From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MLezi-0003gJ-Ly for qemu-devel@nongnu.org; Tue, 30 Jun 2009 11:11:38 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MLezd-0003fz-AM for qemu-devel@nongnu.org; Tue, 30 Jun 2009 11:11:37 -0400 Received: from [199.232.76.173] (port=46097 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MLezd-0003fw-3o for qemu-devel@nongnu.org; Tue, 30 Jun 2009 11:11:33 -0400 Received: from mx20.gnu.org ([199.232.41.8]:25579) by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1MLezc-0005xb-RZ for qemu-devel@nongnu.org; Tue, 30 Jun 2009 11:11:32 -0400 Received: from mail.codesourcery.com ([65.74.133.4]) by mx20.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MLezb-00078n-Ud for qemu-devel@nongnu.org; Tue, 30 Jun 2009 11:11:32 -0400 From: Paul Brook Subject: Re: [Qemu-devel] Re: [RfC PATCH] qdev: rework device properties. Date: Tue, 30 Jun 2009 16:11:27 +0100 References: <1246361503-18254-1-git-send-email-kraxel@redhat.com> <200906301515.45017.paul@codesourcery.com> In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200906301611.29072.paul@codesourcery.com> List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Gerd Hoffmann , Juan Quintela On Tuesday 30 June 2009, Juan Quintela wrote: > Paul Brook wrote: > >> What I want to do, to make usb modular I need two things: > >> > >> a- a way to defino an alias, that "mouse" is equivalent to "QEMU USB > >> Keyboard". One is the qdev name and the other is the name passed to > >> --usbdevice name. > > > > I'm not too bothered about this. We're changing the option, so I don't > > see any particular problem with changing the device names at the same > > time. > > backward compatibility? Not that I care about this one really. This is a new option. There's nothing to be backwards compatible with. > >> b- things like disk are composed of: > >> "disk" : "rest of disk arguments" > >> > > I think this is confusing host configuration with machine configuration. > > This has been discussed before. > > Yes, but it is the same problem. We are going to need something like: > > -usb-hardawre name=bar,... -usbdevice disk:name=bar > > or anything like that, the thing that I mean is that disk is going to > have always a parameter (what disk we mean) and mouse is not going > (necesarely) I'm not sure we're discussing the same thing here. "rest of disk arguments" should not be part of the USB device. The USB device should just have a link to a drive object, which is configured elsewhere (e.g. via a -drive option). Paul