[PATCH v3] usb: roles: fix of node refcount leak in usb_role_switch_is_parent()

[PATCH v3] usb: roles: fix of node refcount leak in usb_role_switch_is_parent()

[Yang Yingliang]

I got the following report while doing device(mt6370-tcpc) load
test with CONFIG_OF_UNITTEST and CONFIG_OF_DYNAMIC enabled:

  OF: ERROR: memory leak, expected refcount 1 instead of 2,
  of_node_get()/of_node_put() unbalanced - destroy cset entry:
  attach overlay node /i2c/pmic@34

The 'parent' returned by fwnode_get_parent() with refcount incremented.
it needs be put after using.

Fixes: 6fadd72943b8 ("usb: roles: get usb-role-switch from parent")
Suggested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
v2 -> v3:
  Remove not needed null pointer check.

v1 -> v2:
  Add description to how is the report generated.
---
 drivers/usb/roles/class.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
index dfaed7eee94f..0650295f261c 100644
--- a/drivers/usb/roles/class.c
+++ b/drivers/usb/roles/class.c
@@ -106,10 +106,13 @@ usb_role_switch_is_parent(struct fwnode_handle *fwnode)
 	struct fwnode_handle *parent = fwnode_get_parent(fwnode);
 	struct device *dev;
 
-	if (!parent || !fwnode_property_present(parent, "usb-role-switch"))
+	if (!parent || !fwnode_property_present(parent, "usb-role-switch")) {
+		fwnode_handle_put(parent);
 		return NULL;
+	}
 
 	dev = class_find_device_by_fwnode(role_class, parent);
+	fwnode_handle_put(parent);
 	return dev ? to_role_switch(dev) : ERR_PTR(-EPROBE_DEFER);
 }
 
-- 
2.25.1

Re: [PATCH v3] usb: roles: fix of node refcount leak in usb_role_switch_is_parent()

[Heikki Krogerus]

Hi,

On Mon, Nov 21, 2022 at 10:46:20PM +0800, Yang Yingliang wrote:
> I got the following report while doing device(mt6370-tcpc) load
> test with CONFIG_OF_UNITTEST and CONFIG_OF_DYNAMIC enabled:
> 
>   OF: ERROR: memory leak, expected refcount 1 instead of 2,
>   of_node_get()/of_node_put() unbalanced - destroy cset entry:
>   attach overlay node /i2c/pmic@34
> 
> The 'parent' returned by fwnode_get_parent() with refcount incremented.
> it needs be put after using.
> 
> Fixes: 6fadd72943b8 ("usb: roles: get usb-role-switch from parent")
> Suggested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>

That's not the appropriate tag in this case. I have not suggested
this patch.

> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
> ---
> v2 -> v3:
>   Remove not needed null pointer check.
> 
> v1 -> v2:
>   Add description to how is the report generated.
> ---
>  drivers/usb/roles/class.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
> index dfaed7eee94f..0650295f261c 100644
> --- a/drivers/usb/roles/class.c
> +++ b/drivers/usb/roles/class.c
> @@ -106,10 +106,13 @@ usb_role_switch_is_parent(struct fwnode_handle *fwnode)
>  	struct fwnode_handle *parent = fwnode_get_parent(fwnode);
>  	struct device *dev;
>  
> -	if (!parent || !fwnode_property_present(parent, "usb-role-switch"))
> +	if (!parent || !fwnode_property_present(parent, "usb-role-switch")) {

Please change that as well like I proposed earlier:

	if (!fwnode_property_present(parent, "usb-role-switch")) {

You don't need to check the parent separately.

> +		fwnode_handle_put(parent);
>  		return NULL;
> +	}
>  
>  	dev = class_find_device_by_fwnode(role_class, parent);
> +	fwnode_handle_put(parent);
>  	return dev ? to_role_switch(dev) : ERR_PTR(-EPROBE_DEFER);
>  }
>  

thanks,

-- 
heikki

Re: [PATCH v3] usb: roles: fix of node refcount leak in usb_role_switch_is_parent()

[Yang Yingliang]

