Re: [PATCH v6 00/11] libxl: ocaml: improve the bindings

[David Scott <dave.scott@eu.citrix.com>]

On 10/12/13 15:42, George Dunlap wrote:
> On Tue, Dec 10, 2013 at 2:34 PM, David Scott <dave.scott@eu.citrix.com> wrote:
>> On 10/12/13 14:10, George Dunlap wrote:
>>>
>>> On 12/10/2013 01:20 PM, Ian Campbell wrote:
>>>> I think the arguments made there still stand, in short it would be
>>>> awesome if xapi could move to using libxl on top of 4.4 and the risks
>>>> are almost entirely contained within this use case, which cannot be
>>>> satisfied by the code as it stands today.
>>>
>>>
>>> Except that that basically calls into question what a "code freeze" is
>>> at all.  At some point we just need to say, "No more, this is what we
>>> have; from now on we work on bug fixes."
>>>
>>> We've decided that PVH dom0 and ARM "physical address space leak" fixes
>>> are blockers for strategic reasons.  Is there a good reason that we
>>> should consider updated OCaml bindings in the same light?
>>>
>>> At this point, the fact that there is only one downstream user
>>> (XenServer) is an argument *against* its inclusion: there is very little
>>> benefit, as XS can simply carry the patches if they want to.
>>
>>
>> A nit-pick:
>
> Not exactly. ;-)
>
>> the downstream user is really 'xenopsd', part of the xapi
>> project. The xapi/xenopsd code is in XenServer and, increasingly, available
>> for other Linux distros (we're trying to do the right thing and make the
>> code easy to package). XenServer could easily carry some patches, but the
>> other distros probably won't. The only workarounds to keep xapi/xenopsd
>> working on non-XenServer distros that I can think of are (i) not using libxl
>> at all [a shame, obviously]; or (ii) depending on a fresh package,
>> 'libxl-ocaml-bindings-fixed' which would be a fork of the in-tree code with
>> the fixes applied and named 'xenlight2' [ugly, not totally sure if it's even
>> possible]. It seems odd to me to decide to ship code which the only user
>> can't actually use ;-)
>
> Right -- so you are arguing that there is in fact a strategic reason
> to get this into 4.4: you want xapi to be able to be easy to package
> and install into distros, and having broken ocaml bindings is a major
> blocker for that.  xapi packages for distros are already in a pretty
> dire state, and not having support until 4.5 could be disastrous.
>
> Would you agree with that assessment?

Yes, I think that's fair.

Thanks,
Dave