The problem at the kernel over 2.6.20

Live forum: http://rt2x00.serialmonkey.com/viewtopic.php?t=4486

cyclon

05-12-2007 08:03:05

All works on the kernel to 2.6.20, but in older kernels such as 2.6.21 or older, I get these messages

Nov 29 011008 mcyclon kernel BUG scheduling while atomic rt73/0x00000002/15424
Nov 29 011008 mcyclon kernel [schedule+109/998] schedule+0x6d/0x3e6
Nov 29 011008 mcyclon kernel [<c03996bb>] schedule+0x6d/0x3e6
Nov 29 011008 mcyclon kernel [<f0d90325>] RTUSBMlmeHardTransmit+0x1b0/0x1ee [rt73]
Nov 29 011008 mcyclon kernel [__mod_timer+165/192] __mod_timer+0xa5/0xc0
Nov 29 011008 mcyclon kernel [<c0122c3c>] __mod_timer+0xa5/0xc0
Nov 29 011008 mcyclon kernel [schedule_timeout+110/139] schedule_timeout+0x6e/0x8b
Nov 29 011008 mcyclon kernel [<c039a2f1>] schedule_timeout+0x6e/0x8b
Nov 29 011008 mcyclon kernel [process_timeout+0/5] process_timeout+0x0/0x5
Nov 29 011008 mcyclon kernel [<c0122ad5>] process_timeout+0x0/0x5
Nov 29 011008 mcyclon kernel [schedule_timeout+105/139] schedule_timeout+0x69/0x8b
Nov 29 011008 mcyclon kernel [<c039a2ec>] schedule_timeout+0x69/0x8b
Nov 29 011008 mcyclon kernel [wait_for_completion_timeout+134/250] wait_for_completion_timeout+0x86/0xfa
Nov 29 011008 mcyclon kernel [<c0399d83>] wait_for_completion_timeout+0x86/0xfa
Nov 29 011008 mcyclon kernel [default_wake_function+0/12] default_wake_function+0x0/0xc
Nov 29 011008 mcyclon kernel [<c0118123>] default_wake_function+0x0/0xc
Nov 29 011008 mcyclon kernel [usb_start_wait_urb+82/137] usb_start_wait_urb+0x52/0x89
Nov 29 011008 mcyclon kernel [<c02ec668>] usb_start_wait_urb+0x52/0x89
Nov 29 011008 mcyclon kernel [usb_control_msg+184/210] usb_control_msg+0xb8/0xd2
Nov 29 011008 mcyclon kernel [<c02ec86b>] usb_control_msg+0xb8/0xd2
Nov 29 011008 mcyclon kernel [update_stats_wait_end+150/185] update_stats_wait_end+0x96/0xb9
Nov 29 011008 mcyclon kernel [<c0116bd9>] update_stats_wait_end+0x96/0xb9
Nov 29 011008 mcyclon kernel [<f0d8a7a4>] RTUSB_VendorRequest+0xc8/0xd5 [rt73]
Nov 29 011008 mcyclon kernel [<f0d8a97b>] RTUSBReadMACRegister+0x31/0x35 [rt73]
Nov 29 011008 mcyclon kernel [<f0d8a9fa>] RTUSBWriteBBPRegister+0x1d/0x89 [rt73]
Nov 29 011008 mcyclon kernel [<f0d84c5d>] AsicSwitchChannel+0x2fc/0x3b2 [rt73]
Nov 29 011008 mcyclon kernel [del_timer+74/99] del_timer+0x4a/0x63
Nov 29 011008 mcyclon kernel [<c01227c9>] del_timer+0x4a/0x63
Nov 29 011008 mcyclon kernel [<f0d85075>] MlmePostRestartStateMachine+0x26/0x2e [rt73]
Nov 29 011008 mcyclon kernel [<f0d87763>] MlmeHandler+0x6d/0x147 [rt73]
Nov 29 011008 mcyclon kernel [<f0d824c0>] MlmeThread+0x60/0x78 [rt73]
Nov 29 011008 mcyclon kernel [<f0d82460>] MlmeThread+0x0/0x78 [rt73]
Nov 29 011008 mcyclon kernel [kernel_thread_helper+7/16] kernel_thread_helper+0x7/0x10
Nov 29 011008 mcyclon kernel [<c01040e3>] kernel_thread_helper+0x7/0x10
Nov 29 011008 mcyclon kernel =======================

