Bug#414998: This may be a regression... (works in 1:0.9.2+r1842.20061207 )

roger roger at computer-surgery.co.uk
Sun Aug 19 19:19:16 UTC 2007


I have now got this problem after upgrading to 1:0.9.3-3 which
seems to be needed by kernel version 2.6.22.

I was running 2.6.18.dfsg.1-12etch2 with madwifi
1:0.9.2+r1842.20061207-2 .

With these verison the madwifi driver would install and run the wireless
fine. The following madwifi related messages during boot:-
julia kernel: input: PC Speaker as /class/input/input1
julia kernel: ath_hal: module license 'Proprietary' taints kernel.
julia kernel: ath_hal: 0.9.18.0 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413, RF5413)Aug 17 19:58:17 julia kernel: Yenta: CardBus bridge found at 0000:04:06.0 [1025:0112]
julia kernel: Yenta: Using CSCINT to route CSC interrupts to PCI
julia kernel: Yenta: Routing CardBus interrupts to PCI
julia kernel: Yenta TI: socket 0000:04:06.0, mfunc 0x013a1b22, devctl 0x64
julia kernel: wlan: 0.8.4.2 (0.9.3)
julia kernel: ath_rate_sample: 1.2 (0.9.3)
julia kernel: ath_pci: 0.9.4.5 (0.9.3)
julia kernel: Yenta: ISA IRQ mask 0x04f8, PCI irq 11
julia kernel: Socket status: 30000006
julia kernel: Yenta: Raising subordinate bus# of parent bus (#04) from #05 to #08
julia kernel: pcmcia: parent PCI bridge I/O window: 0x5000 - 0x5fff
julia kernel: pcmcia: parent PCI bridge Memory window: 0xc3000000 - 0xc30fffff
julia kernel: pcmcia: parent PCI bridge Memory window: 0x88000000 - 0x89ffffff
julia kernel: ACPI: PCI Interrupt Link [LNK2] enabled at IRQ 19
julia kernel: GSI 20 sharing vector 0xE9 and IRQ 20
julia kernel: wifi0: 11b rates: 1Mbps 2Mbps 5.5Mbps 11Mbps
julia kernel: wifi0: 11g rates: 1Mbps 2Mbps 5.5Mbps 11Mbps 6Mbps 9Mbps 12Mbps 18Mbps 24Mbps 36Mbps 48Mbps 54Mbps
julia kernel: wifi0: H/W encryption support: WEP AES AES_CCM TKIP
julia kernel: wifi0: mac 7.8 phy 4.5 radio 5.6
julia kernel: wifi0: Use hw queue 1 for WME_AC_BE traffic
julia kernel: wifi0: Use hw queue 0 for WME_AC_BK traffic
julia kernel: wifi0: Use hw queue 2 for WME_AC_VI traffic
julia kernel: wifi0: Use hw queue 3 for WME_AC_VO traffic
julia kernel: wifi0: Use hw queue 8 for CAB traffic
julia kernel: wifi0: Use hw queue 9 for beacons

After upgrading the kernel, and the madwifi source since I couldn't
make the earlier release buld with kernel 2.6.22 , I get the following
messages when installing the module.

ath_hal: module license 'Proprietary' taints kernel.
ath_hal: 0.9.18.0 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413, RF5413) wlan: 0.8.4.2 (0.9.3)
ath_pci: 0.9.4.5 (0.9.3)
ACPI: PCI Interrupt Link [LNK2] enabled at IRQ 18
ACPI: PCI Interrupt 0000:04:05.0[A] -> Link [LNK2] -> GSI 18 (level, high) -> IRQ 18
wifi%d: unable to attach hardware: 'Hardware revision not supported' (HAL status 13)
ACPI: PCI interrupt for device 0000:04:05.0 disabled

AFAICT both these madwifi source revisions use the same hal.

Interestingly enough checking my log files, I note I did once get this
error message with the previous source release but a reboot fixed it.

Hmm, Investigating further shows that a power cycle fixed it this time
too.

Can I suggest there may be a hardware initialisation in the driver
somewhere?


-- System Information:
Debian Release: 4.0
  APT prefers stable
  APT policy: (500, 'stable'), (50, 'testing'), (20, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.22-1-amd64
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages madwifi-source depends on:
ii  bzip2                         1.0.3-6    high-quality block-sorting file co
ii  debhelper                     5.0.42     helper programs for debian/rules
ii  module-assistant              0.10.8     tool to make module package creati

madwifi-source recommends no packages.

-- no debconf information





More information about the Pkg-madwifi-maintainers mailing list