[MITgcm-devel] (no subject)
Chris Hill
cnh at mit.edu
Thu Jul 22 14:26:15 EDT 2004
Hi
Gianmaria can yu try deleting the two lines
#include "MAIN_PDIRECTIVES1.h"
and
#include "MAIN_PDIRECTIVES2.h"
from the file main.F and see what happens.
Thanks,
Chris
> -----Original Message-----
> From: mitgcm-devel-bounces at mitgcm.org
> [mailto:mitgcm-devel-bounces at mitgcm.org] On Behalf Of Patrick Heimbach
> Sent: Thursday, July 22, 2004 12:32 AM
> To: mitgcm-devel at mitgcm.org
> Cc: sannino at ocean.mit.edu
> Subject: [MITgcm-devel] (no subject)
>
> Hi Chris,
>
> Gianmaria has set up the code on a Cray SV1, and is trying to
> compile in -O3, that means optimize for vectorization.
> The setup is 1 processor and MPI turned off in rhe
> preprocessor (Gianmaria, is that correct?)
>
> Below is the error that he's getting during compile time (in
> optim. level -O1 everything was OK).
>
> My assumption is that some #ifdef needs to be set properly in
> CPP_EEOPTIONS.h. Any idea?
> Or should it be compiled with MPI even for 1 proc?
>
> Cheers
> -p.
>
>
>
> About CRAY, I fixed the problem with the I/O units, but after
> compiled with the -O3 optimization (vectorization) it appears
> another problem, this time during the compilation of MAIN.F:
>
> f90 -O3 -c main.f
>
> IF ( .NOT. eeBootError ) THEN
> ^
> f90-960 f90: ERROR MAIN, File = main.f, Line = 3205, Column = 20
> Variable "EEBOOTERROR" is in a parallel region and must be
> in the shared or private list if autoscope is not specified.
>
> threadIsComplete(myThid) = .TRUE.
> ^
> f90-960 f90: ERROR MAIN, File = main.f, Line = 3215, Column = 9
> Variable "THREADISCOMPLETE" is in a parallel region and
> must be in the shared or private list if autoscope is not specified.
>
>
>
> --------------------------------------------------------
> Patrick Heimbach Massachusetts Institute of Technology
> FON: +1/617/253-5259 EAPS, Room 54-1518
> FAX: +1/617/253-4464 77 Massachusetts Avenue
> mailto:heimbach at mit.edu Cambridge MA 02139
> http://www.mit.edu/~heimbach/ USA
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://dev.mitgcm.org/mailman/listinfo/mitgcm-devel
>
More information about the MITgcm-devel
mailing list