Description of problem: I'm using a laptop with an internal ipw1945 network adapter, and so now with recent rawhide kernels the iwlwifi driver gets loaded. I'm also using a USB network adapter (as the ipw1945 does not work yet), a zd1211rw based adapter. Now with rawhide the zd1211rw_mac80211, iwlwifi *and* the zd1211rw driver get loaded, leading to the oops below. Probably either the zd1211rw *or* the zd1211rw_mac80211 should be turned on in the config, not both. Version-Release number of selected component (if applicable): 2.6.20-1.2960.fc7 How reproducible: Always, usually during boot (which also disables the keyboard). Steps to Reproduce: 1. Boot with a zd1211rw USB dongle attached. 2. Boom. Additional info: Modules linked in: zd1211rw ieee80211softmac ieee80211 ieee80211_crypt zd1211rw_mac80211 cpufreq_ondemand cpufreq_powersave cpufreq_conservative rfcomm l2cap xt_tcpudp ip6t_REJECT ip6table_filter ip6_tables x_tables dm_mirror dm_multipath dm_mod video sbs i2c_ec button dock battery asus_acpi ac ipv6 parport_pc lp parport arc4 ecb blkcipher snd_hda_intel snd_hda_codec snd_seq_dummy snd_seq_oss snd_seq_midi_event fw_ohci snd_seq hci_usb bluetooth iwlwifi fw_core rtc_cmos mac80211 8139cp snd_seq_device rtc_core 8139too rtc_lib snd_pcm_oss mii cfg80211 sdhci mmc_core snd_mixer_oss iTCO_wdt iTCO_vendor_support snd_pcm serio_raw snd_timer snd soundcore i2c_i801 snd_page_alloc i2c_core pcspkr sr_mod cdrom joydev sg ata_piix ata_generic libata sd_mod scsi_mod ext3 jbd mbcache ehci_hcd ohci_hcd uhci_hcd CPU: 0 EIP: 0060:[<c04352f5>] Not tainted VLI EFLAGS: 00010246 (2.6.20-1.2999.fc7 #1) EIP is at queue_work+0x24/0x4d eax: 00000008 ebx: 00000000 ecx: 00000000 edx: f2e7b0b0 esi: 00000000 edi: f2e782e0 ebp: c2498cfc esp: c2498cf4 ds: 007b es: 007b fs: 00d8 gs: 0000 ss: 0068 Process events/0 (pid: 8, ti=c2498000 task=c2494030 task.ti=c2498000) Stack: f2e791e0 f30ab6e0 c2498d1c f8a9271b f2eb8c9c c2498d1c f8a40116 f75691f8 f30ab6e0 f2e782e0 c2498e1c f8a42880 f30ab6e0 f2d01000 f8a4ae7e 00000000 00000002 0000002d 0000002f 00000036 000000d7 00000001 00000000 00000089 Call Trace: [<c04061ed>] show_trace_log_lvl+0x1a/0x2f [<c040629d>] show_stack_log_lvl+0x9b/0xa3 [<c040645d>] show_registers+0x1b8/0x289 [<c040665b>] die+0x12d/0x242 [<c061727e>] do_page_fault+0x3ee/0x4ba [<c0615a14>] error_code+0x7c/0x84 [<f8a9271b>] ipw_rate_scale_rate_init+0xda/0xe2 [iwlwifi] [<f8a42880>] ieee80211_rx_mgmt_assoc_resp+0x571/0x5cb [mac80211] [<f8a437f1>] ieee80211_sta_work+0x99e/0x1482 [mac80211] [<c043514f>] run_workqueue+0x89/0x145 [<c0435b6b>] worker_thread+0xf8/0x124 [<c0438357>] kthread+0xb3/0xdc [<c0405cd7>] kernel_thread_helper+0x7/0x10 ======================= Code: 72 ff ff ff 5b 5d c3 55 89 c1 89 e5 56 53 64 8b 35 04 00 00 00 f0 0f ba 2a 00 19 c0 31 db 85 c0 75 2c 8b 1d a8 06 7f c0 8d 41 08 <39> 41 08 8d 42 04 0f 45 de 39 42 04 74 04 0f 0b eb fe 8b 01 f7 EIP: [<c04352f5>] queue_work+0x24/0x4d SS:ESP 0068:c2498cf4 ADDRCONF(NETDEV_CHANGE): wlan1: link becomes ready wlan1: duplicate address detected!
Can you still recreate this with current rawhide kernels? BTW, I think that summary change was wrong, as the stack trace points directly at the iwlwifi driver...
I have disabled the softmac-based "zd1211rw" driver, in favor of "zd1211rw-mac80211"...
Yes, and this fixes the problem. Thanks John.