AMBER Archive (2008)

Subject: Re: AMBER: compiling PMEMD spits out lots of -tp not supported and hangs @ pmemd.f90

From: Starr Hazard (hazards_at_musc.edu)
Date: Fri Dec 05 2008 - 14:53:14 CST

I'm compiling on the headnode of my cluster.
The compiler is

 ifort -V
Intel(R) Fortran Intel(R) 64 Compiler Professional for applications
running on Intel(R) 64, Version 11.0 Build 20081105 Package ID:
l_cprof_p_11.0.074

which I just downloaded this week.

My cluster top (ganglia) command shows the command running is
fortcom its running @99.88% cpu and using 0.11% memory

command line top

14267 hazards 25 0 60996 16m 7020 R *100 0.1 * 32:29.83 fortcom

So, I do not think I am into swap space. This is the hardware
description from the invoice:

Quad Core Xeon Processor E53452x4MB Cache, 2.33GHz, 1333MHz FSB, PE2950
(222-7341)
Processor: Quad Core Xeon 2nd Processor E5345, 2x4MB Cache, 2.33GHz
1333MHz FSB,
PE2950
Memory: 16GB 667MHz (8x2GB), Dual Ranked DIMMs

cat /etc/redhat-release shows
Red Hat Enterprise Linux AS release 4 (Nahant Update 4)

I compiled pmemd just fine with a 10 something version of ifort.

Starr

Ross Walker wrote:
>> F90_OPT_LO = -tpp7 -O0
>> F90_OPT_MED = -tpp7 -O2
>> F90_OPT_HI = -tpp7 -xP -ip -O3
>> F90_OPT_DFLT = $(F90_OPT_HI)
>>
>
> You can remove the -tpp7 if you want but it is just being ignored right now
> - essentially new versions of the Intel compiler don't support the -tpp
> option and instead just select it automatically. The warnings you are
> getting are just warnings though so this isn't the problem.
>
>
>> make install hums along nicely up to this point and then hangs here for a
>> LONG time (up to 20 minutes)
>>
>> /lib/cpp -traditional -P -I/share/apps/mpich2-intel_11_install/include -
>> DPUBFFT -DBINTRAJ -DMPI -DSLOW_NONBLOCKING_MPI -DDIRFRC_EFS -
>> DDIRFRC_COMTRANS -DDIRFRC_NOVEC -DMKL -DFFTLOADBAL_2PROC pmemd.fpp
>> pmemd.f90
>> ifort -c -auto -tpp7 -xP -ip -O3 pmemd.f90
>> ifort: command line remark #10148: option '-tp' not supported
>>
>
> Which version of ifort is this? Can you do ifort -V. Also what are the specs
> for your actual machine?
>
> I suspect that you may be running out of memory and your machine may be
> swapping like crazy which is why it is taking so long. Try running top and
> see how much memory the compiler is using. If this is a 'brand new' version
> of the Intel compiler there may also be issues with it and we may need to
> find a workaround for it.
>
> More information is needed though before we can diagnose this further.
>
> For reference on my machine - a 2 x quad core Intel E5462 with 16 GB ram
> running Redhat EL4 and ifort 10.1.018 PMEMD compiles in less than a minute.
>
> All the best
> Ross
>
>
> /\
> \/
> |\oss Walker
>
> | Assistant Research Professor |
> | San Diego Supercomputer Center |
> | Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
> | http://www.rosswalker.co.uk | PGP Key available on request |
>
> Note: Electronic Mail is not secure, has no guarantee of delivery, may not
> be read every day, and should not be used for urgent or sensitive issues.
>
>
>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> to majordomo_at_scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu