OS X 10.4.9 not so hot ?

Jonathan Schneider jon at axismilton.ltd.uk
Wed Aug 29 19:35:06 BST 2007

At the risk of turning this mailing list into a Mac fixing one I  
thought I'd post this here as I know there are quite a few of you  
with knowledge of these things.

Basically my iBook used to be dead stable until about June which  
corresponds to the time of the 10.4.9 update. Filesystem checks and  
hardware memory diagnostics reveal nothing.

Now I get panics, not so much when using it, but as often as not when  
it's idle but I suspect only when I've done something in particular  
(USB drive or VLC or Azureus ?) since the last boot in this case  
yesterday. What's more I have several of exactly the same panic, for  
a mutex.

Not knowing exactly how to do it I've loaded the kernel into gdb and  
used x/i commands on the addressed it to put meaning onto the  
backtrace in the panic log. If I've got it right it happens during  
the open system call as you can see below.

Before going into detail has anybody else had similar experiences ?

There's nothing attached, no network filesystem so it does indeed  
look like a bug in the filesystem code. I was wondering/hoping if one  
of you clever kernelly filesystem people might have a bit of a clue  
as to what might be going on from the backtrace below. As I said the  
machine is idle apart from mail and instant messenger applications.  
When things  go bad the window manager (and expose) still kind of  
work and some applications can still be interacted with for a while  
before they all hang. Also emacs dired sort of works for a while.  
Definitely can't start Terminal.

I suppose what I would need to do next is make the panic handler drop  
into gdb when it happens if that's possible.



Wed Aug 22 13:49:12 2007
panic(cpu 0 caller 0x000A4354): attempt to interlock mutex  
(0x028D140C) failed on mutex lock

Latest stack backtrace for cpu 0:
          0x000952D8 0x000957F0 0x00026898 0x000A4354 0x000E5200  
0x000E5FC8 0x000E6408 0x000E8864
          0x000E8140 0x000E89E8 0x00222CE0 0x0023CFF8 0x000FB758  
0x000E9768 0x000F4F0C 0x000ED368
          0x000ED728 0x002AB548 0x000ABB30 0xA9AAABAB
Proceeding back via exception chain:
    Exception state (sv=0x3334D780)
       PC=0x90001F2C; MSR=0x0200D030; DAR=0x26CB0040;  
DSISR=0x40000000; LR=0x0000A930; R1=0xBFFFF4E0; XCP=0x00000030 (0xC00  
- System call)

Kernel version:
Darwin Kernel Version 8.10.0: Wed May 23 16:50:59 PDT 2007;  

[x/i on the backtrace addresses...]
0xa4354 <mlckPatch_isync+384>:  tweq    r4,r4
0xa4340 <mlckPatch_isync+364>:  bne+    0xa4358 <mlckPatch_isync+388>
0x952d8 <print_backtrace+176>:  bne+    cr4,0x953a8 <print_backtrace 
0x957f0 <Debugger+108>: lwz     r0,556(r31)
0x26898 <panic+476>:    lwz     r0,560(r30)
0xa4354 <mlckPatch_isync+384>:  tweq    r4,r4
0xe5200 <vfs_getnewfsid+348>:   mr      r3,r30
0xe5fc8 <vflush+908>:   lwz     r30,84(r31)
0xe6408 <vnode_recycle+300>:    lwz     r2,64(r31)
0xe8864 <vnode_reclaim+156>:    cmpwi   cr7,r30,0
0xe8140 <vfs_event_signal+2980>:        lwz     r0,16(r31)
0xe89e8 <vnode_create+68>:      mr.     r29,r3
0x222ce0 <CompareExtendedCatalogKeys+9740>:     mr.     r30,r3
0x23cff8 <MacToVFSError+24056>: mr.     r30,r3
0xfb758 <VNOP_CREATE+140>:      mr.     r31,r3
0xe9768 <vn_create+348>:        b       0xe97a0 <vn_create+404>
0xf4f0c <vn_open_auth+260>:     cmpwi   cr3,r3,0
0xed368 <chroot+668>:   mr.     r31,r3
0xed728 <open+136>:     lwz     r0,344(r1)
0x2ab548 <unix_syscall+756>:    lwz     r0,20508(r29)
0xabb30 <shandler+272>: li      r3,7

More information about the Ukfreebsd mailing list