Re: [AMBER-Developers] problems with latest updates to CUDA code

From: Jason Swails <jason.swails.gmail.com>
Date: Thu, 10 Feb 2011 11:29:49 -0500

Ah, quite a reasonable thing to try. This eliminates the issues and it
compiles just fine. I'll drop those from the configure script if you
(anyone) would like.

Thanks!
Jason

On Thu, Feb 10, 2011 at 11:16 AM, Scott Le Grand <SLeGrand.nvidia.com>wrote:

> So what happens if you remove ipo from the compilation?
>
>
> -----Original Message-----
> From: Jason Swails [mailto:jason.swails.gmail.com]
> Sent: Thursday, February 10, 2011 08:11
> To: AMBER Developers Mailing List
> Subject: Re: [AMBER-Developers] problems with latest updates to CUDA code
>
> The default install rule does a clean (including a netcdf_clean) first, so
> I
> don't think this is an issue. In any case, the errors I'm getting are
>
> ipo: warning #11020: unresolved gpu_get_nb_energy_
> Referenced in /tmp/ipo_ifortLpW9vm.o
> ipo: warning #11020: unresolved gpu_vdw_correction_
> Referenced in /tmp/ipo_ifortLpW9vm.o
> ipo: warning #11020: unresolved gpu_self_
> Referenced in /tmp/ipo_ifortLpW9vm.o
> ... etc.
>
> which doesn't appear to be netcdf-related.
>
> Thanks!
> Jason
>
> On Thu, Feb 10, 2011 at 11:04 AM, Scott Le Grand <SLeGrand.nvidia.com
> >wrote:
>
> > You've done a make clean right?
> >
> > If you google this issue, this has been seen before:
> >
> > https://www.myroms.org/forum/viewtopic.php?f=31&t=269
> >
> > -----Original Message-----
> > From: Jason Swails [mailto:jason.swails.gmail.com]
> > Sent: Thursday, February 10, 2011 07:59
> > To: AMBER Developers Mailing List
> > Subject: Re: [AMBER-Developers] problems with latest updates to CUDA code
> >
> > I'm getting the same errors on my box
> >
> > ifort 11.1.069, nvcc release 3.2, V0.2.1221
> >
> > GNU compilers (4.4.1) work just fine. Stupid intel...
> >
> > --Jason
> >
> > On Thu, Feb 10, 2011 at 9:29 AM, Ross Walker <ross.rosswalker.co.uk>
> > wrote:
> >
> > > Ross is on travel. I will try it next week.
> > >
> > > > -----Original Message-----
> > > > From: Scott Le Grand [mailto:SLeGrand.nvidia.com]
> > > > Sent: Thursday, February 10, 2011 6:26 AM
> > > > To: amber-developers.ambermd.org
> > > > Subject: Re: [AMBER-Developers] problems with latest updates to CUDA
> > > > code
> > > >
> > > > Works at this end with either GNU or Intel, Ross?
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: case [mailto:case.biomaps.rutgers.edu]
> > > > Sent: Thursday, February 10, 2011 06:15
> > > > To: amber-developers.ambermd.org
> > > > Subject: [AMBER-Developers] problems with latest updates to CUDA code
> > > >
> > > > I'm having trouble with the latest git updates to the cuda code:
> > > >
> > > > make[3]: Leaving directory `/home/case/amber11/src/pmemd/src/cuda'
> > > > ifort -ipo -O3 -no-prec-div -xHost -DCUDA -o pmemd.cuda
> > gbl_constants.o
> > > > gbl_datatypes.o state_info.o file_io_dat.o mdin_ctrl_dat.o
> > > > mdin_ewald_dat.o
> > > > mdin_debugf_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 nextprmtop_section.o
> angles_ub.o
> > > > dihedrals_imp.o cmap.o charmm.o charmm_gold.o -L/usr/local/cuda/lib64
> > > > -L/usr/local/cuda/lib -lcufft -lcudart ./cuda/cuda.a
> > > > /home/case/amber11/lib/libnetcdf.a -shared-intel -Wl,--start-group
> > > > /cottus/opt/intel/Compiler/11.1/069/mkl/lib/em64t/libmkl_intel_lp64.a
> > > > /cottus/opt/intel/Compiler/11.1/069/mkl/lib/em64t/libmkl_sequential.a
> > > > /cottus/opt/intel/Compiler/11.1/069/mkl/lib/em64t/libmkl_core.a
> > > > -Wl,--end-group -lpthread
> > > > ipo: warning #11020: unresolved gpu_get_nb_energy_
> > > > Referenced in /tmp/ipo_ifort1hg2Wz.o
> > > > ipo: warning #11020: unresolved gpu_vdw_correction_
> > > > Referenced in /tmp/ipo_ifort1hg2Wz.o
> > > >
> > > > ....etc
> > > >
> > > > The problem appears to be with gpu.cpp. When compiled with either
> icc
> > > > (which is what the Makefile uses) or icpc, the resulting gpu.o does
> not
> > > have
> > > > any symbols in it(!?!). I can't see the problem. Intel compiler
> > version
> > > > is 11.1.069.
> > > >
> > > > Using gnu compilers seems fine, so I can go that way. But intel
> > > compilers
> > > > *had* been working, up until this last update.
> > > >
> > > > ....dac
> > > >
> > > >
> > > > _______________________________________________
> > > > AMBER-Developers mailing list
> > > > AMBER-Developers.ambermd.org
> > > > http://lists.ambermd.org/mailman/listinfo/amber-developers
> > > >
> > >
> > >
> >
> ----------------------------------------------------------------------------
> > > -------
> > > > This email message is for the sole use of the intended recipient(s)
> and
> > > may
> > > > contain
> > > > confidential information. Any unauthorized review, use, disclosure
> or
> > > > distribution
> > > > is prohibited. If you are not the intended recipient, please contact
> > the
> > > > sender by
> > > > reply email and destroy all copies of the original message.
> > > >
> > >
> > >
> >
> ----------------------------------------------------------------------------
> > > -------
> > > >
> > > > _______________________________________________
> > > > 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
> > >
> >
> >
> >
> > --
> > Jason M. Swails
> > Quantum Theory Project,
> > University of Florida
> > Ph.D. Graduate Student
> > 352-392-4032
> > _______________________________________________
> > 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
> >
>
>
>
> --
> Jason M. Swails
> Quantum Theory Project,
> University of Florida
> Ph.D. Graduate Student
> 352-392-4032
> _______________________________________________
> 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
>



-- 
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
352-392-4032
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Thu Feb 10 2011 - 08:30:04 PST
Custom Search