blk-mq: introduce new queue attribute asyc_dpeth

[PATCH 03/16] blk-mq: remove useless checkings from blk_mq_update_nr_requests()

Posted by Yu Kuai 1 month, 3 weeks ago

From: Yu Kuai <yukuai3@huawei.com>

1) queue_requests_store() is the only caller of
blk_mq_update_nr_requests(), where queue is already freezed, no need to
check mq_freeze_depth;
2) q->tag_set must be set for request_based device, and queue_is_mq() is
already checked in blk_mq_queue_attr_visible(), no need to check
q->tag_set.
3) During initialization, hctx->tags in initialized before queue
kobject, and during del_gendisk, queue kobject is deleted before
exiting hctx, hence checking hctx->tags is useless.

Signed-off-by: Yu Kuai <yukuai3@huawei.com>
---
 block/blk-mq.c | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index b67d6c02eceb..3a219b7b3688 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -4921,24 +4921,15 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
 {
 	struct blk_mq_tag_set *set = q->tag_set;
 	struct blk_mq_hw_ctx *hctx;
-	int ret;
+	int ret = 0;
 	unsigned long i;
 
-	if (WARN_ON_ONCE(!q->mq_freeze_depth))
-		return -EINVAL;
-
-	if (!set)
-		return -EINVAL;
-
 	if (q->nr_requests == nr)
 		return 0;
 
 	blk_mq_quiesce_queue(q);
 
-	ret = 0;
 	queue_for_each_hw_ctx(q, hctx, i) {
-		if (!hctx->tags)
-			continue;
 		/*
 		 * If we're using an MQ scheduler, just update the scheduler
 		 * queue depth. This is similar to what the old code would do.
-- 
2.39.2

Re: [PATCH 03/16] blk-mq: remove useless checkings from blk_mq_update_nr_requests()

Posted by Nilay Shroff 1 month, 3 weeks ago


On 8/14/25 9:05 AM, Yu Kuai wrote:
> From: Yu Kuai <yukuai3@huawei.com>
> 
> 1) queue_requests_store() is the only caller of
> blk_mq_update_nr_requests(), where queue is already freezed, no need to
> check mq_freeze_depth;
> 2) q->tag_set must be set for request_based device, and queue_is_mq() is
> already checked in blk_mq_queue_attr_visible(), no need to check
> q->tag_set.
> 3) During initialization, hctx->tags in initialized before queue
> kobject, and during del_gendisk, queue kobject is deleted before
> exiting hctx, hence checking hctx->tags is useless.
> 
> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
> ---
>  block/blk-mq.c | 11 +----------
>  1 file changed, 1 insertion(+), 10 deletions(-)
> 
> diff --git a/block/blk-mq.c b/block/blk-mq.c
> index b67d6c02eceb..3a219b7b3688 100644
> --- a/block/blk-mq.c
> +++ b/block/blk-mq.c
> @@ -4921,24 +4921,15 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
>  {
>  	struct blk_mq_tag_set *set = q->tag_set;
>  	struct blk_mq_hw_ctx *hctx;
> -	int ret;
> +	int ret = 0;
>  	unsigned long i;
>  
> -	if (WARN_ON_ONCE(!q->mq_freeze_depth))
> -		return -EINVAL;
> -
> -	if (!set)
> -		return -EINVAL;
> -
>  	if (q->nr_requests == nr)
>  		return 0;
>  
>  	blk_mq_quiesce_queue(q);
>  
> -	ret = 0;
>  	queue_for_each_hw_ctx(q, hctx, i) {
> -		if (!hctx->tags)
> -			continue;
It's possible that hctx->tags is set to NULL in case no software 
queues are mapped to the hardware queue. So it seems that this
check is valid. Please see blk_mq_map_swqueue().

