[PATCH 0/2] block, scsi: Small improvements for blk_mq_alloc_queue() usage

[PATCH 0/2] block, scsi: Small improvements for blk_mq_alloc_queue() usage

[John Garry]

A couple of small improvements to not manually set q->queuedata.

I asked Himanshu (cc'ed) to test the bsg-lib.c change as I have no setup
to test.

Based on mkp-scsi 6.10 staging queue at e4f5f8298cf6.

John Garry (2):
  scsi: core: Pass sdev to blk_mq_alloc_queue()
  scsi: bsg: Pass dev to blk_mq_alloc_queue()

 block/bsg-lib.c          | 3 +--
 drivers/scsi/scsi_scan.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

-- 
2.31.1

[PATCH 1/2] scsi: core: Pass sdev to blk_mq_alloc_queue()

[John Garry]

When calling scsi_alloc_sdev() -> blk_mq_alloc_queue(), we don't pass
the sdev as the queuedata, but rather manually set it afterwards. Just
pass to blk_mq_alloc_queue() to have automatically set.

Signed-off-by: John Garry <john.g.garry@oracle.com>
---
 drivers/scsi/scsi_scan.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 463ce6e23dc6..ec54d58fb6c0 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -334,7 +334,7 @@ static struct scsi_device *scsi_alloc_sdev(struct scsi_target *starget,
 	sdev->sg_reserved_size = INT_MAX;
 
 	scsi_init_limits(shost, &lim);
-	q = blk_mq_alloc_queue(&sdev->host->tag_set, &lim, NULL);
+	q = blk_mq_alloc_queue(&sdev->host->tag_set, &lim, sdev);
 	if (IS_ERR(q)) {
 		/* release fn is set up in scsi_sysfs_device_initialise, so
 		 * have to free and put manually here */
@@ -344,7 +344,6 @@ static struct scsi_device *scsi_alloc_sdev(struct scsi_target *starget,
 	}
 	kref_get(&sdev->host->tagset_refcnt);
 	sdev->request_queue = q;
-	q->queuedata = sdev;
 
 	depth = sdev->host->cmd_per_lun ?: 1;
 
-- 
2.31.1

[PATCH 2/2] scsi: bsg: Pass dev to blk_mq_alloc_queue()

[John Garry]

When calling bsg_setup_queue() -> blk_mq_alloc_queue(), we don't pass
the dev as the queuedata, but rather manually set it afterwards. Just
pass dev to blk_mq_alloc_queue() to have automatically set.

Signed-off-by: John Garry <john.g.garry@oracle.com>
---
 block/bsg-lib.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/block/bsg-lib.c b/block/bsg-lib.c
index ee738d129a9f..32da4a4429ce 100644
--- a/block/bsg-lib.c
+++ b/block/bsg-lib.c
@@ -385,13 +385,12 @@ struct request_queue *bsg_setup_queue(struct device *dev, const char *name,
 	if (blk_mq_alloc_tag_set(set))
 		goto out_tag_set;
 
-	q = blk_mq_alloc_queue(set, lim, NULL);
+	q = blk_mq_alloc_queue(set, lim, dev);
 	if (IS_ERR(q)) {
 		ret = PTR_ERR(q);
 		goto out_queue;
 	}
 
-	q->queuedata = dev;
 	blk_queue_rq_timeout(q, BLK_DEFAULT_SG_TIMEOUT);
 
 	bset->bd = bsg_register_queue(q, dev, name, bsg_transport_sg_io_fn);
-- 
2.31.1

Re: [PATCH 1/2] scsi: core: Pass sdev to blk_mq_alloc_queue()

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 2/2] scsi: bsg: Pass dev to blk_mq_alloc_queue()

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 1/2] scsi: core: Pass sdev to blk_mq_alloc_queue()

[Hannes Reinecke]

On 5/24/24 10:48, John Garry wrote:
> When calling scsi_alloc_sdev() -> blk_mq_alloc_queue(), we don't pass
> the sdev as the queuedata, but rather manually set it afterwards. Just
> pass to blk_mq_alloc_queue() to have automatically set.
> 
> Signed-off-by: John Garry <john.g.garry@oracle.com>
> ---
>   drivers/scsi/scsi_scan.c | 3 +--
>   1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
> index 463ce6e23dc6..ec54d58fb6c0 100644
> --- a/drivers/scsi/scsi_scan.c
> +++ b/drivers/scsi/scsi_scan.c
> @@ -334,7 +334,7 @@ static struct scsi_device *scsi_alloc_sdev(struct scsi_target *starget,
>   	sdev->sg_reserved_size = INT_MAX;
>   
>   	scsi_init_limits(shost, &lim);
> -	q = blk_mq_alloc_queue(&sdev->host->tag_set, &lim, NULL);
> +	q = blk_mq_alloc_queue(&sdev->host->tag_set, &lim, sdev);
>   	if (IS_ERR(q)) {
>   		/* release fn is set up in scsi_sysfs_device_initialise, so
>   		 * have to free and put manually here */
> @@ -344,7 +344,6 @@ static struct scsi_device *scsi_alloc_sdev(struct scsi_target *starget,
>   	}
>   	kref_get(&sdev->host->tagset_refcnt);
>   	sdev->request_queue = q;
> -	q->queuedata = sdev;
>   
>   	depth = sdev->host->cmd_per_lun ?: 1;
>   
Reviewed-by: Hannes Reinecke <hare@suse.de>

