* [PATCH v2 1/2] block: don't request a module for the none elevator
2026-11-15 1:58 [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again Qiliang Yuan
@ 2026-11-15 1:58 ` Qiliang Yuan
2026-11-15 1:58 ` [PATCH v2 2/2] block: return early when the elevator doesn't change Qiliang Yuan
2026-11-18 16:07 ` [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again Jens Axboe
2 siblings, 0 replies; 6+ messages in thread
From: Qiliang Yuan @ 2026-11-15 1:58 UTC (permalink / raw)
To: Jens Axboe, Damien Le Moal, Ming Lei, Nilay Shroff, Hannes Reinecke
Cc: linux-block, linux-kernel, Qiliang Yuan, Christoph Hellwig
elv_iosched_store() calls elv_iosched_load_module() for the requested
name before anything else, and loads "<name>-iosched" if no elevator
type of that name is registered. "none" never is, so every write of
"none" to queue/scheduler ends up in request_module("none-iosched").
That runs a modprobe helper which can never find the module, and adds
several milliseconds to every switch to "none", including rewrites of
it by udev rules or tuning tools that set the scheduler on each disk.
Return early for "none", like elevator_change() did before the module
loading moved into elv_iosched_load_module().
Writing queue/scheduler of a null_blk disk 20 times takes (median):
operation before (ms) after (ms) change
none -> none 13.10 9.11 -30%
mq-deadline -> none 13.00 7.97 -39%
Fixes: 734e1a8603128 ("block: Prevent deadlocks when switching elevators")
Signed-off-by: Qiliang Yuan <qiliang.yuan@linux.dev>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Nilay Shroff <nilay@linux.ibm.com>
---
block/elevator.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/block/elevator.c b/block/elevator.c
index 66edf1a5768a8..9d9b2e84a1288 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -775,6 +775,9 @@ static void elv_iosched_load_module(const char *elevator_name)
{
struct elevator_type *found;
+ if (!strncmp(elevator_name, "none", 4))
+ return;
+
spin_lock(&elv_list_lock);
found = __elevator_find(elevator_name);
spin_unlock(&elv_list_lock);
--
2.43.0
^ permalink raw reply [flat|nested] 6+ messages in thread* [PATCH v2 2/2] block: return early when the elevator doesn't change
2026-11-15 1:58 [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again Qiliang Yuan
2026-11-15 1:58 ` [PATCH v2 1/2] block: don't request a module for the none elevator Qiliang Yuan
@ 2026-11-15 1:58 ` Qiliang Yuan
2026-11-16 6:27 ` Christoph Hellwig
2026-11-18 16:07 ` [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again Jens Axboe
2 siblings, 1 reply; 6+ messages in thread
From: Qiliang Yuan @ 2026-11-15 1:58 UTC (permalink / raw)
To: Jens Axboe, Damien Le Moal, Ming Lei, Nilay Shroff, Hannes Reinecke
Cc: linux-block, linux-kernel, Qiliang Yuan
Writing the scheduler a queue already uses to queue/scheduler used to
be a no-op. elevator_change() now only checks for an unchanged elevator
after freezing the queue, and doesn't treat "none" on a queue without
an elevator as unchanged at all, so that case also quiesces the queue
in elevator_switch() and resets nr_requests.
Every such write therefore drains in-flight I/O and waits for an RCU
grace period. udev rules and tuning tools that set the scheduler
unconditionally stall I/O on live disks each time they run.
Check under elevator_lock whether anything would change before
allocating scheduler resources and freezing the queue. The check can
race with another elevator change, which at worst leads to a redundant
switch, so keep the one done under the freeze as well.
Writing the current scheduler to queue/scheduler of a null_blk disk 20
times takes (median):
operation before (ms) after (ms) change
none -> none 9.11 0.08 -99%
mq-deadline -> mq-deadline 9.05 0.08 -99%
4k random reads at queue depth 1 for 10 seconds, while "none" is
written to queue/scheduler every 50 ms:
metric before after change
IOPS 340562 390392 +15%
max latency (us) 16302 1224 -92%
Fixes: b07a889e83355 ("block: move q->sysfs_lock and queue-freeze under show/store method")
Signed-off-by: Qiliang Yuan <qiliang.yuan@linux.dev>
---
block/elevator.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/block/elevator.c b/block/elevator.c
index 9d9b2e84a1288..549cf51543a8b 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -654,10 +654,24 @@ static int elevator_change(struct request_queue *q, struct elv_change_ctx *ctx)
{
unsigned int memflags;
struct blk_mq_tag_set *set = q->tag_set;
+ bool unchanged;
int ret = 0;
lockdep_assert_held(&set->update_nr_hwq_lock);
+ /*
+ * Skip freezing the queue if nothing would change. This can race with
+ * another elevator change, which at worst leads to a redundant switch.
+ */
+ mutex_lock(&q->elevator_lock);
+ if (q->elevator)
+ unchanged = elevator_match(q->elevator->type, ctx->name);
+ else
+ unchanged = !strncmp(ctx->name, "none", 4);
+ mutex_unlock(&q->elevator_lock);
+ if (unchanged)
+ return 0;
+
if (strncmp(ctx->name, "none", 4)) {
ret = blk_mq_alloc_sched_res(q, ctx->type, &ctx->res,
set->nr_hw_queues);
--
2.43.0
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v2 2/2] block: return early when the elevator doesn't change
2026-11-15 1:58 ` [PATCH v2 2/2] block: return early when the elevator doesn't change Qiliang Yuan
@ 2026-11-16 6:27 ` Christoph Hellwig
0 siblings, 0 replies; 6+ messages in thread
From: Christoph Hellwig @ 2026-11-16 6:27 UTC (permalink / raw)
To: Qiliang Yuan
Cc: Jens Axboe, Damien Le Moal, Ming Lei, Nilay Shroff,
Hannes Reinecke, linux-block, linux-kernel
Looks good:
Reviewed-by: Christoph Hellwig <hch@lst.de>
On Sat, Nov 14, 2026 at 08:58:33PM -0500, Qiliang Yuan wrote:
> Fixes: b07a889e83355 ("block: move q->sysfs_lock and queue-freeze under show/store method")
But this fixes tag is not correct, that commit just moved code around.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again
2026-11-15 1:58 [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again Qiliang Yuan
2026-11-15 1:58 ` [PATCH v2 1/2] block: don't request a module for the none elevator Qiliang Yuan
2026-11-15 1:58 ` [PATCH v2 2/2] block: return early when the elevator doesn't change Qiliang Yuan
@ 2026-11-18 16:07 ` Jens Axboe
2026-11-19 1:07 ` Qiliang Yuan
2 siblings, 1 reply; 6+ messages in thread
From: Jens Axboe @ 2026-11-18 16:07 UTC (permalink / raw)
To: Damien Le Moal, Ming Lei, Nilay Shroff, Hannes Reinecke, Qiliang Yuan
Cc: linux-block, linux-kernel, Christoph Hellwig
On Sat, 14 Nov 2026 20:58:31 -0500, Qiliang Yuan wrote:
> Writing the scheduler a queue already uses to queue/scheduler takes
> about 10 ms and freezes the queue, which stalls I/O on live disks every
> time a udev rule or tuning tool sets the scheduler unconditionally.
>
> Patch 1 stops requesting a "none-iosched" module on every write of
> "none". Patch 2 returns before freezing the queue when the elevator
> wouldn't change.
>
> [...]
Applied, thanks!
[1/2] block: don't request a module for the none elevator
commit: c85f570121c94679854b570cae2c0dbcc0eae43b
[2/2] block: return early when the elevator doesn't change
commit: 375a88c6176ffe05e521c2df31b1ceaf4ae5fbaf
Best regards,
--
Jens Axboe
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again
2026-11-18 16:07 ` [PATCH v2 0/2] block: make changing to the current I/O scheduler cheap again Jens Axboe
@ 2026-11-19 1:07 ` Qiliang Yuan
0 siblings, 0 replies; 6+ messages in thread
From: Qiliang Yuan @ 2026-11-19 1:07 UTC (permalink / raw)
To: Jens Axboe
Cc: Qiliang Yuan, Damien Le Moal, Ming Lei, Nilay Shroff,
Hannes Reinecke, linux-block, linux-kernel, Christoph Hellwig
On Wed, 18 Nov 2026 09:07:37 -0700, Jens Axboe wrote:
> Applied, thanks!
>
[...]
> [2/2] block: return early when the elevator doesn't change
> commit: 375a88c6176ffe05e521c2df31b1ceaf4ae5fbaf
This one went in from v2, whose Fixes tag Christoph pointed out was
wrong. v3 changed it to the commit that added the freeze, with the code
unchanged:
Fixes: af2814149883e ("block: freeze the queue in queue_attr_store")
Could you update it in for-next?
v3: https://lore.kernel.org/r/20261116-bug-block-elevator-unchanged-v3-2-cb4dda2dadf4@linux.dev
Thanks,
Qiliang
^ permalink raw reply [flat|nested] 6+ messages in thread