Merge branch 'for-3.16' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq

Pull workqueue updates from Tejun Heo:
 "Lai simplified worker destruction path and internal workqueue locking
  and there are some other minor changes.

  Except for the removal of some long-deprecated interfaces which
  haven't had any in-kernel user for quite a while, there shouldn't be
  any difference to workqueue users"

* 'for-3.16' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
  kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info
  workqueue: remove the confusing POOL_FREEZING
  workqueue: rename first_worker() to first_idle_worker()
  workqueue: remove unused work_clear_pending()
  workqueue: remove unused WORK_CPU_END
  workqueue: declare system_highpri_wq
  workqueue: use generic attach/detach routine for rescuers
  workqueue: separate pool-attaching code out from create_worker()
  workqueue: rename manager_mutex to attach_mutex
  workqueue: narrow the protection range of manager_mutex
  workqueue: convert worker_idr to worker_ida
  workqueue: separate iteration role from worker_idr
  workqueue: destroy worker directly in the idle timeout handler
  workqueue: async worker destruction
  workqueue: destroy_worker() should destroy idle workers only
  workqueue: use manager lock only to protect worker_idr
  workqueue: Remove deprecated system_nrt[_freezable]_wq
  workqueue: Remove deprecated flush[_delayed]_work_sync()
  kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info
  workqueue: simplify wq_update_unbound_numa() by jumping to use_dfl_pwq if the target cpumask equals wq's

3 files changed, 154 insertions, 336 deletions

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 1b22c42e9c2d..a0cc2e95ed1b 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -56,9 +56,8 @@ enum {
 	WORK_NR_COLORS		= (1 << WORK_STRUCT_COLOR_BITS) - 1,
 	WORK_NO_COLOR		= WORK_NR_COLORS,
 
-	/* special cpu IDs */
+	/* not bound to any CPU, prefer the local CPU */
 	WORK_CPU_UNBOUND	= NR_CPUS,
-	WORK_CPU_END		= NR_CPUS + 1,
 
 	/*
 	 * Reserve 7 bits off of pwq pointer w/ debugobjects turned off.
@@ -274,13 +273,6 @@ static inline unsigned int work_static(struct work_struct *work) { return 0; }
 #define delayed_work_pending(w) \
 	work_pending(&(w)->work)
 
-/**
- * work_clear_pending - for internal use only, mark a work item as not pending
- * @work: The work item in question
- */
-#define work_clear_pending(work) \
-	clear_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(work))
-
 /*
  * Workqueue flags and constants.  For details, please refer to
  * Documentation/workqueue.txt.
@@ -340,6 +332,9 @@ enum {
  * short queue flush time.  Don't queue works which can run for too
  * long.
  *
+ * system_highpri_wq is similar to system_wq but for work items which
+ * require WQ_HIGHPRI.
+ *
  * system_long_wq is similar to system_wq but may host long running
  * works.  Queue flushing might take relatively long.
  *
@@ -358,26 +353,13 @@ enum {
  * 'wq_power_efficient' is disabled.  See WQ_POWER_EFFICIENT for more info.
  */
 extern struct workqueue_struct *system_wq;
+extern struct workqueue_struct *system_highpri_wq;
 extern struct workqueue_struct *system_long_wq;
 extern struct workqueue_struct *system_unbound_wq;
 extern struct workqueue_struct *system_freezable_wq;
 extern struct workqueue_struct *system_power_efficient_wq;
 extern struct workqueue_struct *system_freezable_power_efficient_wq;
 
-static inline struct workqueue_struct * __deprecated __system_nrt_wq(void)
-{
-	return system_wq;
-}
-
-static inline struct workqueue_struct * __deprecated __system_nrt_freezable_wq(void)
-{
-	return system_freezable_wq;
-}
-
-/* equivlalent to system_wq and system_freezable_wq, deprecated */
-#define system_nrt_wq			__system_nrt_wq()
-#define system_nrt_freezable_wq		__system_nrt_freezable_wq()
-
 extern struct workqueue_struct *
 __alloc_workqueue_key(const char *fmt, unsigned int flags, int max_active,
 	struct lock_class_key *key, const char *lock_name, ...) __printf(1, 6);
@@ -587,18 +569,6 @@ static inline bool keventd_up(void)
 	return system_wq != NULL;
 }
 
