segmentation fault 2

Accelerated Computing HPC Compilers Legacy PGI Compilers
1

Hi,
i’m stil having problem compiling CALMET downloaded from www.src.com.

I allway got Segmentation fault.

I tried writing

export MPSTKZ=8M
export MPSTKZ=32M
…
export MPSTKZ=128M
export MPSTKZ=unlimited

on a shell but the result is always the same.
It is possible that the problems dipends on my access privilegies to the university cluster??
I had to be a root to avoid this problem??

The cluster have a Red Hat 9 on o Pentium processor.
Linux kali.univpm.it 2.4.20-8 #1 Thu Mar 13 17:54:28 EST 2003 i686 i686 i386 GNU/Linux

Thank you for your help.

2

Hi Ivan,

I downloaded the CALMET source from http://www.src.com/html/verio/download/download.htm but it doesn’t seem to be the same as what you have. The code I found is configured for Windows and doesn’t contain any MPI calls. Do I have the correct package?

Note if you need to adjust the stack size, use your shell’s ‘ulimit’ (bash) or ‘limit’ (csh) commands. MPSTKZ was used for setting an OpenMP thread’s stacksize, but hasn’t been needed since the 6.0 release. Which version of the compilers are you using?

  • Mat
3

Hi Mat,
try downloading the code from http://www.src.com/html/verio/download/mod6_codes.htm. I downloaded it a few days ago so i’m sure we are going to have the same version.

I tryed using ulimit but the result is always the same: Segmentation fault.

Thank you very much.

4

Hi Ivan,

Here’s what I’ve done so far:

  1. Downloaded http://www.src.com/html/verio/download/mod6_Files/CALMET_v6.211.zip
  2. "unzip"ed the package
  3. Copied the files from the “CODE” directory to a “code” directory and transformed the filenames from upper to lower case. (Linux is case sensitive).
  4. Commented out line 2831 of calutils.for
  5. Uncommented out lines 2853 to 2856 of calutils.for
  6. Compiled with ‘pgf90 -O0 -o calmet.out calmet.for’
  7. Then ran ‘calment.out calmet.inp’

The first run failed because the input file had two lines (607 and 613) where the data spanned multiple lines. While I fixed this, the next run’s list file complains about an invalid date. So I’m stuck.

Try compiling the program with “-g” and run it with PGDBG. The debugger should be able to help you determine the cause of the seg fault.

  • Mat
5

Hi Mat,
thank you very much for your help.

I discovered i had a 5.***** version of CALMET. Downloading now the same version as you i have your same problems. So no more segmentation fault.

Hope to solve them reading CALMET user’s guide.

Ivan

6

Please let us know what you find out. I’m sure others are having the same issues.

  • Mat