HR timers soft lockup on resume from suspend to ram
Hi,
With recent git versions, my system won't resume from suspend to ram. At resume time, the kernel hangs just after printing "Enabling non-boot CPUs ...". This is on a Core Duo system (Mac Mini). Looking at the BUG, it looks related to hrtimers.
When the system is hung, and I press on the power button, I get this backtrace (copied from the screen, netconsole won't allow to capture it):
BUG: soft lockup detected on CPU#0! [<c0155c88>] softlockup_tick+0xa8/0x110 [<c0130283>] update_process_times+0x33/0x80 [<c014215b>] tick_sched_timer+0x5b/0xc0 [<c013e3f3>] hrtimer_interrupt+0x143/0x1d0 [<c01172e0>] smp_apic_timer_interrupt+0x50/0x80 [<c0104e4c>] apic_timer_interrupt+0x28/0x30 [<c0372926>] _spin_unlock_irq+0x6/0x30 [<c0370488>] __sched_text_start+0x308/0x880 [<c0372909>] _spin_unlock_irqrestore+0x19/0x30 [<c01378ee>] worker_thread+0x12e/0x170 [<c0120a50>] default_wake_function+0x0/0x10 [<c01377c0>] worker_thread+0x0/0x170 [<c013a88a>] kthread+0xba/0xf0 [<c013a7d0>] kthread+0x0/0xf0 [<c0104fb3>] kernel_thread_helper+0x7/0x14 ======================
I get it every time I press on the power button.
Fred.
Quoted 22 lines · Frédéric RISS
When the system is hung, and I press on the power button, I get this backtrace (copied from the screen, netconsole won't allow to capture it):
BUG: soft lockup detected on CPU#0! [<c0155c88>] softlockup_tick+0xa8/0x110 [<c0130283>] update_process_times+0x33/0x80 [<c014215b>] tick_sched_timer+0x5b/0xc0 [<c013e3f3>] hrtimer_interrupt+0x143/0x1d0 [<c01172e0>] smp_apic_timer_interrupt+0x50/0x80 [<c0104e4c>] apic_timer_interrupt+0x28/0x30 [<c0372926>] _spin_unlock_irq+0x6/0x30 [<c0370488>] __sched_text_start+0x308/0x880 [<c0372909>] _spin_unlock_irqrestore+0x19/0x30 [<c01378ee>] worker_thread+0x12e/0x170 [<c0120a50>] default_wake_function+0x0/0x10 [<c01377c0>] worker_thread+0x0/0x170 [<c013a88a>] kthread+0xba/0xf0 [<c013a7d0>] kthread+0x0/0xf0 [<c0104fb3>] kernel_thread_helper+0x7/0x14 ======================
Quoted 1 line
I get it every time I press on the power button.
Turns, out this wasn't true. In fact multiple power button presses will bring the system back to life. I hadn't the right console loglevel setting when I first tested that.
With a higher loglevel I get this by pressing the power button at least 4 times:
Mar 4 17:43:58 minifunky kernel: [ 580.222727] Enabling non-boot CPUs ... Mar 4 17:43:58 minifunky kernel: [ 597.696477] BUG: soft lockup detected on CPU#0! Mar 4 17:43:58 minifunky kernel: [ 597.698186] [softlockup_tick+168/272] softlockup_tick+0xa8/0x110 Mar 4 17:43:58 minifunky kernel: [ 597.699924] [update_process_times+51/128] update_process_times+0x33/0x80 Mar 4 17:43:58 minifunky kernel: [ 597.701676] [tick_sched_timer+91/192] tick_sched_timer+0x5b/0xc0 Mar 4 17:43:58 minifunky kernel: [ 597.703453] [hrtimer_interrupt+323/464] hrtimer_interrupt+0x143/0x1d0 Mar 4 17:43:58 minifunky kernel: [ 597.705255] [smp_apic_timer_interrupt+80/128] smp_apic_timer_interrupt+0x50/0x80 Mar 4 17:43:58 minifunky kernel: [ 597.707066] [apic_timer_interrupt+40/48] apic_timer_interrupt+0x28/0x30 Mar 4 17:43:58 minifunky kernel: [ 597.708885] [scrdown+27/208] scrdown+0x1b/0xd0 Mar 4 17:43:58 minifunky kernel: [ 597.710684] [worker_thread+248/368] worker_thread+0xf8/0x170 Mar 4 17:43:58 minifunky kernel: [ 