From mboxrd@z Thu Jan  1 00:00:00 1970
From: Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH v2] KVM: x86: Inject pending interrupt even if pending nmi
 exist
Date: Wed, 23 Mar 2016 19:22:21 +0100
Message-ID: <56F2DEDD.5060600@redhat.com>
References: <7CC7D3013F4B10479273CE019E6A8C7A3ED90C@BPXM15GP.gisp.nec.co.jp>
 <56F297A7.1030309@redhat.com> <20160323172159.GA22164@potion.brq.redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Yuki Shibuya <shibuya.yk@ncos.nec.co.jp>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	Jan Kiszka <jan.kiszka@siemens.com>,
	Nobuo Yoshida <yoshida.nb@ncos.nec.co.jp>
To: =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mail-wm0-f68.google.com ([74.125.82.68]:32955 "EHLO
	mail-wm0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751430AbcCWSWY (ORCPT <rfc822;kvm@vger.kernel.org>);
	Wed, 23 Mar 2016 14:22:24 -0400
Received: by mail-wm0-f68.google.com with SMTP id u125so6274691wmg.0
        for <kvm@vger.kernel.org>; Wed, 23 Mar 2016 11:22:24 -0700 (PDT)
In-Reply-To: <20160323172159.GA22164@potion.brq.redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>



On 23/03/2016 18:21, Radim Kr=C4=8Dm=C3=A1=C5=99 wrote:
> > > +		 * nmi pending counter is cleared to prevent skipping
> > > +		 * injectable pending interrupts.
> > > +		 */
> > > +		if (vcpu->arch.nmi_pending && kvm_cpu_has_injectable_intr(vcpu=
)
> > > +					&& kvm_x86_ops->interrupt_allowed(vcpu))
> > > +			vcpu->arch.nmi_pending =3D 0;
> >=20
> > I am not sure I understand this.  Why is it safe to drop nmi_pendin=
g?
>=20
> NMIs are latched (queue length 1) and therefore cannot be pending aft=
er
> an injection.  I think we want to do it unconditionally.

If that is right, process_nmi would be the place where you'd limit the
queue to 1.

Paolo