
> Has anyone dealt with this or a similar issue?<br>
><br>
> Thanks,<br>
> Richard<br>
><br>
><br>
><br>
> _______________________________________________<br>
> CIG-SHORT mailing list<br>
> <a href="mailto:CIG-SHORT@geodynamics.org" target="_blank">CIG-SHORT@geodynamics.org</a><br>
> <a href="http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short" target="_blank">http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short</a><br>
><br>
<br>
_______________________________________________<br>
CIG-SHORT mailing list<br>
<a href="mailto:CIG-SHORT@geodynamics.org" target="_blank">CIG-SHORT@geodynamics.org</a><br>
<a href="http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short" target="_blank">http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short</a><br>
</blockquote></div></div></div><span><font color="#888888"><br><br clear="all"><div><br></div>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener
</font></span></div></div>
<br>_______________________________________________<br>
CIG-SHORT mailing list<br>
<a href="mailto:CIG-SHORT@geodynamics.org" target="_blank">CIG-SHORT@geodynamics.org</a><br>
<a href="http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short" target="_blank">http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short</a><br></blockquote></div><br></div>