dolfin version 1.0.0-1 FTBFS on armhf in precise

[Andreas Moog]

This is a semi-automatic report based on the latest archive rebuild results [1].
Apologies if it reaches you in error.

Excerpt from the buildlog:

-- thread
-- iostreams
-- math_tr1
-- mpi
-- serialization
-- UFC version: 2.0.5
-- Checking for package 'Armadillo'
-- Performing Test ARMADILLO_TEST_RUNS
-- Performing Test ARMADILLO_TEST_RUNS - Success
-- Found Armadillo: /usr/lib/libarmadillo.so
-- Found LibXml2: /usr/lib/arm-linux-gnueabihf/libxml2.so
-- Found PythonInterp: /usr/bin/python (found version "2.7.2+")
-- Found PythonLibs: /usr/lib/libpython2.7.so
-- NumPy headers found
-- Found SWIG: /usr/bin/swig2.0 (found version "2.0.4")
-- Checking for package 'SCOTCH-PT'
-- Performing test SCOTCH_TEST_RUNS

Session terminated, terminating shell...make: *** wait: No child processes. Stop.
make: *** Waiting for unfinished jobs....
make: *** wait: No child processes. Stop.
 ...terminated.
Build killed with signal 15 after 800 minutes of inactivity
******************************************************************************
Build finished at 20120207-2139
FAILED [dpkg-buildpackage died]

The full buidlog can be found at https://launchpad.net/ubuntu/+archive/test-rebuild-20120201/+build/3143713/+files/buildlog_ubuntu-precise-armhf.dolfin_1.0.0-1_FAILEDTOBUILD.txt.gz.

[1] http://people.ubuntuwire.org/~wgrant/rebuild-ftbfs-test/test-rebuild-20120201-precise.html

[Juan Solano (solano)]

I have tried to build this on my pandaboard running precise. The scotch test hangs when calling MPI_Init_thread in scotch_test_lib.cpp

  MPI_Init_thread(0, 0, MPI_THREAD_SINGLE, &provided);

we get stuck on this loop in function opal_condition_wait():

while (c->c_signaled == 0) {
    opal_progress();

This may be related to this issue of openmpi1.4.3 running on arm:

https://bugs.launchpad.net/ubuntu/+source/openmpi/+bug/949044

[Matthias Klose (doko)]

superseded by lp #1268469, now ftbfs everywhere.