From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MHiFS-0004bm-Ek for qemu-devel@nongnu.org; Fri, 19 Jun 2009 13:51:34 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MHiFL-0004X5-Ck for qemu-devel@nongnu.org; Fri, 19 Jun 2009 13:51:31 -0400 Received: from [199.232.76.173] (port=49756 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MHiFK-0004WI-Cv for qemu-devel@nongnu.org; Fri, 19 Jun 2009 13:51:26 -0400 Received: from mx20.gnu.org ([199.232.41.8]:41154) by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1MHiFJ-0006e9-Fi for qemu-devel@nongnu.org; Fri, 19 Jun 2009 13:51:25 -0400 Received: from mail.codesourcery.com ([65.74.133.4]) by mx20.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MHiFI-0002Ii-24 for qemu-devel@nongnu.org; Fri, 19 Jun 2009 13:51:24 -0400 From: Paul Brook Subject: Re: [Qemu-devel] Re: [PATCH 0/10] qdev patches. Date: Fri, 19 Jun 2009 18:51:20 +0100 References: <1245243565-24807-1-git-send-email-kraxel@redhat.com> <4A3A51B9.90207@redhat.com> <4A3BB5F2.9050701@redhat.com> In-Reply-To: <4A3BB5F2.9050701@redhat.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200906191851.21563.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 On Friday 19 June 2009, Gerd Hoffmann wrote: > Hi, > > > Updated patch queue pushed to qdev.v5. Not posting to avoid spamming the > > list too much. Online viewable via gitweb here: > > http://git.et.redhat.com/?p=qemu-kraxel.git;a=shortlog;h=refs/heads/qdev. > >v5 > > Todays update pushed to qdev.v6. > http://git.et.redhat.com/?p=qemu-kraxel.git;a=shortlog;h=refs/heads/qdev.v6 A few comments on specific patches: * qdev: update pci device registration I dislike passing an {array,length} pair. Especially when it requires every user to manually get the right length. * qdev/core: bus list I don't seen any good reason for this. In fact I think it is a major step backwards. A bus is uniquely identified by its name and parent device. * qdev/pci: misc fixes. All uses of the second argument to savevm should go away, not introduce new ones. I'm unconvinced by the dev->name change. If we're using the same value then why does it exist at all? * qdev/pci: hook up i440fx i440fx_init should not exist. c.f. versatile_pci.c * qdev: update pci device registration This is exactly the sort of fake conversion that I don't like, because you still require use of the old hardcoded initialization functions. Convenience wrappers like smc91c111_init are fine (and will naturally disappear when we have a machine config), but you shouldn't be poking directly at device state. In practice there's no way for the user to have more than one set of IDE busses, so I don't see much point pretending we allow this. i.e. remove the hd_table argument altogether and use drive_get_index directly. * qdev: convert all vga Likewise, pci_vga_init needs to go away. * qdev/scsi: add scsi bus support to qdev, convert drivers This still feels wrong, probably because you're using the same thing for both a parallel scsi bus, and for devices (usb-msd) that incorporate scsi functionality directly. The current qemu scsi API is actually a set of point to point links with individual devices. All the bus emulation is local to the host controller. * qdev/usb* I have not looked at these patches.