Messages in this thread | | | Date | Sun, 23 May 2010 23:17:17 +0200 | From | Borislav Petkov <> | Subject | Re: BUG: unable to handle kernel paging request |
| |
From: Markus <M4rkusXXL@web.de> Date: Sun, May 23, 2010 at 10:44:41PM +0200
> It is 0000000001912178. > http://img256.imageshack.us/img256/4370/shotn.gif
True story, the fact that this value is not what it is expected to be judging from the code it happens might actually be a clue:
[64112.502507] Code: 8b bf b0 00 00 00 31 ed 48 8d 50 f0 48 85 c0 48 0f 45 ea 48 85 ff 74 10 48 89 ee e8 8d d3 ff ff 85 c0 0f 8e c5 00 00 00 48 89 eb <49> 8b bd b8 00 00 00 48 85 ff 74 10 48 89 ee e8 6e d3 ff ff 85 All code ======== 0: 8b bf b0 00 00 00 mov 0xb0(%rdi),%edi 6: 31 ed xor %ebp,%ebp 8: 48 8d 50 f0 lea -0x10(%rax),%rdx c: 48 85 c0 test %rax,%rax f: 48 0f 45 ea cmovne %rdx,%rbp 13: 48 85 ff test %rdi,%rdi 16: 74 10 je 0x28 18: 48 89 ee mov %rbp,%rsi 1b: e8 8d d3 ff ff callq 0xffffffffffffd3ad 20: 85 c0 test %eax,%eax 22: 0f 8e c5 00 00 00 jle 0xed 28: 48 89 eb mov %rbp,%rbx 2b:* 49 8b bd b8 00 00 00 mov 0xb8(%r13),%rdi <-- trapping instruction 32: 48 85 ff test %rdi,%rdi 35: 74 10 je 0x47 37: 48 89 ee mov %rbp,%rsi 3a: e8 6e d3 ff ff callq 0xffffffffffffd3ad 3f: 85 .byte 0x85
Code starting with the faulting instruction =========================================== 0: 49 8b bd b8 00 00 00 mov 0xb8(%r13),%rdi 7: 48 85 ff test %rdi,%rdi a: 74 10 je 0x1c c: 48 89 ee mov %rbp,%rsi f: e8 6e d3 ff ff callq 0xffffffffffffd382 14: 85 .byte 0x85
and %r13 contains 00000000019129c0 in the register snapshot. Hmm...
> Serial console is no option, but this "netconsole"-thing looks > promising. I'll set that up and try to catch the next one with this.
That'll help. Also let me make sure: Is this a vanilla kernel or are there any distribution patches on top of it? I see it says 2.6.34-gentoo but just in case. If it is not a vanilla kernel please try a vanilla one from kernel.org.
Also, please send the whole dmesg of the system, there might be something suspicious there and also the output of 'cat /proc/cpuinfo' and 'dmidecode', the last one run as root.
Thanks.
-- Regards/Gruss, Boris.
| |