mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Chanwoo Choi <cwchoi00@gmail.com>
To: Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Bumwoo Lee <bw365.lee@samsung.com>,
	linux-kernel@vger.kernel.org
Cc: MyungJoo Ham <myungjoo.ham@samsung.com>,
	Chanwoo Choi <cw00.choi@samsung.com>
Subject: Re: [PATCH v1 07/14] extcon: Use unique number for the extcon device ID
Date: Mon, 3 Apr 2023 23:52:46 +0900	[thread overview]
Message-ID: <f0a40f3a-2ff6-5529-ad84-1c66f7e381ea@gmail.com> (raw)
In-Reply-To: <20230322144005.40368-8-andriy.shevchenko@linux.intel.com>

On 23. 3. 22. 23:39, Andy Shevchenko wrote:
> The use of atomic variable is still racy when we do not control which
> device has been unregistered and there is a (theoretical) possibility
> of the overflow that may cause a duplicate extcon device ID number
> to be allocated next time a device is registered.
> 
> Replace above mentioned approach by using IDA framework.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  drivers/extcon/extcon.c | 15 ++++++++++++---
>  drivers/extcon/extcon.h |  2 ++
>  2 files changed, 14 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/extcon/extcon.c b/drivers/extcon/extcon.c
> index 14e66e21597f..0d261ec6c473 100644
> --- a/drivers/extcon/extcon.c
> +++ b/drivers/extcon/extcon.c
> @@ -16,6 +16,7 @@
>  
>  #include <linux/module.h>
>  #include <linux/types.h>
> +#include <linux/idr.h>
>  #include <linux/init.h>
>  #include <linux/device.h>
>  #include <linux/fs.h>
> @@ -238,6 +239,7 @@ struct extcon_cable {
>  
>  static struct class *extcon_class;
>  
> +static DEFINE_IDA(extcon_dev_ids);
>  static LIST_HEAD(extcon_dev_list);
>  static DEFINE_MUTEX(extcon_dev_list_lock);
>  
> @@ -1248,7 +1250,6 @@ static int extcon_alloc_groups(struct extcon_dev *edev)
>  int extcon_dev_register(struct extcon_dev *edev)
>  {
>  	int ret, index = 0;
> -	static atomic_t edev_no = ATOMIC_INIT(-1);
>  
>  	ret = create_extcon_class();
>  	if (ret < 0)
> @@ -1275,8 +1276,14 @@ int extcon_dev_register(struct extcon_dev *edev)
>  		dev_err(&edev->dev, "extcon device name is null\n");
>  		return -EINVAL;
>  	}
> -	dev_set_name(&edev->dev, "extcon%lu",
> -			(unsigned long)atomic_inc_return(&edev_no));
> +
> +	ret = ida_simple_get(&extcon_dev_ids, 0, INT_MAX, GFP_KERNEL);


ida_simple_get and ida_simple_remove are deprecated on 
commit 3264ceec8f17 ("lib/idr.c: document that ida_simple_{get,remove}()
are deprecated"). Instead of them, better to use ida_alloc and ida_free 
according to the comments.


> +	if (ret < 0)
> +		return ret;
> +
> +	edev->id = ret;
> +
> +	dev_set_name(&edev->dev, "extcon%d", edev->id);
>  
>  	ret = extcon_alloc_cables(edev);
>  	if (ret < 0)
> @@ -1368,6 +1375,8 @@ void extcon_dev_unregister(struct extcon_dev *edev)
>  		return;
>  	}
>  
> +	ida_simple_remove(&extcon_dev_ids, edev->id);

ditto.

> +
>  	device_unregister(&edev->dev);
>  
>  	if (edev->mutually_exclusive && edev->max_supported) {
> diff --git a/drivers/extcon/extcon.h b/drivers/extcon/extcon.h
> index 15616446140d..877c0860e300 100644
> --- a/drivers/extcon/extcon.h
> +++ b/drivers/extcon/extcon.h
> @@ -20,6 +20,7 @@
>   *			{0x3, 0x6, 0x5, 0}. If it is {0xFFFFFFFF, 0}, there
>   *			can be no simultaneous connections.
>   * @dev:		Device of this extcon.
> + * @id:			Unique device ID of this extcon.
>   * @state:		Attach/detach state of this extcon. Do not provide at
>   *			register-time.
>   * @nh_all:		Notifier for the state change events for all supported
> @@ -46,6 +47,7 @@ struct extcon_dev {
>  
>  	/* Internal data. Please do not set. */
>  	struct device dev;
> +	int id;
>  	struct raw_notifier_head nh_all;
>  	struct raw_notifier_head *nh;
>  	struct list_head entry;

