update for fix gle docs from michele ceriotti

doc/src/fix_gle.txt

@@ -67,9 +67,10 @@ target value as the {Tstart} and {Tstop} arguments, so that the diffusion
 matrix that gives canonical sampling for a given A is computed automatically.
 However, the GLE framework also allow for non-equilibrium sampling, that
 can be used for instance to model inexpensively zero-point energy
-effects "(Ceriotti2)"_#Ceriotti2. This is achieved specifying the
-{noneq} keyword followed by the name of the file that contains the
-static covariance matrix for the non-equilibrium dynamics.
+effects "(Ceriotti2)"_#Ceriotti2. This is achieved specifying the {noneq}
+ keyword followed by the name of the file that contains the static covariance
+matrix for the non-equilibrium dynamics.  Please note, that the covariance
+matrix is expected to be given in [temperature units].
 
 Since integrating GLE dynamics can be costly when used together with
 simple potentials, one can use the {every} optional keyword to
@@ -148,7 +149,7 @@ dpd/tstat"_pair_dpd.html, "fix gld"_fix_gld.html
 1170-80 (2010)
 
 :link(GLE4MD)
-[(GLE4MD)] "http://epfl-cosmo.github.io/gle4md/"_http://epfl-cosmo.github.io/gle4md/
+[(GLE4MD)] "http://gle4md.org/"_http://gle4md.org/
 
 :link(Ceriotti2)
 [(Ceriotti2)] Ceriotti, Bussi and Parrinello, Phys Rev Lett 103,