597.712505] [default_wake_function+0/16] default_wake_function+0x0/0x10 Mar 4 17:43:58 minifunky kernel: [ 597.714317] [worker_thread+0/368] worker_thread+0x0/0x170 Mar 4 17:43:58 minifunky kernel: [ 597.716137] [kthread+186/240] kthread+0xba/0xf0 Mar 4 17:43:58 minifunky kernel: [ 597.717949] [kthread+0/240] kthread+0x0/0xf0 Mar 4 17:43:58 minifunky kernel: [ 597.719737] [kernel_thread_helper+7/20] kernel_thread_helper+0x7/0x14 Mar 4 17:43:58 minifunky kernel: [ 597.721562] ======================= Mar 4 17:43:58 minifunky kernel: [ 606.119695] SMP alternatives: switching to SMP code Mar 4 17:43:58 minifunky kernel: [ 606.121668] Booting processor 1/1 eip 3000 Mar 4 17:43:58 minifunky kernel: [ 606.123518] CPU 1 irqstacks, hard=c04cf000 soft=c04cd000 Mar 4 17:43:58 minifunky kernel: [ 606.135911] Initializing CPU#1 Mar 4 17:43:58 minifunky kernel: [ 606.196496] Calibrating delay using timer specific routine.. 3333.89 BogoMIPS (lpj=1666945) Mar 4 17:43:58 minifunky kernel: [ 606.196503] CPU: After generic identify, caps: bfe9fbff 00100000 00000000 00000000 0000c1a9 00000000 00000000 Mar 4 17:43:58 minifunky kernel: [ 606.196508] monitor/mwait feature present. Mar 4 17:43:58 minifunky kernel: [ 606.196512] CPU: L1 I cache: 32K, L1 D cache: 32K Mar 4 17:43:58 minifunky kernel: [ 606.196514] CPU: L2 cache: 2048K Mar 4 17:43:58 minifunky kernel: [ 606.196517] CPU: Physical Processor ID: 0 Mar 4 17:43:58 minifunky kernel: [ 606.196518] CPU: Processor Core ID: 1 Mar 4 17:43:58 minifunky kernel: [ 606.196520] CPU: After all inits, caps: bfe9fbff 00100000 00000000 00002940 0000c1a9 00000000 00000000 Mar 4 17:43:58 minifunky kernel: [ 606.196526] Intel machine check architecture supported. Mar 4 17:43:58 minifunky kernel: [ 606.196532] Intel machine check reporting enabled on CPU#1. Mar 4 17:43:58 minifunky kernel: [ 606.197027] CPU1: Intel Genuine Intel(R) CPU T2300 @ 1.66GHz stepping 08 Mar 4 17:43:58 minifunky kernel: [ 606.223215] BUG: soft lockup detected on CPU#1! Mar 4 17:43:58 minifunky kernel: [ 606.225588] [softlockup_tick+168/272] softlockup_tick+0xa8/0x110 Mar 4 17:43:58 minifunky kernel: [ 606.228020] [update_process_times+51/128] update_process_times+0x33/0x80 Mar 4 17:43:58 minifunky kernel: [ 606.230450] [tick_periodic+34/128] tick_periodic+0x22/0x80 Mar 4 17:43:58 minifunky kernel: [ 606.232846] [tick_handle_periodic+21/112] tick_handle_periodic+0x15/0x70 Mar 4 17:43:58 minifunky kernel: [ 606.235299] [smp_apic_timer_interrupt+80/128] smp_apic_timer_interrupt+0x50/0x80 Mar 4 17:43:58 minifunky kernel: [ 606.237742] [apic_timer_interrupt+40/48] apic_timer_interrupt+0x28/0x30 Mar 4 17:43:58 minifunky kernel: [ 606.240237] [_spin_unlock_irq+6/48] _spin_unlock_irq+0x6/0x30 Mar 4 17:43:58 minifunky kernel: [ 606.242760] [schedule+776/2176] __sched_text_start+0x308/0x880 Mar 4 17:43:58 minifunky kernel: [ 606.245334] [migration_thread+0/528] migration_thread+0x0/0x210 Mar 4 17:43:58 minifunky kernel: [ 606.247907] [kthread+173/240] kthread+0xad/0xf0 Mar 4 17:43:58 minifunky kernel: [ 606.250478] [kthread+0/240] kthread+0x0/0xf0 Mar 4 17:43:58 minifunky kernel: [ 606.253069] [kernel_thread_helper+7/20] kernel_thread_helper+0x7/0x14 Mar 4 17:43:58 minifunky kernel: [ 606.255677] ======================= Mar 4 17:43:58 minifunky kernel: [ 606.258301] Switched to high resolution mode on CPU 1 Mar 4 17:43:58 minifunky kernel: [ 606.263491] CPU1 is up