Re: [patch 1/2] KVM: hlt emulation should take in-kernel APIC/PIT timers into account

[Avi Kivity <avi@qumranet.com>]

Marcelo Tosatti wrote:
> Timers that fire between guest hlt and vcpu_block's add_wait_queue() are 
> ignored, possibly resulting in hangs.
>
> Also make sure that atomic_inc and waitqueue_active tests happen in the
> specified order, otherwise the following race is open:
>
> CPU0                                        CPU1
>                                             if (waitqueue_active(wq))
> add_wait_queue()                        
> if (!atomic_read(pit_timer->pending))
>     schedule()
>                                             atomic_inc(pit_timer->pending)
>
> Which is not an issue for the APIC timer due to migration logic.
>
>   

Nasty.  I hope we can get Dor's interrupt injection notification 
working, so we don't have to handle these bugs.

> Index: kvm/virt/kvm/kvm_main.c
> ===================================================================
> --- kvm.orig/virt/kvm/kvm_main.c
> +++ kvm/virt/kvm/kvm_main.c
> @@ -752,6 +752,7 @@ void mark_page_dirty(struct kvm *kvm, gf
>  	}
>  }
>  
> +#ifdef CONFIG_X86
>  /*
>   * The vCPU has executed a HLT instruction with in-kernel mode enabled.
>   */
> @@ -765,6 +766,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcp
>   


This breaks ia64 (and shouldn't s390 use this too?)
>  	 * We will block until either an interrupt or a signal wakes us up
>  	 */
>  	while (!kvm_cpu_has_interrupt(vcpu)
> +	       && !kvm_cpu_has_pending_timer(vcpu)
>   

I guess the fix is to stub this out for the other archs.

-- 
Any sufficiently difficult bug is indistinguishable from a feature.


-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone