Re: [PATCH] [PATCH] extcon: driver model release call not needed

From: Chanwoo Choi
Date: Fri Oct 19 2012 - 22:37:11 EST


On 10/20/2012 11:30 AM, anish kumar wrote:
> On Sat, 2012-10-20 at 10:57 +0900, Chanwoo Choi wrote:
>> On 10/19/2012 02:12 AM, anish kumar wrote:
>>> From: anish kumar <anish198519851985@xxxxxxxxx>
>>>
>>> We don't need a release call in this file as we are doing
>>> everything needed in unregister call and we don't have any
>>> more pointer to free up.
>>>
>>> Signed-off-by: anish kumar <anish198519851985@xxxxxxxxx>
>>> ---
>>> drivers/extcon/extcon-class.c | 4 +---
>>> 1 files changed, 1 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/extcon/extcon-class.c b/drivers/extcon/extcon-class.c
>>> index 946a318..cf30eb1 100644
>>> --- a/drivers/extcon/extcon-class.c
>>> +++ b/drivers/extcon/extcon-class.c
>>> @@ -585,9 +585,7 @@ static void extcon_cleanup(struct extcon_dev *edev, bool skip)
>>>
>>> static void extcon_dev_release(struct device *dev)
>>> {
>>> - struct extcon_dev *edev = (struct extcon_dev *) dev_get_drvdata(dev);
>>> -
>>> - extcon_cleanup(edev, true);
>>> + /* We don't have any thing to free here */
>>> }
>>>
>>> static const char *muex_name = "mutually_exclusive";
>>
>> I can't agree this patch. The extcon_dev_release() function is used
>> for dev->release. If some case without calling extcon_dev_unregister(),
>> I think dev->release function is needed to free memory of edev->dev.
> Is it not being released by extcon_dev_unregister?
> I think it is released by that and we will do two times free and
> list_del(&edev->entry) as it is called by extcon_dev_release also.

I think that this patch should modify it as below patch to remove
two call of kfree(). How about you?

diff --git a/drivers/extcon/extcon-class.c b/drivers/extcon/extcon-class.c
index e717bbc..efca0b4 100644
--- a/drivers/extcon/extcon-class.c
+++ b/drivers/extcon/extcon-class.c
@@ -597,9 +597,8 @@ static void extcon_cleanup(struct extcon_dev *edev, bool skip)
#endif
device_unregister(edev->dev);
put_device(edev->dev);
- }
-
- kfree(edev->dev);
+ } else {
+ kfree(edev->dev);

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/