u_mixed = interpolate(Constant((1.,2.,3.)), M)
u_mixed.vector().array()
produces output
array([ 1., 2., 3., 1., 2., 3., 1., 2., 3., 1., 2., 3., 1.,
2., 3., 1., 2., 3., 1., 2., 3., 1., 2., 3., 1., 2., 3.])
so answer is no.
If you're trying to handle somehow assignment to subfunction then note that the issue is now in progress. Also note that standard way of handling assignment to subfunction is for now projection or interpolation. If you want to fiddle directly with DOFs, use helper functions in M.dofmap()
.