From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:39510) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1c4tWu-0004vW-UV for qemu-devel@nongnu.org; Thu, 10 Nov 2016 12:52:54 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1c4tWt-0003MI-2d for qemu-devel@nongnu.org; Thu, 10 Nov 2016 12:52:53 -0500 Received: from mail-wm0-x242.google.com ([2a00:1450:400c:c09::242]:36261) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1c4tWs-0003Lx-RT for qemu-devel@nongnu.org; Thu, 10 Nov 2016 12:52:50 -0500 Received: by mail-wm0-x242.google.com with SMTP id m203so2942669wma.3 for ; Thu, 10 Nov 2016 09:52:50 -0800 (PST) Sender: Paolo Bonzini From: Paolo Bonzini Date: Thu, 10 Nov 2016 18:52:39 +0100 Message-Id: <1478800362-18138-4-git-send-email-pbonzini@redhat.com> In-Reply-To: <1478800362-18138-1-git-send-email-pbonzini@redhat.com> References: <1478800362-18138-1-git-send-email-pbonzini@redhat.com> Subject: [Qemu-devel] [PULL 3/6] target-i386/machine: fix migrate faile because of Hyper-V HV_X64_MSR_VP_RUNTIME List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: ZhuangYanying From: ZhuangYanying Hyper-V HV_X64_MSR_VP_RUNTIME was introduced in linux-4.4 + qemu-2.5. As long as the KVM module supports, qemu will save / load the vmstate_msr_hyperv_runtime register during the migration. Regardless of whether the hyperv_runtime configuration of x86_cpu_properties is enabled. The qemu-2.3 does not support this feature, of course, failed to migrate. linux-BGSfqC:/home/qemu # ./x86_64-softmmu/qemu-system-x86_64 --enable-kvm \ -nodefaults -machine pc-i440fx-2.3,accel=kvm,usb=off -smp 4 -m 4096 -drive \ file=/work/suse/sles11sp3.img.bak,format=raw,if=none,id=drive-virtio-disk0,cache=none \ -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x5,drive=drive-virtio-disk0,id=virtio-disk0 \ -vnc :99 -device cirrus-vga,id=video0,vgamem_mb=8,bus=pci.0,addr=0x2 -monitor vc save_section_header:se->section_id=3,se->idstr:ram,se->instance_id=0,se->version_id=4 save_section_header:se->section_id=0,se->idstr:timer,se->instance_id=0,se->version_id=2 save_section_header:se->section_id=4,se->idstr:cpu_common,se->instance_id=0,se->version_id=1 save_section_header:se->section_id=5,se->idstr:cpu,se->instance_id=0,se->version_id=12 vmstate_subsection_save:vmsd->name:cpu/async_pf_msr hyperv_runtime_enable_needed:env->msr_hv_runtime=128902811 vmstate_subsection_save:vmsd->name:cpu/msr_hyperv_runtime Since hyperv_runtime is false, vm will not use hv->runtime_offset, then vmstate_msr_hyperv_runtime is no need to transfer while migrating. Signed-off-by: ann.zhuangyanying@huawei.com Message-Id: <1478247398-5016-1-git-send-email-ann.zhuangyanying@huawei.com> Signed-off-by: Paolo Bonzini --- target-i386/machine.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/target-i386/machine.c b/target-i386/machine.c index 48037f1..760f82b 100644 --- a/target-i386/machine.c +++ b/target-i386/machine.c @@ -709,6 +709,10 @@ static bool hyperv_runtime_enable_needed(void *opaque) X86CPU *cpu = opaque; CPUX86State *env = &cpu->env; + if (!cpu->hyperv_runtime) { + return false; + } + return env->msr_hv_runtime != 0; } -- 1.8.3.1