From mboxrd@z Thu Jan 1 00:00:00 1970 From: Raghavendra K T Subject: Re: [PATCH RFC V5 3/6] kvm : Add unhalt msr to aid (live) migration Date: Tue, 17 Apr 2012 12:47:17 +0530 Message-ID: <4F8D18FD.2040801__3614.35796735368$1334647083$gmane$org@linux.vnet.ibm.com> References: <20120323080503.14568.43092.sendpatchset@codeblue> <20120323080723.14568.23068.sendpatchset@codeblue> <20120412001517.GB32051@amt.cnet> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20120412001517.GB32051@amt.cnet> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: Marcelo Tosatti , Avi Kivity Cc: Jeremy Fitzhardinge , X86 , KVM , Konrad Rzeszutek Wilk , Greg Kroah-Hartman , linux-doc@vger.kernel.org, LKML , Ingo Molnar , Srivatsa Vaddagiri , Sasha Levin , "H. Peter Anvin" , Virtualization , Xen , Stefano Stabellini List-Id: virtualization@lists.linuxfoundation.org On 04/12/2012 05:45 AM, Marcelo Tosatti wrote: > On Fri, Mar 23, 2012 at 01:37:26PM +0530, Raghavendra K T wrote: >> From: Raghavendra K T >> [...] > > Unless there is a reason to use an MSR, should use a normal ioctl > such as KVM_{GET,SET}_MP_STATE. > > I agree with you. In the current implementation, since we are not doing any communication between host/guest (on this flag), I too felt MSR is an overkill for this. IMO, patch like below should do the job, which I am planning to include in next version of patch. Let me know if you foresee any side-effects. --- diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index aa44292..5c81a66 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -5691,7 +5691,9 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, struct kvm_mp_state *mp_state) { - mp_state->mp_state = vcpu->arch.mp_state; + mp_state->mp_state = (vcpu->arch.mp_state == KVM_MP_STATE_HALTED && + vcpu->pv_unhalted)? + KVM_MP_STATE_RUNNABLE : vcpu->arch.mp_state; return 0; }