[libvirt] [PATCH v3 2/3] qemu: hot-plug of watchdog

[Michal Privoznik]

On 10/05/2017 01:48 PM, John Ferlan wrote:
> 
> 
> On 10/05/2017 04:07 AM, Michal Privoznik wrote:
>> On 10/04/2017 11:20 PM, John Ferlan wrote:
>>>
>>>
>>> On 09/27/2017 08:12 AM, Michal Privoznik wrote:
>>>> https://bugzilla.redhat.com/show_bug.cgi?id=1447169
>>>>
>>>> Since domain can have at most one watchdog it simplifies things a
>>>> bit. However, since we must be able to set the watchdog action as
>>>> well, new monitor command needs to be used.
>>>>
>>>> Signed-off-by: Michal Privoznik <mprivozn at redhat.com>
>>>> ---
>>>>  src/qemu/qemu_alias.c                              | 13 +++-
>>>>  src/qemu/qemu_alias.h                              |  2 +
>>>>  src/qemu/qemu_command.c                            |  2 +-
>>>>  src/qemu/qemu_command.h                            |  4 +-
>>>>  src/qemu/qemu_driver.c                             | 10 ++-
>>>>  src/qemu/qemu_hotplug.c                            | 72 ++++++++++++++++++++++
>>>>  src/qemu/qemu_hotplug.h                            |  3 +
>>>>  src/qemu/qemu_monitor.c                            | 12 ++++
>>>>  src/qemu/qemu_monitor.h                            |  2 +
>>>>  src/qemu/qemu_monitor_json.c                       | 28 +++++++++
>>>>  src/qemu/qemu_monitor_json.h                       |  3 +
>>>>  tests/qemuhotplugtest.c                            |  9 ++-
>>>>  .../qemuhotplug-watchdog.xml                       |  1 +
>>>>  .../qemuhotplug-base-live+watchdog.xml             | 56 +++++++++++++++++
>>>>  14 files changed, 212 insertions(+), 5 deletions(-)
>>>>  create mode 100644 tests/qemuhotplugtestdevices/qemuhotplug-watchdog.xml
>>>>  create mode 100644 tests/qemuhotplugtestdomains/qemuhotplug-base-live+watchdog.xml
>>>>


>>> I'm fine with the rest of the overall design/concepts, I just think you
>>> need to split up a wee bit more and of course add the caps check....
>>
>> Well, I can split it if you want me to, but:
>>
>> a) in the end the code will look the same,
>> b) it doesn't make sense for somebody to backport just a part of it.
>> They'll backport either all of them or none. They might as well just
>> backport this one. Or not.
>>
>> Michal
>>
> 
> Hey - I used those arguments in my head many times ;-) - perhaps even
> the dog has heard them a few times.  I suppose since there's no reason
> to go back and rework in order to add a capability for the command, then
> no need to deal with splitting up any more, so...
> 
> Reviewed-by: John Ferlan <jferlan at redhat.com>

Pushed thanks :-)

I'll post news.xml patch shortly. Should learn myself to include it in
the series.

Michal