I think the Walker group must have fixed the missing comma; it is not in my
latest drop of the source, but I would have certainly expected it to cause a
bunch of grief if it were ever really there. I think the workaround in
pme_setup.fpp was to use a : only, natom, atm_iac in vdw_correct_setup(),
which while not strictly necessary and at one level a nuisance, at another
level makes for cleaner code. The code you list below, I suspect, is some
transitional snapshot and I would guess you need to update. Ross' group may
have more pertinent info, as they are the ones doing the work to get charmm
support into pmemd.
Regards - Bob Duke
----- Original Message -----
From: "Josh Berryman" <the.real.josh.berryman.gmail.com>
To: <amber.ambermd.org>
Sent: Friday, April 16, 2010 9:25 AM
Subject: [AMBER] PMEMD missing type? - bugfix
> Hello, apologies for cross-post from developer's mailing list; but I am
> not
> a developer.
>
> Anyway, the bug in question is where recent versions of ifc (11.1 in my
> case) refuse to compile what the devs insist is perfect fortran, giving
> the
> error message:
>
> pme_setup.f90(415): error #6404: This name does not have a type, and must
> have an explicit type. [NATOM]
> do n = 1, natom
>
> I know that it is not the amber team's responsibility to run after every
> wrinkle in what is increasingly becoming known as a pretty sloppy fortran
> compiler, but for those of us who, as usual, just want a fix NOW the
> changes
> needed to the code are:
>
> in pme_setup.fpp:
>
> module pme_setup_mod
>
> use gbl_datatypes_mod
>
> to:
>
> module pme_setup_mod
>
> use gbl_datatypes_mod
> use prmtop_dat_mod
>
> And also:
>
> in prmtop_dat.fpp:
>
> gbl_dihed_allocsize, next_mult_fac & !26
> nub, nubtypes, & !28
>
> to:
>
> gbl_dihed_allocsize, next_mult_fac, & !26
> nub, nubtypes, & !28
>
> inserting a single comma.
>
>
> Anyway, I hope this helps someone out there
>
> Josh
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 16 2010 - 07:30:03 PDT
