Re: [PATCH v3 2/4] mm: huge_memory: refactor anon_enabled_store() with change_anon_orders()

[Breno Leitao]

On Sun, Mar 08, 2026 at 02:05:30PM -0700, Andrew Morton wrote:
> On Sat, 07 Mar 2026 08:08:06 -0800 Breno Leitao <leitao@xxxxxxxxxx> wrote:
> > --- a/mm/huge_memory.c
> > +++ b/mm/huge_memory.c
> > @@ -316,6 +316,20 @@ static ssize_t enabled_show(struct kobject *kobj,
> >  	return sysfs_emit(buf, "%s\n", output);
> >  }
> >  
> > +enum anon_enabled_mode {
> > +	ANON_ENABLED_ALWAYS,
> > +	ANON_ENABLED_MADVISE,
> > +	ANON_ENABLED_INHERIT,
> > +	ANON_ENABLED_NEVER,
> > +};
> 
> If we're feeling fancy we could use
> 
> 	ANON_ENABLED_ALWAYS = 0,
> 	...
> 
> here, just to make it clear that we iterate over the enum values with
> an integer and we *require* those values.
> 
> > +static bool change_anon_orders(int order, enum anon_enabled_mode mode)
> > +{
> > +	static unsigned long *orders[] = {
> > +		&huge_anon_orders_always,
> > +		&huge_anon_orders_madvise,
> > +		&huge_anon_orders_inherit,
> > +	};
> > +	bool changed = false;
> > +	int i;
> 
> 	enum anon_enabled_mode m;
> 
> might be clearer here?

Thanks Andrew, let me respin it.

--breno