[troy at osc.edu: Re: MPI-IO netCDF?]

[Robert Latham]

Anybody here more familiar with pgi and how to make it like pointer
math?

==rob

----- Forwarded message from Troy Baer <troy at osc.edu> -----

From: Troy Baer <troy at osc.edu>
Subject: Re: MPI-IO netCDF?
Date: Thu, 27 Feb 2003 10:20:43 -0500
Message-ID: <Pine.SGI.4.21.0302271011350.10505-100000 at paladin.osc.edu>
To: Robert Latham <robl at mcs.anl.gov>
cc: Rob Ross <rross at mcs.anl.gov>
X-Spam-Status: No, hits=-4.0 required=5.0 tests=IN_REP_TO,SUBJ_ENDS_IN_Q_MARK,X_NOT_PRESENT version=2.21

I've tried building this in our IA32 environment (PGI compilers, MPICH
ch_gm 1.2.4..8a) and ran into the following set of compile errors:

mpicc -c -DpgiFortran -I.   -DNDEBUG validator.c
PGC-S-0053-Illegal use of void type (validator.c: 149)
PGC-S-0053-Illegal use of void type (validator.c: 212)
PGC-S-0053-Illegal use of void type (validator.c: 219)
PGC-S-0053-Illegal use of void type (validator.c: 266)
PGC-S-0053-Illegal use of void type (validator.c: 310)
PGC-S-0053-Illegal use of void type (validator.c: 388)
PGC-S-0053-Illegal use of void type (validator.c: 395)
PGC-S-0053-Illegal use of void type (validator.c: 524)
PGC-S-0053-Illegal use of void type (validator.c: 531)
PGC/x86 Linux/x86 4.0-2: compilation completed with severe errors
make[3]: *** [validator.o] Error 2

All of these involve printf()s of error messaages that monkey with elements
of a gbp structure.  Any ideas, other than punting and using gcc?

	--Troy
-- 
Troy Baer                       email:  troy at osc.edu
Science & Technology Support    phone:  614-292-9701
Ohio Supercomputer Center       web:  http://oscinfo.osc.edu


----- End forwarded message -----

-- 
Rob Latham
Mathematics and Computer Science Division    A215 0178 EA2D B059 8CDF
Argonne National Labs, IL USA                B29D F333 664A 4280 315B