Thanks,
--Nilay

Re: [PATCH 03/16] blk-mq: remove useless checkings from blk_mq_update_nr_requests()

Posted by Yu Kuai 1 month, 2 weeks ago

Hi,

在 2025/08/14 20:23, Nilay Shroff 写道:
> 
> 
> On 8/14/25 9:05 AM, Yu Kuai wrote:
>> From: Yu Kuai <yukuai3@huawei.com>
>>
>> 1) queue_requests_store() is the only caller of
>> blk_mq_update_nr_requests(), where queue is already freezed, no need to
>> check mq_freeze_depth;
>> 2) q->tag_set must be set for request_based device, and queue_is_mq() is
>> already checked in blk_mq_queue_attr_visible(), no need to check
>> q->tag_set.
>> 3) During initialization, hctx->tags in initialized before queue
>> kobject, and during del_gendisk, queue kobject is deleted before
>> exiting hctx, hence checking hctx->tags is useless.
>>
>> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
>> ---
>>   block/blk-mq.c | 11 +----------
>>   1 file changed, 1 insertion(+), 10 deletions(-)
>>
>> diff --git a/block/blk-mq.c b/block/blk-mq.c
>> index b67d6c02eceb..3a219b7b3688 100644
>> --- a/block/blk-mq.c
>> +++ b/block/blk-mq.c
>> @@ -4921,24 +4921,15 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
>>   {
>>   	struct blk_mq_tag_set *set = q->tag_set;
>>   	struct blk_mq_hw_ctx *hctx;
>> -	int ret;
>> +	int ret = 0;
>>   	unsigned long i;
>>   
>> -	if (WARN_ON_ONCE(!q->mq_freeze_depth))
>> -		return -EINVAL;
>> -
>> -	if (!set)
>> -		return -EINVAL;
>> -
>>   	if (q->nr_requests == nr)
>>   		return 0;
>>   
>>   	blk_mq_quiesce_queue(q);
>>   
>> -	ret = 0;
>>   	queue_for_each_hw_ctx(q, hctx, i) {
>> -		if (!hctx->tags)
>> -			continue;
> It's possible that hctx->tags is set to NULL in case no software
> queues are mapped to the hardware queue. So it seems that this
> check is valid. Please see blk_mq_map_swqueue().

Ok, thanks for the reviw.

I didn't notice this, just wonder how can this happen?
nr_hw_queues > NR_CPUS?

Thanks,
Kuai

> 
> Thanks,
> --Nilay
> 
> .
>

Re: [PATCH 03/16] blk-mq: remove useless checkings from blk_mq_update_nr_requests()

Posted by Nilay Shroff 1 month, 2 weeks ago


On 8/15/25 7:02 AM, Yu Kuai wrote:
> Hi,
> 
> 在 2025/08/14 20:23, Nilay Shroff 写道:
>>
>>
>> On 8/14/25 9:05 AM, Yu Kuai wrote:
>>> From: Yu Kuai <yukuai3@huawei.com>
>>>
>>> 1) queue_requests_store() is the only caller of
>>> blk_mq_update_nr_requests(), where queue is already freezed, no need to
>>> check mq_freeze_depth;
>>> 2) q->tag_set must be set for request_based device, and queue_is_mq() is
>>> already checked in blk_mq_queue_attr_visible(), no need to check
>>> q->tag_set.
>>> 3) During initialization, hctx->tags in initialized before queue
>>> kobject, and during del_gendisk, queue kobject is deleted before
>>> exiting hctx, hence checking hctx->tags is useless.
>>>
>>> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
>>> ---
>>>   block/blk-mq.c | 11 +----------
>>>   1 file changed, 1 insertion(+), 10 deletions(-)
>>>
>>> diff --git a/block/blk-mq.c b/block/blk-mq.c
>>> index b67d6c02eceb..3a219b7b3688 100644
>>> --- a/block/blk-mq.c
>>> +++ b/block/blk-mq.c
>>> @@ -4921,24 +4921,15 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
>>>   {
>>>       struct blk_mq_tag_set *set = q->tag_set;
>>>       struct blk_mq_hw_ctx *hctx;
>>> -    int ret;
>>> +    int ret = 0;
>>>       unsigned long i;
>>>   -    if (WARN_ON_ONCE(!q->mq_freeze_depth))
>>> -        return -EINVAL;
>>> -
>>> -    if (!set)
>>> -        return -EINVAL;
>>> -
>>>       if (q->nr_requests == nr)
>>>           return 0;
>>>         blk_mq_quiesce_queue(q);
>>>   -    ret = 0;
>>>       queue_for_each_hw_ctx(q, hctx, i) {
>>> -        if (!hctx->tags)
>>> -            continue;
>> It's possible that hctx->tags is set to NULL in case no software
>> queues are mapped to the hardware queue. So it seems that this
>> check is valid. Please see blk_mq_map_swqueue().
> 
> Ok, thanks for the reviw.
> 
> I didn't notice this, just wonder how can this happen?
> nr_hw_queues > NR_CPUS?
> 
I think typically having nr_hw_queues > NR_CPUS is not allowed. 
But it's possible to have no software queues are mapped to hctx.
Check this commit 4412efecf7fd ("Revert "blk-mq: remove code for
dealing with remapping queue")

