[Qemu-devel] [PATCH v4 00/30] virtio: cleanup, fix hot-unplug, move to realize

[Qemu-devel] [PATCH v4 00/30] virtio: cleanup, fix hot-unplug, move to realize

[Paolo Bonzini]

This series fixes hot-unplug of virtio devices, which can crash due to
dangling pointer accesses.

The current implementation supports guest-initiated hot-unplug via the
virtio_bus_destroy_device function, but not hot-unplugging the virtio
device by virtue of unplugging its parent container device.

The problem is that the callback for the bus implementation to cleanup
is placed in the wrong place; it is in virtio_bus_destroy_device, which
should be called by the bus, instead of being somewhere in device code.
We need to have the callback in device code (for example in dc->exit),
so that we invoke it on every unplug action, no matter who starts it.

Thus, the series cleans up plugging and unplugging of virtio devices
so that it does not need any help from the bus (patches 2-5).  It then
stops the virtio devices' overriding of dc->exit, moving their cleanup
code to the new exit callback in VirtioDeviceClass (patches 6-11).
Finally, patch 12 can make virtio-pci implement the device_unplugged
callback.

A similar dangling-pointer bug is exposed by this change in virtio-ccw.
Patch 1 avoids this; it is kept at the beginning to ensure bisectability.

Patches 13-30 are Andreas's conversion of virtio to realize/unrealize,
which has better error propagation as shown by the small testcase fix.

v3->v4: rebased, added Andreas's realize patches on top

v2->v3: fix to s390 patch; added Reviewed-by and Cced patch 1 to
	qemu-stable.

v1->v2: remove useless pointer chasing in virtio_pci_notify, add
	patch 1 to fix breakage reported by Cornelia.

Andreas Färber (18):
  virtio-blk-dataplane: Improve error reporting
  virtio-9p: QOM realize preparations
  virtio-blk: QOM realize preparations
  virtio-serial: QOM realize preparations
  virtio-net: QOM realize preparations
  virtio-balloon: QOM realize preparations
  virtio-rng: QOM realize preparations
  virtio-scsi: QOM realize preparations
  virtio: Start converting VirtioDevice to QOM realize
  virtio-9p: Convert to QOM realize
  virtio-blk: Convert to QOM realize
  virtio-serial: Convert to QOM realize
  virtio-net: Convert to QOM realize
  virtio-balloon: Convert to QOM realize
  virtio-rng: Convert to QOM realize
  virtio-scsi: Convert to QOM realize
  virtio: Complete converting VirtioDevice to QOM realize
  virtio: Convert exit to unrealize

Paolo Bonzini (12):
  virtio-ccw: move virtio_ccw_stop_ioeventfd to virtio_ccw_busdev_unplug
  virtio-bus: remove vdev field
  virtio-ccw: remove vdev field
  virtio-pci: remove vdev field
  virtio-bus: cleanup plug/unplug interface
  virtio-blk: switch exit callback to VirtioDeviceClass
  virtio-serial: switch exit callback to VirtioDeviceClass
  virtio-net: switch exit callback to VirtioDeviceClass
  virtio-scsi: switch exit callback to VirtioDeviceClass
  virtio-balloon: switch exit callback to VirtioDeviceClass
  virtio-rng: switch exit callback to VirtioDeviceClass
  virtio-pci: add device_unplugged callback

 hw/9pfs/virtio-9p-device.c      |  43 +++++++-------
 hw/block/dataplane/virtio-blk.c |  30 +++++-----
 hw/block/dataplane/virtio-blk.h |   5 +-
 hw/block/virtio-blk.c           |  47 ++++++++++------
 hw/char/virtio-serial-bus.c     |  33 +++++------
 hw/net/virtio-net.c             |  36 ++++++------
 hw/s390x/virtio-ccw.c           |  83 +++++++++++++++------------
 hw/s390x/virtio-ccw.h           |   1 -
 hw/scsi/vhost-scsi.c            |  45 +++++++--------
 hw/scsi/virtio-scsi.c           |  57 +++++++++----------
 hw/virtio/virtio-balloon.c      |  32 +++++------
 hw/virtio/virtio-bus.c          |  80 +++++++++++++++-----------
 hw/virtio/virtio-mmio.c         |   9 +--
 hw/virtio/virtio-pci.c          | 122 ++++++++++++++++++++++++----------------
 hw/virtio/virtio-pci.h          |   1 -
 hw/virtio/virtio-rng.c          |  43 +++++++-------
 hw/virtio/virtio.c              |  40 ++++++++-----
 include/hw/virtio/virtio-bus.h  |  22 +++++---
 include/hw/virtio/virtio-rng.h  |   2 +
 include/hw/virtio/virtio-scsi.h |   4 +-
 include/hw/virtio/virtio.h      |   8 ++-
 tests/qdev-monitor-test.c       |   4 +-
 22 files changed, 413 insertions(+), 334 deletions(-)

-- 
1.8.3.1

[Qemu-devel] [PATCH 01/30] virtio-ccw: move virtio_ccw_stop_ioeventfd to virtio_ccw_busdev_unplug

[Paolo Bonzini]

Similar to the PCI bug that prompted these patches, virtio-ccw will
segfault after the reworking of hotplug/hot-unplug.  Prepare for
this by moving virtio_ccw_stop_ioeventfd to before the freeing
of the proxy device.

A better place for this could be the device_unplugged callback
for the virtio-ccw bus.  However, we do not yet have a callback
that works: this patch avoids the problem while leaving the tree
bisectable.

Cc: qemu-stable@nongnu.org
Reported-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Suggested-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/s390x/virtio-ccw.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/hw/s390x/virtio-ccw.c b/hw/s390x/virtio-ccw.c
index f93a81c..e8b4547 100644
--- a/hw/s390x/virtio-ccw.c
+++ b/hw/s390x/virtio-ccw.c
@@ -631,7 +631,6 @@ static int virtio_ccw_exit(VirtioCcwDevice *dev)
 {
     SubchDev *sch = dev->sch;
 
-    virtio_ccw_stop_ioeventfd(dev);
     if (sch) {
         css_subch_assign(sch->cssid, sch->ssid, sch->schid, sch->devno, NULL);
         g_free(sch);
@@ -1228,6 +1227,8 @@ static int virtio_ccw_busdev_unplug(DeviceState *dev)
     VirtioCcwDevice *_dev = (VirtioCcwDevice *)dev;
     SubchDev *sch = _dev->sch;
 
+    virtio_ccw_stop_ioeventfd(_dev);
+
     /*
      * We should arrive here only for device_del, since we don't support
      * direct hot(un)plug of channels, but only through virtio.
-- 
1.8.3.1

[Qemu-devel] [PATCH 02/30] virtio-bus: remove vdev field

[Paolo Bonzini]

The vdev field is complicated to synchronize.  Just access the
BusState's list of children.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-bus.c         | 65 +++++++++++++++++++++++++-----------------
 hw/virtio/virtio-mmio.c        |  9 +++---
 hw/virtio/virtio-pci.c         |  2 +-
 include/hw/virtio/virtio-bus.h | 16 ++++++++---
 4 files changed, 57 insertions(+), 35 deletions(-)

diff --git a/hw/virtio/virtio-bus.c b/hw/virtio/virtio-bus.c
index e6b103c..17dd06e 100644
--- a/hw/virtio/virtio-bus.c
+++ b/hw/virtio/virtio-bus.c
@@ -46,8 +46,6 @@ int virtio_bus_plug_device(VirtIODevice *vdev)
     VirtioBusClass *klass = VIRTIO_BUS_GET_CLASS(bus);
     DPRINTF("%s: plug device.\n", qbus->name);
 
-    bus->vdev = vdev;
-
     if (klass->device_plugged != NULL) {
         klass->device_plugged(qbus->parent);
     }
@@ -58,9 +56,11 @@ int virtio_bus_plug_device(VirtIODevice *vdev)
 /* Reset the virtio_bus */
 void virtio_bus_reset(VirtioBusState *bus)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
+
     DPRINTF("%s: reset device.\n", qbus->name);
-    if (bus->vdev != NULL) {
-        virtio_reset(bus->vdev);
+    if (vdev != NULL) {
+        virtio_reset(vdev);
     }
 }
 
@@ -69,62 +69,71 @@ void virtio_bus_destroy_device(VirtioBusState *bus)
 {
     BusState *qbus = BUS(bus);
     VirtioBusClass *klass = VIRTIO_BUS_GET_CLASS(bus);
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
+
     DPRINTF("%s: remove device.\n", qbus->name);
 
-    if (bus->vdev != NULL) {
+    if (vdev != NULL) {
         if (klass->device_unplug != NULL) {
             klass->device_unplug(qbus->parent);
         }
-        object_unparent(OBJECT(bus->vdev));
-        bus->vdev = NULL;
+        object_unparent(OBJECT(vdev));
     }
 }
 
 /* Get the device id of the plugged device. */
 uint16_t virtio_bus_get_vdev_id(VirtioBusState *bus)
 {
-    assert(bus->vdev != NULL);
-    return bus->vdev->device_id;
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
+    assert(vdev != NULL);
+    return vdev->device_id;
 }
 
 /* Get the config_len field of the plugged device. */
 size_t virtio_bus_get_vdev_config_len(VirtioBusState *bus)
 {
-    assert(bus->vdev != NULL);
-    return bus->vdev->config_len;
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
+    assert(vdev != NULL);
+    return vdev->config_len;
 }
 
 /* Get the features of the plugged device. */
 uint32_t virtio_bus_get_vdev_features(VirtioBusState *bus,
                                     uint32_t requested_features)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
     VirtioDeviceClass *k;
-    assert(bus->vdev != NULL);
-    k = VIRTIO_DEVICE_GET_CLASS(bus->vdev);
+
+    assert(vdev != NULL);
+    k = VIRTIO_DEVICE_GET_CLASS(vdev);
     assert(k->get_features != NULL);
-    return k->get_features(bus->vdev, requested_features);
+    return k->get_features(vdev, requested_features);
 }
 
 /* Set the features of the plugged device. */
 void virtio_bus_set_vdev_features(VirtioBusState *bus,
                                       uint32_t requested_features)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
     VirtioDeviceClass *k;
-    assert(bus->vdev != NULL);
-    k = VIRTIO_DEVICE_GET_CLASS(bus->vdev);
+
+    assert(vdev != NULL);
+    k = VIRTIO_DEVICE_GET_CLASS(vdev);
     if (k->set_features != NULL) {
-        k->set_features(bus->vdev, requested_features);
+        k->set_features(vdev, requested_features);
     }
 }
 
 /* Get bad features of the plugged device. */
 uint32_t virtio_bus_get_vdev_bad_features(VirtioBusState *bus)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
     VirtioDeviceClass *k;
-    assert(bus->vdev != NULL);
-    k = VIRTIO_DEVICE_GET_CLASS(bus->vdev);
+
+    assert(vdev != NULL);
+    k = VIRTIO_DEVICE_GET_CLASS(vdev);
     if (k->bad_features != NULL) {
-        return k->bad_features(bus->vdev);
+        return k->bad_features(vdev);
     } else {
         return 0;
     }
@@ -133,22 +142,26 @@ uint32_t virtio_bus_get_vdev_bad_features(VirtioBusState *bus)
 /* Get config of the plugged device. */
 void virtio_bus_get_vdev_config(VirtioBusState *bus, uint8_t *config)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
     VirtioDeviceClass *k;
-    assert(bus->vdev != NULL);
-    k = VIRTIO_DEVICE_GET_CLASS(bus->vdev);
+
+    assert(vdev != NULL);
+    k = VIRTIO_DEVICE_GET_CLASS(vdev);
     if (k->get_config != NULL) {
-        k->get_config(bus->vdev, config);
+        k->get_config(vdev, config);
     }
 }
 
 /* Set config of the plugged device. */
 void virtio_bus_set_vdev_config(VirtioBusState *bus, uint8_t *config)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
     VirtioDeviceClass *k;
-    assert(bus->vdev != NULL);
-    k = VIRTIO_DEVICE_GET_CLASS(bus->vdev);
+
+    assert(vdev != NULL);
+    k = VIRTIO_DEVICE_GET_CLASS(vdev);
     if (k->set_config != NULL) {
-        k->set_config(bus->vdev, config);
+        k->set_config(vdev, config);
     }
 }
 
diff --git a/hw/virtio/virtio-mmio.c b/hw/virtio/virtio-mmio.c
index 29cf284..8829eb0 100644
--- a/hw/virtio/virtio-mmio.c
+++ b/hw/virtio/virtio-mmio.c
@@ -95,7 +95,7 @@ static void virtio_mmio_bus_new(VirtioBusState *bus, size_t bus_size,
 static uint64_t virtio_mmio_read(void *opaque, hwaddr offset, unsigned size)
 {
     VirtIOMMIOProxy *proxy = (VirtIOMMIOProxy *)opaque;
-    VirtIODevice *vdev = proxy->bus.vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
 
     DPRINTF("virtio_mmio_read offset 0x%x\n", (int)offset);
 
@@ -185,7 +185,7 @@ static void virtio_mmio_write(void *opaque, hwaddr offset, uint64_t value,
                               unsigned size)
 {
     VirtIOMMIOProxy *proxy = (VirtIOMMIOProxy *)opaque;
-    VirtIODevice *vdev = proxy->bus.vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
 
     DPRINTF("virtio_mmio_write offset 0x%x value 0x%" PRIx64 "\n",
             (int)offset, value);
@@ -298,12 +298,13 @@ static const MemoryRegionOps virtio_mem_ops = {
 static void virtio_mmio_update_irq(DeviceState *opaque, uint16_t vector)
 {
     VirtIOMMIOProxy *proxy = VIRTIO_MMIO(opaque);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     int level;
 
-    if (!proxy->bus.vdev) {
+    if (!vdev) {
         return;
     }
-    level = (proxy->bus.vdev->isr != 0);
+    level = (vdev->isr != 0);
     DPRINTF("virtio_mmio setting IRQ %d\n", level);
     qemu_set_irq(proxy->irq, level);
 }
diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 7647be8..76b7652 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -943,7 +943,7 @@ static void virtio_pci_device_plugged(DeviceState *d)
     uint8_t *config;
     uint32_t size;
 
-    proxy->vdev = bus->vdev;
+    proxy->vdev = virtio_bus_get_device(bus);
 
     config = proxy->pci_dev.config;
     if (proxy->class_code) {
diff --git a/include/hw/virtio/virtio-bus.h b/include/hw/virtio/virtio-bus.h
index 9217f85..ba0f86a 100644
--- a/include/hw/virtio/virtio-bus.h
+++ b/include/hw/virtio/virtio-bus.h
@@ -72,10 +72,6 @@ typedef struct VirtioBusClass {
 
 struct VirtioBusState {
     BusState parent_obj;
-    /*
-     * Only one VirtIODevice can be plugged on the bus.
-     */
-    VirtIODevice *vdev;
 };
 
 int virtio_bus_plug_device(VirtIODevice *vdev);
@@ -98,4 +94,16 @@ void virtio_bus_get_vdev_config(VirtioBusState *bus, uint8_t *config);
 /* Set config of the plugged device. */
 void virtio_bus_set_vdev_config(VirtioBusState *bus, uint8_t *config);
 
+static inline VirtIODevice *virtio_bus_get_device(VirtioBusState *bus)
+{
+    BusState *qbus = &bus->parent_obj;
+    BusChild *kid = QTAILQ_FIRST(&qbus->children);
+    DeviceState *qdev = kid ? kid->child : NULL;
+
+    /* This is used on the data path, the cast is guaranteed
+     * to succeed by the qdev machinery.
+     */
+    return (VirtIODevice *)qdev;
+}
+
 #endif /* VIRTIO_BUS_H */
-- 
1.8.3.1

[Qemu-devel] [PATCH 03/30] virtio-ccw: remove vdev field

[Paolo Bonzini]

The vdev field is complicated to synchronize.  Just access the
BusState's list of children.

