Re: [PATCH v5 1/3] [media] media-device: check before unregister if mdev was registered

[Javier Martinez Canillas <javier@osg.samsung.com>]

Hello Mauro,

On 12/15/2015 07:41 AM, Mauro Carvalho Chehab wrote:
> Em Fri, 11 Dec 2015 16:36:27 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> On 12/11/2015 03:57 PM, Javier Martinez Canillas wrote:
>>> Most media functions that unregister, check if the corresponding register
>>> function succeed before. So these functions can safely be called even if a
>>> registration was never made or the component as already been unregistered.
>>>
>>> Add the same check to media_device_unregister() function for consistency.
>>>
>>> This will also allow to split the media_device_register() function in an
>>> initialization and registration functions without the need to change the
>>> generic cleanup functions and error code paths for all the media drivers.
>>>
>>> Suggested-by: Sakari Ailus <sakari.ailus@linux.intel.com>
>>> Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
>>> Acked-by: Sakari Ailus <sakari.ailus@linux.intel.com>
>>>
>>> ---
>>>
>>> Changes in v5: None
>>> Changes in v4: None
>>> Changes in v3: None
>>> Changes in v2:
>>> - Reword the documentation for media_device_unregister(). Suggested by Sakari.
>>> - Added Sakari's Acked-by tag for patch #1.
>>>
>>>  drivers/media/media-device.c | 6 ++++++
>>>  1 file changed, 6 insertions(+)
>>>
>>> diff --git a/drivers/media/media-device.c b/drivers/media/media-device.c
>>> index c12481c753a0..11c1c7383361 100644
>>> --- a/drivers/media/media-device.c
>>> +++ b/drivers/media/media-device.c
>>> @@ -577,6 +577,8 @@ EXPORT_SYMBOL_GPL(__media_device_register);
>>>   * media_device_unregister - unregister a media device
>>>   * @mdev:	The media device
>>>   *
>>> + * It is safe to call this function on an unregistered
>>> + * (but initialised) media device.
>>>   */
>>>  void media_device_unregister(struct media_device *mdev)
>>>  {
>>> @@ -584,6 +586,10 @@ void media_device_unregister(struct media_device *mdev)
>>>  	struct media_entity *next;
>>>  	struct media_interface *intf, *tmp_intf;
>>>  
>>> +	/* Check if mdev was ever registered at all */
>>> +	if (!media_devnode_is_registered(&mdev->devnode))
>>> +		return;
>>
>> This is a good check, however, this check will not prevent
>> media_device_unregister() from getting run twice by two
>> different drivers. i.e media_devnode gets unregistered
>> towards the end of at the end of media_device_unregister().
>>
>> In an example case, if bridge driver and snd-usb-aduio both
>> call media_device_unregister(), this check won't help prevent
>> media_device_unregister() being done only once. I think we
>> need a different state variable in struct media_device
>> to ensure unregister is done only once.
> 
> True. We need move the spin_lock() code to be called before
> calling media_device_is_registered().
> 

I see, the check in $SUBJECT was to avoid each driver to have a similar
check for cleanup in its error path but I didn't take into account the
fact that two drivers may share the same struct media_device, as you
explained to me is the case for ALSA and V4L2 MC integration.

> I'm sending such patches.
>

Thanks, I'll review and test them.
 
> Thanks for reporting it.
> 
> Regards,
> Mauro
> 

Best regards,
-- 
Javier Martinez Canillas
Open Source Group
Samsung Research America