I am sorry that i did not consider the memory hotplug case,
and i think we should add a new param to distinguish two different cases.
At 2020-06-30 04:24:29, "Mike Rapoport" <rppt@xxxxxxxxxx> wrote: >On Mon, Jun 29, 2020 at 11:50:15AM -0700, David Rientjes wrote: >> On Tue, 30 Jun 2020, Su Hui wrote: >> >> > remove the redundancy code, the zone_start_pfn >> > is assigned from zone->zone_start_pfn >> > Signed-off-by: Su Hui <sh_def@xxxxxxx> >> >> I don't think this is redundant, it's used by memory hotplug when onlining >> new memory. > >Right, it is: > >$ git grep -wn init_currently_empty_zone mm/memory_hotplug.c >mm/memory_hotplug.c:697: init_currently_empty_zone(zone, start_pfn, nr_pages); > > >> > --- >> > mm/page_alloc.c | 2 -- >> > 1 file changed, 2 deletions(-) >> > >> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c >> > index 3c4eb750a199..3372a8c9fbc4 100644 >> > --- a/mm/page_alloc.c >> > +++ b/mm/page_alloc.c >> > @@ -6215,8 +6215,6 @@ void __meminit init_currently_empty_zone(struct zone *zone, >> > if (zone_idx > pgdat->nr_zones) >> > pgdat->nr_zones = zone_idx; >> > >> > - zone->zone_start_pfn = zone_start_pfn; >> > - >> > mminit_dprintk(MMINIT_TRACE, "memmap_init", >> > "Initialising map node %d zone %lu pfns %lu -> %lu\n", >> > pgdat->node_id, >> > >-- >Sincerely yours, >Mike.