Cc: qemu-stable@nongnu.org
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/s390x/virtio-ccw.c | 80 ++++++++++++++++++++++++++++-----------------------
 hw/s390x/virtio-ccw.h |  1 -
 2 files changed, 44 insertions(+), 37 deletions(-)

diff --git a/hw/s390x/virtio-ccw.c b/hw/s390x/virtio-ccw.c
index e8b4547..ecc80ec 100644
--- a/hw/s390x/virtio-ccw.c
+++ b/hw/s390x/virtio-ccw.c
@@ -57,9 +57,10 @@ static const TypeInfo virtual_css_bus_info = {
 VirtIODevice *virtio_ccw_get_vdev(SubchDev *sch)
 {
     VirtIODevice *vdev = NULL;
+    VirtioCcwDevice *dev = sch->driver_data;
 
-    if (sch->driver_data) {
-        vdev = ((VirtioCcwDevice *)sch->driver_data)->vdev;
+    if (dev) {
+        vdev = virtio_bus_get_device(&dev->bus);
     }
     return vdev;
 }
@@ -67,7 +68,8 @@ VirtIODevice *virtio_ccw_get_vdev(SubchDev *sch)
 static int virtio_ccw_set_guest2host_notifier(VirtioCcwDevice *dev, int n,
                                               bool assign, bool set_handler)
 {
-    VirtQueue *vq = virtio_get_queue(dev->vdev, n);
+    VirtIODevice *vdev = virtio_bus_get_device(&dev->bus);
+    VirtQueue *vq = virtio_get_queue(vdev, n);
     EventNotifier *notifier = virtio_queue_get_host_notifier(vq);
     int r = 0;
     SubchDev *sch = dev->sch;
@@ -97,6 +99,7 @@ static int virtio_ccw_set_guest2host_notifier(VirtioCcwDevice *dev, int n,
 
 static void virtio_ccw_start_ioeventfd(VirtioCcwDevice *dev)
 {
+    VirtIODevice *vdev;
     int n, r;
 
     if (!(dev->flags & VIRTIO_CCW_FLAG_USE_IOEVENTFD) ||
@@ -104,8 +107,9 @@ static void virtio_ccw_start_ioeventfd(VirtioCcwDevice *dev)
         dev->ioeventfd_started) {
         return;
     }
+    vdev = virtio_bus_get_device(&dev->bus);
     for (n = 0; n < VIRTIO_PCI_QUEUE_MAX; n++) {
-        if (!virtio_queue_get_num(dev->vdev, n)) {
+        if (!virtio_queue_get_num(vdev, n)) {
             continue;
         }
         r = virtio_ccw_set_guest2host_notifier(dev, n, true, true);
@@ -118,7 +122,7 @@ static void virtio_ccw_start_ioeventfd(VirtioCcwDevice *dev)
 
   assign_error:
     while (--n >= 0) {
-        if (!virtio_queue_get_num(dev->vdev, n)) {
+        if (!virtio_queue_get_num(vdev, n)) {
             continue;
         }
         r = virtio_ccw_set_guest2host_notifier(dev, n, false, false);
@@ -132,13 +136,15 @@ static void virtio_ccw_start_ioeventfd(VirtioCcwDevice *dev)
 
 static void virtio_ccw_stop_ioeventfd(VirtioCcwDevice *dev)
 {
+    VirtIODevice *vdev;
     int n, r;
 
     if (!dev->ioeventfd_started) {
         return;
     }
+    vdev = virtio_bus_get_device(&dev->bus);
     for (n = 0; n < VIRTIO_PCI_QUEUE_MAX; n++) {
-        if (!virtio_queue_get_num(dev->vdev, n)) {
+        if (!virtio_queue_get_num(vdev, n)) {
             continue;
         }
         r = virtio_ccw_set_guest2host_notifier(dev, n, false, false);
@@ -189,7 +195,7 @@ typedef struct VirtioFeatDesc {
 static int virtio_ccw_set_vqs(SubchDev *sch, uint64_t addr, uint32_t align,
                               uint16_t index, uint16_t num)
 {
-    VirtioCcwDevice *dev = sch->driver_data;
+    VirtIODevice *vdev = virtio_ccw_get_vdev(sch);
 
     if (index > VIRTIO_PCI_QUEUE_MAX) {
         return -EINVAL;
@@ -200,23 +206,23 @@ static int virtio_ccw_set_vqs(SubchDev *sch, uint64_t addr, uint32_t align,
         return -EINVAL;
     }
 
-    if (!dev) {
+    if (!vdev) {
         return -EINVAL;
     }
 
-    virtio_queue_set_addr(dev->vdev, index, addr);
+    virtio_queue_set_addr(vdev, index, addr);
     if (!addr) {
-        virtio_queue_set_vector(dev->vdev, index, 0);
+        virtio_queue_set_vector(vdev, index, 0);
     } else {
         /* Fail if we don't have a big enough queue. */
         /* TODO: Add interface to handle vring.num changing */
-        if (virtio_queue_get_num(dev->vdev, index) > num) {
+        if (virtio_queue_get_num(vdev, index) > num) {
             return -EINVAL;
         }
-        virtio_queue_set_vector(dev->vdev, index, index);
+        virtio_queue_set_vector(vdev, index, index);
     }
     /* tell notify handler in case of config change */
-    dev->vdev->config_vector = VIRTIO_PCI_QUEUE_MAX;
+    vdev->config_vector = VIRTIO_PCI_QUEUE_MAX;
     return 0;
 }
 
@@ -230,6 +236,7 @@ static int virtio_ccw_cb(SubchDev *sch, CCW1 ccw)
     hwaddr indicators;
     VqConfigBlock vq_config;
     VirtioCcwDevice *dev = sch->driver_data;
+    VirtIODevice *vdev = virtio_ccw_get_vdev(sch);
     bool check_len;
     int len;
     hwaddr hw_len;
@@ -272,7 +279,7 @@ static int virtio_ccw_cb(SubchDev *sch, CCW1 ccw)
         break;
     case CCW_CMD_VDEV_RESET:
         virtio_ccw_stop_ioeventfd(dev);
-        virtio_reset(dev->vdev);
+        virtio_reset(vdev);
         ret = 0;
         break;
     case CCW_CMD_READ_FEAT:
@@ -319,7 +326,7 @@ static int virtio_ccw_cb(SubchDev *sch, CCW1 ccw)
             features.features = ldl_le_phys(ccw.cda);
             if (features.index < ARRAY_SIZE(dev->host_features)) {
                 virtio_bus_set_vdev_features(&dev->bus, features.features);
-                dev->vdev->guest_features = features.features;
+                vdev->guest_features = features.features;
             } else {
                 /*
                  * If the guest supports more feature bits, assert that it
@@ -337,30 +344,30 @@ static int virtio_ccw_cb(SubchDev *sch, CCW1 ccw)
         break;
     case CCW_CMD_READ_CONF:
         if (check_len) {
-            if (ccw.count > dev->vdev->config_len) {
+            if (ccw.count > vdev->config_len) {
                 ret = -EINVAL;
                 break;
             }
         }
-        len = MIN(ccw.count, dev->vdev->config_len);
+        len = MIN(ccw.count, vdev->config_len);
         if (!ccw.cda) {
             ret = -EFAULT;
         } else {
-            virtio_bus_get_vdev_config(&dev->bus, dev->vdev->config);
+            virtio_bus_get_vdev_config(&dev->bus, vdev->config);
             /* XXX config space endianness */
-            cpu_physical_memory_write(ccw.cda, dev->vdev->config, len);
+            cpu_physical_memory_write(ccw.cda, vdev->config, len);
             sch->curr_status.scsw.count = ccw.count - len;
             ret = 0;
         }
         break;
     case CCW_CMD_WRITE_CONF:
         if (check_len) {
-            if (ccw.count > dev->vdev->config_len) {
+            if (ccw.count > vdev->config_len) {
                 ret = -EINVAL;
                 break;
             }
         }
-        len = MIN(ccw.count, dev->vdev->config_len);
+        len = MIN(ccw.count, vdev->config_len);
         hw_len = len;
         if (!ccw.cda) {
             ret = -EFAULT;
@@ -371,9 +378,9 @@ static int virtio_ccw_cb(SubchDev *sch, CCW1 ccw)
             } else {
                 len = hw_len;
                 /* XXX config space endianness */
-                memcpy(dev->vdev->config, config, len);
+                memcpy(vdev->config, config, len);
                 cpu_physical_memory_unmap(config, hw_len, 0, hw_len);
-                virtio_bus_set_vdev_config(&dev->bus, dev->vdev->config);
+                virtio_bus_set_vdev_config(&dev->bus, vdev->config);
                 sch->curr_status.scsw.count = ccw.count - len;
                 ret = 0;
             }
@@ -397,9 +404,9 @@ static int virtio_ccw_cb(SubchDev *sch, CCW1 ccw)
             if (!(status & VIRTIO_CONFIG_S_DRIVER_OK)) {
                 virtio_ccw_stop_ioeventfd(dev);
             }
-            virtio_set_status(dev->vdev, status);
-            if (dev->vdev->status == 0) {
-                virtio_reset(dev->vdev);
+            virtio_set_status(vdev, status);
+            if (vdev->status == 0) {
+                virtio_reset(vdev);
             }
             if (status & VIRTIO_CONFIG_S_DRIVER_OK) {
                 virtio_ccw_start_ioeventfd(dev);
@@ -463,7 +470,7 @@ static int virtio_ccw_cb(SubchDev *sch, CCW1 ccw)
             ret = -EFAULT;
         } else {
             vq_config.index = lduw_phys(ccw.cda);
-            vq_config.num_max = virtio_queue_get_num(dev->vdev,
+            vq_config.num_max = virtio_queue_get_num(vdev,
                                                      vq_config.index);
             stw_phys(ccw.cda + sizeof(vq_config.index), vq_config.num_max);
             sch->curr_status.scsw.count = ccw.count - sizeof(vq_config);
@@ -495,7 +502,6 @@ static int virtio_ccw_device_init(VirtioCcwDevice *dev, VirtIODevice *vdev)
     sch->driver_data = dev;
     dev->sch = sch;
 
-    dev->vdev = vdev;
     dev->indicators = 0;
 
     /* Initialize subchannel structure. */
@@ -608,7 +614,7 @@ static int virtio_ccw_device_init(VirtioCcwDevice *dev, VirtIODevice *vdev)
     memset(&sch->id, 0, sizeof(SenseId));
     sch->id.reserved = 0xff;
     sch->id.cu_type = VIRTIO_CCW_CU_TYPE;
-    sch->id.cu_model = dev->vdev->device_id;
+    sch->id.cu_model = vdev->device_id;
 
     /* Only the first 32 feature bits are used. */
     dev->host_features[0] = virtio_bus_get_vdev_features(&dev->bus,
@@ -891,9 +897,10 @@ static unsigned virtio_ccw_get_features(DeviceState *d)
 static void virtio_ccw_reset(DeviceState *d)
 {
     VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
+    VirtIODevice *vdev = virtio_bus_get_device(&dev->bus);
 
     virtio_ccw_stop_ioeventfd(dev);
-    virtio_reset(dev->vdev);
+    virtio_reset(vdev);
     css_reset_sch(dev->sch);
     dev->indicators = 0;
     dev->indicators2 = 0;
@@ -933,9 +940,10 @@ static int virtio_ccw_set_host_notifier(DeviceState *d, int n, bool assign)
 static int virtio_ccw_set_guest_notifier(VirtioCcwDevice *dev, int n,
                                          bool assign, bool with_irqfd)
 {
-    VirtQueue *vq = virtio_get_queue(dev->vdev, n);
+    VirtIODevice *vdev = virtio_bus_get_device(&dev->bus);
+    VirtQueue *vq = virtio_get_queue(vdev, n);
     EventNotifier *notifier = virtio_queue_get_guest_notifier(vq);
-    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(dev->vdev);
+    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
 
     if (assign) {
         int r = event_notifier_init(notifier, 0);
@@ -951,16 +959,16 @@ static int virtio_ccw_set_guest_notifier(VirtioCcwDevice *dev, int n,
          * land in qemu (and only the irq fd) in this code.
          */
         if (k->guest_notifier_mask) {
-            k->guest_notifier_mask(dev->vdev, n, false);
+            k->guest_notifier_mask(vdev, n, false);
         }
         /* get lost events and re-inject */
         if (k->guest_notifier_pending &&
-            k->guest_notifier_pending(dev->vdev, n)) {
+            k->guest_notifier_pending(vdev, n)) {
             event_notifier_set(notifier);
         }
     } else {
         if (k->guest_notifier_mask) {
-            k->guest_notifier_mask(dev->vdev, n, true);
+            k->guest_notifier_mask(vdev, n, true);
         }
         virtio_queue_set_guest_notifier_fd_handler(vq, false, with_irqfd);
         event_notifier_cleanup(notifier);
@@ -972,7 +980,7 @@ static int virtio_ccw_set_guest_notifiers(DeviceState *d, int nvqs,
                                           bool assigned)
 {
     VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
-    VirtIODevice *vdev = dev->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&dev->bus);
     int r, n;
 
     for (n = 0; n < nvqs; n++) {
diff --git a/hw/s390x/virtio-ccw.h b/hw/s390x/virtio-ccw.h
index 96d6f5d..00932c7 100644
--- a/hw/s390x/virtio-ccw.h
+++ b/hw/s390x/virtio-ccw.h
@@ -77,7 +77,6 @@ typedef struct VirtIOCCWDeviceClass {
 struct VirtioCcwDevice {
     DeviceState parent_obj;
     SubchDev *sch;
-    VirtIODevice *vdev;
     char *bus_id;
     uint32_t host_features[VIRTIO_CCW_FEATURE_SIZE];
     VirtioBusState bus;
-- 
1.8.3.1

[Qemu-devel] [PATCH 04/30] virtio-pci: remove vdev field

[Paolo Bonzini]

The vdev field is complicated to synchronize.  Just access the
BusState's list of children.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-pci.c | 110 +++++++++++++++++++++++++++++--------------------
 hw/virtio/virtio-pci.h |   1 -
 2 files changed, 65 insertions(+), 46 deletions(-)

diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 76b7652..15b92e9 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -113,31 +113,40 @@ static inline VirtIOPCIProxy *to_virtio_pci_proxy_fast(DeviceState *d)
 static void virtio_pci_notify(DeviceState *d, uint16_t vector)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy_fast(d);
+
     if (msix_enabled(&proxy->pci_dev))
         msix_notify(&proxy->pci_dev, vector);
-    else
-        pci_set_irq(&proxy->pci_dev, proxy->vdev->isr & 1);
+    else {
+        VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+        pci_set_irq(&proxy->pci_dev, vdev->isr & 1);
+    }
 }
 
 static void virtio_pci_save_config(DeviceState *d, QEMUFile *f)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+
     pci_device_save(&proxy->pci_dev, f);
     msix_save(&proxy->pci_dev, f);
     if (msix_present(&proxy->pci_dev))
-        qemu_put_be16(f, proxy->vdev->config_vector);
+        qemu_put_be16(f, vdev->config_vector);
 }
 
 static void virtio_pci_save_queue(DeviceState *d, int n, QEMUFile *f)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+
     if (msix_present(&proxy->pci_dev))
-        qemu_put_be16(f, virtio_queue_vector(proxy->vdev, n));
+        qemu_put_be16(f, virtio_queue_vector(vdev, n));
 }
 
 static int virtio_pci_load_config(DeviceState *d, QEMUFile *f)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+
     int ret;
     ret = pci_device_load(&proxy->pci_dev, f);
     if (ret) {
@@ -146,12 +155,12 @@ static int virtio_pci_load_config(DeviceState *d, QEMUFile *f)
     msix_unuse_all_vectors(&proxy->pci_dev);
     msix_load(&proxy->pci_dev, f);
     if (msix_present(&proxy->pci_dev)) {
-        qemu_get_be16s(f, &proxy->vdev->config_vector);
+        qemu_get_be16s(f, &vdev->config_vector);
     } else {
-        proxy->vdev->config_vector = VIRTIO_NO_VECTOR;
+        vdev->config_vector = VIRTIO_NO_VECTOR;
     }
-    if (proxy->vdev->config_vector != VIRTIO_NO_VECTOR) {
-        return msix_vector_use(&proxy->pci_dev, proxy->vdev->config_vector);
+    if (vdev->config_vector != VIRTIO_NO_VECTOR) {
+        return msix_vector_use(&proxy->pci_dev, vdev->config_vector);
     }
     return 0;
 }
@@ -159,13 +168,15 @@ static int virtio_pci_load_config(DeviceState *d, QEMUFile *f)
 static int virtio_pci_load_queue(DeviceState *d, int n, QEMUFile *f)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+
     uint16_t vector;
     if (msix_present(&proxy->pci_dev)) {
         qemu_get_be16s(f, &vector);
     } else {
         vector = VIRTIO_NO_VECTOR;
     }
-    virtio_queue_set_vector(proxy->vdev, n, vector);
+    virtio_queue_set_vector(vdev, n, vector);
     if (vector != VIRTIO_NO_VECTOR) {
         return msix_vector_use(&proxy->pci_dev, vector);
     }
@@ -175,7 +186,8 @@ static int virtio_pci_load_queue(DeviceState *d, int n, QEMUFile *f)
 static int virtio_pci_set_host_notifier_internal(VirtIOPCIProxy *proxy,
                                                  int n, bool assign, bool set_handler)
 {
-    VirtQueue *vq = virtio_get_queue(proxy->vdev, n);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+    VirtQueue *vq = virtio_get_queue(vdev, n);
     EventNotifier *notifier = virtio_queue_get_host_notifier(vq);
     int r = 0;
 
@@ -200,6 +212,7 @@ static int virtio_pci_set_host_notifier_internal(VirtIOPCIProxy *proxy,
 
 static void virtio_pci_start_ioeventfd(VirtIOPCIProxy *proxy)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     int n, r;
 
     if (!(proxy->flags & VIRTIO_PCI_FLAG_USE_IOEVENTFD) ||
@@ -209,7 +222,7 @@ static void virtio_pci_start_ioeventfd(VirtIOPCIProxy *proxy)
     }
 
     for (n = 0; n < VIRTIO_PCI_QUEUE_MAX; n++) {
-        if (!virtio_queue_get_num(proxy->vdev, n)) {
+        if (!virtio_queue_get_num(vdev, n)) {
             continue;
         }
 
@@ -223,7 +236,7 @@ static void virtio_pci_start_ioeventfd(VirtIOPCIProxy *proxy)
 
 assign_error:
     while (--n >= 0) {
-        if (!virtio_queue_get_num(proxy->vdev, n)) {
+        if (!virtio_queue_get_num(vdev, n)) {
             continue;
         }
 
@@ -236,6 +249,7 @@ assign_error:
 
 static void virtio_pci_stop_ioeventfd(VirtIOPCIProxy *proxy)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     int r;
     int n;
 
