From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexander Graf Subject: Re: [PATCH 0/3] Introduce per-vm kvm device attributes Date: Thu, 10 Apr 2014 13:37:54 +0200 Message-ID: <53468292.5080200@suse.de> References: <1397128604-50161-1-git-send-email-borntraeger@de.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Paolo Bonzini , KVM , linux-s390 , Cornelia Huck , Dominik Dingel To: Christian Borntraeger , Marcelo Tosatti , Gleb Natapov Return-path: Received: from cantor2.suse.de ([195.135.220.15]:47834 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S935486AbaDJLh4 (ORCPT ); Thu, 10 Apr 2014 07:37:56 -0400 In-Reply-To: <1397128604-50161-1-git-send-email-borntraeger@de.ibm.com> Sender: kvm-owner@vger.kernel.org List-ID: On 10.04.14 13:16, Christian Borntraeger wrote: > Marcelo, Gleb, Alex, > > this is a followup to the recent discussion (re config device) > > To avoid a bunch of new ioctl, lets reuse the device attributes on > the "VM device". All of these attributes are architecture specific. > Patch1 introduces the interface, > Patch2 makes CMMA configurable > Patch3 uses the interface introduced in patch1 > > Marcelo, Gleb, Alex, are you ok with the approach of patch 1? Then we > will continue using that for VM specific controls instead of adding > new ioctls. (Otherwise we will continue with new ioctl). Looks reasonable to me, though I wouldn't be opposed to ENABLE_CAP + specific ioctl for reset in this case either. Whatever works better for you guys. Acked-by: Alexander Graf Alex