Re: [Qemu-devel] [PATCH v4 00/11] convert savevm to use qapi and introduce qmp command

[Wenchao Xia <xiawenc@linux.vnet.ibm.com>]

于 2013-4-11 1:33, Pavel Hrdina 写道:
> On 10.4.2013 19:15, Eric Blake wrote:
>> On 04/10/2013 08:05 AM, Pavel Hrdina wrote:
>>> Here is another proposal how to handle vm snapshots:
>>>
>>> QMP vm-snapshot-save:
>>>      - { 'command': 'vm-snapshot-save',
>>>          'data': { 'name': 'str' },
>>>          'returns': 'SnapshotInfo' }
>>>      - vm-snapshot-save returns an error if there is an existing
>>>        snapshot with the same name
>>>      - you cannot provide an id for a new snapshot
>>>      - on success all information about created snapshot will be
>>> returned
>>>
>>> QMP vm-snapshot-load
>>>      - { 'command': 'vm-snapshot-load',
>>>          'data': { '*name': 'str', '*id': 'int' },
>>>          'returns': 'SnapshotInfo' }
>>>      - one of the name or id must be provided
>>>      - if both are provided they will match only the snapshot with the
>>>        same name and id
>>>      - returns SnapshotInfo only if the snapshot exists.
>>>
>>> QMP vm-snapshot-delete:
>>>      - { 'command': 'vm-snapshot-delete',
>>>          'data': { '*name': 'str', '*id': 'int' },
>>>          'returns': 'SnapshotInfo' }
>>>      - same rules as vm-snapshot-load
>>
>> Missing some form of query-snapshots to list all consistent snapshots
>> that can be loaded or deleted (or is that another series?)
>
> Yes, there is another series for that.
> http://lists.nongnu.org/archive/html/qemu-devel/2013-04/msg00196.html
>
>>
>> Also, while load can only take a consistent snapshot, it might make
>> sense to expose two levels of delete - one that deletes consistent
>> snapshots, and one that deletes a snapshot for a given block device
>> regardless of whether it is consistent across all devices in use by
>> the VM.
>
> I think that Wenchao also works on it. He posted some patches for block
> snapshots about month ago.
>
   yep, the latest version is at
http://lists.nongnu.org/archive/html/qemu-devel/2013-04/msg02525.html
which expose block'image info to management stack. Luckily the two
serial does not conflict each other in main part:),  only for
bdrv_snapshot_find() function.

>>
>>>
>>> HMP savevm:
>>>      - args_type = "force:-f,name:s?",
>>>      - if the name is not provided the HMP command will generates new
>>>        one for QMP command
>>>      - if there is already a snapshot with provided or generated name
>>>        it will fails
>>>      - there will be an optional -f parameter to force saving requested
>>>        snapshot and it will internally use vm-snapshot-delete and then
>>>        vm-snapshot-save
>>>      - all information about created snapshot will be printed
>>>
>>> HMP loadvm:
>>>      - args_type = "name:s?,id:i?",
>>>      - follow the same behavior as the QMP command
>>
>> Except that with HMP, arguments are positional.  You can't provide 'id'
>> in isolation; by listing id:i? second, the parser requires that the
>> first argument encountered is a name.
>>
>> I think a better interface might be:
>>
>> HMP loadvm:
>>      - args_type = "id:-i,name:s?'
>>
>> If just name is given, first do a tag lookup; if that fails do an id
>> lookup.  If the -i flag is given, skip the tag lookup just do an id
>> lookup.
>
> I didn't realize that, thanks for pointing this out.
>
>>
>>>
>>> HMP delvm:
>>>      - args_type = "name:s?,id:i?"
>>>      - same rules as loadvm
>>
>> Same comment as for loadvm.
>>
>
>


-- 
Best Regards

Wenchao Xia