@@ -244,7 +258,7 @@ static void virtio_pci_stop_ioeventfd(VirtIOPCIProxy *proxy)
     }
 
     for (n = 0; n < VIRTIO_PCI_QUEUE_MAX; n++) {
-        if (!virtio_queue_get_num(proxy->vdev, n)) {
+        if (!virtio_queue_get_num(vdev, n)) {
             continue;
         }
 
@@ -257,7 +271,7 @@ static void virtio_pci_stop_ioeventfd(VirtIOPCIProxy *proxy)
 static void virtio_ioport_write(void *opaque, uint32_t addr, uint32_t val)
 {
     VirtIOPCIProxy *proxy = opaque;
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     hwaddr pa;
 
     switch (addr) {
@@ -272,7 +286,7 @@ static void virtio_ioport_write(void *opaque, uint32_t addr, uint32_t val)
         pa = (hwaddr)val << VIRTIO_PCI_QUEUE_ADDR_SHIFT;
         if (pa == 0) {
             virtio_pci_stop_ioeventfd(proxy);
-            virtio_reset(proxy->vdev);
+            virtio_reset(vdev);
             msix_unuse_all_vectors(&proxy->pci_dev);
         }
         else
@@ -299,7 +313,7 @@ static void virtio_ioport_write(void *opaque, uint32_t addr, uint32_t val)
         }
 
         if (vdev->status == 0) {
-            virtio_reset(proxy->vdev);
+            virtio_reset(vdev);
             msix_unuse_all_vectors(&proxy->pci_dev);
         }
 
@@ -335,7 +349,7 @@ static void virtio_ioport_write(void *opaque, uint32_t addr, uint32_t val)
 
 static uint32_t virtio_ioport_read(VirtIOPCIProxy *proxy, uint32_t addr)
 {
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     uint32_t ret = 0xFFFFFFFF;
 
     switch (addr) {
@@ -381,6 +395,7 @@ static uint64_t virtio_pci_config_read(void *opaque, hwaddr addr,
                                        unsigned size)
 {
     VirtIOPCIProxy *proxy = opaque;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
     uint64_t val = 0;
     if (addr < config) {
@@ -390,16 +405,16 @@ static uint64_t virtio_pci_config_read(void *opaque, hwaddr addr,
 
     switch (size) {
     case 1:
-        val = virtio_config_readb(proxy->vdev, addr);
+        val = virtio_config_readb(vdev, addr);
         break;
     case 2:
-        val = virtio_config_readw(proxy->vdev, addr);
+        val = virtio_config_readw(vdev, addr);
         if (virtio_is_big_endian()) {
             val = bswap16(val);
         }
         break;
     case 4:
-        val = virtio_config_readl(proxy->vdev, addr);
+        val = virtio_config_readl(vdev, addr);
         if (virtio_is_big_endian()) {
             val = bswap32(val);
         }
@@ -413,6 +428,7 @@ static void virtio_pci_config_write(void *opaque, hwaddr addr,
 {
     VirtIOPCIProxy *proxy = opaque;
     uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     if (addr < config) {
         virtio_ioport_write(proxy, addr, val);
         return;
@@ -424,19 +440,19 @@ static void virtio_pci_config_write(void *opaque, hwaddr addr,
      */
     switch (size) {
     case 1:
-        virtio_config_writeb(proxy->vdev, addr, val);
+        virtio_config_writeb(vdev, addr, val);
         break;
     case 2:
         if (virtio_is_big_endian()) {
             val = bswap16(val);
         }
-        virtio_config_writew(proxy->vdev, addr, val);
+        virtio_config_writew(vdev, addr, val);
         break;
     case 4:
         if (virtio_is_big_endian()) {
             val = bswap32(val);
         }
-        virtio_config_writel(proxy->vdev, addr, val);
+        virtio_config_writel(vdev, addr, val);
         break;
     }
 }
@@ -455,6 +471,7 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,
                                 uint32_t val, int len)
 {
     VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
 
     pci_default_write_config(pci_dev, address, val, len);
 
@@ -462,8 +479,7 @@ static void virtio_write_config(PCIDevice *pci_dev, uint32_t address,
         !(pci_dev->config[PCI_COMMAND] & PCI_COMMAND_MASTER) &&
         !(proxy->flags & VIRTIO_PCI_FLAG_BUS_MASTER_BUG)) {
         virtio_pci_stop_ioeventfd(proxy);
-        virtio_set_status(proxy->vdev,
-                          proxy->vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK);
+        virtio_set_status(vdev, vdev->status & ~VIRTIO_CONFIG_S_DRIVER_OK);
     }
 }
 
@@ -506,7 +522,8 @@ static int kvm_virtio_pci_irqfd_use(VirtIOPCIProxy *proxy,
                                  unsigned int vector)
 {
     VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
-    VirtQueue *vq = virtio_get_queue(proxy->vdev, queue_no);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+    VirtQueue *vq = virtio_get_queue(vdev, queue_no);
     EventNotifier *n = virtio_queue_get_guest_notifier(vq);
     int ret;
     ret = kvm_irqchip_add_irqfd_notifier(kvm_state, n, NULL, irqfd->virq);
@@ -517,7 +534,8 @@ static void kvm_virtio_pci_irqfd_release(VirtIOPCIProxy *proxy,
                                       unsigned int queue_no,
                                       unsigned int vector)
 {
-    VirtQueue *vq = virtio_get_queue(proxy->vdev, queue_no);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+    VirtQueue *vq = virtio_get_queue(vdev, queue_no);
     EventNotifier *n = virtio_queue_get_guest_notifier(vq);
     VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
     int ret;
@@ -529,7 +547,7 @@ static void kvm_virtio_pci_irqfd_release(VirtIOPCIProxy *proxy,
 static int kvm_virtio_pci_vector_use(VirtIOPCIProxy *proxy, int nvqs)
 {
     PCIDevice *dev = &proxy->pci_dev;
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
     unsigned int vector;
     int ret, queue_no;
@@ -578,7 +596,7 @@ undo:
 static void kvm_virtio_pci_vector_release(VirtIOPCIProxy *proxy, int nvqs)
 {
     PCIDevice *dev = &proxy->pci_dev;
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     unsigned int vector;
     int queue_no;
     VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
@@ -606,8 +624,9 @@ static int virtio_pci_vq_vector_unmask(VirtIOPCIProxy *proxy,
                                        unsigned int vector,
                                        MSIMessage msg)
 {
-    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(proxy->vdev);
-    VirtQueue *vq = virtio_get_queue(proxy->vdev, queue_no);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
+    VirtQueue *vq = virtio_get_queue(vdev, queue_no);
     EventNotifier *n = virtio_queue_get_guest_notifier(vq);
     VirtIOIRQFD *irqfd;
     int ret = 0;
@@ -626,10 +645,10 @@ static int virtio_pci_vq_vector_unmask(VirtIOPCIProxy *proxy,
      * Otherwise, set it up now.
      */
     if (k->guest_notifier_mask) {
-        k->guest_notifier_mask(proxy->vdev, queue_no, false);
+        k->guest_notifier_mask(vdev, queue_no, false);
         /* Test after unmasking to avoid losing events. */
         if (k->guest_notifier_pending &&
-            k->guest_notifier_pending(proxy->vdev, queue_no)) {
+            k->guest_notifier_pending(vdev, queue_no)) {
             event_notifier_set(n);
         }
     } else {
@@ -642,13 +661,14 @@ static void virtio_pci_vq_vector_mask(VirtIOPCIProxy *proxy,
                                              unsigned int queue_no,
                                              unsigned int vector)
 {
-    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(proxy->vdev);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
 
     /* If guest supports masking, keep irqfd but mask it.
      * Otherwise, clean it up now.
      */ 
     if (k->guest_notifier_mask) {
-        k->guest_notifier_mask(proxy->vdev, queue_no, true);
+        k->guest_notifier_mask(vdev, queue_no, true);
     } else {
         kvm_virtio_pci_irqfd_release(proxy, queue_no, vector);
     }
@@ -658,7 +678,7 @@ static int virtio_pci_vector_unmask(PCIDevice *dev, unsigned vector,
                                     MSIMessage msg)
 {
     VirtIOPCIProxy *proxy = container_of(dev, VirtIOPCIProxy, pci_dev);
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     int ret, queue_no;
 
     for (queue_no = 0; queue_no < proxy->nvqs_with_notifiers; queue_no++) {
@@ -688,7 +708,7 @@ undo:
 static void virtio_pci_vector_mask(PCIDevice *dev, unsigned vector)
 {
     VirtIOPCIProxy *proxy = container_of(dev, VirtIOPCIProxy, pci_dev);
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     int queue_no;
 
     for (queue_no = 0; queue_no < proxy->nvqs_with_notifiers; queue_no++) {
@@ -707,7 +727,7 @@ static void virtio_pci_vector_poll(PCIDevice *dev,
                                    unsigned int vector_end)
 {
     VirtIOPCIProxy *proxy = container_of(dev, VirtIOPCIProxy, pci_dev);
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
     int queue_no;
     unsigned int vector;
@@ -739,8 +759,9 @@ static int virtio_pci_set_guest_notifier(DeviceState *d, int n, bool assign,
                                          bool with_irqfd)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
-    VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(proxy->vdev);
-    VirtQueue *vq = virtio_get_queue(proxy->vdev, n);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(vdev);
+    VirtQueue *vq = virtio_get_queue(vdev, n);
     EventNotifier *notifier = virtio_queue_get_guest_notifier(vq);
 
     if (assign) {
@@ -755,7 +776,7 @@ static int virtio_pci_set_guest_notifier(DeviceState *d, int n, bool assign,
     }
 
     if (!msix_enabled(&proxy->pci_dev) && vdc->guest_notifier_mask) {
-        vdc->guest_notifier_mask(proxy->vdev, n, !assign);
+        vdc->guest_notifier_mask(vdev, n, !assign);
     }
 
     return 0;
@@ -770,7 +791,7 @@ static bool virtio_pci_query_guest_notifiers(DeviceState *d)
 static int virtio_pci_set_guest_notifiers(DeviceState *d, int nvqs, bool assign)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
-    VirtIODevice *vdev = proxy->vdev;
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
     VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
     int r, n;
     bool with_irqfd = msix_enabled(&proxy->pci_dev) &&
@@ -864,11 +885,12 @@ static int virtio_pci_set_host_notifier(DeviceState *d, int n, bool assign)
 static void virtio_pci_vmstate_change(DeviceState *d, bool running)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
+    VirtIODevice *vdev = virtio_bus_get_device(&proxy->bus);
 
     if (running) {
         /* Try to find out if the guest has bus master disabled, but is
            in ready state. Then we have a buggy guest OS. */
-        if ((proxy->vdev->status & VIRTIO_CONFIG_S_DRIVER_OK) &&
+        if ((vdev->status & VIRTIO_CONFIG_S_DRIVER_OK) &&
             !(proxy->pci_dev.config[PCI_COMMAND] & PCI_COMMAND_MASTER)) {
             proxy->flags |= VIRTIO_PCI_FLAG_BUS_MASTER_BUG;
         }
@@ -943,8 +965,6 @@ static void virtio_pci_device_plugged(DeviceState *d)
     uint8_t *config;
     uint32_t size;
 
-    proxy->vdev = virtio_bus_get_device(bus);
-
     config = proxy->pci_dev.config;
     if (proxy->class_code) {
         pci_config_set_class(config, proxy->class_code);
diff --git a/hw/virtio/virtio-pci.h b/hw/virtio/virtio-pci.h
index 917bcc5..dc332ae 100644
--- a/hw/virtio/virtio-pci.h
+++ b/hw/virtio/virtio-pci.h
@@ -82,7 +82,6 @@ typedef struct VirtioPCIClass {
 
 struct VirtIOPCIProxy {
     PCIDevice pci_dev;
-    VirtIODevice *vdev;
     MemoryRegion bar;
     uint32_t flags;
     uint32_t class_code;
-- 
1.8.3.1

[Qemu-devel] [PATCH 05/30] virtio-bus: cleanup plug/unplug interface

[Paolo Bonzini]

Right now we have these pairs:

- virtio_bus_plug_device/virtio_bus_destroy_device.  The first
  takes a VirtIODevice, the second takes a VirtioBusState

- device_plugged/device_unplug callbacks in the VirtioBusClass
  (here it's just the naming that is inconsistent)

- virtio_bus_destroy_device is not called by anyone (and since
  it calls qdev_free, it would be called by the proxies---but
  then the callback is useless since the proxies can do whatever
  they want before calling virtio_bus_destroy_device)

And there is a k->init but no k->exit, hence virtio_device_exit is
overwritten by subclasses (except virtio-9p).  This cleans it up by:

- renaming the device_unplug callback to device_unplugged

- renaming virtio_bus_plug_device to virtio_bus_device_plugged,
  matching the callback name

- renaming virtio_bus_destroy_device to virtio_bus_device_unplugged,
  removing the qdev_free, making it take a VirtIODevice and calling it
  from virtio_device_exit

- adding a k->exit callback

virtio_device_exit is still overwritten, the next patches will fix that.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-bus.c         | 19 +++++++++----------
 hw/virtio/virtio.c             |  7 ++++++-
 include/hw/virtio/virtio-bus.h |  6 +++---
 include/hw/virtio/virtio.h     |  1 +
 4 files changed, 19 insertions(+), 14 deletions(-)

diff --git a/hw/virtio/virtio-bus.c b/hw/virtio/virtio-bus.c
index 17dd06e..eb77019 100644
--- a/hw/virtio/virtio-bus.c
+++ b/hw/virtio/virtio-bus.c
@@ -37,8 +37,8 @@ do { printf("virtio_bus: " fmt , ## __VA_ARGS__); } while (0)
 #define DPRINTF(fmt, ...) do { } while (0)
 #endif
 
-/* Plug the VirtIODevice */
-int virtio_bus_plug_device(VirtIODevice *vdev)
+/* A VirtIODevice is being plugged */
+int virtio_bus_device_plugged(VirtIODevice *vdev)
 {
     DeviceState *qdev = DEVICE(vdev);
     BusState *qbus = BUS(qdev_get_parent_bus(qdev));
@@ -64,20 +64,19 @@ void virtio_bus_reset(VirtioBusState *bus)
     }
 }
 
-/* Destroy the VirtIODevice */
-void virtio_bus_destroy_device(VirtioBusState *bus)
+/* A VirtIODevice is being unplugged */
+void virtio_bus_device_unplugged(VirtIODevice *vdev)
 {
-    BusState *qbus = BUS(bus);
-    VirtioBusClass *klass = VIRTIO_BUS_GET_CLASS(bus);
-    VirtIODevice *vdev = virtio_bus_get_device(bus);
+    DeviceState *qdev = DEVICE(vdev);
+    BusState *qbus = BUS(qdev_get_parent_bus(qdev));
+    VirtioBusClass *klass = VIRTIO_BUS_GET_CLASS(qbus);
 
     DPRINTF("%s: remove device.\n", qbus->name);
 
     if (vdev != NULL) {
-        if (klass->device_unplug != NULL) {
-            klass->device_unplug(qbus->parent);
+        if (klass->device_unplugged != NULL) {
+            klass->device_unplugged(qbus->parent);
         }
-        object_unparent(OBJECT(vdev));
     }
 }
 
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 2f1e73b..965b2c0 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -1158,14 +1158,19 @@ static int virtio_device_init(DeviceState *qdev)
     if (k->init(vdev) < 0) {
         return -1;
     }
-    virtio_bus_plug_device(vdev);
+    virtio_bus_device_plugged(vdev);
     return 0;
 }
 
 static int virtio_device_exit(DeviceState *qdev)
 {
     VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
+    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(qdev);
 
+    virtio_bus_device_unplugged(vdev);
+    if (k->exit) {
+        k->exit(vdev);
+    }
     if (vdev->bus_name) {
         g_free(vdev->bus_name);
         vdev->bus_name = NULL;
diff --git a/include/hw/virtio/virtio-bus.h b/include/hw/virtio/virtio-bus.h
index ba0f86a..0756545 100644
--- a/include/hw/virtio/virtio-bus.h
+++ b/include/hw/virtio/virtio-bus.h
@@ -61,7 +61,7 @@ typedef struct VirtioBusClass {
      * transport independent exit function.
      * This is called by virtio-bus just before the device is unplugged.
      */
-    void (*device_unplug)(DeviceState *d);
+    void (*device_unplugged)(DeviceState *d);
     /*
      * Does the transport have variable vring alignment?
      * (ie can it ever call virtio_queue_set_align()?)
@@ -74,9 +74,9 @@ struct VirtioBusState {
     BusState parent_obj;
 };
 
-int virtio_bus_plug_device(VirtIODevice *vdev);
+int virtio_bus_device_plugged(VirtIODevice *vdev);
 void virtio_bus_reset(VirtioBusState *bus);
-void virtio_bus_destroy_device(VirtioBusState *bus);
+void virtio_bus_device_unplugged(VirtIODevice *bus);
 /* Get the device id of the plugged device. */
 uint16_t virtio_bus_get_vdev_id(VirtioBusState *bus);
 /* Get the config_len field of the plugged device. */
diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
index a90522d..59756c2 100644
--- a/include/hw/virtio/virtio.h
+++ b/include/hw/virtio/virtio.h
@@ -127,6 +127,7 @@ typedef struct VirtioDeviceClass {
     /* This is what a VirtioDevice must implement */
     DeviceClass parent;
     int (*init)(VirtIODevice *vdev);
+    void (*exit)(VirtIODevice *vdev);
     uint32_t (*get_features)(VirtIODevice *vdev, uint32_t requested_features);
     uint32_t (*bad_features)(VirtIODevice *vdev);
     void (*set_features)(VirtIODevice *vdev, uint32_t val);
-- 
1.8.3.1

