Messages in this thread | | | Subject | Re: [patch] voluntary-preempt-2.6.8.1-P0 | From | Peter Zijlstra <> | Date | Sun, 15 Aug 2004 20:29:38 +0200 |
| |
On Sun, 2004-08-15 at 13:33 -0400, Lee Revell wrote: > On Sun, 2004-08-15 at 10:01, Peter Zijlstra wrote: > > > It still locks up hard for me when voluntary-preempt=3, however it does > > finish the boot; dmesg attached. The lockup occurs several minutes into > > use; usually by by the time I've started X, launched evolution and > > selected my first imap folder the machine's dead. > > > > If you need more information or want me to try some patches, just let me > > know. > > These look a bit worrisome: > > Aug 15 15:24:11 twins kernel: Total of 2 processors activated (5537.79 BogoMIPS). > Aug 15 15:24:11 twins kernel: WARNING: This combination of AMD processors is not suitable for SMP. > Aug 15 15:24:11 twins kernel: ENABLING IO-APIC IRQs > > Aug 15 15:24:11 twins kernel: I/O APIC: AMD Errata #22 may be present. In the event of instability try > Aug 15 15:24:11 twins kernel: : booting with the "noapic" option. >
Yeah, true. I run with 2 Thunderbirds, not Athlon MPs; AMD does not like that, the CPUs don't share that inclination.
As for the AMD Errata #22, all other kernels run rock solid without noapic, so I don't see why these should differ. I did try noapic though, not on P0 but on one of the later Os.
-- Peter Zijlstra <a.p.zijlstra@chello.nl>
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |