The vmsd code expects the fields structure to be properly terminated,
not NULL. An assigned device should never be saved or restored, and
recent qemu fixes to the no_migrate flag should ensure this, but let's
avoid setting the wrong precedent.
Signed-off-by: Alex Williamson <alex.william...@redhat.com>
---
v2:
- Change to commit log only, was "device-assignment: Add fields to
VMStateDescription". Recent qemu.git no_migrate changes avoid
potential segfault, but this should still be applied for correctness.
hw/device-assignment.c | 5 ++++-
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/hw/device-assignment.c b/hw/device-assignment.c
index e97f565..0925fa8 100644
--- a/hw/device-assignment.c
+++ b/hw/device-assignment.c
@@ -1697,7 +1697,10 @@ static void
assigned_dev_unregister_msix_mmio(AssignedDevice *dev)
}
static const VMStateDescription vmstate_assigned_device = {
- .name = "pci-assign"
+ .name = "pci-assign",
+ .fields = (VMStateField []) {
+ VMSTATE_END_OF_LIST()
+ }
};
static void reset_assigned_device(DeviceState *dev)
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
