AMBER Archive (2004)

Subject: Re: AMBER: Intel fortran compiler and -static switch in older pmemd releases

From: Robert Duke (rduke_at_email.unc.edu)
Date: Fri Apr 02 2004 - 15:35:45 CST

David -
Yes, I played with some of this stuff when dinking around with an AMD64
system and building in IA32 compatibility mode. Point is, there really is
not clear info about what the compiler does by default, what it is sane to
do, and what actually happened when you are done. I looked at this stuff
for a while and was most disappointed; I did not want to spend the time
figuring out all the ins and outs, and was fairly sure that the average
amber user would not want to either. In my opinion, the standards of
documentation and quality for ifc / ifort are just abysmal, but we are stuck
with it because it is the fastest thing out there; of course I am griping to
the wrong folks. So I am mostly in the mode of trying to reduce the extent
that people have grief with this stuff.
Regards - Bob

----- Original Message -----
From: "David E. Konerding DSD Staff" <dekonerding_at_lbl.gov>
To: <amber_at_scripps.edu>
Sent: Friday, April 02, 2004 4:21 PM
Subject: Re: AMBER: Intel fortran compiler and -static switch in older pmemd
releases

> Robert Duke wrote:
>
> >Folks -
> >Just a heads-up on a potential source of grief. In earlier releases of
> >pmemd, I included the -static switch in the ifc machine files because it
> >helped avoid problems with needing to distribute the compiler libraries
on
> >all the machines pmemd was running on. However, with later if not
earlier
> >releases of ifc, there has been a lot of ill-defined grief associated
with
> >the interface to system libraries. Now, in an ideal world, there would
be a
> >clear delineation between the compiler-related libraries and the system
> >libraries, and you would be able to have the right thing happening by
using
> >static compiler libraries and dynamic system libraries. The dynamic
system
> >libraries would present consistent call interfaces and behaviour, and if
> >they didn't, there would be provision for versioning.
> >
> >
>
> Hi Bob,
>
> I don't have time to look into this in detail, but in my experience, the
> linker *is* capable of linking in some libraries statically and others
> dynamically.
> However, this normally applies to user libraries, not compiler or system
> libraries. I think you pass something to the compile like:
>
>
>
> # gcc a.o -Wl,-Bstatic -lfoo -Wl,-Bdynamic -lbar
>
> the switch works that anything after -Wl,-Bstatic will be linked
statically until you hit the -Bdynamic
> line, at which point further libs would be linked dynamically.
>
> Unfrotunately, I suspect you'd have to tell the compiler not to link any
standard libraries in and
> specify them by hand on the final link step, which obviously isn't
desireable.
>
> Dave
>
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu