panic: page fault during install_ap_tramp()

[John Baldwin]

On Thursday 22 April 2004 08:39 am, Ian Freislich wrote:
> Hi
>
> I started getting this panic from Monday this week.  The system is
> an old gigabyte dual pentium II.  Scheduler is SCHED_BSD if that
> makes a difference so early on in the boot.  A while back I got a
> similar panic
> (http://www.freebsd.org/cgi/getmsg.cgi?fetch=1649457+1655104+/usr/local/www
>/db/text/2004/freebsd-current/20040208.freebsd-current) Which, in summary,
> jhb thought was as a result of a missing MADT in the ACPI whatever.  This
> problem in the above link went away and
> has now resurfaced after about 2 months.  I do a fresh world and
> kernel from current sources about every two days.

Hmm, (subject changed for the real panic).  Can you verify if a 4.x SMP kernel 
boots ok on this box?  This might be a pmap bug of some sort.  Peter might 
have a guess so I've cc'd him.

> MPTable: <OEM00000 PROD00000000>
> kernel trap 12 with interrupts disabled
>
>
> Fatal trap 12: page fault while in kernel mode
> cpuid = 0; apic id = 00
> fault virtual address   = 0x1c
> fault code              = supervisor write, page not present
> instruction pointer     = 0x8:0xc05e4035
> stack pointer           = 0x10:0xc0821b38
> frame pointer           = 0x10:0xc0821c30
> 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         = 0 ()
> trap number             = 12
> panic: page fault
> at line 815 in file ../../../i386/i386/trap.c
> cpuid = 0;
> Stack backtrace:
> backtrace(c065c878,0,32f,c067abcf,100) at backtrace+0x17
> __panic(c067abcf,32f,c065807f,c067aa27,1) at __panic+0x15d
> trap_fatal(c0821af8,1c,1010101,1010101,c06b1d00) at trap_fatal+0x376
> trap_pfault(c0821af8,0,1c,c0821ad4,1c) at trap_pfault+0x242
> trap(c0090018,c0c30010,c0c30010,c009f000,c0c3b000) at trap+0x30d
> calltrap() at calltrap+0x5
> --- trap 0xc, eip = 0xc05e4035, esp = 0xc0821b38, ebp = 0xc0821c30 ---
> vm_fault(c0c3b000,c009f000,2,0,c06b1d00) at vm_fault+0x1e5
> trap_pfault(c0821cac,0,c009f000,0,c009f000) at trap_pfault+0x1e4
> trap(18,10,7070010,18e,c06a6e00) at trap+0x30d
> calltrap() at calltrap+0x5
> --- trap 0xc, eip = 0xc06236e3, esp = 0xc0821cec, ebp = 0xc0821d00 ---
> install_ap_tramp(c06c7d80,c0679f7a,0,1,c0512bb7) at install_ap_tramp+0x53
> start_all_aps(c06c7c00,c0679e6f,0,1,8) at start_all_aps+0x45
> cpu_mp_start(c06bbb60,c066a515,0,1,c06a0670) at cpu_mp_start+0x21f
> mp_start(0,81ec00,81e000,81ec00,81e000) at mp_start+0x60
> mi_startup() at mi_startup+0xb5
> begin() at begin+0x2c
> Uptime: 1s
> panic: kthread_create called too soon
> at line 84 in file ../../../kern/kern_kthread.ccpuid = 0;
> Uptime: 1s

-- 
John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org