Re: reboot hanging

I see this also quite frequently (and have seen it for a period of time, 
running current amd64 updated almost each week).
each time i have to hard reboot, and restart in single mode to fsck my 
small boot partition (the rest is zfs, no problem there).
My proc is a core2 6400. I'm using :

nooptions     SCHED_4BSD        # 4BSD scheduler
options     SCHED_ULE        # Newer scheduler
options     PREEMPTION        # Enable kernel thread preemption

But I was using SCHED_4BSD before and it did happen as well.

Sorry no debugging available.

Bruno Damour


Kris Kennaway wrote:
> I get this fairly often when rebooting an amd64 machine.  Looks like
> it's hung trying to do a TLB shootdown.
>
> hammer1# reboot
> Aug  3 20:56:11 hammer1 reboot: rebooted by root
> Aug  3 20:56:11 hammer1 syslogd: exiting on signal 15
> Waiting (max 60 seconds) for system process `vnlru' to stop...done
> Waiting (max 60 seconds) for system process `bufdaemon' to stop...done
> Waiting (max 60 seconds) for system process `syncer' to stop...
> Syncing disks, vnodes remaining...2 2 0 1 1 1 0 0 0 done
> All buffers synced.
>
> [hung here]
>
> KDB: enter: Break sequence on console
> [thread pid 16 tid 100005 ]
> Stopped at      kdb_enter+0x31: leave
> db> wh
> Tracing pid 16 tid 100005 td 0xffffff00010c96a0
> kdb_enter() at kdb_enter+0x31
> siointr1() at siointr1+0xde
> siointr() at siointr+0x31
> intr_execute_handlers() at intr_execute_handlers+0x102
> lapic_handle_intr() at lapic_handle_intr+0x39
> Xapic_isr1() at Xapic_isr1+0x7f
> --- interrupt, rip = 0xffffffff8039d376, rsp = 0xffffffffabc93b70, rbp = 0xffffffffabc93b80 ---
> acpi_cpu_c1() at acpi_cpu_c1+0x6
> acpi_cpu_idle() at acpi_cpu_idle+0x1c4
> cpu_idle() at cpu_idle+0x29
> sched_idletd() at sched_idletd+0x1b1
> fork_exit() at fork_exit+0x14a
> fork_trampoline() at fork_trampoline+0xe
> --- trap 0, rip = 0, rsp = 0xffffffffabc93d30, rbp = 0 ---
> db> show allpcpu
> Current CPU: 1
>
> cpuid        = 0
> curthread    = 0xffffff00010ef9f0: pid 19 "swi4: clock sio"
> curpcb       = 0xffffffffabcb1d40
> fpcurthread  = none
> idlethread   = 0xffffff00010c3000: pid 17 "idle: cpu0"
>
> cpuid        = 1
> curthread    = 0xffffff00010c96a0: pid 16 "idle: cpu1"
> curpcb       = 0xffffffffabc93d40
> fpcurthread  = none
> idlethread   = 0xffffff00010c96a0: pid 16 "idle: cpu1"
>
> cpuid        = 2
> curthread    = 0xffffff00010c9350: pid 15 "idle: cpu2"
> curpcb       = 0xffffffffabc8ed40
> fpcurthread  = none
> idlethread   = 0xffffff00010c9350: pid 15 "idle: cpu2"
>
> cpuid        = 3
> curthread    = 0xffffff00010c6350: pid 3 "g_up"
> curpcb       = 0xffffffffabc7ad40
> fpcurthread  = none
> idlethread   = 0xffffff00010c9000: pid 14 "idle: cpu3"
>
> cpuid        = 4
> curthread    = 0xffffff00010c69f0: pid 13 "idle: cpu4"
> curpcb       = 0xffffffffabc84d40
> fpcurthread  = none
> idlethread   = 0xffffff00010c69f0: pid 13 "idle: cpu4"
>
> cpuid        = 5
> curthread    = 0xffffff00010c6000: pid 12 "idle: cpu5"
> curpcb       = 0xffffffffabc75d40
> fpcurthread  = none
> idlethread   = 0xffffff00010c6000: pid 12 "idle: cpu5"
>
> cpuid        = 6
> curthread    = 0xffffff00010c39f0: pid 11 "idle: cpu6"
> curpcb       = 0xffffffffabc70d40
> fpcurthread  = none
> idlethread   = 0xffffff00010c39f0: pid 11 "idle: cpu6"
>
> cpuid        = 7
> curthread    = 0xffffff00010c36a0: pid 10 "idle: cpu7"
> curpcb       = 0xffffffffabc6bd40
> fpcurthread  = none
> idlethread   = 0xffffff00010c36a0: pid 10 "idle: cpu7"
>
> db> wh 3
> Tracing pid 3 tid 100010 td 0xffffff00010c6350
> cpustop_handler() at cpustop_handler+0x40
> ipi_nmi_handler() at ipi_nmi_handler+0x34
> trap() at trap+0x4d
> nmi_calltrap() at nmi_calltrap+0x8
> --- trap 0x13, rip = 0xffffffff803b01b2, rsp = 0xffffffffab19aff0, rbp = 0xffffffffabc7ab00 ---
> smp_tlb_shootdown() at smp_tlb_shootdown+0xa2
> smp_invlpg_range() at smp_invlpg_range+0x1e
> pmap_invalidate_range() at pmap_invalidate_range+0x59
> pmap_qremove() at pmap_qremove+0x65
> swp_pager_async_iodone() at swp_pager_async_iodone+0x5f
> bufdone() at bufdone+0x9e
> swapgeom_done() at swapgeom_done+0x3d
> biodone() at biodone+0x88
> g_io_schedule_up() at g_io_schedule_up+0x122
> g_up_procbody() at g_up_procbody+0x78
> fork_exit() at fork_exit+0x14a
> fork_trampoline() at fork_trampoline+0xe
> --- trap 0, rip = 0, rsp = 0xffffffffabc7ad30, rbp = 0 ---
> db>
>