Mailing List Archive

Re: [Minios-devel] [MirageOS-devel] [RFC] Unicore Subproject Proposal
Hello,

Anil Madhavapeddy, on jeu. 14 sept. 2017 15:38:54 +0100, wrote:
> On 13 Sep 2017, at 17:13, Samuel Thibault <samuel.thibault@ens-lyon.org> wrote:
> > Anil Madhavapeddy, on mer. 13 sept. 2017 11:11:03 +0100, wrote:
> >> Maintaining a forked MiniOS has been a multi-year source of a maintenance burden for MirageOS,
> >
> > I'm just wondering why this happened?
> >
> > The mini-os repository is open for development, it's just a matter of
> > agreeing on how to implement features :)
>
> We forked it well before mini-os spun out into a separate repository,
> around 5 years ago.

Ok :)

> It's a combination of both I think. I had a very quick look at the latest mini-os
> tree and ran into build problems from the master branch

Ok, so it's just a matter of cleanly working on it.

Samuel

_______________________________________________
Xen-api mailing list
Xen-api@lists.xen.org
https://lists.xen.org/cgi-bin/mailman/listinfo/xen-api
Re: [Minios-devel] [MirageOS-devel] [RFC] Unicore Subproject Proposal [ In reply to ]
Hey Samuel,

On 14.09.2017 16:47, Samuel Thibault wrote:
> Hello,
>
> Anil Madhavapeddy, on jeu. 14 sept. 2017 15:38:54 +0100, wrote:
>> On 13 Sep 2017, at 17:13, Samuel Thibault <samuel.thibault@ens-lyon.org> wrote:
>>> Anil Madhavapeddy, on mer. 13 sept. 2017 11:11:03 +0100, wrote:
>>>> Maintaining a forked MiniOS has been a multi-year source of a maintenance burden for MirageOS,
>>>
>>> I'm just wondering why this happened?
>>>
>>> The mini-os repository is open for development, it's just a matter of
>>> agreeing on how to implement features :)
>>
>> We forked it well before mini-os spun out into a separate repository,
>> around 5 years ago.

We did the same for our ClickOS, MiniCache and Minipython Unikernel: We
used our own fork of MiniOS (4-5 years ago). However, while working on
it, we noticed that there is much more to do than just introducing some
switches and add new features.

We add several functionality to our MiniOS tree (replaced network I/O,
TSO feature, persistent grants for block, select/poll support, proper
heap memory allocation, hooks into low-level code), but support for more
hypervisor stayed forever on our wishlist. We concluded that you should
either intrusively clean-up and reorganize the Mini-OS code (which we
think is non-trivial to do) or you start with a new project and port the
pieces from MiniOS that you need (which is clearly lots of work, too).

We decided for the latter because it also forces us to rethink about
internal APIs and code organization. A proper design should also
simplify adding support for more hypervisors. And because we understand
the needs of Unikernels better today ;-), I think we should consider
this as a rare chance and should make use of it.

>
> Ok :)
>
>> It's a combination of both I think. I had a very quick look at the latest mini-os
>> tree and ran into build problems from the master branch
>
> Ok, so it's just a matter of cleanly working on it.
>
> Samuel
>

Thanks,

Simon

--
============================================================
Simon Kuenzer
??? ??????
Research Scientist,
Networked Systems and Data Analytics Group
NEC Laboratories Europe, Network Research Division
Kurfuerstenanlage 36, D-69115 Heidelberg
Tel. +49 (0)6221 4342-264
Fax: +49 (0)6221 4342-5264
e-mail: simon.kuenzer@neclab.eu
============================================================
NEC Europe Ltd | Registered Office: Athene, Odyssey
Business Park, West End Road, London, HA4 6QE, GB
Registered in England 2832014

_______________________________________________
Xen-api mailing list
Xen-api@lists.xen.org
https://lists.xen.org/cgi-bin/mailman/listinfo/xen-api