[PATCH] block: don't quiesce the queue when setting the default elevator

[PATCH] block: don't quiesce the queue when setting the default elevator

[Qiliang Yuan]

elevator_set_default() runs from blk_register_queue() while the disk is
still being added: the block device is not visible yet and uevents are
suppressed, so no file system I/O can be issued. elevator_change()
freezes the queue and cancels the dispatch work before switching, which
is enough to drain dispatch activity from passthrough requests, as the
comment in elevator_change() already states.

elevator_switch() still quiesces the queue unconditionally, though, so
every disk added with a default elevator waits for an RCU grace period.
Commit 245a489e81e1 ("block: avoid to quiesce queue in
elevator_init_mq") had removed this wait from the default elevator
setup, and it came back when that setup was folded into
elevator_change().

The wait is paid in full by callers that add disks one at a time. ublk
adds the disk of a single-queue device from START_DEV, which takes 7.7ms
at p50 in a 16 vCPU KVM guest, most of it in blk_mq_quiesce_queue()
called from elevator_set_default().

Skip quiescing in elevator_switch() when setting the default elevator.

ublk null target, 1000 single-queue devices created one by one:

                     before      after
  START_DEV p50      7.74ms     0.68ms
  START_DEV p99     14.61ms     1.32ms
  devices/s             114        971

Fixes: 1e44bedbc921 ("block: unifying elevator change")
Cc: stable@vger.kernel.org
Signed-off-by: Qiliang Yuan <odys.yuan@gmail.com>
---
 block/elevator.c | 7 +++++--
 block/elevator.h | 2 ++
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/block/elevator.c b/block/elevator.c
index 2161b6eea680c..da9ba70ac9321 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -573,7 +573,8 @@ static int elevator_switch(struct request_queue *q, struct elv_change_ctx *ctx)
 			return -EINVAL;
 	}
 
-	blk_mq_quiesce_queue(q);
+	if (!ctx->no_quiesce)
+		blk_mq_quiesce_queue(q);
 
 	if (q->elevator) {
 		ctx->old = q->elevator;
@@ -594,7 +595,8 @@ static int elevator_switch(struct request_queue *q, struct elv_change_ctx *ctx)
 	blk_add_trace_msg(q, "elv switch: %s", ctx->name);
 
 out_unfreeze:
-	blk_mq_unquiesce_queue(q);
+	if (!ctx->no_quiesce)
+		blk_mq_unquiesce_queue(q);
 
 	if (ret) {
 		pr_warn("elv: switch to \"%s\" failed, falling back to \"none\"\n",
@@ -731,6 +733,7 @@ void elevator_set_default(struct request_queue *q)
 	struct elv_change_ctx ctx = {
 		.name = "mq-deadline",
 		.no_uevent = true,
+		.no_quiesce = true,
 	};
 	int err;
 
diff --git a/block/elevator.h b/block/elevator.h
index 3eb32516be0b1..3886d97427e33 100644
--- a/block/elevator.h
+++ b/block/elevator.h
@@ -43,6 +43,8 @@ struct elevator_resources {
 struct elv_change_ctx {
 	const char *name;
 	bool no_uevent;
+	/* the disk isn't added yet, so skip quiescing the queue */
+	bool no_quiesce;
 
 	/* for unregistering old elevator */
 	struct elevator_queue *old;

---
base-commit: 551c722f40809618230001baccf219193e22fc5a
change-id: 20260930-bug-block-elevator-init-quiesce-35d73f9b462d

Best regards,
-- 
Qiliang Yuan <odys.yuan@gmail.com>

Re: [PATCH] block: don't quiesce the queue when setting the default elevator

[Qiliang Yuan]

Hi Ming, Jens,

Gentle ping. Since 1e44bedbc921 ("block: unifying elevator change"),
setting the default elevator quiesces the queue again, which
245a489e81e1 had avoided, so every disk added with a default elevator
waits for an RCU grace period. START_DEV of a single-queue ublk device
takes 7.74ms at p50 instead of 0.68ms.

Ming, as the author of 1e44bedbc921, could you take a look when you
have a moment?

Thanks,
Qiliang

Re: [PATCH] block: don't quiesce the queue when setting the default elevator

[Qiliang Yuan]

Hi Ming,

The patch and my ping went to your Red Hat address, which I think you
no longer use, so adding your current one. The patch is here:

https://lore.kernel.org/r/20261001-bug-block-elevator-init-quiesce-v1-1-5c85fa0ec59b@gmail.com

Thanks,
Qiliang

Re: [PATCH] block: don't quiesce the queue when setting the default elevator

[Ming Lei]

