Re: [PATCH 0/2] Implement numa node notifier

[Vlastimil Babka]

On 4/4/25 00:06, Harry Yoo wrote:
> On Thu, Apr 03, 2025 at 03:02:25PM +0200, David Hildenbrand wrote:
>> On 02.04.25 19:03, Oscar Salvador wrote:
>> > On Wed, Apr 02, 2025 at 06:06:51PM +0200, Vlastimil Babka wrote:
>> > > What if we had two chains:
>> > > 
>> > > register_node_notifier()
>> > > register_node_normal_notifier()
>> > > 
>> > > I think they could have shared the state #defines and struct node_notify
>> > > would have just one nid and be always >= 0.
>> > > 
>> > > Or would it add too much extra boilerplate and only slab cares?
>> > 
>> > We could indeed go on that direction to try to decouple
>> > status_change_nid from status_change_nid_normal.
>> > 
>> > Although as you said, slub is the only user of status_change_nid_normal
>> > for the time beign, so I am not sure of adding a second chain for only
>> > one user.
>> > 
>> > Might look cleaner though, and the advantatge is that slub would not get
>> > notified for nodes adquiring only ZONE_MOVABLE.
>> > 
>> > Let us see what David thinks about it.
>> 
>> I'd hope we'd be able to get rid of the _normal stuff completely, it's seems
>> way to specialized.
> 
> Hmm, perhaps we can remove it with as part of this patch series?
> 
> status_change_nid_normal has been used to indicate both 'There is a
> status change' AND 'The node id when the NUMA node has normal memory'.
> 
> But since NUMA node notifier triggers only when there is a state change,
> it can simply pass nid, like patch 2 does. SLUB can then check whether the
> node has normal memory.

Well the state change could be adding movable memory, SLUB checks that
there's no normal memory and thus does nothing. Then normal memory is added
to the node, but there's no new notification and SLUB is left without
supporting the node forever.

But with David's suggestion we avoid that problem.

> Or am I missing something?
> 
>> We added that in
>> 
>> commit b9d5ab2562eceeada5e4837a621b6260574dd11d
>> Author: Lai Jiangshan <laijs@xxxxxxxxxxxxxx>
>> Date:   Tue Dec 11 16:01:05 2012 -0800
>> 
>>     slub, hotplug: ignore unrelated node's hot-adding and hot-removing
>>     SLUB only focuses on the nodes which have normal memory and it ignores the
>>     other node's hot-adding and hot-removing.
>>     Aka: if some memory of a node which has no onlined memory is online, but
>>     this new memory onlined is not normal memory (for example, highmem), we
>>     should not allocate kmem_cache_node for SLUB.
>>     And if the last normal memory is offlined, but the node still has memory,
>>     we should remove kmem_cache_node for that node.  (The current code delays
>>     it when all of the memory is offlined)
>>     So we only do something when marg->status_change_nid_normal > 0.
>>     marg->status_change_nid is not suitable here.
>>     The same problem doesn't exist in SLAB, because SLAB allocates kmem_list3
>>     for every node even the node don't have normal memory, SLAB tolerates
>>     kmem_list3 on alien nodes.  SLUB only focuses on the nodes which have
>>     normal memory, it don't tolerate alien kmem_cache_node.  The patch makes
>>     SLUB become self-compatible and avoids WARNs and BUGs in rare conditions.
>> 
>> 
>> How "bad" would it be if we do the slab_mem_going_online_callback() etc even
>> for completely-movable nodes? I assume one kmem_cache_alloc() per slab_caches.
>>
>> slab_mem_going_offline_callback() only does shrinking, #dontcare
>> 
>> Looking at slab_mem_offline_callback(), we never even free the caches either
>> way when offlining. So the implication would be that we would have movable-only nodes
>> set in slab_nodes.
>> 
>> 
>> We don't expect many such nodes, so ... do we care?
>> 
>> -- 
>> Cheers,
>> 
>> David / dhildenb
>> 
>> 
>