Please note: This project is no longer active. The website is kept online for historic purposes only.
If you´re looking for a Linux driver for your Atheros WLAN device, you should continue here .

Ticket #1801 (new enhancement)

Opened 10 years ago

Last modified 8 years ago

Madwifi AP with a Atheros AR5416 (AR5008) Chip

Reported by: foodoc Assigned to:
Priority: minor Milestone:
Component: madwifi: driver Version: trunk
Keywords: AR5416 Beacons Cc:
Patch is attached: 0 Pending:

Description

Good news everyone... finally, I figured out how to get the beacon xmits on these cards (kind of) working!

See the attached proposal patch. Since I'm a noob with madwifi's source. I want to get some feedback, before any fancy "signed-off-by"/"acked-by" lines (if needed)...

Attachments

madwifi-ar5416-ap-beacons.diff (2.6 kB) - added by foodoc on 02/23/08 01:13:34.
initialize Beacons correctly on AR5008
madwifi-ar5416-ap-beacons-v2.diff (3.4 kB) - added by foodoc on 02/27/08 12:34:59.
a little update!
madwifi-ar5416-ap-beacons-v2-update.diff (3.7 kB) - added by foodoc on 04/19/08 15:11:46.
for r3546
madwifi-ar5416-ap-beacons-v2-update.2.diff (3.7 kB) - added by foodoc on 04/19/08 15:16:01.
for r3546 with default case!

Change History

02/23/08 01:13:34 changed by foodoc

  • attachment madwifi-ar5416-ap-beacons.diff added.

initialize Beacons correctly on AR5008

02/27/08 11:57:33 changed by DrunkenBear

Nice work ;D

02/27/08 12:34:59 changed by foodoc

  • attachment madwifi-ar5416-ap-beacons-v2.diff added.

a little update!

02/27/08 12:37:31 changed by foodoc

here's a little update...

03/07/08 00:35:41 changed by BrainSlayer

doesnt work here. i still see no beacons nor any interrupt

03/07/08 00:56:06 changed by foodoc

well, if you don't even see interrupts then something is very wrong with your setup. you should take a look into /proc/interrupts or dmesg for any spurious interrupts...

03/10/08 00:08:24 changed by DrunkenBear

v1 work great, but not perfect v2 is very unstable

why you change: bt.bt_nextdba = (nexttbtt - 2) >> 3; bt.bt_nextswba = (nexttbtt - 10) >> 3; to: bt.bt_nextdba = (nexttbtt - ath_hal_dma_beacon_response_time) << 3; bt.bt_nextswba = (nexttbtt - ath_hal_sw_beacon_response_time) << 3;

03/10/08 00:27:02 changed by foodoc

well, there's a comment in hal/ah.h (@ struct HAL_BEACON_TIMERS ) next DBA in 1/8th TU... so V1 is obviously "wrong" ;)

anyway, if v2 is unstable you "can" change:

bt.bt_intval = intval;

to (3 should be ok to reduce the IRQ-Load even more!)

bt.bt_intval = intval * 2;

This change is ok as long as you have only one AP (no VAPs!). Because the intval is actually in 1/4th milliseconds !

04/19/08 15:11:46 changed by foodoc

  • attachment madwifi-ar5416-ap-beacons-v2-update.diff added.

for r3546

04/19/08 15:16:01 changed by foodoc

  • attachment madwifi-ar5416-ap-beacons-v2-update.2.diff added.

for r3546 with default case!

04/26/08 00:57:09 changed by foodoc

HINT: disable all atheros specific features (especially XR, Turbo & Framebursting!) This helps a lot, if you don't want NETDEV Watchdog to bark around... 

iwpriv ath0 abolt 0

03/16/10 21:26:17 changed by supermorph

i wanted to ask you guys.

any further progress? i have a wrt300n v2 and am waiting for dd-wrt to become available to it.

i know it has that chip (a site shown me) so if u have any progress could you show the dd-wrt team? (specifically brainslayer) as atheros wont play ball with driver sources.

thanks again, supermorph