madwifi does not work with IBM Z60t

• If you have a key to enable/disable your wlan card, make sure that the card is actually turned on when the driver gets loaded.
• Try to load ath_pci with the "rfkill=0" parameter
• Try to temporarily turn off ACPI.
• Please check the mailing list archives regarding possible solutions of a "HAL status 3"-type problem.

Please tell us whether any of these tips helped.

* I made sure that that the switch was on. I tried in both positions * rfkill=0 produce an error: [4350583.031000] ath_pci: Unknown parameter `rfkill'

Now if I had the source of the hal, I could investigate deeper....

The rfkill module parameter was introduced in r1362. You probably have to update your snapshot.

What about turning ACPI off, did that help? Anything valuable found in the mailing list archives?

Regarding source of the HAL: we all would like to have access to the HAL source. Unfortunately this isn't possible, unless you work for a company that is able to fulfill the necessary requirements.

madwifi-ng does not do anything better with "rfkill=0"

no I haven't tried with ACPI off because I have to interrupt all my work to reboot

Re: the hal, since you don't give me enough info, I can't reply. But if it involve not releasing the code as open source. then I'd better spent my time on the free drivers....

disabling acpi does not help at all.

again, having the source of the hal would help.

We had dozens of discussions about the HAL being binary-only. Refer to this page? for the excerpts. It's been made clear by Atheros that the HAL source won't be made available, and I'm not in a position to change that decision. If that means for you to turn away from MadWifi: the page mentioned above has some pointers on other drivers which don't make use of a closed-source HAL.

i get the same problems with the newer svn builds (after christmas). but my platform is a routerboard 500 (mipsel). in earlier builds i found out that sometimes the used compiler version can cause this problem. but now this wont help anymore. modprobe -r ath_pci modprobe ath_pci (so just reloading) helps normally. but if i now try to do anything with madwifi like iwconfig ath0 ap any or something like that i get dozens of kernel errors like wifi0: hardware error; reseting wifi0: ath_reset: unable to reset hardware: 'Hardware didn't respond as expected'

and stupid things like that. or just in other words. madwifi-ng does not work anymore since my last svn update

Make sure that modprobe -r ath_pci actually removed all MadWifi modules - this isn't always the case. If some modules remain loaded it will lead to the problem you've described.

FYI, I am no longer getting any errors in Dapper (with 2.6.15-11-686) off of ng trunk, however iwconfig reports no wireless interfaces (wireless-tools 27+28pre10-1ubuntu1). I'm confused, but it's Progress(tm) of some sort I suppose.

For kicks, here is the log after loading the module:

[4296031.458000] ath_hal: 0.9.16.13 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413, RF5413, DFS)
[4296031.463000] wlan: 0.8.4.2 (Atheros/multi-bss)
[4296031.469000] ath_rate_sample: 1.2
[4296031.475000] ath_pci: `' invalid for parameter `ath_debug'
[4296039.110000] ath_rate_sample: unloaded
[4296039.112000] wlan: driver unloaded
[4296039.113000] ath_hal: driver unloaded
[4296039.132000] ath_hal: 0.9.16.13 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413, RF5413, DFS)
[4296039.137000] wlan: 0.8.4.2 (Atheros/multi-bss)
[4296039.141000] ath_rate_sample: 1.2
[4296039.149000] ath_pci: 0.9.4.5 (Atheros/multi-bss)
[4296039.151000] ACPI: PCI Interrupt 0000:13:00.0[A] -> GSI 19 (level, low) -> IRQ 74
[4296039.152000] PCI: Setting latency timer of device 0000:13:00.0 to 64
[4296039.153000] ath_attach: devid 0x1014
[4296040.287000] ath_xr_rate_setup: rate table too small (13 > 15)
[4296040.289000] wifi0: 11a rates: 6Mbps 9Mbps 12Mbps 18Mbps 24Mbps 36Mbps 48Mbps 54Mbps
[4296040.290000] wifi0: 11b rates: 1Mbps 2Mbps 5.5Mbps 11Mbps
[4296040.291000] wifi0: 11g rates: 1Mbps 2Mbps 5.5Mbps 11Mbps 6Mbps 9Mbps 12Mbps 18Mbps 24Mbps 36Mbps 48Mbps 54Mbps
[4296040.292000] wifi0: H/W encryption support: WEP AES AES_CCM TKIP
[4296040.293000] wifi0: mac 9.15 phy 6.1 radio 10.2
[4296040.293000] wifi0: Use hw queue 1 for WME_AC_BE traffic
[4296040.294000] wifi0: Use hw queue 0 for WME_AC_BK traffic
[4296040.295000] wifi0: Use hw queue 2 for WME_AC_VI traffic
[4296040.295000] wifi0: Use hw queue 3 for WME_AC_VO traffic
[4296040.296000] wifi0: Use hw queue 8 for CAB traffic
[4296040.296000] wifi0: Use hw queue 9 for beacons
[4296040.297000] wifi0: Atheros 5212: mem=0xa7f00000, irq=74

And this shows up a few times during 'ifup wifi0'

[4296895.273000] ath_hardstart: discard, no node in cb

Allright, I'm typing this via ath0 wireless on my z60t.

Simply following the instructions on the UserDocs/FirstTimeHowTo page got me going in Ubuntu Dapper. I'm not sure where the wifi0 interface came from, but using ath0 works great.

hub, can you give it another try?

Using Revision: 1391 URL: http://svn.madwifi.org/trunk Repository UUID: 0192ed92-7a03-0410-a25b-9323aeb14dbd

Still no success. I tried with the switch in both position, making sure the modules where unloaded.

$sudo iwconfig ath0 ath0 No such device

$sudo iwlist scan lo Interface doesn't support scanning.

sit0 Interface doesn't support scanning.

eth0 Interface doesn't support scanning.

Did you create the interface with wlanconfig?

wlanconfig ath0 create wlandev wifi0 wlanmode sta

I had no luck until I did that.

Actually I reboot after install the modules and it works. Looks like the previous version mess up the hardware.

Closing this ticket, since the reported issues seem to have been resolved.

...and, how is Linux better than Windows...?

@ThinkBaz?: this ticket is not the right place to start an OS flame war.