Thanks,
--Nilay

Re: [PATCH 03/16] blk-mq: remove useless checkings from blk_mq_update_nr_requests()

Posted by Ming Lei 1 month, 2 weeks ago

On Fri, Aug 15, 2025 at 05:29:01PM +0530, Nilay Shroff wrote:
> 
> 
> On 8/15/25 7:02 AM, Yu Kuai wrote:
> > Hi,
> > 
> > 在 2025/08/14 20:23, Nilay Shroff 写道:
> >>
> >>
> >> On 8/14/25 9:05 AM, Yu Kuai wrote:
> >>> From: Yu Kuai <yukuai3@huawei.com>
> >>>
> >>> 1) queue_requests_store() is the only caller of
> >>> blk_mq_update_nr_requests(), where queue is already freezed, no need to
> >>> check mq_freeze_depth;
> >>> 2) q->tag_set must be set for request_based device, and queue_is_mq() is
> >>> already checked in blk_mq_queue_attr_visible(), no need to check
> >>> q->tag_set.
> >>> 3) During initialization, hctx->tags in initialized before queue
> >>> kobject, and during del_gendisk, queue kobject is deleted before
> >>> exiting hctx, hence checking hctx->tags is useless.
> >>>
> >>> Signed-off-by: Yu Kuai <yukuai3@huawei.com>
> >>> ---
> >>>   block/blk-mq.c | 11 +----------
> >>>   1 file changed, 1 insertion(+), 10 deletions(-)
> >>>
> >>> diff --git a/block/blk-mq.c b/block/blk-mq.c
> >>> index b67d6c02eceb..3a219b7b3688 100644
> >>> --- a/block/blk-mq.c
> >>> +++ b/block/blk-mq.c
> >>> @@ -4921,24 +4921,15 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
> >>>   {
> >>>       struct blk_mq_tag_set *set = q->tag_set;
> >>>       struct blk_mq_hw_ctx *hctx;
> >>> -    int ret;
> >>> +    int ret = 0;
> >>>       unsigned long i;
> >>>   -    if (WARN_ON_ONCE(!q->mq_freeze_depth))
> >>> -        return -EINVAL;
> >>> -
> >>> -    if (!set)
> >>> -        return -EINVAL;
> >>> -
> >>>       if (q->nr_requests == nr)
> >>>           return 0;
> >>>         blk_mq_quiesce_queue(q);
> >>>   -    ret = 0;
> >>>       queue_for_each_hw_ctx(q, hctx, i) {
> >>> -        if (!hctx->tags)
> >>> -            continue;
> >> It's possible that hctx->tags is set to NULL in case no software
> >> queues are mapped to the hardware queue. So it seems that this
> >> check is valid. Please see blk_mq_map_swqueue().
> > 
> > Ok, thanks for the reviw.
> > 
> > I didn't notice this, just wonder how can this happen?
> > nr_hw_queues > NR_CPUS?
> > 
> I think typically having nr_hw_queues > NR_CPUS is not allowed. 
> But it's possible to have no software queues are mapped to hctx.
> Check this commit 4412efecf7fd ("Revert "blk-mq: remove code for
> dealing with remapping queue")

It is possible in case of poll_queues, for example:

modprobe null_blk submit_queues=$(getconf _NPROCESSORS_ONLN) poll_queues=2
 
 

Thanks, 
Ming