Re: [PATCHv2 6/6] mac80211: mesh power save basics

[Marco Porsch <marco@cozybit.com>]

On 01/18/2013 12:52 PM, Johannes Berg wrote:
> On Thu, 2013-01-17 at 17:34 +0100, Marco Porsch wrote:
>> On 01/16/2013 11:14 PM, Johannes Berg wrote:
>>> On Mon, 2013-01-07 at 16:04 +0100, Marco Porsch wrote:
>>>
>>>> mode determines when non-peer mesh STA may send Probe Requests and Mesh Peering
>>>
>>> Please break lines to less than 72 characters, I personally prefer
>>> around 60 or so but I'll apply 72 too.
>>>
>>>> +static inline bool ieee80211_has_qos_mesh_ps(__le16 qc)
>>>> +{
>>>> +	return (qc & cpu_to_le16(IEEE80211_QOS_CTL_MESH_PS_LEVEL)) != 0;
>>>
>>> bool means you don't need the !=0 and parentheses.
>>
>> Actually, then it pops the following warning:
>>     CHECK   net/mac80211/mesh_ps.c
>> include/linux/ieee80211.h:584:19: warning: incorrect type in return
>> expression (different base types)
>> include/linux/ieee80211.h:584:19:    expected bool
>> include/linux/ieee80211.h:584:19:    got restricted __le16
>>
>>
>> Other functions in that file have following style:
>> static inline int ieee80211_has_pm(__le16 fc)
>> {
>> 	return (fc & cpu_to_le16(IEEE80211_FCTL_PM)) != 0;
>> }
>>
>> What do you recommend?
>
> I'm clearly not quick enough for you, sorry. I guess we should keep it
> as is then, but I see you removed it entirely later? Anyway ... I just
> wasn't aware sparse would then complain, that's a bit stupid but hey.

Ah sorry. I just realized that
1) all the inlines in that file are about Frame Control and none copes 
with QoS Control
2) I am the only one using QoS Mesh PM and in just a single occasion

So... SNIP =)

--Marco