public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Chanwoo Choi <cw00.choi@samsung.com>
To: arvindY <arvind.yadav.cs@gmail.com>,
	myungjoo.ham@samsung.com, kyungmin.park@samsung.com
Cc: linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org
Subject: Re: [PATCH] PM / devfreq: use put_device() instead of kfree()
Date: Fri, 13 Apr 2018 11:29:34 +0900	[thread overview]
Message-ID: <5AD0160E.7050601@samsung.com> (raw)
In-Reply-To: <5AD012CB.7030003@gmail.com>

On 2018년 04월 13일 11:15, arvindY wrote:
> Hi Chanwoo,
> 
> On Friday 13 April 2018 06:43 AM, Chanwoo Choi wrote:
>> On 2018년 04월 13일 10:03, Chanwoo Choi wrote:
>>> Hi,
>>>
>>> I'm sorry for the late reply.
>>>
>>> On 2018년 03월 30일 20:44, Arvind Yadav wrote:
>>>> Never directly free @dev after calling device_register() or
>>>> device_unregister(), even if device_register() returned an error.
>>>> Always use put_device() to give up the reference initialized.
>>>>
>>>> Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
>>>> ---
>>>>   drivers/devfreq/devfreq.c | 4 +++-
>>>>   1 file changed, 3 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c
>>>> index fe2af6a..a225b94 100644
>>>> --- a/drivers/devfreq/devfreq.c
>>>> +++ b/drivers/devfreq/devfreq.c
>>>> @@ -625,7 +625,8 @@ struct devfreq *devfreq_add_device(struct device *dev,
>>>>       err = device_register(&devfreq->dev);
>>>>       if (err) {
>>>>           mutex_unlock(&devfreq->lock);
>>>> -        goto err_dev;
>>>> +        put_device(&devfreq->dev);
>>>> +        goto err_out;
>>> why do you change the goto postion?
>>> err_out is correct to free the memory of devfreq instance.
>> Sorry. err_dev is correct instead of err_out.
> 
> If you will see the comment for device_register(drivers/base/core.c)
> there is mentioned that 'NOTE: _Never_ directly free @dev
> after calling this function, even if it returned an error!
> Always use put_device() to give up the reference initialized
> in this function instead. Here  put_device() will decrement
> the last reference and then free the memory by calling dev->release.
> Internally put_device() -> kobject_put() -> kobject_cleanup() which
> is responsible to call 'dev -> release' and also free other kobject resources.
> 
> We are releasing devfreq in devfreq_dev_release(). So no need
> to call kfree() again. It'll be redundant.  'err_out' is correct.

You're right. err_out is correct.
put_device() -> dev->release() -> devfreq_dev_release() -> kfree(devfreq)

>>
>>>>       }
>>>>         devfreq->trans_table =    devm_kzalloc(&devfreq->dev,
>>>> @@ -671,6 +672,7 @@ struct devfreq *devfreq_add_device(struct device *dev,
>>>>       mutex_unlock(&devfreq_list_lock);
>>>>         device_unregister(&devfreq->dev);
>>>> +    devfreq = NULL;
>>> It is wrong. If you initialize the devfreq as NULL,
>>> never free the 'devfreq' instance.
> 
> No need to release memory after device_unregister().
> driver core will take care of this. It will release memory
> So no need to call free again.

If you have to initialize the devfreq instance as NULL,
I think that you better to init in the devfreq_dev_release()
as following:

diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c
index fe2af6aa88fc..8c52a13d3887 100644
--- a/drivers/devfreq/devfreq.c
+++ b/drivers/devfreq/devfreq.c
@@ -543,6 +543,7 @@ static void devfreq_dev_release(struct device *dev)
 
        mutex_destroy(&devfreq->lock);
        kfree(devfreq);
+       devfreq = NULL;
 }


>>>
>>>>   err_dev:
>>>>       if (devfreq)
>>>>           kfree(devfreq);
>>>>
>>>
>>
> 
> ~arvind
> 
> 
> 


-- 
Best Regards,
Chanwoo Choi
Samsung Electronics

  reply	other threads:[~2018-04-13  2:32 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20180330114418epcas5p1affd1c3d97e7422dd33e6d3ca73073be@epcas5p1.samsung.com>
2018-03-30 11:44 ` [PATCH] PM / devfreq: use put_device() instead of kfree() Arvind Yadav
2018-04-13  1:03   ` Chanwoo Choi
2018-04-13  1:13     ` Chanwoo Choi
2018-04-13  2:15       ` arvindY
2018-04-13  2:29         ` Chanwoo Choi [this message]
2018-04-13  2:37           ` arvindY
2018-04-13  2:43             ` Chanwoo Choi
2018-04-18  7:57               ` MyungJoo Ham

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5AD0160E.7050601@samsung.com \
    --to=cw00.choi@samsung.com \
    --cc=arvind.yadav.cs@gmail.com \
    --cc=kyungmin.park@samsung.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=myungjoo.ham@samsung.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox