I've been running into regular system lockups during boot, and one kernel panic (flashing num/scroll lock). It seems to be related to booting with the kill-switch enabled (wireless disabled). At least I've not observed it happening if the wireless signal is enabled. This seems to be a new problem with 2.6.19-gentoo-r5 and the latest ipw drivers. I've never enountered it on 2.6.18-gentoo-r4 Reproducible: Sometimes Steps to Reproduce: Unpredictable. Boot with wireless "killed" - sometimes it hangs, sometimes it doesn't. 2.6.19-gentoo-r5 #2 SMP Sat Feb 24 00:19:57 EST 2007 i686 Genuine Intel(R) CPU T2500 @ 2.00GHz GenuineIntel GNU/Linux [I--] [ ] net-wireless/ipw3945-1.2.0 (0) [I--] [ ] net-wireless/ipw3945-ucode-1.14.2 (0) [I--] [ ] net-wireless/ipw3945d-1.7.22-r4 (0) [I--] [ ] net-wireless/wireless-tools-28 (0) [I--] [ ] sys-kernel/gentoo-sources-2.6.19-r5 Feb 27 18:58:54 cardinal BUG: soft lockup detected on CPU#0! Feb 27 18:58:54 cardinal [<c013d15f>] softlockup_tick+0x92/0xb4 Feb 27 18:58:54 cardinal [<c012905f>] update_process_times+0x26/0x5c Feb 27 18:58:54 cardinal [<c01113ea>] smp_apic_timer_interrupt+0x91/0xb7 Feb 27 18:58:54 cardinal [<c01036e7>] apic_timer_interrupt+0x1f/0x24 Feb 27 18:58:54 cardinal [<c02287ff>] delay_tsc+0xb/0x13 Feb 27 18:58:54 cardinal [<c0228833>] __delay+0x6/0x7 Feb 27 18:58:54 cardinal [<f8a2e378>] ipw_bg_alive_start+0x112/0x1c7 [ipw3945] Feb 27 18:58:54 cardinal [<c012f08b>] run_workqueue+0x72/0xec Feb 27 18:58:54 cardinal [<f8a2e266>] ipw_bg_alive_start+0x0/0x1c7 [ipw3945] Feb 27 18:58:54 cardinal [<c012f7a9>] worker_thread+0x11a/0x135 Feb 27 18:58:54 cardinal [<c011ab0a>] default_wake_function+0x0/0xc Feb 27 18:58:54 cardinal [<c012f68f>] worker_thread+0x0/0x135 Feb 27 18:58:54 cardinal [<c0131e6a>] kthread+0xde/0xe2 Feb 27 18:58:54 cardinal [<c0131d8c>] kthread+0x0/0xe2 Feb 27 18:58:54 cardinal [<c0103837>] kernel_thread_helper+0x7/0x10 Feb 27 18:58:54 cardinal ======================= Feb 27 19:03:57 cardinal BUG: soft lockup detected on CPU#0! Feb 27 19:03:57 cardinal [<c013d15f>] softlockup_tick+0x92/0xb4 Feb 27 19:03:57 cardinal [<c012905f>] update_process_times+0x26/0x5c Feb 27 19:03:57 cardinal [<c01113ea>] smp_apic_timer_interrupt+0x91/0xb7 Feb 27 19:03:57 cardinal [<c01036e7>] apic_timer_interrupt+0x1f/0x24 Feb 27 19:03:57 cardinal [<c022007b>] as_dispatch_request+0x18/0x3c1 Feb 27 19:03:57 cardinal [<f8a2e3d4>] ipw_bg_alive_start+0x16e/0x1c7 [ipw3945] Feb 27 19:03:57 cardinal [<c012f08b>] run_workqueue+0x72/0xec Feb 27 19:03:57 cardinal [<f8a2e266>] ipw_bg_alive_start+0x0/0x1c7 [ipw3945] Feb 27 19:03:57 cardinal [<c012f7a9>] worker_thread+0x11a/0x135 Feb 27 19:03:57 cardinal [<c011ab0a>] default_wake_function+0x0/0xc Feb 27 19:03:57 cardinal [<c012f68f>] worker_thread+0x0/0x135 Feb 27 19:03:57 cardinal [<c0131e6a>] kthread+0xde/0xe2 Feb 27 19:03:57 cardinal [<c0131d8c>] kthread+0x0/0xe2 Feb 27 19:03:57 cardinal [<c0103837>] kernel_thread_helper+0x7/0x10 Feb 24 00:49:25 cardinal BUG: soft lockup detected on CPU#0! Feb 24 00:49:25 cardinal [<c013d15f>] softlockup_tick+0x92/0xb4 Feb 24 00:49:25 cardinal [<c012905f>] update_process_times+0x26/0x5c Feb 24 00:49:25 cardinal [<c01113ea>] smp_apic_timer_interrupt+0x91/0xb7 Feb 24 00:49:25 cardinal [<c01036e7>] apic_timer_interrupt+0x1f/0x24 Feb 24 00:49:25 cardinal [<c022007b>] as_dispatch_request+0x18/0x3c1 Feb 24 00:49:25 cardinal [<f8a2b3d4>] ipw_bg_alive_start+0x16e/0x1c7 [ipw3945] Feb 24 00:49:25 cardinal [<f8c6e200>] hsfengine4006_+0x1f0/0x4b0 [hsfengine] Feb 24 00:49:25 cardinal [<c012f08b>] run_workqueue+0x72/0xec Feb 24 00:49:25 cardinal [<f8a2b266>] ipw_bg_alive_start+0x0/0x1c7 [ipw3945] Feb 24 00:49:25 cardinal [<c012f7a9>] worker_thread+0x11a/0x135 Feb 24 00:49:25 cardinal [<c011ab0a>] default_wake_function+0x0/0xc Feb 24 00:49:25 cardinal [<c012f68f>] worker_thread+0x0/0x135 Feb 24 00:49:25 cardinal [<c0131e6a>] kthread+0xde/0xe2 Feb 24 00:49:25 cardinal [<c0131d8c>] kthread+0x0/0xe2 Feb 24 00:49:25 cardinal [<c0103837>] kernel_thread_helper+0x7/0x10 Feb 24 00:49:25 cardinal =======================
We can't do much about bugs found in upstream code. Their bugzilla contains a patch that should solve that, please see http://www.bughost.org/bugzilla/show_bug.cgi?id=1096 . If it fixes the problem for you, the best thing would be to ask them to release an updated version of their package.
Should be fixed in 1.2.1 (see bug 181610).
