From mboxrd@z Thu Jan  1 00:00:00 1970
From: Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: Seeking a KVM benchmark
Date: Wed, 12 Nov 2014 12:56:40 -0500 (EST)
Message-ID: <1257793896.944291.1415815000290.JavaMail.zimbra@redhat.com>
References: <CALCETrXG4WvJWfjsnH7AmXHDKjt+kmiQqvG4X2VZz+g_fARTLA@mail.gmail.com> <20141110173835.GA26770@minantech.com> <5463458C.2070000@redhat.com> <20141112152204.GC26187@minantech.com> <54637C25.5000305@redhat.com> <20141112153230.GD26187@minantech.com> <5463821C.2020900@redhat.com> <CALCETrUOSHqhwZRGG-C4xUNWT5pq5QuPynXykZKEfW8ktbGGdQ@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Cc: Gleb Natapov <gleb@kernel.org>, Avi Kivity <avi.kivity@gmail.com>,
	kvm list <kvm@vger.kernel.org>
To: Andy Lutomirski <luto@amacapital.net>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx5-phx2.redhat.com ([209.132.183.37]:57988 "EHLO
	mx5-phx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751685AbaKLR4o (ORCPT <rfc822;kvm@vger.kernel.org>);
	Wed, 12 Nov 2014 12:56:44 -0500
In-Reply-To: <CALCETrUOSHqhwZRGG-C4xUNWT5pq5QuPynXykZKEfW8ktbGGdQ@mail.gmail.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

> Assuming you're running both of my patches (LOAD_EFER regardless of
> nx, but skip LOAD_EFER of guest == host), then some of the speedup may
> be just less code running.  I haven't figured out exactly when
> vmx_save_host_state runs, but my patches avoid a call to
> kvm_set_shared_msr, which is worth a few cycles.

Yes, that's possible.  vmx_save_host_state is here:

        preempt_disable();

        kvm_x86_ops->prepare_guest_switch(vcpu);  // <<<<
        if (vcpu->fpu_active)
                kvm_load_guest_fpu(vcpu);
        kvm_load_guest_xcr0(vcpu);

        vcpu->mode = IN_GUEST_MODE;

        srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);

and it's a fairly hot function.

Paolo