[Qemu-devel] [PATCH 06/30] virtio-blk: switch exit callback to VirtioDeviceClass

[Paolo Bonzini]

This ensures hot-unplug is handled properly by the proxy.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/block/virtio-blk.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index 13f6d82..7f0440f 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -728,20 +728,18 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
     return 0;
 }
 
-static int virtio_blk_device_exit(DeviceState *dev)
+static void virtio_blk_device_exit(VirtIODevice *vdev)
 {
-    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
-    VirtIOBlock *s = VIRTIO_BLK(dev);
+    VirtIOBlock *s = VIRTIO_BLK(vdev);
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
     remove_migration_state_change_notifier(&s->migration_state_notifier);
     virtio_blk_data_plane_destroy(s->dataplane);
     s->dataplane = NULL;
 #endif
     qemu_del_vm_change_state_handler(s->change);
-    unregister_savevm(dev, "virtio-blk", s);
+    unregister_savevm(DEVICE(vdev), "virtio-blk", s);
     blockdev_mark_auto_del(s->bs);
     virtio_cleanup(vdev);
-    return 0;
 }
 
 static Property virtio_blk_properties[] = {
@@ -753,10 +751,10 @@ static void virtio_blk_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_blk_device_exit;
     dc->props = virtio_blk_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
     vdc->init = virtio_blk_device_init;
+    vdc->exit = virtio_blk_device_exit;
     vdc->get_config = virtio_blk_update_config;
     vdc->set_config = virtio_blk_set_config;
     vdc->get_features = virtio_blk_get_features;
-- 
1.8.3.1

[Qemu-devel] [PATCH 07/30] virtio-serial: switch exit callback to VirtioDeviceClass

[Paolo Bonzini]

This ensures hot-unplug is handled properly by the proxy.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/char/virtio-serial-bus.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/hw/char/virtio-serial-bus.c b/hw/char/virtio-serial-bus.c
index 703f026..a7ede90 100644
--- a/hw/char/virtio-serial-bus.c
+++ b/hw/char/virtio-serial-bus.c
@@ -987,12 +987,11 @@ static const TypeInfo virtio_serial_port_type_info = {
     .class_init = virtio_serial_port_class_init,
 };
 
-static int virtio_serial_device_exit(DeviceState *dev)
+static void virtio_serial_device_exit(VirtIODevice *vdev)
 {
-    VirtIOSerial *vser = VIRTIO_SERIAL(dev);
-    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOSerial *vser = VIRTIO_SERIAL(vdev);
 
-    unregister_savevm(dev, "virtio-console", vser);
+    unregister_savevm(DEVICE(vdev), "virtio-console", vser);
 
     g_free(vser->ivqs);
     g_free(vser->ovqs);
@@ -1004,7 +1003,6 @@ static int virtio_serial_device_exit(DeviceState *dev)
         g_free(vser->post_load);
     }
     virtio_cleanup(vdev);
-    return 0;
 }
 
 static Property virtio_serial_properties[] = {
@@ -1016,10 +1014,10 @@ static void virtio_serial_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_serial_device_exit;
     dc->props = virtio_serial_properties;
     set_bit(DEVICE_CATEGORY_INPUT, dc->categories);
     vdc->init = virtio_serial_device_init;
+    vdc->exit = virtio_serial_device_exit;
     vdc->get_features = get_features;
     vdc->get_config = get_config;
     vdc->set_config = set_config;
-- 
1.8.3.1

[Qemu-devel] [PATCH 08/30] virtio-net: switch exit callback to VirtioDeviceClass

[Paolo Bonzini]

This ensures hot-unplug is handled properly by the proxy.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/net/virtio-net.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index b75c753..93a81eb 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -1570,16 +1570,15 @@ static int virtio_net_device_init(VirtIODevice *vdev)
     return 0;
 }
 
-static int virtio_net_device_exit(DeviceState *qdev)
+static void virtio_net_device_exit(VirtIODevice *vdev)
 {
-    VirtIONet *n = VIRTIO_NET(qdev);
-    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
+    VirtIONet *n = VIRTIO_NET(vdev);
     int i;
 
     /* This will stop vhost backend if appropriate. */
     virtio_net_set_status(vdev, 0);
 
-    unregister_savevm(qdev, "virtio-net", n);
+    unregister_savevm(DEVICE(vdev), "virtio-net", n);
 
     if (n->netclient_name) {
         g_free(n->netclient_name);
@@ -1610,8 +1609,6 @@ static int virtio_net_device_exit(DeviceState *qdev)
     g_free(n->vqs);
     qemu_del_nic(n->nic);
     virtio_cleanup(vdev);
-
-    return 0;
 }
 
 static void virtio_net_instance_init(Object *obj)
@@ -1638,10 +1635,10 @@ static void virtio_net_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_net_device_exit;
     dc->props = virtio_net_properties;
     set_bit(DEVICE_CATEGORY_NETWORK, dc->categories);
     vdc->init = virtio_net_device_init;
+    vdc->exit = virtio_net_device_exit;
     vdc->get_config = virtio_net_get_config;
     vdc->set_config = virtio_net_set_config;
     vdc->get_features = virtio_net_get_features;
-- 
1.8.3.1

[Qemu-devel] [PATCH 09/30] virtio-scsi: switch exit callback to VirtioDeviceClass

[Paolo Bonzini]

This ensures hot-unplug is handled properly by the proxy.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi/vhost-scsi.c            | 11 +++++------
 hw/scsi/virtio-scsi.c           | 15 +++++++--------
 include/hw/virtio/virtio-scsi.h |  2 +-
 3 files changed, 13 insertions(+), 15 deletions(-)

diff --git a/hw/scsi/vhost-scsi.c b/hw/scsi/vhost-scsi.c
index 9e770fb..5e3cc61 100644
--- a/hw/scsi/vhost-scsi.c
+++ b/hw/scsi/vhost-scsi.c
@@ -240,11 +240,10 @@ static int vhost_scsi_init(VirtIODevice *vdev)
     return 0;
 }
 
-static int vhost_scsi_exit(DeviceState *qdev)
+static void vhost_scsi_exit(VirtIODevice *vdev)
 {
-    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
-    VHostSCSI *s = VHOST_SCSI(qdev);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(qdev);
+    VHostSCSI *s = VHOST_SCSI(vdev);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
 
     migrate_del_blocker(s->migration_blocker);
     error_free(s->migration_blocker);
@@ -253,7 +252,7 @@ static int vhost_scsi_exit(DeviceState *qdev)
     vhost_scsi_set_status(vdev, 0);
 
     g_free(s->dev.vqs);
-    return virtio_scsi_common_exit(vs);
+    virtio_scsi_common_exit(vs);
 }
 
 static Property vhost_scsi_properties[] = {
@@ -265,10 +264,10 @@ static void vhost_scsi_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = vhost_scsi_exit;
     dc->props = vhost_scsi_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
     vdc->init = vhost_scsi_init;
+    vdc->exit = vhost_scsi_exit;
     vdc->get_features = vhost_scsi_get_features;
     vdc->set_config = vhost_scsi_set_config;
     vdc->set_status = vhost_scsi_set_status;
diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index 26d95a1..83344ea 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -644,22 +644,21 @@ static int virtio_scsi_device_init(VirtIODevice *vdev)
     return 0;
 }
 
-int virtio_scsi_common_exit(VirtIOSCSICommon *vs)
+void virtio_scsi_common_exit(VirtIOSCSICommon *vs)
 {
     VirtIODevice *vdev = VIRTIO_DEVICE(vs);
 
     g_free(vs->cmd_vqs);
     virtio_cleanup(vdev);
-    return 0;
 }
 
-static int virtio_scsi_device_exit(DeviceState *qdev)
+static void virtio_scsi_device_exit(VirtIODevice *vdev)
 {
-    VirtIOSCSI *s = VIRTIO_SCSI(qdev);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(qdev);
+    VirtIOSCSI *s = VIRTIO_SCSI(vdev);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
 
-    unregister_savevm(qdev, "virtio-scsi", s);
-    return virtio_scsi_common_exit(vs);
+    unregister_savevm(DEVICE(vdev), "virtio-scsi", s);
+    virtio_scsi_common_exit(vs);
 }
 
 static Property virtio_scsi_properties[] = {
@@ -680,10 +679,10 @@ static void virtio_scsi_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_scsi_device_exit;
     dc->props = virtio_scsi_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
     vdc->init = virtio_scsi_device_init;
+    vdc->exit = virtio_scsi_device_exit;
     vdc->set_config = virtio_scsi_set_config;
     vdc->get_features = virtio_scsi_get_features;
     vdc->reset = virtio_scsi_reset;
diff --git a/include/hw/virtio/virtio-scsi.h b/include/hw/virtio/virtio-scsi.h
index 9a98540..206c61d 100644
--- a/include/hw/virtio/virtio-scsi.h
+++ b/include/hw/virtio/virtio-scsi.h
@@ -187,6 +187,6 @@ typedef struct {
                                             VIRTIO_SCSI_F_CHANGE, true)
 
 int virtio_scsi_common_init(VirtIOSCSICommon *vs);
-int virtio_scsi_common_exit(VirtIOSCSICommon *vs);
+void virtio_scsi_common_exit(VirtIOSCSICommon *vs);
 
 #endif /* _QEMU_VIRTIO_SCSI_H */
-- 
1.8.3.1

[Qemu-devel] [PATCH 10/30] virtio-balloon: switch exit callback to VirtioDeviceClass

[Paolo Bonzini]

This ensures hot-unplug is handled properly by the proxy.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-balloon.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
index 9504877..d7a392d 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
@@ -370,16 +370,14 @@ static int virtio_balloon_device_init(VirtIODevice *vdev)
     return 0;
 }
 
-static int virtio_balloon_device_exit(DeviceState *qdev)
+static void virtio_balloon_device_exit(VirtIODevice *vdev)
 {
-    VirtIOBalloon *s = VIRTIO_BALLOON(qdev);
-    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
+    VirtIOBalloon *s = VIRTIO_BALLOON(vdev);
 
     balloon_stats_destroy_timer(s);
     qemu_remove_balloon_handler(s);
-    unregister_savevm(qdev, "virtio-balloon", s);
+    unregister_savevm(DEVICE(vdev), "virtio-balloon", s);
     virtio_cleanup(vdev);
-    return 0;
 }
 
 static Property virtio_balloon_properties[] = {
@@ -390,10 +388,10 @@ static void virtio_balloon_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_balloon_device_exit;
     dc->props = virtio_balloon_properties;
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
     vdc->init = virtio_balloon_device_init;
+    vdc->exit = virtio_balloon_device_exit;
     vdc->get_config = virtio_balloon_get_config;
     vdc->set_config = virtio_balloon_set_config;
     vdc->get_features = virtio_balloon_get_features;
-- 
1.8.3.1

[Qemu-devel] [PATCH 11/30] virtio-rng: switch exit callback to VirtioDeviceClass

[Paolo Bonzini]

This ensures hot-unplug is handled properly by the proxy.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-rng.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c
index b22ccf1..42ca568 100644
--- a/hw/virtio/virtio-rng.c
+++ b/hw/virtio/virtio-rng.c
@@ -190,16 +190,14 @@ static int virtio_rng_device_init(VirtIODevice *vdev)
     return 0;
 }
 
-static int virtio_rng_device_exit(DeviceState *qdev)
+static void virtio_rng_device_exit(VirtIODevice *vdev)
 {
-    VirtIORNG *vrng = VIRTIO_RNG(qdev);
-    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
+    VirtIORNG *vrng = VIRTIO_RNG(vdev);
 
     timer_del(vrng->rate_limit_timer);
     timer_free(vrng->rate_limit_timer);
-    unregister_savevm(qdev, "virtio-rng", vrng);
+    unregister_savevm(DEVICE(vdev), "virtio-rng", vrng);
     virtio_cleanup(vdev);
-    return 0;
 }
 
 static Property virtio_rng_properties[] = {
@@ -211,10 +209,10 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_rng_device_exit;
     dc->props = virtio_rng_properties;
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
     vdc->init = virtio_rng_device_init;
+    vdc->exit = virtio_rng_device_exit;
     vdc->get_features = get_features;
 }
 
-- 
1.8.3.1

[Qemu-devel] [PATCH 12/30] virtio-pci: add device_unplugged callback

[Paolo Bonzini]

This fixes a crash in hot-unplug of virtio-pci devices behind a PCIe
switch.  The crash happens because the ioeventfd is still set whent the
child is destroyed (destruction happens in postorder).  Then the proxy
tries to unset to ioeventfd, but the virtqueue structure that holds the
EventNotifier has been trashed in the meanwhile.  kvm_set_ioeventfd_pio
does not expect failure and aborts.

The fix is simply to move parts of uninitialization to a new
device_unplugged callback, which is called before the child is destroyed.

Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-pci.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 15b92e9..30c9f2b 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -1002,6 +1002,15 @@ static void virtio_pci_device_plugged(DeviceState *d)
                                                       proxy->host_features);
 }
 
+static void virtio_pci_device_unplugged(DeviceState *d)
+{
+    PCIDevice *pci_dev = PCI_DEVICE(d);
+    VirtIOPCIProxy *proxy = VIRTIO_PCI(d);
+
+    virtio_pci_stop_ioeventfd(proxy);
+    msix_uninit_exclusive_bar(pci_dev);
+}
+
 static int virtio_pci_init(PCIDevice *pci_dev)
 {
     VirtIOPCIProxy *dev = VIRTIO_PCI(pci_dev);
@@ -1016,9 +1025,7 @@ static int virtio_pci_init(PCIDevice *pci_dev)
 static void virtio_pci_exit(PCIDevice *pci_dev)
 {
     VirtIOPCIProxy *proxy = VIRTIO_PCI(pci_dev);
-    virtio_pci_stop_ioeventfd(proxy);
     memory_region_destroy(&proxy->bar);
-    msix_uninit_exclusive_bar(pci_dev);
 }
 
 static void virtio_pci_reset(DeviceState *qdev)
@@ -1553,6 +1560,7 @@ static void virtio_pci_bus_class_init(ObjectClass *klass, void *data)
     k->set_guest_notifiers = virtio_pci_set_guest_notifiers;
     k->vmstate_change = virtio_pci_vmstate_change;
     k->device_plugged = virtio_pci_device_plugged;
+    k->device_unplugged = virtio_pci_device_unplugged;
 }
 
 static const TypeInfo virtio_pci_bus_info = {
-- 
1.8.3.1

[Qemu-devel] [PATCH 13/30] virtio-blk-dataplane: Improve error reporting

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Return an Error so that it can be propagated later.

