2.6.0-test3-mm2 - failure to panic

2.6.0-test3-mm2 - failure to panic

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 2966 bytes --]

Last night, I left my laptop running all night, and find this on the terminals this morning:

Message from syslogd@turing-police at Mon Aug 18 04:01:01 2003 ...
turing-police kernel: In idle task - not syncing

So I go digging in the syslogs and find:

Aug 18 04:00:59 turing-police kernel: Unable to handle kernel paging request at virtual address fff95fb8
Aug 18 04:00:59 turing-police kernel:  printing eip:
Aug 18 04:00:59 turing-police kernel: c0238135
Aug 18 04:00:59 turing-police kernel: *pde = 00001067
Aug 18 04:00:59 turing-police kernel: *pte = 00000000
Aug 18 04:00:59 turing-police kernel: Oops: 0000 [#1]
Aug 18 04:00:59 turing-police kernel: PREEMPT
Aug 18 04:00:59 turing-police kernel: CPU:    0
Aug 18 04:01:00 turing-police kernel: EIP:    0060:[acpi_processor_idle+371/485]    Tainted: P   VLI
Aug 18 04:01:00 turing-police kernel: EFLAGS: 00010206
Aug 18 04:01:00 turing-police kernel: EIP is at acpi_processor_idle+0x173/0x1e5
Aug 18 04:01:00 turing-police kernel: eax: 00000a14   ebx: 00000a10   ecx: 00000000   edx: 00000808
Aug 18 04:01:00 turing-police kernel: esi: c138eab0   edi: c138ea00   ebp: fff95fcc   esp: c04b5fb4
Aug 18 04:01:00 turing-police kernel: ds: 007b   es: 007b   ss: 0068
Aug 18 04:01:00 turing-police kernel: Process swapper (pid: 0, threadinfo=c04b4000 task=c04279c0)
Aug 18 04:01:00 turing-police kernel: Stack: c0237fc2 c138ea00 c023007b c04b4000 0009b800 c0107000 c04b5fd8 c010ace7
Aug 18 04:01:00 turing-police kernel:        0001080e c04b5ff8 c04b66f0 c04279c0 00000000 c04db0c8 00000005 c04b644a
Aug 18 04:01:00 turing-police kernel:        c04e4240 0008e000 c010017e
Aug 18 04:01:00 turing-police kernel: Call Trace:
Aug 18 04:01:00 turing-police kernel:  [acpi_processor_idle+0/485] acpi_processor_idle+0x0/0x1e5
Aug 18 04:01:00 turing-police kernel:  [acpi_tb_build_common_facs+8/181] acpi_tb_build_common_facs+0x8/0xb5
Aug 18 04:01:00 turing-police kernel:  [rest_init+0/96] _stext+0x0/0x60
Aug 18 04:01:00 turing-police kernel:  [cpu_idle+47/56] cpu_idle+0x2f/0x38
Aug 18 04:01:01 turing-police kernel:  [start_kernel+379/416] start_kernel+0x17b/0x1a0
Aug 18 04:01:01 turing-police kernel:  [unknown_bootoption+0/248] unknown_bootoption+0x0/0xf8
Aug 18 04:01:01 turing-police kernel:
Aug 18 04:01:01 turing-police kernel: Code: f6 05 71 b7 4f c0 01 75 0f 29 fb 8d 83 ff ff ff 00 25 ff ff ff 00 eb 05 29 fb 8d 43 ff 2b 46 0c 8d 58 fc eb 03 fb eb 6d 8b 4e 1c <8b> 7d ec 85 c9 8b 57 14 74 2f 3b 5e 24 76 2a 8b 46 18 c7 46 30
Aug 18 04:01:01 turing-police kernel:  <0>Kernel panic: Attempted to kill the idle task!
Aug 18 04:01:01 turing-police kernel: In idle task - not syncing

However, we somehow managed to survive the call to panic():

% date
Mon Aug 18 09:16:06 EDT 2003
% uptime
 09:16:09  up 2 days, 11:09,  4 users,  load average: 0.11, 0.23, 0.18
 
Leaving the two questions:

1) Why did acpi_processor_idle() get a bad pointer?
2) Why wasn't the laptop dead as a doornail this morning?

Any ideas?

[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]