-/* used to be different but now identical to flush_work(), deprecated */
-static inline bool __deprecated flush_work_sync(struct work_struct *work)
-{
-	return flush_work(work);
-}
-
-/* used to be different but now identical to flush_delayed_work(), deprecated */
-static inline bool __deprecated flush_delayed_work_sync(struct delayed_work *dwork)
-{
-	return flush_delayed_work(dwork);
-}
-
 #ifndef CONFIG_SMP
 static inline long work_on_cpu(int cpu, long (*fn)(void *), void *arg)
 {
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index a4bab46cd38e..6203d2900877 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -65,15 +65,12 @@ enum {
 	 * be executing on any CPU.  The pool behaves as an unbound one.
 	 *
 	 * Note that DISASSOCIATED should be flipped only while holding
-	 * manager_mutex to avoid changing binding state while
-	 * create_worker() is in progress.
+	 * attach_mutex to avoid changing binding state while
+	 * worker_attach_to_pool() is in progress.
 	 */
-	POOL_MANAGE_WORKERS	= 1 << 0,	/* need to manage workers */
 	POOL_DISASSOCIATED	= 1 << 2,	/* cpu can't serve workers */
-	POOL_FREEZING		= 1 << 3,	/* freeze in progress */
 
 	/* worker flags */
-	WORKER_STARTED		= 1 << 0,	/* started */
 	WORKER_DIE		= 1 << 1,	/* die die die */
 	WORKER_IDLE		= 1 << 2,	/* is idle */
 	WORKER_PREP		= 1 << 3,	/* preparing to run works */
@@ -124,8 +121,7 @@ enum {
  *    cpu or grabbing pool->lock is enough for read access.  If
  *    POOL_DISASSOCIATED is set, it's identical to L.
  *
- * MG: pool->manager_mutex and pool->lock protected.  Writes require both
- *     locks.  Reads can happen under either lock.
+ * A: pool->attach_mutex protected.
  *
  * PL: wq_pool_mutex protected.
  *
@@ -163,8 +159,11 @@ struct worker_pool {
 
 	/* see manage_workers() for details on the two manager mutexes */
 	struct mutex		manager_arb;	/* manager arbitration */
-	struct mutex		manager_mutex;	/* manager exclusion */
-	struct idr		worker_idr;	/* MG: worker IDs and iteration */
+	struct mutex		attach_mutex;	/* attach/detach exclusion */
+	struct list_head	workers;	/* A: attached workers */
+	struct completion	*detach_completion; /* all workers detached */
+
+	struct ida		worker_ida;	/* worker IDs for task name */
 
 	struct workqueue_attrs	*attrs;		/* I: worker attributes */
 	struct hlist_node	hash_node;	/* PL: unbound_pool_hash node */
@@ -340,16 +339,6 @@ static void copy_workqueue_attrs(struct workqueue_attrs *to,
 			   lockdep_is_held(&wq->mutex),			\
 			   "sched RCU or wq->mutex should be held")
 
-#ifdef CONFIG_LOCKDEP
-#define assert_manager_or_pool_lock(pool)				\
-	WARN_ONCE(debug_locks &&					\
-		  !lockdep_is_held(&(pool)->manager_mutex) &&		\
-		  !lockdep_is_held(&(pool)->lock),			\
-		  "pool->manager_mutex or ->lock should be held")
-#else
-#define assert_manager_or_pool_lock(pool)	do { } while (0)
-#endif
-
 #define for_each_cpu_worker_pool(pool, cpu)				\
 	for ((pool) = &per_cpu(cpu_worker_pools, cpu)[0];		\
 	     (pool) < &per_cpu(cpu_worker_pools, cpu)[NR_STD_WORKER_POOLS]; \
@@ -375,17 +364,16 @@ static void copy_workqueue_attrs(struct workqueue_attrs *to,
 /**
  * for_each_pool_worker - iterate through all workers of a worker_pool
  * @worker: iteration cursor
- * @wi: integer used for iteration
  * @pool: worker_pool to iterate workers of
  *
- * This must be called with either @pool->manager_mutex or ->lock held.
+ * This must be called with @pool->attach_mutex.
  *
  * The if/else clause exists only for the lockdep assertion and can be
  * ignored.
  */
-#define for_each_pool_worker(worker, wi, pool)				\
-	idr_for_each_entry(&(pool)->worker_idr, (worker), (wi))		\
-		if (({ assert_manager_or_pool_lock((pool)); false; })) { } \
+#define for_each_pool_worker(worker, pool)				\
+	list_for_each_entry((worker), &(pool)->workers, node)		\
+		if (({ lockdep_assert_held(&pool->attach_mutex); false; })) { } \
 		else
 
 /**
@@ -763,13 +751,6 @@ static bool need_to_create_worker(struct worker_pool *pool)
 	return need_more_worker(pool) && !may_start_working(pool);
 }
 
-/* Do I need to be the manager? */
-static bool need_to_manage_workers(struct worker_pool *pool)
-{
-	return need_to_create_worker(pool) ||
-		(pool->flags & POOL_MANAGE_WORKERS);
-}
-
 /* Do we have too many workers and should some go away? */
 static bool too_many_workers(struct worker_pool *pool)
 {
@@ -791,8 +772,8 @@ static bool too_many_workers(struct worker_pool *pool)
  * Wake up functions.
  */
 