-- 
Best Regards,
Samsung Electronics
Chanwoo Choi


  reply	other threads:[~2023-04-03 14:53 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-22 14:39 [PATCH v1 00/14] extcon: Core cleanups and documentation fixes Andy Shevchenko
2023-03-22 14:39 ` [PATCH v1 01/14] extcon: Fix kernel doc of property fields to avoid warnings Andy Shevchenko
2023-04-03 13:50   ` Chanwoo Choi
2023-03-22 14:39 ` [PATCH v1 02/14] extcon: Fix kernel doc of property capability " Andy Shevchenko
2023-04-03 13:51   ` Chanwoo Choi
2023-03-22 14:39 ` [PATCH v1 03/14] extcon: Use DECLARE_BITMAP() to declare bit arrays Andy Shevchenko
2023-04-03 14:04   ` Chanwoo Choi
2023-03-22 14:39 ` [PATCH v1 04/14] extcon: use sysfs_emit() to instead of sprintf() Andy Shevchenko
2023-04-03 14:10   ` Chanwoo Choi
2023-03-22 14:39 ` [PATCH v1 05/14] extcon: Amend kernel documentation of struct extcon_dev Andy Shevchenko
2023-04-03 14:10   ` Chanwoo Choi
2023-03-22 14:39 ` [PATCH v1 06/14] extcon: Allow name to be assigned outside of the framework Andy Shevchenko
2023-04-03 13:56   ` Chanwoo Choi
2023-03-22 14:39 ` [PATCH v1 07/14] extcon: Use unique number for the extcon device ID Andy Shevchenko
2023-04-03 14:52   ` Chanwoo Choi [this message]
2023-04-05 15:03     ` Andy Shevchenko
2023-03-22 14:39 ` [PATCH v1 08/14] extcon: Switch to use kasprintf_strarray() Andy Shevchenko
2023-04-03 14:58   ` Chanwoo Choi
2023-04-05  8:16     ` Andy Shevchenko
2023-04-05 15:05       ` Andy Shevchenko
2023-03-22 14:40 ` [PATCH v1 09/14] extcon: Use device_match_of_node() helper Andy Shevchenko
2023-04-03 14:14   ` Chanwoo Choi
2023-03-22 14:40 ` [PATCH v1 10/14] extcon: use dev_of_node(dev) instead of dev->of_node Andy Shevchenko
2023-04-03 14:17   ` Chanwoo Choi
2023-03-22 14:40 ` [PATCH v1 11/14] extcon: Remove dup device name in the message and unneeded error check Andy Shevchenko
2023-04-03 14:59   ` Chanwoo Choi
2023-03-22 14:40 ` [PATCH v1 12/14] extcon: Use sizeof(*pointer) instead of sizeof(type) Andy Shevchenko
2023-04-03 14:32   ` Chanwoo Choi
2023-03-22 14:40 ` [PATCH v1 13/14] extcon: Drop unneeded assignments Andy Shevchenko
2023-04-03 15:06   ` Chanwoo Choi
2023-03-22 14:40 ` [PATCH v1 14/14] extcon: Use positive conditional in ternary operator Andy Shevchenko
2023-04-03 14:38   ` Chanwoo Choi
2023-04-05  8:18     ` Andy Shevchenko
2023-03-30 10:11 ` [PATCH v1 00/14] extcon: Core cleanups and documentation fixes Andy Shevchenko
2023-03-31  0:48   ` Bumwoo Lee
2023-04-05 23:17     ` Chanwoo Choi
2023-04-06  0:04       ` Bumwoo Lee
2023-04-06 10:49       ` 'Andy Shevchenko'
2023-04-03 13:58 ` Chanwoo Choi
2023-04-05  8:12   ` Andy Shevchenko

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=f0a40f3a-2ff6-5529-ad84-1c66f7e381ea@gmail.com \
    --to=cwchoi00@gmail.com \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=bw365.lee@samsung.com \
    --cc=cw00.choi@samsung.com \
    --cc=linux-kernel@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

all inboxes | Powered by JetHome®