From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: KVM: x86: do not execute guest code if vcpu not runnable
Date: Sat, 19 Jul 2008 20:41:04 -0300
Message-ID: <20080719234104.GA10435@dmt.cnet>
References: <20080719181143.GA27283@dmt.cnet>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: kvm-devel <kvm@vger.kernel.org>
To: Avi Kivity <avi@qumranet.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([66.187.233.31]:46404 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1753972AbYGSXle (ORCPT <rfc822;kvm@vger.kernel.org>);
	Sat, 19 Jul 2008 19:41:34 -0400
Content-Disposition: inline
In-Reply-To: <20080719181143.GA27283@dmt.cnet>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Sat, Jul 19, 2008 at 03:11:44PM -0300, Marcelo Tosatti wrote:
> 
> If a vcpu has been offlined, or not initialized at all, signals
> requesting userspace work to be performed will result in KVM attempting
> to re-entry guest mode.
> 
> Problem is that the in-kernel irqchip emulation happily executes HALTED
> state vcpu's. This breaks "savevm" on Windows SMP installation, for
> example.
> 
> Also, setting the mpstate to runnable when setting an IRQ via
> pic_irq_request / kvm_apic_set_irq is necessary due to this check on
> emulate_halt:
> 
>                 kvm_vcpu_block(vcpu);
>                 down_read(&vcpu->kvm->slots_lock);
>                 if (vcpu->arch.mp_state != KVM_MP_STATE_RUNNABLE)
>                         return -EINTR;
> 
> Otherwise the vcpu will return to userspace needlessly.

Scratch this, burns 100% CPU.