Re: [RFC V4] cfg80211: introduce critical protocol indication from user-space

[Ben Greear <greearb@candelatech.com>]

On 04/05/2013 12:19 PM, Arend van Spriel wrote:

>>> +    trace_rdev_crit_proto_stop(&rdev->wiphy, wdev);
>>> +    if (rdev->crit_proto_started) {
>>> +        ret = rdev->ops->crit_proto_stop(&rdev->wiphy, wdev);
>>> +        rdev->crit_proto_started = ret != 0;
>>
>> Maybe:  rdev->crit_proto_started = !ret;
>
> Maybe not. If ret == 0, crit_proto_started should be false. If ret != 0, ie. crit_proto_stop() failed, it should remain true. It is a bit of reverse logic. I
> could change it to !!ret or make it more clear using !ret ? false : true. Another idea is changing the return type of crit_proto_stop() to void and always set
> crit_proto_started to false.

Well, I had a hard time figuring out what that code was supposed to do (obviously).

Maybe just:
	if (ret == 0)
		rdev->crit_proto_started = false;

Or just require that the driver not fail this operation ever and always set crit_proto_started to false.

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com