xen: explicitly set hostdev driver.name at runtime, not in postparse

Xen only supports a single type of PCI hostdev assignment, so it is
superfluous to have <driver name='xen'/> peppered throughout the
config. It *is* necessary to have the driver type explicitly set in
the hostdev object before calling into the hypervisor-agnostic "hostdev
manager" though (otherwise the hostdev manager doesn't know whether it
should do Xen-specific setup, or VFIO-specific setup).

Historically, the Xen driver has checked for "default" driver name
(i.e. not set in the XML), and set it to "xen', during the XML
postparse, thus guaranteeing that it will be set by the time the
object is sent to the hostdev manager at runtime, but also setting it
so early that a simple round-trip of parse-format results in the XML
always containing an explicit <driver name='xen'/>, even if that
wasn't specified in the original XML.

The QEMU driver *doesn't* set driver.name during postparse though;
instead, it waits until domain startup time (or device attach time for
hotplug), and sets the driver.name then. The result is that a
parse-format round trip of the XML in the QEMU driver *doesn't* add in
the <driver name='vfio'/>.

This patch modifies the Xen driver to behave similarly to the QEMU
driver - the PostParse just checks for a driver.name that isn't
supported by the Xen driver, and any explicit setting to "xen" is
deferred until domain runtime rather than during the postparse, thus
Xen domain XML also doesn't get extraneous <driver name='xen'/>.

This delayed setting of driver.name of course results in slightly
different xml2xml parse-format results, so the unit test data is
modified accordingly.

Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>

src/libxl/libxl_domain.c

@@ -160,8 +160,15 @@ libxlDomainDeviceDefPostParse(virDomainDeviceDef *dev,
 
         if (hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS &&
             hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI &&
-            pcisrc->driver.name == VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_DEFAULT)
+            pcisrc->driver.name != VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_DEFAULT &&
-            pcisrc->driver.name = VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_XEN;
+            pcisrc->driver.name != VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_XEN) {
+
+            /* Xen only supports "Xen" style of hostdev */
+            virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
+                           _("XEN does not support device assignment mode '%1$s'"),
+                           virDeviceHostdevPCIDriverNameTypeToString(pcisrc->driver.name));
+            return -1;
+        }
     }
 
     if (dev->type == VIR_DOMAIN_DEVICE_VIDEO) {
@@ -986,18 +993,9 @@ libxlNetworkPrepareDevices(virDomainDef *def)
         if (actualType == VIR_DOMAIN_NET_TYPE_HOSTDEV &&
             net->type == VIR_DOMAIN_NET_TYPE_NETWORK) {
             /* Each type='hostdev' network device must also have a
-             * corresponding entry in the hostdevs array. For netdevs
+             * corresponding entry in the hostdevs array.
-             * that are hardcoded as type='hostdev', this is already
-             * done by the parser, but for those allocated from a
-             * network / determined at runtime, we need to do it
-             * separately.
              */
             virDomainHostdevDef *hostdev = virDomainNetGetActualHostdev(net);
-            virDomainHostdevSubsysPCI *pcisrc = &hostdev->source.subsys.u.pci;
-
-            if (hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS &&
-                hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI)
-                pcisrc->driver.name = VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_XEN;
 
             if (virDomainHostdevInsert(def, hostdev) < 0)
                 return -1;
@@ -1007,6 +1005,54 @@ libxlNetworkPrepareDevices(virDomainDef *def)
     return 0;
 }
 
+
+static int
+libxlHostdevPrepareDevices(virDomainDef *def)
+{
+    size_t i;
+
+    for (i = 0; i < def->nhostdevs; i++) {
+        virDomainHostdevDef *hostdev = def->hostdevs[i];
+        virDomainHostdevSubsysPCI *pcisrc = &hostdev->source.subsys.u.pci;
+
+        if (hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS &&
+            hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI &&
+            pcisrc->driver.name == VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_DEFAULT) {
+
+
+            pcisrc->driver.name = VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_XEN;
+
+            /* Q: Why do we set an explicit driver.name here even
+             * though Xen (currently) only supports one driver.name for
+             * PCI hostdev assignment?
+             *
+             * A: Setting the driver name *in the domain XML config*
+             * is optional (and actually pointless at the time of
+             * writing, since each hypervisor only supports a single
+             * type of PCI hostdev device assignment). But the
+             * hypervisor-agnostic virHostdevPrepareDomainDevices(),
+             * which is called immediately after this function in
+             * order to do any driver-related setup (e.g. bind the
+             * host device to a driver kernel driver), requires that
+             * the driver.name be explicitly set (otherwise it wouldn't
+             * know whether to bind to the Xen driver or VFIO driver,
+             * for example).
+             *
+             * NB: If there are ever multiple types of device
+             * assignment supported by Xen, DO NOT CHANGE the value of
+             * driver.name set above when the config is "default" to
+             * anything other than "xen", unless the guest ABI of the
+             * new type is compatible with that of current "xen"
+             * device assignment.
+             *
+             */
+        }
+    }
+
+    return 0;
+}
+
+
 static void
 libxlConsoleCallback(libxl_ctx *ctx, libxl_event *ev, void *for_callback)
 {
@@ -1174,6 +1220,9 @@ libxlDomainStartPrepare(libxlDriverPrivate *driver,
     if (libxlNetworkPrepareDevices(vm->def) < 0)
         goto error;
 
