From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:57085) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WCWUP-0007QQ-8t for qemu-devel@nongnu.org; Sun, 09 Feb 2014 10:40:20 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WCWUG-0007ap-6G for qemu-devel@nongnu.org; Sun, 09 Feb 2014 10:40:13 -0500 Received: from cantor2.suse.de ([195.135.220.15]:55007 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WCWUG-0007Xx-07 for qemu-devel@nongnu.org; Sun, 09 Feb 2014 10:40:04 -0500 Message-ID: <52F7A14F.3050906@suse.de> Date: Sun, 09 Feb 2014 16:39:59 +0100 From: =?ISO-8859-15?Q?Andreas_F=E4rber?= MIME-Version: 1.0 References: <1391877522-17254-1-git-send-email-mark.cave-ayland@ilande.co.uk> <1391877522-17254-2-git-send-email-mark.cave-ayland@ilande.co.uk> <52F79A4B.5060404@suse.de> <52F79DB2.1000401@ilande.co.uk> In-Reply-To: <52F79DB2.1000401@ilande.co.uk> Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCHv2 1/2] sun4m: Add Sun CG3 framebuffer and corresponding OpenBIOS FCode ROM List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Mark Cave-Ayland Cc: Peter Maydell , qemu-devel@nongnu.org, Blue Swirl , Bob Breuer , Anthony Liguori , Paolo Bonzini , Artyom Tarasenko Am 09.02.2014 16:24, schrieb Mark Cave-Ayland: > One thing I'm not sure about is how the QOM stuff interacts with sysbus > - can you quickly point me towards an existing device that does this so > I can understand how this works? The ARM MPCore devices come to mind. But really all you've been asked to do in that aspect is replace dc->init with dc->realize and adapt the signature to make it compile. :) A bonus would be to separate between .instance_init and dc->realize, with MemoryRegion initialization and anything that does not rely on properties going into the initfn, and everything relying on properties being specified or affecting global state (such as RAM registration, VMState registration, ...) going into the realizefn. > These patches have been around for > months and I really want to get them in for QEMU 2.0 if freeze is comin= g > up soon. Apart from Blue's absence I see no major blocker to that. :) I initially wondered if -vga wasn't considered a legacy option, but I figured we can't create SysBus/SBus devices through -device, so no alternative. SBus QOM base class refactorings as discussed for an earlier device can be applied on top and are not strictly needed for 2.0 unless you have more such devices coming up. Cheers, Andreas --=20 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=FCrnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imend=F6rffer; HRB 16746 AG N=FCrnbe= rg