From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH 1/3] KVM: x86: merge kvm_arch_set_irq with kvm_set_msi_inatomic Date: Mon, 2 Nov 2015 18:05:20 +0100 Message-ID: <563797D0.7050109@redhat.com> References: <1446466878-32837-1-git-send-email-pbonzini@redhat.com> <1446466878-32837-2-git-send-email-pbonzini@redhat.com> <20151102145923.GB24667@potion.brq.redhat.com> <56378A62.3090706@redhat.com> <20151102170156.GA28717@potion.brq.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, Alex Williamson To: Radim Krcmar Return-path: Received: from mx1.redhat.com ([209.132.183.28]:46928 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754331AbbKBRF0 (ORCPT ); Mon, 2 Nov 2015 12:05:26 -0500 In-Reply-To: <20151102170156.GA28717@potion.brq.redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 02/11/2015 18:01, Radim Krcmar wrote: >> > Yes. Both because the Virtuozzo people confirmed that kvm_arch_set_irq >> > isn't needed for synic, and because synic is currently broken with APICv. > Thanks. > > (We can add direct delivery for |online vcpus| < X if performance with > low number of VCPUs happens to regress because of the schedule_work,) Yeah, we will change the VFIO interrupt handler to non-threaded, which will do more or less the same (what you lose now through schedule_work, you will recoup by doing things directly in the ISR). Paolo