qmp_command:add command "query-target"

Signed-off-by: NYiqian Wei <yiwei@redhat.com>

qemu_guest_agent: new case to get basic info of vm

Add new case checking provisioning mode

Signed-off-by: Nqingwangrh <qinwang@redhat.com>

Get timezone,hostname and active users info via qga
and check the result in guest.
Signed-off-by: NXiaoling Gao <xiagao@redhat.com>

Make cpu hotplug cases work on s390

disable windows guest for case uefi_check_debugcon

Signed-off-by: NXueqiang Wei <xuwei@redhat.com>

add test for acpi description of serial and parallel ports correctness

Signed-off-by: NMicky Yun Chan <michan@redhat.com>

virtio-serial:Keep chardev open to re-use

Commit during guest reboot

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

New case of cache_sizes_test

Add new case that commit during guest reboot
Signed-off-by: NAihua Liang <aliang@redhat.com>

Add support to commit on both system and data disk
Signed-off-by: NAihua Liang <aliang@redhat.com>

1. s390 supports cpu hotplug but not hotunplug, update to make some
existing cpu hotplug cases work on s390.
2. Also make a simple change so that some windows versions do not check
cpu topology, and highlight a "TBD" of supported win OS list.
Signed-off-by: NYihuang Yu <yihyu@redhat.com>

ept_test: New case to test when ept is off

Support query_named_block_nodes during live commit

virtio_serial:Hot-unplug port during data transferring

Test different cache sizes in a guest.
Signed-off-by: NTingting Mao <timao@redhat.com>

Avoid duplicated deployment of kar on L1 guest

blockdev_commit_query_named_block_nodes: add support
for query_named_block_nodes during live commit
Signed-off-by: NAihua Liang <aliang@redhat.com>

Signed-off-by: NQinghua Cheng <qcheng@redhat.com>

Should clean temporary file even data transfer failed
Signed-off-by: NNana Liu <nanliu@redhat.com>

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

usb_storage: add rerror and werror options for usb-storage device

Signed-off-by: Nybduan <yduan@redhat.com>

usb_hotplug_unplug: add hotplug/hotunplug support for usb-storage

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

* https://github.com/autotest/tp-qemu:
  qemu..migration_with_speed_measurement: Fix py3 int casting
  qemu.migration_with_speed_measurement: Fix cpu_flags location
  qemu..migration_with_speed_meassurement: Wait for mig start
  qemu..migration: Run pre_migrate before each iteration

add testcase for chardev a free port value

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

Signed-off-by: NMicky Yun Chan <michan@redhat.com>

Live commit on multi-disks

blockdev_commit_base: support commit on multi-disks
blockdev_commit: support commit on multi-disks
blockdev_commit: add cfg to support multi-disks
Signed-off-by: NAihua Liang <aliang@redhat.com>

Python3 does not allow casting "1.0" to integer directly, let's use
int(float("1.0")) to avoid "invalid literal for int" ValueErrors.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

For some time the "cpu_flags" test location had been adjusted to include
"src" dir in "provider/cpuflags.py", let's reflect this change in this
test as well.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Currently the code only checks for active migration, but VM might be
also in "setup" phase before it actually starts the migration. Let's use
a well maintained "get_migrate_progress" function to make sure the
migration started.

This comes from a real-world case where "Migration already ended" was
raised with a "setup" status.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Previously the "pre_migrate" function was only executed before the first
iteration, but at least for mig_set_speed it's required before each
iteration.

As for the "check_dma" function, it also makes sense to run it for each
iteration and code-wise should work well, but because of outdated
code-path I have only tested it with slighly modified version (without
dmesg check on virtio_blk).

Note the migration_with_speed_measurement test is not affected as it's
located in a different file.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>