From mboxrd@z Thu Jan  1 00:00:00 1970
From: Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: A problem about optimize the kvm_vcpu_on_spin
Date: Mon, 14 Nov 2016 03:56:15 -0500 (EST)
Message-ID: <822509139.12568024.1479113775757.JavaMail.zimbra@redhat.com>
References: <58291719.4020401@huawei.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8BIT
Cc: Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        "Wubin (H)" <wu.wubin@huawei.com>,
        zhaoshenglong <zhaoshenglong@huawei.com>,
        kvm <kvm@vger.kernel.org>
To: "Longpeng (Mike)" <longpeng2@huawei.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx6-phx2.redhat.com ([209.132.183.39]:49225 "EHLO
        mx6-phx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752596AbcKNI41 (ORCPT <rfc822;kvm@vger.kernel.org>);
        Mon, 14 Nov 2016 03:56:27 -0500
In-Reply-To: <58291719.4020401@huawei.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>


> Intel sdm vol3 ch-25.1.3 (p-1074) said: The “PAUSE-loop exiting” VM-execution
> control is ignored if CPL > 0.
> 
> Does this means that only guest's kernelspace spinlock could causes VMexit
> due to "pause-loop exiting" ?
> 
> If so, I think the kvm_vcpu_on_spin should selects a (currently not
> running) vcpu which cpl=0 to boost, because the cpl>0 vcpus are not
> the lock-holder. (suppose the guest isn't RT-linux)
> 
> What about your opinion ?

Yes, that's a good idea!

Paolo