On 2022/11/22 16:33, Heikki Krogerus wrote:
> Hi,
>
> On Mon, Nov 21, 2022 at 10:46:20PM +0800, Yang Yingliang wrote:
>> I got the following report while doing device(mt6370-tcpc) load
>> test with CONFIG_OF_UNITTEST and CONFIG_OF_DYNAMIC enabled:
>>
>>    OF: ERROR: memory leak, expected refcount 1 instead of 2,
>>    of_node_get()/of_node_put() unbalanced - destroy cset entry:
>>    attach overlay node /i2c/pmic@34
>>
>> The 'parent' returned by fwnode_get_parent() with refcount incremented.
>> it needs be put after using.
>>
>> Fixes: 6fadd72943b8 ("usb: roles: get usb-role-switch from parent")
>> Suggested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
> That's not the appropriate tag in this case. I have not suggested
> this patch.
>
>> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
>> ---
>> v2 -> v3:
>>    Remove not needed null pointer check.
>>
>> v1 -> v2:
>>    Add description to how is the report generated.
>> ---
>>   drivers/usb/roles/class.c | 5 ++++-
>>   1 file changed, 4 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
>> index dfaed7eee94f..0650295f261c 100644
>> --- a/drivers/usb/roles/class.c
>> +++ b/drivers/usb/roles/class.c
>> @@ -106,10 +106,13 @@ usb_role_switch_is_parent(struct fwnode_handle *fwnode)
>>   	struct fwnode_handle *parent = fwnode_get_parent(fwnode);
>>   	struct device *dev;
>>   
>> -	if (!parent || !fwnode_property_present(parent, "usb-role-switch"))
>> +	if (!parent || !fwnode_property_present(parent, "usb-role-switch")) {
> Please change that as well like I proposed earlier:
>
> 	if (!fwnode_property_present(parent, "usb-role-switch")) {
>
> You don't need to check the parent separately.
Yes, fwnode_property_present() has already checked the parent, the 
parent check can be removed.
Need I add your suggest tag when I send v4 to change it as above.

Thanks,
Yang
>
>> +		fwnode_handle_put(parent);
>>   		return NULL;
>> +	}
>>   
>>   	dev = class_find_device_by_fwnode(role_class, parent);
>> +	fwnode_handle_put(parent);
>>   	return dev ? to_role_switch(dev) : ERR_PTR(-EPROBE_DEFER);
>>   }
>>   
> thanks,
>

Re: [PATCH v3] usb: roles: fix of node refcount leak in usb_role_switch_is_parent()

[Heikki Krogerus]

On Tue, Nov 22, 2022 at 04:47:00PM +0800, Yang Yingliang wrote:
> 
> On 2022/11/22 16:33, Heikki Krogerus wrote:
> > Hi,
> > 
> > On Mon, Nov 21, 2022 at 10:46:20PM +0800, Yang Yingliang wrote:
> > > I got the following report while doing device(mt6370-tcpc) load
> > > test with CONFIG_OF_UNITTEST and CONFIG_OF_DYNAMIC enabled:
> > > 
> > >    OF: ERROR: memory leak, expected refcount 1 instead of 2,
> > >    of_node_get()/of_node_put() unbalanced - destroy cset entry:
> > >    attach overlay node /i2c/pmic@34
> > > 
> > > The 'parent' returned by fwnode_get_parent() with refcount incremented.
> > > it needs be put after using.
> > > 
> > > Fixes: 6fadd72943b8 ("usb: roles: get usb-role-switch from parent")
> > > Suggested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
> > That's not the appropriate tag in this case. I have not suggested
> > this patch.
> > 
> > > Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
> > > ---
> > > v2 -> v3:
> > >    Remove not needed null pointer check.
> > > 
> > > v1 -> v2:
> > >    Add description to how is the report generated.
> > > ---
> > >   drivers/usb/roles/class.c | 5 ++++-
> > >   1 file changed, 4 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
> > > index dfaed7eee94f..0650295f261c 100644
> > > --- a/drivers/usb/roles/class.c
> > > +++ b/drivers/usb/roles/class.c
> > > @@ -106,10 +106,13 @@ usb_role_switch_is_parent(struct fwnode_handle *fwnode)
> > >   	struct fwnode_handle *parent = fwnode_get_parent(fwnode);
> > >   	struct device *dev;
> > > -	if (!parent || !fwnode_property_present(parent, "usb-role-switch"))
> > > +	if (!parent || !fwnode_property_present(parent, "usb-role-switch")) {
> > Please change that as well like I proposed earlier:
> > 
> > 	if (!fwnode_property_present(parent, "usb-role-switch")) {
> > 
> > You don't need to check the parent separately.
> Yes, fwnode_property_present() has already checked the parent, the parent
> check can be removed.
> Need I add your suggest tag when I send v4 to change it as above.

No. I still have not suggested this patch. I've only commented it.

thanks,

-- 
heikki