Re: [patch 4/8] KVM: x86: replace hrtimer based timer emulation

From: Avi Kivity <avi@redhat.com>
To: Marcelo Tosatti <mtosatti@redhat.com>
Cc: kvm@vger.kernel.org
Subject: Re: [patch 4/8] KVM: x86: replace hrtimer based timer emulation
Date: Wed, 08 Jul 2009 19:36:42 +0300	[thread overview]
Message-ID: <4A54CB1A.7090900@redhat.com> (raw)
In-Reply-To: <20090708162404.GA6803@amt.cnet>

On 07/08/2009 07:24 PM, Marcelo Tosatti wrote:
>> I wonder if we're really winning with this.  Guests should be
>> scheduled-out-but-not-halted rarely, and in all other cases we need to
>> keep the timer.  A timer comparison on each guest entry might be
>> expensive (maybe not so much with tsc based timers).
>>      
>
> Any activity outside of guest mode that takes more than the period of
> the timer (think 1000Hz) causes unnecessary host load.
>
> Booting a RHEL5 UP without VNC or serial output on an idle host:
>
> timer_int_normal=95416 timer_interrupt_accumulated=873
>
> (and it continues to increase in that rate, roughly 1%).
>
> Now factor in multiple guests, loaded host, and you'll probably see more
> than that.
>    

If you're using qcow, you may be seeing the non-aio accesses.  
Otherwise, I can't think what can cause 1ms latency.  Random host 
process?  long mmu resync?

-- 
error compiling committee.c: too many arguments to function