From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: [RFC][PATCH] KVM: Introduce direct MSI message injection for in-kernel irqchips Date: Mon, 24 Oct 2011 18:10:28 +0200 Message-ID: <4EA58DF4.2050100@siemens.com> References: <4EA13917.7070401@siemens.com> <4EA533B8.4040407@redhat.com> <4EA53BBF.2010704@siemens.com> <4EA54759.902@redhat.com> <4EA554B0.7030808@siemens.com> <20111024124355.GA28822@redhat.com> <4EA563FD.5060308@siemens.com> <4EA56B99.2030201@siemens.com> <20111024144039.GB29886@redhat.com> <4EA57D8B.7020905@siemens.com> <20111024160526.GA30385@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Avi Kivity , Marcelo Tosatti , kvm To: "Michael S. Tsirkin" Return-path: Received: from david.siemens.de ([192.35.17.14]:15773 "EHLO david.siemens.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932769Ab1JXQKc (ORCPT ); Mon, 24 Oct 2011 12:10:32 -0400 In-Reply-To: <20111024160526.GA30385@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 2011-10-24 18:05, Michael S. Tsirkin wrote: >> This is what I have in mind: >> - devices set PBA bit if MSI message cannot be sent due to mask (*) >> - core checks&clears PBA bit on unmask, injects message if bit was set >> - devices clear PBA bit if message reason is resolved before unmask (*) > > OK, but practically, when exactly does the device clear PBA? Consider a network adapter that signals messages in a RX ring: If the corresponding vector is masked while the guest empties the ring, I strongly assume that the device is supposed to take back the pending bit in that case so that there is no interrupt inject on a later vector unmask operation. Jan -- Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux