From mboxrd@z Thu Jan 1 00:00:00 1970 From: Radim =?utf-8?B?S3LEjW3DocWZ?= Subject: Re: [PATCH RFC] KVM: X86: save and load PVCLOCK_TSC_UNSTABLE_BIT when migration Date: Mon, 12 Jun 2017 21:52:10 +0200 Message-ID: <20170612195209.GA16151@potion> References: <1497273796-44720-1-git-send-email-jianjay.zhou@huawei.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: kvm@vger.kernel.org, qemu-devel@nongnu.org, wangxinxin.wang@huawei.com, weidong.huang@huawei.com, arei.gonglei@huawei.com To: Jay Zhou Return-path: Received: from mx1.redhat.com ([209.132.183.28]:35608 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752065AbdFLTwT (ORCPT ); Mon, 12 Jun 2017 15:52:19 -0400 Content-Disposition: inline In-Reply-To: <1497273796-44720-1-git-send-email-jianjay.zhou@huawei.com> Sender: kvm-owner@vger.kernel.org List-ID: 2017-06-12 21:23+0800, Jay Zhou: > Guest using kvmclock will be hanged when migrating from unstable > tsc host to stable tsc host occasionally. > Sometimes, the tsc timestamp saved at the source side will be > backward when the guest stopped, and this value is transferred > to the destination side. The guest at the destination side thought > kvmclock is stable, so the protection mechanism against time > going backwards is not used. > When the first time vcpu0 enters the guest at the destination > side to update the wall clock, the result of > pvclock_clocksource_read will be backward occasionally, > which results in the wall clock drift. > > Signed-off-by: Jay Zhou > --- Hm, are you using KVM that has e3fd9a93a12a (4.9+)? If you get a timestamp from KVM_GET_CLOCK() and pass that to KVM_SET_CLOCK(), then kvmclock should not jump backwards anymore (it could before 4.9, but only if the host had stable tsc). A possible source of this bug in when QEMU recomputes the timestamp to pass to KVM_SET_CLOCK() using TSC and kvmclock page. Can you provide the values used for KVM_GET_CLOCK and KVM_SET_CLOCK when the bug occurs? Thanks.