[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20001201: 20001201: Question: Gempak5.6 SIGSEGV w



Stephen,

Ok, I must have put that fix in before I tarred up the dist that I posted for 
you.
Our HP runs HPUX 11.0B, and I'll try to go through it Monday to see
if anything appears there.

If that doesn't find anything, I'll go through the code profiler on Solaris to 
see if it
points out any non-standard usage, or other memory problems.

Steve Chiswell
Unidata User Support



>From: address@hidden
>Organization: UCAR/Unidata
>Keywords: 200012012252.eB1MqIo27903

>
>
>I have the string*512 and head*512 already in the release:
>
>rwxr-xr-x 504/2000      0 Oct 17 18:13 2000   
>gempak/source/programs/sf/sfchck/
>rw-r--r-- 504/2000   7842 Oct 17 17:20 2000   
>gempak/source/programs/sf/sfchck/sfchck.f
>rwxr-xr-x 504/2000    110 Mar 28 17:06 2000   
>gempak/source/programs/sf/sfchck/sfchck_link
>rw-r--r-- 504/2000   1861 Jun 19 20:03 2000   
>gempak/source/programs/sf/sfchck/sfkdat.f
>rw-r--r-- 504/2000   1023 Apr  3 16:43 1997   
>gempak/source/programs/sf/sfchck/sfkhdr.f
>rw-r--r-- 504/2000    974 Apr  3 16:43 1997   
>gempak/source/programs/sf/sfchck/sfkinp.f
>rw-r--r-- 504/2000   1549 Jun 19 20:03 2000   
>gempak/source/programs/sf/sfchck/sfkosf.f
>rw-r--r-- 504/2000    948 May  1 16:24 2000   
>gempak/source/programs/sf/sfchck/Makefile
>
>I recompiled with -g and used dde, but the core dumps before the first   
>step.
>
>I did a quick estimate of the amount of memory required to initialize the   
>PARAMETERS and variables in GEMPRM.PRM, MCHPRM.HPUX, sfchck.f and   
>ERMISS.FNC and get 9185986 bytes +/- 1KB.
>
>I'll wait for 5.6a, unless you can think of something else.
>
>Stephen Sinnis | Tell:905.566.9511x379
>Pelmorex Inc    | Cell:416.409.1141
>
>
> -----Original Message-----
>From: support [mailto:address@hidden]
>Sent: Friday, December 01, 2000 3:13 PM
>To: address@hidden
>Cc: address@hidden
>Subject: 20001201: Question: Gempak5.6 SIGSEGV w
>
>
>
>
>
>Stephen,
>
>This may be the problem I fixed and reported to NCEP at:
>http://www.unidata.ucar.edu/glimpse/gempak/3935
>
>This fix will be in my 5.6a release which I'm trying to get
>out in the next week. Until then, you can edit your:
>$GEMPAK/source/programs/sf/sfchck/sfchck.f file and change:
>head*132 to head*512
>and
>string*132 to string*512
>
>Then recompile with:
>
>cd $GEMPAK/source/programs/sf/sfchck
>make clean
>make all
>make install
>make clean
>
>See if this fixes your problem.
>
>Steve Chiswell
>Unidata User Support
>
>>From: address@hidden
>>Organization: UCAR/Unidata
>>Keywords: 200012011859.eB1IxNo13196
>
>>
>>
>>Hi.
>>
>>I have searched the e-mail lists for help on this, but have not found   
>any
>>ideas.
>>
>>I have recently installed Gempak5.6 on an HP-UX 10.20 C200.  Everything   
>    
>
>>seems ok so far.  Thank-you.
>>
>>However, I am getting a core dump when I try to use sfchck:
>>
>>Pid 2185 received a SIGSEGV for stack growth failure.
>>Possible causes: insufficient memory or swap space,
>>or stack size exceeded maxssiz.
>>Segmentation fault (core dumped)
>>
>>The current system maxssiz (Maximum Stack Size) is 32 MB (8KB x 4KB per   
>    
>
>>page).  The maxdsiz (Maximum Data Space) is 288 MB (72KB x 4KB per   