-/* Return the first worker.  Safe with preemption disabled */
-static struct worker *first_worker(struct worker_pool *pool)
+/* Return the first idle worker.  Safe with preemption disabled */
+static struct worker *first_idle_worker(struct worker_pool *pool)
 {
 	if (unlikely(list_empty(&pool->idle_list)))
 		return NULL;
@@ -811,7 +792,7 @@ static struct worker *first_worker(struct worker_pool *pool)
  */
 static void wake_up_worker(struct worker_pool *pool)
 {
-	struct worker *worker = first_worker(pool);
+	struct worker *worker = first_idle_worker(pool);
 
 	if (likely(worker))
 		wake_up_process(worker->task);
@@ -885,7 +866,7 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, int cpu)
 	 */
 	if (atomic_dec_and_test(&pool->nr_running) &&
 	    !list_empty(&pool->worklist))
-		to_wakeup = first_worker(pool);
+		to_wakeup = first_idle_worker(pool);
 	return to_wakeup ? to_wakeup->task : NULL;
 }
 
@@ -1621,70 +1602,6 @@ static void worker_leave_idle(struct worker *worker)
 	list_del_init(&worker->entry);
 }
 
-/**
- * worker_maybe_bind_and_lock - try to bind %current to worker_pool and lock it
- * @pool: target worker_pool
- *
- * Bind %current to the cpu of @pool if it is associated and lock @pool.
- *
- * Works which are scheduled while the cpu is online must at least be
- * scheduled to a worker which is bound to the cpu so that if they are
- * flushed from cpu callbacks while cpu is going down, they are
- * guaranteed to execute on the cpu.
- *
- * This function is to be used by unbound workers and rescuers to bind
- * themselves to the target cpu and may race with cpu going down or
- * coming online.  kthread_bind() can't be used because it may put the
- * worker to already dead cpu and set_cpus_allowed_ptr() can't be used
- * verbatim as it's best effort and blocking and pool may be
- * [dis]associated in the meantime.
- *
- * This function tries set_cpus_allowed() and locks pool and verifies the
- * binding against %POOL_DISASSOCIATED which is set during
- * %CPU_DOWN_PREPARE and cleared during %CPU_ONLINE, so if the worker
- * enters idle state or fetches works without dropping lock, it can
- * guarantee the scheduling requirement described in the first paragraph.
- *
- * CONTEXT:
- * Might sleep.  Called without any lock but returns with pool->lock
- * held.
- *
- * Return:
- * %true if the associated pool is online (@worker is successfully
- * bound), %false if offline.
- */
-static bool worker_maybe_bind_and_lock(struct worker_pool *pool)
-__acquires(&pool->lock)
-{
-	while (true) {
-		/*
-		 * The following call may fail, succeed or succeed
-		 * without actually migrating the task to the cpu if
-		 * it races with cpu hotunplug operation.  Verify
-		 * against POOL_DISASSOCIATED.
-		 */
-		if (!(pool->flags & POOL_DISASSOCIATED))
-			set_cpus_allowed_ptr(current, pool->attrs->cpumask);
-
-		spin_lock_irq(&pool->lock);
-		if (pool->flags & POOL_DISASSOCIATED)
-			return false;
-		if (task_cpu(current) == pool->cpu &&
-		    cpumask_equal(&current->cpus_allowed, pool->attrs->cpumask))
-			return true;
-		spin_unlock_irq(&pool->lock);
-
-		/*
-		 * We've raced with CPU hot[un]plug.  Give it a breather
-		 * and retry migration.  cond_resched() is required here;
-		 * otherwise, we might deadlock against cpu_stop trying to
-		 * bring down the CPU on non-preemptive kernel.
-		 */
-		cpu_relax();
-		cond_resched();
-	}
-}
-
 static struct worker *alloc_worker(void)
 {
 	struct worker *worker;
@@ -1693,6 +1610,7 @@ static struct worker *alloc_worker(void)
 	if (worker) {
 		INIT_LIST_HEAD(&worker->entry);
 		INIT_LIST_HEAD(&worker->scheduled);
+		INIT_LIST_HEAD(&worker->node);
 		/* on creation a worker is in !idle && prep state */
 		worker->flags = WORKER_PREP;
 	}
@@ -1700,12 +1618,68 @@ static struct worker *alloc_worker(void)
 }
 
