blk-cgroup: delay freeing policy data after rcu grace period

Currently blkcg_print_blkgs() must hold RCU to iterate blkgs from a
blkcg, and prfill() must hold queue_lock to prevent policy data from
being freed by policy deactivation. As a consequence, queue_lock has to
be nested under RCU from blkcg_print_blkgs().

Delay freeing policy data until after an RCU grace period so prfill() can
be protected by RCU alone.

Signed-off-by: Yu Kuai <yukuai@fygo.io>
Link: https://patch.msgid.link/e20e5d984b41a026d61851966bed35eb094c4bff.1780621988.git.yukuai@fygo.io
Signed-off-by: Jens Axboe <axboe@kernel.dk>

5 files changed, 42 insertions, 6 deletions

diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index d8fdace464b4..0d3e32d246a2 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c
@@ -550,13 +550,20 @@ static void bfq_pd_init(struct blkg_policy_data *pd)
 	bfqg->rq_pos_tree = RB_ROOT;
 }
 
-static void bfq_pd_free(struct blkg_policy_data *pd)
+static void bfqg_release(struct rcu_head *rcu)
 {
+	struct blkg_policy_data *pd =
+		container_of(rcu, struct blkg_policy_data, rcu_head);
 	struct bfq_group *bfqg = pd_to_bfqg(pd);
 
 	bfqg_put(bfqg);
 }
 
+static void bfq_pd_free(struct blkg_policy_data *pd)
+{
+	call_rcu(&pd->rcu_head, bfqg_release);
+}
+
 static void bfq_pd_reset_stats(struct blkg_policy_data *pd)
 {
 	struct bfq_group *bfqg = pd_to_bfqg(pd);
diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h
index f25fecb87c43..cc603ded6ded 100644
--- a/block/blk-cgroup.h
+++ b/block/blk-cgroup.h
@@ -140,6 +140,8 @@ struct blkg_policy_data {
 	struct blkcg_gq			*blkg;
 	int				plid;
 	bool				online;
+
+	struct rcu_head			rcu_head;
 };
 
 /*
diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index 563cc7dcf348..27d2dcaa65f0 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -3050,6 +3050,16 @@ static void ioc_pd_init(struct blkg_policy_data *pd)
 	spin_unlock_irqrestore(&ioc->lock, flags);
 }
 
+static void iocg_release(struct rcu_head *rcu)
+{
+	struct blkg_policy_data *pd =
+		container_of(rcu, struct blkg_policy_data, rcu_head);
+	struct ioc_gq *iocg = pd_to_iocg(pd);
+
+	free_percpu(iocg->pcpu_stat);
+	kfree(iocg);
+}
+
 static void ioc_pd_free(struct blkg_policy_data *pd)
 {
 	struct ioc_gq *iocg = pd_to_iocg(pd);
@@ -3074,8 +3084,8 @@ static void ioc_pd_free(struct blkg_policy_data *pd)
 
 		hrtimer_cancel(&iocg->waitq_timer);
 	}
-	free_percpu(iocg->pcpu_stat);
-	kfree(iocg);
+
+	call_rcu(&pd->rcu_head, iocg_release);
 }
 
 static void ioc_pd_stat(struct blkg_policy_data *pd, struct seq_file *s)
diff --git a/block/blk-iolatency.c b/block/blk-iolatency.c
index 1aaee6fb0f59..cef02b6c5fa9 100644
--- a/block/blk-iolatency.c
+++ b/block/blk-iolatency.c
@@ -1031,13 +1031,21 @@ static void iolatency_pd_offline(struct blkg_policy_data *pd)
 	iolatency_clear_scaling(blkg);
 }
 
-static void iolatency_pd_free(struct blkg_policy_data *pd)
+static void iolat_release(struct rcu_head *rcu)
 {
+	struct blkg_policy_data *pd =
+		container_of(rcu, struct blkg_policy_data, rcu_head);
 	struct iolatency_grp *iolat = pd_to_lat(pd);
+
 	free_percpu(iolat->stats);
 	kfree(iolat);
 }
 
+static void iolatency_pd_free(struct blkg_policy_data *pd)
+{
+	call_rcu(&pd->rcu_head, iolat_release);
+}
+
 static struct cftype iolatency_files[] = {
 	{
 		.name = "latency",
diff --git a/block/blk-throttle.c b/block/blk-throttle.c
index 47052ba21d1b..ffc3b70065d4 100644
--- a/block/blk-throttle.c
+++ b/block/blk-throttle.c
@@ -353,16 +353,25 @@ static void throtl_pd_online(struct blkg_policy_data *pd)
 	tg_update_has_rules(tg);
 }
 
-static void throtl_pd_free(struct blkg_policy_data *pd)
+static void tg_release(struct rcu_head *rcu)
 {
+	struct blkg_policy_data *pd =
+		container_of(rcu, struct blkg_policy_data, rcu_head);
 	struct throtl_grp *tg = pd_to_tg(pd);
 
-	timer_delete_sync(&tg->service_queue.pending_timer);
 	blkg_rwstat_exit(&tg->stat_bytes);
 	blkg_rwstat_exit(&tg->stat_ios);
 	kfree(tg);
 }
 
+static void throtl_pd_free(struct blkg_policy_data *pd)
+{
+	struct throtl_grp *tg = pd_to_tg(pd);
+
+	timer_delete_sync(&tg->service_queue.pending_timer);
+	call_rcu(&pd->rcu_head, tg_release);
+}
+
 static struct throtl_grp *
 throtl_rb_first(struct throtl_service_queue *parent_sq)
 {