From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NKbBw-000134-Ri for qemu-devel@nongnu.org; Tue, 15 Dec 2009 12:28:08 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1NKbBt-0000yT-1K for qemu-devel@nongnu.org; Tue, 15 Dec 2009 12:28:08 -0500 Received: from [199.232.76.173] (port=38652 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NKbBs-0000xv-7z for qemu-devel@nongnu.org; Tue, 15 Dec 2009 12:28:04 -0500 Received: from gecko.sbs.de ([194.138.37.40]:21167) by monty-python.gnu.org with esmtps (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1NKbBr-0003u4-85 for qemu-devel@nongnu.org; Tue, 15 Dec 2009 12:28:03 -0500 Message-ID: <4B27C7B1.9030609@siemens.com> Date: Tue, 15 Dec 2009 18:30:25 +0100 From: Wolfgang Mauerer MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="------------090704050008010304070101" Subject: [Qemu-devel] [PATCH, RFC] Allow adding empty SCSI controllers List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gerd Hoffmann Cc: "qemu-devel@nongnu.org" This is a multi-part message in MIME format. --------------090704050008010304070101 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Hi Gerd, in commit 5b684b5a56e81f6f, you introduced an explicit check to prevent adding SCSI controllers without attached disks to the system. Is there any other method to introduce disk-less controllers into the system? If not, I'd suggest to remove the check (patch is attached) -- there are some situations when you want empty SCSI controllers, for instance for the libvirt hotplug/remove framework currently under development (see http://article.gmane.org/gmane.comp.emulators.libvirt/19043) Thanks, Wolfgang --------------090704050008010304070101 Content-Type: text/x-patch; name="allow_empty_scsi_controllers.patch" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="allow_empty_scsi_controllers.patch" commit c827742224b9a3a0d9dad0ce36c7e59c1a796ec1 Author: Wolfgang Mauerer Date: Tue Dec 15 18:06:19 2009 +0100 Revert "hotplug: fix "pci_add storage if=scsi"" This reverts commit 5b684b5a56e81f6f88234952fe8ed68010c36e19, and also includes some manual adaptions. There are some cases where addding an empty SCSI controller to a system makes sense, so we should not prohibit this option. Signed-off-by: Wolfgang Mauerer diff --git a/hw/lsi53c895a.c b/hw/lsi53c895a.c index 8b8a80b..eee206b 100644 --- a/hw/lsi53c895a.c +++ b/hw/lsi53c895a.c @@ -2103,9 +2103,7 @@ static int lsi_scsi_init(PCIDevice *dev) lsi_soft_reset(s); scsi_bus_new(&s->bus, &dev->qdev, 1, LSI_MAX_DEVS, lsi_command_complete); - if (!dev->qdev.hotplugged) { - scsi_bus_legacy_handle_cmdline(&s->bus); - } + scsi_bus_legacy_handle_cmdline(&s->bus); vmstate_register(-1, &vmstate_lsi_scsi, s); return 0; } diff --git a/hw/pci-hotplug.c b/hw/pci-hotplug.c index 6e42776..075be31 100644 --- a/hw/pci-hotplug.c +++ b/hw/pci-hotplug.c @@ -183,14 +183,10 @@ static PCIDevice *qemu_pci_hot_add_storage(Monitor *mon, switch (type) { case IF_SCSI: - if (!dinfo) { - monitor_printf(mon, "scsi requires a backing file/device.\n"); - return NULL; - } dev = pci_create(bus, devfn, "lsi53c895a"); if (qdev_init(&dev->qdev) < 0) dev = NULL; - if (dev) { + if (dev && dinfo) { if (scsi_hot_add(&dev->qdev, dinfo, 0) != 0) { qdev_unplug(&dev->qdev); dev = NULL; @@ -204,12 +200,15 @@ static PCIDevice *qemu_pci_hot_add_storage(Monitor *mon, } dev = pci_create(bus, devfn, "virtio-blk-pci"); qdev_prop_set_drive(&dev->qdev, "drive", dinfo); - if (qdev_init(&dev->qdev) < 0) - dev = NULL; break; default: dev = NULL; } + if (!dev) + return NULL; + if (dinfo && qdev_init(&dev->qdev) < 0) + return NULL; + return dev; } --------------090704050008010304070101--