From mboxrd@z Thu Jan 1 00:00:00 1970 From: Juan Quintela Subject: KVM call minutes for 2012-10-16 Date: Tue, 16 Oct 2012 16:50:47 +0200 Message-ID: <878vb6mpx4.fsf@elfo.mitica> Reply-To: quintela@redhat.com Mime-Version: 1.0 Content-Type: text/plain To: KVM devel mailing list , qemu-devel@nongnu.org, Alexander Graf , Anthony Liguori Return-path: Received: from mx1.redhat.com ([209.132.183.28]:48792 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752388Ab2JPOu4 (ORCPT ); Tue, 16 Oct 2012 10:50:56 -0400 Sender: kvm-owner@vger.kernel.org List-ID: 2012-10-16 ---------- - cpu as dev making qdev available to all parts of qemu is the best solution (aliguory) how does linux-user people think about it? Eduardo will follow Objections: * conceptually it makes no sense to have devices on *-user * desire of linux-user to maintain the code relativelly small * really they want less dependencies Who is going to merge it? either anthony or andreas through cpu tree - boot and -bootindex (agraf) should this overwrote CMOS on PC (anthony) Always writing nvram makes other things work. -boot c : means boot with legacy device with uefi there is a boot variable -boot c don't make sense out of x86 * make the list of bootable devices visible to the user and store in nvram? At this point I got lost, and Anthony/Alex needs to fill the details of the proposals. - Command line compatibility patches (marcelo) merge then directly or through qdev