Received-SPF: pass (google.com: domain of linux-kernel+bounces-39003-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Sender: Tejun Heo <htejun@gmail.com>
From: Tejun Heo <tj@kernel.org>
To: jiangshanlai@gmail.com
Cc: linux-kernel@vger.kernel.org,
	Naohiro.Aota@wdc.com,
	kernel-team@meta.com,
	Tejun Heo <tj@kernel.org>
Subject: [PATCH 06/10] workqueue: RCU protect wq->dfl_pwq and implement accessors for it
Date: Thu, 25 Jan 2024 07:05:59 -1000
Message-ID: <20240125170628.2017784-7-tj@kernel.org>
In-Reply-To: <20240125170628.2017784-1-tj@kernel.org>
References: <20240125170628.2017784-1-tj@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

wq->cpu_pwq is RCU protected but wq->dfl_pwq isn't. This is okay because
currently wq->dfl_pwq is used only accessed to install it into wq->cpu_pwq
which doesn't require RCU access. However, we want to be able to access
wq->dfl_pwq under RCU in the future to access its __pod_cpumask and the code
can be made easier to read by making the two pwq fields behave in the same
way.

- Make wq->dfl_pwq RCU protected.

- Add unbound_pwq_slot() and unbound_pwq() which can access both ->dfl_pwq
  and ->cpu_pwq. The former returns the double pointer that can be used
  access and update the pwqs. The latter performs locking check and
  dereferences the double pointer.

- pwq accesses and updates are converted to use unbound_pwq[_slot]().

Signed-off-by: Tejun Heo <tj@kernel.org>
---
 kernel/workqueue.c | 64 +++++++++++++++++++++++++++++-----------------
 1 file changed, 40 insertions(+), 24 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index bfb6e951852a..1bca0a4ab9d1 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -304,7 +304,7 @@ struct workqueue_struct {
 	int			saved_max_active; /* WQ: saved max_active */
 
 	struct workqueue_attrs	*unbound_attrs;	/* PW: only for unbound wqs */
-	struct pool_workqueue	*dfl_pwq;	/* PW: only for unbound wqs */
+	struct pool_workqueue __rcu *dfl_pwq;   /* PW: only for unbound wqs */
 
 #ifdef CONFIG_SYSFS
 	struct wq_device	*wq_dev;	/* I: for sysfs interface */
@@ -635,6 +635,23 @@ static int worker_pool_assign_id(struct worker_pool *pool)
 	return ret;
 }
 
+static struct pool_workqueue __rcu **
+unbound_pwq_slot(struct workqueue_struct *wq, int cpu)
+{
+       if (cpu >= 0)
+               return per_cpu_ptr(wq->cpu_pwq, cpu);
+       else
+               return &wq->dfl_pwq;
+}
+
+/* @cpu < 0 for dfl_pwq */
+static struct pool_workqueue *unbound_pwq(struct workqueue_struct *wq, int cpu)
+{
+	return rcu_dereference_check(*unbound_pwq_slot(wq, cpu),
+				     lockdep_is_held(&wq_pool_mutex) ||
+				     lockdep_is_held(&wq->mutex));
+}
+
 static unsigned int work_color_to_flags(int color)
 {
 	return color << WORK_STRUCT_COLOR_SHIFT;
@@ -4325,10 +4342,11 @@ static void wq_calc_pod_cpumask(struct workqueue_attrs *attrs, int cpu,
 				"possible intersect\n");
 }
 
-/* install @pwq into @wq's cpu_pwq and return the old pwq */
+/* install @pwq into @wq and return the old pwq, @cpu < 0 for dfl_pwq */
 static struct pool_workqueue *install_unbound_pwq(struct workqueue_struct *wq,
 					int cpu, struct pool_workqueue *pwq)
 {
+	struct pool_workqueue __rcu **slot = unbound_pwq_slot(wq, cpu);
 	struct pool_workqueue *old_pwq;
 
 	lockdep_assert_held(&wq_pool_mutex);
@@ -4337,8 +4355,8 @@ static struct pool_workqueue *install_unbound_pwq(struct workqueue_struct *wq,
 	/* link_pwq() can handle duplicate calls */
 	link_pwq(pwq);
 
-	old_pwq = rcu_access_pointer(*per_cpu_ptr(wq->cpu_pwq, cpu));
-	rcu_assign_pointer(*per_cpu_ptr(wq->cpu_pwq, cpu), pwq);
+	old_pwq = rcu_access_pointer(*slot);
+	rcu_assign_pointer(*slot, pwq);
 	return old_pwq;
 }
 
@@ -4438,14 +4456,11 @@ static void apply_wqattrs_commit(struct apply_wqattrs_ctx *ctx)
 
 	copy_workqueue_attrs(ctx->wq->unbound_attrs, ctx->attrs);
 
