Re: Fwd: FOSDEM 2019 - Microkernels developer room - CfP

[Samuel Thibault]

Joshua Branson, le jeu. 29 nov. 2018 08:11:05 -0500, a ecrit:
> Samuel Thibault <samuel.thibault@gnu.org> writes:
> > Almudena Garcia, le mer. 28 nov. 2018 14:24:42 +0100, a ecrit:
> >> why not yours? You use to talk about Hurd in FOSDEM every year.
> >
> > I could, sure, I just don't know what I could talk about, and it's good
> > to see various people take the microphone :)
> 
> Just to help you brain storm...And forgive me if these are silly ideas.

Well, these are not silly ideas at all, but for a talk one needs to
have a main story to tell, not only a collection of small stories, and
I'm afraid I have already covered what I can talk about (flexible way
to have hardware support, flexibility for the user), and Justus talked
about virtualization two years ago.

> Maybe you could demo [...]

Demos take a very long time to make them work. Unless people help me
with it (understand: give me a step-by-step process that I'll be able to
just follow) I won't be able to get the time to make them work.

> Perhaps you could outline a roadmap.

Mmm. In the end that's what I could think of from your ideas. I have
thus proposed:

“
A roadmap for the Hurd?

Most people don't realize it, but the Hurd system is actually well
established. Between 75% and 80% of Debian official packages do build
fine, it has mainstream gcc/glibc/llvm support, go and rust ports are
ongoing, it can be installed with the Debian installer and GuixSD and
Arch ports are ongoing...

Yet not so much has been happening within the Hurd itself in the past
couple of years. We have notably added a PCI arbiter, which allows for
both flexible and safe PCI access for end users, and some basic ACPI
support is ongoing. But many exciting features could be achieved with a
bit of work.

This talk will discuss some of these promising features, to give a sort
of ideas roadmap for contributions. Some have implementation sketches
which just need to be polished to be more production-ready, such as
httpfs, mboxfs, or writing translators in more high-level languages
than C. Other features are at early stage, such as adding sound support
through rump, getting complete rid of disk drivers from the kernel by
moving them to userland, or also getting valgrind support. I will also
discuss some promising ideas, such as using libguestfs to get support
for more filesystems.
”

Now, that being written, I very well remember feedback from previous
talks. Basically the question was "Cool project! How do I start?" I
would point to the contribution page of the wiki, and notably its "Small
hack entries", but apparently it is way not detailed enough for people
to give them a try. Obviously, ideally enough I'd have time to develop
them, but I simply don't have it. I mean, for each of these items,
giving more details is not just a matter of writing them, but looking
around what needs to be done, where, which RPCs, etc. I don't have
enough time for that. So, while I'll be happy to give this talk, unless
the community takes a stab at reworking the contribution page into
workable items, I'm afraid it will not drag much more than the previous
years.

Samuel