[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v3 3/3] qemu-iotests: 300: Add test case for modifying persis
From: |
Eric Blake |
Subject: |
Re: [PATCH v3 3/3] qemu-iotests: 300: Add test case for modifying persistence of bitmap |
Date: |
Fri, 12 Feb 2021 13:15:35 -0600 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.7.0 |
On 2/12/21 11:34 AM, Peter Krempa wrote:
> Verify that the modification of the bitmap persistence over migration
> which is controlled via BitmapMigrationBitmapAliasTransform works
> properly.
>
> Based on TestCrossAliasMigration
>
> Signed-off-by: Peter Krempa <pkrempa@redhat.com>
> ---
> tests/qemu-iotests/300 | 91 ++++++++++++++++++++++++++++++++++++++
> tests/qemu-iotests/300.out | 4 +-
> 2 files changed, 93 insertions(+), 2 deletions(-)
>
> + def setUp(self) -> None:
> + TestDirtyBitmapMigration.setUp(self)
> +
> + # Now create another block device and let both have two bitmaps each
> + result = self.vm_a.qmp('blockdev-add',
> + node_name='node-b', driver='null-co')
> + self.assert_qmp(result, 'return', {})
> +
> + result = self.vm_b.qmp('blockdev-add',
> + node_name='node-a', driver='null-co')
The use of null-co with no mention of the read-zeroes option here is a
(minor) semantic conflict with the proposal by Philippe to flip the
default of that option (minor, because all it impacts would be the
execution speed of the test)
https://lists.gnu.org/archive/html/qemu-devel/2021-02/msg04027.html
Not your fault, so I don't mind touching up your additions in the same
manner Philippe proposed (which is safe regardless of whose patch lands
first).
diff --git i/tests/qemu-iotests/300 w/tests/qemu-iotests/300
index 9d4ec6a38195..63036f6a6e13 100755
--- i/tests/qemu-iotests/300
+++ w/tests/qemu-iotests/300
@@ -615,11 +615,13 @@ class
TestAliasTransformMigration(TestDirtyBitmapMigration):
# Now create another block device and let both have two bitmaps
each
result = self.vm_a.qmp('blockdev-add',
- node_name='node-b', driver='null-co')
+ node_name='node-b', driver='null-co',
+ read_zeroes=False)
self.assert_qmp(result, 'return', {})
result = self.vm_b.qmp('blockdev-add',
- node_name='node-a', driver='null-co')
+ node_name='node-a', driver='null-co',
+ read_zeroes=False)
self.assert_qmp(result, 'return', {})
bmaps_to_add = (('node-a', 'bmap-b'),
--
Eric Blake, Principal Software Engineer
Red Hat, Inc. +1-919-301-3226
Virtualization: qemu.org | libvirt.org
- [PATCH v3 0/3] migration: dirty-bitmap: Allow control of bitmap persistence, Peter Krempa, 2021/02/12
- [PATCH v3 1/3] migration: dirty-bitmap: Convert alias map inner members to BitmapMigrationBitmapAlias, Peter Krempa, 2021/02/12
- [PATCH v3 2/3] migration: dirty-bitmap: Allow control of bitmap persistence, Peter Krempa, 2021/02/12
- [PATCH v3 3/3] qemu-iotests: 300: Add test case for modifying persistence of bitmap, Peter Krempa, 2021/02/12
- Re: [PATCH v3 3/3] qemu-iotests: 300: Add test case for modifying persistence of bitmap,
Eric Blake <=
- Re: [PATCH v3 0/3] migration: dirty-bitmap: Allow control of bitmap persistence, no-reply, 2021/02/12
- Re: [PATCH v3 0/3] migration: dirty-bitmap: Allow control of bitmap persistence, Eric Blake, 2021/02/12