From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH v2 4/5] KVM: nVMX: Deliver missed nested-PI notification-vector via self-IPI while interrupts disabled Date: Mon, 11 Dec 2017 23:57:07 +0100 Message-ID: <3db35a7f-47a8-fa4b-c466-10e46eb3c323@redhat.com> References: <1512461786-6465-1-git-send-email-liran.alon@oracle.com> <1512461786-6465-5-git-send-email-liran.alon@oracle.com> <20171206204552.GD3644@flask> <5A292725.1090207@ORACLE.COM> <20171207162638.GA29615@flask> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Cc: kvm@vger.kernel.org, jmattson@google.com, wanpeng.li@hotmail.com, idan.brown@ORACLE.COM, Krish Sadhukhan , Konrad Rzeszutek Wilk To: =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= , Liran Alon Return-path: Received: from mx1.redhat.com ([209.132.183.28]:44312 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752318AbdLKW5L (ORCPT ); Mon, 11 Dec 2017 17:57:11 -0500 In-Reply-To: <20171207162638.GA29615@flask> Content-Language: en-US Sender: kvm-owner@vger.kernel.org List-ID: On 07/12/2017 17:26, Radim Krčmář wrote: > > Keeping the 'kvm_lapic_enabled(vcpu) && vcpu->arch.apicv_active' for > both is fine by me, I just didn't see a reason to check for presence of > sync_pir_to_irr and complete_nested_posted_interrupt, because if they > are NULL, then vcpu->arch.apicv_active must be false. Actually AMD doesn't have them, does it? But adding dummies would be fine for me. Thanks, Paolo