qemu-e2k/tests/qemu-iotests/216
John Snow 72cfb937b8 iotests: remove qemu_io_silent() and qemu_io_silent_check().
Like qemu-img, qemu-io returning 0 should be the norm and not the
exception. Remove all calls to qemu_io_silent that just assert the
return code is zero (That's every last call, as it turns out), and
replace them with a normal qemu_io() call.

qemu_io_silent_check() appeared to have been unused already.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Hanna Reitz <hreitz@redhat.com>
Message-Id: <20220418211504.943969-12-jsnow@redhat.com>
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
2022-04-25 14:30:34 +02:00

118 lines
4.2 KiB
Python
Executable File

#!/usr/bin/env python3
# group: rw quick
#
# Copy-on-read tests using a COR filter node
#
# Copyright (C) 2018 Red Hat, Inc.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
# Creator/Owner: Hanna Reitz <hreitz@redhat.com>
import iotests
from iotests import log, qemu_img, qemu_io
# Need backing file support
iotests.script_initialize(supported_fmts=['qcow2', 'qcow', 'qed', 'vmdk'],
supported_platforms=['linux'])
log('')
log('=== Copy-on-read across nodes ===')
log('')
# The old copy-on-read mechanism without a filter node cannot request
# WRITE_UNCHANGED permissions for its child. Therefore it just tries
# to sneak its write by the usual permission system and holds its
# fingers crossed. However, that sneaking does not work so well when
# there is a filter node in the way: That will receive the write
# request and re-issue a new one to its child, which this time is a
# proper write request that will make the permission system cough --
# unless there is someone at the top (like a guest device) that has
# requested write permissions.
#
# A COR filter node, however, can request the proper permissions for
# its child and therefore is not hit by this issue.
with iotests.FilePath('base.img') as base_img_path, \
iotests.FilePath('top.img') as top_img_path, \
iotests.VM() as vm:
log('--- Setting up images ---')
log('')
qemu_img('create', '-f', iotests.imgfmt, base_img_path, '64M')
qemu_io(base_img_path, '-c', 'write -P 1 0M 1M')
qemu_img('create', '-f', iotests.imgfmt, '-b', base_img_path,
'-F', iotests.imgfmt, top_img_path)
qemu_io(top_img_path, '-c', 'write -P 2 1M 1M')
log('Done')
log('')
log('--- Doing COR ---')
log('')
# Compare with e.g. the following:
# vm.add_drive_raw('if=none,node-name=node0,copy-on-read=on,driver=raw,' \
# 'file.driver=%s,file.file.filename=%s' %
# (iotests.imgfmt, top_img_path))
# (Remove the blockdev-add instead.)
# ((Not tested here because it hits an assertion in the permission
# system.))
vm.launch()
log(vm.qmp('blockdev-add',
node_name='node0',
driver='copy-on-read',
file={
'driver': 'raw',
'file': {
'driver': 'copy-on-read',
'file': {
'driver': 'raw',
'file': {
'driver': iotests.imgfmt,
'file': {
'driver': 'file',
'filename': top_img_path
},
'backing': {
'driver': iotests.imgfmt,
'file': {
'driver': 'file',
'filename': base_img_path
}
}
}
}
}
}))
# Trigger COR
log(vm.qmp('human-monitor-command',
command_line='qemu-io node0 "read 0 64M"'))
vm.shutdown()
log('')
log('--- Checking COR result ---')
log('')
qemu_io(base_img_path, '-c', 'discard 0 64M')
qemu_io(top_img_path, '-c', 'read -P 1 0M 1M')
qemu_io(top_img_path, '-c', 'read -P 2 1M 1M')
log('Done')