Re: [PATCH 1/3] slab: decouple pointer to barn from kmem_cache_node

[Vlastimil Babka (SUSE)]

On 3/13/26 10:27, Harry Yoo wrote:
> On Wed, Mar 11, 2026 at 09:25:55AM +0100, Vlastimil Babka (SUSE) wrote:
>> The pointer to barn currently exists in struct kmem_cache_node. That
>> struct is instantiated for every NUMA node with memory, but we want to
>> have a barn for every online node (including memoryless).
>> 
>> Thus decouple the two structures. In struct kmem_cache we have an array
>> for kmem_cache_node pointers that appears to be sized MAX_NUMNODES but
>> the actual size calculation in kmem_cache_init() uses nr_node_ids.
>> Therefore we can't just add another array of barn pointers. Instead
>> change the array to newly introduced struct kmem_cache_per_node_ptrs
>> holding both kmem_cache_node and barn pointer.
>> 
>> Adjust barn accessor and allocation/initialization code accordingly. For
>> now no functional change intended, barns are created 1:1 together with
>> kmem_cache_nodes.
>> 
>> Signed-off-by: Vlastimil Babka (SUSE) <vbabka@xxxxxxxxxx>
>> ---
>>  mm/slab.h |   7 +++-
>>  mm/slub.c | 128 +++++++++++++++++++++++++++++++++++---------------------------
>>  2 files changed, 78 insertions(+), 57 deletions(-)
>> 
>> diff --git a/mm/slab.h b/mm/slab.h
>> index e9ab292acd22..c735e6b4dddb 100644
>> --- a/mm/slab.h
>> +++ b/mm/slab.h
>> @@ -247,7 +252,7 @@ struct kmem_cache {
>>  	struct kmem_cache_stats __percpu *cpu_stats;
>>  #endif
>>  
>> -	struct kmem_cache_node *node[MAX_NUMNODES];
>> +	struct kmem_cache_per_node_ptrs per_node[MAX_NUMNODES];
>>  };
> 
> We should probably turn this into a true flexible array at some point,
> but that's out of scope for this patchset.

Right.

>> diff --git a/mm/slub.c b/mm/slub.c
>> index 20cb4f3b636d..609a183f8533 100644
>> --- a/mm/slub.c
>> +++ b/mm/slub.c
>> @@ -436,26 +436,24 @@ struct kmem_cache_node {
>>  /*
>> - * Get the barn of the current cpu's closest memory node. It may not exist on
>> - * systems with memoryless nodes but without CONFIG_HAVE_MEMORYLESS_NODES
>> + * Get the barn of the current cpu's memory node. It may be a memoryless node.
>>   */
>>  static inline struct node_barn *get_barn(struct kmem_cache *s)
>>  {
>> -	struct kmem_cache_node *n = get_node(s, numa_mem_id());
>> -
>> -	if (!n)
>> -		return NULL;
>> -
>> -	return n->barn;
>> +	return get_barn_node(s, numa_node_id());
>>  }
> 
> Previously, memoryless nodes on architectures w/ CONFIG_HAVE_MEMORYLESS_NODES
> shared the barn of the nearest NUMA node with memory.
> 
> But now memoryless nodes will have their own barns (after patch 2)
> regardless of CONFIG_HAVE_MEMORYLESS_NODES, and that's intentional, right?

Yeah it improves their caching capacity, but good point, will mention it in
the changelog.

> Otherwise LGTM!
>