On 2023/7/24 17:43, Qi Zheng wrote:
In preparation for implementing lockless slab shrink, use new APIs to dynamically allocate the s_shrink, so that it can be freed asynchronously using kfree_rcu(). Then it doesn't need to wait for RCU read-side critical section when releasing the struct super_block. Signed-off-by: Qi Zheng <zhengqi.a...@bytedance.com>
Reviewed-by: Muchun Song <songmuc...@bytedance.com>