 /**
+ * worker_attach_to_pool() - attach a worker to a pool
+ * @worker: worker to be attached
+ * @pool: the target pool
+ *
+ * Attach @worker to @pool.  Once attached, the %WORKER_UNBOUND flag and
+ * cpu-binding of @worker are kept coordinated with the pool across
+ * cpu-[un]hotplugs.
+ */
+static void worker_attach_to_pool(struct worker *worker,
+				   struct worker_pool *pool)
+{
+	mutex_lock(&pool->attach_mutex);
+
+	/*
+	 * set_cpus_allowed_ptr() will fail if the cpumask doesn't have any
+	 * online CPUs.  It'll be re-applied when any of the CPUs come up.
+	 */
+	set_cpus_allowed_ptr(worker->task, pool->attrs->cpumask);
+
+	/*
+	 * The pool->attach_mutex ensures %POOL_DISASSOCIATED remains
+	 * stable across this function.  See the comments above the
+	 * flag definition for details.
+	 */
+	if (pool->flags & POOL_DISASSOCIATED)
+		worker->flags |= WORKER_UNBOUND;
+
+	list_add_tail(&worker->node, &pool->workers);
+
+	mutex_unlock(&pool->attach_mutex);
+}
+
+/**
+ * worker_detach_from_pool() - detach a worker from its pool
+ * @worker: worker which is attached to its pool
+ * @pool: the pool @worker is attached to
+ *
+ * Undo the attaching which had been done in worker_attach_to_pool().  The
+ * caller worker shouldn't access to the pool after detached except it has
+ * other reference to the pool.
+ */
+static void worker_detach_from_pool(struct worker *worker,
+				    struct worker_pool *pool)
+{
+	struct completion *detach_completion = NULL;
+
+	mutex_lock(&pool->attach_mutex);
+	list_del(&worker->node);
+	if (list_empty(&pool->workers))
+		detach_completion = pool->detach_completion;
+	mutex_unlock(&pool->attach_mutex);
+
+	if (detach_completion)
+		complete(detach_completion);
+}
+
+/**
  * create_worker - create a new workqueue worker
  * @pool: pool the new worker will belong to
  *
- * Create a new worker which is bound to @pool.  The returned worker
- * can be started by calling start_worker() or destroyed using
- * destroy_worker().
+ * Create a new worker which is attached to @pool.  The new worker must be
+ * started by start_worker().
  *
  * CONTEXT:
  * Might sleep.  Does GFP_KERNEL allocations.
@@ -1719,19 +1693,8 @@ static struct worker *create_worker(struct worker_pool *pool)
 	int id = -1;
 	char id_buf[16];
 
-	lockdep_assert_held(&pool->manager_mutex);
-
-	/*
-	 * ID is needed to determine kthread name.  Allocate ID first
-	 * without installing the pointer.
-	 */
-	idr_preload(GFP_KERNEL);
-	spin_lock_irq(&pool->lock);
-
-	id = idr_alloc(&pool->worker_idr, NULL, 0, 0, GFP_NOWAIT);
-
-	spin_unlock_irq(&pool->lock);
-	idr_preload_end();
+	/* ID is needed to determine kthread name */
+	id = ida_simple_get(&pool->worker_ida, 0, 0, GFP_KERNEL);
 	if (id < 0)
 		goto fail;
 
@@ -1758,33 +1721,14 @@ static struct worker *create_worker(struct worker_pool *pool)
 	/* prevent userland from meddling with cpumask of workqueue workers */
 	worker->task->flags |= PF_NO_SETAFFINITY;
 
-	/*
-	 * set_cpus_allowed_ptr() will fail if the cpumask doesn't have any
-	 * online CPUs.  It'll be re-applied when any of the CPUs come up.
-	 */
-	set_cpus_allowed_ptr(worker->task, pool->attrs->cpumask);
-
-	/*
-	 * The caller is responsible for ensuring %POOL_DISASSOCIATED
-	 * remains stable across this function.  See the comments above the
-	 * flag definition for details.
-	 */
-	if (pool->flags & POOL_DISASSOCIATED)
-		worker->flags |= WORKER_UNBOUND;
-
-	/* successful, commit the pointer to idr */
-	spin_lock_irq(&pool->lock);
-	idr_replace(&pool->worker_idr, worker, worker->id);
-	spin_unlock_irq(&pool->lock);
+	/* successful, attach the worker to the pool */
+	worker_attach_to_pool(worker, pool);
 
 	return worker;
 
 fail:
-	if (id >= 0) {
-		spin_lock_irq(&pool->lock);
-		idr_remove(&pool->worker_idr, id);
-		spin_unlock_irq(&pool->lock);
-	}
+	if (id >= 0)
+		ida_simple_remove(&pool->worker_ida, id);
 	kfree(worker);
 	return NULL;
 }
@@ -1800,7 +1744,6 @@ fail:
  */
 static void start_worker(struct worker *worker)
 {
-	worker->flags |= WORKER_STARTED;
 	worker->pool->nr_workers++;
 	worker_enter_idle(worker);
 	wake_up_process(worker->task);
@@ -1818,8 +1761,6 @@ static int create_and_start_worker(struct worker_pool *pool)
 {
 	struct worker *worker;
 
-	mutex_lock(&pool->manager_mutex);
-
 	worker = create_worker(pool);
 	if (worker) {
 		spin_lock_irq(&pool->lock);
@@ -1827,8 +1768,6 @@ static int create_and_start_worker(struct worker_pool *pool)
 		spin_unlock_irq(&pool->lock);
 	}
 
