@Artem: could you make sure your system is fully up-to-date? In particular udev-178? Also, could you up the udev logging with "udevadm control --log-priority=debug" before doing ifconfig wlan0 up.

Lastly, if you boot without enabling any networking daeons (so your interfaces are all down), do you still see any errors in dmesg, or do they only occur when you do "ifconfig up"?

FWIW, I have the same chip (I think) and have not seen any problems:

03:00.0 Network controller: Intel Corporation PRO/Wireless 3945ABG [Golan] Network Connection (rev 02)
Subsystem: Intel Corporation ThinkPad R60e/X60s
Flags: bus master, fast devsel, latency 0, IRQ 47
Memory at edf00000 (32-bit, non-prefetchable) [size=4K]
Capabilities: [c8] Power Management version 2
Capabilities: [d0] MSI: Enable+ Count=1/1 Maskable- 64bit+
Capabilities: [e0] Express Legacy Endpoint, MSI 00
Capabilities: [100] Advanced Error Reporting
Capabilities: [140] Device Serial Number 00-19-d2-ff-ff-29-8a-e0
Kernel driver in use: iwl3945
Kernel modules: iwl3945



[ 7.947504] iwl3945: Intel(R) PRO/Wireless 3945ABG/BG Network Connection driver for Linux, in-tree:s
[ 7.947507] iwl3945: Copyright(c) 2003-2011 Intel Corporation
[ 7.947663] iwl3945 0000:03:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
[ 7.947679] iwl3945 0000:03:00.0: setting latency timer to 64
[ 8.174125] iwl3945 0000:03:00.0: Tunable channels: 13 802.11bg, 23 802.11a channels
[ 8.174133] iwl3945 0000:03:00.0: Detected Intel Wireless WiFi Link 3945ABG
[ 8.174322] iwl3945 0000:03:00.0: irq 47 for MSI/MSI-X
[ 8.203609] ieee80211 phy0: Selected rate control algorithm 'iwl-3945-rs'
[ 16.117376] iwl3945 0000:03:00.0: loaded firmware version 15.32.2.9

@Tom yes, my system is fully up-to-date.
Yes, i see errors with booting without network daemons
Only occur when i do ifconfig wlan0 up. Other interfaces (lo, eth0) are stable.
Yep, 03:00.0 Network controller [0280]: Intel Corporation PRO/Wireless 3945ABG [Golan] Network Connection [8086:4222] (rev 02) is my WiFi chip

@Artem: do you have a file called /lib/udev/rules.d/50-firmware.rules? Or any other files in /lib/udev/rules.d/ not owned by any package? 'pacman -Qo /lib/udev/rules.d/*.rules' will tell you.

Should add: if you do have such files, delete them and try again.

Updated title to reflect actual "faulty" package.

@Tom: o! I found my problem. In /etc/udev/rules.d/ I found old *.rules. After delete them, WiFi is working! Thanks! :)

I seem to have a similar problem on my wife's IBM X41. The wifi device is an "Intel(R) PRO/Wireless 2200". After install of kmod (and deletion of module-init-tools the wifi no longer works. The system needs about 4-5 minutes to boot (it has built in an SSD instead of the harddrive, so it normally boots within 3-4 seconds!). I believe that it is not only a problem with wifi, but also with startup of some other things. This happens with the Arch Linux kernel 3.2.1 as well as all of my vanilla kernels (3.0.4, 3.1.4, and 3.2.0).

dmesg:

[ 2.477950] ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.2.2kmprq
[ 2.477955] ipw2200: Copyright(c) 2003-2006 Intel Corporation
[ 2.643224] ipw2200 0000:04:02.0: PCI INT A -> GSI 21 (level, low) -> IRQ 21
[ 2.643261] ipw2200: Detected Intel PRO/Wireless 2915ABG Network Connection
[ 32.966617] ipw2200: Failed to send TX_POWER: Already sending a command.
[ 33.087353] ipw2200: Failed to send TX_POWER: Already sending a command.
[ 33.208016] ipw2200: Failed to send TX_POWER: Already sending a command.
[ 33.328537] ipw2200: Failed to send TX_POWER: Already sending a command.
[ 33.449044] ipw2200: Failed to send TX_POWER: Already sending a command.
[ 33.449123] ipw2200: Unable to initialize device after 5 attempts.
[ 33.450825] ipw2200: failed to register network device
[ 33.450944] ipw2200 0000:04:02.0: PCI INT A disabled
[ 33.450997] ipw2200: probe of 0000:04:02.0 failed with error -5

I have two custom udev rules files in /etc/udev/rules.d, as attached:

10-network.rules: these rules ensure that my network devices have always the same names ("eth0" and "wlan0").
51-arch.rules: for nvram

These rules (and the rest of the notebook) have always worked as expected until yesterday.

Hello,

I have just the same problem as Stefan Förster (HotblackDesiato) has. After upgrading to kmod the device wifi Intel(R) PRO/Wireless 2200 does not work. I have see if I have a file called /lib/udev/rules.d/50-firmware.rules but I haven't and doing pacman -Qo /lib/udev/rules.d/*.rules shows all files are owned.

