Re: [V5 PATCH 08/26] memcontrol: use N_MEMORY instead N_HIGH_MEMORY

[Michal Hocko <mhocko-AlSwsSmVLrQ@public.gmane.org>]

On Mon 29-10-12 14:08:05, David Rientjes wrote:
> On Mon, 29 Oct 2012, Michal Hocko wrote:
> 
> > > > > N_HIGH_MEMORY stands for the nodes that has normal or high memory.
> > > > > N_MEMORY stands for the nodes that has any memory.
> > > > 
> > > > What is the difference of those two?
> > > > 
> > > 
> > > Patch 5 in the series 
> > 
> > Strange, I do not see that one at the mailing list.
> > 
> 
> http://marc.info/?l=linux-kernel&m=135152595827692

Thanks!

> > > introduces it to be equal to N_HIGH_MEMORY, so 
> > 
> > So this is just a rename? If yes it would be much esier if it was
> > mentioned in the patch description.
> > 
> 
> It's not even a rename even though it should be, it's adding yet another 
> node_states that is equal to N_HIGH_MEMORY since that state already 
> includes all memory.  

Which is really strange because I do not see any reason for yet another
alias if the follow up patches rename all of them (I didn't try to apply
the whole series to check that so I might be wrong here).

> It's just a matter of taste but I think we should be renaming it
> instead of aliasing it (unless you actually want to make N_HIGH_MEMORY
> only include nodes with highmem, but nothing depends on that).

Agreed, I've always considered N_HIGH_MEMORY misleading and confusing so
renaming it would really make a lot of sense to me.
-- 
Michal Hocko
SUSE Labs