From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 2/3] KVM: Emulate MSI-X table in kernel Date: Mon, 17 Jan 2011 18:01:04 +0200 Message-ID: <4D3467C0.4000306@redhat.com> References: <1294309185-21417-1-git-send-email-sheng@linux.intel.com> <1294309185-21417-3-git-send-email-sheng@linux.intel.com> <20110117115447.GA8145@amt.cnet> <201101172018.22478.sheng@linux.intel.com> <4D3433A3.2010305@redhat.com> <20110117124816.GB8985@amt.cnet> <4D343B59.4070507@redhat.com> <20110117155254.GA12350@amt.cnet> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Sheng Yang , "Michael S. Tsirkin" , kvm@vger.kernel.org To: Marcelo Tosatti Return-path: Received: from mx1.redhat.com ([209.132.183.28]:8867 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751932Ab1AQQBJ (ORCPT ); Mon, 17 Jan 2011 11:01:09 -0500 In-Reply-To: <20110117155254.GA12350@amt.cnet> Sender: kvm-owner@vger.kernel.org List-ID: On 01/17/2011 05:52 PM, Marcelo Tosatti wrote: > > > > What a value is written by the guest, which kvm cannot handle itself > > (i.e. a change to anything other than the mask bit), we exit with > > the table and entry ids, so userspace can reread them. > > OK. But regarding access to the MSI-X entry in userspace, it can > only be accessed safely wrt parallel updates by the kernel in the > exit handler. > > Is the exit handler the only location where the MSI-X entry will be > read or written to, in userspace? > Yes. This should be documented, that the kernel owns the msix area as far as writes are concerned. -- error compiling committee.c: too many arguments to function