Re: [PATCH] mm/alloc: fallback to first node if the wanted node offline

[Pingfan Liu]

On Mon, Dec 10, 2018 at 12:00 PM Pingfan Liu <kernelfans@xxxxxxxxx> wrote:
>
> On Fri, Dec 7, 2018 at 11:56 PM Michal Hocko <mhocko@xxxxxxxxxx> wrote:
> >
> > On Fri 07-12-18 22:27:13, Pingfan Liu wrote:
> > [...]
> > > diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c
> > > index 1308f54..4dc497d 100644
> > > --- a/arch/x86/mm/numa.c
> > > +++ b/arch/x86/mm/numa.c
> > > @@ -754,18 +754,23 @@ void __init init_cpu_to_node(void)
> > >  {
> > >         int cpu;
> > >         u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);
> > > +       int node, nr;
> > >
> > >         BUG_ON(cpu_to_apicid == NULL);
> > > +       nr = cpumask_weight(cpu_possible_mask);
> > > +
> > > +       /* bring up all possible node, since dev->numa_node */
> > > +       //should check acpi works for node possible,
> > > +       for_each_node(node)
> > > +               if (!node_online(node))
> > > +                       init_memory_less_node(node);
> >
> > I suspect there is no change if you replace for_each_node by
> >         for_each_node_mask(nid, node_possible_map)
> >
> > here. If that is the case then we are probably calling
> > free_area_init_node too early. I do not see it yet though.
>
> Maybe I do not clearly get your meaning, just try to guess. But if you
> worry about node_possible_map, then it is dynamically set by
> alloc_node_data(). The map is changed after the first time to call

A mistake, it should be node_online_map. and in free_area_init_nodes()
for_each_online_node(nid) {
   free_area_init_node(nid, NULL,..

So at this time, we do not need to worry about the memory-less node.

> free_area_init_node() for the node with memory.  This logic is the
> same as the current x86 code.
>
> Thanks,
> Pingfan