svn commit: r199067 - in head/sys: amd64/amd64 i386/i386
kuriyama at FreeBSD.org
Tue Nov 10 06:11:19 UTC 2009
At Tue, 10 Nov 2009 06:59:11 +0100,
ed at 80386.nl wrote:
> > (1) On MacBook, lock up occured on real hardware? Not on hypervisor?
> Nope. The MacBook does use Apple's Boot Camp, but I can't say whether
> that can be considered to be a hypervisor. What happens, is that the
> system just locks up after probing all the device. It doesn't mount the
> root device anymore.
Hmm, I think Boot Camp is not hypervisor, and should be treated as
real hardware. Which architecture (i386/amd64) you are using?
> > (2) On both system, is there no problem before this commit?
> I bisected it and the revision before this one does boot. I'll
> experiment with the tunables this afternoon. I really had to get some
> sleep last night, so I couldn't investigate it any further.
You can try with hw.clflush_disable="1" in loader.conf. If you can
boot with it, I must be mistaken something (anyway its my fault).
Can you show me CPUID line at boot message (of course from the kernel
you can boot) like this?
> CPU: Dual-Core AMD Opteron(tm) Processor 2210 (1800.07-MHz K8-class CPU)
> Origin = "AuthenticAMD" Id = 0x40f12 Stepping = 2
> AMD Features=0xea500800<SYSCALL,NX,MMX+,FFXSR,RDTSCP,LM,3DNow!+,3DNow!>
Jun Kuriyama <kuriyama at FreeBSD.org> // FreeBSD Project
<kuriyama at s2factory.co.jp> // S2 Factory, Inc.
More information about the svn-src-all