Re: [Xen-devel] [PATCH v2 5/5] golang/xenlight: Add tests host related functionality functions

[George Dunlap <george.dunlap@xxxxxxxxxx>]

On 03/03/17 15:02, Ian Jackson wrote:
> George Dunlap writes ("Re: [PATCH v2 5/5] golang/xenlight: Add tests host 
> related functionality functions"):
>> Right.  The purpose of hand-crafting the code was to get a feel for what
>> a good Go-like output would look like before investing in the IDL.
> 
> That makes sense.
> 
>>  It sounds like you're suggesting that having IDL support would be a
>> prerequisite to getting anything checked in?
> 
> I think it's a prerequisite for it to be hooked into the build by
> default, unless someone can convince me that foreseeable changes to
> the libxl idl will not cause the golang build to break.
> 
>> I'd definitely say havind IDL support would be a prerequisite for
>> declaring the bindings "supported".  I don't think the structures for
>> these functions change so often that it would be a hardship for Ronald
>> or I to change them whenever they broke; and so I would argue it
>> shouldn't be a blocker for getting things into the tree if the code
>> looks good.
> 
> I don't think it's sensible to ask other contributors who change the
> IDL to have to liase with golang experts in order to not break the
> build.  The whole point of the idl system is precisely to make this
> kind of thing automatic.
> 
> OTOH I don't mind it being in-tree if it's not built by default.  Of
> course then it will probably rot, so that's not a very stable
> situation, but if having it in-tree makes the development easier then
> that's fine by me.

Yes, the plan was to have it off by default at very least until we got
configure to detect the presence of a go compiler.  Ronald has already
spent a decent chunk of his time doing Makefile work; I didn't want to
make him spend a big chunk of time poking autoconf as well on a project
which was advertised as programming Go. :-)  Adding "won't break on IDL
changes" is another criteria it would be good to add.

Once the core patches are in-tree, I could start working on some of that
stuff if I have time.  In addition, I keep finding random improvements
to make in Ronald's patches which aren't actually bugs; it would be
nicer for both of us if I could just post patches for those improvements
rather than commenting on his patches and making him re-submit them.

 -George

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel