Description of problem:
sometimes, while migrating to server x, server x crashes

qxl-0: qxl_pre_load: start
qxl-0: qxl_hard_reset: start (loadvm)
qxl-0: qxl_reset_surfaces:
handle_dev_destroy_surfaces:
id 0, group 0, virt start 0, virt end ffffffffffffffff, generation 0, delta 0
validate_virt: panic: virtual address out of range
virt=0x175f99c+0xbf slot_id=1 group_id=1
slot=0x0-0x0 delta=0x0


This happens since in handle_dev_destroy_surface, flush_all_qxl_commands is called. Then, there is an attempt to read the qxl command ring. The command ring appears not to be empty. However, the device hasn't been fully loaded yet and the attempt to process a command and access the pci memory leads to the crash.