Krishna,

Do you have the same problem for smaller systems? Most probably fdf files
are useless in this case, since it is a compiler problem. My wild guess:
check if you are using optimization O3. If you are, then lower it to O2.

Marcos

On Thu, Jul 28, 2011 at 4:45 AM, krishna Mohan Grangadhran Prasanna <
[email protected]> wrote:

>  Dear List,
>
> I am optimizing a Graphene+TCNE ((CN)2C=C(CN)2)) Diels-Alder adduct using
> the .fdf file and geometries attached in this mail.
> I use CG for the minimizing, and it done only the initial step and it got
> crashed.
>
> The Log, error message:
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image              PC                Routine            Line        Source
> siesta             00000000006B01C8  Unknown               Unknown  Unknown
> siesta &nbsp ;           00000000006AC92F  Unknown               Unknown
>  Unknown
> siesta             000000000058C8D3  Unknown               Unknown  Unknown
> siesta             000000000058E46C  Unknown               Unknown  Unknown
> siesta             00000000005DC9CD  Unknown               Unknown  Unknown
> siesta             00000000005DCC86  Unknown           &n bsp;   Unknown
>  Unknown
> siesta             0000000000434DCC  Unknown               Unknown  Unknown
> libc.so.6          0000003F1C61D974  Unknown               Unknown  Unknown
> siesta             0000000000434CD9  Unknown               Unknown  Unknown
> ...
> ...
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> +++++++++++
>
> AND,
> I note the <_.STRUCT_NEXT_ITER> file, which is,
>
>
> ------------------------------------------------------------------------------------------
> $ more 10xGraphene.Final_.STRUCT_NEXT_ITER
>          21.450000000      12.384163052       0.000000000
>          21.450000000     -12.384163052       0.000000000
>           0.000000000       0.000000000      20.000000000
>          210
>   1     6               NaN               NaN               NaN
>   1     6               NaN               NaN               NaN
>   1     6               NaN         & nbsp;     NaN               NaN
>   1     6               NaN               NaN               NaN
>   1     6               NaN               NaN               NaN
> ...
> ...
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
> **Also, the last part of .OUT file is,
>
>
> ------------------------------------------------------------------------------------------
> * Maximum dynamic memory allocated =   109 MB
>
> siesta:                 ==============================
>                             Begin CG move =      1
>       &n bsp;                 ==============================
>
> outcell: Unit cell vectors (Ang):
>        21.450000   12.384163    0.000000
>        21.450000  -12.384163    0.000000
>         0.000000    0.000000   20.000000
>
> outcell: Cell vector modules (Ang)   :   24.768326   24.768326   20.000000
> outcell: Cell angles (23,13,12) (deg):     90.0000     90.0000     60.0000
> outcell: Cell volume (Ang**3)        :  10625.6119
>
> ------------------------------------------------------------------------------------------
>
> Additionally, I checked the FORCE_STRESS file and <siesta: Atomic forces
> (eV/Ang):> from the .OUT file, which seems ok.
>
> So, does anyone know why this crash is occured? - I presumably think that
> (because of _.STRUCT_NEXT_ITER), the Siesta steps into a Wrong geometry (ie.
> because of F90's NaN in X Y Z), and hence the crash. But I dont know why.
>
> Thanks in advance
>
>   *Dr. Krishna Mohan G. P.*
>
> Center for Superfunctional Materials,
>
> POSTECH (Pohang University of Science and Technology),
>
> Republic of Korea.
>
> [email protected] / +821075879321
>

Responder a