Re: [PATCH 1/5] docs/devel: document expectations for QAPI data modelling for QMP

[Markus Armbruster]

Daniel P. Berrangé <berrange@redhat.com> writes:

> Traditionally we have required that newly added QMP commands will model
> any returned data using fine grained QAPI types. This is good for
> commands that are intended to be consumed by machines, where clear data
> representation is very important. Commands that don't satisfy this have
> generally been added to HMP only.
>
> In effect the decision of whether to add a new command to QMP vs HMP has
> been used as a proxy for the decision of whether the cost of designing a
> fine grained QAPI type is justified by the potential benefits.
>
> As a result the commands present in QMP and HMP are non-overlapping
> sets, although HMP comamnds can be accessed indirectly via the QMP
> command 'human-monitor-command'.
>
> One of the downsides of 'human-monitor-command' is that the QEMU monitor
> APIs remain tied into various internal parts of the QEMU code. For
> example any exclusively HMP command will need to use 'monitor_printf'
> to get data out. It would be desirable to be able to fully isolate the
> monitor implementation from QEMU internals, however, this is only
> possible if all commands are exclusively based on QAPI with direct
> QMP exposure.
>
> The way to achieve this desired end goal is to finese the requirements
> for QMP command design. For cases where the output of a command is only
> intended for human consumption, it is reasonable to want to simplify
> the implementation by returning a plain string containing formatted
> data instead of designing a fine grained QAPI data type. This can be
> permitted if-and-only-if the command is exposed under the 'x-' name
> prefix. This indicates that the command data format is liable to
> future change and that it is not following QAPI design best practice.
>
> The poster child example for this would be the 'info registers' HMP
> command which returns printf formatted data representing CPU state.
> This information varies enourmously across target architectures and
> changes relatively frequently as new CPU features are implemented.
> It is there as debugging data for human operators, and any machine
> usage would treat it as an opaque blob. It is thus reasonable to
> expose this in QMP as 'x-query-registers' returning a 'str' field.
>
> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
> ---
>  docs/devel/writing-qmp-commands.rst | 25 +++++++++++++++++++++++++
>  1 file changed, 25 insertions(+)
>
> diff --git a/docs/devel/writing-qmp-commands.rst 
> b/docs/devel/writing-qmp-commands.rst
> index 6a10a06c48..d032daa62d 100644
> --- a/docs/devel/writing-qmp-commands.rst
> +++ b/docs/devel/writing-qmp-commands.rst
> @@ -350,6 +350,31 @@ In this section we will focus on user defined types. 
> Please, check the QAPI
>  documentation for information about the other types.
>  
>  
> +Modelling data in QAPI
> +~~~~~~~~~~~~~~~~~~~~~~

Outline now:

    How to write QMP commands using the QAPI framework
        Overview
        Testing
        Writing a command that doesn't return data
            Arguments
            Errors
            Command Documentation
            Implementing the HMP command
        Writing a command that returns data
-->         Modelling data in QAPI
            User Defined Types
            The HMP command
            Returning Lists

Awkward.  I guess you wanted it next to "User Defined Types", which
makes some sense.

Perhaps minor tweaks (headlines, maybe a bit of text as well) would
suffice:

    How to write QMP commands using the QAPI framework
        Overview
        Testing
        Writing a simple command: hello-world
            Arguments
            Errors
            Command Documentation
            Implementing the HMP command
        Writing more complex commands
            Modelling data in QAPI
            User Defined Types
            The HMP command
            Returning Lists

> +
> +For a QMP command that to be considered stable and supported long term there

"term, there"

> +is a requirement returned data should be explicitly modelled using fine 
> grained

"fine-grained", I think.

> +QAPI types. As a general guide, a caller of the QMP command should never need
> +to parse individual returned data fields. If a field appears to need parsing,
> +them it should be split into separate fields corresponding to each distinct
> +data item. This should be the common case for any new QMP command that is
> +intended to be used by machines, as opposed to exclusively human operators.
> +
> +Some QMP commands, however, are only intended as adhoc debugging aids for 
> human
> +operators. While they may return large amounts of formatted data, it is not
> +expected that machines will need to parse the result. The overhead of 
> defining
> +a fine grained QAPI type for the data may not be justified by the potential
> +benefit. In such cases, it is permitted to have a command return a simple 
> string

There are many existing long lines in this file, so I'm not flagging
yours, except for this one, because it increases the maximum.

> +that contains formatted data, however, it is mandatory for the command to use
> +the 'x-' name prefix. This indicates that the command is not guaranteed to be
> +long term stable / liable to change in future and is not following QAPI 
> design
> +best practices. An example where this approach is taken is the QMP command
> +"x-query-registers". This returns a printf formatted dump of the architecture

Drop "printf".

> +specific CPU state. The way the data is formatted varies across QEMU targets,
> +is liable to change over time, and is only intended to be consumed as an 
> opaque
> +string by machines.
> +
>  User Defined Types
>  ~~~~~~~~~~~~~~~~~~

This file is a tutorial.  It teaches command writing by examples[*].  I
think it should also teach this new class of "ad hoc" QMP commands the
same way.  A section "Writing a debugging aid returning unstructured
text" could go at the very end.


[*] Bit-rotten in places, but that's life for docs.