MIME-Version: 1.0
In-Reply-To: <CAB3XZEckstWFB-2CmWg7JeoyT0bTUgSh5bXqZxADUHiurZgnhw@mail.gmail.com>
References: <1395150804-24090-1-git-send-email-michal.kazior@tieto.com>
	<1395150804-24090-4-git-send-email-michal.kazior@tieto.com>
	<CAB3XZEcRwJ0H_rRSq0kaw8mWLJh2i2msZUmBRBFXZonZSua4kQ@mail.gmail.com>
	<CA+BoTQmNJoMa8p-DnjH9h-NdkTm=6DK2pztCFYGACe8BB5aJ=Q@mail.gmail.com>
	<CAB3XZEckstWFB-2CmWg7JeoyT0bTUgSh5bXqZxADUHiurZgnhw@mail.gmail.com>
Date: Wed, 19 Mar 2014 09:54:47 +0100
Message-ID: <CA+BoTQnNXb=sm=k=fFmYd=e1pOVTRX3MhwMgmhUeWD=2Nuy_8A@mail.gmail.com> (sfid-20140319_095455_559813_B3480B7D)
Subject: Re: [RFC 03/21] mac80211: add max channel calculation utility function
From: Michal Kazior <michal.kazior@tieto.com>
To: Eliad Peller <eliad@wizery.com>
Cc: "linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>,
	Johannes Berg <johannes@sipsolutions.net>
Content-Type: text/plain; charset=UTF-8
Sender: linux-wireless-owner@vger.kernel.org

On 19 March 2014 09:43, Eliad Peller <eliad@wizery.com> wrote:
> On Wed, Mar 19, 2014 at 10:32 AM, Michal Kazior <michal.kazior@tieto.com> wrote:
>> On 18 March 2014 17:17, Eliad Peller <eliad@wizery.com> wrote:
>>> hi,
>>>
>>> On Tue, Mar 18, 2014 at 3:53 PM, Michal Kazior <michal.kazior@tieto.com> wrote:
>>>> The utility function has no uses yet. It is aimed
>>>> at future chanctx reservation management and
>>>> channel switching.
>>>>
>>>> Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
>>>> ---
>>> [...]
>>>
>>>> +int ieee80211_max_num_channels(struct ieee80211_local *local)
>>>> +{
>>>> +       struct ieee80211_sub_if_data *sdata;
>>>> +       int num[NUM_NL80211_IFTYPES] = {};
>>>> +       struct ieee80211_chanctx *ctx;
>>>> +       int num_different_channels = 0;
>>>> +       u8 radar_detect = 0;
>>>> +       u32 max_num_different_channels = 1;
>>>> +       int err;
>>>> +
>>>> +       lockdep_assert_held(&local->chanctx_mtx);
>>>> +
>>>> +       list_for_each_entry(ctx, &local->chanctx_list, list) {
>>>> +               num_different_channels++;
>>>> +
>>> this doesn't seem correct - i think we can have multiple channel
>>> contexts with the same channel (e.g. due to exclusive mode)?
>>
>> Hmm, does it even make sense to consider chanctx mode here?
>>
> i don't think you should consider the chancx mode, but keep track of
> the used channels.
> if you have 2 different channel contexts with the same channel (e.g.
> due to exclusive mode), you have to make sure you don't count the same
> channel twice. you shouldn't care about the actual reason for it.

I disagree. There's a reason why we have exclusive chanctx mode.

If you insist on counting compatible chandefs you can lead to
overcommit if your exclusive chanctx bound interface decides to switch
to another channel.


Michał