new 'beta' version of 'ath' driver available on leaf.

Andrew Atrens atrens at nortelnetworks.com
Fri Apr 8 09:47:52 PDT 2005


Hi Folks,

I've put up a new version of a port of Sam Leffler's ath driver
from FreeBSD-current. Onto leaf -

The files are at -

http://leaf.dragonflybsd.org/~atrens/ath-dragonfly-1.4.2.tar.gz

I started by porting the FreeBSD 5.3 version of 'ath' and merging
in support for the new 0.9.14.9 HAL code from FreeBSD -current,
and over the last  several weeks have been gradually evolving it -
pulling in more and more pieces from the FreeBSD-current version.
I'd say at this point in time it's about 85% -current and 15% 5.3.
Probably the largest 'piece' of the -current code that's still
missing is WPA support.  Pulling that in would mean updating the
DragonFly 802.11 stack, which though could be done is too big a
piece for me to bite off right now.

Having said all that, if you don't care about WPA in the short term,
what I've got is quite solid now I think, I'd say at least 'alpha',
perhaps 'beta' quality.

Here's what dmesg says -

ath_hal: 0.9.14.9 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413)
ath0: <Atheros 5212> mem 0xfafe0000-0xfafeffff irq 11 at device 2.0 on pci2
ath0: mac 5.9 phy 4.3 5ghz radio 3.6
ath0: MAC address: 00:0b:6b:34:b7:7d
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

Turbo mode may (or may not) work - I've been unable to test it. Maybe will
have some time this weekend.

As far as functionality is concerned all operating modes - bss, adhoc, monitor,
and hostap have been tested and work well.

In adhoc and hostap you may experience the 'stuck beacon problem' as
described on mad-wifi (you might want to google for more info), but in
my setup at least stuck beacons aren't a big problem, and they seem to
be very h/w specific (I only see them on one card). For monitor to work
you'll need to patch ifconfig - this is a very small patch - just the
addition of a new 'mediaopt' value of 'monitor'. I don't have a patch handy
but can put it up if someone is interested.

All three rate control algorithms have been ported. 'onoe' has been lightly
tested, 'amrr' heavily, and the new 1.2 version of 'sample', though
I've just started to use it, looks very, very promising.

Once I've got turbo mode tested, and I've done another few cleanup
passes, I'd like to humbly submit this code for review. Right now though
I'm looking for a few hardy testers. :) :)

Have fun and let me know how it works for you!

Cheers,

Andrew.







More information about the Submit mailing list