[gmx-users] Are water molecules restrained "by default"?

[Dawid das]

Dear Gromacs Experts,

I read in Gromacs documentation that
#POSRES_WATER
will be interpreted to restrain position of all water oxygen atoms. However
if I understand it correctly I should define where proper *itp file is
present as it is for restraining heavy atoms of my protein molecule:

; Include Position restraint file
#ifdef POSRES
#include "posre.itp"
#endif

I neither use #include mechanism for water oxygens nor #define.
However I do not get any errors as the documentation suggest when
POSRES_WATER is not defined.
So my question is whether using this what grompp has generated regarding
restraining water oxygens:
#ifdef POSRES_WATER
; Position restraint for each water oxygen
[ position_restraints ]
;  i funct       fcx        fcy        fcz
   1    1       1000       1000       1000
#endif

is enough to actually restrain those atoms? Or my atoms are not restrained
after all?
However if they are, how does Gromacs know what atoms are water oxygen
atoms? Simply it "recognizes" them because POSRES_WATER flag is used and
somewhere in my *top file I include *itp file for water molecues:
#include "charmm27.ff/tip3p.itp"
?

Best wishes,

Dawid Grabarek