Re: [Qemu-devel] [PATCH 1/2] Revert "migration: do not sent zero pages in bulk stage"

[Eric Blake]

On 06/06/2013 03:26 PM, Peter Lieven wrote:

>>> @@ -514,8 +512,8 @@
>>>   ##
>>>   { 'type': 'MigrationStats',
>>>     'data': {'transferred': 'int', 'remaining': 'int', 'total': 'int' ,
>>> -           'duplicate': 'int', 'skipped': 'int', 'normal': 'int',
>>> -           'normal-bytes': 'int', 'dirty-pages-rate' : 'int' } }
>>> +           'duplicate': 'int', 'normal': 'int', 'normal-bytes': 'int',
>>> +           'dirty-pages-rate' : 'int' } }
>> This hunk is questionable.  Removing something that we have previously
>> sent
>> over the wire may break clients that are expecting this field to exist.
>> Rather than reverting the entire patch, you should consider keeping this
>> field present in QMP, even if you now always populate it with 0.
>>
> You are right. I might still account zero pages in the bulk phase to
> give this
> field a meaning. These pages are very likely not written on the
> destination except
> for the cornercases. What do you think?

I don't care if you document a slightly different meaning for the
statistic; as long as it is documented.  Where I do care is removing
something that previous versions provided; I think your later versions
of this series adequately took care of this by reverting only a subset
of the original patch.  I'm okay with the current qemu.git behavior that
always displays returns "skipped":0, or if you post a later patch that
again makes it print a non-zero number that you find useful.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

signature.asc
Description: OpenPGP digital signature