Re: Make NAMD2.7b1 with cuda bug?

From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Wed May 13 2009 - 10:03:44 CDT

On Wed, 2009-05-13 at 10:29 -0400, Charles Taylor wrote:

guys,

i just successfully build a CUDA-2.2 version from the NAMD cvs code
without any NAMD related problems. from looking at the change log
the attached patch is required to fix those compilation problems
properly.

hope this helps,
    axel.

ps.: roman, FYI, i'm getting similar speedups with a GTX 260
card compared to what you reported, but this is with a very
unusual input which seems to keep a lot of work on the CPU
and thus restricts the speedup.

> I had the same problem when I built it. I just moved the declaration
> of idlepoll outside the subroutine so it would be global within the
> scope of the source file. That worked but is doesn't seem to be what
> the developers intended. Something got lost somewhere in the
> distribution.
>
> Charlie Taylor
> UF HPC Center
>
> On May 13, 2009, at 9:56 AM, Roman Petrenko wrote:
>
> > Good question, what is this flag for? When running namd on cuda, it is
> > a required flag.
> >
> >
> > On Tue, May 12, 2009 at 11:48 PM, Lev Lafayette <lev_at_vpac.org> wrote:
> >> Hello NAMDers,
> >>
> >> I have just installed NAMD2.7b with cuda on our cluster. However
> >> during the install process make failed because idlepoll was not
> >> defined in src/BackEnd.C (line 135)
> >>
> >> The install is with openmpi/1.3.2-intel and intel/11.0 compilers.
> >>
> >> The configure command was ./config Linux-x86_64-icc --charm-arch
> >> mpi-linux-x86_64 --with-cuda
> >>
> >> An attempt to make however results with:
> >>
> >> src/BackEnd.C(135): error: identifier "idlepoll" is undefined.
> >>
> >> As a dirty hack I've commented out the relevant lines in src/
> >> BackEnd.C
> >>
> >>
> >> //#ifdef NAMD_CUDA
> >> // if ( ! idlepoll ) {
> >> // NAMD_die("Please add +idlepoll to command line for proper
> >> performance.");
> >> // }
> >> //#endif
> >>
> >> . and it seems to make just fine after this and test jobs work.
> >>
> >>
> >> Regards,
> >>
> >> --
> >> Lev Lafayette (03) 9925 4909, Systems Administrator, Victorian
> >> Partnership for Advanced Computing
> >> "If you are sending a reply to a message or a posting be sure you
> >> summarize the original at the top of the message, or include just
> >> enough text of the original to give a context. This will make sure
> >> readers understand when they start to read your response."
> >> RFC 1855
> >>
> >>
> >
> >
> >
> > --
> > Roman Petrenko
> > Physics Department
> > University of Cincinnati
> >

-- 
=======================================================================
Axel Kohlmeyer   akohlmey_at_cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:57 CST