[gmx-users] intel compiler + amd64: help needed.

[Jones de Andrade]

Hi all!.

First, sorry for the delay in getting in touch again. Bureaucracy. :P

Second, thanks a lot for everybody for advising on this subject. It's
helping me a lot to solve this! :)

Anyway, here is where I'm now:

Erik: I'm sure about the flags. Not using the ones for SSE3, despite my
processor having those. I'll just try to activate those as soon as I get a
compiling procedure. By now, I even using the --disable-x86-64-sse flag for
configure, just to be sure. ;) Also, I tried the "CC_FOR_BUILD=gcc"
environment, but despite it has gone throw the invsqrt compilation, I looked
in and it was compiling everything with gcc, and nothing with icc. :(

Yang: Yes, I'm using the intel compiler with emt64. I reinstalled the
software step by step specifically to be sure about this point. And I
understand that gromacs has an assembly core, and that this is basically the
"heart" of the program speed. But, even thow, 5-10% means something, and
every nano-second counts. Specially when you have a big delay to deal with,
and that this 10% mean 0.25ns. ;)  Even then, after I successfully compile
it we will be able to say for sure how much if extra performance (if it
yelds any) does the intel compiler gives on AMD64 machines. Moreover, as I
said in a long time ago message: I want to use QM/MM with gromacs+cpmd, and
at this point I really dislike the idea of compiling one thing with intel
(cpmd) and the other with gcc (gromacs). Since I can't make cpmd work on
gcc, I'm trying to make gromacs work on intel. ;)

David (last but not the least): I removed the "-ipo" flag from the
compilations. Also, I've added "--without-motif-libraries
--without-motif-includes --without-x" flags to the configure script. On the
other hand, I've been "greping" a bit the directories for the "invsqrt"
calls.... My God, it's not the heart of the program, it's the "sinapses"
among its neurones. Am I wrong, or there is about some thousands of those
calls to be changed? Is it right?

If it is, just let me know, cause I'm tracking those down (finding a script
way to back up original files before edition.) and at the same time trying
to find some sort of "magical flag" to convince icc to *not* use the
"invsqrt" function or the "math.h" file at all (second choice seems to me a
bit too strong). Seems that if it can be done on a flag it will be really
helpfull (and easier) for most of the users. The good new is that there is
no other error message in the ompilation, as I can see (And just a few
warnings to see later, since they look more or less pointless).

By the way, can I ask you what was the "mknb" program for, anyway? :)

Again, thanks a lot to everybody in advance, for all the help an advices
provided up to now! If I manage to finish this one, I promisse to post those
extra benchmarks too, as well as a "tutorial" on "how to compile gromacs on
AMD64 machines with non-sense copilers". ;)

Jones

On 6/1/06, David van der Spoel <spoel at xray.bmc.uu.se> wrote:
>
> > make[3]: ** [grompp] Erro 1
> > make[3]: Leaving directory
> > `/home/johannes/src/gromacs/gromacs-3.3/src/kernel'
> > make[2]: ** [all-recursive] Erro 1
> > make[2]: Leaving directory `/home/johannes/src/gromacs/gromacs-3.3/src'
> > make[1]: ** [all] Erro 2
> > make[1]: Leaving directory `/home/johannes/src/gromacs/gromacs-3.3/src'
> > make: ** [all-recursive] Erro 1
> >
> **********************************************************************************************
> >
> > Trying withou the "-ipo" flag now. It still yelds the same weird warning
> > (warning #147: declaration is incompatible with "double invsqrt(double)"
> > (declared at line 565 of "/usr/local/intel/include/math.h")
>
> rename all gromacs invsqrt calls to gmx_invsqrt
>
> be assured that invsqrt didn't exist in the intel compiler in previous
> releases, but now that it does we'll have to adapt it seems.
>
> ipo is useless for gromacs.
>
> do also turn off X11 at the configure stage
> --
> David.
> ________________________________________________________________________
> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,          75124 Uppsala, Sweden
> phone:  46 18 471 4205          fax: 46 18 511 755
> spoel at xray.bmc.uu.se    spoel at gromacs.org   http://folding.bmc.uu.se
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20060601/c5305d51/attachment.html>