Re: [Qemu-devel] [PATCH 0/5] qapi flattening + some miscellaneous patches

[Markus Armbruster]

Markus Armbruster <address@hidden> writes:

> Gerd Hoffmann <address@hidden> writes:
>
>> On Do, 2015-07-02 at 20:00 +0200, Markus Armbruster wrote:
>>> Gerd Hoffmann <address@hidden> writes:
>>> 
>>> > On Di, 2015-06-23 at 15:32 +0200, Kővágó, Zoltán wrote:
>>> >> I've cherry-picked the qapi related parts from my previous -audiodev
>>> >> patch series, we can hopefully concentrate on one thing at a time.  The
>>> >> most important changes in this patch series are the flattening of the
>>> >> Netdev structures.  This way we can add proper nested structure support
>>> >> into OptsVisitor, without requiring backward-compatibility hacks.
>>> >
>>> > Applies and builds fine, no obvious regressions in testing.
>>> >
>>> > Tested-by: Gerd Hoffmann <address@hidden>
>>> >
>>> > Getting this merged before hard freeze would be great ...
>>> >
>>> > Any takers?  Markus?
>>> 
>>> A first round of review is my best offer, I'm afraid: I'll be away the
>>> next two weeks.
>>> 
>>> Any particular reason why we want this in 2.4?
>>
>> Looked easy as everybody agreed that flattening the qemu structs is a
>> good thing, and merging stuff helps keeping the out-of-tree patch queues
>> smaller ...
>
> Fair enough.
>
> The series does three things, I think:
>
> 1. QAPI struct flattening [PATCH 2+3]
>
> 2. OptsVisitor nesting support [PATCH 1+4]
>
> 3. Nicer qemu_opts_print() [PATCH 5]
>
> Regarding 1., I'm fine with the general approach.  In fact, proper
> netdev_add qapification will need the flattening anyway (if it's
> possible at all, but I'd like to try).  However, PATCH 3 needs a few
> comment edits (could be done on commit), and I'd prefer to avoid the
> type system cheat.  See my review for details.

I'll go offline tonight, back on 22nd.  I think PATCH 2 could go through
Eduardo (NUMA maintainer), and PATCH 3 through Stefan (net maintainer).

> Regarding 2., I have questions, and don't think we should rush it.
>
> Regarding 3., I'm again fine with the general idea, but the "quoting for
> shell" part doesn't look ready.  Take out that part if you want
> something committed *now*.

You could try -trivial for this one.