[PATCH v2] dmaengine: idxd: Remove usage of the deprecated ida_simple_xx() API

[PATCH v2] dmaengine: idxd: Remove usage of the deprecated ida_simple_xx() API

[Christophe JAILLET]

ida_alloc() and ida_free() should be preferred to the deprecated
ida_simple_get() and ida_simple_remove().

This is less verbose.

Note that the upper limit of ida_simple_get() is exclusive, but the one of
ida_alloc_range() is inclusive. Sothis change allows one more device.

MINORMASK is ((1U << MINORBITS) - 1), so allowing MINORMASK as a maximum value
makes sense. It is also consistent with other "ida_.*MINORMASK" and
"ida_*MINOR()" usages.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
Changes v2:
  - remove note about compile tested only, now tested on hw by Fenghua Yu
  - fix some wording in the commit description   [Fenghua Yu]
  
v1: https://lore.kernel.org/all/a899125f42c12fa782a881d341d147519cbb4a23.1702967302.git.christophe.jaillet@wanadoo.fr/
---
 drivers/dma/idxd/cdev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c
index 0423655f5a88..b00926abc69a 100644
--- a/drivers/dma/idxd/cdev.c
+++ b/drivers/dma/idxd/cdev.c
@@ -165,7 +165,7 @@ static void idxd_cdev_dev_release(struct device *dev)
 	struct idxd_wq *wq = idxd_cdev->wq;
 
 	cdev_ctx = &ictx[wq->idxd->data->type];
-	ida_simple_remove(&cdev_ctx->minor_ida, idxd_cdev->minor);
+	ida_free(&cdev_ctx->minor_ida, idxd_cdev->minor);
 	kfree(idxd_cdev);
 }
 
@@ -463,7 +463,7 @@ int idxd_wq_add_cdev(struct idxd_wq *wq)
 	cdev = &idxd_cdev->cdev;
 	dev = cdev_dev(idxd_cdev);
 	cdev_ctx = &ictx[wq->idxd->data->type];
-	minor = ida_simple_get(&cdev_ctx->minor_ida, 0, MINORMASK, GFP_KERNEL);
+	minor = ida_alloc_max(&cdev_ctx->minor_ida, MINORMASK, GFP_KERNEL);
 	if (minor < 0) {
 		kfree(idxd_cdev);
 		return minor;
-- 
2.34.1

Re: [PATCH v2] dmaengine: idxd: Remove usage of the deprecated ida_simple_xx() API

[Lijun Pan]

On 12/19/2023 1:33 PM, Christophe JAILLET wrote:
> ida_alloc() and ida_free() should be preferred to the deprecated
> ida_simple_get() and ida_simple_remove().
> 
> This is less verbose.
> 
> Note that the upper limit of ida_simple_get() is exclusive, but the one of
> ida_alloc_range() is inclusive. Sothis change allows one more device. >
> MINORMASK is ((1U << MINORBITS) - 1), so allowing MINORMASK as a maximum value
> makes sense. It is also consistent with other "ida_.*MINORMASK" and
> "ida_*MINOR()" usages.
> 
> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---

Acked-by: Lijun Pan <lijun.pan@intel.com>


> Changes v2:
>    - remove note about compile tested only, now tested on hw by Fenghua Yu
>    - fix some wording in the commit description   [Fenghua Yu]
>    
> v1: https://lore.kernel.org/all/a899125f42c12fa782a881d341d147519cbb4a23.1702967302.git.christophe.jaillet@wanadoo.fr/
> ---
>   drivers/dma/idxd/cdev.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c
> index 0423655f5a88..b00926abc69a 100644
> --- a/drivers/dma/idxd/cdev.c
> +++ b/drivers/dma/idxd/cdev.c
> @@ -165,7 +165,7 @@ static void idxd_cdev_dev_release(struct device *dev)
>   	struct idxd_wq *wq = idxd_cdev->wq;
>   
>   	cdev_ctx = &ictx[wq->idxd->data->type];
> -	ida_simple_remove(&cdev_ctx->minor_ida, idxd_cdev->minor);
> +	ida_free(&cdev_ctx->minor_ida, idxd_cdev->minor);
>   	kfree(idxd_cdev);
>   }
>   
> @@ -463,7 +463,7 @@ int idxd_wq_add_cdev(struct idxd_wq *wq)
>   	cdev = &idxd_cdev->cdev;
>   	dev = cdev_dev(idxd_cdev);
>   	cdev_ctx = &ictx[wq->idxd->data->type];
> -	minor = ida_simple_get(&cdev_ctx->minor_ida, 0, MINORMASK, GFP_KERNEL);
> +	minor = ida_alloc_max(&cdev_ctx->minor_ida, MINORMASK, GFP_KERNEL);

It shouldn't be an issue though it should be below statement if directly 
converting to ida_alloc_max
+	minor = ida_alloc_max(&cdev_ctx->minor_ida, MINORMASK - 1, GFP_KERNEL);

>   	if (minor < 0) {
>   		kfree(idxd_cdev);
>   		return minor;

Re: [PATCH v2] dmaengine: idxd: Remove usage of the deprecated ida_simple_xx() API

[Fenghua Yu]

On 12/19/23 11:33, Christophe JAILLET wrote:
> ida_alloc() and ida_free() should be preferred to the deprecated
> ida_simple_get() and ida_simple_remove().
> 
> This is less verbose.
> 
> Note that the upper limit of ida_simple_get() is exclusive, but the one of
> ida_alloc_range() is inclusive. Sothis change allows one more device.
> 
> MINORMASK is ((1U << MINORBITS) - 1), so allowing MINORMASK as a maximum value
> makes sense. It is also consistent with other "ida_.*MINORMASK" and
> "ida_*MINOR()" usages.
> 
> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>

Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>

Thanks.

-Fenghua

Re: [PATCH v2] dmaengine: idxd: Remove usage of the deprecated ida_simple_xx() API

[Vinod Koul]

On Tue, 19 Dec 2023 20:33:50 +0100, Christophe JAILLET wrote:
> ida_alloc() and ida_free() should be preferred to the deprecated
> ida_simple_get() and ida_simple_remove().
> 
> This is less verbose.
> 
> Note that the upper limit of ida_simple_get() is exclusive, but the one of
> ida_alloc_range() is inclusive. Sothis change allows one more device.
> 
> [...]

Applied, thanks!

[1/1] dmaengine: idxd: Remove usage of the deprecated ida_simple_xx() API
      commit: 1075ee66a8c19bfa375b19c236fd6a22a867f138

Best regards,
-- 
~Vinod