Tested-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
[AF: Rebased]
Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/block/dataplane/virtio-blk.c | 30 ++++++++++++++++--------------
 hw/block/dataplane/virtio-blk.h |  5 +++--
 hw/block/virtio-blk.c           | 15 +++++++++++++--
 3 files changed, 32 insertions(+), 18 deletions(-)

diff --git a/hw/block/dataplane/virtio-blk.c b/hw/block/dataplane/virtio-blk.c
index f2d7350..1e57f3a 100644
--- a/hw/block/dataplane/virtio-blk.c
+++ b/hw/block/dataplane/virtio-blk.c
@@ -380,8 +380,9 @@ static void start_data_plane_bh(void *opaque)
                        s, QEMU_THREAD_JOINABLE);
 }
 
-bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
-                                  VirtIOBlockDataPlane **dataplane)
+void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
+                                  VirtIOBlockDataPlane **dataplane,
+                                  Error **errp)
 {
     VirtIOBlockDataPlane *s;
     int fd;
@@ -389,33 +390,35 @@ bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
     *dataplane = NULL;
 
     if (!blk->data_plane) {
-        return true;
+        return;
     }
 
     if (blk->scsi) {
-        error_report("device is incompatible with x-data-plane, use scsi=off");
-        return false;
+        error_setg(errp,
+                   "device is incompatible with x-data-plane, use scsi=off");
+        return;
     }
 
     if (blk->config_wce) {
-        error_report("device is incompatible with x-data-plane, "
-                     "use config-wce=off");
-        return false;
+        error_setg(errp, "device is incompatible with x-data-plane, "
+                         "use config-wce=off");
+        return;
     }
 
     /* If dataplane is (re-)enabled while the guest is running there could be
      * block jobs that can conflict.
      */
     if (bdrv_in_use(blk->conf.bs)) {
-        error_report("cannot start dataplane thread while device is in use");
-        return false;
+        error_setg(errp,
+                   "cannot start dataplane thread while device is in use");
+        return;
     }
 
     fd = raw_get_aio_fd(blk->conf.bs);
     if (fd < 0) {
-        error_report("drive is incompatible with x-data-plane, "
-                     "use format=raw,cache=none,aio=native");
-        return false;
+        error_setg(errp, "drive is incompatible with x-data-plane, "
+                         "use format=raw,cache=none,aio=native");
+        return;
     }
 
     s = g_new0(VirtIOBlockDataPlane, 1);
@@ -427,7 +430,6 @@ bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
     bdrv_set_in_use(blk->conf.bs, 1);
 
     *dataplane = s;
-    return true;
 }
 
 void virtio_blk_data_plane_destroy(VirtIOBlockDataPlane *s)
diff --git a/hw/block/dataplane/virtio-blk.h b/hw/block/dataplane/virtio-blk.h
index c90e99f..1750c99 100644
--- a/hw/block/dataplane/virtio-blk.h
+++ b/hw/block/dataplane/virtio-blk.h
@@ -19,8 +19,9 @@
 
 typedef struct VirtIOBlockDataPlane VirtIOBlockDataPlane;
 
-bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
-                                  VirtIOBlockDataPlane **dataplane);
+void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
+                                  VirtIOBlockDataPlane **dataplane,
+                                  Error **errp);
 void virtio_blk_data_plane_destroy(VirtIOBlockDataPlane *s);
 void virtio_blk_data_plane_start(VirtIOBlockDataPlane *s);
 void virtio_blk_data_plane_stop(VirtIOBlockDataPlane *s);
diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index 7f0440f..7552ce3 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -657,6 +657,7 @@ static void virtio_blk_migration_state_changed(Notifier *notifier, void *data)
     VirtIOBlock *s = container_of(notifier, VirtIOBlock,
                                   migration_state_notifier);
     MigrationState *mig = data;
+    Error *err = NULL;
 
     if (migration_in_setup(mig)) {
         if (!s->dataplane) {
@@ -671,7 +672,11 @@ static void virtio_blk_migration_state_changed(Notifier *notifier, void *data)
         }
         bdrv_drain_all(); /* complete in-flight non-dataplane requests */
         virtio_blk_data_plane_create(VIRTIO_DEVICE(s), &s->blk,
-                                     &s->dataplane);
+                                     &s->dataplane, &err);
+        if (err != NULL) {
+            error_report("%s", error_get_pretty(err));
+            error_free(err);
+        }
     }
 }
 #endif /* CONFIG_VIRTIO_BLK_DATA_PLANE */
@@ -681,6 +686,9 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
     DeviceState *qdev = DEVICE(vdev);
     VirtIOBlock *s = VIRTIO_BLK(vdev);
     VirtIOBlkConf *blk = &(s->blk);
+#ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
+    Error *err = NULL;
+#endif
     static int virtio_blk_id;
 
     if (!blk->conf.bs) {
@@ -708,7 +716,10 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
 
     s->vq = virtio_add_queue(vdev, 128, virtio_blk_handle_output);
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
-    if (!virtio_blk_data_plane_create(vdev, blk, &s->dataplane)) {
+    virtio_blk_data_plane_create(vdev, blk, &s->dataplane, &err);
+    if (err != NULL) {
+        error_report("%s", error_get_pretty(err));
+        error_free(err);
         virtio_cleanup(vdev);
         return -1;
     }
-- 
1.8.3.1

[Qemu-devel] [PATCH 14/30] virtio-9p: QOM realize preparations

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Avoid unnecessary VIRTIO_DEVICE().

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/9pfs/virtio-9p-device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/9pfs/virtio-9p-device.c b/hw/9pfs/virtio-9p-device.c
index f0ffbe8..ea21655 100644
--- a/hw/9pfs/virtio-9p-device.c
+++ b/hw/9pfs/virtio-9p-device.c
@@ -49,7 +49,7 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
     FsDriverEntry *fse;
     V9fsPath path;
 
-    virtio_init(VIRTIO_DEVICE(s), "virtio-9p", VIRTIO_ID_9P,
+    virtio_init(vdev, "virtio-9p", VIRTIO_ID_9P,
                 sizeof(struct virtio_9p_config) + MAX_TAG_LEN);
 
     /* initialize pdu allocator */
-- 
1.8.3.1

[Qemu-devel] [PATCH 15/30] virtio-blk: QOM realize preparations

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Rename variable qdev -> dev since that's what realize's argument is called
by convention.

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/block/virtio-blk.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index 7552ce3..15ce6d3 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -683,8 +683,8 @@ static void virtio_blk_migration_state_changed(Notifier *notifier, void *data)
 
 static int virtio_blk_device_init(VirtIODevice *vdev)
 {
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIOBlock *s = VIRTIO_BLK(vdev);
+    DeviceState *dev = DEVICE(vdev);
+    VirtIOBlock *s = VIRTIO_BLK(dev);
     VirtIOBlkConf *blk = &(s->blk);
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
     Error *err = NULL;
@@ -728,14 +728,14 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
 #endif
 
     s->change = qemu_add_vm_change_state_handler(virtio_blk_dma_restart_cb, s);
-    register_savevm(qdev, "virtio-blk", virtio_blk_id++, 2,
+    register_savevm(dev, "virtio-blk", virtio_blk_id++, 2,
                     virtio_blk_save, virtio_blk_load, s);
     bdrv_set_dev_ops(s->bs, &virtio_block_ops, s);
     bdrv_set_buffer_alignment(s->bs, s->conf->logical_block_size);
 
     bdrv_iostatus_enable(s->bs);
 
-    add_boot_device_path(s->conf->bootindex, qdev, "/disk@0,0");
+    add_boot_device_path(s->conf->bootindex, dev, "/disk@0,0");
     return 0;
 }
 
-- 
1.8.3.1

[Qemu-devel] [PATCH 16/30] virtio-serial: QOM realize preparations

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/char/virtio-serial-bus.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/hw/char/virtio-serial-bus.c b/hw/char/virtio-serial-bus.c
index a7ede90..313a5d8 100644
--- a/hw/char/virtio-serial-bus.c
+++ b/hw/char/virtio-serial-bus.c
@@ -891,8 +891,9 @@ static int virtser_port_qdev_exit(DeviceState *qdev)
 
 static int virtio_serial_device_init(VirtIODevice *vdev)
 {
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIOSerial *vser = VIRTIO_SERIAL(vdev);
+    DeviceState *dev = DEVICE(vdev);
+    VirtIOSerial *vser = VIRTIO_SERIAL(dev);
+    BusState *bus;
     uint32_t i, max_supported_ports;
 
     if (!vser->serial.max_virtserial_ports) {
@@ -912,8 +913,9 @@ static int virtio_serial_device_init(VirtIODevice *vdev)
 
     /* Spawn a new virtio-serial bus on which the ports will ride as devices */
     qbus_create_inplace(&vser->bus, sizeof(vser->bus), TYPE_VIRTIO_SERIAL_BUS,
-                        qdev, vdev->bus_name);
-    vser->bus.qbus.allow_hotplug = 1;
+                        dev, vdev->bus_name);
+    bus = BUS(&vser->bus);
+    bus->allow_hotplug = 1;
     vser->bus.vser = vser;
     QTAILQ_INIT(&vser->ports);
 
@@ -961,7 +963,7 @@ static int virtio_serial_device_init(VirtIODevice *vdev)
      * Register for the savevm section with the virtio-console name
      * to preserve backward compat
      */
-    register_savevm(qdev, "virtio-console", -1, 3, virtio_serial_save,
+    register_savevm(dev, "virtio-console", -1, 3, virtio_serial_save,
                     virtio_serial_load, vser);
 
     return 0;
-- 
1.8.3.1

[Qemu-devel] [PATCH 17/30] virtio-net: QOM realize preparations

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Rename variable qdev -> dev since that's what realize's argument is
called by convention.

Avoid duplicate VIRTIO_DEVICE() cast.

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/net/virtio-net.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index 93a81eb..a669f83 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -1488,14 +1488,12 @@ void virtio_net_set_netclient_name(VirtIONet *n, const char *name,
 
 static int virtio_net_device_init(VirtIODevice *vdev)
 {
-    int i;
-
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIONet *n = VIRTIO_NET(vdev);
+    DeviceState *dev = DEVICE(vdev);
+    VirtIONet *n = VIRTIO_NET(dev);
     NetClientState *nc;
+    int i;
 
-    virtio_init(VIRTIO_DEVICE(n), "virtio-net", VIRTIO_ID_NET,
-                                  n->config_size);
+    virtio_init(vdev, "virtio-net", VIRTIO_ID_NET, n->config_size);
 
     n->max_queues = MAX(n->nic_conf.queues, 1);
     n->vqs = g_malloc0(sizeof(VirtIONetQueue) * n->max_queues);
@@ -1535,7 +1533,7 @@ static int virtio_net_device_init(VirtIODevice *vdev)
                               n->netclient_type, n->netclient_name, n);
     } else {
         n->nic = qemu_new_nic(&net_virtio_info, &n->nic_conf,
-                              object_get_typename(OBJECT(qdev)), qdev->id, n);
+                              object_get_typename(OBJECT(dev)), dev->id, n);
     }
 
     peer_test_vnet_hdr(n);
@@ -1562,11 +1560,11 @@ static int virtio_net_device_init(VirtIODevice *vdev)
     nc = qemu_get_queue(n->nic);
     nc->rxfilter_notify_enabled = 1;
 
-    n->qdev = qdev;
-    register_savevm(qdev, "virtio-net", -1, VIRTIO_NET_VM_VERSION,
+    n->qdev = dev;
+    register_savevm(dev, "virtio-net", -1, VIRTIO_NET_VM_VERSION,
                     virtio_net_save, virtio_net_load, n);
 
-    add_boot_device_path(n->nic_conf.bootindex, qdev, "/ethernet-phy@0");
+    add_boot_device_path(n->nic_conf.bootindex, dev, "/ethernet-phy@0");
     return 0;
 }
 
-- 
1.8.3.1

[Qemu-devel] [PATCH 18/30] virtio-balloon: QOM realize preparations

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Rename qdev -> dev since that's what realize's argument is called by
convention. No need to keep more "qdev" around than necessary.

Avoid duplicate VIRTIO_DEVICE() cast.

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-balloon.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
index d7a392d..e6f6b4b 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
@@ -339,8 +339,8 @@ static int virtio_balloon_load(QEMUFile *f, void *opaque, int version_id)
 
 static int virtio_balloon_device_init(VirtIODevice *vdev)
 {
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIOBalloon *s = VIRTIO_BALLOON(vdev);
+    DeviceState *dev = DEVICE(vdev);
+    VirtIOBalloon *s = VIRTIO_BALLOON(dev);
     int ret;
 
     virtio_init(vdev, "virtio-balloon", VIRTIO_ID_BALLOON, 8);
@@ -349,7 +349,7 @@ static int virtio_balloon_device_init(VirtIODevice *vdev)
                                    virtio_balloon_stat, s);
 
     if (ret < 0) {
-        virtio_cleanup(VIRTIO_DEVICE(s));
+        virtio_cleanup(vdev);
         return -1;
     }
 
@@ -357,13 +357,13 @@ static int virtio_balloon_device_init(VirtIODevice *vdev)
     s->dvq = virtio_add_queue(vdev, 128, virtio_balloon_handle_output);
     s->svq = virtio_add_queue(vdev, 128, virtio_balloon_receive_stats);
 
-    register_savevm(qdev, "virtio-balloon", -1, 1,
+    register_savevm(dev, "virtio-balloon", -1, 1,
                     virtio_balloon_save, virtio_balloon_load, s);
 
-    object_property_add(OBJECT(qdev), "guest-stats", "guest statistics",
+    object_property_add(OBJECT(dev), "guest-stats", "guest statistics",
                         balloon_stats_get_all, NULL, NULL, s, NULL);
 
-    object_property_add(OBJECT(qdev), "guest-stats-polling-interval", "int",
+    object_property_add(OBJECT(dev), "guest-stats-polling-interval", "int",
                         balloon_stats_get_poll_interval,
                         balloon_stats_set_poll_interval,
                         NULL, s, NULL);
-- 
1.8.3.1

[Qemu-devel] [PATCH 19/30] virtio-rng: QOM realize preparations

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Rename qdev -> dev because that's what realize's argument is called by
convention. No need to keep more "qdev" around than necessary.

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-rng.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c
index 42ca568..9e2a66a 100644
--- a/hw/virtio/virtio-rng.c
+++ b/hw/virtio/virtio-rng.c
@@ -135,8 +135,8 @@ static void check_rate_limit(void *opaque)
 
 static int virtio_rng_device_init(VirtIODevice *vdev)
 {
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIORNG *vrng = VIRTIO_RNG(vdev);
+    DeviceState *dev = DEVICE(vdev);
+    VirtIORNG *vrng = VIRTIO_RNG(dev);
     Error *local_err = NULL;
 
     if (!vrng->conf.period_ms > 0) {
@@ -148,12 +148,12 @@ static int virtio_rng_device_init(VirtIODevice *vdev)
     if (vrng->conf.rng == NULL) {
         vrng->conf.default_backend = RNG_RANDOM(object_new(TYPE_RNG_RANDOM));
 
-        object_property_add_child(OBJECT(qdev),
+        object_property_add_child(OBJECT(dev),
                                   "default-backend",
                                   OBJECT(vrng->conf.default_backend),
                                   NULL);
 
-        object_property_set_link(OBJECT(qdev),
+        object_property_set_link(OBJECT(dev),
                                  OBJECT(vrng->conf.default_backend),
                                  "rng", NULL);
     }
@@ -184,7 +184,7 @@ static int virtio_rng_device_init(VirtIODevice *vdev)
     timer_mod(vrng->rate_limit_timer,
                    qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) + vrng->conf.period_ms);
 
-    register_savevm(qdev, "virtio-rng", -1, 1, virtio_rng_save,
+    register_savevm(dev, "virtio-rng", -1, 1, virtio_rng_save,
                     virtio_rng_load, vrng);
 
     return 0;
-- 
1.8.3.1

[Qemu-devel] [PATCH 20/30] virtio-scsi: QOM realize preparations

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Rename qdev -> dev since that's what realize's argument is called by
convention. No need to keep more "qdev" around than necessary.

Avoid duplicate VIRTIO_DEVICE() cast.

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi/virtio-scsi.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index 83344ea..a5b957d 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -594,7 +594,7 @@ int virtio_scsi_common_init(VirtIOSCSICommon *s)
     VirtIODevice *vdev = VIRTIO_DEVICE(s);
     int i;
 
-    virtio_init(VIRTIO_DEVICE(s), "virtio-scsi", VIRTIO_ID_SCSI,
+    virtio_init(vdev, "virtio-scsi", VIRTIO_ID_SCSI,
                 sizeof(VirtIOSCSIConfig));
 
     s->cmd_vqs = g_malloc0(s->conf.num_queues * sizeof(VirtQueue *));
@@ -615,9 +615,9 @@ int virtio_scsi_common_init(VirtIOSCSICommon *s)
 
 static int virtio_scsi_device_init(VirtIODevice *vdev)
 {
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
-    VirtIOSCSI *s = VIRTIO_SCSI(vdev);
+    DeviceState *dev = DEVICE(vdev);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(dev);
+    VirtIOSCSI *s = VIRTIO_SCSI(dev);
     static int virtio_scsi_id;
     Error *err = NULL;
     int ret;
@@ -627,10 +627,10 @@ static int virtio_scsi_device_init(VirtIODevice *vdev)
         return ret;
     }
 
-    scsi_bus_new(&s->bus, sizeof(s->bus), qdev,
+    scsi_bus_new(&s->bus, sizeof(s->bus), dev,
                  &virtio_scsi_scsi_info, vdev->bus_name);
 
-    if (!qdev->hotplugged) {
+    if (!dev->hotplugged) {
         scsi_bus_legacy_handle_cmdline(&s->bus, &err);
         if (err != NULL) {
             error_free(err);
@@ -638,7 +638,7 @@ static int virtio_scsi_device_init(VirtIODevice *vdev)
         }
     }
 
-    register_savevm(qdev, "virtio-scsi", virtio_scsi_id++, 1,
+    register_savevm(dev, "virtio-scsi", virtio_scsi_id++, 1,
                     virtio_scsi_save, virtio_scsi_load, s);
 
     return 0;
-- 
1.8.3.1

[Qemu-devel] [PATCH 21/30] virtio: Start converting VirtioDevice to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Temporarily allow either VirtioDeviceClass::init or
VirtioDeviceClass::realize.

Introduce VirtioDeviceClass::unrealize for symmetry.

