APIC Oops on 2.6.19.1

APIC Oops on 2.6.19.1

[Antoine Martin]

As Matt Mackall said:
"So yes, if a user reports a bug that's attributable to a single bit 
memory error that's otherwise unreproduced and unexplained, it's totally 
reasonable to chalk it up to cosmic rays until some sort of pattern of 
reports emerges."

So I guess that the only way to figure out if this is indeed a one-off 
cosmic ray is to post it somewhere public in case someone else sees it?
As there is no APIC mailing list, I am posting to LKML - sorry for the 
line noise, feel free to tell me to post elsewhere (/dev/null?).


Here it is: on a Dual-Opteron Tyan board which is rebooting every hour 
to run some unit tests, I caught this -only once- at boot (partially 
copied by hand):

Pid: 1, comm: swapper Not tainted 2.6.19.1 #1
RIP: 0010:[<ffffffff80272dba>] [<ffffffff80272dba>] 
setup_APIC_timer+0x1e/0xba
RSP: 0000:ffff81007ffa7ec0  EFLAGS: 00000002
RAX: ....


CS: 0010 DS: 0018 ES: 0018 CR0: 0000000000008005003b
CR2: 000000000000000 CR3: 0000000000201000 CR4: 0000000000006e0
Process swapper: (pid: 1, threadinfo fffff81007ffa6000, task 
ffff8100023937a0)
Stack: 000000000be41ca0 ffffffffff806491bc 0000000000000000 000040b400090000
  000000000008e000 000000000090000 0000000000008e000 ffffffffffff80267297
  0000000000000000 fffffff80546280 fffffffff80261a61 00000000000000000000
Call Trace:
  [<ffffffff806491bc>] setup_boot_APIC_clock+0x115/0x11d
  [<ffffffff80267297>] init+0x48/0x306
  [<ffffffff8025bed8>] child_rip+0xa/0x12


Code: 8b 04 25 f0 e0 5f ff 39 d0 73 f5 8b 04 25 f0 e0 5f ff 39 d0
  <0>Kernel panic - not syncing: Attempting to kill init!