On Fri 31-03-17 15:06:41, Hillf Danton wrote:
> On March 31, 2017 2:49 PM Michal Hocko wrote:
> > On Fri 31-03-17 11:49:49, Hillf Danton wrote:
> > [...]
> > > > -/* Can fail with -ENOMEM from allocating a wait table with vmalloc() or
> > > > - * alloc_bootmem_node_nopanic()/memblock_virt_alloc_node_nopanic() */
> > > > -static int __ref ensure_zone_is_initialized(struct zone *zone,
> > > > +static void __ref ensure_zone_is_initialized(struct zone *zone,
> > > > unsigned long start_pfn, unsigned long
> > > > num_pages)
> > > > {
> > > > - if (zone_is_empty(zone))
> > > > - return init_currently_empty_zone(zone, start_pfn,
> > > > num_pages);
> > > > -
> > > > - return 0;
> > > > + if (!zone_is_empty(zone))
> > > > + init_currently_empty_zone(zone, start_pfn, num_pages);
> > > > }
> > > Semantic change added?
> >
> > could you be more specific?
>
> Well, I'm wondering why you are trying to initiate a nonempty zone.
Ups, another fuck up during the initial patch split up. Thanks for
catching that but it would be more helpful to be more specific during
the feedback. I am getting blind to the code I am staring for quite some
time so it was not obvious what you mean here.
Thanks
--
Michal Hocko
SUSE Labs
