workqueue: kick a worker in pwq_adjust_max_active()

If pwq_adjust_max_active() changes max_active from 0 to
saved_max_active, it needs to wakeup worker.  This is already done by
thaw_workqueues().

If pwq_adjust_max_active() increases max_active for an unbound wq,
while not strictly necessary for correctness, it's still desirable to
wake up a worker so that the requested concurrency level is reached
sooner.

Move wake_up_worker() call from thaw_workqueues() to
pwq_adjust_max_active() so that it can handle both of the above two
cases.  This also makes thaw_workqueues() simpler.

tj: Updated comments and description.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index d2ac6cb..79d1d34 100644 (file)
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -3598,6 +3598,12 @@ static void pwq_adjust_max_active(struct pool_workqueue *pwq)
                while (!list_empty(&pwq->delayed_works) &&
                       pwq->nr_active < pwq->max_active)
                        pwq_activate_first_delayed(pwq);
+
+               /*
+                * Need to kick a worker after thawed or an unbound wq's
+                * max_active is bumped.  It's a slow path.  Do it always.
+                */
+               wake_up_worker(pwq->pool);
        } else {
                pwq->max_active = 0;
        }
@@ -4401,13 +4407,6 @@ void thaw_workqueues(void)
        }
        spin_unlock_irq(&pwq_lock);
 
-       /* kick workers */
-       for_each_pool(pool, pi) {
-               spin_lock_irq(&pool->lock);
-               wake_up_worker(pool);
-               spin_unlock_irq(&pool->lock);
-       }
-
        workqueue_freezing = false;
 out_unlock:
        mutex_unlock(&wq_mutex);