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

Re: 20040818: Gempak decoder crashing problem

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 20040818: Gempak decoder crashing problem
  • Date: Wed, 18 Aug 2004 13:09:17 -0400 (EDT)
On Wed, 18 Aug 2004, Unidata Support wrote:

> Are you using the Solaris binary distribution, or build locally?
> My question relates to whether the -O optimization is affecting
> the local build.

Steve,

It's funny you should ask this question.  Let me back up a bit and go into
confession mode.  Last year about this time I built 5.6.k and I decided to
get clever and inserted the '-O' option into the 'Makeinc.solaris' so as
to get everything to run faster and because based on what little I know of
RISC, you should really need to optimize to take advantage of what RISC
architecture offers.  Well of course this was a mistake, as you would have
undoubtedly put it in already if it wasn't going to cause things to break.
I ran into lots of problem with stuff crashing, GARP as well as decoders.
So I rebuilt it taking out the option I had introduced and most things
worked from then on but there were these decoder crashes.  I was not
convinced everything including the gempak libraries were rebuilt and I
assumed that the crashes were because of this.  So one of my objectives in
upgrading to 5.7.2p2 was to build it from scratch as distributed so as to
eliminate over-agressive optimization as a factor.  But evidently I did
not succeed.  Taking a closer look at the distributed 'Makeinc.solaris'
file for 5.7.2p2, I see that there is some optimization stuff there (maybe
was not there at 5.6.k?):

COPT = -DUNDERSCORE -D$(OPSYS) $(GEMINC) $(MOTIFINC) $(XWINCDIR) -O #-g
FOPT =  $(GEMINC) $(GEMINC)/$(OPSYS) -O2 # -g # (bad experience with -O
                                                 aka -O3 and WS6.2)

NCOPT = "FC= " "CXX= "
JASPEROPT = 'CC=$(CC)' 'CFLAGS=-O'

Here is the line from the output of the make which seems to involve
'dcgrib2' (I have the entire make output if you need it.)

f77 -I/weather/GEMPAK5.7.2p2/gempak/include
-I/weather/GEMPAK5.7.2p2/gempak/include/SunOS -O2 -I
/weather/GEMPAK5.7.2p2/gempak/source/bridge/dc
-I/weather/GEMPAK5.7.2p2/gempak/source/gemlib/na
-I/weather/GEMPAK5.7.2p2/gempak/source/gemlib/er -xildoff  dcchecksub.o
dcfillgrid.o  dcflnam.o
  dcflnam2.o  dcgfcyl.o  dcgrib.o  dcggrib.o  dcgnav.o  dcitoc.o
dcogfil.o  dcsgrid.o  dcsubgri
d.o  dcwpgd.o  decode_grib.o  decode_grib1.o  decode_grib2.o
decode_g2pdt.o  decode_g2gds.o  de
code_g2lamb.o  decode_g2ltln.o  dcquasi.o  dcwppg.o  tblist.o  tbsgrid.o
tbsubc.o  qlin.o /weat
her/GEMPAK5.7.2p2/lib/sol/gemlib.a /weather/GEMPAK5.7.2p2/lib/sol/bridge.a
/weather/GEMPAK5.7.2p
2/lib/sol/cgemlib.a /weather/GEMPAK5.7.2p2/lib/sol/gemlib.a
/weather/GEMPAK5.7.2p2/lib/sol/libg
2c.a  /weather/GEMPAK5.7.2p2/jasper/sol/lib/libjasper.a
/weather/GEMPAK5.7.2p2/lib/sol/libpng.a
 /weather/GEMPAK5.7.2p2/lib/sol/libz.a -lm -lsocket -lnsl -lgen -lm -o
dcgrib2

> I'll see if I can create a duplicate of your problem for the 5.7.3
> release I'm working on.

Tom
-----------------------------------------------------------------------------
Tom McDermott                           Email: address@hidden
Systems Administrator                   Phone: (585) 395-5718
Earth Sciences Dept.                    Fax: (585) 395-2416
SUNY College at Brockport