Cheers,

Hannes

Re: [PATCH 2/2] scsi: bsg: Pass dev to blk_mq_alloc_queue()

[Hannes Reinecke]

On 5/24/24 10:48, John Garry wrote:
> When calling bsg_setup_queue() -> blk_mq_alloc_queue(), we don't pass
> the dev as the queuedata, but rather manually set it afterwards. Just
> pass dev to blk_mq_alloc_queue() to have automatically set.
> 
> Signed-off-by: John Garry <john.g.garry@oracle.com>
> ---
>   block/bsg-lib.c | 3 +--
>   1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/block/bsg-lib.c b/block/bsg-lib.c
> index ee738d129a9f..32da4a4429ce 100644
> --- a/block/bsg-lib.c
> +++ b/block/bsg-lib.c
> @@ -385,13 +385,12 @@ struct request_queue *bsg_setup_queue(struct device *dev, const char *name,
>   	if (blk_mq_alloc_tag_set(set))
>   		goto out_tag_set;
>   
> -	q = blk_mq_alloc_queue(set, lim, NULL);
> +	q = blk_mq_alloc_queue(set, lim, dev);
>   	if (IS_ERR(q)) {
>   		ret = PTR_ERR(q);
>   		goto out_queue;
>   	}
>   
> -	q->queuedata = dev;
>   	blk_queue_rq_timeout(q, BLK_DEFAULT_SG_TIMEOUT);
>   
>   	bset->bd = bsg_register_queue(q, dev, name, bsg_transport_sg_io_fn);

Reviewed-by: Hannes Reinecke <hare@suse.de>

Cheers,

Hannes

Re: [PATCH 0/2] block, scsi: Small improvements for blk_mq_alloc_queue() usage

[Himanshu Madhani]

On 5/24/24 01:48, John Garry wrote:
> A couple of small improvements to not manually set q->queuedata.
> 
> I asked Himanshu (cc'ed) to test the bsg-lib.c change as I have no setup
> to test.
> 
> Based on mkp-scsi 6.10 staging queue at e4f5f8298cf6.
> 
> John Garry (2):
>    scsi: core: Pass sdev to blk_mq_alloc_queue()
>    scsi: bsg: Pass dev to blk_mq_alloc_queue()
> 
>   block/bsg-lib.c          | 3 +--
>   drivers/scsi/scsi_scan.c | 3 +--
>   2 files changed, 2 insertions(+), 4 deletions(-)
> 

Changes looks good for this series. I tested basic driver load/unload, 
as my setup is not connected to any Switch.

Tested-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

-- 
Himanshu Madhani                                Oracle Linux Engineering

Re: [PATCH 0/2] block, scsi: Small improvements for blk_mq_alloc_queue() usage

[Martin K. Petersen]

John,

> A couple of small improvements to not manually set q->queuedata.

Applied to 6.11/scsi-staging, thanks!

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: [PATCH 0/2] block, scsi: Small improvements for blk_mq_alloc_queue() usage

[Martin K. Petersen]

On Fri, 24 May 2024 08:48:27 +0000, John Garry wrote:

> A couple of small improvements to not manually set q->queuedata.
> 
> I asked Himanshu (cc'ed) to test the bsg-lib.c change as I have no setup
> to test.
> 
> Based on mkp-scsi 6.10 staging queue at e4f5f8298cf6.
> 
> [...]

Applied to 6.11/scsi-queue, thanks!

[1/2] scsi: core: Pass sdev to blk_mq_alloc_queue()
      https://git.kernel.org/mkp/scsi/c/e7c09df178f7
[2/2] scsi: bsg: Pass dev to blk_mq_alloc_queue()
      https://git.kernel.org/mkp/scsi/c/41b757425203

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: [PATCH 1/2] scsi: core: Pass sdev to blk_mq_alloc_queue()

[Hannes Reinecke]

On 5/24/24 10:48, John Garry wrote:
> When calling scsi_alloc_sdev() -> blk_mq_alloc_queue(), we don't pass
> the sdev as the queuedata, but rather manually set it afterwards. Just
> pass to blk_mq_alloc_queue() to have automatically set.
> 
> Signed-off-by: John Garry <john.g.garry@oracle.com>
> ---
>   drivers/scsi/scsi_scan.c | 3 +--
>   1 file changed, 1 insertion(+), 2 deletions(-)
> 
Reviewed-by: Hannes Reinecke <hare@suse.de>

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                  Kernel Storage Architect
hare@suse.de                                +49 911 74053 688
SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg
HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich