Hello, I compiled Xen unstable changeset 13568 (PAE) and encountered some problems. Test machine: IBM ThinkPad T60, latest BIOS, running FC6. 1. xm dmesg shows the following errors: * APIC error on CPU0: 00(40) * APIC error on CPU0: 00(40) * platform_hypercall.c:142: Domain 0 says that IO-APIC REGSEL is good * irq.c:438:d0 Cannot bind IRQ 2 to guest. In use by ''cascade''. 2. The console keeps showing the following message: "Timer ISR/[0,1]: Time went backwards: delta=NEGATIVE_NUMBER ..." (producted in time-xen.c (kernel sparse) line:651) Seems like a sync problem... 3. The system behaves like it misses interrupts, the mouse and keyboard have some lags... I guess that this happens because of the new kernel (2.6.18), I didn''t experience these problems in kernel 2.6.16.33 4. xend.log shows alot of XenAPI functions that are missing... for example, "API call: VIF.get_io_read_bks not found". Any idea on how to fix it? Btw: I compiled the same changeset/kernel on an Intel vPro computer and it worked perfectly. Thanks in advance, Guy Zana. ____________ Team Leader, Neocleus, Inc. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Sun, Jan 28, 2007 at 08:41:06AM -0500, Guy Zana wrote:> 4. xend.log shows alot of XenAPI functions that are missing... for > example, "API call: VIF.get_io_read_bks not found".These are nothing to worry about, and not the cause of your other problems. Ewan. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Is it a dual core T60? The APIC errors are concerning that particular error (illegal vector) is most often a sign of bad hardware. If it¹s a UP laptop then you might try booting with nolapic noapic nosmp¹ as a set of extra Xen boot parameters. If it¹s SMP then the APIC errors are harder to explain away. Do you know when during boot the APIC errors appear (e.g., before or after dom0 starts to run)? -- Keir On 28/1/07 1:41 pm, "Guy Zana" <guy@neocleus.com> wrote:> Hello, I compiled Xen unstable changeset 13568 (PAE) and encountered some > problems. > Test machine: IBM ThinkPad T60, latest BIOS, running FC6. > > 1. xm dmesg shows the following errors: > * APIC error on CPU0: 00(40) > * APIC error on CPU0: 00(40)_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Yep. It''s a Dual Core T60 (VT is enabled). The APIC error is really after dom0 starts to run, the error appears after the message about relinquishing the console. The problems didn''t appear in changeset 13090 of the unstable tree (non PAE mode). 1. Can it be related to PAE? (hard to see how) 2. Can I use the old kernel (2.6.16.33) with changeset 13568? I didn''t have much problems with that version of the kernel, although I don''t want to mess with inconsistencies... Should I check the latest changeset? Guy. ________________________________ From: Keir Fraser [mailto:Keir.Fraser@cl.cam.ac.uk] Sent: Sunday, January 28, 2007 3:59 PM To: Guy Zana; xen-devel@lists.xensource.com Subject: Re: [Xen-devel] timer_interrupt problem Is it a dual core T60? The APIC errors are concerning - that particular error (illegal vector) is most often a sign of bad hardware. If it''s a UP laptop then you might try booting with ''nolapic noapic nosmp'' as a set of extra Xen boot parameters. If it''s SMP then the APIC errors are harder to explain away. Do you know when during boot the APIC errors appear (e.g., before or after dom0 starts to run)? -- Keir On 28/1/07 1:41 pm, "Guy Zana" <guy@neocleus.com> wrote: Hello, I compiled Xen unstable changeset 13568 (PAE) and encountered some problems. Test machine: IBM ThinkPad T60, latest BIOS, running FC6. 1. xm dmesg shows the following errors: * APIC error on CPU0: 00(40) * APIC error on CPU0: 00(40) _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On 28/1/07 2:27 pm, "Guy Zana" <guy@neocleus.com> wrote:> 1. Can it be related to PAE? (hard to see how) > 2. Can I use the old kernel (2.6.16.33) with changeset 13568? I didn''t have > much problems with that version of the kernel, although I don''t want to mess > with inconsistencies... > > Should I check the latest changeset?Yes, you can run old dom0 kernel with new Xen. That is worth trying. -- Keir _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Hi,
I actually tested changeset 13568 with kernel 2.6.16.33 and it still gives me
that error message ("APIC error on CPUx").
At the other hand, the interrupt lags have disappeared. Tell me if I can assist
you any further with this matter.
Thanks for your help.
Guy.
________________________________
From: Keir Fraser [mailto:Keir.Fraser@cl.cam.ac.uk]
Sent: Sunday, January 28, 2007 4:41 PM
To: Guy Zana; xen-devel@lists.xensource.com
Subject: Re: [Xen-devel] timer_interrupt problem
On 28/1/07 2:27 pm, "Guy Zana" <guy@neocleus.com> wrote:
1. Can it be related to PAE? (hard to see how)
2. Can I use the old kernel (2.6.16.33) with changeset 13568? I
didn''t have much problems with that version of the kernel, although I
don''t want to mess with inconsistencies...
Should I check the latest changeset?
Yes, you can run old dom0 kernel with new Xen. That is worth trying.
-- Keir
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel