Subject: Re: kern/28244: no turbo mode recognised for NetGear WG311T and WG511T
To: None <>
From: List Mail User <>
List: netbsd-bugs
Date: 11/12/2004 04:12:55
	Unfortunately, this isn't a bug. NetBSD and all OS drivers which rely
on the HAL (.../src/contrib/sys/arch/i386/dev/athhal-elf-o.uue) don't support
"Turbo" or "Super" features on 2.4GHz until whenever they get added to the
external module (the #define has been in the header for about three months).
The "Turbo" features do work for 'a' channels (at least on NetBSD, they may
be broken on currents HALs and linux -- see recent thread at the madWiFi
mailing lists).  Also, unmentioned by anyone anywhere I've seen, 108MBs
Atheros based routers and access points do NOW work with the HAL iff, they
are set to "Dynamic Turbo" for 802.11g (this doesn't work at all for 'a',
because of channel shifts) and the firmware has "AR" enabled (not "XR", but
seems to appear at the same 'release' or revision for each vendor);  "XR"
works without "Turbo", though it is ignored (like "Super" features) by the

	Basically, if your cards are the 511T and WG311T, you're SOL for now
on any *nix (ignoring "ndisdriver" & "linuxant").  If your 311 is an 'WAG311',
then you have a different bug, because no 'a' channels or rates show up in
your dmesg;  Otherwise, you have a `minor' bug that your WG311 DOES show
a 5GHz radio in its dmesg (in fact, it makes it look like a 5002X chipset,
where you likely have a 5002G - but who knows, maybe it is a "crippled"
5002X).  Simply, there is a bug, but probably only a minor one in reporting
of the radio revision level (I think the linux driver has recently stopped
reporting radio "4.6" - the 'g'-only 2.4GHz part - as a 5GHz radio) and is
likely in the HAL where you'll have to wait for a fix, and for the new
version to be imported and integrated.

	An example of 108Mbs ("Turbo", not "Super") 802.11a is below:

	% uname -a
	NetBSD cobalt 2.99.10 NetBSD 2.99.10 (COBALT-$Revision: 1.4 $) #778: Wed Nov 10 11:10:10 PST 2004  root@svcs:/sys/arch/i386/compile/COBALT i386
	% dmesg | fgrep ath0
	ath0 at pci2 dev 13 function 0
	ath0: interrupting at ioapic0 pin 21 (irq 11)
	ath0: mac 5.9 phy 4.3 5ghz radio 3.6
	ath0: 11a rates: 6Mbps 9Mbps 12Mbps 18Mbps 24Mbps 36Mbps 48Mbps 54Mbps
	ath0: 11b rates: 1Mbps 2Mbps 5.5Mbps 11Mbps
	ath0: 11g rates: 1Mbps 2Mbps 5.5Mbps 11Mbps 6Mbps 9Mbps 12Mbps 18Mbps 24Mbps 36Mbps 48Mbps 54Mbps
	ath0: turbo rates: 6Mbps 9Mbps 12Mbps 18Mbps 24Mbps 36Mbps 48Mbps 54Mbps
	% ifconfig ath0
        ssid "XXXXX" nwkey *****
        powersave off
        bssid xx:xx:xx:xx:xx:xx chan 50
        address: xx:xx:xx:xx:xx:xx
        media: IEEE802.11 autoselect mode 11a turbo (OFDM54 mode 11a turbo)
        status: active
        inet XXX.XXX.XXX.XXX netmask 0xffffff00 broadcast XXX.XXX.XXX.XXX
        inet6 fexx::xxx:xxff:fexx:xxxx%ath0 prefixlen 64 scopeid 0x2

	Of course, I could be totally wrong.  Otherwise, you could file a
bug on the "dmesg" 5GHz radio report, but it would be a `minor' bug in my
view (Notice, my 5004X above doesn't report the radio quite correctly either;
radio "3.6" is the combined 2.4/5.x GHz unit on the 5004 and some 5002X+
parts, but the dmesg shows no 2.4GHz radio yet 2.4GHz channels are available).

	Paul Shupak