workqueue: Reap workers via kthread_stop() and remove detach_completion

The code to kick off the destruction of workers is now in a process
context (idle_cull_fn()), so kthread_stop() can be used in the process
context to replace the work of pool->detach_completion.

The wakeup in wake_dying_workers() is unneeded after this change, but it
is harmless, jut keep it here until next patch renames wake_dying_workers()
rather than renaming it again and again.

Cc: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 23e424bda7c2f6748e8c6633d7abbb0b955c9bb0..9c5df39854352071216a9fa0c94b8f988d381451 100644 (file)
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -216,7 +216,6 @@ struct worker_pool {
        struct worker           *manager;       /* L: purely informational */
        struct list_head        workers;        /* A: attached workers */
        struct list_head        dying_workers;  /* A: workers about to die */
-       struct completion       *detach_completion; /* all workers detached */
 
        struct ida              worker_ida;     /* worker IDs for task name */
 
@@ -2696,7 +2695,6 @@ static void worker_attach_to_pool(struct worker *worker,
 static void worker_detach_from_pool(struct worker *worker)
 {
        struct worker_pool *pool = worker->pool;
-       struct completion *detach_completion = NULL;
 
        /* there is one permanent BH worker per CPU which should never detach */
        WARN_ON_ONCE(pool->flags & POOL_BH);
@@ -2707,15 +2705,10 @@ static void worker_detach_from_pool(struct worker *worker)
        list_del(&worker->node);
        worker->pool = NULL;
 
-       if (list_empty(&pool->workers) && list_empty(&pool->dying_workers))
-               detach_completion = pool->detach_completion;
        mutex_unlock(&wq_pool_attach_mutex);
 
        /* clear leftover flags without pool->lock after it is detached */
        worker->flags &= ~(WORKER_UNBOUND | WORKER_REBOUND);
-
-       if (detach_completion)
-               complete(detach_completion);
 }
 
 /**
@@ -2816,10 +2809,9 @@ static void unbind_worker(struct worker *worker)
 
 static void wake_dying_workers(struct list_head *cull_list)
 {
-       struct worker *worker, *tmp;
+       struct worker *worker;
 
-       list_for_each_entry_safe(worker, tmp, cull_list, entry) {
-               list_del_init(&worker->entry);
+       list_for_each_entry(worker, cull_list, entry) {
                unbind_worker(worker);
                /*
                 * If the worker was somehow already running, then it had to be
@@ -2835,6 +2827,17 @@ static void wake_dying_workers(struct list_head *cull_list)
        }
 }
 
+static void reap_dying_workers(struct list_head *cull_list)
+{
+       struct worker *worker, *tmp;
+
+       list_for_each_entry_safe(worker, tmp, cull_list, entry) {
+               list_del_init(&worker->entry);
+               kthread_stop_put(worker->task);
+               kfree(worker);
+       }
+}
+
 /**
  * set_worker_dying - Tag a worker for destruction
  * @worker: worker to be destroyed
@@ -2866,6 +2869,9 @@ static void set_worker_dying(struct worker *worker, struct list_head *list)
 
        list_move(&worker->entry, list);
        list_move(&worker->node, &pool->dying_workers);
+
+       /* get an extra task struct reference for later kthread_stop_put() */
+       get_task_struct(worker->task);
 }
 
 /**
@@ -2949,6 +2955,8 @@ static void idle_cull_fn(struct work_struct *work)
        raw_spin_unlock_irq(&pool->lock);
        wake_dying_workers(&cull_list);
        mutex_unlock(&wq_pool_attach_mutex);
+
+       reap_dying_workers(&cull_list);
 }
 
 static void send_mayday(struct work_struct *work)
@@ -3330,7 +3338,6 @@ woke_up:
                ida_free(&pool->worker_ida, worker->id);
                worker_detach_from_pool(worker);
                WARN_ON_ONCE(!list_empty(&worker->entry));
-               kfree(worker);
                return 0;
        }
 
@@ -4863,7 +4870,6 @@ static void rcu_free_pool(struct rcu_head *rcu)
  */
 static void put_unbound_pool(struct worker_pool *pool)
 {
-       DECLARE_COMPLETION_ONSTACK(detach_completion);
        struct worker *worker;
        LIST_HEAD(cull_list);
 
@@ -4917,12 +4923,9 @@ static void put_unbound_pool(struct worker_pool *pool)
 
        wake_dying_workers(&cull_list);
 
-       if (!list_empty(&pool->workers) || !list_empty(&pool->dying_workers))
-               pool->detach_completion = &detach_completion;
        mutex_unlock(&wq_pool_attach_mutex);
 
-       if (pool->detach_completion)
-               wait_for_completion(pool->detach_completion);
+       reap_dying_workers(&cull_list);
 
        /* shut down the timers */
        del_timer_sync(&pool->idle_timer);