From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH v2 6/6] kvm: Level IRQ de-assert for KVM_IRQFD Date: Thu, 28 Jun 2012 15:59:26 +0300 Message-ID: <4FEC552E.40409@redhat.com> References: <20120627044758.23698.249.stgit@bling.home> <20120627051013.23698.91132.stgit@bling.home> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: mst@redhat.com, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, jan.kiszka@siemens.com To: Alex Williamson Return-path: Received: from mx1.redhat.com ([209.132.183.28]:18245 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755415Ab2F1M7a (ORCPT ); Thu, 28 Jun 2012 08:59:30 -0400 In-Reply-To: <20120627051013.23698.91132.stgit@bling.home> Sender: kvm-owner@vger.kernel.org List-ID: On 06/27/2012 08:10 AM, Alex Williamson wrote: > This is an alternate level irqfd de-assert mode that's potentially > useful for emulated drivers. It's included here to show how easy it > is to implement with the new level irqfd and eoifd support. It's > possible this mode might also prove interesting for device-assignment > where we inject via level irqfd, receive an EOI (w/o de-assert), and > use the level de-assert irqfd here. This use case is racy. The guest driver will have shut down the interrupt before EOI, but with what you describe, it will fire again until the eoifd/deassertfd sequence completes. An emulated device will see the guest driver shutting down the interrupt so it's not a problem there. -- error compiling committee.c: too many arguments to function