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio.c         | 42 +++++++++++++++++++++++++++---------------
 include/hw/virtio/virtio.h |  4 +++-
 2 files changed, 30 insertions(+), 16 deletions(-)

diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 965b2c0..4070b37 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -1150,40 +1150,52 @@ void virtio_device_set_child_bus_name(VirtIODevice *vdev, char *bus_name)
     }
 }
 
-static int virtio_device_init(DeviceState *qdev)
-{
-    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
-    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(qdev);
-    assert(k->init != NULL);
-    if (k->init(vdev) < 0) {
-        return -1;
+static void virtio_device_realize(DeviceState *dev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(dev);
+    Error *err = NULL;
+
+    assert(vdc->init != NULL || vdc->realize != NULL);
+    if (vdc->realize != NULL) {
+        vdc->realize(dev, &err);
+        if (err != NULL) {
+            error_propagate(errp, err);
+            return;
+        }
+    } else {
+        if (vdc->init(vdev) < 0) {
+            error_setg(errp, "Device initialization failed.");
+            return;
+        }
     }
     virtio_bus_device_plugged(vdev);
-    return 0;
 }
 
-static int virtio_device_exit(DeviceState *qdev)
+static void virtio_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
-    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(qdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(dev);
 
     virtio_bus_device_unplugged(vdev);
-    if (k->exit) {
+
+    if (k->exit != NULL) {
         k->exit(vdev);
     }
+
     if (vdev->bus_name) {
         g_free(vdev->bus_name);
         vdev->bus_name = NULL;
     }
-    return 0;
 }
 
 static void virtio_device_class_init(ObjectClass *klass, void *data)
 {
     /* Set the default value here. */
     DeviceClass *dc = DEVICE_CLASS(klass);
-    dc->init = virtio_device_init;
-    dc->exit = virtio_device_exit;
+
+    dc->realize = virtio_device_realize;
+    dc->unrealize = virtio_device_unrealize;
     dc->bus_type = TYPE_VIRTIO_BUS;
 }
 
diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
index 59756c2..54c5efe 100644
--- a/include/hw/virtio/virtio.h
+++ b/include/hw/virtio/virtio.h
@@ -124,10 +124,12 @@ struct VirtIODevice
 };
 
 typedef struct VirtioDeviceClass {
-    /* This is what a VirtioDevice must implement */
     DeviceClass parent;
+
+    /* This is what a VirtioDevice must implement */
     int (*init)(VirtIODevice *vdev);
     void (*exit)(VirtIODevice *vdev);
+    DeviceRealize realize;
     uint32_t (*get_features)(VirtIODevice *vdev, uint32_t requested_features);
     uint32_t (*bad_features)(VirtIODevice *vdev);
     void (*set_features)(VirtIODevice *vdev, uint32_t val);
-- 
1.8.3.1

[Qemu-devel] [PATCH 22/30] virtio-9p: Convert to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/9pfs/virtio-9p-device.c | 41 ++++++++++++++++++++---------------------
 1 file changed, 20 insertions(+), 21 deletions(-)

diff --git a/hw/9pfs/virtio-9p-device.c b/hw/9pfs/virtio-9p-device.c
index ea21655..15a4983 100644
--- a/hw/9pfs/virtio-9p-device.c
+++ b/hw/9pfs/virtio-9p-device.c
@@ -41,9 +41,10 @@ static void virtio_9p_get_config(VirtIODevice *vdev, uint8_t *config)
     g_free(cfg);
 }
 
-static int virtio_9p_device_init(VirtIODevice *vdev)
+static void virtio_9p_device_realize(DeviceState *dev, Error **errp)
 {
-    V9fsState *s = VIRTIO_9P(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    V9fsState *s = VIRTIO_9P(dev);
     int i, len;
     struct stat stat;
     FsDriverEntry *fse;
@@ -67,16 +68,16 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
 
     if (!fse) {
         /* We don't have a fsdev identified by fsdev_id */
-        fprintf(stderr, "Virtio-9p device couldn't find fsdev with the "
-                "id = %s\n",
-                s->fsconf.fsdev_id ? s->fsconf.fsdev_id : "NULL");
+        error_setg(errp, "Virtio-9p device couldn't find fsdev with the "
+                   "id = %s",
+                   s->fsconf.fsdev_id ? s->fsconf.fsdev_id : "NULL");
         goto out;
     }
 
     if (!s->fsconf.tag) {
         /* we haven't specified a mount_tag */
-        fprintf(stderr, "fsdev with id %s needs mount_tag arguments\n",
-                s->fsconf.fsdev_id);
+        error_setg(errp, "fsdev with id %s needs mount_tag arguments",
+                   s->fsconf.fsdev_id);
         goto out;
     }
 
@@ -85,8 +86,8 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
     s->ctx.exops.get_st_gen = NULL;
     len = strlen(s->fsconf.tag);
     if (len > MAX_TAG_LEN - 1) {
-        fprintf(stderr, "mount tag '%s' (%d bytes) is longer than "
-                "maximum (%d bytes)", s->fsconf.tag, len, MAX_TAG_LEN - 1);
+        error_setg(errp, "mount tag '%s' (%d bytes) is longer than "
+                   "maximum (%d bytes)", s->fsconf.tag, len, MAX_TAG_LEN - 1);
         goto out;
     }
 
@@ -99,12 +100,12 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
     qemu_co_rwlock_init(&s->rename_lock);
 
     if (s->ops->init(&s->ctx) < 0) {
-        fprintf(stderr, "Virtio-9p Failed to initialize fs-driver with id:%s"
-                " and export path:%s\n", s->fsconf.fsdev_id, s->ctx.fs_root);
+        error_setg(errp, "Virtio-9p Failed to initialize fs-driver with id:%s"
+                   " and export path:%s", s->fsconf.fsdev_id, s->ctx.fs_root);
         goto out;
     }
     if (v9fs_init_worker_threads() < 0) {
-        fprintf(stderr, "worker thread initialization failed\n");
+        error_setg(errp, "worker thread initialization failed");
         goto out;
     }
 
@@ -114,28 +115,25 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
      * use co-routines here.
      */
     if (s->ops->name_to_path(&s->ctx, NULL, "/", &path) < 0) {
-        fprintf(stderr,
-                "error in converting name to path %s", strerror(errno));
+        error_setg(errp,
+                   "error in converting name to path %s", strerror(errno));
         goto out;
     }
     if (s->ops->lstat(&s->ctx, &path, &stat)) {
-        fprintf(stderr, "share path %s does not exist\n", fse->path);
+        error_setg(errp, "share path %s does not exist", fse->path);
         goto out;
     } else if (!S_ISDIR(stat.st_mode)) {
-        fprintf(stderr, "share path %s is not a directory\n", fse->path);
+        error_setg(errp, "share path %s is not a directory", fse->path);
         goto out;
     }
     v9fs_path_free(&path);
 
-    return 0;
+    return;
 out:
     g_free(s->ctx.fs_root);
     g_free(s->tag);
     virtio_cleanup(vdev);
     v9fs_path_free(&path);
-
-    return -1;
-
 }
 
 /* virtio-9p device */
@@ -149,9 +147,10 @@ static void virtio_9p_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_9p_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
-    vdc->init = virtio_9p_device_init;
+    vdc->realize = virtio_9p_device_realize;
     vdc->get_features = virtio_9p_get_features;
     vdc->get_config = virtio_9p_get_config;
 }
-- 
1.8.3.1

[Qemu-devel] [PATCH 23/30] virtio-blk: Convert to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/block/virtio-blk.c     | 24 ++++++++++++------------
 tests/qdev-monitor-test.c |  4 +---
 2 files changed, 13 insertions(+), 15 deletions(-)

diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index 15ce6d3..cb29fdb 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -681,9 +681,9 @@ static void virtio_blk_migration_state_changed(Notifier *notifier, void *data)
 }
 #endif /* CONFIG_VIRTIO_BLK_DATA_PLANE */
 
-static int virtio_blk_device_init(VirtIODevice *vdev)
+static void virtio_blk_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *dev = DEVICE(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
     VirtIOBlock *s = VIRTIO_BLK(dev);
     VirtIOBlkConf *blk = &(s->blk);
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
@@ -692,18 +692,19 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
     static int virtio_blk_id;
 
     if (!blk->conf.bs) {
-        error_report("drive property not set");
-        return -1;
+        error_setg(errp, "drive property not set");
+        return;
     }
     if (!bdrv_is_inserted(blk->conf.bs)) {
-        error_report("Device needs media, but drive is empty");
-        return -1;
+        error_setg(errp, "Device needs media, but drive is empty");
+        return;
     }
 
     blkconf_serial(&blk->conf, &blk->serial);
     s->original_wce = bdrv_enable_write_cache(blk->conf.bs);
     if (blkconf_geometry(&blk->conf, NULL, 65535, 255, 255) < 0) {
-        return -1;
+        error_setg(errp, "Error setting geometry");
+        return;
     }
 
     virtio_init(vdev, "virtio-blk", VIRTIO_ID_BLOCK,
@@ -718,10 +719,9 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
     virtio_blk_data_plane_create(vdev, blk, &s->dataplane, &err);
     if (err != NULL) {
-        error_report("%s", error_get_pretty(err));
-        error_free(err);
+        error_propagate(errp, err);
         virtio_cleanup(vdev);
-        return -1;
+        return;
     }
     s->migration_state_notifier.notify = virtio_blk_migration_state_changed;
     add_migration_state_change_notifier(&s->migration_state_notifier);
@@ -736,7 +736,6 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
     bdrv_iostatus_enable(s->bs);
 
     add_boot_device_path(s->conf->bootindex, dev, "/disk@0,0");
-    return 0;
 }
 
 static void virtio_blk_device_exit(VirtIODevice *vdev)
@@ -762,9 +761,10 @@ static void virtio_blk_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_blk_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
-    vdc->init = virtio_blk_device_init;
+    vdc->realize = virtio_blk_device_realize;
     vdc->exit = virtio_blk_device_exit;
     vdc->get_config = virtio_blk_update_config;
     vdc->set_config = virtio_blk_set_config;
diff --git a/tests/qdev-monitor-test.c b/tests/qdev-monitor-test.c
index 33a8ea4..ba7f9cc 100644
--- a/tests/qdev-monitor-test.c
+++ b/tests/qdev-monitor-test.c
@@ -32,10 +32,8 @@ static void test_device_add(void)
                    "}}");
     g_assert(response);
     error = qdict_get_qdict(response, "error");
-    g_assert(!strcmp(qdict_get_try_str(error, "class") ?: "",
-                     "GenericError"));
     g_assert(!strcmp(qdict_get_try_str(error, "desc") ?: "",
-                     "Device initialization failed."));
+                     "Device needs media, but drive is empty"));
     QDECREF(response);
 
     /* Delete the drive */
-- 
1.8.3.1

[Qemu-devel] [PATCH 24/30] virtio-serial: Convert to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/char/virtio-serial-bus.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/hw/char/virtio-serial-bus.c b/hw/char/virtio-serial-bus.c
index 313a5d8..df96117 100644
--- a/hw/char/virtio-serial-bus.c
+++ b/hw/char/virtio-serial-bus.c
@@ -889,23 +889,24 @@ static int virtser_port_qdev_exit(DeviceState *qdev)
     return 0;
 }
 
-static int virtio_serial_device_init(VirtIODevice *vdev)
+static void virtio_serial_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *dev = DEVICE(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
     VirtIOSerial *vser = VIRTIO_SERIAL(dev);
     BusState *bus;
     uint32_t i, max_supported_ports;
 
     if (!vser->serial.max_virtserial_ports) {
-        return -1;
+        error_setg(errp, "Maximum number of serial ports not specified");
+        return;
     }
 
     /* Each port takes 2 queues, and one pair is for the control queue */
     max_supported_ports = VIRTIO_PCI_QUEUE_MAX / 2 - 1;
 
     if (vser->serial.max_virtserial_ports > max_supported_ports) {
-        error_report("maximum ports supported: %u", max_supported_ports);
-        return -1;
+        error_setg(errp, "maximum ports supported: %u", max_supported_ports);
+        return;
     }
 
     virtio_init(vdev, "virtio-serial", VIRTIO_ID_CONSOLE,
@@ -965,8 +966,6 @@ static int virtio_serial_device_init(VirtIODevice *vdev)
      */
     register_savevm(dev, "virtio-console", -1, 3, virtio_serial_save,
                     virtio_serial_load, vser);
-
-    return 0;
 }
 
 static void virtio_serial_port_class_init(ObjectClass *klass, void *data)
@@ -1016,9 +1015,10 @@ static void virtio_serial_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_serial_properties;
     set_bit(DEVICE_CATEGORY_INPUT, dc->categories);
-    vdc->init = virtio_serial_device_init;
+    vdc->realize = virtio_serial_device_realize;
     vdc->exit = virtio_serial_device_exit;
     vdc->get_features = get_features;
     vdc->get_config = get_config;
-- 
1.8.3.1

[Qemu-devel] [PATCH 25/30] virtio-net: Convert to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/net/virtio-net.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index a669f83..d9e4b8a 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -1486,9 +1486,9 @@ void virtio_net_set_netclient_name(VirtIONet *n, const char *name,
     n->netclient_type = g_strdup(type);
 }
 
-static int virtio_net_device_init(VirtIODevice *vdev)
+static void virtio_net_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *dev = DEVICE(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
     VirtIONet *n = VIRTIO_NET(dev);
     NetClientState *nc;
     int i;
@@ -1565,7 +1565,6 @@ static int virtio_net_device_init(VirtIODevice *vdev)
                     virtio_net_save, virtio_net_load, n);
 
     add_boot_device_path(n->nic_conf.bootindex, dev, "/ethernet-phy@0");
-    return 0;
 }
 
 static void virtio_net_device_exit(VirtIODevice *vdev)
@@ -1633,9 +1632,10 @@ static void virtio_net_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_net_properties;
     set_bit(DEVICE_CATEGORY_NETWORK, dc->categories);
-    vdc->init = virtio_net_device_init;
+    vdc->realize = virtio_net_device_realize;
     vdc->exit = virtio_net_device_exit;
     vdc->get_config = virtio_net_get_config;
     vdc->set_config = virtio_net_set_config;
-- 
1.8.3.1

[Qemu-devel] [PATCH 26/30] virtio-balloon: Convert to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-balloon.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
index e6f6b4b..755e188 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
@@ -337,9 +337,9 @@ static int virtio_balloon_load(QEMUFile *f, void *opaque, int version_id)
     return 0;
 }
 
-static int virtio_balloon_device_init(VirtIODevice *vdev)
+static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *dev = DEVICE(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
     VirtIOBalloon *s = VIRTIO_BALLOON(dev);
     int ret;
 
@@ -349,8 +349,9 @@ static int virtio_balloon_device_init(VirtIODevice *vdev)
                                    virtio_balloon_stat, s);
 
     if (ret < 0) {
+        error_setg(errp, "Adding balloon handler failed");
         virtio_cleanup(vdev);
-        return -1;
+        return;
     }
 
     s->ivq = virtio_add_queue(vdev, 128, virtio_balloon_handle_output);
@@ -367,7 +368,6 @@ static int virtio_balloon_device_init(VirtIODevice *vdev)
                         balloon_stats_get_poll_interval,
                         balloon_stats_set_poll_interval,
                         NULL, s, NULL);
-    return 0;
 }
 
 static void virtio_balloon_device_exit(VirtIODevice *vdev)
@@ -388,9 +388,10 @@ static void virtio_balloon_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_balloon_properties;
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
-    vdc->init = virtio_balloon_device_init;
+    vdc->realize = virtio_balloon_device_realize;
     vdc->exit = virtio_balloon_device_exit;
     vdc->get_config = virtio_balloon_get_config;
     vdc->set_config = virtio_balloon_set_config;
-- 
1.8.3.1

[Qemu-devel] [PATCH 27/30] virtio-rng: Convert to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/virtio/virtio-rng.c         | 24 +++++++++++-------------
 include/hw/virtio/virtio-rng.h |  2 ++
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c
index 9e2a66a..fd61cb2 100644
--- a/hw/virtio/virtio-rng.c
+++ b/hw/virtio/virtio-rng.c
@@ -133,16 +133,16 @@ static void check_rate_limit(void *opaque)
                    qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) + vrng->conf.period_ms);
 }
 
-static int virtio_rng_device_init(VirtIODevice *vdev)
+static void virtio_rng_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *dev = DEVICE(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
     VirtIORNG *vrng = VIRTIO_RNG(dev);
     Error *local_err = NULL;
 
     if (!vrng->conf.period_ms > 0) {
-        qerror_report(QERR_INVALID_PARAMETER_VALUE, "period",
-                      "a positive number");
-        return -1;
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE, "period",
+                  "a positive number");
+        return;
     }
 
     if (vrng->conf.rng == NULL) {
@@ -162,15 +162,14 @@ static int virtio_rng_device_init(VirtIODevice *vdev)
 
     vrng->rng = vrng->conf.rng;
     if (vrng->rng == NULL) {
-        qerror_report(QERR_INVALID_PARAMETER_VALUE, "rng", "a valid object");
-        return -1;
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE, "rng", "a valid object");
+        return;
     }
 
     rng_backend_open(vrng->rng, &local_err);
     if (local_err) {
-        qerror_report_err(local_err);
-        error_free(local_err);
-        return -1;
+        error_propagate(errp, local_err);
+        return;
     }
 
     vrng->vq = virtio_add_queue(vdev, 8, handle_input);
@@ -186,8 +185,6 @@ static int virtio_rng_device_init(VirtIODevice *vdev)
 
     register_savevm(dev, "virtio-rng", -1, 1, virtio_rng_save,
                     virtio_rng_load, vrng);
-
-    return 0;
 }
 
 static void virtio_rng_device_exit(VirtIODevice *vdev)
@@ -209,9 +206,10 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_rng_properties;
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
-    vdc->init = virtio_rng_device_init;
+    vdc->realize = virtio_rng_device_realize;
     vdc->exit = virtio_rng_device_exit;
     vdc->get_features = get_features;
 }
diff --git a/include/hw/virtio/virtio-rng.h b/include/hw/virtio/virtio-rng.h
index debaa15..14e85a5 100644
--- a/include/hw/virtio/virtio-rng.h
+++ b/include/hw/virtio/virtio-rng.h
@@ -18,6 +18,8 @@
 #define TYPE_VIRTIO_RNG "virtio-rng-device"
 #define VIRTIO_RNG(obj) \
         OBJECT_CHECK(VirtIORNG, (obj), TYPE_VIRTIO_RNG)
+#define VIRTIO_RNG_GET_PARENT_CLASS(obj) \
+        OBJECT_GET_PARENT_CLASS(obj, TYPE_VIRTIO_RNG)
 
