[Debichem-devel] GROMACS 4.6-beta1 release

Dominik 'Rathann' Mierzejewski dominik at greysector.net
Thu Dec 20 23:40:34 UTC 2012


Hello Mark,

On Thursday, 20 December 2012 at 00:19, Mark Abraham wrote:
> On Tue, Dec 18, 2012 at 9:18 AM, Susi Lehtola <
> jussilehtola at fedoraproject.org> wrote:
[...]
> > On another note, it seems that the cmake script detects the CPU on the
> > compile system and produces an optimized version for the specific CPU.
> > Naturally this is somewhat problematic for distributions, since e.g. an
> > x86_64 binary should run on all x86_64 systems. Is this the case or not?
> 
> The default build produces an executable that targets the best instruction
> set supported by the build host, and the executables will not run if that
> set is not supported. From the point of view of packaging, that's ugly.
> Either there needs to be multiple packages, or you will need to choose an
> instruction set that is conservative enough to always work.
> -DGMX_CPU_ACCELERATION=None will always work, but will be horribly
> slow. -DGMX_CPU_ACCELERATION=SSE2 probably runs about as fast as the 4.x
> series, but is probably faster through our better use of force-only
> kernels. -DGMX_CPU_ACCELERATION=SSE4.1 is only slightly faster than SSE2.
> And then there is AVX, which is faster still, but stil having teething
> issues.
>
> I'd suggest having an SSE2 and a None, but it really depends on what the
> downstream people want.

For x86_64, SSE2 is supported by all processors, so it shouldn't be an
issue, but for x86_32, the binary should run even on an ancient Pentium
Pro (Fedora x86_32 minimum CPU requirement). Have you considered
implementing runtime CPU detection?

Regards,
Dominik

-- 
Fedora http://fedoraproject.org/wiki/User:Rathann
RPMFusion http://rpmfusion.org | MPlayer http://mplayerhq.hu
"Faith manages."
        -- Delenn to Lennier in Babylon 5:"Confessions and Lamentations"



More information about the Debichem-devel mailing list