RE: kvm/ia64: Fix halt emulation logic.

2008-11-02 Thread Zhang, Xiantao
Marcelo Tosatti wrote: > Hi Xiantao, > > On Wed, Oct 15, 2008 at 09:47:24PM +0800, Zhang, Xiantao wrote: >> +expires = div64_u64(itc_diff, cyc_per_usec); >> +kt = ktime_set(0, 1000 * expires); >> + >> +down_read(&vcpu->kvm->slots_lock); >> +vcpu->arc

Re: kvm/ia64: Fix halt emulation logic.

2008-11-02 Thread Marcelo Tosatti
Hi Xiantao, On Wed, Oct 15, 2008 at 09:47:24PM +0800, Zhang, Xiantao wrote: > + expires = div64_u64(itc_diff, cyc_per_usec); > + kt = ktime_set(0, 1000 * expires); > + > + down_read(&vcpu->kvm->slots_lock); > + vcpu->arch.ht_active = 1; > +

Re: kvm/ia64: Fix halt emulation logic.

2008-10-16 Thread Avi Kivity
Zhang, Xiantao wrote: Attached the updated one. Thanks! Xiantao From 8d827b19ba03dee74edeb5e2bcc23bed7b9f38ea Mon Sep 17 00:00:00 2001 From: Xiantao Zhang <[EMAIL PROTECTED]> Date: Thu, 16 Oct 2008 15:58:15 +0800 Subject: [PATCH] kvm/ia64: Fix halt emulation logic. Common halt halding logic is

RE: kvm/ia64: Fix halt emulation logic.

2008-10-16 Thread Zhang, Xiantao
+ VMX(v, itc_check) = 0; } else { v->arch.timer_pending = 1; } -- 1.5.1 -Original Message- From: Avi Kivity [mailto:[EMAIL PROTECTED] Sent: Thursday,

Re: kvm/ia64: Fix halt emulation logic.

2008-10-16 Thread Avi Kivity
Zhang, Xiantao wrote: smp_call_function_single() will call the function on the current cpu if instructed, so this change is unneeded. Aha, Thanks for pointing this. Yes, it is unnecessary. Could you help to remove it directly ? or need me have a resend. Please resend, since I can't tes

RE: kvm/ia64: Fix halt emulation logic.

2008-10-16 Thread Zhang, Xiantao
Avi Kivity wrote: > Zhang, Xiantao wrote: >> Hi, Avi >> This is the key fix for 2.6.28 merge. Without this patch, guest >> may hang once configured with more than 2 vcpus, it is because x86 >> side changed the halt handling's common logic, but it misses to >> change ia64 side. >> > >> @@ -3

Re: kvm/ia64: Fix halt emulation logic.

2008-10-16 Thread Avi Kivity
Zhang, Xiantao wrote: Hi, Avi This is the key fix for 2.6.28 merge. Without this patch, guest may hang once configured with more than 2 vcpus, it is because x86 side changed the halt handling's common logic, but it misses to change ia64 side. @@ -398,8 +400,11 @@ static int handle_global

RE: kvm/ia64: Fix halt emulation logic.

2008-10-15 Thread Zhang, Xiantao
Forget to mention Jes had tested to boot 16vcpus, and didn't meet any issue. Sorry! :) Acked-by : Jes Sorensen <[EMAIL PROTECTED]> Zhang, Xiantao wrote: > Hi, Avi > This is the key fix for 2.6.28 merge. Without this patch, guest may > hang once configured with more than 2 vcpus, it is beca