[PATCH V2] blk-cgroup: prevent rcu_sched detected stalls warnings while iterating blkgs

From: Yu Kuai
Date: Tue Jul 06 2021 - 21:48:30 EST


We run a test that create millions of cgroups and blkgs, and then trigger
blkg_destroy_all(). blkg_destroy_all() will hold spin lock for a long
time in such situation. Thus release the lock when a batch of blkgs are
destroyed.

blkcg_activate_policy() and blkcg_deactivate_policy() might have the
same problem, however, as they are basically only called from module
init/exit paths, let's leave them alone for now.

Signed-off-by: Yu Kuai <yukuai3@xxxxxxxxxx>
---
changes in V2:
- as suggested by Tejun, rename 'BLKG_DESTROY_BATCH_SIZE' and modify
blkg_destroy_all() only.

block/blk-cgroup.c | 15 +++++++++++++++
1 file changed, 15 insertions(+)

diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 7b06a5fa3cac..575d7a2e7203 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -56,6 +56,8 @@ static LIST_HEAD(all_blkcgs); /* protected by blkcg_pol_mutex */
bool blkcg_debug_stats = false;
static struct workqueue_struct *blkcg_punt_bio_wq;

+#define BLKG_DESTROY_BATCH_SIZE 64
+
static bool blkcg_policy_enabled(struct request_queue *q,
const struct blkcg_policy *pol)
{
@@ -422,7 +424,9 @@ static void blkg_destroy(struct blkcg_gq *blkg)
static void blkg_destroy_all(struct request_queue *q)
{
struct blkcg_gq *blkg, *n;
+ int count = BLKG_DESTROY_BATCH_SIZE;

+restart:
spin_lock_irq(&q->queue_lock);
list_for_each_entry_safe(blkg, n, &q->blkg_list, q_node) {
struct blkcg *blkcg = blkg->blkcg;
@@ -430,6 +434,17 @@ static void blkg_destroy_all(struct request_queue *q)
spin_lock(&blkcg->lock);
blkg_destroy(blkg);
spin_unlock(&blkcg->lock);
+
+ /*
+ * in order to avoid holding the spin lock for too long, release
+ * it when a batch of blkgs are destroyed.
+ */
+ if (!(--count)) {
+ count = BLKG_DESTROY_BATCH_SIZE;
+ spin_unlock_irq(&q->queue_lock);
+ cond_resched();
+ goto restart;
+ }
}

q->root_blkg = NULL;
--
2.31.1