Re: [PATCH] cxl: Fix memory allocation failure test

[Ian Munsie]

Acked-by: Ian Munsie 

Re: [PATCH] cxl: Fix memory allocation failure test

[Frederic Barrat]

Le 30/10/2016 à 20:35, Christophe JAILLET a écrit :

'cxl_context_alloc()' does not return an error pointer. It is just a
shortcut for a call to 'kzalloc' with 'sizeof(struct cxl_context)' as the
size parameter.

So its return value should be compared with NULL.
While fixing it, simplify a bit the code.

Signed-off-by: Christophe JAILLET 
---



Acked-by: Frederic Barrat 

Re: [PATCH] cxl: Fix memory allocation failure test

[Andrew Donnellan]

On 31/10/16 06:35, Christophe JAILLET wrote:

'cxl_context_alloc()' does not return an error pointer. It is just a
shortcut for a call to 'kzalloc' with 'sizeof(struct cxl_context)' as the
size parameter.

So its return value should be compared with NULL.
While fixing it, simplify a bit the code.

Signed-off-by: Christophe JAILLET 


Reviewed-by: Andrew Donnellan 


---
un-compiled because I don't have the required  cross build environment.


Compiles on my system!

--
Andrew Donnellan  OzLabs, ADL Canberra
andrew.donnel...@au1.ibm.com  IBM Australia Limited

Re: [PATCH] cxl: Fix memory allocation failure test

[walter harms]

Am 30.10.2016 20:35, schrieb Christophe JAILLET:
> 'cxl_context_alloc()' does not return an error pointer. It is just a
> shortcut for a call to 'kzalloc' with 'sizeof(struct cxl_context)' as the
> size parameter.
> 
> So its return value should be compared with NULL.
> While fixing it, simplify a bit the code.
> 
> Signed-off-by: Christophe JAILLET 
> ---
> un-compiled because I don't have the required  cross build environment.
> ---
>  drivers/misc/cxl/api.c | 7 ++-
>  1 file changed, 2 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c
> index 2e5233b60971..2b88ad8a2a89 100644
> --- a/drivers/misc/cxl/api.c
> +++ b/drivers/misc/cxl/api.c
> @@ -30,10 +30,8 @@ struct cxl_context *cxl_dev_context_init(struct pci_dev 
> *dev)
>   return ERR_CAST(afu);
>  
>   ctx = cxl_context_alloc();
> - if (IS_ERR(ctx)) {
> - rc = PTR_ERR(ctx);
> - goto err_dev;
> - }
> + if (!ctx)
> + return ERR_PTR(-ENOMEM);
>  

So far i see it is only used 2 times,
To avoid such problems it should be replaced with
kzalloc(sizeof(struct cxl_context), GFP_KERNEL); (from context.c)

or even better:
ctx = kzalloc(*ctx, GFP_KERNEL);

This way the error had been spotted much more early.

just my 2 cents,
re,
 wh

>   ctx->kernelapi = true;
>  
> @@ -61,7 +59,6 @@ struct cxl_context *cxl_dev_context_init(struct pci_dev 
> *dev)
>   kfree(mapping);
>  err_ctx:
>   kfree(ctx);
> -err_dev:
>   return ERR_PTR(rc);
>  }
>  EXPORT_SYMBOL_GPL(cxl_dev_context_init);

[PATCH] cxl: Fix memory allocation failure test

[Christophe JAILLET]

'cxl_context_alloc()' does not return an error pointer. It is just a
shortcut for a call to 'kzalloc' with 'sizeof(struct cxl_context)' as the
size parameter.

So its return value should be compared with NULL.
While fixing it, simplify a bit the code.

Signed-off-by: Christophe JAILLET 
---
un-compiled because I don't have the required  cross build environment.
---
 drivers/misc/cxl/api.c | 7 ++-
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c
index 2e5233b60971..2b88ad8a2a89 100644
--- a/drivers/misc/cxl/api.c
+++ b/drivers/misc/cxl/api.c
@@ -30,10 +30,8 @@ struct cxl_context *cxl_dev_context_init(struct pci_dev *dev)
return ERR_CAST(afu);
 
ctx = cxl_context_alloc();
-   if (IS_ERR(ctx)) {
-   rc = PTR_ERR(ctx);
-   goto err_dev;
-   }
+   if (!ctx)
+   return ERR_PTR(-ENOMEM);
 
ctx->kernelapi = true;
 
@@ -61,7 +59,6 @@ struct cxl_context *cxl_dev_context_init(struct pci_dev *dev)
kfree(mapping);
 err_ctx:
kfree(ctx);
-err_dev:
return ERR_PTR(rc);
 }
 EXPORT_SYMBOL_GPL(cxl_dev_context_init);
-- 
2.9.3