I've run the tests and everything seems to be alright. Just for completeness
and reference for anyone else that comes looking, the system was Yellowdog
Linux running on Apple G5 nodes, compiled with gfortran. The config.h that did
the trick was:
MATH_DEFINES =
MATH_LIBS =
FFT_DEFINES = -DPUBFFT
FFT_INCLUDE =
FFT_LIBS =
NETCDF_HOME =
NETCDF_DEFINES =
NETCDF_MOD =
NETCDF_LIBS =
MPI_DEFINES =
MPI_INCLUDE =
MPI_LIBDIR =
MPI_LIBS =
DIRFRC_DEFINES = -DDIRFRC_COMTRANS -DDIRFRC_EFS
CPP = /usr/bin/cpp
CPPFLAGS = -traditional -P
F90_DEFINES = -DFFTLOADBAL_2PROC
F90 = gfortran
MODULE_SUFFIX = mod
F90FLAGS = -c
F90_OPT_DBG = -g -ftrace=full
F90_OPT_LO = -O0
F90_OPT_MED = -O2
F90_OPT_HI = -O3
F90_OPT_DFLT = $(F90_OPT_HI)
CC = gcc
CFLAGS =
LOAD = gfortran
LOADFLAGS =
LOADLIBS =
Thanks for the help!
-Justin
Quoting Robert Duke <rduke.email.unc.edu>:
> Glad it finally worked! Points for persistence :-) Some of this stuff can
> be hard to work out if you don't do it all the time; for me it is often hard
> because I don't have access to the system.
> Best Regards - Bob
>
> ----- Original Message -----
> From: "Justin A. Lemkul" <jalemkul.vt.edu>
> To: <amber.scripps.edu>
> Sent: Thursday, May 29, 2008 3:12 PM
> Subject: Re: AMBER: Installing PMEMD with Amber10
>
>
> >
> > Thanks for the tip. I got PMEMD to compile and install without any
> > complaints
> > after making the change you suggested. However, I have been unable to
> > test the
> > installation as of yet. I can run all the serial tests interactively on
> > the
> > head node, but anything in parallel requires submission to a queuing
> > system, so
> > I'm waiting my turn. I will report back if anything is amiss.
> >
> > Thanks again,
> > Justin
> >
> > Quoting Robert Duke <rduke.email.unc.edu>:
> >
> >> Okay, I really don't support this stuff (gfortran or the mac for that
> >> matter - I don't have access to any of this mac stuff, and don't see the
> >> point of using pmemd with a slower compiler). However, the error msg you
> >> have here basically says that the fortran code expects the c routine
> >> names
> >> to be decorated with a single underscore. You have specified CFLAGS
> >> = -DDBL_C_UNDERSCORE, which causes the c code to export names with double
> >> trailing underscores, so the next thing I would try is edit your
> >> config.h,
> >> changing the CFLAGS entry to just
> >> CFLAGS =
> >> Then do a make clean and a make and see what happens. And please run the
> >> test suite when you get done with all this - I can't guarantee that any
> >> of
> >> this code actually works as advertised, even if it compiles and links. I
> >> don't really know what the story here is on gfortran on the mac. The web
> >> page references g95/gfortran, but then lists the fortran compiler as g95,
> >> says something about intel; sorry, but I have no idea what the compiler
> >> really is here (I am presuming it is really gfortran, that the script was
> >> really for g95, but different name mangling conventions occur with
> >> gfortran,
> >> that the chip is really intel, or you would be using the ibm compiler,
> >> etc.
> >> etc.). Perhaps the guys that support "amberonmac" will comment?
> >> Good luck - Bob Duke
> >>
> >> ----- Original Message -----
> >> From: "Justin A. Lemkul" <jalemkul.vt.edu>
> >> To: <amber.scripps.edu>
> >> Sent: Thursday, May 29, 2008 9:23 AM
> >> Subject: AMBER: Installing PMEMD with Amber10
> >>
> >>
> >> >
> >> > Hello all,
> >> >
> >> > I am installing Amber10 on our cluster (Apple Xserve G5 nodes). The OS
> >> > on
> >> > the
> >> > nodes is Yellowdog Linux. The installation of AmberTools and Amber10
> >> > proceeded
> >> > with no problems, but I am having some difficulty installing PMEMD. I
> >> > have
> >> > installed lam-7.1.3 that came with Amber10, and am specifying the
> >> > following:
> >> >
> >> > ./configure macosx_ppc gfortran lam pubfft nobintraj
> >> >
> >> > The config.h file is based on one that I found online
> >> > (http://amberonmac.blogspot.com/2008_03_01_archive.html), and is as
> >> > follows:
> >> >
> >> > MATH_DEFINES =
> >> > MATH_LIBS =
> >> > FFT_DEFINES = -DPUBFFT
> >> > FFT_INCLUDE =
> >> > FFT_LIBS =
> >> > NETCDF_HOME =
> >> > NETCDF_DEFINES =
> >> > NETCDF_MOD =
> >> > NETCDF_LIBS =
> >> > MPI_HOME = /home/rdiv1001/amber10
> >> > MPI_DEFINES = -DMPI -DSLOW_NONBLOCKING_MPI
> >> > MPI_INCLUDE = -I$(MPI_HOME)/include
> >> > MPI_LIBDIR = $(MPI_HOME)/lib
> >> > MPI_LIBS = -L$(MPI_LIBDIR) -llamf77mpi -lmpi -llam -ldl -lpthread
> >> > DIRFRC_DEFINES = -DDIRFRC_COMTRANS -DDIRFRC_EFS
> >> > CPP = /usr/bin/cpp
> >> > CPPFLAGS = -traditional -P
> >> > F90_DEFINES = -DFFTLOADBAL_2PROC
> >> >
> >> > F90 = gfortran
> >> > MODULE_SUFFIX = mod
> >> > F90FLAGS = -c
> >> > F90_OPT_DBG = -g -ftrace=full
> >> > F90_OPT_LO = -O0
> >> > F90_OPT_MED = -O2
> >> > F90_OPT_HI = -O3
> >> > F90_OPT_DFLT = $(F90_OPT_HI)
> >> >
> >> > CC = gcc
> >> > CFLAGS = -DDBL_C_UNDERSCORE
> >> >
> >> > LOAD = gfortran
> >> > LOADFLAGS =
> >> > LOADLIBS =
> >> >
> >> >
> >> > When I proceed with 'make install,' I get the following error:
> >> >
> >> > gfortran -c -O3 erfcfun.f90
> >> > gfortran -o pmemd gbl_constants.o gbl_datatypes.o state_info.o
> >> > file_io_dat.o
> >> > mdin_ctrl_dat.o mdin_ewald_dat.o prmtop_dat.o inpcrd_dat.o
> >> > dynamics_dat.o
> >> > img.o
> >> > parallel_dat.o parallel.o gb_parallel.o pme_direct.o pme_recip_dat.o
> >> > pme_slab_recip.o pme_blk_recip.o pme_slab_fft.o pme_blk_fft.o
> >> > pme_fft_dat.o
> >> > fft1d.o bspline.o pme_force.o pbc.o nb_pairlist.o nb_exclusions.o cit.o
> >> > dynamics.o bonds.o angles.o dihedrals.o extra_pnts_nb14.o runmd.o
> >> > loadbal.o
> >> > shake.o prfs.o mol_list.o runmin.o constraints.o axis_optimize.o
> >> > gb_ene.o
> >> > veclib.o gb_force.o timers.o pmemd_lib.o runfiles.o file_io.o bintraj.o
> >> > pmemd_clib.o pmemd.o random.o degcnt.o erfcfun.o nmr_calls.o nmr_lib.o
> >> > get_cmdline.o master_setup.o pme_alltasks_setup.o pme_setup.o
> >> > ene_frc_splines.o
> >> > gb_alltasks_setup.o
> >> > tprmtop_section.o -L/home/rdiv1001/amber10/lib
> >> > -llamf77mpi -lmpi -llam -ldl -lpthread
> >> > prmtop_dat.o: In function `__prmtop_dat_mod__bcast_amber_prmtop_dat':
> >> > prmtop_dat.f90:(.text+0x10bc): undefined reference to `get_bytesize_'
> >> > prmtop_dat.f90:(.text+0x1124): undefined reference to `get_bytesize_'
> >> > prmtop_dat.f90:(.text+0x118c): undefined reference to `get_bytesize_'
> >> > extra_pnts_nb14.o: In function
> >> > `__extra_pnts_nb14_mod__bcast_extra_pnts_nb14_dat':
> >> > extra_pnts_nb14.f90:(.text+0x1bf8): undefined reference to
> >> > `get_bytesize_'
> >> > loadbal.o: In function `__loadbal_mod__start_loadbal_timer':
> >> > loadbal.f90:(.text+0xa0): undefined reference to `get_wall_time_'
> >> > loadbal.o: In function `__loadbal_mod__update_loadbal_timer':
> >> > loadbal.f90:(.text+0xe0): undefined reference to `get_wall_time_'
> >> > runmin.o: In function `__runmin_mod__printe':
> >> > runmin.f90:(.text+0x5fc): undefined reference to `get_wall_time_'
> >> > runmin.f90:(.text+0xa94): undefined reference to `get_wall_time_'
> >> > runfiles.o: In function `__runfiles_mod__prntmd':
> >> > runfiles.f90:(.text+0x2488): undefined reference to `get_wall_time_'
> >> > pmemd.o: In function `MAIN__':
> >> > pmemd.f90:(.text+0xd8): undefined reference to `unlimit_stack_'
> >> > collect2: ld returned 1 exit status
> >> > make[1]: *** [pmemd] Error 1
> >> > make[1]: Leaving directory `/home/rdiv1001/amber10/src/pmemd/src'
> >> > make: *** [install] Error 2
> >> >
> >> > I would be very grateful if anyone could give me any hints on how to
> >> > fix
> >> > this,
> >> > I've been trying for quite some time!
> >> >
> >> > Thanks for your attention.
> >> >
> >> > -Justin
> >> >
> >> > ========================================
> >> >
> >> > Justin A. Lemkul
> >> > Graduate Research Assistant
> >> > Department of Biochemistry
> >> > Virginia Tech
> >> > Blacksburg, VA
> >> > jalemkul.vt.edu | (540) 231-9080
> >> > http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/
> >> >
> >> > ========================================
> >> > -----------------------------------------------------------------------
> >> > The AMBER Mail Reflector
> >> > To post, send mail to amber.scripps.edu
> >> > To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> >> > to majordomo.scripps.edu
> >> >
> >>
> >> -----------------------------------------------------------------------
> >> The AMBER Mail Reflector
> >> To post, send mail to amber.scripps.edu
> >> To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> >> to majordomo.scripps.edu
> >>
> >
> >
> >
> > ========================================
> >
> > Justin A. Lemkul
> > Graduate Research Assistant
> > Department of Biochemistry
> > Virginia Tech
> > Blacksburg, VA
> > jalemkul.vt.edu | (540) 231-9080
> > http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/
> >
> > ========================================
> > -----------------------------------------------------------------------
> > The AMBER Mail Reflector
> > To post, send mail to amber.scripps.edu
> > To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> > to majordomo.scripps.edu
> >
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> to majordomo.scripps.edu
>
========================================
Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul.vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/
========================================
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
to majordomo.scripps.edu
Received on Sun Jun 01 2008 - 06:07:40 PDT
