diff mbox series

[v2] usb: roles: fix of node refcount leak in usb_role_switch_is_parent()

Message ID 20221119091541.1755640-1-yangyingliang@huawei.com (mailing list archive)
State Superseded
Headers show
Series [v2] usb: roles: fix of node refcount leak in usb_role_switch_is_parent() | expand

Commit Message

Yang Yingliang Nov. 19, 2022, 9:15 a.m. UTC
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")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
v1 -> v2:
  Add description to how is the report generated.
---
 drivers/usb/roles/class.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

Comments

Chunfeng Yun Nov. 21, 2022, 7:49 a.m. UTC | #1
On Sat, 2022-11-19 at 17:15 +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")
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
> ---
> v1 -> v2:
>   Add description to how is the report generated.
> ---
>  drivers/usb/roles/class.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
> index dfaed7eee94f..289950e5fcfb 100644
> --- a/drivers/usb/roles/class.c
> +++ b/drivers/usb/roles/class.c
> @@ -106,10 +106,14 @@ 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")) {
> +		if (parent)
> +			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);
>  }
>  
+ Heikki
Chunfeng Yun Nov. 21, 2022, 7:55 a.m. UTC | #2
Hi Yingliang,

On Sat, 2022-11-19 at 17:15 +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")
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
> ---
> v1 -> v2:
>   Add description to how is the report generated.
> ---
>  drivers/usb/roles/class.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
> index dfaed7eee94f..289950e5fcfb 100644
> --- a/drivers/usb/roles/class.c
> +++ b/drivers/usb/roles/class.c
> @@ -106,10 +106,14 @@ 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")) {
> +		if (parent)
> +			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);
>  }
Please use get_maintainer script to find who you should sent to after
generating patches

e.g.:
./scripts/get_maintainer.pl --no-rolestats 00* | gawk '{printf("
--cc=\"%s\" ", $0)} END{printf("\n")}'

Thanks

>
Yang Yingliang Nov. 21, 2022, 8:28 a.m. UTC | #3
On 2022/11/21 15:55, Chunfeng Yun (云春峰) wrote:
> Hi Yingliang,
>
> On Sat, 2022-11-19 at 17:15 +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")
>> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
>> ---
>> v1 -> v2:
>>    Add description to how is the report generated.
>> ---
>>   drivers/usb/roles/class.c | 6 +++++-
>>   1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
>> index dfaed7eee94f..289950e5fcfb 100644
>> --- a/drivers/usb/roles/class.c
>> +++ b/drivers/usb/roles/class.c
>> @@ -106,10 +106,14 @@ 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")) {
>> +		if (parent)
>> +			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);
>>   }
> Please use get_maintainer script to find who you should sent to after
> generating patches
>
> e.g.:
> ./scripts/get_maintainer.pl --no-rolestats 00* | gawk '{printf("
> --cc=\"%s\" ", $0)} END{printf("\n")}'
>
> Thanks
I used get_maintainer.pl to get the maintainer in Linux 6.1-rc5 and sent 
the patch.

./scripts/get_maintainer.pl 
0001-usb-roles-fix-of-node-refcount-leak-in-usb_role_swit.patch
Greg Kroah-Hartman <gregkh@linuxfoundation.org> (supporter:USB 
SUBSYSTEM,blamed_fixes:1/1=100%)
Chunfeng Yun <chunfeng.yun@mediatek.com> (blamed_fixes:1/1=100%)
linux-usb@vger.kernel.org (open list:USB SUBSYSTEM)
linux-kernel@vger.kernel.org (open list)

Thanks,
Yang
>
>>
Heikki Krogerus Nov. 21, 2022, 8:55 a.m. UTC | #4
Hi,

On Mon, Nov 21, 2022 at 07:49:31AM +0000, Chunfeng Yun (云春峰) wrote:
> On Sat, 2022-11-19 at 17:15 +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")
> > Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
> > ---
> > v1 -> v2:
> >   Add description to how is the report generated.
> > ---
> >  drivers/usb/roles/class.c | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
> > index dfaed7eee94f..289950e5fcfb 100644
> > --- a/drivers/usb/roles/class.c
> > +++ b/drivers/usb/roles/class.c
> > @@ -106,10 +106,14 @@ 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")) {
> > +		if (parent)
> > +			fwnode_handle_put(parent);
> >  		return NULL;
> > +	}

fwnode API should be NULL safe, so perhaps like this - clean also the
old condition while at it:

        if (!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);
> >  }
> >  
> + Heikki

Thanks Chunfeng!

cheers,
Yang Yingliang Nov. 21, 2022, 9:01 a.m. UTC | #5
Hi,

On 2022/11/21 16:55, Heikki Krogerus wrote:
> Hi,
>
> On Mon, Nov 21, 2022 at 07:49:31AM +0000, Chunfeng Yun (云春峰) wrote:
>> On Sat, 2022-11-19 at 17:15 +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")
>>> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
>>> ---
>>> v1 -> v2:
>>>    Add description to how is the report generated.
>>> ---
>>>   drivers/usb/roles/class.c | 6 +++++-
>>>   1 file changed, 5 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
>>> index dfaed7eee94f..289950e5fcfb 100644
>>> --- a/drivers/usb/roles/class.c
>>> +++ b/drivers/usb/roles/class.c
>>> @@ -106,10 +106,14 @@ 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")) {
>>> +		if (parent)
>>> +			fwnode_handle_put(parent);
>>>   		return NULL;
>>> +	}
> fwnode API should be NULL safe, so perhaps like this - clean also the
> old condition while at it:
>
>          if (!fwnode_property_present(parent, "usb-role-switch")) {
>                  fwnode_handle_put(parent);
>                  return NULL;
>          }
Thanks for your suggestion, I will send a v3 to change this.

Thanks,
Yang
>
>>>   	dev = class_find_device_by_fwnode(role_class, parent);
>>> +	fwnode_handle_put(parent);
>>>   	return dev ? to_role_switch(dev) : ERR_PTR(-EPROBE_DEFER);
>>>   }
>>>   
>> + Heikki
> Thanks Chunfeng!
>
> cheers,
>
diff mbox series

Patch

diff --git a/drivers/usb/roles/class.c b/drivers/usb/roles/class.c
index dfaed7eee94f..289950e5fcfb 100644
--- a/drivers/usb/roles/class.c
+++ b/drivers/usb/roles/class.c
@@ -106,10 +106,14 @@  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")) {
+		if (parent)
+			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);
 }