Re: ULE panic

From: Morten Rodal <morten_at_rodal.no>
Date: Fri, 9 May 2003 21:55:54 +0200
On Thu, May 08, 2003 at 09:08:20PM +0200, ianf_at_za.uu.net wrote:
> Fatal trap 12: page fault while in kernel mode
> cpuid = 0; lapic.id = 00000000
> fault virtual address   = 0x38
> fault code              = supervisor read, page not present
> instruction pointer     = 0x8:0xc01efc87
> stack pointer           = 0x10:0xcd4dec4c
> frame pointer           = 0x10:0xcd4dec60
> 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         = 14 (swi7: tty:sio clock)
> trap number             = 12
> panic: page fault
> cpuid = 0; lapic.id = 00000000
> Stack backtrace:
> backtrace(c0338cf2,0,c0329d79,cd4deb3c,c0d1abe0) at backtrace+0x17
> panic(c0329d79,c0348f01,c0d19700,1,1) at panic+0x104
> trap_fatal(cd4dec0c,38,1,0,c0d1abe0) at trap_fatal+0x322
> trap_pfault(cd4dec0c,0,38,cd4debe4,38) at trap_pfault+0x232
> trap(cd4d0018,c01e0010,c3ad0010,c0d1abe0,c038da50) at trap+0x3bd
> calltrap() at calltrap+0x5
> --- trap 0xc, eip = 0xc01efc87, esp = 0xcd4dec4c, ebp = 0xcd4dec60 ---
> sched_choose(c038d3c0,c0d1ce40,0,c0d19658,cd4deccc) at sched_choose+0x77
> choosethread(c0d1abe0,cd4dec90,f3e70ca5,df354cc0,ffc00014) at choosethread+0x35
> mi_switch(c038abc0,0,0,0,c0d195a0) at mi_switch+0x17f
> ithread_loop(c0d18180,cd4ded48,0,0,0) at ithread_loop+0x2c8
> fork_exit(c01c5ca0,c0d18180,cd4ded48) at fork_exit+0xb0
> fork_trampoline() at fork_trampoline+0x1a
> --- trap 0x1, eip = 0, esp = 0xcd4ded7c, ebp = 0 ---
> boot() called on cpu#0
> 
> syncing disks, buffers remaining... 

I am seeing the exact same panic (see attached crash log).  The
computer will freeze at the syncing disks point and it is not able to
write a crashdump to the disk.  I have to use the reset button in
order for the system to come up again.

I have tracked this bug down to a simple

   cd /usr/src
   nice +20 make -j 8 clean

on a SMP system.  It varies a bit how far it comes (which suggests a
race condition), last time it got to "===> lib/librpcsvc".

The system is running 5.1-BETA with a kernel built (cvsuped a few
hours prior to this timestamp of course) on

  FreeBSD 5.1-BETA #0: Tue May  6 14:01:03 CEST 2003


-- 
Morten Rodal


Received on Fri May 09 2003 - 10:55:58 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:07 UTC