-	mutex_unlock(&pool->manager_mutex);
-
 	return worker ? 0 : -ENOMEM;
 }
 
@@ -1836,46 +1775,30 @@ static int create_and_start_worker(struct worker_pool *pool)
  * destroy_worker - destroy a workqueue worker
  * @worker: worker to be destroyed
  *
- * Destroy @worker and adjust @pool stats accordingly.
+ * Destroy @worker and adjust @pool stats accordingly.  The worker should
+ * be idle.
  *
  * CONTEXT:
- * spin_lock_irq(pool->lock) which is released and regrabbed.
+ * spin_lock_irq(pool->lock).
  */
 static void destroy_worker(struct worker *worker)
 {
 	struct worker_pool *pool = worker->pool;
 
-	lockdep_assert_held(&pool->manager_mutex);
 	lockdep_assert_held(&pool->lock);
 
 	/* sanity check frenzy */
 	if (WARN_ON(worker->current_work) ||
-	    WARN_ON(!list_empty(&worker->scheduled)))
+	    WARN_ON(!list_empty(&worker->scheduled)) ||
+	    WARN_ON(!(worker->flags & WORKER_IDLE)))
 		return;
 
-	if (worker->flags & WORKER_STARTED)
-		pool->nr_workers--;
-	if (worker->flags & WORKER_IDLE)
-		pool->nr_idle--;
-
-	/*
-	 * Once WORKER_DIE is set, the kworker may destroy itself at any
-	 * point.  Pin to ensure the task stays until we're done with it.
-	 */
-	get_task_struct(worker->task);
+	pool->nr_workers--;
+	pool->nr_idle--;
 
 	list_del_init(&worker->entry);
 	worker->flags |= WORKER_DIE;
-
-	idr_remove(&pool->worker_idr, worker->id);
-
-	spin_unlock_irq(&pool->lock);
-
-	kthread_stop(worker->task);
-	put_task_struct(worker->task);
-	kfree(worker);
-
-	spin_lock_irq(&pool->lock);
+	wake_up_process(worker->task);
 }
 
 static void idle_worker_timeout(unsigned long __pool)
@@ -1884,7 +1807,7 @@ static void idle_worker_timeout(unsigned long __pool)
 
 	spin_lock_irq(&pool->lock);
 
-	if (too_many_workers(pool)) {
+	while (too_many_workers(pool)) {
 		struct worker *worker;
 		unsigned long expires;
 
@@ -1892,13 +1815,12 @@ static void idle_worker_timeout(unsigned long __pool)
 		worker = list_entry(pool->idle_list.prev, struct worker, entry);
 		expires = worker->last_active + IDLE_WORKER_TIMEOUT;
 
-		if (time_before(jiffies, expires))
+		if (time_before(jiffies, expires)) {
 			mod_timer(&pool->idle_timer, expires);
-		else {
-			/* it's been idle for too long, wake up manager */
-			pool->flags |= POOL_MANAGE_WORKERS;
-			wake_up_worker(pool);
+			break;
 		}
+
+		destroy_worker(worker);
 	}
 
 	spin_unlock_irq(&pool->lock);
@@ -2017,44 +1939,6 @@ restart:
 }
 
 /**
- * maybe_destroy_worker - destroy workers which have been idle for a while
- * @pool: pool to destroy workers for
- *
- * Destroy @pool workers which have been idle for longer than
- * IDLE_WORKER_TIMEOUT.
- *
- * LOCKING:
- * spin_lock_irq(pool->lock) which may be released and regrabbed
- * multiple times.  Called only from manager.
- *
- * Return:
- * %false if no action was taken and pool->lock stayed locked, %true
- * otherwise.
- */
-static bool maybe_destroy_workers(struct worker_pool *pool)
-{
-	bool ret = false;
-
-	while (too_many_workers(pool)) {
-		struct worker *worker;
-		unsigned long expires;
-
-		worker = list_entry(pool->idle_list.prev, struct worker, entry);
-		expires = worker->last_active + IDLE_WORKER_TIMEOUT;
-
-		if (time_before(jiffies, expires)) {
-			mod_timer(&pool->idle_timer, expires);
-			break;
-		}
-
-		destroy_worker(worker);
-		ret = true;
-	}
-
-	return ret;
-}
-
-/**
  * manage_workers - manage worker pool
  * @worker: self
  *
@@ -2083,8 +1967,6 @@ static bool manage_workers(struct worker *worker)
 	bool ret = false;
 
