linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] net-sysfs: call dev_hold if kobject_init_and_add success
@ 2019-03-19  2:16 Yue Haibing
  2019-03-19 20:46 ` David Miller
  0 siblings, 1 reply; 2+ messages in thread
From: Yue Haibing @ 2019-03-19  2:16 UTC (permalink / raw)
  To: davem, idosch, alexander.h.duyck, tyhicks
  Cc: linux-kernel, netdev, YueHaibing

From: YueHaibing <yuehaibing@huawei.com>

In netdev_queue_add_kobject and rx_queue_add_kobject,
if sysfs_create_group failed, kobject_put will call
netdev_queue_release to decrease dev refcont, however
dev_hold has not be called. So we will see this while
unregistering dev:

unregister_netdevice: waiting for bcsh0 to become free. Usage count = -1

Reported-by: Hulk Robot <hulkci@huawei.com>
Fixes: d0d668371679 ("net: don't decrement kobj reference count on init failure")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
---
 net/core/net-sysfs.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 4ff661f..8f8b7b6 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -928,6 +928,8 @@ static int rx_queue_add_kobject(struct net_device *dev, int index)
 	if (error)
 		return error;
 
+	dev_hold(queue->dev);
+
 	if (dev->sysfs_rx_queue_group) {
 		error = sysfs_create_group(kobj, dev->sysfs_rx_queue_group);
 		if (error) {
@@ -937,7 +939,6 @@ static int rx_queue_add_kobject(struct net_device *dev, int index)
 	}
 
 	kobject_uevent(kobj, KOBJ_ADD);
-	dev_hold(queue->dev);
 
 	return error;
 }
@@ -1464,6 +1465,8 @@ static int netdev_queue_add_kobject(struct net_device *dev, int index)
 	if (error)
 		return error;
 
+	dev_hold(queue->dev);
+
 #ifdef CONFIG_BQL
 	error = sysfs_create_group(kobj, &dql_group);
 	if (error) {
@@ -1473,7 +1476,6 @@ static int netdev_queue_add_kobject(struct net_device *dev, int index)
 #endif
 
 	kobject_uevent(kobj, KOBJ_ADD);
-	dev_hold(queue->dev);
 
 	return 0;
 }
-- 
2.7.0



^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] net-sysfs: call dev_hold if kobject_init_and_add success
  2019-03-19  2:16 [PATCH] net-sysfs: call dev_hold if kobject_init_and_add success Yue Haibing
@ 2019-03-19 20:46 ` David Miller
  0 siblings, 0 replies; 2+ messages in thread
From: David Miller @ 2019-03-19 20:46 UTC (permalink / raw)
  To: yuehaibing; +Cc: idosch, alexander.h.duyck, tyhicks, linux-kernel, netdev

From: Yue Haibing <yuehaibing@huawei.com>
Date: Tue, 19 Mar 2019 10:16:53 +0800

> From: YueHaibing <yuehaibing@huawei.com>
> 
> In netdev_queue_add_kobject and rx_queue_add_kobject,
> if sysfs_create_group failed, kobject_put will call
> netdev_queue_release to decrease dev refcont, however
> dev_hold has not be called. So we will see this while
> unregistering dev:
> 
> unregister_netdevice: waiting for bcsh0 to become free. Usage count = -1
> 
> Reported-by: Hulk Robot <hulkci@huawei.com>
> Fixes: d0d668371679 ("net: don't decrement kobj reference count on init failure")
> Signed-off-by: YueHaibing <yuehaibing@huawei.com>

Applied and queued up for -stable.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2019-03-19 20:46 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-03-19  2:16 [PATCH] net-sysfs: call dev_hold if kobject_init_and_add success Yue Haibing
2019-03-19 20:46 ` David Miller

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).