[MITgcm-devel] Re: ifort & dome exp.

[Alistair Adcroft]

I ran up against the same problem recently but it is not the data file 
that is the problem it is becaus emsgBUf is too small for problems with 
reasonable numbers of levels.

The fix I came up with is the following:

Index: diagnostics/diag_readparms.F
===================================================================
RCS file: /u/gcmpack/MITgcm/pkg/diagnostics/diag_readparms.F,v
retrieving revision 1.13
diff -r1.13 diag_readparms.F
142c142,144
<         WRITE(msgBuf,*) 'Levels:    ',(levs(l,n),l=1,nlevels(n))
---
 >       do l=1,nlevels(n)
 >         WRITE(msgBuf,*) 'Levels:    ',(levs(l,n))
 >       enddo

The output is not as pretty as Andrea would like but the model gets past 
this point. I haven't checked it in yet.

A.


Jean-Michel Campin wrote:
> Hi Alistair,
> 
> when I use ifort on faulks, the dome experiment fails to run,
> with this error:
> 
> 
>>faulks{build}% cat run.log
>>running ... forrtl: severe (66): output statement overflows record, unit -5, 
>>file Internal List-Directed Write
>>Image              PC        Routine            Line        Source
>>mitgcmuv           0815B7D1  Unknown               Unknown  Unknown
>>mitgcmuv           08120039  Unknown               Unknown  Unknown
>>mitgcmuv           08120630  Unknown               Unknown  Unknown
>>mitgcmuv           08147BA5  Unknown               Unknown  Unknown
>>mitgcmuv           08054FE0  Unknown               Unknown  Unknown
>>
>>Stack trace terminated abnormally.
>>make: *** [output.txt] Error 66
>>
> 
> 
> now if I just uncomment the line:
> #  levels(1,1) = 1.,3.,5.,
> in data.diagnostics, it runs smoothly.
> 
> So, something is not completly right in the way this option:
> # Omit level(1,1) to diagnose ALL levels otherwise list individual levels here
> works with ifort on faulks.
> 
> Can you fix this Pb (since you did this modif) ?
> 
> Thanks,
> 
> Jean-Michel


-- 
Dr Alistair Adcroft            http://www.mit.edu/~adcroft
MIT Climate Modeling Initiative        tel: (617) 253-5938
EAPS 54-1624,  77 Massachusetts Ave,  Cambridge,  MA,  USA