[libvirt] [PATCH v2 2/2] qemu: Use the correct vm def on cold attach

[Laine Stump]

On 06/13/2018 09:00 AM, John Ferlan wrote:
>
> On 06/13/2018 08:36 AM, Ján Tomko wrote:
>> On Wed, Jun 13, 2018 at 08:26:40AM -0400, John Ferlan wrote:
>>>> Or we could focus our energy elsehwere.
>>>>
>>> I like this option...  Fix the one offs when/if they show up and move
>>> on. If someone, some day has the desire to write patches that will
>>> revamp device address assignment, good luck!
>>>
>> I meant not bothering with fixing this bug either.
>>
>> Jano
> That's an option as well, IDC really. It didn't take that much time in
> order to propose the initial solution or the followup. Probably spent
> more energy and time typing and considering things based on review
> questions or comments. If you feel compelled to close the bz, then by
> all means do that. Another viable solution is proposed here.


Nah, I think fixing the current BZ is a reasonable thing to do - current
behavior is clearly wrong, but we don't need to worry about the case I
mentioned until someone else complains (it's been like that for a *very*
long time, and I think I may be the only one who noticed). My apologies
for just dive bombing in with a new corner case without actually
reviewing your code - it was the end of the day and I was tired, but
also remembered running across that problem a long time ago (and
deciding it wasn't worth the effort to fix then either) and just wanted
to make other people aware of it before I forgot again :-P.

I'll go back and actually review the patches now.