+    if (libxlHostdevPrepareDevices(vm->def) < 0)
+        goto error;
+
     if (virHostdevPrepareDomainDevices(hostdev_mgr, LIBXL_DRIVER_INTERNAL_NAME,
                                        vm->def, hostdev_flags) < 0)
         goto error;

src/libxl/libxl_driver.c

@@ -3087,6 +3087,22 @@ libxlDomainAttachHostPCIDevice(libxlDriverPrivate *driver,
 
     VIR_REALLOC_N(vm->def->hostdevs, vm->def->nhostdevs + 1);
 
+    /* The only supported driverType for Xen is
+     * VIR_DOMAIN_HOSTDEV_PCI_DRIVER_TYPE_XEN, which normally isn't
+     * set in the config (because it doesn't need to be), but it does
+     * need to be set for the impending call to
+     * virHostdevPreparePCIDevices()
+     */
+    if (pcisrc->driver.name == VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_DEFAULT)
+        pcisrc->driver.name = VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_XEN;
+
+    if (pcisrc->driver.name != VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_XEN) {
+        virReportError(VIR_ERR_CONFIG_UNSUPPORTED,
+                       _("XEN does not support device assignment mode '%1$s'"),
+                       virDeviceHostdevPCIDriverNameTypeToString(pcisrc->driver.name));
+        goto cleanup;
+    }
+
     if (virHostdevPreparePCIDevices(hostdev_mgr, LIBXL_DRIVER_INTERNAL_NAME,
                                     vm->def->name, vm->def->uuid,
                                     &hostdev, 1, 0) < 0)
@@ -3395,15 +3411,6 @@ libxlDomainAttachNetDevice(libxlDriverPrivate *driver,
 
     if (actualType == VIR_DOMAIN_NET_TYPE_HOSTDEV) {
         virDomainHostdevDef *hostdev = virDomainNetGetActualHostdev(net);
-        virDomainHostdevSubsysPCI *pcisrc = &hostdev->source.subsys.u.pci;
-
-        /* For those just allocated from a network pool whose driver type is
-         * still VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_DEFAULT, we need to set
-         * driver name correctly.
-         */
-        if (hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS &&
-            hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI)
-            pcisrc->driver.name = VIR_DEVICE_HOSTDEV_PCI_DRIVER_NAME_XEN;
 
         /* This is really a "smart hostdev", so it should be attached
          * as a hostdev (the hostdev code will reach over into the

tests/libxlxml2domconfigdata/moredevs-hvm.xml

@@ -43,7 +43,6 @@
     </interface>
     <interface type='hostdev' managed='yes'>
       <mac address='00:16:3e:2e:e7:fc'/>
-      <driver name='xen'/>
       <source>
         <address type='pci' domain='0x0000' bus='0x0a' slot='0x10' function='0x0'/>
       </source>

tests/xlconfigdata/test-fullvirt-pci.xml

@@ -37,13 +37,11 @@
       <model type='cirrus' vram='8192' heads='1' primary='yes'/>
     </video>
     <hostdev mode='subsystem' type='pci' managed='no'>
-      <driver name='xen'/>
       <source>
         <address domain='0x0000' bus='0x01' slot='0x1a' function='0x1'/>
       </source>
     </hostdev>
     <hostdev mode='subsystem' type='pci' managed='no'>
-      <driver name='xen'/>
       <source writeFiltering='no'>
         <address domain='0x0000' bus='0x02' slot='0x00' function='0x0'/>
       </source>

tests/xmconfigdata/test-pci-dev-syntax.xml

@@ -55,13 +55,11 @@
       <model type='cirrus' vram='8192' heads='1' primary='yes'/>
     </video>
     <hostdev mode='subsystem' type='pci' managed='no'>
-      <driver name='xen'/>
       <source>
         <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/>
       </source>
     </hostdev>
     <hostdev mode='subsystem' type='pci' managed='no'>
-      <driver name='xen'/>
       <source>
         <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/>
       </source>

tests/xmconfigdata/test-pci-devs.xml

@@ -55,13 +55,11 @@
       <model type='cirrus' vram='8192' heads='1' primary='yes'/>
     </video>
     <hostdev mode='subsystem' type='pci' managed='no'>
-      <driver name='xen'/>
       <source>
         <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/>
       </source>
     </hostdev>
     <hostdev mode='subsystem' type='pci' managed='no'>
-      <driver name='xen'/>
       <source>
         <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/>
       </source>