[PATCH v5 01/14] slab: add opt-in caching layer of percpu sheaves
Vlastimil Babka
vbabka at suse.cz
Tue Aug 26 01:51:34 PDT 2025
On 8/19/25 06:19, Suren Baghdasaryan wrote:
>> @@ -5624,20 +6561,29 @@ static int init_kmem_cache_nodes(struct kmem_cache *s)
>>
>> for_each_node_mask(node, slab_nodes) {
>> struct kmem_cache_node *n;
>> + struct node_barn *barn = NULL;
>>
>> if (slab_state == DOWN) {
>> early_kmem_cache_node_alloc(node);
>> continue;
>> }
>> +
>> + if (s->cpu_sheaves) {
>> + barn = kmalloc_node(sizeof(*barn), GFP_KERNEL, node);
>> +
>> + if (!barn)
>> + return 0;
>> + }
>> +
>> n = kmem_cache_alloc_node(kmem_cache_node,
>> GFP_KERNEL, node);
>> -
>> if (!n) {
>> - free_kmem_cache_nodes(s);
>
> Why do you skip free_kmem_cache_nodes() here?
It's not necessary as the caller will perform __kmem_cache_release() which
calls free_kmem_cache_nodes()
I have incorporated your other suggestions, thanks!
More information about the maple-tree
mailing list