 /* The Virtio ID for the virtio rng device */
 #define VIRTIO_ID_RNG    4
-- 
1.8.3.1

[Qemu-devel] [PATCH 28/30] virtio-scsi: Convert to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi/vhost-scsi.c            | 33 +++++++++++++++++----------------
 hw/scsi/virtio-scsi.c           | 32 +++++++++++++++-----------------
 include/hw/virtio/virtio-scsi.h |  2 +-
 3 files changed, 33 insertions(+), 34 deletions(-)

diff --git a/hw/scsi/vhost-scsi.c b/hw/scsi/vhost-scsi.c
index 5e3cc61..1f1c9f3 100644
--- a/hw/scsi/vhost-scsi.c
+++ b/hw/scsi/vhost-scsi.c
@@ -196,29 +196,31 @@ static void vhost_scsi_set_status(VirtIODevice *vdev, uint8_t val)
     }
 }
 
-static int vhost_scsi_init(VirtIODevice *vdev)
+static void vhost_scsi_realize(DeviceState *dev, Error **errp)
 {
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
-    VHostSCSI *s = VHOST_SCSI(vdev);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(dev);
+    VHostSCSI *s = VHOST_SCSI(dev);
+    Error *err = NULL;
     int vhostfd = -1;
     int ret;
 
     if (!vs->conf.wwpn) {
-        error_report("vhost-scsi: missing wwpn\n");
-        return -EINVAL;
+        error_setg(errp, "vhost-scsi: missing wwpn");
+        return;
     }
 
     if (vs->conf.vhostfd) {
         vhostfd = monitor_handle_fd_param(cur_mon, vs->conf.vhostfd);
         if (vhostfd == -1) {
-            error_report("vhost-scsi: unable to parse vhostfd\n");
-            return -EINVAL;
+            error_setg(errp, "vhost-scsi: unable to parse vhostfd");
+            return;
         }
     }
 
-    ret = virtio_scsi_common_init(vs);
-    if (ret < 0) {
-        return ret;
+    virtio_scsi_common_realize(dev, &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
     }
 
     s->dev.nvqs = VHOST_SCSI_VQ_NUM_FIXED + vs->conf.num_queues;
@@ -227,17 +229,15 @@ static int vhost_scsi_init(VirtIODevice *vdev)
 
     ret = vhost_dev_init(&s->dev, vhostfd, "/dev/vhost-scsi", true);
     if (ret < 0) {
-        error_report("vhost-scsi: vhost initialization failed: %s\n",
-                strerror(-ret));
-        return ret;
+        error_setg(errp, "vhost-scsi: vhost initialization failed: %s",
+                   strerror(-ret));
+        return;
     }
     s->dev.backend_features = 0;
 
     error_setg(&s->migration_blocker,
             "vhost-scsi does not support migration");
     migrate_add_blocker(s->migration_blocker);
-
-    return 0;
 }
 
 static void vhost_scsi_exit(VirtIODevice *vdev)
@@ -264,9 +264,10 @@ static void vhost_scsi_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = vhost_scsi_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
-    vdc->init = vhost_scsi_init;
+    vdc->realize = vhost_scsi_realize;
     vdc->exit = vhost_scsi_exit;
     vdc->get_features = vhost_scsi_get_features;
     vdc->set_config = vhost_scsi_set_config;
diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index a5b957d..0a41ddb 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -589,9 +589,10 @@ static struct SCSIBusInfo virtio_scsi_scsi_info = {
     .load_request = virtio_scsi_load_request,
 };
 
-int virtio_scsi_common_init(VirtIOSCSICommon *s)
+void virtio_scsi_common_realize(DeviceState *dev, Error **errp)
 {
-    VirtIODevice *vdev = VIRTIO_DEVICE(s);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOSCSICommon *s = VIRTIO_SCSI_COMMON(dev);
     int i;
 
     virtio_init(vdev, "virtio-scsi", VIRTIO_ID_SCSI,
@@ -609,22 +610,19 @@ int virtio_scsi_common_init(VirtIOSCSICommon *s)
         s->cmd_vqs[i] = virtio_add_queue(vdev, VIRTIO_SCSI_VQ_SIZE,
                                          virtio_scsi_handle_cmd);
     }
-
-    return 0;
 }
 
-static int virtio_scsi_device_init(VirtIODevice *vdev)
+static void virtio_scsi_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *dev = DEVICE(vdev);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(dev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
     VirtIOSCSI *s = VIRTIO_SCSI(dev);
     static int virtio_scsi_id;
     Error *err = NULL;
-    int ret;
 
-    ret = virtio_scsi_common_init(vs);
-    if (ret < 0) {
-        return ret;
+    virtio_scsi_common_realize(dev, &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
     }
 
     scsi_bus_new(&s->bus, sizeof(s->bus), dev,
@@ -633,20 +631,19 @@ static int virtio_scsi_device_init(VirtIODevice *vdev)
     if (!dev->hotplugged) {
         scsi_bus_legacy_handle_cmdline(&s->bus, &err);
         if (err != NULL) {
-            error_free(err);
-            return -1;
+            error_propagate(errp, err);
+            return;
         }
     }
 
     register_savevm(dev, "virtio-scsi", virtio_scsi_id++, 1,
                     virtio_scsi_save, virtio_scsi_load, s);
-
-    return 0;
 }
 
 void virtio_scsi_common_exit(VirtIOSCSICommon *vs)
 {
-    VirtIODevice *vdev = VIRTIO_DEVICE(vs);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(dev);
 
     g_free(vs->cmd_vqs);
     virtio_cleanup(vdev);
@@ -679,9 +676,10 @@ static void virtio_scsi_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_scsi_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
-    vdc->init = virtio_scsi_device_init;
+    vdc->realize = virtio_scsi_device_realize;
     vdc->exit = virtio_scsi_device_exit;
     vdc->set_config = virtio_scsi_set_config;
     vdc->get_features = virtio_scsi_get_features;
diff --git a/include/hw/virtio/virtio-scsi.h b/include/hw/virtio/virtio-scsi.h
index 206c61d..da343f1 100644
--- a/include/hw/virtio/virtio-scsi.h
+++ b/include/hw/virtio/virtio-scsi.h
@@ -186,7 +186,7 @@ typedef struct {
     DEFINE_PROP_BIT("param_change", _state, _feature_field,                    \
                                             VIRTIO_SCSI_F_CHANGE, true)
 
-int virtio_scsi_common_init(VirtIOSCSICommon *vs);
+void virtio_scsi_common_realize(DeviceState *dev, Error **errp);
 void virtio_scsi_common_exit(VirtIOSCSICommon *vs);
 
 #endif /* _QEMU_VIRTIO_SCSI_H */
-- 
1.8.3.1

[Qemu-devel] [PATCH 29/30] virtio: Complete converting VirtioDevice to QOM realize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Drop VirtioDeviceClass::init.

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi/virtio-scsi.c      | 3 +--
 hw/virtio/virtio.c         | 6 ------
 include/hw/virtio/virtio.h | 5 +++--
 3 files changed, 4 insertions(+), 10 deletions(-)

diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index 0a41ddb..a208cf3 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -642,8 +642,7 @@ static void virtio_scsi_device_realize(DeviceState *dev, Error **errp)
 
 void virtio_scsi_common_exit(VirtIOSCSICommon *vs)
 {
-    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(dev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(vs);
 
     g_free(vs->cmd_vqs);
     virtio_cleanup(vdev);
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 4070b37..1dba284 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -1156,18 +1156,12 @@ static void virtio_device_realize(DeviceState *dev, Error **errp)
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(dev);
     Error *err = NULL;
 
-    assert(vdc->init != NULL || vdc->realize != NULL);
     if (vdc->realize != NULL) {
         vdc->realize(dev, &err);
         if (err != NULL) {
             error_propagate(errp, err);
             return;
         }
-    } else {
-        if (vdc->init(vdev) < 0) {
-            error_setg(errp, "Device initialization failed.");
-            return;
-        }
     }
     virtio_bus_device_plugged(vdev);
 }
diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
index 54c5efe..6da88c0 100644
--- a/include/hw/virtio/virtio.h
+++ b/include/hw/virtio/virtio.h
@@ -124,12 +124,13 @@ struct VirtIODevice
 };
 
 typedef struct VirtioDeviceClass {
+    /*< private >*/
     DeviceClass parent;
+    /*< public >*/
 
     /* This is what a VirtioDevice must implement */
-    int (*init)(VirtIODevice *vdev);
-    void (*exit)(VirtIODevice *vdev);
     DeviceRealize realize;
+    void (*exit)(VirtIODevice *vdev);
     uint32_t (*get_features)(VirtIODevice *vdev, uint32_t requested_features);
     uint32_t (*bad_features)(VirtIODevice *vdev);
     void (*set_features)(VirtIODevice *vdev, uint32_t val);
-- 
1.8.3.1

[Qemu-devel] [PATCH 30/30] virtio: Convert exit to unrealize

[Paolo Bonzini]

From: Andreas Färber <afaerber@suse.de>

Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/block/virtio-blk.c           | 10 ++++++----
 hw/char/virtio-serial-bus.c     |  9 +++++----
 hw/net/virtio-net.c             |  9 +++++----
 hw/scsi/vhost-scsi.c            | 11 ++++++-----
 hw/scsi/virtio-scsi.c           | 17 +++++++++--------
 hw/virtio/virtio-balloon.c      |  9 +++++----
 hw/virtio/virtio-rng.c          |  9 +++++----
 hw/virtio/virtio.c              | 13 ++++++++-----
 include/hw/virtio/virtio-scsi.h |  2 +-
 include/hw/virtio/virtio.h      |  2 +-
 10 files changed, 51 insertions(+), 40 deletions(-)

diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index cb29fdb..19d0961 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -738,16 +738,18 @@ static void virtio_blk_device_realize(DeviceState *dev, Error **errp)
     add_boot_device_path(s->conf->bootindex, dev, "/disk@0,0");
 }
 
-static void virtio_blk_device_exit(VirtIODevice *vdev)
+static void virtio_blk_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIOBlock *s = VIRTIO_BLK(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOBlock *s = VIRTIO_BLK(dev);
+
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
     remove_migration_state_change_notifier(&s->migration_state_notifier);
     virtio_blk_data_plane_destroy(s->dataplane);
     s->dataplane = NULL;
 #endif
     qemu_del_vm_change_state_handler(s->change);
-    unregister_savevm(DEVICE(vdev), "virtio-blk", s);
+    unregister_savevm(dev, "virtio-blk", s);
     blockdev_mark_auto_del(s->bs);
     virtio_cleanup(vdev);
 }
@@ -765,7 +767,7 @@ static void virtio_blk_class_init(ObjectClass *klass, void *data)
     dc->props = virtio_blk_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
     vdc->realize = virtio_blk_device_realize;
-    vdc->exit = virtio_blk_device_exit;
+    vdc->unrealize = virtio_blk_device_unrealize;
     vdc->get_config = virtio_blk_update_config;
     vdc->set_config = virtio_blk_set_config;
     vdc->get_features = virtio_blk_get_features;
diff --git a/hw/char/virtio-serial-bus.c b/hw/char/virtio-serial-bus.c
index df96117..226e9f9 100644
--- a/hw/char/virtio-serial-bus.c
+++ b/hw/char/virtio-serial-bus.c
@@ -988,11 +988,12 @@ static const TypeInfo virtio_serial_port_type_info = {
     .class_init = virtio_serial_port_class_init,
 };
 
-static void virtio_serial_device_exit(VirtIODevice *vdev)
+static void virtio_serial_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIOSerial *vser = VIRTIO_SERIAL(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOSerial *vser = VIRTIO_SERIAL(dev);
 
-    unregister_savevm(DEVICE(vdev), "virtio-console", vser);
+    unregister_savevm(dev, "virtio-console", vser);
 
     g_free(vser->ivqs);
     g_free(vser->ovqs);
@@ -1019,7 +1020,7 @@ static void virtio_serial_class_init(ObjectClass *klass, void *data)
     dc->props = virtio_serial_properties;
     set_bit(DEVICE_CATEGORY_INPUT, dc->categories);
     vdc->realize = virtio_serial_device_realize;
-    vdc->exit = virtio_serial_device_exit;
+    vdc->unrealize = virtio_serial_device_unrealize;
     vdc->get_features = get_features;
     vdc->get_config = get_config;
     vdc->set_config = set_config;
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index d9e4b8a..2b193bf 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -1567,15 +1567,16 @@ static void virtio_net_device_realize(DeviceState *dev, Error **errp)
     add_boot_device_path(n->nic_conf.bootindex, dev, "/ethernet-phy@0");
 }
 
-static void virtio_net_device_exit(VirtIODevice *vdev)
+static void virtio_net_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIONet *n = VIRTIO_NET(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIONet *n = VIRTIO_NET(dev);
     int i;
 
     /* This will stop vhost backend if appropriate. */
     virtio_net_set_status(vdev, 0);
 
-    unregister_savevm(DEVICE(vdev), "virtio-net", n);
+    unregister_savevm(dev, "virtio-net", n);
 
     if (n->netclient_name) {
         g_free(n->netclient_name);
@@ -1636,7 +1637,7 @@ static void virtio_net_class_init(ObjectClass *klass, void *data)
     dc->props = virtio_net_properties;
     set_bit(DEVICE_CATEGORY_NETWORK, dc->categories);
     vdc->realize = virtio_net_device_realize;
-    vdc->exit = virtio_net_device_exit;
+    vdc->unrealize = virtio_net_device_unrealize;
     vdc->get_config = virtio_net_get_config;
     vdc->set_config = virtio_net_set_config;
     vdc->get_features = virtio_net_get_features;
diff --git a/hw/scsi/vhost-scsi.c b/hw/scsi/vhost-scsi.c
index 1f1c9f3..3983a5b 100644
--- a/hw/scsi/vhost-scsi.c
+++ b/hw/scsi/vhost-scsi.c
@@ -240,10 +240,10 @@ static void vhost_scsi_realize(DeviceState *dev, Error **errp)
     migrate_add_blocker(s->migration_blocker);
 }
 
-static void vhost_scsi_exit(VirtIODevice *vdev)
+static void vhost_scsi_unrealize(DeviceState *dev, Error **errp)
 {
-    VHostSCSI *s = VHOST_SCSI(vdev);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VHostSCSI *s = VHOST_SCSI(dev);
 
     migrate_del_blocker(s->migration_blocker);
     error_free(s->migration_blocker);
@@ -252,7 +252,8 @@ static void vhost_scsi_exit(VirtIODevice *vdev)
     vhost_scsi_set_status(vdev, 0);
 
     g_free(s->dev.vqs);
-    virtio_scsi_common_exit(vs);
+
+    virtio_scsi_common_unrealize(dev, errp);
 }
 
 static Property vhost_scsi_properties[] = {
@@ -268,7 +269,7 @@ static void vhost_scsi_class_init(ObjectClass *klass, void *data)
     dc->props = vhost_scsi_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
     vdc->realize = vhost_scsi_realize;
-    vdc->exit = vhost_scsi_exit;
+    vdc->unrealize = vhost_scsi_unrealize;
     vdc->get_features = vhost_scsi_get_features;
     vdc->set_config = vhost_scsi_set_config;
     vdc->set_status = vhost_scsi_set_status;
diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index a208cf3..6dcdd1b 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -640,21 +640,22 @@ static void virtio_scsi_device_realize(DeviceState *dev, Error **errp)
                     virtio_scsi_save, virtio_scsi_load, s);
 }
 
-void virtio_scsi_common_exit(VirtIOSCSICommon *vs)
+void virtio_scsi_common_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIODevice *vdev = VIRTIO_DEVICE(vs);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(dev);
 
     g_free(vs->cmd_vqs);
     virtio_cleanup(vdev);
 }
 
-static void virtio_scsi_device_exit(VirtIODevice *vdev)
+static void virtio_scsi_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIOSCSI *s = VIRTIO_SCSI(vdev);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
+    VirtIOSCSI *s = VIRTIO_SCSI(dev);
+
+    unregister_savevm(dev, "virtio-scsi", s);
 
-    unregister_savevm(DEVICE(vdev), "virtio-scsi", s);
-    virtio_scsi_common_exit(vs);
+    virtio_scsi_common_unrealize(dev, errp);
 }
 
 static Property virtio_scsi_properties[] = {
@@ -679,7 +680,7 @@ static void virtio_scsi_class_init(ObjectClass *klass, void *data)
     dc->props = virtio_scsi_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
     vdc->realize = virtio_scsi_device_realize;
-    vdc->exit = virtio_scsi_device_exit;
+    vdc->unrealize = virtio_scsi_device_unrealize;
     vdc->set_config = virtio_scsi_set_config;
     vdc->get_features = virtio_scsi_get_features;
     vdc->reset = virtio_scsi_reset;
diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
index 755e188..d9754db 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
@@ -370,13 +370,14 @@ static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)
                         NULL, s, NULL);
 }
 
-static void virtio_balloon_device_exit(VirtIODevice *vdev)
+static void virtio_balloon_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIOBalloon *s = VIRTIO_BALLOON(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOBalloon *s = VIRTIO_BALLOON(dev);
 
     balloon_stats_destroy_timer(s);
     qemu_remove_balloon_handler(s);
-    unregister_savevm(DEVICE(vdev), "virtio-balloon", s);
+    unregister_savevm(dev, "virtio-balloon", s);
     virtio_cleanup(vdev);
 }
 
