Update the order of importing modules by PEP8.

Wrap the codes to function:
  1. Update function find_disk and rename it.
  2. Update function run_sub_test.
  3. Wrap the part of waiting plugging disks completely to
     function wait_plug_disks.
  4. Wrap the part of creating block devices to function
     create_block_devices.
  5. Wrap the part of getting block devices to function
     get_block_devices.
  6. Wrap the part of hotplug or unplug block devices to
     function plug_block_devices.
  7. Wrap the part of format windows disk to function format_disk_win.
  8. Wrap the part of doing io testing on disk to function
     run_io_test.
  9. Update function get_disk_size.
Signed-off-by: NYongxue Hong <yhong@redhat.com>

Signed-off-by: NQianqian Zhu <qizhu@redhat.com>

Signed-off-by: NYumei Huang <yuhuang@redhat.com>

Signed-off-by: NQianqian Zhu <qizhu@redhat.com>

There is a unfriendly log when failed to
hotplug a port. It should give more specific
error info but no more like code error
Signed-off-by: NNana Liu <nanliu@redhat.com>

The expected results for rhel6 and rhel8 are different.
Need to add results check for rhel6.
Signed-off-by: Nyama <yama@redhat.com>

aims to verify all aio tests are passed.
Signed-off-by: Nlolyu <lolyu@redhat.com>

New a case that hot plug a disk then unplug it under the status
of vm is stopping, resume vm at last.
Signed-off-by: NYongxue Hong <yhong@redhat.com>

Ping host when running heavyload stress tool
Signed-off-by: NYu Wang <wyu@redhat.com>

Since ppc doesn't support ide interface, so replace it
with virtio_scsi and virtio_blk.
Signed-off-by: Nzhenyu zhang <zhenyzha@redhat.com>

And sleep a while after host script start, to make sure host script
always start before guest script, cause for pty it must start first
Signed-off-by: NQianqian Zhu <qizhu@redhat.com>

1.Update function "check_host_numa_node_amount" to
  "get_host_numa_node" to get host numa node
  whose node size is non-zero
2.Assign host numa nodes whose node size is non-zero to
  "host-nodes" and "target_num"
Signed-off-by: NXianwang <xianwang@redhat.com>

Signed-off-by: NXiaoling Gao <xiagao@redhat.com>

New cases that boot up two VMs to test scsi block device
with "share-rw=on" and "share-rw=off" option.
Signed-off-by: NYongxue Hong <yhong@redhat.com>

New a case to test scsi device by deleting and scanning.
Signed-off-by: NYongxue Hong <yhong@redhat.com>

Add a new testcase to test QEMU image locking should allow read
operation over the same base image. This case aims to verify that
snapshot in different snapshot chain could be used at the same time.
Signed-off-by: Nlolyu <lolyu@redhat.com>

temporary fix to avoid blockdev updates
Signed-off-by: NXu Tian <xutian@redhat.com>
(cherry picked from commit 1ece20e9b5bdba46495bf7a6d1550348d90518b7)

Check both default and specific speed in guest
Signed-off-by: NYu Wang <wyu@redhat.com>

Move driver microsoft signature check part from
physical_resources_check to win_sigverif case as prewhql
drivers are not msft signed. It's better to move it from
regular cases.
Signed-off-by: Nlijin <lijin@redhat.com>

Add case of Offload Tx/Rx checksum for windows guests
Signed-off-by: NYu Wang <wyu@redhat.com>

Signed-off-by: NQianqian Zhu <qizhu@redhat.com>

1. Support tcp socket
2. Support Windows guest better
3. Make it modulized and multiplexible
Signed-off-by: NQianqian Zhu <qizhu@redhat.com>

Add more cpu models, including Opteron_G5, Opteron_G4, Opteron_G3,
Icelake-Server, Icelake-Client, Cascadelake-Server, Broadwell,
Broadwell-noTSX, Haswell, Haswell-noTSX, IvyBridge, and SandyBridge.
Signed-off-by: NYumei Huang <yuhuang@redhat.com>

The data disk letters will be changed after reboot in windows,
so fetch the letters again.
Signed-off-by: NYongxue Hong <yhong@redhat.com>

Update the following codes:
1. Add support windows in function stress_is_alive.
2. Add check "SHUTDOWN" event in function _change_vm_power.
3. Modify the shell command of shutdown and reboot for windows.
Signed-off-by: NYongxue Hong <yhong@redhat.com>

Need to update the block cases to support blockdev options, since
qemu enable the blockdev options from the certain version.

Updated cases:
 1. block_hotplug
 2. block_resize
 3. cdrom
 4. cdrom_block_size_check
 5. change_media
 6. eject_media
 7. block_hotplug_in_pause
 8. multi_disk_random_hotplug
Signed-off-by: NYongxue Hong <yhong@redhat.com>

New case of update qemu-ga-win without vioser driver installed
Signed-off-by: NXiaoling Gao <xiagao@redhat.com>

Update bin_dict to add support for latest rhel8-related machine type.
Signed-off-by: Nybduan <yduan@redhat.com>

10s is too short, sometimes no results. Need more time to get
output.
Signed-off-by: Nyama <yama@redhat.com>

1. Add resize linux file system and partition steps.
2. Add iozone test after resize.
3. Add resize test on windows system disk.
4. Move many codes to avocado-vt functions,
   like get_disk_size() and extend_volume(), shrink_volume(),
   they all were merged.
Signed-off-by: NPeixiu Hou <phou@redhat.com>

Otherwise sometimes cause unexpected exceptions when case finished
Signed-off-by: NQianqian Zhu <qizhu@redhat.com>

Latest qemu 4.x doesn't support boot option reboot-timeout larger than
65535, it only works with older qemu version.
Signed-off-by: Nybduan <yduan@redhat.com>

Use simple_hotplug() instead of device.hotplug() for q35 support,
'serials' instead of 'virtio_ports' to align new definition. And
remove get device id by name methods since not necessary now
Signed-off-by: NQianqian Zhu <qizhu@redhat.com>

Due to new block_resize file no need disk_update_cmd params,
so delete them, and update a few new block_resize.py needed params.
Signed-off-by: NPeixiu Hou <phou@redhat.com>

With host 100% load, the serial login needs more time, 240s
 isn't enough. For windows guest, the response is too slow,
so no need to support it.
Signed-off-by: Nyama <yama@redhat.com>

Signed-off-by: Nnanliu <nanliu@redhat.com>

1.rewrite function of gagent_check_get_interfaces
2.support windows guest
Signed-off-by: NXiaoling Gao <xiagao@redhat.com>

Signed-off-by: Nyiqian <yiqian@localhost.localdomain>

The windows guest's time zone needs to be the same as host
before test.
Signed-off-by: Nyama <yama@redhat.com>

Add new case to test EPYC, Skylake-Server and Skylake-Client.
Signed-off-by: NYumei Huang <yuhuang@redhat.com>