Re: [Qemu-devel] active block commit bug?

[Markus Armbruster]

Kevin Wolf <address@hidden> writes:

> Am 05.06.2014 um 09:06 hat Markus Armbruster geschrieben:
>> Eric Blake <address@hidden> writes:
>> 
>> > On 06/04/2014 08:09 PM, Fam Zheng wrote:
>> >
>> >>> Sounds like we have an off-by-one condition if empty files behave
>> >>> differently from other files.  We ought to fix that bug (not that your
>> >>> normal guest will ever have a 0-length backing file, but this was what I
>> >>> was trying to use for libvirt's probing of whether active commit is
>> >>> supported)
>> >>>
>> >> 
>> >> Yes, agreed, this special case is only going to make management confused. 
>> >> I
>> >> will send a patch to fix this.
>> >
>> > Thanks.
>> >
>> >> 
>> >> Eric, is this a good way to probe the active commit? I was expecting full
>> >> instrospection of QMP could do it, but I don't know about the
>> >> status of that
>> >> piece of work. Amos, any ideas?
>> >
>> > Introspection already missed qemu 2.0 when active commit was added; and
>> > we're close enough to soft freeze for 2.1 that I'm guessing it will miss
>> > 2.1 as well :(
>> 
>> Almost certainly.  It has non-trivial design issues.  To have a chance
>> to make it into 2.x, it needs to be posted for review early in the 2.x
>> cycle.
>
> Why is nobody discussing these non-trivial design issues then? Patches
> have been posted even in a 1.x development phase, so that's not what is
> blocking us. If anything, it's the missing discussion.

If I remember correctly, review of the patches made us realize
introspection is more complex than initially thought.  We discussed a
proper solution at some length, but concluded implementing it in time
for 2.0 was not in the cards (we were running close to the hard freeze
already).  Unfortunately, nobody has stepped up to implement it in time
for 2.1 either.

To make further progress, we need a sucker^Wvolunteer pushing the
feature.