linux-kernel - [PATCHv2 1/5] devfreq: Fix devfreq_remove_device() to improve the sequence of resource free

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-id: <1399621391-27228-2-git-send-email-cw00.choi@samsung.com>
Date:	Fri, 09 May 2014 16:43:07 +0900
From:	Chanwoo Choi <cw00.choi@...sung.com>
To:	myungjoo.ham@...sung.com, kyungmin.park@...sung.com
Cc:	rafael.j.wysocki@...el.com, nm@...com, jkosina@...e.cz,
	labbe.montjoie@...il.com, a.kesavan@...sung.com,
	sreekumar.c@...sung.com, linux-pm@...r.kernel.org,
	linux-kernel@...r.kernel.org, Chanwoo Choi <cw00.choi@...sung.com>
Subject: [PATCHv2 1/5] devfreq: Fix devfreq_remove_device() to improve the
 sequence of resource free

This patch modify devfreq_remove_device() to improve the sequence of resource
free. If executing existing devfreq_remove_device(), this function always
executes _remove_devfreq() twice. In result, second _remove_devfreq() always
return error value. So, This patch resolves complicated function sequence
as following:

[Flow sequence before modification]
devfreq_remove_device()
   _remove_devfreq(devfreq, false)
      kfree(devfreq);  /* Free devfreq */
      if (!skip ...) { /* skip is false */
         device_unregister(&devfreq->dev)
         put_device(&devfreq->dev);
            ...
            dev->release()
               devfreq_dev_release()
                  _remove_devfreq(devfreq, true) <- Recall to free devfreq
                  /*
		   * Always return error without freeing resource because
		   * already _remove_devfreq() frees the memory of devfreq.
		   */
   }

[Flow sequence after modification]
devfreq_remove_device
   device_unregister(&devfreq->dev)
   put_device(&devfreq->dev);
      ..
      dev->release()
         devfreq_dev_release()
            _remove_devfreq()
               kfree(devfreq); /* Free devfreq */

Signed-off-by: Chanwoo Choi <cw00.choi@...sung.com>
---
 drivers/devfreq/devfreq.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c
index d32c8b6..e960eb0 100644
--- a/drivers/devfreq/devfreq.c
+++ b/drivers/devfreq/devfreq.c
@@ -437,7 +437,7 @@ update:
  * @devfreq:	the devfreq struct
  * @skip:	skip calling device_unregister().
  */
-static void _remove_devfreq(struct devfreq *devfreq, bool skip)
+static void _remove_devfreq(struct devfreq *devfreq)
 {
 	mutex_lock(&devfreq_list_lock);
 	if (IS_ERR(find_device_devfreq(devfreq->dev.parent))) {
@@ -462,11 +462,6 @@ static void _remove_devfreq(struct devfreq *devfreq, bool skip)
 	if (devfreq->profile->exit)
 		devfreq->profile->exit(devfreq->dev.parent);
 
-	if (!skip && get_device(&devfreq->dev)) {
-		device_unregister(&devfreq->dev);
-		put_device(&devfreq->dev);
-	}
-
 	mutex_destroy(&devfreq->lock);
 	kfree(devfreq);
 }
@@ -476,14 +471,12 @@ static void _remove_devfreq(struct devfreq *devfreq, bool skip)
  * @dev:	the devfreq device
  *
  * This calls _remove_devfreq() if _remove_devfreq() is not called.
- * Note that devfreq_dev_release() could be called by _remove_devfreq() as
- * well as by others unregistering the device.
  */
 static void devfreq_dev_release(struct device *dev)
 {
 	struct devfreq *devfreq = to_devfreq(dev);
 
-	_remove_devfreq(devfreq, true);
+	_remove_devfreq(devfreq);
 }
 
 static int devfreq_qos_sanity_check(struct device *dev,
@@ -659,7 +652,8 @@ int devfreq_remove_device(struct devfreq *devfreq)
 	if (!devfreq)
 		return -EINVAL;
 
-	_remove_devfreq(devfreq, false);
+	device_unregister(&devfreq->dev);
+	put_device(&devfreq->dev);
 
 	return 0;
 }
-- 
1.8.0

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