Description:
After install udev-177-3(even 178-1),i got a broken ipw2200 on my laptop (HP dv1118),but could enter into LXDE without WiFi.
Kernel: Linux 3.2.1-1-ARCH(even 3.2.1-2-ARCH)

Sat Jan 21 16:58:19 2012: :: Adjusting system time and setting kernel timezone [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Starting UDev Daemon [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Triggering UDev uevents [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Loading User-specified Modules [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Waiting for UDev uevents to be processed [BUSY] udevd[160]: worker [165] timeout, kill it
Sat Jan 21 16:58:19 2012:
Sat Jan 21 16:58:19 2012: udevd[160]: seq 914 '/devices/pci0000:00/0000:00:1e.0/0000:02:00.0' killed
Sat Jan 21 16:58:19 2012:
Sat Jan 21 16:58:19 2012: udevd[160]: worker [166] timeout, kill it
Sat Jan 21 16:58:19 2012:
Sat Jan 21 16:58:19 2012: udevd[160]: seq 915 '/devices/pci0000:00/0000:00:1e.0/0000:02:06.0' killed
Sat Jan 21 16:58:19 2012:
Sat Jan 21 16:58:19 2012: [DONE]
Sat Jan 21 16:58:19 2012: :: Bringing up loopback interface [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: udevd[160]: worker [165] terminated by signal 9 (Killed)
Sat Jan 21 16:58:19 2012:
Sat Jan 21 16:58:19 2012: udevd[160]: worker [166] terminated by signal 9 (Killed)
Sat Jan 21 16:58:19 2012:
Sat Jan 21 16:58:19 2012: :: Checking Filesystems [BUSY] /dev/sda6: clean, 127100/625856 files, 1003041/2500107 blocks
Sat Jan 21 16:58:19 2012: /dev/sda7: clean, 68395/6512640 files, 22086591/26029308 blocks
Sat Jan 21 16:58:19 2012: [DONE]
Sat Jan 21 16:58:19 2012: :: Remounting Root [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Mounting Local Filesystems [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Activating Swap [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Configuring Time Zone [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Initializing Random Seed [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Removing Leftover Files [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Setting Hostname: arch [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Setting Consoles to UTF-8 mode [BUSY] [DONE]
Sat Jan 21 16:58:19 2012: :: Loading Keyboard Map: us [BUSY] [DONE]
Sat Jan 21 16:58:20 2012: :: Saving dmesg Log [BUSY] [DONE]
Sat Jan 21 16:58:20 2012: INIT: Entering runlevel: 5
Sat Jan 21 16:58:20 2012: :: Starting Syslog-NG [BUSY] [DONE]
Sat Jan 21 16:58:20 2012: :: Starting D-BUS system messagebus [BUSY] [DONE]
Sat Jan 21 16:58:21 2012: :: Starting NetworkManager [BUSY] [DONE]
Sat Jan 21 16:58:21 2012: :: Mounting Network Filesystems [BUSY] [DONE]
Sat Jan 21 16:58:22 2012: :: Starting Cron Daemon [BUSY] [DONE]
Sat Jan 21 16:58:22 2012: :: Starting Open Sound System [BUSY] [DONE]
Sat Jan 21 16:58:23 2012: :: Starting Samba Server [BUSY] [FAIL]

Jan 21 16:58:20 arch kernel: [ 35.702071] ipw2200: Failed to send TX_POWER: Already sending a command.
Jan 21 16:58:20 arch kernel: [ 35.811396] ipw2200: Failed to send TX_POWER: Already sending a command.
Jan 21 16:58:20 arch kernel: [ 35.920617] ipw2200: Failed to send TX_POWER: Already sending a command.
Jan 21 16:58:20 arch kernel: [ 36.029808] ipw2200: Failed to send TX_POWER: Already sending a command.
Jan 21 16:58:20 arch kernel: [ 36.139014] ipw2200: Failed to send TX_POWER: Already sending a command.
Jan 21 16:58:20 arch kernel: [ 36.139115] ipw2200: Unable to initialize device after 5 attempts.
Jan 21 16:58:20 arch kernel: [ 36.143067] ipw2200: failed to register network device

Rollback system to udev-175-1 (keep kmod 4-1)temporarily solved the problem,the computer could run regularly.

I ran into the same problem (IPW2200BG) as Thomas Förster.

Downgrading to udev-175-1 and mkinitcpio-0.8.0-3 is a temporary fix.

It seems we are dealing with two different bugs.

The firmware issue that Artem Lukyanov initially reported for his IPW3945ABG is fixed according to his previous comment.

today,upgrading udev-179-1[testing],the same problem.

Of course it's the same problem. We didn't expect a new udev (or kmod for that matter) to fix this. The problem is still the kernel module pulling in firmware on the init_module call rather than ifup.

I'll close the 3945 bug as fixed. Could people who still have issues with ipw2200 please open a new report, attaching all the relevant logs. I'm not sure what the problem is, but before asking upstream for help it would be nice to have a relatively clear bug report.