2016-06-29 6:55 GMT+08:00 Wanpeng Li :
> 2016-06-29 4:07 GMT+08:00 Paolo Bonzini :
>>
>>
>> On 28/06/2016 20:34, yunhong jiang wrote:
>>> Paolo, thanks for reply.
>>>
>>> Which race window you are talking about? Is it the
>>> kvm_lapic_switch_to_hv_timer()? If yes, hope we will not forgot it once t
2016-06-29 4:07 GMT+08:00 Paolo Bonzini :
>
>
> On 28/06/2016 20:34, yunhong jiang wrote:
>> Paolo, thanks for reply.
>>
>> Which race window you are talking about? Is it the
>> kvm_lapic_switch_to_hv_timer()? If yes, hope we will not forgot it once the
>> lapic timer is not pinned anymore in futur
On 28/06/2016 20:34, yunhong jiang wrote:
> Paolo, thanks for reply.
>
> Which race window you are talking about? Is it the
> kvm_lapic_switch_to_hv_timer()? If yes, hope we will not forgot it once the
> lapic timer is not pinned anymore in future.
Yes, it's that one. This is a good point agai
On Tue, 28 Jun 2016 13:56:38 -0400 (EDT)
Paolo Bonzini wrote:
>
>
> > > diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
> > > index fdc05ae..b15e32a 100644
> > > --- a/arch/x86/kvm/lapic.c
> > > +++ b/arch/x86/kvm/lapic.c
> > > @@ -1454,11 +1454,18 @@ static void start_apic_timer(struc
> > diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
> > index fdc05ae..b15e32a 100644
> > --- a/arch/x86/kvm/lapic.c
> > +++ b/arch/x86/kvm/lapic.c
> > @@ -1454,11 +1454,18 @@ static void start_apic_timer(struct kvm_lapic
> > *apic) /* lapic timer in tsc deadline mode */
> >
On Tue, 28 Jun 2016 14:54:19 +0800
Wanpeng Li wrote:
> INFO: rcu_sched detected stalls on CPUs/tasks:
> 1-...: (11800 GPs behind) idle=45d/140/0 softirq=0/0
> fqs=21663 (detected by 0, t=65016 jiffies, g=11500, c=11499, q=719)
> Task dump for CPU 1:
> qemu-system-x86 R running task
2016-06-28 16:43 GMT+08:00 Paolo Bonzini :
>
>
> On 28/06/2016 08:15, Wanpeng Li wrote:
>> 2016-06-27 21:11 GMT+08:00 Paolo Bonzini :
>>> If the TSC deadline timer is programmed really close to the deadline or
>>> even in the past, the computation in vmx_set_hv_timer can underflow and
>>> cause del
On 28/06/2016 08:15, Wanpeng Li wrote:
> 2016-06-27 21:11 GMT+08:00 Paolo Bonzini :
>> If the TSC deadline timer is programmed really close to the deadline or
>> even in the past, the computation in vmx_set_hv_timer can underflow and
>> cause delta_tsc to be set to a huge value. This generally r
INFO: rcu_sched detected stalls on CPUs/tasks:
1-...: (11800 GPs behind) idle=45d/140/0 softirq=0/0 fqs=21663
(detected by 0, t=65016 jiffies, g=11500, c=11499, q=719)
Task dump for CPU 1:
qemu-system-x86 R running task0 3529 3525 0x00080808
8802021791a0 88021289
2016-06-27 21:11 GMT+08:00 Paolo Bonzini :
> If the TSC deadline timer is programmed really close to the deadline or
> even in the past, the computation in vmx_set_hv_timer can underflow and
> cause delta_tsc to be set to a huge value. This generally results
> in vmx_set_hv_timer returning -ERANGE
If the TSC deadline timer is programmed really close to the deadline or
even in the past, the computation in vmx_set_hv_timer can underflow and
cause delta_tsc to be set to a huge value. This generally results
in vmx_set_hv_timer returning -ERANGE, but we can fix it by limiting
delta_tsc to be pos
11 matches
Mail list logo