Re: [Qemu-devel] [PATCH] qapi/misc.json: Clarify about usage of QMP 'cpu-add'

[Eduardo Habkost]

On Tue, Sep 18, 2018 at 03:23:03PM +0200, Kashyap Chamarthy wrote:
> Eduardo Habkost mentioned on IRC that the intended functionality of QMP
> 'cpu-add' is replaced with a combination of 'query-hotpluggable-cpus'
> and 'device_add'.  And 'cpu-add' is likely to be deprecated in the
> future.
> 
> Add a note in the QAPI schema to reflect that.
> 
> Suggested-by: Eduardo Habkost <address@hidden
> Signed-off-by: Kashyap Chamarthy <address@hidden>
> ---
>  qapi/misc.json | 8 +++++++-
>  1 file changed, 7 insertions(+), 1 deletion(-)
> 
> diff --git a/qapi/misc.json b/qapi/misc.json
> index d450cfef21..178a94b904 100644
> --- a/qapi/misc.json
> +++ b/qapi/misc.json
> @@ -1104,7 +1104,13 @@
>  ##
>  # @cpu-add:
>  #
> -# Adds CPU with specified ID
> +# Adds CPU with specified ID.
> +#
> +# Notes: This command is likely to be deprecated in the future.  The

I suggest we deprecate the command instead of documenting it as
likely to be deprecated.


> +#        way to achieve the intended functionality of 'cpu-add', which
> +#        is to allow CPU hot-plug, is possible with the combination of
> +#        QMP 'query-hotpluggable-cpus' and 'device_add'.  (And
> +#        hot-unplug via 'device_del'.)

I was going to suggest just writing "This command is deprecated
and was replaced by device_add.  See query-hotpluggable-cpus for
details", but it looks like the query-hotpluggable-cpus
documentation doesn't explain how exactly we can use its results.

Igor, is the QMP documentation for query-hotpluggable-cpus the
only documentation we have on CPU hotplug?


>  #
>  # @id: ID of CPU to be created, valid values [0..max_cpus)
>  #
> -- 
> 2.17.1
> 

-- 
Eduardo