@@ -392,7 +393,7 @@ static void virtio_balloon_class_init(ObjectClass *klass, void *data)
     dc->props = virtio_balloon_properties;
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
     vdc->realize = virtio_balloon_device_realize;
-    vdc->exit = virtio_balloon_device_exit;
+    vdc->unrealize = virtio_balloon_device_unrealize;
     vdc->get_config = virtio_balloon_get_config;
     vdc->set_config = virtio_balloon_set_config;
     vdc->get_features = virtio_balloon_get_features;
diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c
index fd61cb2..755fdee 100644
--- a/hw/virtio/virtio-rng.c
+++ b/hw/virtio/virtio-rng.c
@@ -187,13 +187,14 @@ static void virtio_rng_device_realize(DeviceState *dev, Error **errp)
                     virtio_rng_load, vrng);
 }
 
-static void virtio_rng_device_exit(VirtIODevice *vdev)
+static void virtio_rng_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIORNG *vrng = VIRTIO_RNG(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIORNG *vrng = VIRTIO_RNG(dev);
 
     timer_del(vrng->rate_limit_timer);
     timer_free(vrng->rate_limit_timer);
-    unregister_savevm(DEVICE(vdev), "virtio-rng", vrng);
+    unregister_savevm(dev, "virtio-rng", vrng);
     virtio_cleanup(vdev);
 }
 
@@ -210,7 +211,7 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data)
     dc->props = virtio_rng_properties;
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
     vdc->realize = virtio_rng_device_realize;
-    vdc->exit = virtio_rng_device_exit;
+    vdc->unrealize = virtio_rng_device_unrealize;
     vdc->get_features = get_features;
 }
 
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 1dba284..144b9ca 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -1169,12 +1169,15 @@ static void virtio_device_realize(DeviceState *dev, Error **errp)
 static void virtio_device_unrealize(DeviceState *dev, Error **errp)
 {
     VirtIODevice *vdev = VIRTIO_DEVICE(dev);
-    VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(dev);
-
-    virtio_bus_device_unplugged(vdev);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(dev);
+    Error *err = NULL;
 
-    if (k->exit != NULL) {
-        k->exit(vdev);
+    if (vdc->unrealize != NULL) {
+        vdc->unrealize(dev, &err);
+        if (err != NULL) {
+            error_propagate(errp, err);
+            return;
+        }
     }
 
     if (vdev->bus_name) {
diff --git a/include/hw/virtio/virtio-scsi.h b/include/hw/virtio/virtio-scsi.h
index da343f1..42b1024 100644
--- a/include/hw/virtio/virtio-scsi.h
+++ b/include/hw/virtio/virtio-scsi.h
@@ -187,6 +187,6 @@ typedef struct {
                                             VIRTIO_SCSI_F_CHANGE, true)
 
 void virtio_scsi_common_realize(DeviceState *dev, Error **errp);
-void virtio_scsi_common_exit(VirtIOSCSICommon *vs);
+void virtio_scsi_common_unrealize(DeviceState *dev, Error **errp);
 
 #endif /* _QEMU_VIRTIO_SCSI_H */
diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
index 6da88c0..3e54e90 100644
--- a/include/hw/virtio/virtio.h
+++ b/include/hw/virtio/virtio.h
@@ -130,7 +130,7 @@ typedef struct VirtioDeviceClass {
 
     /* This is what a VirtioDevice must implement */
     DeviceRealize realize;
-    void (*exit)(VirtIODevice *vdev);
+    DeviceUnrealize unrealize;
     uint32_t (*get_features)(VirtIODevice *vdev, uint32_t requested_features);
     uint32_t (*bad_features)(VirtIODevice *vdev);
     void (*set_features)(VirtIODevice *vdev, uint32_t val);
-- 
1.8.3.1

Re: [Qemu-devel] [PATCH 23/30] virtio-blk: Convert to QOM realize

[Andreas Färber]

Am 29.11.2013 11:17, schrieb Paolo Bonzini:
> diff --git a/tests/qdev-monitor-test.c b/tests/qdev-monitor-test.c
> index 33a8ea4..ba7f9cc 100644
> --- a/tests/qdev-monitor-test.c
> +++ b/tests/qdev-monitor-test.c
> @@ -32,10 +32,8 @@ static void test_device_add(void)
>                     "}}");
>      g_assert(response);
>      error = qdict_get_qdict(response, "error");
> -    g_assert(!strcmp(qdict_get_try_str(error, "class") ?: "",
> -                     "GenericError"));
>      g_assert(!strcmp(qdict_get_try_str(error, "desc") ?: "",
> -                     "Device initialization failed."));
> +                     "Device needs media, but drive is empty"));
>      QDECREF(response);
>  
>      /* Delete the drive */

Thanks for finding and fixing this qtest failure!

Andreas

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Re: [Qemu-devel] [PATCH 27/30] virtio-rng: Convert to QOM realize

[Andreas Färber]

Am 29.11.2013 11:17, schrieb Paolo Bonzini:
> From: Andreas Färber <afaerber@suse.de>
> 
> Signed-off-by: Andreas Färber <afaerber@suse.de>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> ---
>  hw/virtio/virtio-rng.c         | 24 +++++++++++-------------
>  include/hw/virtio/virtio-rng.h |  2 ++
>  2 files changed, 13 insertions(+), 13 deletions(-)
> 
> diff --git a/hw/virtio/virtio-rng.c b/hw/virtio/virtio-rng.c
> index 9e2a66a..fd61cb2 100644
> --- a/hw/virtio/virtio-rng.c
> +++ b/hw/virtio/virtio-rng.c
> @@ -133,16 +133,16 @@ static void check_rate_limit(void *opaque)
>                     qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) + vrng->conf.period_ms);
>  }
>  
> -static int virtio_rng_device_init(VirtIODevice *vdev)
> +static void virtio_rng_device_realize(DeviceState *dev, Error **errp)
>  {
> -    DeviceState *dev = DEVICE(vdev);
> +    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
>      VirtIORNG *vrng = VIRTIO_RNG(dev);
>      Error *local_err = NULL;
>  
>      if (!vrng->conf.period_ms > 0) {
> -        qerror_report(QERR_INVALID_PARAMETER_VALUE, "period",
> -                      "a positive number");
> -        return -1;
> +        error_set(errp, QERR_INVALID_PARAMETER_VALUE, "period",
> +                  "a positive number");
> +        return;

This matches the rebase I did last night, thanks. However...

>      }
>  
>      if (vrng->conf.rng == NULL) {
> @@ -162,15 +162,14 @@ static int virtio_rng_device_init(VirtIODevice *vdev)
>  
>      vrng->rng = vrng->conf.rng;
>      if (vrng->rng == NULL) {
> -        qerror_report(QERR_INVALID_PARAMETER_VALUE, "rng", "a valid object");
> -        return -1;
> +        error_set(errp, QERR_INVALID_PARAMETER_VALUE, "rng", "a valid object");
> +        return;
>      }
>  
>      rng_backend_open(vrng->rng, &local_err);
>      if (local_err) {
> -        qerror_report_err(local_err);
> -        error_free(local_err);
> -        return -1;
> +        error_propagate(errp, local_err);
> +        return;
>      }
>  
>      vrng->vq = virtio_add_queue(vdev, 8, handle_input);
> @@ -186,8 +185,6 @@ static int virtio_rng_device_init(VirtIODevice *vdev)
>  
>      register_savevm(dev, "virtio-rng", -1, 1, virtio_rng_save,
>                      virtio_rng_load, vrng);
> -
> -    return 0;
>  }
>  
>  static void virtio_rng_device_exit(VirtIODevice *vdev)
> @@ -209,9 +206,10 @@ static void virtio_rng_class_init(ObjectClass *klass, void *data)
>  {
>      DeviceClass *dc = DEVICE_CLASS(klass);
>      VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
> +
>      dc->props = virtio_rng_properties;
>      set_bit(DEVICE_CATEGORY_MISC, dc->categories);
> -    vdc->init = virtio_rng_device_init;
> +    vdc->realize = virtio_rng_device_realize;
>      vdc->exit = virtio_rng_device_exit;
>      vdc->get_features = get_features;
>  }
> diff --git a/include/hw/virtio/virtio-rng.h b/include/hw/virtio/virtio-rng.h
> index debaa15..14e85a5 100644
> --- a/include/hw/virtio/virtio-rng.h
> +++ b/include/hw/virtio/virtio-rng.h
> @@ -18,6 +18,8 @@
>  #define TYPE_VIRTIO_RNG "virtio-rng-device"
>  #define VIRTIO_RNG(obj) \
>          OBJECT_CHECK(VirtIORNG, (obj), TYPE_VIRTIO_RNG)
> +#define VIRTIO_RNG_GET_PARENT_CLASS(obj) \
> +        OBJECT_GET_PARENT_CLASS(obj, TYPE_VIRTIO_RNG)
>  
>  /* The Virtio ID for the virtio rng device */
>  #define VIRTIO_ID_RNG    4

This hunk seems a leftover from v2. How to proceed?

Andreas

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Re: [Qemu-devel] [PATCH 27/30] virtio-rng: Convert to QOM realize

[Paolo Bonzini]

Il 29/11/2013 17:10, Andreas Färber ha scritto:
>> > +#define VIRTIO_RNG_GET_PARENT_CLASS(obj) \
>> > +        OBJECT_GET_PARENT_CLASS(obj, TYPE_VIRTIO_RNG)
>> >  
>> >  /* The Virtio ID for the virtio rng device */
>> >  #define VIRTIO_ID_RNG    4
> This hunk seems a leftover from v2. How to proceed?

Possibility 1: mst removes it

Possibility 2: I remove it and resend

Possibility 3: mst gives Acked-by, I remove it and send pull request

Paolo

Re: [Qemu-devel] [PATCH v4 00/30] virtio: cleanup, fix hot-unplug, move to realize

[Andreas Färber]

Am 29.11.2013 11:17, schrieb Paolo Bonzini:
> This series fixes hot-unplug of virtio devices, which can crash due to
> dangling pointer accesses.
> 
> The current implementation supports guest-initiated hot-unplug via the
> virtio_bus_destroy_device function, but not hot-unplugging the virtio
> device by virtue of unplugging its parent container device.
> 
> The problem is that the callback for the bus implementation to cleanup
> is placed in the wrong place; it is in virtio_bus_destroy_device, which
> should be called by the bus, instead of being somewhere in device code.
> We need to have the callback in device code (for example in dc->exit),
> so that we invoke it on every unplug action, no matter who starts it.
> 
> Thus, the series cleans up plugging and unplugging of virtio devices
> so that it does not need any help from the bus (patches 2-5).  It then
> stops the virtio devices' overriding of dc->exit, moving their cleanup
> code to the new exit callback in VirtioDeviceClass (patches 6-11).
> Finally, patch 12 can make virtio-pci implement the device_unplugged
> callback.
> 
> A similar dangling-pointer bug is exposed by this change in virtio-ccw.
> Patch 1 avoids this; it is kept at the beginning to ensure bisectability.
> 
> Patches 13-30 are Andreas's conversion of virtio to realize/unrealize,
> which has better error propagation as shown by the small testcase fix.
> 
> v3->v4: rebased, added Andreas's realize patches on top
> 
> v2->v3: fix to s390 patch; added Reviewed-by and Cced patch 1 to
> 	qemu-stable.
> 
> v1->v2: remove useless pointer chasing in virtio_pci_notify, add
> 	patch 1 to fix breakage reported by Cornelia.
> 
> Andreas Färber (18):
>   virtio-blk-dataplane: Improve error reporting
>   virtio-9p: QOM realize preparations
>   virtio-blk: QOM realize preparations
>   virtio-serial: QOM realize preparations
>   virtio-net: QOM realize preparations
>   virtio-balloon: QOM realize preparations
>   virtio-rng: QOM realize preparations
>   virtio-scsi: QOM realize preparations
>   virtio: Start converting VirtioDevice to QOM realize
>   virtio-9p: Convert to QOM realize
>   virtio-blk: Convert to QOM realize
>   virtio-serial: Convert to QOM realize
>   virtio-net: Convert to QOM realize
>   virtio-balloon: Convert to QOM realize
>   virtio-rng: Convert to QOM realize
>   virtio-scsi: Convert to QOM realize
>   virtio: Complete converting VirtioDevice to QOM realize
>   virtio: Convert exit to unrealize
> 
> Paolo Bonzini (12):
>   virtio-ccw: move virtio_ccw_stop_ioeventfd to virtio_ccw_busdev_unplug
>   virtio-bus: remove vdev field
>   virtio-ccw: remove vdev field
>   virtio-pci: remove vdev field
>   virtio-bus: cleanup plug/unplug interface

>   virtio-blk: switch exit callback to VirtioDeviceClass
>   virtio-serial: switch exit callback to VirtioDeviceClass
>   virtio-net: switch exit callback to VirtioDeviceClass
>   virtio-scsi: switch exit callback to VirtioDeviceClass
>   virtio-balloon: switch exit callback to VirtioDeviceClass
>   virtio-rng: switch exit callback to VirtioDeviceClass

These 6 patches still fail to acknowledge that by fixing the overridden
exit callback they also fix the issue that I was fixing differently in:
http://patchwork.ozlabs.org/patch/249786/ (v1)
http://patchwork.ozlabs.org/patch/263881/ (v2)

Request to add a sentence to the commit message such as:
"This also fixes a memory leak for bus_name, spotted by Andreas."
(Reported-by would be wrong since I didn't spot the 2nd hotunplug issue)

I will be offline for a couple of days and couldn't complete going
through all patches, so if the end result is working QOM-realize'ified
devices then I'm fine with these going through you, mst or whomever. :)

Thanks,
Andreas

>   virtio-pci: add device_unplugged callback

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Re: [Qemu-devel] [PATCH v4 00/30] virtio: cleanup, fix hot-unplug, move to realize

[Paolo Bonzini]

Il 29/11/2013 11:17, Paolo Bonzini ha scritto:
> This series fixes hot-unplug of virtio devices, which can crash due to
> dangling pointer accesses.
> 
> The current implementation supports guest-initiated hot-unplug via the
> virtio_bus_destroy_device function, but not hot-unplugging the virtio
> device by virtue of unplugging its parent container device.
> 
> The problem is that the callback for the bus implementation to cleanup
> is placed in the wrong place; it is in virtio_bus_destroy_device, which
> should be called by the bus, instead of being somewhere in device code.
> We need to have the callback in device code (for example in dc->exit),
> so that we invoke it on every unplug action, no matter who starts it.
> 
> Thus, the series cleans up plugging and unplugging of virtio devices
> so that it does not need any help from the bus (patches 2-5).  It then
> stops the virtio devices' overriding of dc->exit, moving their cleanup
> code to the new exit callback in VirtioDeviceClass (patches 6-11).
> Finally, patch 12 can make virtio-pci implement the device_unplugged
> callback.
> 
> A similar dangling-pointer bug is exposed by this change in virtio-ccw.
> Patch 1 avoids this; it is kept at the beginning to ensure bisectability.
> 
> Patches 13-30 are Andreas's conversion of virtio to realize/unrealize,
> which has better error propagation as shown by the small testcase fix.

Conny,

I pushed this to github at git://github.com/bonzini/qemu.git, branch
refs/heads/virtio.

It would be great if you can give it a test in case I screwed up a rebase.

Paolo

Re: [Qemu-devel] [PATCH v4 00/30] virtio: cleanup, fix hot-unplug, move to realize

[Cornelia Huck]

On Mon, 09 Dec 2013 18:57:55 +0100
Paolo Bonzini <pbonzini@redhat.com> wrote:

> Il 29/11/2013 11:17, Paolo Bonzini ha scritto:
> > This series fixes hot-unplug of virtio devices, which can crash due to
> > dangling pointer accesses.
> > 
> > The current implementation supports guest-initiated hot-unplug via the
> > virtio_bus_destroy_device function, but not hot-unplugging the virtio
> > device by virtue of unplugging its parent container device.
> > 
> > The problem is that the callback for the bus implementation to cleanup
> > is placed in the wrong place; it is in virtio_bus_destroy_device, which
> > should be called by the bus, instead of being somewhere in device code.
> > We need to have the callback in device code (for example in dc->exit),
> > so that we invoke it on every unplug action, no matter who starts it.
> > 
> > Thus, the series cleans up plugging and unplugging of virtio devices
> > so that it does not need any help from the bus (patches 2-5).  It then
> > stops the virtio devices' overriding of dc->exit, moving their cleanup
> > code to the new exit callback in VirtioDeviceClass (patches 6-11).
> > Finally, patch 12 can make virtio-pci implement the device_unplugged
> > callback.
> > 
> > A similar dangling-pointer bug is exposed by this change in virtio-ccw.
> > Patch 1 avoids this; it is kept at the beginning to ensure bisectability.
> > 
> > Patches 13-30 are Andreas's conversion of virtio to realize/unrealize,
> > which has better error propagation as shown by the small testcase fix.
> 
> Conny,
> 
> I pushed this to github at git://github.com/bonzini/qemu.git, branch
> refs/heads/virtio.
> 
> It would be great if you can give it a test in case I screwed up a rebase.

It might have been a good idea to actually cc: me :)

But yes, it seems to pass my sniff test.