Problems with Linux Build of parallel netcdf

Richard Hedges richard-hedges at llnl.gov
Wed Oct 5 13:16:20 CDT 2005


At 10:52 AM -0700 10/5/05, Richard Hedges wrote:
>I am building the official release of 1.0 on an ia32 Linux CLuster 
>using Intel compilers (version 8.0).
>
>I am seeing a bunch of unexpected failures in the verification 
>suite, in particular  of the tests
>n[c,f]mpi_put_var*_float , getting diagnostics such as "Numeric 
>conversion not representable" or
>"Var  value read not that expected"
>
>I am using the following settings:
>
>    setenv MPICC  "mpiicc -lmpio"
>    setenv MPIF77 "mpiifort -lmpio"
>    setenv F77    ifort
>    setenv FC     ifort
>    setenv CC     icc

Looks like I needed

CFLAGS = FFLAGS =  -O -mp

Now the verification succeeds.



>
>
>The same settings work fine on an ia64 Linux System.
>
>Is there a known problem with the Intel compilers, or is something 
>more needed re. the settings?
>
>
>Thanks,
>
>- Richard
>
>--
>
>
>
>====================================================
>
>Richard Hedges
>Scalable I/O Project
>Development Environment Group - Livermore Computing
>Lawrence Livermore National Laboratory
>7000 East Avenue, MS L-557
>Livermore, CA	94551
>
>v:	(925) 423-2699
>f:	(925) 423-6961
>E:	richard-hedges at llnl.gov


-- 



====================================================

Richard Hedges
Scalable I/O Project
Development Environment Group - Livermore Computing
Lawrence Livermore National Laboratory
7000 East Avenue, MS L-557
Livermore, CA	94551

v:	(925) 423-2699
f:	(925) 423-6961
E:	richard-hedges at llnl.gov




More information about the parallel-netcdf mailing list