 	/*
-	 * Managership is governed by two mutexes - manager_arb and
-	 * manager_mutex.  manager_arb handles arbitration of manager role.
 	 * Anyone who successfully grabs manager_arb wins the arbitration
 	 * and becomes the manager.  mutex_trylock() on pool->manager_arb
 	 * failure while holding pool->lock reliably indicates that someone
@@ -2093,40 +1975,12 @@ static bool manage_workers(struct worker *worker)
 	 * grabbing manager_arb is responsible for actually performing
 	 * manager duties.  If manager_arb is grabbed and released without
 	 * actual management, the pool may stall indefinitely.
-	 *
-	 * manager_mutex is used for exclusion of actual management
-	 * operations.  The holder of manager_mutex can be sure that none
-	 * of management operations, including creation and destruction of
-	 * workers, won't take place until the mutex is released.  Because
-	 * manager_mutex doesn't interfere with manager role arbitration,
-	 * it is guaranteed that the pool's management, while may be
-	 * delayed, won't be disturbed by someone else grabbing
-	 * manager_mutex.
 	 */
 	if (!mutex_trylock(&pool->manager_arb))
 		return ret;
 
-	/*
-	 * With manager arbitration won, manager_mutex would be free in
-	 * most cases.  trylock first without dropping @pool->lock.
-	 */
-	if (unlikely(!mutex_trylock(&pool->manager_mutex))) {
-		spin_unlock_irq(&pool->lock);
-		mutex_lock(&pool->manager_mutex);
-		spin_lock_irq(&pool->lock);
-		ret = true;
-	}
-
-	pool->flags &= ~POOL_MANAGE_WORKERS;
-
-	/*
-	 * Destroy and then create so that may_start_working() is true
-	 * on return.
-	 */
-	ret |= maybe_destroy_workers(pool);
 	ret |= maybe_create_worker(pool);
 
-	mutex_unlock(&pool->manager_mutex);
 	mutex_unlock(&pool->manager_arb);
 	return ret;
 }
@@ -2314,6 +2168,11 @@ woke_up:
 		spin_unlock_irq(&pool->lock);
 		WARN_ON_ONCE(!list_empty(&worker->entry));
 		worker->task->flags &= ~PF_WQ_WORKER;
+
+		set_task_comm(worker->task, "kworker/dying");
+		ida_simple_remove(&pool->worker_ida, worker->id);
+		worker_detach_from_pool(worker, pool);
+		kfree(worker);
 		return 0;
 	}
 
@@ -2361,9 +2220,6 @@ recheck:
 
 	worker_set_flags(worker, WORKER_PREP, false);
 sleep:
-	if (unlikely(need_to_manage_workers(pool)) && manage_workers(worker))
-		goto recheck;
-
 	/*
 	 * pool->lock is held and there's no work to process and no need to
 	 * manage, sleep.  Workers are woken up only while holding
@@ -2440,8 +2296,9 @@ repeat:
 
 		spin_unlock_irq(&wq_mayday_lock);
 
-		/* migrate to the target cpu if possible */
-		worker_maybe_bind_and_lock(pool);
+		worker_attach_to_pool(rescuer, pool);
+
+		spin_lock_irq(&pool->lock);
 		rescuer->pool = pool;
 
 		/*
@@ -2454,6 +2311,11 @@ repeat:
 				move_linked_works(work, scheduled, &n);
 
 		process_scheduled_works(rescuer);
+		spin_unlock_irq(&pool->lock);
+
+		worker_detach_from_pool(rescuer, pool);
+
+		spin_lock_irq(&pool->lock);
 
 		/*
 		 * Put the reference grabbed by send_mayday().  @pool won't
@@ -3550,9 +3412,10 @@ static int init_worker_pool(struct worker_pool *pool)
 		    (unsigned long)pool);
 
 	mutex_init(&pool->manager_arb);
-	mutex_init(&pool->manager_mutex);
-	idr_init(&pool->worker_idr);
+	mutex_init(&pool->attach_mutex);
+	INIT_LIST_HEAD(&pool->workers);
 
+	ida_init(&pool->worker_ida);
 	INIT_HLIST_NODE(&pool->hash_node);
 	pool->refcnt = 1;
 
@@ -3567,7 +3430,7 @@ static void rcu_free_pool(struct rcu_head *rcu)
 {
 	struct worker_pool *pool = container_of(rcu, struct worker_pool, rcu);
 
-	idr_destroy(&pool->worker_idr);
+	ida_destroy(&pool->worker_ida);
 	free_workqueue_attrs(pool->attrs);
 	kfree(pool);
 }
@@ -3585,6 +3448,7 @@ 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;
 
 	lockdep_assert_held(&wq_pool_mutex);
@@ -3605,18 +3469,24 @@ static void put_unbound_pool(struct worker_pool *pool)
 	/*
 	 * Become the manager and destroy all workers.  Grabbing
 	 * manager_arb prevents @pool's workers from blocking on
-	 * manager_mutex.
+	 * attach_mutex.
 	 */
 	mutex_lock(&pool->manager_arb);
-	mutex_lock(&pool->manager_mutex);
-	spin_lock_irq(&pool->lock);
 
