[PATCH next] driver core: auxiliary bus: Fix IS_ERR() vs NULL check in __devm_auxiliary_device_create()

Dan Carpenter posted 1 patch 7 months, 4 weeks ago
drivers/base/auxiliary.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
[PATCH next] driver core: auxiliary bus: Fix IS_ERR() vs NULL check in __devm_auxiliary_device_create()
Posted by Dan Carpenter 7 months, 4 weeks ago
The auxiliary_device_create() function returns NULL.  It doesn't return
error pointers.  Update the checking to match.

Fixes: eaa0d30216c1 ("driver core: auxiliary bus: add device creation helpers")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
---
 drivers/base/auxiliary.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/base/auxiliary.c b/drivers/base/auxiliary.c
index 810b6105a75d..61b876d90b7f 100644
--- a/drivers/base/auxiliary.c
+++ b/drivers/base/auxiliary.c
@@ -491,8 +491,8 @@ struct auxiliary_device *__devm_auxiliary_device_create(struct device *dev,
 	int ret;
 
 	auxdev = auxiliary_device_create(dev, modname, devname, platform_data, id);
-	if (IS_ERR(auxdev))
-		return auxdev;
+	if (!auxdev)
+		return ERR_PTR(-ENOMEM);
 
 	ret = devm_add_action_or_reset(dev, auxiliary_device_destroy,
 				       auxdev);
-- 
2.47.2
Re: [PATCH next] driver core: auxiliary bus: Fix IS_ERR() vs NULL check in __devm_auxiliary_device_create()
Posted by Jerome Brunet 7 months, 4 weeks ago
On Wed 23 Apr 2025 at 11:17, Dan Carpenter <dan.carpenter@linaro.org> wrote:

> The auxiliary_device_create() function returns NULL.  It doesn't return
> error pointers.  Update the checking to match.
>
> Fixes: eaa0d30216c1 ("driver core: auxiliary bus: add device creation helpers")
> Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>

Thanks for catching this mistake Dan.

The thing was initially setup to return error code. Greg asked to
simply return NULL on error and I forgot to re-align the devm variant.

So I think the fix should be to check for NULL as you did but return
NULL too so it is aligned with non-devm variant.

If you wish, I can handle a v2.

> ---
>  drivers/base/auxiliary.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/base/auxiliary.c b/drivers/base/auxiliary.c
> index 810b6105a75d..61b876d90b7f 100644
> --- a/drivers/base/auxiliary.c
> +++ b/drivers/base/auxiliary.c
> @@ -491,8 +491,8 @@ struct auxiliary_device *__devm_auxiliary_device_create(struct device *dev,
>  	int ret;
>  
>  	auxdev = auxiliary_device_create(dev, modname, devname, platform_data, id);
> -	if (IS_ERR(auxdev))
> -		return auxdev;
> +	if (!auxdev)
> +		return ERR_PTR(-ENOMEM);
>  
>  	ret = devm_add_action_or_reset(dev, auxiliary_device_destroy,
>  				       auxdev);

-- 
Jerome
Re: [PATCH next] driver core: auxiliary bus: Fix IS_ERR() vs NULL check in __devm_auxiliary_device_create()
Posted by Dan Carpenter 7 months, 4 weeks ago
On Wed, Apr 23, 2025 at 11:31:56AM +0200, Jerome Brunet wrote:
> On Wed 23 Apr 2025 at 11:17, Dan Carpenter <dan.carpenter@linaro.org> wrote:
> 
> > The auxiliary_device_create() function returns NULL.  It doesn't return
> > error pointers.  Update the checking to match.
> >
> > Fixes: eaa0d30216c1 ("driver core: auxiliary bus: add device creation helpers")
> > Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
> 
> Thanks for catching this mistake Dan.
> 
> The thing was initially setup to return error code. Greg asked to
> simply return NULL on error and I forgot to re-align the devm variant.
> 
> So I think the fix should be to check for NULL as you did but return
> NULL too so it is aligned with non-devm variant.
> 
> If you wish, I can handle a v2.

Of course, it's hard to resist an offer like that but I can send a v2.

regards,
dan carpenter