workqueue: clear POOL_DISASSOCIATED in rebind_workers()

a9ab775bcadf ("workqueue: directly restore CPU affinity of workers
from CPU_ONLINE") moved pool locking into rebind_workers() but left
"pool->flags &= ~POOL_DISASSOCIATED" in workqueue_cpu_up_callback().

There is nothing necessarily wrong with it, but there is no benefit
either.  Let's move it into rebind_workers() and achieve the following
benefits:

  1) better readability, POOL_DISASSOCIATED is cleared in rebind_workers()
     as expected.

  2) we can guarantee that, when POOL_DISASSOCIATED is clear, the
     running workers of the pool are on the local CPU (pool->cpu).

tj: Minor description update.

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 8474e5752f1f8df25d30b9a95712bd55a2caeb73..68461b8d9b39d4f71f86d4c3b6ab95a9483755f8 100644 (file)
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -4535,6 +4535,7 @@ static void rebind_workers(struct worker_pool *pool)
                                                  pool->attrs->cpumask) < 0);
 
        spin_lock_irq(&pool->lock);
+       pool->flags &= ~POOL_DISASSOCIATED;
 
        for_each_pool_worker(worker, pool) {
                unsigned int worker_flags = worker->flags;
@@ -4637,10 +4638,6 @@ static int workqueue_cpu_up_callback(struct notifier_block *nfb,
                        mutex_lock(&pool->attach_mutex);
 
                        if (pool->cpu == cpu) {
-                               spin_lock_irq(&pool->lock);
-                               pool->flags &= ~POOL_DISASSOCIATED;
-                               spin_unlock_irq(&pool->lock);
-
                                rebind_workers(pool);
                        } else if (pool->cpu < 0) {
                                restore_unbound_workers_cpumask(pool, cpu);