-	while ((worker = first_worker(pool)))
+	spin_lock_irq(&pool->lock);
+	while ((worker = first_idle_worker(pool)))
 		destroy_worker(worker);
 	WARN_ON(pool->nr_workers || pool->nr_idle);
-
 	spin_unlock_irq(&pool->lock);
-	mutex_unlock(&pool->manager_mutex);
+
+	mutex_lock(&pool->attach_mutex);
+	if (!list_empty(&pool->workers))
+		pool->detach_completion = &detach_completion;
+	mutex_unlock(&pool->attach_mutex);
+
+	if (pool->detach_completion)
+		wait_for_completion(pool->detach_completion);
+
 	mutex_unlock(&pool->manager_arb);
 
 	/* shut down the timers */
@@ -3662,9 +3532,6 @@ static struct worker_pool *get_unbound_pool(const struct workqueue_attrs *attrs)
 	if (!pool || init_worker_pool(pool) < 0)
 		goto fail;
 
-	if (workqueue_freezing)
-		pool->flags |= POOL_FREEZING;
-
 	lockdep_set_subclass(&pool->lock, 1);	/* see put_pwq() */
 	copy_workqueue_attrs(pool->attrs, attrs);
 
@@ -3771,7 +3638,12 @@ static void pwq_adjust_max_active(struct pool_workqueue *pwq)
 
 	spin_lock_irq(&pwq->pool->lock);
 
-	if (!freezable || !(pwq->pool->flags & POOL_FREEZING)) {
+	/*
+	 * During [un]freezing, the caller is responsible for ensuring that
+	 * this function is called at least once after @workqueue_freezing
+	 * is updated and visible.
+	 */
+	if (!freezable || !workqueue_freezing) {
 		pwq->max_active = wq->saved_max_active;
 
 		while (!list_empty(&pwq->delayed_works) &&
@@ -4103,17 +3975,13 @@ static void wq_update_unbound_numa(struct workqueue_struct *wq, int cpu,
 	 * Let's determine what needs to be done.  If the target cpumask is
 	 * different from wq's, we need to compare it to @pwq's and create
 	 * a new one if they don't match.  If the target cpumask equals
-	 * wq's, the default pwq should be used.  If @pwq is already the
-	 * default one, nothing to do; otherwise, install the default one.
+	 * wq's, the default pwq should be used.
 	 */
 	if (wq_calc_node_cpumask(wq->unbound_attrs, node, cpu_off, cpumask)) {
 		if (cpumask_equal(cpumask, pwq->pool->attrs->cpumask))
 			goto out_unlock;
 	} else {
-		if (pwq == wq->dfl_pwq)
-			goto out_unlock;
-		else
-			goto use_dfl_pwq;
+		goto use_dfl_pwq;
 	}
 
 	mutex_unlock(&wq->mutex);
@@ -4121,8 +3989,8 @@ static void wq_update_unbound_numa(struct workqueue_struct *wq, int cpu,
 	/* create a new pwq */
 	pwq = alloc_unbound_pwq(wq, target_attrs);
 	if (!pwq) {
-		pr_warning("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
-			   wq->name);
+		pr_warn("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
+			wq->name);
 		mutex_lock(&wq->mutex);
 		goto use_dfl_pwq;
 	}
@@ -4599,28 +4467,27 @@ static void wq_unbind_fn(struct work_struct *work)
 	int cpu = smp_processor_id();
 	struct worker_pool *pool;
 	struct worker *worker;
-	int wi;
 
 	for_each_cpu_worker_pool(pool, cpu) {
 		WARN_ON_ONCE(cpu != smp_processor_id());
 
-		mutex_lock(&pool->manager_mutex);
+		mutex_lock(&pool->attach_mutex);
 		spin_lock_irq(&pool->lock);
 
 		/*
-		 * We've blocked all manager operations.  Make all workers
+		 * We've blocked all attach/detach operations. Make all workers
 		 * unbound and set DISASSOCIATED.  Before this, all workers
 		 * except for the ones which are still executing works from
 		 * before the last CPU down must be on the cpu.  After
 		 * this, they may become diasporas.
 		 */
-		for_each_pool_worker(worker, wi, pool)
+		for_each_pool_worker(worker, pool)
 			worker->flags |= WORKER_UNBOUND;
 
 		pool->flags |= POOL_DISASSOCIATED;
 
 		spin_unlock_irq(&pool->lock);
-		mutex_unlock(&pool->manager_mutex);
+		mutex_unlock(&pool->attach_mutex);
 
 		/*
 		 * Call schedule() so that we cross rq->lock and thus can
@@ -4660,9 +4527,8 @@ static void wq_unbind_fn(struct work_struct *work)
 static void rebind_workers(struct worker_pool *pool)
 {
 	struct worker *worker;
-	int wi;
 
-	lockdep_assert_held(&pool->manager_mutex);
+	lockdep_assert_held(&pool->attach_mutex);
 
