[PATCH 08/12] staging: dgap: tty.c: fixes ioctl param list

Mark Hounschell markh at compro.net
Tue Aug 27 17:25:30 UTC 2013


On 08/27/2013 01:20 PM, Lidza Louina wrote:
> On Fri, Aug 23, 2013 at 4:23 PM, Dan Carpenter <dan.carpenter at oracle.com> wrote:
>> On Wed, Aug 21, 2013 at 09:48:38PM -0400, Lidza Louina wrote:
>>> -static int dgap_tty_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd,
>>> +static int dgap_tty_ioctl(struct tty_struct *tty, unsigned int cmd,
>>>                unsigned long arg)
>>
>> Btw, now that this file compiles I see it in my static checker runs.
>>
>> dgap_tty_ioctl() takes a lock but always returns before unlocking.
>>
>>>   {
>>>        struct board_t *bd;
>>> @@ -3423,41 +3423,6 @@ static int dgap_tty_ioctl(struct tty_struct *tty, struct file *file, unsigned in
>>>
>>>                return(-ENOIOCTLCMD);
>>
>> In other words, these return statements should be breaks so we hit
>> the call to:
>>
>>          DGNC_UNLOCK(ch->ch_lock, lock_flags);
>
> Hmm, I looked thru the code a few times, and it seems like
> it unlocks before every return. Can you paste an example
> where it does this?
>

FYI, there is no DGNC_UNLOCK in the dgap driver? Maybe he was looking at 
the dgnc while commenting on the dgap??? Just a guess.

Mark


More information about the devel mailing list