[mpich-discuss] MPICH 3.3.1 tests fail

[Zhou, Hui]

Hi Mark,

Some of the tests will fail due to resource limit or issues that haven’t been addressed yet. I took a look at your test log and the failed tests are all expected. In particular, the `getfence|putfence` tests are with very large non-contiguous data type that results in timeout due to yet to be optimized data loop code. They designed as stress tests that doesn’t affect most applications. The persistent tests are known failures. The bcastlength test is merely a test fail to produce expected error message. So they are all within expectations.

—
Hui Zhou









On Oct 4, 2019, at 7:35 AM, Mark Cunningham via discuss <discuss at mpich.org<mailto:discuss at mpich.org>> wrote:

I have built MPICH 3.3.1 on Ubuntu 18.04 with the gcc-7 compiler suite.  (Build script is provided: build_mpich.)  I have built with both --with-device=ch3:nemesis and --with-device=ch3:nemesis:ofi
with the same outcome in the make test.  A dozen of the tests fail with timeouts.
I ran the test on an 8 processor Thinkstation but what would be beneficial is some sort of sense of what the test results should be.  I've browsed the test results that you've posted but these appear to be summaries that conclude All Tests Passed, without individual timings for each test.

In any case, I would appreciate any guidance that you might be able to provide.  I have provided the config files and the summary.junit.xml file. (On my machine the summary.xml file just reports that there is an error on line 9562.)  The result of mpiexec -info is:
~/build/mpich-3.3.1/src/pm/hydra % ./mpiexec.hydra -info
HYDRA build details:
    Version:                                 3.3.1
    Release Date:                            Wed Jun  5 14:57:33 CDT 2019
    CC:                              gcc
    CXX:                             g++
    F77:                             gfortran
    F90:                             gfortran
    Configure options:                       '--disable-option-checking' '--prefix=/home/share/mpich/mpich-3.3.1' '--with-devce=ch3:nemesis' '--cache-file=/dev/null' '--srcdir=.' 'CC=gcc' 'CFLAGS= -O2' 'LDFLAGS=' 'LIBS=' 'CPPFLAGS= -I/home/mark/buildmpich-3.3.1/src/mpl/include -I/home/mark/build/mpich-3.3.1/src/mpl/include -I/home/mark/build/mpich-3.3.1/src/openpa/src -I/hme/mark/build/mpich-3.3.1/src/openpa/src -D_REENTRANT -I/home/mark/build/mpich-3.3.1/src/mpi/romio/include' 'MPLLIBNAME=mpl'
    Process Manager:                         pmi
    Launchers available:                     ssh rsh fork slurm ll lsf sge manual persist
    Topology libraries available:            hwloc
    Resource management kernels available:   user slurm ll lsf sge pbs cobalt
    Checkpointing libraries available:
    Demux engines available:                 poll select
---------------------------------------------------------
Thanks for your help,
Mark

Mark A Cunningham, PhD
Senior Scientist
Applied Research in Acoustics LLC
Culpeper VA 22701

_______________________________________________
discuss mailing list     discuss at mpich.org<mailto:discuss at mpich.org>
To manage subscription options or unsubscribe:
https://lists.mpich.org/mailman/listinfo/discuss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpich.org/pipermail/discuss/attachments/20191004/7e4de7e7/attachment-0001.html>