[PATCH RFC v3] media: em28xx: Fix race condition between open and init function

Shuah Khan skhan at linuxfoundation.org
Fri Apr 16 23:03:05 UTC 2021


On 4/16/21 1:33 PM, Igor Torrente wrote:
> 
> 
> On 4/15/21 2:25 PM, Shuah Khan wrote:
>> On 4/15/21 8:07 AM, Igor Matheus Andrade Torrente wrote:
>>> Fixes a race condition - for lack of a more precise term - between
>>> em28xx_v4l2_open and em28xx_v4l2_init, by detaching the v4l2_dev,
>>> media_pad and vdev structs from the em28xx_v4l2, and managing the
>>> lifetime of those objects more dynamicaly.
>>>
>>> The race happens when a thread[1] - containing the em28xx_v4l2_init()
>>> code - calls the v4l2_mc_create_media_graph(), and it return a error,
>>> if a thread[2] - running v4l2_open() - pass the verification point
>>> and reaches the em28xx_v4l2_open() before the thread[1] finishes
>>> the deregistration of v4l2 subsystem, the thread[1] will free all
>>> resources before the em28xx_v4l2_open() can process their things,
>>> because the em28xx_v4l2_init() has the dev->lock. And all this lead
>>> the thread[2] to cause a user-after-free.
>>>
>>> Reported-by: kernel test robot <lkp at intel.com>
>>> Reported-and-tested-by: 
>>> syzbot+b2391895514ed9ef4a8e at syzkaller.appspotmail.com
>>> Signed-off-by: Igor Matheus Andrade Torrente <igormtorrente at gmail.com>
>>> ---
>>>
>>> V2: Add v4l2_i2c_new_subdev null check
>>>      Deal with v4l2 subdevs dependencies
>>>
>>> V3: Fix link error when compiled as a module
>>>
>>> ---
>>>   drivers/media/usb/em28xx/em28xx-camera.c |   4 +-
>>>   drivers/media/usb/em28xx/em28xx-video.c  | 300 +++++++++++++++--------
>>>   drivers/media/usb/em28xx/em28xx.h        |   6 +-
>>>   3 files changed, 209 insertions(+), 101 deletions(-)
>>>
>>
>> The changes looks good to me. Have you tried building as a modules and
>> running modprobes and rmmods? You can do that without a device.
>>
> 
> I tried and everything worked fine.
> 

Thank you.

Reviewed-by: Shuah Khan <skhan at linuxfoundation.org>

thanks,
-- Shuah



More information about the Linux-kernel-mentees mailing list