Attempt to detect when we are direct-launched without the necessary P…

[rhc54]

…MI support, and thus are incorrectly identified as being "singleton". Advise the user on the required PMI(x) support and error out.

Signed-off-by: Ralph Castain rhc@open-mpi.org

[rhc54]

@hppritcha @jsquyres @bwbarrett Please see if this meets your requirements.

[rhc54]

Here is what it looks like for SLURM:

$ srun -n 1 ./mpi_spin
--------------------------------------------------------------------------
The application appears to have been direct launched using "srun",
but OMPI was not built with SLURM's PMI support and therefore cannot
execute. There are several options for building PMI support under
SLURM, depending upon the SLURM version you are using:

  version 16.05 or later: you can use SLURM's PMIx support. This
  requires that you configure and build SLURM --with-pmix.

  Versions earlier than 16.05: you must use either SLURM's PMI-1 or
  PMI-2 support. SLURM builds PMI-1 by default, or you can manually
  install PMI-2. You must then build Open MPI using --with-pmi pointing
  to the SLURM PMI library location.

Please configure as appropriate and try again.
--------------------------------------------------------------------------
*** An error occurred in MPI_Init
*** on a NULL communicator
*** MPI_ERRORS_ARE_FATAL (processes in this communicator will now abort,
***    and potentially your MPI job)
[rhc001:189810] Local abort before MPI_INIT completed completed successfully, but am not able to aggregate error messages, and not able to guarantee that all other processes were killed!
srun: error: rhc001: task 0: Exited with exit code 1

[bwbarrett]

I'll give it a try today, but that looks awesome!

[hppritcha]

Im not sure the ALPS check is correct. I'll double check for an env variable to check. Not really important though since configure auto detects for Cray pmi. User would have to explicitly request no PMI support.

[rhc54]

Thanks @hppritcha - I wasn't sure about the ALPS check either, but figured I should at least give it a try. I was more concerned about the singleton case where someone builds OMPI without ALPS support (e.g., ALPS is installed somewhere unexpected) and then launches a job using aprun. We can remove the ALPS code if you feel it isn't needed.

[hppritcha]

@rhc54 this might help:

n17276@kaibab:~>aprun -n 1 env | grep ALPS
PE_PRODUCT_LIST=CRAY_PMI:CRAY_LIBSCI:TOTALVIEW:TOTALVIEW-SUPPORT:GNU:GCC:CRAYPE:CRAYPE_MC12:CRAY_LLM:CRAY_XPMEM:CRAY_DMAPP:CRAY_UGNI:CRAY_UDREG:CRAY_ALPS
CRAY_ALPS_POST_LINK_OPTS=-L/opt/cray/alps/5.2.4-2.0502.9774.31.12.gem/lib64
CRAY_ALPS_INCLUDE_OPTS=-I/opt/cray/alps/5.2.4-2.0502.9774.31.12.gem/include
ALPS_APP_DEPTH=1
ALPS_APP_ID=21657911
ALPS_APP_PE=0

I

[rhc54]

@hppritcha Done - thanks!