1. should use findall to find a str in the output of session.cmd_output
2. match str changed from "IO-APIC-fasteoi" to "IO-APIC  <?>-fasteoi"
Signed-off-by: NXiyue Wang <xiywang@redhat.com>

Netperf: fix some issues

multi_queues_test: Add params for sub_test 'multi_vms_nics'

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

* https://github.com/autotest/tp-qemu:
  cgroup: Allow 5% of extra pages when RSS only limit
  cgroup: Avoid counting RssShmem

virtio-win driver signtool check Test

1. Boot a guest with latest virtio-win.iso and virtio-win.vfd.
2. Run signtool.exe to verify each driver '.sys', '.inf' and 'Wdf'file.
Signed-off-by: NPeixiu Hou <phou@redhat.com>

nvdimm: set vm_mem_limit to avoid vm mem larger than maxmem

Signed-off-by: NWenli Quan <wquan@redhat.com>

Signed-off-by: NWenli Quan <wquan@redhat.com>

set jumbo frame to default when occurs test.error or test.fail with
netperf.jumbo.host_guest test.
Signed-off-by: NWenli Quan <wquan@redhat.com>

qemu_guest_agent: add test case for guest-file qga command

fio_perf: update fio version and fix python3 string issue

The previous fio version is too old and cann't be compiled.
Signed-off-by: Nyama <yama@redhat.com>

seabios_strict: add strict option on/off test

Check boot result when strict is on/off
Signed-off-by: Nybduan <yduan@redhat.com>

Signed-off-by: NYanan Fu <yfu@redhat.com>

live_snapshot_chain: Flush data to disk before running system_reset

QMP command 'system_reset' is issued after creating files in the guest. The
command has an effect similar to the physical reset button on a PC. But
filesystems may be left in an unclean state which causes the loss of data. Run
the command "sync" to flush the data to disk.
Signed-off-by: NPing Li <pingl@redhat.com>

single_driver_install: Add tests for virtio input devices

added a qemu-img convert test

Verify it can successfully convert a enlarged snapshot.
Signed-off-by: NHaotong Chen <hachen@redhat.com>

When run bg_sub_test, it will run script multi_vms_nics.py directly,
so params needed in multi_vms_nics.py with 'params.get()' should
have a default value.
Signed-off-by: NSitong Liu <siliu@redhat.com>

iozone_windows: Delete "mem=2048" from iozone_windows.cfg

iozone_windows cfg set mem=2048 before, but when the guest
min memory large then 2G, like win2016 need min memory 4G,
the job will be canceled.
It's better that let the guest use automatic allocation mem value.
Signed-off-by: NPeixiu Hou <phou@redhat.com>

pci_devices: run "info qtree" after the guest boots up

qemu_option_check: python3 compatible, add e1000e, remove e1000

1. fix python 3 compatible issue
2. add e1000e variant
3. remove e1000 for RHEL8 guest
Signed-off-by: NXiyue Wang <xiywang@redhat.com>

multi_queues_test: fix TypeError for python3

Signed-off-by: NSitong Liu <siliu@redhat.com>

seabios.bin_file: Add new case for seabios

ethtool: fix python3 compatible issue

qmp_pmemsave: Decode process.system_output() to string

Signed-off-by: NXiyue Wang <xiywang@redhat.com>

Add bin file check for seabios
Signed-off-by: Nybduan <yduan@redhat.com>

mutl_vms_nics: fix TypeError for python3

Signed-off-by: NSitong Liu <siliu@redhat.com>

balloon_hotplug: invoke simple_hotplug() to set bus under q35

nic_hotplug: python3 compatible, add e1000e, remove e1000

Signed-off-by: NYanan Fu <yfu@redhat.com>

QEMU "info qtree" output depends on information generated by guest
firmware. One example is the bus number in PCI device address. QEMU
retrieves it from the PCI configuration space of the bridge which
creates the bus that a PCI device is attached to. So if test runs
"info qtree" before firmware finishes initializing pci devices, its
output can be incorrect.

The current code works around this timing issue by sleeping 5 seconds.
That is not reliable and doesn't work on my aarch64 system. So I'd
suggest to change the test to not start until it verifies the guest
has booted up. This will make the test execution time slightly longer
but is reliable.
Signed-off-by: NHuan Xiong <huan.xiong@hxt-semitech.com>