linux-kernel - [PATCH v4 for-6.18/block 1/5] blk-mq-sched: introduce high level elevator lock

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20250908092007.3796967-2-yukuai1@huaweicloud.com>
Date: Mon,  8 Sep 2025 17:20:03 +0800
From: Yu Kuai <yukuai1@...weicloud.com>
To: dlemoal@...nel.org,
	hare@...e.de,
	bvanassche@....org,
	ming.lei@...hat.com,
	tj@...nel.org,
	josef@...icpanda.com,
	axboe@...nel.dk,
	yukuai3@...wei.com
Cc: cgroups@...r.kernel.org,
	linux-block@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	yukuai1@...weicloud.com,
	yi.zhang@...wei.com,
	yangerkun@...wei.com,
	johnny.chenyi@...wei.com
Subject: [PATCH v4 for-6.18/block 1/5] blk-mq-sched: introduce high level elevator lock

From: Yu Kuai <yukuai3@...wei.com>

Currently, both mq-deadline and bfq have global spin lock that will be
grabbed inside elevator methods like dispatch_request, insert_requests,
and bio_merge. And the global lock is the main reason mq-deadline and
bfq can't scale very well.

While dispatching request, blk_mq_get_disatpch_budget() and
blk_mq_get_driver_tag() must be called, and they are not ready to be called
inside elevator methods, hence introduce a new method like
dispatch_requests is not possible.

Hence introduce a new high level elevator lock, currently it is protecting
dispatch_request only. Following patches will convert mq-deadline and bfq
to use this lock and finally support request batch dispatching by calling
the method multiple time while holding the lock.

Signed-off-by: Yu Kuai <yukuai3@...wei.com>
---
 block/blk-mq-sched.c |  9 ++++++++-
 block/elevator.c     |  1 +
 block/elevator.h     | 31 +++++++++++++++++++++++++++++--
 3 files changed, 38 insertions(+), 3 deletions(-)

diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
index d06bb137a743..6f2b0573a8ff 100644
--- a/block/blk-mq-sched.c
+++ b/block/blk-mq-sched.c
@@ -113,7 +113,14 @@ static int __blk_mq_do_dispatch_sched(struct blk_mq_hw_ctx *hctx)
 		if (budget_token < 0)
 			break;
 
-		rq = e->type->ops.dispatch_request(hctx);
+		if (blk_queue_sq_sched(q)) {
+			elevator_dispatch_lock(e);
+			rq = e->type->ops.dispatch_request(hctx);
+			elevator_dispatch_unlock(e);
+		} else {
+			rq = e->type->ops.dispatch_request(hctx);
+		}
+
 		if (!rq) {
 			blk_mq_put_dispatch_budget(q, budget_token);
 			/*
diff --git a/block/elevator.c b/block/elevator.c
index e2ebfbf107b3..e132a24f9ce3 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -146,6 +146,7 @@ struct elevator_queue *elevator_alloc(struct request_queue *q,
 	eq->type = e;
 	kobject_init(&eq->kobj, &elv_ktype);
 	mutex_init(&eq->sysfs_lock);
+	spin_lock_init(&eq->lock);
 	hash_init(eq->hash);
 	eq->et = et;
 
diff --git a/block/elevator.h b/block/elevator.h
index c4d20155065e..4864add27136 100644
--- a/block/elevator.h
+++ b/block/elevator.h
@@ -119,13 +119,13 @@ struct request *elv_rqhash_find(struct request_queue *q, sector_t offset);
 /*
  * each queue has an elevator_queue associated with it
  */
-struct elevator_queue
-{
+struct elevator_queue {
 	struct elevator_type *type;
 	struct elevator_tags *et;
 	void *elevator_data;
 	struct kobject kobj;
 	struct mutex sysfs_lock;
+	spinlock_t lock;
 	unsigned long flags;
 	DECLARE_HASHTABLE(hash, ELV_HASH_BITS);
 };
@@ -133,6 +133,33 @@ struct elevator_queue
 #define ELEVATOR_FLAG_REGISTERED	0
 #define ELEVATOR_FLAG_DYING		1
 #define ELEVATOR_FLAG_ENABLE_WBT_ON_EXIT	2
+#define ELEVATOR_FLAG_DISPATCH_IRQ	3
+
+#define elevator_lock(e)		spin_lock(&(e)->lock)
+#define elevator_unlock(e)		spin_unlock(&(e)->lock)
+#define elevator_lock_irq(e)		spin_lock_irq(&(e)->lock)
+#define elevator_unlock_irq(e)		spin_unlock_irq(&(e)->lock)
+#define elevator_lock_irqsave(e, flags) \
+	spin_lock_irqsave(&(e)->lock, flags)
+#define elevator_unlock_irqrestore(e, flags) \
+	spin_unlock_irqrestore(&(e)->lock, flags)
+#define elevator_lock_assert_held(e)	lockdep_assert_held(&(e)->lock)
+
+static inline void elevator_dispatch_lock(struct elevator_queue *eq)
+{
+	if (test_bit(ELEVATOR_FLAG_DISPATCH_IRQ, &eq->flags))
+		elevator_lock_irq(eq);
+	else
+		elevator_lock(eq);
+}
+
+static inline void elevator_dispatch_unlock(struct elevator_queue *eq)
+{
+	if (test_bit(ELEVATOR_FLAG_DISPATCH_IRQ, &eq->flags))
+		elevator_unlock_irq(eq);
+	else
+		elevator_unlock(eq);
+}
 
 /*
  * block elevator interface
-- 
2.39.2