[MITgcm-devel] compiler optimization on SunOS

[Martin Losch]

Hi Constantinos,

"crash" was not the correct term. We run it for only 1 timestep. Even  
in this first timestep, the result is "ridiculous", please have a  
look a the monitor output (grep %MON STDOUT.0000) or the first output  
of the cg2d solver:
> cg2d: Sum(rhs),rhsMax =   4.37778975048286E-32  3.01076492720156E+23
> (PID.TID 0000.0001)                    cg2d_init_res =     
> 2.98294373127293E+00
> (PID.TID 0000.0001)                       cg2d_iters =    10
> (PID.TID 0000.0001)                         cg2d_res =     
> 4.04220112635106E-02
> cg3d: Sum(rhs),rhsMax =  -9.56179579958416E-15  3.02526346868531E+23
> (PID.TID 0000.0001)                    cg3d_init_res =     
> 1.95235821994126E+01
> (PID.TID 0000.0001)                       cg3d_iters =    10
> (PID.TID 0000.0001)                         cg3d_res =     
> 4.36038214487218E-01
with rhsMax = 1e23, which immediately leads to a "SOLUTION IS HEADING  
OUT OF BOUNDS" message from the monitor output. (not that we limited  
the cg?d iters to 10 in order speed up the runs, but the "cg2d: ..."  
lines are before any iteration within cg2d.

Martin

On 16 Feb 2007, at 16:37, Constantinos Evangelinos wrote:

> On Fri 16 Feb 2007 06:35, Martin Losch wrote:
>
>> I have tried the list of NOOPTFILES from some very old genmake file,
>> NOOPTFILES = barrier.F different_multiple.F external_fields_load.F
>> ini_vertical_grid.F ini_spherical_polar_grid.F ini_cori.F
>> mon_printstats_rl.F mon_printstats_rs.F
>> but that doesn't help.
>>
>> Any further suggestions to add to the list of NOOPTFILES?
>
> I'd try and get a traceback of the sequence of calls leading to the  
> crash and
> put all these files in the NOOPTFILES instead as a first guess.  
> From the
> STDOUT file it appears that the crash happens at the final runtime  
> statistics
> stage?
>
> Constantinos
> -- 
> Dr. Constantinos Evangelinos                    Room 54-1518, EAPS/MIT
> Earth, Atmospheric and Planetary Sciences       77 Massachusetts  
> Avenue
> Massachusetts Institute of Technology           Cambridge, MA 02139
> +1-617-253-5259/+1-617-253-4464 (fax)           USA
>
>
> -- 
> Dr. Constantinos Evangelinos
> Department of Earth, Atmospheric and Planetary Sciences
> Massachusetts Institute of Technology
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel