Re: Panic on boot on my Samsung Sens Pro 500

On Sunday 31 October 2004 06:47 pm, M. Warner Losh wrote:
> # This is RELENG_5
>
> GDB: no debug ports present
> KDB: debugger backends: ddb
> KDB: current backend: ddb
> SMAP type=01 base=0000000000000000 len=000000000009fc00
> SMAP type=02 base=000000000009fc00 len=0000000000000400
> SMAP type=02 base=00000000000e77e2 len=00000000000071d0
> SMAP type=02 base=00000000000f20a8 len=000000000000df58
> SMAP type=01 base=0000000000100000 len=0000000001700000
> SMAP type=02 base=00000000ffff20a8 len=000000000000df58
> Copyright (c) 1992-2004 The FreeBSD Project.
> Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
>         The Regents of the University of California. All rights reserved.
> FreeBSD 5.3-STABLE #1: Sun Oct 31 14:42:40 MST 2004
>    
> imp_at_paco-paco.village.org:/paco/imp/FreeBSD/5x/src/sys/i386/compile/GENERIC
> kernel trap 12 with interrupts disabled
>
>
> Fatal trap 12: page fault while in kernel mode
> fault virtual address   = 0x1c
> fault code              = supervisor read, page not present
> instruction pointer     = 0x8:0xc0761974
> stack pointer           = 0x10:0xc0c21ba4
> frame pointer           = 0x10:0xc0c21c70
> 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 ()
> [thread 0]
> Stopped at      vm_fault+0x1b0: cmpxchgl        %ecx,0x1c(%edx)
> db> trace
> vm_fault(c103a000,c0f75000,1,0,c08dc560) at vm_fault+0x1b0
> trap_pfault(c0c21cd8,0,c0f750f4) at trap_pfault+0x19f
> trap(18,10,10,c1044420,c1045780) at trap+0x315
> calltrap() at calltrap+0x5
> --- trap 0xc, eip = 0xc0f750f4, esp = 0xc0c21d18, ebp = 0xc0c21d44 ---
> end(c1044420,102,c0843640,c08b48f0,c26000) at 0xc0f750f4
> uma_prealloc(c1044420,800,c0c21d84,c07655e4,c10b3000) at uma_prealloc+0x64
> pmap_init(c10b3000,fffff000,c10b3000,c0c21d98,c05eb4c2) at pmap_init+0x67
> vm_mem_init(0,c1ec00,c1e000,0,c043fb65) at vm_mem_init+0x38
> mi_startup() at mi_startup+0x96
> begin() at begin+0x2c
> db>
>
> uma_prealloc + 100 (0x64) looks like:
>
> 0xc077733a <uma_prealloc+94>:   push   %edi
> 0xc077733b <uma_prealloc+95>:   call   0xc0774c78 <slab_zalloc>
> 0xc0777340 <uma_prealloc+100>:  mov    %eax,%ecx
>
> 0xc0f750f4 is outside of the text area (so I suspect that our stalk
> walker is a little hozed :-(.  Also of interest is that I couldn't
> boot on the video console, but had to boot over serial to even get
> this far.
>
> This seems to correspond to:
> 	while (slabs > 0) {
> -->		slab = slab_zalloc(zone, M_WAITOK);
> 		LIST_INSERT_HEAD(&keg->uk_free_slab, slab, us_link);
> 		slabs--;
>
> 	}
>
> But without more specific information about the zalloc, it is hard to
> know what to do.  The crash is 100% repeatable.
>
> Same kernel boots fine on my Dell Inspiron 3000, so it can't be the
> kernel being corrupt.  The same disk boots before and after on the
> i3k (the disk was cloned from the i3k disk...)
>
> Ideas?

I have similar problems on my ultra60 with any kernels other than GENERIC.  I 
have no idea, but this leads me to believe it is a bug in MI code somewhere.

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