On Wed, Sep 30, 2026 at 10:12 PM Qiliang Yuan <odys.yuan@gmail.com> wrote:
>
> elevator_set_default() runs from blk_register_queue() while the disk is
> still being added: the block device is not visible yet and uevents are
> suppressed, so no file system I/O can be issued. elevator_change()
> freezes the queue and cancels the dispatch work before switching, which
> is enough to drain dispatch activity from passthrough requests, as the
> comment in elevator_change() already states.
>
> elevator_switch() still quiesces the queue unconditionally, though, so
> every disk added with a default elevator waits for an RCU grace period.
> Commit 245a489e81e1 ("block: avoid to quiesce queue in
> elevator_init_mq") had removed this wait from the default elevator
> setup, and it came back when that setup was folded into
> elevator_change().
>
> The wait is paid in full by callers that add disks one at a time. ublk
> adds the disk of a single-queue device from START_DEV, which takes 7.7ms
> at p50 in a 16 vCPU KVM guest, most of it in blk_mq_quiesce_queue()
> called from elevator_set_default().
>
> Skip quiescing in elevator_switch() when setting the default elevator.
>
> ublk null target, 1000 single-queue devices created one by one:
>
>                      before      after
>   START_DEV p50      7.74ms     0.68ms
>   START_DEV p99     14.61ms     1.32ms
>   devices/s             114        971
>
> Fixes: 1e44bedbc921 ("block: unifying elevator change")
> Cc: stable@vger.kernel.org
> Signed-off-by: Qiliang Yuan <odys.yuan@gmail.com>
> ---
>  block/elevator.c | 7 +++++--
>  block/elevator.h | 2 ++
>  2 files changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/block/elevator.c b/block/elevator.c
> index 2161b6eea680c..da9ba70ac9321 100644
> --- a/block/elevator.c
> +++ b/block/elevator.c
> @@ -573,7 +573,8 @@ static int elevator_switch(struct request_queue *q, struct elv_change_ctx *ctx)
>                         return -EINVAL;
>         }
>
> -       blk_mq_quiesce_queue(q);
> +       if (!ctx->no_quiesce)
> +               blk_mq_quiesce_queue(q);
>
>         if (q->elevator) {
>                 ctx->old = q->elevator;
> @@ -594,7 +595,8 @@ static int elevator_switch(struct request_queue *q, struct elv_change_ctx *ctx)
>         blk_add_trace_msg(q, "elv switch: %s", ctx->name);
>
>  out_unfreeze:
> -       blk_mq_unquiesce_queue(q);
> +       if (!ctx->no_quiesce)
> +               blk_mq_unquiesce_queue(q);
>
>         if (ret) {
>                 pr_warn("elv: switch to \"%s\" failed, falling back to \"none\"\n",
> @@ -731,6 +733,7 @@ void elevator_set_default(struct request_queue *q)
>         struct elv_change_ctx ctx = {
>                 .name = "mq-deadline",
>                 .no_uevent = true,
> +               .no_quiesce = true,
>         };
>         int err;

Looks good given elevator_set_default() is only called on new added disk,

Reviewed-by: Ming Lei <tom.leiming@gmail.com>



Thanks,
Ming Lei

Re: [PATCH] block: don't quiesce the queue when setting the default elevator

[Nilay Shroff]

On 10/1/26 8:40 AM, Qiliang Yuan wrote:
> elevator_set_default() runs from blk_register_queue() while the disk is
> still being added: the block device is not visible yet and uevents are
> suppressed, so no file system I/O can be issued. elevator_change()
> freezes the queue and cancels the dispatch work before switching, which
> is enough to drain dispatch activity from passthrough requests, as the
> comment in elevator_change() already states.
> 
> elevator_switch() still quiesces the queue unconditionally, though, so
> every disk added with a default elevator waits for an RCU grace period.
> Commit 245a489e81e1 ("block: avoid to quiesce queue in
> elevator_init_mq") had removed this wait from the default elevator
> setup, and it came back when that setup was folded into
> elevator_change().
> 
> The wait is paid in full by callers that add disks one at a time. ublk
> adds the disk of a single-queue device from START_DEV, which takes 7.7ms
> at p50 in a 16 vCPU KVM guest, most of it in blk_mq_quiesce_queue()
> called from elevator_set_default().
> 
> Skip quiescing in elevator_switch() when setting the default elevator.
> 
> ublk null target, 1000 single-queue devices created one by one:
> 
>                       before      after
>    START_DEV p50      7.74ms     0.68ms
>    START_DEV p99     14.61ms     1.32ms
>    devices/s             114        971
> 
> Fixes: 1e44bedbc921 ("block: unifying elevator change")
> Cc:stable@vger.kernel.org
> Signed-off-by: Qiliang Yuan<odys.yuan@gmail.com>
> ---
>   block/elevator.c | 7 +++++--
>   block/elevator.h | 2 ++
>   2 files changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/block/elevator.c b/block/elevator.c
> index 2161b6eea680c..da9ba70ac9321 100644
> --- a/block/elevator.c
> +++ b/block/elevator.c
> @@ -573,7 +573,8 @@ static int elevator_switch(struct request_queue *q, struct elv_change_ctx *ctx)
>   			return -EINVAL;
>   	}
>   
> -	blk_mq_quiesce_queue(q);
> +	if (!ctx->no_quiesce)
> +		blk_mq_quiesce_queue(q);
>   
>   	if (q->elevator) {
>   		ctx->old = q->elevator;
> @@ -594,7 +595,8 @@ static int elevator_switch(struct request_queue *q, struct elv_change_ctx *ctx)
>   	blk_add_trace_msg(q, "elv switch: %s", ctx->name);
>   
>   out_unfreeze:
> -	blk_mq_unquiesce_queue(q);
> +	if (!ctx->no_quiesce)
> +		blk_mq_unquiesce_queue(q);
>   
One minor nit: Since we only unquiesce the queue at the out_unfreeze label,
could you rename it to out_unquiesce while you're at it? I realize this label
predates your change, but the new name would better reflect what happens there.

Otherwise, this change looks good to me.

Reviewed-by: Nilay Shroff <nilay@linux.ibm.com>