Re: [Qemu-devel] [PATCH V4 4/7] qmp: Allow to change password on names block driver states.

[Benoît Canet]

Le Monday 09 Dec 2013 à 12:03:26 (-0500), Luiz Capitulino a écrit :
> On Mon, 9 Dec 2013 17:48:50 +0100
> Benoît Canet <address@hidden> wrote:
> 
> > Le Monday 09 Dec 2013 à 11:41:09 (-0500), Luiz Capitulino a écrit :
> > > On Mon, 9 Dec 2013 17:23:09 +0100
> > > Kevin Wolf <address@hidden> wrote:
> > > 
> > > > > > I'm leaning slightly towards the approach that Benoît took, if only 
> > > > > > for
> > > > > > the naming aspect (that is, I also thought of the idea of a bool 
> > > > > > flag,
> > > > > > but didn't suggest it because I didn't like the implications on the
> > > > > > naming).  But I can live with either approach, if anyone else has a
> > > > > > strong opinion.
> > > > > 
> > > > > Well, we can pick up any descriptive name 'treat-device-as-a-node',
> > > > > 'device-is-a-graph-node'...
> > > > 
> > > > All devices are represented by nodes, so that doesn't make sense.
> > > > If anything, 'interpret-device-name-as-node-name', which at the same
> > > > time makes it pretty clear that we're abusing a field for something it
> > > > wasn't meant for.
> > > 
> > > Having two optionals where they can't be specified at the same time
> > > and can't be left off at the same time is a clear abuse as well.
> > > 
> > > The truth is, both proposals are bad. This makes me think that maybe
> > > we should introduce a block API 2.0 and deprecate the current one
> > > (partly or completely).
> > > 
> > 
> > It took me one year to go from the block filters and block backend
> > requirement to the state where my customer allows me to work on block 
> > filters.
> > 
> > Now if we add to this the new requirement of block API 2.0 I think I will 
> > soon
> > have time to concentrate myself on non qemu projects :(
> 
> I don't think it would be something major as far as code is concerned.
> What can take a lot of time and energy is to define the API. The QMP
> commands implementation would probably be a wrapper around a single (or
> a set of) block functions.
> 
> Again, I can live with what I suggested because I find it simpler
> than your original proposal: no existing field is changed, only one
> field is added, and clients can happily omit it if they don't know
> what it's about.
> 

I already have rewritten the patches to support your version of the commands.

I will let you people decide which version qemu will merge.

Best regards

Benoît