Date: Wed, 10 Oct 2012 19:38:01 +0200 From: Bernhard Schmidt <bschmidt@techwires.net> To: freebsd-net@freebsd.org Cc: "Ronald F. Guilmette" <rfg@tristatelogic.com> Subject: Re: Oops! (was: Support for Intel 5100 WiFi ?) Message-ID: <201210101938.01847.bschmidt@techwires.net> In-Reply-To: <14851.1349645303@tristatelogic.com> References: <14851.1349645303@tristatelogic.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday 07 October 2012 23:28:23 Ronald F. Guilmette wrote: > > In message <201210071040.01104.bschmidt@techwires.net>, > Bernhard Schmidt <bschmidt@techwires.net> wrote: > > >iwn(4) does support 11n, 5GHz and 40MHz channels. Though, it might > >be better to switch to 9.x as it has received many many enhancements. > > OK. I suspected that might be the case. I'm glad that I asked! > > Is the iwn(4) driver in 9.0 pretty good. Or are there some things in > 9.1-RC1 (relating to iwn) that I might possibly want/need? (I prefer > to stick with official releases, if possible.) There is at least one 11n related bug fix 9.1 which isn't available in 9.0. You should be fine with 9.0 if you are not running into this issue (which is pretty hard to trigger I guess). > >Using those features is rather easy, it does so by default ;) > > OK. I want to make sure that I understand what you just said. > > (Please note that I was almost entirely ignorant about the standards and > the current state of technology with respect to ALL of this wireless stuff > up until about three weeks ago, but I have tried my best to learn since > then, so bear with me here.) > > If I have understood you correctly, then you are saying that, with respect > to the decision of whether to use or, alternatively, to not use 11n > (falling back to 11g if necessary), *and* with respect to the decision to > use or not use 5GHz (falling back to 2.4Ghz if necessary), *and* with > respect to the decision of whether to use or not use 40GHz channels > (falling back to 20MHz if necessary), the client does not get to make > any of its own decisions about any of these things, and it must instead > just follow the lead of whatever AP it is trying to connect to. > > Is that correct? Have I understood you correctly? > > (Forgive me. I'me really just still feeling my way through all of this > stuff.) What I was trying to say is, if an AP supports 11n, 11g, 20 and 40MHz channels, by default the setup which does give the best performance is used; which would be 11n+40MHz in this case. If an AP supports just 20MHz channels, iwn(4) is fine with that; same for 11g vs. 11n. Choosing 2GHz over 5GHz is something entirely different. If there is an AP which provides access to the same SSID via either 2GHz or 5GHz, iwn(4) might choose randomly (it's not entirely random, usually it's the first network detected, but close anyway). If you want to have influence on that process, you are able to by setting the appropriate options via ifconfig(8). Either something like *-ht* to disable 11n, or *channel* to restrict the device to a certain channel. -- Bernhard
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201210101938.01847.bschmidt>