From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MGCWZ-0002XI-HA for qemu-devel@nongnu.org; Mon, 15 Jun 2009 09:46:59 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MGCWV-0002WU-JG for qemu-devel@nongnu.org; Mon, 15 Jun 2009 09:46:59 -0400 Received: from [199.232.76.173] (port=49307 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MGCWV-0002WO-9c for qemu-devel@nongnu.org; Mon, 15 Jun 2009 09:46:55 -0400 Received: from mail-qy0-f191.google.com ([209.85.221.191]:46659) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MGCWU-0007xF-OD for qemu-devel@nongnu.org; Mon, 15 Jun 2009 09:46:55 -0400 Received: by qyk29 with SMTP id 29so4568811qyk.4 for ; Mon, 15 Jun 2009 06:46:52 -0700 (PDT) Message-ID: <4A3650C9.1080403@codemonkey.ws> Date: Mon, 15 Jun 2009 08:46:49 -0500 From: Anthony Liguori MIME-Version: 1.0 Subject: Re: [Qemu-devel] Re: [PATCH] Add -no-virtio-balloon command-line option References: <1244661817-3293-1-git-send-email-ehabkost@redhat.com> <4A300891.4060500@codemonkey.ws> <20090610193408.GE18045@blackpad> <4A30D748.8070507@redhat.com> <4A36138C.4060306@redhat.com> <87vdmx69na.fsf@pike.pond.sub.org> <4A36447B.6030509@codemonkey.ws> <4A364F61.8090406@redhat.com> In-Reply-To: <4A364F61.8090406@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: dlaor@redhat.com Cc: Markus Armbruster , qemu-devel@nongnu.org Dor Laor wrote: > Anthony Liguori wrote: >> Markus Armbruster wrote: >>> Dor Laor writes: >>> >>> >>>> Gerd Hoffmann wrote: >>>> >>>>>> This new option may be used to disable the virtio-balloon device. >>>>>> >>>>>> >>>>> ACK, looks good. >>>>> >>>>> And I can drop my lazy patch which just comments out virtio-balloon >>>>> in the code ;) >>>>> >>>> Actually, I rather have the balloon disabled by default. It's like any >>>> other pci device which needs >>>> explicit command line specification. Why consume a pci slot if not >>>> implicitly required? >>>> >>> >>> I agree. The command line is plenty magic without implicitly added >>> PCI devices. >>> >>> Besides, negative options -no-virtio-balloon lead to ugly >>> double-negatives like if (!no_virtio_balloon). >>> >> >> It does no harm to add the device by default so why force the user to >> explicitly enable it? > > It consumes one of the 32 pci devices we have today. It's the same for > not having -usb as default > or any other type of device. Soon we'll need pci_addr= along with it > anyway ;) IMHO, -usb should be default. It wasn't made default when it was introduced because it was known to be buggy. For management tools, explicitly disabling something is easy enough. For humans, I think wanting features is more common than needing more than 32 PCI slots. Regards, Anthony Liguori > Dor >