From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:43289) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1abPYi-000263-3M for qemu-devel@nongnu.org; Thu, 03 Mar 2016 04:28:40 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1abPYe-0001v7-TC for qemu-devel@nongnu.org; Thu, 03 Mar 2016 04:28:36 -0500 Received: from mx1.redhat.com ([209.132.183.28]:43234) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1abPYe-0001v3-Mz for qemu-devel@nongnu.org; Thu, 03 Mar 2016 04:28:32 -0500 References: <1456078260-6669-1-git-send-email-davidkiarie4@gmail.com> <1456078260-6669-3-git-send-email-davidkiarie4@gmail.com> <56CA1975.1020609@web.de> <56D75263.5060404@gmail.com> From: Marcel Apfelbaum Message-ID: <56D803BC.60509@redhat.com> Date: Thu, 3 Mar 2016 11:28:28 +0200 MIME-Version: 1.0 In-Reply-To: <56D75263.5060404@gmail.com> Content-Type: text/plain; charset=iso-8859-15; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [V6 2/4] hw/core: Add AMD IOMMU to machine properties List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: David Kiarie , Jan Kiszka , qemu-devel@nongnu.org Cc: valentine.sinitsyn@gmail.com, mst@redhat.com On 03/02/2016 10:51 PM, David Kiarie wrote: > > > On 21/02/16 23:09, Jan Kiszka wrote: >> On 2016-02-21 19:10, David Kiarie wrote: >>> diff --git a/qemu-options.hx b/qemu-options.hx >>> index 2f0465e..dad160f 100644 >>> --- a/qemu-options.hx >>> +++ b/qemu-options.hx >>> @@ -38,7 +38,7 @@ DEF("machine", HAS_ARG, QEMU_OPTION_machine, \ >>> " kvm_shadow_mem=size of KVM shadow MMU\n" >>> " dump-guest-core=on|off include guest memory in a core dump (default=on)\n" >>> " mem-merge=on|off controls memory merge support (default: on)\n" >>> - " iommu=on|off controls emulated Intel IOMMU (VT-d) support (default=off)\n" >>> + " iommu=amd|intel enables and selects the emulated IOMMU (default: off)\n" >> We should also support "iommu=off" or "none" to explicitly disable it. >> That is consistent with other switches, and maybe there will once be a >> machine type (chipset) with IOMMU enabled by default. > > We could have such but this will not be referenced anywhere in the code as the IOMMU is 'off' by default. Most of the other such switches relate to properties that are 'on' by default. It doesn't matter if is not used, some management systems would prefer to pass an "off" value rather than not adding the property. Also you need a default value. Even in the description you mention it as "off". Thanks, Marcel > >> >> Jan >> >