All of lore.kernel.org
 help / color / mirror / Atom feed
From: Luis R. Rodriguez <lrodriguez@atheros.com>
To: ath9k-devel@lists.ath9k.org
Subject: [ath9k-devel] help
Date: Fri, 7 May 2010 11:21:59 -0700	[thread overview]
Message-ID: <20100507182159.GA2375@tux> (raw)
In-Reply-To: <j2hd78619451005060551j885786aexdae0c9d065bb778c@mail.gmail.com>

On Thu, May 06, 2010 at 05:51:52AM -0700, Devan Rehunathan wrote:
> Hello,
> 
> I am trying to setup a linux wireless access point using the 5GHz spectrum.
> 
> I have the following card:
> 
> http://www.compexshop.com/product_info.php/cPath/57/products_id/290
> 
> installed on a debian machine running 2.6.32.8. I have installed the latest
> versions of the ath9k driver as well as the latest developmental release of
> hostapd (0.7.2). When I run "iw list", I see that all the 5GHz channels are
> either already in use

What do you mean by already in use?

> or disabled.

If it is disabled then you cannot use them.

> While I can manually set the channel to
> a 5GHz, I am unable to set the card to master mode.

Master mode thing is an old wext thing. New 802.11 drivers
use nl80211 and mac80211 drivers can only use nl80211 for
AP mode of operation.

> Using hostapd
> returns errors that the driver is unable to select a 5Ghz channel.

Providing logs might help, but likely this just means you
cannot use AP mode of operation on that channel.

> I've also
> tried changing the CRDA settings to no avail.

You are not supposed to, the settings that already ship should be
valid. If you still proceed on by changing your regulatory settings
you are on your own and you will not get support. For further details
see:

http://wireless.kernel.org/en/developers/Regulatory/statement

> I see that AP mode is
> supported from http://wireless.kernel.org/en/users/Drivers/ath9k

Indeed.

> I know this question has been asked before /msg01368.html. 

Whatever that is.

> But I would
> like to ask if what I'm trying to do is possible with the ath9k current
> driver?

Its open code, you can try to make it sing if you like, but overriding
regulatory settings is simply not supported. However, if you want to help
regulatory compliance that is supported. So for example if you have a JP
latpop that allows channel 13 and come to the US you can set the regulatory
domain to US to disable channel 13, but if instead you have a US card and
you go to JP channel 13 will not be enabled if it wasn't before, even if
you set your regulatory domain to JP. The changing of regulatory domains
on Linux then only helps regulatory compliance. The different regulatory
agencies are pretty strict about these considerations, for more details
on this specific topic you may want to read:

http://wireless.kernel.org/en/developers/Regulatory/CRDA#Changing_regulatory_domains

Granted, I don't particularly think this is intelligent, I think instead
in the future legislation should change to shift liabilty down to the user
if a user does want to indeed modify regulatory settings. But today that is
not the state of affairs so we simply cannot support it.

If you're wondering why we have taken this stance then consider how you can
try to get proper vendor support from 802.11 manufacturers upstream on the
Linux kernel. On our second Linux wireless summit back a few years ago we
tried to address all the concerns vendors had for support. Regulatory was
on the top of the list. So, instead of working on proprietary crap solutions
we figured we could come up with better open technologies for regulatory
compliance, and at the same time get proper vendor support upstream.

http://wireless.kernel.org/en/vendors/VendorSupport

We worked on that for a few years, got it merged and now today we have
proper vendor support from at least Atheros. Whether others will follow
remains to be seen ;)

> Or do I need a patch msg00532.html? Or am I using the wrong driver?

Breaking regulatory rules is simply not supported. If you do have a
channel that is enabled though and it does not work that is different
and for that you should post logs to the respective project list.

Hope that helps.

  Luis

  reply	other threads:[~2010-05-07 18:21 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-05-06 12:51 [ath9k-devel] help Devan Rehunathan
2010-05-07 18:21 ` Luis R. Rodriguez [this message]
2010-05-07 18:28   ` Peter Stuge
2010-05-07 18:39     ` Luis R. Rodriguez
2010-05-07 19:58   ` Devan Rehunathan
2010-05-07 21:34     ` Luis R. Rodriguez
2010-05-07 21:41       ` Devan Rehunathan
2010-05-07 21:47         ` Luis R. Rodriguez
2010-05-07 21:54           ` Luis R. Rodriguez
2010-05-07 22:01           ` Devan Rehunathan
2010-05-07 22:07             ` Luis R. Rodriguez
2010-05-07 22:16               ` Devan Rehunathan
2010-05-07 22:21                 ` Luis R. Rodriguez
  -- strict thread matches above, loose matches on Subject: below --
2009-10-10  7:36 xushaofeng1987

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100507182159.GA2375@tux \
    --to=lrodriguez@atheros.com \
    --cc=ath9k-devel@lists.ath9k.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.