Messages in this thread | | | Date | Wed, 4 Aug 2010 09:33:24 +0200 | From | Ingo Molnar <> | Subject | Re: oops in ioapic_write_entry |
| |
* Eric W. Biederman <ebiederm@xmission.com> wrote:
> This patch definitely does not solve Dave Airlie's problem. That was > clearly caused by a (ACPI being disabled compared to the test case) and a > bad BIOS provided MP table with an invalid apic_id, and our attempt to > lookup the ioapic and failed.
The kernel should not crash obviously, so that's a bug that has to be addressed - or failing any timely resolution the original patch which introduced the regression [if it's a regression] needs to be reverted.
Thanks,
Ingo
| |