[Nek5000-users] nek5000 compilation
stefanke
stefanke at lav.mavt.ethz.ch
Thu Nov 26 10:47:00 CST 2009
Ok it's indeed a non-glibc platform.
Please replace your chelpers.c (source file in the trunk) by
#include <execinfo.h>
#include <stdio.h>
#include <stdlib.h>
#include "jl/fname.h"
#define print_stack FORTRAN_NAME(print_stack, PRINT_STACK)
/* Obtain a backtrace and print it to stdout. */
void print_stack(void)
{
#if defined __GLIBC__
void *bt[50];
int i;
int bt_size = backtrace(bt, 50);
char **symbols = backtrace_symbols(bt, bt_size);
printf ("backtrace(): obtained %d stack frames.\n", bt_size);
for (i=0; i<bt_size; i++) printf("%s\n", symbols[i]);
free (symbols);
#endif
}
and try again.
#Stefan
On Nov 26, 2009, at 5:43 PM, Tamay Ozgokmen wrote:
>
> It is Apple Mac OS X 10.3.9
>
> (sysx is based on a bunch of Macs.. strange but it works OK...)
>
> On Nov 26, 2009, at 11:40 AM, stefanke wrote:
>
>> What's your OS (operating system)?
>>
>>
>> On Nov 26, 2009, at 5:25 PM, Tamay Ozgokmen wrote:
>>
>>>
>>> Hi Stefan,
>>>
>>> I am unable to answer this question since I know nothing about C or glibc :)
>>> I also tried with mpicc (instead of mpixlc) with similar results.
>>> It supposed to be a system based on Power G4 (similar to IBM P4 and P5) chips with
>>> IBM compilers, I am not sure what exactly is installed. The website is (if any use):
>>>
>>> http://www.arc.vt.edu/arc/SystemX/
>>>
>>> I have been running various versions of nek5000 on this for several years.
>>> A makefile that I use to compile an older version of nek500 is attached,
>>> if it can provide any clues on how to fix this...
>>>
>>> Thanks,
>>> Tamay
>>>
>>>
>>> <makefile.sysx>
>>>
>>>
>>> On Nov 26, 2009, at 11:16 AM, stefanke wrote:
>>>
>>>> "/home/ozgokmen/nek5_svn/trunk/nek/chelpers.c", line 1.10: 1506-296 (S) #include file <execinfo.h> not found.
>>>>
>>>> It seems like your running on a non-glibc platform. Right?
>>>>
>>>> #Stefan
>>>>
>>>> On Nov 26, 2009, at 5:05 PM, Tamay Ozgokmen wrote:
>>>>
>>>>>
>>>>> Hi,
>>>>>
>>>>> I put compiler options "mpif77" and "mpixlc" in makenek (below) but it did not go through
>>>>> (compiler.out attached); any suggestions?
>>>>>
>>>>> Thanks!
>>>>> Tamay
>>>>>
>>>>>
>>>>> **********************************************************************************************************
>>>>>
>>>>> #!/bin/bash
>>>>> #
>>>>> # nek5000 build script
>>>>> # (c) 2008,2009 UCHICAGO ARGONNE, LLC
>>>>> #
>>>>>
>>>>> # source path
>>>>> SOURCE_ROOT="$HOME/nek5_svn/trunk/nek"
>>>>>
>>>>> # plug-in list
>>>>> PLUGIN_LIST=""
>>>>>
>>>>> # pre-processor symbol list
>>>>> PPLIST="" # MPIIO BLAS_MXM BG_MXM K10_MXM
>>>>>
>>>>> # Fortran compiler
>>>>> F77="mpif77"
>>>>>
>>>>> # C compiler
>>>>> CC="mpixlc"
>>>>>
>>>>> ******************************************************************************************************************
>>>>> <compiler.out>
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Nek5000-users mailing list
>>>>> Nek5000-users at lists.mcs.anl.gov
>>>>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>>>>
>>>
>>> *******************************************************************************
>>>
>>> Tamay Ozgokmen
>>> Associate Professor
>>> Division of Meteorology and Physical Oceanography
>>> Rosenstiel School of Marine and Atmospheric Science
>>> 4600 Rickenbacker Causeway
>>> Miami, FL 33149-1098, USA
>>> Phone : 305 421 4053
>>> Fax : 305 421 4696
>>> Email : tozgokmen at rsmas.miami.edu
>>> Web : http://www.rsmas.miami.edu/personal/tamay
>>>
>>
>
> *******************************************************************************
>
> Tamay Ozgokmen
> Associate Professor
> Division of Meteorology and Physical Oceanography
> Rosenstiel School of Marine and Atmospheric Science
> 4600 Rickenbacker Causeway
> Miami, FL 33149-1098, USA
> Phone : 305 421 4053
> Fax : 305 421 4696
> Email : tozgokmen at rsmas.miami.edu
> Web : http://www.rsmas.miami.edu/personal/tamay
>
More information about the Nek5000-users
mailing list