Re: [AMBER-Developers] another amber 14 compile problem netcdf

From: Crowley, Michael <Michael.Crowley.nrel.gov>
Date: Fri, 27 Sep 2013 17:53:03 -0600

I am able to compile now on the linux clusters both of which gave the same
tag problem with netcdf.
I added --tag=cc to a couple of lines in
Ambertools/src/netcdf/src/f90/Makefile.in


LTFCCOMPILE = $(LIBTOOL) --tag=cc --mode=compile $(FC) $(AM_FCFLAGS)
$(FCFLAGS)
FCLD = $(FC)
FCLINK = $(LIBTOOL) --tag=cc --mode=link $(FCLD) $(AM_FCFLAGS) $(FCFLAGS) \
        $(AM_LDFLAGS) $(LDFLAGS) -o $.

during make install:

/bin/sh ../libtool --tag=cc --mode=compile ifort -I../libsrc -I. -ip -O3
-xHost -c -o typeSizes.lo typeSizes.f90
libtool: ignoring unknown tag cc

So it ignores the tag but seems to compile and run just fine, whereas
without the --tag, it complains that there isn't one.



Go figure.
Is this something we want to put into the tree?
Mike

-- 
Michael F. Crowley, Ph.D.
Principal Scientist, Biomolecular Sciences Division
National Renewable Energy Laboratory
1617 Cole Blvd.
Golden, CO 80401
303-384-6345 office
303-887-0149 cell
On 9/27/13 4:45 PM, "Crowley, Michael" <Michael.Crowley.nrel.gov> wrote:
>OK, thanks Dan, I will just leave it alone and have my own workaround.
>Mike
>
>-- 
>Michael F. Crowley, Ph.D.
>Principal Scientist, Biomolecular Sciences Division
>National Renewable Energy Laboratory
>1617 Cole Blvd.
>Golden, CO 80401
>303-384-6345 office
>303-887-0149 cell
>
>
>
>
>
>
>
>On 9/27/13 3:17 PM, "Daniel Roe" <daniel.r.roe.gmail.com> wrote:
>
>>I'm not 100% sure but I think mdgx may require '-lnetcdf' since it can
>>be
>>built as a shared library. Probably we will need to have
>>'netcdf_flag' in
>>configure2 set as a separate variable in config.h
>>(e.g. NETCDF_FLAG) instead
>>of having it just be part of FLIBS, and
>>have programs that require
>>'-lnetcdf' (like nab and mdgx) use that.
>>
>>-Dan
>>
>>On Fri, Sep 27, 2013 at 2:51
>>PM, Crowley, Michael
>><Michael.Crowley.nrel.gov> wrote:
>>> Looks like mdgx
>>also has  -lnetcdf hardcoded in its Makefile. Do we just
>>> replace it with
>>$(NETCDFLIB)  ?
>>> Mike
>>> --
>>> Michael F. Crowley, Ph.D.
>>> Principal
>>Scientist, Biomolecular Sciences Division
>>> National Renewable Energy
>>Laboratory
>>> 1617 Cole Blvd.
>>> Golden, CO 80401
>>> 303-384-6345 office
>>>
>>303-887-0149 cell
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> On 9/27/13 1:22 PM, "Crowley, Michael"
>><Michael.Crowley.nrel.gov> wrote:
>>>
>>>>Sorry about the double emails, don't
>>know why NREL's server is doing that.
>>>>--
>>>>Michael F. Crowley,
>>Ph.D.
>>>>Principal Scientist, Biomolecular Sciences Division
>>>>National
>>Renewable Energy Laboratory
>>>>1617 Cole Blvd.
>>>>Golden, CO
>>80401
>>>>303-384-6345 office
>>>>303-887-0149 cell
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>On 9/27/13
>>1:19 PM, "Crowley, Michael" <Michael.Crowley.nrel.gov> wrote:
>>>>
>>>>>Thanks
>>Jason,
>>>>>I will try that also. Just noticed someone added the pgi compilers
>>into
>>>>>the default module load, that may be interfering, who
>>knows.
>>>>>Mike
>>>>>
>>>>>--
>>>>>Michael F. Crowley, Ph.D.
>>>>>Principal
>>Scientist, Biomolecular Sciences Division
>>>>>National Renewable Energy
>>Laboratory
>>>>>1617 Cole Blvd.
>>>>>Golden, CO 80401
>>>>>303-384-6345
>>office
>>>>>303-887-0149 cell
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>On 9/27/13 11:56 AM,
>>"Jason Swails" <jason.swails.gmail.com> wrote:
>>>>>
>>>>>>On Fri, Sep 27, 2013
>>at 1:35 PM, David A Case
>>>>>><case.biomaps.rutgers.edu>wrote:
>>>>>>
>>>>>>> On
>>Fri, Sep 27, 2013, Crowley, Michael wrote:
>>>>>>>
>>>>>>> > This is the netcdf
>>problem, something about a tag:
>>>>>>>
>>>>>>> Does it happen with gnu
>>compilers?  Which version of intel compilers
>>>>>>>are
>>>>>>>  you
>>>>>>>
>>using?
>>>>>>>
>>>>>>
>>>>>>The only time I've seen this before is when compiler
>>environment
>>>>>>variables
>>>>>>are set.  What happens if you
>>try
>>>>>>
>>>>>>unset FC F90 CC CXX F77
>>>>>>
>>>>>>and then try to build?
>>(unsetenv if you are using cshell).
>>>>>>
>>>>>>--
>>>>>>Jason M.
>>Swails
>>>>>>BioMaPS,
>>>>>>Rutgers University
>>>>>>Postdoctoral
>>Researcher
>>>>>>_______________________________________________
>>>>>>AMBER-Dev
>>elopers mailing 
>>list
>>>>>>AMBER-Developers.ambermd.org
>>>>>>http://lists.ambermd.org/mailman/l
>>istinfo/amber-developers
>>>>>
>>>>>
>>>>>________________________________________
>>_______
>>>>>AMBER-Developers mailing
>>list
>>>>>AMBER-Developers.ambermd.org
>>>>>http://lists.ambermd.org/mailman/lis
>>tinfo/amber-developers
>>>>
>>>>
>>>>_____________________________________________
>>__
>>>>AMBER-Developers mailing
>>list
>>>>AMBER-Developers.ambermd.org
>>>>http://lists.ambermd.org/mailman/listi
>>nfo/amber-developers
>>>
>>>
>>> _______________________________________________
>>>
>>AMBER-Developers mailing list
>>> AMBER-Developers.ambermd.org
>>>
>>http://lists.ambermd.org/mailman/listinfo/amber-developers
>>
>>
>>
>>--
>>-------------------------
>>Daniel R. Roe, PhD
>>Department of Medicinal
>>Chemistry
>>University of Utah
>>30 South 2000 East, Room 201
>>Salt Lake City, UT
>>84112-5820
>>http://home.chpc.utah.edu/~cheatham/
>>(801) 587-9652
>>(801)
>>585-6208 
>>(Fax)
>>
>>_______________________________________________
>>AMBER-Developers
>>mailing 
>>list
>>AMBER-Developers.ambermd.org
>>http://lists.ambermd.org/mailman/listinfo/
>>amber-developers
>>
>>
>
>
>_______________________________________________
>AMBER-Developers mailing list
>AMBER-Developers.ambermd.org
>http://lists.ambermd.org/mailman/listinfo/amber-developers
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Sep 27 2013 - 17:00:02 PDT
Custom Search