Nov 29 011010 mcyclon kernel BUG soft lockup detected on CPU#0!
Nov 29 011010 mcyclon kernel [softlockup_tick+153/193] softlockup_tick+0x99/0xc1
Nov 29 011010 mcyclon kernel [<c0143bc5>] softlockup_tick+0x99/0xc1
Nov 29 011010 mcyclon kernel [update_process_times+58/93] update_process_times+0x3a/0x5d
Nov 29 011010 mcyclon kernel [<c0122d06>] update_process_times+0x3a/0x5d
Nov 29 011010 mcyclon kernel [tick_sched_timer+307/387] tick_sched_timer+0x133/0x183
Nov 29 011010 mcyclon kernel [<c0131eba>] tick_sched_timer+0x133/0x183
Nov 29 011010 mcyclon kernel [hrtimer_interrupt+302/460] hrtimer_interrupt+0x12e/0x1cc
Nov 29 011010 mcyclon kernel [<c012e313>] hrtimer_interrupt+0x12e/0x1cc
Nov 29 011010 mcyclon kernel [tick_sched_timer+0/387] tick_sched_timer+0x0/0x183
Nov 29 011010 mcyclon kernel [<c0131d87>] tick_sched_timer+0x0/0x183
Nov 29 011010 mcyclon kernel [tick_sched_timer+0/387] tick_sched_timer+0x0/0x183
Nov 29 011010 mcyclon kernel [<c0131d87>] tick_sched_timer+0x0/0x183
Nov 29 011010 mcyclon kernel [timer_interrupt+11/17] timer_interrupt+0xb/0x11
Nov 29 011010 mcyclon kernel [<c0105cbb>] timer_interrupt+0xb/0x11
Nov 29 011010 mcyclon kernel [handle_IRQ_event+26/63] handle_IRQ_event+0x1a/0x3f
Nov 29 011010 mcyclon kernel [<c0143e3c>] handle_IRQ_event+0x1a/0x3f
Nov 29 011010 mcyclon kernel [handle_level_irq+102/235] handle_level_irq+0x66/0xeb
Nov 29 011010 mcyclon kernel [<c0145138>] handle_level_irq+0x66/0xeb
Nov 29 011010 mcyclon kernel [do_IRQ+114/138] do_IRQ+0x72/0x8a
Nov 29 011010 mcyclon kernel [<c0105320>] do_IRQ+0x72/0x8a
Nov 29 011010 mcyclon kernel [common_interrupt+35/40] common_interrupt+0x23/0x28
Nov 29 011010 mcyclon kernel [<c0103ff3>] common_interrupt+0x23/0x28
Nov 29 011010 mcyclon kernel [acpi_processor_idle+536/873] acpi_processor_idle+0x218/0x369
Nov 29 011010 mcyclon kernel [<c027dadb>] acpi_processor_idle+0x218/0x369
Nov 29 011010 mcyclon kernel [cpu_idle+70/104] cpu_idle+0x46/0x68
Nov 29 011010 mcyclon kernel [<c010234c>] cpu_idle+0x46/0x68
Nov 29 011010 mcyclon kernel [start_kernel+599/604] start_kernel+0x257/0x25c
Nov 29 011010 mcyclon kernel [<c04d29ed>] start_kernel+0x257/0x25c
Nov 29 011010 mcyclon kernel [unknown_bootoption+0/406] unknown_bootoption+0x0/0x196
Nov 29 011010 mcyclon kernel [<c04d2317>] unknown_bootoption+0x0/0x196
Nov 29 011010 mcyclon kernel =======================

What to do?

Thanks in advance!

cyclon

07-12-2007 22:57:14

I found that the problem was related to the parameter "Preemption Model", which was established in "Preemtible kernel (Low-latency Desktop)". When put "Voluntary kernel preemption", the error has disappeared.

Is it something that would make everything work with "Preemtible kernel (Low-latency Desktop)"?

Spy84464

10-12-2007 18:52:05

Hello,
Have you tried a very recent CVS tarball?

Regards,
Romain

cyclon

10-12-2007 19:25:10

Last I tried, it is rt73-cvs-2007120709.

cyclon

11-12-2007 06:02:05

Now I tried rt73-cvs-2007121012 at 2.6.23.3. I have the same result.