Re: More MP problems.

[Adam K Kirchhoff <adamk@xxxxxxxxxxxx>]

Adam K Kirchhoff <adamk@xxxxxxxxxxxx> wrote:
> Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxx> wrote:
>>     I've committed it.  It's a hack but I need to move on.  I'm not 100%
>>     certain this will fix all the issues permanently but if the problem
>>     pops up again we will at least know where to look.

> Well something certainly popped up :-)  My machine was up all night
> without problems, building a few ports.  Just this morning, while
> building more ports, this happened:

> Fatal trap 12: page fault while in kernel mode
> mp_lock = 00000001; cpuid = 1; lapic.id = 01000000
> fault virtual address   = 0x20
> fault code              = supervisor read, page not present
> instruction pointer     = 0x8:0xc01f584c
> stack pointer           = 0x10:0xdbc8cd4c
> frame pointer           = 0x10:0xdbc8cd54
> code segment            = base 0x0, limit 0xfffff, type 0x1b
>                         = DPL 0, pres 1, def32 1, gran 1
> processor eflags        = interrupt enabled,
> resume, IOPL = 0
> current process         = Idle
> current thread          = pri 12 
> interrupt mask          =  <- SMP: XXX
> kernel: type 12 trap, code=0

> CPU1 stopping CPUs: 0x0000000d
>  stopped
> Stopped at      tcp_disconnect+0x8:     movl  0x20(%ebx),%
> db> trace
> tcp_disconnect(0,deda6b04,ff808000,dbc8cd78,c01aa6f7) at
> tcp_disconnect+0x8
> tcp_usr_detach(db0d48a0) at tcp_usr_detach+0x1e
> netmsg_pru_detach(deda6b04) at netmsg_pru_detach+0xf
> tcpmsg_service_loop(0,0,0,0,0) at tcpmsg_service_loop+0x26
> lwkt_exit() at lwkt_exit
> db> reset

It'a already happened a couple more times, now.  At least one of those
times, I had closed an ssh session just before the lockup occurred.  The
most recent time, the entire machien locked up before the debug prompt
showed up, leaving me with:

CPU0 stopping CPUs: 0x0000000e
 stopped
 Stopped at      tcp_dis

So I'll have to wait till I get home and reboot the box before I can try
any further testing.

Adam