-	/* save the previous pwq and install the new one */
+	/* save the previous pwqs and install the new ones */
 	for_each_possible_cpu(cpu)
 		ctx->pwq_tbl[cpu] = install_unbound_pwq(ctx->wq, cpu,
 							ctx->pwq_tbl[cpu]);
-
-	/* @dfl_pwq might not have been used, ensure it's linked */
-	link_pwq(ctx->dfl_pwq);
-	swap(ctx->wq->dfl_pwq, ctx->dfl_pwq);
+	ctx->dfl_pwq = install_unbound_pwq(ctx->wq, -1, ctx->dfl_pwq);
 
 	mutex_unlock(&ctx->wq->mutex);
 }
@@ -4555,9 +4570,7 @@ static void wq_update_pod(struct workqueue_struct *wq, int cpu,
 
 	/* nothing to do if the target cpumask matches the current pwq */
 	wq_calc_pod_cpumask(target_attrs, cpu, off_cpu);
-	pwq = rcu_dereference_protected(*per_cpu_ptr(wq->cpu_pwq, cpu),
-					lockdep_is_held(&wq_pool_mutex));
-	if (wqattrs_equal(target_attrs, pwq->pool->attrs))
+	if (wqattrs_equal(target_attrs, unbound_pwq(wq, cpu)->pool->attrs))
 		return;
 
 	/* create a new pwq */
@@ -4575,10 +4588,11 @@ static void wq_update_pod(struct workqueue_struct *wq, int cpu,
 
 use_dfl_pwq:
 	mutex_lock(&wq->mutex);
-	raw_spin_lock_irq(&wq->dfl_pwq->pool->lock);
-	get_pwq(wq->dfl_pwq);
-	raw_spin_unlock_irq(&wq->dfl_pwq->pool->lock);
-	old_pwq = install_unbound_pwq(wq, cpu, wq->dfl_pwq);
+	pwq = unbound_pwq(wq, -1);
+	raw_spin_lock_irq(&pwq->pool->lock);
+	get_pwq(pwq);
+	raw_spin_unlock_irq(&pwq->pool->lock);
+	old_pwq = install_unbound_pwq(wq, cpu, pwq);
 out_unlock:
 	mutex_unlock(&wq->mutex);
 	put_pwq_unlocked(old_pwq);
@@ -4616,10 +4630,13 @@ static int alloc_and_link_pwqs(struct workqueue_struct *wq)
 
 	cpus_read_lock();
 	if (wq->flags & __WQ_ORDERED) {
+		struct pool_workqueue *dfl_pwq;
+
 		ret = apply_workqueue_attrs(wq, ordered_wq_attrs[highpri]);
 		/* there should only be single pwq for ordering guarantee */
-		WARN(!ret && (wq->pwqs.next != &wq->dfl_pwq->pwqs_node ||
-			      wq->pwqs.prev != &wq->dfl_pwq->pwqs_node),
+		dfl_pwq = rcu_access_pointer(wq->dfl_pwq);
+		WARN(!ret && (wq->pwqs.next != &dfl_pwq->pwqs_node ||
+			      wq->pwqs.prev != &dfl_pwq->pwqs_node),
 		     "ordering guarantee broken for workqueue %s\n", wq->name);
 	} else {
 		ret = apply_workqueue_attrs(wq, unbound_std_wq_attrs[highpri]);
@@ -4853,7 +4870,7 @@ static bool pwq_busy(struct pool_workqueue *pwq)
 		if (pwq->nr_in_flight[i])
 			return true;
 
-	if ((pwq != pwq->wq->dfl_pwq) && (pwq->refcnt > 1))
+	if ((pwq != rcu_access_pointer(pwq->wq->dfl_pwq)) && (pwq->refcnt > 1))
 		return true;
 	if (!pwq_is_empty(pwq))
 		return true;
@@ -4937,13 +4954,12 @@ void destroy_workqueue(struct workqueue_struct *wq)
 	rcu_read_lock();
 
 	for_each_possible_cpu(cpu) {
-		pwq = rcu_access_pointer(*per_cpu_ptr(wq->cpu_pwq, cpu));
-		RCU_INIT_POINTER(*per_cpu_ptr(wq->cpu_pwq, cpu), NULL);
-		put_pwq_unlocked(pwq);
+		put_pwq_unlocked(unbound_pwq(wq, cpu));
+		RCU_INIT_POINTER(*unbound_pwq_slot(wq, cpu), NULL);
 	}
 
-	put_pwq_unlocked(wq->dfl_pwq);
-	wq->dfl_pwq = NULL;
+	put_pwq_unlocked(unbound_pwq(wq, -1));
+	RCU_INIT_POINTER(*unbound_pwq_slot(wq, -1), NULL);
 
 	rcu_read_unlock();
 }
-- 
2.43.0