[CPMD-list] Trying to get MPI to work

Axel Kohlmeyer axel.kohlmeyer at theochem.ruhr-uni-bochum.de
Mon Aug 4 19:12:34 CEST 2003


>>> "MK" == mkosmows  <mkosmows at mailbox.syr.edu> writes:

MK> Dear CPMD community:
 
 
MK> The compile failure will be discussed first.  I used Configure PC-PGI-MPI.  I 
MK> changed only the FFLAG option of -pc64 to -pc 64 (perhaps this could be 
MK> fixed in the distributed Configure script in future distributions).  The error 
MK> I get is 
MK> undefined reference to `mpi_xxxx in function my_xxx within mypara.o.  There 
MK> are 
MK> several of these errors.  Any assistance would be greatly appreciated.

hi mark,

the default configure uses the options '-byteswapio' and '-Msecond_underscore'.
you should remove them for your setup. the second option is for 
g77 compatibility and you don't need this, if you have compiled your
mpi library with pgf77 (or pgf90). 
 
MK> Lastly, I get the following error message when attempting to run the suggested 
MK> test of mpich.  I think it has something to do with ssh.  I have made the ssh 
MK> keys on both systems, and made sure that both systems have each others 
MK> public keys.  I have added the proper hostnames to the mpich machine config 
MK> file on the system running the test.  I have also made sure that the ssh 
MK> server 
MK> allows empty passwords.  Please forgive the lack of proper terminology - first 
MK> time with ssh.

MK> [mark at linux test]$ make testing 
MK> (cd pt2pt ; ./runtests  -check ) 
MK> Failed to run simple program! 
MK> Output from run attempt was 
MK> *** Testing Unexpected messages *** 
MK> bash: line 1: /home/mark/mpich-1.2.5/examples/test/pt2pt/./third: No such file 
MK> or 
MK> directory 
MK> p0_7921:  p4_error: Child process exited while making connection to remote 
MK> process on linux2: 0 

stupid question: did you make sure, that mpich uses ssh and not rsh?
also the test seems to assume, that both machines have a (NFS-)shared 
directory and the test starts from that. is that true?

hope this helps,
     axel.
 
MK> /home/mark/mpich-1.2.5/bin/mpirun: line 1:  7921 Broken pipe             
MK> /home/mark/mpich-1.2.5/examples/test/pt2pt/./third -p4pg 
MK> /home/mark/mpich-1.2.5/examples/test/pt2pt/PI7836 -p4wd 
MK> /home/mark/mpich-1.2.5/examples/test/pt2pt 
MK> *** Testing Unexpected messages *** 
MK> mpirun program was /home/mark/mpich-1.2.5/bin/mpirun 
MK> mpirun command was 
MK> /home/mark/mpich-1.2.5/bin/mpirun -mvhome -np 2 ./third </dev/null >>third.out 
2> &1 
MK> make: *** [runtest] Error 1 
MK> [mark at linux test]$
 
MK> Any insight on either of these problems would be greatly appreciated.  If 
MK> further 
MK> information is required, I will provide as best as possible.
 
MK> Thank you,
 
MK> Mark Kosmowski

MK> Chemistry Department
MK> Syracuse University
MK> mkosmows at syr.edu

MK> _______________________________________________
MK> CPMD-list mailing list
MK> CPMD-list at cpmd.org
MK> http://www.cpmd.org/mailman/listinfo/cpmd-list



--

=======================================================================
Axel Kohlmeyer       e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de
Lehrstuhl fuer Theoretische Chemie          Phone: ++49 (0)234/32-26673
Ruhr-Universitaet Bochum - NC 03/53         Fax:   ++49 (0)234/32-14045
D-44780 Bochum                   http://www.theochem.ruhr-uni-bochum.de
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.



More information about the CPMD-list mailing list