[ath9k-devel] Reg domain won't change and forced to CO (I'm in US)

[Luis R. Rodriguez <lrodriguez@atheros.com>]

On Mon, Feb 02, 2009 at 02:23:23PM -0800, Chris Kennedy wrote:
> 
> On Feb 2, 2009, at 4:13 PM, Chris Kennedy wrote:
> 
> >
> > On Feb 2, 2009, at 3:35 PM, Chris Kennedy wrote:
> >
> >>
> >> I'm using the Linksys WMP300N with broadcom chips in them as clients,
> >> and my AP is a DWA-552
> >> which has the AR5116 in it.  I switch on/off with an Apple Extreme
> >>
> >
> > I mean the AR5416 actually :).
> 
> And here's a good showing of the results I get pretty consistent
> right now in N mode for the ath9k AP,
> first is the Ath9k to the  Linksys300N client, second is the Apple
> Extreme Ap to a Linksys300N client (to
> compare with), and the last is the ath9k to the G client (has an
> intel chipset, uses the ipw2200 driver)...
> 
> 
> Ath9k to the  Linksys300N:
> oz iw-htrate # iperf -c 192.168.254.103 -r

You may want to try

iperf -c <ip> -i 1 -t 100000

Or something similar.

> ------------------------------------------------------------
> Client connecting to 192.168.254.103, TCP port 5001
> TCP window size: 16.0 KByte (default)
> ------------------------------------------------------------
> [  3] local 192.168.254.3 port 33374 connected with 192.168.254.103
> port 5001
> [  3]  0.0-10.0 sec  80.1 MBytes  67.2 Mbits/sec
> ------------------------------------------------------------
> Server listening on TCP port 5001
> TCP window size: 85.3 KByte (default)
> ------------------------------------------------------------
> [  4] local 192.168.254.3 port 5001 connected with 192.168.254.103
> port 1645
> [  4]  0.0-10.0 sec  6.16 MBytes  5.16 Mbits/sec
> 
> 
> Apple Extreme Ap to a Linksys300N :
> oz iw-htrate # iperf -c 192.168.254.105 -r

Apple Extreme AP, with iperf and a shell?

> ------------------------------------------------------------
> Client connecting to 192.168.254.105, TCP port 5001
> TCP window size: 16.0 KByte (default)
> ------------------------------------------------------------
> [  3] local 192.168.254.3 port 52080 connected with 192.168.254.105
> port 5001
> [  3]  0.0-10.1 sec  72.7 MBytes  60.5 Mbits/sec
> ------------------------------------------------------------
> Server listening on TCP port 5001
> TCP window size: 85.3 KByte (default)
> ------------------------------------------------------------
> [  4] local 192.168.254.3 port 5001 connected with 192.168.254.105
> port 1712
> [  4]  0.0-10.0 sec  36.5 MBytes  30.6 Mbits/sec
> 
> 
> 
> ath9k to the G client (has an intel chipset, uses the ipw2200 driver).:
> oz iw-htrate # iperf -c 192.168.254.150 -r
> ------------------------------------------------------------
> Client connecting to 192.168.254.150, TCP port 5001
> TCP window size: 16.0 KByte (default)
> ------------------------------------------------------------
> [  3] local 192.168.254.3 port 37313 connected with 192.168.254.150
> port 5001
> [  3]  0.0-10.1 sec  4.32 MBytes  3.59 Mbits/sec
> ------------------------------------------------------------
> Server listening on TCP port 5001
> TCP window size: 85.3 KByte (default)
> ------------------------------------------------------------
> [  4] local 192.168.254.3 port 5001 connected with 192.168.254.150
> port 39018
> [  4]  0.0-14.5 sec  2.07 MBytes  1.20 Mbits/sec
> 
> 
> 
> 
> So those bi-directional tests show basically what I see.  Of course
> there's times like
> the other day when it all goes to heck and I have 11Mbit/s speeds in
> N mode for
> the Ath9k AP.  I'm using the newest git version of hostapd right now
> (newest wireless-testing), had been going
> back and forth between 0.6.7, which I don't really think much changes
> between those
> two from what I can tell.  It was definitely acting up though the
> other day and it didn't get
> faster for about 12 hours and that was only in G mode, today is the
> first time I've tried N
> mode again and now it seems somewhat decent and like it had been at
> it's best in the past.
> It's almost like switching between the driver/kernel that had not
> worked, somehow kicked the
> card into a bad mood for a day or so, maybe that's crazy though and
> some other factor like
> dBm or something.

What does your iwlist wlan0 scan look like? Busy?

Thanks for your detailed reports and testing!

  Luis