[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20201026134010.GA73258@mtj.duckdns.org>
Date: Mon, 26 Oct 2020 09:41:14 -0400
From: Tejun Heo <tj@...nel.org>
To: qiang.zhang@...driver.com
Cc: jiangshanlai@...il.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] workqueue: replace call_rcu with kfree_rcu
On Wed, Oct 14, 2020 at 04:31:45PM +0800, qiang.zhang@...driver.com wrote:
> diff --git a/kernel/workqueue.c b/kernel/workqueue.c
> index ac088ce6059b..8d4fe649631a 100644
> --- a/kernel/workqueue.c
> +++ b/kernel/workqueue.c
> @@ -3653,11 +3653,6 @@ static struct worker_pool *get_unbound_pool(const struct workqueue_attrs *attrs)
> return NULL;
> }
>
> -static void rcu_free_pwq(struct rcu_head *rcu)
> -{
> - kmem_cache_free(pwq_cache,
> - container_of(rcu, struct pool_workqueue, rcu));
> -}
>
> /*
> * Scheduled on system_wq by put_pwq() when an unbound pwq hits zero refcnt
> @@ -3683,7 +3678,7 @@ static void pwq_unbound_release_workfn(struct work_struct *work)
> put_unbound_pool(pool);
> mutex_unlock(&wq_pool_mutex);
>
> - call_rcu(&pwq->rcu, rcu_free_pwq);
> + kfree_rcu(pwq, rcu);
kfree_rcu() eventually calls kfree() which is different from
kmem_cache_free(). It can only be used for kmalloc'd areas, not
kmem_cache_alloc'd ones.
Thanks.
--
tejun
Powered by blists - more mailing lists