>page).
>> How many KB should they be?
>>
>># sysdef | grep -i max
>>NAME                      VALUE       BOOT        MIN-MAX        UNITS   
>    
>
>>  FLAGS
>>dbc_max_pct                  50          -           -
>>                  -
>>maxdsiz                   73728          -        256-655360     Pages   
>    
>
>>  -
>>maxfiles                     60          -         30-2048
>>              -
>>maxfiles_lim               1024          -         30-2048
>>              -
>>maxssiz                    8192          -        256-655360     Pages   
>    
>
>>  -
>>maxswapchunks               512          -          1-16384
>>             -
>>maxtsiz                   16384          -        256-655360     Pages   
>    
>
>>  -
>>maxuprc                      75          -          3-
>>                  -
>>maxvgs                       10          -           -
>>                  -
>>netmemmax              68489216          -           -
>>                  -
>>
>>pelmops3 43: make all
>>        rm -f *.o
>>         fort77 +E4 +E6 -I/OPS/Apps/GEMPAK5.6/gempak/include
>> -I/OPS/Apps/GEMPAK5.6/gempak/include/HPUX -Wl,+n sfchck.f
>>/OPS/Apps/GEMPAK5.6/lib/hpux/sfchck.a
>>/OPS/Apps/GEMPAK5.6/lib/hpux/gemlib.a
>> /OPS/Apps/GEMPAK5.6/lib/hpux/appl.a
>> /OPS/Apps/GEMPAK5.6/lib/hpux/syslib.a
>> /OPS/Apps/GEMPAK5.6/lib/hpux/cgemlib.a
>> /OPS/Apps/GEMPAK5.6/lib/hpux/gemlib.a -o sfchck
>>sfchck.f:
>>   MAIN sfchck:
>>/usr/ccs/bin/ld: (Warning) At least one PA 2.0 object file (sfchck.o)   
>was
>>detected. The linked output may not run on a PA 1.x system.
>>
>>pelmops3 44: which sfchck
>>no sfchck in /usr/bin /opt/ansic/bin /usr/ccs/bin /var/yp
>>/usr/lib/netsvc/yp /usr/contrib/bin /opt/nettladm/bin /opt/pd/bin
>>/usr/bin/X11 /usr/contrib/bin/X11 /opt/upgrade/bin /opt/hparray/bin
>>/opt/ignite/bin /opt/langtools/bin /opt/fortran/bin /opt/imake/bin
>>/opt/graphics/OpenGL/debugger/bin /opt/aCC/bin /opt/CC/bin
>>/OPS/Apps/GEMPAK5.6/bin/hpux /OPS/Apps/GEMPAK5.6/bin/scripts
>>
>>pelmops3 45: make install
>>        rm -f *.o
>>        install -i sfchck /OPS/Apps/GEMPAK5.6/bin/hpux
>>sfchck installed as /OPS/Apps/GEMPAK5.6/bin/hpux
>>
>>pelmops3 46: which sfchck
>>/OPS/Apps/GEMPAK5.6/bin/hpux/sfchck
>>
>>pelmops3 47: /OPS/Apps/GEMPAK5.6/bin/hpux/sfchck
>>
>>Pid 2185 received a SIGSEGV for stack growth failure.
>>Possible causes: insufficient memory or swap space,
>>or stack size exceeded maxssiz.
>>Segmentation fault (core dumped)
>>
>>root@pelmops3 [/OPS/Apps/GEMPAK5.6/gempak/source/programs/sf/sfchck]
>># dmesg core
>>
>>Dec  1 18:33
>>Magic number wrong (namelist mismatch?)
>>
>>
>>Stephen Sinnis | Tell:905.566.9511x379
>>Pelmorex Inc    | Cell:416.409.1141
>>
>
>**************************************************************************  
>Unidata User Support                                    UCAR Unidata   
>(303)497-8644                                                  P.O. Box   
>address@hidden                                   Boulder, CO   
> --------------------------------------------------------------------------  
>Unidata WWW Service                        http://www.unidata.ucar.edu/   
>**************************************************************************  
>