 	/*
 	 * Restore CPU affinity of all workers.  As all idle workers should
@@ -4671,13 +4537,13 @@ static void rebind_workers(struct worker_pool *pool)
 	 * of all workers first and then clear UNBOUND.  As we're called
 	 * from CPU_ONLINE, the following shouldn't fail.
 	 */
-	for_each_pool_worker(worker, wi, pool)
+	for_each_pool_worker(worker, pool)
 		WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task,
 						  pool->attrs->cpumask) < 0);
 
 	spin_lock_irq(&pool->lock);
 
-	for_each_pool_worker(worker, wi, pool) {
+	for_each_pool_worker(worker, pool) {
 		unsigned int worker_flags = worker->flags;
 
 		/*
@@ -4729,9 +4595,8 @@ static void restore_unbound_workers_cpumask(struct worker_pool *pool, int cpu)
 {
 	static cpumask_t cpumask;
 	struct worker *worker;
-	int wi;
 
-	lockdep_assert_held(&pool->manager_mutex);
+	lockdep_assert_held(&pool->attach_mutex);
 
 	/* is @cpu allowed for @pool? */
 	if (!cpumask_test_cpu(cpu, pool->attrs->cpumask))
@@ -4743,7 +4608,7 @@ static void restore_unbound_workers_cpumask(struct worker_pool *pool, int cpu)
 		return;
 
 	/* as we're called from CPU_ONLINE, the following shouldn't fail */
-	for_each_pool_worker(worker, wi, pool)
+	for_each_pool_worker(worker, pool)
 		WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task,
 						  pool->attrs->cpumask) < 0);
 }
@@ -4776,7 +4641,7 @@ static int workqueue_cpu_up_callback(struct notifier_block *nfb,
 		mutex_lock(&wq_pool_mutex);
 
 		for_each_pool(pool, pi) {
-			mutex_lock(&pool->manager_mutex);
+			mutex_lock(&pool->attach_mutex);
 
 			if (pool->cpu == cpu) {
 				spin_lock_irq(&pool->lock);
@@ -4788,7 +4653,7 @@ static int workqueue_cpu_up_callback(struct notifier_block *nfb,
 				restore_unbound_workers_cpumask(pool, cpu);
 			}
 
-			mutex_unlock(&pool->manager_mutex);
+			mutex_unlock(&pool->attach_mutex);
 		}
 
 		/* update NUMA affinity of unbound workqueues */
@@ -4887,24 +4752,14 @@ EXPORT_SYMBOL_GPL(work_on_cpu);
  */
 void freeze_workqueues_begin(void)
 {
-	struct worker_pool *pool;
 	struct workqueue_struct *wq;
 	struct pool_workqueue *pwq;
-	int pi;
 
 	mutex_lock(&wq_pool_mutex);
 
 	WARN_ON_ONCE(workqueue_freezing);
 	workqueue_freezing = true;
 
-	/* set FREEZING */
-	for_each_pool(pool, pi) {
-		spin_lock_irq(&pool->lock);
-		WARN_ON_ONCE(pool->flags & POOL_FREEZING);
-		pool->flags |= POOL_FREEZING;
-		spin_unlock_irq(&pool->lock);
-	}
-
 	list_for_each_entry(wq, &workqueues, list) {
 		mutex_lock(&wq->mutex);
 		for_each_pwq(pwq, wq)
@@ -4974,21 +4829,13 @@ void thaw_workqueues(void)
 {
 	struct workqueue_struct *wq;
 	struct pool_workqueue *pwq;
-	struct worker_pool *pool;
-	int pi;
 
 	mutex_lock(&wq_pool_mutex);
 
 	if (!workqueue_freezing)
 		goto out_unlock;
 
-	/* clear FREEZING */
-	for_each_pool(pool, pi) {
-		spin_lock_irq(&pool->lock);
-		WARN_ON_ONCE(!(pool->flags & POOL_FREEZING));
-		pool->flags &= ~POOL_FREEZING;
-		spin_unlock_irq(&pool->lock);
-	}
+	workqueue_freezing = false;
 
 	/* restore max_active and repopulate worklist */
 	list_for_each_entry(wq, &workqueues, list) {
@@ -4998,7 +4845,6 @@ void thaw_workqueues(void)
 		mutex_unlock(&wq->mutex);
 	}
 
-	workqueue_freezing = false;
 out_unlock:
 	mutex_unlock(&wq_pool_mutex);
 }
diff --git a/kernel/workqueue_internal.h b/kernel/workqueue_internal.h
index 7e2204db0b1a..45215870ac6c 100644
--- a/kernel/workqueue_internal.h
+++ b/kernel/workqueue_internal.h
@@ -37,6 +37,8 @@ struct worker {
 	struct task_struct	*task;		/* I: worker task */
 	struct worker_pool	*pool;		/* I: the associated pool */
 						/* L: for rescuers */
+	struct list_head	node;		/* A: anchored at pool->workers */
+						/* A: runs through worker->node */
 
 	unsigned long		last_active;	/* L: last active timestamp */
 	unsigned int		flags;		/* X: flags */